Sharon R. Steadman - John Gregory McMahon - Archaeology of Anatolia, Volume IV - Recent Discoveries (2018 - 2020) (2021)

The Archaeology of
Anatolia, Volum IV
Recent Discoveries (2018-2020)
Edited by
Sharon R. Steadman
Gregory McMahon
The Archaeology of Anatolia,
Volume IV
The Archaeology of Anatolia,
Volume IV:
Edited by
Sharon R. Steadman and Gregory McMahon
Cambridge
Scholars
Publishing
The Archaeology ofAnatolia, Volume IV: Recent Discoveries (2018-2020)
Edited by Sharon R. Steadman and Gregory McMahon
This book first published 2021
Cambridge Scholars Publishing
Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK
British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library
Copyright@ 2021 by Sharon R. Steadman, Gregory McMahon and contributors
All rights for this book reserved. No part of this book may be reproduced, stored in a retrieval system, or
transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise,
without the prior permission ofthe copyright owner.
ISBN (10): 1-5275-7601-9

ISBN (13): 978-1-5275-7601-8
TABLE OF CONTENTS
List of Figures ...................................................................................................................................................... viii
List of Tables ........................................................................................................................................................ xiv
Chapter One ............................................................................................................................................................. 1

Introduction to the Archaeology ofAnatolia: Volume IV
Sharon R. Steadman and Gregory McMahon
Part I: Excavations
Chapter Two ............................................................................................................................................................ 4

Ambar Dam Salvage Excavations 2018-2020: Ambar Hoyiik, Gre Filla, and Kendale Hecala
A. TubaOkse
Chapter Three ........................................................................................................................................................ 21

Gokr;eada Ugurlu: A Preliminary Report on the 2018-2020 Field Seasons
Bun;in Erdogu, <;igdem Atakuman, and Nejat Yucel
Chapter Four.......................................................................................................................................................... 33
The Inner Western Anatolian Prehistoric Period Porsuk Culture:
The Kanht~ HoyiikProject 2018 Final Excavation Report
Ali Umut Turkcan and Duygu Ertemin
Chapter Five .......................................................................................................................................................... 45

Kulaks1zlar Revisited: Excavating the Contexts ofa Cha/eolithic Marble Workshop
Turan Takaoglu
Chapter Six ............................................................................................................................................................ 56

Excavations at U~akh Hoyiik: Recent Results
Anacleto D' Agostino, Stefania Mazzoni, and Valentina Orsi
Chapter Seven ....................................................................................................................................................... 75

A Fresh Perspective on the Middle Bronze Age at Tell Atchana, Alalakh: The 2007-2019 Seasons
Murat Akar, K. Ashhan Yener, Miige Bulu, and Tara Ingman
Chapter Eight. ........................................................................................................................................................ 96

2015-2020 Excavations at Elbistan Karahoyiik and Hamzatepe:
A Hittite Settlement and its Cemetery in the Elbistan Plain
Bora Uysal and Ali <;if9i
Chapter Nine ....................................................................................................................................................... 108

Sirkeli Hoyiik: Insights into the Archaeology ofBronze and Iron Age Cilicia
Mirko Novak, Deniz Y ~in, Mirco Brunner, Sinem Hac10smanoglu,
Ekin Koza!, Sabina Kulemann-Ossen, Hannah Monninghoff, Susanne Rutishauser,
Alexander B. Sallee, and Sonke Szidat
Chapter Ten ......................................................................................................................................................... 125

Iron Age Pottery from Oluz Hoyiik, North Central Anatolia
Mona Saba
Chapter Eleven .................................................................................................................................................... 132

Recent Work (2018-2019) at Porsuk-Zeyve Hoyiik in Southern Cappadocia
Claire Barat, Emine Koker Gok9e, Jean-Fran9ois Pichonneau,
Vivien Mathe, and Guillaume Bruniaux
vi Table of Contents
Chapter Twelve ................................................................................................................................................... 146

The Kerkenes Project 2019- 2020
Scott Branting, Joseph W. Lehner, Sevil Baltah Trrpan, Dominique Langis-Barsetti,
Sarah R. Graff, Tuna Kalayc1, Amanda Groff, Lucas Proctor, Jessica Robkin,
Paige Paulsen, and Niliifer Baturayoglu Yoney
Chapter Thirteen .................................................................................................................................................. 155

Sinop Kalesi Archaeological Excavations: The 2017 and 2019 Field Seasons
Owen Doonan, Andrew Goldman, Alexander Bauer, Ulrike Krotscheck, Paolo Maranzana,
Nicole Rose, Jane Rempel, Susan Sherratt, Emine Sokmen, and Huseyin Vural
Chapter Fourteen ................................................................................................................................................. 170

An Analysis ofthe Archaeological and Textual Data from the Podium Hall at Ayanis:
Is this the Sanctuary ofthe Urartian Goddess 'Arubaini?
Mehmet I~lkh and Kenan ¼1k
Appendix: Preliminary Results ofRadiocarbon Dating on Ayanis Castle
Mehmet I~lkh, Alper ~engul, and Oguz Aras
Chapter Fifteen .................................................................................................................................................... 194

The Tuspa Citadel and Mound Excavations, 2010- 2019
Billent Gem;:, Erkan Konyar, and Armagan Tan
Chapter Sixteen ................................................................................................................................................... 210

Hierapolis ofPhrygia 2018-2020: Recent Work and New Perspectives
Grazia Semeraro
Chapter Seventeen ............................................................................................................................................... 229

Recent Discoveries at Komana (2018-2020): The Late Iron Age/Early Hellenistic Levels
D. Burcu Erciyas, Ay~e Batman, and Ashhan <;::ivildag
Chapter Eighteen ................................................................................................................................................. 244

Digging in Doliche 2018-2020
Michael Blomer
Chapter Nineteen ................................................................................................................................................. 256

Roman Bogazkoy: New Discoveries in the Northern Lower City
Dominique Kruger and Sven Kuhn
Chapter Twenty ................................................................................................................................................... 267

Archaeological research at Elaiussa Sebaste (Rough Cilicia) in 2019
Annalisa Polosa
Part II: Surveys
Chapter Twenty-One ........................................................................................................................................... 276

The 2018-2019 Archaeological Survey Results on the Location ofSettlements
Connecting the Mountain Section ofGiresun to Central Anatolia
Salih Kaymak91
Chapter Twenty-Two .......................................................................................................................................... 287

Mountains as Connected Landscapes ofAlterity: Boz Mountain Range and its Piedmont
Omur Harman~ah, Peri Johnson, Shannon Martino,
Muge Durusu-Tannover, and Ben Marsh
Chapter Twenty-Three ........................................................................................................................................ 300

New Assessments ofSettlement Distribution during the Bronze and Iron Ages in Southeast Konya
<;::igdem Maner
Chapter Twenty-Four .......................................................................................................................................... 312

Get Going with Survey Legacy Data. The 2018 and 2019 Archaeological Survey
Results from the Sagalassos Archaeological Research Project
Ralf Vandam, Patrick T. Willett, Christopher Carleton, and Jeroen Poblome
The Archaeology of Anatolia Volume IV: vii
Chapter Twenty-Five ........................................................................................................................................... 322

The Results ofthe 2019 Gargara Archaeological Survey in the South Troad
Nurettin Arslan and Caner Bakan
Part III: The State of the Field
Chapter Twenty-Six ............................................................................................................................................ 338

Archaeobotany in Anatolia
John M. Marston and Lorenzo Castellano
Chapter Twenty-Seven ........................................................................................................................................ 355

From Bones to Genomes: Current Research on the Population History ofPrehistoric Anatolia
Duygu Deniz Kazanc1, N. Ezgi Alt1m~1k, Ayc;a Aydogan, Fusun Ozer,
Elif Surer, <;:'.igdem Atakuman, Mehmet Somel, and Y1lmaz Selim Erda!
Chapter Twenty-Eight ......................................................................................................................................... 372

Cha/eolithic Religion and Ritual on the Anatolian Plateau
Burcu Y1ldmm and Sharon R. Steadman
Contributors to the Volume ................................................................................................................................. 396
Index.................................................................................................................................................................... 406
LIST OF FIGURES
Figure 2-1. Location of Ambar Dam.

Figure 2-2. Ambar Hoyiik Middle Bronze.
Figure 2-3. Chronological Table of Neolithic Sites Based on Calibrated Dates.
Figure 2-4. Gre Filla, General Drone Photo.
Figure 2-5. Gre Filla Late Antique cemetery.
Figure 2-6. Gre Filla Late Antique cemetery small finds.
Figure 2-7. Gre Filla, schematic plan of Pre-Pottery Neolithic B buildings.
Figure 2-8. Gre Filla architecture.
Figure 2-9. Gre Filla PPNB small finds.
Figure 2-10. Kendale Hecala architecture.
Figure 2-11. Kendale Hecala, schematic plan of building levels.
Figure 2-12. Kendale Hecala small finds.
F igure 2-13. Kendale Hecala, Pottery Neolithic period small finds, pottery, and ground stone objects.
Figure 3-1. The location of Ugurlu in the northeastern Aegean region.
Figure 3-2. Pre-pottery Neolithic beads made from nephrite and agate.
Figure 3-3. The semi-subterranean oval hut-like structure in Phase IV.
Figure 3-4. The terrazzo platform.
Figure 3-5. The early pottery at Ugurlu.
Figure 3-6. Monumental architecture of the Pottery Neolithic Phase V.
Figure 3-7. A pyramid-shaped amulet made from Spondylus.
Figure 3-8. Some of the objects found in the ritual pits.
Figure 3-9. "Eared pots" unique to Goks:eada.
Figure 3-10. The burial pit, ca. 5300 BCE.
Figure 3-11. Buildings 11 and 12 located at the highest point of the settlement.
Figure 3-12. Plan of Building 12.
Figure 3-13. A unique sherd with incised deer motif.
Figure 3-14. A building complex featuring separate buildings built adjacent to each other.
Figure 4-1. Regional map with the sites mentioned in the text and the position of Kanht~ Hoyiik.
Figure 4-2. View of Kanht~ Hoyiik in Kuzfmdik Valley from the southeast.
Figure 4-3. Kanht~ Hoyiik site plan of the Chalcolithic hilltop.
Figure 4-4. Trench M-17, Space 18 orthophoto.
Figure 4-5. Mudbrick wall found in Space 18 in the deep sounding.
Figure 4-6. Fire installation found in the deep sounding.
Figure 4-7. Trench 0-16 Space 8 orthophoto.
Figure 4-8. Harris Matrix showing the building 1 and building 2 spaces.
Figure 4-9. Trench N-16, Space 5, part of building 1.
Figure 4-10. Kanht~ Hoyiik selected bone tool typology.
Figure 4-11: Examples from the Early Bronze Age pottery assemblage.
Figure 4-12: Chipped stone industry data from the 2018 excavation season.
Figure 5-1. Map showing the location ofKulaksizlar and other major sites with Chalcolithic finds.
Figure 5-2. View ofKulaksizlar showing the area of excavation, from the south.
Figure 5-3. Fragments of bowls and jars found along with marble objects.
Figure 5-4. Large container with a spout towards the bottom.
Figure 5-5. Partial plan of Sector 1 where workshops are clustered.
Figure 5-6. View of Structure A from the west.
Figure 5-7. A preform of a conical rhyton.
Figure 5-8. Sandstone drill-bits of various diameters.
Figure 5-9. Fragments of nearly finished Kilia figurines broken off at the neck or at the waist.
Figure 5-10. Fragments of nearly finished miniature Kilia figurines.
Figure 5-11. Selected examples of obsidian artifacts from Melos, and the central Anatolian source of Golliidag.
Figure 5-12. Miniature jadeite axe.
Figure 5-13. Fragment of two broken mace heads of black and beige porphyry.
Figure 5-14. Copper pin and a copper slag.
Figure 6-1. Location ofU~akh Hoyiik.
Figure 6-2. Area D: Building III, preliminary plan.
Figure 6-3. Area D; the central portion of Building III (2016).
The Archaeology of Anatolia Volume IV: ix
Figure 6-4. Area D: structures dating to the Middle Iron Age.

Figure 6-5. Area D: the wall/glacis on the southern slope of the hoytik (2016).
Figure 6-6. Area A: Building II (with some structures discussed in the chapter marked).
Figure 6-7. Area A: polychrome stone mosaic floor of Building II.
Figure 6-8. Middle Iron Age pottery from area D, Unit 218.
Figure 6-9. Late Middle Iron Age and Late Iron Age pottery from Area D, Unit 369.
Figure 6-10. Late Iron Age and Hellenistic Period Pottery from Area A, Unit 285.
Figure 7-1. Map showing major second millennium BC Anatolian and Near Eastern sites.
Figure 7-2. Map showing Tell Tayinat (AS 126), Tayinat al-Saghir (AS 127), and Tell Atchana.
Figure 7-3. Map of Tell Atchana.
Figure 7-4. Aerial view of Square 32.57, Local Phase 5g Apsidal Building in Area 1.
Figure 7-5. Plan showing the remains of the earlier palace in Squares 33.32 and 33.53.
Figure 7-6. Aerial view of Square 33.53.
Figure 7-7. AT26395, wall painting fragment from Square 33.53, Local Phase 2.
Figure 7-8. AT26620, wall painting fragment from Square 33.53, Local Phase 2.
Figure 7-9. AT26367, watt painting fragment from Square 33.53, Local Phase 2.
Figure 7-10. Domestic quarters, MBA fortification wall, and the slope cemetery in Area 3, Square 45.44.
Figure 7-11. A selection of burial vessels from the Area 3 cemetery.
Figure 7-12. Selected late MBA ceramics.
Figure 7-13. Selected late MBA ceramics.
Figure 7-14. AT25415. l, Kamares Ware fragment from Square 33.53, Local Phase 3.
Figure 7-15. MBA radiocarbon dates from Tell Atchana.
Figure 8-1. Map and aerial view showing the location of Elbistan Karahoytik and Hamzatepe.
Figure 8-2. Topographic plan ofElbistan Karahoytik.
Figure 8-3. Architectural remains from the 1st layer at the Northwest Slope area of Elbistan Karahoytik.
Figure 8-4. Architectural layers from the Northern Terrace Area at Elbistan Karahoylik:.
Figure 8-5. Cylinder seal from the 4th layer at the Northern Terrace Area at Elbistan Karahoytik.
Figure 8-6. Architectural remains and bronze cauldron from the Northern Terrace Area at Elbistan Karahoylik:.
Figure 8-7. Stamp seal, button seal, and spindle bottle from the Northern Terrace Area at Elbistan Karahoytik
Figure 8-8. The relief-decorated Hittite ceramics from the Northern Terrace Area at Elbistan Karahoytik.
Figure 8-9. Architectural remains of the 8th layer from the Northern Terrace Area at Elbistan Karahoytik.
Figure 8-10. Topographic plan ofHamzatepe.
Figure 8-11. Hamzatepe excavation area.
Figure 8-12. Hamzatepe urns, numbers 16, 17, and 29.
Figure 8-13. Hamzatepe cremation area and stone-surrounded grave.
Figure 8-14. The finds from the stone-surrounded grave at Hamzatepe.
Figure 8-15. Hamzatepe White Painted vessels.
Figure 9-1. Central places in Cilicia during Bronze and Iron Age with position of Sirkeli Hoytik.
Figure 9-2. Plan of Sirkeli Hoytik with excavation sectors.
Figure 9-3. Rock Relief ofMuwattalli II.
Figure 9-4. KDE ofradiocarbon data of different phases from Sirkeli Hoytik.
Figure 9-5. Ortho-image of Sector A with Building Al.
Figure 9-6. Ortho-image of Sector E with Building E1.
Figure 9-7. Ortho-image of Sector D with Building Dl.
Figure 9-8. Ortho-image of Sector F.
Figure 9-9. View of building 01 at the end of the 2019 campaign.
Figure 9-10. Ortho-image of building 01 with nearby street.
Figure 9-11. View of tombs 2-4.
Figure 9-12. OCI pottery.
Figure 9-13. MCI pottery.
Figure 9-14. NCI pottery.
Figure 9-15. Sirkeli Hoytik during the Period NCI (1190-330 BCE).
Figure 10-1. General view of Oluz Hoytik; Map of Iron Age settlements in the KIZllrrmak basin.
Figure 10-2. Dark Age sherds from Oluz Hoytik.
Figure 10-3. Potsherd, early Middle Iron Age, Oluz Hoylik:.
Figure 10-4. Potsherd, early Middle Iron Age, Oluz Hoylik:.
Figure 10-5. Potsherd, Late Iron Age, Oluz Hoytik.
Figure 10-8. Potsherd, Late Iron Age, otuz Hoytik.
Figure 10-9. Cult vessel, Late Iron Age, Oluz Hoylik:.
Figure 10-10. Potsherd, early Middle Iron Age, Oluz Hoytik.
X List of Figures
Figure 10-12. Potsherd, Middle Iron Age, Oluz Hoyuk.

Figure 10-13. Krater fragment, Late Iron Age, Oluz Hoyuk.
Figure 11-1. Map of the archaeological site in the Nigde province.
Figure 11-2. Aerial view of the Zeyve Hoyuk site.
Figure 11-3. Map of the localization of the different "Chantiers" ( excavations areas) at Zeyve Hoyuk.
Figure 11-4. Magnetic anomaly map of Porsuk-Zeyve Hoytik.
Figure 11-5. Identification of the main anomalies visible on the magnetic anomaly map.
Figure 11-6. Map of Chantier 2 and Zone 5.
Figure 11-7. Consolidation works, Chantier 2.
Figure 11-8. Mudbrick workshop.
Figure 11-9. Aerial view, zone 5 (Jean-Fraw;:ois Pichonneau).
Figure 11-10. General plan of zone 5.
Figure 12-1. Map of the location of Urban Block 8, the Kale, and the Cappadocia Gate within Kerkenes.
Figure 12-2. View to the north ofKerkenes.
Figure 12-3. The Kerkenes team in 2020 during the COVID-19 pandemic.
Figure 12-4. Gradiometry survey on top ofthe Kale in 2019.
Figure 12-5. Configuration of trenches within Urban Block 8.
Figure 12-6. Aerial photograph of the extents ofTrench45 in 2019, outlined in white dots.
Figure 12-7. A fragmentary ivory plaque from Trench 45.
Figure 12-8. Jn situ pottery vessels being excavated within Trench 45.
Figure 12-9. The Leica 3D RTC360 laser scanner being used to scan the Cappadocia Gate.
Figure 12-10. Structured-light scanning of a pottery vessel with the Artec Spider in the excavation depot.
Figure 13-1. Plan of Sinop Kale Excavation trenches, 2017.
Figure 13-2. Drone view, Ops. 1 and 4, indicating location of Phase 7 and 6 features.
Figure 13-3. SK Phase 7 wall (Locus 75) and SK Phase 6 wall (Locus 74) in Op. 4.
Figure 13-4. Section drawing showing interrelationship of SK Phases 5- 8.
Figure 13-5. River-worn stone glacis, SK Phase 6.
Figure 13-6. Deposit of ceramics from Op. 1.
Figure 13-7. SK 16384, reconstructed vase and drawing; Hamsi bones.
Figure 13-8. Early Bronze Age figurines (SK 16937 and SK 17089).
Figure 13-9. Section from Op. 2 indicating stratigraphy of the construction fills, Hellenistic wall.
Figure 13-10. Photo and profile of West Slope ware sherd, SK 16331.
Figure 13-11. Session plan of the fortification produced by FARO Scene.
Figure 13-12. An overview of the scans taken by the survey.
Figure 13-13. Autodesk Recap measurements of the south wall of the fortification and the northern part of the wall.
Figure 13-14. pXRF 2016 DataK-Means Analysis.
Figure 13-15. pXRF 2019 Data K-Means Analysis.
Figure 14-1. General view of Ayanis Citadel and Stiphan (Eiduru?) Mountain.
Figure 14-2. Plan of the Temple Complex; Reconstruction of the temple complex at Ayanis.
Figure 14-3. General view of the Podium Hall and back room.
Figure 14-4. Reconstruction of the Podium Hall and back room.
Figure 14-5. The Onyx Podium with decoration in the hall, and its details.
Figure 14-6. The scene of offering beads to the goddess 'Arubaini (?).
Figure 14-7. Beads which were recovered from the back room of the Podium Hall.
Figure 14-8. The inscriptions on the lateral surface of the bulla.
Figure 14-9. The bulla with female names and its impression (lnv.2016 458-A).
Figure 14-10. A bulla without inscription and its impressions/drawing.
Figure 14-11. Drawing of impression of the stamp seal from Toprakkale.
Figure 14-12. Photos and drawings of Cretulae fragments.
Figure 14-13. Stamp seal impression of Cretula; Drawing of stamp seal impression of Cretula.
Figure 14-14. General and plan view of Ayanis Castle and bedrock, view from the south.
Figure 14-15. Location of radiocarbon samples on drawing of Ayanis Castle.
Figure 14-16. Simplified geological map of eastern Lake Van, epicenter distribution of earthquakes.
Figure 14-17. Main and aftershock distribution of Van earthquake and historical earthquakes around Lake Van.
Figure 14-18. Collapsed walls in Temple after Van Earthquake; Extension and cracks on the basaltic stone columns.
Figure 14-19. Collapsed mudbrick wall in the Southern Curtain Wall; Extension crack
Figure 14-20. Location ofradiocarbon samples on aerial photo in detail.
Figure 15-1. Photo of the Tuspa Citadel.
Figure 15-2. Plan showing the Building I and Building II areas.
Figure 15-3. Plan of the entire Tuspa Citadel showing areas described in the text.
Figure 15-4. Underground Chan1ber Grave BG90; The BG90 Rock Tomb.
Figure 15-5. Rock steps connecting the tomb with the citadel above; The Tuspa Fountain.
Figure 15-6. Inside the "Minua sirsini."
The Archaeology of Anatolia Volume IV: xi
Figure 15-7. Tuspa Mound during excavation.

Figure 15-8. The Medieval tombs from TuspaMound.
Figure 15-9. Pottery from Late Iron Age/Post-Urartian layers; Important finds.
Figure 15-10. Hocker burial from the Late Iron Age/Post Urartian layers; Grave 5.
Figure 15-11. Plan showing the Urartian period layers.
Figure 16-12. Photo of Late and Early Urartian walls at Tuspa Mound.
Figure 15-13. The Columned Hall in the Urartian complex; Urartian tablet from the mound.
Figure 15-14. The Early Iron Age pits on the mound.
Figure 15-15. The Early Bronze Age Layers; An andiron from the Early Bronze Age layer.
Figure 16-1. City plan.
Figure 16-2. Insula 104, residential area. Location of the new excavation essays.
Figure 16-3. North Necropolis. Tomb 55, exedra.
Figure 16-4. Plan of the Sanctuary of Apollo, investigations 2018-2020.
Figure 16-5. South area ofNymphaeum. View from a drone; Sondage no. 7 between the terracing wall.
Figure 16-6. South Portico, essay no. 4.
Figure 16-7. Overall plan of the sacrificial area (excavations 2002-2006, 2018).
Figure 16-8. Reconstructive plan of the Nymphaeum.
Figure 16-9. Fault systems across the Sanctuary of Apollo and the Ploutonion.
Figure 16-10. Fragment of a statue from the north portico.
Figure 16-11. Seismic fracture in the South portico.
Figure 16-12. Nymphaeum, north side: foundation arch above the fault.
Figure 16-13. Phrygian necropolis: general plan; Eastern and W estem area of the Phrygian necropolis.
Figure 16-14. Phrygian necropolis: pyre; Rectangular monument: collapse.
Figure 16-15. Phrygian Pottery: jar with geometric decoration; Cooking pot; small mugs; high-footed cups.
Figure 16-16. Fragments of the statue of Athena from the Sanctuary of Apollo; Byzantine unguentaria.
Figure 16-17. Overall view of the Ploutonion (2020).
Figure 17-1. Location of Komana on map.
Figure 17-2. Hamamtepe excavations, aerial view.
Figure 17-3. Hamamtepe excavation areas.
Figure 17-4. Plan drawing of the Hellenistic walls with Early Byzantine walls; Aerial view of the Hellenistic walls.
Figure 17-5. Scarab-shaped stamp seal with Thutmose III (1479- 1429 BCE), Menkhepherre.
Figure 17-6. Silver coin depicting Hera/owls; Bronze coin (Knossos) with labyrinth on reverse.
Figure 17-7. Terracotta figurine and bronze fibula.
Figure 17-8. Three polychrome pitchers from HTP0l, Late Iron Age.
Figure 17-9. Examples of Banded Wares, Komana.
Figure 17-10. Sherds with ivy garlands, Komana.
Figure 17-11. Examples of the imported pottery, Komana.
Figure 17-12. Examples of Hellenistic Colour Coated Ware-A, Komana.
Figure 17-13. Hellenistic Colour Coated Ware-A, Komana.
Figure 18-1. Map of ancient North Syria.
Figure 18-2. View of the hill (Keber Tepe) that was the centre of the urban area ofDoliche.
Figure 18-3. Map ofKeber Tepe (Doliche).
Figure 18-4. Top: Doliche, Map of excavation area field 415.
Figure 18-5. Doliche, Roman baths, room with cold water swimming pool.
Figure 18-6. Doliche, Roman baths, test trench with pilae in hypocaust.
Figure 18-7. Collection of seal impressions from the city archive at Doliche.
Figure 18-8. Doliche, field 417, trenches 17-04/19-07, city archive, remains of foundation walls.
Figure 18-9. Doliche, Terrace church, schematic plan with reconstruction.
Figure 18-10. Aerial view of the terrace church and adjacent buildings in 2020.
Figure 18-11. Doliche, Terrace church, detail of mosaic floor in the nave and northeastern comer of bema.
Figure 18-12. Doliche, Terrace church, Apse with 6th century CE mosaic.
Figure 18-13. Doliche, Terrace church, Northern annex room.
Figure 19-1. The Roman road and settlements in the vicinity ofljattusa.
Figure 19-2. Plan of the Roman structures.
Figure 19-3. Aerial view of the excavation area of the assumed tower, 2018.
Figure 19-4. View of the inner face of the Roman fortification wall.
Figure 19-5. Aerial view of the area at the modem road.
Figure 19-6. Aerial view of the building at M1hraphkaya with the walls of the water basin marked in blue.
Figure 19-7. View of the excavation area in front ofM1hraphkaya.
Figure 19-8. Reconstruction of the building west ofMlhraphkaya.
Figure 19-9. Restored panel of the wall painting from the banquet room.
Figure 19-10. Aerial view of the villa with baths.
Figure 19-11. Mosaic floor.
xii List of Figures
Figure 20-1. General plan of Elaiussa.

Figure 20-2. View of the southern necropolis.
Figure 20-3. Aerial view of the southern necropolis.
Figure 20-4. Aerial view of the northern promontory.
Figure 20-5. Plan of the northern promontory architecture.
Figure 20-6. View of the polygonal wall on the northern promontory.
Figure 20-7. Aerial view of the wall that cuts the wave motif floor mosaic on the northern promontory.
Figure 20-8. GIS map showing the enhancements based on recent work.
Figure 20-9. Plan and photo of the Basilica of the Great Baths, narthex and southern nave.
Figure 20-10. Basilica of the Great Baths.
Figure 21-1. Settlements identified in Sebinkarahisar district during the 2018-2019 survey years.
Figure 21-2: Distribution of the settlements in Sebinkarahisar district, identified in the 2018- 2019 seasons.
Figure 21-3. Ahurcuk Fortress.
Figure 21-4. Golctik settlement.
Figure 21-5. Golctik settlement: ceramics from Hellenistic and Roman periods.
Figure 21-6. Golctik settlement: bead from Roman period.
Figure 21-7. Kabaktepe settlement.
Figure 21-8. Kabaktepe settlement: Hellenistic-Roman ceramics.
Figure 21-9. Kabalctepe settlement: Hellenistic-Roman glass handle.
Figure 21-10. Kiliseta§ settlement.
Figure 21-11. Kiliseta§ settlement: ceramics from the Early Bronze Age and the 2nd millennium BCE.
Figure 21-12. Kiliseta§ settlement: a ceramic piece from the Early Bronze Age.
Figure 21-13. Dikmentepe Fortress and Sebinkarahisar Fortress.
Figure 21-14. Paste and slip color variations of ceramics encountered in the 2018-2019 seasons.
Figure 22-1. Map ofYalburt Yaylas1 Archaeological Landscape Research Project.
Figure 22-2. Map of the western Boz Mountain Range with routes discussed in the text.
Figure 22-3. Map of the geomorphological zones in the western Boz Mountain Range.
Figure 22-4. Map of the incised Bulcuk and Kempos Valleys.
Figure 22-5. Photograph of the documentation of the Byzantine tombstone at a Bulcuk spring.
Figure 23-6. Map of the eroded piedmont.
Figure 22-7. Map of the Tekke Valley.
Figure 22-8. Photograph of the middle promontory of Akkaya Mevkii from the west.
Figure 22-9. Ceramic stamp seal from the middle promontory (locus 3) of Akkaya Mevkii.
Figure 23-1. Map showing surveyed sites of the 2019 field season.
Figure 23-3. Meke Golu.
Figure 23-4. Gokgoz Tepesi.
Figure 23-5. Early, Middle, and Late Iron Age pottery from Sm~an Hoyuk.
Figure 23-6. Rock massif with foundation of a wall to its eastern side at Eski Belkaya Y erle§imi.
Figure 23-7. Boulder with square lewis holes at Eski Belkaya Yerle~imi.
Figure 23-8. The newly discovered hieroglyphic Luwian inscription at Karaoren (Karaoren 2).
Figure 23-9. G1can Kalesi.
Figure 23-10. Middle and Late Bronze Age and Iron Age pottery from G1can Kalesi.
Figure 23-11. View from G1can Kalesi to the southwest towards Karapmar and Hotam1~.
Figure 23-12. Drone image ofDagoren.
Figure 23-13. Dagoren.
Figure 23-14. Ye~iltepe Hoyuk.
Figure 24-1. The study region of Sagalassos, showing all known sites and areas.
Figure 24-2. The aggregated LAMAP model for the study region of Sagalassos for seven archaeological periods.
Figure 24-3. Overview of the surveyed fields for the validation of the LAMAP model.
Figure 24-4. One of the thirteen house remains that were identified at CAM02.
Figure 24-5. Overview of all the investigated fields in 2019.
Figure 25-1. Map of the Troad.
Figure 25-2. Plan of ancient Gargara.
Figure 25-3. View ofKocakaya (Old Gargara), looking from west to east.
Figure 25-4. Kocadag mountain, Hill A Area.
Figure 25-5. Walls of the citadel.
Figure 25-6. Aerial view of the citadel.
Figure 25-7. Warrior relief.
Figure 25-8. Fragments of Aeolic capitals.
Figure 25-9. Architectural terracottas (sima style).
Figure 25-10. Architectural terracottas (geison tiles).
Figure 25-11. Architectural terracottas (geison tiles).
The Archaeology of Anatolia Volume IV: xiii
Figure 25-12. Ruins of a house.

Figure 25-13. a- b: Black-slipped ceramics of Athenian origin; c: Amphorae.
Figure 25-14. Fortification walls (P2).
Figure 25-15. A typical house in the rural surroundings of Gargara.
Figure 25-16. Classical period bronze coins of Gargara (Troia Museum, <;anakkale).
Figure 26-1. Map of sites with published archaeobotanical seed remains from Anatolia.
Figure 26-2. Map of20tli century annual precipitation in Anatolia.
Figure 26-3. Number of sites and number of archaeobotanical samples included in this dataset, by region and period.
Figure 26-4. Frequency and intensity of use of cereal, pulse, and oil and fiber taxa by period.
Figure 26-5. Frequency and intensity of use of the most common fruit taxa by period.
Figure 26-6. Relative abundance of cereal taxa based on summed RI scores by period.
Figure 26-7. Relative abundance of cereal taxa based on summed RI scores by macro-region and macro-period.
Figure 26-8. Relative abundance of pulse taxa based on summed RI scores by period.
Figure 26-9. Relative abundance of pulse taxa based on summed RI scores by macro-region and macro-period.
Figure 26-10. Relative abundance of fruit taxa based on summed RI scores by period.
Figure 26-11. Ubiquity of major fruit ta.xa for three macro-regions where fruits are best attested.
Figure 27-1. Geographic locations of the sites mentioned in this chapter.
Figure 27-2.Principal component analysis of the ancient genomes projected onto modem-day W Eurasian populations.
Figure 27-3. An illustration of population history in SW Asia.
Figure 28-1 Map of Anatolian Plateau showing sites included in this chapter.
Figure 28-2. Plan ofHac1lar IIA "shrines"; Canhasan I, Layer 2B.
Figure 28-3. Plan ofKurw;:ay Level 6A.
Figure 28-4. Plan ofKurw;;ay Level 3, showing Structure 56; Plan of Western Compound at <;ad1r Hoyi.ik.
Figure 28-5. Sketch of figurines from Hac1lar; Sketch of figurines head Canhasan.
Figure 28-6. Sketch of figurines from Gi.ivercinkayas1.
Figure 28-7. Animal figurine from <;adu Hoyi.ik; Stone object; Incised "cube."
Figure 28-8. Drawing of the andiron from the Omphalos Building at <;adu Hoyi.ik.
Figure 28-9. Sketch of vessels from Gi.ivercinkayas1, Ko~k Hoyi.ik, Tepecik <;iftlik.
Figure 28-10. Sketch of human and animal applique at Gi.ivercinkayas1, Tepecik <;iftlik.
Figure 28-11. Sketch of vessel in the form of a tortoise from Ko~k Hoyi.ik.
Figure 28-12. Photo of<;amhbel Tarlas1 II, Grave 16.
Figure 28-13. Photos of Late Chalcolithic infant jar burials at <;ad1r Hoyi.ik.
LIST OF TABLES
Table 3-1. Stratigraphy and Dates of Ugurlu

Table 4-1. Kanht~ Hoyillc cal. C 14 dating data.
Table 4-2. Overview of the construction phases.
Table 4-3.2018 season chipped stone industry data.
Table 5-1. Two radiocarbon dates from Kulaks1zlar.
Table 5-2. A tentative chronological chart of the western Anatolian littoral.
Table 9-1. Cilician Chronology.
Table 9-2. Samples and C14 dates.
Table 14-1. Results of radiocarbon dating (Please see Table 2 for details).
Table 14-2. Details of the radiocarbon dating.
Table 21-1. Settlements identified in the ~ebinkarahisar District in the 2018-2019 seasons.
Table 23-1. Surveyed settlements in 2019 and 2020.
Table 24-1. The newly discovered sites and the validation of the predictive models with aggregated predicted score.
Table 24-2. The re-visited archaeological sites in 2019 with updated information.
Table 25-1. Analysis of the ore.
Table 25-2. Analysis of the clinker.
Table 25-3. Analysis of stones in the furnace wall.
Table 26-1. Table of site names and periods from which quantitative archaeobotanical results have been published.
Table 26-2. Representative Index (RI) scoring matrix.
Table 28-1. List of sites included in the study.
Table 28-2. Chalcolithic periodization used in the study.
CHAPTER ONE
INTRODUCTION TO THE ARCHAEOLOGY OF ANATOLIA: VOLUME JV
SHARON R. STEADMAN
AND GREGORY MCMAHON
It is with tremendous pleasure that we offer the fourth volume in this series that features recent archaeological
fieldwork in Anatolia. As has been noted in previous introductory chapters, the intention was to publish a volume in
this series every two years, commencing with Volume I in 2015. Indeed, Volume II appeared in the fall of 2017, and
Volume III in good order in the fall of 2019. Each volume features detailed and timely excavation and survey reports,
and the new section, "The State of the Field," launched in 2019, offers overviews of current findings in various
disciplines within the archaeological field. At the beginning of 2020 Steadman and McMahon began preparations for
Volume IV. Then came Covid-19 and severely abbreviated, or more often, cancelled, 2020 field seasons. This created
a quandary. Should the two-year sequence be suspended so that Volume IV could capture what would, we hoped, be a
normal 2021 season? In the end the editors put out the call for submissions to the series as usual, deciding to gauge
whether to move forward based on the response to the call. Astonishingly, over 30 directors and specialists answered
the call with an intention to submit to the volume; of these an extraordinary 28 indeed contributed chapters. Steadman
and McMahon view this as a testament to the devotion Anatolianists have to their work.
This volume follows in the footsteps of Volume III, which introduced the A4 format along with color figures. The
larger format allows for increased detail in illustrations, and the use of color in critical images such as displays of
ceramic types and field shots significantly enhances the reader's comprehension of the data presented.
The fourth volume features projects directed by archaeologists residing and working in Turkey, Europe, and North
America. Initial submissions to the volume are first reviewed by the series editors, and then vetted by the series'
editorial review board:
Levent At1c1 (University of Nevada, Las Vegas) Timothy Matney (Akron University)
Claudia Glatz (University of Glasgow) Arkadiusz Marciniak (Adam Mickiewicz University)
Lisa Kealhofer (Santa Clara University) Christopher Roosevelt (Kos; University)
Omtir Harman~ah (University of Illinois, Chicago) Paul Zimansky (Stony Brook University)
Submissions approved by the review board are then submitted to the editorial panel at Cambridge Scholars Publishing
(CSP). Submissions accepted by the CSP editorial panel are once again reviewed by the series editors who make style
and content recommendations to the authors. The chapters included here, therefore, are the result of a rigorous peer-
review process.
The time periods covered in the research presented span the millennia from the Bpi-Palaeolithic to the Byzantine
and Medieval periods. The breadth of Anatolia is represented, with chapters on Urartian sites in the easternmost
reaches of the country to sites resting on the shores of the Aegean Sea. Research along the Black Sea coast is
geographically balanced by reports on excavations in southeastern Anatolia. While many of the chapters include
reports on "traditional" field seasons in 2018 and 2019, readers will be impressed by the innovative ways researchers
used their 2020 field seasons, conducting short and targeted, often problem-solving fieldwork projects, and engaging
in deep and rich analysis of the compendia of data gathered in extensive pre-Covid seasons.
Chapters 2-20 feature information on excavations, and Chapters 21-25 offer the latest on survey work. These 24
excavation and survey chapters can be grouped into four general themes. First, not surprisingly, many chapters focus
on insights that emerged as a result of the targeted work undertaken in the 2020 "Covid season." For instance, new and
detailed analyses of the Chalcolithic marble workshop at Kulaks1zlar (Ch. 5, by Takaoglu), the Iron Age pottery from
Oluz Hoytik (Ch. 10, by Saba), the Middle Bronze Age at Tell Atchana (Ch. 7, by Akar et al.), and the creation of a
digital inventory of artifacts and the magnetic survey at Porsuk-Zeyve Hoytik (Ch. 11, by Barat et al.) significantly
advanced research at these sites. The very exciting discoveries concerning Urartian worship at Ayanis, as well as
brand new radiocarbon dates (Ch. 14, by I~1kh and I~1k, and the I$1kh et al. Appendix), and the Sagalassos project's
use of predictive modeling (Ch. 24, by Vandam et al.) illustrate the types of insights gained during these
unprecedented times.
A second theme that emerged concerned "routes and relations." Many investigators used the last several years to
refine their understanding of how particular regions and places were connected through travel and trade routes. Okse's
salvage excavations in the southeast (Ch. 2, "Ambar Dam Salvage Excavations") detail the relationships between
these sites and the larger region; the work at Goks;eada Ugurlu (Ch. 3, by Erdogu et al.) offers critical insights into
2 Chapter One: Steadman and McMahon
communities along the Aegean coast and the inland areas. A similar approach investigated the Sinop region's relations
to the west and south (Ch. 13, by Doonan et al.). Three survey projects also took on the goal of understanding regional
contacts. Doliche's relationship with settlements in Cilicia and North Syria were the subject of Blamer's work (Ch.
18). Kaymak91 (Ch. 21) conducted an extensive analysis of the connections between the mountainous Giresun region
along the Black Sea coast with central Anatolia, and Harman~ah et al. (Ch. 22) offered a fascinating study ofrelations
between mountain and valley, both in ancient times, and in ethnographic and archaeological understanding.
A third approach taken by authors was to concentrate on refining the phasing of multiple occupational levels at
various sites, using the expanding archaeological toolkit available for such endeavors. Turkcan and Ertemin (Ch. 4)
offered a final review of their work at Kanhta~, including the intricate phasing of the Early Chalcolithic architectural
features. The detailed analyses of stratigraphy and ceramics at Usakh Hoytik (Ch. 6, by D ' Agostino et al.) allowed for
a better understanding of the Iron Age resting atop the Hittite remains, and Novak et al. (Ch. 9, "Sirkeli Hoytik")
offered an invaluable chronological analysis of how Sirkeli Hoytik fits into the Bronze and Iron Ages of Plain Cilicia.
Investigations at Tuspa (Ch. 15, by Gen<; et al.) laid out arguments for the chronological relationships of some of this
site's main architectural features, and research at Komana (Ch. 17, by Erciyas et al.) employed stratigraphic and
ceramic analysis to identify seven phases spanning the Chalcolithic to the Sel9uk periods.
Perhaps not an actual "theme," the fourth category includes what we might call "traditional" reports on excavation
and survey field seasons, especially welcome in these difficult times. Uysal and <;if<;i (Ch. 8) profiled their work at
two sites in the Elbistan Plain, and Branting et al. (Ch. 12) offered the latest findings at the grand site of Kerkenes
Dag1. Recent work at Hierapolis (Ch. 16, by Semeraro) investigated the architectural strategies residents employed to
manage the significant ancient seismic events in the region. Kruger and Killm (Ch. 19) presented an in-depth report on
the most recent work on Roman Bogazkoy, and Polosa (Ch. 20) profiled the recent discovery of a basilica at Elaiussa
Sebaste in Rough Cilicia. Maner's report on her latest Konya region survey data (Ch. 23) highlighted Bronze and Iron
Age settlements, and Arslan and Bakan's survey in the Gargara region (Ch. 25) identified the location of cities missed
by earlier investigators in the region.
The third and final section in the volume, "The State of the Field," includes three chapters that readers will find
truly represent the intention of the section title. Marston and Castellano (Ch. 26) present a marvelous survey of the
latest data on archaeobotanical research across Anatolia. The "From Bones to Genomes" entry (Ch. 27, by Kazanc1 et
al.) is an incredible summary of the cutting-edge research on ancient DNA analysis. The final entry by Yildmm and
Steadman (Ch. 28) offers readers a survey of "Chalcolithic Religion and Ritual" across the Anatolian Plateau.
Many ofus were prevented by the various effects of the Covid-19 pandemic from pursuing our research in Turkey
last summer, and, in many cases, in the summer of 2021 as well. Losing so abruptly the opportunity to do our work,
excavation or survey, on the ground, in country, brought home to us just what a privilege and pleasure it is to pursue
fieldwork in Turkey. In addition to missing the chance to work in the field, we miss our time with colleagues and
friends whom we look forward to seeing during the field season. It is always salutary to live and work in this place
that has been home to such a rich variety of peoples and cultures, over an almost unfathomable span of time; even one
season away from that opportunity is an experience we hope will not be repeated. We therefore look forward, if
anything, even more eagerly to our future work in the field, armed by an ever-widening array of approaches and
technologies, but always with the same goal: a greater understanding of the truly unique cultural legacy provided by
Turkey's rich and amazingly varied archaeological resources.
We would prefer to express our thanks in person as well, but this volume also provides the opportunity to share our
gratitude for the infrastructure provided by Turkey in support of our research. Everyone who enjoys the chance to do
fieldwork is happy to thank the Ktiltfu ve Turizm Bakanhg1 and the Ktiltfu Varhklar1 ve Mi.izeler Genel Mi.idfuli.igi.i for
all they do to facilitate our work. Our thanks as well to all the government representatives, the temsiciler, for their
willingness to work with a remarkably varied range of projects, in locations all over the country. This volume, like its
predecessors, would not be possible without the Ministry of Culture and Tourism, the General Directorate of Cultural
Monuments and Museums, the museum directors, government representatives, Turkish colleagues, and people of
Turkey, and we therefore thank them for making fieldwork of all kinds in Turkey such a uniquely rewarding
expenence.
PART I:
EXCAVATIONS
(Photo courtesy of the <;::adir Hoyuk Project)

CHAPTER Two
AMBAR DAM SALVAGE EXCAVATIONS 2018-2020:

AMBAR HOYUK, GRE FILLA, AND KENDALE HECALA
A. TUBA OKSE
The archaeology of the Upper Tigris Region is marked by salvage excavations carried out in the flood zones of
several dams built on the upper basin of the Tigris River-the lhsu Dam- and a series of smaller irrigation dams
which are being constructed on the tributaries. One of these dams is built on the upper valley of the Ambar c;,:ay, ca. 49
km north of the Tigris basin. Two mounds-Kendale Hecala and Gre Filla-will be submerged; the eastern skirts
-~
.. ~, .- .'f,
~
Figure 2-1. Location of Ambar Dam: a) The Ambar Dam and excavated sites; b) Neolithic settlements in the Upper
Tigris region (prepared by Sakir Can): 1. c;,:ayonu, 2. Til Huzur Yayvantepe, 3. Girikihaciyan, 4. Ambar Hoyuk, 5. Gre
Filla, 6. Kendale Hecala, 7. Hakemi Use, 8. Boztepe, 9. Karavelyan, 10. Salat Cami Yam, 11. Kortik Tepe, 12.
Demirkoy, 13. Hallan c;,:emi, 14. Sumaki Hoyuk, 15. Hasankeyf Hoyuk, 16. Turbe Hoyuk, 17. Gusir Hoyuk, 18.
Boncuklu Tarla; c) View of Gre Filla in foreground and Kendale Hecala in background (Excavation Archive, 2019).
Ambar Dam Salvage Excavations 2018-2020 5
of Ambar Hoyiik will also be affected by the flood (Okse et al. 2020). These mounds are located on the alluvial plain,
which is suitable for dry farming, on the southern skirt of the Uzuncaseki mountain range. The Ambar Valley is on a
natural route crossing the most navigable mountain passages in the Taurus range, leading to the obsidian resources of
Solhan in Bingol (Okse 2020: 13-15, fig. 17). The mounds are located close to each other; Gre Filla is ca. 200 m to
the south of Ambar Hoyiik, and Kendale Hecala is ca. 1000 m south of Ambar Hoyiik (Fig. 2-1 ).
AMBARHOYUK
Ambar Hoyiik covers an area of ca. 0. 7 ha, with its skirts extending to ca. 9 ha. Since the mound will not be
submerged, excavations were carried out in a sounding on the eastern slope. The slope wash includes pot sherds from
the Early Pottery Neolithic period onwards (Okse 2020: 7-11 ). An Ubaid sherd was recorded as a surface find
(Peasnall 2004: 30, fig. 1/1 , 7); however, in the sounding, no Ubaid sherds have been found. The handmade Iron Age
pottery as well as the "festoon ware" found in the fill excavated at Ambar Hoyiik is associated with nomadic
communities that emerged in the Upper Tigris region after the withdrawal of the Middle Assyrian Kingdom in the
middle of the 11 th century BCE. After the Neo-Assyrian conquest in the 9th century BCE, Tusban (Ziyarettepe)
became the provincial centre. The Neo-Assyrian standard vessels found in the fill may indicate an occupation under
Neo-Assyrian supremacy. The site was also inhabited from the Late Antique-Byzantine period (5 th-7 th centuries CE)
onwards.
Figure 2-2. Ambar Hoyiik Middle Bronze: a) Level 5; b) Portable altar foot Ll6/0009/S/04
(Excavation Archive, 2020, 2019).
6 Chapter Two: Okse
The sounding revealed six architectural levels (Okse 2020: 8-9). The upper levels (1-4) date from the Late
Antique period to the Middle Age, and the lower two (levels 5-6) to the Middle Bronze Age (1951-1771 cal. BC).
Level 5 is represented by the entrance and three narrow rooms, probably storage areas within a building complex, and
Level 6 by a quasi-similar building constructed on the bedrock (Fig. 2-2a). The material includes sherds of Red-Brown
Wash Ware, Dark Rimmed Orange Bowls, Habur Painted pottery, and a few terracotta feet belonging to portable
altars (Fig. 2-2b), similar to those found in the piazza at Hirbemerdon (Laneri et al. 2016: 64-67), and in the burnt
building room 2 at Muslumantepe (Ay et al. 2013: fig. 7). The monumental appearance of the buildings may indicate
the existence of a local administrator managing the surrounding fields .
GREFILLA
Gre Filla is ca. 7-8 m in height, covering an area of ca. 0.5 ha. The main settlement period is the Pre-Pottery
Neolithic, presenting a sequence dating to 9300-7500 cal. BC. The Pottery Neolithic period is represented only by pot
sherds; the architecture seems to have been destroyed by graves. After a hiatus of ca. five millennia, the site became a
cemetery (Figs. 2-3 and 2-4).
·e <1,1
...e,: ~
== ;:=
<1,1
~
~
== C.
s < ~ ·i:::
~
== e,: "'...e,:
~
:o <I.I
~
=... =
C. n ~ n se,: p"' ~<
= =
Cal. <I.I :o .c 0
v- = ~ :o ~ :o
<I.I
BCE = =
;:id 1 s v- ·e <~ Period
=
:o
~
-; n
e,:
~
::
·i:::
s :o.......
.!:: = ~<1,1
s •!::"'= :o..0.
'ii
....... ~
"C
e,:
-u~ 'i;> ~
-;
=
- - s
O"
~
~
e,:
u
....e,:
-;
<I.I
~
i=i <
zu
= ~~
e,: e,:
~ ~ =
<I.I <I.I <1,1
v- z i=i c., c., <1,1
c., z u
= ;:id
""' 00 00
II Ubaid
5 000
IIIA
III Pottery
6 000
Neolithic
s
0
3-1
IIIB
1
...
0
I
1-
<1,1
7 2
7 000
...
Oil
Pre-
cell Pottery
8 000 Neolithic
B
grill
9 000
Pre-
Pottery
Neolithic
10 000 A
Figure 2-3. Chronological table of Neolithic sites based on calibrated dates (after Karul 2011: 6; Ozkaya and Co~kun
2011: 103; Erim Ozdogan 201 la: 192-193, 201 lb: 134; Cauvin et al. 2011: 4; Schmidt 2011: 52; Hauptmann 2011:
103; Rosenberg 2011: 80; Tekin 2011: 152, 2017: 183, 189,230).
Gre Filla I: Late Antique Graveyard
Approximately two-thirds of the Late Antique graves are stone cists built mostly with coarsely shaped limestone
slabs; however, some cists had been built with regularly cut rectangular limestones, probably stones used as spoil from
previous levels. Most of the graves include single individuals, although some graves were re-used, and a small number
of graves include two or more individuals. Most of the skeletons are laid in supine position, on their backs, with hands
on the belly. Nearly all graves are east-west oriented with the head in the west. The graves are mostly simple earthen
burials surrounded with rough stones; however, a few include pithoi with skeletons in hocker position (Fig. 2-5a-b ).
One-fifth of the burials are oriented north-south, indicating that non-Christians are buried in the same cemetery. Three
blocks covering one cist grave have reliefs on their surfaces (Fig. 2-5c ), indicating that architectural blocks were
converted to grave covers. Nearly one-third of the graves contain vessels and personal belongings. In two graves Late
Hellenistic coins were found, and in several graves, glass vessels dating to the 3rd-4th centuries CE were recovered
(Fig. 2-6).
Figure 2-4. Gre Filla, general drone photo (Excavation Archive, 2020).
8 Chapter Two: Okse
Figure 2-5. Gre Filla Late Antique cemetery (Excavation Archive, 2019-2020): a) Stone cist grave NS/008/G with
pithos burial; b) Lamax grave 08/132/G; c) Relief stones from grave 08/020/G.
Figure 2-6. Gre Filla Late Antique cemetery small finds (Excavation Archive, 2019-2020): a) pithos of grave
NS/008/G; b) glass vessel 15/0028/U/01; c) glass vessel 16/0015/U/02; d) glass vessel 09/0118/U/01; e) trefoil jug
08/0023/S/04; f) glass vessel N9/0060/U/02.
Gre Filla II-III: Middle Age and Pottery Neolithic Period
The fill in and around the graves includes several sherds, chipped stone artefacts, and terra cotta figurines,
indicating that the graves had been dug into the upper levels overlying the Pre-Pottery Neolithic settlement. The sherds
are composed of coarse and medium-coarse vessels, a few incised (sgraffito) examples, moulded (barbotine) and
glazed sherds dating from the Byzantine to the Early Ottoman periods, and Neolithic sherds dating to ca. 6900-5200
BCE. Based on the presence of a few "Hassunan" sherds and "Husking Trays," Gre Filla was the only Ambar Valley
site inhabited between 6200 and 5700 BCE (Okse 2020: 5-8, 2021: fig. 2-4).
Gre Filla IV: Pre-Pottery Neolithic B
The Pre-Pottery Neolithic B (PPNB) period is represented by a ca. 3.5 m thick archaeological fill. The radiocarbon
dates reveal a total sequence of 9300 to 7550 cal. BC (Fig. 2-3). The structures are composed of 5-14 re-building
levels, indicating the renewing of buildings, a common occurrence throughout the Neolithic Age (Ozdogan and
Ozdogan 1998: 589-592).
The settlement layout in the northern operation
NORTHERN OPERATION SO UTHE RN OPERATION
is marked by three oval subterranean structures,
separated by ca. 9-10 m from each other. Each oval
structure is surrounded by quadrangular buildings
separated by narrow paths connecting them. The
grouping of quadrangular structures around each
. oval structure left the impression of subgroups, each
making use of the subterranean structure in the
Qi£,
middle of their residential circle (Figs. 2-4, 2-7).
Another oval structure and several quadrangular
buildings were also uncovered in the southern
operation. Judging by radiocarbon dates and the
undisturbed stratigraphic position of the oval and
quadrangular buildings, it seems clear that both
• .. . 1~ . " ' . building plan types had been used simultaneously.
• . 't...::1J o I, •
Thus, unlike the situation in (:ayonu (Erim Ozdogan
" 201 la), no periodization of building schemes can be
put forth.
co
M
0
Q
Oval Buildings
"
The oval buildings are ca. 7.5 x 10 m in dimension,
each equipped with four pillars constructed of
limestone and mud mortar. The walls, renewed
multiple times, were not always fully placed one on
top of another, causing an appearance of intertwined
walls. The buildings have slightly accentuated
comers, similar to those in the Round Building
Phase r2-4 at (:ayonu (Erim Ozdogan 201 la: fig. 6)
and in Gusir Hoyuk Phase 1 (Karul 2011: 10, fig.
5).
Building 8 in the southern part of the operation
comprises eight renewal levels (Fig. 2-8a). The inner
Figure 2-7. Gre Filla, schematic plan of Pre-Pottery fill contains an immense amount of stone debris and
Neolithic B buildings (drawn by author). reddish soil belonging to different fill layers with
specific finds, pointing to several "closing" steps.
The uppermost levels 1-2 consist of stone fill overlying the pillars. Level 2 is represented by a quadrangular unit
of 2.92 x 2.23 m, attached to the inner face of the southeast oval wall. The considerable amount of chipped stone
debitage indicates use as a lithics workshop. Level 3 has an earthen floor, and the pillars are visible. The fill contains
broken pieces of a pig skeleton, deer antlers, considerable numbers of beads, malachite pieces, figurines, and ground
stone artefacts. Level 4 is equipped with a stone platform and 11 stone bases for wooden posts. Between the southern
pillars, an installation consisting of a stone-built pit, a large stone slab, and a stone-built platform indicate probable
ritual facilities. In level 5, apart from four lower pillars, a stone-built "altar" and four stone bases have been
uncovered. The fill contains a considerable amount of small clay lumps, indicating ritual behaviour during the
"closing" ceremony. In level 6 four lower pillars, four stone bases, and two stone phallic figures of 47 and 55 cm in
height were found erected near the oval wall, opposite one another. Level 7 contains three stone bases and the lowest
pillars. The entire depth of the structure is ca. 3 m.
10 Chapter Two: Okse
Figure 2-8. Gre Filla architecture (Excavation Archive 2019, 2020). Northern Sector: a) oval building 8 in trench NS;
b) Oval Building 15 in trenches P7-8; c) upper PPNB levels in trenches O-P8-9;
Southern Sector: d) middle PPNB Levels in trenches H-15-6.
Building 7 in the northwestern area is also equipped with four stone-built pillars. The upper stone fill overlying the
pillars contains immense amounts of chipped stone debitage and broken ground stone artefacts. The western half was
uncovered up to the stone paved floor in level 2. In level 3, a quadrangular hearth was unearthed between two pillars.
This level comprises interior partition walls, and pieces of terrazzo floors, as well as a burnt wooden superstructure of
the building.
Building 15, in the northeastern area, is equipped with four pillars, ca. 1 x 1 m in size, built of limestones and
plastered with mud. Eight undecorated stelae, attached to the inner face of the oval wall, are evenly spaced. These are
formed of large stone slabs ca. 1.16-1.35 m high, 40-88 cm wide, and 18-23 cm thick. On the eastern inner face, a
niche measuring 25 x 25 cm included 10 pestles deposited inside (Fig. 2-8b); two quadrangular hearths, ca. 75 cm
below the niche, were built with upright stone slabs. The five layers of fill are characterized by stones, reddish brown
soil, and a significant amount of burnt mudbrick debris. The fill contains animal bones, ground stone artefacts such as
mortars, pestles (one of zoomorphic form), phalli, grinding stones, weights, and chipped stone artefacts including
Mureybet and Nemrik points.
Quadrangular Buildings
The quadrangular structures surrounding each oval structure are mostly composed of cell-planned buildings and
large-room buildings with internal and/or external buttresses, and a few grill-planned buildings, each with a large
room (Fig. 2-8c). Several large rooms were equipped with 1--4 stone bases for wooden posts to support the roof.
Pieces of plaster found in the debris of some buildings bear the negative imprint of the tree trunks used in roof
construction. Some buildings have mud plaster on the walls. Most of the floors are composed of compacted clay;
however, mud-plastered and stone-paved floors are also present. Dense mudbrick debris and mud plaster with reed
traces, recovered in some buildings, provide evidence of the construction of their superstructure.
In the northern operation, a total of 28 quadrangular buildings have so far been uncovered. Nearly all these
buildings are composed of 3-12 levels, built on top of each other, with minor shifts. Since very little material is found
in these levels, the floors may have been swept clean before the new construction.
Cell-planned buildings are occasionally composed of 4-6 small rooms attached to a larger room, some having a
doorway enabling access from the narrow paths. In the northern operation, nine cell-planned buildings have been
uncovered (Buildings 1, 6, 9, 11, 18, 19, 24, 25, and 28) (Fig. 2-8c). Similar buildings are known from CayonU (Erim
Ozdogan 201 la: figs. 34-36, 53), and from the Pre-Pottery Neolithic B settlements in Sumaki HoyUk (7000-6800
BCE) (Erim Ozdogan 201 lb: 24, 37) and Boncuklu Tarla (7500-7000 BCE) (Okse et al. 2014: 104-107; Kartal et al.
2014: 486-487) in the Upper Tigris region. Buildings 19 and 24 have hearths inside, and the floor of Building 11 is
cobble-paved. In Building 19, Nemrik and Mureybet points were evident among the lithic assemblage (Okse 2020: 29,
fig. 6). Approximately 50 terra cotta bird figurines (Fig. 2-9.i-j) and Nevali Cori type points (Fig. 2-9.h) were
collected from Building 11.
e
d
b 5cm
Figure 2-9. Gre Filla PPNB small finds (Excavation Archive, 2019-2020): a) stone human figure H6/0108/R/0l; b)
rubbing stone I5/0052/R/02; c) zoomorphic pestle N9/0529/R/0l; d) stone plug O8/0434/R/03; e) stone plug
H6/0439/R/0l; f) obsidian microlith 08/0433/0/01, g) flint microlith P9/0663/N01; h) Nevali Cori point
O9/0147/N/0l; i. baked clay figurine N9/0245/P/0l; j) baked clay figurine O9/0257/P/0l; k) bone needle
N8/0328/Q/0l; 1) malachite celt with horn handle O8/0434NQ/06.
Buttressed buildings at Gre Filla were planned with one or two large rooms. Internal buttresses are spaced
generally ca. 0.56-1.10 m apart from each other, probably built to support the roof or to provide stability. In the
northern operation, seven buttressed buildings have been uncovered (Buildings 2, 4, 5, 10, 13, 14, and 16). Buttressed
buildings are known from Cafer HoyUk Level XIII-VI (Cauvin et al. 2011: 1-2, figs. 5-6). A wattle-and daub
superstructure is attested for Building 13, though buildings with hearths inside are rarely attested (Building 16). The
floors of Buildings 10 and 16 were cobble-paved, similar to the Cobble-Paved Building Phase cpl-3 at CayonU (Erim
Ozdogan 201 la: figs. 31-33). In Building 2-level 8, small malachite celts were found attached to a horn handle (Fig.
2-9.1); items produced from malachite indicate the progress of primitive experimentation with early metal processing,
7586-7502 cal. BC. Another malachite celt was found in Building 5, dating to 8296-8197 cal. BC. Similar raw and
processed copper and malachite finds were found in CayonU level 7, belonging to the intermediate level between the
grill and cell-planned buildings, in the sub-phases of the Channelled and Cobble-Paved buildings dating to 9000-8600
cal. BC; they were also present at Hallan Cerni (Ozdogan and Ozdogan 1999: 14-18).
Other Gre Filla buildings, with a large room and a row of buildings with one large room and narrow long
chambers, resemble grill-planned buildings with wide distances (ca. 50-120 cm) between the walls (Buildings 17, 20,
26, and 30). All buildings were equipped with hearths inside, and the floors are cobble-paved. In the northern
operation, seven single-chamber buildings have been uncovered (Buildings 3, 9, 12, 21, 22, 23, 27, and 29), similar to
those known from CayonU (Erim Ozdogan 201 la: figs. 53-54, 57-58). The floors of buildings 22 and 23 were cobble-
paved, and Building 22 is equipped with an interior hearth; several mortars were placed under its foundations,
indicating a foundation ritual.
In the southern operation, a 1.2 m thick cultural deposit belonging to six architectural levels is attested. An oval
structure (Building 8) and cell-planned structures occupying areas of ca. 22- 38 m2 have been uncovered. These
structures, with two small and one large room, are sometimes built with rounded corners. Building 1-levels 4- 6, 9-
level 3, and Buildings 14 and 15 have hearths inside, as well as stone paved floors (Fig. 2-8d). The small finds
comprise ground stone products such as mortars, pestles, flat axes, and weights. A Byblos point was found in Building
1-level 4.
Sub-Periods
In the upper sub-period of the northern operation, oval structures 7, 8, and 15 were intentionally filled, covering all
pillars. The uppermost level of Building 8 is dated to 7845-7599 cal. BC. Six quadrangular structures surrounding
each oval structure, comprising 2- 10 building levels, were found in this sub-period. Cell-planned Building 1, and the
large-chambered Building 2, occupy areas of ca. 48- 50 m2, and the buttressed Buildings 3, 4, 10, and 11 are ca. 25-27
m2 in area; these strata are ca. 0.50-1.6 m thick. Building 2-level 8 is dated to 7586-7502 cal. BC, and level 9 to
8017-7788 cal. BC.
The middle sub-periods are represented by the upper levels of the pillared oval structures. Four pillars of Building
8-level 3, dated to 8349-8242 cal. BC, are preserved to heights of 1-1.5 m. The structure is surrounded by buttressed
Buildings 13 and 26 and large chamber Buildings 10 and 16, all composed of2-12 levels. Building 26 includes a total
of five hearths consisting of several levels built on top of each other which were built on the floors of two large rooms.
Building 7 is surrounded by the buttressed Building 5 and large-chamber Building 12; Building 15 is surrounded by
the buttressed Building 14 and cell-planned Building 6. All quadrangular buildings occupying areas of ca. 22-27 m2
are considerably smaller than the two buildings in the upper sub-period; the fill belonging to these levels is ca. 0.30-
1.6 m thick. Building 5 is dated to 8296-8197 cal. BC, and there was an assemblage of artefacts on the path to the
south of Building 7, to 8726-8531 cal. BC. In the southern operation, the uppermost level of Building 2 is dated to
8481- 8293 cal. BC.
The lower sub-period includes the lower levels of the oval buildings in the northern operation. Buildings with one
large room and narrow long chambers resembling grill-planned buildings appear in this sub-period (Buildings 17, 20,
26, and 30). Buildings 16, 19, 24, 27, and 31 are large-chamber structures. No cell-planned or buttressed buildings
have so far been observed. Five buildings (Building 29, 27, 16, 26, and 31) surround Building 8, three (Building 17,
23, and 24) Building 7, and two (Building 19 and 20) Building 15. The lowermost level 8 of Building 6, overlying
Building 19, is dated to 8750- 8556 cal. BC, indicating an earlier date for the lower sub-period. All quadrangular
buildings, occupying areas of ca. 14-32 m2 , are considerably smaller than those in both upper sub-periods; these strata
are ca. 20-70 cm thick. In the southern operation, this period is represented by structures built on the virgin soil. These
partially uncovered buildings are built with mud-plastered walls.
Small Finds
The small finds from these buildings comprise a considerable number of ground stone artefacts produced of basalt,
limestone, and quartzite. These include flat axes, adzes, hammers, pestles, grinding stones, mortars, weight stones, and
rubbing stones. One pestle has a zoomorphic stalk (Fig. 2-9.c), similar to those found at Gusir Hoyiik (Karul 2011: 16,
fig. 17), Hallan <;emi (Rosenberg 2011: 78, fig. 13), and Kortik Tepe (Ozkaya and Co~kun 2011: 122, figs. 24-25).
Stone plugs, also found in these deposits, resemble those from Gusir Hoyiilc (Karul 2011: 16, fig. 18).
The lithic assemblages, including cores and flakes, indicate a chipped stone tool production at the settlement. More
than half of the material is flint; the rest is obsidian. The chipped stone tools comprise a rich assemblage including a
large number of blades, scrapers, hacks, piercing tools, cutting blades, and sickle blades, as well as microliths (Fig. 2-
9.f-g), Nemrik points, crested blades, and Byblos points. In Building 4 several horn fragments with grooves inside,
resembling horn sickles with flint chips, have been found. Other finds consist of terra cotta spindle whorls and
figurines, as well as bone awls.
Gre Filla V: Pre-Pottery Neolithic A
The latest phase of the PPNA period was uncovered in a sounding of 8 x 4 m at the northern edge of the mound,
under destruction caused by earth transport. This period is represented by a ca. 1.5 m thick archaeological fill. The
earliest settlement constructed on virgin soil (Shelmo Formation) appeared at 708.73 m above sea level. Quadrangular
Building 32 and rounded Buildings 33 and 34 were built with stones resting in mud mortar, and the oval buildings
were plastered with mud. Two earth burials accompanied by a few beads and microliths have been uncovered in
Building 34. Similar architecture is also attested in the Round Building Phase r2-3 at <;ayonii (Erim Ozdogan 201 la:
fig. 6), Kortik Tepe (Ozkaya and Co~kun 2011: 111-112, figs. 3-6), and Gusir Hoyiik Phase 2 (Karul 2011: 11-12,
figs. 8-9),
Small finds include a bone awl and spatula, a decorated stone, pestles, beads, and flakes. The presence of Salibiya,
El-Khiam, and Amuk points among the chipped stone assemblage date, these buildings to the later phase of the PPNA
and to the PPNA-PPNB transition. A radiocarbon sample obtained from a drilling core dates these levels to 9299-
9177 cal. BC.
KENDALE HECALA
Kendale Hecala is ca. 3-4 m in height, covering an area of ca. 0.65 ha (Figs. 2-1 0a, 2-11 ). According to the data
obtained from the drill cores taken from the mound slope in 2019, the settlement area seems to be much wider; the ca.
2-4 m thick alluvial fill carried by floods has covered the lower mound slopes, and the archaeological deposit is 4.5 m
in height. Radiocarbon dates give a sequence of 6700-4550 cal. BC, including the Pottery Neolithic period (Period III)
and the Early Ubaid period (Period II) (Fig. 2-3). After a hiatus of ca. 5300 years, the site was inhabited in the Middle
Ages (Period I).
Figure 2-10. Kendale Hecala architecture (Excavation Archive, 2019-2020). Period 1-11: a) 2019, Middle Age and
Ubaid levels; Period II-III: b) 2020, Ubaid and Pottery Neolithic levels; c). Oven in an Ubaidian Building in trench
K8; d) General view of Pottery Neolithic buildings from trench K8 towards trenches J7-8.
Kendale Hecala I: Middle Ages
Underneath the surface soil, three architectural levels represented by simple foundations built with rough stones
have been recognized (Figs. 2-1 0a-b, 2-11 ). The upper levels were strongly affected by agricultural activities
undertaken on the mound; only pieces of foundation walls, stone heaps, pieces of pebble stone-paved floors, hearths,
and pits survived. The buildings are mostly rectangular, occasionally oriented in a northeast-southwest direction, and
separated from each other by narrow streets. Several pits containing shells, fishing net weights, and crab claws, as well
as bones of large and small cattle and tortoise shells, reflect the rich diet of the inhabitants. In a structure associated
with clay plastered beehive-shaped storage units, a number of grinding stones were found, suggesting its function as a
place for preparing and storing food. Burned mudbrick debris scattered in and around ashy soil contains animal bones
and sherds of cooking pots, an assemblage associated with cooking activities.
Kendale Hecala I Kendale Hecala II
'
L _ _ '
_ _ _ ...,
Kendale Hecala IHA Kendale Hecala IIIB-upper levels
"
"
I
//
0 10m
Kendale Hecala IIIB-lower levels NNNNNI
Figure 2-11. Kendale Hecala, schematic plan of building levels

(prepared by Sakir Can and Seyma Cift9i, drawn by author).
The material left behind, including broken heavy stone tools and a small number of pot sherds, suggests a seasonal
settlement. Ground stone artefacts include grinding stones, pestles, mortars, a hand axe, a shaft-loch axe, a hammer,
sling stones, weights, and beads. Terracotta finds consist of spindle whorls and animal figurines, bone objects of pins,
and metal objects of coins, pins, and clothing accessories. The pottery assemblage is composed mainly of coarse and
middle-coarse common ware appearing as pithoi, jars, cooking pots, and bowls (Fig. 2-12.a-f). A small number of
sherds from moulded (barbotine) and glazed vessels belonging to a period from the Islamic period onwards were
recovered. On the other hand, two hocker burials in medieval jars, an infant and an adult, present an extraordinary
tradition within the burial customs of the Middle Ages.
Calibrated radiocarbon dates (771-902 cal. CE), obtained from three locations-a storage pit J8/025/<;::, an ash
deposit around clay walled store units, and the fill from a jar in trench K9-coincide with the Abbasid period.
However, the assemblage also contains vessels dating to the 11 th- 14th centuries CE.
o- - -5cm
Figure 2-12. Kendale Hecala Small Finds (Excavation Archive, 2019-2020).

Period I: a) vase K9/0l 14/S/04; b) cup K9/0l 14/S/04; c) oil lamp K8/0106/S/0l; d) oval plate L8/0015/S/0l;
Period II; e-f) pot sherds from locus J8/0037/S.
Kendale Hecala II: Chalcolithic Period
Two architectural levels have been distinguished within this period. The upper level is represented by quadrangular
mudbrick structures with limestone foundations and indoor hearths and ovens inclined to about 30 degrees (Figs. 2-
lOc, 2-11 ). These are mostly single-roomed mudbrick structures occupying areas of ca. 20-25 m2, some including
smaller inner divisions or extended by small annexes, probably used for storing. Several postholes, observed in some
floors, as well as limestone post-bases, indicate wooden superstructures as temporary roofs. Calibrated dates (4704-
4540 cal. BC) obtained from a hearth (J7 /031/F) coincide with the Ubaid 4 period. In trench K8, one bowl with hand
motives and a small number of sherds decorated by horizontal stripes and wavy lines, resembling the Northern Ubaid
phases 3-4 (Davidson and McKerell 1980: 157; Campbell 2007: 104, 131), were found among coarse tempered clay
trays (Fig. 2-12.e-f) (Okse 2020: fig. 9:9). In the western part of the settlement, two stone structures with thin inner
walls resembling a grill-plan were probably used as store houses.
A structure with external buttresses in trench L 7-8 has pebble stone foundations and pise walls. The ashy fill and
debris of broken mudbricks overlie a three-phased compacted clay floor. On the burned floor (L8/028/M) dating to
4849-4706 cal. BC, plastered hearths have been uncovered. Another mudbrick structure (K9/064/D) is dated to 5011-
4800 cal. BC, both representing Ubaid phase 3. The calibrated date (5226-5032 cal. BC) from a floor (L8/035/T)
represents the Halaf-Ubaid Transition (5200 BCE). The only burial from this period is an infant in a jar placed under a
stone pavement associated with hearths.
The lower level is represented by two quadrangular pise structures with stone foundations, indoor hearths, and
storage units. A multi-roomed structure with stone foundations and indoor ovens with floors, inclined to about 30
degrees in trenches J-K7-8, held Ubaid sherds beside dominant Neolithic sherds, as well as bone awls, terra cotta
spindle whorls, pestles, and shells, the latter probably obtained from the Ambar Valley. The fill between the pise walls
under this structure in trenches J-K7-8 is not thick.
The western part of the settlement seems to have been used for storage facilities. Rows of small quadrangular units
uncovered in trenches J6-7 and K7 indicate a special function, probably as small storage units or basins for fodder.
The grill-planned structures seem to have been renewed frequently. These installations, upraised above the ground,
could also have been open-air bedsteads, as used in northern Mesopotamia until the last century (Coqueugniot 1999:
53, pl. 2; Tekin 2017: 284-285). Underneath one of the grill-planned structures in trench J7, a single-roomed structure
with a semi-circular attachment at the eastern wall was recovered.
Kendale Hecala III: Pottery Neolithic Period
Five architectural levels have been distinguished within this period (Figs. 2-1 Od, 2-11 ). The architecture is
characterized by rectangular pise structures built by lining handmade mud blocks side by side, or each new layer is
lined up, after the previous layer dries. The mud blocks are covered by white plaster, resembling the architecture at
Sumaki Hoyuk 3 (Erim Ozdogan 2011 b: 27-28, 52). The single or 2-3 roomed rectangular structures resemble those
from Tell Sabi Abyad (Nieuwenhuyse et al. 2010: 74-75, 82), Hakemi Use (Tekin 2017: 247, fig. 78), Sumaki Hoyuk
Phase 2-3 (Erim Ozdogan 201 lb: 26-27, figs. 12a-b, 19-20) and <;ayonu Phase IIA (Erim Ozdogan 201 la: 190-191 ,
fig. 5).
These levels mostly produced Dark Faced Burnished Ware (DFBW) as well as pattern burnished vessels,
consisting of deep bowls, hole-mouth jars, and short cylindrical or funnel-shaped jar necks with simple rims (Fig. 2-
13.a-b, d-e) (Okse 2021: fig. 2). The vessels include round-bottomed, spherical, and semi-spherical bowls, short
funnel necked jars, and round-bottomed shallow plates. Similar forms are recorded in <;ayonu horizon 1-2 (Ozdogan
and Ozdogan 1993: 98, fig. 4; Erim Ozdogan and Yalman 2004: 89, fig. 8), Hakemi Use (Tekin 2011: 166, 172, fig. 6,
16, 2017: 244), Sumaki Hoyuk 1-2 (Erim Ozdogan 201 la: 55), Salat Cami Yam 3-2 (Miyake 2011: 146-147, fig. 21)
and Til Huzur Yayvantepe (Caneva 2011: 183, figs. 8-9) in the Upper Tigris region.
Ow - w 5cm
n 0
(1 - - _ ,cm OM M M 5cm
Figure 2-13. Kendale Hecala, Pottery Neolithic period small finds (Excavation Archive, 2019-2020). Pottery: a)
DFBW bowl Fl 1/0055/S/02; b) DFBW bowl J7/0037/R/Ol; c) SW bowl J8/00124/S/02; d) DFBW impressed jar
L7/0024/S/Ol; e) DFBW bowl K8/0090/S/02; f) SW bowl J7/0016/S/Ol; g) SW jar K8/0273/S/Ol; Small Finds: h)
baked clay figurine K9/0026/P/Ol; i) obsidian chopper K6/0039/0 /0l; Bone Objects: j) spatula J8/0106/Q/Ol; k) awl
J8/0186/Q/Ol; 1) awl K7/0087/Q/02; m) awl J7/0121/Q/Ol; Ground Stone Objects: n) celt J8/0190/R/Ol; o) pestle
Fl 1/0004/R/03; p) shaft hole axe J8/0212/R/Ol; r) shaft hole axe J7/0284/R/Ol.
Standard Ware (SW) (Dabaghiyah-Sotto or Proto-Hassuna) is also represented by a great number of vessels and
sherds (Fig. 2-13.c, g) (Okse 2020: 5, figs. 7:1-6, 8:1-2, 2021: fig. 3). These vessels appear in <;ayonu horizon 1-2
Ambar Dam Salvage Excavations 2018- 2020 17
(Ozdogan and Ozdogan 1993: 97-98, figs. 3-4; Erim Ozdogan and Yalman 2004: 72-73, figs. 9-12), Hakemi Use
(Tekin 2011: 168, 172, fig. 17, 2017: Pl. XIB), Salat Cami Yam 3 (Miyake 2011: 135, 147, fig. 18), and Sumaki
Hi:iyuk (Erim Ozdogan 2011b: 37). The coarsely finished Chaff-Tempered Coarse Ware, similar to those at <;ayi:inti
(Erim Ozdogan and Yalman 2004: 73), occur as body sherds. Also, Black Series styles are evident, as in <;ayi:inti
horizon 1-2 (Ozdogan and Ozdogan 1993: 99, fig. 5; Erim Ozdogan and Yalman 2004: 73) and Salat Cami Yam I
(Miyake 2005, 2011: 145).
About two-thirds of the chipped stone tools are made of flint. Most of the pieces are flakes, fewer are blades and
cores. The obsidian assemblage is comprised of small blades and flakes, demonstrating imported raw material. The
artefacts consist mostly of choppers (Fig. 2-13.i), and some large blades seem to have been attached to handles. Bone
awls (Fig. 2-13.j-m) and numerous pestles, grinding stones, mortars, flat and shaft hole axes (Fig. 2-13.n-r), weights,
and spindle whorls are found in these contexts.
Kendale Hecala IIIA: Later Phase
Within two architectural levels, five structures occupying areas of ca. 30 m2 have been identified (Fig. 2-11). In
trenches 8-9, two multi-roomed buildings were exposed, one with an indoor and an outdoor hearth. The structures in
the northern part of the settlement seem to have been single-roomed; only the one in trenches L7-8 has been
completely excavated.
These structures contain painted Halaf sherds (Standard and Fine Standard Ware) (Okse 2020: 7, figs. 9:6- 8, 10,
2021: fig. 4). The fine-pasted, thin-walled, carinated vessels are painted relatively plainly with simple geometric
motifs in brown or black paint. Similar assemblages are known from Boztepe (Parker et al. 2001: 565, 570, 582-583),
Girikihaciyan (Watson and Le Blanc 1990: 6, 31, fig. 4.0, 1- 8), Til Huzur-Yayvantepe (Caneva 2011: 177-178, 183,
fig. 14), Siirt Turbe Hi:iyuk (Saglamtimur and Ozan 2007: 16) and Karavelyan (Tekin 2014). A few DFBW and SW
sherds, decorated with incised lines and impressions (Fig. 2-13 .c), belong to Impressed (pointille) vessels (Fig. 2-
13.d), widespread in the eastern Mediterranean (<;ilingiroglu 2010). A small number ofterra cotta figurines, found in
upper levels, have pointed heads and incised eyes (Okse 2020: fig. 8:4-5), similar to those from Salat Cami Yam
(Miyake 2011: 148) and Hakemi Use (Tekin 2017: Pl. XIVA); figurines of nude females and women with lower
bodies shaped in cylindrical form (Okse 2020: fig. 12:3), similar to Akars;ay Tepe figurines (Ozba~aran and Duru
2011: 201) were also recovered.
The lack of Hassunan sherds at Kendale Hecala indicates the abandonment of the site during this period. Hassunan
sherds found at Gre Filla suggest that residents of this settlement may have moved to Gre Filla. Flood traces, including
silt and pebble stones, might explain the abandonment of Kendale Hecala. Nevertheless, the architectural style
continues from earlier levels to this phase.
Kendale Hecala !JIB: Earlier Phase
The early phase of the Pottery Neolithic period is represented by two levels with pise structures with indoor and
outdoor hearths showing multiple renovation processes. Six buildings represent the upper level (Fig. 2-11). The
structure in trenches J7- 8 is a multi-roomed building with outdoor hearths and an indoor oven. The building, having
an intramural area of ca. 32 m2, consists of a large room with an oven in the northeastern comer and five cells,
probably storage units, in the northern and eastern areas. The compacted clay floor below a ca. 20 cm thick ash deposit
shows four renewal phases, indicating habitation with seasonal intervals. The ash deposit includes shells harvested
from the Ambar Valley, and two miniature DFBW bowls found in situ. A mudbrick wall (J8/060/D) is dated to 5486--
5367 cal. BC. A large-chamber building to its east seems to have been a single-roomed structure. A building in trench
K8 occupies an area of ca. 15 m2 ; the buttressed building to its southeast and two single roomed structures in trenches
L8- 9 occupy areas of ca. 20-42 m2 .
The lower ten structures (Fig. 2-11) dating to 6657- 6497 cal. BC rest on virgin soil, ca. 704 m above sea level. The
buildings, occupying areas of ca. 20-30 m2, are constructed with mud blocks supported by horizontal wooden beams.
A single-roomed structure with two storage cells in trenches J7-8 had an infant burial under the compacted clay floor.
The collapsed walls indicate earthquake damage. The small finds consist ofDFBW and SW sherds, a small number of
beads, weights, pestles, a bone awl and spatula, and sickle-blades, as well as shells and cereals. Further, one or two-
roomed pise structures in trenches K- L7- 9 are associated with indoor and/or outdoor multi-layered hearths. Among
the ash deposits in trench K7-8 was a terracotta platform containing several pot fragments and grinding stones, which
indicate open-air activities for food preparation. Narrow pathways between buildings offered several animal bones,
fish bones, and shells, likely deposited as waste. The pise building with outer buttresses in trenches J7- 8 has a large
room equipped with several ovens and cells, probably storage areas, at the northern wall. The building to the east is
also constructed on the same plan, and only the northern comers of the building to the south are uncovered. A lower
building was identified based on pise walls running in the same directions, suggesting re-building phases. A pise grill-
planned building in trench J7 indicates the early storage function of the western part of the settlement.
Three DFBW sherds with human relief decorations were found in the earlier level 1 (Okse 2019). Mineral Coarse
Ware (Early Mineral Ware) comprises ca. one-third of the assemblage. These thick-walled sherds with dense mineral
tempered pastes were fired at low temperatures. The shapes are mostly "hole-mouth" jars, bowls, straight sided
vessels, and short cylindrical necked globular pots (Okse 2020: 5, fig. 7:7-8). Similar assemblages were uncovered at
Salat Cami Yam I (Miyake 2005, 2011: 142) and in the upper two levels of Sumaki Hoyi.ik, which dates to 6700- 6200
BCE (Erim Ozdogan 201 lb: 37, 55, fig. 33).
CONCLUDING REMARKS
The salvage excavations at these three mounds on the upper basin of the Ambar <;ay represent the archaeology of
the northernmost part of Northern Mesopotamia. The region seems to have culturally been a part of the Upper Tigris
region; on the other hand, some micro-regional differences are evident.
The PPNB architecture at Gre Filla proved that rectangular and oval structures were used simultaneously. The
settlement is planned with rectangular dwellings surrounding central oval buildings, suggesting a social structure of
sub-groups in the community. Terra cotta figurines also reveal some local types without parallels in northern
Mesopotamia. The PN architecture at Kendale Hecala consists mainly of one or two-roomed pise structures organized
by narrow streets with open spaces that include hearths. The PN community seems to have abandoned the site between
6300 and 5900 BCE and moved to Gre Filla; however, in the 6th millennium BCE, both sites were inhabited. Kendale
Hecala presents a continuous sequence in the 5 th millennium BCE as well. Only Ambar Hoyi.ik was inhabited during
the Bronze and Iron Ages. Gre Filla became a graveyard from the Imperial Roman period through the Middle Ages,
and Kendale Hecala and Ambar Hoyi.ik were inhabited throughout the Middle Ages.
ACKNOWLEDGEMENTS
Salvage excavations were financed by the State Waterworks of the Ministry of Energy, and organized by the
General Direction of Cultural Heritage and Museums of the Ministry of Culture and Tourism of Turkey. The
excavations were carried out by a professional team under the co-direction of Vehbi Yurt, the director of the
Diyarbaku Museum, and the author. The preliminary report presented here is produced from the meticulous and
devoted work of team members. Field reports were prepared by archaeologists Mesut Vural, Ozlem Ekinb~, Fevziye
Koncalar, i. Berkan Erdem, Buket Be;;iks;i, tbrahim Ulu, <;agda;; Ozdogan, Serdar Olam, Sevnur <;oban, Songi.il
Yeti;;ir, Giilseher Kizilaslan, Ebru Serbes, Sadi Canbolat, i. Tayfur A;;kar, and Abdiilselam <;etinkaya in Gre Filla; by
$akir Can, $eyma <;ifts;i, Deli! Y~a, Huseyin Aka, Muhammet Akdogan, Mehmet Turk, Emrah Dins;, Rojda Balta,
Sefa Baykara, Sabri Yildiz, Berivan Atabay, Tarlk Gunce, tbrahim Karat~, and Mujdat Ayhan in Kendale Hecala;
and by 1. Tayfur A;;kar in Ambar Hoyi.ik. Reports on antluopological investigations were prepared by Dogan <;akan
and Belgin Asian; pottery by Elif Bozkurt, Memik Kereci, Hiilya Torun, and Kubra Canbolat; lithics by Z. Beyza
Agrrsoy and 1. Berkan Erdem. Sincere thanks also to Azad Giil and Mevlut <;olak, representatives of the Diyarbaku
Museum. The radiocarbon analyses were carried out by the Marmara Research Centre of the Turkish Scientific and
Technical Research Institute (TO-BiTAK-MAM), Kocaeli.
REFERENCES CITED
Ay, Eyyup, Ay;;en Ay, and <;agn Murat Tarhan. 2013. Musliimantepe Kazilari 2005-2008. llzsu Barajz ve HES Projesi
Arkeolojik Kazzlarz 2004- 2008 <;alz~malarz, 277- 306. Diyarbaklr: TC Kiiltilr ve Turizm Bakanhgi Kultur
Var!Iklari ve Milzeler Gene! Mudilrlilgil, Diyarbakrr Muze Mudilrlilgu.
Campbell, Stuart. 2007. Rethirlking HalafChronology. Paleorient 33(1): 103-136.
Caneva, Isabella. 2011. Ti! Huzur-Yayvantepe. In The Neolithic in Turkey 1: N<OW Excavations and N<OW Research.
The Tigris Basin, M. Ozdogan, N. Ba;;gelen, and P. Kuniholm, eds., 173-184. istanbul: Arkeoloji ve Sanat
Yaymlar1.
Cauvin, Jacques, Oliver Aurenche, Marie-Claire Cauvin, and Nur Balkan-Ath. 2011. The Pre-Pottery Site of Cafer
Hoyi.ik. In The Neolithic in Turkey: New Excavations and N<OW Research. The Euphrates Basin, M. Ozdogan, N.
B~gelen, and P. Kuniholm, eds., 1-40. istanbul: Arkeoloji ve Sanat Yaymlan.
Coqueugniot, Eric. 1999. Tell Dja'de el Mughara. In Archaeology of the Upper Syrian Euphrates, the Tishrin Dam
Area. Proceedings of the International Symposium Held at Barcelona, January 28th- 3(Jh 1998, G. de! Olmo Lete
and J.-L. Montero Fenoll6s, eds., 41- 55. Barcelona: Universitat de Barcelona.
<;ilingiroglu, <;iler. 2010. The Appearance of Impressed Pottery in the Neolithic Aegean and its Implications for
Maritime Networks in the Eastern Mediterranean. TUBA-AR 23: 9-22.
Davidson, T.E. and Hugh McKerell. 1980. The Neutron Activation Analysis of Halaf and Ubaid Pottery from Tell
Arpachiyah and Tepe Gawra. Jraq XLII: 155- 167.
Erim Ozdogan, Ash. 201 la. <;ayonil. In The Neolithic in Turkey: N<OW Excavations and New Research. The Tigris
Basin, M. Ozdogan, N. B~gelen, and P. Kuniholm, eds., 185-269. istanbul: Arkeoloji ve Sanat Yaymlar1.
- . 201 lb. Sumaki Hoyi.ik: A New Neolithic Settlement in the Upper Tigris Basin. In The Neolithic in Turkey: New
Excavations and N<OW Research. The Tigris Basin, M. Ozdogan, N. Ba~gelen, and P. Kuniholm, eds., 19- 60.
istanbul: Arkeoloji ve Sanat Yaymlan.
Erim Ozdogan, Ash and Nurcan Yalman. 2004. Katkih Kil Kaplar ve <;anak-<;omlek: <;ayonil <;anak-<;omleksiz ve
<;anak-<;omlekli Neolitik Buluntulari Dzerinden Bir Y orum. TUBA-AR 7: 67-92.
Ambar Dam Salvage Excavations 2018- 2020 19
Hauptmann, Harald. 2011. The Urfa Region. In The Neolithic in Turkey: New Excavations and New Research. The
Euphrates Basin, M. Ozdogan, N. B~gelen, and P. Kuniholm, eds., 85- 138. istanbul: Arkeoloji ve Sanat
Yaymlar1.
Kartal, Metin, Harun T~klfan, Kaan Bulut, and Onur Din9. 2014. Yontma~ Bulgular I~1gmda Yukar1 Dicle
Havzas1'nda Yeni Bir Neolitik Yerle~im: Boncuklu Tarla. Ankara Oniversitesi Di! ve Tarih-Cografya Fakilltesi
Dergisi 54: 489-500.
Karul, Necmi. 2011. Gusir Hoytik. In The Neolithic in Turkey: New Excavations and New Research. The Tigris Basin,
M. Ozdogan, N. Ba~gelen, and P. Kuniholm, eds., 1- 17. istanbul: Arkeoloji ve Sanat Yaymlar1.
Laneri, Nicola, Stefano Valentini, Marta Aquilano, Margherita Dallai, and Martina Massimino. 2016. Chapter 4. Phase
IIIB: The Middle Bronze Age. Hirbemerdon Tepe Archaeological Project 2003-2013 Final Report: Chronology
and Material Culture, N. Laneri, ed., 41- 80. Bologna: BraDypUS. Available at
<http://books.bradypus.net/hirbemerdon_report>
Miyake, Yutaka. 2005. Archaeological survey at Salat Cami Yam. A Pottery Neolithic Site in the Tigris Valley,
South-east Turkey. Anatolica 31: 1-18.
-. 2011. Salat Cami Yan1: A Pottery Neolithic Settlement in the Tigris Valley. In The Neolithic in Turkey: New
Excavations and New Research. The Tigris Basin, M. Ozdogan, N. B~gelen, and P. Kuniholm, eds., 129- 149.
Nieuwenhuyse, Olivier P., Peter M.MG. Akkermans, and Johannes van der Plicht. 2010. Not So Coarse, Nor Always
Plain-The Earliest Pottery of Syria. Antiquity 84: 71-85.
Okse, A. Tuba. 2019. Yukan Dicle Havzasmda Kendale Hecala Neolitik <;ag Stilize lnsan Kabartmah Kap Par9alar1.
SRMKA 1:1-12.
-. 2020. Yukar1 Dicle Havzasmda Ambar <;ay1 Vadisi Yerle~im Tarihi. Olba 28: 1-34.
-. 2021. New Data on the Late Neolithic Pottery from the Northern Upper Tigris Region: Ambar Dam Reservoir.
Neolithic Pottery from the Near East: Production, Distribution and Use, Proceedings of the 2019 Third
International Workshop (Antalya, Turkey), R. Ozbal, M. Erdalktran, and Y. Tonoike, eds., 291- 305. lstanbul: Ko9
University AKMED.
Okse, A. Tuba, Harun T~klran, Metin Kartal, Ahmet GormU~, Erkan Atay, and Nihat Erdogan. 2014. Ihsu Baraj1
ln~aat Sahas1 2012 Kazilar1. 35. Kazi Sonur;larz ToplantlSl 1: 102-112.
Okse, A.Tuba, Ay~in Konak, and Vehbi Yurt. 2020. Ambar Baraj1-Ambar Hoytik, Gre Filla (Ambar I) ve Kendale
Hecala 2018 Kaz1s1. 41. Uluslararasi Kazi, Arafllrma ve Arkeometri Sempozyumu I: 299- 313.
Ozb~aran, Mihriban and Gune~ Duru. 2011. Akar9ay Tepe: A PPNB and PN Settlement in Middle Euphrates-Urfa. In
The Neolithic in Turkey: New Excavations and New Research. The Euphrates Basin, M. Ozdogan, N. B~gelen,
and P. Kuniholm, eds., 165-202. lstanbul: Arkeoloji ve Sanat Yaymlan.
Ozdogan, Mehmet and Ash Ozdogan. 1993. Pre-Halafian Pottery of Southeastern Anatolia, with Special Reference to
the <;ayoni.i Sequence. In Between the Rivers and over the Mountains. Archaeologica Anatolica et Mesopotamica
Alba Palmieri Dedicata, M. Frangipane, H. Hauptmann, M. Liverani, P. Matthiae, and M. Mellink, eds., 87- 103.
Roma: Universita di Roma "La Sapienza".
-. 1998. Buildings of Cult and the Cult of Buildings. In Light on Top of the Black Hill. Studies Presented to Ha/et
(:ambel, G. Arsebi.ik, M.J. Mellink, and W. Schirmer, eds., 581-593. istanbul: Ege Yaymlar1.
-. 1999. Archaeological Evidence on the Early Metallurgy at <;ayoni.i Tepesi. In The Beginnings of Metallurgy, Der
Anschnitt, Beiheft 9, A. Hauptmann, E. Pernicka, Th. Rehren, and D. Yal9m, eds., 13- 23. Bochum: Deutsches
Bergbau-Museum.
Ozkaya, Vecihi and Aytay Co~kun. 2011. Kortik Tepe. In The Neolithic in Turkey: New Excavations and New
Research. The Tigris Basin, M. Ozdogan, N. B~gelen, and P. Kuniholm, eds., 89- 127. istanbul: Arkeoloji ve
Sanat Yaymlan.
Parker, Bradley J., Andrew Creekmore, and Charles Easton. 2001. The Upper Tigris Archaeological Research Project
(UTARP). Excavations and Survey at Boztepe and Intensive Survey at Talav~ Tepe, 1999: A Preliminary Report.
In Salvage Project of the Archaeological Heritage of the lhsu and Carchemish Dam Reservoirs Activities in 1999,
N. Tuna, J. Ozti.irk, and J. Velibeyoglu, eds., 584- 591. Ankara: Middle East Technical University.
Peasnall, Brian L. 2004. 2002 Diyarbak1I Small Streams Archaeological Survey. 21. Ara$hrma Sonur;larz ToplantlSl 2:
29-44.
Rosenberg, Michael. 2011. Hallan <;emi. In The Neolithic in Turkey: New Excavations and New Research. The Tigris
Saglamtimur, Haluk and Ali Ozan. 2007. Siirt-Ti.irbe Hoytik Kaz1S1, On Rapor. Ege Oniversitesi Arkeoloji Dergisi
10(2): 1-32.
Schmidt, Klaus. 2011. Gobekli Tepe. In The Neolithic in Turkey: New Excavations and New Research. The Euphrates
Tekin, Hali!. 2011. Hakemi Use: A Newly Discovered Late Neolithic Site in Southeastern Anatolia. In The Neolithic
in Turkey: New Excavations and New Research. The Tigris Basin, M. Ozdogan, N. B~gelen, and P. Kuniholm,
eds., 151-172. istanbul: Arkeoloji ve Sanat Yaymlan.
-. 2014. Karavelyan'dan birHalafMezan. InArmizzi: Studies in Honor ofEngin Ozgen, A. Engin, B. Helwing, and
B. Uysal, eds., 247-252. Ankara: Asitan.
- . 2017. Tarihoncesinde Mezopotamya: Yeni Yakla~1mlar, Yeni Yorumlar ve Yeni Kronoloji. Ankara: Bilgin Ki.iltilr
Sanat Yaymlan.
Watson, Patty Jo and Steven A. Le Blanc. 1990. Girikihaciyan: A Halajian Site in Southeastern Turkey. California:
Cotsen Institute of Archaeology Press.
CHAPTER THREE
GOK('.EADA UGURLU:
A PRELIMINARY REPORT ON THE 2018-2020 FIELD SEASONS
BUR<;::iN ERDOGU, {;iGDEM ATAKUMAN, AND NEJAT YDCEL
Located on the north Aegean island of Gok9eada (Imbros), the site of Ugurlu is one of the most important
Neolithic settlements in the Aegean. Key research questions for island archaeology include: why did people settle the
islands? How did they manage an insular setting? What were the relations between island societies, and between
islanders and mainlanders (Broodbank 2000; Knapp 2008; Dawson 2014)? These questions have also been central to
the Ugurlu Archaeological Project, which was initiated in 2009. Gok9eada is about 17 km from the Gallipoli Peninsula
and covers an area of 289.5 km2. The prehistoric site of Ugurlu is a low mound about 1 km north of the village of
Ugurlu, on the western part of the island. It is located at 40° 13' 23.64" N and 25° 71' 56.08" E, at an elevation of 16
m above sea level. The site covers an area of approximately 250 x 200 m on a gentle slope at the eastern foot of
Mount Doganh (Isa). The main Ugurlu-Derekoy road cuts through the site. The Pilon stream lies at the eastern part of
the site, and there is also a nearby spring (Fig. 3-1 ).
Figure 3-1. The location ofUgurlu in the northeastern Aegean region showing
its topographic plan and the excavation trenches.
22 Chapter Three: Erdogu et al.
Trial trenches and the systematic surface collection recovered in the 2009 season showed the site of Ugurlu to be
one of the earliest Neolithic sites in the Aegean Islands. The ten seasons of excavations conducted between 2010 and
2020 have provided a wealth of information and a large material record that is currently under study. This chapter
presents the 2018-2020 excavation results. Some earlier results have been previously published (e.g., Erdogu 2011,
2014, 2017b).
During the excavations, six main cultural phases, designated as I-VI (beginning with Phase I at the top; Table 3-1)
have been revealed. A total of 31 radiocarbon dates at Ugurlu clearly prove the long history of occupation at the site,
from 6760 to 4350 BCE. The earliest, Phase VI, is referred to as the "Pre-Pottery Neolithic" in the Ugurlu sequence.
The earliest pottery found at the site appears at around 6600 BCE during Phase VI-V, and its quantity is quite limited.
Phase V is marked by the first Pottery Neolithic, ca. 6600/6500-5900 BCE, and findings indicate some Anatolian
connections. During Phase IV, around 5900/5800 BCE, the settlement expanded, and a local island culture emerged.
After a short transition period between 5500-5300 BCE, fundamental changes in architecture, subsistence economy,
pottery tradition, and symbolism occurred in Phase III, ca. 5300-4900 BCE. Although the site was abandoned around
4300 BCE, there were no abrupt changes recorded in Phase IL
Sea levels and the shoreline in the Aegean were
Calibrated BC dates Phase different in prehistoric times. Starting at the time of
the Last Glacial Maximum (around 21,000 years
6760-6600 BCE VI
ago), sea levels were about 120 m lower than today.
6600-6500 BCE VI-V The islands of Gok9eada, Bozcaada, Lemnos, and
6500-5900 BCE Samothrace were connected to the mainland. Sea
V
levels were 100 m lower at the time of the Older
5900-5500 BCE IV Dryas (ca. 17,000 to 16,000 years ago), and
5500-5300 BCE IV-III Gok9eada, together with Lemnos, began to form as
an island around 14,000 years ago (Perissoratis and
5300-4900 BCE III Conispoliatis 2003; Chalkioti 2016). The rise of
4900-4500 BCE III-II Aegean Sea levels continued at a slower pace and
reached approximately 18 to 12 m lower than today
4500-4350 BCE II at the beginning of the Neolithic period, about 8,500
Table 3-1. Stratigraphy and dates ofUgurlu years ago (Ozbek and Erdogu 2014).
THE PRE-POTTERY NEOLITHIC OCCUPATION
The earliest occupation (Phase VI) at Ugurlu is located in the eastern part of the settlement, close to the Pilon
stream, and is characterised by the absence of pottery. The thickness of the pre-pottery deposit is about 1 m. Ugurlu VI
is dated by three AMS determinations, with the "Start Boundary" around 6760 BCE (95.4%), and the "End Boundary"
around 6528 BCE (95.4%) based upon a Bayesian calibration. The earliest occupation at Ugurlu was excavated over
an area of about 18 m2 and is divided into seven layers. A circular structure measuring 1.4 m in diameter was found in
the upper first layer. It has a central hearth with a diameter of 30 cm and is surrounded by thick lime 10-15 cm in
width. The floor of the structure was plastered with sandy mud with elastic calcite and ash crystals, and stones were
placed on the skirts of the structure to support possible stakes.
A thin white-coloured floor with patches of ash
and a hearth with a 30 cm diameter was surrounded
by thick lime plaster, as well as a 10 cm diameter
posthole, supported with stones, which was revealed
in the second layer. The excavated area is probably
inside a building, but we do not know its plan since
the building walls are outside the excavated area.
Animal horns and skulls were discovered on the
floor. Four stone beads of agate and nephrite (Fig.
3-2), two bone awls, and a shaft straightener were
also found on the floor. Below the second layer, a
burnt layer, with an extremely dense concentration
of animal bones and enormous amounts of flint-
knapping waste products, of the same raw materials,
was discovered in the third layer.
Figure 3-2. Pre-Pottery Neolithic beads made A building structure was observed in the fourth
from nephrite and agate. layer, and now we are presented with an enclosed
settlement that was surrounded by drystones, the so-
called "glacis." It is a semi-subterranean oval hut-like structure sized 2.70 x 1.40 m, comprising a pise-lined wall
stretching outside the limit of excavation (Fig. 3-3). Two postholes lie along an axis. The floor was made of mud
plaster probably with calcareous sediment. In addition, patches of thick lime floor are observed on the inner upper
sides. It also has traces of a central hearth. Five shell beads, a bone awl, and a large shaft straightener made from
Gok9eada Ugurlu 23
nephrite were found in the structure. Excavation revealed a dense concentration of animal bones deposited just outside
of the structure (in the fifth layer). It seems likely that this was primarily refuse, deliberately deposited in the aftermath
of feasting (?). A large oven about 1.5 m in diameter was also found. At least five cattle horns were found at the
border of the building.
A part of a semi-subterranean oval hut-like structure, probably 3 min diameter, was also found in the sixth layer,
but again extending outside the limit of excavation. It has patches of flooring as well as daub fragments, with the
impression of straw and reed, interpreted as wall debris. White lime-plastered wall fragments painted in red were also
found. Ten worked bones and five stone beads were discovered on the floor as well as debris. The seventh layer was
excavated in a very narrow area; it probably contains the remains of a structure that was destroyed by fire.
Figure 3-3. The semi-subterranean oval hut-like structure in Phase IV.

Denis Guilbeau, who works on the chipped stones of Ugurlu VI, pointed out that pressure blades are quite
numerous, including large ones (Guilbeau et al. 2019). The majority were made by the standing pressure flaking
mode. Some may have been made by the lever pressure mode. The use of these pressure technique modes indicates
that the flintknappers were highly skilled. Tools are very rare, and most consist of flakes and blades used directly
without any retouch or with a lateral retouch. The lateral retouch of some blades is so deep and regular that they can
be considered backed blades. Sickle inserts are quite frequent. The other tools, such as scrapers, denticulates, drills,
and burins, are less frequent. Except for a few examples of obsidian and rock crystal, the chipped stone industry is
mainly made of local flint. Macroscopic examination and XRF analysis show that at least ten obsidian artefacts came
from Melos. However, macroscopic examination of three obsidian blades-not yet chemically analysed-show an
Anatolian origin. The chipped stone industry of Ugurlu was similar to western Anatolian Pre-Pottery Neolithic sites,
such as Ulucak VI, and was quite different from other regions of Anatolia, the Levant, and even from other parts of the
wider Aegean.
Extremely dense concentrations of animal bones were studied by Levent Atic1, who has identified domestic sheep,
goat, pig, and cattle, indicating that a four-tiered animal management regime existed. Plant samples consisting of
cereal grains were also collected by flotation.
THE EMERGENCE OF POTTERY
The earliest pottery found at the site appears around 6600 BCE (Phase VI-V), and its quantity is quite limited
(only about 30 sherds) (Erdogu and Atakuman 2021). Phase VI-Vis represented in two trenches in the eastern part of
the settlement, namely BB20-21 and CC21. Clear archaeological features were observed in Trench BB20-21. A
shallow oval depression with sharp edges, with a northeast-southwest alignment, was found. Although no walls can be
seen, this area seems to be part of a structure. A very thin, partially traceable white-coloured floor is visible. An oval
sunken mud-plastered basin (ca. 60 x 50 cm), and a hearth, ca. 35 cm in diameter, were associated with the floor. This
structure has a terrazzo platform made with burnt lime and pebbles; it was coloured red and was polished (Fig. 3-4). A
partly damaged platform, approximately 50 cm thick, extends outside the limit of excavation. Three bone spatulas, two
stone chisels, a mortar, and a shell bead were found on the floor. In Trench CC21, no architectural structures were
identified, with the possible exception of scattered clusters of stones directly overlying the virgin soil.
Figure 3-4. The terrazzo platform.
J
I/
3cm
4 5
Figure 3-5. The early pottery at Ugurlu.

Generally, the pottery is characterized by a complete absence of plant inclusions and a large amount of mineral
inclusions. This pottery has been sorted out into two main ware groups: Red Slipped Burnished Ware (RSBW) and
Dark Face Burnished Ware (DFBW). The surface of RSBW vessels appears to have a dark red coloured slip and light
burnish. A rim sherd from a globular bowl of closed form was recovered (Fig. 3-5). Alongside the RSBW, there are
unburnished reddish/buff sherds which are similar to the former, except in the surface finish. The DFBW consists of
greyish orange, dark brown, and dull black coloured wares. Greyish orange ware is characteristic, although it
disappears in later phases. Ware texture is fine, and the wall thickness is rather thin. The surface colour is mostly light
grey, and a thin pink-coloured slip is applied on the exterior surfaces, which generally show a mottled appearance on
the surface. Only a single rim sherd of this ware, a globular bowl of closed form, was found. Dark brown ware is also
characteristic with its highly burnished surface and thin walls. Black coloured sherds with dull burnish are very rare.
Gok9eada Ugurlu 25
MONUMENTAL STRUCTURE OF THE POTTERY NEOLITHIC
The earliest example of monumental architecture discovered in the Aegean was found at Ugurlu, dating to around
6000 BCE. This structure is interesting primarily because of its construction and plan. Parallel stone walls of more
than 1 min height and ca. 70 cm in thickness lie approximately 1.80 m apart, and a narrow rectangular space (room?)
of 4.4 x 1.8 m lies between two walls. The southern wall is about 7.5 m long and has an 80 cm wide doorway.
An obelisk made up of two pieces of stone, consisting of a slightly curved stone with a pyramid-shaped stone on its
top were placed at the structure's doorway (Fig. 3-6). The pyramid-shaped stone was stylized into a round-topped
pyramidal form. Flat wedge holes on the stone suggest that it had been taken from a quarry. The partly destroyed front
face had been carved as a rectangular protrusion. Such protrusions can be seen on prehistoric anthropomorphic steles,
which hint that the obelisk may symbolize a stylized human. A pyramid-shaped amulet made from Spondylus was
found hidden beneath the obelisk (Fig. 3-7) (Gemici and Atakuman 2021; Atakuman et al. 2018).
AN
. - -' - ~
0
J I
Figure 3-6. Monumental architecture of the Pottery Neolithic Phase V.
Around 5900 BCE, the front of the south wall

was back filled, and a 5 x 4 m structure (Building 2)
was built. The obelisk was no longer functional, and
the pyramid-shaped stone was overturned. The
previous doorway was turned into a fireplace.
• Massive exterior buttresses were attached to the
southern wall of Building 2. The thickness of the
eastern and western walls is about 1 m. The small
size of the room (9 .2 m2) and the thick walls and
massive exterior buttresses indicate that the building
had an upper story. The northern wall unusually
stretches about 3.5 m towards the east with a
parallel wall about 1 m wide creating a courtyard. A
sherd with human motif in relief and a head made of
animal bone from an acrolithic figurine were found
Figure 3-7. A pyramid-shaped amulet in the building. A copper bead associated with the
made from Spondylus. building is a unique find (Erdogu 2017a).
CULTURAL TRANSFORMATION IN THE MID-6TH MILLENNIUM BCE
Around 5600/5500 BCE in Anatolia and southeastern Europe, significant changes occurred both in the material
aspects of the cultural life and in the society itself, but not in any standard format (<;::evik and Erdogu 2019). The site
of Ugurlu witnessed major cultural transformations around 5500 BCE, but the level and amount of habitation was
rather limited until 5300 BCE. Around 5300 BCE changes occurred in pottery production, subsistence economy,
settlement organisation, and building plans.
Figure 3-8. Some of the objects found in the ritual pits.

Thirty-five pits, associated with a rich variety of artifacts, as well as human and animal bones, were excavated in
the western area of the settlement (Karamurat et al. 2021 ). The inner walls and bottoms of the pits were plastered with
yellow-coloured clay, between 30 and 50 mm thick. The pits were circular in shape, some as deep and wide as 1 m.
Within most of the pits, large amounts of pottery sherds, animal bones (mostly sheep and goat), and various rare
objects were discovered. Among the objects found in the pits were large numbers of bone objects, such as awls,
smoothers, worked horns and worked teeth, stone objects, such as grinding stones, chisels, and axes, and Spondylus
objects, such as rings and/or bracelets. Also in these pits were special purpose anthropomorphic pottery vessels, such
as "eared pots" and polypod vessels, as well as a substantial number of fragmented anthropomorphic figurines
(Gemici and Atakuman 2021; Atakuman et al. 2018) (Figs. 3-8 and 3-9). The pits are located around a communal
building, which is partly damaged and measures ca. 7 x 7 m. The floor of the building was plastered with burnt lime
mixed with soil and sediment. Traces of red paint remain on parts of the floor surface and also near the entrance. A
large bull's horn was found on the threshold of the building. In front of the building was a burial pit with the skeletons
Gok9eada Ugurlu 27
of eleven people, including adults, children, and adolescents (Fig. 3-10). The bodies may have been thrown into the pit
and covered by large stones. A partial skeleton and human finger bones were also found in some of the other pits. Two
AMS dates from the highest and lowest skeletons, as well as two other AMS dates from the communal building, yield
a result around 5300 BCE (Boz and Erdogu 2019).
5cm
Figure 3-9. "Eared pots"

Figure 3-10. The burial pit, ca. 5300 BCE.
unique to Gok9eada.
Two very large structures sharing the same walls were excavated at the highest point of the settlement. The largest
structure, Building 12 (ca. 8 x 8 m), was built earlier, and a structure was then later added on its eastern side (Building
11, ca. 7 x 7 m) (Fig. 3-11 ). An additional wall was constructed on the eastern side of Building 11, and a courtyard
was later added to the south, along with other additional walls.
Figure 3-11. Buildings 11 and 12 located at the highest point of the settlement.
Building 12, which is trapezoidal in shape, has multiple layers of a compact white-coloured plaster floor (Fig. 3-
12). Postholes, supported by stones, were found on the floor. The elaborately constructed western wall is about 1 m
thick, while other walls are about 70 cm in width. The entrance lies on the southwestern portion of the structure. A
circular hearth measuring 90 x 90 cm was located in the south central part of the building. There is an area adjacent to
the north wall of the building surrounded by thin stone walls on three sides. A niche with traces of red paint on its
slightly depressed floor surface lies in the centre. Next to it there is a lime-based platform bounded by thin walls. In
the northwest comer of the building, there is another area with storage vessels and grinding stones.
ci
()
Post Hol e
•Post Hole
Figure 3-12. Plan of Building 12.
Nothing was found on the completely clean floor

of the building except a unique sherd with an
incised deer motif on it (Fig. 3-13). The building
was deliberately filled after functions within it
ceased, and the ritual pits which included rare and
unusual objects, described above, were apparently
dug in it.
Building 11, which was, as noted above, added
to Building 12, also has multiple layers of a
compact white-coloured plaster floor. A large part
of its floor had been destroyed by two large pits as
well as by later intrusions and surface activities. A
centrally placed entrance lies on the southern long
wall. Within Building 11 are two hearths, ca. 80 x
80 cm in diameter, facing each other in the middle
of the structure, with rectangular platforms on the
Figure 3-13. A unique sherd with incised deer motif. sides.
Gok9eada Ugurlu 29
Another large plastered pit measuring 2 x 1.3 m was dug into the building floor, probably after functional use
ceased. Other disturbances of Building 11 include circular stone paved silos which were also built after the building
fell out of use, indicating that this area was perhaps later converted to another function.
A large building complex (Building 3), about 11 x 18 m in size, was excavated in the eastern part of the settlement.
The building was quite sophisticated, built with drystone walls and yellow-coloured plaster clay floors. The building
complex consists of three separate units (B.3.1-B.3.3), which were built adjacent to each other (Fig. 3-14). The largest
building unit, B.3.1, lies on the east side and has four rooms of different sizes. The largest room measures 6 x 4.5 m;
the comers of the room were paved with stone slabs. The northwestern comer of the room yielded a raised stone
platform of 80 x 70 cm in size. Several grinding slabs and a mortar buried in the floor were found next to the platform.
In the southwest comer of the room two large bone tools-an awl and a chisel-as well as a stone axe, were found. A
large, well-made Spondylus bracelet was also found in the southwest comer of the room. A narrow doorway lies in the
northwestern comer of the room, from which one stepped over a raised threshold to enter another room ca. 3 x 4.5 m
in size. A platform with dimensions of 80 x 85 cm and a mortar placed on the floor were also found in this room. In
the southern area of the largest B.3.1 room, there is another room measuring ca. 4 x 2 m, which is reached by passing
through a doorway. Eight broken bone tools, three Spondylus bracelets, and shell beads were found in this small room;
a human skull was also found in its threshold. There is another small room in the northeast comer of building unit
B.3 .1. The floor of this room was made of compacted soil, and nothing was found in it.
Figure 3-14. A Building Complex featuring separate buildings built adjacent to each other.
Although building B.3.2 in the middle part of the Building Complex has smaller dimensions compared to the
others, the northern part was not excavated due to destruction. Only two rooms of the building were excavated. A
large storage container and a plastered storage pit with a width of 1 m and a depth of 1.50 m were found in the 4 x 2 m
room on the southwestern side where the entrance of the building is located. Two stone chisels of nephrite were
discovered in this room. The other excavated room is approximately 2 x 3 m in size; a stone platform measuring 85 x
50 cm lies in the northwest comer. A mortar and a large grinding stone were found next to the platform.
Building unit B.3.3, located in the western portion of the Building Complex, has been heavily damaged by the
irrigation channel. At least three rooms have been identified; no in situ finds were recovered.
DISCUSSION AND CONCLUDING REMARKS
The beginning of the Neolithic way of life in the Aegean is still a matter of debate, as there appears to be no
consensus regarding its chronological onset or the modalities of its process (eg., Perles 2001; Perles et al. 2013;
Reingruber 2011; Dawson 2014; Kotsakis 2014; Sampson 2014, 2019; Horejs et al. 2015; <;ilingiroglu 2017; Douka et
al. 2017; Erdogu 2017a; Horejs 2019; <;evik and Erdogu 2020; Atakuman et al. 2020). These discussions show that
the mechanism of the Neolithic transition in the Aegean is a complex issue. For the beginning of the Neolithic in the
Aegean, it is reasonable to think that there was some kind of interaction between indigenous peoples and newcomers.
Recent research has provided important evidence for pre-Neolithic communities in this region (Atakuman et al. 2020;
<;ilingiroglu et al. 2020). Late Pleistocene and Early Holocene finds were also discovered in Goki:;eada. In addition,
"typical" Neolithic features, such as sedentism, burial custom, and animal management exist in the pre-Neolithic so-
called Mesolithic Aegean (e.g., at Maroulas; see Sampson et al. 2010).
The origins of the chipped-stone tradition in the earliest Neolithic Aegean periods are not yet clear. The chipped
stone industry from early Ugurlu was characterized by pressure blades. The complete absence of the pressure
technique from this region before the Neolithic is noteworthy. This technique was mastered in southeastern Turkey
and is attested at Mersin-Yumuktepe as early as the first half of the seventh millennium BCE (Altmbilek-Algill 2011).
On the other hand, the toolkits at the sites in this southeastern region show distinctive features that are absent in
Ugurlu. The chipped stone industry of early Ugurlu is similar to western Anatolian Pre-Pottery Neolithic sites and is
quite different from other regions of Anatolia and the Levant (Guilbeau et al. 2019).
Excavations at Ugurlu have shown that a four-tiered animal management system, consisting of domestic sheep,
goats, cattle, and pigs, was present from the beginning of occupation. This is compatible with western Anatolian
trends but is different from central and northwestern Anatolia. Isotope analysis has shown that the origin of domestic
animals brought to the island is also western Anatolia (Pilaar-Birch et. al. 2019). Based on our knowledge of the raw
materials that do not occur naturally on Goki:;eada, there is good evidence in pre-pottery layers ofUgurlu for the long-
distance circulation and exchange of artefacts. The exchange networks that brought them to the island are best
documented by obsidian, agate, and nephrite.
According to wide consensus, pottery emerged in Anatolia around 7000 BCE, as is also found in northern
Mesopotamia and the northern Levant (Ozdogan 2009; Ozdol 2012). Pottery has been reported at the 9th millennium
BCE site of Boncuklu in central Anatolia (Fletcher et al. 2017), as well as at Demirkoy along the Upper Tigris
(Rosenberg and Peasnall 1998); however, these early attempts to manufacture pottery did not continue in this or the
wider region until well over a millennium later. Around 6600 BCE, pottery began to be adopted by the Ugurlu
community as is found at other settlements in western Anatolia such as at Ulucak, Barcm, and <;ukurii:;i (<;evik and
Vuru~kan 2020; Gerritsen et al. 2013; Horejs et al. 2015), at least 400 years later than other regions of Anatolia. This
development may have been related to a period when communities with different social, symbolic, and economic
practices began to exchange goods, people, and food, thereby creating and negotiating social divisions. Throughout
the entire sequence of Neolithic occupation at Ugurlu, the pottery was produced by utilizing the same methods and
raw materials, which indicates an entirely local origin for the introduction of pottery at the site. At the same time, the
first ceramics at Ugurlu seem to have emerged without an episode of trial and error, which indicate a pre-existing
knowledge of pottery production, perhaps through these exchange mechanisms.
The monumental structure at Ugurlu is extraordinary in that obelisks related to architecture do not appear
elsewhere in the Aegean and the Balkans. On the other hand, a free-standing stone with red ochre on it was found at
the site ofYabalkovo, Bulgaria, dating to ca. 6000 BCE (Roodenberg et al. 2014). A large monumental building ca.
8.5 x 9.5 m was also found at Koutroulou Magoula, Greece, dating to ca. 6000 BCE (Hamilakis et al. 2017). Ugurlu's
monumental structure displays features and finds that differ distinctly from those recovered from the domestic
architecture. The function of this structure remains unclear, but its size and form suggest communal and/or ceremonial
activity as the reason for its construction. Architecture indicative of communal activity and of a monumental nature is
reminiscent from some features of other Anatolian and Near Eastern monumental structures (Atakuman 2014). The
monumental structure at Ugurlu is probably associated with ritual activities that served as community regulatory
mechanisms during the Neolithic.
We suggest that archaeological remains at the mid-6th millennium BCE Ugurlu site demonstrate the early
development ofproto-urbanisation on the island. The settlement was divided into three sections, a large building in the
centre, a residential building complex in the east, and a ritual building and pits in the west. The structure in the centre
is not only architecturally complex but also seems to have served multiple functions, ranging from residential through
administrative and economic, to ceremonial and cultic. In the west, the large and richly filled pits were placed in an
open area that served as the courtyard of a special building where a collective human burial pit was also located. The
communal building and pits at Ugurlu may be associated with public events triggered by seasonal cycles, during
which symbolic acts of social regeneration were performed (Karamurat et al. 2021). However, much as we may wish
to disentangle secular from religious initiatives, administrative from ceremonial functions, or ideological from cultic
Goks;eada Ugurlu 31
purposes, it is unlikely that we will ever be able to distinguish satisfactorily among all these deeply entwined, closely
inter-related aspects of the mid-6th millennium BCE society at Ugurlu.
All of this indicates that Ugurlu is one of the most important prehistoric sites in the northeastern Aegean. Since
only part of the site has been excavated, it stands as an exceptional archaeological reserve for future study.
ACKNOWLEDGEMENTS
The Ugurlu Excavation Project is supported by the Turkish Ministry of Culture and Tourism, and also by the
Turkish Science Foundation (TUBiTAK project no: 114K271), University of Thrace Scientific Research Funds,
University of Antalya Scientific Research Funds, Middle East Technical University Scientific Research Funds, and the
Turkish Historical Society. We wish to thank Terry Christenson for his kind corrections to the language in this paper.
Also, we gratefully acknowledge all the members of the Ugurlu Excavation Project for their hard work which made
this project possible.
REFERENCES CITED
Altmbilek-Algul, <;::iler. 2011. Chipped Stone Industry ofYumuktepe: Preliminary Results from "The Early Neolithic"
Phase. Anatolia Antiqua 19: 13- 25.
Atakuman, <;::igdem. 2014. Architectural Discourse and Social Transformation During the Early Neolithic of Southeast
Anatolia. Journal of World Prehistory 27(1): 1--42. DOI: 10.1007/s10963-014-9070-4
Atakuman, <;::igdem, Hasan Can Gemici, Cansu Karamurat, and Burs;in Erdogu. 2018. Gokyeada-Ugurlu Hoyuk'te
bulunan Neolitik ve Kalkolitik Donem Figi.irinler. 33. Arkeometri Sonur;lan ToplantlSl 33: 175-187.
Atakuman, <;::igdem, Burs;in Erdogu, Hasan C. Gemici, ismail Baykara, Murat Karakos;, Paolo Biagi, Elisabetta
Starnini, Denis Guilbeau, Nejat Yucel, Didem Turan, and Murat Dirican. 2020. Before the Neolithic in the
Aegean: The Late Pleistocene and the Early Holocene Record ofBozburun-Southwest Turkey. Journal of Island
and Coastal Archaeology.
DOI: https://doi.org/10.1080/15564894.2020.1803458
Boz, Ba~k and Burs;in Erdogu. 2019. A Sixth-Millennium BC Burial Pit at Ugurlu on the Island of Goks;eada.
Antiquity 93(369): 1-5.
Broodbank, Cyprian. 2000. An Island Archaeology of the Early Cyclades. Cambridge: Cambridge University Press.
Chalkioti, Areti. 2016. Reconstructing the Coastal Configuration of Lemnos Island (Northeast Aegean Sea, Greece)
Since the Last Glacial Maximum. In Geoarcheologie des fies de Mediterranee, M . Ghilardi, ed., 109-118. Paris:
CNRS Editions.
yevik, Ozlem and Burs;in Erdogu. 2019. Multiple Faces of Changes in 5600/5500 cal. BC Anatolia and Thrace.
Anatolica 45: 1- 16.
- . 2020. Absolute Chronology of Cultural Continuity, Change and Break in Western Anatolia Between 6850-5480
cal. BC: The Ulucak Hoyuk Case. Mediterrannean Archaeology and Archeometry 20(1 ): 77-92.
yevik Ozlem and Osman Vuru~kan. 2020. Ulucak Hoyuk: Pottery Emergence in Western Anatolia. Documenta
Praehistorica XLVII: 96-109.
<;::ilingiroglu, <;::iler. 2017. The Aegean before and after 7000 BC Dispersal: Defining Patterning and Variability. Neo-
Lithics l (16): 32--41.
<;::ilingiroglu, <;::iler, Malgorzata Kaczanowska, Janusz K. Kozlowski, Berkay Dins;er, Canan <;::akrrlar, and Didem
Turan. 2020. Between Anatolia and the Aegean: Epipalaeolithic and Mesolithic Foragers of the Karaburun
Peninsula. Journal ofField Archaeolozy. DOI: https://doi.org/ 10.1080/00934690.2020. l 786929
Dawson, Helen. 2014. Mediterranean Voyages: The Archaeology of Island Colonisation and Abandonment. Walnut
Creek: Left Coast Press.
Douka, Katerina, Nikos Efstratiou, Mette Marie Hald, Peter Steen Henriksen, and Alexandria Karetsou. 2017. Dating
Knossos and the Arrival of the Earliest Neolithic in the Southern Aegean. Antiquity 91(356): 304- 21.
Erdogu, Burs;in. 2011. A Preliminary Report from the 2009 and 2010 Field Seasons at Ugurlu on the island of
Goks;eada. Anatolica 37: 45- 65.
- . 2014. Goks;eada Ugurlu Archaeological Project: A Preliminary Report from the 2011- 2013 Field Seasons.
Anatolica 40: 157-178.
- . 2017a. Whither the Aegean Neolithic. In Going West? The Dissemination of Neolithic Innovations Between the
Bosporus and the Carpathians. A. Reingruber, Z. Tsirtsoni, and P. Nedelcheva, eds., 29--42. London: Routledge.
- . 2017b. A Preliminary Report on the Earliest Neolithic Levels at Ugurlu on the Island of Gokceada. Anatolica 43:
71- 82.
Erdogu, Burs;in and <;::igdem Atakuman. 2021. The Early Pottery from Ugurlu on the Island of Goks;eada, North-
eastern Aegean. In Neolithic Pottery from the Near East: Production, Distribution and Use, Proceedings of the
2019 Third International Workshop (Antalya, Turkey), R. Ozbal, M. Erdalk1ran, and Y. Tonoike, eds. 321- 327.
istanbul: Ko<; University AKMED Publications.
Fletcher, Alexandra, Douglas Baird, M. Spataro, and Andy Fairbairn. 2017. Early Ceramics in Anatolia: Implications
for the Production and Use of the Earliest Pottery. The Evidence from Boncuklu Hoyiik. Cambridge
ArchaeologicalJournal 27(2): 351-369.
Gemici, Hasan Can and <;igdem Atakuman 2021. The World of Figurines in the Neolithic and Cha/eolithic Aegean.
The Case of Ugurlu Hoyiik on Gokc;eada (Imbros). Oxford: BAR Publishing, International series 3021.
Gerritsen, Fokke A., Rana Ozbal, and Laurens Thissen. 2013. The Earliest Neolithic Levels at Barcm Hoyuk,
Northwestern Turkey. Anatolica 39: 53-92.
Guilbeau, Denis, Nurcan Kayacan, <;iler Altmbilek-Algiil, Burs;in Erdogu, and Ozlem <;evik. 2019. A Comparative
Study of the Initial Neolithic Chipped-stone Assemblages ofUlucak and Ugurlu. Anatolian Studies 69: 1-20.
Hamilakis, Yannis, Nina Kyparissi-Apostolika, Thomas Loughlin, Tristan Carter, James N. Cole, Yorgos Facorellis,
Stella Katsarou, Aggeliki Kaznesi, Areti Pentedeka, Vasileios Tsamis, and Nicolas Norzin. 2017. Koutroulou
Magoula in Phthiotida, Central Greece: A Middle Neolithic Tell Site in Context. In Communities, Landscapes, and
Interaction in Neolithic Greece, Proceedings of the International Conference, Rethymno 29- 30 May, 2015, A.
Sarris, E. Kalogiropoulou, T. Kalayci, and L. Karimali, eds., 81-96. International Monographs in Prehistory 20.
New York and Oxford: Berghahn Books.
Horejs, B., B. Milic, F. Ostmann, U. Thanheiser, B. Weninger, and A. Galik. 2015. The Aegean in the Early 7th
Millennium BC: Maritime Networks and Colonization. Journal of World Prehistory 28: 289-330.
Horejs, Barbara. 2019. Long and Short Revolutions towards the Neolithic in Western Anatolia and Aegean.
Documenta Praehistorica XLVI: 68-83.
Karamurat, Cansu, <;igdem Atakuman, and Burs;in Erdogu. 2021. Digging Pits and Making Places at Ugurlu During
the Sixth Millennium BC. Oxford Journal ofArchaeology 40(1 ): 23-42.
Knapp, Bernard A. 2008. Prehistoric and Protohistoric Cyprus. Identity, Insularity, and Connectivity. Oxford: Oxford
University Press.
Kotsakis, Kostas. 2014. Domesticating the Periphery: New Research into the Neolithic of Greece. Pharos 20(1): 41-
73.
Ozbek, Onur and Burs;in Erdogu. 2014. Initial Occupation of the Gelibolu Peninsula and the Island of Goks;eada
(Imbroz) in the Pre-Neolithic and Early Neolithic. Eurasian Prehistory 1(1-2): 97-128.
Ozdogan, Mehmet. 2009. Earliest Use of Pottery in Anatolia. In Early Farmers, Late Foragers, and Ceramic
Traditions: on the Beginning ofPottery in the Near East and Europe, D. Gheorghiu, ed., 22-43. Newcastle Upon
Tyne: Cambridge Scholars Publishing.
Ozdol, Serap. 2012. The Development and Traditions of Pottery in the Neolithic of the Anatolian Plateau. BAR
International Series 2439. Oxford: BAR Publishing.
Perissoratis, Constantine and Nikolaos Conispoliatis. 2003. The Impacts of Sea-Level Changes During Latest
Pleistocene and Holocene Times on the Morphology of the Ionian and Aegean Seas (SE Alpine Europe). Marine
Geology 196: 145- 156.
Perles, Catherine. 2001. The Early Neolithic in Greece. Cambridge: Cambridge University Press.
Perles, Catherine, Anita Quiles, and Helene Valladas. 2013. Early Seventh-Millennium AMS Dates from Domestic
Seeds in the Initial Neolithic at Franchthi Cave (Argolid, Greece). Antiquity 87: 1001-1015.
Pilaar-Birch, Suzanne E., Levent At1c1, and Burs;in Erdogu. 2019. Spread of Domestic Animals Across Neolithic
Western Anatolia: New Stable Isotope Evidence from Ugurlu Hoyuk, the Island ofGoks;eada, Turkey. PLOS One,
October 10:
DOI: https://doi.org/10.1371/joumal.pone.0222319
Reingruber, Agathe. 2011. Early Neolithic Settlement Patterns and Exchange Networks in the Aegean. Documenta
Praehistorica 38: 291- 305.
Roodenberg, Jacob, Krassimir Leshtakov, and Vanya Petrova, eds. 2014. Yabalkovo. Volume 1 (Maritsa Project 2).
Sofia: Ars et Technica Explicatus.
Rosenberg, Michael and Brian L. Peasnall. 1998. A Report on Soundings at Demirkoy Hoyuk: an Aceramic Neolithic
Site in Eastern Anatolia. Anatolica 24: 195-207.
Sampson, Adamantios. 2014. The Mesolithic of the Aegean basin. In La transition neolithique en Mediterranee: actes
du colloque Transitions en Mediterranee, ou comment des chasseurs devinrent agriculteurs, Toulouse, 14-15 avril
2011. C. Marren, T. Perrin, and J. Guilaine, eds., 189- 207. Aries: Editions Errance.
- . 2019. Paleolithic and Mesolithic Sailors in the Aegean and the Near East. Cambridge: Cambridge Scholars
Publishing.
Sampson, Adamantios, Malgorzata Kaczanowska, and Janus K. Kozlowski. 2010. The Prehistory of the Island of
Kythnos (Cyclades, Greece) and the Mesolithic Settlement at Maroulas. Krakow: The Polish Academy of
Sciences.
CHAPTER FOUR
THE INNER WESTERN ANATOLIAN PREHISTORIC PERIOD PORSUK CULTURE:

THEKANLITA$HOYUKPROJECT2018FINALEXCAVATIONREPORT
ALi UMUT TORKCAN AND DUYGU ERTEMiN
Anatolian archaeology demonstrates an extensive settlement pattern for the Early Chalcolithic periods in the
southern and northern regions of the peninsula, for example, in the Konya plain (West <;::atalhoyuk, Can Hasan I), the
Lake District (Hacilar, Kuru9ay), in eastern Cappadocia (Tepecik-<;::iftlik, Ko~khoyuk), in the Aegean (Ulucak
Hoyuk), and the eastern Marmara region (Fikirtepe, Ihpmar, and Aktoprakhk). However, settlements become fewer
and unknown in the northern parts of western Anatolia, including the Sakarya, K1z1hrmak, and Porsuk River basins
(Fig. 4-1) (Turkcan 2018). The topography of the upper Porsuk area comprises wide valleys cut by sometimes high
hill reliefs, with many tributaries, especially in the western region, which encompasses Eski~ehir. This complex
topography provided opportunities for many types of habitation, including both farming and herding as well as
hunting during the 6th millennium BCE (Turkcan 2018: 564-565). Three Chalcolithic sites in the Upper Porsuk region,
Kanhta~, Orman Fidanhg1, and Keskaya, are noteworthy as hilltop settlements. Kanhta~ Hoyuk, as the best preserved
Early Chalcolithic settlement in the Porsuk River basin, provides valuable information for the period and the region.
+
• Archaeological Site
-River
D Water Body
0
• CAPPADOCJA
•
Kuru~ay \') I(ONYA :LAIN Tepecik-<;:iftlik
Hactlar tJ" •
• <;:atalhoyiik West 0
0 0
0.
50 100
Figure 4-1. Regional map with the sites mentioned in the text
and the position of Kanhta~ Hoyuk in northwestern Anatolia.
Kanht~ Hoyuk ("bloody stone" in English), which leans against a sharp outcropping of basalt bedrock, is situated
in a small valley called Kuzfmd1k, located in the Eski~ehir/inonu Province, at the edge of a mountainous region in
northwestern Anatolia (Turkcan 2011) (Fig. 4-2). The Chalcolithic character ofKanhta~ Hoyuk was established during
the survey of the region in the 1990s by Turan Efe (Efe 1990), who also conducted a rescue excavation project
between 1992 and 1994 for another Porsuk River basin settlement, Orman Fidanhg1, situated on a steep rocky hillside
34 Chapter Four: Turkcan and Ertemin
near the Porsuk River (Efe 2001). Based on that work he developed the concept of an "Upper Porsuk Culture."
Kanhta~ Hoyuk has Early Chalcolithic habitation cultural material similar to that characteristic at the settlement of
Orman Fidanhg1, and provides a series of C 14 calibrated dates for the first time in this area (Table 4-1) (Turkcan et al.
2017).
The Kanhta~ Hoyuk Excavation Project, led by
Ali Umut Turkcan from Anadolu University, began
research in the area, and on the mound, as an
. .
- ·•·
intensive survey research for three seasons from
. -~->' ' 2008 until 2012 (Turkcan 2011). This survey
covered an area of ca. 6 km2, including major
portions of the inonu Valley, Kuzfmdik Valley, and
the Kuzfmdik Dam area. The survey uncovered
promising results for future research at Lower
Palaeolithic to Early Ottoman rural settlement
patterns; Early Bronze Age sites were found mainly
around Kanhta~ and also on the eastern slopes of
Kanhta~ Hoyuk. This intensive survey revealed an
extraordinary number of marble bracelet pieces
demonstrating their production processes, a rich
array of chipped stone tools with many raw material
types (e.g., opal, quartz, chalcedon, and radiolarite),
Figure 4-2. View ofKanhta~ Hoyuk in and a rich ceramic assemblage with painted and
Kuzfmd1k V allev from the southeast. incised decorated examples, particularly identified
during the survey between the years 2008-2009.
Afterwards, following the survey, the project
became a systematic excavation of the mound
LAB T R ENCH
SAM PL E UN I T UN I T SAM PLE
CA L BCE between 2013 and 2018, using a single context
NO D ESCR I PTION DESCRIPTION
excavation and recording system to understand the
BETA 0-15 200b Stone waU rubble Animal bone 5835-5825/5810-5715
Porsuk settlement patterning and its cultural mater-
BETA 0- 15 200a Stone wall rubble Burnt wood 5720-5630 ial contextually within space and time. The project
BETA N-17 301 Subsurface deposit Burnt cob
2880-2830/2820-
2660/2650-2635
came to an end in 2019 with a study season, during
BETA N-17 302b Room fill Clay(bumt plaster) 5320-5215 which intra-site spatial distribution of the settlement
BETA N-17 302a Room fill Burnt wood 5630-5530
was studied, and Harris Matrices of the excavated
areas and important contexts were produced and
BETA N-17 303 Room fill Animal bone 5670-5620
elaborated upon. In the 2018 final excavation
BETA 0-17 501 Subsurface deposit Anirnal bone 5645-5605
season, four constructional phases of the site were
BETA 0-17 502f Room fill Burnt wood 5710-5620 identified according to intra-site settlement patterns;
BETA 0-17 502e Room fill Bumi wood 5630-5530 in addition, several production activities are attested,
BETA 0-16 601 Subsurface deposit Burnt wood 5735-5645 such as that of marble bracelets, pigment, chipped
BETA 0-16 605 Ashy layer 8l011t animal bone 5665-5615 stone tools, and a rich ceramic assemblage, all
identified during the survey and the excavations
(Baysal et al. 2015; Turkcan 2009; Turkcan et al.
Table 4-1. Kanhta~ Hoyuk cal. Cl4 dating data. 2017).
THE 2013-2017 KANLITAS EXCAVATIONS IN THE HILLTOP AREA:

THE EARLY CHALCOLITHIC PERIOD 1
During the first year of the excavations, work focused on the southwestern section trenches (0-14, 0-15, and N-
15), situated in the northwestern portion of the hilltop area, revealing the architectural features of Kanhta~ Hoyuk.
These include different wall construction techniques such as the use of mudbrick, employed in one example on
sandstone slabs along with stone foundations (0-14), rock cut spaces, and stone walls, along with open spaces. These
latter contain hearth features surrounded by stone pavement and compact mud-packing layers, and were believed to
have been used as a work area, mainly for marble bracelet production.
In 2014, excavations focused on the southeastern section of the hill in trenches N-16, N-17, 0-16, 0-17, and P-16,
which revealed domestic spaces with stone foundations and mud-plastered walls. Excavations continued in these
trenches in 2015, while new trenches in the east (NO-18), north (M-16/17, M-18) and west (N-14) sections were also
1http://www.kanlitas.com/assets/files/2016.pdf, http://www.kanlitas.com/assets/files/2017 .pdf,

http ://www.kanlitas .com/assets/files/2013.pdf, http://www.kanlitas.com/assets/files/2014.pdf,
http://www.kanlitas.com/assets/files/2015.pdf. All detailed excavation reports can be accessed on the website (www.kanlitas .com);
four years of excavations have been published periodically in Kutahya il Y1lhg1 Books (Turkcan 2015, 2016 ; Ttirkcan et al. 2017) .
The Inner W estem Anatolian Prehistoric Period Porsuk Culture 35
opened. Building 1 is identified by partition walls that create six different spaces in the 2015 season. In some of these,
postholes suggest that the space was only partially roofed; mud wall foundations were accompanied by three small
posts that perhaps supported a grass thatch roof. Another contribution to understanding the site stratigraphy was made
with a deep sounding excavation in 2015. This revealed five different earlier sequential depositions in the southeastern
comer of trench M-1 7.
Intensive magnetometer and GPR surveys were conducted around the site, especially on the western and eastern
sloping fields, by Mahmut Drahor and his team in the 2015 excavation season. 2 Magnetometer studies in the western
field revealed a heavily burned large ca. 20 x 5 m rectangular building, with a large oval hearth in the middle, on the
same axis (southwest-northeast) as the Chalcolithic buildings unearthed on the settlement's hilltop. This might
suggest that this heavily burned structure was a freestanding building, contemporary with the hilltop architecture.
In 2016, excavations continued in trenches Ml6/l 7 and N-17, while larger trenches in the east (NO-19) and south
(Pl 7/18), were excavated to understand the terracing system and use of bedrock in architecture. Also, the
abandonment processes in Building 1 were clarified.
In 2017, excavations focused on the northern section of the hilltop in trenches L15, Ll6/17, and M-15. Excavation
in trench L-17 revealed the first burial discovered at Kanhta~ Hoyuk, which belonged to a young adult woman in a
fully flexed position and covered with stones. It appeared that the burial cut truncated the northern wall of Building 1.
Also, in trench L-16, a burnt ceramic kiln with channels in its base (for the placement of pots), belonging to an earlier
architectural phase, was exposed.
THE 2018 FINAL EXCAVATION SEASON RESULTS
The purpose of the 2018 excavations 3 was to understand the sequential stratification between the construction
phases of the buildings and spaces (Fig. 4-3). With this aim, excavations focused on the trenches where the
connections between spaces and deposits were observed (M-17, 0-16, N-16, Pl5, P-16, and M-14), along with the
eastern trench IJ-25 down the hill on the eastern slope, where an Early Bronze Age (EBA) occupational phase of the
mound was revealed. Excavations in trench M-17 helped in understanding the spatial connections between M-17, N-
17, and 0-17, while areas excavated in trenches 0-16 (Sp.8) and 0-17 (Sp.4) were understood to be the same space,
with stone foundation walls packed with yellowish mud layers. These connections clarified the existence of another
building (Building 2) which Building 1 abutted on its west. Also, deep sounding work revealed another wall dating
earlier than the Building 2 construction.
14 15 16 17 18 19 25
-- . ---- ~ - - - I ---,
I
I
I
L
z%2859
' -
M
i:tf'" N
:~:·
0
&;i(m
p
Figure 4-3. Kanhta~ Hoyuk site plan of the Chalcolithic hilltop.
2
http://www.kanlitas.com/assets/files/2015 .pdf; see especially pp. 131-132.
3
http://www.kanlitas.com/assets/files/2018.pdf.
THE SUMMIT AREA EXCAVATIONS:

THE EARLY CHALCOLITHIC PERIOD SETTLEMENT
TrenchM-17
Trench Ml6/l 7 was excavated in the 2015 and 2016 excavation seasons; several spaces (Sp.10, Sp.11) were
exposed, along with important features such as a fire installation, storage bin, and refuse pits.
In 2018, excavations continued, but only in Trench M-17 (5 x 5 m); the western part (M-16) was excluded from
excavation work. The aim of the excavations in this trench was to reach deeper levels and to better understand the
stratigraphy of the mound better, since in the 2015 and 2016 seasons, deep soundings which had already been
undertaken in the southeastern comer of this trench had revealed a stone wall, initially thought to be part of an
enclosure wall surrounding the settlement.
Excavations on the walls previously exposed in 2015 continued in 2018. The grey mud wall, named U.652, is
abutted by the yellowish mud wall, F.800. It was important to understand the construction sequences of these walls
since the U.652 wall feature belongs to Sp.11 of Building 1. As a result, it became clear that the F.800 yellowish mud
wall was constructed on a stone wall foundation which begins in the southwestern comer of trench M-18, extends to
the southeastern comer of trench M-17, and curves to trench N-17, to connect with the Sp.3 yellowish mud wall
(Ertemin and Sahin 2018: 25-29).
Another important architectural discovery was the layer upon which the grey mud wall rests. It is now understood
that the eastern wall in Sp.11 (F .652) extends to trench N-17, abuts the stone foundation mud wall, and was
constructed after wall F.800 was built. Also, the fact that these two wall features abut each other and extend into
trench N-17 helped revealed another space belonging to Building 1, between Sp.5 in trench N-16 and Sp.11 in trench
M-17. The fill in this newly identified space was excavated in the 2015 season (U.410); the texture was identified as a
silty ashy midden-like layer, similar to the fill of Sp.5 (U.408).
After the walls of Sp.11 were excavated and the layer upon which they sit was exposed, the top layer of another
earlier yellowish mud wall (U.658), with a posthole (U.662) in the middle of it, was exposed.
Space 18 and The Deep Sounding
Excavations in trench M-17 revealed important architectural and sequential knowledge about the construction of
the spaces and buildings. When the excavations started in the trench, the aim was to execute a deep sounding to
understand the stratigraphy and the construction of the stone wall found in the 2015 deep sounding, thought at the time
to be an enclosure wall surrounding the settlement.
After understanding the upper layer of this stone
foundation connection with Sp.3, work shifted from
the deep sounding to Space 18, which had been
identified and partly excavated in the 201 7
excavation season. Space 18 is situated north of this
stone wall foundation (F.800) in the southern half of
trench L-17; the northern part of the space was
identified in 2016 and named Space 14. Since the
northern and southern parts of this space were
identified in different trenches in different
excavation seasons, different space numbers were
assigned. Space 18 refers to the southern area, while
Space 14 refers to the northern part. Walls in this
space were made with simple, thin (25 cm wide)
yellowish mud (Fig. 4-4). The southern part of the
fill in this space, situated in trench M-1 7, was
excavated in combination with unit number 667.
After excavating 4 cm of fill, an oven (F.655) was
Figure 4-4. Trench M-17, Space 18 Orthophoto. identified in the southwestern comer. After Sp.4,
Space 18 is the second space within which an oven
has been identified. It should be noted that a packed mud platform (Y.2YL) was Hientlhect m the northwestern comer ot
this space in 2016. In general, since this space included domestic features related to food preparation practices, it was
identified as a domestic space, thought to be used during the construction of Building 1.
After the identification of this space, a sounding was excavated in the southern area of the space, situated on the
northern part of trench M-1 7. One of the most important features found in the deep sounding was a brick wall,
oriented in an east-west direction, with a stone foundation and proper yellowish mudbrick and grey-coloured mortar
(Fig. 4-5). The upper part of the wall, with brick-and-mortar superstructure, had been truncated, and mudbrick was
scattered throughout Unit 670; the western and eastern sections of the deep sounding clearly showed the brick-and-
mortar wall construction truncated in the middle. Under the sandy, silty mixed layer, which included scattered
The Inner Western Anatolian Prehistoric Period Porsuk Culture 37
fragments of bricks in the layer, a midden-like dark, ashy layer (U.671) covering the stone foundation of the mudbrick
wall (F.659) was identified. After this stone foundation was excavated, the base (U.677) of the hearth feature (F.657)
was found abutting the bedrock on the north, situated at the very bottom of the deep sounding. After revealing this
hearth feature (Fig. 4-6), the deep sounding work in trench M-17 ended. Many yellow and red pigments were
recovered from U.671 and U.677; ground stone artefacts with pigment traces were also found in the units.
Figure 4-5. Mudbrick wall found in Figure 4-6. Fire installation

Space 18 in the deep sounding. found in the deep sounding.
TrenchM-14
The 5 x 5 m M-14 trench is situated next to trench N-14 on the north and M-15 on the east. Just like trench N-14,
excavated in 2015, the eastern half of the deposition in this trench sloped downward to the west, with a 2 m elevation
differential. Excavations began on the highest point in the eastern half of the trench with unit number U. 750.
The topsoil (U.750) was cleaned and scraped for 5 cm, within which fragments of marble bracelets were found.
Due to the increase in the number of artefacts found in this layer, excavations continued into unit number U.751; an
ashy deposit in the southeastern comer of this unit contained a large flat rock bearing red pigments. This comer
(U.752) was excavated, and red pigment samples were taken. Although there was no identifiable feature in this trench,
the many artefacts found in this eastern half of the trench, such as marble bracelet fragments, drilling cores for bead
making, and magnesite and sandstone bracelet fragments, may have been prepared as training pieces. Also, in U.751,
marble bracelet roughcasts, fragments, core drillings, polishing stones, chipped stone tools, and mussel shells were
found. The layer found under U.751, on the northeastern side of the trench, was designated U.753; an in situ grinding
stone and hammer stone were found next to each other.
The artefacts found in the trench suggested that this area was used as an open space work area, especially given its
position north of trench N-14 (an open space work area) and west of M-15 (a workshop area); it seems clear that
trench M-14 was used as an open space work area as well.
Given the collection of related units and features between the workshop spaces, including U.752 situated in the
southeastern comer of trench M-14, which is in tum related to the F.652 posthole in the northeast area of Trench N-14,
and the F.003 platform in the northwestern comer of trench N-15 (the open space workshop area), it can be suggested
that open space workshop areas were concentrated on the western part of the settlement.
TrenchN-16
Excavation work in Trench N-16, last excavated in 2015 when the southern portion of Building 1 was revealed,
continued in the 2018 season. In 2015, a yellowish mud wall abutting the southern walls in Sp.5 and Sp.6 was
identified and recorded as F.400. In the 2018 season the aim of the excavations was to understand the construction
technique and sequence of this wall, situated north of trench 0-16. F.400 was excavated with unit number 411; after
excavating the top mud layer, a stone foundation wall abutting the western stone wall (F.200) of Sp.8 in trench 0-16
was revealed.
Another important discovery in this trench was the U.415 cluster of finds. This cluster unit had a midden-like silty
ashy soft fill containing several artefacts such as chipped stone tools, a hammer stone, a broken loom weight, and large
animal bones, which suggested butchery activity. A total of 60 liters of large and small size stones were recovered
from the fill, some of which were intentionally put on the bottom of this feature before it was filled with stones and
artefacts.
38 Chapter Four: Tiirkcan and Ertemin
Trench 0-16
The 2014 excavations in trench 0-16 revealed a space (Sp.8) along with a hearth feature abutting the southern wall
of this space. In the 2018 excavation season, the purpose of the excavations in this trench was to record and excavate
the hearth and wall features, and then proceed deeper to find the earlier levels of architectural elements. Since the
appearance of these above-mentioned features was different from what was revealed in 2014, it was decided to record,
photograph, and draw these features with new unit and feature numbers. The new designation F.700 replaced F.200, a
north-south oriented wall in Sp.8 to the north, F.701 became F.201, an east-west oriented wall also in Sp.8 to the
south, and F.702 became the hearth (formerly F.202) (Fig. 4-7). Excavations in the trench therefore began with U.702,
the top layer within the hearth. Within this fill, scattered red pigment, a worked stone with red pigment marks, which
was later identified as a colour palette, a ground stone tool with red pigment traces, and a pottery base were found.
Figure 4-7. Trench 0-16 Space 8 orthophoto.
Excavations then proceeded to expose a second burnt ashy fill (U.703) in the hearth, as well as a truncated base
(U.704), a cut, and the stone foundation of the hearth (U.705, U.706). This allowed us to understand that the hearth
was built shortly after the stone foundation walls of Sp.8 were built, and that the hearth abutted the southern wall
(F.701) of Sp.8.
Excavations proceeded to focus on the southern half of the trench. Deposition in this area, which is dark grey and
ashy (U.709 and U.710), is thought to be an open waste area or work space between the northern part of trench P-16
(U.610) and Sp.8 in trench 0-16. The western part of the trench north of U.710 was identified as a stone pavement,
consisting of a sandy, yellowish packing mud mixed with small pebble stones. This layer (U. 711) was identical to the
U.201 yellowish packing layer observed in trench 0-15. This context may be interpreted as an open work area related
to the F.702 hearth, and stone pavements (yellowish mud-packing layers), creating a platform with ashy deposits
around it in trenches 0-16, P-16, and 0-15. This ashy layer (U.710) produced animal bones, chipped stone tools, a
pottery sherd, a needle, and a mussel shell.
In the 2014 season it was discovered that the northwestern comer of this trench had been cut due to illegal
excavations. The limit of this truncation was U.711 to the south. The area east of this truncation was excavated as unit
number 712 and later 714. During the excavations ofU.714, red pigments and a worked stone with red pigment traces
were found; a sample of the pigment (U.714-Sample no. 3) was taken, and the stone was recorded as a special find
(U.714-Xl). Excavations in this layer revealed an earlier feature, a circle made with stones west of wall F.700.
The Space 8 and Space 4 Connection
Another important area of excavation work was in U.716, conducted in Sp.8. The purpose of excavation in this
layer was to understand the connection/relation between Sp.8 and Sp.4 to the east in trench 0-17. Excavation ofU.716
revealed the stratigraphic relationship of Sp.8 and Sp.4. The mud wall construction in Sp.4 and the stone walls
defining Sp.8 were revealed during different excavation seasons and in different trenches, and thereby recorded with
different space numbers. After observing stone foundations under the mud layer of the wall in Sp.4, and the same
deposition in colour and texture in both Sp.8 and Sp.4 (U.716), it became clear that Sp.4 and Sp.8 are a unified space.
Following this realization, the fill in this area was excavated as a single context (U.718); many special artefacts were
found, such as a piece of hammer stone and a scapula, in the southeastern part of the unit, while the northern area
revealed a cluster of special finds. This cluster included two phallus-shaped stone pestles, three relatively large
ceramic pieces, an arrowhead, a worked stone, and an animal figurine with a broken head.
Trench P-16
Trench P-16 was excavated very briefly to understand the open work area deposition and its connection to trench
0-16. Layer U.610, exposed in 2014, was re-opened with new unit number U.715. This ashy layer is packed with
smaller size stones and is 15-16 cm deep; red pigments from the northern section of this unit were found and sampled
(U.715-Sample no. 3). It should be noted that the northern limit of this unit was trench 0-16, where many examples of
red pigments and worked stones with pigment traces on them were found. The boundary between trenches P-16 and
0-16 was excavated as unit number U. 71 7; a partial animal jaw bone (U. 71 7-X 1), a worked animal bone (U. 717-X2),
a needle (U.717-X3), a spindle whorl (U.717-X5), and a mussel shell (U.717- X6) were found along with a polishing
stone (U.717-X4), in the vicinity of where the majority of the ceramic fragments were recovered. These find contexts,
and the connection between the ashy layers of trenches P-16 and 0-16, suggest that this area was an open work space,
perhaps a narrow passageway to the other open work spaces in trenches N-14, N-15, N-16, M-14, 0-14, and 0-15.
Trench P-15
Excavations in Trench P-16 revealed a block of bedrock extending from the southeast to the northwest in the
middle of the trench, which was later identified as the northern wall. This wall is the southern limit of the open area
and passageway described above. The discovery of this wall made it necessary to extend the excavation work into
trench P-15, situated west of trench P-16 and northwest of trench 0-15. Initial scraping of this revealed bedrock
situated at the southern limit of the trench; the bedrock was shaped to create a wall similar to one found in trench P-16.
These two bedrock walls were connected by a stone foundation wall situated in the eastern part of the trench creating a
rectangular-shaped space, which was later named Space 17.
Space 17
The fill in space 17, U.721 and later U.722, was 25 cm in depth and had no architectural features within the space.
However, interestingly, a hammer stone (U.721-Xl), a needle (U.721-X2), and worked stones (U.722-X2, X4), which
are thought to be pigment palettes, were found. Excavations in the eastern part of the trench revealed a V-shaped
collapsed stone wall, which could be the entrance to this area. Excavations in this space were not finished when the
excavation season ended, and the floor of this space was not identified, nor were architectural features discovered.
Fairly large and rectangular in size, with strong bedrock walls, this space may have been used as a domestic house, but
more excavation work is needed to support this idea.
TRENCH I-J25: THE EASTERN SLOPE AREA EXCAVATIONS

AND THE RECOVERY OF THE EBA PERIOD AT KANLITAS HOYUK
Trench I-J25 (5 x 10 m), situated on the eastern slope of the hoyillc, revealed Early Bronze Age occupation. The
evidence for the EBA period was first discovered in systematic collections during the 2008- 2009 years of intensive
survey made on the mound which recovered diagnostic pieces of EBA II ceramics (Tiirkcan 2011).
This trench was excavated for the first time in the 2018 season; work began in the northern part of the trench (I-25,
5 x 5 m) with U.550 as the surface scrape. This revealed mixed material culture, with the majority of the ceramics
belonging to the Bronze Age and some to the Early Chalcolithic; pieces were eroded from the summit area due to
slope wash. It is determined that U.550 was not a secure context due to the natural erosion of the soil on the slope.
Under this mixed layer, a secure deposition (U.551) revealed portions of the burnt base of a truncated hearth. The
northwestern comer of the trench was excavated (U. 553) to the level of the layer upon which the hearth was situated;
however, no other features were found in the trench other than this truncated fire installation.
The southern section of the trench (J-25, 5 x 5 m) was excavated simultaneously with the northern section (I-25),
and an accumulation of a thick orange/yellow mud layer was found (U.555). Many characteristic EBA ceramics were
found in this unit. Two of them (U555-X2 and X3) were nearly complete vessels, with neck and handles. The eastern
part of trench J-25 revealed a cut with an ashy and soft fill (U.556). From this fill a horn (U.556-Xl), many ceramic
sherds and animal bones, along with two animal figurines (U.556-Xl 1, X12), were found. Eight liters of soil were
taken as a flotation sample from this fill, which later revealed seeds of Triticum dicoccum (Emmer wheat) from
Poaceae (wheat) Familia, Lens culinaris (lentil) from Fabaceae (legume) Familia, and Vicia ervilia (tare) (Kavak and
(::akan 2018).
Excavations in the yellow mud layer (U.555) continued. Noteworthy is a standing carved stone from the bedrock
in this packed layer. Considering the cut with a deposit consisting of special finds (U.556) on the eastern edge of it, it
was believed that this packing layer covered a possible burial space.
40 Chapter Four: Tlirkcan and Ertemin
FINAL THOUGHTS ON THE CONSTRUCTION PHASES
The final excavation season revealed the architectural character of the Chalcolithic mound with two buildings
constructed using different techniques and materials, consisting of several spaces along with open space work areas
with fire installations and mud platforms/packed layers (Fig. 4-8). An Early Bronze Age II occupation of the site on
the eastern slope of the hoylik was also investigated. Here we aim to present the final thoughts on the construction
phases of the Chalcolithic occupation.
803 S . 10
Figure 4-8. Harris Matrix showing the Building 1 and Building 2 spaces.
The Kanhta~ Hoylik hilltop settlement was founded on a large rock outcrop, and the people of Kanhta~ used a
terracing system to construct the buildings and workshop areas. Excavations between 2013 and 2018 demonstrated
that the spaces were constructed on top of earlier features such as walls and fireplaces belonging to an earlier
construction phase. Another important construction characteristic of Kanhta~ is the use of different materials and
techniques in space and building construction such as rock cutting, stone foundation with mud on top, and mud
building with supporting posts, clearly demonstrating that the choices for construction materials and techniques
differed between phases and according to the function of spaces such as domestic spaces, buildings, and open area
workshops.
This interesting difference in construction choices, terracing, and lack of distinct floors in some spaces made it
harder to understand the phasing of the site. Prior to the 2018 season, the phasing of the site had been tentatively
interpreted. This interpretation suggested that the latest level found on top of the hill was the thick ashy layer exposed
in trenches N-14, N-15, and 0-15, where the spaces were identified by the rock cut technique. These trenches were
situated at the highest point of the mound; under this thick ashy layer rested undisturbed stone walls throughout the
trenches, leading to the possibility of an abandonment of the site after an extensive fire event. This ash level was
interpreted as the final occupation level in the western part of the settlement, associated with Level I (Turkcan 2014).
Excavations during the 2014 season revealed other rectangular or trapezoidal spaces in trenches N-16, N-17, 0-16,
0-17, and P-16 with different wall construction techniques such as mudbrick walls with stone foundations, and pise
walls, some of which covered stone foundation walls and rock cut spaces. These buildings were furnished with
domestic features such as storage spaces for crops, and ovens and hearths (Turkcan 2014). This different style of
construction and the presence of domestic features led to the designation of a Level II occupational level (Turkcan
2016). Level III was identified in the deep sounding work in trenches M-16/17 with a stone pavement and the midden
layer underneath as a packing layer (Turkcan 2016).
In the 2018 excavation season, the complete phasing of the settlement was decided according to construction
sequences, Harris matrices of site stratigraphy, and C14 dating results (BETA laboratory). The Kanhta~ Hoylik Early
Chalcolithic occupation situated on top of the mound has been divided into four construction phases (Table 4-2).
Levels Description Trenches/Spaces

M-16 ( U.802, U.805)
IA Bu i lding 1 abandonment
L- 16 ( F.257)
M-16, M-17, L- 16, N-16
1B Building 1 construction
(Space 5,6,10,11,15, 19)
IC Space 18 const ruction and occupati on M-17, L-17 (Space 18, Space 14)
M-17, N-17, 0 - 16, (Bu i ld ing 2)

Bui lding 2 construction and occupation
IIA P17/18, 0 -16, N-15, M-15, L-15,
Open space work areas
M- 18, NO- 18 (Open space)
0 -14, 0 - 15, P- 16/17

11B Bed rock spaces
(Space 1,2,3, 17)
Ill Deep sounding be low Space 18 M- 17 (F.659)
IV Deep sounding fire install ation M-17 (F.657)
Table 4-2. Overview of the construction levels.

Level I refers to the latest construction activity on the mound and is divided into three sub-phases: Phase IA, Phase
IB, and Phase IC. Phase IA represents the abandonment activities of Building 1 such as the truncation of the south
wall in Sp.6, midden pits (U.802, 805), and the truncating of the floor in Sp.10 in the south. The northern wall of
Sp.10 is also truncated by a burial cut (F.257). In addition, the southern wall of Sp.6, extending from trench N-16 to
M-15, belonging to Building 1, is interpreted as a later cut that occurred during abandonment activities.
Phase 1B refers to the construction phase of Building 1. This rectangular, light grey mud pise building consists of
six spaces (Sp. 5-6, 10-11, 15, and 19), divided by partition walls. Space 5 demonstrates the use of posts at the comer
of walls by the clear traces ofpostholes (Fig. 4-9). During excavations in 2018 in trench M-17, it was understood that
the eastern wall of Space 11 abutted the western wall of Building 2, which revealed that the construction of Building 1
happened after the construction of Building 2. This was an important discovery that helped in understanding the
construction sequences.
Phase IC refers to Space 14 in trench L-17 and
Space 18 in trench M-17. These were excavated in
different seasons, so they were given different
designations; however, they belonged to the same
construction as one unified space with its southern
wall abutting the northern stone foundation wall of
Building 2. Since this space was constructed earlier
than Building 1 and later than Building 2, it was
necessary to give it a separate phase name
representing a different sequence in construction
activities.
Level II refers to the sequence when rock cut
spaces and Building 2 were constructed and used
along with open space work areas. Different from
Level I, which was constructed with light grey
material using the pise technique, Level II construc-
tion included stone foundations with yellowish
Figure 4-9. Trench N-16, Space 5, part of Building 1. mudbrick on top, and the construction of spaces by
manipulating and shaping the bedrock ridges.
Level II refers to the sequence when rock cut spaces and Building 2 were constructed and used along with open
space work areas. Different from Level I, which was constructed with light grey material with pise technique, Level II
construction activities included the use of stone foundations with yellowish mudbrick on top, and the construction of
spaces by manipulating and shaping the bedrock ridges. Level II is divided into IIA and IIB. Level IIA is the sequence
when Building 2 was constructed and used. This rectangular building consists of six spaces divided by partition walls
as well (Sp.4, 8, 12-13, and 20-21). Since the building extends across eight different trenches, and these trenches were
excavated in different seasons, until the exploration of the stone foundations, these areas were interpreted as separate
spaces. In 2018, with the exploration of stone foundation walls in trenches M-17, 0-16, and 0-17, it was possible to
identify the single Building 2 structure across these spaces.
Building 2 is the most important structure demonstrating internal domestic features such as storage depots for
crops, an oven, and a platform. Also, interestingly, this building was surrounded by large hearths (approximately 110
x 150 cm) in the south, north, and east. As it was abutting Building 1 on the west and surrounded by hearths on the
other three sides, this construction choice may be interpreted as a building strategy, perhaps related to the
reinforcement of the terrace system.
It is also important to discuss the open space work areas within this phase. These open spaces were found in
several trenches, such as N-15, M-15, L-15, L-16, Pl 7/18, and NO-18. The characteristic feature of these spaces are
the hearths, production debris from marble bracelet and bead production, flintknapping debitage, and stone pavement
with thick packing deposit/yellowish mud platforms. Trench N-15 has the best example of these characteristics
(Turkcan 2013). A narrow passage starting from trench P-17 and extending to trench 0-16 leads the way to this
workshop area in trench N-15, with a stone pavement and packed platforms along with a hearth. Another important
open space work area was found in trenches L-15 and L-16. Fire installations found in these spaces were much bigger
than the hearths (110 x 150 cm in general), and the one in L-16 included a fire chamber with channels, which was
later interpreted as a ceramics kiln. The construction of these open space areas was attributed to Level IIA; however, it
seems likely that their use continued into construction Level IA.
Level IIB refers to the rock cut spaces, Sp.1-3 and 17, that were found in trenches 0-14, 0-15, P-15, and P-16.
These spaces are situated on the southern highest point on top of the mound. Although these spaces were interpreted as
the latest construction on the site during the first two years of the project, after understanding the terracing system and
comparing Cl4 dates from samples belonging to trench 0-15 and Building 2, it is now known that the construction of
these spaces occurred much earlier. In 2018, traces of cutting and manipulating the bedrock to create walls for a space
were also observed in Sp.17. The unique construction technique of these spaces, as well as their Cl4 dates, place them
earlier than the occupational sequence of Building 2, and they are therefore identified as Level IIB. Whether these
spaces continued to be used during Level IIA or not is still a question waiting to be answered.
42 Chapter Four: Tiirkcan and Ertemin
Level III refers to the construction sequence found during the deep sounding work in Sp.18, in trench M-17, in the
2018 season. A proper mudbrick wall with mortar, of at least four courses, and with an east-west oriented diagonal
stone foundation, was found in the deep sounding. Up until the 2018 excavation season, apart from the rock cut
spaces, wall features found at Kanhta~ consisted of pise, mud slab, and mudbrick, whether on a stone foundation or
not. However, the preservation of the mortar was poor in most cases. This mudbrick wall on a stone foundation clearly
showed another sequence of construction under Sp.18. Continued excavation of this wall in the deep sounding
revealed an ashy layer with a hearth feature under the stone foundation of the wall.
This discovery demonstrated an earlier sequence in which the use of the fire installation ceased, and a wall
belonging to a building was constructed on top. Continued work in the deep sounding in trench M-17 revealed earlier
deposition under the stone pavement, demonstrating earlier sequences. Due to this work in the deep sounding, Level
IV defines an earlier sequence found under the mudbrick wall construction including the deposition under the
pavement. Level III and IV, revealed during deep sounding excavations, require more work to understand their extent
on the settlement.
SPECIAL FINDS AND MATERIAL ANALYSES

FROM THE 2018 EXCAVATION SEASON
In terms of cultural material identification and analysis, between the 2013 and 2017 seasons, many production
practices were identified, such as chipped stone and marble bracelet industries, which were analyzed contextually. It
should be noted that the marble bracelet industry displayed the whole chaine repertoire of manufacturing processes,
including tools derived from local raw material sources such as magnesite grinding tools. These tools, probably the
first in Anatolia as well as in the Near East, were introduced in an earlier publication (Baysal et al. 2015).
In the 2018 excavation season, material culture analysis was promising for the identification of several other
production practices. For example, bone tool typology (awls, half epiphase-based awls, worked bone, waste bead core,
spatulas, antler haft, weaver's shuttle, bark extractor, and a geometric designed antler), and production techniques
(grinding, splitting, cutting, heat treatment, polishing, drilling, perforating) were identified (Fig. 4-1 0.a-b). It should
be noted that there are many signs of pigment production on many grinding tools, pestles, and on a small pot.
a) a) b)
b)
c) d)
Figure 4-10. Kanhta~ Hoyiik selected bone Figure 4-11. Examples from the Early Bronze Age
tool typology: a) examples of spatula; b) pottery assemblage found in trench IJ-25: a) Dark Faced
examples of antler hafts. Burnished Ware; b) Dark mouthed Ware; c) Brown
slipped and burnished; d) Red slipped and burnished.
A total of 23 pigment samples from different trenches and contexts were analyzed using SEM-EDX (Scanning
Electron Microscopy-Energy Dispersive X-ray Spectroscopy) in 2018 (Gtizel, 2018). These samples, collected during
excavations between 2013 and 2017, were mostly found on ground stones, pestles, and inside a pot. Analysis revealed
that the red and yellow colour were derived from ochre and hematite raw materials. These raw materials were
observed in the survey studies situated on the hill opposite to Kanhta~.
Another important contribution from the 2018 excavation season was the revealing of an EBA II presence on the
hoyiik based on its cultural materials, mainly the pottery, chipped stone, and figurine assemblages, which were
recovered in secure contexts. The especially rich ceramic assemblage dating to the EBA II (Fig. 4-11.a---d) consisted of
six whole vessels and 957 sherds with red and brown slipped decoration on rims, handles, and necks. According to
-Onan (2018: 110), this assemblage demonstrates a similar typology to the ceramic assemblages in Early Bronze Age II
levels at Kiilltioba, Seyitomer Hoyiik, Demircihoyiik, Demircihoyiik Sanket, Aizonai, Hoyiiktepe ve Kuru<;ay, Troya,
Kusura, and Beycesultan.
The Inner Westem Anatolian Prehistoric Period Porsuk Culture 43
This season was also very important for understanding the transformations within the chipped stone industry
through the various levels. Recovered in the 2018 excavation season, similar to previous years, were a flint nodule,
flint cores, a plunging blade, pieces with cortex, debris, blanks, and tools (Table 4-3). The lithic artefacts obtained
between the 2013 and 2017 excavation seasons at Kanhta~ Hoyuk revealed the Early Chalcolithic chipped stone
industry as a blade industry; however, the lithic artefacts uncovered in 2018 suggest the existence of a flake-dominated
industry in the early levels (Kolankaya Bostanc12018) (Fig. 4-12.a-b).
Re duction
Flint Opal Chalce dony Radio larite Obs idian TOTAL
Seque nce
Nodule I - - - - 1
Core 7 - 1 - 8
Plunging Blade 1 - - - - 1
De bris 806 2 19 72 29 - 1124

Blade 32 7 4 2 - 45
Blade-like Flake 40 2 1 - 43
Flake 69 12 6 5 - 92
Blade let I I I - 3
Tool 30 - - 2 l 33
Table 4-3.2018 season chipped stone industry data demonstrating

the reduction sequence and production according to raw materials.
a) b)
■■■■■ ■ ■ ■ ■ ■
Figure 4-12. Chipped stone industry data from the 2018 excavation season:
a) flint knives; b) flint and radiolarite ad hoc tools.
CONCLUSION
Owing to its strategic location, well-preserved architectural features, and open space workshop areas, along with
the material culture and production practices such as chipped stone tool production, decorated pottery assemblages
displaying Pre-Vinca affinities, outstanding marble bracelet industry with whole chaine repertoire of manufacturing
tools, frequent red pigment use, and production of many ground stone assemblages, the Kanhta~ Hoyuk prehistoric
settlement provides a great source of information for the understanding of Porsuk Culture in inner northwestern
Anatolia and the Chalcolithic character of the region in the 6th millennium BCE. Although the project came to an end
in 2019 with a study season, further research on the recovered artefacts is still ongoing.
ACKNOWLEDGEMENTS
The Kanhta~ Hoyuk Excavation Project is grateful for the contributions of D0<;. Dr. Adnan Baysal (Ankara
-Oniversitesi), D0<;. Dr. Emma Baysal (Ankara -Oniversitesi), Prof. Dr. Onur Ozbek (~anakkale Onsekiz Mart
-Oniversitesi) as ground stone specialists; D0<;. Dr. Neyir Bostanc1 Kolankaya (Hacettepe -Oniversitesi) chipped stone
specialist, Ar~. Gor. Dr. Hande Bulut (Dtizce -Oniversitesi) bone tool specialist, Ogr. Gor. N azan -Onan (Dumlupmar
-Oniversitesi) ceramic specialist, Dr. Ogr. Gor. Ayten ~ahk (~anakkale Onsekiz Mart -Oniversitesi) mineralogy and
petrography specialist, and Prof. Dr. Mahmut Drahor (9 Eylul -Oniversitesi) geophysics specialist, all of whom
participated in this research. Our thanks as well to Ogr. -Oyesi Salih Kavak of the ~ukurova -Oniversitesi Biology
department, Prof. Dr. Halil ~akan, laboratory supervisor, Ar~. Gor. Sevim Kurtuldu (Anadolu -Oniversitesi), the mayor
of inonU Kadir Bozkurt, dear friends from the village, Hahl ~avdar and Nuray ~avdar, and the field archaeologists
and student assistants who made the excavation work possible throughout the years. We would also like to thank Sean
Doyle for preparing the regional map for this publication.
REFERENCES CITED
Baysal, Emma, Adnan Baysal, Ali Turkcan, and Adam Nazaroff. 2015. Early Specialized Production? A Chalcolithic
Stone Bracelet Workshop at Kanht~, Turkey. Oxford Journal ofArchaeology 34: 235-257.
Efe, Turan. 1990. Three Early Sites in the Vicinity of Eski~ehir: Asmainler, Kanht~ and Kes Kaya. Anatolica 16: 31-
60.
- . 2001. The Salvage Excavations at Orman Fidanlzgi: A Cha/eolithic Site in Inland Northwestern Anatolia. lstanbul:
TASK.
Ertemin, Duygu and Semih ~ahin. 2018. M-17 As;masi Raporu. In Kanlzta~ Hoyiik (Eski~ehirllnonii) 2018 Yilz Kazi
<;alz~malarz Rapor, A. Turkcan, S. Kurtuldu, and D. Ertemin, eds., 25-44. T.C. Killtur ve Turizm Bakanhgi Kultur
Varhklari ve Muzeler Genel Mudurlugu. Retrieved from (http://www.kanlitas.com/assets/files/2018.pdf)
Guzel Emine, T. 2018. Kanhta~ Hoyuk Kazismdaki Pigmentlerin Analizi. In Kanlzta~ Hoyiik (Eski~ehirllnonii) 2018
Yzlz Kazi <;alz~malarz Rapor, A. Turkcan, S. Kurtuldu, and D. Ertemin, eds., 137-142. T.C. Killtur ve Turizm
Bakanhgi Killtur Varhklan ve Muzeler Gene! Mudurlugu.
Kavak, Salih and Hali! <;akan. 2018. Kanht~ Hoyuk 2018 Sezonu Arkeobotaniksel <;ah~malan. In Kanlzta~ Hoyiik
(Eskfyehirllnonii) 2018 Yzlz Kazz <;alz~malarz Rapor, A. Turkcan, S. Kurtuldu, and D. Ertemin, eds., 131-136. T.C.
Killtur ve Turizm Bakanhgi Kultur Varhklari ve Muzeler Genel Mudurlugu.
Kolankaya Bostanci, Neyir. 2018. Kanht~ Hoyuk Yontmat~ Endustrisi 2018 Yih Raporu. In Kanlzta~ Hoyiik
(Eskfyehirllnonii) 2018 Yzlz Kazz <;alz~malarz Rapor, A. Turkcan, S. Kurtuldu, and D. Ertemin, eds., 90- 107. T.C.
Killtur ve Turizm Bakanhgi Kultur Varhklari ve Muzeler Genel Mudurlugu.
Turkcan, Ali U. 2009. Preliminary remarks on Kanht~ (Eski~ehir) Chalcolithic Pottery. Seres09, I. International
Ceramic, Glass, Porcelain, Enamel Glaze and Pigment Congress, 148-153.
- . 2011. Kanht~ Hoyuk ve Civari (lnonu/Eski~ehir) Yuzey Ar~tirmas1. Ara~llrma Sonuc;larz Toplantrsz 28: 303-
328.
- . 2013. N-15 As;ma Kazi <;ah~malan. In Kanhta~ Hoyiik (Es!qehirllnonii) 2013 Kazz <;alz~malarz Rapor, A.
Turkcan, ed., 21-33. T.C. Killtur ve Turizm Bakanhgi Killtur Varhklari ve Muzeler Gene! Mudurlugu.
- . 2014. Eski~ehir, Kanht~ Hoyuk Kazi <;ah~malan. Kazz Sonuc;larz Toplanhsz 36(2): 685-697.
- . 2015. Kanht~ Hoyuk (Esk~ehir/lnonu) 2013-14 Kazi <;ah~malar1. In Kutahya Miizesi 2014 Y1llig1 JI, S. Onan,
ed., 197- 226. Kutahya: TC. Kutahya Valiligi ii Killtur ve Turizm Mudurlugu Kutahya Muzesi Mudurlugu.
- . 2016. Kanhta~ Hoyuk (Esk~ehir/lnonu) 2015 Yih Kazi <;ah~masi Sonus;lar1. In Kiitahya Miizesi 2015 Y1ll1gr 111, S.
Onan, ed., 179-216. Kutahya: TC. Kutahya Valiligi ii Killtur ve Turizm Mudurlugu Kutahya Muzesi Mudurlugu.
- . 2018. At the Crossroads: Changing Chalcolithic Settlement Patterns in Phrygia. In Communities in Transition: The
Circum-Aegean Area During the 5th and 4th Millennia BC., S. Dietz, F. Mavridis, and T. Takaoglu, eds., 556- 566.
London: Oxbow Books.
Turkcan, Ali U., Neyir Kolankaya Bostanci, Onur Ozbek, Halil <;akan, and Salih Kavak. 2017. Kanht~ Hoyuk
(Esk~ehir/lnonu) Kaz1S1 2016 Kazi <;ah~malar1. In Kutahya Miizesi 2016 Y1ll1g1 IV, S. Onan, ed., 233-266.
Kutahya: TC. Kutahya Valiligi ii Killtur ve Turizm Mudurlugu Kutahya Muzesi Mudurlugu.
Onan, Nazan. 2018. Kanht~ Hoyuk 2018 Yih Sezonu ilk Tuns; <;agi <;anak <;omlegi Raporu. In Kanlzta~ Hoyiik
(Eski~ehirllnonii) 2018 Yzh Kazz <;alz~malarz Rapor, A. Turkcan, S. Kurtuldu, and D. Ertemin, eds., 108-113. T.C.
Killtur ve Turizm Bakanhgi Killtur Varhklari ve Muzeler Genel Mudurlugu.
CHAPTER FIVE
KULAKSIZLAR REVISITED:
EXCAVATING THE CONTEXTS OF A CHALCOLITHIC MARBLE WORKSHOP
TURAN TAKAOGLU
The fifth millennium BCE marble workshop site of Kulaks1zlar, identified through surface surveys in the 1990s
near a synonymous village close to the town of Akhisar in central western Anatolia (Fig. 5-1 ), has so far been the only
site that may provide an understanding of how marble figurines and vessels were manufactured in prehistoric times.
The site of Kulaks1zlar lies about 100 km east of the Aegean coast on a sloping ground that flanks the Akhisar Plain
from the east. Past surveys carried out at this very unusual site yielded surface finds composed of debris related to the
various stages of manufacturing products of high craftsmanship such as the so-called Kilia figurines, conical beakers,
and bowls with either straight or slightly curving sides (Din9 1996a, 1996b; Takaoglu 2001, 2002, 2005). The conical
rhyton type, characterized by a rather elongated tapering body with two symmetrically opposed vertical lugs and
horizontal perforation for suspension just below the rim, was one of the trademarks of the late fifth millennium BCE
societies of western Anatolia; these vessels may have been popular among the populations of certain Aegean islands
as well (Takaoglu and Bamyac1 2018). The schematic Kilia figurine type, characterized by large heads which contrast
with a thin flat body, and arms sharply bent at the elbows and pointing upwards, was another hallmark of the fifth
millennium BCE in western Anatolia.
Dagdcrc ®
• Kulaks1zlar
lake J\,Jarmara
c,Jiz (Hennos
0
@Ulucak
®Yq ilova
Kii9iikMenderes (Kayster)
AEGEAN
@ Qukuri<;i H oyi.ik
eander,y_) _ __
Aph ·oclisias @
~ cpccik
Figure 5-1. Map showing the location of Kulaks1zlar

and other major sites with Chalcolithic finds in central-western Anatolia.
Based upon analyses of the surface finds alone, it was argued elsewhere by the present author that marble-working
at Kulaks1zlar was an example of a specialized craft production carried out at the village level by a group of
independent craftsmen who intensively produced the artifacts for regional exchange (Takaoglu 2005). A general
model was also offered for the marble production at Kulaks1zlar based on a careful analysis of the distribution and
46 Chapter Five: Takaoglu
technological aspects of surface finds; this model postulated that pre-urban craft specialization could take a very
complex form, rather than being a simple and casual activity carried out during the spare time left over from basic
agriculture-based subsistence pursuits (Takaoglu 2005). During the analysis of surface finds, Kulaks1zlar marble
working was considered an example of a specialized craft activity because the manufacturing debris displayed four
diagnostic features: segregation of production, volume of output, little variability in the finished form of artifacts, and
manufacturing techniques (standardization). The marble figurines and vessels from Kulaks1zlar can be classified as
precious artifacts, widely exchanged and highly valued. Movements of these symbolically important objects may have
been either direct (inter-personal or inter-societal reciprocal exchanges) or indirect (middleman trade through
wandering pastoralists or itinerant traders). Kilia figurines and conical vessels could have been suitable for gift
exchanges on a personal or communal basis.
Despite the importance of Kulaks1zlar, the site remained unexcavated for nearly three decades until 2018, when
the opportunity arose to conduct a two-year rescue excavation project with the permission of the Directorate of
Antiquities and Museums of Turkey. The aim was to expand and elaborate the general knowledge of this marble
workshop site derived from the analysis of surface finds, and to throw new light on the chronology of the site.
Kulaks1zlar has been among the most unfortunate archaeological sites in Turkey ever since the marble Kilia figurines
came to the notice oflooters and art dealers in the 1990s, if not earlier. It also suffered from deep plowing activities as
much as it suffered from serious plundering. Two seasons of rescue excavations conducted at Kulaks1zlar in 2018 and
2019 by the present author now allow us to address several issues requiring a clearer understanding about the marble
working at Kulaks1zlar. Such issues include the temporal setting of the site within the western Anatolian chronology,
architectural buildings and features that could be related to marble-working activities, and the external relations of the
site. Throwing light on these issues through the brief excavations at Kulaks1zlar in this sense may complement and
elucidate several of the poorly understood aspects of this curious site.
DATING THE SITE
Two seasons of excavations conducted in 2018 and 2019 confirmed that Kulaks1zlar was a single-period flat
settlement, dated to a period between ca. 4500-4250 BCE based on radiocarbon dating (Tubitak-688, Tubitak-1341).
The thickness of the cultural layer overlying the limestone bedrock ranged between 20-50 cm, depending on the
topography of the site that undulates westwards towards the fertile Akhisar Plain (Fig. 5-2). As far as the areas
excavated are concerned, this single cultural layer was directly covered by modem soil, pointing to the total absence
of periods represented other than the Middle Chalcolithic period. This dating is important because the Kilia figurines
have often been erroneously dated to the Early Bronze Age based on both their similarities to the Cycladic figurines
and their status as early relics appearing in contexts of later date. Not a single potsherd that could be attributed to the
Early Bronze Age was noted during two seasons of excavations conducted at Kulaks1zlar.
Figure 5-2. View ofKulaks1zlar showing the area of excavation, from the south.
The Middle Chalcolithic period has a distinctive pottery tradition throughout the western Anatolian littoral, which
also finds close parallels with various sites of this period on the Aegean islands; Kulaks1zlar was one of the few sites
that first helped to define this pottery tradition in this area. The pottery of Kulaks1zlar is quite homogeneous in terms
of fabric, surface treatment, and shape. Although the pottery is monochrome, there is considerable variation in surface
Kulaks1zlar Revisited 47
color, which ranges from reddish-brown to various shades of grayish brown. Certain pots were seemingly finely
smoothed and even burnished (Fig. 5-3). The variation observed in the surface color of these pots must have derived
from the uncontrolled temperature of the firing. The most characteristic pottery type is a bowl with upraised high
handles. Knob-like projections or relief decorations frequently appear on the upraised handles. This type of bowl
comprises a striking 28% of the pottery repertoire alone. Mushroom handles are another common handle type for
bowls used at Kulaks1zlar. In certain cases, crescentic lugs or pellets are attested on either the exterior or interior of
bowls just below the rim. Necked jars having vertical strap handles with knobs are also very common. Some of the
jars have incised decorations dominated by chevrons, pointille, plastic bands with vertical incisions, knobs, and relief
decoration mainly characterized by bands with finger impressions. Four examples of body fragments with a spout
have also been recorded, along with a large container, the latter possibly representative of similar examples to which
these spouts once belonged (Figs. 5-3.16 and 5-4).
2 3 4 5
7 8 9 10 11
12 13 14 15 16
O 5cm
law-----' law-----' law-----'
Figure 5-3. Fragments of bowls (1-8) and jars (9-16) found along with marble objects.
0 30 cm.
Figure 5-4. Large container with a spout towards the bottom.

The Middle Chalcolithic pottery of Kulaks1zlar has identical parallels (see Table 5-1) with phase III (in revised
site chronology) at Ye~ilova in Izmir, which has radiocarbon dates of 4340- 4230 BCE (Derin and Caymaz 2018: figs.
51.6- 51.7; Derin 2020: 2). Similar pottery is also at home at Liman Tepe in the Izmir region, which provided dates
ranging between 4600 and 4200 BCE (Tuncel and Sahoglu 2018: figs. 53.9- 53.11). The recently excavated Troadic
site of Gulpmar has also yielded examples of the same pottery tradition in phase III, which also has similar
radiocarbon dates (Takaoglu and Ozdemir 2018). Phase IV at <;ine-Tepecik also belongs to this horizon, since, in
addition to the pottery of Middle Chalcolithic character, this important site, which is situated on a strategic location
near one of the tributaries of the River Meander, yielded a series of examples representing marble Kilia figurines and
conical rhyta (Gunel 2018: figs. 55.3- 55.10). Likewise, Malkayas1 Cave in the Latrnos Mountains also produced
pottery of Middle Chalcolithic character, along with fragments of marble Kilia figurines (Peschlow-Bindokat and
Gerber 2012: figs. 40- 45). Moreover, similarities also exist between the pottery ofKulaks1zlar and several sites on the
eastern Aegean islands such as Emporio on Chios and Tigani on Samos, Kalythies on Rhodes, and Vathy Cave on
Kalymnos. Thus, it is safe to place the occupation of Kulaks1zlar temporally in the ca. 4500- 4250 BCE range, falling
roughly in the second sub-period of the Middle Chalcolithic (Table 5-2).
t/J
t:
"Cl O" Tavaba$I
~ -\:, Be$ik Liman
o· Cal. Gillpmar Kumtepe Ugurlu Ulucak Ye$ilova Yass,hiiyi.ik Kulaks,zlar Lower
~ Sivritepe Tepe
0..
o· BC A Cave
0..
5600
I 5500 LIi
t"rl
~
'< 5400
n
::,- 5300
"'[ 5200 III
g.: 2 5100 II IV
;;·
5000
4900
4800
3::::
0.: I 4700
0..
ii"' 4600
n
::,- 4500 lll ll
"'
"S'-
~ 2
4400
4300 I]
VII
;;·
4200
Table 5-1. A tentative chronological chart of the western Anatolian littoral

during the Early and Middle Chalcolithic periods based on C 14 dates from the excavated sites.
Laboratory Reference Nature of Sample Location Date (BP) Uncalib. Calibrated Date at 2a
Tlibitak-1341 Animal Bone Structure C 5609 ± 31 BP 4498 - 4360 BCE
Tlibitak-688 Animal Bone Structure A 5448 ±31 BP 4351- 4249 BCE
Table 5-2. Two radiocarbon dates from Kulaks1zlar.
As a result, marble working at Kulaks1zlar could be viewed as a phenomenon of the Middle Chalcolithic period, a
significant stage in western Anatolian cultural history that witnessed significant changes and transformations in most
aspects of socioeconomic and cultural life, including an increase in population and settlement numbers, appearance of
new settlements in places other than alluvial plains, developments in marble and copper-working technologies, the
appearance of new pottery forms and styles, and a rise in the degree of cultural interactions and long-distance
exchange.
LOCI OF MANUFACTURE
The homogeneity of the surface pottery assemblage and manufacturing debris, resulting from the various stages of
marble figurines and vessels recorded throughout the surface of the site, was previously used to hypothesize that
Kulaks1zlar was a short-lived flat settlement and that production took place simultaneously in several different loci
during the time of the site's occupation. This theory needed to be tested with data derived from excavations. As far as
the areas excavated are concerned, at least three sectors of architectural features, separated from each other by empty
spaces, have been noted at Kulaks1zlar. Among them, Sector 1 was excavated in 2018 and 2019, while Sector 2 has
been kept in reserve for future archaeological studies. The field of Sector 3 appears to have been severely damaged by
the plowing activities of farmers during the last decades.
A 2m
Kulaks1zlar - Sector I
Figure 5-5. Partial plan of Sector 1 where workshops are clustered.
In Sector 1, marble working seems to have been carried out in isolated small U-shaped enclosures, though other
forms of architectural features such as raised stone platforms were also used in open spaces around them (Fig. 5-5). It
is unlikely that these enclosures were used as living units since one end of the building was open, and the majority of
finds from their interiors and immediate surroundings consisted mainly of manufacturing debris related to the
production of figurines or vessels. It can easily be deduced from the analysis of pottery and marble working-related
finds that these buildings were contemporary and do not represent different time periods of occupation.
The most notable architectural feature of Sector 1 is probably a small U-shaped enclosure lying on an east-west
axis, named Structure A (Fig. 5-6). Only a single row of foundations of this 3.5 m long and 2.8 m wide building has
survived due mainly to the intensive field-plowing and tobacco cultivation of the land until two decades ago. The
Figure 5-6. View of Structure A from the west.

stone foundation of this building was laid directly upon the limestone bedrock, which is about 40-50 cm below the
surface level. There are some indications of a superstructure, including traces of postholes identified in several
locations adjacent to the outer faces of the foundation wall in the form of slight indentations for the positioning of
wooden posts. A scatter of lumps of clay that appear to represent the wattle-and-daub wall-building technique was
identified in the space north of Structure A and west of Structure B. Some of these lumps of clay preserve impressions
of twigs and split planks. These lumps do not show a pattern, suggesting that they resulted from a collapse.
Two small platforms created by a pavement of small flat stones rest in front of and inside the apsidal end of the
building. Nearly one hundred marble fragments representing the various stages of conical rhyton manufacture were
found both around these stone-paved features and in areas just outside the apsidal end of the building, some of which
are illustrated in Fig. 5-7. Most of these fragments belong to pieces broken during the hollowing out of their interiors.
O 5cm
"-e-e,, lw----wl lwwl
Figure 5-7. A preform ofa conical rhyta (1) pointed bases (2-5), and rim fragments (6-9), broken during the
hollowing out of the interiors found in Structure A, Grid J 18, q4.
Nearly two dozen sandstone drill-bits of varying diameters have also been found along with these marble waste by-
products related to conical rhyton manufacture (Fig. 5-8). Over a dozen obsidian artifacts from Melian and central
Anatolian Golludag sources were found in a cache adjacent to the northern wall near the stone-paved platform inside
the enclosure. The finds recovered both from the inside and just outside the apsidal end of the building seem to
indicate that this specific area saw activity focused on the production of marble conical rhyta.
Part of another enclosure (Structure B) was identified to the east of Structure A with a single row of stones. Again,
due to extensive plowing in the past, most parts of this building have disappeared from the archaeological record. The
scatter of lumps of clay belonging to a superstructure employing the wattle-and-daub wall-building technique could
be related to Structure B.
3 4
5 6 7 8 9
0 5 cm
Figure 5-8. Sandstone drill bits of various diameters

used to hollow out the interior of conical rhyta from Structure A.
Part of another enclosure (Structure B) was identified to the east of Structure A with a single row of stones. Again,
due to extensive plowing in the past, most parts of this building have disappeared from the archaeological record. The
scatter of lumps of clay belonging to a superstructure employing the wattle-and-daub wall-building technique could
be related to Structure B.
The foundation of a third enclosure exists nearly 5 m west of Structure A (Fig. 5-5). This building, named
Structure C accordingly, yielded numerous unfinished fragments of Kilia figurines, broken mainly during the
formation of their delicate necks, indicating that the production of Kilia figurines took place in another building near
those focused on the manufacture of conical rhyta (Fig. 5-5). The remaining stones and the footprint of the foundation
indicate that this structure was also U-shaped. It seems that an area of the interior adjacent to the apsidal end was
paved with flat stone slabs, most likely forming a place for marble working. A series of fragmentary preforms of Kilia
3 4
5 6
7 8 9 10
5cm
Figure 5-9. Fragments of nearly finished Kilia figurines broken off at the neck (1 - 6), or at the waist (7- 10),
from Structure C and its surroundings.
1 2 3 4
0 3cm
Figure 5-10. Fragments of nearly finished miniature Kilia figurines.

figurines, as well as nearly finished figurines broken during the refinement stage of manufacture, were found here
(Figs. 5-9 and 5-10). The lack of tools other than several abrasives made of sandstone, emery, and pumice in this area
could indicate that this small building was where the last stages of Kilia figurine manufacture took place.
A kind of platform lined with small stones also exists at the western end of Structure C. A flat grinding stone slab
was found on the southwest comer of this platform. Several sandstone drill-bits, as well as over two dozen marble
rims, bodies, and base fragments of conical rhyta broken during manufacture, were found in front of this platform. A
preform of a complete conical rhyton was also recorded above this platform, along with additional broken fragments.
A nearly complete large two-handled container with a spout towards the base was found in a collapsed position in
front of this raised platform. Similar large jars have often tentatively been linked to chums used in dairy products,
such as the conversion of fatty milk into butter. The spout is thought to have served as an outlet to remove
overpressure, while the mouth of the jar is closed by a piece of leather. Although residue analysis conducted on this
large container at the archaeometry laboratory of <;anakkale Onsekiz Mart University unfortunately did not provide
any evidence regarding its use in milk-related production, the possibility remains that it may be associated with tasks
linked to marble-working. The location of this large container, capable of holding 50 liters of water, near the stone
platform around which numerous waste by-products related to conical rhyton manufacture were found, leads one to
consider that it may have been utilized in certain production activities related to marble working (e.g., to hold water
which aided in the operation of the sandstone drill-bits during the hollowing out of the vessel interiors). A copper pin
was also found near this jar, which will be mentioned below.
A rich assemblage of stone tools encountered in the excavated areas alongside the waste by-products of marble
working indicates that the craftsmen of Kulaks1zlar developed complex procurement strategies to obtain raw materials
for the tasks for which they were intended. Sandstone was brought to the site to produce drill-bits; emery and pumice
for abrasives; basalt and gabbro for axes and hammers; obsidian, flint, jasper, and radiolarite for chipped stone tools;
and andesite and schist for grinding stones. This diversity in the raw materials used was something that made the craft
production here remarkable. The coexistence of different types of stone within the circumscribed area in which
Kulaks1zlar was located played a major role in the development of craft production at the site. Nevertheless, the raw
materials were gathered not only from those found naturally in the immediate environment of the site but also from
distant locations as far as the Cycladic islands to the west and central Anatolia to the east.
EXTERNAL RELATIONS OF THE SITE
The products of the Kulaks1zlar marble workshop were distributed both on intra-regional and inter-regional levels.
Several sites in central western Anatolia have already yielded marble objects that can be traced to the Kulaks1zlar
workshop. The marble finds of Kulaks1zlar origin found at the site of Dagdere could be considered an archaeological
manifestation of intra-regional exchange between the settlers of Kulaks1zlar and other settlements located around it
during this period (Takaoglu 2017). Such discoveries as Dagdere clearly show that not only sites on the alluvial
Akhisar plain, but also other sites located at high elevations on the mountains encircling the plain, demanded and
valued the marble products of Kulaks1zlar. The products of Kulaks1zlar were also part of an inter-regional network
including an exchange system that covered most parts of western Anatolia from Be~ik-Sivritepe in the western coast
of the <;anakkale region in the northwest to the Karain Cave in the Antalya region to the south during the Middle
Chalcolithic period.
1 2 3 4 5 6 7 8 9 10
0 5cm
Figure 5-11. Selected examples of obsidian artifacts from Melos (1-8),

and the central Anatolian source of Golludag (9-10).
It would appear from the recovery of excavated exotic artifacts that Kulaks1zlar was also part of a larger inter-
regional network of communication during this period. For instance, among the excavated 360 chipped stone artifacts,
62 are of obsidian; 52 pieces originated in Melos, and for 10 pieces the source was Golludag in central Anatolia. Two
of the Melian obsidian pieces are represented by cores, while the remaining ones are mainly blades and flakes (Fig. 5-
11 ). Several routes probably existed simultaneously for the spread of Melian obsidian. One of them followed a land-
based route along the Buyuk Menderes (Meander) River. The high amount of Melian obsidian recorded at Miletus I
indicates that the site played a key role in contacts between the Cyclades and inner western Anatolia as early as the
Late Chalcolithic period. Due to its important location at the mouth of the Meander, Miletus may have also been
responsible for the distribution of Melian obsidian arriving by sea to inland sites located along the Meander River
valley, indicated by their large percentages at (;ine-Tepecik, Beycesultan, and Aphrodisias (Niemeier 2000: 125-27;
Kouka 2014: 53, 2019: 239). The site of (;ukuri9i, located in a tributary valley of the Ku9uk Menderes (Kayster)
River, also played a major role comparable to that of Miletus in the transmission of Melian obsidian by sea (Schwall
et al. 2020: 15). The Melian obsidian likely reached Kulaks1zlar through another land-based route following the Gediz
(Hermos) River basin, as finds from sites such as Liman Tepe and Bakla Tepe indicate. The recovery of marble
conical rhyta fragments at Liman Tepe, and the head of a marble Kilia figurine at Ulucak, verifies that this route was
in operation as early as the late fifth millennium BCE (Tuncel and Sahoglu 2018: 523; b. (;evik, Pers. Comm.).
Other than the obsidian evidence, several ground stone artifacts excavated at Kulaks1zlar may also have import at
the site. One of them is a miniature jadeite axe (Fig. 5-12). The beveled edge of this finely polished miniature axe
shows some scars from use. This piece presumably originated in the Cycladic island of Syros, one of the best-known
sources in the Aegean. Jadeite axes of possibly Syros origin were reported at the Neolithic site of (;ukuri9i near
Ephesus in the Izmir region (Schwall et al. 2020). A network of communication in Melian obsidian was already
operating as early as the Neolithic period in western Anatolia, and the discovery of jadeite axes of Syros origin at
(;ukuri9i now demonstrates that this network may have also included artifacts made of jadeite. This raw material from
Syros was evidently in continuous demand from the Neolithic to the Early Bronze Age (S0rensen et al. 2017). The
recovery of both obsidian from Melos and jadeite from Syros at Kulaks1zlar could be evidence of the continuation of
such a prevailing network of trade in the late 5th millennium BCE.
0 3cm
: : 0
2
5cm
Figure 5-12. Miniature jadeite axe. Figure 5-13. Fragment of two broken mace heads of
black and beige porphyry (1-2).
Two examples of a mace head of black and beige porphyry excavated at Kulaks1zlar also deserve mention here
(Fig. 5-13). This type of stone mace head, with a perforation in the middle, would have originally been affixed to a
wooden shaft. The closest parallel for these two stone mace heads from Kulaks1zlar was surprisingly reported at
Sardis during the excavations of the Roman era Wadi B Temple, although it was viewed as an object from earlier
times mixed in deposits of the Roman stratum (Greenewalt 2007: 746, fig. 12). In terms of raw material, the close
similarity between the two mace heads from Kulaks1zlar and the one from Sardis is intriguing. Although typologically
comparable mace heads are particularly common at Early Bronze Age Anatolian sites as symbolic items signifying
authority, there is no reason to argue that the Kulaks1zlar and Sardis examples could not be early in date. Stone mace
heads made their first appearance in Anatolia as early as the earliest phases of the Neolithic period, albeit in small
numbers. The Kulaks1zlar examples of Middle Chalcolithic date could be imports acquired through an exchange,
perhaps in return for marble objects locally produced at the site.
Other products may have arrived at Kulaks1zlar through trade. One copper slag and pin excavated at Kulaks1zlar
indicate that copper was also recorded at the site (Fig. 5-14). Although archaeological evidence for copper is rare in
western Anatolia during the Middle Chalcolithic, its presence is known to us from several settlements such as Ugurlu
III on Gok9eada (Imbros), Kumtepe A, Gulpmar III, and Liman Tepe VII (B. Erdogu, Pers. Comm.; Gabriel 2006:
356; Glider et al. 2021: fig. 2; Tuncel and Sahoglu 2018: 524-25). The four copper tools from the Middle Chalcolithic
phase III settlement at Gulpmar are known to be arsenical copper containing 1% arsenic. The copper was obtained by
heating and smelting the sulfur-bearing polymetallic ore, which is a more complex metallurgical process than the
collecting and smelting of oxide-containing ores. During roasting and smelting, much of the arsenic was lost, and only
1% of it remained in the metal. The lack of evidence of either slag or crucibles for smelting at Gulpmar suggests that
the metal of the tools might have been brought in as rod forms with rectangular cross-sections to the site, where they
might have been shaped by applying heating-forging-annealing cycles to obtain tools with the necessary
mechanical properties. The recovery of slag at

Kulaks1zlar may imply that either actual smelting
took place at the site or that the slag was introduced
to the site as a resource so that it could be subjected
to a secondary process.
The two copper examples from Kulaks1zlar have
not yet been subjected to archaeometric analyses to
determine if any relationship exists with the copper
artifacts found at other Middle Chalcolithic sites
such as Giilpmar III, Kumtepe A, Ugurlu II, and
Liman Tepe VII. Despite the rarity of copper
artifacts in western Anatolia, the available data may
be accepted as evidence for on-site copper-working
during this period. Because it requires a specific skill
and knowledge, copper-working may have been a
specialized activity that could be undertaken by
1 2 itinerant craftsmen. The relative complexity involved
0 5cm
in producing copper objects, combined with the lack
of local copper sources in every part of western
Anatolia, implies that there was some mechanism by
which the technology (specialists and resources)
Figure 5-14. Copper pin (1) and a copper slag (2). arrived at the settlements.
CONCLUDING REMARKS
One of the most significant contributions of excavations conducted at Kulaks1zlar is that now it is possible to date
the site and its remains to ca. 4500- 4250 BCE, which in turn allows us to place the marble figurines and vessels
found at sites other than Kulaks1zlar into this sequence. One needs to consider that there may have been other marble
workshops producing conical rhyta in this period. There are indications that another workshop specializing in the
production of conical rhyta might have existed somewhere in northwestern Anatolia (Takaoglu and Bamyac1 2018).
The discovery of different spectra of imported artifacts made from exotic raw materials such as obsidian from
Melos and the central Anatolian source of Golltidag, as well as jadeite, porphyry, and copper, confirm that Kulaks1zlar
was part of a well-developed intra- and inter-regional network of exchange during the last stages of the Middle
Chalcolithic period. These exotic objects may have been part of the trade network that facilitated the movement of
precious marble figurines and vessels of the Kulaks1zlar workshop in western Anatolia and beyond. It would be
wonderful to consider these exotic items as exchange objects obtained in return for marble figurines and vessels if
there were also additional material culture datasets attesting to wider external relations. If the marble objects were
manufactured for a non-economic purpose at Kulaks1zlar, then one may expect that they were transported far from
their place of production through an inter-societal or inter-individual exchange to regulate social relations, although
this assumption needs archaeological demonstration.
It is also possible to picture where the marble working activity was taking place from an architectural point of
view. A series of small U-shaped enclosures and their related features such as stone-paved platforms and raised stone
platforms, discovered during excavations, appear to be related entirely to marble working. As far as the areas
excavated are concerned, the architecture at Kulaks1zlar is virtually devoid of finds which would attest to domestic
life, such as ovens, hearths, bins, and storage and cooking vessels. Noteworthy exceptions are the animal bones.
Nearly 380 animal bones recorded during the excavations may be closely related to diet. Among the fauna]
assemblage composed of cattle, sheep and goats, red deer, and fallow deer, cattle (72%) was the most common animal
at the site. Several trial trenches opened in close proximity to production areas sought evidence regarding dwellings of
craftsmen and their families as well as aspects of domestic life. Most of these trial trenches, unfortunately, yielded
only scatters of potsherds, manufacturing errors of marble figurines and vessels, and the tools used to produce them.
This may indicate that dwellings were most likely situated in one of the unexcavated areas in the vicinity of the
production areas.
The site of Kulaks1zlar can be viewed as an important site demonstrating what categories of archaeological
evidence to expect when encountering a production site focused on the production of figurines and vessels in the
archaeological record. Kulakstzlar in this context may constitute a reference in certain respects for those future
archaeologists who will discover and excavate those undiscovered marble workshops located in the Cyclades.
ACKNOWLEDGMENTS
I would like to thank the Directorate of Antiquities and Monuments of Turkey and the Institute for Aegean
Prehistory (INSTAP) for their support for the conducting of two seasons of excavations at Kulaks1zlar. I would also
like to acknowledge the kind support of Harun Gi.illi.i, Li.itfi Bkici, Bren Si.ilek, and Alper Dogan from the Manisa
Archaeology Museum for facilitating these two seasons of archaeological excavations at Kulaks1zlar. Onur Bamyac1,
Abdulkadir Ozdemir, Serhan Mutlu, Mera! Mutlu, Omer Can Yildmm, Murat Ay, and Mustafa Kase also deserve
special thanks for their valuable contributions during the fieldwork. The project was also supported by the Scientific
Research Projects division of the <;anakkale Onsekiz Mart University (BAP, Project no: SBA-2020-3372).
REFERENCES CITED
Derin, Zafer. 2020. izmir-Ye~ilova Hoyi.igi.i. jzmir Ara~tzrmalarz Dergisi 12: 1-10.
Derin, Zafer and Tayfun Caymaz. 2018. The Chalcolithic Period at Ye~ilova Hoyi.ik. In Communities in Transition:
The Circum-Aegean Later Neolithic Stage (ca. 5000/4800-3200/3000 BC), S. Dietz, F. Mavridis, Z. Tankosic, and
T. Takaoglu, eds., 499- 505. Oxford: Oxbow.
Dins;, Rafet. 1996a. 1994 Y1h Akhisar-Kulaks1zlar Menner Atolyesi Yi.izey Ar~trrmas1. XIII. Ara~hrma Sonur;:larz
Toplantzsz 1: 11-41.
- . 1996b. Kulaks1zlar Menner idol atolyesi ve s;evre ara~t1nnalar1. XIV Ara~tzrma Sonur;larz Toplanhsz 2: 255-282.
Gabriel, Utta. 2006. Bin Blick zuri.ick: Das fiinfte Jahrtausend vor Christus in der Troas. In Troia: Archaologie eines
Siedlungshilgels und seiner Landschaft, M. Korfmann, ed., 355-360. Mainz: Zabem.
Greenewalt, Crawford H. 2007. Sardis: Archaeological Research and Conservation Projects in 2005. Kazz Sonur;:larz
Gi.ider, Omit, Turan Takaoglu, and Abdulkadir Ozdemir. 2021. Middle Chalcolithic Copper Tools from Gi.ilpmar in
Northwestern Anatolia: An Archaeometric Approach. Archaologisches Korrespondenzblatt 51: 155- 170.
Gi.inel, Sevins;. 2018. Prehistoric Culture in <;;ine-Tepecik Hoyi.ik and its Contribution to the Archaeology of the
Region. In Communities in Transition: The Circum-Aegean Later Neolithic Stage (ca. 5000/4800-3200/3000 BC) ,
S. Dietz, F. Mavridis, Z. Tankosic, and T. Takaoglu, eds., 541-550. Oxford: Oxbow.
Kouka, Ourania. 2014. Past Stories- Modem Narratives: Cultural Dialogues between Bast Aegean Islands and the
West Anatolian Mainland in the 4th Millennium BC. In Western Anatolia before Troy: Proto-Urbanisation in the
4th Millennium BC?, B. Horejs and M. Mehofer, eds., 43-64. Wien: Austrian Academy of Sciences Press.
- . 2019. Local and Imported in Action: Western Anatolian and Cycladic Figurines at Early Bronze Age Miletus. In
Beyond the Cyclades: Early Cycladic Sculpture in Context from Mainland Greece, the North and East Aegean, M.
Marthari, C. Renfrew, and M.J. Boyd, eds., 237-49. Oxford: Oxbow.
Niemeier, W.-D. 2000. Milet: Knotenpunkt im bronzezeit-lichen Metallhandel zwischen Anatolien und der Agais?
In.Anatolian Metal, Vol. 1, 0. Yals;m, ed., 125- 36. Der Anschnitt Suppl. 13. Bochum: Deutsches Bergbau
Museum.
Peschlow-Bindokat, Anneliese and Christoph Gerber. 2012. The Latmos-Be~pannak Mountains. In The Neolithic in
Turkey, Volume 5: New Excavations & New Research, M. Ozdogan, N. Ba~gelen, and P. Kuniholm, eds., 67- 115.
Schwall, Christoph, Michael Brandl, Tatjana M. Gluhak, Bogdana Milis;, Lisa Betina, Lasse Sorensen, Danilo Wolf,
and Barbara Horejs. 2020. From Near and Far: Stone Procurement and Exchange at <;ukuris;i Hoyi.ik in Western
Anatolia. Journal ofLithic Studies 7: 1-25.
Sorensen, Lasse, Pierre Petrequin, Anne-Marie Petrequin, Michel Errera, Barbara Horejs, and Frederic Herbaut. 2017.
Les limites sud-orientales des jades alpins (Grece et Turquie). In JADE. Objets-signes et interpretations sociales
des jades a/pins dans /'Europe Neolithique, P. Petrequin, B. Gauthier, and A-M. Petrequin, eds., 491-520.
Besans;on: Presses universitaires de Franche-Comte et du Centre de Recherche Archeologique de la Vallee de
l'Ain.
Takaoglu, Turan. 2001. 1999 Y1h Kulaks1zlar Menner Atolyesi Ar~t1rmas1. 18. Ar~trma Sonur;:larz Toplantzs1 II:
157- 168.
- . 2002. Chalcolithic Marble Working at Kulaks1zlar in Central-Western Anatolia. Turkish Academy of Sciences
Journal ofArchaeology 5: 71-94.
- . 2005. A Cha/eolithic Marble Workshop at Kulaks1zlar in Western Anatolia. An Analysis of Production and Craft
Specialization. BAR International Series 1358. Oxford: Archaeopress.
- . 2017. Middle Chalcolithic Finds from Dagdere in the Akhisar/Manisa Region. Anadolu/Anatolia 43: 1-14.
Takaoglu, Turan and Abdulkadir Ozdemir. 2018. The Middle Chalcolithic Period in the Troad: A New Look from
Gi.ilpmar. In Communities in Transition: The Circum-Aegean Area During the 5th and 4th Millennia BC, S. Dietz,
F. Mavridis, Z. Taknosic, and T. Takaoglu, eds., 479-490. Oxford: Oxbow.
Takaoglu, Turan and A. Onur Bamyac1. 2018. On the Marble Pointed Rhyta: New Evidence from Ye$iltepe irl North-
Western Anatolian Hinterland. In Communities in Transition: The Circum-Aegean Later Neolithic Stage (ca.
5000/4800-3200/3000 BC), S. Dietz, F. Mavridis, Z. Tankosic, and T. Takaoglu, eds., 493-500. Oxford: Oxbow.
Tuncel, Riza and Vas1f Sahoglu. 2018. The Chalcolithic of Coastal Western Anatolia: A View from Liman Tepe,
izmir. In Communities in Transition. The Circum-Aegean Later Neolithic Stage (ca. 5000/4800--3200/3000 BC),
S. Dietz, F. Mavridis, Z. Tankosic, and T. Takaoglu, eds., 513- 529. Oxford: Oxbow.
CHAPTER SIX
EXCAVATIONS AT USAKLI HOYUK: RECENT RESULTS
ANACLETO D 'AGOSTINO, STEFANIA MAZZONI, AND VALENTINA ORSI
A SITE IN THE UPLANDS OF CENTRAL ANATOLIA 1
U~akh Hoyuk is located at the base of a narrow valley oriented along an east-west axis, bordered by hills, in the
heart of the Anatolian plateau; this region saw the rise of the first urban settlements during the 2nd millennium BCE
and rests at the center of the territory defined in the Hittite texts as the "Upper Land" (Fig. 6-1 ). The development of
the ancient settlement owes much to the possibility of exploiting the potential water supply of two streams, the Kotii
Dere and the Egri Oz Dere, flowing a few dozen metres northwest and northeast of the site, as well as the nearby
springs. Also critical was its position along a communication axis that would play a central role over the centuries.
The reciprocal visibility between the site and the peak of Kerkenes Dag1, which must have always been an important
reference point in movements between the plateau and Cappadocia, may have helped to define U~akh's function and
importance.
Figure 6-1. Location ofU~akh Hoyuk (prepared by Y. Ozarslan using an ArcGIS Online basemap).
During the most recent campaigns the archaeological team has been working to reconstruct the shape of the
settlement at the time of the Hittite kingdom, and to understand the sequence of occupation and development after the
changes that occurred between the Late Bronze Age and the Iron Age (Mazzoni et al. 2019). The foci of recent
seasons of work (2017-2020) have been the southern portion of the lower city and the southern slope of the high
1
This section was contributed by A. D' Agostino, as were the following two sections describing Area D and Area A.
Excavations at U~akh Hoyuk 57
mound, where two large buildings of Hittite date have been found and which must rest under remains of later periods.
The results achieved so far suggest that the Hittite settlement was founded on a site that already had a long history of
occupation, spreading out between the highest point, the mound (2 ha), to the large ovoid terrace (10 ha). The Iron
Age settlement is found on the high mound. In the transition from the Late Bronze to the Iron Age, the duality
between the citadel and the lower city (terrace) as areas featuring public buildings seems to lose importance in favour
of a new Iron Age pattern centred on the citadel, which was probably fortified.
AREA D: THE HITTITE BUILDINGS AND THE IRON AGE SEQUENCE
Building III
The excavations in Area D, on the southern slope of the mound, allowed us to expose a sequence of layers
spanning from the Late Bronze to the Late Iron Age, containing many building phases and contextual materials.
Within the trench, the exposed layers belong to three main levels: the large building at the bottom, the remains on top
of it, and a large stone structure. The latter is probably a wall surrounding the hill; at a certain point late in its
existence, having partially collapsed, it fulfilled the function of a glacis. Further traces of a later level of occupation
have been identified on top of the glacis, but they have been documented only minimally. The lower level dates to the
Late Bronze Age, those above to the Iron Age.
The southern side of the hoyuk is occupied by Building III, the perimeter wing of which is about 50 m in length
and which has been exposed along the contours of the slope. Some portions of the floors, the foundation walls, and the
large stone wall that supports the terrace on which the building is erected have been preserved (Fig. 6-2). The external
retaining wall (USS) and the fill between and below the foundations appear to be based on a natural hill of yellow
unconsolidated sediments at an altitude of 1134.40 m asl. Erosion along the slope has obliterated the external part of
the building, and only the lower course of the perimeter wall delimiting the groundwork is preserved, together with the
foundations and small portions of floors higher up on the slope.
+ + + + + + + + +
Figure 6-2. Area D: Building III, preliminary plan (with some structures discussed in the chapter marked).
The plan shows walls and structures belonging to different phases (2019).
The building was adapted to the irregular profile of the pre-existing natural hill by filling in height differences and
terracing with the use of soil, both natural and from earlier phases of occupation. The floors of the building close to
the north section of the trench are about 10 m higher than the base of the perimeter wall, which is made of large
boulders located downslope and projects outward atop an outcrop of virgin soil. In fact, as the external walking
surface south of the perimeter wall is many metres below that of the inner floors and walls, we have to imagine that
when the building was still standing, a huge wall (constructed of granite boulders in the lower part, and of mudbrick,
probably plastered, in the upper part) rose before those who approached the site from the south, from the base of the
mound, producing a remarkable visual effect. 2
The building has a western arm, projecting southwards, with a large space of about 12.30 m on the east-west axis
delimited by two walls, and a central and eastern part divided into rooms of various size and shape, only partially
excavated. In general, the elevation is preserved for a few centimetres, but erosion and later events have damaged
almost all the structures.
2
The difference in altitude between the ground level south of the mound and the base of the wall USS is about 26 metres.
58 Chapter Six: D' Agostino et al.
The 2017-2019 excavations of the western and eastern sectors of Building III made it possible to discover new
portions in addition to the structure's central body investigated in the 2015 and 2016 campaigns; these new areas
consist of rooms arranged in two rows and oriented west-northwest to east-southeast (Mazzoni et al. 2019, 2018). In
the western half of the structure, the remains of the building are largely eroded and only partially preserved where the
large Iron Age stone wall overlapped them, while the southern perimeter wall-the continuation of wall USS made of
granite boulders-has been lost here.
We have found traces of a foundation wall (US354 and US353) along the slope, and a red soil layer resulting from
burnt mudbricks (US348) largely washed away; running parallel to this is a portion of a room with a plastered floor.
The cracked, whitish plastered floor (US233) contains large black patches, the consequence of a fire, traces of which
have been found throughout the building; the floor is partially covered by whitish or black ashes.
At the approximate centre of the excavation area, several foundation walls were fom1d (US362 and US312),
oriented north-south; also discovered was the northern side of an east-west oriented wall (US332) lying on top of the
perimeter wall USS, which here presents a well-preserved external fayade of large granite boulders. The floors
associated with these walls are not preserved, and pits from later periods obliterate the flat area near the edge of the
excavation trench. The foundation walls of Building III have external faces made of a row of medium and small-sized
stones, with a core consisting of stones, rubble, and earth.
As for the eastern squares, two additional rooms were identified in the course of the 2017 season, as well as part of
the wall marking the south side, which was all but destroyed by erosion along the sloping surface. The floor of the
southern room US434 was not preserved, while portions of the foundation wall separating this from the northern room
US433 are still visible. However, we do not know how these were connected with each other via a doorway. The
mudbrick wall US182, separating this room from room US473, is in a very bad state of preservation and has a floury
texture. Two pits, of which only the lower levels are preserved, have obliterated the eastern and southern parts of the
room. Clear traces of intense fire can be noted on all of the walls and floor, characterised by the baking of mudbricks
and melting of the clay surface on the beaten earth floor, with bubbles and vitrification at different points on the
pavement (D'Agostino 2020). This indicates the very high temperature reached during the fire, made more fierce by
the collapse of the ceiling and the resulting "kiln effect" that produced the above-mentioned consequences on the
remains of the building. The effects of the fire on the building material are differentiated. The mudbricks of the wall
were red with a brown to black core, while the plaster covering them had a homogeneous white colour with some
smoke spots, but unlike the floor, the wall plaster was not melted. Some parts of the floor, where the plaster was still
preserved, appeared shattered, blackened, or with greenish nuances; other areas lacking plaster were of a buff colour,
and yet other portions had lost the superficial vitrified slip and showed only the lower part of the plaster layer with a
reddish colour. These effects depended upon the particular conditions of the fire and other unknown factors, rather
than the results of a different composition of coating materials; these are apparently the same, based on the appearance
of the white plaster preserved on the walls and floors of other rooms such as US472, US432, and US456, respectively,
in the central and western body of the building. Along the eastern limit of room 433, the mudbrick wall (US415) is
partially preserved, and its plaster has collapsed, almost slipped, onto the floor.
In a posthole at the centre of the room (US424), the well-preserved lower part of a burned post (US425), with a
diameter of 50 cm, probably supported the roof. According to the recent results of a dendrochronological study, the
post from room 433 is a fragment of cedar; radiocarbon tests and wiggle-matching analysis provided a date between
the end of the 15th and beginning of the 14th centuries BCE, within a range of few decades (Gminska-Nowak et al.
2021).
The red (US548) and ashy soil of the fill is not very coherent, and the remnant of collapsed walls consists of gritty
and powdery soil (US587) that bears traces of combustion, along with the presence of mudbricks, either fragmentary
or completely destroyed and burnt, as well as ash, coal, and slag. This fill was disturbed by the tmmelling of a large
and branched fox's den, which removed a substantial part of the archaeological deposit.
In this eastern sector, wall USS appears without granite boulders, which were probably reused in ancient times or,
in some cases, had slipped down to the base of the hoyillc, as is the case of the stone found in operation 2 (D 'Agostino
and Orsi 2015: 41, pl. 11 ). In this area, wall USS consists of medium and small stones at the base. Its thickness of
about 1.2 m also appears to be residual compared to the thickness of 1.7 m which is found to the west, where the wall
is best preserved.
Another interesting piece of information from the 2017 excavations relates to the perimeter wall built of large
granitic boulders and preserved in the central body of the building. Its alignment is in fact not linear; the wall recedes
progressively towards the north, proceeding from west to east, probably to adapt to the curve of the natural hill. The
plan shows a segmented profile of the front, with the western sector protruding southwards and the central and eastern
sectors progressively receding northward (each sector recedes a few dozen centimetres compared to the previous one)
(Fig. 6-2). This irregularity of the profile is quite traditional in Hittite public architecture. The western segment,
delimiting the large area US456, is lost, and only the orthogonal wall (US359), connecting the lost part of wall USS,
indicates its projection. Here, the ground filling the foundation of Building III was revealed; the team sectioned
approximately 20 m of the foundation's front in order to confirm that wall USS did not continue.
The eastern and western limits of Building III have not yet been reached; this is due to the curve of the hill where
the remains appear to be more eroded, as demonstrated by the recent results of excavations. The plot map generated by
the 2017 geoelectric prospection on the southeastern slope of the hoyillc shows some rectilinear anomalies that can be
interpreted as walls delimiting two rooms, aligned with the orientation of the other walls of Building III, and
presumably developing eastward beyond the eastern limit of the trench.
No materials from the original phase remain in place, which probably indicates an abandonment of the building.
Once abandoned, Building III suffered a violent fire that burned the wooden structure of the building, baking and
crumbling the mudbricks in the wall. The vitreous layer of some floors (US27 in room 59, and in part USl 76 in room
473 and 420 in room 433) certainly resulted from the partial melting of the surface of the clay, which suggests very
high temperatures (Fig. 6-3). After this event the building fell into disrepair.
Figure 6-3. Area D; the central portion of Building III (2016). From the bottom, rooms 472,473, and 59.
Above the burnt remains of Building III, layers and wall/glacis dating to the Iron Age. Taken from east-southeast.
The Iron Age Layers
A long sequence of layers overlaps the burned remains of Building III, which include floors, pits, domestic
installations, and small walls. These demarcate a different occupation dating to the Middle Iron Age and are
characterised by domestic functions and possibly small productive installations connected with daily activities, at least
in this part of the site. Other pits and small specific deposits date to the Early Iron Age (Orsi 2020: 282-283), mainly
documented at the site by the presence of sherds in secondary contexts. From the evidence collected so far, we can
suggest that at the tum of the 1st millennium BCE a clear change in the shape of the settlement occurred, as the
dwellings were concentrated on top of the citadel, which, at some point, was surrounded by a wall and a stone glacis
that probably presented the impression of a fortified mound.
Some of the exterior areas of Building III are directly covered by the rough stones comprising the wall/glacis that
encompassed the slope of the citadel, and date to somewhere between the end of the Middle and the Late Iron Age; in
the more internal area, these remains are covered by a sequence of floors and stone walls belonging to houses of the
Middle Iron Age.
Pits of various sizes obliterate various remains of Building III and represent the first evidence of reoccupation
dating to the Iron Age. Walls of dwellings built with small and medium-sized stones on the slope attest to a
subsequent phase of greater density on the mound (Fig. 6-4). Only small portions of these dwellings are preserved
within the limits of the excavation area, and these are partly eroded by the soil having slid down the slope. There are
remains of walls belonging to some housing units, probably built at slightly different elevations. Found inside a room,
north of wall 239, on the beaten earthen floor (US254) on which layers of ash were also accumulated, was a painted
60 Chapter Six: D 'Agostino et al.
ceramic fragment in the Ali~ar IV style, with animal silhouettes. To the east, in the same room, an ash pit has yielded
many animal bone fragments. A partial tand1r (US300), and remains of burned clay, in relation to a floor with traces
of fire and abundant ashes, as well as remains of animal bones, qualify the space in front (US267) as an area intended
for domestic and kitchen activities; several floors of beaten earth overlap in this area. Textile tools have been found in
relation to several pits (Bonacchi 2020: 64). To the east, sections of walls (US247, US229, US274) belonging to
another house built on the slope feature portions of floors of beaten earth covered by ash (US248 and US271 ), and the
remains of a fireplace (US270).
---====----- 2Meters
N
Figure 6-4. Area D: structures

dating to the Middle Iron Age.
Figure 6-5. Area D: the wall/glacis on the southern slope of the hoyiik (2016).
Above room 433 in Building III, a sequence of floors and walls present a similar type of occupation (D' Agostino
and Orsi 2020: 163, figs. 2-4). The various reconstructions over time along the slope compromised the integrity of the
layers, and the construction at different heights, with soil movement and terracing, makes the interpretation of the
remains more complicated. There are six main phases characterised by remains of domestic architecture, with portions
of walls, floors, fireplaces, and walkways exposed over an area of a few metres. Some retaining walls, close to the
northern limit of the area, continued to be used and were restored over the course of time. The fifth phase from the
bottom is identified by the presence of a large accumulation of small and medium-sized stones (US374), to be
considered part of the structure identified in the central and western part of Area D that covered the upper part of the
hoyuk, probably a glacis, of which several sections have been excavated in the adjacent sector (US87, US200, US300)
(Fig. 6-5). However, it appears less structured here than in the central and westernmost portions. A burnt wooden
sample found inside the wall, probably the fragment of a construction timber, provided a date of the 8th-5 th centuries
BCE (Gminska-Nowak et al. 2021). The geolectric prospection, which focused on the southeastern slope of the hoyuk,
revealed an even larger anomaly in the upper part of the slope that can be interpreted as the continuation of the stone
glacis beyond the eastern limit of Area D, or another section of the wall/glacis connected with the part exposed so far.
At the top of the sequence, and directly below the topsoil, remains of other walls delimiting two rooms, and earthen
floors, indicate the more recent phase of occupation (phase 6).
The sequence of layers in Area D documents a continuous occupation focused on the mound, driven by the need to
use all the space available, including areas on the slope. The western squares are more greatly eroded. Here several
pits (US297-327, US325) cut through the red layer resulting from the decay of the burnt mudbricks, and the plastered
floor (US23 3), belonging to the wall of Building III. A little further east, the remains of a section of wall made from
small and medium-sized stones and burnt and fragmentary mudbricks, and a platform of stones, identify an occupation
phase that reused a floor of Building III. A possible posthole lined with stones cut in the plastered floor, apparently
aligned with posthole US428, has to be considered as being in use during this phase, although we cannot exclude it as
being a reuse of an original posthole of the Hittite building.
AREAA
The activities in Area A were focused on the large Building II, on the portion of the flat terrace northeast of the
hoyuk. The building has an irregular and apparently asymmetrical profile with different wings, some of which project
from the main body of the structure, which takes up at least 2025 m 2 ( 45 m per side) (Fig. 6-6). The results of geo-
physical prospection indicate that it has to be wider.
... In fact, not all the area at the base of the mound
produces geophysical anomalies because of the
presence of later deposits and soil slipping from the
hoyuk, making it impossible to collect more
evidence on its western extension. This means that
the building might continue toward the hoyuk. At
the same time, the limits on the eastern, northern,
and southern sides have not yet been located, either
through geophysical examination or excavations.
What remains are the foundation walls made from
large granite boulders lying on a sub-structure of
middle-sized rough stones. Only a small portion of
the original inner floors has been found, and this
makes it difficult to suggest any functions for the
different groups of rooms exposed, except for a
possible storage space identified by long and parallel
rooms in the northwestern wing. The difficulty of
understanding the use of these spaces is increased by
the post-depositional events in which the original
accumulation of soils and materials was compromised
after the abandonment, and stones and bricks were
removed from the remains of Building II for new
constructions in later centuries. However, its plan,
dimension, and monumentality suggest that Building
II was an important public building in the Late
Bronze Age city.
Over the past few years, investigation has
concentrated on the eastern and northern sides. In
Figure 6-6. Area A: Building II (with some structures particular, between the 2017 and 2020 campaigns,
discussed in the chapter marked). The plan shows walls the eastern sector of the building was explored, close
and structures belonging to different phases (2019). to the southern limit reached in 2016 (D 'Agostino
and Orsi 2020: 164-165, figs. 10-11). Here a wall 3.6 m wide, oriented from west-northwest to east-southeast, with
the external faces preserved for a few centimetres (US245-US250), projects out for 11 m from the main south-
southwest/north-northeast alignment represented by wall US220, and delimits what is most likely a wing of the
building extending eastward. The wall is different in appearance to what has been found so far. Between the external
faces of large stones there is a fill consisting mainly of smaller stones with only a few large boulders, and it appears as
though the large boulders have been removed, exposing the sub-foundation. The eastern edge of the wall lies on a
platform of a clayish soil, reddish in colour, with white limestone inclusions that, at first glance, appeared to be natural
soil.
Here and further north, in relation to the boulders accumulated on the surface, the walls are more poorly preserved
than in other parts of the building and have been dismantled in the past, when it was used as a quarry for construction
materials. Probably the sloping curve of the hill made the boulders easier to detect, and to reuse them for some
purpose in ancient times or, more recently, to remove them as they created obstacles to ploughing.
Figure 6-7. Area A: polychrome stone mosaic floor of Building IL
Further north, the enlargement of the excavation in what appears to be a courtyard bounded on the west by the
main body of the building has produced an interesting and unexpected discovery. Portions of a cobbled stone
pavement (US256: 7.29 x 3 m) have been found in the form of a mosaic created by means of coarser pebbles and
crushed stones arranged in groups by colour (white, black or dark blue, and red, including two yellow pieces) and
following triangular patterns aligned in bands and delimited by frameworks of alternating colours (D' Agostino 2019a)
(Fig. 6-7). Apparently, these are stones other than the granite used for the construction of Building IL The pavement
has the same north-northeast/south-southwest orientation as the eastern walls of Building IL Further to the east two
other decorative bands of stones, white, dark blue, and black in colour and in various shapes, run parallel to pavement
US256. Both bands are 12 cm wide; one is about 5 m long and is composed of three lines of stones, and one is 2 m
long and has two lines of stones. The pieces of stone are laid on a soil base that covers the sub-foundations of the walls
and forms the basis of the pavement. According to the evidence at hand, the pavement must be related to the main
usage phase of Building II and dates to the Late Bronze Age, as it resembles earlier mosaic floors of this kind made
with small stones of different colours documented in Anatolia and probably throughout the Near East.
Excavations at U~akh Ho)'lilc 63
Close to the stones accumulated on the surface, in the area first investigated in 2014, excavation was resumed in
2018 in order to follow the continuation of some foundation walls already identified and only partially exposed. Here
several walls form the comer of a projecting wing (US 122 parallel to US 111, and US 139/140 connecting it to US 111),
which develops northwards and is currently covered by stone boulders out of place. Here too we note the presence of
many medium-sized stones in the sub-foundations as if the construction with large boulders had been partially
dismantled, being located at the point where the ancient slope must have started to descend, such as is the situation
with the southern wall (US245-US250) described above.
Another excavation area concerned the northern limit of the building. Some aligned boulders arranged in two rows
and oriented north-northeast/south-southwest (US294 and US312), perpendicular to the northern wall of Building III,
are part of the original phase of use. Here we have observed the presence of structures belonging to a later phase,
which we can preliminarily date to the Hellenistic/Roman period; these lean against the displaced boulders of Building
III. Evidence of re-use of the foundation walls as a solid basis for the later houses has also been observed in the nearby
western squares excavated in 2016, where new walls are also recognizable in their different building technique and the
laying ofrough, small, and medium-sized stones (Mazzoni et al. 2018: 70, figs. 5--6).
During the last season of work in 2020, a stratigraphic test trench allowed us to observe that a 2 m thick platform
of clayish soil lies beneath the foundation level of the south portion of Building II, probably to create an artificial
terrace intended to level and reinforce the building site. This sub-structure is made of thick layers of clay of a dark
red/brown colour with limestone inclusions (US260) that alternate with thinner layers of dark grey clay, the base of
which is formed by a clayey layer, green and grey in colour. The soil, which was probably transported from elsewhere,
is characterised by an accentuated plasticity, and is similar to natural clay layers. A similar structure was documented
in relation to Building III on the citadel, and traces of some preparatory work have been revealed also in a small test
sounding on the northeastern side of the building, opened in 2013 (D'Agostino 2019b: 70-71, figs. 25 and 29;
Mazzoni and D'Agostino 2015: 168-169). This evidence is lacking on the northern side of the building, where a
second test sounding revealed a layer of grey soil, parts of which are sandy and more than 2 m thick, but show no
other preparatory work. This choice to differentiate the solutions for the construction platform likely depended on the
nature and consistency of the ground where Building II was founded.
CERAMIC EVIDENCE AND CERAMIC TRADITIONS3
The long sequence of occupation brought to light at U~kh Ho)'lilc has provided a rich corpus of pottery, much of
which is being investigated against the background of central Anatolian cultural horizons.
The Ceramic Sequence ofArea D
In Area D, the earliest set of materials derives from the Building III foundation layers which, although related to a
secondary context of deposition, offer a reference post quem for the building's construction. The assemblage (Orsi
2019: 112-124, 2020: 278) testifies to post-depositional events already occurring in ancient times. The corpus, in fact,
includes: red slip and geometric-painted handmade wares dating to the second half of the Early Bronze Age and to the
Transitional period; red slip wheelmade ware characteristic of the Assyrian colonies and Old Hittite periods; and
common Late Bronze Age typologies. Many parallels for the late Bronze Age assemblage are to be ascribed to the
early stages of the Hittite ceramic sequence (see Schoop 2011: 242-243), here including some special classes like the
gold wash ware and gold wash ware imitations. However, a C14 date provided by the remains of a burned post found
in a posthole on the floor of the building attests that the latest pottery may also include more recent specimens, dating
to the first half of the 14th century BCE.4 As far as the phase of use of the building is concerned, unfortunately, the
inventory of artefacts lying on the floor of the building was quantitatively quite limited and confined to potsherds
ranging from medium to medium-poor states of preservation.
The large quantity of Late Bronze Age materials recovered from the Area D post-Building III levels, as well as
from the burnt layers in Area C, although evidently displaced from the spot where they first entered the archaeological
record, are quite likely to be correlated with a phase of life of Building III and, at least some of them, with the last
phase of life of the building (Orsi 2020: 279- 281 ). AC 14 sample from deposits above Building III (Unit 411) giving a
range between the 14th century and the first half of the 13th century BCE at 95% probability, attests to the occupation
of the building in the late stages of the Hittite sequence and gives a reference ante quem for the dating of the most
recent set of Late Bronze Age ceramics.
In relation to Iron Age phasing in central Anatolia, there is no definite agreement. A fine-grained correlation
between archaeological and historical sequences is challenging, and the range of absolute dates available for the
period is not free of issues. For these reasons, Iron Age phases are usually site-specific (Summers 2008; Kealhofer and
Grave 2011). In north central Anatolia, however, the Early Iron Age is mainly equated with the period from the 12th to
10th centuries BCE; the Middle Iron Age with the 10th/9th to 8th centuries BCE; and the Late Iron Age with the 8th/7th to
3
This section was written by Valentina Orsi.
4
The carbon sarnple-USHO66-has a date range between 1415 and 1363 BCE calibrated (Gmmska-Nowak et al. 2021), which
falls between the Middle and Late phase ofthe Hittite ceramic sequence.
the second half of the 4th century BCE (based on Genz 2011; Seeher 2018; but see also Kealhofer and Grave 2011 for
slightly different ranges).
Clean Early Iron Age contexts at U~akh Area D are limited to some pits and small deposits, but potsherds showing
close similarities with Early Iron Age north central Anatolian ceramic assemblages are also scattered as residuals in
different Middle Iron Age layers (Orsi 2020). The main distinctive aspect of this ceramic assemblage is the handmade
forming technique. Two main groupings are attested which appear quite different from one another: the cluster of
plain wares, which may be further distinguished on a functional basis between a common production-the plain and
plain-burnished ware- and a kitchen production, and the cluster of red painted and decorated wares. The plain ware
production is quite rough; the combination of low-capital-intensive technological choices-like the handmade
building process and low-control firing process-together with the morphological inventory limited to simple
utilitarian shapes, mainly with generalised functions, suggests ceramic production processes that were organised on a
relatively small scale, probably at a household level of production.
The same considerations relating to production hold true for the painted wares, but in this case an increase of the
capital invested in technology-as suggested by possible refinement phases on the wheel or by the more accurate
firing process or clay preparation-is not to be excluded. This would point to different workshops, but it is not clear if
these are to be distinguished on the diachronic or contextual level. 5
Regarding chronology, the largest set of plain handmade samples finds good comparisons with the intermediate
and, secondarily, with the late stage ofBtiyi.ikkaya Early Iron Age inventories, mainly dating between the 11 th and 10th
century BCE (Genz 2004: 15-16; Seeher 2018: 89), or, more precisely, between the 11 th and early (Seeher 2018: tab.
1) or mid-10th century BCE (Seeher 2018: 107). Comparisons for most of U~akh's red-painted samples are found in
the Buyi.ikkaya Early Iron Age late phase and with Buyi.ikkale unstratified materials, probably to be dated to the 10th
century BCE. The U~akh Early Iron Age occupation thus spans the 11 th and 10th centuries BCE. A possible beginning
ofU~akh Hoyuk's Early Iron Age occupation in the 12th century BCE is, however, uncertain.
The Middle Iron Age ceramic inventory is quite rich. The pottery is largely wheelmade, but the use of mixed
techniques also seems to be widespread, especially for large vessels like kraters. Matte painted decoration is extremely
common: darkish-coloured paint predominates widely, but the use of bichrome paint, red and black, is also attested
(Fig. 6-8.5). 6
The production is rather varied, with main categories including darkish paint on plain, red fabrics (Fig. 6-8.2--4);
on reddish-brown, burnished fabrics (Fig. 6-8.1); on plain yellowish or buff fabrics (Fig. 6-8.7, 11); or on yellowish
slipped fabrics (Fig. 6-8.8- 9, 12- 13, 15). The use of burnishing is quite widespread, especially in association with
unslipped surfaces (Fig. 6-8. l ). The painted patterns are mostly geometric: typical examples are bowls with slightly
thickened rim adorned by a series of transversal bands and stripes (Fig. 6-8.1). More complex patterns, mainly located
on the everted rim of finer bowls, are also attested, like lozenges-plain or cross-hatched-chevrons, zig-zag rows,
festoons, fishbone and butterfly motifs, and wavy and lace lines. Furthermore, more complex patterns may be located
on the upper part of the body of closed shapes (Fig. 6-8.2--4), among which are a combination of ladder motifs,
chevrons, solid and cross-hatched triangles, concentric circles, chessboards, pendent wavy lines, diamonds, and
lattices. The tree motif is also attested (Fig. 6-8.2).
In addition to geometric painted patterns, naturalistic, silhouette motifs, typical of Ali~ar IV style pottery, are also
present (Fig. 6-8.6-15). 7 Although the painted silhouettes present close similarities between them in terms of
production, only two main variants may be distinguished at the moment in the depiction of animals: one more
"naturalistic," entirely solid, made of accurate, curved lines (Fig. 6-8.7, 13); and one more stylised, made of rather
straight, segmented traits, characterised by wider lines and a typical rendering of the heads, simply outlined and non-
solid (Fig. 6-8.14-15). Ali~r IV style pottery from Area D does not constitute a homogeneous production. In fact, in
addition to relatively "high standard" samples, made of regularly shaped vessels- primarily thrown on the wheel-
with carefully smoothed and slipped surfaces, and intensively decorated (Fig. 6-8.13), "coarser" variants are also
attested, associated with more irregularly walled vessels, with bare and simply smoothed surfaces, and less intensive
decoration (Fig. 6-8. 6-7). The partition of space and the distribution of filling patterns-the typical concentric
circles-in the two silhouette style vessels (Fig. 6-8.7 and 13), for example, are remarkably different; in 6-8.13
concentric circles, of small size, are arrayed according to rather regular spacing, covering the entire area of the panel
delimited by the straight, multiple lines, while in Fig. 6-8.7, as far as we can observe on the preserved section, the
space does not seem to be divided into horizontal panels, and the concentric circles, larger and thicker than in 6-8.13,
are arrayed in sparse, vertical rows. All this points to different workshops, different productions within the same
workshop, or different chronological spans. The possibility of differentiated production finds a parallel at Bogazkoy
(Genz 2000: 38).
5
The fact that plain and painted wares are found together in different Early Iron Age contexts at U~ak:11, however, would suggest
that the two variants, even if related to different chronological spans, at least for a period must have coexisted.
6
A bichrome painted decoration style is already attested in the southeastern central Anatolian Early Iron Age (Genz 2003: 185),
found at Kaman-Kalehoyiik in levels lid 1- 3-attributed to ca. 12th-10th centuries BCE-and Porsuk IV (Matsumura 2008a). The
bichrome painted decoration at Bogazkoy starts to be attested with the Buyiikk:ale-II stage which, following the Buyiikk:aya stage,
represents the second phase of the local Middle Iron Age sequence (Genz 2006: 133), and is considered to be dated from the ca.
mid-9th to the 8th century BCE (Seeher 2018: 141).
7
For silhouette style pottery from Area C, see Orsi 2019: 125, figs. 102, 105.
--
5
15
11 12
■ ■ ■
Figure 6-8. Middle Iron Age pottery from area D, Unit 218 (group no. l); Unit 279, locus 2 (nos. 2-3); Unit 322 (no.
4); Unit 102 (no. 5). Selection of silhouette style potsherds from area D (nos. 6-15).
••
13
CII ■ "' I
--.
16
17
-.
19
22
•• •• ••
-
25
Figure 6-9. Late Middle Iron Age and Late Iron Age pottery from Area D, Unit 369 (nos. 1-9);
Unit 87 (nos. 10-13, 18, 21); Unit 461 (no. 14); Unit 464 (no. 19); Unit 383 (nos. 15-17);
Unit 370 (nos. 20, 22, 24); and Unit 95 (nos. 23, 25).
The appearance of silhouette or Ali~ar IV style pottery in north central Anatolian contexts occurs toward the mid-
9th century BCE.8 Even though post-depositional events seem to have profoundly affected the depositional sequence, a
C14 sample from pit 330- USHO14- dating between the end of the 11 th and 10th centuries BCE, although not
particularly precise, may offer at least a general reference post quem for the appearance of silhouette style pottery at
U~akh Area D (Gminska-Nowak et al. 2021). This is particularly relevant as all the silhouette style potsherds so far
recovered belong to layers above pit 330, which, by contrast, contained a red-painted potsherd probably to be ascribed
to the Early Iron Age ceramic horizon.
As regards the Late Iron Age, C14 dates available so far are still imprecise. For instance, a C14 sample from
carbonised wood retrieved from the stone glacis (Unit 87) described above gave a wide-ranging date of763-486 BCE
at 95% probability; this places the sample either in the Late Iron Age or as early as a late phase of the Middle Iron Age
(Gminska-Nowak et al. 2021). The Late Iron Age ceramic inventory of Area D is characterised by the presence of
plain burnished wares, red slip wares (for these wares see Orsi 2019: 28), and painted wares. Among the painted
typologies, series of simple, horizontal painted bands in brown or darkish brown colour are quite common, largely
attested in association with closed shapes, either small or medium-large in size (Fig. 6-9.2-5, 12-13). Unfortunately,
most of the samples are poorly preserved body sherds, but a large portion of a jar has been reconstructed, and is
characterised by a slightly convex base, elongated body, and vertical handles (Fig. 6-9.19). A similar jar with an
elongated body and banded decoration was discovered at Ali~ar Ho)'lilc, mound D, level 2 (Schmidt 1931: 134, fig.
182). Simple banded decoration finds wide comparisons at Late Iron Age north central Anatolian sites;9 some
similarities with infrequent U~akh samples (Fig. 6-9.5), however, can be found among typical western traditions like
Lydian Fine Ware jars.10 Small body sherds, moreover, might also be portions of simple panelled decorations (see
below), made of monochrome vertical and horizontal stripes, which constitute a further distinctive characteristic of
Late Iron Age central north Anatolian sites. 11
A second typology is represented by samples with bichrome--already attested in the Middle Iron Age--or
polychrome decoration on surfaces, mainly associated with a lustrous finish (Fig. 6-9.6, 10, 14-18, 20-22, 25). The
polychrome rendering is achieved by the combination of fabrics, slips, and paint of different colours. The paint,
always thick and usually brilliant, is primarily monochrome, black in colour, or bichrome, black and red, with the red
colour frequently tending to violet. The painted pattern is applied over a white or yellowish-white base rendered as a
band or a wider panel; it is not always clear if the white base of the panel is a true paint or a dense slip. The geometric
painted patterns include simple horizontal bands and lines, composite series of triangles and parallel lines (Fig. 6-
9.15- 17), chessboards (Fig. 6-9.20-21 ), and more complex panels of alternated latticed panels between straight lines
and lace motifs (Fig. 6-9.14). The panelled decoration is a typical component of Late Iron Age central north Anatolian
ceramic inventories (Kealhofer and Grave 2011: 421), either in the form of simple panels made ofreddish or brownish
stripes over a whitish base, or associated with more complex geometric or naturalistic motifs. 12
Among painted motifs, large concentric circles in brown paint are attested on body sherds of large vessels with
brownish, lustrous surface (Fig. 6-9.24) or yellowish-buff slip (Fig. 6-9.23). Concentric circles are quite common as a
8
Silhouette style pottery at Bogazkoy starts with the Buyiikkale IIb stage, the second Middle Iron Age phase in the local sequence.
Its dating to the first half of the 9th centwy BCE, however, is unfortunately not supported by internal absolute data, not available for
this part of the sequence, but by possible parallels with central Anatolian sequences. In particular, the sequence at Gordion is taken
as a reference: considering that silhouette style pottery at Gordion belongs to levels dating back to the mid-9th centwy BCE and
considering that Gordion is definitely outside the silhouette style pottery core area, the appearance of this ware at Bogazkoy may be
hypothesised to be slightly earlier (Seeher 2018: 144). Dating it to the first half of the 9th centwy BCE would be further confirmed
by the Cl4 sequence at Kaman-Kalehoyilk, where silhouette style pottery is associated with phase 2c of the local sequence
(Matsumura and Omori 2010: 446).
9
See for example Kerkenes Dag1 (Schmidt 1929: fig. 62: K 106, K 113, K 114, K 112, fig. 61: K 101, K 103), Buyukkale le (see
for example Bossert 2000: 96, pl. 56: 601); Ali9ar Hoyuk (Schmidt 1933: pl. 3; pl. 4: 186; von der Osten and Schmidt 1930: 255,
fig. 225; von der Osten 1937: 59, fig. 65: 5; 60, fig. 67: 3) Kaman-Kalehoyuk IIa6-7 (Matsumura 2005: pl. 222: KL87-484),
KU$akh/Sarissa (Powroznik 2010: pl. 68), Ma$at Hoyuk ( Ozguc 1982: fig. K: 8), and <;adir Hoyuk (Kealhofer et al. 2010: 75).
10
A similarity in terms of paint and surface colours and finishing, in fact, is visible between U$akh painted type n. 44 and
specimens from the 6th century BCE Sardis inventory (Cahill and Greenewalt 2020: https://www.levantineceramics.org/wares/785-
lydian-fine-ware, and see https://s3 .amazonaws.corn/lux-production/misc_ images/images/OOO/OOO/l 90/large/open-uri20201105-4-
https://s3 .amazonaws.corn/lux-production/misc_ irnages/irnages/OOO/OOO/ l 90/large/open-uri20201105-4- lh9xspg.jpg? 1604582301 ).
The small size ofthe U9akh samples, however, prevents a complete comparison.
11 For which see for example Oyrnagaac (Y1lmaz 2016: fig. 31: 7- 8); Ma at Hoyuk (Ozguc 1982: pl. 65: 5); Eskiyapar
9
(Bayburtluoglu 1979: 182, fig. 35) Buyiikkale lev. 1 (Bossert 2000: 37, pl. 18: 168; pl. 107: 174); lev. la (Bossert 2000: 72, pl. 38:
348); Buyukkale NW slope (Genz 2006: 113, fig. 18: l; 114, pl. 16: 4); Kerkenes Dag1 (Schmidt 1929: fig. 54, K 85); Karnan-
Kalehoyuk (Matsumura 2005: pl. 278.KL88- l 70, unstratified).
12
For the more simple panels, see for example Buyiikkale lev. 1 (Bossert 2000: 37, pl. 17: 163, pl. 18: 167; 30, pl. 102: 30) and
Kelkenes Dag1 (Schmidt 1929: fig. 60: K 62, K27). For the complex motifs, see for example Buyukkale lev. la (Bossert 2000: 72,
pl. 111: 354-355; 139, pl. 139: 1252-1253; 140, pl. 140: 1285, 1287; 141, pl. 140: 1291); Buyiikkale lev. 1 (Bossert 2000: 97, pl.
56: 614; 72, pl. 113: 359-360); Alaca Hoyuk (<;inaroglu and <;elik 2009: 96, fig. 8); Ma~t Hoyuk (Ozguc 1978: pl. 79: 3, pl. 80:
11, pl. 82: 1-4); Kerkenes Dag1 (Schmidt 1929: fig. 60: K 27; Summers 2021: lOTR22UOlpot01; 11TR24U04pot02,
http://www.kerkenes.metu.edu.tr/kerkl/07finds/InPottery/painted.html#Jars); Ali$M Hoyuk (von der Osten 1937: 61, fig. 68: 2- 3,
5; 68, fig. 71: 4, 11); Kaman-Kalehoyuk (Matsumura 2008b: figs. 14-15); Oluz Hoyuk (Donmez and Naza-Donmez 2009: fig. 28);
and Oyrnagaac (Y1lmaz 2016: fig. 31: 2--6). For the Yozgat area see also Kealhofer et al 2010 and Kealhofer and Grave 2011: 427.
filling in the Middle Iron Age painted pottery, but wide variants seem to be well attested in the Late Iron Age, 13 so that
a more accurate analysis is required in order to define whether Area D samples should be properly ascribed to the Late
Iron Age ceramic corpus or be considered as residual Middle Iron Age samples.
The Ceramic Sequence of Area A
The pottery associated with most of the deposits between and above the remains of the large stone walls of
Building II is largely mixed; ranging from the Early Bronze Age to the Roman period, it reflects the long sequence of
events and post-depositional events that affected the area during the building's construction and after its abandonment.
A reference post quern for the construction of Building II is provided by the ceramic inventory from the test sounding
made in room 126 below the Building II foundation walls; similarly to the Building III foundation layers, the materials
from the Area A test sounding also reflect a sequence of post-depositional events which occurred in antiquity, but the
date range of the most recent set of materials seems likely to be attributed to the period between the 15th and 14th
centuries BCE (Orsi 2018, 2020: 276-278). No materials have been retrieved in situ that may be directly related to
Building II, but the recovery of some miniature vessels in the fills in the central-northern sectors might be connected
to a sacred function of the building (D'Agostino et al. 2020: 136-138).
In the northern sectors of the area, the Late Bronze Age Building II stone foundations served as a base on which to
build new, smaller walls and structures, which probably should be assigned to the Hellenistic and Roman periods
(D'Agostino 2019b: 70, fig. 22; Orsi 2019: 97-98). In the central and southern sectors of Area A, flimsy and irregular
stone installations (D'Agostino 2019b: 70, figs. 20--21) testify to some form of temporary occupation of the area
during the Iron Age. Despite the relatively high incidence of older, residual materials, the deposits immediately above
the large stone foundations seem largely to be attributed to the Late Iron Age and Hellenistic periods. 14
Monochrome and polychrome painted wares with carefully smoothed or burnished surfaces constitute distinctive
types. 15 The state of preservation of the potsherds-largely body-sherds-varies greatly, but in most cases the painted
decoration appears to be associated with closed shapes or large-sized vessels. Typical patterns include simple bands
and stripes in red or reddish-brown paint (Fig. 6-10.2); simple and relatively coarse panelled patterns made of vertical
and horizontal lines (Fig. 6-10.9); bands (Fig. 6-10.1) in red paint; and different variants of more complex panelled
decorations (Fig. 6-10.3- 8). In most cases, they are probably to be related to white panels framed by red (Fig. 6-10.3,
6) or brown lines (Fig. 6-10.4, 7-8). Among the painted patterns, series of concentric circles in brown paint over a
white base constitute a common component (Fig. 6-10.4), finding extensive comparisons in Late Iron Age north
central Anatolian contexts. 16 Fragments with brown paint over a white panel (Fig. 6-10.7- 8) are likely to be related to
more complex panelled decorations with geometric or naturalistic patterns also typical of Late Iron Age north central
Anatolia (see note 13 above for references and discussion).
Among painted decorations, bichrome red and brown painted patterns over a white/yellowish-white base are also
attested; the geometric pattern reproducing a series of brown painted cross-hatched lozenges framed by red bands
located on a reworked potsherd (Fig. 6-10.5) finds close comparisons in Buyillckale level 1, where the same pattern is
attested in bicoloured decoration (Bossert 2000: 37, 72, 130, and see plates: 164, pl. 17, 170, pl. 106, 351, pl. 112 [in
Building 35], 1276, pl. 140). Similar motifs in monochrome paint on white panels and bands are also widely attested
in Late Iron Age contexts, 17 but the matte paint on an Area A sample also recalls Middle Iron Age bichrome painted
potsherds.
Fine and depurated ceramics with carefully smoothed or polished surfaces and painted decoration of red (Fig. 6-
10.16), and red and white bands (Fig. 6-10.15) may be related to the ware groups known as Central Anatolian
Hellenistic Banded Ware or Hellenistic Polychrome Ware of the Galatian type (Lohner-Urban 2018). This specific
variant of fine, banded ware, which is considered typical of the 2nd and first half of the 1st century BCE, would testify
to U~akh's Middle Hellenistic-Early Roman ceramic tradition. 18
13 See for example Buyiikkale NW slope (Genz 2006: 113, pl. 16: 3); Ma~at Hoyuk (Ozg(i1, 1982: fig. L: 5); Kerkenes Dag1
(Summers 2021: 101R22UO 1pot0l, http://www.kerkenes.metu.edu.tr/kerkl/07finds/InPottery/painted.html#Jars); Ali~ar Hoylik

(Schmidt 1933: pl. 9: 77, b 230:15); Ku~h-Sarissa (Powroznik 2010: pl. 68: 1-3); and Gordion (Dusinberre et al. 2020, Gordion
YH53892.l, in 'Provincial Anatolian Lydianizing wares', for which see Dusinberre and Lynch 2020, Provincial Anatolian
"Lydianizing" wares).
14
The presence of Middle Iron Age ceramic types in superficial layers disturbed by ploughing activities, however, attests to the
possibility of earlier forms of utilization of the area.
15 Traces of finishing with a hard tool are occasionally visible on the surfaces; more commonly the surfaces are well smoothed and
characterised by low porosity.

16
Compare for example with Oluz Hoylik (Donmez and Naza-Donmez 2009: fig. 34; Donmez 2015: fig. 22); Alaca Hoylik (Ko~ay
and Akok 1973: pl. 13: Al. r. 28); Kerkenes Dag1 (Summers 2021: 101R22U0lpot01 and spec. 11TR24U04pot02
(http://www.kerkenes.metu.edu.tr/kerkl/07finds/InPottery/painted.htrnl#Jars); and Ali~ar Hoyuk (Schmidt 1933: pl. 9: b 743:16, b
1180:07; von der Osten 1937: 60, fig. 67: 1, 5, 7, 9 and spec. Schmidt 1933: 43, fig. 44, pl. 9: b 742:8; von der Osten 1937: 50, fig.
44: C 128; 58: fig. 64: 2; 60, fig. 67: 11).
17
For which see for example Alaca Hoyuk (Ko~y and Akok 1966: pl. 71); Kerkenes Dag1 (Kerkenes inventory K93.061,
http://www.kerkenes.metu.edu.tr/kerkl/07finds/InPottery/painted.htrn1#Jars; 93SURFU0Opot01); Ali~ar Hoylik (von der Osten
1937: 50, fig. 44: c 128; 54, fig. 51: 3, fig. 58: 8); and K~akh-Sarissa (Powroznik 2010: pl. 67: 7).
18 For related ware types from the surface survey see D'Agostino and Orsi 2015: 76-77, 177-178.
■ ■
7 9
■--
12
L■
--
Figure 6-10. Late Iron Age and Hellenistic Period Pottery from Area A, Unit 285 (nos. 1-5); Unit 180 (no. 6); Unit
297 (nos. 7-9); Unit 295 (nos. 10-14); Unit 227 (no. 15); and Unit 238 (no. 16).
Slightly coarser variants ofred banded ware ceramics (Fig. 6-10.10-14), including shallow bowls with red stripes
and wide bands (Fig. 6-10.10- 11), instead seem more likely to be interpreted as earlier variants, closer to typologies
like the West Anatolian Banded Ware, probably to be dated from the late 4 th to the mid-2nd century BCE, 19 or to more
southern families of red banded potteries dating from the Achaemenid to the early Hellenistic period. 20 The banded
decoration, in general, is typical of Hellenistic central Anatolian tableware inventories, but as may also be appreciated
in the long pottery sequence from U~akh, it has a long tradition in central Anatolian ceramic production, denoting a
relatively high conservatism.
SOME REMARKS ABOUT THE ARCHITECTURE OF

U$AKLI HOYUK/ZIPPALANDA IN THE HITTITE PERIOD21
The recent activities in the area of the largest building so far excavated at the site (Building II in Area A, Fig. 6-6)
have substantially confirmed the preliminary remarks on its official nature, as indicated by its monumental aspect, the
technique of stone-working, the general size, and its having been constructed on an artificial terrace on the eastern side
of the lower town. The practice of raising on artificial platforms has also been documented in Building III in Area D
on the southern side of the central mound: the acropolis of the old settlement ofU~akh. Here, as a high elevation was
already provided by the natural limestone hilltop, we can speculate on the possibility that terracing was necessary for
levelling the ground but also for isolating and separating the basement of the structure. This may also have been the
deciding factor for raising Building II on an artificial platform, combined with the aim of providing it with a high
elevation and prominence in the urban landscape of the lower town.
Interpreting Building II (Mazzoni et al. 2019: 61, see plan 3) on the basis of its architecture and plan requires
comparison to the Bogazkoy/ljattusa temples, especially to the earlier temples (2, 3, 5) characterized by a somewhat
irregular layout of the different units (Schachner 2011: 176- 177; Krause 1940). Hittite sacral architecture is
documented by monumental buildings composed of units opening onto square and often pillared courtyards erected
over artificial platforms or terraces with massive stone foundations and plinths at the base of the walls made of large
boulders and dressed stones. These architectural features connect U~akh Building II to the known Hittite temples,
especially those in the capital. The recent work at ijattusa has shown that Temple 1 was built "on a complex system of
terraces" (Schachner 2020: 111, figs 2, 5) aimed also at creating different elevations and giving a higher location and
visibility to the temple; in the same way, the older temples 2- 5 in the Upper City were also built on terraces
(Schachner 2020: 112). These have been dated later than Temple 1, whose first construction is attributed to the
transition from the Old to the Middle Hittite period, in the passage from the 17th to 16th century BCE (Schachner 2020:
148-150). The older temples of the Upper City (2, 3, 5) were built after Temple 1, and consequently help to anchor the
construction of Building II at U~kh in the period of the emergence of Hittite sacral architecture, probably to the phase
of the first spread of sacral architectural models with sophisticated massive stone-working techniques in central
Anatolia. The results obtained by the recent excavations in the capital and in its regional centres indicate a gradual rise
in centralization which produced, besides military and political empowe1ment and territorial control, a great
investment in planning monumental architecture to host the display and performance of power. U~akh/Zippalanda
belonged to the inner land of ijatti and played a highly significant religious role in the ritual activities which were an
integral part of the representation oflegitimacy of the Hittite state (Torri 2015, 2019).
Other characteristics connect the two buildings, which were both decorated with elements of colour to embellish
walls and floors, although different techniques and gemes were employed: painted plaster in Building III and
polychrome mosaics in Building II. Not only can they be considered as emblematic of the official character and
function of both structures, but their presence at U~akh shows a direct link with the artisans of the Hittite capital, at
least for the use of painted plaster, while the mosaic floor reveals instead an element of autonomy, at least at this point
in our research. It is worth stressing that both gemes attest to the diffusion in Hittite Anatolia of Aegean innovations of
architectural decoration that had spread to Egypt and the Levant since the 17th- 16th centuries BCE; at that time ijattusa
and its main regional centres were included in the international network of interaction linking the kingdoms of the
eastern Mediterranean.
In Areas C and D fragments of painted plaster with red and black geometric motifs on a whitish ground were
collected from different spots in the burnt deposit, coming (probably for Area C, more surely in Area D) from
Building III, whose southern side was excavated in the years 2015- 2019 (D'Agostino 2019b: 72- 73, fig. 56),22 and
which has tentatively been identified as a palace. The fragments of polychrome plaster have been compared with
remains from Bogazkoy/ljattusa and Ma~at Hoyuk and other Anatolian centres, recently re-examined (von Ruden and
Jungfleisch 2017); see the fragments with floral designs from Buyiikkale (von Ruden and Jungfleisch 2017: fig. 2;
19
This specific group of semi-fine potteries, considered typical of western central Anatolia- as far as the Halys river to the East-
are attributed on the basis of the Gordion sequence to a time span ranging from the late 4th to the mid-2nd century BCE (Stewart
2010: 152-153; and 2017 on West Anatolian Hellenistic Banded Ware).
20
As seen for example at Knuk Hoyiik KH-P IIIB (lev. Al.3) and KH-P IIIA (lev. Al.2), dating from the 6th to the mid-2nd century
BCE (Derada and Trameri 2021).
21
This section was contributed by Stefania Mazzoni.
22 Iolanda Cacozza is preparing a dissertation on the U~akb Hoyiik painted plaster; she presented a poster at the 12th ICAANE at
Bologna: 'Another Plaster on the Wall: the evidences ofU~akl1 Hoyiik, a preliminary analysis' .
Neve 1993: 627, fig. 16) and the geometric motifs from Temple 9 (von Ruden and Jungfleisch 2017: fig. 5; Neve
1999: pl. 3 lc) and Temple 5 (von Ruden and Jungfleisch 2017: fig. 7; Neve 2001: pl. 48). Worth mentioning are some
technical features, such as the impressions of string lines which were originally set into the damp plaster (Brysbaert
2002, 2008; von Ruden and Jungfleisch 2017: 70-73) that connect the Hittite specimens to the well documented
frescos of the palaces of Tell Atchana/Alalakh, Qatna, Tell Kabri in the Levant, and Tell ed-Dab'a/Avaris in the
Egyptian delta. These in turn derived from Late Minoan models for fresco technology (unknown to the Near East) and
the figurative repertoire, possibly through the circulation of artisans among these courts during the 17th- 16th centuries
BCE,2 3 anticipating the spread of the "International Style" and its context of diplomatic interrelations. 24
Another geme and technique of decoration using colour-attested this time apparently only at U~kh-is a
polychrome stone mosaic floor documented in the eastern hall or court of Building II in Area A; it consists of a
stratum of irregular small stones and some pebbles of white, red, and black-blue colour laid on the beaten earth and
shaped in rectangular spaces or registers containing triangles. These are surrounded by a frame, also made of stones,
inserted vertically into the earth (D' Agostino 2019a: figs. 5- 6) (Fig. 6-7). The origin of pebble mosaics has been
traced back to the 14th century Mycenaean area, with examples from Tiryns of LH IIIA period (Salzmann 1982: 5, 8,
Cat. No. 129), pre-dating by many centuries the Iron Age (9th-8th century BCE) diffusion of mosaics to Assyria and
the Assyrian provincial towns and to Phrygia, where mosaics were characterized by more sophisticated techniques and
geometric patterns (Bunnens 2016). To the Tiryns examples we may now add the more fine geometric pebble mosaic
of the "Quartier Nu" (west of the Agora) of Mallia.25 This large complex consisted of three wings opening onto a
central court; in the second architectural phase of the LM IIIA2-B 1 period the court was covered entirely with blue
sea-pebbles, and in its southeast angle along the north fas;ade of the southern wing, probably reconstructed as a portico
with columns, a polychrome pebble mosaic decorated the area of the portico. As J. Driessen and A. Famoux write:
here, a square area with central column base was especially decorated with a fine geometric pattern showing
lozenges and spirals, made up of grey, blue and white beach pebbles pushed in a clay floor, thus forming what
is for the moment the finest mosaic floor of the Bronze Age Aegean. Somewhat larger pebbles are placed in
alignments, forming nine horizontal east-west registers, six vertical north-south registers and nine diagonal
lines, whereas small randomly placed pebbles fill in the space between and outside, and four spirals are
preserved in the central side panels (Driessen and Famoux 1994: 61, Pl. III.3; see also Driessen and Farnoux
2012: 300).26
This description helps to underline the technical and decorative parallelism of the Late Minoan and Hittite
mosaics. Coloured pebbles are not so frequent in the area of U~akh, so we can therefore reasonably admit a local
adoption of a foreign technique, possibly on the wave of the diffusion of Aegean-style painted plaster on the walls. In
any case, the presence of this quite exotic element of decoration adds further evidence to support the hypothesis of the
official nature of Building IL
In conclusion, even though the isolated presence of a coloured mosaic at U ~kh may indicate a local line of
cultural contacts with the Aegean, it is more reasonable to consider together the use of painted plaster and coloured
mosaic floor. Both decorative features can be better investigated in the framework of the process of the
monumentalisation of the Hittite capital, where models of architectural decoration, including sculpture, were
introduced and innovated by an original Hittite art language; in the same context, the monumental temples of Ijattusa
established the model of a Hittite sacred architecture for the towns of central Anatolia.
REFERENCES CITED
Bayburtluoglu, Inci. 1979. Eskiyapar Phryg s;ag1. In VIII. Turk Tarih Kongresi, Ankara 11- 15 Ekim 1976, Kongreye
sunulan bildiriler, Cilt 1: 293- 303.
Bonacchi, Alice. 2020. Textile Production in Central Anatolia between the 2nd and the 1st Millennium BC: Analysis of
Tools and Contexts. Asia Anteriore Antica 2: 53-77.
Bossert, Eva-Maria. 2000. Die keramik ph,ygischer zeit von Bogazkoy: Funde aus den Grabungskampagne 1906,
1907, 1911, 1921, 1931-1939 und 1952- 1960. Bogazkoy-Ijattusa: Ergebnisse der Ausgrabungen 18. Mainz am
Rhein: Philipp von Zabern.
Brysbaert, Ann. 2002. Common Craftsmanship in the Aegean and East Mediterranean Bronze Age. Preliminary
Technological Evidence with Emphasis on the Painted Plaster. A.zypten und Levant 12: 95- 107.
- . 2008. The Power of Technolozy in the Bronze Age Aegean and East Mediterranean. The Case ofPainted Plaster.
Monographs in Mediterranean Archaeology 12. London/Oakville: Equinox.
23
The higher date of the destruction of the palace of Tell Kabri dated to the end of the MB Wtransition to MB III around 1700 BCE
on the basis of Cl4 data and in accordance with recent Cl4 data from Tell ed Dab'a suggests as a consequence also a high date for
the diffusion of the Aegean style painted plaster in the East Mediterranean (Hoflmayer et al. 2016).
24
For the exchange of artisans, the main reference is still Zaccagnini 1983. For the "International Style," see Feldman 2006.
25
I am sincerely grateful to Ian Driessen who contributed suggestions and materials to better illustrate these comparisons.
26 The use of extensive pebble flooring with imbedded column bases is also documented at Sissi (Driessen et al. 2011, inter alia 99,
114, fig. 5.41 ).

Bunnens, Guy. 2016. Neo-Assyrian Pebble Mosaics in their Architectural Context. In The Provincial Archaeolozy of
the Assyrian Empire, J. McGinnis, D. Wicke, and T. Greenfield, eds., 59- 70. Cambridge: Ziyaret Tepe
Archaeological Trust.
Cahill, Nick, and Crawford H. Greenewalt Jr. 2020. Lydian Fine Ware. The Levantine Ceramics Project,
https://www.levantineceramics.org/wares/785-lydian-fine-ware.
<;inaroglu, Aykut and Duygu <;elik. 2009. Alaca Hoyi.ik 2007 Y1h Kaz1s1. Kazz Sonuc;larz Toplantzs1 30(3): 91-104.
D'Agostino, Anacleto. 2019a. A Mosaic Floor from the Late Bronze Age Building II of U~akh Hoyillc, Central
Turkey. Antiquity 93/372: 1- 8.
- . 2019b. A Stratigraphic Summary. In Exploring a Site in the North Central Anatolian Plateau: Archaeological
Research at U~akh Hoyiik (2013-2015), S. Mazzoni, A. D'Agostino, and V. Orsi, eds. Asia Anteriore Antica 1:
70- 91.
- . 2020. Tracing Fire Events and Destructions of Late Bronze Age Date: The End of the Hittite Building on the
Citadel of U~akh Hoytik. In Anatolia Between the 13th and the 12th Century B.C.E., S. de Martino and E.
Devecchi, eds., 69-93. Eothen 23, Collana di Studi sulle Civilta dell'Oriente Antico. Firenze: LoGisma Editore.
D'Agostino, Anacleto and Valentina Orsi. 2015. The Archaeological Survey: Methods and Materials. In The U~akh
Hoyiik Survey Project (2008- 2012). A Final Report, S. Mazzoni and F. Pecchioli Daddi, eds., 35- 343. Studia
Asiana 10. Firenze: Firenze University Press.
- . 2020. Preliminary Report on the 2018 Excavation Season at U~akh Hoyillc (Yozgat). Kazz SonU<;lan Toplanhs1
41(1): 161-172.
D'Agostino, Anacleto, Valentina Orsi, and Giulia Torri. 2020. Looking for Traces of Cultic Practices at U~akh Hoyillc.
Some Remarks about Buildings, Texts and Potsherds. News from the Lands of Hittites-Scientific Journal for
Anatolian Research 3-4, 2019-2020: 113-142.
Derada, Martina and Andrea Trameri. 2021. Continuity of Local Ceramic Traditions at K1mk Hoyillc (Southern
Cappadocia): Late Achaemenid Pottery from the 'NW Building.' Paper presented at the 12th ICAANE, Bologna
2021.
Donmez, Sevket. 2015. Preliminary Results on the Hellenistic and Iron Age Phases at Oluz Hoyillc. In Recent Studies
on the Archaeology of Anatolia, E. Lafh and S. Patac1, eds., 255-272. BAR International Series 2750. Oxford:
Archaeopress.
Donmez, Sevket and E. Emine Naza-Donmez. 2009. Amasya-Oluz Hoyillc Kazis1 2007 Donemi <;::ah~malar1: Ilk
Sonm;lar. Belleten LXXIIl/267: 395-421.
Driessen, Jan and Alexandre Farnoux. 1994. Mycenaeans at Malia? Archaeolozy 1: 54---64.
- . 2012. A House Model from Mallia. In: Kretes Minoidos. Tradizione e identita minoica tra produzione artiRianale,
pratiche cerimoniali e memoria de/ passato. Studi offerti a Vincenzo La Rosa per ii suo 70° compleanno, F.
Carinci, N. Cucuzza, P. Militello, and 0. Palio, eds., 299-310. Studi di Archeologia Cretese 10. Padova: Bottega
d 'Erasmo.
Driessen Jan, Ilse Schoep, Frank Carpentier, Isabelle Crevecoeur, Maud Devolder, Florence Gaignerot-Driessen,
Piraye Haciguzeller, Valasia Isaakidou, Simon Jusseret, Charlotte Langohr, Quentin Letesson, and Aurore Schmitt.
2011. Excavations at Sissi, II. Preliminary Reports on the 2009-2010 Campaigns. Louvain: Presses Universitaires
de Louvain.
Dusinberre, Elspeth and Kathleen Lynch. 2020. Provincial Anatolian "Lydianizing" wares. The Levantine Ceramics
Project, https://www.levantineceramics.org/wares/792-provincial-anatolian-lydianizing-wares.
Dusinberre, Elspeth, Kathleen Lynch, and Mary M. Voigt. 2020. Gordion YH53892.l. The Levantine Ceramics
Project, https://www.levantineceramics.org/vessels/23209-gordion-yh53892- l.
Feldman, Marian H. 2006. Diplomacy by Design: Luxury Arts and an "International Style" in the Ancient Near East,
1400-1200 BCE. Chicago: University of Chicago Press.
Genz, Hennann. 2000. Die Eisenzeit in Zentralanatolien im Lichte der keramischen Funde vom Buyukkaya in
Bogazkoy/ljattusa. TUBA-AR III: 35-54.
- . 2003. The Early Iron Age in Central Anatolia. In Identifying Changes: The Transition from Bronze to Iron Ages in
Anatolia and its Neighbouring Regions, Proceedings of the International Workshop, Istanbul, November 8- 9,
2002, B. Fischer, H. Genz, E. Jean, and K. Koroglu, eds., 179-191. Istanbul: Turk Esh;:ag Bilimleri Enstitusii.
- . 2004. Biiyiikkaya. I. Die Keramik der Eisenzeit. Funde aus den Grabungskampagnen 1993 bis 1998. Bogazkoy-
ijattusa: Ergebnisse der Ausgrabungen XXL Mainz: Philipp von Zabern.
- . 2006. Die eisenzeitliche Besiedlung im Bereich der Grabungen am mittleren Biiyukkale-Nordwesthang 1998-
2000. In Ergebnisse der Grabungen an den Ostteichen und am mittleren Biiyiikkale-Nordwesthang in den Jahren
1996-2000, J. Seeher, ed., 98-158. Bogazkoy-Berichte 8. DeutschesArchaologisches Institut.
- . 2011. The Iron Age in Central Anatolia. In The Black Sea, Greece, Anatolia and Europe in the First Millennium
BC, G.R. Tsetskhladze, ed., 331-368. Colloquia Antiqua 1. Leuven, Paris, Walpole MA: Peeters.
Gmiii.ska-Nowak, Barbara, Anacleto D'Agostino, Yasemin Ozarslan, Valentina Orsi, Anastasia Christopoulou,
Stefania Mazzoni, Dnal Akkemik, and Tomasz Wazny. 2021. Dendrochronological Analysis and Radiocarbon
Dating of Charcoal Remains from the Multi-Period Site of U~akh Hoyillc, Yozgat, Turkey. Journal of
Archaeological Science: Reports 38.103078.
Excavations at U~akh Hoyillc 73
Hoflmayer, Felix, Assaf Yasur-Landau, Eric H. Cline, Michael W. Dee, Brita Lorentzen, and Simone Riehl. 2016.
New Radiocarbon Dates from Tell Kabri Support a High Middle Bronze Age Chronology. Radiocarbon 58(3):
599-613.
Kealhofer, Lisa and Peter Grave. 2011. The Iron Age on the Central Anatolian Plateau. In The Oxford Handbook of
Ancient Anatolia, S.R. Steadman and G. McMahon, eds., 415-442. Oxford: Oxford University Press.
Kealhofer, Lisa, Peter Grave, Ben Marsh, Sharon Steadman, Ronald L. Gorny and Geoffrey D. Summers. 2010.
Patterns of Iron Age Interaction in Central Anatolia: Three Sites in Yozgat Province. Anatolian Studies 60: 71-92.
Ko~y, Hamit Z. and Mahmut Akok. 1966. Alaca Hoyiik Kazzsz. 1940-1948'deki <;alz~malarz ve Ke~iflere ait j/k
Rapor. Ausgrabungen von Alaca Hoyiik. Vorbericht iiber die Forschungen und Entdeckungen von 1940-1948.
Turk Tarih Kurumu Yaymlanndan V. Seri- Sa. 6. Ankara: Turk Tarih Kurumu.
Ko~y, Hamit Z. and Mahmut Akok. 1973. Alaca Hoyiik Kazzsz. 1963- 1967 <;alz~malarz ve Ke~zjlere ait j/k Rapor.
Alaca Hoyiik Preliminary Report on Research and Discoveries 1963-1967. Ankara: Turk Tarih Kurumu Basllllevi.
Krause, Karl. 1940. Bogazkoy Tempel V Ein Beitrag zum Problem der hethitischen Baukunst. Istanbuler Forschungen
Bd. 11, Keilschrifturkunden aus Bogazkoy. Berlin: Archaologisches Institut des Deutschen Reiches.
Lohner-Urban, Ute. 2018. Central Anatolian Hellenistic Banded Ware. The Levantine Ceramics Project,
https://www.levantineceramics.org/wares/522-central-anatolian-hellenistic-banded-ware
Matsumura, Kimiyoshi. 2005. Die eisenzeitliche keramik in Zentralanatolien aufgrund der Grund/age der Ausgrabung von
Kaman-Kalehoyiik. Ph.D. dissertation, Freie Universitat Berlin.
- . 2008a. The Early Iron Age in Kaman-Kalehoyillc: The Search for its Roots. In Fundstellen. Gesammelte Schriften
zur Archaologie und Geschichte Altvorderasiens ad honorem Hartmut Kuhne, D. Bonatz, RM. Czichon, and F.J.
Kreppner, eds., 41-50. Wiesbaden: Harrassowitz Verlag.
- . 2008b. A Note on Anatolian Iron Age Ceramic Chronology: Black Lustrous Ware with Diamond Faceting.
Anatolian Archaeological Studies 17: 175-184.
Matsumura, Kimiyoshi and Takayuki Omori. 2010. The Iron Age Chronology in Anatolia Reconsidered: The Results
of the Excavations at Kaman-Kalehoyillc. In Proceedings of the (Jh International Congress on the Archaeology of
the Ancient Near East, May 5th-J(Jh 2008, 'Sapienza'-Universita di Roma, P. Matthiae, F. Pinnock, L. Nigro, and
N. Marchetti, eds., 443-455. Wiesbaden: Harrassowitz Verlag.
Mazzoni, Stefania and Anacleto D'Agostino. 2015. Research at U~kh Hoyillc (Central Anatolian Plateau). In The
Archaeology of Anatolia: Current Work (2013-2014), S.R. Steadman and G. McMahon, eds., 149-179.
Cambridge: Cambridge Scholars Publishing.
Mazzoni, Stefania, Anacleto D'Agostino, and Valentina Orsi. 2018. New Results from the Excavation Season 2016 at
U~akh Hoyillc (Y ozgat). Kazz Sonuc;larz Toplantzsz 39(3): 69-82.
- . 2019. Exploring a site in the North Central Anatolian Plateau: Archaeological Research at U~akh Hoyillc (2013-
2015). Asia Anteriore Antica 1: 57-142.
Neve, Peter. 1993. Die Ausgrabungen in Bogazkoy-Ijattusa 1992. ArchaoloRischer AnzeiRer 1993: 621- 652.
- . 1999. Die Oberstadt von flattusa. Die Bauwerke I. Das zentrale Tempelviertel. Bogazkoy-Ijattusa 16. Berlin:
Gebr. Mann.
- . 2001. Die Oberstadt von flattusa. Die Bauwerke II. Die Bastion des Sphinxtores und die Tempelviertel am Konigs-
und Lowentor. Bogazkoy-Ijattusa 17. Mainz am Rhein: Philipp von Zabem.
Orsi, Valentina. 2018. Reading the Late Bronze Age Ceramic Evidence at U~akh Hoyillc (Central Turkey). The Pottery
from the Area A Test Sounding. Anatolica 44: 179- 211.
- . 2019. Ceramics and Materials. In Exploring a Site in the North Central Anatolian Plateau: Archaeological
Research at U~akh Hoyiik (2013-2015), S. Mazzoni, A. D'Agostino, and V. Orsi, eds. Asia Anteriore Antica 1:
92- 142.
- . 2020. The Transition from the Bronze to the Iron Age at U~akh Hoyillc: The Ceramic Sequence. In Anatolia
between the 13th and the 12th Century BCE. S. De Martino and E. Devecchi, eds., 271- 316. Eothen 23. Firenze:
LoGisma Editore.
Ozgui;, Tahsin. 1978. M~at Hoyiik kazzlarz ve c;evresindeki ara~tzrmalar/Excavations at M~at Hoyiik and
investigations in its vicinity. Ankara: Turk Tarih Kurumu Basllllevi.
- . 1982. Ma~at Hoyiik II. Bogazkoyiin kuzeydogusunda bir Hitit merkezi. A Hittite Centre Northeast of Bogazkoy.
Ankara: Turk Tarih Kurumu.
Powroznik, Klaus J. 2010. Die Eisenzeit in Ku~aklz. Kusakli-Sarissa Band 5. Rahden/Westf.: Marie Leidorf GmbH.
Salzmann, Dieter. 1982. Untersuchungen zur den Antiken Kieselmosaiken von den Anfangen bis zum Beginn d.
Tesseratechnik. Archaologische Forschungen 10. Berlin: Berlin Deutsches Archaologisches Institut, Mann Verlag.
Schachner, Andreas. 2011. Hattuscha. Auf der Suche nach dem sagenhaften Grossreich der Hethiter. Munchen: C.H.
Beck.
- . 2020. The Great Temple at Ijattusa. Some Preliminary Interpretations. In Cult, Temple, Sacred Spaces, Cult
Practices and Cult Spaces in Hittite Anatolia and Neighbouring Cultures. ProceedinRS of the First International
HFR Symposium, Mainz, 3- 5 June 2019, S. Gorke and Ch. W. Steider, eds., 105- 158. Studien zu den Bogazkoy-
Texten Bd. 66. Wiesbaden: Harrassowitz Verlag.
Schmidt, Erich F. 1929. Test Excavations in the City ofKerkenes Dagh. American Journal of Semitic Languages and
Literatures 45(4): 221-274.
-. 1931. Anatolia through the Ages. Discoveries at the Alishar Mound 1927-9. OIC 11. Chicago: The University of
Chicago Press.
- . 1933. The Alishar Huyuk. Seasons of 1928 and 1929. Part II. OIP 20. Chicago: The University of Chicago Press.
Schoop, Ulf-Dietrich. 2011. Hittite Pottery: A Summary. In Insights into Hittite History and Archaeology, H. Genz
and P. Mielke, eds., 241- 274. Colloquia Antiqua 2. Leuven: Peeters.
Seeher, Jurgen. 2018. Bityiikkaya 11. Bauwerke und Befunde der Grabungskampagnen 1952-1955 und 1993-1998. Mit
Beitriigen von Ulf-Dietrich Schoop und Sven Kuhn. Bogazkoy-ijattusa 27. Berlin/Boston: De Gruyter, GmbH.
Stewart, Shannan M. 2010. Gordion After the Knot: Hellenistic Pottery and Culture. Ph.D. dissertation, Department of
Classics, University of Cincinnati.
-. 2017. West Anatolian Hellenistic Banded Ware. The Levantine Ceramics Project, accessed on 10 April 2021,
https://www.levantineceramics.org/wares/480-west-anatolian-hellenistic-banded-ware
Summers, Geoffrey D. 2008. Periodisation and Technology in the Central Anatolian Iron Age. Archaeology, History
and Audiences. Ancient Near Eastern Studies 45: 202- 217.
- . 2021. Excavations at the Cappadocia Gate: Kerkenes Final Reports 1. OIP 145. Chicago: The Oriental Institute.
Torri, Giulia. 2015. Epigraphic Evidence about Zippalanda. In The U~aklz Hoyiik Survey Project (2008-2012). A Final
Report, S. Mazzoni and F. Pecchioli Daddi, eds., 365- 367. Studia Asiana 10. Firenze: Firenze University Press.
-. 2019. Did the Storm God of Zippalanda have a Mother or a Wife? Remarks about the Cults of Katabba and the
Sun Goddess of the Earth in Zippalanda and Ankuwa. Asia Anteriore Antica 1: 217- 224.
von der Osten, Hans H. 1937. The Alishar Hityiik Seasons of 1930-32, Part III. OIP 30. Chicago: The University of
Chicago Press.
von der Osten, Hans H. and Erich F. Schmidt. 1930. The Alishar Hityiik Season of 1927, Part 1. OIP 6. Chicago: The
University of Chicago Press.
von Ruden, Constanze and Johannes Jungfleisch. 2017. Incorporating the Other: A Transcultural Perspective on Some
Wall Painting Fragments from Hattusha. Byzas 23: 61-83.
Yilmaz, Mehmet A. 2016. Iron Age Pottery. In Archiieologische Forschungen am Oymaagac; Hoyiik/Nerik 2011-
2015, R.M. Czichon, J. Klinger, P. Hnila, D.P. Mielke, H. Bohm, C. Forster, C. Griggs, M. Kahler, G.K. Kunst, M.
Lehman, B. Lorentzen, S. Manning, K. Marklein, H. Marquardt, S. Reichmut, J. Richter, C. Rossner, B. Sadiklar,
K. Seufer, R. Sobott, I. Traub-Sobott, H. von der Hosten-Woldenburg, M. Weber, H. Wolter, and M.A. Y1lmaz.
Mitteilungen der Deutschen Orient-Gesellschafi 148: 68-73.
Zaccagnini, Carlo. 1983. Patterns of Mobility among Ancient Near Eastern Craftsmen. Journal of Near Eastern
Studies 42(4): 245-264.
CHAPTER SEVEN
A FRESH PERSPECTIVE ON THE MIDDLE BRONZE AGE AT TELL ATCHANA,

ALALAKH:THE2007-2019SEASONS
MURAT AKAR, K. ASLIHAN YENER, MDGE BULU, AND TARA INGMAN
Tell Atchana, Alalakh, holds a unique position in the 2nd millennium BC 1 political, economic, and cultural history
of Anatolia, the Near East, and the Levant. Located in a buffer zone that connects distant and distinct regions along the
southernmost fringe of the Anatolian peninsula in modem Turkey, the site has the potential to provide answers to a
long list of research questions that dominate the literature, from the late 3rd millennium BC collapse, to 2nd millennium
BC urbanization, to the transition from the 2nd to !81 millennium BC (Fig. 7-1). This includes themes from craft
production to imperial strategies and systems of exchange, as well as controversial topics such as climate change-
induced collapse, urbanization, and mobility patterns.
=========:::::iKilometers
65 130 260 390
Figure 7-1. Map showing major second millennium BC Anatolian and Near Eastern sites
(©Alalakh Excavations Archive, map by M. Akar).
This chapter exclusively deals with, and provides an overview of, the recent archaeological fieldwork (2007-2019)
that sheds light on the late Middle Bronze Age (MBA, hereafter) sequence, when the site reached a flourishing state
under the hegemony of the Kingdom of YamlJad as the capital city of the kingdom later called Mukis. This is defined
as a period of prosperity, urban expansion, and establishment of cross-cultural contacts generally accepted to end ca.
1650 BC, according to the Middle Chronology, with the early Hittite campaigns in the region (Yener 2007).
The beginnings of this urban expansion period, on the other hand, remain in need of further exploration, as the end
of the 3rd millennium BC is marked by the presence of numerous destructions, collapse, and abandonment layers in a
wide geographical framework; in a regional framework, it may be defined by declines in the number of settlements
(see for instance, Schwartz and Nichols 2006; Kuzucuoglu and Marro 2007; Weiss 2017). This is observed through
key sites and their surroundings, such as Kiiltepe (Kanes) in central Anatolia, Tell Mardikh (Ebla) in the northern
Levant, and Tell Leilan in the IJabur Valley. The succeeding revival and re-establishment phases at the
aforementioned sites point to urban expansions, often linked to the formation of territorial city-state kingdoms and
1
As a policy, Tell Atchana Excavations uses "BC" in order to be consistent with the high number of calibrated Cl4 dates.
76 Chapter Seven: Akar et al.
their successful exploitation of production and surplus supply, as well as to the supra-regional trade networks that
were the hallmark of the MBA (Schwartz 2006; Vener 2007; Laneri and Schwartz 2011; Kulakoglu 2011; Akar and
Kara2020).
In the Amuq Valley, this transition is still ambiguous, due to the fragmented nature of the archaeological data
between the two tell sites, Atchana and Tayinat. Tell Tayinat revealed the major 3rd and 1st millennium BC sequence
(Welton et al. 2011; Harrison 2013), whereas Tell Atchana was mainly occupied throughout the 2n d millennium BC
(Vener 2013a). This disconnected view lies partly in the modem topography and partly in how these two tells were
approached by archaeologists as distinct entities. The recent discovery of Iron Age levels at Tell Atchana has shown
that the Iron Age city expanded beyond the limits of Tell Tayinat (Vener 2013a; Montesanto and Pucci 2019-20).
These new archaeological finds further suggest that the temporal shifts observed between the two locations from the
3rd millennium to the 1st millennium BC created a multiproxy city which developed according to changes in the
riverbed of the Orontes. This is clearly visible in various boring/coring projects, which have revealed several diverse
riverbed channels and the presence of lower towns expanding beyond the visible tells, including the small mound of
Tayinat al-Saghir (Batiuk 2007; Horowitz et al. 2019; Av~ar et al. 2019, 2020). Therefore, the larger cityscape of Tell
Atchana and Tell Tayinat should then be seen as the palimpsest of the multilayered history of a "Megacity" shaped by
the Orontes River (Fig. 7-2, Vener 2005: 4, 2013a: 22).
Figure 7-2. Map showing Tell Tayinat (AS126), Tayinat al-Saghir (AS127), and Tell Atchana (AS136)
representing different loci of a Megacity (©Alalakh Excavations Archive, photo by M. Akar).
The view, on the other hand, is more elusive for the Early- Middle Bronze Age transition. At Tell Mardikh (Ebia),
a late 3rd millennium BC Palace G text refers to a dependent city called A-la-la-!Ju-um in various forms, likely
associated with the 2nd millennium BC city of Alalakh (Matthiae 1978; Astour 1992; Vener 2005; Archi 2006, 2020).
This textual evidence may then be geo-referenced to the Early Bronze Age remains at Tell Tayinat, as Tell Atchana,
according to Woolley's temple sounding, revealed no Early Bronze (EB) IVA (2500- 2300 BC) remains contemporary
to the Tell Mardikh (Ebia) Palace G (see Mellink 1957; Batiuk and Horowitz 2010 contra Woolley 1955). This
suggests that the heart of the city moved, yet the name continued to be used in the Middle and Late Bronze Ages
(Vener 2005: 4).
At Tell Atchana, except for the broad-scale horizontal exposure of the monumental Level VII Palace, temple, and
gate complex, the earlier MBA remains and their stratigraphic and chronological correlations were left ambiguous in
Woolley's final publication (1955). Due to the high water table, the stratification pit that was dug through the southern
end of the Level VII Palace in squares L-K 13- 15 reached 13.75 m below the top of the mound. The excavation area
became progressively smaller due to the depth and the difficulty of removing soil, and Levels XVI-XIV were
investigated only in 10 x 8 m exposures. Level XVII was not explored, due to its being under water (Woolley 1955:
10- 11, fig. 2). The stratification pit, the temple sounding, the Level IV Palace soundings, and the Site H and Trench F
A Fresh Perspective on the Middle Bronze Age at Tell Atchana 77
excavations generated data for the MBA but with limited exposures. Today, the temple sounding resembles a meteor
crater, which is partially filled with decayed mudbrick detritus and which is visible down to Level XVI, revealing an
uninterrupted sequence of temples. The Palace stratification pit reached Level XVII, where the earlier house/palace
complexes of the city were explored. The third deep exploration was in the Level IV Palace room 22, which revealed a
severely burnt workshop with substantial architecture, whereas Site H and Trench F revealed the city fortifications
(Woolley 1955: 106-110, 133, pl. XXII). Unfortunately, the limited documentation from the MBA soundings, the
absence of elevation data, and the inaccessible field records have prompted the complicated task of re-exploring
Woolley's MBA stratigraphic sequence. This operation became more complex with the long, steady, slow-changing
nature of the material culture, often creating difficulties in defining index markers for chronological implications.
Nevertheless, the earliest levels encountered in the temple sounding at Tell Atchana point to the presence of
monumental architecture, reflecting signs of a well-organized administrative system indicative of urbanization
(Woolley 1955: 36, fig.19).
Indirectly, the signs of a strong administrative system are also evident in the recent excavations conducted at the
peripheral site of Toprakhisar Hoyuk in the Altmozu highland, located roughly 10 km from Alalakh. The presence of
an administrative building there, dated to the early 20 th century BC, is regarded as the reflection of the business of
olive oil and wine production operated under the management of the kings of Alalakh (Akar and Kara 2020). The high
number of dispersed small and medium sized MBA sites across the Amuq Valley points, on the other hand, to an
efficient organization of agricultural practice (Bulu 2017a). According to recent sedimentary corings around Tell
Atchana, the documented charcoal intensity may indicate a significant increase in agricultural activities during the
MBA (Av~ar et al. 2019). Thus, intra- and offsite data confirm that the Amuq Valley and its surroundings benefited
from various aspects of urbanization at the beginning of the MBA.
As Sydney Smith (1949: 8) stated, Tell Atchana provided the potential of establishing linkages to historical events
from stratified archaeological layers. In this delicate task, scholarly work was predominantly concentrated on
redefining Woolley's chronology through his stratigraphical attributions (see Mellink 1957; Gates 1981; McClellan
1989). Apart from Heinz's (1992) in depth re-examination reconstructing the developmental patterns of the MBA
material culture and chronology, no further studies were conducted to understand the early 2nd millennium BC
sequence. This has become one of the primary research goals of the recent excavations at Tell Atchana.
THE NEW EXCAVATIONS (2007-2019)

The continuous alterations of the Royal Precinct
throughout the MBA are best observed in the ruins
of the burnt gateway of Level VII, lower in
elevation than the Level VII Palace and Temple,
which were built on higher ground in the northeast
comer of the Royal Precinct. While a gentle rising
grade from the gateway to the palace and temple
complex is to be expected, for a careful observer,
the relationship between the still-standing walls of
the MBA Level VII Palace and the Late Bronze Age
Level IV Palace is somewhat puzzling. Although
there seems to be at least 200 years between these
two palace complexes, today one can look at the
western fa9ade of the Level VII Palace by standing
in the southeast comer of the much younger Level
IV Palace. In fact, the floor level of the Level VII
Palace southern wing is on the same level as the
Level IV Palace. This is due to major MBA
construction projects that ultimately altered the
topography of the Royal Precinct, where several
platforms and terracing operations transformed this
area into a stepped terrace, creating a puzzling
stratigraphic effect.
According to new research conducted on a
broader area investigating the Late Bronze (LB) II
stratigraphy at Tell Atchana, a new site sequence is
presented using "Periods" rather than Woolley's
"Level" attributions (Y ener 2013a; Yener et al.
2019). This is still in progress for the MBA. Thus,
for the MBA, the Period and Level designations are
in accordance with Woolley, meaning that Period 7-
Figure 7-3. Map of Tell Atchana (©Alalakh 8-9 is equivalent to Level VII-VIII-IX.
Excavations Archive, map by M. Akar).
Fourteen excavation squares have been placed along an east-west orientation cross-cutting the Royal Precinct of
Alalakh (Area 1), where stratigraphic correlations can be obtained with the standing architectural remains of Levels IV
and VIL Three excavation units yielding MBA levels were located in the Level VII (Squares 33.32 and 33.53) and
Level IV Palace courtyards (Square 32.57). While Square 33.32 and Square 32.57 were successful in exploring the
stratigraphic sequence from pre-Level VII to pre-Level IV Palace contexts, Square 33.53, expanding to Rooms 16, 17,
18, 22, 34, and 35, revealed new data regarding the construction history of the Level VII Palace. The excavations
conducted on the northeastern slope of the tell, on the outskirts of the Royal Precinct in Area 3 (Squares 45.44 and
45.45), revealed a sequence of city fortification systems from MB II to LB I and a slope cemetery. The stratigraphic
excavations conducted in the southwestern elevated portion of the tell in Area 4 recently revealed MBA levels, which
will be detailed elsewhere once the contextual study is finalized (Fig. 7-3).
Area 1, Square 32.57
The deep sounding located in the courtyard of the Level IV Palace has generated data regarding the poorly
understood, problematic phases represented by Levels VI and V in Woolley's chronology. The Local Phases 1-3 in
these squares represent the LB I levels. Local Phase 4 is a short-lived MB II-LB I transitional phase defined by
sacrificial pits as part of a termination ritual over the remains of the partially exposed monumental building complex
(Local Phase 5) that dates to the MBA (Yener 2015, 2017). This long-lasting building revealed seven different phases
defined according to continuous alterations, the raising of floors, and various modifications in the arrangement of
spaces. Of most particular interest is the presence of a curvilinear southern wall (in Local Phases 5b, f, and g), which
added a remarkable character; it is defined as the "Apsidal Building" (Yener 2015; Yener et al 2020; cf. Fig. 7-4).
While Local Phase 5 revealed excellent levels of preservation, with walls standing up to 3 m tall, the continuous
alteration of the rooms resulted in the clearance of floors prior to the repacking of new floors, leading to a limited
presence of materials that can be used to attribute functional designations to partially exposed rooms. Yet, from its
adherence to a sacred space in each phase, its unique architectural features in close proximity to the Ishtar Temple, and
ritual objects, the building is of significant importance and likely functioned as a temple dedicated to one of the gods
or goddesses of Alalakh, further discussed below (Y ener 2015; Yener et al. 2020: 9).
Figure 7-4. Aerial view of Square 32.57, Local Phase 5g Apsidal Building in Area 1
(©Alalakh Excavations Archive, photo by M. Akar).
Area 1, Squares 33.32 and 33.53
The Level VII Palace is an extraordinary structure that expands over a 100 x 30 m wide area. Together with the
Northern Palace at Tell Mardikh (Ebla), it represents the characteristics of palace architecture in the region during the
rule of the Amorite dynasty (Matthiae 2002). It is located in the northeast sector of the Royal Precinct, incorporated
into the city wall on the east. The rooms in the north (1-8) are at a much lower elevation and are interpreted to be the
residential quarters, the audience chamber, and the throne room, with courtyards providing access. The staircase
located in the northeast comer of the palace (room 3) indicated that a tower was built in conjunction with the city wall.
The elevation rises south of courtyard 9 via a staircase, where one could reach the more utilitarian sectors of the
palace. This creates approximately 2 m of elevation difference within the same building. Occupied by several rooms
and courtyards, the southern wing of the palace has facilities that served as workshops and for cooking and storage,
along with the enigmatic addition of the royal hypogeum (Woolley 1955: 93, fig. 35).
Square 33.32 is located in the large lower courtyard in the north, whereas Square 33.53 expanded across several
rooms on the high terrace in the southern sector of the palace (Fig. 7-5). These two squares now provide a fine-tuned
stratigraphy for understanding the earlier MBA levels, as well as the subsequent step-by-step construction phases of
the Level VII Palace. The results acquired are thus altering our understanding of commonly accepted paradigms, such
as the origins of wall painting practices in the eastern Mediterranean.
>\ ~
~
evel VII Gateway
Level IV Palace
N
Legend
Period 4/Level IV
. . . Period 7/Level VU
~ New Exposures
0 5 10 ~ ~ ~
-=:-==---===- -Meters
Figure 7-5. Plan showing the remains of the earlier palace in Squares 33.32 and 33.53 in relation to the
Level VII Palace (©Alalakh Excavations Archive, plan by O.H. K1rman).
The excavation conducted in Square 33.32 revealed the presence of an earlier palace complex with at least three re-
building phases (3a, b, c). The severely burnt earliest Local Phase 3c revealed two rooms, one partially exposed,
functioning as the cooking, storage, and serving quarter of a large complex (Bulu 2016). The thickness of the walls
showed that the structure was at least two stories high. Of particular importance, the southern wall likely functioned as
a terrace wall, since beyond this point, the continuation of the building was exposed at a much higher elevation in
Square 33.53 in the south. This showed that the terraced nature of the Level VII Palace is due to previous construction
projects that altered the topography.
Signs of burning were also encountered in Square 33.53; yet, due to the extreme width of the mudbrick walls
exposed, the spaces left for investigating living surfaces were minimal (Fig. 7-6). Nevertheless, the square provided an
understanding of the general layout of the previous palace. The northwest-southeast wall likely functioned as the
exterior wall, since trash deposits were revealed to the west in both squares. Of particular importance, this trash
deposit revealed a Middle Minoan IIB-IIIA Kamares Ware fragment, indicating that Alalakh was directly or indirectly
involved in a network of exchange that included the Aegean world (Koehl 2020: 204, fig. 2).
In relation to the later Level VII Palace, a construction deposit (Local Phase 2) was documented in both squares.
While the leveling fill was relatively homogenous in Square 33.32, the rubbish fill deposit in Square 33.53 revealed
the presence of various objects, including ivory inlays, composite statue fragments, and non-joining wall painting
fragments. Sealed by the thick concrete-like floor of Level VII, there is no doubt that these fragments belong to a
context earlier than the burnt Level VII Palace. The stratigraphic and chronological significance of these findings will
be discussed elsewhere in detail in future publications.
Cleaning operations conducted in the Level VII Palace revealed further interesting artifacts that were not registered
by Woolley for almost a century. These include the recovery of a tablet documenting a witness list (Lauinger 2014),
vitrified limestone fragments of a winged sculpture (Yener 2013b), and, recently, the fragment of a BIBRU in central
Anatolian style, likely depicting a bull.
The Earl~r MBA .

Palace Walls
Figure 7-6. Aerial view of Square 33.53 (©Alalakh Excavations Archive, photo by M. Akar).
New Wall Painting Fragments from the 2018-2019 Seasons
Our soundings during the 2015-2019 seasons in the Alalakh Level VII Palace also unearthed more than 70 high-
quality new wall painting fragments from MB II contexts in Square 33 .53. Radiocarbon analyses have dated the
fragments to be within a span from 1780-1680 BC (Yener 2021; cf. Fig. 7-15). Most were burnished and painted, and
some had figurative patterns, although the designs cannot as yet be reconstructed. In total, the fragments themselves
vary in size and quality of preservation, yet this small assemblage adds valuable information about the different phases
of the palace decorated with frescoes, the chronology of the palace, and the techniques used in the plaster fragments
themselves.
Wall painting fragments at Alalakh were first discovered by the previous excavator, Sir Leonard Woolley (1955:
228) in the Level VII Palace and have often been considered as examples of the earliest so-called fresco-secco
techniques in western Asia. Re-analyzing some of these fragments in the Ashmolean Museum, Brysbaert (2002) has
described their fresco techniques as the presence of fingernail impressions, particular tool and brush marks, the use or
absence of specific pigments, and the use of lime plaster. This also confirms the original Woolley analysis (Barker
1955), which had concluded that the wall frescos were painted while the plaster was still wet. All of these attributes
have been seen on the new fragments as well.
The wall painting fragments from Woolley' s excavation have gained special attention due to the continuing
discussion about the technique's assumed origins in the Aegean (Niemeier and Niemeier 2000: 780-781) and the
question of traveling craftspeople in the eastern Mediterranean serving as the vehicle of transmission for the
iconography (Niemeier and Niemeier 2000; von Ruden 2020; Koehl 2013, 2020). Yener (2021) has recently argued
that the canonical "Aegean" griffin motif at Alalakh was thus attributed based on mistaken interpretations. Newly
accessed UCL Library Special Collections archival photos argue against this reconstruction and suggest rather an
image of a winged deity, reflecting its ancient Near Eastern roots. Furthermore, according to Woolley, the Level VII
Palace is to be dated over 100-150 years older than the Knossos paintings, making it difficult to assume influence
from the west; he pointed instead to north Syria as the place from which the tradition of wall painting emerged.
Usually overlooked, it is important to note that fresco fragments were found in earlier Level IX (dated to the 19th
century BC) and Level VIII palace contexts as well. Woolley (1955: 31-32) notes that the red and white painted wall
plaster fragments came from the floor of Level IX Building A and were of "Pompeian" red color, although a pattern
could not be distinguished.
The first new fresco fragments emerged in the palace construction fill (Local Phase 2, see above). To judge by the
thickness of the painted lines and the great precision of the brush strokes in the application of the paint, the artistic
skill of the painters demonstrates a good control of the brush. Only three examples of the new wall painting fragments
are described here.
Figure 7-7. AT26395, wall painting fragment from Figure 7-8. AT26620, wall painting fragment from
Square 33.53, Local Phase 2 (©Alalakh Excavations Square 33.53, Local Phase 2 (©Alalakh Excavations
Archive, photo by M. Akar). Archive, photo by M. Akar).
AT26395 (Fig. 7-7) is a large polychrome fragment from Square 33.53. Various pigments were preserved, such as
dark blue, dark yellow, dark red, black, and cream, on polished plaster. Two red bands are seen on the upper and lower
part of the fragment and are delineated with string marks. Clearly visible are finger smudges on the bottom band. One
dark blue triangle and one dark blue stump emanate vertically from the upper red band. A dark yellow, curvilinear s-
shape, whose vertical appendage is overpainted by the red band, has a black tapering brush stroke, which runs parallel
to it. Portable x-ray fluorescence (pXRF) 2 analysis revealed that cobalt and iron minerals were used as pigments.
Significantly, the dark blue triangular design was made with a cobalt-rich pigment. Preliminary analysis revealed that
the dark red color is iron, while the dark yellow ochre color was created by using crushed glass pigmented by iron
oxide.
AT26620 (Fig. 7-8) from Square 33.53 is a fragment with two black parallel lines with a curvilinear motif in black
framing a faded light grayish-blue interior. Originally the center was a much more vibrant blue color, given the
Figure 7-9. AT26367, wall painting fragments from Square 33.53, Local Phase 2
2
XRF, Oxford Instruments XMET 5100; measurement time was at 120sec on mining_le_fp with 25sec at 45KeV; remainder at
13KeV.
copper content revealed in the pXRF analysis. Iron oxide/ochre and the black border pigments contained iron. The
lower design may depict the petal of a flower.
AT26367 (Fig. 7-9) from Square 33.53 consists of several smaller fragments painted with a dark red pigment over
a cream-colored white background. Cross-hatched darker, black-brown patterns may portray mudbricks stacked like a
wall, although the architectonic staggering in mudbrick depictions in Aegean paintings is missing. According to the
size of the designs, this appears to be a miniature fresco.
Area 3, Squares 45.44 and 45.45
Area 3 was opened initially to explore the fortification system on the northeastern slope of the mound (Y ener and
Yaz1c10glu 2010: 24), and four squares have been explored here to date (Squares 45.44, 45.45, 45.71, and 45.72).
These are the earliest phases thus far excavated, and are preliminarily dated to the late MBA: Local Phase 4 in Squares
45.71 and 45.72, which consists of a casemate fortification wall with five transitional MB II-LB I graves dug into it
(Yener and Yaz1c10glu 2010: 25-26); Local Phase 5 in Squares 45.44 and 45.45, where the casemate fortification wall
has also been identified, along with two rooms of a likely domestic structure to the west (Ingman 2014: 62-63, 2017).
Five graves have been found in the rooms west of the wall, with more burials placed to its east in the cemetery area
(Fig. 7-10). While this extramural cemetery in Squares 45.71 and 45.72 is interspersed with phases of architecture (the
fortification wall in Local Phase 4 and a kitchen/workshop complex in Local Phase 2; Yener and Yaz1c10glu 2010: 24-
29), in Squares 45.44 and 45.45 it is placed directly against the eastern, outer side of the fortification wall and consists
of an open space with no architecture. This has been extensively subjected to post-depositional disturbances,
particularly slope wash, erosion, and modem farming activities (Ingman 2017: 249). Preliminary dating of the burials
in Squares 45.44 and 45.45, therefore, was accomplished with the use of a 3D GIS model that used identified fill
layers within the cemetery area and reference points from the wall itself, such as dateable foundations, to propose a
phasing sequence of the cemetery in these squares. This placed the majority of excavated graves within the LB I, with
the earliest burials dating to the late MBA (Ingman 2017). However, recent radiocarbon dates from a sample of the
individuals buried in the cemetery suggest that burial in the cemetery began much earlier, perhaps as early as the MB
I, and that the majority of these graves may in fact date to the MB II (Ingman et al. 2021; Skourtanioti et al. 2020).
Further analysis is needed to resolve this question, and a conservative approach is therefore taken here, with
discussion only of those burials which can be confidently assigned to the MBA by both stratigraphic and radiocarbon
data.
Figure 7-10. Domestic quarters, MBA fortification wall, and the slope cemetery in Area 3, Square 45.44
Tell Atchana has one of the largest 2nd millennium BC burial assemblages in the region, with 342 graves
documented to date (Ingman 2020a). Although the majority of these have been dated to the Late Bronze Age, 31 have
been identified as belonging to the MBA, most of which were found in an extramural cemetery in Area 3, where 134
burials have been excavated to date (Ingman 2017, 2020a). The MBA burials are mostly single, simple pit graves
(n=26), which dominate the assemblage at the site in all areas and time periods (67% of the total burials), as well as
multiple, simple pit burials (n=3), one infant pot burial, and one constructed tomb, the so-called royal hypogeum
found by Woolley within the Level VII Palace (Ingman 2020a; Woolley 1955: 95-97). Sixteen of the burials were
found inside the city; those discovered by Woolley were under floors in the Royal Precinct, in Site H on the southern
slope, and in his exploratory Trench A (Woolley 1955: 221-222), and those found by Yener were in Square 45.44 to
the west of the fortification wall, including four inside the two room contexts that back up against the city wall; one
was dug into the foundations of the city wall (Ingman 2014: 63). The remaining 15 MBA graves were found in the
extramural cemetery, making the intra-/extramural distribution in this period nearly a 50% split.
Figure 7-11. A selection of burial vessels from the Area 3 cemetery

Grave goods are common in the MBA burials (Fig. 7-11 ), with 71 % of the graves including at least one object,
although most have only one (32.3%) or two (22.6%) objects; "rich'' graves are rare (Ingman 2020a: 172). As is true
for MBA and LB I graves generally, pottery makes up the majority of grave goods in the MBA burials (Ingman
2020a: 168-169), which comprises just over two-thirds of the grave goods assemblage (67.8%). Piriformjuglets (Fig.
7-12.26) are the most common shape (n=IO; 25% of the pottery), followed by shoulder goblets (n=6; 15% of the
pottery, Fig. 7-12.14) and short-neck jars (n=6; 15% of the pottery, Fig. 7-12.24). These three shapes are almost
exclusively associated with graves at Tell Atchana, leading to their identification as "funerary vessels" (Ingman
2020b), and together they make up close to 40% (37.3%) of the total grave goods in the MBA (Ingman 2020a: fig.
4.22). Most of the rest of the grave goods in this period consist of personal adornments (18.6%) such as pins, rings,
and beaded jewelry, made of both metal (mainly copper alloys, though an infant female [LI 52] was also found with a
silver ring and a piece oflead wire) and stone or vitreous materials.
The Late MBA Ceramics
The ceramic repertoire of Periods 9, 8, and 7 is fairly homogeneous, with some contextual differences regarding
the appreciation and consumption of different ware and shape types. The ceramics were predominantly manufactured
by the use of a rotary kinetic energy (hereafter RKE, Roux 2019: 54). Although it is not always possible to reconstruct
the fashioning techniques macroscopically since their traces are concealed by finishing techniques, traces of wheel-
throwing can be seen on some of the small and medium-sized vessels. Larger vessels, on the other hand, were likely
produced on a turntable. Hand modeled ceramics that were manufactured without RKE are rarely encountered and are
mainly limited to a few body sherds that are likely non-local (Bulu 2016: 305), whereas very large pithoid jars and
some of the Syro-Cilician Ware vessels were produced with the combination of both fashioning techniques (Bulu
2021: 217-219). The vast majority of the ceramics were wet-smoothed, and further surface treatments are less
frequent. Burnishing with and without the use of RKE is occasionally encountered in various ware types, but it is the
associated characteristic of Gray Burnished Ware. Among the decorative techniques, simple line or comb incisions are
mainly applied to jars and kraters (Fig. 7-13.1-2, 7-8). Although attested in rare cases, potter's marks applied as
incisions prior to the firing of the vessel are sometimes found on bowls and jars (Fig. 7-12.4). Added rope decoration
to the shoulder (sometimes also to the lower part) is almost exclusively seen on cooking pots and pithoid jars (Fig. 7-
13.13, 15). Application of painting appears as two main traditions: Syro-Cilician Ware and Banded Ware. The less
frequent third tradition is the Sgraffiato Ware, which combines the application of paint and incisions. All ware types
have medium- to hard-fired fabrics; low-fired fabrics are attested only by some of the trays.
3
z ---------<
J
5
7
4
2
,
c::; ,_7 \\Y ! l' 6
7 9
/
~
~
I
.7 \ 7~
12
8 10
~
13
7
C
\ 11
[S ED
14 16
c:::::~----------1
~ -7 rT) ~ l
19 15 17
/ j!i
20
-~
--- 10cm
18
'
e====~:L
21
23
22
24 25
CD cD26 27 28
Figure 7-12. Selected late MBA ceramics. l-AT19010.14 (SCW), 2-AT23311 (SW): s-curve bowls; 3-AT24347.l, 4-
25415.17: SW hook-rimmed bowls; 5-AT25423.l, 6-AT25409.9: SW carinated bowls; 7-AT25083.l, 8-AT27108.21:
SW rounded shallow bowls; 9-AT25411.3: SW strainer bowl; 10-AT25093.l, ll-AT25409.2: SW hemispherical
bowls; 12-AT25091.6, 13-AT26998.l: SW lids; 14-AT18961: SF shoulder goblet; 15-AT25490.3: SF s-curve cup; 16-
AT19261: SW biconical cup; 17-AT25096.2 (SF), 18-AT26501.l (SG): hemispherical cups; 19-AT25475.2, 20-
AT25496.2: CWO high footed bowls; 21-AT25409.7, 22-AT23664.l: MI trays; 23-AT17665.l: SCW closed shape;
24-AT15712: SW short-neck jar; 25-AT14646: SF biconicaljar; 26-AT14699: GB piriformjuglet; 27-AT15754, 28-
AT19263: SW globular juglets (©Alalakh Excavations Archive).
-) I):)
,,
,':
~}
3
=~~~~===~~~o~=~=
2
)T~ 4
5
D 6
- ------ -- - - - - -
"" "' .,.,, <a,C::, ... "' "' ">- '>C::, ... ""., .,, ~
~
7
IT\ 12
10
) ~~ 11
13
-----
14 0 --- 20cm 15
Figure 7-13. Selected late MBA ceramics. 1-AT9257.3 (SW), 2-AT4247.1 (SW), 3-AT6124.3 (SCW): kraters; 4-
AT25423.2: SW jug; 5-AT19024.2 (SCW), 6-AT12719 (SW): pitchers; 7-AT8381, 8-AT23669, 9-AT25415.7: SW
globular jars; 10-AT18356: BW globular jar; 11-AT27124.1: BW high-neck jar; 12-AT8384, 13-AT23349: MI
cooking pots; 14-ATI 1792: SW globular pithoid jar; 15-ATI 1765: SW hemispherical pithoid jar (©Alalakh
Excavations Archive).
The ware and shape typology of LB II ceramics of Tell Atchana have been extensively studied and published
(Horowitz 2019), whereas those of LB I and MBA are still ongoing and have only partially been published (Horowitz
2015, 2017; Bulu 2016). Different ware types of the MBA and LBA ceramics that have been macroscopically
classified as local products have also been confirmed through the implementation of Neutron Activation Analysis
(NAA) and ceramic petrography, demonstrating that they were produced with the exploitation of locally available
calcareous clays and tempering agents at Tell Atchana (Bulu 2012, 2021; Gutsuz et al. 2017); experimental studies
support this (Morrison and Horowitz 2016).
Some ware types have a very long tradition at the site, first appearing during the MBA and continuing throughout
the 2nd millennium BC without any major differences in the fabric characteristics. The prevailing ware type among
them is Simple Ware (SW), which is characterized by a semi-fine to medium-coarse calcareous fabric with river sand
added as temper. The MBA shape types that appear in Simple Ware are numerous. Among the open shapes, the s-
curve bowl (Fig. 7-12.2) is the prevailing bowl type that appears in all MBA contexts. The second most frequent bowl
type is the hook-rimmed bowl (Fig. 7-12.3, 4). Other bowl types that appear in far fewer numbers are rounded shallow
bowls (Fig. 7-12.7, 8), hemispherical bowls (Fig. 7-12.10, 11), and carinated bowls (Fig. 7-12.5, 6). In comparison to
bowls, cups are much less frequent; the retrieved examples ares-curve (Fig. 7-12.15), biconical (Fig. 7-12.16), and
hemispherical (Fig. 7-12.17) types. Constituting a rather intermediate shape between open and closed vessels, kraters
(Fig. 7-13.1, 2), and especially the biconical subtype with a distinct carination on the shoulder, are also commonly
attested in all MBA contexts. Among the closed shapes, the prevailing jar type is the globular jar; while medium-
sized, wide-mouthed (Fig. 7-13.7) and narrow-mouthed (Fig. 7-13.8) globular jars are the most frequently attested
ones, small-sized globular jars and those with an ovoid body (Fig. 7-13.9) are much rarer subtypes. The other jar types
are high-neck jars, as well as the small-sized short-neck jars (Fig. 7-12.24) and biconicaljars (Fig. 7-12.25). Examples
of jugs (Fig. 7-13.4), pitchers (Fig. 7-13.6), and juglets (Fig. 7-12.27, 28) have also been retrieved. Lastly, the storage
related large pithoid jars were also manufactured with Simple Ware fabric. Hemispherical pithoid jars with a rail rim
(Fig. 7-13.15) are very typical of the MBA, but globular pithoid jars (Fig. 7-13.14) are also commonly found. On
some examples of the latter type, the base has a single perforation, which might be associated with cheese making or
beer brewing (Ellison 1984; Gates 1988: 68). Within the shape repertoire of Simple Ware, rare shapes such as lids
(Fig. 7-12.12-13) and strainer bowls (Fig. 7-12.9) are also attested.
The very fine, likely levigated variant of Simple Ware is classified as Fine Simple Ware (SF), which was mainly
used to manufacture thin-walled bowls, cups, shoulder goblets (Fig. 7-12.14), or small-sized jars. Painted pottery of
the MBA is also classified as subvariants of Simple Ware, since it shares the same properties regarding the paste
preparation, fashioning, and finishing techniques, as well as firing (Horowitz 2015: 165). The most characteristic
painted pottery of the MBA contexts is Syro-Cilician Ware (SCW). It is characterized by its distinct geometric, floral,
figural, and animal motifs that are applied to specific vessel shape types, and it constitutes one of the materialized
reflections of early interregional connectivity between different and distant regions during the MBA (see Seton-
Williams 1953; Tubb 1983; Bagh 2003, 2013; Bulu 2017b). The fabrics of the Syro-Cilician Ware vessels range from
very fine to medium-coarse. Although the vast majority of the Syro-Cilician Ware vessels were manufactured with the
use ofRKE, recent technological analysis results showed that some of the pitchers and kraters were manufactured with
a combination of hand modeling and RKE (Bulu 2021 ). The most commonly attested shape types are pitchers with the
typical "eye" decoration (Fig. 7-13.5), kraters (Fig. 7-13.3), ands-curve bowls (Fig. 7-12.1), but examples of carinated
bowls, rounded shallow bowls, cups, globular jars, short-neck jars, juglets, side-spouted jars, and animal-shaped
vessels are also attested.
The second painted pottery tradition is Banded Ware (BW). In contrast to the LBA examples that are characterized
by broad horizontal bands (Horowitz 2019: 197), the MBA examples are typically decorated with either simple
horizontal lines or with an arrangement of thicker bands that are framed by thinner lines. Complete/partially complete
examples and diagnostic sherds of Banded Ware vessels are very rare, and they are mostly represented by globular jars
(Fig. 7-13.10), high-neck jars (Fig. 7-13.11), and short-neck jars, although fragments that belong to cups and shallow
bowls are also found. The last and least common painted pottery tradition of the MBA is Sgraffiato Ware (SG), the
typical vessel shape of which is the hemispherical cup (Fig. 7-12.18). In this style, thin, straight, and wavy lines were
horizontally scratched on the broad painted bands, revealing the fabric color underneath.
The other MBA ware types with specific surface treatments are Gray Burnished Ware (GB) and Red Slip Ware
(RS). Gray Burnished Ware has a very fine to semi-fine fabric with no or very rare visible inclusions. The surfaces of
the vessels were highly burnished, and the dark gray color was acquired through reducing atmospheric conditions
during the firing of the vessels. The most typical vessel shape types are the hook-rimmed bowl, often with one or more
incised lines on top of the rim, and piriform juglets (Fig. 7-12.26). Red Slip Ware is one of the least common ware
types during the MBA. Although the majority of the retrieved examples are non-diagnostic body sherds of open and
closed vessels, there are also diagnostic sherds that belong to shallow bowls, biconical cups, and globular jars.
Cookware assemblages of MBA Tell Atchana consist of two distinct ware types. The prevailing one is calcite
tempered (MI), which constitutes a chronological marker for MBA, as opposed to the shell tempered one that is
characteristic of the LBA (Horowitz 2019; Horowitz and <;akular 2017). The most common shape type is the wide-
mouthed cooking pot with a globular or carinated body (Fig. 7-13.12, 13); two opposing handles added to the rim is a
frequently attested characteristic. Calcite tempered trays (Fig. 7-12.21, 22) are also very typical of the MBA, with the
top of the rim mainly adorned with incised rope decoration or with a thick groove. The second but less frequent
cookware is grit tempered (CWO), which is mainly found as fragments of shallow bowls (Fig. 7-12.19- 20) or body
sherds of cooking pots. The high foot fragments that are found in this fabric indicate that the bowl fragments likely
belong to high-footed bowls that were previously named "champagne-cups" by Woolley (1955: 329, Type 113).
The overall late MBA local ceramic assemblage from Tell Atchana finds parallels in various sites in a wider
geographic setting. These include Ebla/Tell Mardikh, Hama (Nigro 2002a, 2002b ), and Qatna/Tell Mishrifeh (Iamoni
2012) in western Syria, Umm el-Marra (Curvers et al. 1997: figs. 23-24, 20.9-10; Schwartz et al. 2000: fig. 8, 2003:
figs. 29-30, 33), and Tell Hadidi (Dornemann 1992, 1979: figs. 20-23) in the Syrian part of the middle Euphrates,
Oylum Hoyiik in the Kilis Plain (Ozgen and Helwing 2001: abb. 7, 16-19; Engin 2020: fig. 11), and Zincirli in the
Karasu Valley (Morgan and Soldi 2021: figs. 10, 15, 19-22), as well as Tarsus Gozlukule (Goldman 1956: pls. 287,
291,293,295, 297-298, 368-370, 372) and Kinet Hoyuk (Gates 2000, 2010: fig. 8, 2011: fig. 10, 12) in Cilicia.
Tfi:'i:-----r--~~lfJlr--J
;------ - "\\
//
//
0 5cm
Figure 7-14. AT25415.1, Kamares Ware fragment from Square 33.53, Local Phase 3
(©Alalakh Excavations Archive).
Among the non-local ceramics retrieved from the late MBA contexts, the recovery of a Middle Minoan IIB-IIIA
Kamares Ware hemispherical cup fragment (Fig. 7-14) is intriguing for a number ofreasons, not only because it is the
very first example retrieved throughout the excavation history at the site, but also since its presence in the broader
eastern Mediterranean setting is limited to a few sites (Koehl 2020: 203, fig. 2a-b; Pucci et al. 2020: 152).
Furthermore, while the earliest contacts between Alalakh and Crete have mainly been attributed to the technical and
iconographic similarities between the wall paintings from the Level VII Palace and those of the Minoan Palaces ( see
the discussion on this topic in the wall paintings section here), the recovery of a Kamares Ware cup from a context
predating the latest phase of the Level VII Palace testifies to the presence of interactions between Alalakh and the
Aegean much earlier. The late MBA contexts also yielded several other body sherds with likely non-local fabrics,
which include bichrome slipped and bichrome painted examples that are not known among typical local painted and
slipped wares at the site. Future archaeometric research will be conducted on the non-local fabrics to shed light on the
extent of these external contacts.
OxCal v4.4 .3 Bronk Ramse 2021 · r:5 Atmos heric data from Reimer et al 2020
The Level V II Palace

Construction Deposi t
The Apsidal Building
R_ Date AT13 52
R_Date AT11 54
The Earlier MBA Palace
R_Date AT11 56
2400 2200 2000 1800 1600
Calibrated date (calBC)
Figure 7-15. MBA radiocarbon dates from Tell Atchana. AT26099 (TUBiTAK MAM YDBE AMS Laboratory;
TUBiTAK-0491) and AT26083 (TUBiTAK-0492) are from Square 33.53, Local Phase 2, Level VII Palace
construction deposit. AT12888 (Beta Analytic; BETA-409551) is from a room context and AT13752 (BETA-409550)
is a short-lived sample from a mixed street deposit in Square 32.57, Local Phase 5b, apsidal building. ATl 1754
(BETA-409552) and A Tl 1756 (BETA-409553) are from Square 33.32, Local Phase 3c, pre-Level VII palace. The
raw dates were calibrated using OxCal v.4.4.3 software (Bronk Ramsey 2020) based on the IntCal20 atmospheric
curve (Reimer et al. 2020).
DISCUSSION
The urbanized, complex nature of the city during the MBA is well defined through all excavation areas, revealing
the presence of a long-lasting, strong administrative system capable of conducting large-scale building programs from
palaces to temples and city fortifications. This is partially defined through deep soundings, providing a glimpse of the
arrangement of the space and the city's public monuments. The C14 dates acquired from these distinct building phases
provide a temporal framework that stretches across a wide span due to continuous and long-lasting usage of the
building phases discussed (Fig. 7-15). Unfortunately, owing to the massive size of the buildings exposed, squares were
often mostly occupied by thick mudbrick walls, leaving limited space for exploration of the living surfaces.
Continuous re-floorings and the absence of stone foundations in most cases further complicate the matter.
Nevertheless, several important points can be presented regarding the exposed MBA contexts and their significance in
the wider Anatolian, Near Eastern, and eastern Mediterranean setting.
With regards to the Apsidal Building exposed in Square 32.57, the rectangular buildings with curvilinear walls are
a common feature of the Aegean and western Anatolian Early Bronze Age, though at a much smaller scale; this
tradition is minimally noted in central and southeastern Anatolia and eastward (Yener 2015). The multi-functional
buildings of utilitarian nature such as the Early Bronze Age Rounded Building (Level 4) at Tell al Raqa'i, along the
middle 1Jabur River, is an example of the architectural traditions of a northern Mesopotamian rural community where
the structure was dedicated to processing, storing, and administrative management of grain supply (Schwartz and
Curvers 1992). In the upper parts of the JJabur, the remarkable monumental underground stone structure exposed in
close proximity to the palace of Tupkis in the city of Tell Mozan, Urkes (ca. 2300 BC) is another unique architectural
example that may perhaps signal Hurrian construction practices in cult and ritual buildings in the late Early Bronze
Age. According to its excavators, the stone-covered underground structure (abi) served as a sacred space for the spirits
of the netherworld in accordance with the Hurrian rituals known from later Hittite sources (Buccellati and Kelly-
Buccellati 2007: 142). Furthermore, recent research (Yener et al. 2020) suggests that the complete plan of the Apsidal
Building at Alalakh may resemble the later 3rd millennium BC in antis plan of Temple C at the middle Euphrates site
of Tel Bi 'a, Tuttul (Miglus and Strommenger 2002: 102, Taf. 124). All these examples may perhaps then hint at
connections to the east, in accordance with the Hurrian and Amorite practices that intrude into the Amuq Valley at the
beginning of the 2nd millennium BC, also suggested through recent excavations at Toprakhisar Hoyiik (Akar and Kara
2020). Thus, speculatively, the presence of an apsidal building in the late MBA at Alalakh may represent temple
building practices with northern Mesopotamian origins. This is in accordance with the limited number of small finds
retrieved from the building. The presence of a particular mold-made female figurine with a tall headdress is regarded
as representing an early Hurrian-style iconographic predecessor to LBA examples from Anatolia (Yener 2015: 489).
The strong connections to northern Mesopotamian late 3rd millennium BC architecture at Tell Atchana may then be
linked to the 4.2 K BP climate event leading to habitat tracking groups, including Hurrians and Amorites (Weiss 2014;
Burke 2017: 296; Akar and Kara 2020). Such a statement also finds support in the new genomic research conducted
on the individuals from the MBA cemetery at Tell Atchana, which points to the presence of genetic contributions to
the population that could have northern Mesopotamian origins (Skourtanioti et al. 2020: 1168).
The funerary practices at Tell Atchana in the MBA also fit within a larger picture of Near Eastern ritual practices
in this period, showing connections to regions both near and farther flung. Although the collective rock-cut and
residential tombs that appear commonly in both the Levant and Mesopotamia, respectively ( e.g., Cradic 2017; Gonen
1992; Hallote 1995; Keswani 2012; Laneri 2014; Morandi Bonaccosi 2011), are not found at Tell Atchana, burials
from other sites in northern/northwestern Syria are very similar in terms of locations, types, and grave goods. At sites
such as Oylum Hoyiik (Engin 2020), Tell Afis (Di Michele and Pedrosi 2012), Tell Tuqan (Ascalone 2014), Tilmen
Hoyiik (Duru 2003), and Tell Leilan (Weiss et al. 1990), simple pit graves have been recovered in association with
city fortifications, as seen in the extramural cemetery at Tell Atchana, and all of these graves have a similar range of
grave goods and resemble each other in terms of body and object positionings. This association between pit graves and
fortifications begins in the early MBA (Felli 2012) and seems to represent a uniquely northern Syrian phenomenon
that may reflect the region's intermediary position-both geographical and perhaps cultural-between the Levant and
Mesopotamia on the one hand, where intramural burials were the general rule of the time, and Anatolia on the other,
where a strong tradition of extramural and off-site burial prevailed for much of the Bronze Age (e.g., Akyurt 1998;
Bachhuber 2015; Emre 1991).
There are also indications at Tell Atchana that post-funerary rituals may have been carried out for the dead similar
to the kispum rituals that are well-known (and textually attested) from sites like Mari (e.g., Barrett 2007; Jacquet 2012;
Jonker 1995; MacDougal 2014; Pfalzner 2015; Pitard 1996; Salles 1995). Texts describe these rituals as dedicated to
caring for and remembering the dead in the form of continued provisioning with food and drink at regular, designated
festivals, and although most of the texts describe these practices in a royal context, there is growing archaeological
evidence that corresponds to the performance of similar rituals in non-royal contexts, such as vessels and animal bones
deposited in the fills of tomb shafts and pits, as well as depressions and installations that seem to have been designed
for pouring libations (Horwitz 2001; Keswani 2012: 187; Lange 2012; Wissing 2012; Wygnanska 2014: 46--48).
Despite the often highly disturbed nature of the extramural cemetery fill, there are indications that funerary feasts
and/or post-funerary offerings may have been conducted here, particularly vessel shapes such ass-curve bowls that do
not appear in the grave good assemblage at Tell Atchana but which are common in MBA graves at nearby sites, such
as Qatna (Morandi Bonacossi 2011; Morandi Bonacossi et al. 2009), Tell Tuqan (Ascalone 2014), and Ebla (Matthiae
1984), as well as farther south in the Levant (Baker 2012: 92- 94). Fragments ofbasins, two examples of pot stands (or
perhaps portable hearths?) and a box-shaped, burnished vessel, all found in the cemetery fill, may also indicate the
practice of offering ceremonies or funerary feasts taking place here (for more details, see Ingman 2020a: 148-157).
Perhaps the most intriguing evidence for such practices at Tell Atchana comes from the Level VII Palace and the Shaft
Grave, where the surrounding rooms of the palace contained funerary vessels (Woolley 1948: 16, 1955: 95-104),
indicating that these rooms may have served as a locus for the performance of funerary rituals (Ingman 2020a: 55-64).
The discovery of well-preserved fresco fragments in Square 33.53 in the southern sector of the Level VII palace
has now begun to shed light on the question of early 2nd millennium BC wall painting practices in the eastern
Mediterranean and their chronological relevance to similar practices in the Aegean world. It is well known that the
palette of red, black, and white wall painting over plaster has a very early appearance dating back to the Aceramic
Neolithic and Chalcolithic in Anatolia, the Levant, Mesopotamia, and inner Syria. This stylistic tradition continued
into the 3rd millennium BC as evidenced by tri-color figurative and geometric wall paintings at Tell el-Halawa B, Tell
Mumbaqat, and Tell es-Sweyhat (Dunham 1993 and references), as well as at Ebla (Pinnock 2019: fig. 16).
An antecedent to the MB II multi-colored frescos at Alalakh is the use of polychromatic pigments on wall
paintings (especially blues and greens) at Tell Burak (Kamlah and Sader 2010: pl. 115) dated to the 19th century BC.
These "proto-frescos" are described as having Egyptian stylistic influences as well as underscoring the technique of
preliminary paintings applied on wet plaster and final paintings applied on secco.
Polychrome paintings at Mari make their appearance in the 19th century BC with the so-called investiture scene,
given its blue sky, the multi-colored sphinxes, and the blue bird with the drooping lotus flower (Parrot 1958: pl. A;
Nunn 1988; Nunn and Piening 2020). It is described as tempera on a mud wall and partly white-washed lime gypsum
plaster. The emphasis scholars place on the use of lime plaster for frescos is misplaced since at issue is local geology.
That is, the painters and preparers of the walls used whatever was available in their own region. At Alalakh, the region
was filled with limestone, and therefore lime plaster was used; at Mari, gypsum was plentiful in the region, and thus it
became the background of preference. Furthermore, enhancing the relevance of Mari to Alalakh, in her examination of
Mari wall paintings, Muller (2018: fig. 15) has found that the blue pigment at Mari was crushed vitrified Egyptian
blue faience, imitating lapis lazuli. This use of faience, mimicking the color blue, mirrors the crushed glass and faience
applied on the Alalakh paintings. With reflected light, what a dazzling, sparkling Level VII Palace it must have been.
By the early 2nd millennium BC, polychrome wall painting was well established in western Asia. We would submit
that, with the iconographic preference at Alalakh for local and Egyptian themes so prevalent in the Level VII seals, as
well as the tradition of polychrome Ancient Near Eastern fresco motifs, Alalakh identified predominantly with
regional styles in general, and Egypt and Mari in particular, more than with the Aegean.
In terms of ceramic evidence, although the late MBA corpus is characterized as a homogeneous assemblage as a
whole, contextual analysis conducted in Areas 1 and 3 indicates that there were some differences in the appreciation
and consumption of some shape and ware types within and between the two areas. For instance, while the s-curve
bowl is the prevailing bowl type attested in Squares 32.57 and 33.32, hook-rimmed bowls, as well as rounded shallow
bowls, are by far the most frequently encountered bowl types in Square 33.53. Local Phase 5 of Square 45.44 in Area
3 is a rather small MBA exposure, but, while the hook-rimmed bowls are completely absent from its domestic
contexts, rounded shallow bowls outnumber s-curve bowls. These results would indicate different eating and/or
serving-related traditions seen within different sectors of the Royal Precinct, as well as between different areas of the
settlement during the MBA. A similar distinction can also be seen in the distribution of painted pottery styles. The
predominant painted pottery style retrieved from Squares 33.53 in Area 1 and 45.44 in Area 3 is Banded Ware, as
opposed to the prevailing Syro-Cilician Ware assemblages retrieved from Squares 33.32 and 32.57 (Bulu 2021: 269-
270, 275). In addition to the typical Syro-Cilician Ware examples, a closed vessel sherd depicting a defecating equid
from Square 45.44 clearly constitutes a unique example (Fig. 7-12.23). The band decoration below resembles the
framing pattern of the Syro-Cilician Ware examples, but the animal motif, and its posture, is extraordinary. This
phenomenon may suggest that they could have been trying to imitate the Syro-Cilician Ware vessels used in the Royal
Precinct in a rather unusual way, or it may reflect different painted pottery production and/or consumption practices
applied outside the Royal Precinct (Bulu 2021: 230).
CONCLUSIONS
After two decades of new archaeological fieldwork at Tell Atchana, the political, economic, and cultural
complexity of the Kingdom of Mukis and its capital city Alalakh has been broadly investigated and intensively
published for the Late Bronze Age (see Yener et al. 2019). The MBA, on the other hand, remains to be further
investigated via the generation of new data from various districts of the tell and through the re-evaluation of the data
collection from the 1930s- 1940s excavations. This short chapter presents only a glimpse of the ongoing research into
the MBA, revealing the urbanized and cosmopolitan nature of a city located in the borderlands. The research at
Alalakh has also proven the far-reaching potential of the re-evaluation of old excavation data for redefining commonly
accepted and well-rooted assumptions such as the origins of wall painting practices in the Anatolian and eastern
Mediterranean world.
ACKNOWLEDGEMENTS
The Tell Atchana, Alalakh Excavations were funded by the Turkish Ministry of Culture and Tourism, INSTAP
(Institute for Aegean Prehistory), FAVAE (Fund for Amuq Valley Archaeological Expeditions), and Hatay Mustafa
Kemal University (BAP #21.A.003). The archaeometric research is funded by TUBiTAK (Project #l 14K766). The
recent geoarchaeological research is funded by AKMED (Project #2018/P.1019) and HMKU Scientific Research
Project Grant (BAP #18. M. 018). The authors thank all members of the Tell Atchana excavation team.
REFERENCES CITED
Akar, Murat and Demet Kara. 2020. The Formation of Collective, Political and Cultural Memory in the Middle
Bronze Age: Foundation and Termination Rituals at Toprakhisar Hoyillc. Anatolian Studies 70: 77- 103.
Akyurt, i. Metin. 1998. MO. 2. Binde Anadolu'da Olil Gomme Adetleri. Bestattungssitten Anatoliens im zweiten
vorchristlichen Jahrtausend (Zusammenfassung). Ankara: Turk Tarih Kurumu Bas1mevi.
Archi, Alfonso. 2006. Alalakh al Tempo del Regno di Ebla. In Tra Oriente e Occidente: Studi in onore di Elena Di
Filippo Balestrazzi, D. Morandi Bonacossi, E. Rova, F. Veronese, and P. Zanovello, eds., 3-5. Padova: Sargon.
- . 2020. Linguistic and Political Borders in the Period of the Ebla Archives. In Alalakh and its Neighbors:
Proceedings of the 15th Anniversary Symposium at the NfnV Hatay Archaeology Museum, June 10-12, 2015, K.A.
Yener and T. Ingman, eds., 31-40. Leuven: Peeters.
Ascalone, Emico. 2014. The Long Life of the Dead. A Middle Bronze IB Necropolis at Tell Tuqan, Syria. In Tell
Tuqan Excavations and Regional Perspectives: Cultural Developments in Inner Syria from the Early Bronze Age
to the Persian/Hellenistic Period, F. Baffi, R. Fiorentino, and L. Peyronel, eds., 189-225. Salento: Universita del
Salento.
Astour, Michael C. 1992. Alalakh. In Anchor Bible Dictionary, I., David Noel Freedman, ed., 142- 145. New York:
Doubleday.
Av~r, Ula~, Murat Akar, and Charlotte Pearson. 2019. Geoarchaeological Investigations in the Amuq Valley of
Hatay: Sediment Coring Project in the Environs of Tell Atchana. In The Proceedings and Abstracts Book, 72nd
Geological Congress of Turkey with international participation, H. Sozbilir, <;. Ozkaymak, B. Uzel, 0. Sumer, M.
Softa, <;. Tepe, and S. Eski, eds., 798- 802. Ankara: TMMOB Jeoloji Muhendisleri Odas1 Yaymlar1.
- . 2020. Hatay Amik Ovas1 Jeoarkeoloji Projesi: Sediman Karotu <;ah~malarmm On Sonur;:lar1. Arkeometri Sonur;lan
Bachhuber, Christoph. 2015. Citadel and Cemetery in Early Bronze Age Anatolia. Monographs in Mediterranean
Archaeology 13. Sheffield, UK: Equinox.
Bagh, Tine. 2003. The Relationship Between Levantine Painted Ware, Syro/Cilician Ware and Khabur Ware and the
Chronological Implications. In The Synchronisation of Civilisations in the Eastern Mediterranean in the Second
Millennium BCE. Proceedings of the SCIEM 2000 Euro-Conference in Haindorf, 2nd-7th ofMay 2001, M. Bietak,
ed., 219- 238. Wien: Verlag der Osterreichischen Akademie der Wissenschaften.
- . 2013. Levantine Painted Ware from Egypt and the Levant. Wien: Osterreichischen Akademie der Wissenschaften.
Baker, Jill L. 2012. The Funeral Kit: Mortuary Practices in the Archaeological Record. Walnut Creek, California:
Left Coast Press.
Barker, M. 1955. Fragments of Mural Paintings. Examination of the Fragments of Mural Paintings from Atchana. In
Alalakh: An Account of the Excavations at Tell Atchana in the Hatay: 1937-1949, L. Woolley, ed., 233- 234.
Oxford: Society of Antiquaries.
Barrett, Caitlin E. 2007. Was Dust their Food and Clay Their Bread? Grave Goods, the Mesopotamian Afterlife, and
the Liminal Role ofinana/Ishtar. Journal ofAncient Near Eastern Religions 7(1): 7- 65.
Batiuk, Stephen D. 2007. Ancient Landscapes of the Amuq: Geoarchaeological Surveys of the Amuq Valley: 1999-
2006. Journal of the Canadian Society for Mesopotamian Research 2: 51-57.
Batiuk, Stephen D. and Mara T. Horowitz. 2010. Temple Deep Sounding Investigations 2001- 2006. In Tell Atchana,
Ancient Alalakh. Volume 1: 2003- 2004 Excavation Seasons, K.A. Yener, ed., 161-168. istanbul: Koi;: University
Press.
Bronk Ramsey, Christopher, 2020. OxCal Program v. 4.4.4 Radiocarbon Accelerator Unit. University of Oxford,
Oxford, U.K. (Available at https://cl 4.arch.ox.ac.uk/oxcal.html.)
Brysbaert, Ann. 2002. Common Craftmanship in the Aegean and East Mediterranean Bronze Age: Preliminary
Technological Evidence with Emphasis on the Painted Plaster from Tell el-Dab'a, Egypt. Egypt and the Levant 12:
95- 107.
Buccellati, Giorgio and Marilyn Kelly-Buccellati. 2007. Urkesh and the Question of the Hurrian Homeland. Bulletin
of the Georgian National Academy ofSciences 75(2): 141- 151.
Bulu, Miige. 2012. Interpreting an Intact Kitchen Context from Middle Bronze Age Alalakh: Its Organization and
Function. Unpublished M.A. thesis, Archaeology and History of Art, Koi;: University, istanbul.
- . 2016. An Intact Palace Kitchen Context from Middle Bronze Age Alalakh: Organization and Function. In
Proceedings of the 9th International Congress on the Archaeology of the Ancient Near East, 9-13 June 2014,
Basel, R.A. Stucky, 0. Kaelin, and H.-P. Mathys, eds., 301-314. Wiesbaden: Harrassowitz Verlag.
- . 2017a. A New Look at the Periphery of the Hittite Empire: Re-evaluating Middle and Late Bronze Age
Settlements of the Amuq Valley in the Light of Ceramics. In Places and Spaces in Hittite Anatolia I: Hatti and the
East. Proceedings of an International Workshop on Hittite Historical Geography in Istanbul, 25th-26th October
2013, M. Alparslan, ed., 185-208. istanbul: Tllrk Eski<;ag Bilimleri Enstitiis-U.
- . 2017b. A Syro-Cilician Pitcher from a Middle Bronze Age Kitchen at Tell Atchana, Alalakh. In Overturning
Certainties in Near Eastern Archaeology. A Festschrift in Honor of K. Ashhan Yener, <;. Maner, M.T. Horowitz,
and AS. Gilbert, eds., 101-116. Leiden: Brill.
- . 2021. Production and Consumption of Syro-Cilician Ware at Tell Atchana, Alalakh: A Technological and
Functional Analysis. Unpublished Ph.D. dissertation, Archaeology and History of Art, Ko<; University, istanbul.
Burke, Aaron A. 2017. Amorites, Climate Change, and the Negotiation of Identity at the End of the Third Millennium
B.C. In The Late Third Millennium in the Ancient Near East. Chronology, C14, and Climate Change, F.
Hoflmayer, ed., 261-307. Chicago: University of Chicago Press.
Cradic, Melissa S. 2017. Embodiments of Death: The Funerary Sequence and Commemoration in the Bronze Age
Levant. Bulletin of the American Schools of Oriental Research 377: 219-248.
Curvers, Hans H., Glenn M. Schwartz, and Sally Dunham. 1997. Umm el-Marra, a Bronze Age Urban Center in the
Jabbul Plain, Western Syria. American Journal ofArchaeology 101(2): 201- 239.
Di Michele, Angelo and Maria Elena Pedrosi. 2012. Fortification and Burial Grounds in Tell Afis (Syria) between
Middle Bronze Age and Late Bronze Age. Paper read at Broadening Horizons 3. Conference of Young
Researchers Working in the Ancient Near East.
Dornemann, Rudolph H. 1979. Tell Hadidi: A Millennium of Bronze Age City Occupation. Annual of the American
Schools of Oriental Research 44: 113-151.
- . 1992. Early Second Millennium Ceramic Parallels Between Tell Hadidi-Azu and Mari. In Mari in Retrospect.
Fifty Years ofMari and Mari Studies, G.D. Young, ed., 77-112. Winona Lake, Indiana: Eisenbrauns.
Dunham, Sally. 1993. A Wall Painting from Tell al-Raqa'i, North-east Syria. Levant 25: 127-143.
Duru, Refik. 2003. UnutulmuJ Bir BaJkent Ti/men/A Forgotten Capital City Ti/men. lstanbul: Tursab K-Ultfu
Yaymlar1.
Ellison, Rosemary. 1984. The Uses of Pottery. Iraq 46(1): 63-68.
Emre, Kutlu. 1991. Cemeteries of Second Millennium B.C. in Central Anatolia. In Essays on Ancient Anatolian and
Syrian Studies in the 2nd and 1st Millennium B.C., H.I.H.P.T. Mikasa, ed., 1-15. Wiesbaden: Otto Harrassowitz
Verlag.
Engin, Atilla. 2020. Oylum Hoyuk and Alalakh: Cultural Relations in the Second Millennium BCE. In Alalakh and its
Neighbors: Proceedings of the 15th Anniversary Symposium at the New Hatay Archaeology Museum, June 10-12,
2015, K.A. Yener and T. Ingman, eds., 275-303. Leuven: Peeters.
Felli, Candida. 2012. Funerary Practices from the End of the Early to the Middle Bronze Age in Northwestern Syria:
The Middle Euphrates Valley. In (Re-)Constructing Funerary Rituals in the Ancient Near East. Proceedings of the
First International Symposium of the Tubingen Post-Graduate School "Symbols of the Dead" in May 2009, P.
Pfalzner, H. Niehr, E. Pernicka, and A. Wissing, eds., 79-110. Qatna Studien Supplementa 1. Wiesbaden:
Harrassowitz Verlag.
Gates, Marie-Hemiette. 1981. Alalakh Levels VI and V: A Chronological Reassessment. Syro-Mesopotamian Studies
4(2): 11-50.
- . 1988. Dialogues Between Ancient Near Eastern Texts and the Archaeological Record: Test Cases from Bronze
Age Syria. Bulletin ofthe American Schools ofOriental Research 270: 63-91.
- . 2000. Kinet Hoyuk (Hatay, Turkey) and MB Levantine Chronology. Akkadica 119-120: 77-101.
- . 2010. 2008 Season at Kinet Hoyuk (Ye~il-Dortyol, Hatay). Kazi Sonw;lan Toplantzsi 31: 303- 320.
- . 2011. 2009 Season at Kinet Hoyuk (Ye~il-Dortyol, Hatay). Kazi Sonu<;lan ToplantlSl 32: 182- 195.
Goldman, Hetty. 1956. Excavations at Gozliikule, Tarsus. Princeton: Princeton University Press.
Gonen, Rivka. 1992. Burial Patterns and Cultural Diversity in Late Bronze Age Canaan. American Schools of
Oriental Research Dissertation Series 7. Winona Lake, Indiana: Eisenbrauns.
Gutsuz, P1nar, Mustafa Kibaroglu, Gfusel Sunal, and Sinem Hac1osmanoglu. 2017. Geochemical Characterization of
Clay Deposits in the Amuq Valley (Southern Turkey) and the Implications for Archaeometric Study of Ancient
Ceramics. Applied Clay Science 141: 316- 333.
Hallote, Rachel S. 1995. Mortuary Archaeology and the Middle Bronze Age Southern Levant. Journal of
Mediterranean Archaeology 8(1): 93-122.
Harrison, T.P. 2013. Tayinat in the Early Iron Age. In Across the Border: Late Bronze- Iron Age Relations between
Syria and Anatolia. Proceedings of a Symposium held at the Research Center of Anatolian Studies, Ko<;
University, Istanbul May 31-June 1, 2010, K.A.Yener, ed., 61-87. Amsterdam: Peeters.
Heinz, M. 1992. Tell Atchana/Alalakh: Die Schichten VII-XVII. Alter Orient und Altes Testament Band 41. Kevelaer:
Verlag Butzon & Bercker.
Horowitz, Mara T. 2015. The Evolution of Plain Ware Ceramics at the Regional Capital of Alalakh in the 2nd
Millennium BCE. In Plain Pottery Traditions of the Eastern Mediterranean and Near East: Production, Use, and
Social Significance, C. Glatz, ed., 153-82. Walnut Creek: Left Coast Press.
- . 2017. Pot-marks as a Feature oflnterregional Connectivity at Tell Atchana-Alalakh: Evidence from the 2006-12
Excavations. In Questions, Approaches, and Dialogues in Eastern Mediterranean Archaeology: Studies in Honor
of Marie-Henriette and Charles Gates, E. Kozal, M. Akar, Y. Heffron,<;. <;ilingiroglu, T.E. Serifoglu, C. <;almlar,
S. Unlusoy, and E. Jean, eds., 307- 329. Munster: Ugarit-Verlag.
- . 2019. The Local Ceramics of Late Bronze II Alalakh. In Tell Atchana, Alalakh Volume 2: The Late Bronze II City.
2006-2010 Excavation Seasons, K.A. Yener, M. Akar, and M.T. Horowitz, eds., 193-249. Istanbul: Kos;
University Press.
Horowitz, Mara T. and Canan <;akrrlar. 2017. Novel Uses of Wild Faunal Resources at Transitional Middle-Late
Bronze Age Tell Atchana. In Overturning Certainties in Near Eastern Archaeology. A Festschrift in Honor ofK.
Aslzhan Yener, <;. Maner, M.T. Horowitz, and A.S. Gilbert, eds., 222- 244. Leiden and Boston: Brill.
Horowitz, Mara T., Lynn Dodd, Adam Green, and Derek Ryter. 2019. Survey and Geophysical Research at Tell
Atchana, 2006--2010. In Tell Atchana, Alalakh Volume 2: The Late Bronze II City. 2006-2010 Excavation
Seasons, K.A. Yener, M. Akar, and M. T. Horowitz, eds., 93- 125. istanbul: Kos; University Press.
Honvitz, Liora Kolska. 2001. Animal Offerings in The Middle Bronze Age: Food for the Gods, Food for Thought.
Palestine Exploration Quarterly 133(2): 78- 90.
Iamoni, Marco. 2012. The Late MBA and LBA Pottery Horizons at Qatna. Innovation and Conservation in the
Ceramic Tradition of a Regional Capital and the Implications for Second Millennium Syrian Chronology. Udine:
Forum.
Ingman, Tara. 2014. Mortuary Practices at Tell Atchana, Ancient Alalakh in the Middle and Late Bronze Ages.
Unpublished M.A. thesis, Kos; University, istanbul.
- . 2017. The Extramural Cemetery at Tell Atchana, Ancient Alalakh and GIS Modeling. In Overturning Certainties
in Near Eastern Archaeology: A Festschrift in Honor ofK. Aslzhan Yener, <;. Maner, M. Horowitz, and A. Gilbert,
eds., 245-258. Leiden and Boston: Brill.
- . 2020a. Identity and Changing Funerary Rituals at Tell Atchana, Alalakh: Mortuary and Isotopic Analyses.
Unpublished Ph.D. dissertation, Kos; University, istanbul.
- . 2020b. Mortuary Practices and GIS Modeling at Tell Atchana, Ancient Alalakh. In Alalakh and its Neighbors:
Proceedings of the 15th Anniversary Symposium at the New Hatay Archaeology Museum, June 10- 12, 2015, K.A.
Yener and T. Ingman, eds., 389--406. Leuven: Peeters.
Ingman, Tara, Stefanie Eisenmann, Eirini Skourtanioti, Murat Akar, Jana Ilgner, Guido Alberto Gnecchi Ruscone,
Petrus le Roux, Rula Shafiq, Gunnar Neumann, Marcel Keller, Cacilia Freund, Sara Marzo, Mary Lucas, Johannes
Krause, Patrick Roberts, K. Ashhan Y ener, and Philipp W. Stockhammer. 2021. Human Mobility at Tell Atchana
(Alalakh), Hatay, Turkey During the 2nd Millennium BCE: Integration of Isotopic and Genomic Evidence. PLOS
ONE 16(6): e0241883.
Jacquet, Antoine. 2012. Funerary Rites and Cult of the Ancestors During the Amorite Period: the Evidence of the
Royal Archives of Mari. In (Re-)Constructing Funerary Rituals in the Ancient Near East. Proceedings of the First
International Symposium of the Tubingen Post-Graduate School "Symbols of the Dead" in May 2009, P. Pfalzner,
H. Niehr, E. Pemicka, and A. Wissing, eds., 123- 136. Qatna Studien Supplementa 1. Wiesbaden: Harrassowitz
Verlag.
Jonker, Gerdien. 1995. The Topography of Remembrance: The Dead, Tradition and Collective Memory in
Mesopotamia. Leiden: Brill.
Kamlah, Jens and Helene Sader. 2010. Deutsch-libanesische Ausgrabungen auf Tell el-Burak sudlich von Sidon
Vorbericht nach Abschluss der siebten Kampagne. Zeitschrift des Deutschen Palastina-Vereins 126: 93-115.
Keswani, Priscilla. 2012. Urban Mortuary Practices at Enkomi and Ugarit in the Second Millennium BCE. In (Re-)
Constructing Funerary Rituals in the Ancient Near East. Proceedings of the First International Symposium of the
Tubingen Post-Graduate School "Symbols of the Dead" in May 2009, P. Pfalzner, H. Niehr, E. Pemicka, and A.
Wissing, eds., 182- 203. Qatna Studien Supplementa 1. Wiesbaden: Harrassowitz Verlag.
Koehl, Robert B. 2013. The Near Eastern Contribution to Aegean Wall Painting and Vice Versa. In Cultures in
Contact: From Mesopotamia to the Mediterranean in the Second Millennium B.C., J. Aruz, S. Graff, and Y. Rakic,
eds., 172-178. New York: Metropolitan Museum of Art.
- . 2020. Alalakh and the Aegean: Five Centuries of Shifting but Enduring Contacts. In Alalakh and its Neighbors:
Proceedings of the 15th Anniversary Symposium at the New Hatay Archaeology Museum, June 10- 12, 2015, K.A.
Yener and T. Ingman, eds., 201-223. Leuven: Peeters.
Kulakoglu, Fikri. 2011. Kultepe-Kanes: A Second Millennium B.C.E. Trading Center on the Central Plateau. In The
Oxford Handbook of Ancient Anatolia, 10,000-323 B.C.E., S.R. Steadman and G. McMahon, eds., 1012-1030.
Oxford: Oxford University Press.
Kuzucuoglu, Catherine and Catherine Marro, eds. 2007. Societes Humaines et Changement Climatique a la fin du
troisieme millenaire: une crise a-t-elle eu lieu en Haute Mesopotamie ? Actes du Collogue de Lyon (5-8 decembre
2005). istanbul: Institut Fran9ais d'Etudes Anatoliennes-Georges Dumezil.
Laneri, Nicola. 2014. Locating the Social Memory of the Ancestors: Residential Funerary Chambers as Locales of
Social Remembrance in Mesopotamia During the Late Third and Early Second Millennia BCE. In Contextualising
Grave Inventories in the Ancient Near East. Proceedings of a Workshop at the London 7th ICAANE in April 2010
and an International Symposium in Tubingen in November 2010, Both Organised by the Tubingen Post-Graduate
School "Symbols of the Dead", P. Pfalzner, H. Niehr, E. Pernicka, S. Lange, and T. Koster, eds., 3- 10. Qatna
Studien Supplementa 3. Weisbaden: Harrassowitz Verlag.
Laneri, Nicola and Mark Schwartz. 2011. Southeastern and Eastern Anatolia in the Middle Bronze Age. In The Oxford
Handbook of Ancient Anatolia. 10,000- 323 B.C.E., S.R. Steadman and G. McMahon, eds., 337- 360. Oxford:
Oxford University Press.
Lange, Sarah. 2012. Food and Libation Offerings for the Royal Dead in Ugarit. In (Re-)Constructing Funerary Rituals
in the Ancient Near East. Proceedings of the First International Symposium of the Tubingen Post-Graduate School
"Symbols of the Dead" in May 2009, P. Pfalzner, H. Niehr, E. Pernicka, and A. Wissing, eds., 161-181. Qatna
Studien Supplementa 1. Wiesbaden: Harrassowitz Verlag.
Lauinger, Jacob. 2014. Witnessing at Old Babylonian Alalah: A New Level VII Witness List from the Kos; University
Excavations at Tell Atchana/Alalah. Revue d'Assyriologie 108: 25-40.
MacDougal, Renata. 2014. Remembrance and the Dead in Second Millennium BCE Mesopotamia. Unpublished Ph.D.
dissertation, University of Leicester, Leicester.
Matthiae, Paolo. 1978. Preliminary Remarks on the Royal Palace of Ebia. Syro-Mesopotamian Studies 2(2): 13-33.
- . 1984. New Discoveries at Ebia: The Excavation of the Western Palace and the Royal Necropolis of the Amorite
Period. The Biblical Archaeologist 47(1): 18-32.
-. 2002. About the Formation of the Old Syrian Architecture. In OfPots and Plans: Papers on the Archaeology and
History of Mesopotamia and Syria Presented to David Oates in Honor of his 75th Birthday, L. al-Gailani Werr, J.
Curtis, H. Martin, A. McMahon, J. Oates, and J.E. Reade, eds., 191-209. London: NABU publications.
McClellan, Thomas L. 1989. The Chronology and Ceramic Assemblages of Alalakh. In Essays in Ancient Civilization
Presented to Helene J Kantor, A. Leonard, Jr. and B.B. Williams, eds., 181-212. Chicago: University of Chicago
Press.
Mellink, Machteld J. 1957. Alalakh: An Account of the Excavations at Tell Atchana in the Hatay, 1937-1949,
Review. American Journal of Archaeology 61(4): 395-400.
Miglus, Peter A. and Eva Strommenger. 2002. Tall Bi 'a I Tuttul - VIII. Stadtbefestigungen, Hauser und Tempel.
Saarbrilcken: Saarbrilcker Druckerei und Verlag.
Montesanto, Mariacarmela and Marina Pucci. 2019- 20. The Iron Age at Alalakh. Archaeology & History in the
Lebanon 50-51: 93-135.
Morandi Bonacossi, Daniele. 2011. The Middle Bronze Age Necropolis at Mishrifeh. In Interdiszipliniire Studien zur
Konigsgruft von Qatna, P. Pfalzner, ed., 11-37. Qatna-Studien 1. Wiesbaden: Harrassowitz Verlag.
Morandi Bonacossi, Daniele, M. da Ros, G. Gama, M. Iamoni, and M. Merlino. 2009. The 'Eastern Palace' and the
Residential Architecture of Area T at Mishrifeh/Qatna: Preliminary Report on the 2006- 2008 Excavation
Campaigns of the Italian Component of the Syro-Italian Archaeological Project. Mesopotamia: Rivista di
Archeologia, Epigrafia e Storia Orientale Antica 44: 61-112.
Morgan, Kathryn R. and Sebastiano Soldi. 2021. Middle Bronze Age Zincirli: An Interim Report on Architecture,
Small Finds, and Ceramics from a Monumental Complex of the 17th Century B.C.E. Bulletin of the American
Schools of Oriental Research 385 (DOI 10.1086/711910).
Morrison, Jerolyn E. and Mara T. Horowitz. 2016. Field-Based Experiments Replicating Ceramic Fabrics: Late
Bronze Age Cookwares from Two Mediterranean Sites. In Integrative Approaches in Ceramic Petrography, M.F.
Ownby, LC. Drue, and M.A. Masucci, eds., 177-195. Salt Lake City: University of Utah Press.
Muller, Beatrice. 2018. Contextes techniques et historiques des peintures murals du Grand Palais Royal de Mari. Une
mise au point. In Tracing Technoscapes: The Production of Bronze Age Wall Paintings in the Eastern
Mediterranean. International Congress on the Archaeology of the Ancient Near East, J. Becker, J. Jungfleisch, and
C. von Ruden, eds., 41-84. Leiden: Sidestone Press.
Niemeier, Wolf-Dietrich and Barbara Niemeier. 2000. Aegean Frescoes in Syria-Palestine: Alalakh and Tel Kabri. In
The Wall Paintings ofThera: Proceedings of the First International Symposium, S. Sherratt, ed., 763- 800. Athens:
Thera Foundation.
Nigro, Lorenzo. 2002a. The Middle Bronze Age Pottery Horizon of Northern Inner Syria on the Basis of the Stratified
Assemblages of Tell Mardikh and Rama. In Ceramique de L 'Age du Bronze en Syrie I: La Syrie du Sud et la
Vallee de l'Oronte, M. Al-Maqdissi, V. Matoi:an, and C. Nicolle, eds., 97-128. Beyrouth: Institut Frans;ais
D ' Archeologie du Proche-Orient.
-. 2002b. The MB Pottery Horizon of Tell Mardi.kb/Ancient Ebla in a Chronological Perspective. In The Middle
Bronze Age in the Levant. Proceedings of an International Conference on MB IIA Ceramic Material, Vienna, 24th-
26th of January 2001, M. Bietak, ed., 297-328. Wien: Verlag der Osterreichiscen Akademie der Wissenschaften.
Nunn, Astrid. 1988. Die Wandmalerei und der glasierte Wandschmuck im A/ten Orient. Handbuch der Orientalistik.
Siebente Abteilung: Kunst und Archaologie 1,2 B6. Leiden: Brill.
Nunn, Astrid and Heinrich Piening, eds. 2020. Mesopotamian Sculpture in Colour. Gladbeck: PeWe-Verlag.
Ozgen, Engin and Barbara Helwing. 2001. Ausgrabungen auf dem Oylum Hoyilk, 1997- 2000. Zweiter vorlaufiger
Bericht. Istanbuler Mitteilungen 51: 61-136.
Parrot, Andre. 1958. Mission Archeologique de Mari. Le Palais. 2. Peintures murals. Paris: Librarie Orientaliste Paul
Geuthner.
Pfalzner, Peter. 2015. A House of Kings and Gods-Ritual Places in Syrian Palaces. In Cult and Ritual on the
Levantine Coast and its Impact on the Eastern Mediterranean Realm: Proceedings of the International
Symposium, Beirut 2012, A.M. Maila-Afeiche, ed., 413-442. BAAL Hors-Serie 10. Beyrouth: Bulletin
d'Archeologie et d'Architecture Libanaises.
Pinnock, Frances. 2019. The Royal Palace G of Early Syrian Ebia: Structure and Functions. In Ancient Egyptian and
Ancient Near Eastern Palaces, Vol. II, M. Bietak, P. Matthiae, and S. Prell, eds., 67- 79. Wiesbaden: Harrassowitz
Verlag.
Pitard, Wayne T. 1996. Care of the Dead at Emar. In Emar: The History, Religion, and Culture of a Syrian Town in
the Late Bronze Age, M.W. Chavalas, ed., 123-140. Bethesda: CDL Press.
Pucci, Marina, Ekin Koza!, and Robert B. Koehl. 2020. Thoughts on the Reception and Rejection of Aegean and
Cypriot Ceramics in the Amuq Plain during the Late Bronze and Iron Ages. In Communication Uneven.
Acceptance of and Resistance to Foreign Influences in the Connected Ancient Mediterranean, J. Driessen and A.
Vanzetti, eds., 147-169. AEGIS 20. Louvain: Presses universitaires de Louvain.
Reimer, Paula J., William E.N. Austin, Edouard Bard, Alex Bayliss, Ppaul G. Blackwell, Christopher Bronk Ramsey, Martin
Butzin, Hai Cheng, R. Lawrence Edwards, Michael Friedrich, Pieter M. Grootes, Thomas P. Guilderson, Iika Hajdas,
Timothy J. Heaton, Alan G. Hogg, Konrad A. Hughen, Brend Kromer, Sturt W. Manning, Raimund Muscheler, Jonathan G.
Palmer, Charlotte Pearson, Johannes van der Plicht, Ron W. Reimer, David A. Richards, E. Marian Scott, John R. Southon,
Christian S. M. Turney, Lukas Wacker, Florian Adolphi, UlfBtintgen, Manuela Capano, Simon M. Fahrni, Alexandra
Fogtmann-Schulz, Ronny Friedrich, Peter Kohler, Sabrina Kudsk, Fusa Miyake, Jesper Olsen, Frederick Reinig, Minoru
Sakamoto, Adam Sookdeo and Sarah Talamo 2020. The IntCal20 Northern Hemisphere Radiocarbon Age Calibration
Curve (0-55 cal kBP). Radiocarbon 62: 725-757.
Roux, Valentine. 2019. Ceramics and Society. A Technological Approach to Archaeological Assemblages. Cham:
Springer.
von Ruden, Constance. 2020. Reconsidering the Alalakh Paintings within their Levantine Context. In Alalakh and its
Neighbors: Proceedings of the 15th Anniversary Symposium at the New Hatay Archaeology Museum, June 10-12,
2015, K.A. Yener and T. Ingman, eds., 141-169. Leuven: Peeters.
Salles, Jean-Frans;ois. 1995. Rituel Mortuaire et Rituel Social a Ras Shamra/Ougarit. In The Archaeology of Death in
the Ancient Near East, S. Campbell and A. Green, eds., 171-184. Oxbow Monographs 51. Oxford: Oxbow Books.
Schwartz, Glenn M. 2006. From Collapse to Regeneration. In After Collapse: The Regeneration of Complex Societies,
G.M. Schwartz and J.J. Nichols, eds., 3-17. Tucson: The University of Arizona Press.
Schwartz, Glenn M. and Hans Curvers. 1992. Tell al-Raqa'i 1989 and 1990: Further Investigations at a Small Rural
Site of Early Urban Northern Mesopotamia. American Journal ofArchaeology 96: 397--419.
Schwartz, Glenn M., Hans H. Curvers, Fokke A. Gerritsen, Jennifer A. Maccormack, Naomi F. Miller, and Jill A.
Weber. 2000. Excavation and Survey in the Jabbul Plain, Western Syria: The Umm el-Marra Project 1996- 1997.
American Journal ofArchaeology 104(3): 419--462.
Schwartz, Glenn M., Hans H. Curvers, Sally Dunham, and Barbara Stuart. 2003. A Third-Millennium B.C. Elite
Tomb and Other New Evidence from Tell Umm el-Marra, Syria. American Journal of Archaeology 107(3): 325-
361.
Schwartz, Glenn M. and John J. Nichols, eds. 2006. After Collapse: The Regeneration of Complex Societies. Tucson:
The University of Arizona Press.
Seton-Williams, M. Veronica. 1953. A Painted Pottery of the Second Millennium from Southern Turkey and Northern
Syria. Iraq 15(1): 56-68.
Skourtanioti, Eirini, Y1lmaz S. Erda!, Marcella Frangipane, Francesca Balossi Restelli, K. Ashhan Yener, Frances
Pinnock, Paolo Matthiae, Rana Ozbal, Ulf-Dietrich Schoop, Farhad Guliyev, Tufan Akhundov, Bertille Lyonnet,
Emily L. Hammer, Selin E. Nugent, Marta Burri, Gunnar Neumann, Sandra Penske, Tara Ingman, Murat Akar,
Rula Shafiq, Guilio Palumbi, Stefanie Eisenmann, Marta d'Andrea, Adam B. Rohrlach, Christina Warinner,
Choongwon Jeong, Philipp W. Stockhammer, Wolfgang Haak, and Johannes Krause. 2020. Genomic History of
Neolithic to Bronze Age Anatolia, Northern Levant and South Caucasus. Cell 181: 1158- 1175.
Smith, Sidney. 1949. The Statue of Idrimi. Occasional Publications of the British Institute of Archaeology at Ankara 1.
London: British Institute of Archaeology in Ankara.
Tubb, Jonathan N. 1983. The MB IIA Period in Palestine: Its Relationship with Syria and Its Origin. Levant 15: 49-
62.
Weiss, Harvey. 2014. The Northern Levant During the Intermediate Bronze Age: Altered Trajectories. In Oxford
Handbook of the Archaeology of the Levant, A.E. Killebrew and M.L. Steiner, eds., 367-387. Oxford: Oxford
University Press.
- . ed. 2017. Megadrought and Collapse from Early Agriculture to Angkor. New York: Oxford University Press.
Weiss, Harvey, Peter M.M.G. Akkermans, Gil J. Stein, Dominique Parayre, and Robert Whiting. 1990. 1985
Excavations at Tell Leilan, Syria. American Journal ofArchaeology 94(4): 529- 581.
Welton, Lynn, Stephen D. Batiuk, Timothy P. Harrison, David R. Lipovitch, and Mairi M. Capper. 2011. Tell Tayinat
in the Late Third Millennium. Recent Investigations of the Tayinat Archaeological Project 2008- 2010. Anatolica
37: 147-185.
Wissing, Anne. 2012. Ritual Aspects of Middle Bronze Age Burial Practices in the Hurrian City of Urkesh. In
(Re)Constructing Funerary Rituals in the Ancient Near East. Proceedings of the First International Symposium of
the Tubingen Post-Graduate School "Symbols of the Dead" in May 2009, P. Pfalzner, H. Niehr, E. Pernicka, and
A. Wissing, eds., 111-121. Qatna Studien Supplementa 1. Wiesbaden: Harrassowitz Verlag.
Woolley, C. Leonard. 1948. Excavations at Atchana-Alalakh, 1939. The Antiquaries Journal 28(1, 2): 1-19.
- . 1955. Alalakh: An Account of the Excavations at Tell Atchana in the Hatay, 1937-1949. Reports of the Research
Committee of the Society of Antiquaries of London 18. London: Oxford University Press.
Wygnanska, Zuzanna. 2014. The Ancestor Cult in the Middle Bronze Age at Tell Arbid, Syria. In Contextualising
Grave Inventories in the Ancient Near East. Proceedings of a Workshop at the London 7th ICAANE in April 2010
and an International Symposium in Tubingen in November 2010, Both Organised by the Tubingen Post-Graduate
School "Symbols of the Dead," P. Pfalzner, H. Niehr, E. Pemicka, S. Lange, and T. Koster, eds., 39--49. Qatna
Studien Supplementa 3. Weisbaden: Harrassowitz Verlag.
Yener, K. Ashhan. 2005. The Amuq Valley Regional Projects. In The Amuq Valley Regional Projects. Vol.I, Surveys
in the Plain of Antioch and Orontes Delta, Turkey, 1995-2002, K.A. Yener, ed., 1-23. Chicago: Oriental Institute
Publications, University of Chicago.
- . 2007. The Anatolian Middle Bronze Kingdoms and Alalakh: Mukish, Kanesh and Trade. Anatolian Studies 57:
151-160.
- . 2013a. New Excavations at Alalakh: The 14th-12th Centuries BC. In Across the Border: Late Bronze- Iron Age
Relations between Syria and Anatolia. Proceedings of a Symposium held at the Research Center of Anatolian
Studies, Ko<; University, Istanbul May 31-Junel, 2010, K.A. Yener, ed., 11-36. Leuven: Peeters.
- . 2013b. Recent Excavations at Alalakh: Throne Embellishments in Middle Bronze Age Level VIL In Cultures in
Contact: From Mesopotamia to Mediterranean in the Second Millennium B.C., Joan Aruz, Sarah B. Graff, and
Yelena Rakic, eds., 142- 153. New York: The Metropolitan Museum of Art.
- . 2015. A Monumental Middle Bronze Age Apsidal Building at Alalakh. In NOSTOI. Indigenous Culture Migration
and Integration in the Aegean Islands and Western Anatolia During the Late Bronze and Early Iron Age, N.
Stampolidis, <;. Maner, and K. Kopanias, eds., 485--498. lstanbul: Ko9 University Press.
- . 2017. Cult and Ritual at Late Bronze Age II Alalakh: Hybridity and Power under Hittite Administration. In 5emes
Rencontres d'Archeologie de L 'IFEA. Hittitology Today: Studies on Hittite and Neo-Hittite Anatolia in Honor of
Emmanuel Laroche's 100th Birthday, A. Mouton, ed., 215-224. lstanbul: Institut Fran9ais d'Etudes Anatoliennes
Georges Dumezil.
- . 2021. Some Thoughts about Middle Bronze Age Alalakh and Ugarit: Reassessing an Alalakh Wall Painting with
Archival Data. In Ougarit, 90 ans apres, V. Matoian ed., 1-16: https://www.mission-ougarit.fr/parution-rso-xxvi/.
Yener, K. Ashhan and G. Bike Yaz1c10glu. 2010. Excavation Results. In The Amuq Valley Regional Projects:
Excavations in the Plain of Antioch: Tell Atchana, Ancient Alalakh, Vol. I: The 2003-2004 Excavation Seasons,
K.A. Yener, ed., 11--49. lstanbul: Kay University Press.
Yener, K. Asllhan, Murat Akar, and Mara T. Horowitz, eds. 2019. Tell Atchana, Alalakh Volume 2: The Late Bronze
11 City. 2006-2010 Excavations. istanbul: Ko9 University Press.
Yener, K. Ashhan, Murat Akar, and Tara Ingman. 2020. Alalakh in the Past, Atchana in the Present: Situating Site and
City. In Alalakh and Its Neighbours. Proceedings of the 15th Anniversary Symposium at the New Hatay
Archaeology Museum, 10- 12 June 2015, K.A. Yener and T. Ingman, eds., 3- 12. Leuven: Peeters.
CHAPTER EIGHT
2015-2020 EXCAVATIONS AT ELBiSTAN KARAHOYUK AND HAMZATEPE:

A HITTITE SETTLEMENT AND ITS CEMETERY IN THE ELBiSTAN PLAIN
BORA UYSAL AND ALi {;iF<;i
The archaeological site of Elbistan Karahoyuk is located in the center of Karahuytik neighborhood about 10 km
northwest of the Elbistan district in Kahramanmara~ province. The site is situated on the east bank ofHurman stream-
a tributary of the Ceyhan River-in the southwestern portion of the Elbistan Plain, north of Mount Sar. Currently,
with its preserved height of 21 m and covering an area of 450 x 300 m, Elbistan Karahoyuk is one of the largest
mounds in the region (Fig. 8-1 ).
The cemetery site of Hamzatepe, associated with the site of Elbistan Karahoyuk, is located on the west bank of
Burman stream very close to the mound. The natural hill known as Hamzatepe, 1236 m asl, is situated at the northeast
end of the rolling hills between the Karahoyuk and Izgm neighborhoods. The area that has yielded archaeological
remains is a ridge with an altitude of 1214 m, located at the northeastern tip of the hill; this locality is also known as
Hamzatepe. This 48-m-high natural rocky outcrop is spread over an area of approximately 380 x 380 m (Fig. 8-1 ).
lf, ~ =- AF$1N
■
•
HamzaTe
Figure 8-1. Left: Map showing the location of Elbistan Karahoyuk and Hamzatepe;
Right: Aerial view of Elbistan Karahoyuk and Hamzatepe.
The earliest habitation at Elbistan Karahoyuk, which lies at a strategic location in the Elbistan Plain, dates back to
the Early Bronze Age. Research has shown that the site was an affluent settlement center during the Middle and Late
Bronze Ages and therefore must have been one of the important cities in this region during the Old Assyrian Trade
Colony and Hittite periods. The site remained a prominent center during the Iron Age, after which its importance
seems to have waned, although habitation continued during the Hellenistic and Roman periods. Although various
suggestions have been put forward regarding the ancient name of the site (Yigit 1997: 1-14; Pehlivanoglu 2019: 145-
172), so far there is no consensus on its historical localization.
The earliest archaeological investigations at Elbistan Karahoyuk date to the tum of the 20 th century. Firstly, test
excavations were conducted at the site in 1906 by Hugo Grothe as part of his explorations in the region that lies
between Kayseri and Kahramanmara~ provinces (Grothe 1911: CCLXXXXIII-CCLXXXIV). Later on, Elbistan
Karahoyuk was also visited by Hans Heinrich von der Osten in 1929 during his explorations in the region (von der
Osten 1930: 115-116, figs. 124-125).
The first systematic excavations at Elbistan Karahoyuk were conducted in 1947 by Tahsin and Nimet Ozgtiy.
Excavations were carried out in two areas for about one month. The main trench was placed in the center of the
mound in a 20 x 20 m area, and a 10 x 5 m test trench was excavated in the southern sector of the mound (Ozguy
1948: 230; Ozguy and Ozguy 1949: 19). Today, the main excavation area is still clearly recognizable as a wide pit in
the center of the mound; however, traces of the test trench have disappeared altogether.
2015- 2020 Excavations at Elbistan Karahoyiik 97
During the expedition led by the Ozgiic; team, excavations reached a depth of 5.60 m from the topsoil, and six
main settlement levels were identified. From top to bottom, Level 1 dates to the Roman period, and Levels 2 to 5 to
the Iron Age. The Iron Age is divided into two main phases, in which Levels 3 and 2 represent a settlement that was
established in the late 7th century BCE, lasting for a long period throughout the Hellenistic and transitioning into the
Roman periods. The earlier Levels 4 and 5, on the other hand, represent a settlement that was part of the Neo-Hittite
cultural sphere and was established in the 11 th century BCE, lasting until the late 8th century BCE. The earliest
excavated settlement level is Level 6, which dates to the later phase of the Late Bronze Age, that is the Hittite Empire
period (Ozgiic; 1948: 230- 231; Ozgiic; and Ozgiic; 1949: 20- 21, 34- 36). Undoubtedly, the most significant finding of
the 1947 expedition is a stela inscribed with Luwian hieroglyphs (the "Karahoyiik Stela"), which was found in Level
5 (Ozgiic; 1948: 231; Ozgiic; and Ozgiic; 1949: 22- 25, 52- 53, plans 3-4, pl. VIII- X, XLIX- LI; Masson 1979: 225-
241; Nowicki 1981: 251 - 273; Hawkins 1993: 273- 279; Woudhuizen 2003: 211 - 224; Quickelberghe 2013: 253- 263).
After a long interruption, investigations at Elbistan Karahoyiik were reinitiated in 2014 with geomagnetic and geo-
radar surveys conducted under the directorship of the Kahramanmara~ Museum. Following the survey, excavations
began under the directorship of the Kahramanmara~ Museum in 2015 and continued until 2019 with the scientific
leadership of Bora Uysal as project coordinator. In the 2020 season, excavations were conducted under the
directorship of Bora Uysal. This chapter presents the results of the 2015- 2020 excavations at Elbistan Karahoyiik and
2017- 2018 excavations at Hamzatepe.
2015-2020 EXCAVATIONS AT ELBiSTAN KARAHOYUK

I J K L M N O P R S T U V Y In 194 7, in their description of the mound,
Z AA AB AC AD AE AF
excavators note the following: the contour of the

mound was even on all slopes; the houses of
Karahiiyiik village covered the southern and
southwestern parts of the mound; the remaining
parts of the mound were severely damaged; and the
only part that seemed intact was the core of the
mound (Ozgiic; and Ozgiic; 1949: 17- 18). In 2015,
the mound was found to be further disturbed by the
impact of modem development. The modem village
settlement has by now spread all across the mound;
an asphalt road goes around the skirts of the mound,
and large portions of archaeological deposits,
especially on the northern side, have been cut away.
Moreover, deposits have moved downslope from
north to south due to natural erosion. Except for the
northern aspect of the mound, where the soil has
been removed, all its slopes have lost their original
height and contours due to natural factors. Despite
all these circumstances that threaten the mound,
archaeological deposits of the earlier chronological
periods have remained intact, sealed under the thick
Iron Age settlement layer. With the re-initiation of
archaeological work at the site, all modem buildings
on the mound were nationalized. Buildings will be
t demolished according to planned procedures, after

which the mound will resume a physical appearance
approximating its original state before modem
settlement (Uysal 2017: 255, 2019: 100; Uysal and
(.:ifc;i 2017: 36, 2019a: 397).
Renewed excavations at Elbistan Karahoyiik
were carried out in two nearby areas, located north
of the main excavation trench of 1947, and were
labeled the Northwest Slope and North Terrace (Fig.
8-2). Excavations in the Northwest Slope Area were
Figure 8-2. Topographic plan of Elbistan Karahoyiik. carried out in the 2015- 2016 seasons, and
excavations on the North Terrace continued
throughout the 2015- 2020 seasons. In 2017, work on the slope was stopped, and the North Terrace Area, which
yielded more fruitful results, became the main focus of investigations.
Seven building levels were identified in the Northwest Slope trenches G7, H7, 17, H8, and 18. No Roman period
remains were encountered in this area, although the topmost settlement level identified during the 1947 expedition
dated to this period. Most probably, the Roman period settlement lies in the areas south of our excavation trench.
98 Chapter Eight: Uysal and <;:if9i
Level 1 in the new trench represents a settlement that was founded in the Late Iron Age and was inhabited into the
Hellenistic period. Level 2 represents an earlier phase of the Late Iron Age, while Levels 3-4 date to the Middle Iron
Age, and Level 5 represents the Early Iron Age. Levels 6-7 date to the Late Bronze Age. In this stratigraphic
sequence, Levels 3, 4, and 5 belong to the period of the Neo-Hittite kingdoms in the regional chronology. In the
Northwest Slope Area, only Level 1 was excavated horizontally, while other levels were exposed in restricted areas.
Architectural remains of these earlier levels are not preserved well enough to delineate building plans; ceramic sherds
excavated from these levels show that the deposits are mixed. Therefore, chronological assessment of Levels 5, 6, and
7 is still pending and will require further work (Uysal 2017: 255-256, fig. 4, 2019: 101-101; Uysal and <;:if9i 2017:
38-39, figs. 5-7, 2018: 569-570, figs. 3-4).
Figure 8-3. Architectural remains from the 1st layer at the Northwest Slope area ofElbistan Karahi:iyiik.
Level 1 on the Northwest Slope is documented just below the topsoil in trenches H7, HS, 17, and 18. Although
architectural remains in certain sections have been washed downslope by erosion, this building level is mostly well-
preserved (Fig. 8-3). Wall foundations are built from small and medium stones with a mudbrick superstructure that is
preserved up to 7-8 courses (ca. 80 cm high) in certain sections. Excavated remains belong to a north-south oriented
large house with at least four rooms and an attached courtyard. A hearth feature and a large storage jar found in one of
the rooms indicates that this was the food preparation and cooking area. The hearth has a rectangular form with
rounded comers and a low ledge, resembling a tray. However, intense ash deposits excavated from around this
rectangular feature, as well as the burnt, red coloring of the mudbricks in the nearby wall, show how the feature was
used intensively as a hearth. Additionally, a bread oven (tandzr) and a mortar were found in the middle of the
courtyard. Apart from the architectural remains, many randomly placed trash pits were found across the excavated
areas of Level 1 (Uysal 2017: 256, fig. 5, 2019: 101; Uysal and <;:if9i 2017: 38, fig. 3).
The North Terrace Area, where renewed excavations at Elbistan Karahi:iyiik are currently focused, covers an area
measuring 60 x 30 m. As we were informed by the local inhabitants of the Karahiiyiik neighborhood, originally a
small and relatively flat area existed at this spot, which was then widened and flattened into a terrace by bulldozers.
The soil that was removed from the rear and lateral sides of the terrace was then piled up at its front. Eight settlement
levels were identified in the North Terrace Area excavations in trenches N9, 09, Ml 0, NlO, Ll 1, Ml 1, Nll, Ll2, and
Ml2 (Fig. 8-4a). There are Hellenistic-Roman period remains visible below the surface in the area that is located
between the 1947 trench and the new trenches. These remains constitute Level 1 in this as yet unexcavated area. The
remains of this level could not be documented in the excavated trenches due to the removal of these deposits during
terracing in the recent past (Uysal 2017: 255-256, fig. 6; Uysal and <;:if9i 2017: 39, 2018: 571). As such, the
stratigraphic sequence of the excavated area is rendered as follows: it begins with Level 2, Late Iron Age; it continues
with Level 3, Middle Iron Age; Level 4 is the Early Iron Age; Level 5 is Late Bronze Age III; Level 6 is Late Bronze
Age II; and Level 7 is Late Bronze Age I. Levels 3 and 4 represent the Neo-Hittite kingdom period, while Levels 5, 6,
and 7 represent the Hittite Empire period. Investigations on the chronological assessment of Level 8 are ongoing.
Level 2 was documented in trenches Lll-Ll2, and Ml2, where walls with stone foundations and mudbrick
superstructure were uncovered; these are preserved in certain sections up to 9-10 courses, reaching a height of about 1
m. At the same time, however, the remains are not well enough preserved as a whole to reveal a clear horizontal plan
(Fig. 8-4b). A characteristic feature of this level that dates to the Late Iron Age is the presence of multiple hearths and
bread ovens. The southeastern portion of the excavated area must have been used for daily chores or as a workshop. A
2015-2020 Excavations at Elbistan Karahoyuk 99
large grinding stone was found in the center of this area, and a vat (bathtub) was found placed in a wall comer in this
section (Uysal 2017: 256, fig. 7; Uysal and Cif9i 2018: 570-571, fig. 5).
Figure 8-4. a) Architectural layers from the Northern Terrace Area at Elbistan Karahoyuk; b) Architectural remains
from the 2nd layer at the Northern Terrace Area; c) Architectural remains from the 3rd layer at the Northern Terrace
Area; d) Architectural remains from the 4th layer at the Northern Terrace Area.
Level 3 is documented in trenches MlO-Ml 1 and NlO-Nl 1, where Level 2 deposits have been removed during
terracing so that Level 3 lies directly below a thick mixed deposit (Fig. 8-4c ). The architectural remains of this phase
that dates to the Middle Iron Age are well preserved in the western portion of the excavated area, while they become
scantier in the eastern areas. Wall foundations are built from small and medium rubble stones, and the mudbrick
superstructure is preserved in certain sections up to 6--7 courses (about 70 cm high). The architectural layout of this
phase is oriented north-south, with parallel and perpendicular walls. A rectangular structure with two or three rooms
can be clearly delineated in trench Ml 1. A long and narrow division wall has been added to the eastern wall of the
room in the southwest (Uysal 2017: 256, fig. 8; Uysal and Cif9i 2017: 39-40, fig. 8, 2018: 571, fig. 9, 2019a: 398,
figs. 3-4).
Level 4, which dates to the Early Iron Age, is
documented in trenches LlO, Ml 0, and NlO-Nl 1
(Fig. 8-4d). Architectural remains are well preserved
in the northwestern portion of the excavated area.
Walls with stone foundations and 9-10 courses of
mudbrick are preserved to a height of over 1 m in
certain sections. There are small storage areas inside
the rectangular planned rooms extending in an east-
west direction, and bread ovens are found in outdoor
spaces (Uysal 2017: 256; Uysal and Cif9i 2018: 571,
fig. 10, 2019a: 398, fig. 5, 2019b: 411-412, figs. 3-
4).
A wide variety of material culture finds were
recovered from excavated Iron Age levels in both the
Figure 8-5. Cylinder seal from the 4th layer at the Northwest Slope and North Terrace excavation areas
Northern Terrace Area at Elbistan Karahoyuk. (Uysal and Cif9i 2017: 40, figs. 9-12, 2018: 570-
100 Chapter Eight: Uysal and (:if9i
571, fig. 8, 2019b: 412, figs. 5-6). The most notable object among the small finds is a steatite cylinder seal that bears
a hieroglyphic inscription in two panels, which was found in Level 4 on the North Terrace (Fig. 8-5). Inscriptions
identify three personal names. This seal is the only other epigraphic find from the site since the discovery of the
inscribed Elbistan Karahoyuk stela (Uysal 2017: 256, fig. 9, 2019: 101; Alparslan and Uysal 2018; Uysal and (:if9i
2018: 571, fig. 11).
Iron Age ceramics can be categorized into two main groups as monochrome and painted wares. Monochrome
wares constitute the majority and display diverse paste characteristics and varying surface colors. Some examples are
decorated with simple grooves or incisions. The most notable group among the monochrome wares are highly
burnished, grey/black lustrous wares. Painted wares constitute a small group and are characterized by geometric
designs executed in dark colors on a light-colored background (Uysal 2019: 101-102; Uysal and <;if9i 2017: 38-39,
fig. 4, 2018: 570, fig. 2).
Figure 8-6. a) Architectural remains from the 5th layer at the Northern Terrace Area at Elbistan Karahoyuk; b)
Architectural remains from the 6th layer at the Northern Terrace Area; c) Architectural remains from the 7th layer at
the Northern Terrace Area; d) Bronze Cauldron from the 5th layer at the Northern Terrace Area.
Level 5, documented in trenches Ml0 and Nl0-Nll on the North Terrace, dates to the Late Bronze Age III, which
corresponds to the late phase of the Hittite Empire period. Architectural remains of this level are well preserved in the
southern portion of the excavation area (Fig. 8-6a). Walls are built of mudbrick above stone foundations, and
rectilinear architectural units are oriented in a northeast-southwest direction. Only the foundations are preserved in
some sections of the walls. In most parts, wall foundations consist of two parallel rows of large and medium-sized
stones with rubble fill in between. On the other hand, the floors of the central room and the two adjacent areas are
carefully and tightly paved with pebbles and medium-sized stones (Uysal 2017: 256, figs. 10-11; Uysal and <;if9i
2018: 571-572, figs. 12-13).
Level 6, documented in trenches Ml0 and Nl0, dates to the Late Bronze Age II, which corresponds to the middle
Hittite Empire period. The architectural remains of this phase are poorly preserved (Fig. 8-6b). A stone pavement
segment, oriented northeast-southwest, and disturbed wall foundations, were revealed in the eastern portion of the
excavated area. Wall foundation remains, a bread oven, and refuse pits were encountered in the western portion of the
excavation area (Uysal and <;if9i 2019b: 412, fig. 7).
Level 7, documented in trenches N9-Nl 0, dates to the Late Bronze Age I, which is the early phase of the Hittite
Empire period. The architectural remains of this phase are also poorly preserved (Fig. 8-6c). Similar to Level 6, Level
7 remains consist of a stone pavement segment, oriented northeast-southwest, and scanty remains of stone wall
foundations.
A notable find among the material culture remains of the Late Bronze Age, unearthed on the North Terrace, is a
large bronze cauldron found on the Level 5 stone pavement in trench Nl 1 (Fig. 8-6d). Such large metal vessels are
2015-2020 Excavations at Elbistan Karahoyuk 101
better known from the Old Assyrian Trade Colony period and Iron Age, while they are rarely attested at Hittite period
sites. So far, the bronze cauldron from Elbistan Karahoyuk is unique in the archaeological record of this period (Uysal
2017: 256, fig. 10, 2019: 102; Uysal and Cif9i 2018: 572, fig. 13).
Also notable are one stamp seal and one button
seal, both bearing hieroglyphic signs, found in the
Level 7 deposits in trench Nl0 (Fig. 8-7a). The stamp
seal is made of steatite; it has a rounded form, and the
hieroglyphic signs on its stamping surface have been
abraded due to frequent use. The button seal is made
of a purplish-brown colored stone and is circular and
flat. The button seal has two stamping surfaces
bearing hieroglyphs. One side is too damaged for the
signs to be intelligible, whereas three clear signs are
observable on the other side. Similar seals bearing
hieroglyphs were also found during the 1947
excavations (Ozgu9 and 6zgu9 1949: 42).
In general, Late Bronze Age ceramics of Elbistan
Karahoyuk display typical characteristics of ceramic
assemblages from other known Hittite period centers
in Anatolia. A spindle bottle in Red Lustrous
Wheelmade Ware found in Level 6 is significant for
showing the site's connections with the eastern Medi-
a terranean (Fig. 8-7b). Although Red Lustrous
Wheelmade Ware ceramics are widely attested at
b Hittite sites, a spindle bottle is a special vessel form
that is less frequently found (Mielke 2007: 155-165;
Schoop 2011: 254).
In addition to imported wares, another special
ware attested at the site is relief-decorated Hittite
ceramics. Three sherds with relief decorations were
found in the North Terrace area (Fig. 8-8). One of the
sherds was found in trench Ml0 during the removal
of Level 5 remains in a fill between Levels 5 and 6
(Fig. 8-8.a).
To produce the figure in low relief, first, the
ceramic paste was rolled into a thin plaque; then a
mold bearing the negative model of the figure was
impressed onto this plaque; and finally, the cut-out
Figure 8-7. a) Stamp seal and button seal from the 7th impression was attached to the surface of the ceramic
layer at the Northern Terrace Area at Elbistan vessel before drying. Most probably, this complex
Karahoyuk; b) Spindle Bottle from the 6th layer at the figure was impressed multiple times in a row on the
Northern Terrace Area. surface of a tall vase. The surface of the vessel is
coated with red slip, while the light-colored paste
of the impressed figure is left plain. The figure consists of three ammal tigures standmg on a two-row checkerboard-
paved floor. A lion walking towards the left with its head turned backwards is depicted on the left, and a gazelle
walking in the same direction facing back is depicted in the center. The figure on the right is a lion walking towards
the right; however, only the tail and the rear legs are preserved.
Within the repertoire of relief and applique-decorated Hittite ceramics, fragments with handmade figures are more
intriguing, because these figures belong to large narrative compositions. Two such fragments of Old Hittite vases that
date to the latest phase of the Middle Bronze Age were found at the site during the 1947 excavations (Ozgu9 and
6zgu9 1949: 38-40, pl. 47, 1-2; 6zgu9 1957: 70-71, fig. 3; Boehmer 1983: 42, figs. 12, 32). Also, two fragments
decorated with this technique were found in recent excavations in trenches N9 and 09. However, these fragments do
not come from secure archaeological contexts, as they were found during the excavation of topsoil.
The sherd found in trench N9 is red-brown-slipped and bears a male figure walking towards the right, depicted in
relief (Fig. 8-8.b). The man's long robe is light cream-colored. The top of the head, the lower part of the body below
the waist, and the tip of the raised left arm are broken. The position of this arm suggests that the hand may have been
formed in the gesture of supplication. The right arm is folded over the chest, and the hand is seen holding the hilt of a
staff or a weapon.
The sherd found in trench 09 is slipped in reddish clay; the applique consists of two figures walking towards the
right (Fig. 8-8.c). The figures are dressed in long robes in light cream color. Of the male figure in the front, only the
back of the head, a small portion of the body, and the right arm is preserved. A large portion of the body and the left
arm of the second figure is preserved, but its gender cannot be determined. The figure in the front extends a curved
arm holding the figure behind by the wrist. The fingers of the figure in the front are detailed by parallel lines, while
102 Chapter Eight: Uysal and (:if9i
the other figure's hand bears no details. The positioning of the arms in these two human figures is reminiscent of the
first two figures of the group of four in the second frieze of the inand1ktepe vase (Ozgu9 1988: 20, fig. 27, pl. H2, 47,
1, 54, 2).
Figure 8-8. The relief-decorated Hittite Figure 8-9. Architectural remains of the 8th layer from the
ceramics from the Northern Terrace Area Northern Terrace Area at Elbistan Karahoyuk.
at Elbistan Karahoyuk.
Level 8 was documented in trench N9 in the North Terrace (Fig. 8-9). Architectural remains of this phase are
orientated northwest-southeast and are built directly upon a layer of fluvial flood deposit. The floor level of this phase
in the northeastern sector of the trench is covered with a deposit of silt and small river pebbles that is densely packed
in certain areas and loose in others. This silty deposit is even thicker in the southwestern sector of the trench. The
foundation of a wall, built of small stones, was unearthed in the north, running across the length of the trench. A door
is located in this wall segment, close to the northwestern comer of the area. Door jambs are placed on both sides of
the doorway. The comer of another rectangular room, built of mudbrick walls on stone foundations, was unearthed in
the southeastern sector of the trench. Above the room, the extension of another wall foundation can be seen in the
section.
Trench N9 is located in an area of the mound where archaeological deposits have eroded and been severely
disturbed. Part of the Level 8 phase in this area was exposed directly below the topsoil. For this reason, finds from
this level, and especially the ceramics, are chronologically mixed. A reliable dating for this level can only be proposed
after the excavation of the same level in adjacent trenches. Only then will we be able to assess whether Level 8
represents the earliest phase of the Late Bronze Age or the latest phase of the Middle Bronze Age.
In the report on our reconnaissance survey conducted for assessing the current destruction at Elbistan Karahoyuk
prior to the initiation of excavations in 2015, we had indicated that an asphalt road encircles the skirts of the mound
and that archaeological deposits have been disturbed by soil removal especially on the northern side of the mound
(Uysal 2017: 255; Uysal and (:if9i 2017: 37). While this irreversible damage is most unfortunate, it also allows us to
draw certain observations about settlement sequence. The vertical section cut on the northeastern side of the mound
by bulldozers reveals fluvial deposits in the top layers, and a burnt level is visible further below. The silty deposit
visible in this section and the silty deposit encountered in trench N9 excavations below Level 8 are at the same
elevation. We believe that further investigation of this riverine deposit in future seasons has the potential to reveal
interesting observations about the settlement history of the mound.
2017-2018 HAMZATEPE EXCAVATIONS
In the 1947 Elbistan Karahoyuk excavations report, Hamzatepe is mentioned as a natural extension of the Soluk
Tepe hill further beyond, and it is indicated that the ridges of the rolling hills in this area are spotted with tumuli
(Ozgu9 and 6zgu9 1949: 16, 20). It seems that over this half century, time has taken a toll on Hamzatepe, where no
tumulus is visible today. The top of this rocky hill has been terraced, and a water depot has been built on top. Local
inhabitants informed us that ancient architectural remains were encountered during the construction of the terraces
and the depot. Albeit few, ceramic sherds are still visible on the hill and the vicinity. However, the northeastern aspect
of the hill that overlooks the mound presents a strikingly different view. Burnt human remains and ceramic sherds are
clearly visible on the surface on the northeastern slopes (Uysal 2019: 102; Uysal and <;if9i 2018: 572-573, fig. 14).
First, a sounding trench was opened on the northeastern side of the hill, which was soon enlarged horizontally
upon encountering urns immediately below the topsoil (Fig. 8-10). A cremation area with a grave encircled by stones,
an inhumation burial, 29 urns, and various architectural remains were revealed in trenches HlO, IlO, JlO, and KlO at
Hamzatepe. The majority of the human skeletal remains found scattered during excavations are burnt, but there are
also skeletal elements, which have not been exposed to fire, among the remains (Uysal and (:if9i 2019a: 399-400, fig.
8, 2019b: 413).
2015-2020 Excavations at Elbistan Karahoytik 103
HI,-. X'.
2 I~-~\~\
3 - ' ~~ \ \ \~\
-
I\
"'✓~
I \
"'
i \\
\
I\ I\
I
I './
Figure 8-10. Topographic plan ofHamzatepe.
Architectural remains encountered in the trenches consist of several rows of wall foundations built with small and
medium stones (Fig. 8-11 ). Close to the middle of the excavated area, the stone foundations of a wall built directly above
bedrock, oriented northwest-southeast, are clearly visible. A platform is located to the east of these wall foundations. The
western side of the platform follows a straight line, while it sits directly on bedrock on all other sides. The wall
foundations and the stone foundation of the platform are placed at a perpendicular angle to one another; however, their
relationship cannot be understood, since a wide area in between is severely disturbed (Uysal and <;:'.if9i 2019a: 400, fig. 8).
Figure 8-11. Hamzatepe excavation area.

104 Chapter Eight: Uysal and <;ifc,;i
Figure 8-12. Hamzatepe urns, numbers 16, 17, and 29.
Nine of the urns were found nearly intact (Fig. 8-12). While most of the remaining urns are broken, skeletal
remains have still been partially preserved in situ in them. The urns were situated directly on bedrock at a shallow
depth from topsoil and supported by small stones. The natural bedrock is compact in certain areas, while it is
calcareous and can easily be chipped and carved in other spots. There is no particular vessel form that was designated
to be used as an um. Except for the two open-mouth jars, the others urns consist of necked jars with globular forms.
No rim sherds were discovered during excavations. The majority of the jars have rounded bases, while fewer have a
flat or ring base. In the few examples that are preserved to a sufficient height for the handle joints to be observable,
the jars appear to have single or double vertical handles. No evidence was recovered that documented how the vessels
were closed. The distribution of the urns and the fact that neither the urns nor the wall foundations stratigraphically
cut one another suggest that the urns may have been buried in conformity to the layout of the structures (Uysal and
<;ifc,;i 2019a: 400, figs. 9- 12, 2019b: 413, figs. 11- 12).
As a result of the damage caused at the top of the hill, the archaeological deposits of Hamzatepe have washed
downslope. For this reason, excavated deposits are very mixed, and ceramics dating to multiple periods, from the
early 2nd millennium BCE to the Classical period, are found together. In addition to ceramic finds, material culture
remains encountered in the area with the cremation
burials include beads, spindle whorls, terracotta
weights, and arrowheads made of iron and bronze.
Grave goods accompanying the human remains are
attested only in four of the urns. However, it is
likely that some of the small objects found in the
area were originally placed in the urns that were
found broken during excavations. Indeed, some of
these finds show signs of exposure to fire, such as
discolorations, scorching, and vitrification (Uysal
and <;ifc,;i 2019a: 399-400, 2019b: 413).
The most intriguing find at Hamzatepe is the
cremation area that features a grave encircled by
stones, encountered in trench II0 (Fig. 8-13). At the
initial stage of excavations, the grave appeared as a
pile of stones measuring 70 cm in diameter. As
excavations progressed, this stone pile was
Figure 8-13. Hamzatepe cremation area and understood to be situated inside a "terracotta"
stone-surrounded grave. platform that is oriented northeast- southwest. The
2015-2020 Excavations at Elbistan Karahoytik 105
rectangular platform has rounded comers; it is 4 to 5 cm thick and measures 2 x 1.1 m. The grave encircled by stones
is built leaning against the upper part of the northeastern side of the platform. When the grave was opened, various
grave goods, including a ceramic bottle, three pieces of silver jewelry, and a button seal were found accompanying
cremated human skeletal remains (Fig. 8-14). The surface of the ceramic bottle is vitrified due to exposure to intense
fire. Similarly, beads and spindle whorls found in the grave show signs of exposure to high temperatures. Silver
jewelry items found in the grave display exquisite workmanship and mastery of the filigree technique (Uysal 2019:
102-103; Uysal and <;:'.ifi;;i 2019a: 401, figs. 13-15).
Figure 8-14. The finds from the stone-surrounded grave at Hamzatepe.

Ceramic evidence for the chronological period represented at Hamzatepe comes from two vessels found in um 17
and um 29 (Fig. 8-15). Both are examples of White Painted Ware, which is commonly attested in Cyprus and along
the eastern Mediterranean coast. According to the ceramic chronology of Cyprus, this ware group dates to the
Cypriot-Geometric III/Cypriot-Archaic I period, which corresponds to the Middle Iron Age in Anatolian chronology
(Gjerstad 1948: 48-49). Based on these imported vessels and typology of the local urns, Hamzatepe Cemetery can be
dated to the Middle Iron Age. At the same time, it is possible that some of the urns belong to the Late Iron Age, while
others yet possibly date to the Early Iron Age (Uysal and <;:'.ifi;;i 2019a: 401, fig. 12, 2019b: 413, fig. 12; <;:'.ifi;;i 2019).
Figure 8-15. Hamzatepe White Painted vessels.

106 Chapter Eight Uysal and <;if9i
During the 1947 Elbistan Karahoyillc excavations, two simple inhumations of adult individuals buried in hocker
position were discovered under the floors of houses in Iron Age levels (Ozgli9 and Ozgu9 1949: 22, 54- 56, pl. VII, 4-
5, LII, 1-4). These sub-floor graves indicate that at least some communities inhabiting the region during the Iron Age
practiced intramural burial. At the same time, however, excavations at Hamzatepe have revealed that extramural
cremation was also practiced contemporaneously.
We may be reminded here that in the archaeological record of the Iron Age cremation is a well-attested funerary
practice at Phrygian and Urartian sites. However, archaeological remains of cremation within the Neo-Hittite cultural
sphere have until now only been documented at Yunus Cemetery located at Carchemish in Gaziantep province
(Woolley 1939). Hamzatepe now represents the second example of a Neo-Hittite cremation cemetery. Findings from
Hamzatepe excavations suggest that the site was used as a local cemetery. Investigations in the vicinity have
documented urns located in other areas of the rocky outcrop, and further excavations will be planned in selected areas
in future seasons.
CONCLUDING REMARKS
Although renewed excavations at Elbistan Karahoyillc are only at an early stage, results have already been
significant and promising. The documentation of Late Bronze Age levels in a stratified archaeological sequence
below the thick Iron Age deposits in the North Terrace Area is a significant advancement for the archaeology of the
region. There is no other archaeological site with excavated Hittite period levels within the boundaries of
Kahramanmar:1§ province. As excavations progress deeper, documenting the earlier settlement sequence of the
mound, the site bears the potential to become a key reference site for the settlement history of the region. Apart from
the fruitful results of the excavations on the mound, the unexpected discovery of the extramural cemetery at
Hamzatepe has undoubtedly enriched the project's findings. As a preliminary observation, the presence of imported
(or imitated) wares both at Elbistan Karahoyillc and Hamzatepe is worthy of note, as it demonstrates that the Elbistan
Plain was connected with the eastern Mediterranean cultural sphere during this period.
Finally, we would like to note that the ongoing effort to establish a museum in Elbistan has been completed. The
Elbistan City Museum was opened to visitors on June 19, 2021. Archaeological finds from the 1947 expedition and
the recent excavations have now been transported to the city center of Elbistan district. Archaeological materials from
Elbistan Karahoyillc and Hamzatepe are exhibited in the land in which they were found.
REFERENCES CITED
Alparslan, Metin and Bora Uysal. 2018. Bin Rollsiegel mit anatolischen Hieroglyphen aus Elbistan-Karahoyillc.
Zeitschriftfiir Assyriologie und vorderasiatische Archiio/ogie 108(2): 235- 239.
Boehmer, Rainer Michael. 1983. Die Reliejkeramik von Bogazkoy, Grabungskampagnen 1906-1912, 1931-1939,
1952-1978. Bogazkoy-Battusa 13. Berlin: Gebr. Mann.
<;if9i, Ali. 2019. Two White Painted Vessels and Dating of Hamza Tepe (Elbistan-Karahoyillc) Cemetery. Colloquium
Anatolicum 18: 1-14.
Gjerstad, Einar. 1948. The Swedish Cyprus Expedition, 4/2. The Cypro-Geometric, Cypro-Archaic and Cypro-
Classical Periods. Stockholm: The Swedish Cyprus Expedition & Humanistiska Fonden.
Grothe, Hugo. 1911. Meine Vorderasien Expedition 1906 & 1907, 1: Die fachwissenschaftlichen Ergebnisse. Leipzig:
K.W. Hiersemann.
Hawkins, John David. 1993. The Historical Significance ofKarahoyillc (Elbistan) Stele. In Nimet Ozgilr; 'e Armagan.
Aspects of Art and Iconography: Anatolia and its Neighbors. Studies in Honor of Nimet Ozgiir;, M.J. Mellink, E.
Porada, and T. Ozgu9, eds., 273-279. Ankara: Turk Tarih Kurumu Bas1mevi.
Masson, Emilia. 1979. La stele de Karahoyillc-Elbistan: nouvel examen. In Florilegium Anatolicum, Melanges ojferts
aEmmanuel Laroche, 225- 241. Paris: Editions de Boccard.
Mielke, Dirk Paul. 2007. Red Lustrous Wheelmade Ware from Hittite Contexts. In The Lustrous Wares of Late
Bronze Age Cyprus and the Eastern Mediterranean, I. Hein, ed., 155-168. Papers of a Conference, Vienna 5th-6th
November 2004. Vienna: Austrian Academy of Sciences Press.
Nowicki, Helmut. 1981. Bemerkungen zur hier.-luw. Inschrift von Karahoyillc-Elbistan. Zeitschrift fur vergleichende
Sprachforschung 95(2): 251- 273.
Osten, Hans Heinrich von der. 1930. Explorations in Hittite Asia Minor 1929. Oriental Institute Communications 8.
Chicago: The University of Chicago Press.
Ozgu9, Tahsin. 1948. Elbistan Ovas1'ndaki Tetkik Gezileri ve Karahoyillc Kaz1s1. Belleten 12(45): 226-232.
- . 1957. The Bitik Vase. Anadolu 2: 57- 78.
- . 1988. inandzktepe, Eski Hitit <;agznda Onemli Bir Killt Merkezi. An Important Cult Center in the Old Hittite
Period. Ankara: Turk Tarih Kurumu Bas1mevi.
Ozgu9, Tahsin and Nimet Ozgu9. 1949. Tiirk Tarih Kurumu Tarafindan Yapzlan Karahoyiik Hafriyatz Raporu 1947.
Ausgrabungen in Karahoyiik, Bericht iiber die im Auftrage der Tilrkischen Geschichts-Komission im 1947
durchgefiihrten Ausgrabungen. Ankara: Turk Tarih Kurumu Bas1mevi.
2015- 2020 Excavations at Elbistan Karahoyuk 107
Pehlivanoglu, Tolga. 2019. La(hu)wazantiya: Ticari ve Dini Bir Merkezin Tarihs;esi ve Lokalizasyonu. Archivum
Anatolicum 13(2): 145- 172.
Quickelberghe, Etienne Van. 2013. Reflexions Autour de la Stele de Karahoyuk (Elbistan). Le Museon: Revue
d'Etudes Orienta/es 126(3--4): 253-263.
Schoop, Ulf-Dietrich. 2011. Hittite Pottery: A Summary. Insights into Hittite History and Archaeology, H. Genz and
D.P. Mielke, eds., 242-273. Leuven, Paris, Walpole: Peeters.
Uysal, Bora. 2017. Elbistan Karahoyuk'te Yap1lan Yeni Kaz1lar Hakkmda Genel Bir Degerlendirme. In Samsat 'tan
Acemhoyiik'e Eski Uygarlzklarzn jzinde, Aliye Oztan 'a Armagan. From Samosata to Acemhoyiik: Trailing the
Ancient Civilizations, Studies Presented to Honour of Aliye Oztan, S. Ozkan, H. Hury1lmaz, and A. Tlirker, eds.,
255-263. lzmir: Ege -Oniversitesi Bas1mevi.
- . 2019. Elbistan Ovas1'nda Yer Alan Bir Hitit Yerle~mesi: Elbistan Karahoyuk. In Uluslararasi Antik <;ag
Doneminde Mara~ Sempozyumu, 17-18 Kasim 2017, C. Kabakc1, 1. Solak, A. Ceylan, 0. Dumankaya, M. Canh,
and 0.Y. Ova, eds., 98- 104. Kahramanmaraf Kahramanmar~ Buyillc~ehir Belediyesi.
Uysal, Bora and Ali yifs;i. 2017. Elbistan Karahoyuk Kazis1 2015. 38. Kazi Sonuc;larz ToplantlSl 2: 35--46.
- . 2018. Elbistan Karahoyillc Kaz1s1 2016. 39. Kazi Sonuc;larz ToplantlSl 1: 569-578.
- . 2019a. Elbistan Karahoyuk Kaz1s1 2017. 40. Kazz Sonuc;lan ToplantlSl 1: 397--410.
- . 2019b. Elbistan Karahoyuk Kaz1s1 2018. 41. Kazi Sonuc;larz ToplantlSl 1: 411--421.
Woolley, Leonard. 1939. The Iron-Age Graves of Carchemish. University of Liverpool Annals of Archaeology and
Anthropology 26: 11-37.
Woudhuizen, Frederik Christiaan. 2003. The Luwian Hieroglyphic Inscription on the Stele from Karahoyuk-Elbistan.
Ancient West & East 2(2): 211-224.
Yigit, Turgut. 1997. Eski Anadolu Kentlerinden Luhuzatia/Lawazantiya'nm Tarihi ve Lokalizasyonu -Ozerine.
Belleten 41/230: 1-14.
CHAPTER NINE
SiRKELi H6YUK:
INSIGHTS INTO THE ARCHAEOLOGY OF BRONZE AND IRON AGE CILICIA
MIRK0 NOVAK, DENIZ YA~iN, MIRC0 BRUNNER, SiNEM HACI0SMAN0GLU, EKIN

K0ZAL, SABINA KULEMANN-OSSEN, HANNAH MbNNINGH0FF,
SUSANNE RUTISHAUSER, ALEXANDER E. S0LLEE, AND SbNKE SZIDAT
Sirkeli Hoyiik is one of the largest Bronze and Iron Age sites in Plain Cilicia (Greek Kilikia Pedias, Latin Cilicia
Campestris), a fertile, water-rich, alluvial landscape in the south of present-day Turkey. The region was known under
various names during the Bronze and Iron Ages: Kawa, Kizzuwatna, Hiyawa, Qawa, Que. Situated at the interface of
larger neighbouring regions like Anatolia, Syro-Mesopotamia, the Levant, and Cyprus, its status changed several
times. At different times, it represented either an independent principality, a vassal state, or an integral part of great
empires like those of the Hittites, the Mittani, the Assyrians, or the Babylonians. As one of the key sites in Cilicia, the
research at Sirkeli Hoyiik aims at shedding light on the colourful cultural history and material expressions of this
region.
0
co
M
0
M
0
0 r-...
M M
0
r-...
M
0
r-...
0 M
r-...
M
0
M
0
0 ID
M M
0
ID
M
A
40 20 0 40
--====---
■ ■=:::::i_ _ _ _ km
3800 ~ site
~ Om ,.. CITY
Figure 9-1. Central places in Cilicia during the Bronze and Iron Ages, showing the position of Sirkeli Hoyiik.
Map by Susanne Rutishauser (2021, April 7). Zenodo: http://doi.org/10.5281/zenodo.4670332).
The site is situated 40 km east of Adana on the left bank of the Ceyhan River, precisely at the point where the river
passes through the Misis Mountains (Fig. 9-1 ). It is well known as the location of a relief showing the Hittite king
Muwattalli II (ca. 1290-1272 BCE) and was excavated successively by John Garstang in 1936 (Garstang 1937 and
Sirkeli Hoyuk 109
1938), Barthel Hrouda in 1992-1996 (Hrouda 1997), and Horst Ehringhaus in 1997 (Ehringhaus 1999). The current
project was initiated in 2006, first as a cooperation between the universities of Tubingen and Canakkale. In 2011, the
project was transferred to Bern University (Novak et al. 2019).
Investigations since 2006 have focused on a range of topics including urbanism, chronology, and cultural contacts
of the site, as well as its position within the surrounding settlement system. The results of the current project have not
only elucidated the settlement's historic and cultural development, but also contribute valuable information regarding
the cultural history of the entire region.
OUTLINE OF PREVIOUS WORK1
The actual mound measures 350 x 300 m and rises about 30 m above the surrounding plain (Fig. 9-2). It is
subdivided into an almost circular higher section in the south, the so-called "Inner Citadel," and the lower, trapezoid
"Outer Citadel" to the north. The latter is also referred to as the "Plateau." Its eastern flank is formed by a rock ridge, a
spur of the Misis Mountains, running northwards up to the southern bank of the Ceyhan River (Novak 2021). Two
rock reliefs from the Hittite Imperial period, the better preserved one depicting Great King Muwattalli II (ca. 1290-
1272 BCE), were carved onto the ridge's eastern face at the mound's northeastern tip (Ehringhaus 2005: 95-101;
Kozal and Novak 2017; Fig. 9-3).
Sirkeli Hoyuk 2019
..
Topography
·.. ..... ..
. •.. .
: .. . ..
•
. .
.
..
. .
.
.... . ..--- . .
.. -------- Ceyhan
.....
. .. ..
,.
................. ...
•
0 200
Figure 9-2. Plan of Sirkeli Hoyuk, with red marks showing excavation areas of
Garstang, Hrouda, and Ehringhaus, and blue marks showing recent excavation sectors
(Susanne Rutishauser, Sirkeli Project, Bern University).
Numerous smaller trenches distributed all over the citadel mound were opened during the 1936 (Ahrens 2014) and
1992-1997 excavations (cf. plan in Novak et al. 2019: 249, fig. 206), but only one larger building was discovered and
partly exposed close to the rock reliefs in what is today called Sector E (Ehringhaus 1999). The current project has
been focusing on fewer areas, which were, however, excavated both on a larger scale and more intensively regarding
the documentation of artefacts and building remains, as well as conducting the systematic collection of samples for
scientific analyses (Novak et al. 2019). In the lower city, the city wall and a city gate are being investigated in
Sector F; in addition, a small sounding was dug at the northern foot of the citadel (Sector B) in order to confirm the
course of the city wall suggested by the geophysical survey. In the Outer citadel area, larger buildings were at least
partially uncovered in Sectors A in the northwest and C in the centre; furthermore, Sector E was created to gather
further information on the large stone building that had previously been discovered during the 1992-1997 excavations
just next to the rock reliefs. Another large building was uncovered in Sector D on the hilltop of the Inner citadel,
Building D 1, centred on a courtyard.
1
This section is authored by Mirko Novak and Deniz Yll{iin Meier.
110 Chapter Nine: Novak et al.
Figure 9-3. Rock relief ofMuwattalli II (Alexander Ahrens, Sirkeli Project, Bern University).
From the beginning, the current project's excavations were accompanied by non-invasive methods such as remote
sensing, as well as geophysical and field surveys (Sollee et al. 2018; Novak 2020). This interdisciplinary approach has
provided evidence that the Bronze and Iron Age settlement was a complex urban landscape consisting of a fortified
citadel on the main mound and an extensive walled lower town. Outside the walled city were workshop areas with
ovens. The citadel was not only separated physically from the lower city, but was also subdivided into two parts by
means of a wall and a height offset. Surveys and geomagnetic prospections on the opposite side of the river, conducted
by a Freiburg and Munich team directed by S. Halama, revealed an extended suburb, occupied during the 2nd and 1st
millennia BCE (Halama et al. 2019).
According to the results as of 2017, Buildings Al in Sector A and Dl in Sector D were obviously Bronze Age
foundations that were modified and re-used in the Iron Age. While Building Dl was apparently abandoned towards
the end of the 8th century BCE, the area of Building A 1 remained occupied until the Hellenistic period. The transition
from the Late Iron Age (Late Babylonian and Persian periods) to the Hellenistic period can be identified in Sector C,
where a large building, referred to as Building Cl, was erected at the beginning of the Hellenistic period. Settlement
traces from the period after the 1st century BCE are not attested at Sirkeli Hoyuk.
Sirkeli Hoyillc 111
Since 2017, continuing archaeological investigations in all areas produced further information on the development
of all the areas that had been excavated to that point. Moreover, there now is clarity about the southern boundary of
the settlement and its relation to the large rock spur of Beks;i Kultibesi located just southwest of the main mound, two
issues that the project had not been able to tackle until the 2017 campaign.
CHRONOLOGY
Cilician Regional Chronology
A series of workshops held between 2014 and 2017 led to the creation of a comparative stratigraphy of all Bronze
and Iron Age sites in Cilicia (Cilician Chronology Group 2017), which serves as the foundation of a regional
chronology that uses archaeological material and scientific data, as well as historical information, and follows the
systematic terminology of the ARCANE project (Novak et al. 2020). The 2nd and 1st millennia BCE are divided into
the Old-Cilician (OCI), Middle-Cilician (MCI), Neo-Cilician (NCI), and Late-Cilician (LCI) periods; each one of
these periods is divided into further subperiods: OCI covers the time span between ca. 2050 and ca. 1560 BCE, MCI
ca. 1560 to ca. 1190, NCI ca. 1190 to ca. 330, and LCI ca. 330 to ca. 50 BCE (Table 9-1).
Period Dates Conventional Sirkeli Hoyiik

ECI4 2420-2250 EBAIVa
ZXI
ECIS 2250- 2050 EBAIVb
OCil 2050- 1950
MBAI
OCI2 1950-1700 Z X I P VII / U VIII
OCI3 1700-1560
MBAII
MCil 1560-1522
ZIX/PVI
MCI2 1522-1420
LBAI
MCI3 1420-1350 ZVIII/PVI
MCI4 1350-1190 LBAII ZVII/ P VI
NCil 1190-1130 LBA III / IA Ia Z VII /P VI/ UVII
NCI2 1130-950 IA lb ZVI/PV /UVI
NCI3 950-720 Z V /UV-III
IAII
NCI4 720- 609 PIV/UII
NCIS 609-539
IA III ZIV/PIII/UI
NCI6 539-330
LCI 1 330-50 Hellenistic Z 111-11
Table 9-1. Cilician Chronology.
Radiocarbon Dates2
During the 2013- 2016 excavations at Sirkeli Hoyillc, charcoal samples were taken from the various sectors and
assigned to individual architectural phases, construction periods, and cultural stages on the basis of stratigraphic
sequences (Szidat 2019). These charcoal samples were chemically cleaned by removing all remaining soil material
and converted into graphite through combustion and a reduction. The graphite was then analysed for radiocarbon
(C14) measurement using an accelerator mass spectrometry in the Laboratory for Analysis of Radiocarbon with AMS
(LARA) at the University of Bern (Szidat et al. 2014). The validation of the resulting radiocarbon dates was carried
out in a separate study with wood samples whose ages had previously been determined by dendrochronology. These
measurements confirmed that the dating results are correct within the stated measurement uncertainties.
2
This section is authored by Sonke Szidat, Mirco Bnumer, and Susanne Rutishauser.
Period Dates Sample Code Lab Unkalibr.

14C-Age
(BP)
OCI-MCI2 1350-1190 Si16-D0336 BE- 3343±20
SE-D0281 6014.1.1
Sil6-A0039 BE- 3338±20
SE-A0500 6020.1.1
Si16-A0054 BE- 3223±20
SE-A0508 6022.1.1
Si13-D0182 BE- 3191±20
SE-D0204 6005.1.1
NCI 1 1190- 1130 Si15-A0137 BE- 2964±19
SE-A0479 6018.1.1
Si16-D0382 BE- 2929±20
SE-D0399 6011.1.1
Si15-A0153 BE- 2916±20
SE-A0479 6019.1.1
NCI2 1130-950 Si16-D0324 BE- 2934±20
SE-D0340 6012.1.1
Si15-D0276 BE- 2929±20
SE-D0287 6010.1.1
Si15-D0287 BE- 2917±20
SE-D0287 6009.1.1
Sil3-A0096 BE- 2903±20
SE-A0460 6016.1.1
Si16-D0274 BE- 2901±20
SE-D0372 6015.1.1
Si15-D0196 BE- 2897±20
SE-D0261 6007.1.1
Si15-D0222 BE- 2873±20
SE-D0268 6008.1.1
Si16-D0345 BE- 2855±20
SE-D0396 6170.1.1
NCI3 950-720 Si16-D0325 BE- 2875±33
SE-D0136 6013.1.1
Si13-D0172 BE- 2880±20
SE-D0198 6006.1.1
Si15-F0186 BE- 2844±20
SE-F0204 6024.1.1
Si16-F0260 BE- 2758±43
SE-F0293 6030.1.1
Si16-F0288 BE- 2775±20
SE-F0302 6032.1.1
Si16-F0242 BE- 2774±20
SE-F0278 6029.1.1
Si16-F0272 BE- 2756±20
SE-F0295 6031.1.2
Si16-F0192 BE- 2743±20
SE-F0279 6027.1.1
Si15-F0204 BE- 2739±20
SE-F0210 6025.1.1
NCI4 720-609 Si16-F0176 BE- 2763±20
SE-F0269 6026.1.1
Si16-F0202 BE- 2717±34
SE-F0269 6028.1.1
Table 9-2. Samples and C14 dates.
Table 9-2 shows the radiocarbon dates in detail. The measurement results are given as uncalibrated Cl4 ages (BP)
in years before 1950. A calendar age is determined from these raw data using the current international calibration
Sirkeli Hoyuk 113
MCI 3-4 NCI 1 NCl2 NC 3 NCI 4
0.020
0.015
0.005
0.000
-1500 -1000
KOE_MCl3-4 ■ KDE_NCl1 ■ KDE_NCl2 ■ KOE_NCl3 ■ KDE_NCl4
Figure 9-4. KDE of radiocarbon data of different phases from Sirkeli Hoyuk
and their correspondences with or deviations from historical data.
curve IntCal20 (Reimer et al. 2020). The calibration of the ages was carried out with a confidence interval of 95 .4%.
In Fig. 9-4, these calendar ages are also compared with the time spans of the corresponding chronological periods. For
this purpose, a Bayesian Sequence Phase Model in OxCal was chosen (Bronk Ramsey 2009). To highlight the
individual phases, the modelled kernel density was calculated for each phase, visualised with OxcAAR, and displayed
on a timeline (Bronk Ramsey 2017; Hinz et al. 2021).
For the most part, the samples from the same chronological periods show a fairly high temporal homogeneity and
can be assigned to intervals covering 160 to 370 years. The samples analysed here range from 1668 to 1427 BCE for
Middle-Cilician 3---4, from 1257 to 1084 BCE for Neo-Cilician 1, from 1132 to 1026 BCE for Neo-Cilician 2, from
1075 to 846 BCE for Neo-Cilcian 3, and from 898 to 816 BCE for Neo-Cilician 4. Fig. 9-4 clearly shows, on the one
hand, that for the Neo Cilician 1, 2, and 3 periods, the kernel densities of the calculated phases from the relative
regional chronology are consistent. However, on the other hand, for the Middle-Cilician 3---4 and Neo-Cilician 4
periods, the radiocarbon dates are two to three centuries older than the historical classification for the respective
periods. It should be noted here that charcoal samples can be subject to the old wood effect, especially when whole
logs are burnt. In such cases, it can happen that the results of the analysis refers to the mean age of the wood and not-
as desired from an archaeological point of view-the time when the wood was burnt. The difference can be as much
as several centuries, which is why charcoals (as in the present case) should only be used for rough, not fine
chronological dating. It remains curious, however, why such a large old wood effect should have occurred in the
Middle-Cilician 3---4 and Neo-Cilician 4 periods, while it is practically negligible for Neo-Cilician 1, 2, and 3. At a
minimum, selective verification of the radiocarbon dating of the charcoals by analyses of short-lived plant material
such as seeds, leaves, or twigs is necessary to support the regional chronology presented here by absolute dating.
REPORT ON RESULTS FROM 2018 THROUGH 2020
The Outer Citadel: Sector A and Sector E3
During the 2nd millennium BCE, the two exposed ends of the flat trapezoid plateau of the Outer Citadel were built
up with larger buildings whose walls were largely made of stone. Building Al in the northwest was erected at the
transition between the OCI and MCI periods on older OCI-period predecessors (Fig. 9-5). In its original state, it had a
flagstone pavement attested in at least four rooms. In the course of the MCI period, the floor level was raised, and the
stone pavement was replaced by clay and plaster floors. Furthermore, some walls were removed and repaired. This
amounts to at least three construction phases that can be distinguished for Building Al during the MCI period. Two
large storage vessels were found embedded in the most recent floor, still in situ. The assemblage of this phase also
3
This section is authored by Mirko Novak.
included further ceramic vessels, as well as numerous metal objects, weaving weights, terra cotta figurines, and the
stone head of a composite figurine (Novak 2021: 110, fig. 12). In the NCI period, the building was first partially re-
used and later built over by smaller houses. Despite the fragmentary state of preservation of the later building remains,
it was possible to prove a continuous occupation of the area until the late 1st century BCE (LCI period). Parts of
inventories, consisting of vessels and various types of artefacts, were recovered from the NCI and LCI phases. The
walls of the NCI period consisted of mudbrick layers resting on stone bases. While the latter were still found in situ,
the remains of the mudbrick superstructures were discovered in a step trench north of Building Al, where they had
collapsed onto the northern slope of the citadel. Underneath the collapse of the mudbrick walls, several narrow stone
walls were found running parallel to each other at different heights along the slope. They were also aligned with the
terrain in this area and therefore must have been retaining walls forming a steep glacis during the NCI period.
Sirkeli Hoyiik
Sektor A N
09/201 9 A
Sirkeli Hoyiik
Sektor E N
09/2019 A
!Om
Figure 9-5. Ortho-image of Sector A with

--- 0 10m
Figure 9-6. Ortho-image of Sector E with Building El

Building Al (Marosch Novak, (Fabrizio Biitscher and Marosch Novak,
Sirkeli Project, Bern University). Sirkeli Project, Bern University).
At the northeastern tip of the plateau, Building El, located above the rock face with the reliefs, which had already
been exposed in the 1990s (Ehringhaus 1999), was investigated again during the 2018 and 2019 campaigns (Fig. 9-6).
The baulks that the excavations in the 1990s had left standing were removed layer by layer to gain a better
understanding of the building's stratigraphy. Additionally, the area to the west of the building was further exposed in
order to clarify whether Building El continues in this direction. It turned out that the actual stone building was
founded at the beginning of the MCI on top of the remains of an older building from the OCI. Moreover, evidence
now suggests that Building El was abandoned at the end of the MCI at the latest. The sparse remains of the NCI and
LCI periods show no relation to the older architecture, hence representing a significant difference from the situation
observed in Building Al. With the early dating of Building El, the question of its relationship to the rock reliefs and
to three shallow cups on the upper surface of the rock face arises again. Contrary to previous assumptions, the building
was not erected at the same time as the reliefs were placed, but must have existed for some time before the images
were carved into the face of the rock.
The Inner Citadel: Sector D 4
A larger building located on the highest point of the Inner Citadel, which was already recognised on satellite
images before the excavations began, is being investigated in Sector D (Fig. 9-7). It was founded in the 12th century
4
Sirkeli Hoyuk 115
BCE (early NCI 2) on MCI-period remains and remained in use until the late 8th century BCE (NCI 3), before it was
overbuilt by a squatter layer of NCI 4 date. Although NCI 5-6 and LCI 1 sherds attest to a more recent occupation of
the Inner Citadel, these have left no architectural traces, at least not in Sector D (Kulemann-Ossen and Monninghoff
2019). Several rooms have already been uncovered, and various floors have been identified. Among the installations,
two elaborately walled drainage channels stand out. Very extensive pottery inventories were found in several rooms of
the building, partly mixed with larger quantities of animal bones. The function of the building could not yet be
determined.
Sirkeli Hoyiik
Sektor D N
09/2019 A
10 m
Figure 9-7. Ortho-image of Sector D with Figure 9-8. Ortho-image of Sector F (J. Heim, S.
Building Dl (Marosch Novak, Sirkeli Project, Gur, and Alexander E. Sollee, Sirkeli Project,
Bern University). Bern University).
The Lower Town: Sector F, Southern and Western Lower Town 5
In 2012, a geophysical survey revealed that the ancient settlement was not confined to the hoyuk, but also
encompassed an extensive lower town to the southeast of the main mound. To gain information on the dating and
development of this part of the settlement, an excavation area named "Sector F" (Fig. 9-8) was opened at a spot where
the magnetometry survey suggested the presence of a city gate (cf. Fig. 9-2). Since then, archaeological work in this
area has elucidated the development of this area (Sollee 2019: 86-106; Sollee et al. 2020).
The oldest period of occupation identified in this area of excavation dates to the transition between periods OCI
and MCI (ca. 17th-15 th centuries BCE). Presently, the exposure of these structures may be far too limited to allow
further conclusions about them, but their discovery strongly suggests that the settlement already featured a lower town
during that time. This area was abandoned in the 15 th century BCE and was not resettled until the 11 th century BCE.
Following this initial, probably small-scale and very loose resettlement, the lower town was transformed into an
essential part of the ancient settlement during the 10th century BCE, when it was equipped with a 3-4 m wide
fortification wall. This was accompanied by the establishment of new houses and workshops within the city limits.
Traces of fire indicate that this period of occupation ended in a catastrophic event (perhaps an enemy attack), which
was followed by a profound reorganisation of the residential quarter. New houses were erected on top of the remains
of the older ones. Most notably, however, the fortification system was redesigned. It now featured two 2-3 m strong
fortification walls separated by an approximately 5.5 m wide empty space, as well as a four-pier gate, an arrangement
5
This section is authored by Alexander E. Sollee.
that finds close parallels to other Neo-Hittite sites like Karkami~ or Sam'al (cf. Woolley 1921; Koldewey 1898;
Naumann 1971: 291- 302). Large stone blocks that were probably quarried in the Upper Town (Bek<;:i Kulubesi; see
below) must have made the town defenses an impressive display of political and military power.
While the 9th century BCE fortification system generally maintained its appearance until the late 7th century BCE,
two major changes have been identified in the residential quarter. The first occurred probably in the earlier part of the
8th century BCE. At this time, a large building (F 1), constructed of massive stone walls and featuring a large courtyard
with two column bases replaced older, smaller houses. However, a ~ century BCE sherd layer covering its remains
suggests that Building Fl was already abandoned at the end of the 8th century BCE (NCI 3). The lower town was
finally abandoned in the late 7th century BCE (NCI 4), which coincides with, and may be related to, the demise of the
Neo-Assyrian Empire.
In 2017 and 2018, a geophysical prospection and a surface survey revealed that the lower town of Sirkeli Hoyuk
was not limited to the field southeast of the main mound, but continued southwards beyond the railroad tracks (cf. Fig.
9-2). This part of the lower town, which is referred to as the "southern lower town," was a densely occupied quarter of
the ancient settlement. Furthermore, at least two large streets, one probably coming from a gate at the southern end of
the quarter and the other one from the east, as well as a number of open spaces, can be identified. Further geoelectric
measurements indicate that the southern lower town was surrounded by the 9th- 7th century BCE fortification system.
This fits well with the results of the field survey, as the bulk of ceramics gathered from the surface of the southern
lower town date to periods NCI 3-4 (ca. 950-609 BCE). Hence, the magnetometry image probably displays the NCI
3-4 settlement. While the amount of later (i.e., Hellenistic and Roman) sherds found in the southern lower town
survey is negligible, material dating to the MCI period made up for a significant percentage of the collected ceramics.
Hence, the southern lower town must have already been occupied during the second half of the 2nd millennium BCE,
and indeed geoelectric profiles indicate at least two major layers of occupation in this area.
A third part of the lower town of Sirkeli Hoyuk was identified west of the main mound (cf. Fig. 9-2). A segment of
the double city wall representing this area's northern border had already been recognised earlier (Hubner and Hemeier
2019: fig. 9). Evidence gathered in 2018 suggests that it continued westward, forming a shallow arc connecting the
northwest ends of the main mound and Bek<;:i Kulubesi (cf. Fig. 9-15). In contrast to the other two segments, the
western lower town of Sirkeli Hoyuk was not densely occupied. Only a few buildings, as well as a number of streets,
can be made out on the magnetometry image. Since large amounts of NCI and LCI pottery were collected from the
surface in this area (Novak et al. 2019: figs. 44-45), the structures could date to either of those periods, but their
orientation advocates an NCI date.
The Upper Town: Sector O (Bekri Kuliibesi) 6
Bek<;:i Kulubesi is a mountainous outcrop located only a few metres southwest of the main mound of Sirkeli Hoyuk
(cf. Fig. 9-2). This rock spur is divided into three crests (north, west, and south) separated by small valleys. During his
work at Sirkeli Hoyuk, Barthel Hrouda had already surveyed this rocky elevation and documented several rock-cut
tombs and cist graves along its slopes (Hrouda 1998). This led to Bek<;:i Kulubesi being regarded as a Roman or
Hellenistic period necropolis. However, the results of the investigations headed by Alexander E. Sollee, in cooperation
with the main project, have demonstrated that the rock spur represented the settlement's Upper Town area during the
early 1st and perhaps already late 2nd millennia BCE. Most of the ceramic material collected from the surface dates to
the NCI period, which is therefore identified as the period during which the area experienced its most intensive
occupation. A substantial amount of MCI sherds hint at an even earlier start of settlement activities on the
mountainous outcrop.
Excavations exposed a large building (Building 01; Fig. 9-9) which featured a number of chambers that were cut
into the natural terraces of the bedrock. They were probably subterranean and may have therefore served as storage
basements. Additional rooms were created by constructing narrow stone-footed walls perpendicular to the terraced
bedrock faces. The material recovered from the building's floors, when some chambers had already been subdivided
by the addition of small stone walls, mainly dates to the period from the 8th to ~ century BCE (NCI 3-4). However,
since the earliest levels of Building 01 have not yet been reached, the edifice could turn out to have been founded
(considerably) earlier. Excavations just southeast of Building 01 furthermore revealed the remains of a street (Fig. 9-
10). It consisted of a mix of clay and crushed limestone that was applied onto the bedrock's surface and is therefore
similar to the floor of the gate exposed in Sector F.
Furthermore, the investigation of three previously looted tombs (Fig. 9-11) allowed us to document the structure of
the burial chambers. Each tomb featured a separate entrance shaft ("dromos") ending in a rectangular opening that led
into a rock-cut dome-like chamber with semicircular niches on three sides. In the centre of each chamber, just next to
the entrance, a rectangular shaft, reaching down about 1 m, was located. Interestingly, all three tombs were
interconnected through small holes. The purpose of these small openings, which would have been large enough for a
person to squeeze through, and were part of the original design, remains unclear for now. While no solid evidence
regarding the date of the tombs was gathered, their typological similarities with Late Bronze Age and Iron Age tombs
6
This section is authored by Alexander E. Sollee.
Sirkeli Hoyiik 117
0 1 2 3 4 5
I I I I
Figure 9-9. View of Building 01 at the end of the Figure 9-10. Ortho-image of Building 01 with
2019 campaign (Alexander E. Sallee, Sirkeli Project, nearby street (Alexander E. Sallee, Sirkeli
Bern University). Project, Bern University).
Figure 9-11. View of tombs 2--4 (Marosch Novak, Sirkeli Project, Bern University).
in Cyprus (Schuster Keswani 2012), the low amount of Hellenistic and Roman period material around them, as well
as missing traces of Roman or Hellenistic period tools, may suggest that the area had already been used as a cemetery
during the earlier part of the 1st millennium BCE.
The uncovering of the area surrounding the rock tombs led to a further unexpected discovery, as it turned out that
the tombs were cut into a former quarry (cf. Fig. 9-11 ). The dimensions of the rectangular cavities left by the removed
blocks correspond with those of the ashlars used to construct the city wall, which was partially exposed in the
southeast lower town ("Sector F," see above).
Investigation of the Rock Reliefs 7
A second Hittite relief, carved onto the same rock face as the relief of Muwattalli II, had already been discovered
in 1994 (Ehringhaus 2005: 95-101). It was suggested that the second relief had been chiselled out in antiquity. More
detailed investigations of the rock face, where the two reliefs are located, were carried out from 2017 to 2019. While a
team of the University of Naples, led by M. Marazzi, scanned the area using state-of-the-art technology, two stone
conservators (C. Schneider and K. Zimmermann) examined the erosion damage.
The documentation not only produced improved images of the two previously-known reliefs, but also provided
evidence that the rock face had been prepared for a third relief in between the two known ones. However, the third
relief was never executed (Marazzi et al. 2019). Moreover, the new scans prove that the second relief showed
Muwattalli's son and successor, Mursili III (Ur1Ji-Tessup). However, contrary to what has been assumed since the
discovery of this relief, this image was not deliberately chiselled out in an act of damnatio memoriae, but washed out
by natural weathering.
7
CERAMIC SEQUENCE
Pottery of OC/8
The Old-Cilician period dates between ca. 2050-1560 BCE and is hypothetically divided into three phases in terms
of historical developments, defined as Old-Cilician 1 (ca. 2050-1950 BCE), 2 (1950-1700 BCE), and 3 (1700-
1560 BCE). Throughout this period, the pottery
shows consistent characteristics (Fig. 9-12). Syro-
Cilician Painted Ware (Fig. 9-12.1-2), Standard
Ware (fine, medium, coarse), Red Slipped and
Burnished, Brown Slipped and Burnished, Grey
Burnished Ware, Red Gritty Ware, Mineral Cook
Ware, and Cypriot imports (White Painted and Red-
on-Black/Red) are represented in the OCI assemblage
of Sirkeli Hoyuk. Syro-Cilician Painted Ware has
two subtypes. One is a handmade type with red
gritty fabric, cream colored slip and red-brown
2
paint. The second is a wheelmade type without slip,
Si94-H0019.001 Si93/ 40 but with brown and red-brown paint. Geometric
motifs represent the dominant type of decoration at
Sirkeli Hoyuk. The painted, slipped, and burnished
wares are comprised only of table wares, including
bowls, cups, jars, and jugs used for eating, drinking,
and serving. The painted/slipped and burnished
wares are decorated versions of Standard Ware,
which share the same form typology. Thus, the most
common ware in the assemblage is the Standard
Ware, which has an undecorated smoothed surface
(Fig. 9-12.3). Regarding forms, bowls (hemispherical,
3 3 s-formed, carinated, pedestalled, conical), deep
Si17-E0039 Si94-H0149 bowls (carinated), one-handled cups, goblets, jugs
(plain rim and trefoil), bottles (high neck and
globular), jars (carinated and high neck), cooking
\Ocm l
pots (high neck and globular), and pithoi are
Figure 9-12. OCI pottery (Photos by Laura Simons, represented at Sirkeli Hoyuk during the OCI.
Sirkeli Proiect. Bern Universitv).
Pottery of MC/9
The Middle-Cilician Period dates to ca. 1560-1190 BCE and is divided into four phases: Middle-Cilician 1 (1560-
1500 BCE), Middle-Cilician 2 (1500-1420/1400 BCE), Middle-Cilician 3 (1420/1400-1350/1300 BCE), and Middle-
Cilician 4 (1350/1300-1190 BCE). Unlike during the Old-Cilician period, a number of changes can be observed in the
pottery assemblages of the Middle-Cilician phases. Middle-Cilician 1 essentially represents a continuation of Old-
Cilician pottery traditions, but the share of Standard Ware increases while Syro-Cilician Painted Ware decreases. A
new appearance in this phase are two rim fragments that bear cylinder seal impressions, tentatively identified as
Luwian Hieroglyphic inscriptions (personal communication, Annick Payne). Middle-Cilician 2 marks a clear break in
the typology of the period's pottery assemblage. In this phase, Syro-Cilician Painted Ware disappears completely and
is replaced by another elaborate ware. Furthermore, Black-Impressed Ware appears. This particular type was only
found in this phase. Standard Ware continues to be the prevailing ware. Deep conical bowls with internal rounded rims
are the typical form of this phase. In Middle-Cilican 3 and 4, deep conical bowls become rarer, whereas shallow bowls
with internal rims and plates with plain or raised rims become more common (Fig. 9-13). This transition from deep
bowls to shallow bowls parallels the developments observed in the Upper City of Bogazkoy (Muller-Karpe 1988:
161-162, pl. 50). Along with Standard Ware, Red Slip, Brown Slip, Banded Ware, Painted Ware, and Cooking Ware
are represented among the locally produced wares. Red Lustrous Wheelmade Ware and Cross-hatched Ware are
imports from Western Cilicia. White Slip II and Monochrome Ware represent Cypriot imports of the Middle Cilician
period. A distinct feature of the Middle Cilician Period pottery is the application of potter's marks on vessels before
firing (Fig. 9-13.3), usually on Standard or Banded Ware (for previous publications on Old and Middle Cilician
pottery see Kozal 2013, 2019a, 2019b, 2019c, 2020, forthcoming; Kozal et al. 2019a; 2019b; 2019c).
8
This section is authored by Ekin Kozal.
9 This section is authored by Ekin Kozal.
Sirkeli Hoyuk 119
Areal 6.2, various MCI-plates
Areal 6.2, various MCI-shallow bowls
-----
3
Si1 6-O0377
Figure 9-13. MCI pottery (Laura Simons, Sirkeli Project, Bern University).
Pottery ofNCP 0
The pottery repertoire of the Neo-Cilician periods 2-4 comprises a multitude of decorated and undecorated wares
that develop significantly from the 11 th to the 7th century BCE due to contacts with neighbouring regions and
expressions of local craftsmanship (Fig. 9-14). Since 2006, NCI pottery has been found at Sirkeli Hoyuk in all
excavation sectors. Within its large corpus, NCI 3 and 4 assemblages, especially those from the 8th century, are well
attested, while NCI 2 pottery is scarcely found. The continuity of the stratigraphic sequence is nevertheless mirrored in
the continuous development of pottery shapes and decoration, recently strengthened by radiocarbon dates from phases
NCI 2 and 3 (cf. Table 9-2 and Fig. 9-4).
Since 2018, we have correlated specific innovations in ceramic production and consumption with the chronology
of the site (Kulemann-Ossen and M6nninghoff2019; M6nninghoff2020). Excavation sectors D and F especially have
yielded adequate stratigraphic data to present hereafter some general remarks on the pottery repertoire and its
development.
10
This section is authored by Sabina Kulemann-Ossen and Hannah Mi.inninghoff.
NCI2
During the NCI 2 period (late 12th to mid-10th century BCE), undecorated wares were continuously modelled after
MCI types like inverted rim bowls or pots with everted rims. Similar developments were traced in the 12th and 11 th
centuries BCE at other Cilician sites like Tarsus-Gozlukule (Hanfmann 1963: 94) and Kilise Tepe (Bouthillier 2014:
152). In addition, several types of coarsely red or brown painted wares developed at that time. The moment when
Cypriot style painted pottery arrived at Sirkeli Hoyuk cannot be dated exactly. The oldest examples come from phase
9 in sector D. Straight-sided bowls, painted with dark brown or dark grey horizontal bands at the rim, represent the
most common type in this period (Fig. 9-14.1). In
t
the late 11 th or early 10th century BCE contexts
sherds of table ware are found alongside MCI type
undecorated wares.
Sil 8-D0276.003 Sil 5-D0007.041-044+095
NCI 3-4
During the 10th to 7th centuries BCE, a part of the
,
3
Si13-F0086 repertoire of Cypro-Geometric and Cypro-Archaic
4 pottery was assimilated into the local pottery
Sil 5-D0027.181+182
repertoire and reproduced with certain preferences,
forming a "Cypro-Cilician" style (Fig. 9-14.2, 4, 9,
) 12). The painted wares, first belonging to the White
5 Painted (WP) and Bichrome, later the Black-on-Red
Si1 7-F0128.010
(BoR) varieties, comprise a large part of the ceramic
\ assemblages especially in the 9th century BCE. The
J7 \ l
'' Fi
most common decorated vessel types are bowls of
a 6
I
7
different varieties and pouring vessels. Painted
larger transport or storage vessels are rare, however.
Si16-F0012 Si17-D0300.172+ 173+ 175
Cypriot styles replace many local Bronze Age
traditions of undecorated pottery, but some Cilician
traits develop parallel to this (Fig. 9-14. 7, 10 for
8
Sil 9-F0093.236 particular Cilician shapes of the 8th century).
Throughout the NCI an increase in popularity of red
slip surface treatment is fueled by the dissemination
9 10 of Red Slip traditions throughout the eastern
Sil 7-D0300.152 Sil 7-F0l 75.001 Mediterranean. A shallow bowl with horizontal
handles is the most common shape. The incised
decoration of Cypriot style fluted ware is
11 12 furthermore frequently combined with shallow bowl
Sil 9-D0122 Sil 8-50096.001
types with a red slip. The well attested 8th century
BCE pottery assemblage is characterized by further
assimilation of some Cypriot stylistic features of
._..___ vessel shapes and decorations. Furthermore, the
proportion of undecorated ware increases, and
entirely new vessel shapes are incorporated (Fig. 9-
13 14.11-12). Throughout all excavation areas, Late-
Si19-F0093.227
Assyrian and Post-Assyrian pottery is only rarely
Figure 9-14. NCI pottery (Gabriele Elsen-Novak, found in NCI contexts (Kulemann-Ossen 2019a:
Sirkeli Project, Bern University). 190, 2019b: 316).
Archaeometric Analysis of the Potteries 11
Several ceramic assemblages from Sirkeli Hoyuk were selected and investigated by using archaeometric methods
to improve the understanding of the development of the site's pottery assemblage as well as to gain further
information on interregional connections. Among the ceramic assemblages at Sirkeli Hoyuk, one of the selected
groups consists of MCI Plain Ware (n=37) and NCI Plain Ware (n=21), the most common pottery groups found in
both periods at Sirkeli Hoyuk. The goals are to find out whether the choice of raw material or production techniques
related to Plain Ware were standarised or changed during these periods. Furthermore, the archaeometric studies set out
to clarify whether the Plain Wares were produced locally or not, in order to shed light on the structure of the ceramic
industry in Cilicia during the late 2nd and early pt millennia BCE. Additionally, samples of NCI White Painted Ware
11
This section is authored by Sinem Hac10smanoglu.
Sirkeli Hoyuk 121
(n=42), one of the most common pottery types among Cypriot Style ceramics at Sirkeli Hoyuk, were selected to
investigate whether they were local imitations produced in Cilicia or imported from Cyprus, which would contribute
valuable information for the reconstruction of cultural-economic relations between Cilicia and Cyprus at that time.
Finally, specimens of NCI Assyrian ware (n=40) were selected for analysis. The goal here was to gather information
on socio-cultural relations between Cilicia and northern Mesopotamia during the Iron Age and to establish the degree
to which pottery production in Plain Cilicia may have been influenced by the Neo-Assyrian Empire.
Besides ceramic samples, clay samples (n=54) were systematically collected at different spots in the Ceyhan Plain
(Yukan Ova) to explore compositional characteristics of the possible local clay deposits, and to obtain clay reference
materials that can be used in further archaeometric studies of ancient ceramics in the region. In order to answer these
questions, archaeometric studies at Sirkeli Hoyuk aimed at (1) establishing raw material characteristics and
classifications, (2) determining the sources of raw material (provenance) of the selected potteries, (3) exploring the
procurement pattern of the raw material, (4) correlating continuity/discontinuity of production techniques during the
Late Bronze and Iron Age, and (5) tracing chronological differences regarding the use of clay sources and raw
material properties. For these purposes, the chemical composition of the selected samples was determined by LA-ICP-
MS (Laser Ablation Inductively Coupled Plasma Mass Spectroscopy). Mineralogical contents and textural analysis
were investigated using Thin Section Petrography, XRD (X-ray diffraction); SEM (Scanning Electron Microscopy)
was applied to obtain information on raw material characteristics and production processes (i.e. firing temperature) of
ceramics as well as for provenance analysis.
According to the preliminary results of the archaeometric analysis, which are being conducted in the framework of
a doctoral thesis by Sinem Hac10smanoglu (University of Tubingen), it seems that the MCI and NCI Plain Ware from
Sirkeli Hoyuk was produced from local calcareous clays available in the Ceyhan Plain (Hac10smanoglu et. al., in
preparation). Mineralogical and geochemical analysis indicate that the selected samples have almost the same
compositional characteristics, though some variations have been observed as well. This might be related to different
paste preparation processes related to vessel functions and/or due to the local compositional variation of the clay
sources used in their production (Hac10smanoglu et. al., in preparation).
CONCLUSIONS AND OUTLINE OF UPCOMING WOR.K12
~ • wall
- pit
- high-temperature area
- citywall
street
street-wall
- - - citywall
- ditch
100 200 300 400 500 600 m
'K/':.,..
~ ·{
) . .·
/"'~ ..~
... .
: ..... . .
~, "
. . ---.. .
_
~ J
Figure 9-15. Sirkeli Hoyuk during the Period NCI (1190-330 BCE) with Citadel, Lower Town, Upper Town,
Suburb, and workshop areas (Susanne Rutishauser and Joelle Heim, Sirkeli Project, Bern University).
12
This section is authored by Mirko Novak and Deniz Ya~in Meier.
The work of the 2018-2020 seasons at Sirkeli Hoyillc has yielded important new insights. Firstly, the picture of the
urban layout has been significantly emiched by the discovery of the Upper Town, which shows the extent and
complexity of the urban landscape in a new light (Fig. 9-15). A dynamic development of the Upper Town from quarry
to settlement area and burial ground can be observed. The complex cityscape is reminiscent in many aspects of the
structure of the Hittite capital ijattusa on the one hand, and the position and structure of the citadel of the urban
construction of the Iron Age cities of the "Neo-Hittite" culture on the other. Secondly, structures such as Buildings Al
and El for the 2nd millennium BCE, and Dl as well as the city wall with the gate for the pt millennium BCE, attest to
elaborate and partly monumental architecture that indicates a highlighted significance of the site, although it was never
the capital of the respective political entity in Plain Cilicia. Thirdly, the long occupation history and the observed and
documented stratigraphies in the various sectors offer the possibility of reconstructing the chronology of the entire
region and the development of its material culture in great detail. The find material also makes it possible to determine
the external cultural impulses and contacts to which Sirkeli Hoyillc and Cilicia were subjected in the course of the 2nd
and 1st millennia BCE. These reflected in no small measure the political conditions of the respective eras. It is hoped
that further research will provide more information on all these aspects. Moreover, the research at Sirkeli Hoyiik
demonstrates the potential of a multidisciplinary approach in exploring large, complex ancient cityscapes by using
invasive and non-invasive methods alike.
ACKNOWLEDGEMENTS
We would like to thank the Ministry of Culture of the Turkish Republic for their generous permission to carry out
archaeological research at Sirkeli Hoyillc and all their staff for their tireless support of our work. We would also like to
thank the Turkish Railway Company, which gave us the former railway station as an expedition house and supported
its conversion into a research centre. The city of Ceyhan helped us to set up the visitor park. We thank the Swiss
National Science Foundation for its generous support of the project. Furthermore, we thank the Fritz-Thyssen-
Stiftung, the University of Bern's Initiator Grant initiative, the Max-Freiherr-von-Oppenheim-Stiftung, the LMU's
Graduate School "Distant Worlds" as well as the Munchener Universitatsgesellschaft for providing additional funds
for the investigations carried out in the lower town and Upper Town areas. Ekin Koza! would like to extend thanks to
the Alexander von Humboldt Foundation, and Freiburg and <;anakkale Onsekiz Mart Universities for their support.
Finally, we thank all the staff, students, and workers from Turkey, Switzerland, and Germany for their dedicated work.
REFERENCES CITED
Ahrens, Alexander. 2014. John Garstang at Sirkeli Hoyiik, Cilician Plain, in 1936-1937. Anatolica 40: 47-60.
Bouthillier, Christina. 2014. From the Late Bronze Age into the Iron Age. In Further Work at Kilise Tepe, 2007-2011:
Refining the Bronze to Iron Age Transition, C. Bouthillier, C. Colantoni, S. Debruyne, C. Glatz, M.M. Hald, D.
Helsop, E. Koza!, B. Miller, P. Popko, N. Postgate, C.S. Steele, and A Stone, eds. Anatolian Studies 64: 148- 157.
Bronk Ramsey, Christopher. 2009. Bayesian Analysis of Radiocarbon Dates. Radiocarbon 51 (1): 33 7-360.
- . 2017. Methods for Summarizing Radiocarbon Datasets. Radiocarbon 59(6): 1809-1833.
Cilician Chronology Group 2017. A Comparative Stratigraphy of Cilicia. Results of the First Three Cilician
Chronology Workshops. Altorientalische Forschungen 44(2): 1- 37.
Ehringhaus, Horst. 1999. V orlaufiger Bericht i.iber die Ausgrabung auf dem Sirkeli Hoyillc, Provinz Adana/fi.irkei im
Jahre 1997. Istanbuler Mitteilungen 49: 83-140.
- . 2005. Gotter, Herrscher, Inschriften. Die Felsreliefs der hethitischen Groflreichszeit in der Tiirkei. Mainz: Verlag
Philipp von Zabern.
Garstang, John. 1937. Explorations in Cilicia. The Neilson Expedition: Preliminary Report I. Annals of Archaeology
and Anthropology ofthe University ofLiverpool 24: 52-68.
- . 1938. Explorations in Cilicia. The Neilson Expedition: Preliminary Report II. Liverpool Annals of Archaeology
and Anthopo/ogy 25: 12- 23.
Hac10smanoglu, Sinem, Mustafa Kibaroglu, Ekin Koza!, Hannah Monninghoff, and Joachim Opitz. In preparation.
Exploring the Impact of Socio-Political Transformations between The Late Bronze and the Iron Age on Ceramic
Production through the Archaeometric Analysis of Plain Ware from Sirkeli Hoyiik. Manuscript in Preparation.
Halama, Simon, Ekin Koza!, Natascha Kreutz, Martin Renger, and Deniz Y~in Meier. 2019. Pilotsurvey in der
Vorstadt. In Sirkeli Hoyiik. Ein urbanes Zentrum am Puruna-Pyramos im Ebenen-Kilikien. Vorbericht der
schweizerisch-tiirkischen Ausgrabungen 2006-2015, M. Novak, E. Koza!, and D. Ya§in, eds., 121- 146. Schriften
zur Vorderasiatischen Archaologie 13. Wiesbaden: Harrassowitz Verlag.
Hanfmann, George M.A. 1963. The Iron Age Pottery of Tarsus. In Excavations at Gozlii Kule, Tarsus, Vol. III, The
Iron Age, H. Goldman, ed., 18-332. Princeton, N.J.: Princeton University Press.
Hinz, Martin, Clemens Schmid, Daniel Knitter, and Carolin Tietze. 2021. oxcAAR: Interface to "OxCal" Radiocarbon
Calibration (version 1.1.0) https://CRAN.R-project.org/package=oxcAAR
Hrouda, Barthel. 1997. Vorlaufiger Bericht i.iber die Ausgrabungsergebnisse auf dem Sirkeli Hoyi.ik/Si.idti.irkei von
1992- 1996. Istanbuler Mitteilungen 47: 91- 150.
Sirkeli Hoyillc 123
- . 1998. Survey in der Umgebung von Sirkeli Hoyillc, 1994. In Light on Top of the Black Hill.· Studies Presented to
Ha/et r;ambel, G. Arsebillc, M. Mellink, and W. Schirmer, eds., 427-433. lstanbul: Ege Yaymlan.
Hubner, Christian, and Birthe Hemeier. 2019. Ergebnisse der geophysikalischen Prospelction. In Sirkeli Hoyiik. Ein
urbanes Zentrum am Puruna-Pyramos im Ebenen-Kilikien. Vorbericht der schweizerisch-tiirkischen Ausgrabimgen
2006-2015. M. Novak, E. Kozal, and D. Y~in, eds., 53- 60. Schriften zur Vorderasiatischen Archaologie 13.
Wiesbaden: Harrassowitz Verlag.
Koldewey, Robert. 1898. Die Architelctur von Sendschirli. In Ausgrabungen in Sendschirli IL Ausgrabungsbericht
und Architektur, Orient-Comite zu Berlin, ed., 103- 171. Berlin: DeGruyter (reprint 2013).
Kozal, Ekin. 2013. Exploring Sirkeli Hoyillc in the Late Bronze Age and its Interregional Connections. In Across the
Border: Late Bronze-Iron Age Relations between Syria and Anatolia, Proceedings of a Symposium Held at the
Research Center of Anatolian Studies, Ko<; University, jstanbul May 31- June 1, 2010, K.A. Yener, ed., 213- 224.
Ancient Near Eastern Studies 42. Leiden/Boston: Brill.
- . 2019a. Selctor A, Keramik der Kulturstufen OCI und MCI. In Sirkeli Hoyiik. Ein urbanes Zentrum am Puruna-
Pyramos im Ebenen-Kilikien. Vorbericht der schweizerisch-tiirkischen Ausgrabungen 2006-2015. M. Novak, E.
Koza!, and D. Ya~in, eds., 186-188 and 197-199. Schriften zur Vorderasiatischen Archaologie 13. Wiesbaden:
- . 2019b. Untersuchungen zur Keramik aus den Ausgrabungen der Jahre 1992 bis 1996. Keramik der Kulturstufe
MCI aus Areal 6/2 Nord aus der Phase 8/7. In Sirkeli Hoyiik. Ein urbanes Zentrum am Puruna-Pyramos im
Ebenen-Kilikien. Vorbericht der schweizerisch-tiirkischen Ausgrabungen 2006--2015, M. Novak, E. Koza!, and D.
Ya~in, eds., 263-270 and 275-287. Schriften zur Vorderasiatischen Archaologie 13. Wiesbaden: Harrassowitz
Verlag.
- . 2019c. Selctor D, Keramik der Kulturstufen MCI und NCI 1. In Sirkeli Hoyiik. Ein urbanes Zentrum am Puruna-
Pyramos im Ebenen-Kilikien. Vorbericht der schweizerisch-tiirkischen Ausgrabungen 2006-2015, M. Novak, E.
Koza!, and D. Ya~in, eds., 309-315 and 322-330. Schriften zur Vorderasiatischen Archaologie 13. Wiesbaden:
- . 2020. OCI Pottery from Sector F. In A Sallee et al., The Development of the Southeast Lower Town of Sirkeli
Hoyillc. A Preliminary Assessment Based on the 2013-2019 Campaigns. Altorientalische Forschungen 47(2): 243-
244.
- . Forthcoming. Late Bronze Age Pottery Assemblages from Sirkeli Hoyillc. In Ceramic Identities at The Frontiers of
the Empires. The Regional Dimension ofPottery Production in Late Bronze Age Northern Syria and Anatolia, S.
Mazzoni, M. Pucci, and F. Ventury, eds. Pisa: Edizioni ETS.
Koza!, Ekin and Mirko Novak. 2017. Facing Muwattalli: Some Thoughts on the Visibility and Function of the Rock
Reliefs at Sirkeli Hoyillc, Cilicia. In Questions, Approaches, and Dialogues in Eastern Mediterranean Archaeology:
Studies in Honor of Marie-Henriette and Charles Gates, E. Koza!, M. Akar, Y. Heffron, <;. <;ilingiroglu, T.E.
~erifoglu, C. <;aklrlar, S. Unlusoy, and E. Jean, eds., 371- 388. Alter Orient und Altes Testament 445. Munster:
U garit-Verlag.
Koza!, Ekin, Natascha Kreutz, Sabina Kulemann-Ossen, Ursin Raffainer, and Deniz Y~in Meier. 2019a.
Keramiktypologie und -entwicklung. In Sirkeli Hoyiik. Ein urbanes Zentrum am Puruna-Pyramos im Ebenen-
Kilikien. Vorbericht der schweizerisch-tiirkischen Ausgrabtmgen 2006--2015, M. Novak, E. Koza!, and D. Ya~in,
eds., 338-358. Schriften zur Vorderasiatischen Archaologie 13. Wiesbaden: Harrassowitz Verlag.
Koza!, Ekin, Natascha Kreutz, Sabina Kulemann-Ossen, and Deniz Y~in Meier. 2019b. Unterstadtsurvey. In Sirkeli
Hoyiik. Ein urbanes Zentrum am Puruna-Pyramos im Ebenen-Kilikien. Vorbericht der schweizerisch-tiirkischen
Ausgrabungen 2006-2015, M. Novak, E. Koza!, and D. Y~in, eds., 64-85. Schriften zur Vorderasiatischen
Archaologie 13. Wiesbaden: Harrassowitz Verlag.
Koza!, Ekin, Natascha Kreutz, and Deniz Ya~in Meier. 2019c. Pilotsurvey in der Vorstadt, Keramik. In Sirkeli Hoyiik.
Ein urbanes Zentrum am Purima-Pyramos im Ebenen-Kilikien. Vorbericht der schweizerisch-tiirkischen
Ausgrabungen 2006--2015, M. Novak, E. Koza!, and D. Ya~in, eds., 125-130 and 132-136. Schriften zur
V orderasiatischen Archaologie 13. Wiesbaden: Harrassowitz Verlag.
Kulemann-Ossen, Sabina. 2019a. Die assyrisierende Keramik aus Selctor A. In Sirkeli Hoyiik. Ein urbanes Zentrum
am Puruna-Pyramos im Ebenen KiliJ.rien. Vorbericht der schweizerisch-tiirkischen Ausgrabungen 2006-2015, M.
Novak, E. Koza!, and D. Y~in, eds., 190-192. Schriften zur Vorderasiatischen Archaologie 13. Wiesbaden:
Harrasowitz Verlag.
- . 2019b. Keramik der Kulturstufen NCI 2 bis NCI 5. In Sirkeli Hoyiik. Ein urbanes Zentrum am Purima-Pyramos
im Ebenen Kilikien. Vorbericht der schweizerisch-tiirkischen Ausgrabungen 2006- 2015, M. Novak, E. Koza!, and
D. Ya~in, eds., 315-321. Schriften zur Vorderasiatischen Archaologie 13. Wiesbaden: Harrasowitz Verlag.
Kulemann-Ossen, Sabina and Hannah Monninghoff. 2019. Hybridity of Style: Iron Age Pottery from Sirkeli Hoyillc.
Studi Micenei ed Egeo-Anatolici, Nuova Serie 5: 111-145.
Marazzi, Massimiliano, Natalia Bolatti Guzzo, and Leopoldo Repola. 2019. Neue Untersuchungen zu den Reliefs von
Sirkeli. Altorientalische Forschungen 46(2): 214- 233.
Monninghoff, Hannah. 2020. NCI Pottery from Sector F. In A Sallee et al., The Development of the Southeast Lower
Town of Sirkeli Hoyillc. A Preliminary Assessment based on the 2013-2019 Campaigns. Altorientalische
Forschungen 47(2): 244-260.
Muller-Karpe, Andreas. 1988. Hethitische Topferei der Oberstadt von Hattusa. Ein Beitrag zur Kenntnis spiit-
groj3reichzeitlicher Keramik und Topferbetriebe unter Zugrundelegung der Grabungsergebnisse vom 1978- 82 in
Bogazkoy. Marburger Studien zur Vor- und Fruhgeschichte 10. Marburg/Lahn: Hitzeroth.
Naumann, Rudolf. 1971. Architektur Kleinasiens. Von ihren Anfiingen bis zum Ende der hethitischen Zeit. Zweite,
iiberarbeitete Aujlage. Revised edition. Tubingen: Wasmuth.
Novak, Mirko. 2020. Sirkeli Hoyillc. The Discovery and Exploration of a Complex Urban Landscape in Iron Age
Cilicia. In Metallurgica Anatolica. Festschriftfar Onsal Yalr;zn anliisslich seines 65. Geburtstags, H.G. Yali;:m and
0. Stegemeier, eds., 209- 224. Bochum: Bergbaumuseum.
- . 2021. Sirkeli Hoyillc. A Central Site and Gateway Community in Kawa/Kizzuwatna/Hiyawa. In Nf!Ws from the
Lands of the Hittites, M. Marazzi, S. di Martino, and C. Mora, eds., Scientific Journal for Anatolian Research, Vol.
3-4,2019- 2020: 143- 168.
Novak, Mirko, Ekin Kozal, Sabina Kulemann-Ossen, and Deniz Y ~in Meier. 2020. Both Sides of the Amanus. Cilicia
and Amuq: A Comparative Chronology. In Alalakh and its Neighbors: Proceedings of the 15th Anniversary
Symposium at the Nf!W Hatay Archaeology Museum, June 10-12, 2015, K.A. Yener and T. Ingman, eds., 367-388.
Ancient Near Eastern Studies Supplement Series 55. Leuven: Peeters.
Novak, Mirko, Ekin Kozal, and Deniz Y~in Meier, eds. 2019. Sirkeli Hoyiik. Ein urbanes Zentrum am Puruna-
Pyramos im Ebenen Kilikien. Vorbericht der schweizerisch-tiirkischen Ausgrabungen 2006-2015. Schriften zur
V orderasiatischen Archaologie 13. Wiesbaden: Harrassowitz Verlag.
Reimer, Paula J., William E.N. Austin, Edouard Bard, Alex Bayliss, Paul G. Blackwell, Christopher Bronk Ramsey,
Martin Butzin, Hai Cheng, R. Lawrence Edwards, Michael Friedrich, Pieter M. Grootes, Thomas P. Guilderson,
Irka Hajdas, Timothy J. Heaton, Alan G. Hogg, Konrad A. Hughen, Bernd Kromer, Sturt W. Manning, Raimund
Muscheler, Jonathan G. Palmer, Charlotte Pearson, Johannes van der Plicht, Ron W. Reimer, David A. Richards,
E. Marian Scott, John R. Southon, Christian S. M. Tumey, Lukas Wacker, Florian Adolphi, UlfBuntgen, Manuela
Capano, Simon M Fahrni, Alexandra Fogtmann-Schulz, Ronny Friedrich, Peter Kohler, Sabrina Kudsk, Fusa
Miyake, Jesper Olsen, Frederick Reinig, Minoru Sakamoto, Adam Sookdeo, and Sahra Talamo. 2020. The
IntCal20 Northern Hemisphere Radiocarbon Age Calibration Curve (0-55 cal kBP). Radiocarbon 62: 725-757.
https://doi.org/10.1017/RDC.2020.41.
Schuster Keswani, Priscilla. 2012. Mortuary Practices and Burial Cults in Cyprus from the Bronze Age Through the
Early Iron Age. British School at Athens Studies 20: 313-330.
Sollee, Alexander E. 2019 Sektor F. Architektur und Stratigrafie. In Sirkeli Hoyiik. Ein urbanes Zentrum am Puruna-
Pyramos im Ebenen Kilikien. Vorbericht der schweizerisch-tiirkischen Ausgrabungen 2006- 2015, M. Novak, E.
Kozal, and D. Y~in, eds., 86--103. Schriften zur Vorderasiatischen Archaologie 13. Wiesbaden: Harrasowitz
Verlag.
Sollee, Alexander E., Susanne Rutishauser, Christian Hubner, Birthe Hemeier, and Mirko Novak. 2018. Fernerlcundung,
geophysikalische Prospektion und Ausgrabungen am Sirkeli Hoyillc (TR). Die Wiederentdeckung des antiken
Kummanni/Kisuatni. In Naturwissenschaftliche Methoden in der Archiiologie, Th. Burri and R. Stapfer, eds., 112-
135. Mitteilungen der Naturforschende Gesellschaft in Bern NF 75. Bern: Naturforschende Gesellschaft.
Sollee, Alexander E., Hannah Monninghoff, Ekin Kozal, Doga Karakaya, Joelle Heim, and Selin Gur. 2020. The
Development of the Southeast Lower Town of Sirkeli Hoyillc. A Preliminary Assessment Based on the 2013-2019
Campaigns. Altorientalische Forschungen 47(2): 215-285.
Szidat, Sonke. 2019. Ergebnisse der Radiokarbondatierungen. In Sirkeli Hoyiik. Ein urbanes Zentrum am Puruna-
Pyramos im Ebenen-Kilikien. Vorbericht der schweizerisch-tiirkischen Ausgrabungen 2006-2015, M . Novak, E.
Kozal, and D. Y~in, eds., 43-45. Schriften zur Vorderasiatischen Archaologie 13. Wiesbaden: Harrassowitz
Verlag.
Szidat, Sonke, Gary A. Salazar, Edith Vogel, Michael Battaglia, Lukas Wacker, Hans-Amo Synal, and Andreas
Turler. 2014. 14C Analysis and Sample Preparation at the New Bern Laboratory for the Analysis of Radiocarbon
with AMS (LARA). Radiocarbon 56: 561-566. https://doi.org/ 10.2458/56.17457.
Woolley, Charles Leonard. 1921. Carchemish. Report on the Excavations at Djerablus on Behalf of the British
Museum. Part 11: The Town Defenses. London: British Museum.
CHAPTER TEN
IRON AGE POTTERY FROM 0LUZ HOYUK,

NORTH CENTRAL ANATOLIA
MONA SABA
When observed across the entirety of central Anatolian cultures, Iron Age painted pottery culture emerged in the
12th century BCE, often referred to as the Dark Age inside the K1z1hrmak bend, and continued uninterrupted until the
Hellenistic period (2nd century BCE). This has been demonstrated by the archaeological finds uncovered in settlements
such as Bogazkoy, Alaca Hoyiik, <;ad1r Hoyiik, Eskiyapar, Ma~at Hoyiik, and especially at Oluz Hoyiik (Fig. 10-1 ).
The emergence of Iron Age pottery in the K1z1hrmak basin, including in the north central and central Anatolian
regions, seems to be related to the political developments of the period. It is understood that with the collapse of the
Hittite kingdom in 1180 BCE, ijattusa (Bogazkoy) became rather deserted. Contemporary with this abandonment
process are settlement traces found at Biiyiikkaya and the Temple District in the Upper City; these settlers were far
fewer in number than those who had abandoned Bogazkoy (Genz 2004; Parzinger and Sanz 1992: 33). The
architecture, pottery, and small finds that belong to the post-Hittite settlements at Bogazkoy are representative of a less
advanced culture compared to their predecessors.
KARADENiZ
.-
Sivas
/J ~
Figure 10-1. Left: general view of Oluz Hoyiik; Right: map of Iron Age settlements in the Kiz1hrmak basin.
The painted ware encountered in very small numbers at the beginning of the Early Iron Age at Biiyiikkaya (Genz
2004: Taf. 19) increases over the course of this period, emerging as the precursor of the painted ware that is termed the
Ali~ar IV group at the onset of the Middle Iron Age. It serves as a geographical indicator for the first manufacturing
period of the painted pottery tradition in the K1z1hrmak basin Iron Age. This process can be observed in Architecture
Layer 7A at Oluz Hoyiik, which includes small finds as well as this pottery (Donmez and Abazoglu 2018: 81-99,
2019: 237-260). Oluz Hoyiik, therefore, is far more informative on this ceramic and chronological issue than is
Bogazkoy.
A new architectural layer (7B) identified at Oluz Hoyiik in 2014, which we date to the Dark Age, speaks to the Hittite
collapse in the lands of the ijakmis Kingdom in the Ye~ihrmak basin, which forms the border of the ijatti Land and the Upper
Land. Evidence suggests that the collapse may not have occurred with a violent war or destruction. The pottery fragments,
characterized by painted decoration with thick bands and geometric and vegetal motifs (Donmez and Abazoglu 2018: fig.
21 ), medium burnished and self-slipped, were uncovered on the floor of a simple courtyard made of pressed earth in Layer 7B
(late 13th-early 12th centrny BCE) alongside sherds dated to the Hittite kingdom period (Donmez and Abazoglu 2018: fig.
19, 20). The fact that differing pottery groups were found together in Layer 7B, dated to the last years of the Hittite kingdom,
should be evaluated as a reflection of a cultural change in ceramics. The coexistence of old traditional and new culture pottery,
despite no traces of fire or destruction in the 7B architectural layer, indicates that post-Hittite occupation continued peacefully.
This occupational phase is known as the "Collapse Period of the Hittite Culture" at Oluz Hoyiik. The presence of these pottery
126 Chapter Ten: Saba
groups at Oluz Hoy(llc is very important in terms of showing that the "Hittite Kingdom Collapse Period" took place in
settlements in the middle and northern parts of the KlZllumak basin.
Excavatioos during the 2015 season, carried out in Architectural Layer 7A (11111-lOth centuries BCE), located above Layer
7B, are a first for Oluz Hoy(llc and the Amas.ya region. Here we discovered a wall belonging to a mudbrick building containing
a simple hearth (Donmez and Abazoglu 2018: fig. 22). The steep slope of the mound rests just north of the mudbrick
structure; the hearth was built just to the south in the interior of the structure. Due to erosion associated with the northern
slope, most of the wall and other remains related to the hearth have disappeared. In the Gordion 7B and 7A (1100-950 BCE)
layers dating to the Darlc Age, hearths were unearthed in single-room simple and coarse residences built with the pise technique
(Voigt and DeVries 2011: fig. 2.8a). In light of the discoveries at Oluz Hoy(llc, it appears that there are similarities between the
KlZtlumak basin sites and those west of the KlZtlumak in terms of building design in the Darlc Age, on the basis of the use of in-
room hearths.
Other than in Architectural Layer 7A at Oluz Hoy(llc, Darlc Age pottery has been found at Dogantepe (Donmez 2017: fig.
8a-b), Eskiyapar (Bayburtluoglu 1979: 293- 303), and <;adrr Hoy(llc (Ross 2010: fig. 5b). Architectural remains dating to the
Early Iron (Dark Age) are rare in the Kizilumak basin, until recently known only from discoveries at Bogazkoy-Buyiildcaya
(Genz 2000: 35-54, 2001: 1-2, 2003: 179-191, 2004), and at <;adrr Hoy(llc (Ross et al. 2019), the latter primarily industrial
in nature. Excavations carried out by J. Seeber at Buyiildcaya (Seeber 2010: 220-229) revealed the existence of a settlement
dating back to the 1211i_1011i centuries BCE. At the beginning of the Early Iron Age, the pottery from Bi.iy(llckaya was described
as having three characteristics; it was handmade, darlc colored and poorly burnished, with coarse clay, plant, and pebble
inclusions. The fact that these coarse vessels, which occur later than the better-made Hittite pottery, consist only of plain
examples, definitely indicates that there was no paint-decorated tradition in the early phase of the Dark Age, the exact length of
which cannot be detennined.
Although the date in the mid-Darlc Age period at which the paint decoration emerged at Bi.iy(llckaya is not known, the
approximate beginning of the 11 lh century BCE stands out in the stratification and chronological table prepared by Hermann
Genz (Genz 2004: tab. 1). The projected starting date (lllh century BCE) for the Bi.iy(llckaya Darlc Age pottery, usually
characterized by triangles with interior dots, zig-zags, and simple lines in light red or reddish-brown, which were painted on buff
or dark buff vessels, has not been documented in any publication. In addition to this dating problem, speculation that the
ceramics in question may have originated in the Early Bronze Age, or that they could be related to the Kaska people,
was offered first by Genz and then by Seeber (Genz 2001: 2; Seeber 2010: 226-227).
The period between the Early Bronze Age and the Dark Age paint-decorated traditions is a very long period of
about 1000 years. Such a long-lasting painted-pottery tradition in the KlZlhrmak basin is so far unknown. Therefore,
establishing a link between the Early Bronze Age II painted pottery tradition (Intermediate, Ali~~tr III, <;iradere, and
Delice) and the Bi.iyi.ikkaya Dark Age painted pottery, besides negatively affecting the Bogazkoy chronology, creates
a problematic and untenable situation. However, the suggestion that the Dark Age painted pottery should be
associated with Kaska identity is not a problem, in our opinion, and, on the contrary, is a correct approach.
It is a distinct possibility that the painted ware tradition found in Architecture Layer 7B at Oluz Hoyi.ik continued
into Architectural Layer 7A even if the decorative style changed, and even if a chronological gap intervened. The
Oluz Hoyi.ik 011 Architectural Layer pottery, dated to the beginning of the Middle Iron Age, which follows these
developments in the Early Iron Age, is identified as part of the tradition forming the prototype of Ali~ar IV pottery.
When this process is evaluated together with the developments at Bogazkoy-Bi.iyi.ikkaya, it demonstrates that the
Amasya-<;orum region is the place where the K1z1lrrmak basin Iron Age paint-decorated pottery and Ali~ar IV
tradition started.
It can be concluded that the tradition of painted pottery with triangular, festoon motifs, and figural decoration, with
its emergence during the Early Iron Age and uninterrupted continuation into the Hellenistic period, was a product that
emerged from the unique internal dynamics of the region, produced by societies living within the K1z1lrrmak bend.
Within the development of the KlZllrrmak basin painted-ware tradition, in terms of its basic features, including the
many compositions of geometric and vegetal motifs in colors of brown and buff, the triangles employed as decorations
since the Dark Age stand out in their distinctiveness. Triangle motifs continued to be used as a complementary
element and main motif of the pottery decoration in the 9 th and 8 th centuries BCE, i.e., the Middle Iron Age. Since the
8 th century BCE, the widespread use of the festoon motif in compositions or as the main decorative element in the
composition has also enriched the Iron Age painted pottery of the K1zilrrmak basin.
Simple triangular motifs filled with dots (Fig. 10-2), which are the forerunner examples of K1z1hrmak basin
painted ware, dating back to the 11 th- 10th centuries BCE, point to the extensive history of this type of painted
decoration at Oluz Hoyi.ik. With the recovery of similar examples of this decoration at Eskiyapar (Bayburtluoglu 1979:
293- 305), Bogazkoy-Bi.iyi.ikkaya (Genz 2004), and <;adir Hoyi.ik (Ross 2010: fig. 5b), as well as at Iron Age
settlements near Oluz Hoyi.ik, it is shown that the point of origin of the painted triangle ware is the Amasya-<;orum-
y ozgat region, i.e., the northern part of the K1z1hrmak bend. 1
Kraters with large triangle decorations (Figs. 10-3 and 10-4) which were uncovered at Oluz Hoyi.ik and dated to the
end of the Early Iron Age (first half of the 9th century BCE) show that this decoration was used in the transitional
1
Triangle decoration resembling Triangle-Ware is encountered in the Kmhrmak basin Late Iron Age settlements of Sulusaray
(6zcan 1992: 169, Res. 4 a- b, Sek. 2), Ma~at Hoyii.k (6zgu~ 1982: 60, Lev. 76/ 1 a--<l), Turhal Kalesi (Durbin 1971: Fig. 3/4),
Kaman-Kalehoyii.k (Omura 1992: Lev. V/5, VI/2, Res. II/6), Ali~ar Hoyii.k (von der Osten 1937: Pl. V/2), <;:engeltepe (Una! 1968:
126, Res. 46, Sek. 17, 20) and Zeyve Hoyii.k-Porsuk (Dupre 1983: Pl. 79/127, 83/ 176, 84/ 189, 90/238, 927255).
Iron Age Pottery from Oluz Hoytik 127
period to the Middle Iron Age (850- 650/600 BCE), in larger scale and on the larger vessels. The triangle motifs (Figs.
10-5 and 10-6), which were extensively used until the 6th century BCE, with differing design properties, began to
change, gaining common properties and becoming the main decorative element in the composition from this date
onward. Triangle decorations dating to after the 6th century BCE, with their ends dangling down and their insides
painted or decorated with diagonal cage motifs (Figs. 10-7 and 10-8), demonstrate that this design became the basis of
a new decorative regime in painted ware at the onset of the Late Iron Age. Examples of pottery decorated with this
kind of design are present at Oluz Hoytik by the 5th and 4th centuries BCE (Fig. 10-9).
0 5cm
Figure 10-2. Dark Age sherds from Oluz Hoytik.

Figure 10-3. Potsherd, Early Middle Iron Age, Oluz Hoyiik. Figure 10-4. Potsherd, Early Middle Iron Age,
Oluz Hoyiik.
Figure 10-6. Potsherd, Late Iron Age, Oluz Hoyiik.
Figure 10-5. Potsherd, Late Iron Age, Oluz Hoyiik.
Figure 10-7. Potsherd, Late Iron Age, Oluz Hoyiik. Figure 10-8. Potsherd, Late Iron Age, Oluz Hoyiik.
Iron Age Pottery from Oluz Hoyuk 129
Figure 10-10. Potsherd, early Middle Iron Age,

Oluz Hoyuk.
OLUZ07037
Figure 10-9. Cult vessel, Late Iron Age, Oluz Hoyuk. Figure 10-11. Potsherd, Late Iron Age, Oluz Hoyuk.
Figure 10-12. Potsherd, Middle Iron Age, Figure 10-13. Krater fragment, Late Iron Age, Oluz Hoyuk.
Oluz Hoyuk.
Besides triangle decoration, festoon decoration, which was not used in the 11 th century BCE, emerged as a
supplementary element to the composition (Fig. 10-10) at Oluz Hoyillc by the 8th century BCE. The festoon motif was
employed without the changing of its shape until the 4th century BCE, and was preferred as the main decorative
element (Fig. 10-11) in comparison to the changes occurring in the triangle design in the 6th-5th centuries BCE.
The painted-ware group known as Ali~ar IV is decorated with compositions of deer, ibex, and bird motifs in
mostly stylized and silhouetted technique, and concentric circles and rays, or sometimes geometric design
compositions. This ware constitutes examples of pottery in the KlZlhrmak bend and areas to its south in the Middle
Iron Age. Before the excavations at Oluz Hoyillc, other than at Ma~t Hoyillc (Ozgui;; 1982: pl. 69/4, 70/2-8z) and
Bolus-Aktepe (Ozgui;; 1978: pl. 73/ 1-3), no evidence of deer silhouette figures of Ali~ar IV type had been found at
sites in north central Anatolia. Although the Oluz Hoyillc excavations have provided important data on this subject, as
of yet, examples of vessels decorated with deer silhouette have been lacking. On the other hand, ray patterns used on
the neck of vessels decorated in Ali~ar IV style can be seen in Oluz Hoyillc levels dating to the 9th century BCE (Fig.
10-12).
Research has shown that Middle Iron Age workshops which produced uniform decorative techniques in that period
exhibited important changes in the styles of pottery emerging in the KlZlhrmak bend and regions to its south in the
Late Iron Age. For example, in place of the deer silhouette motif, figures of line decoration in varying styles, or with
more natural deer figures, made using reserve techniques, emerged.2 Deer figures in this later style, reflecting Ali~ar
IV techniques, have been found on a krater from the Oluz Hoyillc 2B Architectural Layer (Fig. 10-13), dated to the 5th
century BCE.
Oluz Hoyillc has emerged as an important settlement at which the triangular and festoon ware, as well as the Ali~ar
IV Style and Late Iron Age tradition, can be found in uninterrupted sequence in Iron Age stratified architectural layers
(7B, 7A, 6B, 6A, SB, SA, 4B, 4A, 3, 2B, and 2A), ranging from the late 2nd millennium BCE into the Hellenistic
period. The systematic archaeological excavations at Oluz Hoyillc, which began in 2007, have provided the first fully
stratified and traceable Iron Age sequence capable of offering substantial data on the painted pottery tradition in the
K1z1hrmak basin and its vicinity.
REFERENCES CITED
Akurgal, Ekrem. 1955. Phrygische Kunst. Ankara: Archaeologische Institut der Universitat.
Bayburtluoglu, Inci. 1979. Eskiyapar Frig <;ag1. Turk TarihKongresiBildirileri 8(1): 293- 305.
Bossert, Eva-Maria. 2000. Die Keramik Phrygischer Zeit von Bogazkoy: Funde aus dem Grabungskampagnen 1906,
1907, 1911, 1912, 1931-1939 und 1952-1960. Bogazkoy-ljattusa: Ergebnisse der Ausgrabungen 18. Mainz am
Rhein: Philipp von Zabern.
Donmez, ~evket. 2017. Amasya-0/uz Hoyiik. Kuzey-Orta Anadolu'da Bir Akhamenid (Pers) Yerl~mesi. 2009-2013
<;ah~malarz Gene/ Degerlendirmeler ve Onsonur;lar. Amasya: Arnasya Valiligi.
Donmez, ~evket and Fidane Abazoglu. 2018. K1z1hrmak Havzas1 Demir <;ag, <;anak-<;omlek Geleneginin Kokeni
-OzerineDu~unceler. TUBA-AR 23: 81-99.
- . 2019. Hitit Somas1 Kuzey-Orta Anadolu: Oluz Hoyillc'te Karanhk <;agile hgili Yeni Bulgular. In IX Uluslararasz
Hititoloji Kongresi Bildirileri/Acts of the JXlh International Congress ofHittitology, Volume I, 1. Cilt, A. Sue!, ed.,
237-260. <;orum: <;orum Valiligi.
Dupre, Sylvestre. 1983. Porsuk I. La Ceramique de /'Age du Bronze et de / 'Age du Fer. Memoires (Institut Frani;;ais
d 'Etudes Anatoliennes d ' Istanbul) 20. Paris: Editions Recherche sur !es civilizations.
Durbin, Gail E.S. 1971. Iron Age Pottery from the Provinces ofTokat and Sivas. Anatolian Studies 21: 91-124.
Genz, Hermann. 2000. Die Eisenzeit in Zentralanatolien im Lichte der Keramischen Funde vom Biiyiikkaya in
Bogazkoy/Hattusa. TUBA-AR 3: 35-54.
- . 2001. Bogazkoy'unErkenDemir<;ag1.Eskir;ag Bilimleri Enstitiisii Haber/er 12: 1- 2.
- . 2003. The Early Iron Age in Central Anatolia. In Identifying Changes: The Transition from Bronze to Iron Ages in
Anatolia and Its Neighbouring Regions. Proceedings of the International Workshop, B. Fischer, H. Genz, E. Jean,
and K. Koroglu, eds., 179- 187. Istanbul: Turk Eskii;;ag Bilimleri Enstitiisii.
- . 2004. Die Keramik der Eisenzeit. Funde aus den Grabungskampagnen 1993 bis 1998. Mainz am Rhein: Philipp
vonZabern.
Omura, Sachihiro. 1992. 1990 Y1h Kaman-Kalehoyillc Kazilan. Kazi Sonur;larz Toplanhsz 8(1): 319-336.
Ozcan, Birsen. 1991. Sulusaray-Sebastopolis Antik Kenti. Miize Kurtarma Kazzlarz Semineri 1: 261-308.
- . 1992. Sulusaray-1990 Kaz1s1. Miize Kurtarma Kazzlarz Semineri 2: 167-200.
Ozgui;;, Tahsin. 1971. Demir Devrinde Kiiltepe ve Civarz. Ankara: Turk Tarih Kurumu Yaymlarmdan.
- . 1978. Ma~at Hoyiik Kazzlarz ve <;evresindeki Ara~tzrmalar. Ankara: Turk Tarih Kurumu Yaymlanndan.
2
The settlements where !he painted ware of this style wilh deer decorations were uncovered are· Bogazkoy-Buyiikkale (Bossert
2000: Taf. 140/1287, 1289, 1290), Alaca Hoyiik (Akurgal 1955: Taf. 28/a), Ali~ar Hoyiik (von der Osten 1937: Fig. 42/c 89), Ma~t
Hoyiik (Ozgu~ 1982: Lev. 74/4, 8), Sulusaray-Sebastopolis (Ozcan 1991: Res. 16, Sek. 5), Buyiiktepe Hoyiik (Sagona 1992: Fig.
6/11) and Kiiltepe (Ozgil~ 1971: Lev. XX/1-2, XXI/1 a-b, XXII/1-2).
Iron Age Pottery from Oluz Hoyuk 131
- . 1982. Ma~at Hoyiik II. Bogazkoy 'iin Kuzeydogusunda Bir Hitit Merkezi. Ankara: Turk Tarih Kurumu
Yaymlarmdan.
Parzinger, Hermann and Rosa Sanz. 1992. Die Oberstadt von Hattusa. Hethitische Keramik aus dem Zentralen
Tempelviertel. Funde aus dem Grabungen 1982-1987. Bogazkoy-ljattu!fa: Ergebnisse der Ausgrabungen des
Deutschen Archaologischen Instituts 15. Berlin: Gebruder Mann Verlag.
Ross, Jennifer C. 2010. <;adrr Hoyuk: The Upper South Slope 2006-2009. Anatolica 36: 67-87.
Ross, Jennifer C., Gregory McMahon, Yagmur Heffron, Sarah E. Adcock, Sharon R. Steadman, Benjamin S.
Arbuckle, Alexia Smith, and Madelynn von Baeyer. 2019. Anatolian Empires: Local Experiences from Hittites to
Phrygians at <;adu Hoyuk. Journal ofEastern Mediterranean Archaeology and Heritage Studies 7(3): 299-320.
Sagona, Antonio. 1992. Excavations at Biiyuktepe Hoyuk 1991. Second Preliminary Report. Anatolian Studies 42:
29-46.
Seeher, Jurgen. 2010. After the Empire. Observations on the Early Iron Age in Central Anatolia. In ipamati
kistamati pari tumatimis. Luwian and Hittite Studies Presented to J. David Hawkins on the Occasion of His 70th
Birthday, I. Singer, ed., 220-229. Te!Aviv: Institute of Archaeology, Tel Aviv University.
Ona!, Ahmet. 1968. 1966 <;enge!tepe (Y ozgat) Sondaj1 Omaporu. Tiirk Arkeoloji Dergisi 15(1): 119-142.
Voigt, Mary M. and Keith DeVries. 2011. Emerging Problems and Doubts. In The New Chronology of Iron Age
Gordian, C.B. Rose and G. Darbyshire, eds., 23-48. Philadelphia: University of Pennsylvania Press.
von der Osten, Hans H. 1937. The Alishar Huyiik Seasons of 1930- 32. Part IL Chicago: Oriental Institute Press.
CHAPTER ELEVEN
RECENT WORK (2018-2019) AT PORSUK-ZEYVE HO YUK

IN SOUTHERN CAPPADOCIA
CLAIRE BARAT, EMiNE KOK.ER GbK<;::E, JEAN-FRAN<;::OIS PICHONNEAU,

VIVIEN MATHE, AND GUILLAUME BRUNIAUX
The Porsuk-Zeyve Hoyi.ik project (Nigde prefecture) in southern Cappadocia is the oldest French excavation still in
operation in Turkey. The archaeological site is situated on Zeyve Hoyi.ik, which belongs to the small southern
Cappadocian village of Porsuk (in the Nigde province, 14 km east of the subprefecture of Uluk:1~la), and rests at the
base of the central part of the Taurus Mountains (Bolkar Daglan) (Fig. 11-1) (Dupre 1983: 13; Pelon & Dupre 1987:
14). It is located 40 km from the Cilician Gates (Gi.ilek Bogaz1), the famous pass between the Anatolian plateau and
Cilicia (Lebreton 2013).
Figure 11-1. Map of the archaeological sites in Nigde province (Jean-Frarn;ois Pichonneau).
Recent Work (2018-2019) at Porsuk-Zeyve Hoyiik 133
The site of Zeyve is a tabular hoyiik, 400 x 180 m in size, situated at an average elevation of 1300 m asl, with an
area of 4 ha (Dupre 1983: 13; Pelon and Dupre 1987: 17). The hoyiik consists of a conglomerate table upon which
rests 8 m of archaeological strata (Dupre 1983: 13; Pel on and Dupre 1987: 18) (Fig. 11-2).
Figure 11-2. Aerial view of the Zeyve Hoyiik site (DAO Jean-Franc;ois Pichonneau).
The first to identify the site of Zeyve Hoyiik (without giving it a name) in Porsuk village was the Scottish
archaeologist W.-M. Ramsay. He visited the area in 1891 and again in 1902 (Ramsay 1903: 401--403; Dupre 1983:
13-14; Beyer 2012: 47). He identified an archaeological site thanks to the presence of marble and cut stones, but he
was not able to describe more than that due to the presence of crops. Having observed and deciphered three milestones
in the Porsuk cemetery, Ramsay identified the site as Colonia Faustiniana, founded in 176 CE by the Roman emperor
Marcus Aurelius, at the place where his wife, Faustina the Younger, died on their return from Syria (Ramsay 1903:
401--403; Pelon and Dupre 1987: 16).
The Swiss Assyriologist and Hittitology pioneer E.-O. Forrer visited the site in 1926 (Forrer 1937: 146-149; Dupre
1983: 14, 127-128; Pelon and Dupre 1987: 16; Beyer 2012: 47) and identified Hittite pottery fragments as "pre-
Roman," as well as the remains of a vaulted structure which appeared to be Hittite. Forrer identified the site as Dunna
(from Hittite texts), Tunna (from Assyrian sources), Dana (from Xenophon, Anabasis 1.2) and Tynna (from
Ptolemy1), next to the site of Faustinaupolis, founded by emperor Marcus Aurelius in 176 CE (Forrer 1937: 146-149;
Dupre 1983: 14, 127-128; Pelon and Dupre 1987: 16). After that, the site of Zeyve Hoyi.ik was not investigated by
archaeologists during the first half of the 20th century.
A new opportunity occurred in 1960 when a bulldozer, while building a track giving access to the neighboring
gypsum quarry, cut the western end of the hoyiik and substantially shaved the top off of it. The action exhumed a big
sandstone block bearing a Neo-Hittite hieroglyphic inscription (Dupre 1983: 14; Pelon and Dupre 1987: 17). The
inscription was brought to the Nigde museum in July 1960 (Beyer and Laroche-Traunecker 2017: 230, n. 6), and it
was there that E. Laroche (Strasbourg University) came to know of it (Beyer and Laroche-Traunecker 2017: 230).
After becoming director of the French Institute of Archaeology in Istanbul in 1964, Laroche took steps to obtain
excavation permission from the Turkish authorities for the site of Zeyve Hoyi.ik. Finally, after a survey campaign in
1968, at Laroche's request, 0. Pelon took responsibility for the archaeological mission of Porsuk-Zeyve Hoyi.ik in
1969 (Dupre 1983: 14-15; Beyer and Laroche-Traunecker 2017: 230).
The Neo-Hittite hieroglyphic inscription discovered at Zeyve Hoyi.ik in 1960 mentioned General Parahwaras, who
was pleased with himself due to the favor of the god Sarmas and that of King Masaurhisas; the inscription was
published in 1969 (Hawkins 1969: 99-109; Dupre 1983: 14) and dated to the second half of the 8th century BCE
(Beyer 2012: 47). The discovery of the inscription drew the attention of archaeologists to the Porsuk region. J.D.
Hawkins visited the region after the discovery of the inscription, and later, P. Meriggi (Padua University), visited
during his third trip to Anatolia in 1962 (Meriggi 1963: 283-284). Meriggi mentioned the discovery of a Latin
inscription on a now missing funerary altar (Meriggi 1963: 284) bearing the epitaph of the Roman centurion Titus
Sempronius Augustinus and dated to the reign of Marcus Aurelius (161-180 CE) (Tibiletti 1963: 300-303; Pelon and
Dupre 1987: 16).
In 1969, excavations began at Zeyve Hoyiik under the direction of Pelon, who opened three areas (called
"Chantiers") (Fig. 11-3): Chantier I, at the foot of the hoyiik, where a Late Roman building was brought to light;
1
Based on the mention ofTynna in the itinerary found on a grave stone (CIL, VI, 5076).
134 Chapter Eleven: Barat et al.
Chantier II on the upper western side of the hoyuk, partially damaged by a bulldozer in 1960, where a defense system
was discovered; and Chantier III, where an exploratory excavation was undertaken on the southern side of the site in
1968, which revealed stone foundations of the surrounding wall. During this 1969 campaign, fortification walls and a
gate system, known as the "Hittite postern," were removed. They are the most representative remains of the site. Pelon
led excavations at Porsuk from 1969 to 1977 (Pelon 1970, 1972, 1976, 1978, 1982), then from 1986 to 1989 (Pelon
1992), and then in 2002, totaling 15 campaigns (Pelon 2005; Beyer 2012: 47). During this period, Chantier IV was
opened at the northern end of the site, where Hellenistic and Roman remains and foundations of gypsum chambers
were brought to light.
00 01 0'2 0:, 104 05 06 07 08 109 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Ui 21 28 29 30 31 :n 33 34 35 36 31 38 39 40 41 42 43 44 45 46 41 48 49
Nord A
B
I C
[)
E
I'
G
fl
K K
L I.
M M
N N
0 0
Q Q
R R
s s
T T
u lJ
V
w w
X X
y y
z z
0 25 50m
00 01 02 d3 IM o·s 6 07 08 109 10 11 12 13 14 IS 16 17 18 19 20 21 Z2 23 24 25 26 27 28 29 :io 31 32 33 34 3; 36 :H 38 39 40 41 42 43 44 45 46 41 48 4
Figure 11-3. Map of the localization of the different "Chantiers"

(excavation areas) at Zeyve Hoyuk (after Beyer 2012: 47, Fig. 3).
Professor Dominique Beyer (Strasbourg University) succeeded Pelon in 2003 and led 13 archaeological campaigns
up to 2015 (Beyer 2006, 2007, 2008, 2009, 2011; Beyer et al. 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2012, 2013,
2014, 2015, 2016), continuing the excavations in the four areas previously opened by Pelon. In 2016, Claire Barat,
Associate Professor at Valenciennes and Hainaut-Cambresis University, now Hauts-de-France Polytechnic University,
became the director of the Porsuk-Zeyve Hoyuk excavation, with the collaboration of Dr. Emine Koker G6k9e, Hac1
Bekta~ Veli University, as assistant director; they led their first campaign in 2017 (Barat and Koker G6k9e 2019). This
chapter presents the 2018-2019 work at Porsuk-Zeyve Hoyuk (see also Barat et al. 2020).
DIGITAL INVENTORY OF THE ARCHAEOLOGICAL DEPOT AND ARCHIVES
One of the most urgent missions of the Porsuk-Zeyve Hoyuk archaeological team was the inventory of the
archaeological depot. This depot was built in 1970-1971, at the present excavation house location (Pelon 1972: 303-
304), during the first campaigns led by 0. Pelon. It was destroyed in 1984 by mine blast fragments from the
neighboring gypsum quarry (Pelon 1992: 306-308). A new depot was built out of reach of the mine blast fragments,
in the southern area of the hoyuk (Pelon 1992: 309, fig. 6). This depot contained all the mission's equipment and some
of the archaeological objects (architecture fragments, basalt millstone, and human bones). The ceramic material,
initially stored on the site, was brought to the Nigde museum in 1977 (Pelon 1992: 306, n. 8). The new depot,
consisting of two rooms, continued to be used under the direction of Beyer (2003-2015). A new excavation house was
built in 1993 on the western slope of the hoyuk, at the location of the former depot; one of its rooms was designated as
a depot, mainly for small objects.
During these forty-five years of archaeological excavations, no inventory of the excavation material had been
undertaken. A systematic and computer-based inventory of the depot began in 2018, under the direction of J.-F.
Pichonneau, an archaeologist in the Regional Service of Archaeology at the Regional Directorate of Cultural Affairs
of Nouvelle Aquitaine (Bordeaux), in the French Ministry of Culture and Communication. The mission applied the
inventory system following the norms of the French Ministry of Culture and Communication (Simon-Millot 2012).
The objects were repackaged, in accordance with Nigde museum standards, in integrated plastic boxes, type Curver
Recent Work (2018-2019) at Porsuk-Zeyve Hoyuk 135
Unibox 20 lt (432 x 350 x 165 mm)-29 lt (432 x 355 x 245 mm). These new containers were labeled using a
continuous numbering system, associated with one or more forms detailing their contents. The ceramic material was
repackaged in plastic bags. The faunal material was separated from the ceramic material and also repackaged in plastic
bags. The same procedure was applied to the other material, whatever its nature. Human fragments were treated and
packaged separately. The aim of the digital inventory is to facilitate access to excavation material for future
researchers.
In addition to the transition and transmission of the digital data during the change of directorship from Beyer to
Barat in 2015-2016, a number of the Pelon and Beyer mission excavation archives were also stored in the research
laboratory Archimede (UMR 7044, CNRS-Strasbourg University) (Beyer and Stahl 2015: 22-24), and not digitized.
Thanks to collaboration between the research laboratory CALHISTE (EA 4343, Valenciennes and Hainaut-Cambresis
University) and Archimede laboratory, an exhaustive digitization of the Porsuk-Zeyve Hoyuk excavation archives was
undertaken in 2018. A total of 18,901 documents were listed, classified, digitized, and indexed (1208 black and white
or colour photographs, 377 negatives, 1058 slides, 1528 drawings, and 14,730 digital photographs). The destination of
these digitized documents is a future open access database (Zanella et al. 2017).
MAGNETIC SURVEY 2
In 2018, a magnetic survey covered an area of 3 .2 ha, i.e., the whole of Zeyve Hoyuk, except for the areas opened
up by the Chantier II, Chantier II sud, and Chantier IV excavations (see Fig. 11-4). The aim of this non-invasive
survey was to map the remains of buried but near-surface archaeological structures, and thus to highlight the
organisation of urban development.
Figure 11-4. Magnetic anomaly map of Porsuk-Zeyve Hoyuk (Guillaume Bruniaux and Vivien Mathe).
The site is located at an altitude of 1300 m asl, shaped as an elongated hillock 16 m high, 400 m long (east/west)
and approximately 180 m wide (north/south). The hoyuk, which dips slightly to the east, is located at the confluence
of two rivers flowing to the northeast. The slopes are steep, except for a depression in the southeastern part of the site.
According to the geomorphological study by Kuzucuoglu (1997), Zeyve Hoyuk is composed of a thick layer of
conglomerate resting on a gypsum substrate. This conglomerate is covered by 8 meters of archaeological deposits. The
origin of the conglomerate seems to be fluvial (sediments laid by torrential water flows) and gravitational
2 This section was contributed by V. Mathe and G. Bruniaux.

(erosion/colluvial deposits). The petrography of the conglomerate is heterogeneous but associated with rocks
characterizing the Bolkar Daglan massiflocated south of the site. Its petrographic composition can be subdivided into
two categories: magnetic rocks (ophiolites, serpentinites, granodiorites) and non-magnetic rocks (marble, quartzite,
gypsite). These different rocks can be found in archaeological layers and on the surface.
The presence of materials with magnetic properties on or in the ground causes a local deformation of the earth's
magnetic field called a magnetic anomaly. Most archaeological structures that create associated anomalies are:
• burnt or heated structures such as hearths, kilns, and ceramic heaps;

• depressions such as ditches, pits, and postholes;
• wall foundations composed of stones that have a weaker magnetic signal than the ground (e.g. quartzite,
gypsite, and limestone);
• wall foundations composed of basic and ultra-basic rocks (e.g., basalts, gabbros, and serpentinites) which have
a high natural magnetization.
Magnetic prospecting mainly highlights magnetic anomalies less than 1 m below the surface because the magnetic
field strength decreases rapidly with the distance between the source and the sensor.
The magnetic survey protocol employed on Zeyve hoyuk uses a GSMP-35G magnetometer (GEM System)
consisting of two potassium vapour sensors with a sensitivity of 0.3 pT (manufacturer's reference). The sensors are
positioned in the horizontal plane, 50 cm apart and approximately 30 cm above the ground. The magnetometer records
the magnetic field at a rate of 20 measurements per second. The measurements are positioned by a GPS antenna
placed above the sensors. The acquired magnetic data are processed (Bruniaux et al. 2017) and then interpolated to a
10 x 10 cm grid using the kriging method.
Figure 11-5. Identification of the main anomalies visible on the magnetic anomaly map
(Guillaume Bruniaux and Vivien Mathe).
The various magnetic disturbances and features delineated in the acquisition protocol have been listed in Fig. 11-5.
It is essential to begin the list of anomalies by isolating those that are not of archaeological interest. This is notably the
case for a set of high intensity magnetic dipoles forming parallel and perpendicular straight lines (Fig. 11-5: black
circles). These magnetic sources correspond to metal stakes marking an old square grid (10 m square). The iron
structure of the excavation deposit and the numerous pieces of metal present in the vicinity of the construction
generate a very intense anomaly (Ml) extending over a radius of several tens of meters. A second very intense
Recent Work (2018- 2019) at Porsuk-Zeyve Hoyuk 137
anomaly (M2) extends to the east of Chantier II sud. It originates both from the fence surrounding the excavation area
and from the baked brick wall of the fortification.
To these anomalies created by highly magnetic and well identified sources, one must add the disturbances created
by the acquisition protocol and the data processing. Tangential to the boundary of the depression in the centre of the
site, the white dotted lines (Fig. 11-5) mark a strong discontinuity between a less magnetic zone (to the west) and a
more magnetic zone (to the east). This discontinuity is located at the boundary between two acquisition zones. It is
created by a variation in the height of the sensors in relation to the ground at a point where the slope is very marked.
Two other discontinuities are visible in the depression and mark the limits of the survey areas (black dotted lines).
These two boundaries are artefacts of the processing carried out to correct for temporal variations in the earth's
magnetic field. They are not taken into account for the interpretation of the magnetic map.
The map of magnetic field intensity variations shows the presence of numerous anomalies, many of them linear
and rectilinear (Fig. 11-4). A major spatial orientation emerges (southwestlnortheast) mainly in the northern third of
the site and around the excavation depot. This direction is consistent with that of the remains uncovered by the
excavation in Chantier IV; it also corresponds to the direction of the northwestern flank of the plateau. On the other
hand, on the southern half of the plateau, other directions appear, sometimes overlapping the previous one, which
complicates the interpretation of the map.
To the southeast, the excavations in Chantiers II and II sud revealed the presence of a rampart composed of
terracotta bricks and gypsum blocks. The terracotta materials have a high thermo-remanent magnetization and an
increased magnetic susceptibility. They are therefore one of the most intense magnetic sources that can be encountered
on the plateau: they most often create a high positive anomaly. On the other hand, gypsum sedimentary rocks are very
weakly magnetic. Consequently, their presence in the soil creates a negative anomaly. The rampart uncovered by
Chantiers II and II sud shows a successive use of these two materials: brick for the base of the construction, which is
older, and gypsum for the upper parts.
Anomalies 11-1 to 11-4 correspond to high values ranging from 50 to over 100 nT (Fig. 11-5). They indicate
structures made of baked bricks, probably elements of the rampart whose width could be estimated here at 6 to 7 m.
The irregularity of the southern limit of anomaly 11-4 suggests that the wall is in a poor state of preservation on the
southern slope. Linear anomalies II-5, with low magnetism (about 10-15 nT), are superimposed on the southeast end
of the highly magnetic anomaly II-4 (about 100 nT). These are walls made of gypsum and forming the upper part of
the rampart; some of them are visible on the ground surface.
The geophysical survey revealed the presence of numerous positive and linear magnetic anomalies, some of them
measuring more than 100 m in length. On the one hand, we can distinguish the 2 to 3 m wide anomalies, crossing the
plateau longitudinally and forming some curves (Fig. 11-5, al to a4, magenta bands). The signal intensity, weaker
than for anomalies II-1 to II-4, could correspond to mudbrick structures (walls?). However, the shape of these
anomalies trends instead in favor ofroads. The anomalies al, a2, and a3 are interrupted at the level of the depression.
It can be assumed that these structures are either eroded or masked by a greater thickness of colluvial sediment
covering them. It is indeed likely that, for example, al and a3 are one and the same structure. A fifth band, a5,
intersects a3 at right angles.
These broad linear anomalies are complemented by a large number of narrower positive and linear anomalies (Fig.
11-5, magenta lines). One, measuring over 130 min length (cl), parallels the northern edge of the plateau and lies in
the continuation of the western half of al. Most of the others (bl to b8), on the other hand, are approximately
perpendicular to anomalies al to a4 and cl. From a magnetic point of view, these anomalies have the characteristics of
mudbrick walls or ditches filled with fine sediments, but they may also correspond to narrow circulation axes (about 1
mwide).
Numerous negative and linear anomalies were also mapped (Fig. 11-5, yellow lines). To the north, the identification
of the source of the anomalies was facilitated by the proximity ofChantier IV. Indeed, the anomalies follow the same
directions as the structures uncovered by the excavation. The very angular shapes of these anomalies and their low
magnetic intensities indicate that they are gypsite or quartzite walls. On the one hand, there are small square structures
of about 6 m on each side (dl to d4), and on the other hand, more complex structures such as el to e4. To the west of
the depression, comparable anomalies are found, but the orientations of the structures are variable. Some of them
follow directions close to the structures uncovered in Chantier IV (southwest/northeast, northwest/southeast). This is
the case of the large complex e5 or e6 and e7, but not of the anomalies fl to f6, oriented according to the cardinal
directions (east/west, north/south).
Magnetic surveys on Zeyve Hoyuk in Porsuk have revealed numerous magnetic anomalies between the
excavations of Chantiers II and II sud to the west of the mound, and Chantier IV to the east. The analysis of the
magnetic map allowed us to discern several sets of anomalies according to their spatial organisation, their shape, and
their location.
Along the continuity of Chantiers II and II sud, the survey highlights the layout of the brick rampart. Six to seven
m wide, it completely blocks the southwestern face of the spur, i.e., the face that is least naturally protected by the
slope. This rampart is topped on the southeast by a stone construction. The two materials used here correspond to two
periods of occupation of the site; the brick constructions are from the Hittite period, whereas the stone constructions
are more likely to be from the Roman period. Such stone wall constructions are very numerous in the vicinity of
Chantier IV, along the northwestern slope, but also to the south, around the excavation site. They are mostly oriented
southwest/northeast and northwest/southeast like those found in Chantier IV. Based on the nature of the materials used
and the main directions of the structures, it seems reasonable to hypothesize that the plateau was densely occupied in
the Roman period and that there was a real urban framework.
Numerous linear anomalies interpreted as roads, brick walls, and depressions (ditches or wall recovery trenches)
complete the indications of occupation of the site revealed by the magnetic surveys. It is probable that some of these
structures are from the Hittite period, but this hypothesis needs to be confirmed. It is indeed essential to have recourse
to targeted archaeological surveys in order to clarify the interpretations of the results of the prospection and to verify
the hypotheses put forward following this operation.
CONSOLIDATION, PRESERVATION WORK, AND MUDBRICK WORKSHOP
During the 201 7 summer campaign, an initial cleaning was carried out in previous excavation areas (Chantier II
and Chantier IV) and in access pathways (Barat and Koker Gok9e 2019: 509). Particular attention was paid to the
establishment of circulation routes.
In 2018, a cleaning of the "Hittite pastern" area, in Chantier II was undertaken, both because this is the first,
impressive, area encountered by visitors to the site, and also due to the need to secure this area. In 2019, the cleaning
of squares F05 and F06 in Chantier II took place in order to make the Iron Age fortification visible from the road
going to Zeyve Hoyiik, in an effort to highlight the value of the site. After this work, consolidation activities were
carried out in Chantier II, prior to restoration and reconstruction.
Figure 11-6. Map of Chantier 2 and Zone 5 (Jean-Fran9ois Pichonneau).
Consolidation work was conducted in Chantier II beginning in 2018. This included consolidating Iron Age walls
utilizing the dry masonry technique (squares F05, F06, and H05), particularly inside the curved corridor. In 2019,
parallel to the archaeological excavations led in Zone 5 (see Fig. 11-6), stabilisation and consolidation work took
place in Chantier II, in squares H04, G04, and H05 (see Fig. 11-7), in preparation for a restoration project.
Figure 11-7. Consolidation work, Chantier 2 (Jean-Frarn;ois Pichonneau).
The chosen zone for this work was the northwestern part of the Hittite fortification wall and tower (Late Bronze
Age), which was reused in the Iron Age. Initially, the tower and fortification were cleaned in preparation for
consolidation. For stabilisation and protection, mudbricks from a stock created in the 2018 summer campaign (see
below) were used. These mudbricks correspond to the later building phase of the fortification wall. The first building
of the fortification consisted of mudbricks of larger size. This earlier wall had been destroyed and rebuilt with
mudbricks of smaller size. The northern comer of the tower, (stabilized during the 2019 summer campaign), had been
destroyed during a second fire which destroyed the fortification wall, which was then rebuilt with stones during the
Early Iron Age. It was thus decided to leave this Early Iron Age restoration intact for the stabilisation of the north
comer of the tower.
The wall between the tower and the curved corridor was consolidated. At the bottom of this wall were sandstone
bases, and above, mudbrick elevations, baked during the fires endured by the fortification. These baked mudbricks
have melted due to time and exposure to the elements.
The chosen technique for preservation was to encase the original fortification using mudbricks constructed in 2018,
in an eight-row matrix of mudbricks. This consolidation/stabilisation and protection technique was a preliminary test
as a restoration project. The materials used are earth, water, and straw, rather than concrete. This
consolidation/stabilisation/protection action is part of an environmentally responsible approach, respectful of local
skills and materials and based on traditional mudbrick construction. In fact, the use of mudbrick is found in the
vernacular architecture of the Porsuk region, a piedmont zone at the base of the Taurus Mountains. The old village of
Porsuk (winter village), abandoned during the 1980s for the new village of Porsuk (summer village), due to a
conversion of agricultural activities (a transition from pastoralism to arboriculture), provides beautiful examples of
traditional mudbrick houses. In the future, the archaeological mission of Porsuk-Zeyve Hoyuk intends to promote and
utilize its excavation workers' skills.
In 2018, a mudbrick workshop was initiated on the excavation grounds in order to create a stock of mudbricks for
the preservation/consolidation work planned for 2019 (see Fig. 11-8). The 1300 mudbrick reserve was used for the
preservation of the tower and Hittite fortification walls in Chantier II (squares H04, G04, and H05) described above
(seeFig.11-7).
The composition for the creation of the 1300 mudbricks was the same as that used in 2014 3 :
• 30 wheelbarrows of earth (1800 lt)

• 3 wheelbarrows of straw
• 250-300 lt of water
3
The mission was able to observe the state of conservation of the mudbricks made in 2014 and thereby choose the best
composition.
Figure 11-8. Mudbrick workshop (Emine Koker Gok<;e).
The adobe texture was trampled by booted workers for about two hours and then left for twelve hours. The texture
was then moulded in 40 x 18 x 12-13 cm moulds (Beyer et al. 2015: 287) to create the mudbricks. Afterwards, the
mudbricks were dried in the sun for one week. A minimum of forty days of drying is necessary before utilisation.
During the 2019 summer campaign, more than 3000 mudbricks were manufactured.
ZONE 5 EXCAVATIONS
Zone 5 excavations took place during the 2019 summer campaign. Geographically, zone 5 is adjacent to Chantier
II (Fig. 11-9). This new zone, located in squares H06, H07, 106, and 107, corresponds to a space previously excavated
Figure 11-9. Aerial view, zone 5.
by Pelon at the beginning of the 1970s (Pelon 1972: 305-306) and by Tibet and Beyer between 2008 and 2012 (Beyer
et al. 2009: 343-349, 2010: 235-238, 2012: 178-186, 2013: 201-213). At the beginning of the new excavations, zone
5 presented only some walls on the surface. After weeding, and revealing the various archaeological levels, zone 5
was divided into seven sectors, numbered from 1 to 7. The boundaries of the sectors were defined by walls linking
these sectors (Fig. 11-10).
3m
-r
1I
1I
' I I
··· ·· ······················1···1··· ·····1···················
I I I I
I
I
I
I
I
I 1
I I
1 I
33 1 I
I , ' : I I I
Figure 11-10. General plan of zone 5 (Jean-Frarn;ois Pichonneau).
Sector 1 is the space situated between walls 01, 02, and 16. On its northern side, it is limited by a 2 m wide bench,
preserved between the zone 5 excavations and Tibet's earlier sounding excavation in Chantier II. However, for better
coherence, the data from Tibet's sounding excavations have been integrated into that of zone 5. The first
archaeological levels brought to light in sector 1 (5001, 5036, and 5087) are alluvium deposits where pithoi (P04, P06)
were discovered. Under these levels, we find a destruction stratum with a few mudbrick fragments, wall primer, and
pithoi. This destruction level lies on alluvial ground corresponding to the last stratum excavated in this sector, but also
contemporary with walls 02, 22, 36, and 23. In the northern area of sector 1, a narrow structure (35 cm) appeared,
composed on the plan by wall units 22, 36, and 33. These low walls, made of bricks and little stones, are covered by
an earthen layer, coated in whitewashed plaster. The ceramic material found in these strata is limited in quantity, but
allows dating to the beginning of the 1st century BCE for the building and occupation of this first building level of
housing (level 1). Subsequently, the walls of level 1 were systematically knocked down and leveled off for the
construction of a new building, constituted by walls 01, 02, 03, and 16. These walls have an average thickness of 75
cm and were constructed of big gypsum blocks with an earthen mortar. The footings of the walls of this new level 2
construction were placed directly on the structures of level 1. The occupation strata contemporary to level 2 were
previously excavated by Pelon in the 1970s and also during the sounding excavation made in the northern area by
Tibet between 2008 and 2012.
Sector 2 is separated from sector 1 by walls 02 and 36. Stratigraphically, sector 2 is similar to sector 1, but it was
designated as a separate sector because the structure built in level 2 had important modifications; the basement,
arranged to gather the reserves stored in the pithoi, was modified, and wall 02 was razed to extend this storage space.
The foundations of these new structures 01, 02, 03, and 16 are, in part, composed of large river stones, which were
laid down directly on level 1 structures. However, walls 01 and 03 simply leaned against walls 05 and 20. The
superstructure of these walls is composed of gypsum block bases. Due to the nature of gypsum, the cut of the blocks
did not fit that of the river stones. Pithoi 01 and 03 belong to this last level and were inserted into deposit 5001, also
covering wall 02. However, deposit 5086, present in sectors 1 and 2, leans on wall 02, and we can see the same
phenomenon for destruction strata 5087 and 5088, covering level 1. In the southern area of sector 2, a hearth, leaning
back against wall 05 and pithos 03, was revealed. Pithos 03 was installed in the domestic hearth after the destruction
and backfilling of this baking structure. The destroyed hearth preserved the internal walls of the baking chamber. The
hearth walls were built with small gypsum stones joined by an earthen mortar, similar to the other built structures. The
baking chamber, 1.20 min diameter, was totally filled by blocks. One of these blocks, entirely burnt and polished, on
which pithos 03 rested, could have functioned as a support element on the floor. The somewhat burnt state of the walls
of the baking chamber allows us to estimate that the top temperatures reached were not very high. We can therefore
infer a domestic use for this hearth. The fire installation consisted of a simple pit which does not conform to any
particular structure. The pit holds ashes and charcoal from its last use. Belonging to level 1 of this housing, the hearth
is situated in a space defined as a courtyard, opening onto a cooking area, sector 3, and to rooms, to the north. In
sector 2, the strata contemporary to the hearth activity are deposits 5022 and 5032. These occupational strata are
composed of alluvium containing a very high percentage of charcoal particles. Above, stratum 5031 contains some
elements of burnt clay coming from the destruction of the hearth. This ensemble is covered by deposit 5001, found
everywhere in zone 5, as if this space had been elevated and reconfigured.
Sector 3 also presents the two levels of construction described for sector 1 and 2. Level 1 is composed of structures
05, 19, 21, 28, and 34. Wall 28, with an average width of 80 to 90 cm, is built with gypsum blocks, constituting, in the
lower part, relatively well-ordered bases. The impressive wall 28 forms a corner with wall 05, which was originally
meant to be the western boundary of the first house of level 1, but was raised and extended as a terrace wall between
sectors 5 and 6 in the second level. Inside the sector 3 space, we find platform 21, forming a platform or a ramp that
gives access to a threshold in wall 28. This platform is composed ofmudbricks, small gypsum stones, and sandstone
slabs. Against this platform, pithos 07, small in size, was placed. Pithos 08, also installed in level 1, against wall 05,
was covered during the building of wall 03 in level 2. Between wall 28 and platform 21, a paving was constructed
which continues a little bit to the south, along the platform, and moves on to matrix 5032, on which charcoal layer
5022 rests. In the southwestern comer of sector 3 there was a small quadrangular pit. In the pit fill we found a small
ceramic bottle, a pot base intended for cooking, and small gypsum stones. In the northeastern comer of sector 3,
between platform 21 and wall 03, two counterweights were brought to light in the soil of level 1. Here we have noted
the presence of a small posthole in the matrix of 5032. On the southern area, against platform 21, and in matrix 5032,
some burnt clay stains combined with an ashy deposit are the only evidence of fireplace activity. In sector 3, the whole
oflevel 1 is covered by stratum 5031, comparable to a destruction stratum, above which deposit 5001 was laid.
Sectors 1, 2, and 3, in level 1 and then in level 2, present completely different housing types. Level 1 is composed
of three rooms, resting against enclosure wall 27; these three rooms open, in the south, to an interior courtyard,
leading on to a space that could have been a kitchen and perhaps also a weaver's workshop in the northeastern comer
of sector 3. In this courtyard, but against wall 05, which marks the southern boundary of the housing area, a domestic
hearth had been constructed. The nature and narrowness of the level 1 walls do not suggest that this housing had a
second floor. The interior walls of the rooms were covered by a layer of earthen surface on which was placed a layer
of whitewash. The ceramic material, limited in quantity in the excavated strata, is attributable to the end of the 2nd
century BCE. In fact, occupation strata contained sigilated Aretine ceramics, among which we found fragments of
plates with annular stems, dated between 40 and 15 BCE. This ceramic material is quite comparable with that defined
by C. Abadie-Reyna! for phases 4 and 5 in the 1989 Porsuk excavations (Abadie-Reyna! 1992: 373- 376). Level 1 of
zone 5, excavated in 2019, can be associated with phase 5 of the 1989 excavations in Chantier IV, and level 2 of zone
5, the second and last occupation of the zone, can be associated with phase 4 of the 1989 excavations in Chantier IV.
During the 1' t century BCE, this first type of housing, the plan and layout of which resembles Hellenistic housing, was
totally razed in order to move on to a new type of house. This new house was provided with a cellar where pithoi
intended for surplus were stored. The walls of the new house were much wider and built with shapeless gypsum
blocks, the bases of which sit directly on level 1 walls. The walls of the new house were covered by an earthen
surface. Wooden stairs, the base of which were situated in sector 6, allowed access to the second storey rooms through
a gallery. The plan of this house is reminiscent of a small urban building, with storage and shops on the first floor.
Sector 4 is a space situated between the paved pathway of sector 7 and walls 4 and 28 (enclosure wall). The
excavation stopped on the destruction strata of the level 1 remains. Wall 35, 50 cm wide, was built using gypsum
stones. It was totally destroyed, then filled in for the construction of the paved pathway of sector 7. This space stayed
open. It borders the path in the first level and forms a natural spillway for grey and rain water draining to the north.
Sector 5, situated in the southwestern comer of zone 5, presents two building levels and is delimited by walls 10,
11, and 12. The first level is composed of gypsum stone bases to walls 10, 11, and 12, joined by an earthen mortar.
The ceramic material collected there suggests that this first housing was built during the 1st century BCE. Wall 12, 65
cm wide, is a raw earthen wall, with sides covered by a whitewash primer. The building of wall 12 with raw earth
required the use of form panels. This technique consists of producing a mobile form (mould) the same thickness as the
wall; the interior of the form is filled with adobe and then packed down. A threshold was created in wall 10, allowing
access to sector 6 between walls 06 and 08. This ensemble, forming the first level of sector 5, was planned, and a new
structure, this one built of big blocks of gypsum, was constructed according to the same plan as the previous building.
There are two openings for this building: one in the north, through wall 11, giving direct access to a paved terrace in
front of the pathway of sector 7, and the other, through wall 10, maintained from the earlier level. In the southwestern
comer of sector 5, pithos 06 was placed in the fill stratum of the second level of the building.
Sector 6 is a space situated in the southern part of zone 5 and can be defined as a passage between the last two
levels of construction. A sandstone threshold slab, preserved between walls 06 and 08, marks the access to sector 5.
Wall 07, made of gypsum stones, corresponds, from its original construction, to a terrace wall, which is also an
elongation of wall 04. Small walls 08 and 09, 40 cm wide, are the bases which supported wooden stairs leading to
private apartments on the second floor. Between walls 08 and 01, a large sandstone slab (slab 15) marks the base of
this stairs. It is clear that wall 06 experienced three stages of construction and repair during its expansion. In its first
iteration, this wall was a light structure composed of small stones set in an earthen mortar, the sides of which were
covered by a primer. In the second stage, only the portion of the wall in contact with walls 05 and 07 was built up. The
final stage of walls 06 and 07 represents the final reconstruction in sector 5.
Sector 7 was previously excavated in 1969 by Pelon (Pelon 1972: 306). It features the remains of a slab that slopes
significantly toward the north. It may correspond to the remains of the pathway (mentioned above) that has a north-
south orientation that allows access to the summit of the site. In sector 7, wall 35, destroyed then filled, passed under
the pathway, in a western direction. During the 1969 excavation, little information was recorded about the
archaeological strata of this sector. The western part of the pathway was bordered by a large gypsum wall of which
only a few piled stones remain.
CONCLUSION
At the conclusion of the excavation campaign in zone 5, it became clear that there is a transition between a
Hellenistic housing model (level 1) to a Roman housing model (level 2) with very different plans. These two levels of
construction correspond to phases Ia, lb, and le as defined by Tibet in the test excavation in square H7 (Beyer et al.
2013: 203- 209). Phase Ia corresponds to the 1960 destruction and leveling. This assessment permits us to understand
the transformation of zone 5, in which the first constructions attributable to phase le are built according to a
Hellenistic plan. Through level 1, zone 5 presents a concept of housing adopted and developed during the Hellenistic
period. The most characteristic element of this house is the presence of a courtyard around which space is organized.
It is a central space, initially designed for open-air domestic activities. The rooms are built around this space and were
erected as needed, showing therefore an agglutinated character. However, the plan of the later levels Ia and lb
corresponds precisely to a Roman model, more urban than rural, like a small insula a tabernae provided with only one
or two floors.
This dating of levels of construction is not surprising from a political point of view for the southern part of
Cappadocia. The site of Zeyve Hoyuk is indeed situated at the southern extremity of the Kingdom of Cappadocia, a
friend of Rome after the peace of Apamea in 18 8 BCE. The Kingdom of Cappadocia, like other Anatolian kingdoms
(Pontus, Bithynia) became an allied and client kingdom to Rome in the Late Hellenistic period, when Rome began to
create provinces in Anatolia (e.g., creation of the province of Asia in 129 BCE after the bequest of the Kingdom of
Pergamon; creation of a command, provincia, in Cilicia in 100 BCE, actual provincialisation of Cilicia in 78- 74 BCE)
(Ferrary 2004: 780-781). The site of Zeyve Hoyuk is only 40 km distant from the Cilician Gates (now Gulek Pass or
Giilek Bogazz); between the l't century BCE and l't century CE, if this pass was really creating a political frontier
between a client kingdom and a Roman province, we can suppose it was in no way a commercial or cultural frontier.
Roman tableware was imported into the Porsuk region and surely served, as elsewhere, as a marker of social
distinction. The Kingdom of Cappadocia was made into a province between 15 and 17 CE (Sartre 1991: 36, 1995:
170), under the reign of Emperor Tiberius, and thus Rome assumed the political administration of the territory in
which the site of Zeyve Hoyuk was located. It is therefore normal for the second level of construction (1st -2nd century
CE) to find more Roman ceramics and a clearly Roman housing model.
REFERENCES CITED
Abadie-Reynal, Catherine. 1992. Porsuk. Rapport sur la campagne de fouilles de 1989. Chantier Est. Syria 69: 249-
377.
Barat, Claire and Emine Koker Gokye, 2019. Porsuk-Zeyve Hoyuk 2017 Y1h Kaz1s1. Kazz Sonuc;larz Toplantzsz 40(1):
505- 516.
Barat, Claire, Emine Koker Gokye, Jean-Franyois Pichonneau, Vivien Mathe, Guillaume Bruniaux, and Romain
StoraL 2020. Porsuk-Zeyve Hoyuk: rapport preliminaire des campagnes 2018 et 2019. Anatolia Antiqua XXVIII:
145-172.
Beyer, Dominique. 2006. Zeyve Hoyi.ik (Porsuk)-The excavations 2004. Kazi Sonuc;Iarz Toplantm 2 (2): 65-72.
- . 2007. Zeyve Hoyillc (Porsuk), 2005. Kazi Sonuc;Iarz Toplantzsz 28 (1): 629- 638.
- . 2008. Zeyve Hoyillc (Porsuk), 2006. Kazi Sonuc;Iarz Toplantw 29 (2): 107-116.
- . 2009. Zeyve Hoyillc (Porsuk), 2007. Kazi Sonuc;larz Toplantzsz 30 (1): 385-392.
- . 2011. Zeyve Hoyillc (Porsuk) Excavations in 2009. Kazz Sonuc;larz Toplantzsz 32(4): 394-402.
- . 2012. Zeyve Hoyillc-Porsuk. Bilan des recherches sur les niveaux du Bronze et du Fer. In Archeologies et espaces
parcourus, Ji'''' rencontres archeologiques de l'IFEA, Istanbul, 11-13 novembre 2010, 0. Herny, ed., 47-56.
istanbul: Institut Frans;ais d'Etudes Anatoliennes-Georges Dumezil.
Beyer, Dominique and Frans;oise Laroche-Traunecker. 2017. Le site de Zeyve Hoyillc-Porsuk aux epoques hittite et
neo-hittite. Remarques sur la succession des systemes defensifs. In L'hittitologie aujourd'hui: Etudes sur
l'Anatolie hittite et neo-hittite a /'occasion du centenaire de la naissance d'Emmanuel Laroche, 5,mes rencontres
archeologiques de l'IFEA, Istanbul, 21-22 novembre 2014, A. Mouton, ed., 229-244. istanbul: Institut Frans;ais
d'Etudes Anatoliennes-Georges Dumezil.
Beyer, Dominique and Marie Stahl. 2015. Les archives de la composante d'archeologie orientale de l'UMR 7044.
Archimede 2: 20-40.
Beyer, Dominique, et al. 2004. Porsuk: rapport sommaire sur la campagne de fouilles de 2003. Anatolia Antiqua XII:
267-281.
Beyer, Dominique, et al. 2005. Porsuk (Zeyve Hoyillc): rapport sommaire sur la campagne de fouilles de 2004.
Anatolia Antiqua XIII: 295-318.
Beyer, Dominique, Isabelle Chalier, Frans;oise Laroche-Traunecker, Stephane Lebreton, Julie Patrier, and Aksel Tibet.
2006. Zeyve Hoyillc (Porsuk): rapport sommaire sur la campagne de fouilles de 2005. Anatolia Antiqua XIV: 205-
244.
Beyer, Dominique, Isabelle Chalier, Frans;oise Laroche-Traunecker, Julie Patrier, and Aksel Tibet. 2007. Zeyve
Hoyillc (Porsuk): rapport sommaire sur la campagne de fouilles de 2006. Anatolia Antiqua XV: 289-314.
- . 2008. Zeyve Hoyillc (Porsuk): rapport sommaire sur la campagne de fouilles de 2007. Anatolia Antiqua XVI: 313-
344.
Beyer, Dominique, Isabelle Chalier, Frans;oise Laroche-Traunecker, Stephane Lebreton, and Aksel Tibet. 2009. Zeyve
Hoyillc (Porsuk): rapport sommaire sur la campagne de fouilles de 2008. Anatolia Antiqua XVII: 317-349.
Beyer, Dominique, Isabelle Chalier, Fabrice De Backer, Frans;oise Laroche-Traunecker, Stephane Lebreton, and Aksel
Tibet. 2010. Campagne 2009 de la mission archeologique de Zeyve Hoyillc (Porsuk). Anatolia Antiqua XVIII:
215- 242.
Beyer, Dominique, Isabelle Chalier, Frarn;:oise Kirner, Julie Patrier, and Aksel Tibet. 2012. Zeyve Hoyi.ik-Porsuk:
rapport preliminaire de la campagne 2011. Anatolia Antiqua XX: 177-203.
Beyer, Dominique, Frans;oise Laroche-Traunecker, Julie Patrier, and Aksel Tibet. 2013. Zeyve Hoyillc-Porsuk: rapport
preliminaire de la campagne 2012. Anatolia Antiqua XXI: 201- 234.
Beyer, Dominique, Isabelle Chalier, Frans;oise Kirner, Frans;oise Laroche-Traunecker, and Aksel Tibet. 2014. Zeyve
Hoyillc-Porsuk: rapport preliminaire de la campagne 2013. Anatolia Antiqua XXII: 327-342.
Beyer, Dominique, Can Karavul, Frans;oise Laroche-Traunecker, and Aksel Tibet. 2015. Rapport preliminaire sur !es
travaux de la mission archeologique de Zeyve Hoyillc-Porsuk 2014. Anatolia Antiqua XXIII: 275-290.
Beyer, Dominique, Isabelle Chalier, and Frans;oise Kirner. 2016. Rapport preliminaire sur !es travaux de la mission
archeologique de Zeyve Hoyillc-Porsuk 2015. Anatolia Antiqua XXIV: 253-280.
Bruniaux, Guillaume, Vivien Mathe, Frans;ois Leveque, Adrien Camus, and Vincent Ard. 2017. Data Processing
Chain to High Spatial Resolution Magnetic Survey: Application on the Neolithic Site of Le Pontet (Charente-
Maritime, France). Archaeological Prospection 24: 1- 14.
Dupre, Sylvestre. 1983. Porsuk I. La ceramique de /'Age du Bronze et de /'Age du Fer. Paris: Editions Recherches sur
les Civilisations.
Ferrary, Jean-Louis. 2004. Rome, !es Balkans, la Grece et !'Orient au deuxieme siecle av. J.-C. In Rome et Ia conquete
du monde mediterraneen, Tome 2, Genese d 'un empire, C. Nicolet, ed., 729-788. Paris: Presses Universitaires de
France.
Forrer, Emil 0. 1937. Kilikien zur Zeit des Hatti-Reiches. Klio 30: 135-186.
Hawkins, John D. 1969. A Hieroglyphic Hittite Inscription from Porsuk. Anatolian Studies 19: 99- 109.
Kuzucuoglu, Catherine. 1997. Le site de Porsuk: contexte geomorphologique local et regional. Anatolia Antiqua V:
19-33.
Lebreton, Stephane. 2013. Les Portes de Cilicie. In Lieux de memoire en Orient grec a l'epoque imperiale, A.
Gangloff, ed., 305-331. Bern-Berlin-Brussels: Peter Lang.
Meriggi, Piero. 1963. Terzo viaggio anatolico. Oriens Antiquus 2: 275- 299.
Pelon, Olivier. 1970. Rapport preliminaire sur la premiere campagne de fouilles a Porsuk-Uluk1~la (Turquie). Syria
47: 279- 286.
- . 1972. Rapport preliminaire sur la deuxieme et la troisieme campagne de fouilles a Porsuk-Uluk1~la (Turquie) en
1970 et 1971. Syria 49: 303-317.
- . 1976. Cinq campagnes de fouilles a Porsuk (1969-1976). VIII. Turk Tarih Kongresi (I. Cilt): 233-237. Ankara:
Turk Tarih Kurumu.
- . 1978. Six campagnes de fouilles aPorsuk (Turquie meridionale) de 1969 a 1977. Comptes rendus des seances de
l'Academie des Inscriptions et Belles-Lettres 122(2): 347- 359.
- . 1982. La fouille de Porsuk-Uluk~la (1969- 1977). Travaux et Recherches en Turquie Vol. 1: 75-77. Leuven:
Peeters Publishers.
- . 1992. Quatre campagnes de fouilles aPorsuk (Turquie meridionale) de 1986 a 1989. Syria 69: 305- 347.
- . 2005. Une fouille hittite au pied du Taurus. In Archeologies, vingt ans de recherchesfranr;aises dans le monde, P.
Mongne, ed., 198-201. Paris: Maisonneuve et Larose.
Pelon, Olivier and Sylvestre Dupre. 1987. Une fouille frarn;aise au pied du Taurus. Archeologia 221: 14- 25.
Ramsay, William R. 1903. Cilicia, Tarsus and the Great Taurus Pass. The Geographical Journal 22(4): 357-410.
Sartre, Maurice. 1991. L 'Orient romain. Provinces et societes provinciales en Mediterranee orientale d 'Auguste aux
Severes (31 avant J-C.- 235 apres J-C.) . Paris: Seuil.
- . 1995. L 'Asie Mineure et l'Anatolie d'Alexandre a Diocletien. JVe siec/e av. J-C./Jlle siec/e ap. J-C. Paris:
Armand Colin.
Simon-Millot, Rolande. 2012. Un inventaire, des inventaires. Les nouvel/es de /'archeologie 130: 1-4.
Tibiletti, Gianfranco.1963. Nota all' iscrizione latine di Porsuk. Oriens Antiquus 2: 300-303.
Zanella, Sandra, Jean-Paul Brun, Martine Denoyelle, Pierre Rouillard, and Stephane Verger. 2017. Les archives de
fouilles. Modes d 'emploi. Paris: College de France.
CHAPTER TWELVE
THE KERKENES PROJECT 2019-2020
SCOTT BRANTING, JOSEPH W. LEHNER, SEViL BALTALI TIRPAN,

DOMINIQUE LANGIS-BARSETTI, SARAH R. GRAFF, TUNA KALAYCI,
AMANDA GROFF, LUCAS PROCTOR, JESSICA ROBKIN, PAIGE PAULSEN,
AND NiLUFER BATURAYOGLU YONEY
The archaeological site of Kerkenes covers more than 2.5 km 2 of a low mountaintop in central Anatolia. The
irregular landscape is enclosed by a 7 km-long stone fortification circuit, accessed by seven monumental city gates,
and is covered by a fabric of urban blocks, independent structures, and communal spaces. While original surveys and
test excavations in the late 1920s deemed the site to be "post-Hittite, pre-Classical" (Summers and Summers 1998:
178), the current Kerkenes Project has narrowed the period of occupation to the later part of the Middle Iron Age,
specifically the mid- to late-7th and early 6th centuries BCE. The short period of occupation is a notable feature of the
settlement and has been a major factor in the methodologies employed during each season of archaeological research.
While work in the 2019 and 2020 seasons was limited in scope by permitting issues and the COVID-19 pandemic, our
efforts remained focused on exploring the social organization of the ancient city (Fig. 12-1).
BACKGROUND OF KERKENES
Located at the heart of central Anatolia, 11 km
southwest of modem Sorgun in the province of
Yozgat, the site of Kerkenes sits atop one of the
highest ridges of the area. The exposed mountaintop
of Kerkenes Dag1 overlooks much of the Kanak Su
river basin. Faraway landmarks are visible beyond
this basin, such as the peak of Erciyes Dag1 to the
south and the foothills of the Pontus Mountains to the
north (Fig. 12-2). Important long-term north-south
and east-west trade routes intersected in the Kanak
Su basin, just east of modem Sorgun. One could even
hypothesize that the founders of Kerkenes chose the
location for the city because it would allow them to
extend some command and control over those
crossroads (Anderson 1903: 25; Branting 1996).
Whether the opportunity to control valuable trade
nexuses played a role in the foundation of the city,
the original settlers must have had very good reasons
to establish themselves where they did, for the site of
Kerkenes presented them with some important
challenges. For one, no settlement yet discovered
existed prior to the Iron Age foundation, meaning that
the 271 ha city and its 7 km circuit of stone
fortifications had to be built from scratch. Dozens, if
not hundreds of granite outcrops had to be cleared
before construction could begin on a large scale, and
adequate sources of water had to be identified on the
Kerkenes 2019-2020 mountaintop. Since most agricultural fields were in
the surrounding valleys, some 200 m below,
~ 11 - .,:;om Kerkenes' population had to either develop its own
fields along the ridge's slopes or secure a reliable
supply of food from the nearby villages. All told, the
Figure 12-1. Map of the location of
efforts and resources expended to found and settle
(1) Urban Block 8 (2) the Kale, and
Kerkenes would have been enormous, and yet it
(3) the Cappadocia Gate within Kerkenes.
remains unclear what prompted such an endeavor.
The Kerkenes Project 2019-2020 147
Figure 12-2. View to the north ofKerkenes.
While the area within the bend of the K1z1hrmak was well documented during the Bronze Age, this situation ended
abruptly with the fall of the Hittite Empire. Neighboring regions such as Urartu, Tabal, and the Neo-Hittite polities of
southeastern Anatolia continued to appear in written sources during the following Iron Age, either in indigenous
records or more often in Neo-Assyrian ones, but these records only contain extremely vague and disparate mentions of
the situation prevailing within the heart of central Anatolia. Despite being surrounded by eminent neighbors, Urartu to
the east, Tabal to the south, Phrygia to the west, and towards the end of the period, the Greek colonies of the Black
Sea to the north, the political and cultural realities of the region within the bend of the K1z1hrmak remain something of
a mystery. Nevertheless, important Iron Age settlements such as Bogazkale, Ali~ar, and Kerkenes have been identified
across the region (Kealhofer et al. 2009; von der Osten 193 7a, 193 7b ). The scale of some of these sites as well as a
rich material culture suggest that the bend of the K1z1hrmak remained a vibrant region even after it no longer served as
the Hittite imperial heartland (Ross et al. 2018, 2019).
The dearth of contemporary written sources mentioning this area greatly complicates the identification of
Kerkenes. Przeworski first suggested that the ruins on the Kerkenes Dag1 might be Pteria, a city mentioned by
Herodotus I.76 (Przeworski 1929). Unfortunately, the Greek historian was vague in his account, limiting himself to an
ambiguous geographical description, to the point where various other sites have been proposed over the years as the
true location of Pteria (Texier 1839: 209-25; Donmez 2004; Summerer 2007: 29). Nevertheless, should Herodotus's
brief mention of Pteria be accurate, Kerkenes represents the most viable candidate. However, one of the most glaring
omissions in the Greek account pertains to the cultural affiliation or affiliations of the city. The 2003 discovery of an
inscription in the paleo-Phrygian language in the Palatial Complex at Kerkenes, additional Phrygian religious
iconography, and Phrygian graffiti across the city, strongly suggest that the rulers and at least some of the city's
population were Phrygians (Summers 2006). However, the larger size of Kerkenes relative to Gordion, the Phrygian
capital, and its presence on the very edge of Phrygia's sphere of influence raises even more questions.
Regardless of its size or prosperity, Kerkenes was a tragically short-lived settlement. All excavated areas to date
display a single layer of occupation, with minimal traces of modifications or repairs. Geophysical surveys carried out
across the site likewise show a clear lack of significant architectural phasing. All evidence points to Kerkenes having
been founded, built, and inhabited for no more than two or three generations before its destruction, putting the origin
of the city somewhere in the mid- to late 7th century BCE. While the foundation date of Kerkenes remains somewhat
nebulous, its end is far better understood. According to Herodotus, Pteria was captured by Croesus of Lydia in his
final campaign against Cyrus of Persia, in the mid-540s BCE. The text briefly mentions how Croesus took control of
the city before engaging in an inconclusive battle with the Persians. What is described in this account would be
supported by archaeological evidence from Kerkenes, which shows that the city was looted, destroyed, and abandoned.
While the exact fate of the inhabitants cannot be ascertained, other than two skeletons found crushed beneath
collapse within the Cappadocia Gate (Summers 2021 ), it should be noted that there is no evidence of occupation
following the Iron Age abandonment until the Classical periods, and no evidence that attempts at salvage were made
after the fire, supporting the idea of a swift and decisive depopulation. There are, however, telling signs that the event
was violent. From the Palatial Complex to Urban Block 8, statues and vessels were broken and scattered prior to the
fire, as fragments not only underwent differential burning, but joining pieces of the same object can also be found
meters apart. Once the city had been emptied of most of its valuable contents and inhabitants, arsonists put the city to
the torch. Since only two human casualties have thus far been found, we know the fire was not part of the initial
attack, but rather the final act of the destruction of the city. As Herodotus does not explicitly mention the fire, it is
possible that it may have taken place at the hands of Cyrus' men instead. Regardless of who started the fire, it ravaged
148 Chapter Twelve: Branting et al.
much of the site intensely, as shown in both excavations and geophysical survey (Summers and Summers 1998). The
short-lived nature of the settlement offers archaeological advantages. Without complex stratigraphy, large areas can be
excavated relatively rapidly. Moreover, the extensive burning as well as the absence of superimposed architectural
phases make geophysical survey techniques highly effective. In the case of Kerkenes, gradiometry and electrical
resistance are allowing the reconstruction oflarge portions of the city's original plan on a scale that would have been
impossible to achieve with excavations alone (Summers et al. 1996; Branting 2004; Langis-Barsetti 2013).
PRIOR ARCHAEOLOGICAL RESEARCH
Although stories and legends about the ruins on the Kerkenes Dag1 had been part of local folklore for generations
(Bittel 1960/1961 ), archaeological work at the site did not begin until 1926, over 20 years after the first assessment of
the ruins (Anderson 1903). From 1926 to 1928, Kerkenes was the target of three very brief exploration missions by
H.H. von der Osten and Erich Schmidt, then in charge of the nearby Alishar Expedition. Although cursory, their work
produced the first map of the walled city (von der Osten 1928), and the first excavations (Schmidt 1929). Since their
stated goal had been to find a monumental settlement contemporary with the Hittite capital of 1-Jattusa, their interest in
Kerkenes quickly faded once it became clear that the ruins dated to after the fall of the Hittite Empire.
The city waited until 1993 to be once more at the center of a research campaign, when Geoffrey D. Summers and
co-director Fran9oise Summers initiated the modem Kerkenes Dag Project, the precursor to the current Kerkenes
Project. Given the intimidating size of the site, remote sensing techniques were prioritized from the start, with the goal
of not only producing extensive maps of the remains, but also providing clues to the most promising areas for
excavation. Techniques used included blimp and hot air balloon photography, satellite images, ground penetrating
radar (GPR), electromagnetic induction (EMI), conductivity, gradiometry, GPS, and electrical resistance surveys
(Summers and Summers 2010). The gradiometry and GPS micro-topographic surveys cover the entire Iron Age city,
while the electrical resistance survey now covers 83.5 ha, or ca. 30% of the entire site. The work undertaken was
ground-breaking and enabled computational approaches to its analysis that continue to be employed today (Kvamme
2003: 437). The Kerkenes Project strives not only to apply but also pioneer new methods to explore this ancient city,
an ethos introduced by the Summers and continued under the current director, Scott Branting, starting in 2014
(Branting and Summers 2002; Branting 2004; Summers et al. 1999, 2007).
With data detailing the location of both exposed and buried remains, Summers and his team began excavations in
1996. Efforts were divided between an early exploratory campaign designed to assess the condition and nature of the
archaeological remains throughout the site, as well as larger excavations targeting specific structures within the city.
Over the following 17 years, roughly 7000 m 2 were exposed, mainly in the so-called Palatial Complex (Summers
forthcoming) and at one of the city's seven gates, dubbed the Cappadocia Gate by Summers, due to its position
overlooking the northern reaches of the Cappadocian plain (Summers 2021). With these two monumental projects
concluded in 2011, the project was able to shift its research focus towards understanding the broader social and
economic organization of the city.
2019-2020 SEASONS: SCOPE OF WORK
The 2019 and 2020 seasons were vast departures from typical seasons of the Kerkenes Project over the prior 28
years. The 2020 season was heavily affected by the COVID-19 pandemic, as was so much of the world (Fig. 12-3).
Figure 12-3: The Kerkenes team in 2020 during the Figurel2-4. Gradiometry survey on top of the
COVID-19 pandemic. Kale in 2019.
The Kerkenes Project 2019- 2020 149
Travel restrictions and social distancing limited work at the site to one staff member over 11 days in mid to late
July. However, even the 2019 season was significantly shortened from a typical season due to bureaucratic issues
affecting projects in the region. Work in 2019 began in mid-June and continued through July, cutting in half the
planned season. Throughout these seasons the flexibility and dedication of the two government representatives, Ertan
Ylimaz from the Karaman Museum in 2019, and Gokhan Akkul from the Batman Museum in 2020, were critical
factors in the successes that the project was able to attain.
Research during the 2019 season focused on extending the excavations within one of the urban blocks in the
northern area of the city. In addition, a gradiometry survey was conducted on the Byzantine Kale (castle) on the
eastern side of the city through a collaboration with Kos; University. This work was complemented in both seasons
with site conservation efforts and with work in the artifact depots of the excavation facilities. In addition, several
specialist studies continued during both 2019 and 2020.
GEOPHYSICAL SURVEY
In a typical season, four to six weeks of electrical resistance survey are possible during the months of May and
June. This has been revealing, on average, about 6 ha of the buried city plan each year. The timing of the field season
is the main limiting factor, with sufficient soil moisture critical to acquiring data. With the melting of the snows and
spring rains there is usually enough moisture remaining in the soil to undertake the electrical resistance survey until
early to mid-June. By the time the project arrived in the field in 2019 and in 2020, the ground was too dry to
successfully carry the electrical current. Therefore, neither season proved fruitful in extending this form of mapping.
However, other geophysical mapping techniques do not require soil moisture. Gradiometry survey has been used to
great effect in previous seasons well into the late summer months. This technique has been applied to almost all of
Kerkenes in past years and was highly successful in revealing detailed information about buried buildings that were
burned during the destruction of the city. The one remaining area of the site not yet surveyed with gradiometry is the
Byzantine Kale.
In 2019, given the impossibility of undertaking electrical resistance survey and because of an opportunity to access
a magnetometer afforded us by colleagues at Kos; University, we were able to test this form of survey up on the Kale
(Fig. 12-4). While the Kale presents logistical issues, new developments have made it an enticing target. Over the last
few years, we have used a range of geophysical survey techniques to develop a plan of the castle and investigate the
possible presence of an earlier manifestation of it contemporary with the Iron Age city (Branting et al. 2019: 101-
102). It also provided a good training environment for students and faculty from Kos; University in preparation for a
survey they were to carry out elsewhere in Turkey. The results, obtained over two days within a 1200 m2 area,
revealed a wall and the edge of the large water cistern. This is encouraging and lays the groundwork for completing a
gradiometry survey of the Kale in future seasons.
EXCAVATION
Excavations at Kerkenes over this period only took place during the 2019 season. The 2020 season was focused on
conservation efforts since only one of our usual team could participate in fieldwork because of the COVID-19
pandemic. During the 2019 season, all excavations took place within Urban Block 8 (UB8), near the northernmost
point of the city. This is a large urban block, covering ca. 6000 m2 • Excavations in UB8 started with test trenches in
1996, though it has been the site of annual excavations only since 2014. The focus over this span of time has been on
wide area excavations, exposing the entirety of the buried buildings and open spaces to better understand the
utilization of these structures and spaces during the Iron Age settlement. By the end of the 2019-2020 seasons ca.
1800 m2 ofUB8 had been excavated.
One trench was opened during the shortened 2019 season, Trench 45 (TR45). This was positioned to the north of
Trench 41 (TR41) and adjoins Trench 29 (TR29) and Trench 31 (TR 31) to the west (Fig. 12-5). This area was of
interest for two reasons. First, the eastern edge of TR45 encompasses the northeastern extents of the UB8 enclosure
wall, all the way back to the city wall to the north. In the early years of the project, it had been postulated that a ring-
road might have been constructed along the inner face of the city wall to facilitate the movement of troops. However,
the high stone talus that remained following the collapse of the city wall blocked geophysical prospection from
identifying the exact relationship between the northern end of the urban block, the potential road, and the city wall.
Therefore, excavation was needed. Second, two freestanding single-roomed structures, identified in this area of the
geophysical surveys, are apparently unique within UB8 in terms of their architecture and positioning. Their position
alone in the northeastern comer of the urban block is intriguing.
While the shortened season precluded the excavation of the entirety of TR45, including the interiors of the two
buildings, a large I-shaped area of the trench was exposed (Fig. 12-6). This area measured 3-3.5 min width, 13.5 m
east to west, and 29.5 m north to south. It included the entirety of the northern end of the eastern urban block wall and
area just inside the walls defining the eastern extent of TR29 and the northern extent of TR4 l. Within the western
extent of this area a small section of stone paving was found, presumably leading up to the southernmost of the two
unexposed buildings. The southern section of the I-shaped area revealed the threshold in the wall between TR45 and
TR41, a granite staircase, and a raised platform to the south of the unexcavated building. The eastern section exposed
the top of the urban block wall and a narrow space along the interior of the wall.
Figure 12-5. Configuration of trenches Figure 12-6. Aerial photograph of the extents of
within Urban Block 8. Trench 45 in 2019, outlined in white dots. The
outlines of the two buried buildings are noted as
white overlays. A small section of the inner face
of the city wall, which the urban block wall abuts,
is seen in the far northern end of the trench.
I,, , , 1, , ,, I ! cm
19TR45U19bon01
Figure 12-7. A fragmentary ivory plaque from Trench 45.
Few finds were discovered in the western section, though the analysis of soil samples collected from the surface
might yield more clues in future seasons. Over 571 liters of soil samples were collected in 2019 from across TR45.
The southern section proved more interesting, with the burnt remains of the wooden door in the wall between TR41
and TR45 found lying on top of the granite staircase. In addition, a number of artifacts were found on top of the raised
platform. These included three fragmentary ivory plaques, two of which were carved (Fig. 12-7), and a copper alloy
arrowhead. In the eastern section, just inside the urban block wall, a significant cluster of smashed in situ pottery was
The Kerkenes Project 2019-2020 151
found along with a few antler and ivory inlays (Fig. 12-8). Elucidating the relationship of these artifacts to the two
buildings within this area must await further excavation.
Despite the limited time, one definitive conclusion
was reached by the excavations in TR45 and the
exposure of the 1 m-wide urban block wall as it ran
north to the city wall. Rather than stopping short of
the city wall and turning back to the west, the urban
block wall continues right up to the fortifications,
abutting directly against them. It is now clear that the
idea of an interior ring-road just inside the city wall is
incorrect, with at least one example now demonstrating
that some of the urban blocks incorporated the city
wall as one of their exterior walls. Without this ring-
road, the rapid movement of troops along the
perimeter of the city would have been more difficult
and circuitous but not impossible.
CONSERVATION
Conservation of the architecture on site and of

objects in the excavation depots was an important
facet of the work in 2019 and represented most of the
work in 2020. On site conservation efforts in 2019
focused primarily on the Cappadocia Gate, though
additional work was undertaken on cleaning and
conservation within UB8, the Palatial Complex, and a
stretch of city wall at the entrance of the site. The
work in the Cappadocia Gate was complicated by the
delayed start to the field season and a series of very
heavy rainstorms in early June of 2019. During one
of these rainstorms, just after the team arrived at the
excavation house in Sahmurath, a portion of one of
the walls of the inner chamber of the Cappadocia
Gate fell inward. This was a section adjacent to an
area of wall fall that occurred in 2011 after
excavation in 2010. The main conservation difficulty
with the Cappadocia Gate stems from the fact that
when the entire gate was burned during the
destruction of the city in the 540s BCE, the timber
Figure 12-8. In situ pottery vessels beams that held the gate were subjected to intense
being excavated within Trench 45. heat, affecting its structural integrity. The lower
courses of the walls had remained standing in the
millennia since then simply because the upper walls collapsed and infilled the area against the wall faces. Excavations
within the Cappadocia Gate between 1999 and 2011 then removed this collapse, a necessity for fully understanding
the architecture and use of one of the city gates. However, limits imposed on backfilling as a method of preservation
precluded returning this stabilizing fill to prevent the future collapse of the structurally compromised walls. In
partnership with faculty and students in the Department of Architecture at Abdullah GUI University in Kayseri, a team
from Rekare Mimarhk in Istanbul was brought to the site shortly after the collapse to do a detailed 3D scan of the gate
with a Leica 3D RTC360 laser scanner (Fig. 12-9). These data, together with earlier 3D models and plans of the gate
(Branting et al. 2017: 170, 2019: 109), which has been monitored over the last several years, are being used to assess
the structural integrity of this section of the Cappadocia Gate and of the gate as a whole. The wedge of collapse from
2011 did stabilize the Middle Tower of the gate, and initial results suggest that this new wedge of collapse will have a
similar effect along the connecting wall and the East Tower.
Meanwhile, in the excavation depots, conservation of both freshly excavated material and material from prior
seasons of excavation were expertly tended to by the project's conservator, Soran Avcil. Yearly conservation of iron
material within the depots is a constant task, involving both monitoring the storage climate and cleaning the iron, as
necessary. We were very fortunate to be able to have Mr. Avcil on site in 2020 to continue this important work,
despite the pandemic. Even basic tasks such as changing silica gel within storage boxes can have a profound effect on
the long-term preservation of material.
In 2019, we were also able to continue the time intensive process of floating and analyzing soil samples from the
excavations. Particularly in artifact rich contexts, such as areas of TR40, the soil sampling program has greatly aided
in the recovery of very small object fragments. On-site dry sieving is impossible due to the heavy clay soil of
Kerkenes, so a wet sieving protocol for select contexts has become standard practice. 3D scanning of artifacts with an
Artec Spider structured-light 3D scanner also continued throughout 2019 (Fig. 12-10).
Figure 12-9. The Leica 3D RTC360 laser scanner Figure 12-10. Structured-light scanning of a pottery vessel
being used by the Rekare Mimarhk team to scan with the Artec Spider in the excavation depot.
the Cappadocia Gate.
SPECIALIST STUDIES
A hallmark of the Kerkenes Project has always been methodological innovation. There are currently several
projects either in progress or in development that are being carried out in collaboration with institutions around the
world. In 2019-2020, several team members continued the ongoing work of creating a more detailed building-by-
building plan of the buried city from the combined geophysical datasets collected over the past 28 years. This work
not only increases our understanding of the city but is also foundational to several computational projects currently
underway, including a social network analysis of portions of the city, statistical comparisons of potential households
and storage locations within the city, simulations of transportation and planning in the city, and computer simulations
of the final fiery destruction of the city.
Work by the team has also focused on the analysis of artifacts, including production practices in their manufacture
and the origins of their raw materials. This includes ongoing laboratory analysis of samples from the ceramic corpus at
Arizona State University, of amber inlays and metals at the University of Sydney, and of a glass bead discovered in
UBS in 2011 at the University of Sydney and the University of Central Florida. The glass bead analysis has been
partially funded through a National Science Foundation (NSF) Major Instrumentation Grant.
Developmental work has also continued during 2019-2020 on a range of tools and methods for use in future work
at Kerkenes, but with a vision of expanding their availability to archaeological projects around the world. These
projects include the use of augmented reality, neural networks, machine learning, artificial intelligence, and robotics.
Among them is the open-source software, named DATCH, meant to enable users wearing augmented reality headsets
to make scaled drawings of plans and sections while on site (ucfdatch.org). 2019-2020 saw the completion of the
prototype DATCH software through a National Endowment for the Humanities (NEH) Level II Digital Humanities
Advancement Grant (DHAG) and the recent awarding for the next phase in the development of this software through a
new NEH Level III DHAG Grant. Another project, started in 2019 in collaboration with colleagues in the Department
of Mathematics and Computer Science at the Universite du Quebec a Trois-Rivieres, is investigating the use of
machine learning and neural networks to help identify joins among groups of broken ceramics. Finally, new projects
are just being explored involving artificial intelligence and robotics with the University of Leiden.
Not all of the project's work is focused on the distant past. An important facet of our efforts over the past decade
was ethnographic research, especially the exploration of interactions between the project and the people in the village
of $ahmurath and their perceptions of the project and of the site (Baltah T1rpan 2019). This builds on ethnographic
work carried out at Kerkenes and in the surrounding region for almost a century (Morrison 1939; Ergenekon 1999).
However, this work is even more critical today, and we hope it will help lay a better foundation for developing a more
wholistic understanding of the Kerkenes Dag1 in the distant past, the more recent past, the present, and the future.
CONCLUSION
Despite the difficulties posed by the COVID-19 pandemic, the multifaceted Kerkenes Project continues to expand
our understanding of this important archaeological and cultural site. While on-going aspects of the project, such as
geophysical survey and excavation, were curtailed during this period, the innovative efforts of the project continued.
As we eventually emerge from the pandemic, the work done in the 2019-2020 seasons should lay the groundwork for
an important expansion of the project in the decades to come.
The Kerkenes Project 2019- 2020 153
ACKNOWLEDGEMENTS
The Kerkenes Project could not operate without the support of the Turkish Ministry of Culture and Tourism,
officials in Y ozgat and Sorgun, the village of $ahmurath, our financial sponsors, and the entire te8lll. Special
acknowledgement must be made of the two Ministry Representatives, Ertan Y1lmaz from the Kar8lllan Museum in
2019, and Gokhan Akkul from the Batman Museum in 2020, without whom the project would have been unable to
move forward through these difficult times. In addition, the Yozgat Museum, and its director Omer Y1lmaz, greatly
helped to facilitate this work. We are also very grateful to the current Yozgat Governor, Ziya Polat, the former Yozgat
Governor, Kadir <;aklr, the current Sorgun District Governor, Bahadrr Gune~, the former Sorgun District Governor,
Dr. Mustafa Altmpmar, the Sorgun Mayor, Mustafa Ekinci, the $ahmurath Mayor Turan Ba~filrk, and the Sorgun
Administrative Director Metin Kayhan.
The 2019-2020 seasons of the Kerkenes Project were financially supported by the Merops Foundation, the
University of Central Florida, the NSF Grant Award # 1624105, "Investigation Into The Social Organization Of An
Early City" and NSF Grant Award # 1726636, 'State of the Art Imaging X-Ray Photoelectron Spectroscopy:
Integrating Nanotechnology, Catalysis, Manufacturing, Physics and Archaeological Research with Education', the
Berelsman Foundation, Archie Wainright, the NEH Digital Humanities Advancement Grants HAA-256218 and HAA-
277278, and an anonymous private donor.
REFERENCES CITED
Anderson, John George Clark. 1903. Studia Pontica I. A Journey of Exploration in Fontus. Bruxelles: H. L8lllertin,
Libraire-Editeur.
Baltah T1rpan, Sevil. 2019. Rupture Between the Archaeological Sites and Local Communities. In Public
Archaeology: Theoretical Approaches and Current Practices in Turkey, I. Glirsu, ed., 51- 56. C8lllbridge:
McDonald Institute for Archaeological Research.
Bittel, Kurt. 1960/1961. Legenden vom Kerkenes-Dag (Kappadokien). Oriens 13/14: 29-34.
Branting, Scott. 1996. The Ali~ar Regional Survey 1993-1994: A Preliminary Report. Anatolica 22: 145-15 8.
- . 2004. Iron Age Pedestrians at Kerkenes Dag: An Archaeological GIS-T Approach to Movement and
Transportation. Ph.D. dissertation, Department of Anthropology, State University of New York at Buffalo.
Branting, Scott and Geoffrey D. Summers. 2002. Modelling Terrain: The Global Positioning System (GPS) Survey at
Kerkenes Dag, Turkey. Antiquity 76: 639-640.
Branting, Scott, Joseph W. Lehner, Sevil Baltah Trrpan, Sarah R. Graff, John M. Marston, Tuna Kalayc1, Yasemin
Ozarslan, Dominique Langis-Barsetti, Lucas Proctor, and Paige Paulsen. 2017. The Kerkenes Project 2015- 2016.
In The Archaeology of Anatolia Volume II: Recent Discoveries (2015-2016). S.R. Steadman and G. McMahon,
eds., 154- 175. Newcastle upon Tyne: C8lllbridge Scholars Publishing.
Branting, Scott, Joseph W. Lehner, Sevil Baltah-Tirpan, Dominique Langis-Barsetti, Tuna Kalayc1, Sarah R. Graff,
Lucas Proctor, Nilufer Baturayoglu Yoney, Burak As1liskender, Canan <;akrrlar-Oddens, and John M. Marston.
2019. The Kerkenes Project 2017- 2018. In The Archaeology of Anatolia, Volume III: Recent Discoveries (2017-
2018). S.R. Steadman and G. McMahon, eds., 99-111. Newcastle upon Tyne: Cambridge Scholars Publishing.
Donmez, $evket. 2004. Akalan Hakkmda Yeni Gozlemler. Anadolu Ara~tzrmalarz 17(1 ): 67- 92.
Ergenekon, Beg~en. 1999. The Ethnoarchaeology of $ahmurath Village by Kerkenes Excavations in Turkey.
Caesaraugusta 73: 169- 175.
Herodotus. (1987). The History: Herodotus. D. Grene, translator. Chicago: University of Chicago Press.
Kealhofer, Lisa, Peter Grave, Hermann Genz, and Ben Marsh. 2009. Post-collapse: The Re-emergence of Polity in
Iron Age Bogazkoy, Central Anatolia. Oxford Journal ofArchaeology 28(3): 275- 300.
Kv8lllille, Kenneth L. 2003. Geophysical Surveys as Landscape Archaeology. American Antiquity 68(3): 435--457.
Langis-Barsetti, Dominique. 2013. Results of the 2011 Geophysical Survey at Kerkenes Dag, Central Anatolia.
Anatolia Antiqua 21: 69-86.
Morrison, John A. 1939. Ali~ar: A Unit of Land in the Kanak Su Basin of Central Anatolia. Ph.D. dissertation,
Department of Geography, University of Chicago.
Przeworski, Stefan. 1929. Die Lage von Pteria. Archiv Orientalni 1: 312-315.
Ross, Jennifer C., Sharon R. Steadman, Gregory McMahon, Sarah E. Adcock, and Joshua W. Cannon. 2018. When
the Giant Falls: Endurance and Adaptation at <;adir Hoyuk in the Context of the Hittite Empire and its Collapse.
Journal ofField Archaeology 44(1): 19-39.
Ross, Jennifer C., Gregory McMahon, Yagmur Heffron, Sarah E. Adcock, Sharon R. Steadman, Benjamin S.
Arbuckle, Alexia Smith, and Madelynn von Baeyer. 2019. Anatolian Empires: Local Experiences from Hittites to
Phrygians at <;adu Hoyuk. Journal ofEastern Mediterranean Archaeology and Heritage Studies 7(3): 299- 320.
Schmidt, Erich F. 1929. Test Excavations in the City on Kerkenes Dagh. American Journal of Semitic Languages and
Literatures 45(4): 83-92.
Summerer, Latife. 2007. Greeks and natives on the southern Black Sea Coast in Antiquity. In The Black Sea; Past,
Present and Future, G. Erkut and S. Mitchell, eds., 27-36. London: British Institute at Ankara.
Summers, Geoffrey D. 2006. Phrygian Expansion to the East: Evidence of Cult from Kerkenes Dag. Baghdader
Mitteilungen 37: 647- 658.
- . 2021 Excavations at the Cappadocia Gate. Kerkenes Final Reports I. Chicago: Oriental Institute Press.
- . Forthcoming. Excavations at the Palatial Complex. Kerkenes Final Reports II. Chicago: Oriental Institute Press.
Summers, Geoffrey D. and Frarn;:oise Summers. 1998. The Kerkenes Dag Project. In Ancient Anatolia: Fifty Years'
Work by the British Institute of Archaeology at Ankara, R. Matthews, ed., 177-194. London: British Institute of
Archaeology at Ankara.
- . 2010. From Picks to Pixels: Eighty Years of Development in the Tools of Archaeological Exploration and
Interpretation, 1927-2007, at Kerkenes Dag in Central Turkey. In Proceedings of the 6th International Congress
on the Archaeology of the Ancient Near East, May 5th -10th 2008, Sapienza-Universita di Roma. Volume 2
Excavations, Surveys and Restorations: Reports on Recent Field Archaeology in the Near East, P. Matthiae and L.
Romano, eds., 669-683. Wiesbaden: Harrassowitz.
Summers, Geoffrey D., Frarn;:oise Summers, Nilufer Baturayoglu, Omur Harman~ah, and Elspeth R. McIntosh. 1996.
The Kerkenes Dag Survey: An Interim Report. Anatolian Studies 46: 201-234.
Summers, Geoffrey D., Fran9oise Summers, and Scott Branting. 2007. Kerkenes News 9, 2006. Ankara: METU Press.
Summers, Geoffrey D., Fran9oise Summers, David Stronach, and Musa ()zcan. 1999. Kerkenes News 2, 1999.
Ankara: METU Press.
Texier, Charles. 1839. Description de l 'Asie Mineure: faite par ordre du gouvernement fram;ais en 1833-183 7. Paris:
Le Ministere de I 'Instruction Publique.
von der Osten, Hans Henning. 1928. An Unnoticed Ancient Metropolis of Asia Minor. Geographical Review 18: 221-
274.
- . 1937a. The Alishar Hiiyiik: Seasons of 1930-32, Part 11. Chicago: University of Chicago Press.
- . 1937b. The Alishar Hiiyiik: Seasons of 1930-32, Part III. Chicago: University of Chicago Press.
CHAPTER THIRTEEN
SiNOP KALESi ARCHAEOLOGICAL EXCAVATIONS:

THE 2017 AND 2019 FIELD SEASONS
OWEN DOONAN, ANDREW GOLDMAN, ALEXANDER BAUER, ULRIKE KROTSCHECK,

PAOLO MARANZANA, NICOLE ROSE, JANE REMPEL, SUSAN SHERRATT,
EMiNE S6KMEN, AND HDSEYiN VURAL
The Sinop Regional Archaeological Project (SRAP) continued with a third campaign of excavations at the site of
Sinop Kalesi in July 2017, followed by a study season in 2019. The excavation builds on more than a decade of survey
and environmental research in Sinop and ties in with the longer-term regional project through ongoing environmental
studies, ceramic analyses, and regional-scale archaeological research. Major goals of the first phase of excavations
(2015-2017) include:
1) to test the hypothesis that the maritime culture in the Black Sea, prior to the foundation of a Greek colonial
settlement, laid the foundations for cultural and economic developments during Ionian and subsequent colonial
systems;
2) to examine the evolution of the city wall in relation to the development ofurban-peri-urban settlement;
3) to establish a stratified sequence of ceramics spanning proto-historic-medieval periods along the central north
coast of Anatolia.
The present chapter summarizes results that are presented in full detail in a forthcoming volume, Sinop Kale
Excavations, 2015-2017 (Doonan et al. forthcoming). 1 For an overview of the 2015-16 field seasons please see our
contribution in Archaeology ofAnatolia: Recent Discoveries vol. II (Doonan et al. 2017).
THE 2017 FIELD CAMPAIGN
The third season of the Sinop Kale Excavations' initial three-year campaign (2015-2017) continued our
preliminary exploration of the stratigraphic sequence between the northern section of the Hellenistic fortification wall
and the western scarp of the Kale. The 2017 season made progress on several outstanding problems identified in our
most recent report in this series, particularly the nature and stratigraphy of the wall initially identified as "Structure l,"
the earliest architectural feature recorded at the site (Doonan et al. 2017). The fieldwork was carried out to document
the stratigraphy of the earliest recorded settlement in the city and the Kale as well as to test and refine the Sinop Urban
Settlement Sequence (SUSS), a general categorization scheme structured to outline the major periods of settlement
activity currently recognized at Sinop (Table 13-1).
Excavation carried out in four trenches distributed across the project's allocated investigation area (Urban Parcel
Sinop 2 pafta 8 ada 44, henceforth Parcel 2.8.44; see Fig. 13-1) unearthed evidence for several dynamic episodes of
terracing and fortification construction across the zone. By the end of the Sinop Kalesi 2017 field campaign the team
had identified ten principal stratigraphic phases (Sinop Kale [SK] Phases 1-10), from sterile soil (SUSS 10) and traces
of Bronze Age cultural material (SUSS 9) to the Ottoman period (SUSS 1). The excavation of Sinop Kale Phases 6
and earlier are particularly noteworthy, representing the first stratigraphic investigations of ancient Sinope for pre-
Hellenistic (SUSS 5) occupation.
Parcel 2.8.44 proved useful for exploring Bronze Age through Hellenistic settlement activity on the Kale primarily
for its convenient access to Sinop's earlier strata. Leveling activities inside the parcel, undertaken in the 1960s to
create a minibus (dolm~) parking lot adjacent to the city's old bus station, removed an estimated 3 m of post-
Hellenistic accumulation, affording the team direct, unimpeded access to Sinop's Hellenistic and pre-Hellenistic strata
1
The volume includes a general overview of the articulation of the Sinop Kale Excavations with the Sinop SUIVey (Doonan), a
detailed stratigraphy of the Sinop Kale excavations (Goldman), a synthesis of the evolution of the city plan (Doonan and Vural),
analyses of the city wall in Hellenistic (Rempel and Doonan) and Early Byzantine (Maranzana and Domzalski) phases, and
overviews of the handmade (Bauer and Sherratt; Rose et al.), and colonial ceramics (Krotscheck and Rempel), floral and faunal
remains (Piskin), and analysis of Prof. Ludwig Budde's archive of the 1951-53 excavations in Sinop (Krotschek, et al.).
156 Chapter Thirteen: Doonan et al.
SUSS PHASE SK PHASE PERIOD DATES

0 = 0 MODERN/REPUBLICAN 1923 to present
1 = 1 OTTOMAN 15 1h-20 th century CE
2 = 1 SULJUK 13 1h-15 1h century CE
3 BYZANTINE
3D - Late Byzantine 11 th-13 th century CE
3C - Middle Byzantine 1Oti,_ 11 th century CE
3B - Byzantine Dark Ages 7th-9 th century CE
3A = 2 Late Roman/E. Byzantine Late 3rd- 7th century CE
4 ROMAN
4C = 3 High/Middle Empire 2" d-mid-3'd century CE
4B = 3 Early Empire 46 BCE- I st century CE
4A = 4 Late Republic (siege to colony) 70--46 BCE
5 HELLENISTIC
5B - Pontic Kingdom 183-70 BCE
5A = 5 Early Hellenistic c. 300-183 BCE
6 = 6 CLASSICAL C. 479-c. 300 BCE
7 = 7? ARCHAIC/GREEK SETTLEMENT C. 630-479 BCE
8 = 8? EARLY IRON AGE c. 1200/1100-c. 630 BCE
9 = 9 BRONZE AGE c. 3000-c. 1200/1100 BCE
10 = 10 STERILE SOIL --
Table 13-1. Chronological table outlining SUSS and associated SK Phases

(based on chronological framework in Doonan 2004).
(SUSS 5A and earlier). Lamentably, the Kale's

Sinop Kale Excavation
2015-17 Roman and post-Roman strata (SUSS 1-4) in the
North Tower
Parcel 2.8.44
parcel were demolished, as was the external terrace
=Tower A
C. Merrony
surface to the west of, and contemporary with, the
Hellenistic curtain wall (SUSS 5A), a surface that lay
0 5 10m
ca. 10-15 cm above the modem parking lot, as
evinced by the elevation of the threshold in the sortie
door of Tower A. However, their removal aided our
initial objectives, as did the Sinop Museum's
trenching and clearing operations at the southern end
of the parcel in 2012. This exposed, undeveloped,
-------. urban area provided the SKE with comparatively
--
easy access to the settlement's earliest identified
areas of occupation, a long sequence of robust
settlement activity, and a sensitive urban liminal
zone, an interface between urban astu and peri-urban
chora at Sinope's westernmost boundary.
The four-week 2017 field season in Parcel 2.8.44
-------. saw excavation continue in three pre-existing
trenches (Operations 1, 2, and 4) and in a new 2 x 2
-------. m sondage (Op. 5) on the eastern side of the
Hellenistic fortifications (Fig. 13-1 ). The objective
-------. for Op. 5, which cut into a sloping, exposed surface
Op. 2 abutting the old bus station parking lot, was to
-------. determine whether any trace of the Kale's later
-------. occupation periods (SUSS 1-4) remained at one of
-------
-------
the highest preserved elevations on the modem Kale.
-------
------- GI]
Additionally, the trench lay ca. 20-25 m west of
where E. Akurgal and L. Budde situated a sondage
'' ------- during their 1951-53 investigations (Akurgal and
Budde 1956; Erzen 1956: lev. XXVIII); it was hoped
z -------. that the new trench would shed light on the results of
all
0 -------.
their earlier, largely unpublished fieldwork. The Op.
5 results were disappointing, as limitations of time
and resources forced its closure at a depth of only ca.
-------. 1.45 m from the modem surface. Most contexts
proved to be mixed, spanning the early Hellenistic
Figure 13-1. Plan of Sinop Kale excavation period to the Ottoman eras (SUSS 1-6); associated
trenches, 2017 (drawing by A. Goldman). ceramics and objects included a black-and-red
Sinop Kalesi Archaeological Excavations 157
fusiform unguentarium of Early Hellenistic date, and a nearly intact, yellow- and brown-glazed oil lamp of the 12th-
13th century CE. However, below a pair of modem, sandy leveling fills and several thin lenses of reused fill, ca. 1.05
m from the modem surface, were two small features: a circular trash pit (ca. 1.1 m in diameter) and a small, ephemeral
hearth located in the middle of the trench. Although they proved impossible to date with any precision, these lower,
better-preserved strata suggest that continued work in the parking lot area holds significant future potential.
The main objective of the 2017 season was a continuation of stratigraphically secure, vertical clearing in Ops. 1, 2,
and 4, to record deposits of the early Hellenistic (SUSS 5A) and pre-Hellenistic periods (SUSS 6-10). Ops. 1 and 4
are a set of contiguous east-west trenches which measure 6 m (north-south) x 10 m (east-west) and lie just south of
the base of Tower A. Both trenches begin ca. 1 m west of the Hellenistic fortification wall and extend westwards
towards the Kale's steep scarp for nearly the entire width of the parcel. Their general stratigraphic profile was
disrupted by several significant later features, specifically the Byzantine fortification circuit (ca. 2.5 m wide) and a
mortared 19th -century terrace wall along the western baulk. Despite these later intrusions, the parcel proved to contain
well preserved evidence for pre-Hellenistic activity on the Kale, particularly in SK Phases 6 and 7, which lay just
below the modem parking lot surface. Op. 2 lies ca. 10 m south of Op. 4's southern baulk, and this north-south
oriented trench covered an area of 4.5 m (north-south) x 6 m (east-west) by the end of the 2017 season. It is situated
directly against the Hellenistic wall, where its foundation blocks and construction trench had been partially exposed
by the Sinop Museum. As a result, Op. 2 proved ideal for relatively rapid access to both the foundation deposits of the
Hellenistic wall and the pre-Hellenistic strata (SUSS 5 and earlier), eventually reaching a layer of reddish, sterile soil
(SUSS 10) at a depth of ca. 2.6 m below the parking lot surface.
SK Phase 2
Byzantine
Circuit Wall
SK Phase SA
Fortification
Ditch (Locus 24)
Figure 13-2. Drone view, Ops 1 and 4, indicating location of Phase 7 and 6 features (Sinop Kale Excavations).
The investigation of SK Phases 6 and 7 in Ops. 1, 2, and 4 revealed dynamic construction activities on the western
edge of the Kale which altered considerably the area's topography and function prior to the construction of the
Hellenistic curtain wall (Fig. 13-2). The surface of SK Phase 7, the earlier of the two strata, rests ca. 1.45 m below the
modem surface and represents the earliest substantial occupation level explored on the Kale. Although a deeper
sounding was undertaken in Op. 4, to a depth of ca. 2.15 m, the small area of coverage (ca. 1 x 0.7 m) proved the
existence of a lower stratum, SK Phase 8, but little about its character or date. SK Phase 7, on the other hand,
contained a substantial north-south running wall of unmortared rubble (Op. 4 ~ 75), ca. 75-80 cm in width and
preserved in places up to four courses in height (ca. 50-55 cm). Preserved between the Byzantine wall foundations on
the west and the Hellenistic wall foundation trench on the east, this feature extends from Op. 4's southern baulk
northwards into Op. 1, terminating at its north baulk for a total preserved length of 11.5 m (Fig. 13-3). The wall's
original function remains unclear, although its replacement by increasingly large and sophisticated defensive barriers,
in SK Phases 6 and 5A, suggests that it also served as part of the settlement's early defensive network.
On either side of this wall lay a flat, contemporary, exterior surface with no discernible features. The eastern side
in both Ops. 1 and 4, however, was found covered by a thick blanket of rubble debris which lay alongside and directly
abutting the wall (Fig. 13-4). The relatively even distribution of this flattened rubble, which matched in composition
the preserved stone matrix of the long wall, suggests that the wall's upper portions were carefully demolished,
tumbled to the east and equably redeployed to form a level base fill. Its total volume indicates that the long wall must
originally have been a substantial barrier approaching at least a meter in height, if not more. Its careful destruction
suggests that it was demolished as part of a well-planned and executed rebuilding effort on the Kale, the initiation of
the succeeding SK Phase 6. The homogenous character of the overlying soil, a thick (ca. 75 cm), uniform layer of dark
brown fill seemingly dumped in place as part of a single episode, reinforces this impression.
Figure 13-3. SK Phase 7 wall (Locus 75) and SK Phase 6 wall (Locus 74) in Op. 4 (photo by A. Boccaccio).
Operation 4
2017 1:20
North Section
NW NE
Corner Approxi mate Level of SK Phase 5 Hellenistic Terrace Surface Corner
Byzantine Wall
Rubble Fill and
SK Phase S
Block Facings
(SK Phase 2) ..-------r
1
Hellenist ic
Wall
Byza ntine Wall Foundation
Foundation Trenches Tre nch
SK Phase
?Wall
1m 2m (Locus 75)
Figure 13-4. Section drawing (indicated on Figure 2), showing interrelationship of SK Phases 5-8
(drawing by A. Goldman).
The construction of SK Phase 6 appears to have transformed the western Kale into a far more substantial and
monumental defensive zone. In Ops. 1 and 4, our investigation exposed two significant, likely interrelated features
bisected by the Byzantine fortification wall (Fig. 13-2). To the west was a large, stone-built structure, which stretched
in a roughly north-south direction from the southern baulk of Op. 4 for a total distance of 7.75 m in length, to the
Byzantine wall foundation trench in Op. 1 (Fig. 13-5). In 2017, a 1.2 x 1.35 m sondage, cut to a depth of 1.7 m from
the modem surface in Op. 4, revealed the full vertical extent of the feature , which consists of nine preserved, inwardly
slanting courses, with large, irregular stones set into the base (probably below ground) and increasingly smaller, more
uniformly rectilinear river-worn stones stacked towards the top. The stones vary widely in size, with a maximum
length of 50 cm and a width of 25 cm. The wall appears to have been unmortared and the exterior face unplastered. In
the baulk between Ops. 1 and 4, atop the river-worn stones, were traces of degraded and burnt mudbrick (see Doonan
et al. 201 7: fig. 9-1 7). This set of alternating red and grayish deposits extended southwards over 1.2 m along the top of
the wall, and degraded mudbrick melt or collapse was also traceable in lenses ca. 20 cm thick lying to the west of the
wall face. Upon its discovery in 2015, this feature ("Structure l") was suspected to represent a portion of an
Early/Middle Iron Age (i.e., pre-colonial) pit house, of the type characterized by a use of river-worn stones,
recognized nearby during the cleaning of the western Kale scarp in 2000. More broadly exposed, it is now understood
that this impressive feature is a low, slanting glacis wall, the river-worn stones placed in a secondary-use context
(probably salvaged from earlier structures in the area) and capped by a mudbrick superstructure, most of which was
subsequently removed during construction activities associated with SK Phase 5.
I y',, -
Figure 13-5. River-worn stone glacis, SK Phase 6 (photo by A. Boccaccio).
The second significant feature of SK Phase 6, on the eastern side of the Byzantine wall, is a fragmentary, north-
south oriented line of stones (Locus 74) which stretches across portions of Op. 1 and 4 for an approximate length of
ca. 7.5 m (Figs. 13-2 and 13-3). The feature is poorly preserved, lying only ca. 15 cm below the parking lot surface. A
deep east-west ditch-a subterranean component of the SUSS 5A Hellenistic fortification system (see below)-
truncates its northern end in Op. 1, while its southern end is terminated near the midpoint of Op. 4 by its intersection
with the Byzantine wall trench. In contrast to the well-built glacis, its construction quality is variable, as suggested by
its fragmentary remains, which consist of two courses composed of irregular, unmortared sandstone and ashlar
fragments ca. 50-60 cm in width, placed in a line that is a single row thick. Cut into the dark brown clay fill dumped
above the SK Phase 7 surface, the feature is contemporary with the river-stone glacis and aligns in a roughly parallel
direction (Fig. 13-2). It is thus preliminarily conjectured that these two features represent the exterior faces of the
same structure: a massive, double-faced wall, roughly ca. 4.5 m thick, which delineated the western, landward edge of
the pre-Hellenistic city. In its complete form, this fortification wall would have consisted of: a stone socle/glacis on
the western, outer facing side, set more deeply into the Kale slope; a more simple, rough stone packing along the
eastern face; a thick earth-packed core; and a superstructure of mudbrick. A construction project of this magnitude
would have required significant resources and planning on the part of the Sinopean community. Such activities are
seemingly borne out by the careful intentional tumbling and burial of its antecedent wall (in SK Phase 7), followed by
the importation of significant fill to raise up the Kale surface on the eastern side of the wall between ca. 0.75-1.0 m.
The result would have been a truly impressive barrier, similar to the type ofreconstructive and/or replacement efforts
known at numerous other Mediterranean and Black Sea settlements, including Old Smyrna, Massalia, and
Myrmekeion in the Kimmerian Bosporos (Frederiksen 2011: 46--47).
Since relatively few artifacts can be associated
with SK Phase 7, it has been difficult at present to
ascertain its date or that of the succeeding stratum.
Present evidence suggests that the transition between
the two phases took place in the 6th century BCE.
An imported, painted archaic Ionian amphora
handle (SK 16938) from a deposit associated with
the deconstruction of the SK Phase 7 wall in Op. 4
dates to the mid-6 th century BCE (Krotscheck and
Rempel forthcoming). In addition, a Lydian streaky
ware sherd with bent swastika design in reserve
against a reddish-brown slipped field, dating to the
7th-mid-6 th ca. BCE, was found embedded in the
remaining mudbrick atop the glacis. Removal of the
rubble in Op. 1 exposed a series of broken, mostly
handmade vessels mixed in with the rubble deposit
Figure 13-6. Deposit of ceramics from Op. 1 (Fig. 13-6). Since all were in a smashed yet relatively
(photo by A. Boccaccio). complete condition, it appears likely that they
represent trash tossed into the area during the
demolition activities associated with the end of SK
Phase 7, either just prior to or in tandem with the
creation of the debris field from the long wall
deconstruction. One of these pots, a large, round
handmade storage vessel (SK 16384), with a
greyish-brown burnished exterior, was placed
upside-down near the top of the debris field, its rim
and base broken away prior to its deposition (Fig.
13-7a-b). Analysis of the soil inside revealed traces
a of its original contents, a dozen vertebrae from
anchovies (Turkish hamsi), a species which remains
a cherished local delicacy in Sinop even today (Fig.
13-7c). At present, then, a preliminary dating of the
strata seems to support relating SK Phase 7 with the
Archaic (i.e., early colonial) period (SUSS 7), with
the construction of the more robust fortifications in
SK Phase 6 dating to either Late Archaic (SUSS 7)
or Classical (SUSS 6) periods of Sinope.
In all three trenches, the 2015-2017 excavations
provided significant data about SK Phase 5/SUSS
SA, the stratum associated with the construction of
Sinop's best-preserved ancient monument. As was
the case with SK Phase 7, the Kale was transformed
yet again with the truncation of existing monuments
Figure 13-7. a and b) SK 16384, reconstructed vase
(e.g., the glacis wall) and the construction of a
and drawing; c) Hamsi bones (photos by A. Boccaccio
higher, wider, westward-facing terrace. The latter
and drawing by T. Ross).
was achieved by adding a thick, greyish-green clay
fill to the immediate west of the glacis face, a
deposit that was somewhat surprisingly found to contain Bronze Age and Iron Age material, including large amounts
of fragmentary handmade pottery, flint blades, spindle whorls, and three statuette fragments, two of which (SK 1693 7
and 17089) resemble Early Bronze Age figurines from Ko9umbeli and Ali~ar (Fig. 13-8), (cf. Atakuman 2017: 86, 91,
Figs. le and Sb; also see Bauer and Sherratt forthcoming). While Bronze Age material has been recorded elsewhere on
the Sinopean peninsula (Doonan 2004: 51-56; I~m 1998), these new finds represent its first scientific documentation
in the vicinity of Sinop itself. While the original context of the grey fill is not currently known, the soil is likely to
have been harvested from a source on or adjacent to the Kale. Consequently, although a primary Bronze Age deposit
has not been located to date on the Kale, we have preliminarily assigned SK Phase 9 to designate a Bronze Age
presence, one which can be generally associated with Sinop's SUSS 9.
Q ..
SK 16937 SK 17089
0 5cm 0 5cm
Figure 13-8. Early Bronze Age figurines (SK 16937 and SK 17089) (drawings by T. Ross).
Tremendous resources were brought to bear on the task of creating the SUSS 5A Hellenistic circuit wall system,
which included a curtain wall over 300 m long, with bevel-comer towers at each end, and at least four further towers
punctuating the curtain at strategic intervals. Our work has revealed, for the first time, the full morphology of this
wall, such as the existence of at least one subterranean defensive ditch (in Op. 1, Locus 24), at least 7 m long (east-
west), 2 m wide, and ca. 2 m in depth. The trench was filled with a tightly compacted deposit of burnt charcoal and
ash, animal bones, pottery sherds of coarse and common ware, iron fragments, significant amounts of broken pan and
cover tiles (some stamped), and 13 large ballista balls (ca. 40-46 cm in diameter). In 2017, a large, still legible bronze
coin (YH 16550) was recovered lying upon the ditch' s floor, a pt_century BCE issue of Amisos with parallels dating
to ca. 85- 65 BCE (cf. SNG BM Black Sea 1212, AE 22 mm, 12.64 gr., 1213-14 and SNG Stancomb 701 var). In light
of this find, our current hypothesis is that the mixed fill could represent debris from the siege of Sinop by Lucullus in
70 BCE.
NW NE
corner corner
Mode rn Co ncrete Hell enistic
Wa ll
/ Fou nda ti on
Bl ocks
Hellenistic
Pit Locu s 10 1 Wa ll
Foundat ion
Trench
SK Ph ase 5
Operation 2
2017 1:20
North Section
SK Phase 10:
0 0.5 lm Sterile Soil
Figure 13-9. Section from Op. 2 indicating the stratigraphy of the construction fills,
Hellenistic wall (drawing by A. Goldman).
The 2017 excavations also revealed substantial evidence relating to the Hellenistic wall's construction processes,
particularly through our work in Op. 2, which has provided the first stratigraphic evidence for the dating of the wall
(Fig. 13-9; see Rempel and Doonan forthcoming). The first stage of construction involved digging a wide trench ca. 3
m from the line of the wall, with a slope of approximately 45 degrees, and a deeper cut ca. 1.5 m from the base of the
SK 16331
0 5cm
Figure 13-10. Photo and profile of West Slope Ware sherd, SK 16331 from the construction fill
of the Hellenistic wall in Op. 2 (photo and drawing by A. Boccaccio and T. Ross).
wall, with a depth of ca. 2 m below the modem surface. Initial analyses of the pottery from the Op. 2 foundation
trench suggests that all diagnostic material, including imported material from the Aegean, which includes a significant
quantity of black slip pottery (e.g., Attic West Slope Ware; YH SK 16331; Fig. 13-10), comes from the 3rd century
BCE or earlier. While building fills typically contain material that covers a long date range, and is characterized by
mixed fragments (as is the case at Sinop), and while a full analysis of the tile and amphora fragments is still underway,
it seems likely, based on the predominance of 4th-3 rd century BCE imports, and the absence of key diagnostics from
the 2nd century BCE, that the fortification wall was actually built in the third century BCE-likely in the second
quarter or middle of that century. This discovery represents a significant chronological shift, as previous scholars had
identified its construction with the period when Sinope was the capital of the Pontic Kingdom, and specifically with
the reigns of one of two kings: Phamakes I, who made Sinope the capital of the Pontic Kingdom in 183 BCE; or
Mithridates VI, the last king of Pontos, who was born in Sinope and reigned from ca. 120-63 BCE (Bryer and
Winfield 1985: 70, 76-7; Barat 2010: 45; Crow 2014: 39). Our results now indicate that this elaborate fortification
system is unconnected with these rulers and that it was built nearly a century before the conquest of Sinope by
Phamakes I (183 BCE).
THE 2019 STUDY SEASON: LASER SCANNING OF SINOP KALESi
Three-dimensional surveys have become widely used for the documentation of archaeological sites. They enable
researchers to recreate and visualize large monuments against the risk of damage and to provide the possibility of
sharing for academics and the public (English Heritage 2007). During the 2019 study season, the fortification walls of
Sinop Kalesi, encompassing the Hellenistic to Ottoman phases, were scanned with a Faro Focus M70 laser scanner
made by the architectural firm Odaba~1 Mimarhk Restorasyon LTD STi. Scanning and modeling of the Sinop
fortification walls presented unique challenges. The necessity of scanning the fa9ade of the inner and the outer walls
separately required a great deal more time in order to connect the sessions and achieve complete point clouds. For a
satisfactory alignment, overlapped features are required via multiple scans of the target area. For example, in order to
combine session data from the inner part of the south wall with session data from the outer portion, approximately 200
sessions had to be precisely linked and the margin of error minimized. The other problem was that the fortification
wall was divided by a street on the east-west axis, and this interruption caused a discontinuity by separating the
northern parts of the fortification from the rest.
In order to overcome these logistical challenges, architectural surveying was followed by first scanning and
modeling the entire south wall, 70% of the east and west walls, then the entire north wall, and the remaining 30% of
the east and west walls separately. By combining two coherent models together, the sequential sessions set up along
the street were used to ensure the precise alignment of these two models. In the modeling of the fortification walls,
sessions were set up in 259 different locations, each collecting 360 degrees of data from an area of about 70 m radius
and obtaining GPS, altimeter, and compass data for each session (Fig. 13-11 ). These data were integrated into the
point data and ensured the precision of the study.
FARO Scene software was used to model the sessions. Three different methods are used for combining laser
scanning data in Scene: target based, top-view based, and cloud-to-cloud registration. During the modeling of the
Sinop fortification walls, the target based method was not used in order to avoid errors that may occur due to the
protruding surfaces of the walls. Instead, a draft model was created by using the top-view based method based on the
GPS data and alignment of the vertical elements. Then cloud-to-cloud registration was applied to provide much more
precise consolidation. As a result, modeling of an area of approximately 60,000 m 2 (300 x 200 m) was completed with
an error margin of2.85 cm.
>-'-d.___,..-- - - ,
~ ... ,_
" , n
,J
' ,~ \ '\
Figure 13-11. Top: Session plan-red line represents the street that divides the fortifications in two;
Bottom: plan of the fortification produced by FARO Scene (plans by c;. Kilm9 ).
After the session merging process was completed over FARO Scene, it was transferred to the Auto Desk Recap
program for viewing, analyzing, and creating 2D architectural drawings (Fig. 13-11 ). Recap has created a platform
where the entire area can be viewed through the point cloud, and in addition, 360-degree panoramic high-resolution
images can be used from the session's position. In other words, it reproduced the Sinope fortification wall in a virtual
environment. These applications have produced a high-precision measurable model that is the basis of the ongoing re-
study of the architectural remains (Figs. 13-12, 13-13, and 13-14) and forms the basis of the site plan our team is
employing as we prepare to expand excavations over the entire Sinop Kalesi site.
Figure 13-12. An overview of the scans taken by the survey (scan by<;. K1hn9).
Figure 13-13. Autodesk Recap measurements of the south wall of the fortification (photo by<;. Klhn9).
Figure 13-14. Autodesk Recap measurements of the northern part of the wall (photo by<;. K1hrn;).
CERAMIC ANALYSES USING PORTABLE X-RAY FLUORESCENCE (pXRF)
One important aim of the Sinop Kalesi Excavations has been to investigate whether pre-colonial networks of
interaction can be documented at the Sinop Kalesi site, and at what points Sinop may have functioned as a maritime hub
in the Black Sea (Doonan et al. 2017). Ceramic investigations offer a particularly interesting insight into movements of
people and goods into Sinop. While ceramic vessels are sometimes the focus of trade/exchange, either as commodities or
containers, ceramic production often unfolds locally, reflecting the technological practices of discrete sub-communities
(Gosselain 1992, 2000). This report summarizes general findings that will be reported in detail in our forthcoming
volume (Doonan et al. forthcoming). For the contextual discussion of handmade assemblages from Sinop Kalesi see
Bauer and Sherratt (forthcoming). For discussion of the methodological and theoretical issues related to our use of
portable x-ray fluorescence (pXRF) for geochemical compositional analysis in the field see Rose et al. (forthcoming).
This analysis focuses on ceramics collected through excavations at the Sinop Kalesi site in the 2015-2017 field
seasons, with pXRF analysis conducted in the field laboratory in both 2016 and 2019. Briefly, pXRF testing protocol
targeted the ceramic fabric through a fresh break, and each sample was run at two settings on two test spots to expand
the element range targeted as well as to account for possible matrix effects caused by coarse, heterogeneous fabrics
(see Forster et al. 2011; Shackley 2011; and Frahm 2018 for discussion). Sampling in 2016 focused on a collection of
90 mostly handmade open bowls decorated with horizontal bands of bosses, impressed decorations, or rope-like relief
(Fig. 13-15). These wares are typical of traditions known from points around the Black Sea, and appear in a diversity
of style that suggests non-local engagement at Sinop in the pre-colonial and early colonial periods (Bauer and Sherratt
forthcoming). Clay from a bed exposed by an erosional scarp approximately 500 m westward along the shoreline from
the Kale site was sampled and tested as a possible local clay source.
Three clusters of clay sources were identified through a series of cluster and principal component analyses (Rose et
al. forthcoming). Group 1 is the predominant cluster, including over half (53.3%) of the dataset as well as the local
beach clay sample. Though most samples in this group are visibly consistent with local traditions, a few samples
marked in the field as possibly non-local also cluster here. For example, SK15240.3 demonstrates parallels to Early
Transcaucasian ware, which, though originating in the eastern Black Sea area, is known across the Near Eastern world
in the Early Bronze Age, including several sites in the Sinop area (Bauer and Sherratt forthcoming; see also Thissen
1993; Bauer 2006).
Group 2 comprises 22.2% of the dataset and is again typologically consistent with local traditions with few
exceptions. Two samples (SK15611 and SK 15225.4) are possibly Early Transcaucasian Culture (also known as the
Kura-Araxes culture, 3rd millennium BCE) and Kizil Koba culture (Crimea, 1st millennium BCE) (Bauer and Sherratt
forthcoming). However, given the geological heterogeneity of the Sinop region, it is very likely that this group also
represents local sources.
CLUSPLOT( cldata )
-.::I"
j.
-
N
C
Q)
C
N
0
a. 15
E
0 0
()
N
I
-4 -2 0 2 4 6
Component 1
These two components explain 49.88 % of the point variability.
Figure 13-15. pXRF 2016 Data K-Means Analysis (chart byN. Rose).
Group 3 is overall rarer, represented by only 11 samples and marked compositionally by lower iron and higher
calcium levels. Most of these samples are technologically and decoratively distinct from the rest of the dataset,
including painted and slipped wares. Typological analysis in the field had suggested Iron Age, Hellenistic, and/or
Byzantine periods for several of these sherds. All sherds with typological parallels to Iron Age sites in the Bafra
district 100 km east of Sinop appear in this cluster, in addition to the Ionian and Kizil Koba types mentioned above.
Dispersed clustering within this group may indicate that these may in fact point to multiple non-local sources. We plan
to further investigate this group during the 2022 field season.
In 2019, sampling expanded to include a wider range of types and chronological horizons, encompassing both pre-
colonial and colonial horizons, resulting in the collection of 106 samples (Fig. 13-16). Four geochemical clusters were
identified, likely suggesting the addition of new clay sources in the colonial era. Samples appear more evenly divided
among the four groups, with the largest cluster (Group 3) only containing 26.4% of the dataset. Of particular note is
CLUSPLOT( cldata)
-.:::t"
-
N
C
Q)
c::
N
0 0
a.
E
0
u NI
-.:::t"
I
-4 -2 0 2 4
Component 1
These two components explain 44 .34 % of the point variability.
Figure 13-16. pXRF 2019 Data K-Means Analysis (chart by N. Rose).
SK 15214.26, which was not only noted as bearing decoration unique among Sinop regional finds but also having
decorative and fabric parallels to at least one find from Olbia (Bauer and Sherratt forthcoming). Group 4, representing
25.3% of the dataset, is the only clearly local cluster, given its close association with the local beach clay. None of the
special types appear in this cluster save a black slipped incised sherd (SK 16771). Among the assemblages analyzed in
2019, Group 1 appears to be most clearly marked as non-local. While most samples in this cluster do not overly
diverge stylistically or technologically from the rest of the dataset, it also includes a number of samples marked in the
field as possible imports or bearing non-local decorative or finishing elements, including an Ionian painted sherd (SK
16635) and Phrygian Lustrous Black Fine ware (SK 15905.23; see Bauer and Sherratt forthcoming for discussion).
Though we are limited in our ability to compare the 2016 and 2019 datasets because we were unable to maintain
internal consistency, a clear distinction between the two datasets that necessitates discussion is the number of clusters
determined in the analysis. It is proposed here that the smallest cluster in the latter (Group 2), comprising 18.1 % of the
2019 dataset, represents an expanded exploitation of the Sinop landscape in the colonial period. This cluster is
dominated by our special finds, including one clearly Hellenistic ceramic, an Ionian painted ware, and a possibly
Phrygian bichrome ware. Furthermore, the few tiles and amphorae tested in 2019 cluster into Group 2. It is proposed
here that this cluster may represent sources outside the immediate vicinity of Sinop Kalesi, perhaps in the Demirci
Valley or on Boztepe, where Hellenistic and Roman kilns producing such finds are well known (summarized in
Doonan 2004). As such, further testing is needed.
Ultimately, this research suggests that a great majority of the pre-colonial ceramics found to date at Sinop Kalesi
were produced locally in the Bronze Age, despite the typological diversity. In the pre-colonial and early colonial era,
pots were produced locally by makers who practiced distinct technological/stylistic traditions. The producers of these
vessels perhaps occupied the site for seasonal fishing and produced pots as needed for everyday use. However, as
movements around the Black Sea transitioned from a subsistence to mercantile focus, new ceramic industries
developed in the Sinop promontory and a greater number of ceramic imports begin to appear.
CONCLUSIONS
Remarkable results were obtained relating to all our goals in the 2017 and 2019 field campaigns. In sum, we
docun1ented: (1) more substantial evidence of an Early Bronze Age occupation within the boundaries of the city of
Sinop; (2) a more nuanced understanding of the diverse assemblage of Iron Age ceramics at the site; (3) a firmer
dating and interpretation of the stratigraphy related to the city wall; and (4) a more detailed study of the architectural
development of Sinop Kalesi by mapping the full set of walls on the site.
The Early Bronze Age presence at the site was confirmed by finds of three figurines with stylistic affinities to
examples from ikiztepe and central Anatolia, demonstrating beyond a doubt that there was some kind of occupation of
the site in the third millennium BCE. Finds of fish bones, particularly hamsi, within handmade closed vessels from the
early Iron Age phases further support our contention that the site served as a fishing camp in its early phases. The
pXRF analyses of handmade wares demonstrated that many ceramic types that are more typical of the north and west
coasts of the Black Sea were actually made of local clays. This result emiches our model of itinerant fisher-folk in the
Black Sea taking advantage of the seasonal migrations of important food species including bonito (palamut) and
anchovies (hamsz).
A more complex sequence of building phases has been documented for Structure 1, the glacis of the Archaic
colonial wall. It now appears that there were at least two major building phases of a defensive curtain in this location.
Ceramic finds within the fills of these phases suggest that the second phase of the wall was built in the 6th century
BCE, while the date for the underlying phase remains obscure. The form of the wall is comparable to Archaic walls at
Smyrna and other sites, with an outer glacis protecting a rubble-filled base topped by a mudbrick construction.
Stratigraphic investigations into the foundations of the Hellenistic city wall clarified our initial results relating to
the building of the city wall, which we now believe was constructed during the first half of the 3rd century BCE,
approximately a century earlier than previously believed. The re-dating of the Hellenistic wall is established by the
stratigraphic investigations in Op. 2, and this revision places the wall's construction in the politically and militarily
competitive early generations of the Hellenistic period. The fine construction and innovative features of the Hellenistic
wall fit in well with the time when Demetrios Poliorketes and his contemporaries were engaged in a grand arms race
across the eastern Mediterranean and western Asia.
The debris dumped into the defensive trench set before the Hellenistic walls was clearly dated by numismatic
evidence to the second or third decades of the 1' 1 century BCE, increasing the likelihood that this debris, including
roof tiles, charred wood, ballista balls, and animal bones, may be related to the Lucullan conquest of the city in 70
BCE.
These results of the 2017 and 2019 field seasons at Sinop Kalesi build upon and refine our understanding of the
site and its implications for understanding the pre-colonial to colonial transitions in the Black Sea region and lay the
groundwork for expanding our investigations at the hub of Black Sea trade and communications.
ACKNOWLEDGEMENTS
We would like to thank the Turkish Republic Ministry of Culture for granting us permission to excavate the site of
Sinop Kalesi in 2017 under the administration of the Sinop Museum, and the helpful oversight of Mr. Denizhan
Demiroglu (Sinop Museum). We acknowledge Sinop Mayor Balo Ergiil and the city council for granting us the use of
this plot for an initial period of ten years and former Sinop ii Ozel tdaresi General Secretary Mehmet Yuzer for
granting the use of three container structures as office and storage space. Mr. Cemalettin Kaya, Chair of the Sinop
Tourism Board and Mr. Hasan Onur, Chair of the Sinop Historical and Cultural Research foundation, provided
invaluable assistance. Current Sinop Mayor Bari§ Ayhan and Sinop parliamentary representative Bari~ Karadeniz have
provided invaluable support for our continued research at the site and planned expansion of our research program.
Major funding for the 2017 and 2019 field seasons was provided by a three-year grant from the National Endowment
for the Humanities (RZ-51768) and the British Institute of Archaeology at Ankara, in addition to California State
University Northridge, Queens College, Gonzaga University, and the Friends of Sinop Kalesi Excavations.
REFERENCES CITED
Akurgal, Ekrem and Ludwig Budde. 1956. Vorliiufiger Bericht iiber die Ausgrabungen in Sinope. Turk Tarih Kurumu
V. Ser. No. 14. Ankara: Turk Tarih Kurumu Yaymlanndan.
Atakuman, <;igdem. 2017. Figurines of the Anatolian Early Bronze Age: The Assemblage from Koc;umbeli-Ankara.
Anatolian Studies 67: 85-108.
Barat, Claire. 2010. La ville de Sinope, reflexions historiques et archeologiques. Ancient Civilizations from Scythia to
Siberia 16: 25--64.
Bauer, Alexander A. 2006. Between the Steppe and the Sown: Prehistoric Sinop and Inter-regional Interaction along
the Black Sea Coast. In Beyond the Steppe and the Sown, D.L. Peterson, L.M. Popova, and A.T. Smith, eds., 225-
246. Leiden: Brill.
Bauer, Alexander and Susan Sherratt. Forthcoming. The Handmade, Pre-Colonial and Para-Colonial Ceramics from
the Citadel of Sinop, Turkey: Some Preliminary Observation. In Sinop Kale Excavations, 2015-2017, 0. Doonan,
A. Bauer, and E. Sokmen, eds. 1stanbul: Ege Yaymlar1.
Bryer, Anthony and David Winfield. 1985. The Byzantine Monuments and Topography of the Pontos. Washington,
D.C.: Dumbarton Oaks.
Crow, James. 2014. Sinop, the Citadel Walls, Description and Commentary. In Legends ofAuthority: The 1215 Seljuk
inscriptions ofSinop Citadel, Turkey, S. Redford, ed., 21-60. istanbul: Koc; University Press.
Doonan, Owen. 2004. Sinop Landscapes: Exploring Connection in the Hinterland of a Black Sea Port. Philadelphia:
University of Pennsylvania Museum Publications.
Doonan, Owen, Alexander Bauer, and Emine Sokmen, eds. Forthcoming. Sinop Kale Excavations, 2015- 2017.
istanbul: Ege Yaymlar1.
Doonan, Owen, Huseyin Vural, Andrew Goldman, Jane Rempel, Susan Sherratt, Ulrike Krotscheck, and Paolo
Maranzana. 2017. Sinop Kale Excavations: The 2015-16 Report. In The Archaeology of Anatolia Volume II:
Recent Research (2015-2016), S.R. Steadman and G. McMahon, eds., 178-199. Newcastle Upon Tyne:
Cambridge Scholars Publishing.
English Heritage. 2007. 3D Laser Scanning for Heritage: Advice and guidance to users on laser scanning in
archaeology and architecture. Swindon: English Heritage Publishing.
Erzen, Arif. 1956. Sinop kaz1s1 1953 y1h c;ah~malar1. Tiirk Arkeoloji Dergisi 6(1): 1- 14.
Forster, Nicola, Peter Grave, Nancy Vickery, and Lisa Kealhofer. 2011. Non-destructive Analysis Using PXRF:
Methodology and Application to Archaeological Ceramics. X-Ray Spectrometry 40: 389- 398.
Frahm, Ellery. 2018. Ceramic Studies Using Portable XRF: From Experimental Tempered Ceramics to Imports and
Imitations at Tell Mozan, Syria. Journal ofArchaeological Science 90: 12-38.
Fredericksen, Rune. 2011. Greek City Walls of the Archaic Period, 900-480 BC. Oxford: Oxford University Press.
Gosselain, Olivier P. 1992. Technology and Style: Potters and Pottery Making Among the Bafia of Cameroon. Man
27: 559- 586.
- . 2000. Materializing Identities: An Africanist Perspective. Journal of Archaeological Method and Theory 7:187-
217.
1~m, Mehmet A. 1998. Sinop Region Field Survey. Anatolia Antiqua 6: 95- 139.
Krotscheck, Ulrike and Jane Rempel. Forthcoming. Pre-Hellenistic Greek and Regional Imported Ceramics from the
Citadel of Sinop, Turkey. In Sinop Kale Excavations, 2015- 2017, 0. Doonan, A. Bauer, and E. Sokmen, eds.
istanbul: Ege Y aymlar1.
Rempel, Jane and Owen Doonan. Forthcoming. The Hellenistic Fortifications of Ancient Sinope: Construction,
Chronology, and Reconstruction. In Sinop Kale Excavations, 2015- 2017, 0. Doonan, A. Bauer, and E. Sokmen,
eds. istanbul: Ege Yaymlar1.
Rose, Nicole, Alexander Bauer, Berk Siileyman, Paige Plannette, and Spencer Hurst. Forthcoming. A Preliminary
Examination of Handmade Ceramics from Sinop Kale NW through Portable X-Ray Fluorescence Spectrometry. In
Sinop Kale Excavations, 2015- 2017, 0. Doonan, A. Bauer, and E. Sokmen, eds. istanbul: Ege Yaymlar1.
Shackley, M. Steven, ed. 2011. X-Ray Fluorescence Spectrometry (XRF) in Geoarchaeology. New York: Springer.
SNG BM Black Sea: Sy/loge Nummorum Graecorum, Great Britain, Volume IX, British Museum, Part 1: The Black
Sea. London, 1993.
SNG Stancomb: Syllof!:e Nummorum Graecorum, Great Britain, Volume XI, The William Stancomb Collection of
Coins ofthe Black Sea Region. Oxford, 2000.
Thissen, Laurens. 1993. New Insights in Balkan-Anatolian Connections in the Late Chalcolithic: Old Evidence from
the Turkish Black Sea Littoral. Anatolian Studies 43: 207~237.
CHAPTER FOURTEEN
AN ANALYSIS OF THE ARCHAEOLOGICAL AND TEXTUAL DATA

FROM THE PODIUM HALL AT AYANiS:
IS THIS THE SANCTUARY OF THE URARTIAN GODDESS 'ARUBAINI?
MEHMET ISIIZLI AND KENAN lSIK
Between the 9th and 7th centuries BCE, the Urartian kingdom, which ruled an immense geographical region,
including today's eastern Anatolia, Armenia, and northwest Iran, established its capital city at Tuspa (Van Fortress),
and was a particularly strong presence along the eastern shores of Lake Van. In the last period of the kingdom, an
important Urartian centre arising from these shores was the city of Ayanis. Following the ensuing discussion of the
Podium Hall, we present the latest radiocarbon dates of the Ayanis Castle in an Appendix.
Ayanis was founded by Rusa Argisti!Ji (Rusa son of Argisti), one of the last Urartian kings, who revived the
kingdom's military power and undertook major construction projects. Archaeological excavations carried out here
over the past 30 years have made very important contributions to Urartian archaeology, especially in the square or
tower temple (in Urartian: susi) in the centre of the citadel, which was built in the name of the Urartian supreme god
1:Jaldi, along with the pillared hall surrounding it. The temple also has one of the longest Urartian inscriptions ever
discovered. The best-preserved structures of Ayanis are within the temple complex; one of these, the "Podium Hall,"
was unearthed in recent excavations. This Podium Hall, located just east of the Temple of 1:Jaldi, along with its
material culture, presents some very extraordinary discoveries in terms of Urartian cult architecture and religious
practices, because this is a structure where the outstanding elements are feminine, not masculine. The available data
indicate that the Ayanis Podium Hall could be the place of worship for the goddess 'Arubaini, consort of the chief god
1:Jaldi. In this chapter, the textual and archaeological data obtained from the Podium Hall will be evaluated, and its
relevance to the goddess 'Arubaini and her cult analysed. Thus, the goddess 'Arubaini cult, which is little known in
Urartian archaeology, will be examined in light of these new data. Preliminary reports on this special structure have
been published in previous Archaeology of Anatolia volumes; this chapter completes the series of three (see I~1kh
2017; I~1kh et al. 2019).
RECENT FIELD WORK AT AYANiS CITADEL: THE PODIUM HALL
The Urartian citadel of Ayanis is located 35 km to the north of the city of Van, built on a natural rocky hill rising
on the eastern shore of Lake Van (Fig. 14-1).
Figure 14-1. General view of Ayanis Citadel and Suphan (Eiduru?) Mountain.
Analysis of the Podium Hall at Ayanis 171
According to the building inscription found in front of the monumental gate of the citadel, it was named RusalJinili
Eidurukai (Foundations of Rusa in front of Mount Eiduru) after its founder ( CTU I. A 12-1 VI, 9), the Urartian King
Rusa Argisti1Ji 1 (673/72-652 BCE), who was responsible for a series of architectural and agricultural projects during
his reign. More than thirty years of excavations have revealed many important structures in the citadel and the outer
town, but the most important of these is undoubtedly the temple complex in the citadel (Fig. 14-2a-b). This complex
consists of a core temple and different sized spaces attached to it. The square-planned temple room in the centre of the
complex is dedicated to 1-Jaldi, who was the head of the Urartian pantheon.
•
• @
• • •.,•
.
I
• r
._, ,1..
.,.i.!I
CD --..-
• ~
{!) Temple Area
The Podium Hall •
O Podium
Onyx Pavement
• • @ The Pillared Area • Completed Wall
Figure 14-2. a) Plan of the temple complex; b) Reconstruction of the temple complex at Ayanis.
The longest known (as of now) Urartian building inscription decorates the front of this temple room and the walls
of its entrance corridor, and it has been perfectly preserved. The temple building itself is surrounded by a pillared
courtyard. There are seven rooms on the ground in the southern part of this courtyard, with eight rectangular rooms
1 In the Urartian king list, "Argisti's son Rusa" is chronologically II or III. Rusa is controversial. For these discussions see Salvini
2012: 111-134; Fuchs 2012: 135-161; Seidl 2012: 177-181. For this reason, the expression "Rusa Argisti!Ji" in Urartian is used for
"Rusa" in this chapter.
172 Chapter Fourteen: I~1kh et al.
lined up side-by-side under the floor. These rooms, which contained a large number of metal artefacts dedicated to the
god 1-Jaldi, were described as "temple rooms" by the excavator (Cilingiroglu 2001: 37-65; Batmaz 2015: 183-195). In
addition, there are places which are still waiting to be excavated on the north side of the temple complex. Thanks to
recent excavations in this area, an extraordinary structure has been added to this complex. A large hall, located in the
eastern part of the complex, has been uncovered, adjacent to the core temple (susi). This building, consisting of a hall
and a back room to the south, measures 22 x 8 m (Figs. 14-3 and 14-4).
The building has been named "Podium Hall" 2 due to an extraordinarily beautiful marble podium inside it. The new
podium is similar to the podium in the cella of the Ayanis susi in its general features, except that this new one is
Figure 14-3. General view of the Podium Hall and back room.
2 It is referred to as "Hall with Podium" in our previous publications.

Figure 14-4. Reconstruction of the Podium Hall and back room.
larger and better preserved. There are also some differences in ornamental details. This podium, which is adjacent to
the south wall of the hall, measures 2.50 x 2.60 m and is 76 cm above the ground (Fig. 14-5). The new podium is
larger than the podium in the susi, which measures 1.75 x 0.75 m. The foundation of the podium in the Podium Hall is
made of mudbrick and wood and is covered by onyx slabs. Undoubtedly the most important feature of this podium is
its decoration. On the upper surface of the podium, just as on the podium in the cella, scenes of trees of life and
mythical creatures in antithetic stances are seen (Fig. 14-5). On the side surfaces of the podium, just like the cella
walls, the decoration is created by use of the intaglio technique. In the meantime, gold-bronze, clay, and stone
decorative objects found around the podium and in the hall indicate that the building was looted after the destruction
of the castle.
Figure 14-5. The onyx podium with decoration in the hall, and its details
(with traces of a furniture fixture on the drawing).
To the south of the hall there is a back room, measuring 4.5 x 8 m, which is accessed through a portico doorway
(Fig. 14-2a). The floor of the room is mostly paved with onyx blocks, which is similar to the floor covering on the
western side of the podium in the hall. Apart from these areas, the entire floor of the podium hall is paved with
mudbrick blocks. Traces of wooden remains on the walls of the back room point to a special interior decoration (1~1kh
et al. 2019).
THE PODIUM HALL IN THE TEMPLE COMPLEX (IJALDI E.BARA)
As stated above, the most distinctive architectural element of the Podium Hall is the podium itself, with its elegant
ornamentation. The podium in the hall is exactly the same as the podium in the cella (susi), in terms of technique and
decoration, indicating that the Podium Hall should also be considered an area for religious activities, like the temple.
However, the decoration alone is not enough to define the function of this space. Along with the podium, a number of
archaeological and epigraphic discoveries support the religious character of the hall. First of all, the location of the
Podium Hall in the temple complex indicates the religious function of the building in terms of architectural integrity.
The hall is connected with the pillared courtyard in which the core temple (susi) is found (Fig. 14-2a). There are two
entrances on the west wall of the hall. The first doorway towards the north, with its marble floor and blue painted
walls, was unearthed during early excavations. In addition, bronze building discs with inscriptions and an onyx base
were found on both sides of this door. This rich context clearly shows that it is the entrance to a significant structure.
In other words, the susi temple, the pillared courtyard around it, and the Podium Hall to the east, might have been built
as a unit, as a specific and coherent concept. The equivalent of this architectural complex in the Urartian inscriptions
was most likely the expression "E.BARA." The Urartian equivalent of the Sumerogram E.BARA must be the word
"iarani" (CTUV. 392, 473). The word "iarani" appears in the Kelishin stele (CTU I. A 3-11), which is one of the
early i~scriptions of the kingdom, and is also found in the Palu rock inscription (CTUI. 5-53). As a matter of fact, just
as the E.BARA were made in the name of the god IJaldi, the "iarani" was built in the name of the god IJaldi, as in the
example of the Palu inscription. A suggestion was made earlier by Zimansky in regard to the descriptions ofE.BARA
in Bastam and <;::avu~tepe, pointing out that E.BARA referred to the whole site and not just a single building.
(Zimansky 1985: 75). However, according to Salvini, the word E.BARA "describes the temple complex (the whole
sanctuary) in Ayanis" (Salvini 2007: 46). Indeed, although the expression E.BARA in the Urartian inscriptions means
"sanctuary" in particular, it must have been used to describe the "temple area/complex" in a broader sense.
King Rusa ArgistilJi, in his inscriptions relating to the construction of Ayanis Castle, gives two general
expressions, E.BARA for the temple complex and E.GAL for the castle or fortifications, rather than giving the names
of individual building units (CTUI. A 12-9; CTUI. 12-1 VI). In this case, the susi temple dedicated to the god IJaldi
is one of the structures within a temple comple~ (E.BARA) consisting of units with different functions.
The best example of the definition of E.BARA is the Upper Castle (Yukan Kale) temple complex in <;::avu~tepe
Castle (SardurilJinili) with its isolated location. Although the Upper Castle was built within the scope of the
construction program of <;::avu~tepe Castle, it shows that this place was allocated only for the IJaldi E.BARA due to
the height of the hill on which it is located, its distance, and its architectural disconnection with the rest of the castle.
The construction of this temple complex dedicated to the god IJaldi (E.BARA) was also particularly emphasized in the
god Irmusini temple inscription (CTU I. A 9-17). As at <;::avu~tepe, E.BARA at Ayanis must be defined as the
E.BARA of the "god IJaldi" (CTU I. A 12-1 II) as well. From this point of view, the entire religious complex
consisting of the susi temple, the pillared courtyard around it, and the Podium Hall which was unearthed at Ayanis,
can be defined as E.BARA.
Another remarkable point about the phrase E.BARA is that this expression is equivalent to another Sumerogram.
Indeed, the expression E.BARA (CTU I.A 10-5 Ro) in the Urartian part of the Topzawa Stele, belonging to the
Urartian King Rusa I (735/30---714 BCE) is referred to as E.DINGIRMEs in the Assyrian version of the same
inscription (CTUI. A 10-5 Vo). This word can be translated directly as "House of gods/ goddesses." In the Urartian
inscriptions, it is stated that the E.BARA were established in the name of Haldi. However, based on the equation
E.BARA = E.DINGIRMEs, it may well be considered that other gods or goddesses were worshipped in the temple
complexes in question.
Similarly, the most important evidence for multiple worship of deities specific to Ayanis comes from the susi
temple inscription. In the inscription in question, the ritual of sacrifice to a special group of gods/goddesses, including
the god Eiduru, directly related to Ayanis, is mentioned (CTU I. A 12-1, I 10-11; II 1-2). This information brings to
mind the possibility of the existence of cult places belonging to other gods or goddesses in the temple complex, along
with Jjaldi.
At this point, we tum our attention to the goddess 'Arubaini, who is the deity most often mentioned after Jjaldi in
the Ayanis temple inscription. Archaeological data related directly or indirectly with 'Arubaini, in the form of
inscribed, pictorial, and ritual objects, were obtained from the newly-discovered Podium Hall.
The Podium Hall: The Cult Space of the Goddess 'Arubaini?
'Arubaini, Uarubaini (read Warubani), or 'Arubani, the consort of the chief god Jjaldi, is a very important goddess
in the Urartian pantheon. She is called Bagbartu or Bagmashtu in the Assyrian account of the sack ofMu~~ir (RINAP
2: no:1, 7, 65, 82). 'Arubaini is the first goddess to be listed on the Urartu pantheon list in Meher Kap1 (CTU I. A 3-1).
As is customary with goddesses, her sacrificial animal is the cow (GU4.AB). In Urartian inscriptions, the rituals of
Jjaldi and his wife (in Urartian sila) 'Arubaini, are frequently mentioned together; for inscriptions see CTU II. 260---
261. The most detailed explanation of these rituals appears in the inscription of the Ayanis susi temple. Unfortunately,
an important part of this long inscription is not fully understood due to many hapax legomena in the text.
Partial duplicates of this inscription were found at Karmir-Blur (Teisebai URU), Kef (Jjaldiei URU) and even
Bastam (Rusai DRU.TUR), which were also fortresses founded by Rusa Argistil).i. An important point here is that,
unlike the situation at the other centres in question, the Ayanis temple inscription was found intact and firmly in place
within an archaeological context. The main theme of this long inscription is mostly the rituals of Jjaldi and '.~baini,
and it indicates that these intense worship rituals were performed in the temple area of the citadel (E.BARA).
According to the susi inscription (CTUI. A 12-1) rituals were performed "susikai" = "in front of the temple room of
susi" and in connection with the structures named "sirbani" and "adunusi," which are known only from the
inscriptions ofRusa Argistil}i (CTU II: 11, 181- 182). The fact that there is no mention of the construction of separate
adun,usi and sirbani structures in the inscriptions is probably because they were considered within the scope of
E.BARA. The adunusi, whose function is not known exactly, is thought to be the place where votive weapons were
held or stored according to the passages in the inscription (Salvini 2001: 260-261). In the temple area, the places that
comply with this description are where votive weapons were frequently found and unearthed, the eight storage rooms
built underground in the south of the susi temple building (I~ikh et al. 2020).
Regarding the sirbani, a passage of the temple inscription mentions a ritual of sacrifice (SUM) performed in front
ofJjaldi (tialdikai) and 'Arubaini ('Arubainikai). The mention of the sirbani in the continuation of these lines suggests
the possibility that the ritual in question was related to this structure, or even performed there (CTU I. 12-1 II, 9-10).
Perhaps this building might be the Podium Hall. As a matter of fact, the most frequently mentioned building after susi
in the temple inscription is the sirbani. Until now, different interpretations have been offered about the function of the
sirbani. Diakonoff suggested that the word sirbani was probably derived from the root "se-" which means "shepherd"
in Urartian, and was "the structure in which animals were held before they were sacrificed to the gods" (see Diakonoff
1991: 15, fn. 26). However, according to Salvini, this building must be sought in the area of the sanctuary, and
perhaps could be the portico (see Salvini 2001: 260, fn. 11).
It should be emphasised that, at the time of these publications, the Podium Hall in the Ayanis temple complex had
not yet been discovered. In other words, in all known Urartian citadels, there were no examples of structures that were
so close to each other for religious and cult purposes. Today, the best-known temple complex in Urartu is the one at
Ayanis.
Based on these details, it can be argued that the sirbani is a structure located in the Ayanis temple complex and
associated with the "'Arubaini cult" together with Jjaldi. Indeed, the temple inscription mentions that sacrifices were
offered in front of this god and goddess. It is probable that this involved a cult statue ( or relief) of the god and goddess
(tialdi and 'Arubaini). Although no cultic statue has been found in the excavations so far, gods and goddesses are
usually depicted standing on sacred animals or mythological creatures, or sitting on a throne in depictions/scenes of
worship.
The meaning of the BARA logogram in the expression E.BARA, which is used to define the space, supports this
suggestion. The Assyrian equivalent of the BARA logogram is the word parakku. It means "dais, sanctuary, shrine,
divine throne room" (CAD P). Meanings such as "dais, pedestal, plinth, divine throne" here probably refer to the
ornamented onyx podium both in the susi and in the Hall. On the other hand, it has been suggested that the Podium
Hall could be a "throne hall" based on the Assyrian examples (I~ikh et al. 2019: 97). However, considering both the
extraordinary decorations on the podium and the expressions in the temple inscription, they more than likely carried
cult sculptures representing 1-Jaldi and 'Arubaini, rather than a royal throne. Important evidence in this regard are the
traces of a furniture leg which have been observed on the podium in the hall. The size of four round hollows and their
position on the podium indicate that they may belong to some kind of pedestal rather than a throne. Presumably they
belonged to the base of a pedestal carrying the cult statue of the god or goddess.
In 714 BCE, the 1-Jaldi temple in Mu~a~ir was ransacked, and a large number and variety of temple items were
carried off by the Assyrian king Sargon IL When reading through the list of loot from this campaign, it can be seen
that there are many large-sized cult materials such as gold decorated silver beds, bronze-silver libation cauldrons, and
bronze cultic sculptures (RINAP 2: no. 63, 65, 82). The 1-Jaldi E.BARA at Ayanis also may well have been a place
where rituals were performed and rich cult items were stored. The rich findings of the Podium Hall, which are likely
to be the remnants of these items, partially support this suggestion. In addition, in situ findings from other places in
the Ayanis temple complex made in previous years are also supportive in this regard. The back room of the Podium
Hall may be the room where the cult items of the temple were placed or stored. The isolated location and interior
spatial arrangement of the room also support this suggestion, along with the fact that similar types of votive items
were stored in specific places in the temple.
The Goddess of Beads
At Urartian sites, excavated beads normally come from graves. It can be said that the use of beads in the traditions
dating to pre-Urartian periods has a religious purpose beyond aesthetics, and some data suggest that the bead is an
item associated with the cult of the goddess 'Arubaini. One of the strongest pieces of evidence for this is an agate bead
which is preserved in the Van museum collections. The inscription on the bead states: "Argisti brought a gift from
ErialJi land to the goddess Uarubaini" (CTU IV. D 8-1). This king is most probably Argisti I (785-756 BCE), and
ErialJi Land is located on the modem Shirak Plain (Din9ol and Kavakh 1982). As a matter of fact, we know that the
Urartian king who conquered ErialJi Land was Argisti I. Furthermore, in a short inscription on a carnelian bead which
Figure 14-6. a) The scene of offering beads to the goddess 'Arubaini (?) sitting on the throne (after Seidl 2004:
Taf. 65); b-c) Noble woman/goddess ('Arubaini?) sitting on the throne holding a string of beads in her hand
(after Seidl 2004: Abb. 102/sm-12,13).
is held in a private collection, the expression "To Uaruba(ini), Sarduri" is mentioned (Walker 1978). It is understood
from this cryptic inscription that this bead is dedicated to the goddess 'Arubaini. Bead-offering scenes are depicted on
some glyptic Urartian art works, and in these scenes, gifts are presented, especially by women, to a goddess or noble
woman sitting on a throne. Among these gifts, strings of beads attract particular attention (Fig. l 4-6a-c ). The seated
goddess/noble woman often is depicted wearing string beads as jewellery, or carrying a string of beads in one hand
(Fig. 14-6b-c) (Kellner 1991; Seidl 2004; <;avu~oglu 2014).
In addition, Piotrovsky suggests that the bronze woman figurine with exaggerated bead strings now in the Erebuni
Museum, Yerevan, and the female figures sitting on the throne on the Karmir-Blur and Toprakkale medallions
represent the goddess 'Arubaini (Piotrovsky 1966). These examples show that beads are one of the most important cult
objects which were offered to the goddess 'Arubaini. In previous excavations at the citadel of Ayanis, a total of 29
beads were found in the susi temple area. Most of these beads are of cylindrical or spherical form and are made of
semi-precious stones such as carnelian (Saglamtimur et al. 2001). During the most recent excavations, 84 additional
beads, resembling the 29 previously discovered, both in material and form, were recovered from the Podium Hall,
particularly in the back room (I~Ikh et al. 2019) (Fig. 14-7).
Figure 14-7. Beads which were recovered from the back room of the Podium Hall.
The fact that no human remains have been found in the area where the beads were unearthed clearly shows that
they cannot have been worn as jewellery. Beads in this quantity in such a confined space must be related to Urartian
religious practices here. In this context, the special position of the back room where the beads were found, together
with the Podium Hall, suggests that they were votive items. Just as the many inscribed bronze votive items found
around the susi temple were dedicated to the god 1-Jaldi, the aforementioned beads were probably offered to the
goddess 'Arubaini. Not all of these beads have to be inscribed. As a matter of fact, strings including many beads are
seen in the depictions. The suggestion of the offering of beads to 'Arubaini is strengthened by the epigraphic evidence
for the presence of 'Arubaini, to which we now tum.
SEAL-IMPRESSED AND CUNEIFORM-INSCRIBED

BULLAE FROM THE PODIUM HALL
In addition to the beads, another important link between the 'Arubaini cult and the Podium Hall can be established
through the inscribed and stamped clay objects found in the space. Undoubtedly, the most interesting group among the
finds recovered from the Podium Hall consists of cuneiform- and seal-imprinted clay objects. This valuable and
meaningful group, found during the 2016 excavation campaign, consists of two bullae with stamped-cuneiform
inscriptions, and three bullae with stamps only; in addition, three cretulae 3 fragments were found in 201 7. This group
provides invaluable information on the function of the hall with the podium and its relationship with the 'Arubaini
cult.
The Inscribed Bullae
1. Inscribed Bulla (Inv.2016 456-A)
This drop-shaped bulla has a height of 1.4 cm and a width of 3 .1 cm. There is no seal impression on the upper and
lower surfaces of this bulla. The upper surface of the bulla, which is the colour of its cream-coloured paste, is broken.
The two-line inscription is on the lateral surface of the bulla (Fig. 14-8a):
1 D['a]-ru-ba<i-ni>
2 ~-sa-lJi-na-di KURe.
Commentary: In the first line, the name of the goddess 'Aruba<ini> is mentioned. The Urartian name of Ayanis
Fortress is shown on line 2 of the bulla as ~sahina KURe. = ~sahina KUReidurukai "RusalJinili (city/fortress) before
Mount Eiduru." Here, the "KURe." signs are used as an abbreviation for the expression KUREidurukai (in front of Mount
3
Cretulae are clay objects larger than bullae and with seals or inscriptions. Cretulae were applied directly to containers, probably
vessels (Salvini 2001: 279).
Eiduru). Similar abbreviations are common, especially on clay inscriptions of the Rusa Argisti!Ji period (Salvini and
Wegner 2014). Rusa=IJi=na=di: here Rusa (Norn.) with case-suffixes Lok+Pl+Dir. KURe. as an abbreviation of
Eiduru=kai is parsed as Eiduru (Norn.) with postposition suffix -kai "before." Thus, this short inscription can be
translated as "(For) goddess 'Arubaini, in/to Rusa!Jinili in front of Mt. E[iduru]."
This inscribed bulla shows that an item or property was sent to the goddess 'Arubaini' of Rusa!Jinili (with -di
directive suffix), (the fortress of Ayanis). Thus, it is understood that at Ayanis, there was a cult to the goddess
'Arubaini, as well as to the god 1-Jaldi, who was the supreme figure ofUrartian religion.
~1~ J>~ ~ ~ ~ 4w~

i>- <bY<bYW m
b
Figure 14-8. a) The inscriptions on the lateral surface of the bulla (Inv.2016 456-A);
b) The inscription on the lateral surface of the bulla (Inv.2016 458-A).
2. Inscribed Bulla (Inv.2016 458-A)
This drop-shaped bulla has a height of 1.5 cm and a width of 2.4 cm. The three-line inscription is on the lateral
surface of the bulla (Fig. 14-8b):
1 MUNus,a-ri-i-ni MUNUS.LAGAR.KUR
2 MUNUSqa-qa-a-ni MUNUSKALAG. <GA>
3 MUNUSil-zi-"tu"-[u?]-ni TA
Commentary: In the first line 'arini is a feminine personal name, as shown by the determinative MUNUS
"woman." The title of this woman is "MUNUS.LAGAR.KUR," which consists of Sumerograms. MUNUS.LAGAR=
murub (ABZ 458); MI.LAGAR=murub2 (amiltu) "woman, free woman" (ME 551). But when MUNUS.LAGAR.KUR,
which is here mentioned for the first time in an Urartian inscription, is examined, MUNUS is used as a logogram and
determinative meaning "woman." The last ideogram, KUR, is widely used in the Urartian inscriptions as a
determinative in the meaning of"land" and "mountain" (ABZ 366), while at the same time it means "palace" (Grekyan
2016). The ideogram LAGAR or lagar/lukur = MUNUS.ME (nadi:tu) "a kind of priestess" (ABZ 554). Thus, when the
combination of MUNUS.LAGAR.KUR on the bulla is evaluated together, it can be translated as "Miss 'Arini,
Priestess of the Palace." However, this translation, which is compatible with the rare use of the expression "KUR" in
Urartu as "palace," is one very possible meaning. As a result, we are dealing with a kind of priestess's name.
In line 2, the name "Qaqani" is mentioned as a second woman, again with the MUNUS determinative. The signs
following Qaqani can be transcribed as MUNusKALAG. <GA>. We know this title from the Old Assyrian/Babylonian
written documents (CAD D). MUNUsKALAG. <GA>= Akkadian dannatu (ABZ 554); dannatu means "fortress" as well
as "famine, distress" (CAD D; Schramm 2010). Accordingly, the second line can be translated as "Miss Qaqani,
Woman of the Fortress." Considering the definition of "MI/MUNUS.E.GAL" (Woman of the Fortress) in the queen
Sumerograms, especially in the Assyrian inscriptions, there may even be a possibility that the "Qaqani" in this bulla is
a queen or Urartian royal family member. However, it is clear that we need a little more data.
The third and last line mentions another woman named Ilzituni, identified with MUNUS. Interestingly, this name
is followed only by the TA sign. This sign most likely has the logogram function here. This logogram usually
corresponds to Akkadian "istu/issu/ultu" prepositions, meaning "from, out of, since, after, by" (ABZ 139; ME 248;
CAD I-J). According to the flow of the text, the TA logogram may be used here to means"- from or - by". Accordingly,
this line can be translated as: "(Sent) by Miss Ilzituni (for) Miss 'Arini Priestess of Palace (and) Miss Qaqani Woman
of Fortress." This translation is the most likely, and any other explanation for the presence of the TA sign here would
present difficulties. Before discovery of this bulla, only a few women's names were known in Urartian studies. Two
were queens: Tariria and Qaquli. This bulla adds three more women's names associated with the goddess 'Arubaini
cult. Lastly, the "palace" (KUR) and "fortress" (KALAG. <GA>) mentioned in the female titles in the text would be
Ayanis Castle.
The Seal Impressions
There are four seal-imprinted bullae which have been recovered from the hall with the podium. The first of these
bullae (Jnv.2016 458-A), which includes several women's names, is described above. There is in addition a stamp-seal
impression consisting of some symbols on the upper and lower sides of this bulla (Fig. 14-9a-c). Exactly the same
stamp seal made impressions on the other three bullae found.
j~~
"~~~?,:~;~-~
Figure 14-9. The bulla with female names and its impression (Inv. 2016 458-A).
Left: Upper surface; Right: Lower surface.
3. Stamped Bulla (Inv.2016 457-A)
There is no inscription on the side of this drop-shaped bulla. On the lower and upper surfaces there is the stamp
seal impression mentioned above (Fig. 14-l0a)
4. Stamped Bulla (Inv.2016 455-A)
Similarly, there is no inscription on the lateral part of the drop-shaped blade. On the lower and upper surfaces of
these, there is the stamp seal impression mentioned above (Fig. 14-l0b).
5. Stamped Bulla (Inv.AYN.16. VI.C40)
This is a fragment of a bulla with only one face preserved. There is the aforementioned seal impression on this face
(Fig. 14-l0c).
C
Figure. 14-10. a) A bulla without inscription and its impressions/drawing (Inv.2016 457-A);
b) A bulla without inscription and its impressions/drawing (Inv.2016 455-A);
c) Fragment ofa bulla with its impression/drawing (Inv. AYN.16.VI.C40).
We are presented with a uniform seal impression consisting of the same figures, which can also be seen on the
bullae above. The scene in all of these impressions, some of which are broken or faint, consists of several
hieroglyphs/linear symbols. In the upper left is a linear arc-shape, possibly a stylized ox-yoke form. This form/symbol
is similar to the stylized horns, boat, and perhaps omega forms in Mesopotamian iconography. Parallels to this symbol
are also seen on a bronze bowl from Toprakkale (Lehmann-Haupt 1906: fig. 71) and a few vessel fragments
(Lehmann-Haupt 1910-1931: 579).
A possible bull head in profile at the front of the ox-yoke form is seen on the seal impression. This symbol, with its
distinctive crest, is also reminiscent of a horse's head; however, in parallel examples, this symbol is seen more as a
bull's head. The last figure in the seal-imprinted scene is a fish with a tail and fins in the lower part of the scene. A
goat's head, with its horns bent backwards, should have been attached to the front of the fish's head, and, indeed, this
goat head figure is most clearly seen on bullae nos. 3 and 4. This highly-stylized goat-fish or carp-goat (in Akkadian
$Uburmasu) figure often appears in connection with the Ea cult in Mesopotamian iconography. It should be noted that
in Karmir-Blur, Kef, and Ayanis, which are well-known centres in the Rusa ArgistilJi period, we also find figures
associated with the Ea cult (Batmaz 2019). These have been called fish-men (in Akkadian kulullu). In Mesopotamia,
the "fish-goat" figure has been depicted with a goat-headed sceptre, especially on kudurru and seals, since the time of
the Third Dynasty of Ur (Oman 2005). At this point, we should mention the striking depiction of a sceptre with the
head of a goat or bull on a stamp seal impression from Toprakkale, also associated with the Rusa ArgistilJi period (Fig.
14-lla) (Lehmann-Haupt 1910-1931).
b
Figure 14-11. a) Drawing of impression of the stamp seal from Toprakkale (redrawn after Lehmann-Haupt, 1910-31:
833); b) Drawing of impression of a cylinder seal from Toprakkale (redrawn after Wartke 1993); c) Drawing of an
impression of a stamp seal from Toprakkale (redrawn after Ayvazian 2006: TK 25).
In fact, the closest parallels to the bull head and ox-yoke symbols of the bullae imprints from the Podium Hall are
seen on a cylinder-stamp seal impression which was found in Toprakkale (Wartke 1993: Abb. 89; Ayvazian 2006: TK
25). In the worship scene on the surface of this cylinder seal, a person with the symbols behind him and holding a
sceptre, stands at the front of a castle or building (temple?) with two bastions on its comers (Fig. 14-11 b ). The stylized
sceptre in this person's hand may have a goat/bull head. In the stamp part of the seal, only the bull head and ox-yoke
symbols are seen (Fig. 14-11 c).
Another cylinder-stamp seal with several of these symbols is preserved in the Pierpont Morgan Museum (Seidl
1979: 144, Abb.l). We do not know exactly what the symbols repeated on the seals in question mean. This is a
problem related to the fact that the Urartian hieroglyph/linear script, which has a limited number of examples, is
largely undeciphered.
Cretulae Fragments (Inv. AYN.17. VI.AOS; AYN.17. VI.A32; AYN.16. VI.A217)
Three cylindrical and stamped clay fragments/cretulae were found during the excavations in the Podium Hall at
Ayanis Citadel, and when the seal impressions on these items are combined, a scene is presented, consisting of two
mixed creatures and two figures with their backs turned, holding strings of beads in their hands, and pictured around a
tree oflife (Fig. 14-12a---c).
These figures, albeit only partially preserved, are typologically similar to women or eunuchs. A few cuneiform
signs on two of these pieces indicate that the cylinder seal was inscribed in the lower and upper bands, but not enough
survives for any kind of reading. One of these cretulae (Fig. 14-12c) has been stamped with what is probably the
round end of a cylinder seal. In this extraordinary impression, encountered for the first time in Urartu, a woman in a
long dress and wearing a pompom hat, her long hair pulled back, plays a tambourine with both hands (Fig. 14-13).
The scene gives her the appearance of being in motion.
Right behind this figure are the bull head and ox-yoke shaped signs seen on the bullae and stamp seal impressions
of the Podium Hall. These two symbols show that the seal impression in question may be related to the bullae and seal
impressions of the Podium Hall.
The tree of life and fertilization scenes are widely depicted in Urartian art, although the end-stamp impressions of
these cylinders are most often centaurs. It is interesting that here the stamp impression is a scene of a woman playing a
tambourine accompanied by the bull head and ox-yoke symbols. In fact, depictions of women playing tambourines are
most often observed in scenes on Urartian bronze belts (Seidl 2009). The figure is a novelty in Urartian glyptic art.
This aside, the main point is the repetition of the symbols on the impression with those on the bullae from the Podium
Hall. It may well be that all seals and bullae found at the Podium Hall were symbolically related to each other and
used for the same purpose.
,,;;-;--;-;f...{j}~':2,::,_;;z_;(::,
, ......
_ ..,JI
-: _.,
"..t -"'-
~{!?€
. - ~~.,-:,
:~
31: ,
~:.:,.
,
:,,--::
,
-, , ,
-
Figure 14-12. Photos and drawings of cretulae fragments; a) Inv. AYN.17.VI.A05;

b) Inv. A YN.l 7.VI.A32; c) Inv. AYN.16.VI.A217.
Figure 14-13. Stamp seal impression and drawing of cretula.
CONCLUSION
The recently discovered Podium Hall, with its architectural features and extraordinary finds, makes a significant
contribution to Urartian archaeology. First of all, this new building group is notew?rthy for its location within the
1-Jaldi temple complex. It must have been encompassed within the concept of 1-Jaldi E.BARA, which is mentioned in
the inscriptions and characterizes the entire temple complex. The susi temple inscription with its description of the
rituals to be performed here correlates well with the architecture and small finds in the temple complex. It seems likely
that the Podium Hall was the sirbani building, frequently mentioned in the temple inscription, where 1-Jaldi and
'Arubaini were jointly represented. As a matter of fact, the Podium Hall in the temple complex was the second largest
area, after the "at the front of susi (susikai)," where mass religious ceremonies could be held. Although this place
cannot be defined as a temple in its fullest sense, it should be seen as the place where rituals dedicated to the goddess
'Arubaini were held, together with those of 1-Jaldi. In this sacred place, the goddess 'Arubaini is more prominent. The
bulla with the expression "to the goddess 'Arubaini in Ayanis Castle," which has been found in this space, is
important proof in this respect. Another important piece of inscribed evidence is the bulla containing many female
names, unique for Urartu. It is the first time that so many female names and titles have been found on an inscription in
Urartian written documents. Also, for the first time, we see the name of an Urartian priestess CArini) in this
document. In the same document, we see the name of another woman (Qaqani) who was likely a queen or a member
of the Urartian royal family. In an inscription on a bulla which was found in previous excavations in Ayanis Castle,
the name MUNUsnama, which is thought to be another woman's name, is mentioned (CTU IV. CB Ay-53). These
inscribed pieces of evidence have provided clearer data about the function of the area in the context of the spatial-
material culture relationship.
Accordingly, it can be stated that there is an 'Arubaini cult at Ayanis in terms of both spatial and religious
organisation. This cult was probably not independent of the 1-Jaldi cult. On the contrary, the deities appear to have
been worshipped jointly. The susi temple inscription supports this, and at this point we cannot speak of an independent
"'Arubaini worship" or"' Arubaini temple."
Certainly, there might have been differences in ritualistic details such as cult offerings and sacrifice, and we know
from other inscriptions that, unlike 1-Jaldi, a cow was sacrificed to "'Arubaini." We also know that offerings of strings
of beads were identified as being devoted to 'Arubaini. Dozens of beads which were found in the hall probably reflect
this. In fact, the existence of beads dedicated to gods and goddesses in ancient Mesopotamia was known long before
Urartu. They were mostly beads of agate or carnelian stones. One of the earliest examples of inscribed beads found in
Susa belongs to Sulgi (2094-2047 BCE), king of the Third Dynasty of Ur. These carnelian beads are dedicated to the
goddesses Ninlil (RIME 3/2) and Ningal (Potts 1999). An agate bead devoted to Ninlil belongs to the Kassite period
(ca. 1595-1155 BCE) (MMA 125 1995), and another agate bead with an inscription from the Kassite period is
dedicated to the goddesses Namma and !star (Potts 1999). These examples suggest that votive beads in Urartu may be
a continuation of the Mesopotamian tradition. According to the available data, we do not know exactly whether beads
were devoted to more than one goddess in Urartu, as in Mesopotamia. The limited examples available only belong to
the goddess 'Arubaini. As a result, we cannot determine for now whether this tradition is a practice belonging only to
the 'Arubaini cult.
At the moment, we have little detailed information about the organisation and practices of this 'Arubaini cult.
However, it is quite possible that certain duties in this cult were performed by women in accordance with the female
character of the goddess. We certainly can talk about a cult in which women were more prominent. At this point, we
again mention the name of the priestess ('Arini) on a bulla in this place. Another important factor is the inscription on
a linear and cuneiform etching on a bulla which was found at Ayanis in previous years that reads "LU D'aruba," that
is, "Man of 'Arubaini" (CTUIV. CB Ay-39, 191-192).
184 Chapter Fourteen: I~lkh et al.
In fact, the determinative "LU" in this statement refers to the occupation of attendant. In this statement, it is not
clear whether the official in question was a woman or a man. As a result, we interpret this as "a person related to the
'Arubaini cult." Another important point regarding this issue: it is possible that those involved in the 'Arubaini cult
organisation, or its worshippers, were using seals with a common composition. Glyptic evidence obtained from the
Podium Hall points in this direction. The seal impressions on five bullae and one cretula separately must have
belonged to persons involved with this cult within, or associated with, the cultic organisation. These symbols on the
prints of multiple seals were probably signs or attributes symbolizing a god/goddess or their cults. At this point, we
should recall the worship scene on the Toprakkale seal depicting the front of a castle/temple, and a stage with the bull
head and ox-yoke symbols.
The bullae with these seal impressions and their location seem to be connected to a cult of the goddess 'Arubaini.
In addition, it is understood that seal impressions with the same scene came from at least two separate stamp seals.
When the stamp seal impression on the cretula is added to them, we realize that we are seeing a new standard stamp
seal impression in Urartian glyptic art, alongside the king ofRusa and LDczyuli seals.
When all the data are taken into consideration, the Podium Hall with its feminine associations was quite likely a
cult place for 'Arubaini. This extraordinary building, however, could not have been built just for this cult alone, but
must have served other religious and royal activities as well. Naturally 'Arubaini must have shared this cult place with
her consort, the chief god Ijaldi, for whom the entire temple complex was built. However, the main worship area of
Ijaldi is presumably the susi temple and its pillared courtyard. In fact, this was an area where there were many votive
weapons devoted to Ijaldi, emphasizing power, war, and therefore the Urartian male principle. In the upcoming
seasons, this issue will become clearer with new archaeological discoveries from both Ayanis and from Urartu
generally.
ACKNOWLEDGEMENTS
We are grateful to Prof. Paul Zimansky for reading the manuscript and sharing his opinions with us about the text.
We also give thanks to Ms. Jan Bailey (B.A Hons) for checking the English text.
APPENDIX
PRELIMINARY RESULTS OF RADIOCARBON DATING ON AYANIS CASTLE
MEHMET I$IKLI, ALPER ~ENGUL, AND Oouz ARAS
The many castles in the Kingdom of Urartu are spread over a wide area of approximately 222,000 km2, centering
on the Van Basin, including Lake Sevan and Lake Urmia (Salvini 1995; <;ilingiroglu 1997). Rusa, son of Argisti
(685-645 BCE), who was the most prominent of the Urartian kings, initiated many reforms during this period,
building at least five large castles/cities in the Lake Van basin. One of these, Ayanis Castle, is located 35 km north of
Van, on the east coast of Lake Van, north of today's Ayanis village, 300 m inland from the lake. The magnificent
castle was built on rocky ground at an altitude of 250 m above the lake level and 1866 m above sea level (Fig. 14-14).
The castle is a one-day walk from Tuspa, which is the Urartian state's capital city.
Ayanis City was destroyed by a great earthquake; almost sealed by the collapse of monumental mudbrick
structures, it became uninhabitable. Therefore, the city remained intact to the present day, complete with the original
artefacts. Ayanis Castle presents us with many problems in spite of its rich cultural inventory: When was the
castle/city established?; When and how was it destroyed?; Is there a relationship between the end of the fortress and
that of the Urartian kingdom?; Are the end of the castle and the death of Argisti' s son Rusa simultaneous?; Should
discussion of a "Post-Urartu period" ensue? Until now, there were two basic criteria used to date Ayanis: inscriptions
and dendrochronological analysis, as no radiocarbon analysis had been carried out for Ayanis. In recent fieldwork,
radiocarbon samples were taken in order to resolve the questions highlighted above. In this study, the first results of
these analyses will be presented.
Tharlks to the building inscription on the monumental citadel gate of Ayanis Castle, it is known that its founder
was Rusa, son of Argisti. Based on textual data and dendrochronological dating, the foundation of the castle is
determined to be 673/672 BCE (<;ilingiroglu 2001; Manning et al. 2001), and it is considered the last castle built by
Rusa in the kingdom. After this time, the kingdom grew weak.
The decades of excavations at Ayanis, carried out in two areas, namely the citadel and the Outer City, have
unearthed: the Temple Area, Podium Hall, Domestic Spaces, Hall with Pier, Monumental Door, East and West
Storage Rooms, South-North-East and West Walls, and a portion of the outer city (<;ilingiroglu and Salvini 2001;
<;ilingiroglu and I~1kh 2014). The Ayanis Castle excavations have raised several questions regarding the foundation of
Analysis of the Podium Hall atAyanis 185
the castle, the end of the castle and the kingdom, and the period of its founder. According to some researchers,
Argisti's son Rusa, whose reforms and castle-building brought the Kingdom ofUrartu to its highest point, may have
also been the person who dragged the kingdom into collapse with these activities. The purported 40-year reign of
Argisti's son Rusa is calculated based on Assyrian written sources, but the certainty of these years is a matter of
debate, namely whether the founder of Ayanis was Rusa II or Rusa III.
(Limestone)
Figure 14-14. General and plan view of Ayanis Castle and bedrock, view from the south.
Another question requiring clarification is the connection bet\veen the end of Ayanis and the end of the kingdom.
Written and archaeological data indicate that the Kingdom ofUrartu was largely ineffective after Rusa. Unknown is
whether the Ayanis Castle was still standing, or destroyed and abandoned, after Rusa's reign. We do know that life
continued in the outer city of Ayanis for a while, but the situation at the citadel is not clear. Urartian written
documents had largely ceased by this point, perhaps indicating the weakening or demise of central authority. After
Rusa, the royal line of Urartu cannot be followed clearly, though some sources suggest that a few members of the
royal family or princes may have held on to some weak vestiges of power.
The possible connection between the end of Ayanis Castle and the end of the kingdom highlights the importance of
understanding why and how Ayanis Castle was destroyed. Excavations have revealed large ruins and fire traces in
ahnost all areas in Ayanis Castle, undoubtedly caused by a big earthquake. This issue will be analyzed in more detail
below. Identifying the date of this earthquake will help solve the uncertainties regarding the end of the kingdom as
well as the castle.
As noted above, dates for the Urartian Kingdom have primarily depended on Assyrian texts. Urartian inscriptions
on the castles name the king and his father, which are then cross-referenced with Assyrian textual data that mention
these kings or their castles. This leads to controversy regarding the chronological order of Urartian kings and their
precise dates of reign. Until recently, no C14 dates were available from this region. The lack of reliable radiocarbon
methods in.previous decades and the availability of textual sources are the primary reasons for the lack of C14 data.
When excavations at Ayanis Castle commenced in 1989 the focus was on dendrochronological methods, given the
availability of wood samples. However, a new era of radiocarbon dating of samples has now commenced.
RADIOCARBON DATING SAMPLES
Temple Area: AYN-OJ
The temple areas established in the central point of Urartian cities mostly consist of a cella and a courtyard with
piers surrounding it. The Ayanis Temple Courtyard, which hosted religious ceremonies, consists of 12 piers lining
mudbrick walls which are 5 m high and 4 min thickness. Two of these piers are adjacent to the cella; the other 10 are
arranged symmetrically on three sides of the temple (<;ilingiroglu and Salvini 2001). The northern piers had been built
on bedrock and were partially recovered, but the southern piers, lacking a bedrock foundation, were completely
destroyed, likely by the earthquake, making samples from this location a necessity. Excavations had revealed a
doorway (originally identified as a niche) in the northern area of the courtyard. A charcoal sample (AYN-01; Fig. 14-
,
15) was recovered from this doorway, situated in the northern wall, which leads from the Temple Courtyard to the
open area beyond.
AYN-02 .
N AYN-03 ~.J
. ,
.• AYN-05
..,
AYN-04
• ' sample Location
Figure 14-15. Location ofradiocarbon samples on drawing of Ayanis Castle.
Cella: AYN-05
Urartian cella temple structures, with their unique styles, are located on the hills of the sites; they are generally
square-shaped with rhizalite comers and vary in size, the latter likely dependent on the size of the bedrock available.
The Ayanis Castle cella., built of mudbrick walls on a stone foundation, measures 13 x 13 m on its exterior and 4.6 x
4.6 m on the interior. The northern walls, situated on an east- west axis, were set directly on the bedrock; the southern
wall, however, rested on about 5 m of foundation walls (<;ilingiroglu and Salvini 2001). We know that the structure
was built by Rusa., son of Argisti, based on the 16 m long inscription engraved on the front walls of the cella. Samples
were taken to verify Rusa., son of Argisti's years of sovereignty (685-645 BCE), and to identify when the cella was
built. The charcoal sample (AYN-05; Fig. 14-15) was taken from the northern wooden beams found between the
basalt blocks and the mudbrick walls.
Podium Hall: AYN-06
The Podium Hall, discussed in the main text, has a generally rectangular plan and consists of a large main room
and a smaller back room. The main room, which forms the center of the hall, is 22 x 8 m; a doorway gives access to a
small back room measuring 4.5 x 8 m (l§ikh et al. 2019; I§ikh and Aras 2020). Of importance is to determine the
chronological relationship between the Podium Hall and the cella., which differ somewhat architecturally. A charcoal
wood sample (AYN-06; Fig. 14-15) was taken from the water drainage channel in the Podium Hall.
Southern Fortification Areas: AYN-04
The city walls surrounding the Ayanis Castle constitute one of the most beautiful examples of Urartian military
architecture. The southern fortifications were built of ashlar basalt stones carved into the bedrock. The city walls, built
using the rustication technique, were shaped according to the slope of the bedrock; a mudbrick superstructure sat upon
stone foundations. These tall city walls had to be earthquake-resistant; however, it is clear that even these walls could
not prevail against the final earthquake. The top half of the mudbrick superstructure toppled in toto to the base of the
fortification wall. The earthquake also claimed human victims in this area; along with one medallion, three seals and
one fibula were found, indicating that these individuals may have belonged to the noble class. The burnt nature of
some of these bones indicate that these people were exposed to intense fire; this evidence, along with other data, offers
proof that a fire raged throughout the area following the earthquake (Aras and Be~ikr;;i forthcoming). Dating of the
human remains (AYN-04; Fig. 14-15) will supply both the approximate date of the collapse of the castle and the date
of the earthquake that destroyed it.
Northern Fortification Areas: AYN-021AYN-03
Limestone blocks belonging to the northern wall were found during the excavations carried out on a steep slope
and in a narrow area in the northern part of the citadel. Three rows of these kyklobic limestone blocks, which are very
similar to the eastern wall in terms of construction technique and material, have been preserved; their heights range
from 50 to 75 cm. The height of the protected part of the fortification foundations resting on the bedrock is 2.26 m. In
front of the wall was a 2-m deep cultural deposit containing animal bones, cuneiform bullae, stone work, bone
artefacts, bronze artefacts, and many ceramic pieces, all covered with a dense layer of ash. A crack in the east and
west sections of the North Wall provided two soil samples (AYN-02 and A YN-03; Fig. 14-1 5). These do not represent
a fault plane, but are thought to have a direct connection with the earthquake.
ACTIVE TECTONICS AND EARTHQUAKE POTENTIAL

OF AY ANIS CASTLE ENVIRONS
Although the castle rests on resistant bedrock, it is located in a very active zone in terms of seismicity (Fig. 14-16).
Most of the earthquake records belonging to both the instrumental period (earthquakes from 1900 to the present) and
historical (generally from 1100-1900 CE, determined from historical catalogues) are known. There is little data on
earthquakes prior to these periods, although old earthquakes have been detected in palaeoseismology studies on active
faults in the region. The two most active faults in the region are the Alakoy and the Van (Fig. 14-17), the latter just 14
km south of Ayanis. On October 23, 2011, the Van Fault ruptured; this event, known as the Van Earthquake M= 7. l ,
became the largest earthquake known in the instrumental period and further damaged the Ayanis Castle (Poyraz et al.
2011).
g
0
N-r,:;;;;=a::================.---
~ Legend
- - - ---=----=:---~r
Magnitude
• 4,0 - 45
Geology
t •
e 4.5 - 5.0
- Q~temary UndifferMIJaled dt:1stics
~ Quaternary Lavas
N •
g e
0
5.0 - 55 ~ Plo-Quatemary *ustrine-ftuvletl le deposus
v -
0
...
M •
•
5.5 • 15.0
6.0 • 6 5
• Paleoeenll-M1oeene Clas!fcs and ~rbonates
Upper Cret"°"""' Ol)hiollic melange
0 *• 6.5 • 7.1 - Paloozole--Mesozooe
Mw - 7. 1 Van Eanhquake - -
g l":,, --::.:---------""""'.!--~---~.,.......,r/ft·
~-
a... "'"'"
...
N
••
0
0
0
co -
co
...
N
0
•
0
0
o -
co
...
N
0
0
0
•• •
N-
...
"-
N Lake Van
I I I I I I
336000 345000 354000 363000 372000 381000
Figure 14-16. Simplified geological map of eastern Lake Van, epicenter distribution of earthquakes after the Van
Earthquake, and major active faults around Ayanis Castle (MTA 2002, Emre et al 2013; $engiil et al. 2019).
Figure 14-17. Main and aftershock distribution of Van earthquake

and historical earthquakes around Lake Van (Sengiil et al 2019).
Effects of the Van Earthquake 2011 (M=7.1) on Ayanis Castle
Earthquake effects on Ayanis Castle can be considered for both the instrumental and historical periods. In the
recent earthquake on 23 October, 2011, and in previous historical-period tremors, the temple, curtain mudbrick walls,
and interior walls, which are the structural elements of the castle, were damaged, and the columns carrying the upper
floor were damaged.
Given that excavations take place every summer in Ayanis Castle, the parts of the Ayanis Castle affected by the
Van Earthquake in 2011 could be identified immediately. During the earthquake, serious collapse was observed
especially in the east-west trending walls of the temple area (Fig. 14-18, left). In the opposite direction, the walls
standing north-south had little damage, and only small cracks developed.
Figure 14-18. Left: Local collapsed brick walls in the Temple area after the Van Earthquake in 2011;
Right: Extension and cracks on the basaltic stone columns of the Temple area.
Another group of buildings where fractures and separations are most common are the structures, approximately
150 x 150 cm, which are described as columnar pillars, which carried the weight of the upper floor of the castle,
which is not preserved (Fig. 14-18, right). The outer surfaces of the lower parts of these structures, which remained
intact at the bottom, were covered with basalt blocks, and the inner and upper parts were filled with adobe material.
The mudbrick section was heavily affected by atmospheric conditions over time and began to melt. However, the
basalt blocks have managed to contain all the effects of deformation. The type of damage observed in these pillars has
generally developed as block explosion, separation, and cracking of comer blocks. There are also pillars that have
completely collapsed. However, this destruction did not happen during the 2011 Van Earthquake.
Large-scale damage and fractures, not recorded in instrumental and known historical earthquakes, were observed
in excavations on both curtain walls, the columns of the citadel area, and the northern slope of the castle. Some of this
damage was seen on the walls built with mudbricks up to 4-5 m high and forming the upper parts of the curtain walls.
These walls were broken and toppled due to gravity along some planes (Fig. 14-19, left). These walls extend and are
parallel to the mapped active faults.
Figure 14-19. Left: Collapsed mudbrick wall in the Southern Curtain Wall, displacement plane (dashed white line)
and movement (arrow), view from west; Right: An extension crack filled by topsoil in the Northern pit.
An extension joint with a dip/dip direction of 325/80 was observed in a pit opened on the northern slope of the
Castle area (Fig. 14-19, right). This extension does not provide direct fault plane data. It has a geometrically "V"-
shaped structure and closes progressively.
This extension was observed on the west wall of the pit. In the east wall of the same pit, some similar but not
identical cracks have been observed. During the 2011 Van Earthquake, it was observed that some of the fractures on
the surface occurred in the form of lateral spreading and other mass movements. However, the fractures, especially on
the slopes of the heights or hills, due to gravity, are directly related to earthquakes. Therefore, it is thought that this
type of rupture on the northern slope of the castle developed as a result of mass movement down the slope triggered
by an earthquake.
PRELIMINARY RESULTS OF RADIOCARBON DATING
The samples (Fig. 14-20) A YN-01, A YN-05, and A YN-06 derived from charcoal/charred wood pieces, A YN-04
from human bone, and A YN-02 and AYN-03 were soil from the northern trench. Analyses were performed at
TOBiTAK MAM Earth and Marine Sciences Institute Laboratory. Analysis results and obtained age data are given in
Table 14-1 and in detail in Table 14-2.
Figure 14-20. Location of radiocarbon samples on aerial photo in detail (red rectangles).
According to the analysis results, while some age data are very close to each other, some indicate a period out of
the castle's usage time. The period of the Urartian king Rusa, son of Argisti, who built the castle, is between 685---<i45
BCE. Bone sample (AYN-04) from the obtained age data is in this range. The charcoal samples AYN-01 and AYN-05
and the soil sample A YN-02 point to approximately the same time.
Radiocarbon Age Calibrated Age

Sample Material
!l:rBP} !l'.r cal. BC-AD}
AYN-01 Charcoal 2799 ± 25 1016 - 894 (%94.4)
AYN-02 Bulk soil 2793 ± 26 1279 - 1113 (%95.4)
AYN-03 Bulk soil 1842 ± 25 121 - 240 (%91.0)
AYN-04 Human Bone 2519 ± 25 650 - 544 (%51.2)
AYN-05 Charcoal 2821 ± 26 1044 - 910 (%95.4)
AYN-06 Charcoal 2404 ± 29 544 - 401 {%85.9}
Table 14-1. Results ofradiocarbon dating (Please see Table 2 for details).
AYN-01 AYN-02
g
i
l s
i
If I
a:
"' 2600
2400
1'00 1300 1200 1100 1000 000
--(l(al,IO) --(....0)
M
M M
AYN-03 AYN-04
8 i
I i 2400
l i ~ 2300
2200
--~) 301 ..
, ... 2100
~
900 TOO
~~
500 ...
--p(aloO())
2 Si ma Kalibras nu
Takv,m Ya , Olas1hk
MS 121 - 240 %910
MS 87 - 107 %44
AYN-05 AYN-06
AYN-06 14C_YA~I( 404, )
2600
% g
i
~
i
t
~
I
! ~
2600
!
22()0
2000
2000
2AOO
1200 1100 1000 000 100
700 600 500 400 300 200
800
--(1(11MO)
- - -(l(alMO)
2 S ma Kalibras nu
Takvim Ya 1 Olas1hk
M 1044 - 910 %954
Table 14-2. Details of the radiocarbon dating.

CONCLUSION
Although the dates of the wood samples AYN-01 and AYN-05 taken from the Temple Courtyard are consistent
with each other, they date to much earlier than 680-670 BCE, which is thought to be the foundation date of the castle.
Archaeological data show that long and strong timbers are needed to cover large areas. Therefore, considering that the
C14 results reveal the age of the trees, we can speculate that these dates are consistent with the possible age of the
wood used in the castle construction. Dendrochronology and C 14 tests on the same wood sample can increase the
accuracy of C14 tests and offer more precise dates with the double verification method. In the future, the project aims
to achieve such results.
The AYN-02 sample is a soil sample taken from the extension joint in a northern pit. It is relevant that the age of
this sample matches that from the two charcoal samples. Further, given that the AYN-03 sample was taken from a
layer close to the surface, it is not surprising to find a later date compared to the others; it is also possible that it is
contaminated with additional current carbon isotopes from the surface.
Although there are still debates about how life in Ayanis Castle ended and how the castle was burned, the general
view is that the castle may have been destroyed as a result of a large earthquake. Human bones found in the southern
fortification wall surrounding the stadium indicate that there were people under the ruins. These discoveries are the
first collective bone group found at Ayanis Castle. The AYN-04 sample taken from these bones helps date the
collapse of the castle. This is also important in terms of determining the date of the possible massive earthquake that
caused the overall destruction of the castle.
REFERENCES CITED
ABZ=R. Borger. 1978. Assyrisch-Babylonische Zeichen/iste. Alter Orient und Altes Testament Band 33. Neukirchen-
Vluyn: Neukirchener Verlag.
Aras, Oguz and Buket Be~ikiyi. Forthcoming. The Problem ofHorse-Riding Tribes Within The Archaeology ofEastern
Anatolia, and the Traces ofScythians Within Ayanis Castle. Leuven: Peeters Publishers.
Ayvazian, Alina. 2006. Urartian Glyptic: New Perspectives. Unpublished Ph.D. dissertation, University of
California, Berkeley. Ann Arbor: ProQuest Dissertations Publishing.
Batmaz, Atilla. 2015. Ayanis Kalesi'ndeki IJaldi Tapmag1'nm Depo Odalar1. In Uluslararas1 DojJ;u Anadolu- Giiney
Kajkasya Kiiltiirleri Sempozyumu: Bildiriler II/International Symposium on East Anatolia- South Caucasus
Cultures: Proceedings 11, M. I~1kh and B. Can, eds., 183-195. Newcastle upon Tyne: Cambridge Scholars
Publishing.
-. 2019. Protective Clay Figurines in the Urartian Fortresses. In Over the Mountains and Far Aw01. Studies in Near
Eastern History and Archaeology Presented to Mirjo Salvini on the Occasion ofhis 8(Jh Birthd01, P.S. Avetisyan,
R. Dan, and Y.H. Grekyan, eds., 58- 70. Oxford: Archaeopress.
CAD=The Assyrian Dictionary ofthe Oriental Institute ofthe University of Chicago. Chicago: Oriental Institute Press.
CTU I=M. Salvini. 2008. Corpus dei testi urartei, vol. I. Le iscrizioni su pietra e roccia. Documenta Asiana VIIUI.
Rome: CNR, Istituto di studi sulle civilta dell'Egeo e del Vicino Oriente.
CTU II=M. Salvini. 2008. Corpus dei testi urartei, vol. II. Le iscrizioni su pietra e roccia thesaurus. Documenta
Asiana VIIUI. Rome: CNR, Istituto di studi sulle civilta dell'Egeo e del Vicino Oriente.
CTU IV= M. Salvini. 2012. Corpus dei testi urartei, vol. IV. Jscrizioni su bronzi, argilla e altri supporti, nuove
iscrizioni su pietra paleograjia generale. Documenta Asiana VIII/IV. Rome: CNR, Istituto di studi sulle civilta
dell 'Egeo e del Vicino Oriente.
CTU V=M. Salvini. 2018. Corpus dei testi urartei, vol. V. Revisione de/le epigraji e nuovi testi su pietra e roccia
(CTU A), dizionario urarteo, schizzo grammaticale de/la lingua urartea, I Testi. Paris: Editions de Boccard.
<;avu~oglu, Rafet. 2014. Urartu Kemerler/Urartian Belts. istanbul: Rezan Has Miizesi.
<;ilingiroglu, Altan. 1997. Urartu Kralhg1 Tarihi ve Sanah. Izmir: Ya~ar Egitim ve Kiiltiir Vakf1 Yaymlan.
- . 2001. Temple Area. In Ayanis I: Ten Years' Excavations in Rusahinili Eiduru-kai, 1989- 1998, Altan <;ilingiroglu
and Mirjo Salvini, eds., 15-24. Documenta Asiana 6. Rome: Istituto per gli studi micenei ed egeo-anatolici CNR.
<;ilingiroglu, Altan and Mehmet I~Ikh. 2014. 25. Y1lmda Ayanis Kalesi Kazilar1 - Dun, Bugiin ve Gelecek. In
Anadolu'nun Zirvesinde Turk Arkeolojisinin 40. Y1h, H. Kasapoglu and M.A. Y1lmaz, eds., 309-324. Ankara:
Bilgin Kiiltfu Sanat.
<;ilingiroglu Altan and Mirjo Salvini, eds. 2001. Ayanis I: Ten Years' ofExcavation at Rusahinili Eiduru-kai, 1989-
1998. Rome: Istituto per gli studi micenei ed egeo-anatolici CNR.
Diakonoff, I.M. 1991. Sacrifices in the City of Teiseba (UKN 448): Lights on the Social History of Urartu.
Archaologische Mitteilungen aus Iran 24: 13-21.
Dini;ol, Ali M. and Ersin Kavakh. 1980 [1982]. Van Bolge Miizesinde Bulunan Yaz1th Bir Urartu Boncugw'Ein
beschrifteter urartaeischer Stein-kugel. Anadolu Ara~flrmalarz 8: 231- 234.
Emre, Omer, Tamer Duman, Selim Ozalp, Hasan Elmac1, Seyda Olgun and Fuat Saroglu. 2013. Annotated Active
Fault Map of Turkey. Scale 1: 1.250.000. General Directorate of Mineral Exploration and Exploration (MTA),
Special Publication Series-30, Ankara, Turkey.
192 Chapter Fourteen: I~lkh et al.
Fuchs, Andreas. 2012. Urartu in der Zeit. In Biainili-Urartu. The Proceedings of the Symposium held in Munich 12-14
October 2007, S. Kroll, C. Gruber, U. Hellwag, M. Roaf, and P. Zimansky, eds., 135- 161. Leuven: Peeters.
Grekyan, Yervand. 2016. A Note on the Toprakkale Tablet. Nouvelles Assyriologiques Breves et Utilitaires Mars/1:
54-56.
I~lk, Kenan and Mehmet I~1kh. 2015. Inscribed Bullae and Bone Layer Discovered at the Ayanis FortressNan in
2014. Altorientalische Forschungen 42(2): 142-152.
I~lkh, Mehmet. 2017. A Key Site in Urartian Archaeology: Recent Fieldwork at the Ayanis Castle (Rusag.inili Eiduru-
Kai). In The Archaeology of Anatolia, Volume II: Recent Discoveries (2015- 2016), S.R. Steadman and G.
McMahon, eds., 116-135. Newcastle upon Tyne: Cambridge Scholars Publishing.
I~lkh, Mehmet, Atila Tilrker, Oguz Aras, Ay~egtil Akm Aras, Mehmet Ali Ozdemir, and Gtil~ah Oztilrk. 2019. A New
Space in the Ayanis Citadel: The Hall with Podium. A Preliminary Report on the Excavations of 2014-2018. In
The Archaeology of Anatolia, Volume 111: Recent Discoveries (2017- 2018), S.R. Steadman and G. McMahon,
eds., 84-98. Newcastle upon Tyne: Cambridge Scholars Publishing.
I~lkh Mehmet, Haza! Ocak, and Buket Be~ikc;i. 2020. Van Ayanis Kalesi 2018 Y1h Kazi ve Onanm <;ah~malar1. 41.
Kazz Sonur;larz Toplantzsz, Cilt 3: 145-155.
I~lkh, Mehmet and Oguz Aras. 2020. A New Royal and Religious Space in Ayanis Citadel: The Hall with Podium. In
Proceedings of the 11th International Congress on the Archaeology of the Ancient Near East: Vol. 1: Mobility in
the Ancient Near East. Images in Context. Archaeology as Cultural Heritage. Engendering Near Eastern
Archaeology. Societal Contexts of Religion. Shaping the Living Space, A. Otto, M. Herles, and K. Kaniuth, eds.,
161- 170. Wiesbaden: Harrassowitz Verlag. doi:10.2307/j.ctv10tq3v9.
Kellner, H.J. 1991. Giirtelbleche aus Urartu. Prahistorische Bronzefunde, Abteilung 12, Bd. 3. Stuttgart: Steiner.
Lehmann-Haupt, C.F. 1906. Materia/ien zur alteren Geschichte Armeniens und Mesopotamiens. Abhandlungen der
Koniglichen Gesellschaft der Wissenschaften zu Gottingen, Phil.-Hist. Klasse IX: Berlin.
-. 1910-1931. Armenien, einst undjetzt, T. 2. Berlin-Leipzig: B. Behrs-Verlag.
Manning, Sturt W., Bernd Kromer, Peter Ian Kuniholm, and Maryanne W. Newton. 2001. Anatolian Tree Rings and
New Chronology for the East Mediterranean Bronze-Iron Ages. Science (21 Dec.) 294(5551): 2532-2535.
ME=R Labat and F. Malbran-Labat. 1988. Manuel d'Epigraphie Akkadienne. Signes, Syllabaire, Jdeogrammes (6e
edition). Paris: Librairie Orientaliste Paul Geuthner.
MMA 125=Annual Report ofthe Trustees ofThe Metropolitan Musewn ofArt 125 (1994-1995). 1995. New York.
MTA, 2002. 1:500 000 olc;ekli Ttirkiye Jeoloji Haritalan, Van paftas1. MTA Gene! Mudurlugu, Ankara.
Oman, Tallay. 2005. The triumph of the symbol: Pictorial representation of deities in Mesopotamia and the biblical
image ban. Fribourg-Gottingen: Academic PressNandenhoeck Ruprecht.
Piotrovsky, Boris Borisovich. 1966. Urartu Dini. Dil-Tarih ve Coffe-afya Fakiiltesi DerRisi 23(1-2): 37-52.
Potts, D.T. 1999. The Archaeology ofElam: Formation and Transformation of an Ancient Iranian State. Cambridge:
Cambridge University Press.
Poyraz, Selda, M. Alper Sengtil, and Ali Pmar. 2011. 23 Ekim 2011 Van-Tabanh Depremi Kaynak Mekanizmas1 ve
Sismotektonik Yorumu. Istanbul Yerbilimleri DerRisi: 24(2): 129-139.
RLME 3/2=1997. Royal Inscriptions of Mesopotamia: Early Periods, Ur 111 Period (2112-2004 BC), (312), Douglas
Frayne, ed. Toronto: University Press of Toronto.
RINAP 2=G. Frame. 2021. The Royal Inscriptions ofSargon II, King ofAssyria (721-705 BC). The Royal Inscriptions
of the Neo-Assyrian Period vol. 2, G. Frame, B.L. Eichler, K. Radner, and S. Tinney, eds. University Park,
Pennsylvania: The Pennsylvania State University Press.
Saglamtimur, Haluk, Gulriz Kozbe, and Ozlem <;evik. 2001. Small Finds. In Ayanis I: Ten Years' Excavations in
Rusahini/i Eiduru- kai, 1989- 1998, A. <;ilingiroglu and M. Salvini, eds. 219- 250. Documenta Asiana VI. Rome:
Istituto per gli Studi Micenei ed Egeo-Anatolici CNR.
Salvini, Mirjo. 1995. Geschichte und Kultur der Urartaer. Darmstadt: WBG Verlag.
-. 2001. The Inscriptions of Ayanis. Cuneiform and Hieroglyphic. In Ayanis I: Ten Years' Excavations in Rusahini/i
Eiduru-kai, 1989-1998, A. <;ilingiroglu and M. Salvini, eds. 251-319. Documenta Asiana VI. Rome: Istituto per
gli Studi Micenei ed Egeo-Anatolici CNR.
-. 2007. Die urartaische Tontafel VAT 7770 aus Toprakkale. Altorientalische Forschungen 34(1): 37-50.
- . 2012. Das Corpus der urartaischen Inschriften. In Biainili-Urartu The Proceedings of the Symposium held in
Munich 12-14 October 2007, S. Kroll, C. Gruber, U. Hellwag, M. Roaf, and P. Zimansky, eds., 111-134. Leuven:
Peeters.
Salvini, Mirjo and Ilse Wegner-Haas. 2014. Einfiihrung in die urartaische Sprache. Wiesbaden: Harrassowitz.
Schramm, W. 2010. Akkadische Logogramme. Gottinger Beitrage zum Alten Orient Band 5. Gottingen:
Universitatsverlag Gottingen.
Seidl, Ursula. 1979. Die Siegelbilder. InBastam I: Ausgrabungen in den urartaischen Anlagen 1972-1975, W. Kleiss,
ed., 137-149. Teheraner Forschungen Berlin: Gebr. Mann Verlag.
- . 2004. Bronzekunst Urartus. Mainz: Phillip von Zabem.
-. 2009. Musik und Tanz in Urartu. In A/tan <;ilingiroglu'na Armagan. Yukarz Denizin Kzyzsznda Urartu Krallzgz'na
Adanmz~ Bir Hayat (Studies in Honour of A/tan <;ilingiroglu. A Life Dedicated to Urartu on the Shores of the
Upper Sea), H. Saglamtimur, E. Abay, Z. Derin, U.A. Erdem, A. Batmaz, F. Dedeoglu, M. Erdalk1ran, B.M.
B~turk, and E. Konak91, eds., 607-619. Istanbul: Arkeoloji ve Sanat Yaymlan.
-. 2012. Rusa son ofErimena, Rusa son of Argi§ti and Rusahinili/Toprakkale. JnBiainili-Urartu. The Proceedings of
the Symposium held in Munich 12- 14 October 2007, S. Kroll, C. Gruber, U. Hellwag, M. Roaf, and P. Zimansky,
eds., 177- 181. Leuven: Peeters.
~engill, M. Alper, ~ule Giirboga, ismail Akk:aya, and Ali Ozvan. 2019. Deformation Patterns in the Van Region
(Eastern Turkey) and their Significance for the Tectonic Framework. Geologica Carpathica 70(3): 193- 208.
Walker, C.B.F. 1978. Texts and Fragments. Journal of Cuneiform Studies 30(4): 234-249.
Wartke, Ralf-Bernhard. 1993. Urartu-das Reich am Ararat. Kulturgeschichte der antiken Welt 59. Mainz: Phillipp
vonZabern.
Zimansky, P. 1985. Ecology and Empire: The structure of the Urartian State. Chicago: Oriental Institute Press.
-. 2012. Urartu as Empire: Cultural Integration in the Kingdom of Van. Jn Biainili-Urartu. The Proceedings of the
Symposium held in Munich 12- 14 October 2007, S. Kroll, C. Gruber, U. Hellwag, M. Roaf, and P. Zimansky, eds.,
101-110. Leuven: Peeters.
CHAPTER FIFTEEN
THE TUSPA CITADEL AND MOUND EXCAVATIONS, 2010-2019

BDLENT GEN<;::, ERKAN KONYAR, AND ARMAGAN TAN
THE TUSPA CITADEL
Tuspa, the capital of the Urartian Kingdom, identified today with the Van fortress on the eastern shore of Lake Van,
rises on a conglomerate rock and extends approximately 1250 min length (east-west), is 70-80 m wide north to south,
and is approximately 100 m high (Fig. 15-1 ). Various monumental buildings in the citadel area, such as the palace, rock
tombs, and city walls, are present in this rocky area, having been built during the development of the Urartian Kingdom.
In addition, the cuneiform instructions on the building blocks, stelai, and other inscriptions, containing various
Figure 15-1. Photo of the Tuspa Citadel.

narratives describing zoning and construction activities, and annals, are also found in this area of the capital. During
research carried out on the Tuspa Citadel since 2010, the focus has predominantly been on documentation and
architectural drawings (Fig. 15-2 and 15-3). In addition to these, excavations were carried out at various points on the
citadel, and new data reflecting the architectural traditions and transformations in the kingdom, such as the BG90
Underground Chamber Tomb, the Tuspa Urartian Fountain, and the Tuspa sirsini, were revealed.
I
\
_ _ _ _ _ ,.... 201,
~.,.,.,_.....
••
Figure 15-2. Plan showing the Building I and Building II areas.
The Tuspa Citadel and Mound Excavations 195
Figure 15-3. Plan of the entire Tuspa Citadel showing areas described in the text.
THE UNDERGROUND CHAMBER GRAVE: BG90
The BG90 underground carved rock tomb, which we unearthed in an area close to the eastern excavations on the
Tuspa Citadel, offers a different interpretation ofUrartian royal tombs. BG90, carved into the bedrock, generated new
discussions due to its unusual location and plan in comparison to other royal tombs. At issue is its chronological position
in Urartian tomb typology, based on its location, and the similarities to, and differences from, other royal tombs in the
citadel.
The tomb is located approximately 40 m east of the East Trench, 35 m south of the area where the niched stele
collection and Assyrian inscription are located; it was the first rock tomb unearthed at Tuspa Citadel (Fig. 15-4). It
exhibits a similar tradition with the grave type defined as Urartian Common Graves, based on features such as its
underground location and plan. At the entrance of the tomb, there is a pit-shaped dromos with a dimension of 2.28 x
2.15 m, with a depth of2.5 m, carved into the bedrock. The dromos leads to a door opening of 1.08 x 0.76 min a frame
measuring 1.60 x 1.30 m, and it is 32 cm deep; it serves as the entrance to the main room to the east. The rectangular
tomb chamber measures 5.80 x 3.15 m and in places is 2.60-2.80 m deep. A wall niche, 50 cm deep and 40 cm high, is
located 1 m above the tomb floor; it extends across the entire southern and eastern walls of the burial chamber, and onto
a part of the northern wall as well. The upper part of the niche inclines upward about 70 cm. From the upper edge of the
niche to the ceiling, the side walls rise at a right angle of approximately 80 cm. Thus, the depth of the main room reaches
ca. 2.80 m in some areas.
Due to its location and structural features, the burial chamber has been interpreted as quite different from the other
royal tombs in the citadel. At this point, its placement within the Urartian tomb typology and chronology is controversial.
The loss of data on the tomb's original discovery makes it necessary to assess it based solely on its typological and
architectural features. Also, its proximity to the Assyrian inscribed niche (CTU A 1-2) to the north of the tomb, which
survives in a highly damaged condition, suggests that its possible relationship to the niche should also be taken into
account.
The tomb's main room is accessed from the east by a single step carved into the bedrock; a framed door leads from
the dromos-shaped anterior chamber (Fig. 15-4, right). The door also has a shaft slot located on the dromos side.
Therefore, it could be opened and closed with a single wing which probably was made of wood. In Urartian chamber
graves outside the citadel, grave doors accessed from the dromos were usually closed with a flat stone. However, in this
citadel example, there is a more advanced entrance system, using door wings such as the types found at the
196 Chapter Fifteen: Gen9 et al.
Figure 15-4. Left: Underground Chamber Grave BG90; Right: The BG90 Rock Tomb.
Founders (Kurucular), Neft Kuyu, and East Chambers (Dogu Odalan) examples in the Van Fortress (Piotrovsky 1966:
302-309).
The niche that extends to the entire southern and eastern walls and the eastern part of the north wall of the tomb main
chamber shows similarities with the tomb chamber in the Cremation Tomb at Tuspa Citadel (Konyar 201 la: 216). The
unfinished state of the niche in the north wall can be explained by the fact that the bedrock in this area is quite damaged.
Niches are unique to the Urartians; a similar example is found in the Argisti I tomb. It is known that cremation vessels
(urns), or gifts for the dead, were placed in the niche in the Cremation Tomb. Therefore, it can be assumed that the niche
surrounding the walls of the burial chamber in the BG90 rock tomb was dedicated to the placement of cremation vessels
and grave gifts. Interpreting the BG90 rock tomb together with the Cremation Tomb at the citadel is important as it
shows that various types of tombs appealing to different burial traditions were used simultaneously at the Tuspa Citadel
during the Urartian period.
The structural features of the tomb are not clear. Usually, the top cover of rock-cut burial chambers is carved from
the bedrock. However, there is no bedrock layer that could form the top of the BG90 rock tomb chamber. Although no
evidence of the subsequent destruction of the existing top cover can be seen on the surface, the fragments of large and
small rocks dumped into the burial chamber are noteworthy. It should also be noted that the bedrock in this region is
soft and susceptible to fragmentation. The underground stone graves are built by creating false arches or installing long
flat flagstones. However, this is not possible in this tomb as the upper opening is approximately 3.5 min size. A drainage
channel on the upper floor extends from just east of the burial chamber, passing across the bedrock, to a point south of
the bedrock.
THE TUSPA FOUNTAIN
When examining the bedrock on which the Tuspa Citadel was built, the first features encountered are characteristic
Urartian structures such as fortification wall foundations and monumental rock-cut tombs. When viewing the rocky
foothills from the plain, mostly "T"-shaped, and occasionally rectangular rock-cut niches, both of which surround the
entire citadel, stand out. One of the rectangular niches, located in the area known as "Horhor Water" in the northwest
of the old Van city and southwest of Van Citadel, differs from the other niches; it features rock steps to the north
connecting it to the citadel above (Fig. 15-5, left). The area was excavated to understand the function and period of this
niche in connection with the citadel. The excavations revealed that this niche was a fountain/water structure (Konyar et
al. 2013a: 196--199). This niche, carved into the bedrock, is 1.42 m high, 1.04 m wide, and 65 cm deep. It features
another opening in the middle, largely rectangular in shape but narrower to the north. The opening extends towards the
niche through the rock cavity on an incline, and extends outward from the rocky area. A small pool, 95 x 75 cm in size
and 18 cm deep, carved into the rock, rests 90 cm below the niche. In the northeast comer of this pool, which is almost
square in shape, there is a small "V" -shaped water channel 180 cm long, 12-20 cm wide, and 13 cm deep. This channel
was carved into the bedrock just like the niche and the pool. The south side of both the channel and the pool is bordered
by bedrock (Konyar et al. 2015: 74).
At the channel's east end, an intact jar was found in situ, resting in a groove. Another pithos was found in situ, albeit
in a fragmented state, 50 cm east of the jar. This pithos was also seated in a specifically carved cavity in the bedrock.
From a point 40 cm east of the pithos, the bedrock continued on a downward slope. It is clear that this structure, located
in an area where natural water resources are present today, and probably were in the Urartian period as well, functioned
as a fountain (Fig. 15-5, right). Similar examples of water structures (taramanili) built together with Urartian inscriptions
have been previously documented (CTU A 5-58, 59). Although there are no inscriptions present at this Van Fortress
structure, based on the masonry technique and the stairs leading up to the citadel, it is reasonable to speculate that it was
built in the Urartian period and was continuously used until the water source dried up (Konyar et al. 2015).
Figure 15-5. Left: Rock steps connecting the tomb with the citadel above;
Right: The Tuspa Fountain, built in the Urartian period.
THE TUSP A SIRSINI
The structure known as the sirsini of the Urartian king Minua (810-785/80 BCE), located at the entrance of the rock
chamber, consists of a single room on the northern slope of the Tuspa site. This structure has been the subject of various
research endeavors since it was discovered by Schulz in 1827. New data on the function of the space were revealed with
the studies carried out in the building in 1989-1990 and then again in 2018. During the first scientific excavations in
the sirsini area between 1989-1990, a rock platform
of2.50-l.50 x 3.50 m was excavated in a rocky area
90 cm above ground level to the east of the building
entrance. A stele slot of 60 x 50 cm, with a depth of
20 cm, was found in the center of the platform
(Tarhan and Sevin 1993: 844, fig. 3). The sirsini,
which constitutes a unique case in Urartian
architecture, is filled with a dense rubble and garbage
layer that has accumulated over the years due to its
unprotected state. As a result of the excavation and
cleaning we have started in this area, we have
documented and examined the structure in more
detail (Konyar et al. 2019a: 177-178, figs. 20-22).
The interior dimensions of the structure were found
to be 20-30 m in length on the east-west axis, and
8.20 m wide north-south. Its height is 2.50 m in
general, narrowing to 2 m at the entrance. The
Figure 15-6. Inside the "Minua sirsini."
interior, therefore, has an area of 166 m 2 • Today, the
east-west entrance of the building, which is 8.45 m wide and 2 m high, has also been cleaned, and the floor has been
uncovered (Fig. 15-6). There is an area that has a slight inclination towards the north, approximately 90 x 55 m from
the entrance.
Excavations revealed that beginning 1.5 m in front of the entrance, the bedrock continued on a 70% slope towards
the north, and that large stones were lined up in two rows to the east and west of the entrance, at the beginning of the
slope. A sounding, 7.5 m (east-west) by 2.4 m (north-south), was conducted to better understand this area.
It became clear that the entrance of the sir§ini structure was originally not as wide as it is today, and that the wall
marks on the ground indicated that the entrance had been narrowed. Traces of possible stele and niche slots were also
found on the northern side of the narrowed entrance wall, indicating the original presence of a stele or an inscription.
The stele/inscription erected here was most likely a duplicate of other sir§ini inscriptions. That this area contained
inscriptions, together with its design and location on the slope of the Tuspa Citadel, which is the political-religious
symbol of the Urartian Kingdom, indicate that this structure was of importance. In addition, the sirsini should be
evaluated in the context of the 5000 m 2 wide terrace area (barnizi) at its front rather than solely as a rock-cut structure.
The stele slot, located on a platform built into the bedrock east of the entrance to the sirsini area, clearly indicates
that a stele was erected here. This platform is also suitable for the placement of objects other than stelai. Sacrificial
ceremonies are known to have taken place in front of stelai, in addition to the wine libations in Urartian ceremonies.
The same may have been the case here. Furthermore, it is likely that the barnizi area in front of the sirsini, mentioned
in the inscriptions, was a place where sacrificial animals were kept, to be sacrificed in the sirsini en masse as offerings
to gods or goddesses on religious ceremony days. As a matter of fact, in the first lines of the Me her Kap1 inscription,
which lists the Urartian gods/goddesses and the number of sacrificial animals to be slaughtered to them, the chief god
IJaldi alone required the sacrifice of 17 bulls and 34 sheep (CTU A 3-1, 4). It may be assumed that these sacrifices were
offered to the gods in the Urartian capital Tuspa before expeditions. Special places must have been needed for the
slaughter of such large numbers of animals. It seems unlikely that these mass sacrifice rituals would be performed in
Urartian temples due to the complexities of spatial logistics and health concerns.
When all the data are evaluated together it is clear that the rock platform in front of the sirsini, along with the stelai,
constitutes a unique complex particular to the capital. This complex can be considered an area where sacrificial animals
were kept for a short time, and then sacrificed.
TUSPA MOUND (VAN FORTRESS MOUND)
The first studies conducted at the Tuspa Mound, which stretches along the northern part of the Tuspa Citadel, were
carried out in 1939 by Kirsopp and Silva Lake (Lake 1940). Lake conducted soundings in the Old Van city south of
Van Fortress, and on the eastern part of Tuspa Mound to the north of the fortress. Researchers noted the presence of
Early Bronze Age and Urartian pottery found in the sounding on the mound (Korfmann 1977: 173-200, Taf. 111/2, IV/2;
1982: 195; Lake 1940: 179-191). After a long gap, ArifErzen and his team excavated a sounding on the western end
of the mound and reported the presence ofUrartian building foundations and pottery (Erzen et al. 1960: 20, 1963: 35).
The short-lived excavations conducted between 1989-1991 revealed a two-phased Urartian structure, defined as
Late (Phase I) and Early (Phase II), associated with the Urartian settlement on the mound (Tarhan and Sevin 1990: 433-
434; Tarhan 1994: 41-44, fig. 15). After these earlier projects, we renewed excavations on the Tuspa Mound in 2010,
which continued uninterrupted until 2019 (Fig. 15-7). Studies were carried out to understand the settlement periods and
Figure 15-7. Tuspa Mound during excavation.

structure levels on the mound. As a result, layers belonging to the Medieval and Modem Ages, post-Urartian/Late Iron
Age, Urartian, Early Iron, and Bronze Age were investigated, and important data belonging to the relevant periods and
layers were revealed.
MEDIEVAL AND MODERN AGE LAYERS
The Medieval and Modem Age layers, which form the latest layers of the Tuspa Mound, consist of a cemetery area
that covers the entire surface of the mound and the two phases of Medieval architecture. A large number of burials are
found in the cemetery, which was continuously used from the Medieval period to the beginning of the 20th century. The
simple earthen graves in this area were between 0.50 and 1.30 m in depth. All graves were in an east-west direction.
The posture of the skeletons unearthed indicates two different burial traditions. In the first group (Christian), skeletons
were laid on their backs facing east with the head to the west. The hands were clasped on the chest for women, and the
abdomen for men. Unlike Islamic burials, they were buried with jewelry. In the second group, consisting of Islamic
burials, the dead were buried without coffins, except in rare cases. An array of in situ iron nails and hooks unearthed in
one example shows that coffin-like caskets may have been used. In the Islamic burials, the skeletons positioned in the
east-west direction were laid on their right sides, with their head towards the west. Thus, the body and face of the person
faced south, towards the direction of qibla (toward Mecca); the hands were positioned on the sides, and the body was
wrapped in shrouds. The head was mostly supported by soil. Stones and adobe were placed behind the body to prevent
slipping and to hold the body sideways in the direction of qibla. In Islamic burials, if the soil is hard, a 40-50 cm wide
hole was opened on an east-west axis in the lower southern end of the grave pit, in the direction of the qibla, suitable
to the dimensions of the body; the dead were laid there (this practice is called Shami or lahd). After this, the burial was
covered with thin, flat stones, wood, or adobe. Especially in cases where flat stones and wood are used, the pieces were
Figure 15-8. The Medieval tombs from Tuspa Mound.
placed over the body on an incline. If the soil is not hard, the floor of the grave pit was dug so that the body could be
placed sideways, and the burial was completed by covering the body w ith stones, adobe, wood, or reed (this process is
called shakk). In the cemetery, two in situ skeletons were found buried facing north. There were also instances where
more than one body was placed in the same burial pit. Flat sandstone slate plates were used as headstones and footstones
on the top layer of most graves. A lthough rare, mudbrick was also used to support and cover the sides of some of the
grave pits. The covering techniques and other structural features of the grave pits can provide an understanding of the
chronology and burial customs in the cemetery (Konyar 2012: 413-414). Graves are typologically diverse here. The top
layers consist mostly of flat sandstones, with some of the graves including head and footstones. Except for an example
200 Chapter Fifteen: Gem; et al.
with a cross motif, no writing or decoration was found on these stones. Excavations revealed that the graves unearthed
during the study were of five different types (Konyar et al. 2012: 221), outlined below.
Type 1-Earthen graves covered with flagstones: These types of burials are commonly covered with semi-worked
flagstones (Fig. 15-8, photo on lower right). Grave pits are either elliptical or rectangular. The rectangular-shaped graves
feature a burial pit on the bottom where the grave width is enlarged. Adults are buried about 80-100 cm below the
surface, while children and infants were found to be buried 20-35 cm deep.
Type 2-Simple earthen graves without flagstones: The grave pits are elliptical and without flagstones covering the
graves (Fig. 15-8, middle photos at bottom). The boundaries of the pit can be easily identified in damaged graves. Grave
depth is 80-100 cm for adults and 20-35 cm for infants and children. In some graves, flagstones, or poplar/willow wood
material, 5-10 cm in diameter, was found at a depth of 50-60 cm. These would be used to cover the dead before filling
the grave with soil. These wooden remains, sometimes placed in an orderly manner, were found in situ.
Type 3-Mudbrick cist graves: These types of graves are formed by laying mudbrick blocks, 30 x 20 x 8 cm in size,
over the grave pits. In some of these tombs, the walls were built by placing mudbricks on top of each other using the
corbelling technique. Hence, the tomb walls narrow at the top. Mudbrick walls were observed on one or more of the
inner sides of the graves.
Type 4-Stone cist graves: The burial pits of this type were surrounded with semi-worked stones to form a simple
sarcophagus for the deceased. This type of burial has been found in very small numbers.
Type 5-Burials with gravestones, lined with partially dressed small stones, and filled with rubble in the middle (Fig.
15-8, photo on lower left): In these types of burials, the pit depth is usually over 1 m. The deceased was placed into the
pit and covered with flagstones which were then covered with soil. This soil layer would then be covered with rubble
and surrounded by medium-sized stones.
The large number of grave pits suggests that there had been intense burial activity for quite a long time in the
Medieval and Modem periods (Fig. 15-8, top plans). For this reason, graves that intersect and damaged each other were
frequently encountered. New graves were usually dug with care so as not to damage the old burials. When they did, the
skeletal remains belonging to the previous burial were collected in a separate comer of the grave pit. The similarity of
structural features and lack of elevational differences make it hard to distinguish graves chronologically. However,
Christian burials were more frequently found in the lower levels, at times intersecting with Islamic burials.
Consequently, it would not be wrong to assume that the cemetery was mainly utilized by Christians in the early periods,
but that Islamic burials became more frequent towards later periods (Konyar et al. 2013b: 129-130).
Small finds, placed as gifts, were found in situ in some of the graves. A bracelet made of 41 blue beads on the right
wrist of a child burial in trench N27 is one of them. The burial of a young adult female with 12 glass bracelets on her
arm in trench N24 is also noteworthy (Konyar et al. 2019a: 170-172). Other finds such as rings and glass bracelets
recovered from some of the graves have also survived to the present day in good condition. Two glass anklets were
found in situ on the ankles of a young child in the M28 tomb. Such burial items can be found, albeit rarely (Konyar
2012: 415; Konyar et al. 2019b: 66).
The grave fill often contains large amounts of pottery and other types of finds belonging to earlier layers of the
mound. Early Bronze Age Karaz pottery, Urartian ceramics with red burnish, and grooved ware, a small number of
painted and cream-plastered ceramics from the Late Iron Age, and glazed, sgra.ffito-decorated ceramic material in
various forms, mostly in small pieces, from the Medieval period, are typical examples encountered in grave fills. Finds
of glazed and decorated Medieval pottery, Byzantine coins, print decorated light clay pottery from the Selc;;uk period,
and Ottoman tobacco bowl fragments indicate that these two different traditions of burials continued throughout the
Medieval and Ottoman periods. In these layers, an 11th-century coin (X. Konstantinos: 25 December 1059- 21 May
1067-Constantinople), which was not in situ in the grave pits, points to a possible chronology for the Medieval layers.
In addition, the discovery of an almost complete Islamic period vessel with imprinted decoration, dating to the 12th- 13th
century is also noteworthy (Konyar 2012: 413-415).
MEDIEVAL ARCHITECTURE
Several Medieval architectural features were found under the cemetery layer in trenches M27-M30 starting from the
west end ofTuspa Mound (Konyar et al. 2012: 222). This faint presence of architecture, defined as the Early (Phase II)
and Late (Phase I) phases, was severely damaged by the grave pits. Although it was heavily damaged by early phase
burials, Phase I features a more intelligible architectural plan. The 80 cm thick Phase I walls, just below the cemetery
layer, indicate a large structure. These walls were built with large, semi-dressed stones and packed with small rubble
stones; mud was used as a binding agent. The walls, heavily damaged by the grave pits, seemingly belong to a large
space with a square plan, extending on a northeast-southwest axis. The function and chronology of the structure could
not be determined conclusively due to the lack of architectural integrity, the absence of in situ finds, and the lack of
solid pottery contexts.
The architecture of Phase II consists of interrelated rooms with a more distinctive plan compared to the architecture
of Phase I. The 80-90 cm thick walls were built with a single row of small stones. One or two lines of mudbricks can
be seen on some parts ofthe walls. These walls were also extensively damaged by the grave pits. The structure contains
several architectural features of varying dimensions, formed by main and partition walls extending on a northeast-
southwest/southeast-northwest axis. Tand1r oven remains were encountered in both levels of Medieval remains. The
features and walls unearthed in four trenches point to a building pattern that may cover a large part of Area A located
on the western part of the mound (Fig. 15-3). Almost all the pottery material found in the mound layers has been
intermingled due to the damage from the graves. While glazed and unglazed ceramics constitute the majority of these,
Urartian and Early Bronze Age pottery has also been found in large numbers (Konyar et al. 2013b: 130, 2019b: 66--67).
THE LATE IRON AGE/POST-URARTIAN LAYER
Architectural remains and in situ finds related to Late Iron Age/post-Urartian levels were unearthed in the west
trenches (N20, N21) and the eastern trenches of Area A (in trenches M25, M26, and M27) on the TuspaMound (Konyar
0 )Q~(.)
00000
ouooo
C
■ ■ ■ 15cm
Figure 15-9. Top: Pottery types from the Late Iron Age/post-Urartian layers;
Bottom: Some of the important finds from the Late Iron Age/post-Urartian layers: a) alabaster seals;
b) terracotta and bone amulets; c) bronze jewelry; d) carnelian beads; e) bronze fibulae.
et al. 2017: 127-130) (Figs. 15-2, 15-3). Excavations revealed the layer under the two Medieval phases, an orange-
colored thick clay filling, which was seen in the sections and understood to be located just above the Urartian layers.
This level, located just above the Urartian strata, provides important data regarding the layer defined as Late Iron
Age or post-Urartian, throughout the mound. The southern part of trench M25 revealed floors and double rows of walls
made of 45 x 45 cm mudbrick blocks laid roughly in a north-south direction, similar to those found in the M26 trench.
A wall 1 m in width, built with two rows of mudbrick blocks measuring 50 x 50 and 50 x 25 cm, unearthed in the
southern area of trench L26, is also stratigraphically important as it is located immediately above the Urartian structures.
Remains of this mudbrick wall, and a two-phased associated floor, can clearly be traced through trenches M25-M27,
L26, and N25, located in the eastern section of Area A; the wall rested on top of the in situ findings and the Urartian
strata. Numerous ornaments and pottery were unearthed in the burnt areas on this floor. The change in forms of pottery
and ware groups after the Urartian period is particularly noteworthy (Fig. 15-9a).
A large group of intact or restorable artifacts belonging to a post-Urartian period was uncovered in situ on the
preserved floor of stone chips. Numerous remarkable bronze artifacts such as beads of various types, needles, tweezers,
buttons, and fibulae were found. Also, beads and pendants made of glass, carnelian, and marble, and seals made of
alabaster were recovered. Amulets made of bone and terracotta with depictions of various creatures on them were also
collected. A large number of cylindrical bronze beads, hand-shaped pendants, and almost intact miniature vessels with
cream slip are some of the important finds from this layer (Fig. 15-9b). A vessel with a single vertical handle and the
depiction of a mountain goat on its spout, found in situ in trench M26, is another important find that can be evaluated
in relation to this layer. In terms of typology and ornament style, this unique, almost-intact vessel, has parallels to those
found in northwestern Iran.
Triangle-Festoon Wares with cream slip, dating to the Late Iron Age, made up a very small part of the overall
material finds of this layer. Additionally, the pottery found in this strata can be generally classified into five ware groups
which are currently under study.
LATE IRON AGE/POST-URARTIAN CEMETERY STRATA
Eight hocker burials, described below, were

unearthed during the Tuspa Mound excavations. This
may mean that the mound was used partially or
completely for burial purposes (necropolis?) after the
Urartian period (Fig. 15-l0a-b). The graves
belonging to these hocker-shaped burials were
occasionally dug in a way that damaged the mudbrick
walls of the Urartian architecture (Konyar 2011b:
153, 2012: 418; Konyar et al. 2018: 143-145, 2019a:
172-173).
Grave 1, in trench M25, is a Late Iron Age semi-
hocker burial. It was unearthed in a section near the
northeastern comer of a rectangular planned room
belonging to the Urartian building level; the grave is
built of mudbrick blocks, and rests just below the
layer belonging to the post-Urartian period. The
decedent, facing slightly south, had hands clasped on
the chest, with the knees bent in the hocker position.
The skeleton is presumed to belong to an adult male.
Grave 2 dates to the Late Iron Age and is located at
the comer where the western and northern walls of the
building noted above intersect. The grave is 110 cm
long, and the south side was bordered by five rows of
stones. This individual was lying face down with legs
pulled to the abdomen, facing slightly north. The
skeleton, thought to be an adult female, was 83 cm
long and 49 cm wide. This burial seems to be
contemporary with the hocker burial unearthed in the
excavations carried out by Kirsop and Silva Lake at
Anah K1z in 1938-1939 (Korfmann 1977: Levha 8-
1/2). Grave goods included with the skeletal remains
included bronze earrings, a bronze bracelet found on
the right arm, a bronze necklace, and a fibula on the
lower part of the neck. A number of various types of
beads were also found. An alabaster stamp seal
depicting a running goat or deer figure in the stamp
part was also uncovered, most likely hung around the
neck with the beads.
Figure 15-10. a) Hocker burial from the Late Grave 3 is a poorly preserved skull, also dating to
Iron Age/post-Urartian layers; b) Grave 5, Semi- the Late Iron Age, found in the upper layers of the
hocker burial from the Late Iron Age/post- entrance to the northwestern room of the Urartian
Urartian layers; note plate under left arm. complex. A bronze earring was found with the skull.
Grave 4 is a hocker burial, unearthed in the western rear rooms of the Urartian complex, just outside of the middle
room's western wall. The semi-hocker skeleton is thought to belong to a male, though the skull was not found. A
carinated bow1 fragment .c ut in half and Late Iron Age pottery with cream slip were also found next to the burial.
Grave 5 features skeletal remains of a female buried in the semi-hocker style. This burial was located in the
southernmost room of the colwnned hall ·i n trench M25; her grave had damaged a part of the western wall. The remains
lay in an east west direction and were found to be in a quite damaged condition. Th.e re were earrings in the ears, and
bronze bracelets with iron spirals were found on both wrists. In the neck area, a bead, a spiral-shaped bronze pendant,
and tweezers or a tweezer-shaped pendant, which we think may have hung around the neck, and a small fibula, were
found. Just below the left arm of the skeleton was a small, slightly crested plate with tile paste (Fig. 15-1 Ob). In add1tion,
a 2.07 cm semi-spherical object made of diorite stone with a thin bronze piece passing through the middle had been
placed on the plate. Similar examples to a fibula resting on the lower part of the right shoulder of the skeleton are known
from Ziwiye and the south Caucasus (Muscarella 1965: pl. 58, figs. 3, 6). This example bears a particular resemblance
to a specimen in the Adana museum (Ogiin 1979: 183, Abv. 6). The fact that the grave was dug into the Urartian layer,
partially destroying the Urartian wall, reinforces the connection of the tomb with the Late Iron Age. The broken and
missing rim of the plate under the left arm of the skeleton indicates that the plate may have been placed in the grave
m
already a broken form as a burial gift. Grave 6 is a hocker burial unearthed in Trench M24. No skull was found in
this grave, which intersected the western wall of the northernmost of the Urartian structures. The skeleton, lying on its
left side, was also poorly preserved. Unlike Grave 5, no grave gifts were found in this burial. This grave was located
within the wester.n door opening of the Urartian structure, which we discovered as excavations proceeded.
Grave 7 is another hocker burial, found in trench N24. The grave was dug into the western wall of an Urartian
building known as structure number 2; the graves lies in a south to north direction. This grave is also very badly
preserved.
Grave 8 ·i s a hocker burial, found ·i n trench Nl9, and extends in a north south direction. The grave destroyed a part
of the western foundation. wall as well at its southern corner; a pithes burial lacking a secure proveni.ence, most likely
disturbed, was found just south of this burial. A broken fibula was found near the highly damaged hocker burial. The
pithes burial, which was unearthed at the corner of the same wall, provided a richer collection of burial gifts, including
bronze and silver crescent-shaped earrings, bronze rings, and beads made of agate, carnelilll), and glass.
IBEURARTIANSTRUCTURELEVELS
The studies carried out in the western part ofTuspa Mound between 1989 and 1991 identified (Urartian) Early
Architectural Phase Ile, thought to be associated with the Middle Iron Age, and defined as a large complex within the
Urartian Architectural Layers, and the IIb (Urartian) Late Ar,chitectural Phase, a relatively weak architecture layer (Sevin
2012: 361 362).
With the resumption of the excavations at Tuspa Mound in 2010, old trenches and structural foundations in the
western area of the mound were cleaned and re-exposed. In the new excavations carried out in area N21 andN20 in the
western area of the mound (Konyar 2011b: 147 166, 2012: 409 428; Konyar et al. 2012: 219 245, 2013a: 193 210,
2017: 127 142), characteristic Urartian finds were found under the 8th century BCE (Tarhan 2011: 329) "Early
Architectural Phase" (Konyar, et al. 2012: 224). As a result of the ongoing excavations in this area, it is now understood
that the Ile Urartian structures, defined as the early architectural phase in previous studies, were actually built on an
Urartian ruin (Konyar et al. 2013a: 194 195). Therefore, when data from the 1989 1991 studies and those from the
current excavations are correlated, a new "Urartian Structure Levels" stratigraphy emerges (Fig. 15-11).
{
I
- . . . . .( f f _ . . _ _ 2019
,l VuforlA$$ Moull4
/A lAYfltSOf~n:AIOO
Figure 15-11. Plan showing the Urart1an penod layers.

THE LATE URARTIAN STRUCTURE LEVEL
Excavations carried out in trenches M23-M26 and N23-N26, east of Area A, revealed the remains of a large building
complex belonging to the Late Urartian Structure Level, including copious amounts ofUrartian pottery just below the
Post-Urartian layers (Fig. 15-12).
Figure. 15-12. Photo of Late and Early Urartian walls at Tuspa Mound.
The scattered and broken pieces of pottery, pithoi, and stone bowls in various forms, grinding stones, and grain
samples obtained from the hearth in one of the rooms belonging to this building, provide clues regarding the use of this
structure in daily life. In addition, the fact that pieces of the same vessels were located at different points in the space
indicates that these vessels may have been broken on purpose and scattered. Furthermore, the discovery of seals and a
large number of bronze nails and objects, as well as beads in different forms, indicates that the area was used for a
variety of purposes (Konyar et al. 2017: 133-134, fig. 9).
THE EARLY URARTIAN STRUCTURE LEVEL
As excavations continued in the rooms and floors of this Late Urartian Structure Level complex, fill associated with
the mudbrick ruins and burnt layers of the Early Urartian Structure Level were found, revealing the building foundations
of the Early Urartian Structure Level. The structures show that the Late Urartian Structure followed a plan similar to
that of the Early Urartian Structure Level.
A Monumental Building and Columned Hall
The Urartian period building complex in trenches M24-M26 and N24-N26, in which the Early and Late Urartian
Structure Levels were excavated, consists of 10 separate but connected units laid out in an approximately square plan
of20 x 20.50 m (Fig. 15-13a). The building was built on 3-4 rows of semi-dressed stone foundations upon which rested
walls built with 10-12 cm thick mudbrick blocks 48-50 x 30 cm in size. In the center of the complex, there is a hall
with columns at each comer (four in total), and a narrower hall/workshop of approximately the same length to the south
and north of this hall. There are four rooms connected by doorways in the eastern part of the building group and three
rooms connected by passages in the western part. The outer and inner walls are almost the same thickness and are
approximately 1.30 m.
As the outer floors and drainage channels were excavated to the east and west of the building, it became clear that
this was a special structure and that its front and rear functioned as independent structures with no other construction
near them. In addition, finds such as bullae, vessel fragments with cuneiform inscriptions, various bronze materials,
tablets (1~1k 2014: 176-179) (Fig. 15-13b), seals (Konyar et al. 2017: 131-137, fig. 9), and high-quality pottery obtained
from various rooms and its surroundings during the excavation of this building, reinforce the impression that the building
was dedicated to a special purpose. Consequently, it is now understood that the lower settlement at Tuspa contained
various public buildings related to the citadel. Thus this complex, located in the Early Iron Age layer, actually constitutes
one of the first structures of the Urartian Kingdom, specific to the capital Tuspa. The presence of a colonnaded hall is
also significant in that it serves as the first example of an architectural style which would later be applied in many diverse
forms in the area as the kingdom expanded.
Figure 15-13. a) The Columned Hall in the Urartian complex; b) Urartian tablet from the mound.
The hall in the Urartian complex, especially with its column bases, draws attention to aspects of size, design, and
function. Located in the center of the complex, this columned hall has a rectangular plan of9 x 7 m and stretches in an
east-west direction. The mudbrick walls with stone foundations are approximately 1.30 meters wide, and their current
height varies between 1.20 and 1.50 m. In front of the northern wall of the hall, there is a platform/bench 7.30 m long,
60 cm wide, and 50 cm high. This bench was built in front of the northern wall of the place using single and partly
double rows of stones and covered with adobe. Column bases are located near the comers of the building. Column bases
are 50-51 cm in diameter. The upper part of a column base, with a diameter of 35 cm, on which probably a wooden
column sat, was present in the middle of the bases. A similar column base was found in the area in the west of the mound
in 1991 (Tarhan and Sevin 1992: 424, fig. 17).
The features of the columned hall, particularly the way column bases are placed close to the comers of the rectangular
room, the bench in front of the northern wall, the adobe platform at the base of the bench and in between two bases, and
the fact that the floor is paved with mudbrick blocks, all attest that this is a special structure. The presence of a
rectangular altar/totem base-like platform, made of small and flat stones measuring 54 x 37 cm, in front of the western
wall, and the absence of any furnace or tandzr oven to heat the room, further enforces this idea. It is highly probable that
this room, which was intended for use by large gatherings of people, had cultic functions. The platform in front of the
western wall supports this interpretation. The columned hall and connected rooms unearthed at Tuspa Mound
demonstrate that our understanding of the purpose of the structures in the lower settlements should be revised. This
work at Tuspa has demonstrated that various public buildings were built outside the citadel, in the lower city (I~ik and
Gen9 2012: 72-79). The structure in which the columned hall is located may have also served cultural and religious
purposes in the lower settlement before the 1-Jaldi cult became widespread (Gen9 2016: 67-76).
When the in situ finds unearthed across the structure are evaluated as a whole, the special characteristic of this space
is further revealed. The small rooms and halls, pieces of pottery, storage containers, a silo, hearth, and tand1r oven, and
the remains of wheat, barley, and the bones of small cattle indicate the function of the structure in daily life. A cuneiform
tablet, bullae, bull head-shaped applique, and some fine-ware pottery, as well as numerous seals, provide clues about
the inhabitants of the building in association with the administrative mechanism. A large number of bronze jewelry
pieces unearthed from various areas, pottery fragments defined as Urartian palace wares, and unique vessel types, are
among the other finds that point to the special function of the building.
In the courtyard, in front of the area where the drainage channel to the west of the Urartian building complex is
located, various Urartian finds were also recovered, which were fragmented and scattered over the Late Urartian floor.
Notable among these are two broken snake-head bracelets (one of which is especially twisted) from these materials,
bronze fragments from a belt (Kellner 1991: 159, fig.15), various ring-shaped bronze pieces, a decorative object made
of diorite, and partial and completed beads and various container parts.
The Early Urartian layer was investigated after documenting the stone chips fill and surface of the Late Urartian
Structure Level in the western part of the complex. Various finds such as Urartian pottery pieces, bronze pieces, beads,
an obsidian arrowhead, and an iron arrowhead were found in this fill. A ceramic piece with cuneiform inscriptions found
in the fill is especially significant. The inscription is on the rim of a vessel fragment and reads "[ ar-gis] ti - e-I," which
states that the vessel is "Argisti's." The cuneiform vessel fragment is important in terms of associating this building
with the royal culture. In the Urartian culture, previous examples of bronze vessels with cuneiform labels revealed the
name of the king to which the vessels belong.
THE EARLY IRON AND BRONZE AGES
In trench N23, the fill that formed the Urartian floor, consisting of pebbles and stone chips of various sizes, the depth
of which reached 80 cm in places, was removed, and the pre-Urartian cultural layers were investigated. The eastern
sections clearly showed that this Urartian stone fill directly overlaid the Early Iron Age pits in places. This indicates a
particularly short time-frame between the Urartian
period and the weak Early Iron Age layer to which
these pits belong (Fig. 15-14).
The presence of a very hard ground made of
compacted soil just below the Urartian layer has been
documented. It is understood from stratigraphic data
and some pottery pieces that this layer belongs to the
Early Iron Age. During the examination of this layer,
a large pit cut with a diameter of 2.70 m was
unearthed in the southwestern part of trench N23. In
addition, another pit measuring 1.30 x 0.63 m was
excavated in the western section of the trench. The
pits in the western section of the N23 trench were dug
into the Early Bronze Age layers during the Early
Figure 15-14. The Early Iron Age pits on the mound. Iron Age.
No architectural remains in this compacted soil and pebble layer have been found in trench N23 so far. However,
during excavations in 2017, a contemporary to this layer was located in the N26-N27 step trenches, consisting of a
poorly-preserved wall line with a two-room plan located just above the Early Bronze Age layer, the latter consisting of
mudbrick debris. These wall lines and the pits that cut into the Early Bronze Age layers belong to the Early Iron Age.
Fragments of grooved ware, attested since the Early Iron Age, were encountered in connection with these architectural
elements.
After removing the Early Iron Age layer pits in trench N23 and the faint architectural elements encountered in the
N26-N27 trenches, the Early Bronze Age layer consisting of mudbrick debris was revealed throughout the trench. In
the southern section of trench N23, an Early Iron Age layer rests on a two-phase Bronze Age layer. In trenches N26 and
N27, an undulating surface used as an open-air area, with a length of 12 m in an east-west direction and a width of 6 m
in the north-south direction was discovered, along with various associated finds (Fig. 15-15a-b).
b
B-N57 VK12503
0 20cm.
Figure 15-15. a) The Early Bronze Age Layers; b) an andiron from the Early Bronze Age layer.
CONCLUSION
During the excavations carried out at the Tuspa Citadel, many important structures such as the underground chamber
tomb defined as BG90, the Tuspa fountain, and the Tuspa sirsini were unearthed. The BG90 rock tomb on the citadel,
in particular, provided important information about the transformation of royal tomb tradition, location, and design, as
it is the first rock tomb unearthed in archaeological excavations on the citadel. Important data regarding water-feature
architecture of the Urartian period were obtained from the Tuspa fountain excavations. It is believed that the word
"taramani" or "taramanili" encountered in the Minua period inscriptions located north of Van Citadel means "fountain
or water source" in the Urartian language. However, no water-related architecture has been found in the excavations in
and around this area north of the citadel. It is not known whether the "Minua Fountains," described as built by Minua
in the Van Citadel, are related to the fountain structure located to the south. The studies carried out in the Tuspa sirsini
focused on understanding the importance and function of this area to the citadel. The area with clear borders on the
208 Chapter Fifteen: Gem; et al.
rocky slope and a rock room with stelai in front of it may be where animals brought for religious rituals were kept for a
short time, perhaps in preparation for ceremonies before being sacrificed.
At the Tuspa Mound, the layers belonging to the Modem Age, Medieval period, post-Urartian, Urartian, Early Iron
Age, and Bronze Age were studied, and the architectural and material data from these layers were documented in detail.
The latest layer of Tuspa Mound was used as a cemetery in the late Medieval period until the Modem Age. Numerous
graves were excavated belonging to the Medieval and Modern Age layers. Prior to this period, a two-phase Medieval
architectural layer was documented. The intensive use of the mound as a cemetery in the later periods caused significant
damage to the earlier Medieval, Late Iron Age, and Urartian layers. The post-Urartian layer is the most damaged layer
on the mound due to its use as a necropolis during the later periods. Nonetheless, architectural remains and important
finds belonging to this layer were unearthed. Hocker burials and associated finds representing this period are particularly
noteworthy. Urartian structure levels were unearthed just below the post-Urartian/Late Iron Age layer.
The mound consists of at least two structure levels (the Early and Late Urartian), with additional repair phases.
Important data were obtained by completing the plan of the Urartian complex of20 x 20.50 m, whose wall thickness
reached 1.30 meters in places, and which consisted of 10 rooms, including a colonnaded hall believed to be used in the
Early and Late Urartian structure levels. As the outer floors and drainage channels were unearthed in the east and west
of the structure, it became clear that it was a special structure, and that its front and rear were not developed on purpose.
Additionally, the various finds such as bullae, inscribed vessel fragments, tablets, seals, and fine ware uncovered from
various rooms and its surroundings during the excavation period of this building reinforce the impression that the
building is a unique structure. This indicates the existence of public buildings and units related to the citadel in the lower
settlement of Tuspa. The pre-Urartian Early Iron Age layer has also been documented, and two structure levels
belonging to the Bronze Age under this layer have been identified for now. Especially the Early Iron Age layer we
unearthed in trench N23 has remained relatively intact due to the fact that the Urartians did not build in this area. Future
studies in the upcoming years will reveal even more significant data about the pre-Urartian Early Iron Age and Bronze
Age periods.
REFERENCES CITED
Erzen, Afif, Emin Bilgir;:, YusufBoysal, and Baki Ogtin. 1960. 1959 Toprakkale Sondajlan ve Bolgedeki <;ah~malar.
Turk Arkeoloji Dergisi 10(2): 5-22.
- . 1963. Van <;evresi 1963 <;ah~malan. Turk Arkeoloji Dergisi 12(2): 34- 36.
Genr;:, Billent. 2016. The Door ofljaldi in Pagan/Ye~ilahr;: and A New Approach on Susi Temple. Aramazd/Armenian
Journal ofNear Eastern Studies IX(2-2015): 67-76.
I~Ik, Kenan. 2014. Van Kalesi Hoyiigu Kaz1larmda K~fedilen Urartu Yaz1h Belgeleri. Colloquium Anatolicum XIII:
173-83.
I~Ik, Kenan and Btilent Genr;:. 2012. The Location of URU Tuspa and A New Inscribed Column Base Belonging to
Ispuini, King ofUrartu. Aramazd/Armenian Journal ofNear Eastern Studies VII(l): 72- 79.
Kellner, Hans-forg. 1991. Grouping and Dating of Bronze Belts. In Urartu: A Metalworking Center in the First
Millenium BCE, R. Merhav, ed., 142-61. Jerusalem: Israel Museum.
Konyar, Erkan. 201 la. Urartu'da Mezar Tipleri ve Gomu .Adetleri/Tomb Types and Burial Traditions in Urartian. In
Urartu: Dogu'da Degi~im/Fransformation in the East, K. Koroglu and E. Konyar, eds., 206- 231. istanbul: Yap1
Kredi Yaymlar1.
-. 201 lb. Excavations at the Mound of Van Fortress/Tuspa. Colloquium Anatolicum 10: 147-66.
-. 2012. Van-Tu~pa A~ag1 Yerle~esi Van Kalesi Hoyiigu Kazilar1. Kazi Sonuc;larz ToplantlSl 33(3): 409-28.
Konyar, Erkan, Can Avc1, Btilent Genr;:, Riza Gurler Akgun, and Arnlagan Tan. 2013a. Excavations at the Van Fortress,
the Mound and the Old City of Van in 2012. Colloquium Anatolicum XII: 193- 210.
Konyar, E., Ismail Ayman, and Can Avc1. 2012. Excavations at the Mound of Van Fortress- 2011. Colloquium
Anatolicum XI: 219-45.
Konyar, E, i. Ayman, C. Avc1, D. Yigitp~a, B. Genr;:, and R.G. Akgtin. 2013b. Van Kalesi Hoyiigu 2011 Y1h
<;ah~malar1. Kazz Sonuc;lan Toplantzsz 34(2): 127-36.
Konyar, Erkan, Billent Genr;:, Can Avc1, Riza Gurler Akgun, and Arnlagan Tan. 2015. Van Kalesi/Tu~pa Horhor
<;e~mesi. Colloquium Anatolicum 14: 73- 81.
Konyar, Erkan, Billent Genr;:, Can Avc1, and Armagan Tan. 2017. The Van Tuspa Excavations 2015-2016. Anatolia
Antiqua XXV: 127-42.
-. 2019a. Excavations at the Old City, Fortress, and Mound of Van: Work in 2018. Anatolia Antiqua XXVII: 169-83.
-. 2019b. Eski Van ~ehri, Kalesi ve HoyiiguKazilar1 2018. Kazz Sonuc;larz Toplantzsz 41(2): 65-83.
Konyar, Erkan, Btilent Genr;:, H. Banu Konyar, Arnlagan Tan, and Can Avc1. 2018. Excavations at the Old City, Fortress,
and Mound of Van: Work in 2017. Anatolia Antiqua 26: 143-53.
Korfrnann, Manfred. 1977. Die Ausgrabungen von Kirsopp und Silva Lake in den Jahren 1938 und 1939 am Burgfelsen
von Van (Tuspa) und in Kalecik. Berytus. Archaeological studies 25: 173-200.
-. 1982. Tilkitepe: Die ersten Ansiitze priihistorischer Forschung in der ostlichen Tiirkei. Tubingen: Ernst Wasmuth.
Lake, Kirsopp. 1940. Vanda Yap1lan Hafriyat, 1938. Tiirk Tarih, Arkeologya ve Etnografya Dergisi 4: 179- 91.
Muscarella, Oscar White. 1965. A Fibula from Hasanlu. American Journal ofArchaeology 69: 233-40.
The Tu~pa Citadel and Mound Excavations 209
Ogun, Baki. 1979. Urartaische Fibeln. In Akten des VII. Internationalen Congresses fiir Iranische Kunst und
Archiiologie, Miinchen, 7.- 10. September 1976, 178- 188. Archaologische Mitteilungen aus Iran, Erganzungsband
6. Berlin: Dietrich Reimer Verlag.
Piotrovsky, Boris B. 1966. 11 Regn.a di Van: Urartu. Roma: Edizioni dell'Ateneo.
Sevin, Veli. 2012. Van Bolgesinde Post-Urartu Donemi: Y1kmt1lar Ozerinde Yeni Bir Ya~am. Belleten LXXV-276:
353-70.
Tarhan, M. Taner. 1994. Recent Research at the Urartian Capital Tushpa. Tel Aviv 21(1): 22-57.
- . 2011. B~kent Tu~pa/The Capital City Tushpa. In Urartu: Dogu'da Degi~im/Transformation in the East, K. Koroglu
and E. Konyar, eds., 286--333. istanbul: Yap1 Kredi Yaymlan.
Tarhan, M. Taner and Veli Sevin. 1990. VanKalesi ve Eski Van ~ehri Kaz1lar1-1989. Kazi Sonuc;lan Toplantzsi XII(2):
429- 56.
-. 1992. Van Kalesi ve Eski Van ~ehri Kaz1lar1-1991. Kazi Sonu<;lan ToplantlSl 14(1): 407-29.
- . 1993. Van Kalesi ve Eski Van ~ehri Kaz1lan 1990 Y1h <;ah~malan. Belleten LVII(220): 843- 61.
CHAPTER SIXTEEN
HIERAPOLIS OF PHRYGIA 2018-2020:

RECENT WORK AND NEW PERSPECTIVES
GRAZIA SEMERARO
This chapter presents the results of the most recent campaigns conducted by the Archaeological Mission at
Hierapolis of Phrygia. After a brief introduction on the history of studies, the results of research carried out in recent
years (2018- 2020), and the multidisciplinary approaches to the study ofHierapolis of Phrygia, are also presented.
Founded in the Hellenistic period through the initiative of the Seleucid rulers, the city stands on a terrace
overlooking the Lykos Valley, a tributary of the Meander, in an extraordinary landscape originating from the presence
of important natural phenomena, linked to the strong seismicity of the area. Thermal waters at high temperatures,
along with poisonous gases, flow from fractures in the ground. The area is rich in limestone, which shaped the white
travertine "waterfalls."
The landscape described by Vitruvius and Strabo (Ritti 1985: 16, 2017: 7) is still recognisable in the travertine
canals running along the slopes of the fertile countryside. Literary sources also contain a rich repertoire of references
about the seismicity of the area by mentioning the numerous earthquakes that affected Hierapolis and other towns in
the Lykos Valley (Ritti 1985: 23-26, 2017: 11-15). Such tales are mainly related to the Hellenistic and Roman ages,
when the area became much more urbanized with the foundation of the Macedonian colonies Hierapolis and
Laodikeia.
Very few literary sources refer to the period before Greek colonisation, when the only important city in the region
seems to have been Colossae. We know from Herodotus that the area fell on the border between Lydia, Caria, and
Phrygia, marked by a stele set up in the city of Kydrara by Croesus, King of Lydia. Many scholars have long
discussed the unknown location ofKydrara, suggesting that it was in the territory occupied by Hierapolis (Ritti 2017:
270, with references).
HISTORY OF RESEARCH
Research at the site of Hierapolis began in the late 1950s with a series of projects related to the Roman (theatre,
North agora, plateia) and Byzantine towns (Martyrion). Initially, Paolo Verzone led the Italian Mission, after which
Daria De Bernardi Ferrero directed the work beginning in 1982, both professors of the Polytechnic of Turin. Between
2000 and 2015, the Mission was directed by Francesco D ' Andria (University of Salento).
Over the past 20 years, a rich programme of investigation, together with the extensive application of
interdisciplinary methodologies, produced numerous results, making Hierapolis one of the best-known Hellenistic
Roman cities in Asia Minor (Fig. 16-1). Specific research programmes involved investigating the place of worship
dedicated to the polyadic deity Apollo, residential structures and nymphaea, and the study of the territory and geology,
in particular its seismic peculiarities (Scardozzi 2015).
Important discoveries such as the tomb of St. Philip highlighted the Christian phase of occupation (D'Andria
2016-2017), while the identification of the Ploutonion (D'Andria 2013) allowed recognition of one of the complexes
that made the city famous in antiquity. This research permitted researchers to better define the chronological and
topographical problems linked to the city's plan and its development over time, which can be summarised as follows.
Pre-Greek Foundation and The Hellenistic Age (3rd-JS' Centuries BCE)
Little evidence about the pre-Greek occupation of the site was derived from the investigations prior to 2015. This
was rectified by the excavation of the Phrygian necropolis, begun as an emergency intervention by the Museum of
Denizli in 2016, and then continued by the Mission in the following years (see below) (Fig. 16-1, see no. 1).
The Hellenistic phase (3rd-1 st century BCE), which is the foundation of the Greek city, is poorly documented in the
archaeological record because of the later occupational phases (D' Andria 2001). Research has shown that the tumulus
tombs in the North and South Necropolis date to the early stages of the city. Before the construction of the great
commercial agora, the northern area of the city was dedicated both to agricultural and artisanal production activities
(Semeraro 2017, with references). Very interesting data came from the sacred areas in the early stages of the city
(Sanctuary of Apollo: Semeraro 2007; Ploutonion: D ' Andria 2019b).
Hierapolis of Phrygia 2018-2020 211
-I.
A
150 300
I) Ph rygian Necropo li s
_./
2) North Necro pol is
3) Poo l a nd product ive
structu re ,',,
4) North-West Necro poli s
5) so ca ll ed Baths-C hurch
6) North T heatre
7) Mi ddl e Byza nt ine
Churc h and Cemetery
8) North Frontinu s Gate
9) Pl ate ia
23
:~
I 0) No rth Agora
11 ) Stoa-bas ili ca
12) Turkis h Houses ...
I
13) Ea rl y- Byzantin e Wa ll s 27) Ploutonion
14) Nym ph ae um of th e Tri tons 28)
15) Great Building 29)
Marble Stoa
Church with Pillars (
16) Byzantine Baths 30) lnsula I 04
17) Cathedral 3 1) Theatre
I 8) North-East Necropolis 32) Church above the Theatre
19) Octagonal Baths 33) East Necropolis
20) St Philip Martyrion 34) so-called Bouleuterion
2 1) St Philip Church 35) so-called Macellum
22) Castellum Aquae 36) Gymnasium
23) Seljuk Fortress 37) South-West Necropolis
24) Great Baths 38) South Frontinus Gate
25) Civil Agora 39) South- East Necropoli s
26) Sanctuary of Apoll o 40) South Bridge
Figure 16-1. City plan (adapted from Scardozzi 2015).
The Augustan and Julio-Claudian Age
Between the 1st century BCE and the 1st century CE, numerous monuments along the axis of the great plateia
embellished the city (Fig. 16-1) (Scardozzi 2015). Marble buildings decorated the sacred areas, in particular in the
Sanctuary of Apollo and Ploutonion. The first phase of the theatre dates back to this period, as do other important
monuments such as the gymnasium and the civil agora. The epigraphy testifies to the commitment of wealthy citizens
who, as euergetes (benefactors), provided the funding for the construction of monuments in honour of the imperial
house.
212 Chapter Sixteen: Semeraro
The Antonine and Severan Age
Between the second half of the 2nd and the first half of the 3rd century CE, Hierapolis went through a period of
renewed splendour. The magnificent f~ade of the stoa basilica, along the eastern side of the commercial agora (Fig.
16-1, see no. 11 ), is one of the most significant monuments of this period, completed under the reign of Antoninus
Pius (D'Andria and Rossignani 2012).
The theatre, with a rich scaenae frons decorated with narrative stories in honour of Apollo and Artemis (D 'Andria
and Ritti 1985), dates back to the Severan period, as well as do important monuments such as the Nymphaeum of the
Tritons (Campagna 2018). Even the Sanctuary of Apollo underwent a radical renovation that involved the construction
of the Nymphaeum and the reconstruction of Building A in the Sanctuary of Apollo (Semeraro 2014).
The Proto-Byzantine and Middle Byzantine Ages
The Proto-Byzantine period is the best documented archaeologically. Between the 4th and 6th centuries CE,
Hierapolis experienced great vitality, in spite of the damage caused by seismic events.
The presence of an important Christian community led to the building of numerous churches within the urban
landscape, while a new monumental pole linked to the cult of the Apostle Philip was raised on a hill to the east of the
city (D'Andria 2017, 2016-17).
The research permitted us to understand the numerous transformations of the main buildings of the imperial age,
including the northern agora, which had been abandoned, and was at this time once again used as an artisanal area.
Important changes involve the sacred areas that had been dedicated to pagan (non-Christian) cults, such as the
partial destruction of the sanctuary of the polyadic deity, Apollo (Semeraro 2007), and the downsizing of the
Ploutonion (D'Andria 2019a). Study of the housing area (fusula 104, see Fig. 16-1, no. 30) revealed, however, that
there was an occupational continuity in the residential sectors, which were renovated and decorated with :frescoes and
mosaics made by local workers (ZaccariaRuggiu 2012, 2019).
fu the Middle Byzantine Age, after the middle of the 7th century CE, the city went through a long phase of deep
transformation, which featured the abandonment of the public areas and a slow ruralisation. fu addition to the effects
of seismic events, changes in the community social structure led to changes in occupational patterns in the territory,
including in the Middle Byzantine village, the necropolis, and rural buildings, in correspondence with the now
abandoned public complexes (Arthur 2006; D'Andria 2018b).
The Selruk Age
Between the 12th and 13th centuries the settlement centre constricted to the area once occupied by the Roman baths.
A Sel9uk castle was built on the edge of the terrace on which stood the ancient Roman city, to control the vast Lykos
Valley (Arthur 2006).
The sanctuary of Saint Philip continued to be a destination for pilgrimages, including the famous pilgrimage of
Federico Barbarossa, who crossed a city now in ruins, to those documented by recent excavations in the East
Necropolis (D' Andria 2017; Ahrens and Brandt 2016).
RESULTS FROM THE 2018 TO 2020 SEASONS
Over the last few years (2018-2020), research has focused on the reconstruction of urban landscapes through the
development of several lines of investigation aimed at:
1- Studying the spatial organisation of selected contexts, in relation to the city's urban landscape, in different
occupation periods.
2- Exploring the relationship with natural phenomena, in particular those of a seismic nature.
3- Exploring the most ancient phases, in relation to the settlement modalities of the territory in the pre-Greek
phase.
The following reports on our activities in these three areas.
Spatial Organisation
This line of inquiry has been developed through investigations in the residential sectors, in the necropolis, and in
the sacred area of the Sanctuary of Apollo.
After the publication of the work carried out between 1990 and 2012 (Zaccaria Ruggiu 2019) the exploration of
fusula 104 was resumed in order to complete the survey of the unexcavated portion in the western part of the block
(Fig. 16-2). The project1 has as its objective the definition of the internal organisation of the block, the defining of the
chronological issues, and the definition of the plan of the "House of the Painted Inscription." The name comes from
the extraordinary discovery of the inscription that reproduces the complete text of the Psalm of Manasse, a text of
great importance for the reconstruction of the spiritual climate of the early Byzantine age (D' Andria et al. 2005-
2006). The interpretation of the small room on whose walls the inscription is painted can be further explored thanks to
the understanding of the entire living space. The new excavations allowed us to complete the plan of the house, and to
identify a fourth dwelling, still to be explored, in the remaining area of the block.
Figure 16-2. Insula 104, residential area. Location of the new excavation sondages.
Figure 16-3. North Necropolis. Tomb 55, exedra.
1
Headed by I. Romeo (University of Florence), in collaboration with Davide Amesano and students from the University of
Florence and University of Pamukkale.
The theme of spatial organisation has also been at the centre of the work carried out in recent years in the
necropolis. A survey was undertaken in order to produce a detailed map that would allow a 1:500 scale display of the
entire North Necropolis. 2 This work has yielded a cartographic base for linking inscriptions on the various sarcophagi
in the necropolis, in concert with the publication of the epigraphic corpus. 3 This cartographic instrument has an
incalculable value for capturing knowledge of the enormous epigraphic and architectural heritage contained in the
North Necropolis ofHierapolis, one of the largest in the ancient world.
Other work included a sector of the necropolis, which was the subject of a detailed survey for the exploration of
the horizontal stratigraphy. 4 The use of this part of the necropolis begins in the Hellenistic Age, with the construction
of a tumulus tomb, and continued throughout the Imperial Age. Subsequent enlargements and subdivisions can be
partially reconstructed thanks to epigraphic data (Anguissola and Costa 2020). The internal paths and the connection
with the road were brought to light, along with the exedra structures (Fig. 16-3) that characterize numerous funerary
complexes in Hierapolis. They are an important tool for studying the use of funeral space by citizens during religious
ceremonies.
The numerous problems concerning the spatial organisation and development of the Sanctuary of Apollo,
dedicated to the polyadic deity, have been at the centre of research in the last two years. Although the excavations
conducted between 2001 and 2011 allowed us to define the layout of the sanctuary in its various phases, and to study
the main buildings, much remains to be done in order to complete the study, and to articulate the role of the sacred
area within the landscape of the city. One of the main problems is to define the structure of the sanctuary in the early
stages.
The surveys carried out in 2018-2019 (Fig. 16-4) 5 allowed us to assign the great terracing wall (US 2001) that
defines the sanctuary to the west to the Hellenistic period (Figs. 16-Sa and 16-Sb), where the travertine staircase that
connects the lower and the median terrace opens.
\
\
'\
\
'\
\
\
\
\
\
\
\
\
\
\
\
\
\ ," -:, ,..
y '
\ ,.- r-"
\ /:,,.
\,.,"
\
\
\
\
\
/J,, O 5 10 m
\\ \ N
Figure 16-4. Plan of the Sanctuary of Apollo, investigations 2018-2020.
2
Scientific studies being carried out by G. Scardozzi (National Research Council) in collaboration with Imma Di Taranto.
3
Study conducted by Tullia Ritti and Giuseppe Scardozzi.
4
The project is headed by A. Anguissola (University of Pisa) and Silvana Costa.
5
Survey team of collaborators: Valentina Aquilino (responsible), Gaia Sabetta, and Vincenzo Ria, with students from the
University ofSalento and University of Pamukkale.
The excavation of a trench (Fig. 16-4, see no. 7, and l 6-5b) permitted us to explore the foundation structures both
of the travertine staircase and the terracing wall (US 2001). Sherds from some black-glazed cups, from the layers of
the foundation walls, are datable to the 2nd century BCE and represent the most ancient materials. The excavation in
this sector is not complete and will continue in the coming years in order to acquire additional stratigraphic
information on the foundations of the Doric order portico that, according to a stylistic study (Ismaelli 2009), appears
connected to the next phase, the Julio-Claudian age. In this period, the sanctuary acquired a monumental aspect,
thanks to the construction of temples B and C, and the second phase of temple A, all connected through an imposing
system of marble staircases. Marble was also used for the fa9ades of the stoai which define the temenos along the
western and eastern side; however, travertine was preferred for the external structures such as the rear wall and the
colonnade, which divide the sanctuary from the southern area (Fig. 16-4, see no. 4; Fig. 16-6).
a b
Figure 16-5. a) South area ofNymphaeum. View from a drone;
b) Sondage no. 7 between the terracing wall (US 2001) and the travertine staircase.
Figure 16-6. South Portico, sondage no. 4.

The new research also made it possible to define the limits of the sacrificial area, identified in past years, in the
area north of temple B (Fig. 16-4, see no. 2). It is one of the most important spaces in the sanctuary since the
stratigraphy indicates peculiar cultic activities (Semeraro 2007, 2012, 2014). It is characterized by numerous
structures related to the use of fire, most of them consisting of small pits full of charcoal remains, installations related
to the ritual of libation, and small hearths. These features are concentrated near a travertine block, with traces of fire
on the walls, which had been used as an altar for animal sacrifice (Fig. 16-6, see US 530).
Most of the structures are linked to the burning of vegetable materials (Fiorentino and Solinas 2008). The organic
residue analyses carried out on the cooking pots have also shown their use for the preparation of herbal infusions
(Semeraro and Notarstefano 2013; Notarstefano 2012). These data are of great interest for the reconstruction of the
cultic practices; they are also relevant to the famous inscription of chresmoi, in honour of Apollo Kareios, found in the
1960s (see discussion in Notarstefano 2012: 132).
The research conducted in 2018 allowed us to recognize the northern and eastern limits of the sacrificial area,
excluding its continuation behind the sacred building (Fig. 16-7). The abandonment of the area in the 1'1 century CE
was also confirmed.
Legend
ashes
■ charcoal
thermic alteration
traces
terracotta pits
Building B
(Temple of Apollo)
J 0--=='m
Figure 16-7. Overall plan of the sacrificial area (excavations 2002-2006, 2018).
It is important to remember that this context dates back to the most ancient phases of the site, as it was verified
through the analysis ofa section in the western area (Semeraro 2012: 309-311). This section consists ofa large cut
made in the Byzantine age, to place a deposit of architectural marble blocks after the destruction of the sanctuary (Fig.
16-7).
The location of the important sacrificial area north of the temple has not yet been clearly explained. Its placement
was evidently dictated by choices made in the Hellenistic age, before the construction of temple B. However, it is
important to point out that use of the area continued even after the construction of the Temple of Apollo.
The rituals performed in the sacrificial area attest to some peculiarities, such as their chthonic character, and the
presence of strong references to the world of vegetation; these can be ascribed to the permanence of local traditions in
the nature of the Macedonian settlement. The peculiar epiclesis ofKareios associated with the polyadic deity is linked
to the pre-Hellenic Phrygian period of occupation (Semeraro 2012).
Also notable is that no trace of a monumental altar has yet been found in the area in front of the temple (Semeraro
2016b). The research conducted in 2018 and 2019 in this area permitted us to discard the hypothesis that a
monumental altar must have existed here, confirming the peculiarity of this sacred context.
Further data useful for understanding the layout of the sanctuary come from the investigations in the area of the
Nymphaeum (Fig. 16-4, nos. 8-9). A large fountain was built during the Severan age on the edge of the upper terrace,
radically changing the layout of the sanctuary, and creating an imposing separation from the lower terrace, where a
Hierapolis of Phrygia 2018- 2020 217
marble theatron was built in the Augustan and Julio-Claudian phases (Semeraro 2016a). Recent research has yielded
important results. 6 First, a second pool at the front has been identified, clarifying the water distribution system (Fig.
16-8b). It was also possible to verify the creation of the system of porticoes enclosing the area in front of the
Nymphaeum. In addition, during the same period, the theatron was restructured in a completely renovated space.
Previous research showed that the upper rows of seats had been removed (Semeraro 2016a). Despite the size
reduction, the theatron is still the dominant element of the lower terrace, placed at a point that affects the location of
the pool in front of the Nymphaeum.
Figure 16-8. Reconstructive plan of the Nymphaeum: (a) main basin; (b) location of the second basin; porticoes.
This vast reorganisation of the area cannot be separated from ritual motivations that reflected remarkable
innovation in the modes of worship. Whereas temple A was rebuilt for the third time on the upper terrace, the new
arrangement of the lower terrace was now focused on the complex of monuments constituted by the Nymphaeum-
theatron- stoai, a set of structures dedicated to specific functions, though these are not yet entirely clear.
In this new phase, water played a fundamental role, but equally central was the area in front of the theatron, the
only space where it is possible to hypothesize that ritual actions (performances?) might have taken place.
On the other hand, the theatron and stoai seem to be connected to rituals for varied levels of participation by the
audience of devotees; a small number could find a place on the marble seats of the theatron, while a larger number of
spectators could be housed in the side porticoes. This viewing area arrangement is connected to the upper terrace
6
Carried out by the University of Messina, under the leadership of Lorenzo Campagna, with collaborators Marta Venuti and
Alessio Toscano Raffa.
through the narrow travertine staircase, also functional as a route available to a limited number of participants
(Semeraro 2016b). The whole system seems designed to feature ceremonies that included a dimension of mystery that
could be the key to understanding the new layout of the sanctuary. This interpretation could be supported by the rise
of interest in mystery cults in Asia Minor during this period (and the rise of interest in mystery cults in the
Hierapolitan context in particular), which cannot be separated from the important role played by the cult of the "Great
Mother" (Semeraro 2016a).
Architectural Relationships with Seismic Phenomena
The most recent research (2018-2019) has concerned the study of seismic phenomena. Particularly important is the
identification of evidence for earthquakes in late antiquity, recognized in the stratigraphy, thus increasing the already
abundant documentation related to seismic activity in the region (for an overview see Kumsar et al. 2016; Marabini
and Scardozzi 2015).
Recent excavations have highlighted new
elements in the Sanctuary of Apollo, permitting us
to reconstruct the relationship between the
construction of the three main buildings and the
underlying fault line (Semeraro 2014). This work
recognized the reality of a "sacralization" of the
natural phenomenon that characterizes the whole
area and the nearby Ploutonion (Fig. 16-9). The
fracture below the sanctuary is the same one that
created the "sacred cave" in the Ploutonion and the
nearby rooms intended for the ritual practices of
incubation, according to the interpretation proposed
by D' Andria (D' Andria 2018a).
The south portico ensures the connection
between the two areas. In fact, only the western wall
fully closes this building, while the central and
eastern areas feature a series of travertine pillars that
allowed persons coming from the south to enter into
the sanctuary (Fig. 16-6).
The excavation carried out in the central area
(Fig. 16-4, see no. 4; Fig. 16-6) revealed a level
dating to the Byzantine age ((jlh century CE). A wall
(4210) inside the portico was rebuilt with a massive
reuse of building materials, such as a fragment of a
marble statue, probably representing a juvenile
· •i Apollo, e.g., the Sauroctonos (Fig. 16-l0a-b) .
Below the Byzantine age deposits there is clear
evidence of a fracture in the ground due to a seismic
event (Fig. 16-11 ). The position of one of the
travertine pillars (at the rear part of the portico) is
---="==='•' - - •"==="m also attributable to a structural failure caused by
seismic activity. Since the remaining part of the
UTM /WGS 114 zone !SN
Figure 16-9. Fault systems across the sanctuary of structure is still standing, the collapse of the column
Apollo and the Ploutonion. is to be connected to a greater "fragility" of the
subsoil at this point, due to the presence of the
previous fracture discovered by our excavation. The seismic event that caused the collapse of the column probably
occurred after the phase of use of the rooms in the porch area (6 th century CE).
The entire area of the sanctuary was affected by structural fractures and crevices, more or less intensely, which
were created during the life of the sanctuary, reflecting the peculiar fragility of the subsoil. One of the objectives of
the present research is to understand the seismic dynamics, through geophysical prospecting carried out by the
University of Padua (Valluzzi et al. 2019). 7 Nevertheless, some fractures were present and quite visible at the time of
construction of the sanctuary. The data that emerged from this study of the foundation structures of the Nymphaeum
are very important to our understanding of the area.
Research carried out inside the Nymphaeum pool and in the north area highlighted the path of the fault that
diagonally crosses the building (Fig. 16-9). It is evident that it was formed due to the same seismic events that caused
7
This research is headed by M .R Valluzzi. Geophysical prospections are under the direction of Rita Deiana (University of Padua).
a b
Figure 16-10. Fragment of a statue from the north portico: a) during the excavation; b) after the first restoration.
the collapse of architectural elements within the

fracture itself. Nevertheless, as noted above, an
opening in the ground already existed at the time of
construction of the foundations that support the
building. In fact, the arches built to the north and
south in effect "cross over" the fracture (Fig. 16-
12).
The function of the arches was to give stability
to the structures above. Proof of the relationship
between fault and the foundation of the arches can
be found in the excavation outside the Nymphaeum
(Fig. 16-4, see no. 9). This highlighted the
continuation of the arched structure to the north. It
should be pointed out that the arched structures on
the north side belong to a 1st century building, the
foundations of which were reused for the
construction of the Nymphaeum. The system of
arched foundations, therefore, dates to the time of
the creation of the sanctuary. Comparable structural
solutions were adopted for the construction of
temple A, also existing above an opening in the
ground (Semeraro 2007: 199 ff., Fig. 33 [US 436],
Figs. 44--45). Furthermore, the same construction
solution was also used in the Ploutonion area
(D' Andria 2018a: 104, Fig. 12). These observations
allow us to hypothesize that the architects of the
Hellenistic age adopted the arched solution not only
to "overcome" the fractures, but also because they
were aware of the role played by arched structures
in ensuring the building's resistance to seismic
stress.
This research has enabled us to understand how
the problems of continuing seismic events were
solved, allowing buildings to withstand numerous
earthquakes. These findings provide a basis for
addressing the theme of seismic-based architectural
Figure 16-11. Seismic fracture in the South portico. measures and coexistence with earthquake-prone
reg10ns.
Figure 16-12. Nymphaeum, north side: foundation arch above the fault.
The Phrygian Settlement
An important discovery concerns the identification of the Phrygian phase dating to the Iron Age, never
documented at the site before this research. A necropolis, dating to the 8th- 7ili century BCE, has been explored through
extensive excavations over the last three years.
The first traces of this phase were identified in 2016 thanks to emergency excavations conducted by the Museum
of Denizli, at the north entrance to the archaeological area (Fig. 16-1, see no. 1). 8 This work led to the discovery of a
series of circular structures made of stones (Fig. 16-13a-b), together with two sets of objects related to cremation
burials. Earthenware pots contained the human remains, which were accompanied by small objects.
In the following years, exploration continued under the auspices of the campaigns conducted by the present
archaeological mission in September and October of each year. 9 A large area was opened to the east of the circular
structures, permitting us to acquire interesting topographic data. The excavation brought to light a roadway and a wall
structure that functions as a boundary to the funerary space (Fig. 16-13a and c).
Even more interesting is the discovery, in 2018, of the remains ofa funeral pyre (Fig. 16-14a); the remains of the
wooden trunks used for cremation were still preserved. Next to the pyre, a pit full of charcoal and ashes was
identified. Palaeobotanical analyses 10 permitted us to identify the fuel used in these contexts: fig for the pyre and pine
wood for the pit. Since no other elements have been identified in the pit, we must hypothesize a use related to the
ritual of cremation, probably the burning of scented essences to counteract the negative effects of the fumes from the
fire.
In 2019 the excavation of a unique monument identified in the southwest comer of the area (Figs. l 6- l 3a, l 6- l 3c,
and l 6- l 4b) allowed us to further expand the data related to the spatial organisation of the area. The monument is a
rectangular structure, made with large travertine blocks, very crumbly and irregular on the surface. The excavation
8 Thanks to Huseyin Baysal, former Director of the Museum ofDenizli and to the archaeologists responsible for the excavation:
Elvan Altmta~, Meral Tarhan, and Birgul Camoglu Gunaydm. A full report is forthcoming.
9 The team of collaborators includes Vito Giannico (responsible), Raffaele Rizzo, and Giuseppe Di Canio.
°
1
Conducted in the Paleobotany Lab of the University ofSalento. I thank prof. G. Fiorentino and M. Primavera for the preliminary
data.
r--- 2016 ---..... 2019

1
' .....
/' .,, road
.
m• /
US61
lf ,~ /
i
i
i'
i /
!
/
i
i
; Legend
i Metamorphic rock Marble
I IIIILimestone - Shale
i
I . .River pebble
- Fired clay
ravertine
,.;;-::\i1<·{Modem trenches /·--- 6
N
IIIIIIConglomeratc rock /
6m
Figure 16-13. a) Phrygian necropolis: general plan; b) Eastern area of the Phrygian necropolis (2016 excavation);
c) Western area of the Phrygian necropolis (2018-2019).
Figure 16-14. a) Phrygian necropolis: pyre; b) Rectangular monument: collapse.

O 6 cm
O 3 cm
~
Figure 16-15. Phrygian Pottery: 1) jar with geometric decoration; 2) cooking pot; 3) small mug; 4) cup.
focused on the study of the internal collapse, in order to understand the dynamics that led to the destruction of the
structure (Fig. 16-14b). This analysis supports the hypothesis that the collapse should be ascribed to a seismic event.
The position of block-fall in fact suggests a collapse due to movement that affected the entire structure. If further
analyses in the upcoming campaigns confirm this hypothesis, we would be faced with a rare archaeological
documentation of an earthquake in the Iron Age. The preliminary study of ceramics seems to indicate a chronology
between the 8th and 7th century BCE, although a few fragments that suggest a more recent period (6th century BCE)
cannot be excluded.
The main shapes of Phrygian pottery are described in Fig. 16-15. The large jar with linear decoration (Fig. 16-15.1)
is one of the oldest artefacts from the site (for similar shapes in the Middle Iron Age, see Ytlmaz 2016: 71, Fig. 30.6).
Numerous sherds belong to pots used as cremation urns. Made ofmicaceous clay, poorly purified, often blackened
outside, they are characterized by globular walls and everted lip (Fig. 16-15.2). Small mugs (Fig. 16-15.3) and
hemispherical cups are also attested both in refined clay and in a coarse fabric similar to cooking pots (Fig. 16-15.4).
Based on a preliminary examination, the pots have typological comparisons with Middle Phrygian ceramics (see
e.g., Henrickson 2007: 194 f.). The data are still under study, and physical anthropological analyses, although planned,
are not yet available. The identification of comparisons for the type of structures attested in the Iron Age necropolis is
also quite difficult because of the complete lack of funerary evidence dating to this period in the territory near
Hierapolis and in the Lykos Valley.
Due to the circumstances of discovery, it is unclear whether the circular structures were covered by a mound,
although such a circumstance is very probable. Therefore, we should envision the presence of an earthen mound; the
absence of visible traces is likely due to the action of run-off and erosion by atmospheric agents.
The discovery of the Iron Age necropolis sheds new light on the settlement before Macedonian colonisation, about
which archaeological data are very scarce. A Phrygian settlement has been identified and excavated in nearby
Laodikeia, where the entire sequence of phases back to the Chalcolithic is known ($i~ek 2017: 6-7). Some
structures, perhaps belonging to the settlement, date back to the Iron Age, as do numerous fragments of the typical
Phrygian pottery with geometric decoration, attested at Hierapolis only by a few specimens.
Other contexts related to the Iron Age in the territory are known thanks only to surface surveys (Castrianni and
Scardozzi 2016), while stratigraphic data regarding the materials related to the various phases and individual contexts
are lacking.
At Hierapolis materials comparable to those found in the necropolis come from the excavation in the medieval
necropolis in the area of the North Agora (Fig. 16-1, see no. 7) (D'Andria 2019b: 487). These residual fragments,
however, represent useful clues connected to the presence of an Iron Age context.
Studies and Research
The materials from the 2018- 2019 excavations were the subject of preliminary cataloguing during these fieldwork
seasons, and this study continued during the 2020 campaign. The analyses of the archaeozoological finds from the
Ploutonion11 provided important results, allowing a better understanding of the rituals practiced in the sanctuary. Very
interesting are the data from the preliminary study (De Grossi Mazzorin and Minniti 2021 ), such as the identification
of numerous bird remains that recall the practices described by Strabo (birds thrown into the sacred cave, see D'Andria
2013). Also noteworthy is the presence of species related to chthonic cults (dogs, snakes, and rare specimens offish).
Among the research projects on individual classes of materials is the work on the sculptures of the Sanctuary of
Apollo, started in 2017, with the purpose of creating a systematic edition of the sculptures found in the years 2000-
2011. 12 The study of this group of materials (151 fragments), however, is particularly difficult due to the very
fragmentary state of conservation.
Most of the sculptures adorning the sacred complex were in fact broken during the systematic destruction of part of
the sanctuary (in particular temple B, dedicated to Apollo) in the Proto-Byzantine phase (Semeraro 2007). Emblematic
is the case of a statue of Athena (Fig. 16-16a), found in very small fragments in the area in front of temple B. The
preliminary study of the 54 fragments permitted us to recognize that it fit a statue type of Athena Parthenos (see the
statue from Pergamon now in Berlin: LIMC, s.v. Athena, vol. II, n. 230, p. 978) and to reconstruct its larger than life
size dimensions.
Numerous sculptural fragments were reused in later structures, such as the already published torso of Eros, a copy
of an original by Lysippus (Pellino 2012), and the above-mentioned fragment of a statue depicting a juvenile Apollo,
probably the Sauroctonos type, discovered in 2019 (Fig. 16-10).
The assemblage also includes the discoveries in the first phase of the research; 165 statues come from the
sanctuary area. Although none of the sculptures are in their original position, the large number of fragments can help
us to imagine, although only partially, the furnishings that decorated the largest city sanctuary and the related
euergetic activities, well evidenced by the numerous inscriptions. In addition, a study of the materials from the
11
Curated by Jacopo De Grossi Mazzorin and Claudia Minniti (University of Salento).
12
Under the supervision ofKatia Mannino (University of Salento). The research is also conducted by Vincenzo Ria, as a doctoral
thesis.
Hellenistic and Roman periods was carried out with the aim of constructing an atlas of ceramic productions, which
includes the main ceramic classes attested at Hierapolis. 13
E
A B
G
Particularly interesting is the study and
archaeometric analysis of the Palaeochristian fusiform
C
unguentaria (Fig. l 6- l 6b), one of the most challenging
projects developed in recent years (Aquilino 2019).
Found in urban, residential, funerary, and religious
milieux, these vessels evoke contexts of the ritual
use of holy water and blessed anointing oil, but also
healing or cosmetic practices. The study, focused on H
the identification of fatty acids and biomarkers using D
gas chromatography-mass spectrometry, aims at the
'11
characterisation of the unguentaria contents and the
reconstruction of the production phases of the
contents, a balm. K L
Molecular biomarkers are treated like any other
archaeological indicator, in order to obtain
information about the composition and the b
production cycle of the medicines, their Figure 16-16. a) Fragments of the statue of
formulation, and the treatment reserved for every Athena from the sanctuary of Apollo;
single ingredient. The use of a wide range of resins b) Byzantine unguentaria.
(Burseraceae spp.; Pinaceae spp.; Styrax officinalis;
Pistacia Lentiscus), plant oils (Ricinus communis; Olive oil, seeds oil), and animal fats, was verified. The techniques
of cold maceration and enjleurage were used for the production of the remedies that could have a fluid, or thick and
smooth, formulation. Some biomarkers allow hypothesizing the use of steam distillation. The biomolecular data were
also subjected to multivariate statistical analysis of hierarchical aggregation in order to establish the recipes. The
results were compared with medical and pharmaceutical literary sources (Aquilino et al. forthcoming).
Restorations
Between 2018 and 2020, the restorations of the Ploutonion were completed and those of the Nymphaeum of the
Sanctuary begun. Both projects have financial support from FIAT-Tofa~ and the Ko9 Foundation. In the Ploutonion
area, the project that began in 2017 continued, 14 aimed at the integrated restoration of the theatron and the creation of
visitor access. In 2019-2020, a path for people with motor disabilities was completed (Fig. 16-17).
In the area of the Nymphaeum, conservation work has begun on the wall structure with the removal of weeds and
cleaning of the marble architectural elements. Restoration work has also begun as well as an enhancement project
focused on the monument and on the arrangement of the marble blocks.
13
Research on ceramics is being conducted by Vito Giannico, Florinda Notarstefano, and Valentina Aquilino (University of
Salento).
14
The project is under the scientific responsibility of Francesco D' Andria (University of Salento), with a team of collaborators: Pio
Panarelli, Sara Bozza, Arif Duygun, and Kadir Ozel.
Figure 16-17. Overall view of the Ploutonion (2020).
Geomatics
During the 2018-2019 campaigns, the geomatics team carried out research with varied and new technologies
aimed at developing the 3D survey. 15 The overall three-dimensional model of the entire archaeological area of
Hierapolis (RU Polytechnic of Turin) was created. The most suitable detection technique for this 3D documentation at
the city-scale is photogrammetry from drones; in particular, the recent innovative availability on the market of a fixed-
wing drone model, with tilting camera and capable of acquiring oblique images (Ebee x with SODA 3D camera, by
SenseFly), has made this aim achievable. The model was generated thanks to an overall flight over the entire city
carried out in 2019. The drone is also equipped with an integrated GNSS----Global Navigation Satellite System, which
allowed us to experiment with direct photogrammetry algorithms, without using topographic support points. This
acquisition is very important. Indeed, the model provides a precise representation of the current state of Hierapolis'
cultural heritage.
CONCLUSIONS AND OUTLINE OF UPCOMING WORK
The main research topics discussed in the previous paragraphs will be the subject of further study in the next
excavation campaigns. In particular, we plan to extend the exploration of the oldest phases of the Sanctuary of Apollo,
through the investigation of the vast area south of the Nymphaeum.
15
Work was carried out under the responsibility of Antonia Spano, in collaboration with Filiberto Chiabrando, Lorenzo Teppati
Lose, and Giulia Sammartano (Polytechnic of Turin).
The investigation in this area will also allow the reconstructing of the modifications related to the insertion of the
Nymphaeum in the Sanctuary. The overall layout of the area can also be better understood through the exploration of
the lower terrace, where the two stoai and the method of connection with the plateia remain to be unearthed (Fig. 16-
8). The exploration of the lower terrace, however, is quite problematic because of the geological peculiarities of the
area: a thick layer of travertine which covers the large space in front of the theatron and up to the plateia, below
which the water from the underground springs flows.
Insights on archaeoseismological aspects will undoubtedly be at the centre of the research that will continue
through collaboration with geophysicists and geologists. The objective ofresearch on the effects of seismic events is a
key aspect of the research in the next years also through the review of the documentation produced by the previous
research carried out at Hierapolis. One of the most important aspects of this research lies in the opportunity to
understand how the Hierapolitan community lived for centuries with the risk of recurrent earthquakes, developing
reaction strategies, but also preventing seismic-based destruction phenomena. The theme of archaeoseismology will
also be addressed in relation to the Phrygian necropolis, in order to deepen our understanding of the geological nature
of the context and to better understand the collapse dynamics of the Iron Age "rectangular building" discussed above.
Another long-term objective of the research is the archaic phase survey. In addition to the exploration of the
necropolis, the program for the next campaign will also provide a series of sondages in the North Agora in order to
clarify the possible presence of an archaic stratification. Surface and geophysical exploration aimed at locating the
Phrygian settlement will continue as well, in order to verify the hypotheses formulated so far.
ACKNOWLEDGEMENTS
I wish to thank the General Directorate for Museums and Cultural Monuments of the Turkish Ministry of Culture,
in the person of the General Director Murat Giiriil and Gokhan Y azg1, of the Deputy Directors Ali Riza Altunel and
Y ahya Co~kun, and the Division of Excavations, Melik Ayaz and Nihal Metin. For the support of the work carried out
in Hierapolis in the years 2018- 2020 I thank the Directors of the Denizli Museum Hilseyin Baysal (2018), Cevdet
Seviw; (2019), and Nail Uyar (from 2019), the Directors of the Culture Office of Denizli province Gulsiim <;elebi
(2018), Mustafa Kaptan (2019), and Turban Veli Akyol (2020).
Numerous researchers and students from the University of Salento (direction), Pamukkale (deputy direction),
Florence, Padua, Pisa, Rome-Sapienza, Messina, Polytechnic of Turin, Amasya, Istanbul, and CNR participated in the
work conducted by the Mission between 2018 and 2020. I thank the Deputy Directors inci Tiirkoglu (2018) and
Co~kun Da~bacak (since 2019), and the officials of the Ministry of Culture Birgiil Camoglu Giinaydm (2018), Omer
Severoglu (2019), and ~ebnem Gulliisay (2020) for their collaboration.
As always, the mission's activity was supported by the University of Salento, and by the contribution of the
Ministry of Foreign Affairs and International Cooperation. In recent years, the mission obtained the contribution of the
PRIN 2015 project Prot. 2015Y8X3KT (Archaeology of urban landscapes in Asia Minor between late Hellenism and
the Byzantine period. Multidisciplinary approaches to the study of Hierapolis of Phrygia) funded by the Ministry of
University and Research. Essential support for the restoration activities was ensured by FIAT-Tof~ and K~-Vakf1,
whom I thank in the people of the CEO Cengiz Eroldu and the Director Omer M. Kos;.
For the sponsorship of the activities, I also thank Quarta Caffe (Lecce), and Komiirciloglu Mermer (Denizli).
Photographic references of materials: Archives of the Archaeological Mission in Hierapolis. For the collaboration in
image processing, I would like to thank Dr. Vito Giannico, Valentina Aquilino, and Vincenzo Ria.
REFERENCES CITED
Ahrens, Sven and Rasmus Brandt. 2016. Excavations in the North-East Necropolis of Hierapolis 2007- 2011. In
Hierapolis di Frigia VIII 1- 2, F. D'Andria, M.P. Caggia, and T. Ismaelli, eds., 395-414. lstanbul: EgeYaymlan.
Anguissola, Anna and Silvana Costa. 2020. I sarcofagi de/la necropoli nord di Hierapolis in contesto: due casi di
studio. Studi Classici e Orientali 66: 301- 328. DOI 10.12871/978883339362920
Aquilino, M. Valentina. 2019. JI contributo de/le analisi dei residui organici a/lo studio Junzionale dei contenitori per
profumi, unguenti e farmaci, Ph.D. thesis (Universita del Salento, Dipartimento di Beni Culturali Dottorato in
Scienze Del Patrimonio Culturale, XXXI Cicio AA. 2017/2018).
Aquilino, Valentina, Grazia Semeraro, and Florinda Notarstefano. Forthcoming. Late Roman Unguentaria from
Hierapolis of Phrygia. Molecular Evidences of the Actual Use. In Unguentarium. A Terracotta Vessel form in the
Hellenistic, Roman and Early Byzantine Mediterranean. An International Symposium, izmir, Dokuz Eyliil
Universitesi, 17-18 May 2018.
Arthur, Paul R. 2006. Byzantine and Turkish Hierapolis (Pamukkale): An Archaeological Guide (Hierapolis
Archaeological Guides). istanbul: Ege Yaymlar1.
Campagna, Lorenzo. 2018. JI Ninfeo dei Tritoni. Hierapolis di Frigia XL istanbul: Ege Yaymlar1.
Castrianni, Laura and Giuseppe Scardozzi. 2016. Insediamenti protostorici del territorio di Hierapoli di Frigia:
l'Hoyiik Tepe (Akkoy) e l'abitato di Can Pmar. In Hierapolis di Frigia VIII(l), F. D'Andria, M.P. Caggia, and T.
Ismaelli, eds., 35- 51. istanbul: Ege Yaymlar1.
D'Andria, Francesco. 2001. Hierapolis of Phrygia. Its Evolution in Hellenistic and Roman Times. In Urbanism in
Western Asia Minor. New Studies on Aphrodisias, Ephesos, Hierapolis, Pergamon, Perge and Xanthos, D. Parrish
ed., Journal ofRoman Archaeology Supplementary Series 45: 96- 115.
-. 2013. Il Ploutonion a Hierapolis di Frigia. Istanbuler Mitteilungen LXIII: 157-217.
-. 2016- 17. Hierapolis alma Philippum. Nuovi scavi, ricerche e restauri nel santuario dell'Apostolo. Rendiconti
Pontificia Accademia Romana di Archaeologia 89: 129-202.
- . 2017. The Sanctuary of St Philip in Hierapolis and the Tombs of Saints in Anatolian Cities. In Life and Death in
Asia Minor in Hellenistic, Roman and Byzantine Times. Studies in Archaeology and Bioarchaeology, J.R. Brandt,
E. Hagelberg, G. Biomstad, and S. Ahrens, eds., 3-18. Oxford: Oxbow Books.
-. 2018a. The Ploutonion ofHierapolis in Light of Recent Research (2013-17). Journal ofRoman Archaeology 31:
91- 129.
-. 2018b. Hierapolis di Frigia. Una agro-town medio-bizantina? In "Di Bisanzio dirai cio che e passato, cio che
passa e che sara. " Scritti in onore di Alessandra Guiglia, S. Pedone and A. Paribeni, eds., 153- 172. Roma: Bardi
Edizioni.
-. 2019a. The Ploutonion ofHierapolis in the 5th-7th c. A.D. American Journal of Archaeology 32: 505-515.
- . 2019b. The Cult of Cybele in Hierapolis of Phrygia. In Phrygia in Antiquity: From the Bronze Age to the
Byzantine Period, G.R. Tsetskhladze, ed., 479-500. Leuven: Peeters.
D 'Andria, Francesco, Maria P. Caggia, and Tommaso Ismaelli, eds. 2012. Hierapolis di Frigia V Le attivita de/le
campagne di scavo e restauro 2004-2006. istanbul: EgeYaymlan.
- . 2016. Hierapolis di Frigia VIII 1-2. Le attivita de/le campagne di scavo e restauro 2007- 2011. istanbul: Ege
Yaymlar1.
D'Andria, Francesco and Tullia Ritti. 1985. Hierapolis, Scavi e ricerche II. Le sculture de/ teatro. I rilievi con i cicli
di Apollo e Artemide. Archaeologica 54. Roma: Giorgio Bretschneider.
D'Andria, Francesco and Maria Pia Rossignani. 2012. La stoa-basilique de Hierapolis de Ph,ygie. Architecture et
contexte urbain. In Basiliques et agoras de Grece et d'Asie Mineure, L. Cavalier, R. Descat, and J. Des Courtils,
eds., 127-152. Memoires, 27. Bordeaux: Ausonius Editions.
D 'Andria, Francesco, Annapaola Zaccaria Ruggiu, Tullia Ritti, Giovanni Battista Bazzana, and Remo Cacitti. 2005-
2006. L'iscrizione dipinta con la preghiera di Manasse a Hierapolis di Frigia (Turchia). Rendiconti Pontificia
Accademia Romana di Archaeologia 78: 349-449.
De Grossi Mazzorin, Jacopo and Claudia Minniti. 2021. Bird and Other Animal Sacrifice in the Ploutonion of
Hierapolis, Phrygia (Turkey): Some Results from Two Votive Deposits. In Roman Animals in Ritual and Funerary
Contexts, Proceedings of the 2nd Meeting of the Zooarchaeology of the Roman Period Working Group, Basel 1'1-
4th February, 2018, S. Deschler-Erb, U. Albarella, S. Valenzuela Lamas, and G. Rasbach, eds., 39-52. Kolloquien
zur Vor- und Fri.ihgeschichte 26. Frankfurt: Romano-Germanic Commission (RGK).
Fiorentino, Girolamo and Francesco Solinas. 2008. Micro-stratigraphical and Archaeobotanical Approaches to
Investigating Ash Deposits in the Apollo Sanctuary in Hierapolis. In l. Uluslaras1 Antik Donemde Kehanet
Apollon'un Anadolu Kultleri Sempozyum Bildirileri (1 st International Symposium on Oracle in Antiquity and the
Cult of Apollo in Asia Minor), l 7-20th August 2005, Ege University, izmir. Arkeoloji Dergisi XII(2): 89-98.
Hemickson, Robert C. 2007. Phrygian Pottery. In Friglerin gizemli uygarhg1llhe Mysterious Civilization of the
Phrygians (Exhibition Catalogue),~- ~enttirk, ed., 189-200. istanbul: Yap1 ve Kredi.
Ismaelli, Tommaso. 2009. Architettura dorica a Hierapolis di Frigia. Hierapolis di Frigia III. istanbul: Ege Y aymlar1.
LIMC"" Lexicon Iconographicum Mythologiae Classicae (1981-)
Kumsar, Halil, Omer Aydan, Celal ~im!;,ek, and Francesco D'Andria. 2016. Historical Earthquakes that Damaged
Hierapolis and Laodikeia Antique Cities and Their Implications for Earthquake Potential of Denizli Basin in
Western Turkey. Bulletin ofEngineering Geology and the Environment 75(2): 519- 536.
Marabini, Stefano and Giuseppe Scardozzi. 2015. La ricerca geoarcheologica a Hierapolis. In Atlante 2015aaNuovo
at/ante di Hierapolis di Frigia. Cartografia archeologica della citta e de/le necropoli. Hierapolis di Frigia VII, G.
Scardozzi, ed., 227-268. istanbul: Ege Yaymlar1.
Notarstefano, Florinda. 2012. Ceramica e alimentazione. L 'analisi chimica dei residui organici nelle ceramiche
applicata ai contesti archeologici. Beni Archeologici-Conoscenza e Tecnologie 10. Bari: Edipuglia.
Pellino, Giuseppe. 2012. Eros in area Apollinis. Una copia dell'Eros di Lisippo tipo Tespie e l'arredo scultoreo del
santuario di Apollo. In Hierapolis di Frigia V, F. D ' Andria, M.P. Caggia, and T. Ismaelli, eds., 359-372. istanbul:
Ege Yaymlar1.
Ritti, Tullia. 1985. Hierapolis. Scavi e Ricerche I. Fonti letterarie ed epigrafiche. Archaeologica, 53. Rome: Giorgio
Bretschneider.
- . 2017. Storia e Istituzioni di Hierapolis. Hierapolis di Frigia IX. istanbul: Ege Yaymlari.
Scardozzi, Giuseppe, ed. 2015. Nuovo at/ante di Hierapolis di Frigia. Cartografia archeologica de/la citta e de/le
necropoli. Hierapolis di Frigia VIL istanbul: Ege Yaymlan.
Semeraro, Grazia. 2007. Ricerche archeologiche nel santuario di Apollo (Regio VII) (2001- 2003). In Hierapolis di
Frigia I. Le attivita de/le campagne di scavo e restauro 2000-2003, F. D'Andria and M.P. Caggia, eds., 169-209.
istanbul: Ege Y aymlar1.
-. 2012. Ricerche nel Santuario di Apollo. In Hierapolis di Frigia V, F. D ' Andria, M.P. Caggia, and T. Ismaelli,
eds., 293- 324. istanbul: Ege Yaymlan.
- . 2014. Archaeology of the Cult in the Sanctuary of Apollo in Hierapolis. In Frail Meandro e il Lico. Archeologia e
storia in un paesaggio anatolico. Atti del W orshop Intemazionale, Universita 'La Sapienza- Roma', 30 Marzo
2012, F. Guizzi, ed., Scienze dell'Antichita 20.2: 11- 29.
- . 2016a. Ricerche nel Santuario di Apollo (2007-2011). In Hierapolis di Frigia VIII 1-2, F. D'Andria, M.P. Caggia,
and T. Ismaelli, eds., 191- 220. istanbul: Ege Yaymlar1.
- . 2016b. The Sanctuary of Apollo in Hierapolis: Building Activities and Ritual Paths. In Ancient Quarries and
Building Sites in Asia Minor. Research on Hierapolis in Phrygia and Other Cities in South-western Anatolia:
Archaeology, Archaeometry, Conservation, T. Ismaelli and G. Scardozzi, eds., 777- 785. Bari: Edipuglia.
- . 2017. Phrygia Hierapolis' inde iiretim arkeolojisi: Yeniveriler ve ara~t1rma perspektifleri/Archeologia della
produzione a Hierapolis di Frigia: Nuovi dati e prospettive di ricerca. In Economia, Produzione e societa. Atti
dell'ottava Edizione del Convegno "Contributo italiano a scavi, ricerche e studi nelle missioni archeologiche in
Turchia." Journal ofArchaeology and Art, Arkeoloji ve Sanat, 154, Ocak- Nisan 2017: 97-110.
Semeraro, Grazia and Florinda Notarstefano. 2013. Integrated Methodologies for the Study of Cultural Contexts. Case
Studies from the Mediterranean Area. Malta and Hierapolis. In Archeologia dei luoghi e delle pratiche di culto.
Atti del convegno, Cavallino, 26-27 gennaio 2012, L. Giardino and G. Tagliamonte, eds., 57-68. Bari: Edipuglia.
~i~ek Celal. 2017. Urban Planning of Laodikeia on the Lykos in the Light of New Evidence. In Landscape and
History in the Lykos Valley. Laodikeia and Hierapolis in Phrygia, C. ~im~ek and F. D 'Andria, eds., 1-51.
Newcastle upon Tyne: Cambridge Scholars Publishing.
Valluzzi, Maria Rosa, Claudia Marson, Sabrina Taffarel, Matteo Salvalaggio, Rita Deiana, and Jacopo Boaga. 2019.
Structural Investigations and Modelling of Seismic Behaviour on Ruins in the Monumental Area of Hierapolis of
Phrygia. RILEM Bookseries 18: 1849-1857, doi: 10.1007/978-3-319-99441-3_198
Yilmaz, Mehmet A. 2016. Iron Age Pottery. In Archiiologische Forschungen am Oymaaga<; Hoyiik/Nerik 2011-2015,
R. Czichon et al., eds. Mitteilungen der Deutschen Orient-Gesellschaft zu Berlin 148: 68-74.
Zaccaria Ruggiu, A. 2012. Un quartiere residenziale: l 'insula 104. In Hierapolis di Frigia V, F. D'Andria, M.P.
Caggia, and T. Ismaelli, eds., 419-442. lstanbul: Ege Yaymlar1.
- . 2019. Le abitazioni dell'insula 104 a Hierapolis di Frigia. Hierapolis di Frigia XII. lstanbul: Ege Yaymlar1.
CHAPTER SEVENTEEN
RECENT DISCOVERIES AT KOMANA (2018-2020):

THE LATE IRON AGE/EARLY HELLENISTIC LEVELS
D. BURCU ERCiYAS, AYSE BATMAN, AND ASLIHAN C::iViLiDAG
The ancient site of Komana is located in the central Pontus region, next to the Iris River (Ye~ihrmak) in the province
of Tokat (Fig. 17-1). Its strategic position in the eastern extension of the fertile Dazimonitis plain on the road from
ancient Dazimon (Tokat) to Neocaesareia (Niksar) benefited the site economically through its extensive life span. The
roads reaching the Black Sea coast through Amaseia (Amasya) in the north (Ramsay 1890: 158), crossroads for many
important routes, and through Neocaesareia in the Lycus Valley in the northeast (Hogarth and Munro 1893: 734),
provided Komana access to the sea, while a possible route towards Sebasteia (Sivas) (Wilson 1960: 242) in the south
connected the site with Cappadocia and Cilicia further south. In addition to its well-connected position within an
interrelated settlement system, Komana prospered due to the rich resources available in terms of agriculture and
metallurgy.
Amaseia•
Alacahoyuk• • Neocaesareia
Eskiyapar• K1zkayas1• 0 Komana Pontica
Gordione Bogazkiiy•
U~akl1 Hoyuk••Tavium eSebasteia
Kaman Kalehoyuk •
Komana Cappadociae
Kin1k Hoyuk•
+
~
0
100 200
Figure 17-1. Location ofKomana on map.
The site was most renowned under Mithradates VI (120-63 BCE), as it was a prominent sanctuary dedicated to the
goddess Ma, but the origins of this sacred site extended well before the Hellenistic period, possibly dating back to the
Hittites (Casabonne 2003). The goddess had warrior attributes with two epithets: aniketos and nikephoros, and was later
associated with Enyo and Bellona when the cult was transferred to Rome (Strabo 12.2.3; for an anecdote from Sulla's
life, Plutarch Sulla 9.2).
Komana was unique due to the ancient character of the cult, predating the Hellenistic and Roman periods, and can
be considered to be comprised of Semitic, Hittite, Persian, Greek, and Roman characteristics. The administrative and
economic organisation at the sanctuary was also unique in that there was an appointed priest under the Mithradatids
who was "second in rank" to the king, and the economy was a temple-based one in which the revenues were collected
230 Chapter Seventeen: Erciyas et al.
at the temple to be redistributed. The temple was popular for the bi-annual festival known as the celebration of "the
exoduses of the goddess" (Strabo 12.3.32). Casabonne likens these festivals to Hittite traditions (Casabonne 2003: 281-
283). Strabo speaks extensively about Komana and describes the site as an emporion, possibly referring to the enlivened
trade during the festivals and describes it as a replica of another Komana in Greater Cappadocia.
There is an extensive bibliography on possible identifications for the cult and the goddess (Casabonne 2003, 2009;
Mutlu 2016), detailing its peculiar structure with hierodouloilhierai (Budin 2009), the significance of the site for the
Pontic royalty and as a temple state (Sokmen 2005; Ballesteros Pastor 2016), and its assimilation into the Roman
province (Olshausen and Biller 1984; Marek 1993; Mitchell 1993). However, the sources are strictly historical, with
very little archaeological evidence to support them. This has been the main motivation behind the project which was
developed at Komana in 2004.
ARCHAEOLOGY AT KOMANA
The first five years of the project (2004-2008) were dedicated to better understanding the landscape of Komana,
including the wider territory, and recording multi-period sites discovered during extensive surveys. At specific sites,
intensive and systematic collecting was also performed, but the primary goal of the survey was general reconnaissance.
These surveys enabled the team first of all to locate ruins in the immediate vicinity of, and belonging to, the sanctuary,
and to observe surface scatter patterns which might be helpful in identifying locations of specific sections of the site or
buildings; and secondly to gain a perspective on the multi-period settlement patterns within what was the Roman
territory ofKomana (Erciyas 2012). During the surveys, the number of Middle Byzantine sites far exceeded those dating
to the rest of the periods, due to their higher visibility of recognizable architectural remains; indeed, the excavations
later indicated a denser settlement throughout the Middle Byzantine period at Komana and the environs (Erciyas 2014;
Erciyas and Sokmen 2010; Tatbul and Erciyas 2019). In the course of doing the surveys only a handful of sites dating
to the Hellenistic period were discovered.
Figure 17-2. Hamamtepe excavations, aerial view.
Excavations at Komana began in 2009. The operation was mostly focused on a mound locally known as Hamamtepe
(Fig. 17-2). The mound is located next to the Ye~ihrmak on flat land. The modem road to Niksar passes by the mound,
splitting the overall site into two parts and paving over ruins which extend towards the terraced hills to the west. These
terraces were predominantly settled during the Byzantine, Sel9uk, and Ottoman periods. A number of Hellenistic period
graves are also present, one of the most prominent being the rock-cut grave frequently mentioned by travellers
(Anderson 1903; Cumont and Cumont 1906). The site is split once more by the river itself, and the modem road to
Recent Discoveries at Komana (2018-2020) 231
Almus to the east. The plain and hills east of the Y e~ihrmak, and the occasional valleys, bear hints of the Hellenistic
and Roman period at the site. The two villages on the east side, Balhdere and Doll-Uk, are crowded with architectural
spolia from Hellenistic and Roman times. The eastern hills were also necropoleis during the Roman period. Tumuli are
scattered across the tops of the hills on this side.
Figure 17-3. Hamamtepe excavation areas.
Hamamtepe is a rocky hill (consisting of metamorphic rock), and the bedrock rises up at different locations, which
necessitates a good understanding of the rather dynamic relief of the site. This is a major factor in stratigraphy, since
the use of bedrock in the different periods seems prominent. The archaeological work concentrated on four major areas,
but the most prominent data for the multi-layered settlement comes from the so-called HTP0l, the excavation area at
the centre of the hill (Fig. 17-3). The Hellenistic period finds mostly come from this central section.
Hamamtepe Chronology
The chronology ofHamamtepe can be presented as seven separate levels (Erciyas 2019).
1. The Ottoman Level (Rural Village Settlement) (17th - 18th Century CE)
The uppermost cover of the entire mound is a thin layer of dusty, dark brown soil which includes lots of irregular
small-sized grey stones, some ceramics, mostly Ottoman green glazed, some tiles, rarely coins, and vegetation.
Immediately under this soil are wall foundations (56-99 cm wide) of the Ottoman village settlement. The foundations
are made of one to two layers of the grey stones that frequently appear in the topsoil. At a few places, where a more
vigorous leveling was necessary, the foundations were built as tall as about 1 m or 4 rows of grey stones. The floors are
beaten earth; material in situ is rare, and spolia are re-used as bases for wooden posts supporting roofs. In the excavated
area we can recognise four large buildings, with two units in the south and two others to the north. The archaeological
contexts suggest that the Ottoman village was abandoned in the 18th century, when the buildings were cleared of the
daily objects and personal belongings.
2. The Dani~mend/Sel<;uk Phase (Fortified Rural Settlement) (12 th- 14th Century CE)
The Dani~end/Selc;:uk buildings are immediately under the Ottoman wall foundations. The wall foundations of the
buildings in this phase are built with large and varied irregular stones (67.78-91.97 cm). They are mostly l to 2 rows in
height, similar to the Ottoman foundations. The distinguishing character of these structures is the presence of tandzr
type ovens in the rooms. In addition to the ovens, there are storage pits, pithoi, rock-cut pits, and cesspits. The
architecture is suggestive of a production centre or workshop on the hill during the Selc;:uk period; the variety of the
finds together with production waste of various materials support this assessment.
Three doctoral dissertations have been completed on the archaeology of this period. Tatbul (2017) identified the
function of the Dani~mend/Selc;:uk buildings through the spatial distribution of archaeological and environmental data.
Er (2020) conducted provenance analysis on the glazed pottery from the Selc;:uk period in his thesis demonstrating traits
of local production, which was later supported by Karasu's thesis (2020), which dealt with the Selc;:uk period pottery
from a chronological and stylistic point of view and placed the site in a central position in terms of pottery production
and trade.
3. Middle Byzantine Level (Cemetery with Two Chapels) (1 Jfh-J 2th Century CE)
The hill was used as a cemetery during the Middle Byzantine period. Thus far 133 graves have been excavated. At
the centre of the cemetery two identical chapels were unearthed (Church A and B). These three-aisled small chapels
must have been briefly used during visits to the dead. Small ceremonies during the burial could have been held here as
well. Although small, they were highly ornamented with decorated interiors and exteriors. There were frescoes covering
the entire ceiling and upper parts of the walls, and tiles adorning the pillars or walls of the bema. Architectural terra
cottas were heavily used in accordance with the style of the period. The bronzes recovered in and around the churches
are further evidence of elaborate decoration in the church in the form of chandeliers, lamps, ceremonial crosses, and the
like.
4. The Early Byzantine Phase (Fortified Rural Settlement) (7th-8th Century CE)
Very little is known from this period in terms of artefacts. Underneath the cemetery there is a prominent building
level characterised by mortared walls standing as high as 2 min some places. Within the small rooms enclosed by these
walls are pithoi. The most prominent feature of this phase is surely the fortification wall surrounding the hill. It must
have been built around the 7th century CE but was repaired and reused during the Selc;:uk period.
5. The Roman Phase (2nd-3rd Century CE)
The Roman period at the site is only represented by a scarce number of coins, architectural spolia, three inscriptions
(also used as spolia), and some ceramics from contexts which could not be associated with the built enviromnent.
6. The Late Iron Age/Hellenistic Phase (4 th- 2nd Century BCE)
Partial walls disturbed by Early Byzantine walls and archaeological layers above and around the walls, with a fairly
mixed ceramic assemblage and other artefacts, characterise the Late Iron Age/Hellenistic phase. The structural remains
and archaeological finds from these layers make up the main subject of this chapter and are discussed below in detail.
7. The Late Cha/eolithic-Early Bronze Age (3000-2700 BCE)
Ceramics that could be dated to the Late Chalcolithic and Early Bronze Age have been found mixed with Late Iron
Age and Hellenistic ceramics in some layers excavated between 2017- 2020.
The Late Iron Age/Hellenistic Period Levels at Hamamtepe
Up to and including the Roman period, the artefacts were recovered in relation to building levels enabling a
reconstruction of the stratigraphy and the settlement plan of Hamamtepe during the 900 years of habitation at the site.
Prior to our recent work, the Early Byzantine levels had been the earliest known occupation at the site, with prominent
mortared walls and the fortification walls. While material culture which could be dated to the Roman and Hellenistic
periods was present in mixed contexts, mostly due to illegal trenches dug across the top of the mound, and in later fills,
in situ architecture had been missing. The 2018 field season revealed a new group of wall foundations just below the
Early Byzantine walls, including an increasing number of formerly unfamiliar ceramics and Hellenistic period coins
(Fig. l 7-4a-b). These walls also displayed a completely different orientation from the later four levels; while the general
layout favoured a north-south, east-west orientation at Medieval Komana, these new walls were laid in a northeast-
southwest direction. It had been assumed that the inhabitants of the upper four levels (Byzantine through Ottoman) each
had the knowledge of the previous period's building remains, and for the most part reused them, either as building
material or as foundations for their own buildings, resulting in a continuation of the same orientation across the mound.
Between the Early Byzantine and Hellenistic levels, though, there is a clear hiatus at least in this part of the mound.
Overall, the newly discovered walls, with their different orientation, a significant change in the colour, texture, and
matrix of the soil encompassing the walls, and a new ceramic assemblage, indicated that we were digging through a
new and formerly unknown level at the site.
The new level was observed in two trenches at the
northeast end of the excavated grids of HTP0 1. These
Hcllcoistic\Valls
two trenches (297/608 and 307/608) are adjacent to
each other, and although the given numbers of the
layers (T) for each trench do not appear to correspond
due to differences in how the contexts are configured,
they contain the same soil and artefact composition.
Excavations first started in 297/608 (10 x 10 m). In
2017, a new set of mortared walls appeared in the
eastern section of the trench, which resulted in the
extension of excavations into 307/608 (5 x 10 m).
These two trenches are also where excavation could
continue deeper due to the decreasing elevation of the
bedrock towards the north. Immediately under the
foundations of the Middle Byzantine period Church
B, a very hard layer consisting of pebbles, broken
pieces of small stones, shells, and a mixed ceramic
assemblage characterised the transition from the
Byzantine to the Hellenistic period (Tl0 in 297/608
and TS and Tl 7 in 307/608). This layer must have
been a fill brought in with the intention of levelling
the uneven surfaces around the church as a pre-
construction preparation. The layer extended to a
fairly large area from the west of Church B to about
20 m east of it. In 2020 an Egyptian scarab-shaped
stamp seal was discovered amid the pebbles (Fig. 17-
5). This steatite scarab bears the prenomen of
Thutmose III (14 79-1429 BCE), Menkhepherre.
Thutmose III of Dynasty XVIII had a very long reign,
and his prenomen frequently appears on seals found
all across the Mediterranean and the Near East long
Figure 17-4. a) Plan drawing of the Hellenistic walls with after his reign, as late as the Late Period (664-525
the Early Byzantine walls; b) Aerial view of the BCE) (Jaeger 1982; Peck 1987). Therefore, dating
Hellenistic walls with the Early Byzantine walls. these seals based on the prenomen has been
problematic. A further complication is the
identification of Menkheperre. The name could also
refer to Menkheperre the High Priest of Dynasty XXI;
it has also been used to recall the name of the god
Amun. With such complications it is difficult to date
this seal exactly; however, a later date within the span
of the use of this prenomen on seals seems more
viable for the Komana example, considering the
general characteristics of the layer in which it was
found. The ceramics from this layer still need to be
studied.
Underneath this fill a transitional deposit was
recognised in both trenches. This deposit is dark
brown to reddish in colour and includes a large
number of ceramics (T9, 10, 11 in 307/608 and Tl2,
14, 17 in 297/608). Within Tl2 in 307/608 a more
consistent level in terms of ceramic assemblages was
5cm revealed. This is also the level at which the previously
mentioned walls started to appear. These dry walls
have three in situ sections and are trapped between
Figure 17-5. Scarab-shaped stamp seal with Thutmose III the foundations of the mortared walls (Fig. l 7-4b). It
(1479-1429 BCE), Menkhepherre possibly is assumed that parts of the walls were removed when
dating to the Late Period (664-525 BCE). the foundation trenches for the later walls were dug.
Due to their fragmented state, it is difficult to recognise a plan, though the presence of a single room may be suggested.
The walls are not well preserved in general, as they have only one or two rows of stones left. Their construction is rather
clumsy. The stones are varied in size and irregular in shape, and they seem to have come from different sources indicated
by their colour variance (grey, red, white). Among these, the presence of a peculiar type of red stone is remarkable. This
red, chalky, and fragile stone was already observed to have been used in the construction of the Middle Byzantine
churches and the Early Byzantine fortification wall; a probable source was identified at some distance away in the
mountains to the east of the site. Their discovery in these early walls suggested that this stone was in use as early as the
Hellenistic period and that possibly the Byzantine period material originated from the Hellenistic buildings.
The disturbance caused by the later mortared walls on the earlier walls and the archaeological deposit is enormous.
In addition, excavating was difficult in these confining spaces. The division of the excavation into four unavoidable
spaces (due to mortared walls) also resulted in the creation of various layers; however, a consistency within the deposits
above and around the early walls was observed, despite slight changes, through layers 12 and 15. Large numbers of
ceramics and animal bones were found in the fill and the layers which surround the early walls. Both the ceramics and
the animal bones significantly changed with these levels, indicating a major period shift at the site. The pottery dating
to the 12th- 14th centuries CE, which was composed of glazed sherds and coarse ware, ceded its place to a wide range of
painted and plain pottery within a large chronological bracket extending from the Late Chalcolithic through the Early
Hellenistic period. The animal bones likewise demonstrated a change in animal husbandry and diet, with an increased
number of sheep, goat, and pigs. A sheep's tooth was sent to TUBiTAK MAM for Cl4 dating in 2019; the calibrated
date came back as ca. 264-93 BCE (Report No. 82325108-125.05-40/3593).
These layers also contained a good number of coins (at least 50) which are helpful in dating. Among these coins
only a handful could be read due to the poor level of preservation. Despite discrepancies in the ceramic assemblages in
terms of periodisation, the coins present a consistent pattern. Their size, weight (despite variations due to preservation),
material (mostly bronze), shape, and even deterioration pattern indicate homogeneity in the group regardless of their
illegibility. The most common type among those which could be cleaned and conserved has on its obverse a leather cap
on the left and eight-rayed star and a bow facing inward on the reverse (Price 1993: SNG-BM I, 977). A second type
has a head in leather cap and a quiver on the reverse (could be Amisos) (125-100 BCE; Price 1993: SNG-BM I, 1136).
A third variation features those with a bow in a case on the obverse with the eight-rayed star and bow facing inward on
the reverse (130-100 BCE; Price 1993: SNG-BM I, 976). The fourth variation are those with a leather cap on the left
and a bow facing inward on the obverse, and an eight-rayed star and a bow on the reverse (130-100 BCE; Stancomb
2000: SNG, 644). These variations are among the "Anonymous" types of Hellenistic period coins of Pontus, dating
roughly to between 130-100 BCE. Unlike later examples these coins do not bear the name of the minting city.
There are two unique and earlier coins from the layers. The first one is a silver (reduced?) siglos (4.09 gr) (KARP18-
HTP01-253 [T21]) (Fig. l 7-6a) This coin has Hera on its obverse (unclear whether in decorated or turreted stephane),
and an owl with outstretched wings on the reverse. Unfortunately, the coin is not well preserved and details are not
legible. Sigloi with Hera and owl have been attributed to Amisos, with examples dated to the 4th and 3rd-2nd centuries
BCE. A reduced siglos may indicate that this is from the later group and may date to the 3rd and 2nd century
c=•--====---=~scm
Figure 17-6. a) Silver coin (Amisos) depicting
Hera/owls with outstretched wings (3 rd-2nd century
BCE); b) Bronze coin (Knossos) with labyrinth on the
--- 5cm
Figure 17-7. Terra cotta figurine (top)

and bronze fibula (bottom).
reverse (ca. 221 BCE).
Recent Discoveries at Komana (2018- 2020) 235
BCE, making it one of the earliest coins discovered at the Komana excavations thus far. Finally, there is a bronze coin
from Knossos/Crete (KARP18-HTP01-050 [T12]) (Fig. 17-6b). Unfortunately, the obverse is badly weathered, but it
has a labyrinth and KN [0] on its reverse. Bronze coins with labyrinths on their reverse were minted at Knossos from
around 300 BCE through the end of the 2nd and early 1st century BCE (For Cretan coins see Svoronos 1890: 189); for
labyrinths Jackson 1971 and Ackermann 2005). This coin is 22 mm in diameter and 8.41 grams, which falls under
Denomination A in Jackson's typology (1971: 294) and is possibly a Zeus/labyrinth type. This group dates to around
221 BCE (Jackson 1971: 293-294).
In addition to the coins, four fragments of terracotta figurines and a bronze fibula were found in these layers, which
could help with dating (Fig. 17-7). The fragments belong to female figurines as indicated by the folds of their drapery.
One example is more complete than the others. The chest of a veiled female is represented wearing a chiton with a v-
shaped collar. A braided strand of her hair falls on her left shoulder. The himation is probably pulled over her head. All
pieces are mould-made and very fragmentary. Female figurines are rather common during the Hellenistic period. The
Komana figurines may be compared to female figurines from Daskyleion (<;evirici 2005: 60). The bronze fibula is a
rather interesting piece of evidence from these layers. Parallels may be found among the Galatian fibulae which have
been identified as the "Middle La Tene" type of material culture (Muller-Karpe 1988; Darbyshire et al. 2000; Co~kun
2014). La Tene style fibulae are divided into two groups. The first group is dated to the 3rd and early 2nd centuries BCE
with examples distributed to a wider region beyond the core Galatian settlement area. The second group is later, and
dated to the late 2 nd and 1st centuries BCE. They are found within the core settlement area of the Galatians. The Komana
fibula is most similar to a specimen at Tufanbeyli-Sar, the site of Komana Cappadocia. The Sar fibula belongs to the
earlier group.
In 2020, a new wall, which is much wider and better constructed, became apparent to the north of the trench (1.6 m
wide, 3.7 m long). While this wall follows the same orientation as the first group, it is about 50 cm deeper and may
belong to another earlier building. The characteristics of the soil deposit also changed with the wall (T25). This new
layer had a darker brown colour and was firmly packed. Ceramics include examples which could be dated to as early as
the Late Chalcolithic period. Such early ceramics have been found in other/upper layers and will be briefly mentioned
in the following section. Future excavations towards the north are promising for further understanding this complex
building level.
LATE IRON AGE/HELLENISTIC POTTERY
The excavations in the HTP0l area at Hamamtepe, between 2017 and 2020, as described above, clearly demonstrated
at Komana the presence of both a previously unknown but much anticipated level, that is the Hellenistic period, and the
potential of reaching pre-Hellenistic occupation levels at the site. The ceramics from the excavated layers were only
partially studied, but a more comprehensive examination will become a part of Ay~e Batman's upcoming dissertation.
Still, most diagnostics were drawn and photographed, and a comprehensive literature review was conducted in order to
position Komana ceramics within the geographical and chronological context. The study included assemblages from
western Anatolia, central Anatolia, Cappadocia, north central Anatolia, and the Pontus region. Both the probable imports
identified in the assemblage and the general international character of Komana as a sanctuary site required an
investigation of a wider geography beyond Anatolia.
We have decided to include some in-depth evaluation of the ceramics from these layers since assemblages from
securely dated Hellenistic levels in north central Anatolia are rare (with the exceptions of Oluz Hoytik [Donmez et al.
2009] and Hadrianopolis [Kan Sabin 2019]) and the Komana material posed a great potential for expanding our
knowledge about the archaeology of Hellenistic Pontus.
Ceramics from the relevant layers included fine, medium, and coarse wares. The coarse wares are identified mainly
with the body sherds of a few plain amphorae and three broken amphora handles without stamps. The cooking pots,
which are mostly black due to exposure to fire, constitute the main percentage of the medium wares. More than twenty
diagnostic pieces of cooking pots were mixed in the layers. While the cooking pots indicate domestic activity, their
haphazard distribution in the layer does not necessarily identify the spaces as domestic. Among the fine wares,
monochrome, bichrome, and polychrome groups were identified. The fine wares included plates, fish plates, medium
and small-sized cups like echinus bowls, small handled kylixes, and unguentaria. A few pieces were classified as
probable imports according to their different fabric, decorations, smooth/perfect textures, and quality of their slips.
The Late Iron Age
The first group of pottery within the Late Iron Age is the polychrome group which has orange fabric with
brown/black and red decorations on white painted panels. There are about 15 diagnostic pieces with this kind of
decoration. Among these there are three fragmented pitchers (Fig. 17-8). The decoration on these pitchers includes
black, red, and white lines and upside down triangles on buff/white bands. While on one the triangles are cross-hatched
in black, on the other they are painted red. One of the three pitchers has a peculiar basket decoration on the body, for
which a comparandum could not be found. The rest of the examples have similar linear decorations; two pieces have
black and red dots in a band defined by black lines at the top and bottom.
- - i"Ctn
Figure 17-8. Three polychrome pitchers from HTP0l, Late Iron Age.
Examples of polychrome painted panelled pitchers were found in north central and central Anatolia during the Late
Iron Age at sites like Bogazkoy (Matsumura 2005: 338) and Kaman-Kalehoyiik (Matsumura 2000: 126; Genz 2000).
Similar examples originating from the Eskiyapar and Bogazkoy excavations are also on exhibit at the <;::orum Museum.
At both Kaman-Kalehoyiik (Matsumura 2008: 177) and Yass1hoyiik/Gordion (Henrickson 1993: 132) similarly
decorated spouted jars were dated to between the early 7th to 6th centuries BCE. Comparanda for the first jug were found
at Km1k Hoyiik Level V (end of early Iron Age- beginning of Middle Iron Age (11 th- 9th c. BCE) (Ergurer 2016: 92) and
the Gordion Destruction Level (Sams 1974: 148).
There is also a group of bichrome pottery from these layers. These have red or brownish red bands around the
exterior of their rims. Red-banded wares are pretty common across central Anatolia in the Late Iron Age (Kealhofer and
Grave 2011: 421 ), and good comparanda for this material can be found at Govezli Tepesi Hoyiik, near Karaman (Ergurer
2018: 69).
These examples from Komana seem to comfortably fit into the pottery traditions of central Anatolia in the Late Iron
Age. The frequency of the use of panel decorations, polychrome techniques, and Greek inspired shapes were interpreted
as the result of interactions with the west during the Late Iron Age (Kealhofer and Grave 2011: 421). The limited
character of the data from this period makes it difficult to reconstruct Komana's relationship with western Anatolia, but
one could certainly suggest relations with sites to the west of Komana.
The Hellenistic Period
In addition to the Late Iron Age ceramics, another group was found in these layers which has not been discovered at
Komana before. These included both painted examples and a group of plain wares with a light orange fabric as
predecessors of sigillata. This group is pretty large and requires a thorough study and thus is not included in this chapter.
However, it may be important to note that a number of drinking vessels and plates in this group have dark orange and
sometimes brown slip, with dribble marks on their exteriors. On some of these plates, rouletting decoration and stamped
palmettes are present.
Banded Wares
The most characteristic type of ceramics among the Hellenistic group is plates, fish plates, and bowls with painted
bands on their interior (Fig. 17-9). Almost all layers both from 207/608 and 307/608 have these sherds. The fabric of
the pottery group varies from orange to light orange, and the bands are mostly red and orange on buff with a light slip.
The bands, which vary in thickness, must have been drawn with a compass. Among these, there are a few exceptions;
one has three white bands on a dark brown slip, and a rim has black, orange, brown, and white bands on a smooth orange
background. The decoration on the rim is on both the interior and the exterior of the bowl.
This type of banded pottery is commonly found in Hellenistic contexts of central Anatolian sites, especially within
the K1z1hrmak basin, and is referred to by a variety of names. Zahn identified this group as "Galatian Ware," which
includes floral motifs in addition to bands (Zahn 1907: 231). Akarca slightly disagreed with the term "Galatian Ware"
and identified Kara Samsun as the original centre of production, with examples from Akalan, Ali~ar, Bogazkoy, Sivas,
Kayseri, and Sivastopol (Akarca 1960: 145). Zoroglu broadened the term by calling it "K1z1hrmak Basin Ware" (Zoroglu
1986: 459). More recently, Ozsait and Ozsait conducted a study of the polychrome banded wares, gathering all data
from sites in the K1z1hrmak basin and demonstrated that this type was also in use in the central Black Sea region
especially around Amasya (Ozsait and Ozsait 2003: 332) U. Lohner Urban included the "Central Anatolian Banded
Ware" at Tavium-Yozgat in the "K1z1hrmak Basin Ware/Galatian Ware" (Urban 2019: 3). Stewart (Stewart 2010: 100)
investigated the "Galatian/K1z1hrmak Basin Ware" at Gordion in two groups: "East Anatolian Banded Ware" which
was found at Bogazkoy, Ali~ar, Alacahoyiik, Buyiik Nefeskoy, Amasya, and Pmarh (Maier 1963: 220; Donmez
(
7
I ■ ■ ■ ■ ■ 10 cm
Figure 17-9. Examples of Banded Wares, Komana.
et al. 2009), and "West Anatolian Banded Ware" which was found at Gordion, <;::atalhoyiik, Ankara, and Adapazarh
(Stewart 2010: 100). The southernmost examples of banded wares come from Komana Cappadocia (Korsulu 2014).
Similar to the "West Anatolian Banded Ware," these only have bands as decoration. The Komana Cappadocia examples
differ from the central Anatolian ones in the colours used for the bands as well as forms (Korsulu 2014: 92), suggesting
closer links with coastal cities as they more closely resemble those at Komana Pontica.
Figure 17-10. Sherds with ivy garlands, Komana.
The exact dating of the banded wares is a difficult task especially for the Black Sea area due to the lack of stratified
excavations. Maier dated the Bogazkoy examples to the Late Iron Age/Late Phrygian period (Maier 1963), while Urban
(2019) dated the Tavium-Yozgat examples to 250-50 BCE. In relation to the Tavium group, D' Agostino suggested a
date between the 2nd century BCE and 1st century CE for the banded ware at U~akh Hoyuk (D'Agostino 2019). While a
number of scholars suggested that such banded pottery was also produced in the Middle and Late Iron Ages in central
Anatolia, they agreed on a Hellenistic date for them (Ozsait and Ozsait 2003); Stewart (2010) dated the Gordian group
more specifically to ca. 240-189 BCE. The Komana banded wares have only bands as decoration similar to Stewart's
"West Anatolian Banded Ware" group and those at Komana Cappadocia. Since the dating margin is fairly large between
the mid-3 rd century BCE and 1st century CE, it would be difficult to suggest a more specific date for the Komana
examples except that there are coin data from the end of the 3rd through the end of the 2nd century BCE, a C 14 date of
ca. 264-93 BCE, and the bronze fibula dated to the 3rd and early 2nd centuries BCE.
In addition to the very characteristic banded wares there are three polychrome fragments with ivy garlands (Fig. 17-
10). They all have orange fabric and red decoration on a white slip. The larger body fragment is decorated with red ivy
leaves with black stems around them. The second and third are much finer pieces and have ivy garlands in addition to a
band of waves on one and checkerboard on the other. Ivy garlands on pottery during the Hellenistic period is a distinctive
characteristic of the Attic West Slope pottery. The Attic originals have white motifs on a black surface, while here we
have brown and black on buff. The execution of the ivy leaf is also rather rough compared to the West Slope examples.
Floral decorations, including ivy garlands, are also present on some examples from the so-called K1z1hrmak Basin Ware.
A complete amphora from Bogazkoy (Zoroglu 1986: 227, fig. 13), another piece found near Amasya at Akkaya (Ozsait
and Ozsait 2003: Pl.I, 6) and two bowls at the Amasya Museum, are among these examples (Donmez 2001). Donmez
dated the introduction of ivy garlands on central Anatolian pottery to the 5th century BCE, suggesting cultural continuity
in pottery production from the Late Iron Age to the Hellenistic period (Donmez 2001: 93-94). The ivy garlands could
have been introduced to the central Anatolian pottery tradition by the West Slope style pottery travelling to the inland
areas. Rotroff identified two workshops of West Slope pottery: one at Pergamon and another, the so-called "Ivy Platter
Workshop", the exact location of which is still uncertain, that produced and even exported to coastal Asia Minor, the
Black Sea, Cyprus, Alexandria, Syria, and Palestine (Rotroff2002: 97). Examples of variations in the execution of ivy
leaves and garlands are found at Daskyleion (Bulut 2013: 78, 79) and Gordian. At Gordian, the Early Hellenistic levels
not only have local productions of ivy leaf decorated vessels, but also a few imports with ivy and even olive garlands
(Stewart 2010: 182-183). Ifwe consider that the West Slope "style" was well known in inland Anatolia, with imports
and local variations, then a link between the inclusion of the ivy garlands in the floral motifs of the Banded Ware may
indeed be suggested.
Imported Pottery
Among the pottery recovered from the Late Iron Age/Hellenistic layers at Komana there are a few possible imports.
The earliest is a small fragment with polychrome geometric design on both the interior and the exterior (Fig. 17-1 I.a).
The fragment has buff fabric, and the motifs are painted in brown and red on a buff slip. This piece resembles
"Ephesianizing Ware," a group of pottery found at Daskyleion, which has linear decorations with additions of "dog-
tooth shapes" and squares interpreted as imitations of the Ephesian Ware (Gurtekin Demir 2002: 116). According to
Gurtekin Demir, Ephesianizing Ware has only been attested at Daskyleion and Sardis so far (Gurtekin Demir 2002: 116)
and can be dated to the second half of the 7th-first half of the 6th century BCE. If this is indeed an example of
Ephesianizing Ware, then Komana may be added to that list. After all, Komana Pontica was under the
b C
c:::::i--===---===is cm
Figure 17-11. Examples of imported pottery, Komana.
control of the Daskyleion satrapy during the Achaemenid period. Despite the similarity and the very attractive
identification of this piece as "Ephesianizing," we would also like to note that the dog-tooth pattern on the exterior and
the possible floral decoration on the interior has parallels to the K1Z1hrmak Basin/Galatian Ware. The zigzags appear on
a piece from K1zkayas1 I, west of Amasya (Ozsait and Ozsait 2003: Pl. IV, III), and the black leaf with the black fill
very closely resembles the leaf and bud ornaments described by Zoroglu, especially on one piece from Eskiyapar
(Zoroglu 1986: 226, fig. l 0). However, this suggestion is made with reservations.
A unique piece of ceramic from the assemblage is the rim of a deep bowl, or possibly a pitcher (Fig. 17-11 .b ). It has
a gray fabric and no slip. The exterior is decorated with an incised garland motif. The closest comparanda for this piece
comes from Gordion's Hellenistic levels. The Gordion examples are on pitchers (Stewart 2010: 210-211). Stewart
attributes the use of stems, ivy leaves, and garlands to the West Slope technique of the Hellenistic period (Stewart 2010:
210), but the use of gray fabric and goat depictions to the Phrygian tradition (Stewart 2010: 210). Such incised tendrils
are also present on a "West Slope Ware" fragment at Daskyleion (Bulut 2013: fig. 15.30, 16.43).
Another possible imported piece belongs to the base ofa possible plate (Fig. 17-11.c) This piece has orange fabric
with a reddish brown slip and brown/black brush bands across the interior. Towards the base of the plate there is a
stamped palmette. Similar pottery named "Delicate Banded Ware" (3 rd-2nd centuries BCE) was found at Priene (Heinze
et al. 2018: 807). While both the style and the unprecedented fabric suggest that this could indeed be an import, further
study is needed.
In the Late Iron Age/Hellenistic layers there is a group of small cups with handles (Figs. 17-12 and 17-1 3). These
are mostly orange in fabric with a brown slip which was only applied to the rim, including the handles and the interior,
with a dribble effect on some pieces (Fig 17-1 3.c). The most distinctive characteristic of this ceramic group is the shape
of its handle. The middle of the handle is pressed to the rim of the bowl with the help of a finger. In addition to the
slipped fragments (Fig.17-13a), there are a few that are unslipped with coarser handles (Fig.17-13.e-f). These cups
closely resemble the so-called Hellenistic Colour Coated Ware-A (Hayes 1991 : 27- 31) produced in and distributed from
Rhodes (Domzalski 2007: 166). These are distributed across the Aegean (Rotroff 1997: 117-118) and more so to the
Black Sea. Examples were found at sites such as Nymphaion, Olbia, and Tanais (Domzalski 2007: 171). The widespread
distribution may indicate other production centres for this vessel type. Domzalski dated them to the 3rd to 2nd century
BCE (Domzalski 2007: 171). While some of the Komana pieces must have been imported, others may be local
imitations.
C d
e f
r==-- -==i- -=~ scm
Figure 17-12. Hellenistic Colour Coated Ware-A, Komana.
a d
b e
~
CZ f
I~
-- - -
C
- 1 0cm
Figure 17-13. Examples of Hellenistic Colour Coated Ware-A, Komana.

Finally, in the ceramic assemblage, there are a few pieces of Attic black glaze, and several possibly Atticized pieces
were discovered, including two pieces of red slipped omphaloi and a few pieces of mould-made bowls with relief
decoration.
CONCLUSION
In this chapter, the most recent discoveries at Komana that include Late Iron Age/Hellenistic contexts have been
introduced. These contexts emerged under the Early Byzantine levels and have only been excavated partially with very
fragmentary architecture accompanying the archaeological deposits rich in Hellenistic pottery, coins, and other finds.
Until 2018, the Medieval levels, spanning from the 18th century Ottoman village settlement to the Early Byzantine
fortified citadel, have provided ample information on the Byzantine and Turkish settlement at the site. However, the
Hellenistic period, which constitutes possibly the most intriguing period at the site, could not be archaeologically
revealed in spite of an abundance of textual evidence. The new depths reached through excavations at the centre of the
mound between 2018 and 2020 revealed previously unknown ceramic and animal bone assemblages, increasing
numbers of Hellenistic coins, and other dispersed artefacts. Architectural remains with a complete new orientation,
distinguishable from later levels, accompanied the finds attesting the new levels. The architecture is very limited and
thus not very informative at this stage; however, a preliminary evaluation of the ceramic and other data from the relevant
layers has demonstrated the potential of understanding Komana both within its local and regional contexts.
The ceramics studied so far fall within the chronological frame of the Late Iron Age/Hellenistic period. On the one
hand, there is a group ofpolychrome pottery, mostly pitchers with comparanda in the central Anatolian Iron Age ceramic
tradition, and on the other hand there is a very large group ofpottery, painted and plain, which can be dated to the Early
Hellenistic period. In particular, the banded wares among the Hellenistic group seem significant for understanding
Komana' s position within north central Anatolia. Komana is so far the easternmost site where banded wares were found
in dateable contexts. It is also the first site located in the Y e~ilrrmak basin at which these wares have been discovered.
This indicates that the banded ware spread further east and is not necessarily limited to the KlZlhrmak basin, and that
Komana had close contacts with its western neighbours. Relations with sites like Gordion and even possibly Daskyleion
situate Komana in larger and international networks not only reaching the Black Sea coast, as one would expect, but
also land routes further to the west. Although very limited, possible imports from western Anatolia and the Aegean
strengthen this proposition. The provincial character of Komana in the Roman period and its delay in
Hellenization/Romanization have been previously discussed (Erciyas, forthcoming). Similarly, the transition from Late
Iron Age ceramic traditions to the early Hellenistic (from Achaemenid rule to the Mithradatids) may have been rather
slow or indiscernible, whereas the Late Iron Age pottery presented above may have continued to be in use until much
later, possibly into the 3rd century BCE. So far, the evidence from these layers indicates that we are dealing with contexts
from the 3rd through the 2nd centuries BCE with complex inter-regional relations. The figurine fragments are also
intriguing, since their discovery may in the future shed light on cultic activity and worship at Komana. Increasing
numbers of finds from these levels could also contribute to our understanding of the Achaemenid period in Pontus.
Overall, the Hellenistic layers at Komana promise to contextualise the site within an exciting historical geography and
fulfill Ballesteros Pastor's (2016) identification of the site as a "cultural crossroads."
ACKNOWLEDGEMENTS
Conducting research on a completely new assemblage of material culture for our research group has been a difficult
task during Covid-19 due to restricted or no access to libraries here in Ankara, and restrictions on travel. Therefore, the
support of our colleagues has been twice as important in composing this chapter. We would like to thank Dr. Geoffrey
Summers, Professor Nick Cahill, Professor Y~ar Ersoy, Associate Professor Marie-Henriette Gates, Associate
Professor Altay Co~kun, and Dr. Peter Talloen, for looking at our pottery and artefacts, and sharing their views as well
as recommending literature and sending e-resources. Our pottery was drawn by Tugba Tekin, and illustrations were
prepared by Murat Erciyas, to whom we are grateful.
REFERENCES CITED
Ackermann, Delphine. 2005. La representation du labyrinthe sur Jes monnaies de Cnossos: origine, evolution et
signification. Bulletin de !'Association suisse d'archeologie classique 2005: 33-36.
Anderson, J.G.C. 1903. Studia Pontica I: A Journey of Exploration in Pontus. Brussels: H. Lamertin.
Akarca, A~k1dil. 1960. Hellenistik <;agda Yerli Pontus Keramigi. Tiirk Tarih Kurumu Yaymlan V: 142-146.
Ballesteros Pastor, Luis. 2016. Comana Pontica in Hellenistic Times. A Cultural Crossroads. In Between Tarhuntas and
Zeus Polieus. Cultural Crossroads in the Temples and Cults of Graeco-Roman Anatolia, M. de Hoz, J.P.S.
Hernandez, and C.M. Valero, eds., 47-75. Colloquia Antiqua 17. Leuven: Peeters Publishers.
Budin, Stephanie. 2009. Strabo's Hierodules: Corinth, Comana and Eryx. In Tempe/prostitution im Altertum: Fakten
und Fiktionen, T.S. Scheer, ed., 198-220. Berlin: Verlag Antike.
Bulut, Hillya. 2013. West Slope Ware from Daskyleion. Jstanbuler Mitteilungen 63: 75- 127.
Casabonne, Olivier. 2003. Syro Anatolica Scripta Minora N.1: Les deux Komana et les exodes de Ma. Le Museon
116(3-4): 281- 283.
- . 2009. Kataonia, Melitene, Kummanni and The Problem ofKomana. Acta Orientalia Belgica XXII: 181-188.
Co~kun, Altay. 2014. Latene-Artefakte im hellenistischen Kleinasien: ein problematisches Kriterium filr die
Bestimmung der ethnischen Identitat(en) der Galater. Jstanbuler Mitteilungen 64: 129-162.
Cumont, Franz and Eugene Cumont. 1906. Studia Pontica II: Voyage d' exploration Archeologique dans le Pont et la
Petite Armenie. Brussels: H. Lamertin.
<;evirici, Figen. 2005. Daskyleion'da Ele Ges;en TapmanKadm Figi.irinleri. OLEA XI: 49- 74.
D ' Agostino, Anacleto. 2019. U~akh Hoyilk: Halklar ve b1rakt1klar1 somut izlerin Orta Anadolu platosundaki bul~ma
noktas1 / Incontro di genti e impronte materiali a U~akh Hoyilk, nell'altipiano centro-anatolico. Arkeoloji ve Sana!
16: 31--42.
Darbyshire, Gareth, Stephen Mitchell, and Levent Vardar. 2000. The Galatian Settlement in Asia Minor. Anatolian
Studies 50: 75-97.
Domzalski, Krzysztof. 2007. Changes in Late Classical and Hellenistic Fine Pottery Production in the Eastern
Mediterranean as Reflected by Imports in the Pontic Area. In The Black Sea in Antiquity. Regional and Interregional
Economic Exchanges, V. Gabrielsen and J. Lund, eds., 161- 182. Black Sea Studies 6. Aarhus: Aarhus University
Press.
Donmez, Nazan, Emine Emel, and $evket Donmez. 2009. Oluz Hoyilk Kaz1s1 ikinci Donem (2008) <;ah~malan.
Colloquium Anatolium VIII: 125-170.
Donmez, $evket. 2001. Amasya Miizesi'nden Boya Bezekli tki <;anak I~1gmda KlZlhrmak Kavsi Ges; Demir ve
Helenistik <;aglar1 <;anak-<;omlegine Yeni Bir Bak:1~. TUBA-AR 4: 89-99.
Er, Mehmet B. 2020. An Archaeometrical Investigation on Provenance and Technological Properties of Seljuk Period
Pottery from Komana (Tokat). Ph.D. thesis. Ankara: Middle East Technical University.
Erciyas, Deniz B. 2012. Komana ve <;evresinde <;aglar Boyu Yerle~im. In Turkiye'de Arkeometrinin Ulu <;marlarz.
Pro/Dr. Ay Melek Ozer ve ProfDr. $ahinde Demirci'ye Armagan, A.A. Akyol and K. Ozdemir, eds., 163- 171.
istanbul: Homer Kitabevi.
- . 2014. Komana'da (Sissiye) 2010-2012 Y11lar1 Kazi <;ah~malan. 35. Kazz Sonuc;larz Toplantzsz 35(1): 281-291.
- . 2019. Archaeology at Komana. In Komana Small Finds, D.B. Erciyas and M. Acara Eser, eds., 1--46. Settlement
Archaeology Series 7. tstanbul: Ege Yaymlan.
- . Forthcoming. Urbanization and Romanization of Komana. In Understanding Transformations: Exploring the
Northern Central Anatolia in Antiquity (c. 4th/3rd century BCE - 4th/5 th century CE), E. Sokmen and A. Schachner,
eds. BYZAS 26. istanbul: Ege Yaymlar1.
Erciyas, Deniz B. and Emine Sokmen. 2010. Komana Antik Kenti Arkeolojik Ar~t1rma Projesi 2008 Y1h Raporu. 27.
Ara~tzrma Sonuc;larz Toplantzsz (Denizli) 2: 355-374.
Ergi.irer, Hatice. 2016. Km1k Hoyilk Demir <;ag Boyahlarmda Bezeme Gelenegi. OLEA XXN: 67- 170.
- . 2018. Govezli Tepesi Hoyi.ik Demir <;ag1 Seramikleri TUBA-AR 22: 63- 93.
Genz, Hermann. 2000. Die Eisenzeit in Zentralanatolien im Lichte der keramischen Funde vom Biiyi.ikkaya in
Bogazkoy/Hattusa TUBA-AR III: 35-54.
Gi.irtekin, Demir and Rafet Giil. 2002. Lydian Painted Pottery at Daskyleion. Anatolian Studies 52: 111-143.
Hayes, John Walker. 1991. Paphos. Vol. III: The Hellenistic and Roman Pottery. Nicosia: Republic of Cyprus.
Heinze, Lars, Marie-Christine Junghans, Laura Picht, and Ursula Mandel. 2018. "Feine Reifenware" aus Priene.
Vorlaufige Uberlegungen zu einer wenig bekannten Keramikgattung des 3./2. Jhs. v. Chr. 9th Scientific Meeting on
Hellenistic Pottery, Thessaloniki, 5-9/12/2012, M. Kazakou and E. Kotsou, eds., 807-815. Athens: Tuµino
Apxmo1-.,oyucrov ITopov Kut A1rUAAO,ptroc:r£rov ~um0uv<ITt ~11µoc;1suµmrov.
Hemickson, Robert Carl. 1993. Politics, Economics, and Ceramic Continuity at Gordion in the Late Second and First
Millennia B.C. Social and Cultural Contexts of New Ceramic Technologies. In Ceramic and Civilization VI, W.D.
Kingery, ed., 89-176. Ohio: American Ceramic Society.
Hogarth, David G. and John A.R. Mumo. 1893. Modern and Ancient Roads in Eastern Asia Minor. London: Royal
Geographical Society.
Jackson, Anne. 1971. The Chronology of the Bronze Coins ofKnossos. Annual ofthe British School at Athens 66: 283-
95.
Jaeger, Bernard. 1982. Essai de classification et datation des scarabees Menkheperre: Prix de la confederation
internationale des negociants en ceuvres d'art 1979. Fribourg/Gottingen: Editions UniversitairesNandenhoeck
Ruprecht.
Kan $ahin, Giilseren. 2019. Hadrianopolis SeramikBuluntularz. Ankara: Ti.irk TarihKurumu.
Karasu, Yunus E. 2020. Komana Anadolu Sels;uklu <;ag1 Seramikleri. PhD. thesis. Ankara: Ankara Hac1 Bayram Veli
Oniversitesi.
Kealhofer, Lisa and Peter Grave. 2011. The Iron Age on the Central Anatolian Plateau. In The Oxford Handbook of
Ancient Anatolia, 10.000- 323 B.C.E., S.R. Steadman and G. McMahon, eds., 415--422. New York: Oxford
University Press.
Korsulu, Hatice. 2014. KappadokiaKomana' s1 Hellenistik Donem Seramikleri. Cedrus II: 89-133.
Maier, Franz George. 1963. Bemerkungen zur Sogenannten Galatischen Keramik von Bogazkoy. Jahrbuch des
Deutschen Archaeologischen lnstituts 78: 218- 255.
Marek, Christian. 1993. Stadt, ,.fra und Territorium in Pontus-Bithynia und nord-Galatia. Istanbuler Forschungen Band
39. Tubingen: Ernst Wasmuth Verlag.
Matsumura, Kimiyoshi. 2000. On the Manufacturing Techniques of Iron Age Ceramics from Kaman-Kalehoyuk.
Anatolian Archaeological Studies XVIII: 119-135.
- . 2005. Die eisenzeitliche Kerarnik in Zentralanatolien- aufder Grund/age der Ausgrabung von Karnan-Kalehoyiik.
Ph.D. dissertation, Freie Universitat Berlin.
- . 2008. A note on Anatolian Iron Age Ceramic Chronology: Black Lustrous Ware with Diamond Faceting. Anatolian
Archaeological Studies XVII: 175- 184.
Mitchell, Stephen. 1993. Anatolia: Land, Men and Gods in Asia Minor JI. Oxford: Clarendon Press.
Mutlu, Serdal. 2016. Tanrn;:a Ma ve Kappadokia Komana's1. Phase/is II: 311-322.
Muller-Karpe, Andreas. 1988. Neue galatische Funde aus Anatolien. Istanbuler Mitteilungen 38: 189- 199.
Olshausen, Eckart and Joseph Biller. 1984. Historisch-geographische Aspekte der Geschichte des Pontischen und
Armenischen Reiches:Teil 1, Untersuchungen zur historischen Geographie von Pontos unter den Mithradatiden.
Wiesbaden: Ludwig Reichert Verlag.
Ozsait, Mehmet and Nesrin Ozsait. 2003. La ceramique dite "Galate" du basin du KlZllrrmak. Anatolia Antiqua/Eski
Anadolu XI: 323- 342.
Peck, William H. 1987. Review of Bernard Jaeger, Essai de classification et datation des scarabees Menkheperre.
Journal ofNear Eastern Studies 46(3): 235- 236.
Price, Martin J. 1993. SNG BM Black Sea I Sy/loge Nummorum Graecorurn: Volume IX The British Museum. Part I:
The Black Sea. London: British Museum Press.
Ramsay, William M. 1890. The Historical Geography ofAsia Minor. London: John Murray.
Rotroff, Susan Irene. 1997. The Athenian Agora. Vol. 29, Hellenistic Pottery Athenian and Imported Whee/made
Tableware and Related Material Part 1: Text. Athens: The American School of Classical Studies at Athens.
- . 2002. West Slope in the East. In Ceramiques hellenistiques et romaines. productions et diffusion en Mediterranee
orientale (Chypre, Egypte et cote syro-palestinienne), F. Blonde, P. Ballet, and J.F. Salles, eds., 97- 115. Lyon:
Maison de !'Orient et de la Mediterraneen Jean Pouilloux.
Sams, Gilbert Kenneth. 1974. Phyrgian Painted Animals: Anatolian Orientalizing Art. Anatolian Studies 24: 169-196.
Sokmen, Emine. 2005. Temple States of Pontus: Comana Pontica and Zela. M. Sc. thesis. Ankara: Middle East
Technical University.
Stancomb, William. 2000. Sy/loge Nummorum Graecorurn British Series 11. The William Stancomb Collection. Coins
of the Black Sea Region. Oxford: Oxford University Press.
Stewart, Shannan. 2010. Gordian After the Knot: Hellenistic Pottery and Culture. Ph.D. dissertation, University of
Cincinnati.
Svoronos, Ioannis N. 1890. Numismatique de la Crete ancienne. Macon: Habelt.
Tatbul, Mustafa N. 2017. Identifying Medieval Komana in the 12th- 13th Centuries Throughout Spatial Analysis of
Archaeological Data with aMultidiscipliany Approach. Ph.D. Thesis. Ankara: Middle East Technical University.
Tatbul, Mustafa N. and Deniz B. Erciyas. 2019. Evaluation of the Recent Finds at Komana from the Early and Middle
Byzantine Period. In Settlements and Necropoleis of the Black Sea and its Hinterland in Antiquity, Gocha R.
Tsetskhladze and Sumer Atasoy eds., 272-280. Oxford: Archaeopress Publishing.
Urban, Ute Lohner. 2019. Zur Neudefinition der ' Galatischen Keramik' als, Hellenistic Central Anatolian Banded Ware.
In Rahmen des Levantine Ceramics Project, Forum Archaeologiae 9/11/2019 (http://farch.net).
Wilson, David R. 1960. The Historical Geography of Bithynia, Paphlagonia and Pontus in the Greek and Roman
Periods. Ph.D. thesis. Oxford: Oxford University.
Zahn, Robert. 1907. Die bei den Ausgrabungen in Boghaz-Koi gefundenen Tonscherben. Wochenschrift fur klassische
Philologie 24: 638-642.
Zoroglu, Levent. 1986. K1z1hrmak Havzas1 Kaplarmm Bic;:im ve Sus Geli~imine Ornekler. Turk Tarih Kurumu Yaymlarz
IX: 459-472.
CHAPTER EIGHTEEN
DIGGING IN DOLICHE 2018-2020
MICHAEL BLOMER
In the Roman period Doliche was a city on the border between the regions of Commagene and Cyrrhestice in the
province of Syria (Fig. 18-1 ). It occupied a strategic position at the junction of arteries which connected the Amanus
passes and Cilicia in the west, the Euphrates crossing at Zeugma in the east, Commagene in the north, and important
North Syrian urban centres such as Cyrrhus, Hierapolis, and Beroea in the south. Today, the site is located at the
northern fringes of the buzzing Turkish city of Gaziantep in southeast Turkey.
Hie .Singara
a
.Chai :::,,
Hatra.
Palmyra.
0 100
Qpr.v.f-11 ,: .Baalbek
Figure 18-1. Map of ancient North Syria (Permission of @Forschungsstelle Asia Minor).
The ancient settlement spreads over a shallow natural hill at the edge of a well-watered plateau (Fig. 18-2). Ancient
quarries and a large number of tombs have been dug into a limestone ridge that faces the city to the west. The ridge is
an extension of the Duluk Baba Tepesi mountain that towers over the ancient city. The main sanctuary ofDoliche, the
shrine of Jupiter Dolichenus, occupied the summit of this mountain (Winter 2014, 2017). The sanctuary preceded the
foundation of the city by hundreds of years. The cult goes back to the early 1st millennium BCE and flourished in the
Late Iron Age (Blomer 2015; Messerschmidt forthcoming), whereas no evidence of a pre-Hellenistic occupation has
been found in the city area so far. The whole surface of the city area is littered with lithic artefacts from the
Palaeolithic and, most importantly, the Neolithic/Chalcolithic periods, but the site appears to have been deserted in the
Bronze Age and Iron Age.
Throughout the Hellenistic and Roman periods, Doliche was a second-tier city of North Syria, which does not
figure prominently in the extant written sources. Consequently, knowledge about its history and urban development
was strictly limited until the archaeological exploration of the site started. Before that, Doliche was mainly recognised
for the cult of Jupiter Dolichenus, which attracted a large number of followers in various parts of the Roman Empire
(Blomer and Winter 2012). Excavations at the sanctuary have corroborated that Doliche was a supra-regional religious
centre and a destination for religious travel in the second and early third centuries CE.
In Late Antiquity Doliche became the seat of a bishop and retained its role as a regional urban centre. After the
Arab conquest, the city was located in the frontier zone between the Byzantine Empire and the Caliphate (Eger 2012).
When the Byzantine Empire regained control over North Syria in the 10th century CE, Doliche was raised to the status
Digging in Doliche 2018-2020 245
Figure 18-2. View of the hill (Keber Tepe) that was the centre of the urban area ofDoliche, from south
(Permission of@Forschungsstelle Asia Minor).
Figure 18-3. Map ofKeber Tepe (Doliche) with excavation areas marked
of capital of an eponymous small border province. In the Crusader period, the Latin counts and countesses of Edessa
were in control of the city. However, after the Sels;uks conquered Doliche in 1156 CE, the site was abandoned and
never reoccupied. Since then, natural degradation, extensive pillaging, and agricultural use have obliterated almost all
246 Chapter Eighteen: Blomer
traces of ancient and medieval occupation. Because of the scarcity of remains, interest in the exploration of the site
had been limited throughout the 20 th century (Wagner 1982).
Most parts of the city area are under protection today, but continuous agricultural activity that entails intensive
tilling, field clearance, and the planting of new crops accelerate the destruction of the ancient and medieval heritage.
Moreover, the growth of the city of Gaziantep over the last two decades, and the urbanisation and industrialisation of
the areas surrounding Doliche, have resulted in a significant increase in illicit excavation activity. In view of the
imminent threats to the site, a new archaeological project was launched in 2015 to investigate the ancient city. The
aim is to develop a fine-grained urban biography of Doliche, from the Hellenistic period to the Middle Ages, to test
assumptions about urban development in the region. Work at the site fully commenced in 2017. For the first project
phase, which ends in 2021, three work plans were defined. The first plan includes an intensive survey of the city area
in combination with geophysical mapping and remote sensing to gain insights into the urban structure (Whybrew
forthcoming; Balkaya et al. 2021 ). The other two work plans include two areas that have been selected for excavations
to explore the Roman and Late Antique city and its material culture (Blomer et. al. 2019; Blomer and Winter 2021).
One area is located on the plateau at the southeastern edge of Keber Tepe hill, where a public centre of the Roman
Imperial period had been identified (Fig. 18-3). The main features identified so far are public baths and the city
archive. The second area where large-scale excavations took place is on the south slope of Keber Tepe. The key
discovery here is a Late Antique Christian basilica, which has been dubbed the terrace church due to its location. The
size of the church and some features of its architecture suggest that this was the main ecclesiastical building of
Doliche.
Figure 18-4. Top: Doliche, Map of excavation area field 415 (baths, archive, and lime kiln);
Bottom: Aerial view of field 415 (baths, archive, and lime kiln) (Permission of@Forschungsstelle Asia Minor).
A PUBLIC BATH BUILDING
A large building complex that covers the western part of the excavation area turned out to be a bath building of the
Roman Imperial period (Fig. 18-4). It occupies a rectangular space of at least 2000 m 2 • While it is not yet possible to
ascertain the details of the ground plan, the excavated features and the results of the geophysical survey suggest that
the baths were axially symmetrically organised and followed the model of Roman imperial baths (Vanesse 2015).
Various large rooms can be identified, but because of extensive ancient pillaging, erosion, and modem ploughing,
little more than the foundations of the floors survived, which are made of mortar and rubble. Since the terrain
descended to the south, the size of the foundations increases significantly in this direction and can reach a height of up
to 1.9 m.
Figure 18-5. Doliche, Roman baths, room with cold water Figure 18-6. Doliche, Roman baths, test trench
swimming pool, 2nd/3 rd century CE (Permission of with pilae in hypocaust, 2nd/3 rd century CE
@Forschungsstelle Asia Minor). (Permission of@Forschungsstelle Asia Minor).
Best preserved so far is a 160 m 2 large room with a cold water swimming pool in the northwest comer of the
building. The pool is 72 x 35 min size and has a mosaic floor made of white tesserae (Fig. 18-5). The depth is 80 cm.
Two steps running along the northern side as well as circular steps in the southeastern and southwestern corners gave
access to the water. The pool was originally framed by a colonnade and surrounded by a corridor with mosaic floor.
The latter is partially preserved and displays geometric patterns that date the floor to the Roman Imperial period. To
discharge water the pool is connected to a sewer that feeds into a network of channels crossing under the foundations
of the baths.
In the southern part of the building, test trenches have revealed the remains of underfloor heating systems, i.e.,
hypocausts. This confirms the location of the warm and hot rooms in this area of the baths. So far, two different rooms
can be distinguished, but the limited extent of the test trenches does not allow us to draw conclusions about the size or
the exact purposes of these rooms. In one trench, regular grids of circular brick columns (pilae) that supported the
floor are still partly preserved (Fig. 18-6). The collapse between the brick columns contained fragmented square tiles
(bipedales), which were placed on top of the columns. Chunks of mortar and fragments of a polychrome geometric
mosaic show that this area was tessellated. In the second trench, only a floor of square brick tiles is preserved.
The intensive pillaging resulted in an almost complete loss of elements of the original furniture.
Apart from the mosaics, frequent finds of marble revetment plaques and profiles can be associated with the interior
decoration of the baths. Furthermore, numerous fragments of an entablature of the Ionic order, made of local
limestone, have been retrieved. The uniform small size of the fragments suggests that they were deliberately smashed
for burning in lime kilns (see below).
Despite the extensive destruction and poor state of preservation, the discovery of the Roman baths at Doliche is
significant. A local tradition of public bathing has not been traced in the region, and therefore the presence of Roman
baths, the adoption of Roman architectural templates, and the employment of Roman building techniques and
materials prove how Doliche was embedded in the Roman koine. The unassuming ruins shed light on the changing
patterns of daily routine in Doliche under the empire and demonstrate how Roman ideas of urban life percolated into
the Syrian hinterland. They also highlight access to resources and knowledge, and point to the entanglement of civic
elites in trans-urban networks. This is not surprising per se and mirrors developments in other provinces of the Roman
Empire (Kelly 2013), but for most cities of Roman North Syria the archaeological evidence is so scanty that no
conclusions about the impact of Roman hegemony on the urban landscape can be drawn (Vanesse 2015).
EXPLORING THE CITY ARCHIVE
A main goal of the project was to locate and investigate the city archive of Doliche. The existence of an archive
had been postulated already in 1940 by Henri Seyrig, who attributed a group of sealings (bullae) from the art market
to the city of Doliche (Seyrig 1940). While it must be assumed that all cities of the Hellenistic and Roman East
possessed archives to keep administrative and legal documents, very few of them can be traced archaeologically
(Coqueugniot 2013). This is usually only possible if the archive was destroyed by fire and subsequently abandoned.
The fire annihilated the documents, but preserved the sealings, made of unburnt clay, which were attached to them.
Consequently, the discovery of massive concentrations of sealings allows for an identification of an archive. Although
the documents cannot be retrieved, the sealings and their images convey important information about the city, its
administration, and the urban cultural and religious milieu (Schreiber 2021). The occurrence of burnt sealings
originating from Doliche strongly suggests that the local city archive had been destroyed by fire too. This became
even more likely when the number of published sealings with images that connected them to Doliche and Jupiter
Dolichenus dramatically increased in the second half of the 20 th century (WeiB 1992; 6nal 2011).
Figure 18-7. Collection of seal impressions from the city archive at Doliche
The precise location of the archive within the city area of Doliche remained elusive until local knowledge and test
trenches dug in 2010 helped to narrow down the source of the bullae to an area east of the bath complex mentioned
above. A first trench was excavated in 2017, and by wet sieving the spoils, 1200 sealings and fragments of sealings
were retrieved. In 2019, the excavation area was extended, and after weeks of wet sieving the new trench had yielded
another large batch of sealings, increasing the total number to more than 4000 (Fig. 18-7). This assemblage is of great
importance for ancient archival studies and will open up new avenues of research in the future (Schreiber 2021).
Moreover, the discovery of a huge number of sealings in a very narrow space unambiguously corroborates the location
of the archive at this very spot. In other trenches in the vicinity, less than a dozen sealings have been retrieved.
Notwithstanding the successful identification of the archive, the archaeological context did not meet the general
expectations. Only scant traces of the associated building have survived. They consist of massive ashlar walls that
were built directly on the bedrock (Fig. 18-8). In the east, up to three courses of ashlar masonry of a north-south
running wall have survived and stand up to a height of 1.1 m, but preservation decreases to the south, where it meets a
1.2 m wide east-west running wall which can be traced over a length of 7 m. A short section of a parallel wall has
been excavated in a test trench further north. This aligns to a large rectangular room enclosed by massive walls, which
were built directly on the bedrock. Bedding trenches had to be dug to compensate for the irregular rock surface. The
absence of any traces of a floor or a floor layer indicates that the extant walls were essentially part of foundations
situated below the ancient floor level.
The room clearly was part of a larger building complex, but neither the layout nor the total dimensions of the latter
can be ascertained. The main reason for this is that at Doliche even substantial ashlar walls do not show up in
geophysical mapping. They are usually made of a very soft and clayish local limestone, the properties of which are
similar not only to the bedrock, but also to the superimposed deposits of rubble. Therefore, they remain largely
untraceable. This strictly limits the scope and efficiency of non-invasive fieldwork methods not only in this area, but
for the whole city (Balkaya et al. 2021). 1 Consequently, only large-scale excavations will shed further light on the
character of the building complex that housed the archive.
Apart from the scant traces of walls, nothing but
a single very thick layer of sealing-rich
heterogenous soil and large ashlar blocks, that
obviously had been part of the surrounding walls,
was registered. This very disordered situation, and
the sporadic finds of modem waste, even at the
bottom of the trench, point to a destruction of the
archive by illicit digging. Apparently, an excavator
had been used, which not only explains the massive
scale of distortion, but also the large number of
sealings which had been overlooked by the robbers.
Based on the extant evidence, a likely scenario is
that a main destruction event unfolded in 1998,
when thousands of sealings surfaced in Turkish and
international private and museum collections (Onal
2011 ). In an initial event, a group of skilled robbers
must have discovered a still largely undamaged
destruction layer, which allowed them to collect a
huge number of seal impressions in a short time.
Nonetheless, the identification of the sealings
required a certain degree of knowledge and
experience. In a subsequent event, however,
probably after the rumour of the discovery had
spread, a team of unskilled robbers (re)opened or
extended the robbed area with an excavator and
demolished everything. The result was a massive
hole in the ground which was then backfilled, again
with an excavator, either by the robbers themselves
or by the field owners who needed it closed to till
the field. The backfill contained thousands of
sealings and bespeaks of the ineptitude of the
second party of robbers. In the process, not only
was the stratigraphy hopelessly disturbed, but also
the architectural context. The little that can be
Figure 18-8. Doliche, field 417, trenches 17-04/19-07, concluded about the architectural setting of the
city archive, remains of foundation walls archive is that the documents were kept in at least
(Permission of@Forschungsstelle Asia Minor). one large room, which was part of a monumental
public building complex. A comparable situation
has been observed at Zeugma, where the excavators retrieved more than 30,000 sealings from a basement at the edge
of the agora which was consumed by fire (Onal 2018). At both sites, it cannot be determined whether the archival
room(s) were still actively used at the time of the destruction or served for the storage of old and outdated files.
1
In contrast to the ashlar walls, mortar and brickwork resonate strongly. In the area of the bath building, floors and foundations can
be traced easily as long as they are close to the surface. The clayish soil absorbs the signals so quickly that results start to become
unreliable at a depth of no more than 1.2-1.5 m.
CONQUEST AND CRISIS
A massive fire devastated the baths and the adjacent archive building. Finds from the baths suggest that this
happened in the mid-3 rd century CE. The most likely scenario is that the destruction was the result of the conquest of
the city by the Sasanian army under Persian Great King Shapur I in 253 CE (Millar 1993: 159-67; Winter and Dignas
2007: 40-3). This is corroborated by the fact that Doliche is mentioned among the cities captured by the king in the
Great Inscription of Shapur I at Naqsh-e Rustam in Iran (Huyse 1999).
The havoc that the Persian invasions wreaked can be traced at many sites of North Syria and Mesopotamia. The
closest parallel is Zeugma on the Euphrates, where the evidence unmistakably shows how the Persian conquest led to
the complete destruction of the city and disrupted urban life for decades (Aylward 2013: 29-31; Abadie-Reynal 2015).
At Doliche, too, the devastating consequences of the Persian invasion can be traced. The baths and the city archive
had been part of the city's public centre, but it appears that no attempts were made to repair and rebuild them after the
Persian conquest. It is not even clear if it was reoccupied at all. Late Antique pottery, glass finds, and coins have been
found in large numbers, but the only building that can be securely dated to the period after the 3rd century destruction
is a large lime kiln, which has been excavated in the slope south of the archive complex. The kiln appears to have been
part of a larger industrial zone, and indicates that the remains of the former public buildings were dismantled and
burned into lime. The systematic reuse, recycling, and upcycling of building material in late antiquity might explain
why so little of the buildings survives and why even the foundation walls of the baths have been removed.
The abandonment of the city centre reveals that the Persian conquest was extremely disruptive. Doliche did not
cease to exist, but the city that eventually rose from the ashes in the 4th century looked profoundly different from the
city of the Imperial period. The archaeological evidence points to disintegration and discontinuity. This is suggestive
not only of a decrease in economic power, but also of a steep population decline. A likely scenario is that the
enslavement and mass abduction of people from Syria that Shapur I mentions in the Great Inscription at Naqsh-e
Rustam had a much more profound impact on Doliche and other cities of the region than has been assumed so far.
DOUCHE IN LATE ANTIQUITY: THE TERRACE CHURCH
In the second excavation area, only Late Antique and early Byzantine contexts have been exposed (Blomer et. al.
2019: 117-126). The main discovery is a large three-aisled basilica (Figs. 18-9 and 18-10). Colonnades separated the
9.3 m wide nave from the aisles, which are 4 m wide. The columns supported an architrave, not arcades as was the
case in most Syrian churches. Since the total height of base, column, capital, and architrave add up to only 3.6 m, a
superimposed second order of columns must have existed, but no remains of the latter have been identified. The
results of geophysical prospection suggest that the total length of the church is approximately 45 m. To the south, a 4
m wide corridor has been added to the aisle, which can be interpreted as a porticus. Corresponding to Syrian custom,
the main entrances to the church were located at this side. However, since the condition of the southern parts of the
church is very fragmentary, observations about the layout of this area must remain tentative.
I
I_ - - - - - - - - - - - - - - - - ,
I
I
I
I I
---
0
-- 10
Doliche, Terrace Church

20m I
I
I
I
Figure 18-9. Doliche, Terrace church, schematic plan with reconstruction

Figure 18-10. Aerial view of the terrace church and adjacent buildings in 2020
Figure 18-11. Doliche, Terrace church, detail of mosaic floor in the nave and northeastern comer ofbema,
late 4th century CE (Permission of@Forschungsstelle Asia Minor).
Partly preserved is the mosaic decoration of the nave (Fig. 18-11 ). The delicate, strictly geometrical design
suggests that it was laid out no later than the tum of the 4th to the 5th century CE. Close parallels can be found in
Antioch and Apameia (Balty 1995). The mosaic is superimposed on an older mosaic floor, which likewise covered the
whole nave and displays geometric patterns.
Very interesting is the discovery of a bema, a peculiar raised platform in the central nave, that was fenced off by
screens and decorated with a mosaic floor. Bemata are known exclusively from Late Antique churches of northwest
Syria. In each town, one church, usually the most significant and largest, had a bema within it. Its precise function,
however, is a hotly debated topic (Loosley 2012). Apparently, the clergy were seated on the bema during parts of the
service, facing the sanctuary. The presence of a bema in Doliche comes as a surprise. So far, this feature is almost
exclusively attested in the territory of Antioch, an ancient metropolis that commanded a large hinterland. Doliche,
however, was located relatively far from Antioch in a separate province, Augusta Euphratesis. Moreover, none of the
many churches of the region around Doliche is equipped with a bema (Hellenkemper 1978). This feature therefore
suggests a connection between the region of Antioch and Doliche in the late 4th century CE and sheds new light on the
complex networks of ritual praxis in late antiquity. Alternatively, it might be necessary to rethink the connection
between the North Syrian bema and the see of Antioch.
·-
-/_f_'t~..-.. . .-.. . . ~ -'- -~ - =;..,. _

Figure 18-12. Doliche, Terrace church, Apse with 6th century CE mosaic
In 2020, the sanctuary of the church with the apse was completely excavated (Fig. 18-12). At least three main
building phases can be distinguished. In the first phase, only a threshold separated the nave from the sanctuary, which
was decorated with a geometric mosaic floor. In a second phase, the sanctuary was raised by laying large rectangular
limestone slabs. Grooves along the western edge of the raised floor show that chancel screens closed off the sanctuary
except for a narrow central passage. In the final phase, the sanctuary and the apse were raised again. The area towered
70 cm over the nave and was accessible by a narrow central staircase. A mosaic decorated the apse. It is still very well
preserved, although the removal of ashlar blocks and an illegal excavation caused major damage to the central and
southwestern part of the floor. The main field shows a Nilotic river scene that is populated with fish, birds, and
papyrus or lotus plants. The motif is well attested in late antiquity and occurs in various churches of the Levant (Balty
1995: 245-254). Three ornamental zones surround the main scene. The inner zone is decorated with a five-stranded
guilloche. This is followed by a broad zone with an opulent multi-coloured perspective swastika meander. It frames
square fields with birds, and at the apex is a triangular field that displays a peacock. The outer zone consists of a
strapwork made of three interlaced bands (guilloche, rainbow stripe, and ribbon twist). The strapwork forms
alternately small circles and spindles. The latter are filled with different kinds of fish and a shrimp. Like the birds in
the perspective swastika meander, the fish refer to the Nilotic theme of the main field. The strapwork and the Nilotic
scene point to an origin in the 6th century CE. The high quality of the floor suggests that in this period Doliche was
still closely connected to the artistic centres of the Syrian coast, most importantly to Antioch.
Figure 18-13. Doliche, Terrace church, Northern annex room from east
The apse is flanked by two annex rooms, which seem to be later additions. The northern room could be accessed
through doors from the northern aisle and the apse. In its current state, the room is very simple (Fig. 18-13). The floor
is made of beaten earth and the masonry of the northern and eastern walls is crude and irregular. Three large stone
chests are lined up in front of the northern wall. At some point, a water channel was dug in the floor. It crosses the
room from east to west and continues in the northern aisle, where it cuts through the mosaic. The purpose of the
channel is still enigmatic, but it seems to have been constructed when the church was still in use.
Of the southern annex room little more than a simple mosaic floor made of large, mainly white tesserae has
survived. The floor stands in stark contrast to the badly preserved, but very elaborate polychrome mosaic of the
adjacent southern aisle. The situation clearly suggests a later building phase or renovation. This also explains why the
room is wider than the southern aisle and why the southern wall is running perpendicularly to the church. In doing so,
it follows the orientation of recently discovered rooms that were added to the eastern wall of the church. So far, three
rooms and a corridor can be distinguished. The walls consist of reused material, but the substantial size of the units,
the existence of two colonnades, a large tessellated floor, and direct access to the annex rooms of the church indicate
that the later additions were part of a larger ecclesiastical ensemble.
It seems that large parts of the church fell out of use in the 7th century CE. Destruction layers in parts of the
northern aisle point to a large fire that ravaged the church, while the collapse of the northern wall appears to have
resulted from an earthquake. Yet the evidence for the late phases of the church is as complex as it is rich. Further
studies of the finds and various small transformations are necessary before any definite conclusions about the final
phase can be drawn.
CONCLUSION
The results of the excavations at Doliche in 2018-2020 have shed new light on urban life in Doliche and the
material culture of North Syria in a diachronic perspective. Three aspects of the city's biography, which came to the
fore during the most recent fieldwork campaigns, have been highlighted in this chapter. The study of the public baths
has revealed a strong Roman impact on the cityscape in the Imperial period. Not only the layout, but also materials
and technology used for the building show how the city was firmly anchored in the Roman koine. Another significant
discovery is the evidence for the Persian conquest of 253 CE. The destruction of the city and a subsequent decrease in
population numbers significantly disrupted urban life at Doliche. The city recovered, but the cityscape changed
profoundly. Parts of the former city centre were never reoccupied.
A symbol of the Late Antique recovery is the terrace basilica, a newly discovered Christian church. After the
original construction in the (later) 4th century CE, the building underwent various renovations that highlight changes in
liturgical praxis in the early Byzantine period. The precise observation of these changes and the study of the finds
from the basilica promise to add important information on the development of Christian cult in North Syria.
ACKNOWLEDGEMENTS
Fieldwork at Doliche is conducted by the Asia Minor Research Centre, Munster University, Germany, in
cooperation with the Turkish Ministry for Culture and Tourism; it is funded by the Deutsche Forschungsgemeinschaft
(DFG).
REFERENCES CITED
Abadie-Reynal, Catherine. 2015. Les fouilles de sauvetage de Zeugma: un bilan des resultats. Journal of Roman
Archaeology 28: 821- 845.
Aylward, William. 2013. The Rescue Excavations at Zeugma in 2000. In Excavations at Zeugma, conducted by
Oxford Archaeology, W. Aylward, ed., 1- 54. Los Altos, CA: Packard Humanities Institute.
Balkaya, <;aglayan, Yunus Levent Ekinci, Olcay <;akmak, Michael Blomer, Julia Arnkens, and Mehmet Ali Kaya.
2021. A Challenging Archaeo-Geophysical Exploration through GPR and ERT Surveys on the Keber Tepe, City
Hill ofDoliche, Commagene (Gaziantep, SE Turkey). Journal ofApplied Geophysics 186: 1- 17.
Balty, Jeannine. 1995. Mosai'ques antiques du Proche-Orient. Chronologie, iconograp hie, interpretation. Annales
litteraires de l'Universite de Besan9on Centre de Recherches d'Histoire Ancienne 140. Paris: Les Belles Lettres.
Blomer, Michael. 2015. Religious Continuity? The Evidence from Doliche. In Religious Identities in the Levant from
Alexander to Muhammed. Continuity and Change, Achim Lichtenberger, R. Raja, and M. Blomer, eds., 129-141.
Contextualizing the Sacred 4. Turnhout: Brepols.
Blomer, Michael, Dilek <;obanoglu, and Engelbert Winter. 2019. Die Stadtgrabung in Doliche. Zu den Ergebnissen
der Feldarbeiten 2015-2018. Jstanbuler Mitteilungen 69: 103-186.
Blomer, Michael and Engelbert Winter, eds. 2012. Juppiter Dolichenus. Vom Lokalkult zur Reichsreligion,
Orientalische Religionen in der Antike 8. Tubingen: Mohr Siebeck.
_ _, eds. 2021. Exploring Urbanism in Ancient North Syria. Fieldwork in Doliche 2015- 2020. Preliminary Reports,
Doliche Urban Excavations 1. Berlin: DeGruyter.
Coqueugniot, Gaelle. 2013. Archives et bibliotheques dans le monde grec. Edifices et organisation. Ve siecle avant
notre ere- IIe siecle de noire ere. BAR International Series 2536. Oxford: Archaeopress.
Eger, Asa 2012. The Spaces between the Teeth: A Gazetteer of Towns on the Islamic-Byzantine Frontier. Istanbul: Ege
Yaymlar1.
Hellenkemper, Hansgerd. 1978. Kirchen und Kloster in der nordlichen Euphratesia. In Studien zur Religion und
Kultur Kleinasiens. Festschriftfiir Friedrich Karl Dorner zum 65. Geburtstag am 28. Februar 1976, LS. ~ahin, E.
Schwertheim, and J. Wagner, eds., 389-414. EPRO 66. Leiden: Brill.
Huyse, Philipp. 1999. Die dreisprachige Jnschrift Siibuhrs I. an der Ka 'ba-i Zardust (SKZ). Corpus inscriptionum
Iranicarum 3.1. London: School of Oriental and African Studies.
Kelly, Amanda. 2013. Roman Bathhouses on Crete as Indicators of Cultural Transition. The Dynamics of Roman
Influence. In Creating Ethnicities & Identities in the Roman World, A. Gardner, E. Herring, and K. Lomas eds.,
131- 167. Bulletin of the Institute of Classical Studies, Suppl. 120. London: Institute of Classical Studies.
Loosley, Emma. 2012. The Architecture and Liturgy of the Berna in Fourth- to-Sixth-Century Syrian Churches.
Leiden: Brill.
Messerschmidt, Wolfgang, ed. Forthcoming. Duluk Baba Tepesi I. Das eisenzeitliche Heiligtum. Dolichener und
Kommagenische Forschungen 11 , Asia Minor Studien. Bonn: Dr. Rudolf Habelt.
Millar, Fergus. 1993. The Roman Near East. Cambridge, MA: Harvard University Press.
Onal, Mehmet. 2011. Die Tonbullae von Doliche. In Von Kummub nach Te/ouch. Archaologische und historische
Untersuchungen in Kommagene, E. Winter, ed., 247-279. Dolichener und Kommagenische Forschungen 4, Asia
Minor Studien 64. Bonn: Habelt.
- . 2018. Die Siegelabdrucke von Zeugma. Dolichener und Kommagenische Forschungen 10, Asia Minor Studien 85.
Bonn: Habe lt.
Schreiber, Torben. 2021. Searching for the Archive of Doliche. Preliminary Report on the Archive Repertoire. In
Hellenistic Sealings and Archives, B. van Oppen de Ruiter and R. Wallenfels, eds., 121-147. Turnhout: Brepols.
Seyrig, Henri. 1940. Cachets d'archives publiques de quelques villes de la Syrie romaine. Melanges de l'Universite
Saint-Joseph 23(2): 85- 107.
Vannesse, Michael. 2015. L'architecture balneaire dans la province romaine de Syrie: quelques cas d'etude. In Zeugma
VI: La Syrie romaine. Permanences et transferts culturels, C. Abadie Reynal and J.-B. Yon, eds., 97-121. Travaux
de la Maison de l'Orient et de la Mediterranee 68. Lyon: Maison de l'Orient et de la Mediterranee Jean Pouilloux.
Wagner, Jorg. 1982. Neue Denkmaler aus Doliche. Ergebnisse einer archaologischen Landesaufnahme im
Ursprungsgebiet des Iupiter Dolichenus. Bonner Jahrbucher 182: 133- 166.
WeiB, Peter. 1992. Neue Tonsiegel von ,Doliche.' Chiron 22: 171- 194.
Digging in Doliche 2018- 2020 255
Whybrew, Sebastian. Forthcoming. The Urban Survey in Doliche 2017-2019. In Exploring Urbanism in Ancient
North Syria. Fieldwork in Doliche 2015- 2020, M. Blamer and E. Winter, eds. Doliche Urban Excavations I.
Berlin: DeGruyter.
Winter, Engelbert, ed. 2014. Kult und Herrschaft am Euphrat. Dolichener und Kommagenische Forschungen 6, Asia
Minor Studien 73. Bonn: Habelt.
- , ed. 2017. Vom eisenzeitlichen Heiligtum zum christlichen Kloster. Neue Forschungen aufdem Diiliik Baba Tepesi.
Dolichener und Kommagenische Forschungen 9, Asia Minor Studien 84. Bonn: Habelt.
Winter, Engelbert and Beate Dignas. 2007. Rome and Persia in Late Antiquity. Neighbours and Rivals. Cambridge:
Cambridge University Press.
CHAPTER NINETEEN
ROMAN BOGAZKOY: NEW DISCOVERIES IN THE NORTHERN LOWER CITY
DOMINIQUE KRUGER AND SVEN KOHN
Settled by mercenary Celtic tribes since the 3rd century BCE, Galatia was bequeathed to Rome by its last ruler
Amyntas in 25 BCE. It became an imperial province, with Ankyra as its capital and with ever-changing borders. The
long-standing martial culture of its inhabitants and its central location within Asia Minor made Galatia an important
place of recruitment for Rome's legions, and an equally important military staging area (Bennet 2019). Galatia
therefore found itself in a peculiar political and socioeconomic situation from the late 1st century BCE to the 1st
century CE, which is mirrored in the local architecture.
The area investigated is situated in central Anatolia, within the bend of the KlZlhrmak, about 20 km north of
Tavium (modem Buyuknefes), the capital of the Galatian tribe of the Trokmer. It includes the city area ofIJattusa, the
Bronze Age capital of the Hittites, and its immediate surroundings. During the excavations and surveys in the ruins of
IJattusa, which have been ongoing for over 100 years (for an overview see Schachner 2011 ), there have been, time and
again, indications of Roman settlements, both in the ancient city itself and in its surroundings. These, however, have
only recently been more precisely located. New excavations carried out in the Northern Lower City of IJattusa since
2014 have provided some insights into the settlement structures.
5km
'\J Settlement with Necropolis
V Workshops
Figure 19-1. The Roman road and settlements in the vicinity ofIJattusa (Bogazkoy-Expedition DAI, S. Kuhn).
Roman Bogazkoy 257
THE IMPERIAL ROAD AND SETTLEMENTS

IN THE SURROUNDINGS OF MODERN BOGAZKOY
Noteworthy in Bogazkoy's wider region is the clustering of villages and farmsteads around the Roman road that
connected Tavium with Amasia to the north, which passed approximately 5 km west of the village (Bittel 1985) (Fig.
19-1 ). Thanks to the discovery of several milestones along the course of the road-the earliest from the reign ofN erva
(e.g., French 2012: nos. 51 [CJ, 54 [B])-at least part of the road can be dated to the 1'1 century CE. Only the road's
general course is known, and details are still unclear. The motivation for undertaking such an extensive infrastructure
project might have been based on furthering Roman military and economic interests. The construction work itself
most likely fell, as was common, to the Roman soldiers based in the province. Some sections of the road around
Bogazkoy were secured and controlled by so-called way stations. Based on their construction at strategic locations,
these are interpreted as small garrisons placed directly by the roadside. None have been thoroughly explored, but an
analysis of the pottery found during surveys confirms that they were in use at the same time as the road they
monitored. Clear evidence of the presence of Roman troops in these structures has, however, only been recorded in
those located on the western slope of Tilkilitepe, in the form of a Roman legionnaire's helmet found in the vicinity
(Bittel 1985: figs. 20-23). These finds support the theory of the existence of a more permanent Roman presence which
would have been established as early as the first decades following the incorporation of the region into the empire. The
question remains, however, whether these way stations were the only garrisons in the area.
PREVIOUS WORK
In the 1960s, W. Dehn undertook a survey in which he was able to identify several settlements of the Hellenistic-
Roman period. R. Czichon subsequently published a map with several sites in the vicinity of Budakozu and their
Figure 19-2. Plan of the Roman structures (Bogazkoy-Expedition DAI, K. Czamitzki).

258 Chapter Nineteen: Kruger and Kuhn
location (Czichon 1997). He undertook further surveys in the immediate vicinity of Bogazkoy between 1996 and
1998, during which he identified numerous other potential settlement sites of the Hellenistic and imperial periods
(Czichon 2004; Kuhn 2014). Remains of Roman settlements have also been found on the so-called Buyukkale, an
inner-city rocky outcrop that served as the Hittite kings' official residence (Bittel 1932: 13 f.; Neve 1982: 170-172). A
Hellenistic-Roman necropolis stretches from the rock Kesikkaya (1-Jattusa) south of the Great Temple into the
Northern Lower City, containing over 200 graves dating from the 3rd century BCE to the 4th century CE; this was
mainly excavated during the late 1960s and early 1970s. Even today, each year's excavations reveal more graves,
which are even found in the immediate vicinity of the Roman imperial structures (Pickard et al. 2017; Kuhn 2021).
Since 2014, extensive excavations have been taking place in the Northern Lower City of 1:Jattusa, the ancient
capital of the Hittite Empire (Fig. 19-2). The results of geophysical surveys commissioned some years prior
(Schachner 2009: 41-46) as well as the remains of a wall of opus caementicium visible on the surface indicated that,
besides the Hittite remains, Roman structures might be found in the area of the so-called M1hraphkaya-a large rock
with a prominent niche carved into its surface, which was thought to be part of a Byzantine chapel (Bittel and
Naumann 1952: 34, 114).
In the first year, the excavations at "M1hraphkaya" revealed the walls of an extensive complex of Roman buildings,
which seems to be connected, at least chronologically, to the already-known Hellenistic-Roman necropolis.
THE ROMAN FORTIFICATIONS-A MILITARY CAMP?
Three major building complexes have been found

so far: first, parts of military fortifications; second, a
large pool or water basin with an adjoining, richly
decorated banquet hall; and third, a villa with an
attached bath (Fig. 19-2). Of these three complexes,
the fortifications are the oldest structures, and may
also answer the question asked above, namely if
there were more garrisons in the area.
Initially, two walls with courses that were
partially traceable above ground were more closely
investigated (Kruger et al. 2015: 88-90). They run
parallel to each other, with a southwestern-
northeastern orientation, and are mostly composed
of Hittite building elements in secondary use. As
quickly became apparent, the ashlars did not form
two separate walls, but were in fact the outer faces
of a single ca. 6 m-wide wall. On the basis of the
still visible walls, the excavated sections,
geophysical surveys, and topographical indicators,
this wall can be traced over a distance of at least 400
m. It forms two roughly right angles and continues
in the direction of the modem village, where we lose
the trail. Because of the dimensions of this
fortification, its regular structure and the rectangular
internal spacing, we proposed the interpretation of
the structure as a Roman military camp.
In order to test this theory, we began excavating
at several key locations. We unearthed a comer of
the fortifications, two internal rooms, and possibly
the remains of a gate. With only two rooms known
(Kruger et al. 2019: 65 f.), we can still offer only
minimal information in regard to their use, or the
shape or layout of the inner structures, but they can
certainly be assigned to the same construction phase,
with an orientation and construction technique
similar to the fortification wall.
For now, the northeastern comer and the area
adjacent to the modem road are far more promising.
Both the fortifications and the hypothetical gate are
connected with the northern Hittite fortifications
(Fig. 19-2), and differ markedly from the above-
Figure 19-3. Aerial view of the excavation area of the described, 6 m-wide wall in regard to their design
assumed tower, 2018. North is to the top and building technique. The northeastern comer
(Bogazkoy-Expedition DAI, M.S. Ozturk).
Roman Bogazkoy 259
seems to have been taken up by a tower on the inner side which was later rebuilt and modified. Immediately southwest
of the Hittite wall, a roughly 4 m-wide wall was constructed, which is connected to the 6 m-wide Roman fortification
wall. Right at this point the 4 m-wide wall turns to the south at a right angle. There is also some evidence for two
interconnected rooms(?) inside or on top of this 6 m-wide wall (Fig. 19-3). The discovery of a collapsed wall that had
fallen from north to south and came to rest in the interior of the eastern "room" indicates that this area had either
formed an open space to begin with (i.e., an interior room), or it had already been entirely dismantled when the wall
collapsed (Kruger et al. 2019: 68-70).
Figure 19-4. View of the inner face of the Roman fortification wall. North is to the top left
(Bogazkoy-Expedition DAI, Y. Dallal).
The area adjacent to the modem road (Fig. 19-4) could have housed a gate, as is suggested by a number of finds. In
1967, during a small excavation in this area, a large wall running southeast to northwest came to light, which was
subsequently interpreted as part of the Hittite city wall (Neve 2004: 177). However, it is actually the inner face of the
Roman fortification wall, consisting of re-used Hittite ashlars, traceable over a length of 20 m (Fig. 19-5). On top of
this wall a building had been erected at a later time-Peter Neve calls it a "Hellenistic house"- the small finds within
date it to the 3rd century CE at the latest (Kuhn 2016: 68-70, figs. 36, 37). In 2020, a large room was excavated in the
southern part of this area, which contains a significant number of architectural elements and installations (Kruger et al.
2021 ). Its location and features suggest that it may be the tower of a gate of the presumed military camp. While only a
relatively small number of finds has been made until now, these, as well as the results of radiocarbon analyses, date
the erection of the fortifications in general to the 1st century CE, possibly to the first half.
Figure 19-5. Aerial view of the area at the modem road (Bogazkoy-Expedition DAI, Y. Dallal).
It remains unclear why a Roman military complex would have been constructed here in Bogazkoy during the early
imperial period. The building project might have been connected to the incorporation of the region into the Roman
Empire (in 25 BCE) and the wider goal of pacifying the new province. The site's strategic position is undeniable; the
camp at Bogazkoy would have been both in sight and in reach of two mountain passes to the west and east (Schachner
2020: 406), as well as the aforementioned road leading northeast, and one of its functions would surely have been
securing and safeguarding both. Its location on a hillside and therefore slightly higher up than the terrain to the north
must have offered sentinels a wide-ranging view of the surrounding area. The ruins of the former Hittite capital were
used as a quarry, to which the many reused Hittite stones testify. The camp could also have functioned as a supply
post, with materials and provisions being distributed via the nearby road.
Since we lack historical sources concerning the region at that time, it is still unknown which and how many troops
were stationed at the military structure. Its function should be viewed in light of the general situation the Romans
faced in central Anatolia at the beginning of their reign. Especially for the early period of Roman rule, when the
province was still not entirely pacified and external enemies were not far, the existence of a Roman military camp
close to a road of the imperial period and in the wider vicinity of an important crossroads can come as no particular
surprise, and would fit nicely into the historical context of 1st century CE Galatia.
THE BUILDING AT MIHRAPLIKAYA
In the 2nd century CE the military camp might already have been abandoned and partly covered by other buildings.
New structures-this time made out of opus caementicium-were constructed on top of the fortifications or in their
immediate vicinity. Why the camp was decommissioned is completely unclear due to our lack of written sources,
leaving us only to assume that neither the camp nor its soldiers were of any continuing importance to the tactical and
strategic situation in the region. Now, in seemingly more peaceful times, the focus shifted to the construction of
civilian living spaces in the same area, continuing the settlement. In the southeastern area, two large building
complexes were erected. These complexes can be dated to the 2nd century CE by radiocarbon dating and art historical
compansons.
,,.
" . -,._-,...-+-t--;-'c-,-\t--;--;---1---1,--,-,.---t--+--1-\-H~-H--I ~
,,.
~ ~ ~"---71='-+~~77'~""'' g
0
3010 3020 3030 3040 3050 3060 3070 3080 3090 3100 31 10 3120
Figure 19-6. Aerial view of the building at M1hraphkaya with the walls of the water basin marked in blue
(Bogazkoy-Expedition DAI, M.S. Ozturk compiled by S. Kuhn).
The complex west of the rock of M1hraphkaya consists of a large water basin measuring about 18 x 55 m (Figs.
19-2 and 19-6), which was partly incorporated into the underlying Hittite structures. The northern wall of the basin
stands on the edge of the slope that leads down to the river Budakozu. M1hraphkaya is situated at one of the lowest
points on the terrain of the ancient city of 1-Jattusa; therefore erosion is always an important factor of taphonomy. The
northern wall of the basin is completely washed away, but it had an estimated height of about 4 m. Right in front of
Roman Bogazkoy 261
the niche carved into the rock face of M1hraphkaya (Fig. 19-7), the basin is divided into a complex system of smaller
pools, water channels, and walls. Figure 19-8 shows a possible reconstruction of the complex. Its central element is an
Figure 19-7. View of the excavation area in front of M1hraphkaya (Bogazkoy-Expedition DAI, M.S. Oztiirk).
Figure 19-8. Reconstruction of the building west ofM1hraphkaya (Bogazkoy-Expedition DAI, 0. Bruderer).
area on the right side of the niche. It is surrounded by walls on three sides, and is built directly next to the rock, using
its southwestern fa<;ade as support. Right in front of this room we found a large number of roof tiles in association
with other architectural elements like Doric capitals and fragments of unfluted columns. Amongst the rubble in front
of the room and also still in situ on one of the walls, we found fragments of painted wall plaster. One of the painted
panels, with a colourful floral motif, has already been restored (Fig. 19-9). A water pipe leads the water through the
southern wall of the room from higher urban areas into the basin. The hole for the water pipe is higher than the usage
horizon of the room. This leads to the assumption that the room was located below the surface of the surrounding area.
Figure 19-9. Panel of the wall painting from the banquet room (restored by K. Radezky)
(Bogazkoy-Expedition DAI, K. Czarnitzki).
Roman Bogazkoy 263
The exact function of these installations at M1hraphkaya is still unclear, and the lack of contemporary comparisons
makes it difficult to specify their purpose. However, the lack of objects associated with the practice of any kind of
known cult allows the interpretation as a purely recreational structure and/or one meant to impress visitors. The niche
and its function also remain mysterious. No plinth or base indicate the presence of a statue of any kind.
THE VILLA WITH BATHS
In the northwestern comer of the large basin, we found another portico and a set of steps which led up to another
building complex (Fig. 19-6, western excavation area directly next to the blue line). The column bases of this portico
are in fact repurposed capitals, which seem to have been scrapped due to defects or flaws (Kruger et al. 2020: 75 figs.
31, 33). One must presume that these capitals were the work of a local artisan lacking the experience of a more skilled
craftsman. The complex situated west of the water basin and its associated structures can, at least at the moment, be
interpreted as a Roman villa rustica, even though other interpretations are not unreasonable.
Figure 19-10. Aerial view of the villa with baths

(Bogazkoy-Expedition DAI, M.S. Oztiirk and Y. Dallal compiled by D. Kruger).
As can be seen on the eastern part of the aerial photograph (Fig. 19-10), the wide defensive wall with its outer
facings of Hittite ashlars is clearly visible and was incorporated into the complex. The excavated complex so far
consists of a number of apsidal rooms, some of which have been heated via hypocaust systems, while others are in fact
the remains of bathing pools. A mosaic floor deserves special mention; roughly 6 x 6 m in size (Fig. 19-11 ), it is
composed of geometrical patterns of coloured bands and diamond shapes, surrounding a central panel showing red,
black, and blue hexagons. The southeastern wall of the large apsidal room in the north and another wall west ofit was
built in the opus reticulatum technique, the use of which is extremely rare in the eastern provinces of the Roman
Empire (Spanu 1996: 924-931 fig. 1). Also notable is the widespread use of marble decorations, especially as wall
decoration or as a floor, which is not native to the region and must therefore have been imported. Lying on top of the
mosaic we also found a composite capital, one side of which bears great resemblance to a capital of the Tuscan order.
If this comparison was intended, however, it was once again executed quite poorly.
These architectural features appear to be a deliberate attempt by the builders to copy contemporary Roman
imperial architectural trends. While they were still largely unknown in Galatia at that time, most of them can be found
not only in Rome but also in various variants and forms in the cities of the Roman East. The mediocre or at times even
deficient workmanship, however, speaks to our case for local craftsmen, who were still unpracticed in recreating these
architectural forms and building techniques. Where exactly these artisans and the comparatively large workforce
necessary for these constructions lived, and who paid their wages, can as of now only be speculated. In the case of the
construction and maintenance of the military camp, the official imperial Roman administration may have played a
greater role. The military structures were, after all, directly connected to the incorporation of Galatia into the Roman
Empire, and the site was possibly of additional importance as a supply base and garrison to guard the road and nearby
mountain passes. In contrast, the construction of an extravagantly furnished villa more likely stemmed from the
building efforts of a wealthy local resident or perhaps a Roman administrator. For now, the question of the identity
and function of the villa's owner must remain unanswered.
Figure 19-11. Mosaic floor (Bogazkoy-Expedition DAI, S. Kuhn).
Its sheer size, elaborate layout, and rich decor, as well as the current lack of specific rooms that one would expect
to find in a Roman villa, would also support the interpretation of the complex as a Roman public bath. Since we still
have not found any contemporary residential buildings in the vicinity, one would have to ask for which clientele this
bath was intended. Several possible explanations come to mind. Among these, the most likely would seem to be that
either the bath was built for and by the inhabitants of a yet-to-be-discovered nearby settlement, or that it was created
and used in a kind of very unusual joint venture by the inhabitants of the surrounding area. Whoever the structure's
intended clients may have been, its construction made the installation and maintenance of a constant water supply a
necessity. A substantial number of water channels-taking the form of simple stone channels set without the aid of
mortar (within the military structures), channels coated with hydraulic mortar, and clay pipes-have indeed been
found in the Northern Lower City in recent years, bringing water to the site from the southeast (Schachner and Kruger
2019: 59; Kruger et al. 2020: 33 f.; Kruger et al. 2021). Water is available in the Hittite city ruins due to the geological
features in particular on the valley side of the rock formations in the city (Schachner and Wittenberg 2012: 251). It
turned out that some of the water pipes discovered in earlier excavations, which were previously considered to be
Byzantine, can be associated with this Roman imperial distribution system.
CONCLUDING REMARKS
The three complexes-the presumed military camp, the villa, and the area west of M1hraphkaya-were rebuilt and
reconstructed more than once, and were in use until the 4th century CE (Fig. 19-2). On top of the fortifications of the
1st century CE, walls of opus caementicium were constructed in the 2nd century CE, comprising several small rooms-
possibly storage spaces. Their southern boundary can be traced over 40 m on the inner shell of the Roman
fortification. AU-shaped structure was excavated in front of the rooms to the southwest, which probably includes two
bases and coloured wall plaster. This suggests a high quality interior similar to the banquet room at M1hraphkaya.
This, as well as the simultaneous reconstruction and integration of the camp wall into the villa complex, clearly shows
a fundamental change in the function and utilisation of the military structures. Some other alterations were probably
undertaken only a few decades after the construction of the villa and the large water basin. To cite a few examples,
walls of opus caementicium cover the older opus reticulatum walls in some areas. Also, walls constructed of quarry
stones and in dry masonry were built on top of the mosaic floor, which both covered the expensive floor and
fundamentally changed the use and layout of the room. Similar developments can be found in the northeastern comer
of the camp, where new walls were inserted, forming new rooms. The many phases of reconstruction can perhaps be
explained by local forces and elites who rose in prominence and began to transform the layout of the buildings
Roman Bogazkoy 265
according to their own needs and preferences, once the central power's influence or interest in Bogazkoy began to
wane.
In this context, the simple, parallel, and interconnected walls made of quarry stones in the western area of the baths
are also noteworthy (Fig. 19-2). They form part of a relatively large and long, rectangular, possibly three-aisled
structure, which at first glance bears some resemblance to an agricultural, commercial, or storage building.
The coins found in the urban area of the ancient city, which were minted in various localities between Rome and
Antioch on the Orontes (Kuhn 2021 ), speak for a lively and frequent exchange between Roman Bogazkoy and the
provinces of the Roman Empire. This fits the small number of pottery sherds that were found, among which small,
imported bowls appear most frequently (Ubben 2020, 2021). The pottery, and brick stamps, and stonemason's marks
that have been found point to various workshops. These have not been located yet, however, thus making both local
productions and imports possible. Noteworthy are also the many glass fragments that came to light, especially in the
western sector of the villa. Among these are fragments of blue, green, white, and transparent glass, which belong to
common as well as prismatic bottles, grooved vessels, and cups. The raw materials necessary for their production were
imported and processed on site (Nakai et al. 2014). As a surprise came the discovery of several fragments of window
glass (Kruger et al. 2021 ); this is the first time that this type of glass has been found in Galatia.
When and why the Roman structures described here were abandoned in the late 4 th century CE remains an open
question. There are, however, some indications of a major destructive event, which may have damaged or destroyed
the building complexes and led to their subsequent abandonment. Finally, whether their erstwhile inhabitants and
occupants remained in the vicinity or gave up the area entirely remains a mystery.
ACKNOWLEDGEMENTS
We would like to thank Professor Dr. Andreas Schachner and the Bogazkoy-Project for the support and the
possibility to work on the Roman structures as well as our colleagues in the field for the fruitful cooperation. We
would also like to thank the DFG for the financial support of the long-term excavation program in the northern Lower
City and on the terrace west of M1hraphkaya.
REFERENCES CITED
Bennet, Julian. 2019. The Annexation of Galatia Reviewed. Ada/ya 22: 223-257.
Bittel, Kurt. 1932. Die James Simon-Grabung in Bogazkoy September 1931. Mitteilungen der Deutschen Orient
Gesellschaft 70: 1-24.
- . 1985. Beobachtungen an und bei einer romischen Straj3e im ostlichen Galatien. Heidenheim: Offsetdruck.
Bittel, Kurt and Rudolf Naumann. 1952. Bogazkoy-ljattusa 1. Architektur, Topographie, Landeskunde und
Siedlungsgeschichte. Stuttgart: W. Kohlliammer Verlag.
Czichon, Rainer M. 1997. Studien zur Regionalgeschichte von ljattusa/Bogazkoy 1996. Mitteilungen der Deutschen
Orient Gesellschaft 129: 89- 102.
- . 2004. Studien zur Regionalgeschichte von Hattuscha/Bogazkoy vom Chalkolithikum bis zur Byzantinischen Zeit.
Habilitation Julius-Maximilians-Universitat Wiirzburg.
French, David. 2012. Roman Roads and Milestones of Asia Minor. 3.2 Galatia. BIAA Electronic Monographs 2.
Ankara:
https:/lbiaa.ac.uk/ckeditor/filemanager/userfiles/ electronic_publications/previews/3 .2_ Galatia_summary. pdf
Kruger, Dominique, Sven Kuhn, and Andreas Schachner. 2015. Ausgrabungen in der nordlichen Unterstadt bei
M1hraphkaya. In A. Schachner, Die Ausgrabungen in Bogazkoy-ljattusa 2014. Archiiologischer Anzeiger 2015(1):
83-92.
- . 2019. Ausgrabungen an der romischen Befestigungsanlage im Norden der Stadt und auf der Terrasse westlich von
M1hraphkaya. In A. Schachner, Die Ausgrabungen in Bogazkoy-ljattusa 2018. Archiiologischer Anzeiger 2019(1):
65-83.
- . 2020. Strukturen und Befunde der romischen Kaiserzeit. In A. Schachner, Die Ausgrabungen in Bogazkoy-ljattusa
2019. Archiiologischer Anzeiger 2020(1): 31-44.
Kruger, Dominique, Serkan Durdemir, and Nurcan Y1ld1z. 2021. Die Befestigungsanlage des 1. Jhs. n. Chr. In A.
Schachner, Die Ausgrabungen in Bogazkoy-ljattusa 2020. Archiiologischer Anzeiger 2021(1): 27- 30.
Kuhn, Sven. 2014. Ein Do,f in Galatien. Bogazkoy-Hattuscha in Hellenismus und Kaiserzeit. Magisterarbeit,
Eberhard Karls Universitat, Tiibingen.
- . 2016. Ein monumentales Gebaude vor Mthraphkaya. In A. Schachner, Die Ausgrabungen in Bogazkoy-ljattusa
2015. Archiiologischer Anzeiger 2016(1): 12- 23.
- . 2021. Isolation und Kommunikation. Ressourcennutzung und Wirtschaftsriiume im nordlichen Zentralanatolien in
Hellenismus und Kaiserzeit. Ph.D. dissertation, Eberhard Karls Universitat, Tiibingen.
Nakai, Izumi, Mayumi Matsuzaki, Daichi Sawamura, Yoshinari Abe, and Andreas Schachner. 2014. Chemical
Characterization of Roman and Early Byzantine Glass from Bogazkoy/ljattusa and its Vicinity. Jstanbuler
Mitteilungen 64: 237-260.
Neve, Peter. 1982. Biiyilkkale, die Bauwerke. Grabungen 1954- 1966, Bogazkoy-ljattusa 12. Berlin: Gebriider Mann.
266 Chapter Nineteen: Kruger and Kiihn
- . 2004. Die hethitischen Stadtmauern von Hattufa eine Bestandsaufnahme. Architectura 34: 169-182.
Pickard, Catriona, Claudia Caldeira, Ninke Harten, Handan Usttindag, Ulf-Dietrich Schoop, Laszlo Bartosiewicz, and
Andreas Schachner. 2017. Reconstructing Iron Age to Roman Period Diet from Bioarchaeological Remains:
Preliminary Results from Bogazkoy, North-Central Anatolia. In Innovation versus Beharrung. Was macht den
Unterschied des hethitischen Reichs im Anatolien des 2. Jahrtausends v. Chr.? A. Schachner, ed., 239- 255. Byzas
23. lstanbul: Ege Yaymlar1.
Schachner, Andreas. 2009. Geophysikalische Untersuchungen und innerstadtische Gelandebegehungen. In A.
Schachner, ed., Die Ausgrabungen in Bogazkoy-Ijattusa 2008. Archiiologischer Anzeiger 2009(1 ): 40---46.
- . 2011. Au/der Suche nach dem Groftreich der Hethiter. Miinchen: C.H. Beck.
- . 2020. The Power of Geography. Criteria for Selecting the Location of Hattusa, the Capital City of the Hittite
Empire. In talugaes wittes. Ancient Near Eastern Studies Presented to Stefano de Martino on the Occasion of his
65 th Birthday, M. Cammerosano, E. Devecchi, and M. Viano, eds., 399---420. Kasion 2. Munster: Zaphon.
Schachner, Andreas and Dominique Kruger. 2019. Befunde der romischen Kaiserzeit auf der nordlichen Terrasse. In
A. Schachner, Die Ausgrabungen in Bogazkoy-Ijattusa 2018. Archiiologischer Anzeiger 2019(1 ): 58-59.
Schachner, A., and H. Wittenberg. 2012. Zu den Wasserspeichern in Bogazkoy/Ijattusa und der Frage ihrer Befullung.
In Wasserwirtscha.ftliche Innovationen im archiiologischen Kontext. Von den priihistorischen Anfiingen bis zu den
Metropolen der Antike, F. Klimscha, R. Eichmann, Chr. Schuler, and H. Fahlbusch, eds., 245-255. Menschen -
Kulturen - Traditionen. Studien aus den Forschungsclustern des Deutschen Archaologischen Instituts, vol. 5.
Rahden/Westf.: Verlag Marie Leidorf.
Spanu, Marcello. 1996. L'opus reticulatum e mix.tum nelle province asiatiche. In L 'Africa romana 11. Atti dell'XI
convegno di studio Cartagine, 15-18 dicembre 1994, M. Khanoussi, P. Ruggeri, and C. Vismara, eds., 923-939.
Pubblicazioni del Dipartimento di storia dell'Universita degli studi di Sassari 28. Ozieri: II Torchietto.
Ubben, Mareke. 2020. Bin erster Uberblick zur romischen Keramik der nordlichen Unterstadt. In A. Schachner, Die
Ausgrabungen in Bogazkoy-Ijattusa 2019. Archiiologischer Anzeiger 2020(1): 44---45.
- . 2021. Die romische Keramik von Bogazkoy: Beobachtungen der Grabungskampagne 2020. In A. Schachner, Die
Ausgrabungen in Bogazkoy-Ijattusa 2020. Archiiologischer Anzeiger 2021(1): 39---44.
CHAPTER TWENTY
ARCHAEOLOGICAL RESEARCH AT ELAIUSSA SEBASTE

(ROUGH CILICIA) IN 2019
ANNALISA POLOSA
Elaiussa Sebaste is a port city on the coast of Rough Cilicia, halfway between Seleucia on the Calycadnus to the
west and Soli-Pompeiopolis to the east. It appears to have been the site of one of the mints of Seleucus VI at the
beginning of the 1st century BCE and is known to have been among the territories that were granted to Archelaus of
Cappadocia in the Augustan period; it was this king who changed its name to Sebaste and made it his residence. It
then became a part of the domain of Antiochus IV of Commagene, who ruled there from 3 8 to 72 CE, when the
province of Cilicia was formally established. As is the case at many other sites in the region, the city must have been
abandoned in the second half of the 7th century CE. The excavations at Elaiussa have been conducted by a team from
the Sapienza University of Rome beginning in 1995 (Equini Schneider 1999, 2004, 2010), and have brought to light
several monumental complexes in the ancient city, settled on a promontory that once was an island, with the mainland
in front (to the west) ofit (Fig. 20-1), though it is now connected to the mainland.
There is a temple on the mainland dating to the
1st century CE, as well as a public space, constituted
by a theatre, a bathing complex, and a commercial
area, dating to the beginning of the 2nd century CE.
This area was then transformed into a basilica!
complex in the early Byzantine period and has been
excavated. Structures that have been investigated on
the promontory, in the area of the two harbor
basins, include a bathing complex, a huge
residential building developed around a central
court, and a residential and workshop quarter.
Discovered among the workshops was one of the
largest kilns in the area, apparently dedicated to the
production of transport amphorae.
Research has been largely influenced by the
need to preserve the archaeological site, threatened
by intensive building activities tied to the touristic
development of the region. In recent years
... ·: 7
excavations have been focused on the
reconstruction of the ancient city's topographical
grid, in order to tie together the monumental
quarters. These quarters are scattered over an area
of nearly 23 ha and are destined to become an
archaeological park. Research goals also include the
reconstruction of the history of Elaiussa Sebaste
with regard to the different stages of the site's
development.
With these research goals in mind, excavations
have been conducted in one area of the necropoleis,
as these can be considered as a guideline to better
understand the urban development process; work
@ ."""""""""""""""""=== has also been carried out on the northern
promontory, the original core of the ancient city,
Figure 20-1. General plan ofElaiussa. and in the public area, which underwent radical
transformations between early Roman and
Byzantine times.
On the mainland, along the coastline, at sea level and on terraces set on the hill, rock cut and built tombs in the
necropoleis can be seen. These are landmarks of the ancient site, as they appear, for example, in the travelers'
268 Chapter Twenty: Polosa
drawings of Laborde and Langlois (Borgia 2003; Laborde and Laborde 1838; Langlois 1861). Recent research
campaigns have focused on one sector of the southwestern necropolis, which had not yet been investigated (Fig. 20-2).
Cleaning of the funerary buildings on the higher terraces showed that, at least in one case, they were surrounded by
Figure 20-2. View of the southern necropolis .
........,,=...........="""'-:-=--:.,®
Figure 20-3. Aerial view of the southern necropolis. The walking path is on the right of the photo.
Archaeological Research at Elaiussa Sebaste 269
a temenos; generally, the tombs had been plundered in antiquity and have been re-used in modem times (functioning
as storage rooms by current local residents), so that the only finds are the funerary beds and, occasionally, scanty
remains of mosaics. However, the discovery of a lid and part of the coffin of a sarcophagus made of Proconnesian
marble and decorated with dancing Nikai and garlands, dating to the first half of the 2nd century CE (Fomace 2019),
gives us a chronological baseline for the floruit of this sector of the necropolis.
The lower terraces feature tombs that are cut into the rocky bank and completed with architectural elements on the
fa9ade; these contain rock-cut funerary beds or plain sarcophagi. Investigations showed that the different terraces
communicated through stair systems. Soundings fronting the southeastern sector of the necropolis revealed a layer of
beaten earth, 16 Roman feet wide, with a preparation underneath made of mortar and limestone shards, which could
be interpreted as a walking path (Fig. 20-3). Further east, a deep sounding revealed a wall on which a thick layer of
murex fragments was leaning, suggesting a possible activity of purpura extraction. Magnetometry showed that other
structures, oriented with the graves, may have partially occupied the space at the front of the necropolis. Some of the
buildings in the necropolis were re-used for a different function in Late Antiquity, as proven by the find of a small kiln
in one of the tombs.
=
'
~ ----- C
--f'~
·····1•
I '
I
I
~------\___ 7
:,,
I
□
I
I
I
lo
I
I
I
I
\tll
I
dQ
I[
Figure 20-4. Aerial view of the northern

20-5. Plan of the northern promontory architecture.
promontory. The signal tower can be found in the
lower portion of the photo.
Partial investigation on the northern promontory in previous years brought to light the remains of a small
Byzantine church. While the edge of the promontory is rocky, a sand dune has accumulated immediately to its south,
and from this a structure roughly circular in shape has emerged. This had been interpreted as a lighthouse (Fig. 20-4),
but can now be tentatively interpreted as a signal tower of post-Byzantine times. The excavations in this area brought
to light the remains of a huge quadrangular building, whose structures are partly re-used in the circular structure, the
plan of which will be better understood after the completion of excavations (Fig. 20-5).
The building revealed a series of rooms facing the sea, built in opus quadratum; the floor of one of the rooms is
decorated with a mosaic featuring a wave pattern made of black and white tesserae. A sounding has shown that these
buildings were at least partly founded on the earliest walls, constructed of polygonal masonry (Fig. 20-6). The
polygonal technique is attested in Late Hellenistic structures at Elaiussa, such as, for example, in the city walls
(Tempesta 2013). The rooms on the sea front probably experienced severe erosion from wind and water over the
course of time, and a new intervention modified the plan of the building, through the erection of a wall made of small
blocks, running from north to south, and cutting the wave pattern mosaic (Fig. 20-7).
To the west, a series of pillars and elements of collapsed arches have been discovered, in primary position, that
could be interpreted as associated with a courtyard, or as sub-structures, as has been found in some examples of Late
Antique domestic buildings at neighboring sites like Karabakh or Okuzlu (Aydmoglu and C,::akmak 2011; Mimaroglu
and Aydmoglu 2017; Fornace 2018). Many architectural elements of the arches, as stated above, were re-used in the
masonry of the circular structure.
Figure 20-6. View of the polygonal wall on the northern promontory.
0 4m
Figure 20-7. Aerial view of the wall that cuts the wave motif floor mosaic on the northern promontory.
Ceramics and other materials are still under study. Huge quantities of spacers (amphora necks) have been found;
their function, whether for production activity, or for architectural purposes (as in the filling of the arches ' extrados),
has yet to be established.
Deeper study is also needed to understand the impact of the construction of the small church on the promontory's
edge, and the chronological relationship between the latter and the structures which came to light in the campaigns
ongoing since 2017. Research in this area is of particular relevance because this was the sector where the city's
earliest core developed. For this reason, along with excavations, a survey of the visible structures on the promontory
has been carried out, and the results have been recorded on the GIS map and in the archive (Fig. 20-8).
Figure 20-8. GIS map showing the enhancements based on recent work.
Excavations involved the public area, where a basilica (known as the Basilica of the Great Baths, Figs. 20-8 to 20-
10) had been built in the palaestra of the bathing complex, which was a part of the public area. The 2018 and 2019
campaigns were dedicated to the narthex aTea and the southern nave. The church has been investigated since 2011
(Naspi and Conti 2014); it has three naves, with an apse in the central nave and one in the northern nave, which houses
a baptistery; several burials have been found, among which is the tomb of the nephews of Tarsus's metropolites,
according to the inscription on the cover slab (Borgia 2014), and one close to the transept, in line with the altar.
The naves were connected through passages featuring colonnades of white marble columns. The church is also
equipped with an annex, accessible through a door in the northern nave. This northern door was then filled in, and, in
a later phase, the annex was reached through a door opening towards the service rooms in the apse area. The northern
annex has no built floor; the fa9ade of a rock-cut tomb dating to the earliest phase of the necropoleis is left visible, and
raw rock protrudes into the annex, whose space is divided into three sections by built arches. The basilica's floors are
made of carpets of opus sectile; in the central and southern nave, fragments of a polychrome mosaic have also been
discovered, probably built during the first phase of the church.
In the 2018 and 2019 campaigns, the narthex and the church's fas;ade were excavated, together with the southern
nave. In the southern nave, also paved with carpets of polychrome opus sectile, a huge burial was found, the opening
of which consists of a limestone slab equipped with a bronze ring; it contains the bones of several individuals. It has
also been possible to excavate the southern entrance of the church, which had been created by re-cutting blocks from
the outer wall of the bath structure to create a doorway.
The narthex has a stairway in the northern comer, allowing access to the upper floor, under which was a burial in a
plain coffin (the cover slab was found broken, and the burial pillaged). The basilica' s fas;ade has two doors, on axis
with the colonnades dividing the aisles, while, from the narthex, the entrance to the aisles was possible through three
doors. The door in the central nave was provided with a green marble threshold.
The church was abandoned, as were almost all the monumental complexes of the ancient city, soon after the
middle of the 7th century CE. After the collapse of the church, probably between the 12th and 13th centuries CE, the
central nave was cleared of debris, and a small chapel was erected in the central apse.
Attention has been paid to the church's consolidation, specifically to activities in preparation for a partial
anastylosis of the colonnade, and to facilitate the accessibility of the monument. To this end, the columns that had
been placed on the stylobate after the collapse of the building have been removed, and the stylobate itself has been
cleaned and consolidated, in order to allow the replacing of the better-preserved columns in their original position.
The floors have also been cleaned and consolidated, and covered with small pebbles, in order to preserve them, and to
enable, at the same time, visitor access to the building.
@ ," """"""""""",,__________
Figure 20-9. Plan and photo of the Basilica of the Great Baths, narthex and southern nave.
Figure 20-10. Basilica of the Great Baths.

The modem layout of this area, and particularly the modem road leading from the coastline to the upper part of the
village and other villages in the inland, represent an obstacle both to archaeological research (the road has been
temporarily closed and slightly modified to allow the complete excavation of the basilica), and the understanding of
the relation between this monument and the double-apse basilica built in Byzantine times in the Roman "agora"; the
two churches are so close to each other that they probably represent a basilical complex, conceived as a unity, though
the road separating the two areas must follow a route that existed at least in Roman times.
Unfortunately, the Covid-19 pandemic did not permit us to conduct a 2020 excavation season; nevertheless, the
Turkish unit of the team was able to carry out study of ceramics and osteological materials, and we plan to resume
fieldwork in 2021.
REFERENCES CITED
Aydmoglu, Umit and Umit <;akmak. 2011. A Rural Settlement in Rough Cilicia-Isauria Region: Karakabakh. Adalya
11: 71-101.
Borgia, Emanuela. 2003. Archaeology in Cilicia in the Ancient Travellers' Notes. Olba 7: 41-77.
- . 2014. Elaiussa Sebaste: note su un'iscrizione funeraria dalla basilica bizantina delle Grandi Terme. Scienze
dell'Antichita 20: 59-69.
Equini Schneider, Eugenia, ed. 1999. Elaiussa Sebaste I. Campagne di scavo 1995-1997. Roma: L'Erma di
Bretschneider.
- , ed. 2004. Elaiussa Sebaste IL Un porto tra oriente e Occidente. Roma: L'Erma di Bretschneider.
- , ed. 2010. Elaiussa Sebaste III. L'agora romana. istanbul: Zero Books.
Fomace, Chiara. 2018. Le tecniche edilizie a Elaiussa Sebaste e nel suo territorio: dall'eta ellenistica all'eta proto-
bizantina. Ph.D. thesis, Sapienza Universita di Roma.
- . 2019. Nikai "danzanti" dalla necropoli sud-occidentale di Elaiussa Sebaste. In LEBALTH· Studies in Honour of
Eugenia Equini Schneider, A. Polosa, H.A. K1Z1larslanoglu, and M. Oral, eds., 355-364. istanbul: Ege Yaymlar1.
Laborde, Alexandre and Leon Laborde. 1838. Voyage en Orient par le comte Alexandre de Laborde, redige et publie
par M Leon de Laborde. Paris: Firmin Didot.
Langlois Victor. 1861. Voyage dans la Cilicie et dans !es montagnes du Taurus execute pendant !es annees 1852-
1853. Paris: B. Duprat.
Mimaroglu, Sinan and Umit Aydmoglu. 2017. Daghk Kilikia'da ge9 antik donem k1rsal yerle~im duzenlemesi:
Okuzlu omegi. In Rural Settlements and Urban Centres in Mediterranean During Antiquity. Symposium
Proceedings, Mersin 2016, D. Aydmoglu and A. Morel, eds., 122-141. Mersin: Mersin Universitesi Yaymlar1.
Naspi, Ambra and Marco Conti. 2014. Considerazioni preliminari sulla "basilica delle Grandi Terme" di Elaiussa
Sebaste. Scienze dell'Antichita 19: 51-57.
Tempesta, Claudia. 2013. La citta e le mura: il caso di Elaiussa Sebaste. Scienze dell'Antichita 19: 569-590.
PART II:
SURVEYS
(Photo courtesy of the ~adir Hi:iyiik Project)

CHAPTER TWENTY -ONE
THE 2018-2019 ARCHAEOLOGICAL SURVEY RESULTS

ON THE LOCATION OF SETTLEMENTS CONNECTING
THE MOUNTAIN SECTION OF GIRESUN TO CENTRAL ANATOLIA
SALiH KAYMAK<;:'.I
Our research area is focused on the Giresun province in the eastern Black Sea region. The western border of the
region is the Melet River in the Mesudiye district of the Ordu province (Melanthios) (Ozsait 1995: 483-496, 2005:
263-276.) The eastern borders of Giresun province are the Alucra and Siran districts (Fig. 21-1). At the southeastern
border are the Su~ehri and Akmcilar districts of Sivas province, which also form the eastern borders of central
Anatolia, and the Kelkit River (Lykos) basin (Magie 1950: 177-178; Olshausen and Biller 1984: 137; Kaymak91
2017a). This area is located within the geographical borders of the Kingdom of Pontus (Arslan 2007: 16; Kmae1 2015:
185; Senyurt and Ak9ay 2016: 224).
ORDU
metre
30')4
2800
·2400
2000
1600
1 200
,oo Si VA S
400
i ~ are t l e r
1- Manast,r Ycrlq imi 11 - Alw;lutarla Ycrlc~imi 21- Kliscbdcni Ycrlqimi Ycrlc~im y cri 2018-2019
+
2- Koc;unbai1-Ku*c;u Ycrlqim i 12- KavakliSmta~ Ycrlqimi 22- GOlcUk Yerlc~irni
- - il sm1r1
3- Gcmik.lya Ycrlc*imi 13- ~cb inkarahisar Kalcsi 23- Yumru c,1kra~ Kale Ycrlc~imi Kale
· ·---- il ~e S1 111r1
4- Kayaba*1 Ycrl qimi ·14- Kaleta~ Ye rl e~imi 24- Kurntcpc Mcvkii
Ti..imi.ili.is - - Akarsu
5- K1s1k Yerlc~imi 15- Kurbantcpc Ycrlqim i
16- Kurbantcpc TumulU sU
25- Kab,1ktcpc ycrlc~imi
26- Ah urcu k Kale Ycrlc~imi
~ "'
6- Kalccik Kalcsi '-"" Bmj/ Gol
7-Kaleba~i Ycrle~imi-Mczarl1g1 17- Pohanta~Ycrlqimi 27- Kl iscmrla Ycrl c~imi
8- lso la Tepe Yerlq imi 18- Kliscta~ Ycrlqimi 28 Dikmcntcpc Ycrlc~im i
9- Kanhkaya Mczarhk Alam 19- Kliscburnu Nckropo l Alam
10- Ga\lur Tepe Kuts.11Al,1111 km 10 20
20- Aslan~.lh Nckropoh.i
Figure 21-1. Settlements identified in Sebinkarahisar district during the 2018-2019 survey years.
The 2018-2019 Archaeological Survey Results 277
The Kingdom of Pontus is surrounded by Colchis in the east, Paphlagonia in the west, where the Halys River
(K1z1hrmak) forms the natural border, Pontos Euxeinos (Black Sea) in the north, and Cappadocia in the south. The
archaeological surveys of Giresun province and its districts, and Giresun Island, were carried out in 2018 and 2019;
also surveyed was Sebinkarahisar, which is located in the mountainous part of Giresun province and rests on the
natural routes connecting the eastern Black Sea region to central Anatolia. The Sebinkarahisar district is defined as the
inner Black Sea region (izb1rak 2001: 326), and is also situated in the Upper Kelkit region. The primary goal of this
work was to complement the surveys we conducted in the coastal and mountainous Alucra and <;amoluk districts of
Giresun province (Kaymak91 2019). Our work in the Sebinkarahisar district encompassed the historical natural routes
connecting the coastal part of Giresun to the Kelkit basin (Lykos), with particular focus on connections dating to the
Early Bronze Age, the 2nd millennium BCE, and the Hellenistic-Roman periods; we also investigated Early Bronze
Age and 2nd millennium routes to central Anatolia. In addition, we investigated the extent of the Stepped Tunnel
architecture (see Kaymak91 2019) that we associate with the Mithradates Dynasty (Mithradates VI Eupator of the
Kingdom of Pontus ). Our research, begun in previous seasons, continued in Sebinkarahisar-Su~ehri (Sivas ),
Sebinkarahisar-Koyulhisar (Sivas), Sebinkarahisar-Akmcilar (Sivas), and Sebinkarahisar-Mesudiye (Ordu).
As noted previously (Kaymak91 2019), our survey of settlements in the area shows that the region was densely
populated in the Early Bronze Age and the 2nd millennium BCE; these settlements in Sebinkarahisar were
predominantly located on hilltops, slopes, and on mounds formed on the hills. Settlements are mostly in sheltered
locations in the mouths of valleys, and on the slopes overlooking streams or water resources. Apart from this, mound
type settlements are found on flat areas at high elevations. Low-lying settlements are almost nonexistent. Ceramics
recovered from the 2nd millennium settlements are quite similar to ceramic finds obtained from central Anatolian sites
(Sivas Su~ehri, Akmcilar) in terms of both form and technique. Based on previous and current survey data, it is also
clear that the Early Bronze Age, 2nd millennium BCE, and Iron Age settlements in the mountainous region are denser
than in the north-coastal area. Mountains in the eastern Black Sea and Giresun serve as a barrier between the coastal
zone and mountainous area, weakening the cultural interaction between the two.
THE 2018-2019 SURVEYS: SITES AND REGIONS INVESTIGATED
Our 2018-2019 surveys were conducted in the southeast of the region that constitutes a large part of the
mountainous part of Giresun province. Survey work conducted in the Sebinkarahisar district in 2019 identified
numerous settlements: Kaleta~ Settlement, Kurbantepe Settlement, Kurbantepe Tumulus, Pohanta~ Settlement,
Kiliseta~ Settlement, Kiliseburnu Cemetery Area, Aslan~ah Necropolis Area, Kilisebeleni Settlement, Golcuk
Settlement, Yumurcakta~ Fortress Settlement, Kumtepe-Girls' Cemetery, Kabaktepe Settlement, Ahurcuk Fortress,
Figure 21-2. Ahurcuk Fortress.

278 Chapter Twenty-One: Kaymak91
Kilisetarla Settlement, and Dikmen Tepe Fortress. In our previous research in the region, we visited Kalecik Fortress,
Kaleba~1, Aluclu Tarla, K1s1k, the Kavakhgmta~ settlements, Sebinkarahisar Fortress, and isola Hill Settlement
(Kaymakp 2019: 218-230, 2020: 49-52) (Fig. 21-1). Our 2018-2019 research primarily began in the area within the
borders of Sebinkarahisar-Akmcilar-Sivas and Sebinkarahisar-Su~ehri-Sivas. The Ahurcuk Fortress was found at a
distance of 10 km southwest of the district center (Fig. 21-2). It is located within the borders of Ahurcuk village, on a
hill lying 300 m north of the Su~ehri-Sebinkarahisar road, which connects central Anatolia to the eastern Black Sea
region. It is in a position that likely controlled the natural route in ancient times. Surrounded by fertile plains on all
sides, the fortress presented examples of pottery dating to the Early Bronze Age, the 2nd millennium BCE, the Late
Iron Age, and the Roman period. It has an elevation of 50 m from the base to the peak.
Figure 21-3. Golctik settlement.
Research continued at Sebinkarahisar Fortress in

2019, which was first investigated in 2018. The
fortress, located southeast of the district center
during the 2018 season, was built on a high and
sheltered hill (Prokopios: 111.IV.6-9; von Gall 1967:
507; Bryer and Winfield 1985: 145-151; Karpuz
1989: 4). Apart from the Stepped Tunnel that we
identified in the fortress, rock-cut boat tombs with
dimensions of approximately 2 x 1 m were also
revealed. Dating to the Roman period (Cumont and
Cumont 1906: 82), these tombs are positioned in
doubles or triples, at a distance of ca. 10-15 m from
one another. There are steps carved into the rock
between the tombs. To the southeast of these tombs
lie rock-carved cisterns with dimensions of 2 x 2 m.
Though the fortress was built on the Early Bronze
Age settlement (Karpuz 1989: 326), pottery from the
2nd millennium BCE, the Late Iron Age, and
Hellenistic and Roman periods was also encountered.
The Golciik settlement, situated within the
borders of Tepeltepe village, 20 km southeast of the
Sebinkarahisar district center, is approximately 200
x 200 m in size (Fig. 21-3). It is located in an area
surrounded by fertile agricultural fields north of the
Kelkit River. Agriculture is carried out in the area,
so the extent of the settlement could not be
determined. The settlement's surface produced
materials from the 2nd millennium BCE and the
Figure 21-4. Golciik settlement: ceramics from Hellenistic to Roman periods (Figs. 21-4 and 21-5).
Hellenistic and Roman periods.
Gavurtepe Sanctuary, which we identified in 2018, is

5 km southeast of the settlement. Although illicit
digging has damaged this area, the sanctuary shows
similarities with the Mavlavita Chapel in Trabzon
(Bryer and Winfield 1985: 214-217, East Wall and
Apse Arch, Manglavita, Rock-cut Chapel). The
Kilisebeleni settlement was discovered on a
predominant hill 1.5 km northwest of Tepeltepe
village. The c;at River flows at a distance of 800-
900 m northeast of the settlement. A small number
of Early Bronze Age and 2nd millennium BCE to
Roman period ceramics were encountered in the
area.
Our research continued at the Kabaktepe
settlement (Fig. 21-6), which is located within
Akviran village, 20 km northeast of Sebinkarahisar,
Figure 21-5. Golcuk settlement: 5 km south of Akmcilar district, and 7 km southwest
bead from Roman period. of Su~ehri district. It is situated on a prominent hill
on the border of Sivas-Akmcilar-Su~ehri and Giresun-Sebinkarahisar. Surrounded by fertile agricultural lands on the
edge of the Kelkit River, the settlement measures 300 x 300 m in area. Ceramics and a few items of other material
culture from the Early Bronze Age, the 2nd millennium BCE, and Roman period were found on the surface (Figs. 21-7
and 21-8). Overlooking the Kelkit basin and its roads to the southeast and the Sebinkarahisar-Su~ehri basin to the
northeast, there are also fertile agricultural fields west of the settlement and on the edge of the Kelkit River.
Figure 21-6. Kabaktepe settlement.
Yumurcakta~ Fortress was identified northeast of Kabaktepe settlement, again on the edge of the Kelkit River,
located in a rocky area on a pronounced hill. The western terrace of the fortress is an area suitable for settlement. The
modem dam structure right next to the fortress has raised the water level to the northern skirts of the fortress. A 4 m-
long row of walls, with khorasan mortar, was found below the peak. Lying on a north-south axis, it was formed with
simple cut stones and is 70 cm thick. There is masonry east of the hill, and a cistern was also found on the northern
slopes, approximately 2 x 2 m in size. The northern section of the cistern has been completely destroyed by illicit
excavation. Plaster residue 0.50 mm in depth can be seen inside it. The cistern's depth is around 3 m. The peak at this
fortress site is 50 x 50 min area and has been terraced. The agricultural area is 100 x 100 min size. On the terraces of
the fortress, ceramics from the Early Bronze Age, the 2nd millennium BCE, the Iron Age, and Hellenistic and Roman
periods were found.
280 Chapter Twenty-One: Kaymak91
Figure 21- 7. Kabaktepe settlement: Figure 21-8. Kabaktepe settlement:

Hellenistic-Roman ceramics. Hellenistic-Roman glass handle.
Kurbantepe Tumulus is located in Ta~cih village, which is approximately 25 km northwest of Sebinkarahisar
district on the Sivas-Su~ehri border. A controlled excavation was carried out by the Giresun Museum in the northern
area of the tumulus in 2018. However, no architectural remains or material culture were found. Later, the dromos
entrance of the tumulus was unearthed through illegal excavation carried out on the northwestern section of the hill.
Measurements were taken on the entrance to the tumulus, but adequate information could not be gathered on the
section of the room as the waste soil had been thrown into the tumulus during these illegal excavations.
Kiliseta~ settlement, which is an important settlement within Ta~ph Village, was found to the north of the Kelkit
River (Fig. 21-9). The settlement spreads over a very large area. We encountered a rock-carved sanctuary 2 x 1 min
size and 60 cm deep on a large hill to the northwest. A similar damaged area was revealed to the north of this area.
Figure 21-9. Kiliseta~ settlement.

Our survey revealed that there were water channels in the rock connecting the sacred areas. Another 3-step area
was found between the sanctuary and the settlement. The approximate length of the steps is 25 x 200 cm with a depth
of 5 cm. An illicit pit was dug at the end of this stepped area (2 x 1 min area with a depth of 150 cm). Pottery from
the Early Bronze Age, the 2nd millennium BCE (Figs. 21-10 and 21-11 ), the Late Iron Age, and Hellenistic and Roman
periods was found at the settlement. Most of the ceramics are undecorated, though a few pieces of painted pottery
from the Iron Age and Roman period were encountered. Some dark-faced vessels from the Early Bronze Age have
incised decoration on their surfaces (Fig. 21-10). It is estimated that these types of decorated vessels reflect a tradition
peculiar to the region.
. - .
~•:;t;.:..--1 .
,,
. ..;,t
~~-~
. .:·
Figure 21-10. Kiliseta~ settlement: ceramics from the Figure 21-11. Kiliseta~ settlement: a ceramic piece
Early Bronze Age and 2nd millennium BCE. from the Early Bronze Age.
Figure 21-12: Dikmentepe Fortress and Sebinkarahisar Fortress.
Located on a natural hill within Aslan~ah village, west of Ta~e1h village and on the border of Sivas Su~ehri, the
Kilisebumu Roman Necropolis area has an important position in the valley, situated at the junction point of the
U9k6prti and Kelkit Rivers. These flattened tombs of the necropolis have a simple rectangular plan. The southern half
of the hill was completely destroyed as a result of soil removal. Tiled tombs are clearly exposed in the section, and the
tomb tiles and skeletons are scattered around due to present-day activities. Illicit digging pits were also encountered on
the upper part of the hill. Pottery from the Early Bronze Age, the 2nd millennium BCE, and Roman period were
identified. In addition, a large number of human bones were revealed. On the surface of some dark-faced vessel
282 Chapter Twenty-One: Kaymaks;1
sherds, there are incised decorations peculiar to this region. As seen from the cross-section, it is clear that the
settlement is in a higher location, and ceramic pieces have accumulated at the foot of the hill due to natural forces such
as rain and wind, and that the area has largely been destroyed. Measurements could not be taken because the borders
of the settlement could not be determined.
Our work continued at Dikmentepe Fortress, located within Ekecek village (Fig. 21-12), 5 km west of the district
center. The fortress, which overlooks the Giresun-Sivas-Akmcilar route from an elevated point, is located at a 4 km
distance from Sebinkarahisar Fortress. It is situated on the road connecting the central Anatolia region to the eastern
Black Sea region through the Kelkit basin. The summit is rocky, and illicit digging is present. The lower skirts of the
summit are flat. A khorasan mortar wall structure surrounding the hill was identified, but its exact dimensions could
not be determined due to the destruction created by natural causes and illicit excavations. The hill is conical, and the
summit can be reached by a modem road.
NEW EVALUATIONS ON ROUTES CONNECTING THE MOUNTAINOUS

REGION OF GiRESUN PROVINCE TO CENTRAL ANATOLIA
Archaeological studies in Giresun province, in the eastern Black Sea region (Sagona and Sagona 2004: 147, 150-
151, 154-155), began with archaeological surveys and excavations carried out in 2009 (Doksanalti et al. 2010: 143-
162). These projects were mostly carried out in the coastal area ofGiresun province and on Giresun Island (Doksanalt1
et al. 2011: 117- 145; Doksanalt1 and Ekici 2017: 389--403.) Research on the mountainous part of the region was
initiated in 1944 by 1.K. Kokten, but no settlements were identified (Kokten, 1944: 678). However, during our
research in <;amoluk, Alucra, and Sebinkarahisar in this area of Giresun and in the Kelkit basin, many important
settlements dating to the Early Bronze Age, the 2nd millennium BCE, the Early and Late Iron Ages, and the Hellenistic
and Roman periods, were revealed (Kaymaks;1 2017a: 99- 105, 2020: 33- 59) (Table 21-1).
No. Settlement Name District Villae.e
1 Kaletas Settlement Sebinkarahisar Kaletas
2 Kurbanteoe Settlement Sebinkarahisar Tasc1h
3 Kurbantepe Tumulus Sebinkarahisar Ta~1h
4 Kilisetarla Settlement Sebinkarahisar Tasc1h
5 Kiliseta.s Settlement Sebinkarahisar Tasc1h
6 Ki!isebumu Cemetery Area Sebinkarahisar Aslansah
7 Aslansah Necropolis Area Sebinkarahisar Aslansah
8 Kilisebeleni Settlement Sebinkarahisar Teoelteoe
9 Golci.ik Settlement Sebinkarahisar Teoelteoe
10 Yumurcaktas Fortress Settlement Sebinkarahisar Yumurcaktas
11 Kumtepe - Girls' Cemetery Sebinkarahisar Yumurcaktas
12 Kabaktepe Settlement Sebinkarahisar Akviran
13 Ahurcuk Fortress Sebinkarahisar Ahurcuk
14 Pohantas Settlement Sebinkarahisar Derekov
15 Dikmen Tepe Fortress Sebinkarahisar Ekecek
16 Sebinkarahisar Fortress Sebinkarahisar Merkez
Table 21-1. Settlements identified in the Sebinkarahisar District in the 2018-2019 seasons.
The topography of Giresun province is elevated and rough. The eastern Black Sea and Giresun Mountains, which
lie parallel to the coast, make it difficult to travel between the coastal zone and the interior mountainous area. In
previous research carried out in the Sebinkarahisar district (Sagona and Sagona 2004: 147, 150-151, 154-155), many
important settlements were dated to the Early Bronze Age, the 2nd millennium BCE, and the Hellenistic and Roman
periods. The presence of materials from foreign cultures among the ceramics collected from the surface of the
settlements located on natural roads and passages in Sebinkarahisar indicates that these roads may have been used for
commercial purposes (Okse 2006: 16.).
Our most recent research focused on the western part of the district in order to expand the results obtained from the
2017-2018 surveys we conducted in the eastern section (Kaymaks;1 2019: 218-230). As noted in the introduction, our
aim in the ongoing research in the area is to reveal the Early Bronze Age, 2nd millennium BCE, and Hellenistic--
Roman connections between the interior and coastal areas. Further, we have continued our investigation of the stepped
tunnel architecture that we associate with the Mithradates Dynasty (Strabo Xll.3.28; Arslan 2007: 4).
The Sebinkarahisar district (Baykara 2000: l; Dan1k 2004: 52), densely populated in prehistory (izb1rak 2001:
326), was also well-known in later times (Hamilton 1842 I: 262; Xenophon, Anabasis V.3.24; Pliny, Natural History
VI.4; Arrian, Periplus 17.4). It is situated in the upper Kelkit region and extends over a large area with mountains,
valleys, and plateaus. The Canik Mountain range (Karagol Mountain) and Giresun Mountains to the north, which have
a high and steep topography, are surrounded by the Sar1s;is;ek-Berdiga Mountain range to the southeast, the southern
slopes of the Kelkit basin to the south, and Egme to the southwest.
In Sebinkarahisar, we attempted to identify the settlements that show parallels to the Kelkit River basin, which is
an extremely efficient water resource and provides for agricultural productivity, and is the territory that forms the
northeastern border of the region. Constituting a large part of the mountainous part of Giresun, Sebinkarahisar is a
geographical transition point between the eastern Black Sea region and central Anatolia. The region, which has a very
rough topographic structure, is on the border of the ancient territory of the East Kaska (von Schuler 1965: 62; Murat
1998: 436), with an eastern border at IJayasa. Surveys carried out here revealed that the natural routes passing through
the region may have been used to connect the eastern Black Sea region to eastern and central Anatolia for ages. The
roads beginning in the southern part of Sebinkarahisar district extend to Akmcilar (Sivas) and pass through the Kelkit
basin to the middle Black Sea region via Su~ehri, Koyulhisar, and Re~adiye. Highways that are in use today also
follow these natural routes.
According to the material culture from the Gavur Fortress and Topalaksag settlement (Giresun-Alucra), which
were identified in the archaeological surveys we conducted in previous years, the extensive Karaz culture extends over
the northwestern area of Giresun-Alucra and C,::amoluk, and according to research in Sivas, it extends to Zara, Kangal,
and Yild1zeli; research in Mesudiye district of Ordu suggests that the cultural border reaches Mesudiye (Ozsait 1992:
357-376, 1993: 311-330, 1994: 459-482, 1995: 483-496). We can say that the Karaz-Early Transcaucasian Culture
extends over the Sebinkarahisar district of Giresun through Siran and goes on to cover Sivas-Su~ehri, and from there
stretches out to the southwest (Meier-Aksoy 2012). The Gavur Fortress settlement is located between Siran and
Sebinkarahisar, and serves as support for the route we have identified that travels from the eastern Black Sea to central
Anatolia. Siran, Alucra, Sebinkarahisar, and Su~ehri are positioned on a natural road-valley within the Siran-Kelkit
basin. Based on the survey data collected from the region and its surroundings (Ozsait 2005: 263-276), we have
previously suggested that the Karaz-Early Transcaucasia Culture extended to the southwestern part of Su~ehri district
beyond Sebinkarahisar, and then continued to the northwest, to Mesudiye district (Ak9ay et al. 2019: 171-192;
Kaymak91 2017b: 45-46).
Grafik Ba~l1g1
60
so
40
30
20
•••• 1•• 1 ■ ■I
10
■ S YR 4/3 (Reddish 8rown)·(K1rm1z1m51 kahverensi) ■ S YR 4/4 (Reddish 8rown)•{Kirm1z1m51 kahverengi)
■ S YR 4/2 (Dark Reddish Grav) -(Koyu krrm,nmg gri) ■ S YR 6/6 (Reddish YellowHK1rm,nmsr san)
■ S YR S/4 (Reddish 8rown),(Kirm,z1mg kahverengi) ■ S YR 5/6 (Yellowish Red){Sanms, kirmm)
■ 5 YR 6/4 (Light Reddish BrownHApk krrmmmg kahvereng1) ■ S YR S/S (Reddish Brown) -(Kirmmmg kahverengi)
■ S YR 5/1 (Grav)·( Gri) ■ 7.5 YR 6/4 (light BrownHAt,k kahverengi)
■ 7.S YR S/6 (S<rong Brown){Koyu kahverengi) ■ 7.5 YR S/4 (S<rong BrownHKoyu kahverengi)
■ 7.5 YR 6/6 (Reddish Yellow){Kirm111mg san) ■ 7.5 YR 7/6 (Reddish YellowHKirm,vmg san)
■ 7.S YR 4/3 (Brown),(Kahverengi) ■ 7.5 YR 6/4 (BrownHKahverengi)

■ 7.S YR 7/4 (PinkHPembe) ■ 7.S YR S/4 (Brown),(Kahverengi)
■ 2.S YR 4/4 (RedHKirm,z1) ■ 2.S YR 6/6 (light Red) -(A1;1k kirmw)
■ 2.5 YR 5/6 (Red)·(Kirm,n) ■ 2.5 YR S/8 (Red){Kirmm)
■ 2.S YR 7/6 (Light Red) -(A1;1kkirmm) ■ 2.S YR S/4 (Reddish BrownHKrrmmms, kahverengi)
Figure 21-13. Paste and slip color variations of ceramics encountered in the 2018-2019 seasons
(based on the Munsell Catalogue).
We learn about the state of the mountainous area in the 2nd millennium BCE from the records of the Hittite king
Mursili II (1344-1306 BCE). The records state that cities such as Dukkamma (Sebinkarahisar) and Aripsa (Giresun)
were first occupied by the Hittites. During this period, the region including Gumu~hane and Sebinkarahisar was
known as the Azzi country. During the Roman period, the military road that crossed Anatolia in an east-west direction
ended in Satala (Sadak village of Kelkit), extended to Siran through Colonia (Sebinkarahisar), and then continued on
to Kelkit. While Sebinkarahisar, which was an important settlement center during the Roman Empire period, is
mentioned in these records, the city was also called Koloneia or Coloneia, and was founded by Pompeius Magnus who
defeated Mithradates during the collapse of the Pontic State in the 1st century BCE (Ramsay 1960: 56).
Our survey work identified numerous examples of ceramics (Fig. 21-13) and noted that the Early Bronze Age
pottery from the mountainous part of Giresun was handmade and had thin dark gray and reddish brown paste (2.5 YR
5/6 [Red]; 7.5 YR 5/4 [Strong Brown- Dark brown]), with a very high quality black slip. These ceramics from the
Giresun region have similar properties to those recovered during our research in Glimii~hane, Ordu, and Sivas. This
suggests that these areas were occupied by the same/related cultural groups who had settled along these transportation
routes. Most of the pottery that we dated to the 2nd millennium BCE in the settlements that we surveyed is red, reddish
brown (2.5 YR 5/6 [Red]; 7.5 YR 5/4 [Strong Brown-Dark brown]; 5 YR 6/4 [Light Reddish Brown]), or self-slipped
and burnished. The paste of these wares has mostly brown and red tones. Among the pottery, there are sherds from the
Iron Age, mostly dating back to the middle and late phases of this era. These samples, which are thought to be oflocal
production, are very characteristic of this region. Hellenistic and Roman ceramics are represented by a few settlements
and examples. These ceramics were well-fired and thin, and had a reddish yellow paste (5 YR 6/6 [Reddish Yellow]).
The density of these wares is higher, especially in the Golciik and Kabaktepe settlements in the mountainous area.
CONCLUSION
The settlements that we identified in Sebinkarahisar district are large-sized hilltop and slope settlements that have a
dense dispersal across the landscape. As noted above, the eastern Black Sea region and Giresun Mountains serve to
weaken cultural relations, as they form geographical barriers between the coastal and mountainous zones. The Early
Bronze Age and 2nd millennium BCE settlements are located in deep river valleys extending from south to north,
along the route of historical natural roads traversing the Giresun Mountains. The settlements are mostly concentrated
on the banks of the streams feeding the Kelkit River; there is a decrease in the density of settlements in the forested
areas due to the difficulties of the geographical terrain.
Inhabited right after the Early Bronze Age and the 2nd millennium BCE, small settlements demonstrate a scattered
pattern that is comprised of structures built in close proximity to one another. The Late Iron Age and Hellenistic-
Roman settlements, which were close to one another, were mostly located in the plains near the streams feeding the
Kelkit River. No traces of settlement were identified in the eastern and southern areas of Sebinkarahisar district, which
constitutes a large part of the mountainous terrain. North of K1hs;kaya Dam, 2nd millennium BCE settlements are
almost nonexistent. This region was likely covered by dense forests in the past. The 2nd millennium settlements are
instead concentrated on the slopes bordering the northern Kelkit basin.
The settlements that we have identified in our research in the eastern part of the Kelkit basin are, as noted, in the
deep north- south valleys, and are also located at a high altitude on the historical natural route stretching between the
Giresun Mountains. These factors make it very difficult to prevent illicit excavation in these areas. Because most of
the cultural assets are situated in mountainous forest areas, they are distant from village centers and are isolated due to
the topography. This enhances the level of destruction on the tumuli and other archaeological sites. Information is
consistently relayed to local governments and law enforcement agencies regarding destruction in these areas.
Our current understanding of the region, based on the surveys carried out in the Sebinkarahisar mountainous area,
which was first settled in the Late Chalcolithic, is that the region was densely inhabited during the Early Bronze Age
and 2nd millennium BCE, and that Late Iron Age, Hellenistic and Roman period settlements, though fewer in number,
were concentrated in the western part of Sebinkarahisar, on the border of Su~ehri and Akmc1lar-Sivas. When the
current provincial and district borders are ignored, it becomes clear that the higher areas in the east had a larger
number of the Early Bronze Age and 2nd millennium BCE settlements, while the lower regions toward the west
(central Anatolia) feature mostly Late Iron Age, Hellenistic, and Roman settlements.
During the 2019 survey season, we focused on the connection points of Sebinkarahisar, Sebinkarahisar-Alucra,
$ebinkarahisar-Akmc1lar (Sivas), and Sebinkarahisar-Su~ehri (Sivas), which form the mountainous part of Giresun
province. In general, the settlements in eastern Sebinkarahisar and the settlement areas up to Akmcilar and Su~hri on
the western border show parallels with the settlements that feature Early Bronze Age and 2nd millennium BCE
cultures. The fortresses situated on the natural route along the Kelkit River, extending on an east-west axis, are in very
close proximity and face one another. Settlements north of the Kelkit River and parallel to it were mostly identified as
slope settlements.
Though Sebinkarahisar is at the junction of the road leading to central Anatolia via Giresun, Kumbet, and Sivas
and the road from Kelkit to <;oruh Valley (Strabo, Geographica XII.III.18), its mountainous and rough landscape
prevents transportation to its outlying areas. As a result of this geographical feature, the district is only suitable for
transportation routes which travel from the west, and is connected to the west via Su~ehri-Koyulhisar-Erbaa. The
shortest and best road connecting the Giresun coastal area to the interior is the modem road between Giresun and
Sebinkarahisar, which was used as a caravan route in historical times. We think that this route was used in the same
way in ancient times as well (Darkot 1988: 789). Undoubtedly, Giresun's connection to the south is realized by the
Giresun-Dereli-$ebinkarahisar road. This road connects to Giimii~hane through the Alucra district and to Sivas,
Erzincan, and Tokat, which are located further south. Therefore, our research has highlighted an important passage
connecting the eastern Black Sea region to eastern and central Anatolia.
The isola Hill, K1slk settlement, and Sebinkarahisar Fortress that we identified in 2018, and Ahurcuk Fortress,
Golciik settlement, Yumurcakta~ Fortress, Kiliset~ settlement, and Kabaktepe settlement, which we identified in
2019, are situated on the prominent hills that overlook the route connecting the eastern Black Sea region to central and
The 2018- 2019 Archaeological Survey Results 285
eastern Anatolia. These settlement units are established at points on the north and south of the Kelkit River, at a
distance of 20- 30 km from one another. This distance between the settlements could be traveled by foot during the
periods of occupation (Okse 2006: 19). The road reaches the districts of Alucra, <;amoluk, Gumii~hane, $iran, and
Kelkit (<;igdem 2012: 32) via $ebinkarahisar in the east and extends from there to Erzurum via Bayburt. In the west, it
is connected from $ebinkarahisar to Akmc1lar, Su~ehri, and Koyulhisar districts in Sivas province, and continues to
the Re~diye district in Tokat province. Therefore, the fact that isola Hill (Giineygoren), K1s1k settlement (Balcana),
and the $ebinkarahisar Fortress were found in places along these routes confirms that they were in existence. The road
continues towards the city of Sebasteia (Sivas) to the southwest of our research area, which was also an important
Byzantine-period military center in the region. The city of Sebasteia was a meeting point at the intersection of military
roads extending in different directions. The Sebasteia road passed through the cities of Komana (Tokat) to the north
and Nikopolis ($ebinkarahisar/Giresun) to the northeast, and continued to the city of Kerasous (Giresun), which is
located within the larger city ofNikopolis (Ramsay 1960: 22-23, 216-242).
ACKNOWLEDGEMENTS
The archaeological surveys we conducted in the region under the permission of the General Directorate of Cultural
Heritage and Museums were supported by the Turkish Historical Society, Erzincan Binali Yildmm University
Scientific Research Coordinatorship, Giresun Govemorate, Giresun Archaeology Museum, $ebinkarahisar District
Govemorate, and $ebinkarahisar Municipality. I would also like to thank Giresun Museum Director Hulusi Giilec; and
Expert Gokhan Giirnal, who always contributed to our research with a solution-oriented cooperation. I would like to
thank Archaeologist Serpil Sandalc1, who participated in our research as the Ministry Representative, and my team
members Archaeologists Giine~ Ozdal, Mehmet Serdar D~demir, Tugc;e Merve Zina!, hem Pekacar, and Elif
<;aglayan, and to all other institutions and organizations and their employees who provided support to us.
REFERENCES CITED
Arrian. 2005. Arrianus 'un Karadeniz Scyahati (Arriani Periplus Ponti Euxini). Yor. M. Arslan, trans. istanbul: Odin
Yaymc1hk.
Akc;ay, Atakan, S. Yucel $enyurt, Seda Kara, Leyla Yorulmaz, Umut Zoroglu, and A. Emirhan Bulut. 2019. Ordu-
Melet Irmag1 Havzas1 Yuzey Ar~tirmas1 2018 Y1h <;ah~malar1. Ara~flrma Sonu9larz Toplantzs1 37(1 ): 171- 192.
Arslan, Murat. 2007. Mithradates VI Eupator: Roma 'nm Biiyiik Dii~mam. istanbul: Odin Yaymc1hk.
Baykara, Tuncer. 2000. Bir Selc;uklu Devri Turk $ehri Olarak Karahisar-1 $arki veya $ebinkarahisar. $ebinkarahisar I.
Tarih ve Kiiltiir Sempozyumu, 1-4. istanbul.
Bryer, Anthony and David J. Winfield. 1985. The Byzantine Monuments and Topography of the Pontos I. Dumbarton
Oaks Studies 20. Washington, D.C.: Dun1barton Oaks.
<;igdem, Siileyman. 2012. Eskic;ag'dan Ortac;ag'a Gilinii~hane. Gumii~hane: Gilinu~hane Valiligi, ii Kiiltiir
Miidiirliigii.
Cumont, Franz and Eugene Cumont. 1906. Voyage d'Exploration Archeologique dans le Pont et la Petite Armenie.
Studia Pontica II. Bruxelles: H. Lamertin.
Dan1k, Ertugrul. 2004. $ebinkarahisar Kalesi. Sana! Tarihi Dergisi 13(2): 51- 78.
Darkot, M. Besim. 1988. Giresun. Tiirkiye Diyanet Vakfi ls/am Ansiklopedisi, 4. Cilt. istanbul: Milli Egitim Basimevi.
Doksanalt1, M. Ertekin, Erdogan Asian, and t. Mete Mimiroglu. 2010. Giresun Iii ve Giresun Adas1 Arkeolojik Yuzey
Ara~t1rmalar1. Ara~hrma Sonu9larz Toplantzs1 28(2): 143-162.
- . 2011. Giresun iii ve Giresun Adas1 Arkeolojik Yuzey Ara~tumalar1. Ara~tzrma Sonu<;larz ToplantlSl 29(2): 117-
145.
Doksanalt1, M. Ertekin, and Makbule Ekici. 2017. Giresun-Aretias/Khalkeritis Adas1 Kaz1lan Seramik Buluntular1.
Pamukkale Oniversitesi Sosyal Bilimler Enstitiisii Dergisi 27: 389-403.
Hamilton, William John. 1842. Researches In Asia Minor, Pontus, and Armenia with Some Account of Their
Antiquities and Geology, Volume I. London: John Murray.
tzbuak, Re~at. 2001. Tiirkiye. istanbul: Milli EgitimBakanhg1 Yaymlar1.
Karpuz, Ha~im. 1989. ;Jebinkarahisar. Ankara: Kiiltiir Bakanhg1.
Kaymakc;1, Salih. 2017a. Ke/kit Havzasz'mn Eski<;ag Tarihi ve Arkeolojisi. Konya: Komen Yaymlar1.
- . 2017b. Dogu Karadeniz Bolgesinde Karaz-Erken Transkafkasya Kiiltiirii' ne ait Yeni Bir Yerle~im: Alucra-Gavur
Kalesi, Giresun. Tiirkiye Bilimler Akademisi Arkeoloji Dergisi (TUBA-AR) 21: 25- 34.
- . 2019. 2017- 2018 Results of Archaeological Survey in Districts Bordering the Kelkit Basin of Giresun. In The
Archaeology of Anatolia, Volume Ill: Recent Discoveries (2017- 2018) , S.R. Steadman and G. McMahon, eds.,
218-231. Newcastle Upon Tyne: Cambridge Scholars Publishing.
- . 2020. Arkeolojik Yiizcy Ara~hrmalan l~1gmda Giresun. Ankara: Bilgin Kiiltiir Sanat Yaymlan.
Kmac1, Mesut. 2015. Antikc;agdaPontos Bolgesi'nin Tarihsel Cografyas1. Cedrus III: 181- 229.
Kokten, i.K. 1944. Orta, Dogu ve Kuzey Anadolu'da Yapilan Tarih Oncesi Ar~tirmalar. Belleten 8/32: 659-680.
Magie, David. 1950. Roman Rule in Asia Minor to the end of the Third Century after Christ I-II. Princeton: Princeton
University Press.
Meier-Aksoy, Y~in. 2012. Dogu Karadeniz'de Erken Transkafkasya Kulti.iri.i: Ard Btilge Dzerinden Bir
Degerlendirme. Uluslararas1 Dogu Anadolu Gilney Kafkasya Kiilturleri Sempozyumu, Bildiriler 2: 258- 265.
Murat, Leyla. 1998. Hitit Dilnyasmda Ga~kalarm Yeri. In 111. Ulus/ararasz Hititoloji Konwesi Bildirileri, <;orum, 16-
22 Eyliil 1996, 435--443. Ankara.
Olshausen, E. and J. Biller. 1984. Historisch-geographische Aspekte der Geschichte des Pontischen und Armenischen
Reiches. Tei/ 1, Untersuchungen zur historischen Geographie von Pontos unter den Mithradatien. Beihefte zum
Tiibinger Atlas des V orderen Orients 29/ 1, Reihe B. Wiesbaden: Dr. Ludwig Reichert.
Okse, A. Tuba. 2006. K1Z1lrrmak ve Fuat Havzalarm1 Birbirine Baglayan Eski Kervan Yollar1. Bilig 34: 15- 32.
Ozsait, Mehmet. 1992. 1990 Y1lmda Ordu-Mesudiye yevresinde Yap1lan Yiizey Ara~hrmalar1. Ar~tzrma Sonur;lan
- . 1993. 1991 Y1h Ordu-Mesudiye Yiizey Ar~tlfillalar1. Ara~tzrma Sonur;lan Top/antzsz 10: 311- 330.
- . 1994. 1993 Y1h Ordu-Mesudiye ve Sivas-Koyulhisar Yiizey Ar~tumalar1. Ara~tzrma Sonur;lan Toplantzsz 12:
459--482.
- . 1995. 1994 Y1h Antalya-Korkuteli Yiizey Ar~hrmalar1. Ar~tmna Sonur;Iarz Toplantzsz 13: 293-315.
- . 2005. 2003 Y1h Amasya, Samsun ve Ordu Illeri Yiizey Ar~trrmalan. Ar~tzrma Sonur;larz Toplantm 12(2): 263-
276.
Pliny the Elder. 1938-1971. Natura/is Historia. vols. 1-X. H.R. Rackham, W.H.S. Jones, and D.E. Eichholz, trans.
Cambridge: Loeb Classical Library.
Ramsay, William M. 1890. The Historical Geography ofAsia Minor. London: John Murray.
Ramsay, William. M. 1960. Anadolu'nun Tarihi Cografyasz. Istanbul: Milli Egitim Bas1mevi. (Turkish translation of
William M. Ramsay, The Historical Geography ofAsia Minor. London: John Murray, 1890.
Sagona, Antonio and Claudia Sagona. 2004. Archaeology at The North-East Anatolian Frontier, I: An Historical
Geography and a Field Survey of the Bayburt Province. Ancient Near Eastern Studies Supplement Series 14.
Louvain and Paris: Peeters.
Strabo. 1993. Antik Anadolu Cografyasz, (Books XII, XIII, XIV) (Geographika). Adnan Pekman, trans. Istanbul:
Arkeoloji ve Sanat Yaymlar1.
~nyurt, Silleyman Y. and Atakan Akyay. 2016. Kurul Kalesi (Ordu) VI. Mithradates Donemi Yerle~imi Uzerine On
Degerlendirmeler. Seleucia 6: 221-248.
von Gall, Hubertus. 1967. Zu den kleinasiatischen Treppentunneln. Archiio/ogischer Anzeiger 82: 504-527.
von Schuler, E. 1965. Die Kaskiier: Ein Beitrag zur Etnographie des a/ten Kleinasien. Untersuchungen zur
Assyriologie und vorderasiatischen Archaologie 3. Berlin: De Gruyter.
Xenophon. 2015. Anabasis. Ari Cokona, trans. Istanbul: Turkiye 1~ Bankasi Kiiltur Yaymlan.
CHAPTER TWENTY-Two
MOUNTAINS AS CONNECTED LANDSCAPES OF ALTERITY:

Boz MOUNTAIN RANGE AND ITS PIEDMONT
0MUR HARMAN~AH, PERI JOHNSON, SHANNON MARTINO,

MDGE DURUSU-TANRIOVER, AND BEN MARSH
Ku§ ur;maz kervan ger;mez, karanhk tuttu yollan,

goziin gonliin kararml§, sen nasil gecesin hey gidi,
Bugdaysiz, r;avdarsiz kara ekmege benzersin.
Yildizlarm, hani yildizlarm, r;ir;eklerin nerdeler?
Ka/bin neden durmu§ riizgan kesilmi§ degirmen gibi?
Suya indi r;akallar, suya indi sogiit dallan,
Barakmuslu Mezarligi kimildamr ir;in ir;in.
Ben ne inim, ne cinim, hen bir garip ademim.

Barakmuslu Koyii'nden Seliimsiz oglu Bekir,
Benim babam, benim dedem yiidellerde oldiiler,
iki elim klZll kanda, sekiz bogaz avcuma bakar...
bir giden bir daha donmez, gitti gider ...
yddelde olmek istemem, dedem gibi, babam gibi ...
Barakmuslu koyiindenim, Seliimsiz oglu Bekir'im
hen bu koyde dogmu§um, bu koyde olmek isterim...
Atilla ilhan, "Barakmuslu Mezar1Jg1" (1954)
Mountains are often groundlessly thought of as romantic backwaters lacking in development and civility, and
portrayed as unruly places to pass through by academics working under the influence of ideologies of the state. When
traveling through the plain of Ilgm, Konya, in the summer of 1895 the German orientalist Friedrich Sarre thought that
Ilgm's modem caravanserai was noteworthy to describe (1896: 23). In contrast, in his writing, the narrow mountain
gorge upstream of the village ofBalkt, in the mountains to the south ofllgm, was a sublime landscape with steep red
slopes forested in the dark greens of magnificent cedar trees, a place of contemplation. 1 The village of Balkt is
precisely the place where the valley becomes less steep and less sublime, and therefore, the romanticism of the
traveler now turns to the landscape's picturesque aspects. Sarre describes how the village's guest house was on the
bank of a clear stream and surrounded by willow and poplar trees. Built of rubble masomy, the guest house's roughly
hewn stairs on the outside balanced the delicate woodwork on the inside. Yet his travel companions could not sleep
indoors because of the threat of Circassian horse thieves in the village (Sarre 1896: 112- 114). Sarre's writings are
exemplary of the romantic separation of the sublime and the picturesque from urban development as well as the
application of other binaries such as mountains as resource versus urban space as sites of industry.
Such binaries of the urban and the rural, or the perception of civilized lowlands and crude shepherds and loggers,
do not adequately account for the linear ecologies that intimately connect the plains to the mountains. In this chapter
we advocate for the significance of these connecting ecologies that resist the colonial or statist marginalization of
mountain peoples and places. These connecting linear ecologies are substantive landscapes of everyday movement,
the flow of water, taskscapes, and interconnected land use, and are not limited to roads and routes. Although we do
discuss routes, they are not sufficient to explain these lines of connectivity, which have more to do with the flow of
1
Sarre emphasizes the visual aspects of the landscapes that he passes through such as the "effektvollen Hintergrund filr das Grun
der priichtigen Cedem" in this gorge above " ungemein rnalerisch" Balla (1896: 112- 113); Anamas (Dedegiil) Mountain is the best
example of the influence of the aesthetics of the sublime and picturesque on his writing and photography (1896: 135- 137; pis. 54--
56).
288 Chapter Twenty-Two: Harman~ah et al.
water, people, and animals in an ecology that is political. We argue below that particular mounds and settlements on
the foothills play a critical role in making such connectivity possible.
The perception that mountains are full of "Circassian horse thieves" and that they are difficult to pass through is a
bias from theories on state development and the control of marginal mountain areas. Likewise, mountains are often
written out of narratives of history which consistently prioritize archives produced by imperial elites who hold onto
administrative centers in the lowlands. Academic perspectives on ancient communities of the mountains tend to
associate them with "landscapes of terror" (e.g., Matthews 2004). In these scenarios, marginalized mountain peoples
are presented either as "tribal" threats to urbanized elites of the prosperous plains and lowland and river valleys, or
impediments to regional circulation (Borden and Purcell 2000: 80). Such perspectives are produced under the
influence of urban archives; they are typical of uncritical characterizations of mountains from an elitist bias and have
to be taken with a grain of salt. Archaeological survey evidence, strengthened by ethnohistorical research, presents a
far more even-handed perspective on life in the mountains. In this chapter we point to the intimately entangled nature
of lowlands and mountains in the local context of west central Anatolia. This chapter is a modest attempt to bring
back mountains as complex and connected landscapes of alterity and to invite mountains back to their place within
settlement history.
The lines of the poem "Barakmuslu Mezarhg1" which we have quoted as an epigraph to this chapter is by the
prominent social realist poet Atilla ilhan (b. 1925-d. 2005), who spent part of his childhood in Ilgm during the late
1930s when his father was appointed as the town's governor. The poem evokes the mountainous landscape where
Barakmuslu is located and the experience of living in such a landscape from an emic, non-statist perspective. The
alterity of the place of burial and of the mountainous landscape are powerfully evoked in ilhan's poem both as sites of
deep history and of abandonment. The cemetery of Barakmuslu is a memorial site that anchors people to the place and
to the land with deep rootedness, whereas its eerie state of abandonment poetically evokes the disquiet of this
landscape.
KEY
• SURVEYED SITES
-;:;Rivers & streams
ca. 1960 /
- -Relict river ,,.,,.
YALBURT
0
YAYLASI •
Figure 22-1. Map ofYalburt Yaylas1 Archaeological Landscape Research Project

area with perennial rivers and streams ca. 1960 (Peri Johnson).
Mountains as Connected Landscapes of Alterity 289
In eight seasons of fieldwork since 2010, the Yalburt Yaylas1 Archaeological Landscape Research Project (Yalburt
Survey Project) has investigated the Ilgm Plain and the neighboring upland and lowland landscapes, with a particular
focus on the political ecology of water during the Hittite Empire period, when two major inscribed water monuments
were built in this imperial borderland landscape (Johnson and Harman~ah 2015, 2019; Harman~ah et al. 2017). These
two hieroglyphic monuments, which were both built at the time ofTudlJaliya IV (1237-1209 BCE), are the Yalburt
Yaylas1 Sacred Pool, located in pastoral uplands north of the Ilgm Plain, and the Koyliitolu Yayla Earthen Dam,
located in the agricultural lowlands to the east of the plain (Fig. 22-1). This chapter focuses on a previously unstudied
and archaeologically relatively unknown upland zone that runs through the southern part of the Yalburt Survey
Project area, directly south of the Ilgm Plain.
This zone is formed by the northwest-southeast stretching Boz Mountain Range that separates the Ilgm Plain from
the valleys that connect the Konya Plain to Bey~ehir Lake. The Boz Mountain Range rises 900 m from the Ilgm Plain
through deeply incised valleys before reaching verdant, spring filled karst uplands around the mountain peaks. This
wet and lush landscape supplies the well-watered Ilgm Plain and stands in stark contrast to the dry, eroded uplands
around Yalburt Yaylas1 that run through the northern part of the Yalburt Survey Project area. Fieldwork in the uplands
and the lower slopes of the southern zone took place during the 2011 and 2018 seasons (Harman~ah and Johnson
2013, 2020). Survey evidence from these field seasons shows that settlement has concentrated in the piedmont below
the incised valleys since the Chalcolithic period (ca. 6000-3000 BCE). The deeply incised valleys of the Boz
Mountain Range carried considerable amounts of sediment into the Ilgm Plain during the Chalcolithic period and
again after the Boz Mountain Range became densely settled in the Roman period. Fieldwork in 2018 documented two
significant 2nd millennium BCE sites in these valleys, Lebbor c;::aym and Akkaya Mevkii, which have substantially
reconfigured our understanding of the dynamics of this landscape, and its relationship to the Ilgm Plain. This work
complemented the 2011 fieldwork and subsequent study seasons in 2017 and 2019 that documented the settlements at
the mouths of the incised valleys, such as the high Mahmuthisar mound. This overview offers a preliminary
discussion of the diachronic settlement history of this landscape, both as an ecology of fragmented and diverse valley
settlements, and as a water-rich and resilient landscape of forests, orchards, vineyards, cereal fields, and pastoralism.
COMPLEMENTARY LANDSCAPES: LINEAR ECOLOGIES AND INFRASTRUCTURES OF

WATER, MOVEMENT (ANIMAL/HUMAN), LAND USE, AND SETTLEMENT
In a forthcoming article, several authors of this chapter have advocated for a survey research focus on medium-
scale landscapes as the scale of human experience (Harman~ah et al. forthcoming). In that article we argue that
archaeological survey fieldwork and landscape archaeology's tools of engagement are best suited for working at the
medium scale, while we presented sample narratives of landscape description and biography for three distinct
ecologies: the Ilgm Plain, the Bulasan River Valley and Koylutolu Yayla lowlands, as well as the eroded pastoral
uplands ofYalburt Yaylas1.
I .)
~\-a Sebiller ,,,. •SARAYCIK HCYOK
~\e •GAVURUN OE~IRMENi
~\6 . suNGURBEY
lhsaniye Meydanll
-<:\ T0RBESI U~HA HClYO~O
\'··.. Karasevin,; •
OREN
\,,. c;E~MESI
MAHMUTHiSAR
TepekOy
KEY
SURVEYED SITES
, Mode rn villages
Ri~'{:.SiiJhreams
L Routes
f_ /
Figure 22-2. Map of the western Boz Mountain Range with routes discussed
in the text and 10 m contour intervals (Peri Johnson).
In the present chapter, we link our study of the medium scale to complementary landscapes, and investigate the
connections between the Ilgm Plain and the green upland landscapes of the southern piedmont and mountains (Fig.
22-2). Using the Yalburt Survey Project evidence, we argue that this connectivity builds a powerful infrastructure in
the form oflinear ecologies of movement, flow of water, land use, and settlement that move up and down between the
plain and the uplands of the Boz Mountain Range along distinct incised valleys. These connections are not limited to
roads, routes, and pathways, but involve precisely located key settlements such as Nodalar Hoyuk and Mahmuthisar
(see discussion below) which provide the necessary cultural and material links, flows of water as a shared vital matter,
long-term practices oftranshumance and seasonal movement, and distinct uses of the agricultural landscape balancing
cereal cultivation and fruit orchards. 2
The main argument of this study is, therefore, the interconnectedness and codependency of mountains and
lowlands, which is the distinct landscape pattern that emerges from our survey results. We build our argument based
on the proximity of lowlands and mountains in this instance, rather than accepting the conventional narrative of their
separation. We resist the temptation to assume that mountain settlements are remote, unsophisticated, and backward
places in contrast to the literate, politically active, and civilized lowland sites, with a critical eye on the urban archives
that actively propagate such biased vision.
SURVEYING THE MOUNTAINS: NOTES ON FIELD METHODS
The Ilgm Plain was known as one of the most well-watered landscapes of the Konya region. Our previous
publications about the lowlands of Ilgm, however, have not dealt with the landscape that watered the plain itself
(Harman~ah et al. 2017, 2018; Johnson and Harman~ah 2019; Harman~ah and Johnson 2020). Rather we have
engaged with the plain itself and its crucial connecting role across regions, thanks to its Salt Road and its historical
significance as a site of military transit and conflict. For our work on upland landscapes in the survey zone, we
concentrated on the northern karst uplands around Yalburt Yaylas1, the namesake of the project, a landscape of
depressions with swallow holes where water flows into the limestone and travels away subterraneously. This chapter,
however, turns to another upland landscape that is very different from Yalburt Yaylas1's uplands; this landscape is
about depressions where water flows out, the melting snowpack of mountain peaks, and the sources of Ilgm' s
formerly copious waters, that is, the southern Boz Mountain Range and its piedmont.
This landscape begins with the parallel mountain ridges with the exposed peaks ofMorbel at 1942 m, and Dikmen
at 1938 m, both of which are about 900 m above the Ilgm Plain (Fig. 22-3). Interspersed between these northwest-
,,l
•SARAYCIK HdYOK
•GAVURUN OECIRMENI
. suNGURBEY ihsaniye Meydanll

T0RBESI U~HA HOYOCO
I
I
BUL<;UK Karasevinc; • C:ubuk I
Orman6zU
~dYUev~:~rtif •
• KUYUSU
Eide!; OREN
l;E~MESI
I
I
I
Bukuk ~~.&'"
\ S6~Ut6zU
Osmanc1k/
Yazllca Yagllca /
Pirali
LEBBOR c;AYIRI
Ba~k6y KEMPOS • G6kt;:eyurt
Jj
/
KEY
SURVEYED SITE
Modern villages
'!;Rivers & streams
ca.1960
~ , \ 4KMA
/
Figure 22-3. Map of the geomorphological zones in the western Boz Mountain Range that
correspond to chapter sections, with 10 m contour intervals (Peri Johnson).
2
Whereas Xavier de Planhol articulated a connection fostered by transhumance between the mountains bordering the Anatolian
plateau and the Antalya Plain, transhumance forms only part of the entangled relationship between the mountains and the plains of
Ilgm. Hermann Wenzel, the geographer who did fieldwork on transhumance in the region around Ilgm a full decade before De
Planhol's fieldwork to the southwest, however, emphasized habitat (Lebensraum) to such an extent that he excluded any discussion
of connectivity, and his writing is not very helpful to this study (De Planhol 1958; Wenzel 1937). Conflicts such as between
Balki's woodworkers and pastoralists who would want to graze in Balk1's commons, and landscape typologies such as descriptions
of forests degraded by pastoralism, take center stage in Wenzel's writing about the Boz Mountain Range (1937: 20, 25-27).
southeast running ridges are karst uplands fed by the streams and springs of the Boz Mountain Range. Although the
geology of the southern uplands is more variable than the northern karst uplands, the landscape is mostly composed of
forests, cereal fields, and undulating meadows with depressions where springs emerge. 3 Many of the upland streams
such as those upstream of Belekler drain to the west before flowing into the Balki-Battal River and eventually the
Ilgm Plain. Regrettably, the Yalburt Survey Project's attention to the archaeology of the sediment, which buried the
Ilgm Plain, has led to an unintentional neglect of these uplands, which the project has not fully surveyed, although
interviews and remote sensing indicate that these uplands are full of archaeological settlements.4
We often found that we were methodologically unprepared and short of the necessary time to survey these upland
landscapes. It is important to note that they require both significant dedicated fieldwork under difficult conditions
with low visibility, and alternative techniques to the locus system of ceramics collection and intensive survey that the
project applied in the lowlands. The very dispersed and flat settlements of these areas, their untilled land surfaces,
sparse and heavily eroded ceramics, and rough terrain often simply did not align with the collection method we
designed for elsewhere in the survey area. We ended up deciding not to devote the necessary time because many of
these sites are Roman and/or medieval and consequently are beyond the central research objectives of the survey
project. A survey that focuses on the Roman period or later would definitely have to develop alternative survey
methodologies for such forest, field, and meadow upland landscapes.
THE INCISED VALLEYS: BULCUK AND KEMPOS VALLEYS
Bulcuk and Elde~ are two prominent and prosperous villages in the piedmont that are supplied with abundant
springs today; both are locations where the deeply incised valleys come down and meet the Ilgm Plain (Fig. 22-4).
KEY
.,,. NQDf.\LAR
• DSURVEYED SITES HOYUK
Modern villages
ILGJN PLAIN
;;; Rivers & streams
ca. 1960
~ Springs
1-Routes
FAZ DAGLI MEVKii •
o
.....~ = 1--~ =52KM &
Figure 22-4. Map of the incised Bulcuk and Kempos Valleys (Peri Johnson).
3
The western Boz Mountain Range does not have copper and mercruy ore deposits similar to S1zma and Ladik (Y1ld1z and Bailey
1978). (Copper was mined first in the EBA and mercruy in the Roman period, although the mercruy ore, cinnabar, may have been
mined in the Neolithic period as a pigment.)
4
Paved asphalt roads to these uplands pass through the area between Balk:1 and Yukar1yigil that is not within the survey boundary,
and the 50 km distance from where we have been hosted during all fieldwork seasons, the Ilgm Sugar Factory, meant that the
southern uplands were logistically difficult.
Kempos (G6k9eyurt), however, is a relatively large historic settlement in a densely forested mountainous area, located
at a copious spring that serves as the headwaters of the Kempos River (see discussion below). At Bulcuk, the Yalburt
Survey Project team surveyed Bulcuk Hoyugu, a fortified Hellenistic and Roman settlement on a natural promontory
immediately north of the village. Bulcuk River flows just to the west of the village, where abundant springs are also
located. The reused Byzantine tombstone of a family's two daughters was documented adjacent to the water gushing
from three spouts of one of the springs (Fig. 22-5). In such a favorable place, this settlement connects both to the
agricultural abundance of the surrounding piedmont and nearby fields in the Ilgm Plain. Upstream of Bulcuk is the
deeply incised Kozlu Valley, where movement is only to local pastures and forest, and topography hinders regional
movement. In the fan of the Kempos River near Elde~, the prominent lowland mound of Nodalar Hoyuk was
surveyed. This small but tall mound has a long and fairly uninterrupted sequence of settlement from the Late
Neolithic-Early Chalcolithic through the Hellenistic period. 5 The settlement's longevity is often associated with its
strategic location in the southern Ilgm Plain as a "lowland administrative center" with surrounding well-watered
agricultural lands; however, these lands are watered by the Kempos River that connects the settlement to diverse
ecologies upstream and underlies Nodalar Hoyuk's dependence on the piedmont and mountains.
Figure 22-5. Photograph of the documentation of the Byzantine tombstone at a Bulcuk spring, 30 July, 2011
(Yalburt Yaylas1 Archaeological Landscape Research Project).
The project team surveyed several of the incised valleys between the piedmont and the mountains. Two
settlements that were extensively surveyed indicate both the potentials and drawbacks of surveying the valley and
mountain landscapes: Faz Dagh Mevkii and Gokseki Mevkii (for the latter see below). The survey team visited the
first of these two settlements, Faz Dagh Mevkii, partly to investigate the relationship of the plains to the mountains
and the connectivity to the south, and partly because of an enigmatic structure on 1950s series NATO standard
topographic maps.
Faz Dagh Mevkii may have been settled as early as the Chalcolithic period and again beginning in the Roman
period, extending into the medieval period. 6 Straddling a spur on the western slope of the deeply incised Kozlu Valley
upstream of Bulcuk, Faz Dagh Mevkii has a view of the Ilgm Plain in the distance to the north. Adjacent to the spur
are several springs and depressions that hold water. The visible bedrock of the spur is limestone with recrystallized
limestone above and quartzite below; the springs emerge where the limestone meets the quartzite. Despite
reforestation trenches not more than 10 years old when we visited in 2016, thick grasses hindered visibility except on
the spur itself, where most of the ceramics were noted. It is the probable Chalcolithic settlement at this site that turned
out to be the most significant for our research.
5
Through ceramics Nodalar Hoyuk is known to be occupied in the Late Neolithic-Early Chalcolithic, Middle Chalcolithic, Late
Chalcolithic, EBA, MBA & LBA, MIA-LIA, and Hellenistic period, with Roman ceramics at a spring nearby.
6
Ceramics were not collected and only preliminarily studied at the settlement by Peri Johnson with section drawings and
photographs. Without Shannon Martino's more thorough analysis, the question of an EBA settlement unfortunately must remain
open.
One of the potentials of studying the upland settlements is the fact that the geomorphological and archaeological
evidence these settlements supply is often relevant to the Holocene anthropogenic landscape changes in the plains.
The survey project has documented two major phases of alluvial sedimentation in the plains: the first beginning in the
Chalcolithic period, and the second taking place after the Roman period (Harman~ah et al. forthcoming). The density
of the Chalcolithic ceramics at Faz Dagh Mevkii is indicative of settlement on the spur-that is, it was not a place of
seasonal activity-and of participation in the settlement expansion to the uplands. This settlement expansion in the
uplands resulted in deforestation, which then was the primary cause for the first phase of alluvial sedimentation. Such
interdependent histories and ecologies of settlement between lowlands and uplands is significant to pursue, as
illustrated by the case of Faz Dagh Mevkii.
The enigmatic structure that originally drew our team to Faz Dagh Mevkii, and which is visible in satellite
imagery, measures 86 m by 50 m. The structure turned out to be a vast monument, arranged in six massive courtyards
in two rows of three, that rests on the most extensive remains at the settlement. The absence of fine ware sherds of
Roman date makes our dating of this most extensive period of settlement at the site with numerous building
foundations very difficult. These foundations are associated with Roman roof tiles and are visible both on the ground
surface and in satellite imagery. This settlement was probably pastoral in nature, with gardens around the depressions
similar to a contemporary pastoral complex with gushing springs located upslope 900 m to the south. The ceramic
assemblage had only one fine ware example: a small light green glazed sherd with white engobe that possibly dates
the enigmatic structure. We tentatively interpret the structure as a Byzantine refuge built during the Abbasid invasions
of the 9th century CE, and date the glazed sherd to the Middle Byzantine period (10th-11 th centuries CE). The structure
would be contemporaneous with the hidden Byzantine dwellings carved within the cliff at Gavurun Degirmeni where
the Balk1-Battal River debouches into the Ilgm Plain (see Fig. 22-6). Faz Dagh Mevkii's view of the plain and
difficulty of access support this interpretation. During the Byzantine period the plain was unruly and the piedmont
was civilized. In contrast to regional connectivity, the enigmatic structure then exemplifies the interdependence of the
pastoral mountains with the settlements in the piedmont and plain below.
KEY ILGIN PLAIN
• tsURVEYED SITES SARAYCIK HOYUK
Modern villages
.,. Rivers & streams
ic ca. 1960
.., Springs
:-Routes
Harmanyaz1
Ge~it
LEBBOR <;:AYIRI :,r:
'v
Figure 22-6. Map of the eroded piedmont (Peri Johnson).
294 Chapter Twenty-Two: Hatm~ah et al.
The routes within the Boz Mountain Range are, unsurprisingly, not associated with the incised valleys such as the
Kozlu Valley, but rather with the northwest- southeast running ridges between them. The northwest- southeast route
over the Boz Mountain Range from Bulcuk starts in the piedmont at the fortified Hellenistic and Roman promontory
settlement of Bulcuk Hoyilgii. The route ascends the ridge running on the east slope of Bulcuk before crossing to
Kempos where it merges with the route from Elde~. It is not very clear if this route from Bulcuk is more than a local
route at any time. The route after Kempos ascends between the two parallel mountain ridges to a pass at an elevation
of 1686 m before descending to the Derbent area (Saracik-Beylik Valleys) near Eli Kesik Han (see Fig. 22-2).
The only evidence for the existence of Roman settlements along the Kempos River to the east of Bulcuk is the
spolia that were spotted in the buildings of Elde~ and Kempos, the two villages that bracket the incised Kempos
Valley. This is an important valley to discuss with respect to vertical infrastructures of settlement, agriculture, and
movement between mountain and plain landscapes with very narrow intervening piedmont. Kempos stands as a
crossroads at a copious spring emerging from the base of a prominent dolomitic limestone outcrop. 7 Surrounded by
terraced orchards, gardens, fields, and a cave for aging cheese in goatskins, Kempos has an abundance of the fruits
and animal husbandry of a Boz Mountain Range mountain village. However the only archaeological evidence we
could document around the village was the Roman spolia. In contrast, Elde~ rests on bedrock at the edge of the Ilgm
Plain with a medieval site (Kurban Tepesi) on a rise above the village. Downstream, the Kempos River meanders
through its fan to Nodalar Hoyilk, known to be settled through the Hellenistic period. Nodalar Hoyilk, with its long
history of settlement, arises not only due to its strategic location in the plain along the Kempos River, but also tharlks
to the abundance of fruit coming down from the mountains.
THE ERODED PIEDMONT: THE BALKI-BATTAL RIVER VALLEY
Moving west ofBulcuk, the next linear landscape ecology that we explore is the Balk1-Battal River Valley and the
geologically distinct erodible landscape of the piedmont that extends northwest from Bulcuk to where the Balk1-Battal
River debouches into the Ilgm Plain and all the way to Balk1 in the southwest (see Fig. 22-3). This is a zone occupied
today by the territories of the villages of Ormanozu, <;atak, Gostere, Harmanyaz1, Ger;:it, and Balk1. Here the piedmont
is constituted of lacustrine bedrock vulnerable to erosion and forms a friable landscape that contrasts with the much
earlier carbonate of the peaks and incised valleys of the Kozlu and Kempos Valleys. To slow down the erosion, this
landscape has been gradually reforested in recent years and therefore has been radically transformed by the state. For
this reason, this is a difficult archaeological landscape to survey; even the Ottoman settlements and features are
difficult to document, let alone the description of much earlier landscape features. Our survey, however, did document
that these now reforested slopes of the piedmont once constituted a pastoral landscape with extensive orchards and
vineyards. Interspersed between the forests today are cereal fields. The landscape differs from elsewhere in the range
also in the infrequency of copious springs and the presence of alternate routes to the paved road that ascends through
the incised valley with fairy chimneys, south of Balk1, described as magnificent by Sarre. The project surveyed two
places settled at a distance from the floodplain of the river: Antike Mevkii and Lebbor <;aym, both with nearby recent
reforestation. We also documented several sites adjacent to the verdant floodplain (Fig. 22-6).
Antike Mevkii is a small ridgetop field 500 m south of Ormanozii on the summit of Koru Tepesi. The site is
composed of reforested heavily eroded slopes to the west, and interspersed cereal fields and forest to the east. The
ceramics are similar to the slopes on which they are found, heavily eroded, but a few diagnostic Hellenistic grey ware
and Roman common ware sherds were collected. Although a pastoral area in the Ottoman period, settled with
Circassian refugees in the nineteenth century, it seems probable that Antike Mevkii participated in a Hellenistic
expansion into the piedmont based on both pastoralism and viticulture, although without evidence of the winepresses
frequently found in areas with less friable bedrock, the viticulture is highly conjectural. By the Roman imperial
period, residents of more prosperous settlements than Antike Mevkii were drinking plenty of wine in thin-walled ware
and local terra sigillata cups.8 A comparable settlement was surveyed on the ridgetop between Bulcuk and Elde~ at the
Uyuz Kuyusu spring (see Fig. 22-4); however, erosion subsequent to Antike Mevkii's settlement has altered the
groundwater, and the springs near Antike Mevkii are in the valley floor to the east.
Unlike the relatively marginal and archaeologically dime-a-dozen ridgetop settlements of Antike Mevkii and Uyuz
Kuyusu, the settlement of Lebbor <;aym is centered around springs in the valley floor of a tributary of the Balk1-
Battal River, less than 3 km from the village ofGer;:it, through a gorge. Settlement spreading widely across a series of
gentle slopes on the eastern and western sides of the valley dates to the Hellenistic and Roman periods. However,
earlier phases dating to the Middle Chalcolithic, Early Bronze Age (EBA ca. 3000- 2000 BCE), and the 2nd
millennium BCE are restricted to the western side, which rises to a low ridgetop with a view over rolling cereal fields
extending to the river at Balk!. The springs are at a boundary between the eroded hills, which we imagine as vineyards
in the 2nd millennium,9 and the rolling fields around the broadening of the river valley at Balk!. Although Lebbor
<;aym's location relates to the character of its place, Ottoman local routes crossing the piedmont to <;atak and
southwards, and climbing through the Beloren Valley to Yukar1r;:igil, meet at Lebbor <;aym (see Figs. 22-2 and 22-6).
7
Although the Elde~-Kempos- Tepekoy road is paved today, for much of the twentieth century and earlier, Kempos was a
crossroads of pack animal trails.
8
The Roman period settlements have an abundance of these cups, but wine amphorae are rare.
9
Grape pollen is a significant part of the Bey~ehir Occupation Phase (Woodbridge et al. 2019).
The settlement reorients us away from the river towards the friable piedmont with its more unstable yet gradual
slopes.
The tentacles of settlement reaching into the piedmont around the Balk1-Battal River Valley in the Chalcolithic
period, such as at Lebbor <;aym, signal the start of the soil erosion of the very vulnerable western piedmont that has
carried massive quantities ofred sediment from the lacustrine bedrock into the fan of the Balki-Battal River. This fan
buries Sarayc1k Hoyuk, which is a Chalcolithic-EBA (and later) mound located at the mouth of the Balk1-Battal River
where the piedmont meets the Ilgm Plain. This location also includes the hidden rock-cut Byzantine dwellings at
Gavurun Degirmeni. Well into the Ilgm Plain, the fan then spreads as far as the Hittite new foundation at Boz Hoyuk
and the nearby Hellenistic urban center of Tyriaion, two of the Yalburt Survey Project's "lowland administrative
centers."
It is not only water but also the sediment that it carries that links the Ilgm Plain and the Boz Mountain Range.
Along the narrow river corridor that this sediment has traveled, few settlements have been surveyed. The first two are
at nodes where the valley opens up. The first, a chipped stone quarry called P~akoy Degirmeni, with Roman and later
settlement, all heavily damaged by reforestation, lies right at the confluence of the Battal and Delikoy Bogazi. The
second is a fortified promontory settlement of the Late Iron Age (LIA ca. 550-330 BCE) and Hellenistic period
known as Asarh Burnu, which defends a crucial turning point along the Delikoy Bogaz1 upstream of the confluence,
at a setting very similar to Bulcuk Hoytigii. From the northern cluster of settlements around Saraycik Hoyuk and
Gavurun Degirmeni, moving southwest towards Gostere, Gec;:it, and Balk1, no other settlements were surveyed along
the river up to Balki. A small historic town with extensive Ottoman gardens watered by canals, Balk1 encircles an
undateable mound encased by the cobblestone paved streets of the contemporary town and crowned by a spectacular
Ottoman mosque displaying a wealth of spolia in its masonry walls. A tumulus and cist grave cemetery of Hellenistic
and Roman date lies above the floodplain on the western side of the river, but the settlement belonging to this
cemetery was not located. The ruins at Adaras, 1.5 km west of the cemetery and beyond the survey boundary, are
thought to be a source of the spolia in Balki, but Adaras has only been surveyed epigraphically, and Balkl was
probably also settled around the same time. The transition of the piedmont to the valleys to the west of the river
further dispersed settlements and movement away from the floodplain.
THE TEKKE VALLEY
In this final section of the chapter, we move to the southeastern corner of the Ilgm Plain, in order to discuss the
Tekke Valley and conclude our presentation of the linear connecting ecologies between the lowlands, piedmont, and
mountains in the southern zone of the project's survey area. This valley is home to some of the most significant
settlements and lines of movement for the landscape history of the region before the Hellenistic period that involve
places with deep settlement histories both close to the lowlands like Mahmuthisar and touching the mountains like
Akkaya Mevkii and Barakmuslu (Fig. 22-7). The multiperiod and impressive mound at Mahmuthisar defines the
archaeological landscape of the twin towns of Mahmuthisar and Beykonak. The latter town is also known for the
sacred shrine ofDedigi Dede, which has deep roots in the Roman and medieval periods (Bakirer and Faroqhi 1975).
We also return to Barakmuslu and the poem by Atilla Ilhan with which this chapter started, "Barakmuslu Mezarhg1,"
which voices our approach to the uplands so well.
The southern edge of the Ilgm Plain arcs along a fault line from near Boz Hoyuk in the northwest to Mahmuthisar
in its southeastern corner, where the Tekke River debouches into the plain. Although cut by the Bulcuk and Kempos
Rivers today, this arcing edge of the piedmont allows movement around the wet Ilgm Plain and channels movement
into the incised Tekke Valley. Unlike the adjacent Kempos Valley, the Tekke Valley cuts through a lacustrine
piedmont to the east of the carbonate bedrock of the central incised valleys. This carbonate bedrock, however, is an
aquifer that supplies the numerous springs around Mahmuthisar, particularly at Ye~ilgol and the Hat1lh Spring within
Beykonak. At the Ye~ilgol spring, we surveyed a Late Iron Age or Hellenistic gabled rock-cut tomb carved into the
cliff above the source.
The archaeological site of Mahmuthisar is a high, lozenge-shaped mound, which we documented in 2011 as
continuously settled from the Middle Bronze Age (MBA 20th- 17th centuries BCE) onwards. 10 Similar to the other
sites near where rivers debouche into the Ilgm Plain such as Nodalar Hoyuk, in 2011, Mahmuthisar appeared to
connect the plain to the piedmont and mountains locally, for example, the inhabitants of Barakmuslu to their wheat
fields and pastures in the mountains. Fieldwork in 2018 indicated that this route also functioned as an imperially
defended route during the Hittite Empire. The evidence for the existence of this route emerges from fieldwork at two
settlements near Barakmuslu: Gokseki Mevkii and Akkaya Mevkii.
Similar to Faz Dagh Mevkii discussed above, Gokseki Mevkii is one of those elusive upland sites where the
methodologies of the Yalburt Survey Project were not very helpful. Our limited survey at the site revealed that this is
a small Byzantine agricultural settlement spread across terraced wheat fields near springs at the uppermost boundary
10
A Late Chalcolithic- EBA sherd has been published from the 1994 survey directed by Hasan Bahar (Bahar et al. 1996: 26, pl.
96.1); the Yalburt Survey Project identified diagnostic MBA, LBA, EIA, MIA, LIA, Hellenistic, Roman imperial, and late Roman
ceramics; Byzantine spolia; and early Ottoman ceramics. In 2011 the effluent of milk barns covered profiles visible in 1994.
Mahmuthisar
~MAHMUTHiSAR
Beykonak
YE~iLGOL
• SAMUDUN
HbYOGO
c;:e~mecik
AKKAY.A
MEVKII
Bara kmuslu • •
KEY obsidian
Guneypmar
• nsuRVEYED SITES
Modern villages
~ Rivers & streams
ca. 1960
.., Springs
:-Routes GOKSEKi MEVKii
o--===-1--===2KM A
Figure 22-7. Map of the Tekke Valley (Peri Johnson).
Figure 22-8. Photograph of the middle promontory of Akkaya Mevkii

from the west, 20 July, 2018 (OmUr Harman~ah).
of the piedmont, where the incised Atuc;uran Valley begins. Among the very limited assemblage of the surface finds
at Gokseki Mevkii were pithos and oxidized common ware sherds and a glass bead. Heading upstream of Barakmuslu
to Gokseki Mevkii and beyond revealed Barakmuslu's alternative local route over the Boz Mountain Range between
the Dikmen and Atuc;uran Peaks with a pass at an elevation of 1733 m, at no higher an elevation than the paved road
today that runs to the east of peaks clustered around Atuc;uran (see Fig. 22-2). 11
It is at Barakmuslu itself, however, where the surprises lay for us in thinking about the significance of the Tekke
Valley. Barakmuslu is a deeply historical mountain village that rests on the slope of an andesite dome with obsidian
and surrounding tuff. Conversations with the village elders held memories of a relationship with the Dibek Degirmeni
Mevkii down in the Ilgm Plain, at an earlier phase of settlement which reemphasized for us the continued movement
between the lowlands and the uplands. The ancient settlement complex of Akkaya Mevkii is an archaeological
settlement site spread across three sedimentary promontories that are located approximately 800 m east and
downslope of Barakmuslu, and closely overlooking the start of the Tekke Valley (Fig. 22-8). These promontories are
separated from each other by deep and lush stream beds where water flows from springs. The site has a long
settlement history from the Late Neolithic-Early Chalcolithic period to the Roman period.
The earliest settlement at Akkaya Mevkii is located on the middle promontory and runs across two surveyed loci.
Immediately above this middle promontory is a barren tuff slope that has been quarried for stone. The analysis of the
ceramic wares from loci 2 and 3 on the middle promontory is still preliminary; however, the earliest sherds have an
abundance of volcanic glass from local production. 12 The Akkaya Mevkii EBA ceramic wares do not share in this
tempering technique which probably relates to the EBA preference for less gritty wares. Although local in production,
these preferences indicate connectivity from an early date. The early wares from Akkaya Mevkii suggest a Late
Neolithic or Early Chalcolithic date, making this site one of the earliest settlements in the survey region.
The Late Neolithic date for this portion of the site is further bolstered by a ceramic stamp seal that was found in
locus 3. In the ploughed agricultural field labeled as locus 3 by the survey team, a well-baked ceramic stamp seal was
found (Fig. 22-9). The seal has a pyramidal shape with a pinched handle (1.70 cm high) and a perforated hole for a
string to be hung on the neck of the handle. The sealing face is roughly square (1.45 x 1.45 cm) with slightly concave
edges. The seal design is geometric with intersecting horizontal and vertical lines and a small pierced hole in the
middle. The seal weighs 2.5 g and has a light brown color. Comparable ceramic stamp seals with geometric designs
from the Neolithic mound of (:atalhoyuk in the <;::umra Plain on the (:ar~amba River fan suggest a Late Neolithic date
for the seal (Turkcan 2005; Atakuman 2013). The middle promontory continued to be settled during the EBA.
Figure 22-9. Ceramic stamp seal from the middle promontory (locus 3) of
Akkaya Mevkii (Yalburt Yaylas1 Archaeological Landscape Research Project).
11
A regional route from Pirali to Tepekoy in the south runs to the east of the survey area (see Figure 22-2). The periods when this
eastern route was operational are, however, unknown. This route has an unsurveyed fort near its pass at an elevation of 1739 m.
Another more well-known Byzantine and earlier northwest-southeast route from south of Kadmham to north of Konya begins 6
km east of this Tepekoy road. Kestel is the fortress that defends the northwestern ascent of this route, which follows the valley of
Y-iikselen after the pass (1746 m) above Kestel.
12
Although pumice layers from the volcanic eruptions associated with the Barakmuslu dome appear in mudstone in the east and
west of the survey area (Ko9 et al. 2017), the high percentage of volcanic glass in the Barakmuslu Chalcolithic wares is
noteworthy.
298 Chapter Twenty-Two: Hatm~ah et al.
After an apparent hiatus in the MBA, the settlement at Akkaya Mevkii returns again during the Late Bronze Age
(16th- 13th centuries BCE) on top of the middle promontory but then also expands to terraces on its slopes and to the
southern promontory (locus 1). A mound at the highest place in the middle of the southern promontory is composed
of the remnants of a burnt building with vitrified mudbrick. Although with less precipitous cliffs than the middle
promontory, the southern promontory overlooks the routes branching from the start of the Tekke Valley, particularly
the ridge route parallel to the incised Atm;:uran Valley. After a lapse in settlement during the Early and Middle Iron
Ages, the northern promontory (loci 4 and 5) was settled during the Late Iron Age or the early Hellenistic period.
Although no defensive walls were documented, the eastern cliffs of the middle and northern promontories are steep, a
dramatic topography coupled with springs.
In the incised valleys of Bulcuk and Kempos and the eroded piedmont of the Balk1-Battal River Valley, we have
not seen Chalcolithic or earlier sites as extensive as Akkaya Mevkii being settled at higher elevations. We would like
to argue that the site also plays a significant role in maintaining the connectivity between the Ilgm area and Konya
Plain to the southeast as well as to Derbent to the south and Bey~ehir Lake to the southwest. Considering that the
settlement ofMahmuthisar was one of the most important sites in the Ilgm Plain during the Late Bronze Age, and was
located at the strategic point at the tail end of the Tekke Valley, Yalburt Survey Project's dating of the newly
documented Akkaya settlement now provides the missing link in making these connections and pulling routes away
from the lowlands and pointing towards passes over the mountains. When we consider the more recent history of this
linear connecting ecology of settlements, both the twin towns of Mahmuthisar and Beykonak, and the village of
Barakmuslu present us with a deeply rooted historical landscape of connectivity between the uplands and the
lowlands as well as regional routes.
CONCLUSION
In this chapter, we offered a narrative about the connected ecologies between the Ilgm Plain and the piedmont and
mountain zones of the Boz Mountain Range. This is a complex archaeological landscape that has received but slight
interest from archaeologists in the past before the work of the Yalburt Survey Project. In our own fieldwork
experience, we frequently realized and acknowledged the limitations of our project's methodological strategies of
survey as we gradually moved from the Ilgm Plain to the Boz Mountain Range. While our field methods were
effective in the lowlands, they were less so along the valleys and ridges of the piedmont landscapes. We addressed the
challenges in working in the forested mountainous zones at sites such as Faz Dagh Mevkii or Gokseki Mevkii.
We presented our narrative of these landscapes by analyzing three distinct zones, the incised valleys ofBulcuk and
Kempos, the eroded piedmont region of the Balk1-Battal River Valley, and finally the Tekke Valley that links
Mahmuthisar to Akkaya Mevkii in a local landscape that finds itself part of regional connectivity. We argued that
these are complementary landscapes that efficiently bind the places and peoples of the lowlands and the uplands.
Discussing the incised valleys ofBulcuk and Kempos, we focused on the important distinctions of settlement histories
between the lowlands and uplands. While the forested upland zones in the Konya region are usually associated with a
density of settlements during the Hellenistic, Roman, and Byantine periods, the Chalcolithic settlement at Faz Dagh
Mevkii pointed to an expansion of settlement into the uplands in the Chalcolithic Period during the Holocene episode
of forest cutting and soil erosion that led to an early event of human-induced alluviation in the lowlands. The eroded
piedmont region of the Balk1-Battal River Valley presented us with an extraordinary past landscape of pastoralism as
well as extensive zones of former orchards and vineyards. The settlement history in this zone presents a complex
infrastructure of settlement from Balk1 all the way to Sarayc1k Hoyillc where the river enters the plain and on to Boz
Hoyillc in the plain. Finally, the Tekke Valley offered us a form of connectivity that went beyond the local lowland-
upland connections, especially during the time of the Hittite Empire, when the Konya region became a significant
borderland zone between the Hittite urban center of Ikkuwaniya, the Jj:ulaya River Land, and Pedassa (Harma~ah et
al. forthcoming). In all three of these linear medium-scale landscape biographies, we intended to illustrate the
codependency and interconnectivity of lowland, piedmont, and mountainous places and peoples, without downplaying
the complexity and particularity of each landscape.
Atilla 1lhan's evocative poem "Barakmuslu Mezarhg1" voices an honest and equitable sentiment of the mountains
from the educated perspective of an urban child in the 1930s. Here the mountain does not appear to be picturesque,
sublime, or backward in time as it would appear in the colonial, marginalizing narratives of academic writing, but the
mountain flourishes as a powerful place of alterity, and also a place of injustice and sorrow, which are narrated
through the perspective of the dead ancestors in the Barakmuslu cemetery. The poem addresses the historically well-
known political practice of marginalization and injustice towards the people and culture of the mountains. Our
landscape biography of the history of the Boz Mountain Range and its relationship to the Ilgm Plain also attempts to
take this politics seriously.
REFERENCES CITED
Atak:uman, <;igdem. 2013. Deciphering Later Neolithic Stamp Seal Imagery of Northern Mesopotamia. Documenta
Praehistorica 40: 247-264.
Bahar, Hasan, Gungor Karauguz, and Ozdemir Koi;:ak. 1996. Eskfr;ag Kanya Ar(1$flrmalarz I (Phrygia Paroreus
Bolgesi: Amtlar, Yerl~meler ve Kiiciik Buluntular). istanbul: H. Bahar.
Balmer, Omtir and Suraiya Faroqhi. 1975. Dedigi Dede ve Tekkeleri. Belleten 39: 447-471.
De Planhol, X. 1958. De la plaine pamphylienne aux lacs pisidiens; nomadisme et vie paysanne. Paris: Librairie
Adrien-Maisonneuve.
Harman~ah, Omfu and Peri Johnson. 2013. Pmarlar, Magaralar, ve Hitit Anadolu'sunda Krrsal Peyzaj: Yalburt
Yaylas1 Arkeolojik Yuzey Ar~trrma Projesi (Ilgm, Konya), 2011 Sezonu Sonui;:lan. Ara~tzrma Sonuc;larz
Toplantzsz 30(2): 73-84.
- . 2020. Yollar ve Guzergiihlarm Siyasi Ekolojisi: Yalburt Yaylas1 ve <;evresi Arkeolojik Yuzey Ar~trrma Projesi
2018 Sezonu. Ara~tzrma Sonuc;larz Toplantzsz 37(3): 189- 207.
Harman~ah, Omur, Peri Johnson, Ben Marsh, and Muge Durusu-Tannover. 2017. Lake-Places, Local Hydrology and
the Hittite Imperial Projects in the Ilgm Plain: Yalburt Yaylas1 Archaeological Landscape Research Project 2015-
2016 Seasons. In The Archaeology of Anatolia: Recent Discoveries (2015-2016) Volume II, S.R. Steadman and
G. McMahon, eds., 302-320. Newcastle upon Tyne: Cambridge Scholars Publishing.
Harman~ah, Omtir, Peri Johnson, Muge Durusu-Tarmover, and Ben Marsh. 2018. Tuz Yolu Boyunca Ilgm Ovas1:
Yalburt Yaylas1 ve <;evresi Yuzey Ar~hrma Projesi 2015 Sezonu Sonui;:lar1. Ara~tzrma Sonuc;larz Toplantm
35(2): 331-353.
Harman~ah, Omur, Peri Johnson, Muge Durusu-Tannover, and Ben Marsh. Forthcoming. The Archaeology of Hittite
Landscapes: A View from the Southwestern Borderlands. Journal of Eastern Mediterranean Archaeology and
Heritage Studies.
Horden, Peregrine and Nicholas Purcell. 2000. The Corrupting Sea: A Study of Mediterranean History. Oxford and
Malden, MA: Blackwell.
Johnson, Peri and Omur Harman~ah. 2015. Landscape, Politics, and Water in the Hittite Borderlands: Yalburt Yaylas1
Archaeological Landscape Research Project 2010-2014. In The Archaeology of Anatolia: Recent Discoveries
(2013- 2014) Volume I, S.R. Steadman and G. McMahon, eds., 255- 277. Newcastle upon Tyne: Cambridge
Scholars Publishing.
- . 2019. The Political Ecology of Roads and Movement: The Yalburt Yaylas1 Archaeological Landscape Research
Project 2018 Season. In The Archaeology of Anatolia: Recent Discoveries (2017-2018) Volume III, S.R.
Steadman and G. McMahon, eds., 193- 204. Newcastle upon Tyne: Cambridge Scholars Publishing.
Koi;:, Ayten, Nuretdin Kaymakc1, Douwe J.J. van Hinsbergen, and Klaudia F. Kuiper. 2017. Miocene Tectonic
History of the Central Tauride Intramontane Basins, and the Paleogeographic Evolution of the Central Anatolian
Plateau. Global and Planetary Change 158: 83-102.
Matthews, Roger. 2004. Landscapes of Terror and Control: Imperial Impacts in Paphlagonia. Near Eastern
Archaeology 7(4): 200- 211.
Sarre, Friedrich. 1896. Reise in Kleinasien, Sommer 1895; Forschungen zur seldjukischen Kunst und Geographie des
Landes. Berlin: Dietrich Reimer.
Turkcan, Ali Umut. 2005. Some Remarks on <;atalhoyuk Stamp Seals. In Changing Materialities at <;atalhoyiik:
Reports from the 1995-99 Seasons by Members of the <;atalhoyiik Teams, I. Hodder, ed., 175- 186. <;atalhoyuk
Research Project Volume 5. BIAA Monograph No. 39. London: British Institute at Ankara.
Wenzel, H. 1937. Forschungen in lnneranatolien, vol. 2; die Steppe als Lebensraum. Kiel: Schmidt & Klaunig.
Woodbridge, J., C.N. Roberts, A. Palmisano, A. Bevan, S. Shennan, R. Fyfe, W.J. Eastwood, A. Izdebski, C. <;ak1rlar,
H. Woldring, N. Broothaerts, D. Kaniewski, M. Finne, and I. Labuhn. 2019. Pollen-Inferred Regional Vegetation
Patterns and Demographic Change in Southern Anatolia through the Holocene. The Holocene 29(5): 728-741.
Ytld1z, M. and E.H. Bailey. 1978. Mercury Deposits in Turkey. Washington, D.C.: United States Government Printing
Office.
CHAPTER TWENTY-THREE
NEW ASSESSMENTS OF SETTLEMENT DISTRIBUTION

DURING THE BRONZE AND IRON AGES IN SOUTHEAST KONYA
{;iGDEM MANER
The stratigraphic and archaeological evidence of the Anatolian Bronze and Iron Ages in south central Anatolia,
specifically the Konya region, is comparatively unknown in comparison to other regions of Anatolia. The reason for
this is that in the Konya region only Boncuklu Hoyillc, (:atalhoyillc, and Karahoyillc (Konya) have been excavated. The
excavation at Karahoyillc by Sedat Alp (1994) is the only excavation of a Bronze Age site and represents some of the
cultural levels later than the prehistoric sites of Boncuklu and (:atalhoyiik. However, the excavations at Karahoyiik
were not really systematic; hence the results of its stratigraphy need to be handled with care. No excavation south of
(:atalhoyillc and in the southeastern Konya plain has yet been conducted; it would be important to excavate one of the
larger hoyillcs (settlement mounds) in the above-mentioned region to understand the stratigraphy and Bronze and Iron
Age cultures of the southern and southeastern Konya plain.
The evidence of the Bronze and Iron Age settlements and monuments of southeast Konya is mainly based on
survey results. A group of surveys in the southeast of the Konya plain conducted by (chronologically) James Mellaart
(1954, 1955, 1958, 1963), Semih Gi.ineri (1989- 91, 1990), and Hasan Bahar (2001, 2002, 2019; Bahar and KO(;ak
2008; Bahar and Ki.ii;:uk:bezci 2012) investigated not only certain regions but mainly visible hoyiiks. The southeastern
region of the Konya plain was not really included in the survey projects mentioned above and remained uninvestigated
until the Konya Eregli Survey Project (henceforth KEYAR) was initiated in 2013. This survey project has
systematically investigated Eregli, Karapmar, Halkapmar, and Emirgazi, which are located in the southeast of the
Konya region. The survey region is located north of the Bolkar Mountains (middle part of the Taurus Mountains), east
of Nigde, west of Ayranc1 (Karaman), and south of (:umra. The region under investigation is located on important
road networks, as from here the Mediterranean, Levant, and Mesopotamia, as well as the north, west and east of
Anatolia can be reached. The aim is to map the Bronze and Iron Age settlements, monuments, geo-environment, and
road networks and to illuminate the settlement pattern during these periods. Between 2013 and 2020, 87 settlements,
two tumuli, and one cave have been surveyed. With every season an additional research branch has been added to the
survey, such as geomorphology, geophysics, oral history, public archaeology, and a landscaping project.1 In the first
part of this contribution an overview of the results of the 2019 and 2020 survey seasons is offered, and in the second
part assessments on the settlement distribution in the southeastern Konya plain will be provided.
SUMMARY OF 2019-2020 SURVEY SEASONS
During the 2019 and 2020 survey seasons sixteen settlements were systematically investigated (Table 23-1, Figs.
23-1 and 23-2). The 2019 field season took place between 16- 29 May, 2019 (Maner 2020). The main focus of the
season was to continue the survey of the Karapmar district. The region of Meke Goli.i, Meke Dag1 and the craters, the
yaylas west, east, and south ofMeke Goli.i (Girginli Yaylas1, Ethemin Yaylas1, and Serpek Yaylas1),2 and the region
between Karaman and Karapmar, were investigated. There were no traces of ancient settlements or monuments around
Meke Goli.i or Meke Dag1. The craters are covered with dark grey and brown reddish volcanic pumice stone. Meke
Goli.i was one of the major salt sources from the Ottoman period until 1952 (Barutoglu 1961). For several years the
lake has been dried out, and the depots where the salt was stored have been closed. The crater lake is covered with a
thin white muddy crust, and one does not get the impression that this was once a major salt source (Fig. 23-3). The
Meke Goli.i is a landmark in the Konya plain and can likely be equated with the saltlick (lild) mentioned in the frontier
description in the treaty between Kuruntija and Tutgalija IV (Maner 2021a).
Four hoyillcs have been surveyed (#76-79; Table 23-1, Fig. 23-1) (Maner 2020). Three of them are located in the
Karapmar district and one in the Eregli district. A~ag1 Gondelen Hoyiik (#76), which was noticed during the 2018
season, is located on the main road from Eregli to A~ag1 Gondelen and is used as a cemetery. The hoyillc is around
1
An article in Turkish and English on eveiy season each year has been published. The English article is published in Anatolia
Antiqua and the Turkish in Ara§flrma Sonur;:lan Toplantis1 published by the Ministry of Culture and Tourism General Directorate
of Antiquities and Heritage. Furthermore see Maner 2017, 2019a, and 2019b, Maner and Kuruyay1rli 2018.
2The yaylas are named after the families who O'Ml them.
New Assessments of Settlement Distribution 301
Settlement Settlement Name Province, Altitude Size (ha)

Number District
76 A~ag1 Gondelen Eregli, A~ag1 1040 0,1
Hoyiik Gondelen
77 Gokgoz Tepesi Karapmar, Serpek 1014 1,56
Hoyiigii Yaylas1
78 Sm;an Hoyiik Karapmar, 1043 1,57
Girginli Yaylas1
79 Halibintol Hoyiik Karapmar, Serpek 1004 0,8
Yaylas1
80 Karaoren Emirgazi, 1319 48,1
Yerlesimi Karaoren
81 G1can Kalesi Emirgazi, 1539 2,2
Karaoren
82 Eski Belkaya Eregli, Belkaya 1196 37,2
Yerlesimi
83 Salma Tepe Karapmar, 1057 6,1
Oymah
84 Demirtepe Karapmar, 1135 0,4
Oymah
85 Adaca Ag1l Karapmar, 1102 0,5
Karakol Oymah
86 Dagoren Karapmar, 1732 20,2
Oymah
87 Oymah Asar Kale Karapmar, 1183 1,8
Oymah
88 Rakka Hoyiik Karapmar, 1075 2,6
Sazhpmar
89 Ye~iltepe Hoyiik Karapmar, islik 1009 4
90 Ovac1k Krater Karapmar, 1548 58,3
Yerlesimi Yesilyurt
91 Alitepe Hoyiigii Karapmar Merkez 1023 4,9
Table 23-1: Surveyed settlements in 2019 and 2020.
Investigated Site (2019): 76 . A~ag1 Gonde len HoyOk 77. Gokgoz Tepesi HoyOgO 78. Sm;:a n HoyOk 79. Hali bi ntol HoyOk
302 Chapter Twenty-Three: Maner
Investigated Site (20 20): 80. Kara6ren Yerle~imi 81. G1can Ka lesi 82. Eski Belkaya Yerle~imi 83. Salma Tepesi 84. Demirtepe 85. Adaca Ag1I Karakol
86. DaQOrenYerle~imi 87. OymalI Asar Kale 88. Rakka HOyUk 89. Ye~i ltepe HOyUk 90. Ovacik KraterYerle~imi
91. Alitepe Hoyogu
Figure 23-3. Meke Golii (drone image, September 2020).
124 m (north-south) by 92 m (east-west) large and around 2 m high. The pottery found in animal holes inside the
cemetery and on the slopes dates to the Early Bronze Age and Iron Age. Gokgoz Tepesi (#77), also known as Govezli
Hoyiik, is located 19.5 km south from Karapmar and 2.4 km north ofKavuklar village (Kamman). Guneri (1989-91:
103, fig. 38) and Brean A~km (A~kin et al. 2018) previously surveyed this site and noticed a major Late Bronze Age
and Iron Age settlement. The hoyiik is around 356 m (north-south) by 440 m (east-west) in size and around 7 m high
(Fig. 23-4). A rampart is visible around the settlement, which suggests the presence of a fortification wall. No
architectural remains were observed. Pottery was collected in three sample areas and includes Metallic Ware from the
Early Bronze Age, as well as ceramics from the Middle and Late Bronze Age, and Early-Late Iron Age .
.......-
Figure 23-4. Gokgoz Tepesi (drone image, May 2019).
Figure 23-5. Early, Middle, and Late Iron Age pottery from Sm;an Hoyiik.
Sm;an Hoyiik (#78) is located 6 km west of Gokgoz Tepesi and 20.3 km south of Karapmar. The settlement is
visible from Gokgoz Tepesi. The site is 450 m (north-south) by 350 m (east-west) and around 10 m high. The pottery
dates to the Chalcolithic, Early Bronze Age, Middle and Late Bronze Age (?), and Iron Age (Fig. 23-5).
Halibintol Hoyiigii (#79; also known as Serpek Hoyiik) is situated within Serpek Yaylas1 and 19.8 km south of
Karapmar. In comparison to the other hoyiiks in Karapmar this is a small settlement; it measures 308 m (north-south)
by 273 m (east-west) and is around 4 m high. The hoyuk is used for agriculture, and the surface is therefore disturbed
from ploughing. The collected pottery dates mainly to the Middle and Late Iron Age.
The 2020 season took place between 16 August and 13 September, 2020. The main regional focus of this survey
season was unsurveyed areas of the Eregli, Emirgazi, and Karapmar districts (Table 23-1, Fig. 23-2; #80-91). 3 In the
Eregli district Eski Belkaya Y erle~imi (#82) was reinvestigated. During the 2014 field season a short investigation
Figure 23-6. Rock massif with foundation of a wall to its eastern side at Eski Belkaya Yerle~imi.
was conducted, which was interrupted because of bad weather conditions. The ruins of Eski Belkaya Y erle~imi are
located on the northern part of Ansama Dag1 (also known as Kotii Dag) at 1196 m asl, 3 .9 km northwest of Belkaya.
The foundations of buildings which are located here belong to an Islamic settlement, which is verified by an Islamic
cemetery and pottery. Around 70 m to the north a rock massif is located, which is around 46 m (north-south) by 35 m
(east-west) in size. This rock massif is of interest (Fig. 23-6) because foundations on its eastern side suggest that they
belong to a protective wall, which once probably surrounded the massif. The foundation of this wall is made of large
Figure 23-7. Boulder with square lewis holes at Eski Belkaya Yerle~imi.
boulders (ca. 1.5-2 m long, ca. 1-1.5 m wide and ca. 0.5-1 m high), which are reminiscent of cyclopean walls.
Systematically positioned holes on the rock massif suggest that it was used as a stone quarry. Some rocks have square
lewis holes (dowel holes) (Fig. 23-7), a technique known from Iron Age sites. Square lewis holes are positioned on top
of the reliefs from Carchemish, Zincirli, and Sak9eg6zu (Ozyar 1998: 636). The lewis holes are used to place pegs and
tenons.
3
For a detailed report of the 2020 season see Maner 2021 b.
Pottery around the rock massif dates to the

Middle and Late Iron Age, and one Late Bronze
Age sherd was found. The lewis holes and the Iron
Age pottery suggest that it might have been used as
a stone quarry and workshop during the Iron Age
and possibly as early as the Late Bronze Age.
Around 170 m north of this rock massif, on top of a
hill, the remains of a fort were discovered in 2014
(#39 Sar1y6ren Kalesi; Maner 2014: 267-268);
however as no pottery was found there it is not
really possible to date the structure. The masonry
suggests a date ranging from the Early Bronze Age
to Late Iron Age.
A major discovery was made at Karaoren-an
inscribed block with hieroglyphic Luwian (henceforth
Karaoren 2) (Maner 2021 b ). Karaoren is located on
the northern part of Karacadag (district Emirgazi) at
around 1319 m asl and around 6.5 km south-
southwest of Emirgazi (Fig. 23-2). The village is
situated next to a large ruin. We had previously
investigated the site in 2016 (Maner 2017). This
newly discovered inscription is probably a part of
the missing inscription from Karaoren (henceforth
Karaoren 1) for which the Ministry of Culture and
Tourism of Turkey and Interpol have been
conducting a search since 2016 4 (Maner et al. 2021).
Figure 23-8. The newly discovered hieroglyphic Luwian The newly discovered hieroglyphic Luwian
inscription at Karaoren (Karaoren 2). inscription was used as a door jamb at the left comer
of a village house in Karaoren (Fig. 23-8). The
inscribed slab is 82 x 44 x 23.5 cm in dimensions and is made of basalt, with a light grey rose colour. The inscription
is made in relief technique; after the removal of the block in December 2020, it became clear that not only did the
narrow side have two lines of inscription, but the back part, which was not visible before removal, was also inscribed.
Karaoren 1 contains information on military campaigns to the west (Maner et al. 2021 ), Karaoren 2 also mentions
military campaigns and refers to the Emirgazi altars. 45 Both likely date to the time period of the Hittite Great King
TutlJalija IV (ca. 1237-1209 BCE).
The discovery of Karaoren 2 invites us to rethink the meaning of the monuments which were discovered at
Emirgazi by T. Callander, William Ramsay, and Sedat Alp (Masson 1979). Three altars and two inscribed blocks were
found at various places at Emirgazi, which is north of Karacadag and situated on the southern slope of Ansama Dag1.
The inhabitants of Karaoren point out that their ancestors told them that the ruins had been dismantled and shipped to
various places such as Emirgazi, Karapmar, Eregli, Hotam1~, Konya, and even Nigde. This might raise the question
whether the Emirgazi monuments might originally have come from Karaoren and whether Emirgazi was in fact not
their original location. An alternative is that the Hittite settlement which is probably beneath the Roman and Byzantine
foundations at Karaoren extended all the way up to Emirgazi. The strategic location of Karaoren, and also Emirgazi, is
quite important as it controls the road from Nigde and Tyana to Karapmar and Aksaray. Hence it would make sense
that the path between Karacadag and Ansama Dag1 would be protected from both sides. Klaus Belke and Marcel
Restle suggested that the Byzantine Bisutum Hyde is located at Goloren on Karacadag (Belke 1984: 174). The Hittite
city Uda is generally equated with Hyde (Forlanini 2017: 241, fu. 26), though the grounds for this are not strong.
However, Goloren, in comparison to Karaoren, is a small and less important site; I have recently suggested that Hyde
might be located at Karaoren, implying also that the Hittite city Uda was situated here as well (Maner 2021b).
An important fort known as G1can Kalesi (#81) is located 2.9 km south of Karaoren on the slope of Karacadag.
This fort had not been investigated, but it is briefly mentioned by ibrahim Giindiiz in his book about Karapmar (1980:
198). The fort is 206 m (northwest-southeast) x 72 m (east-west) in area and is located at 1540 m asl (Fig. 23-9).
It takes around 30 minutes to reach G1can Kalesi from Karaoren. Massive fortification walls made of medium to
large size roughly shaped stone boulders in cyclopean masonry protect the fort. The walls are in some places
preserved up to a height of 2-2.5 m. The drone image indicates towers and rooms within the fort. The fort is
positioned on a natural hill, and outcrops within the fort were probably used for additional outlooks. There are no
architectural remains on these outcrops; however, it might be possible that some towers made of wood or mudbrick
were placed here. The pottery indicates that it was used during the Middle and Late Bronze and Iron Ages (Fig. 23-
10).
4
https://kvmgm.ktb.gov .tr/TR-162927/konya-ili-emirgazi-ilcesi-karaoren-koyunden-kaybolan-lu- .html (17.07.2021).
5
Publication in preparation by <;igdem Maner and Mark Weeden.
(
.- ''
-6
o 5cm
Figure 23-9. G1can Kalesi Figure 23-10. Middle and Late Bronze Age
(drone image, August 2020). and Iron Age pottery from G1can Kalesi.
Figure 23-11. View from G1can Kalesi to the southwest towards Karapmar and Hotam1~.
There is no doubt that G1can Kalesi was related to the Hittite settlement at Karaoren. The fort overlooks the whole
northern side of Karacadag to Hasandag. To the southwest one overlooks the dried out Karapmar Lake, Karapmar to
Hotam1~, and probably on clear days, all the way to K1zildag and Karadag (Fig. 23-11 ). To the east it controls the
eastern peaks of Karacadag, and the roads to Beyoren and Dagoren.
For many years the plan has been to explore Dagoren (#86); however, the steep slope of Karacadag (and the
exhausting memory of the hike from Ovac1k to Dag GolU in 2016 and 201 7), as well as the distant location of Dagoren
discouraged me from accomplishing this previously. Given that 2020 was going to be the final survey
season, a hike up to Dagoren was unavoidable. Dagoren is located at 1732 m asl and is situated around 3.5 km north-
northeast of Oymah. The climb from Beyoren to Dagoren took around 4 hours, guided by a local shepherd. Dagoren is
mentioned for the first time in the report by Callander, who came here in 1904 (Callander 1906: 176-177). Getrude L.
Bell visited Dagoren in 1906 (Ramsay and Bell 1909). Both published a few photographs of the ruins, which show an
extensive Byzantine settlement. The fact that the Hieroglyphic Luwian inscription was once used as a spolium in a
Late Antique structure at Karaoren increased our curiosity to see whether there would be pre-Roman structures,
pottery, monuments, or inscriptions at Dagoren. Belke and Restle visited Dagoren in the 1970s during their research
trips for their publication for the Tabula lmperii Byzantini series (Belke 1984: 155). They suggested that Dagoren is
the refuge settlement of Hyde. The drone images taken in September 2020 of Dagoren (Fig. 23-12) show a planned
settlement, and hence it might be a settlement used for a longer time period rather than a refuge settlement, which
would have been built in haste, as suggested by Belke and Restle (Belke 1984: 155).
Figure 23-12. Drone image ofDagoren (September 2020).
Upon our arrival on Karacadag, parts of a foundation of a long wall leading from the top edge of Karacadag
towards Dagoren were visible. The foundations might belong to a protective wall. Dagoren is situated in a shallow
depression and protected by a hill on the eastern side. The ruins at Dagoren are spread over an area of around 20 ha.
The ground is covered with boulders. The walls of some buildings are preserved up to 3-4 min height (Fig. 23-13).
Stone blocks with incised grape leaves, spiral motifs, and inscriptions (ancient Greek script) are scattered on the
ground. Hence almost no pottery was visible; however two sherds dating to the Middle Iron Age were found. Some of
the ruins are used by shepherds as pens and also huts for themselves. Shepherds from Oymah come and spend several
months with their animals here.
A chain of forts has been discovered in the Bolkar Mountains in the district of Halkapmar (Maner 2017). Most of
these date to the Middle and Late Iron Age. The purpose of these forts was to control the road networks on the Taurus
Mountains and into the Konya plain and down to the Mediterranean. Another chain of three forts was discovered
southeast of Oymah. These are Demir Tepe (#84), Adaca Agil Karakol (#85), and Oymah Asar Kale (#87) (Fig. 23-
2). The pottery from these sites all dates to the Middle and Late Iron Age. They seem to protect the road along the
eastern side of Karacadag towards Emirgazi. This must have been one of the main roads from Eregli to Emirgazi and
also to Bor.
Two hoytiks have been systematically surveyed: Rakka HoyUk (#88) and Ye~iltepe HoyUk (#89). Rakka HoyUk is
mainly used as a cemetery; hence only the slopes were surveyed. The settlement is around 179 m (north-south) x 166
m (east-west) in size and around 2 m high; it is located inside the district town Sazhpmar. The pottery dates to the
Early Bronze Age, Late Bronze Age, and Iron Age. Y e~iltepe Hoytik is located 1.9 km northeast of islik, on the
highway from islik to Karapmar. The hoyUk is ca. 2 m high and 126 m (north-south) x 136 m (east-west) in size (Fig.
23-14). It seems to be a significant settlement, as it is on the main road from Karaman to Karapmar. The pottery dates
to the Chalcolithic, Early and Late Bronze Ages, and Iron Ages.
Figure 23-13. Dagoren.
Figure 23-14. Ye~iltepe Hoyuk.
OBSERVATIONS ON SETTLEMENT DISTRIBUTION
The results of the KEYAR survey show that during the Early Bronze Age, settlements were located on the shores
of lakes, such as Karapmar Golu, Hotam1~ Golu, and Akgol. The hunting grounds must have been one of the reasons
why they chose these locations. Furthermore, there were a few settlements in the alluvial fans between Karacadag and
the Eregli-Bor plain. A preferred region was Halkapmar, probably because of the fertile fields, the fresh-water spring
ofivriz and Delimahmutlu, the copper and silver mines, and the road networks to the south. Most of the Early Bronze
Age settlements continue also into the Middle and Late Bronze Ages. The earliest evidence of habitation on
Karacadag dates to the Middle Bronze Age.
Labarnas (tlattusili I?) claims to have conquered every area in the Lower Land during his reign (also Ijupisna and
Tuwanuwa) (Kessler 1975: 500). The main reason must have been to secure the route over the Taurus Mountains
down to Syria. Hence the southeastern Konya region has a very important strategic meaning for the Hittites. This is
also supported by the Emirgazi and Karaoren monuments.
During the Middle Iron Age there was a settlement explosion, as all of the surveyed sites revealed Middle Iron Age
pottery. During this period forts were built along important road networks, such as the crossroads at the Bolkar
Mountains to the south or the roads along the marshes. These forts create a sort of network, which seems to be typical
for the Iron Age. During the Iron Age the region was part of the Kingdom of Tabal (Aro 1998) and also under the
influence of various other civilizations. With the climatic changes (dry conditions and no harvest) in the Hittite
heartland (within the K1z1hrmak range) people might have migrated to various regions of Anatolia, including
southeastern Konya. This could be one explanation for the increase of the settlements. Also, the geopolitical location,
fertile fields, metal-rich mountains, and communication connections in every direction must have attracted people to
settle here.
Today the landscape of the southeastern part of the Konya plain is arid and dusty, and one can take the highway
from Konya or Aksaray to Adana directly, and also to other provinces. But the palaeo-environmental landscape of the
Bronze and Iron Ages of the southeastern Konya plain was different, as some areas were covered with marshes
(Kuzucuoglu 2019; Kuzucuoglu et al. 2018). The location of the settlement indicates which areas were suitable for
habitation. An interesting fact is that the size of hoyiiks decreases from <;umra to Karapmar and from Karapmar to
Eregli. The marshes of the Hotam1~ and Akgol must have prevented the creation of large settlements, and in some
areas no settlements at all could be established.
CONCLUSION AND FUTURE WORK
The results of KEYAR illuminate the ancient landscape and settlement pattern and preferences from the Early
Bronze to the Late Iron Age. The location of the settlements follows a) the geoenvironment, b) strategic points, and c)
substructure of agriculture and animal husbandry. This region was important from the beginning of human history
onwards,5 as it was a setting of migration, military conflicts, subsistence, and economic gain. An excavation of a
hoyiik south of <;umra or north of Konya is a desideratum, as it would help us understand the stratigraphy of the post-
<;atalhoyiik Konya plain.
Initially, 2020 was plarmed as the final KEYAR season. However, the discovery of the inscription at Karaoren
(Karaoren 2) and the unfinished research and investigation at Karaoren and Dagoren has led to a plarmed final season
in October 2021. During this campaign the areas of Karacadag (the hills east of Karaoren and Dagoren), the forts to
the south (Mennek, Kes;i Kalesi, and Se Kalesi), and the villages on Karacadag will be investigated for spolia with the
hope of illuminating especially the Hittite and Iron Age archaeology and history of Karacadag. Furthermore,
geophysical prospections are plarmed at Zengen P1narb~1 Mevkii, where a rectangular indentation next to a dried-out
spring cave was discovered during the 2014 season (#22, Maner 2015). Indentations of this type are known from the
Late Hittite period as bases for stele (for example in ivriz; Bier 1976). A modern water pool a few meters away from
the spring suggests that there might have been a forerunner for the pool. Pools dating to the Hittite period are known
for example from Eflatun Pman in Bey~ehir. Geophysical prospection will shed light on the question of whether there
was a forerunner for this pool, water canals, and the whole area during the Hittite period.
ACKNOWLEDGEMENTS
I am thankful to the Ministry of Culture and Tourism General Directorate of Antiquities and for the necessary
permissions to work in the region, the ministry representatives Mahmut Altuncan from the Konya Eregli Museum
(2019) and Enver Akgiin from the Konya Museum (2020). Moreover, I would like to thank team members Dr. Belgin
Aksoy (2020), Muhip <;ark1 (2019- 2020), Dr. Emre Kurus;ayuh (2020), Zeynep Ku~dil (2019), Francesco Tessa
(2019), Dicle Yaz (2020), and Sadiye Kaya our driver. The survey is generously supported by Kos; University, Avis
Turkey, Ozkos;lar Otel, and Derya Lokantas1. I also owe thanks to Ian Rutherford for proofreading the English.
REFERENCES CITED
Alp, Sedat. 1994. Korrya Civarznda Karahoyiik Kazzlannda Bulunan Silindir ve Damga Miihilrleri. Ankara: Turk
Tarih Kurumu Bas1mevi.
5 The Aceramic settlement remains at Yagmapmar Prehistorik Y erle~imi (Yagmapmar Prehistoric Site) which have been discovered
during the 2018 field season are an important indicator (Maner 2019b, 2019c).
Aro, Sanna. 1998. Tabal. Zur Geschichte und materiel/en Kultur des zentralanatolischen Hochplateaus von 1200 bis
600 v. Chr. Unpublished Ph.D. dissertation, University of Helsinki.
A§km, Brean, Mehmet Kurt, Mehmet Alkan, Hatice Korsulu, H. Ertug Ergiirer, and Hatice Ergiirer. 2018. Karaman ili
ve Mersin iii Mut il<;esi antik donem yerle§imleri ve yerle§im di.izenlemesi arkeolojik yi.izey ar~tumas1 2016 y1h
<;ah§malan. Ara~tzrma Sonuc;lan ToplantlSl 35(1): 411-424.
Bahar, Hasan. 2001. Ara§t1rmac1 Ogretmen ibrahim Gi.indi.iz'i.in Ki§iligi; Karapmar <;evresi Eski<;ag Ar~tirmalan. In
Karapmar Sempozyumu, Konya, Y. Ki.i<;i.ikdag, ed., 147-163. Konya: Konya Karapmar Belediyesi.
- . 2002. Konya ve Karaman Illeri Yi.izey Ar~tlrmalar1 2000. Ar~tmna Sonuc;lan Toplantzsz 19(2): 257- 270.
- . 2019. Some Remarks on the Early and Middle Iron Age Pottery and Settlements in the Konya Region. In
Crossroads: Konya Plain from Prehistory to the Byzantine Period/Kav~aklar: Prehistorik <;ag'dan Bizans
Donemine Konya Ovasz, <;. Maner, ed., 207- 223. istanbul: Ege Yaymlar1.
Bahar, Hasan and Ozdemir Ko<;ak. 2008. The Transition from Bronze to Iron Age in Lycaonia and its Vicinity. In
Proceedings of the 4th International Congress of the Archaeology of the Ancient Near East, Volume 2: Social and
Cultural Transformation: The Archaeology of Transitional Periods and Dark Ages Excavation Reports, H. Ki.ihne,
R.M. Czichon, and F.J. Kreppner, eds., 9-10. Wiesbaden: Harrassowitz Verlag.
Bahar, Hasan and Hatice G. Ki.i<;i.ikbezci. 2012. 2010 Y1h Konya ve Karaman illeri ile il<;eleri Arkeolojik Yi.izey
Ara§t1rmas1. Ara~tzrma Sonuc;lan Toplantzsz 29: 97-116.
Barutoglu, Omer H. 1961. Tiirkiye Tuz Yataklan. Bi/imsel Madencilik Dergisi 1(2): 68- 78.
Belke, Klaus. 1984. Ga/alien und Lykaonien. Tabula Imperii Byzantini 4. Wien: Verlag der Osterreichischen
Akademie der Wissenschaften.
Bier, L. 1976. A Second Hittite Relief at Ivriz. Journal ofNear Eastern Studies 35(2): 115-126.
Callander, T. 1906. V. Explorations in Lycaonia and Isauria, 1904. In Studies in the History and Art of the Eastern
Provinces of the Roman Empire, W.M. Ramsay, ed., 157-180. Aberdeen: University of Aberdeen.
Forlanini, Massimo. 2017. South Central: The Lower Land and Tarhuntassa. In Hittite Landscape and Geography, M.
Weeden and L.Z. Ullmann, eds., 239- 252. Leiden and Boston: Brill.
Gi.indi.iz, Ibrahim. 1980. Biitiin Yonleriyle Karapmar. Konya: Kuzucular Ofset.
Gi.ineri, Semih. 1989-91. Orta Anadolu Hoyi.ikleri, Karaman-Eregli Ara§tlrmalar1. Tiirk Arkeoloji Dergisi 28-29: 97-
114.
- . 1990. Orta Anadolu Hoyi.ikleri, Karapmar, Cihanbeyli, Sarayoni.i, Kulu Ar~tumalan. Ara~tzrma Sonuc;larz
Toplantzsz 7: 323- 39.
Kessler, Karlheinz. 1975. Hupisna. In Real/exikon der Assyriologie, Vierter band Ha-a-a -Hystaspes, Erich Ebeling
and Ernst F. Weidner, eds., 500. Berlin: De Gruyter.
Kuzucuoglu, Catherine. 2019. Geomorphological Landscapes in the Konya Plain and Surroundings. In Landscapes
and Landforms of Turkey, C. Kuzucuoglu, A. <;iner, and N. Kazanc1, eds., 353-368. Cham: Springer.
Kuzucuoglu, Catherine, Mi.isli.im Demir, Ali Giirel, Jean-Pascal Dumoulin, and <;igdem Maner. 2018. Son Buzul
Maksimum (LGM)-Gey Holosen Eregli Havzasmm Paleo<;evresel Evrimi: Jeolojik indikatorler Yard1miyla ilk
Sonu<;lar. Omer Halisdemir Oniversitesi Miihendislik Bilimleri Dergisi 7(3): 1214-1219.
Maner, <;igdem 2014. Preliminary Report on the First Season of the Konya-Eregli (KEYAR) Survey 2013, Anatolia
Antiqua 22: 343-360.
Maner, <;igdem 2015. Preliminary Report on the Second Season of the Konya-Eregli (KEYAR) Survey 2014,
Anatolia Antiqua 23: 249- 273.
Maner, <;igdem 2017. From the Konya Plain to the Bolkar Mountains: The 2015-2016 Campaigns of the KEYAR
Survey Project. In The Archaeology of Anatolia Volume II: Recent Discoveries (2015-2016), S.R. Steadman and
G. McMahon, eds., 347- 373. Newcastle upon Tyne: Cambridge Scholars Publishing.
- . 2019a. Networks, Crossroads and Interconnections in the Eregli Plain during the Bronze and Iron Ages. In
Crossroads: Konya Plain from Prehistory to the Byzantine Period/Kav~aklar: Prehistorik <;ag'dan Bizans
Donemine Konya Ovasz, <;. Maner, ed., 83-106. istanbul: Ege Yaymlar1.
- . 2019b. Inside Tarhuntassa: A Systematic Survey of Karapmar, Konya. The 2017-2018 Field Seasons of the
KEYAR Survey Project. In The Archaeology of Anatolia Volume III: Recent Discoveries (2017- 2018), S.R.
Steadman and G. McMahon, eds., 205-217. Newcastle upon Tyne: Cambridge Scholars Publishing.
- . 2019c. Preliminary Report on the Sixth Season of the Konya Eregli, Karapmar, Halkapmar and Emirgazi Survey
Project (KEYAR) 2018. Anatolia Antiqua 27: 161-168.
- . 2020. Preliminary Report on the Seventh Season of the Konya Eregli, Karapmar, Halkapmar and Emirgazi Survey
Project (KEYAR) 2019. Anatolia Antiqua 28: 173- 183.
- . 2021a. Eine bedeutende Salzquelle in Karapmar Konya - Meke Goli.i (Meke Maar) und ein
Gleichsetzungsvorschlag filr liki ,,Salzlecke" im Staatsvertrag des Kuruntija und Tutl].alija IV. Altorienta/ische
Forschungen 48(2). Forthcoming.
- . 2021b. Preliminary Report on the Eighth Season of the Konya Eregli, Karapmar, Halkapmar and Emirgazi Survey
project (KEYAR) 2020. Anatolia Antiqua 29. Forthcoming.
Maner, <;igdem and Emre Kuru<;ayuh. 2018. ivriz Ambarderesi K1zlar Oglanlar Saray1 (Manastm) Magaras1'nda
Ara§tirmalar. Belleten 82.295: 785-802.
Maner, <;igdem, Mark Weeden, and Metin Alparslan 2021. Archaologische Forschungen auf dem Karacadag und eine
hieroglyphenluwische Inschrift aus Karaoren. Altorientalische Forschungen 48(2). Forthcoming.
Masson, Emilia. 1979. Les inscriptions louvites hieroglyphiques d'Emirgazi. Journal des Savants 1979: 3-52.
Mellaart, James. 1954. Preliminary Report on a Survey of Pre-Classical Remains in Southern Turkey. Anatolian
Studies 4: 175- 240.
- . 1955. Iron Age Pottery from Southern Anatolia. Belleten 19: 115-36.
- . 1958. Second Millenium Pottery from the Konya Plain and Neighborhood. Belleten 22: 311---45.
- . 1963. Early Cultures of the South Anatolian Plateau, II: The Late Chalcolithic and Early Bronze Ages in the
KonyaPlain. Anatolian Studies 13: 199- 236.
Ozyar, Ash. 1998. The Use and Abuse of Re-use at Karkemish. In Light on Top ofthe Black Hill. Studies Presented to
Halet <;ambel. Karatepe'deki I~1k. Ha/et <;ambel'e sunulan Yazilar, G. Arsebillc, M.J. Mellink, and W. Schirmer,
eds., 633-640. istanbul: Ege Yaymlar1.
Ramsay, William M. and Gertrude L. Bell 1909. The Thousand and One Churches. London: Hodder and Stoughton.
CHAPTER TWENTY-FOUR
GET GOING WITH SURVEY LEGACY DATA.

THE 2018 AND 2019 ARCHAEOLOGICAL SURVEY RESULTS
FROM THE SAGALASSOS ARCHAEOLOGICAL RESEARCH PROJECT
RALF VANDAM, PATRICK T. WILLETT,

W. CHRISTOPHER CARLETON, AND JEROEN POBLOME
The Sagalassos Project has a track record of more than two-and-a-half decades of archaeological surface
prospection, in which we have implemented various methodologies and worked on different scales, to address the
specific research questions at hand (see Vandam et al. 2019b for a full overview of the survey history). This long
history of survey research has resulted in the creation of a unique regional archaeological database comprising more
than 704 sites at 289 geographical locations spanning the Middle Palaeolithic to the Late Ottoman period (Fig. 24-1 ).
This enables a wide range of survey-related research topics for the project, including archaeological predictive
modelling, which we have initiated in recent years (Willett et al. 2021). Such methods can be extremely useful in
expansive and inhomogeneous landscapes like that of the Sagalassos study area, which is a patchwork of different
ecological zones comprising disparate economic potentials.
Legend: Intensive survey areas

• Known sites D Derekoy-Hisar
D Suburban D Burdur Plain
D Bereket D Bagsaray
Figure 24-1. The study region of Sagalassos, showing all known sites and
the areas where intensive archaeological surveys were organized.
The aim of our predictive research is two-fold. First, by creating a reliable model of archaeological potential, the
survey project will have a means of predicting the likely presence of sites and the settlement history of a given
location in the study area, allowing future survey efforts to focus on the areas with the greatest archaeological
potential. This can save valuable time and energy, especially in the more difficult terrain of the mountainous study
area where accessibility is a major challenge. As a resource for informing intensive field walking campaigns, it can
also allow the project to efficiently home in on certain lacunae in our knowledge, whether chronological or spatial,
with a high degree of reliability. This enables us to create links between the areas that have been the focus of previous
surveys, and build a better understanding of areas that are more difficult for researchers to access, such as areas far
from roads or under heavy cultivation. Secondly, modelling the archaeological potential of the landscape will provide
valuable insights into the suitability for human occupation of the many landscapes that comprise the study area.
Ancient settlement patterns can act as a strong indication for certain environmental conditions, such as arability and
water availability, and will help inform the reconstruction of the ancient environment and previous patterns of land-
Get Going with Survey Legacy Data 313
use (e.g., Ginau et al. 2019; Schmidt et al. 2012; Toth et al. 2011). The presence or absence of sites during certain
periods as the environment changed over time can be a broad indication of the general environmental conditions
affecting our survey areas. Overall, this initiative will give us a better understanding of the entire territory of
Sagalassos, in regards to both the archaeological and environmental histories. With this in mind, prior to the 2018
Sagalassos field campaign, models were created to predict the archaeological potential of the landscapes comprising
the 1200 km2 study region. During the subsequent field campaign, these models were tested by pedestrian ground-
truth survey to assess their reliability. In total, twelve previously unknown sites were discovered over the course of the
survey, many representing multi-period artefact concentrations, which we will discuss in detail in this chapter.
In addition, in 2019 we initiated a re-documentation survey of previously known sites that were recorded in the
1990s during extensive surveys conducted by the project (Vanhaverbeke and Waelkens 2003). It has been illustrated
that re-visiting sites has great benefit and is crucial for many reasons (Attema et al. 2020). In our case, re-visitation
allowed us to improve the accuracy of our databases regarding geographical positioning (due to advancement in
technology), chronology, scale, and site nature, thereby improving future outcomes for data-driven approaches.
Importantly, re-visitation also provided an opportunity to update assessments of the conditions of earlier documented
sites in the study area, with an initial focus on the sites of the Aglasun Valley. These condition updates are essential
for heritage resource management. Overall, this work has allowed us to contextualize the survey efforts ofrecent years
within the valley and to complete our documentation of the valley as a whole. In 2013 we undertook similar research
in the southwestern part of the study area with great success (Waelkens et al. 2014). Through the course of these
efforts, it has become clear to us that these activities are necessary in order to conduct synthetic research on the
countryside.
THE 2018 CAMPAIGN
Predictive Modelling
Computational methods of predictive modelling were introduced into the field of archaeology in the 1970s and
have been implemented to varying degrees of success (Wansleeben and Verhart 1997; Wheatley and Gillings 2002;
van Leusen et al. 2005). Typically, these models attempt to indicate the probability of the presence of archaeological
sites in areas that have not previously been directly sampled based on analysis of a set of known site locations. The
main assumption behind predictive modelling is that the location of archaeological remains in the landscape is not
random but relates to certain characteristics of the cultural and natural environment. In the case of the Sagalassos
project, a method was needed that could create reliable predictive surfaces for a highly varied landscape and across a
wide expanse of time and archaeological possibilities.
The type of models that were created for the study area are known as locally-adaptive models of archaeological
potential (LAMAP; Carleton et al. 2012, 2017). LAMAP was developed as an alternative to regression and weights-
of-evidence based approaches, and is a method of prediction that can function well in inhomogeneous landscapes.
Rather than attempt to predict the location of archaeological sites, LAMAP instead aims to estimate the archaeological
potential, or relative suitability, of land parcels for specific human activities defined by a set of known sites, in an area
of interest. Reliable predictive modelling is thus able to mitigate the time and resource constraints often faced by
archaeological survey projects by providing a means of gauging the archaeological potential of a landscape based on
data from previous research conducted there.
The initial stage of building the model for our research area consisted of entering the locations of all known sites
and indicating the natural and cultural landscape features that might have influenced the choices of ancient people to
settle there for each period. Environmental characteristics such as elevation, slope, and proximity to waterways, as
well as cultural variables like proximity to urban centres and ancient roadways, were used to compare between the
known locations of sites and the so-far undocumented parcels of land within the territory. The model divided the
entire 1200 km2 study area into 50 x 50 m "cells" and compared each cell with the information from the already
known sites. This resulted in maps describing the probabilities that these as yet undocumented locations have
landscape and cultural characteristics similar to those of the known site locations. The degree of similarity implies the
level of potential suitability for the human activities described by the set of known sites. The degree of congruence
was represented on a scale of 0-5 "quintiles" created by binning the original continuous similarity measures into five
bins with equal portions of the study area represented in each bin. A "O" indicates a low level of similarity between a
given undocumented location and the known sites, which implies that the activities represented at the known sites
were very unlikely to have taken place in the target undocumented location. Conversely, a "5" indicates a high degree
of similarity, implying that the target undocumented location would have been very suitable for whatever activities
occurred at the known site locations. Seven individual models were created using LAMAP, one for each of the major
periods represented in the archaeological record of the territory since the mid-Holocene, and distinguished by a
distinct shift in settlement pattern and/or material culture: Late Prehistory (ca. 6500 to 2500 BCE), Iron Age-Archaic
(ca. 1150 to 546 BCE), Achaemenid-Hellenistic (ca. 546 to 25 BCE), Roman Imperial (ca. 25 BCE to 300 CE), Late
Antique (ca. 300 to 700 CE), Byzantine (ca. 700 to 1200 CE), and Late Ottoman (ca. 1700 to 1921 CE). After these
seven models were created for the territory, they were then combined to produce a map that indicated the overall
314 Chapter Twenty-Four: Vandam et al.
aggregated score for each 50 x 50 m cell (Fig. 24-2). This aggregated score ranged from Oto 35, with "35" being the
highest archaeological potential.
Figure 24-2. The aggregated LAMAP model for the study region of Sagalassos for seven archaeological periods.
The grey scale represents archaeological potential on a 0-35 scale with lighter shades representing higher potential.
Pedestrian Validation Survey
The reliability of the models for our study area was then tested by pedestrian validation survey during the 2018
campaign. This entailed surveying a random sample of cells from within the prediction areas and comparing the
results to the scale of potential as predicted by the model. In order to achieve a representative sample, the survey
aimed to document multiple 50 x 50 m cells for each value, within the entire range of potentials predicted by the seven
individual models, in addition to 2-3 cells per aggregated value of 0-35 from the aggregate model. The sample cells
that were surveyed came mostly from the regions within the Sagalassos study area that have produced the greatest
amounts ofpalaeoenvironmental data (i.e., Aglasun Valley, Gravgaz, Mamak, Bugduz, and Bereket; see Bakker 2019;
Verstraeten et al. 2019; and Fig. 24-3). The survey methodology was a variation on the standard
Figure 24-3. Overview of the surveyed fields for the validation of the
LAMAP model in the territory of Sagalassos.
Sagalassos survey method (see Vandam et al. 2017, 2019c), in this case entailing three surveyors walking transects 1
m wide and 50 m long while maintaining 25 m distance between, and walking due south to north rather than following
modern field contours. Using a total collection strategy within each transect, all finds from a single cell were bagged
together. New sites that were identified through the survey were documented, and an extra grab sample of the surface
materials was taken if needed to secure the dating and nature of the site. Based on these finds, we aimed to attain an
interpretation and date of the discovered concentrations.
Results
Over the course of the 2018 campaign, a total of 101 fields were surveyed: 93 cells from the model, plus 8 other
fields adjacent to the cells where we identified archaeological remains. This corresponds with a surveyed area of about
0.27 km2 • Due to the validation process requiring the surveying of the entire range of potentials produced by the
models, many fields with low potentials were documented as having little or no archaeological materials on them,
while others with high potentials produced abundant finds, with some exceptions in both cases. In total, 12 previously
unknown sites were discovered over the course of the survey, many represented by multi-period surface artefact
concentrations (Table 24-1 ). In terms of the number of finds, the survey results translate to 1209 sherds collected, 328
Site code Villa2e Dating Interpretation Potential

CAN16 MAMAK Chalcolithic (small) Village 9
CAN17 MAMAK Late Ottoman Pastoral Farmstead 25
KAA09 GRAVGAZ Byzantine Farmstead 23
C,::ENDIK-
SUL05 Byzantine - Late Ottoman Farmsteads 19
SULUDERE
BAYINDIR-
YAK03 Roman - Byzantine - Late Ottoman (small) Village 26
BUGDUZ
BAYINDIR- EBA - Iron Age - Hellenistic -
BUGOS Farmsteads 28
BUGDUZ Roman
BEREKET- Chalcolithic - Roman - Byzantine -
KOK02 (small) Village 32
KOKEZ Late Ottoman
EASTERN
Late Antique - Byzantine - Late
DER20 AGLASUN Farmsteads 19
Ottoman
VALLEY
EASTERN
Hellenistic - Late Antique -
DER21 AGLASUN Farmsteads 16
Byzantine - Late Ottoman
VALLEY
EASTERN
CAM0l AGLASUN Palaeolithic Possible Camp Site 8
VALLEY
EASTERN
CAM02 AGLASUN Late Antique - Byzantine (small) Village 14
VALLEY
EASTERN
Chalcolithic - Roman - Late
CAM03 AGLASUN Rock Shelter 22
Ottoman
VALLEY
Table 24-1. The newly discovered sites through the validation of the
predictive models with their aggregated predicted score.
tiles counted, and 38 lithic artefacts collected. Among the periods documented were Early Holocene, Late
Chalcolithic, Classical-Hellenistic, Roman Imperial, Late Antique, Byzantine, and Late Ottoman. The best represented
period was the Late Ottoman period and to a lesser degree the Chalcolithic, Late Antique, and Byzantine periods. The
attested sites ranged from rock shelters to stone-built structures and water wells. The results per research area are
discussed in detail below.
Mamak
The Mamak area was selected as a research area because it had never been intensively investigated
archaeologically, rather only subject to the extensive surveys of the Sagalassos team in the 1990s (Vanhaverbeke and
Waelkens 2003) and bordered the previous intensive survey research area around Derekoy-Hisar (Vandam et al.
2019c). Currently the area is used mainly for clay extraction, as was also the case in the past (Poblome et al. 2002). It
is an area where environmental testing has been carried out previously and which possibly holds future potential.
During the 2018 campaign we mainly focused on the southern plain, where 10 fields were surveyed in total,
comprising areas predicted to hold both high and low potentials, resulting in the discovery of two new sites.
The first, dubbed CAN16, has been dated to the Chalcolithic period. The site is located in the southern foothills 2.5
km east of modem day Mamak, on top of a moderate hill, providing a clear view over the Mamak plain. This location
had a low aggregated LAMAP value of just 9/35, though it was estimated to have high potential for certain periods:
4/5 for Achaemenid-Hellenistic and 3/5 for Late Ottoman. Despite having a low predicted potential for Late
Prehistory of 1/5, a large concentration of prehistoric pottery was encountered there. Due to the slope and the dense
vegetation, the full extent of the site has been hard to establish, but it is estimated to be roughly 100 x 70 m in size.
Most of the collected sherds were merely body fragments, but they shared strong characteristics with the prehistoric
materials found over the previous two years in the Derekoy-Hisar survey (Vandam et al. 2017, 2019c).
The second site documented in the Mamak area, CANl 7, was encountered on a terraced hillslope with a fairly
high aggregated LAMAP value of 25/35, 1.5 km southeast ofMamak itself. Upon investigation, it was revealed to be
a Late Ottoman pastoral site. This result is interesting in consideration of the LAMAP predictions per period, which
although quite high for Late Prehistory at 515, and 4/5 for three other periods, was only 2/5 for the Late Ottoman
period. At the site, the remains of several structures were located in an open area with a moderate slope. Among the
structures documented were an animal pen, a large "house" construction, and rubble walls. A limited amount of
pottery was also collected, including water jug fragments and building materials such as tiles, as well as pieces of
production waste on the site and in the immediate surroundings. All the materials at the site could be dated to the Late
Ottoman period.
Gravgaz
The next area of focus for the survey was Gravgaz, near the modem village of Tekkekoy. Here we investigated 13
fields, located in areas that were predicted by our modelling to comprise both high and low archaeological potentials.
The Gravgaz area has been the subject of both extensive palynological and sedimentological inquiry (e.g., Bakker et
al. 2012) but has never been subjected to an intensive archaeological survey. Through the course of the survey
activities in this campaign, we documented one new site at a location of moderately high aggregated LAMAP
potential of23/35. At KAA09, a new find cluster was identified, located in the foothills along a small temporary water
stream. Currently the site is mainly overgrown by shrubs, which greatly hindered the survey. The cluster was rather
small in extent and can be estimated to be ca. 70 x 80 m. The identified materials suggest that they date to the
Byzantine period, and preliminarily the site can be interpreted as a farm. These results coincide with per period
LAMAP predictions of 4/5 and 3/5 for the Late Antique and Byzantine periods respectively.
<;-endik-Suludere
The Biigdiiz River has been the subject of extensive sedimentological investigations by the Sagalassos project
(Verstraeten et al. 2019). In the <;endik-Suludere area, the river forms an alluvial fan and terminates into Lake Burdur.
The area itself had not previously been the subject of intensive archaeological research, though the area westwards
around Hacilar-Diiger was investigated intensively from 2010-2012 (e.g,. Vandam et al. 2019a). A new survey,
therefore, presented an interesting opportunity to compare the archaeology between both areas. During the 2018
campaign, 13 fields were surveyed in the <;endik-Suludere area. These fields investigated in the vicinity of Lake
Burdur had been predicted to have relatively low archaeological potential and yielded a very limited number of finds,
likely the result of fluctuating lake levels throughout the past. One new site, noted as SUL05, was discovered during
the survey however, at a location with a 19/35 aggregated LAMAP value, close to Suludere <;ay1. Here, we discovered
a small concentration of mainly Byzantine pottery (including the typical pattern burnished sherds) and tiles
characteristic of a farmstead. Late Ottoman materials were also collected at the site. These results fit fairly well with
the per period LAMAP predictions of 3/5 for both the Byzantine and Late Ottoman periods. Currently the location is
overgrown and not recently ploughed, which hindered our field walking activities. The find cluster seems to be rather
small, approximately 70 x 50 m. The site may have perhaps once been bigger, but the many levelling and ground
works in the surroundings are likely now obscuring the site's true extent.
Baymdzr-Bugduz
The 2018 survey campaign also focused on parts of the Biigdiiz River catchment further upstream, near the
Baymdu and Biigdiiz areas. The location itself had never previously been the subject of an intensive archaeological
survey. A total of 16 fields were surveyed in areas of both high and low archaeological potential. In the high potential
areas of the lowlands, two new sites were identified alongside the rivers.
At the site of YAK03, close to the University of Burdur, we found a multi-period artefact scatter with material
from the Roman Imperial up through the Late Ottoman periods. The most represented period, however, was the
Byzantine Dark Ages (?-9th century CE), as evidenced by typical pattern-burnished sherds. These results fit well with
the LAMAP predictions, which although only moderately high in aggregate at 26/35, were predominately higher for
the individual periods encountered: 3/5 for the Roman, 5/5 for the Late Antique, and 4/5 each for the Byzantine and
Late Ottoman periods. Notable was the high number of tiles at the site, which were extremely fragmented due to
intensive agricultural activity, as the area is currently fully cultivated. The size of the site was rather extensive,
measuring approximately 300 x 200 m, and can possibly be interpreted as a small village settlement.
In Biigdiiz, a small concentration of archaeological materials was also found at site BUGOS, which was located on
an old river bank about 150 m from the present day Biigdiiz River and 1.5 km northeast of the modem village of
Biigdiiz. The find scatter was comprised of materials from different periods, but mainly Achaemenid-Hellenistic
materials were identified. In addition, a typical Iron Age painted sherd and a burnished Early Bronze Age sherd were
collected. The extent of the settlement was small, at about 100 x 80 m, and can preliminarily be interpreted as a farm,
but the intensive agricultural activities in the area may have obscured this. These results coincided quite well with the
predicted LAMAP values of 28/35 overall, 4/5 for Late Prehistory, 5/5 for the rarely encountered Iron Age, and 4/5
for the Achaemenid-Hellenistic period.
Bereket-Kokez
The Bereket-Kokez area has been the subject of both environmental and archaeological research (Kaptijn et al
2013). Most of these endeavors were concentrated in the area south of the modem village of Bereket. In 2018, we
wanted to test the predictive models in the areas farther north of these research areas. In total, 16 fields across a variety
of landscape types comprising the valley, foothills, and highlands were surveyed. These efforts revealed a new artefact
scatter in the foothills at Kokez.
Here at the site of K0K02, about 1.5 km north of the modem village of Kokez, a large concentration of
archaeological materials was found. It is located in the foothills, near the road to Bereket, and east of the Biigdiiz
River, with a high LAMAP aggregate value of 32/35. The scatter was rather large and spread out over several fields,
measuring approximately 200 x 100 m in extent. While no structures were identified at this site, the pottery collected
there ranged from Roman Imperial up to the Late Ottoman period. Most of the materials were red-slipped wares from
Sagalassos, but Bagsaray red-slipped wares were also found. A wide range of shapes were also represented, including
amphorae, bowls, and cup fragments, leading to the interpretation of the scatter as a small village. Immediately north
of and adjacent to this location, a concentration of Late Chalcolithic pottery was also documented. The materials were
related to those found at the Late Chalcolithic sites in the Burdur Plain (Vandam et al. 2019b). Besides per period
LAMAP values of 3/5 for Late Prehistory and 4/5 for the Iron Age, all other periods were predicted to have the
highest archaeological potential of 5/5, which was representative of the range and quantity of materials that were
recovered from the site.
Eastern Aglasun Valley
The Aglasun Valley is of particular importance to the project as it comprises the immediate surroundings of the
Sagalassos settlement itself. The valley has been the focus of several archaeological and environmental research
programs (see for instance Vermoere 2004; Vandam et al. 2017, 2019c). During the 2018 campaign, the survey team
worked in the eastern Aglasun Valley around the villages of Aglasun, Yazir, Derekoy, Camhdere, and A~ag1yumruta~
over the course of two days. In total, 29 fields were surveyed which had been selected from areas where no previous
intensive archaeological research had taken place, while also attempting to representatively sample the various
constituent landscape types. A reoccurring pattern noted by this survey was that the Derekoy area was particularly
densely occupied during the Late Antique-Byzantine period. Several new find scatters were identified along the
Aglasun River: DER20, DER21, and CAM02. Furthermore, new prehistoric sites were also identified at CAMOl and
CAM03, confirming the high level of human activity during the Palaeolithic-Early Holocene documented during
previous surveys (Vandam et al. 2017, 2019c).
The area is characterized by its narrow valley which the Aglasun River has significantly transected. Two small
concentrations of finds were identified by our team at DER20, next to the river near the modem village of Derekoy. It
is an area where only moderate archaeological potential was predicted, with an aggregated LAMAP value of 19/35
and little easily accessible land. This may explain the close proximity of the sites, as they were located just a few
hundred metres from one another. The materials collected at the sites can be predominantly attributed to the Byzantine
period (2/5 LAMAP value), but Late Antique (3/5) and Late Ottoman (3/5) materials have also been attested there. At
DER21, a few Hellenistic sherds were also observed, despite relatively lower LAMAP predictions of 16/35 in
aggregate and 2/5 for the Achaemenid-Hellenistic period specifically. Noteworthy were the many water-rolled sherds,
which can be explained by the proximity of the Aglasun River. These sites can most likely be interpreted as farms,
displaying a wide range of finds, from building materials to bowls and dolia fragments.
At Camhdere, in the hills towards Yumrut~, we discovered a small Late Antique-Byzantine village site at
CAM02. The site size can be estimated to be around 300 x 250 m. We believe that the road to Yumrut~ has probably
cut into the site, as on both sides of the road archaeological materials have been observed. At the site we have
identified 13 in situ structure (possibly house) remains and limestone rubble walls, alongside building material
concentrations (Fig. 24-4). Some of the structures were most likely re-used later, as many Late Ottoman materials
were also found within and around them. Some showed traces of illegal excavation activities. Most of the collected
materials at CAM02 could be dated to the Late Antique period, but Byzantine and Ottoman materials were also
recovered. In addition, a bronze cross and the lead base of a small vessel were also found. In the surrounding fields, a
few carved blocks and a possible door opening were documented. Overall, these results were not well predicted by the
LAMAP models, as the aggregated potential at CAM02 was 14/35, and although high potentials of 5/5 and 4/5 were
predicted for the Iron Age and Roman periods respectively, the Late Antique and Byzantine periods were both
assigned 0/5 values and the Late Ottoman was 2/5.
Figure 24-4. One of the thirteen house remains that were identified at CAM02.
Near the northeastern border of the village of Camhdere, also close to the Aglasun River, a small lithic artefact
concentration was found at CAM0l. The concentration consisted of both tools, such as borers, and production
materials, including chips, flakes, and flake cores. The precise nature of the site is unclear, but the wide range of
artefacts might support the interpretation of a camp site. Based on the stylistic traits of the lithics, we estimate that the
concentration belongs to the Palaeolithic. The location had the lowest aggregated LAMAP value of all the sites
encountered in this campaign at 8/35, but it should be noted that the modelling did not include any Palaeolithic sites in
the training dataset, because too few sites of the period are known from the study area.
Lastly, in the area of Camhdere, we recorded a rock shelter on a high-altitude plateau. Although it was located in
an area that was not selected for testing of the predictive models, we investigated it due to its apparent potential as a
promising location for human activity. The rock shelter continues to be used currently by shepherds, but we were
nevertheless able to discover archaeological materials within and outside of the threshold. Most of the collected
materials were Late Ottoman in date and probably also relate to pastoral activities. In addition, prehistoric lithic
artefacts were also recovered, as well as Roman Imperial cookware fragments. Two wells, likely dating to the Late
Ottoman period, were documented 200 m east of the rock shelter, and are possibly related to the site. Later
comparison with the LAMAP predictive surfaces showed a mixed degree of cohesion, with the moderately high
aggregated value of 22/35 and 4/5 for Late Prehistory but just 2/5 for both Roman and Late Ottoman periods, though
the physical nature of a rock shelter makes it a feature that requires a specific training dataset which was not
considered in the creation of these models.
THE 2019 CAMPAIGN
Pedestrian Re-Documentation Survey
In the following 2019 campaign, sites previously known in the Aglasun Valley that had been documented through
earlier extensive, non-systematic surveys were targeted for intensive surface prospection. These efforts followed the
standard Sagalassos survey methodology, with tract-walkers following the modem natural or man-made contours of
fields surveyed. Survey tracts were 1 m, spaced 20 m apart and segmented into 50 m long "plots," with finds from
each plot bagged together while employing a total collection strategy (see Vandam et al. 2017, 2019c). The total
number of 50 m 2 plots walked per site was determined by artefact fall-off patterns, and aimed to ensure that the entire
concentration of surface material was documented. Adhering to the standard approach allows these results to be
incorporated into and compared with the results of other previous and recently conducted surveys.
In total, six sites, which had been documented in earlier non-intensive efforts, were selected within the wider
Aglasun Valley to be revisited. The sites were selected because their documentation was incomplete, meaning that
they were missing a proper interpretation of their dating, dimensions, exact location, or nature. By revisiting and
reinvestigating the sites and the area around them, we hoped to resolve these questions. The priority of the survey was
on the sites located directly within the Aglasun Valley and on the sites about which we possessed the least
information.
Results
In total, the team was able to investigate 117 fields within the Aglasun Valley (Fig. 24-5). This comprised an area
of 0.53 km 2 , in which we discovered archaeological materials from various time periods and of various natures. A
short overview of the updated interpretations of the resurveyed sites is presented in Table 24-2.
41L,50
Baskoy
1,il.117
..
61
Aglasun
47""49
10Pi:13 34'3
4'1fio Derekoy •
Camhdere
•
Legend
Sagalassos - Fields 201 9 intensive survey
• Village l!!!I Fields 201 9 extensive survey
1.7 3.4km
A
Figure 24-5: Overview of all the investigated fields in 2019.
Site code Village Dating Interpretation
AGL02 AGLASUN Destroyed

(Roman Imperial - Late Antique -
(small) Village
AGL08 AGLASUN Byzantine) Late Ottoman
(Roman Imperial - Late Antique)
Church with associated village
BAS03 BASKOY Byzantine - Ottoman
BAS05 BASKOY (Roman) - Byzantine - Late Ottoman Burial ground
Roman Imperial - Late Antique -
(small) Village
KOY0l K6Y6NU Byzantine - Late Ottoman
YAZ02 YAZIRKOY Late Antique - Byzantine Village
Table 24-2. The revisited archaeological sites in 2019 with their updated information.
CONCLUSION
The long history of survey research by the Sagalassos Project has created a unique regional archaeological
database upon which various future research efforts such as predictive modelling, as we have now initiated, can be
informed. On the other hand, the nature of legacy data presents certain methodological challenges. The datasets
derived from individual survey projects differ in resolution, accuracy, and sampling intensity, making it difficult to
synthesize the information of these various datasets in an efficient way. To facilitate this, the data should be
standardized on fundamental grounds. The recent results of the revisiting and re-documentation activities around
previously documented sites have further demonstrated the need to work out the biases within our dataset in order to
improve the results of future research making use of this legacy information, including synthetic and modelling
approaches.
ACKNOWLEDGMENTS
This research was supported by the Research Fund of the KU Leuven and the Research Foundation Flanders. This
work would not have been possible without the help of all the participants of the 2018 and 2019 Sagalassos survey.
We are grateful to the Ministry of Culture and Tourism of the Republic of Turkey, and their representatives: $enay
Ocal and Hakan Dins;, for the survey permission and aid during the fieldwork campaigns.
REFERENCES CITED
Attema, Peter, John Bintliff, Martijn van Leusen, Philip Bes, Tymon de Haas, Damjan Donev, Wim Jongman, Eva
Kaptijn, Victorino Mayoral, Simonetta Menchelli, Marinella Pasquinucci, Steve Rosen, Jesus Garcia Sanchez,
Luis G. Soler, D. Stone, Gijs Toi, Frank Vermeulen, and Athanasios Vionis. 2020. A Guide to Good Practice in
Mediterranean Surface Survey Projects. Journal of Greek Archaeology 5: 1- 62.
Bakker, Johan. 2019. Sagalassos Under the Weather. Exploring the Impact of Climate Change on Landscape and Land
Use. In Meanwhile in the Mountains: Sagalassos, J. Poblome, E. Torun, P. Talloen, and M. Waelkens, eds., 293-
301. istanbul: Yap1 Kredi Yaymlar1.
Bakker, Johan, Etienne Paulissen, David Kaniewski, Veronique De Laet, Gert Verstraeten, and Marc Waelkens. 2012.
Man, Vegetation and Climate During the Holocene in the Territory of Sagalassos, Western Taurus Mountains, SW
Turkey. Vegetation History and Archaeobotany 21(4-5): 249-266.
Carleton, W. Christopher, James Conolly, and Gyles Ianonne. 2012. A Locally-adaptive Model of Archaeological
Potential (LAMAP). Journal of Archaeological Science 39(11): 3371- 3385.
Carleton, W. Christopher, Kong F. Cheong, Dan Savage, Jack Barry, James Conolly, and Gyles Iannone. 2017.
Reports A Comprehensive Test of the Locally-Adaptive Model of Archaeological Potential (LAMAP). Journal of
Archaeological Science: Reports 11: 59-68.
Ginau, Andreas, Robert Schiestl, and Jurgen Wunderlich. 2019. Integrative Geoarchaeological Research on Settlement
Patterns in the Dynamic Landscape of the Northwestern Nile Delta. Quaternary International 511: 51- 67.
Kaptijn, Eva, Jeroen Poblome, Hannelore Vanhaverbeke, Johan Bakker, and Marc Waelkens. 2013. Societal Changes
in the Hellenistic, Roman and Early Byzantine Periods. Results from the Sagalassos Territorial Archaeological
Survey 2008 (Southwest Turkey). Anatolian Studies 63: 75-95.
Poblome, Jeroen, Patrick Degryse, Willy Viaene, Raoul Ottenburgs, Marc Waelkens, Roland Degeest, and Jean Naud.
2002. The Concept of a Pottery Production Centre. An Archaeometrical Contribution from Ancient Sagalassos.
Journal ofArchaeological Science 29(8): 873-882.
Schmidt, Isabell, Marcel Bradtmoller, Martin Kehl, Andreas Pastoors, Yvonne Tafelmaier, Bernhard Weninger, and
Gerd-Christian Weniger. 2012. Rapid Climate Change and Variability of Settlement Patterns in Iberia During the
Late Pleistocene. Quaternary International 274: 179- 204.
Toth, Peter, Peter Demjan, and Kristina Griacova. 2011. Adaptation of Settlement Strategies to Environmental
Conditions in Southern Slovakia in the Neolithic and Eneolithic. Documenta Praehistorica 38: 307-322.
van Leusen, Martijn, Jos Deeben, Daan Hallewas, Paul Zoetbrood, Hans Kamermans, and Philip Verhagen. 2005. A
Baseline for Predictive Modelling in the Netherlands. In Predictive Modelling f or Archaeological Heritage
Management: A R esearch Agenda, P.M. Van Leusen and H. Kamennans, eds., 25- 93. Amersfoort: National
Service for Archaeological Heritage.
Vandam, Ralf, Patrick T. Willett, and Jeroen Poblome. 2017. Living on the Margins: First Results from the Derekoy
Archaeological Survey of the Sagalassos Project in the Western Taurus Mountains. In Archaeology of Anatolia
Volume 11, S.R. Steadman and G. McMahon, eds., 321-46. Newcastle upon Tyne: Cambridge Scholars Publishing.
Vandam, Ralf, Branko Music, and Igor Medaric. 2019a. Contextualizing Kurus;ay Hoyillc Assessing the Unexplored
Late Chalcolithic Landscape Near the Beginning of Early Social Complexity in SW Turkey. Journal of Field
Archaeology 44(1): 1-19.
Vandam, Ralf, Eva Kaptijn, and Jeroen Poblome. 2019b. Reflecteren op 25 jaar archeologische prospectieonderzoek
in het studiegebied van Sagalassos (Zuidwest-Turkije). Tijdschrift voor Mediterrane Archeologie 60: 115-123.
Vandam, Ralf, Patrick T. Willett, and Jeroen Poblome. 2019c. The Results of the 2017 Derekoy Archaeological
Survey by the Sagalassos Project in the Western Taurus Mountains. In Archaeology of Anatolia Volume III, S.R.
Vanhaverbeke, Hannelore and Marc Waelkens, eds. 2003. The Chora ofSagalassos. The Evolution of the Settlement
Pattern from Prehistoric until Recent Times. Studies in Eastern Mediterranean Archaeology V. Turnhout: Brepols.
Vermoere, Marleen, ed. 2004. Holocene Vegetation History in the Territory of Sagalassos (Southwest Turkey): A
Palynological Approach. Studies in Eastern Mediterranean Archaeology VI. Turnhout: Brepols.
Verstraeten, Gert, Maarten van Loo, Nils Broothaerts, Bert Dusar, and Koen D'Haen. 2019. Sagalassos Under the
Weather. Exploring the impact of Climate Change on Landscape and Land Use. In Meanwhile in the Mountains:
Sagalassos, J. Poblome, E. Torun, P. Talloen, and M. Waelkens, eds., 301-313. istanbul: Yap1 Kredi Yaymlar1.
Waelkens, Marc, Eva Kaptijn, Branko Music, Jeroen Poblome, and Julian Richard. 2014. The 2013 Archaeological
and Geophysical Surveys in the Territory of Sagalassos. Ara~tzrma Sonur;:larz Toplantzlarz 3 2(2): 131-150.
Wansleeben, Milco and Leo Verhart. 1997. Geographical Infmmation Systems: Methodological Progress and
Theoretical Decline? Archaeological Dialogues 4(1): 53-64.
Wheatley, David and Mark Gillings, eds. 2002. Spatial Technology and Archaeology: The Archaeological Applications
of GIS. London: Taylor and Francis.
Willett, Patrick T., W. Christopher Carleton, Ebru Torun, Ralf Vandam, and Jeroen Poblome. 2021. Modeling
Archaeological Potentials in SW Anatolia: A Tool for Planning Sustainable Futures at Ancient Sagalassos. In
Critical Archaeology in the Digital Age, K. Garstki, ed., 129-149. Los Angeles, CA: UCLA Cotsen Institute of
Archaeology Press.
CHAPTER TWENTY -FIVE
THE RESULTS OF THE 2019 GARGARA ARCHAEOLOGICAL SURVEY

IN THE SOUTH TROAD
NURETTiN ARSLAN AND CANER BAKAN
A new project was begun in 2018 with the aim of investigating the rural area around Assos and the ancient cities in
its vicinity. The boundaries of this project, which will last for five years, are limited to the cities of Lamponia and
Gargara to the east of Assos, and Hamaxitos to the west. The objectives of the research are fourfold; firstly, recording
the rural (Chora) area of Assos and remains in neighboring poleis, examining the settlements to find archaeological
evidence, drawing up city plans, and lastly, determining cultural relations between the poleis.
In the first season of the project, a study was made of Gargara, which is located 19 km east of Assos. The detected
archaeological remains were digitally transferred to a map with the help of tools such as GNSS (Global Navigation
Satellite System) and a drone (UAV).
RESEARCH HISTORY
In the southern Troad there are two cities bearing the same name, one on Kocadag Mountain (Old
Gargara/Plaigargara) and the other on the coast to the south of Ankh village (new Gargara/Gargara) (Fig. 25-1). This
TENEDOS
~
SKEPSIS
.
Kolonai
ANTANDROS
A 0 10 20 30 40km
---====----==== (I
Figure 25-1. Map of the Troad.

Results of the 2019 Gargara Archaeological Survey 323
I -
Residential
area
Residential
B {RA3)
area .
RA2) .
.. P3
Kiln .
. ·.-:·_.- .--~. . .
p ·~-==---
area
.· .. .
Residential . · . ··: · ..
·. · ,, . ·.. I
/ p3
I
,..,.,_
I '1
\ • .,J
...
0 1+00 200m
' . . · ·, .·
• House
Figure 25-2. Plan of ancient Gargara.
article discusses Old Gargara on Kocadag. The mountain was visited by several scholars in the 19th century. It was
deemed to be insignificant compared to other cities in the region, and therefore little attention was paid to it. The
geologist J.S. Diller, who was working on the Assos excavations, searched the region and discovered the remains of
two cities (Lamponia and Gargara). An ai1icle about these two cities was written by J.T. Clarke, head of the Assos
excavations, but he erroneously identified the remains on Kozlu Mountain as Gargara and the remains on Kocadag
mountain as Lamponia (Clarke 1888: 298-300, figs. 10-11; Judeich 1901: 124). Moreover, it was stated by Clai·ke
that apart from the polygonal walls on Kocadag, there were no archaeological remains that could be dated after the 4th
century BCE (Clarke 1888: 315-316).
E. Fabricius was another scholar who studied Gargara, in 1888. His notes were published by W. Judeich, who
came to the region in 1896 but did not visit Kocadag. Fabricius gave the following information about the city:
There are ancient ruins on Odjak Kaya/ Kocadag (780 m), which is 7 km from the shore. There ai·e extensive
pieces of pottery and pithos on the slopes of the mountain. The ancient road to the city must have passed
thrnugh the high cliffs. On the slopes of the mountain, there must have been houses built on the flat areas
behind the polygonal terrace walls, made of numerous cyclopic stones. The hill descends slightly to the north
and forms an elliptical plateau. The east and north sides of the hill are very steep. The 500-meter-long walls
that encircle the plateau at the edge of this hill are 1 meter high. The rough stones used in the walls, which have
an average width of3 meters, were placed very orderly. There is a 2.35-meter-wide gate on the edge of the wall
to the west and a rectangular tower on the right. The citadel is divided into two unequal parts by a wall. In the
smaller area at the highest part of the mountain to the south are the foundations of a large building that might
be a temple. All of the ruins must belong to an early period. No Hellenistic, Roman or Christian remains were
observed (Judeich 1901: 121).
Judeich, who accepted the remains on Kocadag as Gargara, did not find any remnants that could claim to be New
Gargara(Judeich 1898: 541-542).
W. Leaf conducted a five-week expedition in the Troad in 1911 in order to investigate the cities mentioned by
Strabo. Leaf did not see Kocadag, and he used the information that Clarke had given about Old Gargara (Leaf 1923:
258-263). J. Cook, who investigated Kocakaya on 1 September 1959, accepted that the remains he saw were Old
Gargara. Cook observed that the high hill (A; see Figs. 25-2 and 25-3) to the east of the mountain, and the sloping area
to the north of it, were surrounded by two different walls. He mentions many dry-stone walls in the form of terraces
on the lower slopes. Cook drew a plan of the citadel, which he found interesting, and accurately described the tower
gates to the west and east of the citadel, as well as steps that reached up to the fortified walls. Based on the pieces of
pottery he found, he proposed that Gargara was abandoned in the second half of the 4th century BCE (Cook 1973:
255-261, fig. 11).
324 Chapter Twenty-Five: Arslan and Bakan
Figure 25-3. View ofKocakaya (Old Gargara), looking from west to east.
R. Stupperich, who was a member of the more recent Assos excavation team, visited Gargara in 1989 and 1991.
During these visits, Stupperich discovered some Aeolic capitals and a relief block with a warrior depicted on it; he
brought these items back to the excavation house. In his article on the warrior relief, Stupperich describes the area
where the Aeolic capitals were found as the sanctuary, but he did not sketch the new remains on the plan originally
drawn by Cook (Stupperich 1995: 127). In 1991, A. Akalm followed the route of Cook and visited Kocakaya and
Zindankale. Akalm, who found two architectural terra cotta fragments belonging to the Archaic period, and gave
incorrect dimensions of the walls (erroneously stated to be 3 m wide and 5 m high), could not locate New Gargara
(Akalm 2007). One author of this chapter, who is the current head of the Assos excavations, also discovered many
architectural terracotta fragments during his visits to Gargara (Arslan 2020: 129-136).
THE 2019 SURFACE SURVEY
The remains of Old Gargara (Fig. 25-2) are located 19 km east of Assos, on Kocakaya hill (741.72 m high), 6 km
from the sea. Kocadag Mountain consists of two elevations, one in the east and one in the west, with a deep valley
running between these two peaks. The Satneios River flows 4 km north of the mountain. Ancient ruins are spread over
the eastern (A) and western (B) hills and the valleys between them, as well as the flat areas to the north and east of the
mountain. (Fig. 25-3). Gargara's northern, southern, and western sides are quite steep. The topographical structure of
the mountain was obviously taken into account in the planning of the defense walls. Only the ancient ruins in area A
are well-preserved and visible. Remains in other parts of the site consist of small fragments of the city and terrace
walls with piles of stones belonging to houses.
The Hill A Area
This area (Figs. 25-2 and 25-4) consists of a small hill (A) to the south, oval in shape, with a slight inclination
towards the north. The hill is surrounded by a polygonal stone wall, and there are remains of a building on its summit.
The limestone blocks of the walls of this building can be observed around the illicit excavation pit dug by treasure
hunters. Fabricius saw this ruin during his visit in 1888 and suggested that it might be a temple (Judeich 1901: 121).
The building was probably less decayed in those years. Since the temple is to the north, it is thought that the building
in question is not a temple. Traces of two more structures (5.5 x 3.8 and 6.2 x 4.4 m) were discovered on the southern
slope of the hill. A flat area was created by cutting into the bedrock for these two structures. Their remains have not
been preserved except for a few foundation stones; therefore, it is not possible to identify these remains
archaeologically. No materials related to daily life were found on the top of the hill.
Figure 25-4. Kocadag mountain, Hill A Area.
The area to the north of the small hill (approximately 3050 m2) was surrounded by a fortification wall (Pl, see Fig.
25-2), forming a citadel with a D-shaped plan. The walls of this citadel have polygonal masonry, similar to the other
walls in the ancient city (Fig. 25-5). Although the outer surface of the stone blocks was left rough, without dressing,
they were crafted more carefully than the other walls (Schulz 2000: 17). The walls of this citadel, which is the best-
preserved building at Gargara, are 2 m wide. The inner parts of the emplekton walls were filled with small stone
fragments.
Figure 25-5. Walls of the citadel.
The walls of the D-shaped citadel are supported by five internal props. The struts are placed 35 m apart from each
other and are approximately 7 m long and 1.5 m wide. On the edge of the northern buttress (Fig. 25-4), there is a
staircase that leads to the wall. In addition to the two monumental gates on the west and east sides of the citadel (see
Fig. 25-2), a third small gate also exists on the west side (Fig. 25-6). To the east, the width of the gate (G 1), with a
rectangular (6.5 x 6.24 m) tower on both sides, is 4.2 m. The towers adjacent to the narrower (2.35 m) western gate
(G2) are different from each other. No tower structure was observed near the third gate (G3), which is the smallest
(1. 7 m) in the west.
Only a few wall fragments have survived from the building remains inside the citadel. Unfortunately, these
preserved pieces are not sufficient to allow us to comment on the plan and function of the structures to which they
once belonged. However, the neatly cut stone blocks unearthed in the illicit excavation pits inside the citadel indicate
the existence of other buildings here. The well-cut limestone blocks found scattered around the vicinity remind us of
the presence of important buildings in the citadel; however, excavations are required in order to construct plans of
these buildings.
The warrior relief in the citadel, discovered by Stupperich, is quite striking (Fig. 25-7). The blocks on both sides of
this human-sized relief (1.4 m in height) are missing. A hoplite, known from Greek vases, is depicted on the
rectangular block. He was probably carrying a shield in one hand and a spear in the other (Stupperich 1995: 131, fig.
3). However, the helmet type with palmette decoration on the head of the figure does not exist in Greek art
(Stupperich 1995, 133). The back and side faces of the relief were left rough; this can be explained by the frieze being
placed in a wall. It is thought that there were similar reliefs on both sides of the existing relief (Stupperich 1995: 135).
It is inferred that the warrior relief, which is dated to the middle of the 6th century BCE, adorned a building inside the
citadel or one of the gates of the citadel (Stupperich 1995: 135-137).
Figure 25-6. Aerial view of the citadel.
'
'
l
I ,..- - - - - - - .-
,,
- _,, I
I\
1,
,,
,,
-=. . ~-:·-c:::_-
;
':." ------
,.. -
- 11
l1
,,
,;>::-,--- . . . . . . . .
'I
I
....
'-
'
' '
'
',_ ----------------J':,
:: ~ • ,' :,, I ' \.
'~-:. _____ ;: / ', \

- ,7 :
,, '
'' ''
,-
'
'r -
""
Figure 25-7. Warrior relief.

The Sanctuary
For the construction of a building 30 m north of the citadel (see Figs. 25-2 and 25-4), the bedrock was roughly cut
into a "U" shape, and a wide wall was built around it. In this area surrounded by a wall, rich finds consisting of
numerous architectural stone blocks, roof tiles, Aeolic capitals, and architectural terra cottas were found. The well-
preserved wall west of the area is 31 m long and has a polygonal pattern. It is understood that this space was planned
as a sanctuary, surrounded by a wall (temenos), and that a temple was built inside it.
-- - --
0
Figure 25-8. Fragments of Aeolic capitals.

100cm
In the sanctuary were discovered a few more examples similar to the Aeolic limestone capitals found by
Stupperich in previous years and brought to the Assos excavation house. Unlike the examples in Neandria and Lesbos,
these column heads are undecorated. The volutes are also formed quite simply, compared to other sites. There are
dowel holes under the heads carved from soft limestone. These column heads differ in size and shape (Fig. 25-8).
Despite the Aeolic capitals having been found in the sanctuary, evidence for the use of stone columns is lacking.
Therefore, it is suggested that columns made of wood were used in the temple. From Caria to the coast of Thrace,
Aeolic capitals were used over a very wide area (Mallwitz and Schiering 1968: 135-142; Radt 1970: 238-242;
Schefold 1939: 42-52; Akurgal 1983: 88-94; Cook and Nichols 1998: 93-96, 200-203; Betancourt 1977: 73-88;
Clarke 1886: 136-148; Wiegartz 1994: 125; Rose 1995: 86-88, 2008: 416; Ba~aran 2000: 157-170). The Aeolic
capitals in Gargara were dated to the last quarter of the 6th century BCE by Wiegartz due to their similarity to
specimens in Mytilene, Klopedi, and on the islands north of the Aegean Sea (Wiegartz 1994: 125). However,
Stupperich considers that these heads are earlier. (Stupperich 1995: 133).
Figure 25-9. Architectural terra cottas (sima style).
The painted roof tiles spread over a wide area in the sanctuary are of Lakonian type. The rich samples of
architectural terracotta consist of sima, geison, antefix, and cladding plates (Figs. 25-9, and 25-10). Decorating the
multi-colored architectural terra cottas are geometrical designs and plants, as well as animals (lions and bull) and
human depictions. Medusa and olive trees are observed on the antefix pieces (Kjellberg 1940: 57-59, 154, pls. 18, 9;
Winter 1993: pls. 106 and 110; Akerstrom 1966: 26, pl. 12, 1). Similar to the Aeolic capitals, the pieces decorated
with Ionic kymation also do not show an integrated structure. This can be attributed to the lack of standard
workmanship during the construction of this temple, or to the fact that it was repaired very soon after construction.
The sima pieces should be dated to 560-525 BCE (Kjellberg 1940: 64, 66, 80-81, 88, 154, pls. 22, 27, 33.19 40.33,
50; Akerstrom 1966: 95, pl. 46.1-2; Boehlau and Schefold 1940: 164, pls. 22, 27, 28, 37; Hood et al. 1954: 144, pl.
8.105; Cook 1973: 335, pl. 61.8; Ramage 1978: 24-25, fig. 71.35, 73.37). The resemblance of the lion and bull figures
to depictions on the friezes of the Assos Athena temple (540 BCE) is quite striking (Wescoat 2012: 138-150, pls. 80-
87).
0 10cm
0 10cm
0 10cm
Figure 25-10. Architectural terra cottas (geison tiles).

The terra cottas found at Gargara attest that the structure in the sanctuary, or perhaps more likely in the temple, had
rich architectural decoration. The use of this type of architectural terra cotta in the Archaic Age is quite common in the
cities of the Troad (Akerstrom 1966: 5-21, pls. 3-8; Cook 1973: 334-35, pl. 61, pl. 62, 4, 7, 15, 10, pl. 63, 16; Arslan
2009: 80, fig. 5, 2017: 140, fig. 21.2).
It is not easy to make suggestions regarding the plan of the temple without an archaeological excavation of the
sanctuary. However, it is possible to suggest a plan that had a rectangular cella and a single row of wooden columns in
the middle, like the Athena temple in Neandria about 40 km north ofGargara (Koldewey 1891: 22, 24, figs. 52, 55).
The Hill B Area
The western and northern sides of the second hill (B), 200 m west of the sanctuary, are quite steep. It is believed
that the 62.6-m-long polygonal walls on the southern slope, facing the city, are terrace walls. On the flat area
extending east-west on the hill are the remains of a wall, of which only small sections have been preserved. There are
no roof tiles, architectural terra cotta fragments, or monumental remains in this area. The few walls on the hill should
be recognized as belonging to houses.
Residential Areas RAJ and RA2
The area where the houses are most concentrated in the city is in the southern part of the valley between hills A
and B. There are ruined houses close to each other in the more gently sloped areas on all sides of the valley. Due to the
fact that these regions are covered with trees, it was not possible to photograph these remains, especially from the air.
The coordinates of this debris were taken with GNSS/CORS, and the number and location of houses was also
recorded. In this way, it was possible to obtain a general idea about these residential structures. Apart from the
remains on the slopes in the area where the houses are located, long walls have also been identified. These are
regarded as terrace walls.
Figure 25-11. Ruins of a house.
The valley between the two hills would have been selected as a residential area because it is sheltered from the
strong northern winds and has a more moderate slope. The total number of house ruins detected in the valley, which
extends from north to south, is 236 (RAl ). Eight of the ruins in question are larger than the others. There are 36
residential structures and terrace walls in the RA2 area, with a slight slope towards the south in the west of the valley.
Small rubble stones were used as building material in most of the houses. Among the ruins of these houses, a few
pieces of limestone blocks attract attention. These were used as lintels on doors and windows. Only a few houses were
built entirely from properly-cut limestone blocks, as in the structures inside the citadel. As a result of the illegal
excavations carried out by treasure hunters in one of these houses, many blocks of stone belonging to walls were
unearthed (Fig. 25-11 ). However, unlike the sanctuary, no roof tiles or architectural terra cotta fragments were found
in the residential areas.
0 5cm
G' 19-22
G'19 L200
C 2019-021
G' 19 L640 G' 19L640
b
(
G'19 L 206 G'19 -20
5cm
>-
~ - - - - - - - - - <~OCM
C
Figure 25-12. a-b) Black-slipped ceramics of Athenian origin; c) Amphorae.
Grinding stones, pottery, and loom weights used in daily life were found in the residential area. The shiny and
black slipped examples among the ceramics must be of Athenian origin (Fig. 25-12a-b). Interestingly, forms such as
the kylix and kantharos are not encountered among the ceramic samples from the city. Black slipped pottery fragments
strongly support the city's last occupation date of 350-325 BCE (Sparkes and Talcott 1933: 295, 302, Kat Nr. 830,
832, 942, figs. 8-9). Most of the amphorae, which must date to the same period as the black slip vessels, are ofThasos
origin (Cook 1973: 259) (Fig. 25-12c). This also reveals the weak relationship of Old Gargara with more distant
poleis.
The Outer Fortifications (P2)
On hill A there is a second fortification wall extending from the east of the hill to north of hill B (Fig. 25-2). The
second wall (P2) forms a large forecourt (11,070 m 2) east of the fortress (Pl), and this area is entered through the 2.5
m wide gate to the north. The polygonal walls, with an average width of 1.30 m (Fig. 25-13), were made more roughly
than the citadel walls (Pl). No remains were discovered in the area inside the wall (i.e., the courtyard), where most of
the walls have been completely destroyed. A flattened space of 1.65 x 1 m was cut into the bedrock south of the
citadel gate (Gl). It is conjectured that this was probably created to make an area to place a monument. The pathway
to the east of the second wall extends to the gate. This must have been the road used for transportation in and out of
the city in ancient times.
Figure 25-13. Fortification walls (P2).
The Outer Fortifications (P3)
A third fortification was found on the flat area on the northern and eastern skirts of Kocadag. As seen on the plan
(Fig. 25-2), the aim of these walls was to protect the inhabitants of the outermost ring of the city, and especially in the
area weakest against attacks. The structure and thickness of the walls, which are now completely destroyed, are
unknown. However, it is possible to trace a line marked by the piles of stones. The deep riverbed east of the settlement
area protected by the wall also provides a natural defense against external attacks. Enclosed within the third wall (P3),
remnants of houses and bakeries were found, in addition to a possible sanctuary, although this has not been
determined for certain.
Residential Area RA3
Small stones in the residential area inside the third fortification and on the flat area at the foot of the mountain must
be the remains of houses. Although the area is much broader than other residential areas, there are fewer houses. The
remains of a large building on the terrace to the east of the area attracts attention. Only the northern walls of the
building are partially preserved, and its function is not fully known. Two architectural terra cotta fragments were
recovered from this area, but no roof tiles were found. It is thought that either the terra cotta in question belonged to
this large ruin of unknown function, or the terracotta was moved here from the temple.
The thick-edged pottery fragments, which are common in residential area RA3, must be pieces of storage vessels.
However, it is difficult to comment on their date. The type of pottery found among the houses located in the valley has
not been found in this area. Archaeological data suggest that residents of lower status lived in this area, which is more
vulnerable in the face of external threats.
Metalworking Furnaces
Three furnaces built close to each other were found in the area east of the outer fortification (P2). Around the
furnaces where sections of the walls are partially visible, it is possible to see dense clinker waste and stones in the wall
that have melted due to high temperature. Samples taken from the melted stone, clinker, and an iron deposit east of the
furnaces were analyzed. The expert who analyzed these samples stated that the metal processed here was iron (Tables
25-1-25-3). A stone mold and a large iron object near the furnaces are perhaps other evidence of blacksmith activity
here. The examination of the area 200 m east of Gargara resulted in the conclusion that iron ore was obtained from
this locality.
Minerals % Minerals %
SiO2 61 ,040 Minerals % SiO2 78,99
Al2O3 20,100 Fe2O3 70,73 Al2O3 9,770
K2O 8,580 SiO2 21 ,944 Fe2O3 7,610
Fe2O3 4,100 CaO 2,335 K2O 1,900
Na2O 2,210 Al2O3 2,066 MgO 0,362
CaO 2,020 MgO 1,436 TiO2 0,356
MgO 0,731 K2O3 0,621 Na2O 0,248
TiO2 0,666 Na2O 0,292 CaO 0,154
P2O5 0,228 TiO2 0,109 SO3 0,148
MnO 0,104 s 0,095 BaO 0,146
Table 25-1. Analysis Table 25-2. Table 25-3. Analysis of
of the ore. Analysis of the clinker. stones in the furnace wall.
GENERAL CHARACTERISTICS OF HOUSES
Archaeological excavations are required to determine the plan and general characteristics of the houses in Gargara.
However, it is possible to get a sense of the houses by considering the ruins in Gargara and rural house architecture in
the local region (Fig. 25-14). As stated above, stone blocks made from a few pieces oflimestone were found in each
pile of stones forming the ruins of the houses. No fragments of tiles were found in the residential areas to prove that
they were covered with a tile roof. Houses in Gargara are therefore thought to be similar to country houses used only
in a certain period of the year, in terms of their construction technique and plan features. These houses consist of a
single room with a rectangular plan. In houses, with a flat roof, mud was used as the binding material for the walls.
The houses were low, with an average height of 2.2 m, and had a door 1.6 m high and a small window (50 x 40 cm).
The houses were simply constructed, and a stone block or thick wood was used as a lintel above the doors and
windows.
Figure 25-14. A typical house in the rural surroundings ofGargara.
CONCLUSION
Ancient writers mention a sanctuary called Gargaron or Gargaros on the peaks of Mount Ida, as well as the city of
Gargara (Homer, Iliad VIII.48, XIV.292, 352; Ptolemy, Geographia V.2.5; Pliny V.122; Pomponius Mela, De
Chorographia 1.93; Stephanos Byzantios, Ethnika, s.v. fapyapa; Alkman, frg. 154; Strabo, Geographica XIII.1, 51,
56, 58; Vergil, Georgics 1.100; Pseudo-Plutarch, De jluviis XIII.3; Mitchell 2004: 1007). Nymphis reports that due to
the cold weather in Gargara, the inhabitants moved down to the Gargaros plain, and that the old city was named
Palaiagargaros (Nymphis, frg. 10). The common opinion of scholars is that the ancient city on Kocadag should be
recognized as Old Gargara and that the city was abandoned in the 4th century BCE. Available evidence proves that Old
Gargara was a polis; it was a member of the Delian League and also minted coins. Gargara paid almost 4500
drachmas of tax, as recorded in the Delian League's lists of 452/1 BCE, 448/7 BCE, 447/6 BCE, 443/2 BCE, 432/1
BCE, 439/8 BCE, 430/29 BCE and 428/7 BCE, and a large number of coins are known to have been struck by
Gargara in the 5th century BCE (Mitchell 2004: 1007-1008; SNG Miinchen 1991: Kat. Nr. 188-92, pl. 7) (Fig. 25-15).
Two significant inscriptions about New Gargara were found at Troy. In the first inscription, dating to 306 BCE,
Gargara is referred to as a member of a union (the Confederation of Athena Ilias), formed in Ilion. The second
inscription, dated 190 BCE, records that Malousios of Gargara was honored for his kindness to the Temple of Athena
and to the Confederation of Athena Ilias in Ilion (Frisch 1975: Nr.l; Meier 2012: 323-326). However, neither of these
inscriptions are related to the city of Gargara on Kocadag.
0 2cm
Figure 25-15. Classical period bronze coins ofGargara (Troia Museum, Canakkale).
After the research carried out at Old Gargara on Kocadag Mountain, important information was obtained about the
character of the Archaic and Classical Age poleis; in addition, a map of the city was drawn. Ceramic samples found in
the ancient city support the notion that Gargara was abandoned suddenly before the Hellenistic age (Cook 1973: 249,
256). That po leis sometimes moved to another location in the past has also been encountered with the cities of Perkote
and Skepsis in the Troad. However, some of the inhabitants of these two cities continued to live in the old city (Arslan
2009: 81). In this respect, however, these circumstances are supported by archaeological findings, but this was not the
case for Gargara.
The ancient city of Gargara was founded on the side of a mountain and its foothills, lacking flat areas but rendering
it quite sheltered from external attacks. The limestone that formed the mountain was used as the building material for
the construction of structures. The only structures now standing in the ancient city are the fortification walls of the
citadel. In Gargara, polygonal walls (Lesbos walls) were preferred, which were widely used in many cities in the
Archaic Age (Acheilara 2020). The cut stone blocks scattered around the sanctuary and residential area suggest that
some buildings have orthogonal masonry. The orthogonal walls must belong to the Late Classical Age, indicating the
last occupation phase of the city. As mentioned previously, systematic archaeological excavations are necessary to
identify and date the city walls and other remains more accurately. However, widescale illegal excavations have
caused the last standing remains of the ancient city to be rapidly destroyed.
Polygonal bonded defensive walls, Aeolic capitals in the sanctuary, architectural terra cotta, and the warrior relief
are undoubtedly archaeological evidence representing the Archaic Age of Gargara. The Classical age findings of
Gargara consist of ceramics, loom weights, and grinding stones. Athenian and black glazed pottery and Thasos
amphorae show that Gargara had relations with distant poleis. Since Old Gargara was located far from the coast, the
relations of Gargara with overseas poleis must have been established through the closest port city, which was Assos.
Although two pieces of Byzantine period ceramics belonging to the 12th century CE were found during the survey in
Kocadag, the evidence indicates that Old Gargara was abandoned after the last quarter of the 4th century BCE, as
stated by Cook and others. The Aeolic capitals in Old Gargara and the polygonal walls, also known as Lesbos walls,
are important evidence demonstrating that the city was an Aeolian settlement (Strabo 13.1.58).
Gargara is flanked by hills and small valleys and is estimated to have included a wide expanse of land.
Ruschenbusch calculated that a typical polis could have lands with an area of about 25-100 km 2 , and this figure was
also accepted by Hansen (Ruschenbusch 1985: 258; Hansen 2004: 71). Gargara should also be classified as a normal
polis. The city was poor in terms of farmland but had extensive pastures for animal husbandry and perhaps forests.
Due to the land ecology around Gargara, it is thought that the main source of income of the city was animal husbandry
(Tenger 1995: 150-151). Apart from animal husbandry, it is possible that the city acquired income from the timber in
the forests. Ancient sources from the Hellenistic Age and later, mentioning that Gargara had fertile soil, must be
referring to New Gargara, which had productive land running along the coast where small streams flowed.
The amount of tax/tribute paid by the Greek poleis to the Delian League was calculated according to their
population. It is also accepted that there is a relationship between the amount of tribute and the port revenues of the
polis as well as its arable agricultural land (Ruschenbusch 1983: 125). Ruschenbusch calculated the population of
Gargara, which paid 4,500 drachmas of tax, as 2400, and stated that 600 of them were citizens. Balcer calculated a
similar number of citizens at Gargara, 563 people (Ruschenbusch 1983: 142; Tenger 1995: 159; Balcer 1984: 436).
The citizens in the poleis are considered to constitute a quarter of the population (Ruschenbusch 1985: 258). Two
hundred ninety-four houses were discovered during the research carried out in Gargara, of which some were destroyed
or perhaps still under the soil. The possibility of small settlements in the rural area around the city's ancient
environment should not be overlooked as well. However, overall, the figures suggested by scholars for the population
of Gargara appear reasonable and acceptable.
According to Ruschenbusch, Greek poleis were not all large regional states, socially, economically and politically,
like Athens and Corinth. Some poleis were simple villages whose economic income was based on agriculture and
whose inhabitants were farmers (Ruschenbusch 1978: 7- 13). This theory of Ruschenbusch also applies to Old
Gargara. Coins, the Delian League's lists, the temple with Aeolic capitals, and the polygonal fortification walls are
sufficient features to describe Gargara as a Greek polis. Despite this, Old Gargara's weak commercial and cultural
relations with distant and close poleis reflect the character of a large village whose citizens were farmers. Not a single
inscription proving the use of writing was found in Old Gargara. There is also no evidence of the existence of an
agora, theater, bouleuterion, or similar public structures which are indispensable elements of a Greek polis.
Old Gargara is an example of a village-polis that was self-sufficient and whose residents were farmers. The
systematic surface survey conducted for the first time in Old Gargara allowed us to get to know the city better and to
describe the village polis more accurately with archaeological data. However, the answers to many questions are still
unknown, such as the plan of the houses and the temple, the exact date of the walls, the location of the necropolis, and
how the water needs of the city were met. Among these, the discovery of the necropolis will definitely emich our
knowledge of the city.
ACKNOWLEDGEMENTS
We would like to thank Dr. A. Ozme, representative of the Turkish Ministry of Culture and Tourism for this
surface survey, and also give thanks to research assistants Omer Can Y1ldmm and Murat Ay, archaeologist Ses:kin
Akyiyek, and Osman <;::apalov, archaeologist from Troia Museum, geologist Suat Karak:1§, and Omer Tatar for their
help with the translation.
REFERENCES CITED
Acheilara, Lillian. 2020. The Lesbian Masonry in the Island of Lesbos. In Urbanism and Architecture in Ancient
Aiolis. Proceedings of the International Conference from 6-9 April 2017 in <;anakkale, E.-M. Mohr, K. Rheidt,
and N. Arslan, eds., 279-289. Asia Minor Studien 95. Bonn: Dr. RudolfHabelt.
Akerstrom, Ake. 1966. Die architektonischen Terrakotten Kleinasiens. Lund: Gleerup.
Akurgal, Ekrem. 1983. Alt-Smyrna- Wohnschichten und Athenatempel. Ankara: Turk Tarih Kurumu Bas1mevi.
Akahn, Ay~e Gul. 2007. Troas'da Arkaik Kokenli Bir Yerle~irn: Gargara. In Patronus. Co~kun Ozgiinel'e 65. Y~
Armagam, E. Oztepe and M. Kad1oglu, eds., 21-17. istanbul: Homer Kitabevi.
Arslan, N. 2009. Kuzey Troas Bolgesi Yuzey Ara~tumalan: Perkote ve Palaiperkote'nin Lokalizasyonuna ili~kin
Sorunlar. TUBA-AR Turkish Academy ofSciences Journal ofArchaeology 12: 77- 87.
- . 2017. Surface Surveys in the Northern Troad and the Identification of <;::iglitepe as Ancient Arisbe. Anatolian
Studies 67:129-144.
- . 2020. Gargara-Ein archaisches Heiligtum in der Troas. In Von der Kunst, ein Bauwerk zu verstehen:
Perspektiven der Bau- und Stadtbaugeschichte von der Antike bis zur Gegenwart. Festschrift far Klaus Rheidt zum
65. Geburtstag, A. Druzynski v. Boetticher, A. Wunderwald, and P.I. Schneider, eds., 129- 136. Gottingen: Hubert
& Co. GmbH und Co. KG Buch Partner.
Balcer, Jack Martin. 1984. Sparda by the Bitter Sea: Imperial Interaction in Western Anatolia. Brown Judaic Studies
52. Chico: Scholars' Press.
Ba~aran, Sait. 2000. Aeolische Kapitelle aus Ainos (Enez). Istanbuler Mitteilungen 50: 157-170.
Betancourt, Philip P. 1977. The Aeolic Style in Architecture- A Survey of its Development in Palestine, the
Halikarnassos Peninsula, and Greece, 1000-500 B.C. Princeton: Princeton University Press.
Boehlau, Johannes and Karl Schefold, eds. 1940. Die Bauten, Larisa am Hermos I. Die Ergebnisse der Ausgrabungen
1902- 1934. Berlin: Verlag Walter de Gruyter & Co.
Clarke, Joseph Thacher. 1886. A Proto-Ionic Capital from the Site of Neandreia (II). The American Journal of
Archaeology and ofthe History ofthe Fine Arts 2/2: 136- 148.
- . 1888. Gargara, Lamponia and Pionia- Towns of the Troad. The American Journal of Archaeology and of the
History ofthe Fine Arts 4(3): 291-319.
Cook, John M. 1973. The Troad: an archaeological and topographical study. Oxford: Clarendon Press.
Cook, John and Richard V. Nichols. 1998. Old Smyrna Excavations- The Temples ofAthena. London: British School
of Athens.
Frisch, Peter. 1975. Die Inschriften von Ilion. Die Inschriften griechischer Stadte aus Kleinasien 3. Bonn: Habelt.
Hansen, Mogens Herman. 2004. Territory and Size of Territory. In An Inventory of Archaic and Classical Poleis,
M.H. Hansen and T.H. Nielsen, eds., 70-73. Oxford: Oxford University Press.
Hood, M. Sinclair F., John Boardman, and John Kinlock Anderson. 1954. Excavation on the Kofina Ridge, Chios. The
British School at Athens 49: 123- 182.
Judeich, Walther. 1898. Bericht uber eine Reise irn nordwestlichen Kleinasien. Sitzungsberichte der Preuj3ischen
Akademie der Wissenschaften zu Berlin XXXVl(2): 436-555.
-. 1901. Gargara und der Altar des idaischen Zeus. Jahreshefte des Osterreichischen Archiiologischen Institutes in
Wien 4: 111- 125.
Kjellberg, Lennart. 1940. Larisa am Hermos- Die architektonischen Terrakotten. Larisa II. Stockholm: Kung!.
Koldewey, Robert. 1891. Neandria-Programm zum Winckelmannsfeste der Archaologischen Gesellschaft zu Berlin,
Bd. 51. Berlin: Reimer.
Leaf, Walter. 1923. Strabo on the Troad. Cambridge: The University Press.
Mallwitz, Alfred and Wolfgang Schiering. 1968. Der alte Athena-Tempel von Milet. Istanbuler Mitteilungen 18: 90-
143.
Meier, Ludwig. 2012. Die Finanzierung ojfentlicher Bauten in der hellenistischen Polis. Berlin: Verlag Antike.
Mitchell, Stephen. 2004. Troas. In An Inventory ofArchaic and Classical Poleis, M.H. Hansen and T.H. Nielsen, eds.,
1007- 1017. Oxford: Oxford University Press.
Radt, Wolfgang. 1970. Siedlungen und Bauten au/ der Halbinsel von Halikarnassos. Ti.ibingen: Wasmuth.
Ramage, Andrew. 1978. Lydian Houses and Architectural Terracottas. Archaeological Exploration of Sardis, vol. 5.
London: Harvard University Press.
Rose, Charles Brian. 1995. The 1994 Post-Bronze Age Research and Excavation at Troia. Studia Troica 5: 81-105.
- . 2008. Separating Fact from Fiction in the Aeolian Migration. Hesperia 77. 399-430.
Ruschenbusch, Bamberg. 1978. Untersuchungen zu Staat und Politik in Griechen/and vom 7.-4. Jh. v. Chr. Eberhard:
Fotodruck u. Verlag GmbH.
-. 1983. Tribut und Biirgerzahl im ersten athenischen Seebund. Zeitschriftfi-ir Papyrologie und Epigraphik 53: 125-
143.
-. 1985. Die Zahl der griechischen Staaten und Arealgrosse und Burgerzahl der "Normalpolis." Zeitschrift fur
Papyro/ogie und Epigraphik 59: 253-263.
Schefold, Karl. 1939. Das Aolische Kapitell. Jahreshefte des Osterreichischen Archiiologischen Instituts in Wien 31:
42-52.
Schulz, Armin. 2000. Die Stadtmauern von Neandreia in der Troas. Asia Minor Studien 38. Bonn: Habelt.
SNG Miinchen. 1991. Sy/loge Nummorum Graecorum. Deutsch/and: Staat/iche Mi-inzsammlung Munchen. Heft 19.
Troas-Lesbos. Nr. 1-813. Miinchen: Hirmer Verlag.
Sparkes, Brian A. and Lucy Talcott. 1933. Black and Plain Pottery: of the 6th, 5th and 4th Centuries B.C. The
Athenian Agora 12. Princeton: American School of Classical Studies at Athens.
Stupperich, Reinhard. 1995. Ein archaisches Kriegerreliefaus Gargara. Asia Minor Studien 16: 127- 138.
Tenger, Bernhard. 1995. Phoroshohe und Bevolkerungszahl. Die Athener Tributlisten als Indikator fur die GroBe der
Einwohnerschaft einer Polis? Studien zum antiken Kleinasien III. Asia Minor Studien 16: 139-160.
Wescoat, BonnaDaix. 2012. The Temple ofAthena at Assos. Oxford: Oxford University Press.
Wiegartz, Hans. 1994. Aiolische Kapitelle-Neufunde 1992 und ihr Verhaltnis zu den bekannten Stticken. Asia Minor
Studien 11: 117- 132.
Winter, Nancy A. 1993. Greek Architectural Terracottas- From the Prehistoric to the end of the Archaic Period.
Oxford: Clarendon Press.
PARTIII:
THE STATE OF THE FIELD
(Photo courtesy of the c;:actir Hoyiik Project)

CHAPTER TWENTY-SIX
ARCHAEOBOTANY IN ANATOLIA
JOHN M. MARSTON AND LORENZO CASTELLANO
Research in archaeobotany, the study of plant remains from archaeological sites, across Anatolia has significantly
expanded in the last 25 years since the last systematic survey of the region was published (Nesbitt and Samuel 1996a).
Our study builds on this survey and other, more recent but also more chronologically and regionally focused reviews
(e.g., Fairbairn 2021; Riehl 2014; Riehl and Nesbitt 2003) to present the current state of archaeobotanical
understanding of the agricultural history of Anatolia. In this chapter, we present the results of a systematic literature
review that encompasses all archaeobotanical studies published to date for the entirety of contemporary Turkey, which
span the Epipalaeolithic through the Medieval period. We note trends that demonstrate the ongoing expansion of
archaeobotany at post-Neolithic sites and on Turkish-led research projects as significant changes over the last 25 years
and argue that further growth of archaeobotanical studies across Anatolia will necessitate regular updates to this
review.
In the first systematic attempt at a synthesis of archaeobotanical publications from Turkey, Nesbitt and Samuel
(1996a: 91) identified 20 current excavations with ongoing archaeobotanical research, as well as 19 projects
completed by that date, studied by 30 archaeobotanists. They include the study of macroscopic plant remains (seeds,
fruits , tubers, and wood) as well as microscopic remains (i.e., phytoliths) and residues among these studies. While
their bibliography provides references to published botanical results from the sites mentioned and major crops are
listed for each, Nesbitt and Samuel do not include either qualitative or quantitative comparisons of those results by
either region or period. A subsequent study by Riehl and Nesbitt (2003) focuses on comparing agriculture in the
Aegean region of Greece and Turkey to other Near Eastern sites (including those in central and southeastern Anatolia)
during the Late Bronze and Iron Ages. In this study, the authors compiled a qualitative comparison of crops on a site-
D "'
.O() •...,,.
,,.,.<:::t r;r"
:i, ~
J O site included in the review
i
Central Anato lia
r8) Site excluded from the review
Eastern Anatolia
Med iterranean egean
North ern Anatolia Marmara
Southeastern Anato lia ~ ransitional Mediterran ean
0 200 400 800
km
Figure 26-1. Sites with published archaeobotanical seed remains, both those included and excluded from this study, by
region. Site codes correspond to Table 26-1 and ESM 26-1. 1 Regions delineated after Atalay 2014.
1
ESM = Electronic Supplementary Material. This file is hosted in perpetuity on tDAR (the Digital Archaeological Record) and
available by searching tDAR.org for this chapter title or tDAR id 461278. It is also available via its DOI:10.48512/XCV8461278
Archaeobotany in Anatolia 339
by-site basis, by both region and period, allowing them to identify regional agricultural signatures that differentiated
Aegean agriculture from that practiced further east and thereby how Near Eastern agricultural systems, but not those
of the Aegean, changed significantly during the Iron Age. Fairbairn (2021) adopts a similar approach, with a narrower
area of focus (all of Asia Minor, excluding eastern and southeastern Turkey) but a broader chronological range (the
Early Chalcolithic through the Late Bronze Age), also comparing the importance of crops among sites using
qualitative metrics. Notably, in this review Fairbairn considers other lines of evidence for reconstructing agricultural
systems, including compiling zooarchaeological evidence for major animal domesticates as well as botanical data, and
other archaeological evidence for agricultural technologies employed.
A distinct approach to synthesizing archaeobotanical data from Anatolia is that of Riehl (2014). Drawing on data
from the ADEMNES database (Riehl 2021) that compiles quantitative archaeobotanical and zooarchaeological data
for the entire Near East, Riehl was able to conduct quantitative comparisons of 35 Anatolian sites dated between the
Pre-Pottery Neolithic B and the Middle Bronze Age. Using multivariate statistical analysis, Riehl distinguished
regional zones of similarity: one spanning western and central Anatolia and the other in eastern Anatolia, with
southeastern Anatolia and the Middle Euphrates as a transitional zone between the two. She also considers differences
in weed floras across the region and the implications of this patterning for understanding regional trade and exchange
in agricultural products.
Our review maintains the broad geographic and temporal scope of Nesbitt and Samuel's (1996a) early synthesis,
while following Riehl (2014) in using only quantitative published results from which to draw conclusions about
similarities and differences over space and time. We restrict our review to cover only seed assemblages (excluding
wood charcoal, microscopic botanical, isotope, and residue analyses) and focus solely on definite or probable
agricultural products in order to focus the evidence for farming systems that is most widely available and usefully
comparative among the greatest number of sites possible. In total, we include 145 chronologically distinct datasets
from 86 sites within the modem boundaries of Turkey (Fig. 26-1, Table 26-1). While it is only possible to include
summary data and discussion of selected datasets in this review, we include a full inventory of sites included, with
chronological extent of occupation and archaeobotanical sampling, geographic coordinates, and bibliographic
references (including reports from which we draw data and those from which we do not) in an electronic
supplementary file hosted on tDAR (ESM 26-1; Marston and Castellano 2021).
Periodization Northern Marmara Aegean Mediterranean Transitional Med. Central Southeast en Eastern
Epi paleolithic Karain B; Okuzini KOrtikTepe Hallan<;:emi

(pre 9,700 BCE)
Aceramic Neolith ic ~1kl1 H6yi.ik; (ayOnU Tepesi;

(c. 9,700 - 7,000BCE) Boncuklu; GObekli Tepe;
<;:atalhOyUk; Gritille; Gusir
Pmarba~1 HOyOk; Cafer HOyi.i;
Neval1 <:;:ori; Ktirtik
Tepe
Ceramic Neolithic Ulucak Mersin -YOmOktepe Erbaba; HOyOcek <;atalhOyOk <;ayOnU Tepesi;
(c. 7,000-6,000BCE) Sumaki HOyUk
Early Chalcolithic ilip1nar Dumuztepe CatalhOyi.ik

(c. 6,000 - 5,400BCE)
Middl e Chalcol ith ic Kumtepe Can Hasan I Kenan Tepe; Tell ~van-Fatmal i
(c. 5,400-4,S00BCE) Kurdu; Girikihac1yan Kalecik
Late Chalcolithic ikiztepe Kumtepe Bakla Tepe; «;ukurir;i Kurur;ay HOyUk «;ad 1r HOyUk; Hacinebi; Yarrm Arslantepe; ~van-
(c. 4,500-3,000BCE) HOyUk; Li man Tepe «;am l i bel Tarlas1 HOyUk <;ayboyu; ~van -
Fatmali Kalecik; Sos
HOyUk
Early Bronze Age ikiztepe; Oymaagar; Troy BaklaTepe; <;ukurir;i Mersin-YUmUktepe; DemircihUyUk; Titri~ HOyUk; Mezraa Arslantepe; ~van-
(c. 3,000 - 2,000BCE) HOyUk; Li man Tepe; Tell Tayinat KUIIUoba; KUltepe- HOyUk; Gre Virike; ~van Kale; ~van-
Yenibademli HOyUk Kanesh Tilba~ar HOyUk; Ta~kun Mevkii;
Ziyaret Tepe imanoglu HOyUk
Midd le Bronze Age ikiztepe Troy KiliseTepe Beycesultan Bogazk6y; Boyal1 Hirbemerdon Tepe; Sos HOyUk
(c. 2,000-1,600BCE) HOyUk; BUklukale; Mezraa HOyUk; Sal at
Gord ion; Kaman- Tepe; Ziyaret Tepe
Kalehoyuk; Ki.Htepe-
Kanesh
Late Bronze Age Oymaagar; Troy Kaymakr;:1 Kifise Tepe; Kinet Bogazk6y; Gord ion; Karkemish; Tille ~van-~van Kale
(c. 1,600-1,200BCE) HOyUk; Tatarll Ku~kl1; Ortak6y HOyUk; Ziyaret Tepe
HOyUk; Tell Atchana
Iron Age Oymaagar; Daskel eion;Troy Miletus KiliseTepe; Sirkeli; Gordian; Ku~akl1; Ziyaret Tepe; Zeviya Ayanis; Sos HOyUk;
(c. 1,200-300/200BCE) Tell Tayinat; Kinet Kerkenes Tivilki; Karkemish; Patnos; Yoncatepe
HOyUk Ziyaret Tepe; Tille
HOyUk
Helle n istic Daskeleion Ephesos Tatarll H6y0k Pessinonte; Gordian Karkemish; Tille ~van-~van Kale
(c. 300/200-1 BCE) HOyUk
Roman Oymaagar;* Ephesos Gordian; Pessinonte ll1su HOyUk; Zeugma ~van-~van Kale
(c. 1-400CE)
Medieval I Oymaagar;; Komana KUr;Ukyal1 Hierapolis Ki net HOyUk-Tupra~ Sagal assos; Gordian; Pessinonte Gritille; Ziyaret Tepe; ~van-~van Kale
(c. 400-1100/200CE) Field; Mersin- Beycesultan Karkemish
YUmUktepe
Me d ieval II Komana Daskeleion KiliseTepe; Kinet Kaman -Kalehoyuk; Mezraa H6y0k; Gre ~van-~van Kale;
(after 1100/1200 CE} HOyOk•TOpra~ Field Gordian Virike ~an~Ta~kun Kale
Table 26-1. Summary table of site names and periods from which quantitative archaeobotanical results have been
published and included in this study (see ESM 26-1 for excluded sites and periods, with bibliography). *Oymaaga9
reports a combined Hellenistic/Roman phase, which we consider here among other data from the Roman period.
340 Chapter Twenty-Six: Marston and Castellano
ENVIRONMENTAL BACKGROUND
The climate, precipitation, and vegetation distribution of Turkey is complex and variable, as it lies at the nexus of
several weather systems and at the boundary between three major plant communities (Turun9oglu et al. 2018; Zohary
1973). The orographic influence of Turkey's mountain ranges, especially the Pontic range of the north and the Taurus
range of the south, promote distinct vegetation communities along the Mediterranean coast, Black Sea coast, and
across the central Anatolia plateau (Atalay 2018). On the basis of climate and vegetation, Turkey can be divided into
eight eco-regions: the Aegean, Marmara, Mediterranean, and Transitional Mediterranean (the region of upland
southwestern Turkey also known as the "Lake Region"), as well as central, northern, eastern, and southeastern
Anatolia (following Atalay 2014). We use these physiographic regions as the basis for grouping sites, as shown in Fig.
26-1. It should be noted that this definition incorporates a degree of ecological and cultural diversity within single eco-
regions: for example, eastern Anatolia includes both the Upper Murat-Van district and the Upper Euphrates Valley.
Unfortunately, finer geographic distinctions are not possible given the limited number of records available.
125 250
2324
mm
0
a mm
160
140
120
100
60
@ Ephesus (Aegean) @Zeugma (SE-Anatolia)
@Mersin (Mediterranean) @ Bogazkoy (C-Anatolia)
20 © lkiztepe (N-An atolia) ® Ayanis (E-Anatolia)
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOVDEC
b
Figure 26-2. a) Main map: modeled annual precipitation 1970-2000, in mm, across Anatolia.
Physiographic regions and sites follow Fig. 26-1; b) Inset: seasonal precipitation at six locations associated
with archaeological sites; letters correlate seasonal graphs with locations on main map. Values are extracted
from WorldClim2 (1970-2000 average), 30-seconds spatial resolution (Fick and Hijmans 2017).
The complex physical geography of the Anatolian peninsula has direct implications for the climate of each eco-
region, with differences in rainfall and temperature in large part defined by proximity to the seacoast and elevation
(Turke~ 2003) (Fig. 26-2). As expected by their location, the western and southern coastal regions experience a typical
Mediterranean climate, with mild, wet winters and hot, dry summers. Due to the influence of humidity tracks from the
Black Sea, the northern coast is wet throughout the entire year, with an oceanic climate defined by wet, mild-to-cold
winters and wet, warm summers (Turke~ 2003). Compared to the coastal regions, significantly lower amounts of
rainfall reach central Anatolia, due to a combination of orographic effects (determined by the Taurus and Pontus
chains) and localized winter high-pressure systems, both of which are responsible for partially blocking winter
Mediterranean storms. The resulting climate in the plateau of central Anatolia is cold semi-arid, with an average
rainfall in the non-mountainous regions of the plateau between 280 and 400 mm/year (Fig. 26-2a). The distribution of
precipitation in central Anatolia mirrors the typical Mediterranean seasonality, except with a relatively higher
contribution of spring rains to the annual total (Fig. 26-2b).
These differences in both the amount and timing of moisture, as well as topography and soil development,
structure the botanical landscapes of Anatolia, both in terms of natural vegetation and agricultural potential. Certain
crops, such as olives, are frost sensitive and thus are generally restricted to Mediterranean and Aegean agriculture,
though pollen and archaeological evidence suggests that olive arboriculture may have been expanded opportunistically
into highland areas during warmer climatic periods (England et al. 2008; Kaniewski et al. 2007; Vermoere et al. 2003).
Other crops, especially annuals, are restricted only by precipitation, both total annual amounts and seasonal variability
(Ci9ek 2003), allowing irrigation to expand potential areas of cultivation. More substantial barriers to agriculture are
posed by heavily forested areas, steep slopes, and poor or thin soils, although archaeological evidence provides
counter-examples of farming in such areas from time to time (e.g., Casana 2008). When considering the entire
Holocene period, as we do in this study, climate change must also be acknowledged as a factor that led to significant
differences in agricultural potential from the Early Holocene through the mid-Holocene climatic optimum and Late
Holocene aridification, although this topic is much more complex than can only be briefly addressed here (see, for
example, Roberts et al. 2016).
METHODS
Sites with published archaeobotanical data were identified through a systematic review of the published literature.
Once identified, we chose sites for inclusion in our analytical dataset based on the following criteria:
l. Archaeobotanical data reported include seed (carpological) data

2. Seed data are quantitatively presented by either count or weight
3. The chronological phasing of the archaeobotanical samples is sufficiently precise to attribute their results to
periods no broader than the periods we defined (Table 26-1)
4. The report of the data is publicly available; theses and dissertations not available in online repositories were
excluded
5. Underwater or shipwreck contexts were excluded, given the substantial differences in formation processes that
produced and preserved these exceptional contexts, and that their location of production may not have been in
Anatolia
6. Reports that duplicated results from another report were excluded when they did not offer new data; we prefer
the original in such cases, though when later reports do not distinguish new and previously published data, we
use the more recent and complete report
If a report passed all of the criteria above, then it was included in our analysis, regardless of number of samples or
number of botanical finds. Sites from which seed data are reported, but which do not meet the other four criteria above,
were excluded but do appear on Fig. 26-1 as excluded sites and are listed with a full bibliography, chronology, and
geographic coordinates in ESM 26-1. In total, we include 146 chronologically distinct datasets from 86 sites; an
additional 29 sites are excluded but listed in ESM 26-1 and shown on Figs. 26-1 and 26-2.
A major challenge in comparing data across sites, time periods, analysts, and era of publication is that botanical
terminologies vary in ways both minor and significant. We addressed this challenge in several ways. We standardize
botanical names for crops following Zohary et al. (2012), save for cereals in which we follow the earlier taxonomy
used in the Flora of Turkey (Davis 1965-2000). When identifications were presented at a taxonomic level more
specific than the ones we adopted, we reduced that level of identification-e.g., two-row barley (Hordeum vulgare
subsp. distichum) was reduced to hulled barley (Hordeum vulgare). When identifications were presented at uncertain
levels of identification, we reduced those to the genus or (sub-)family level-e.g., Triticum cf. dicoccum was reduced
to Triticum sp., and Vicia/Lathyrus was reduced to "Pulse indeterminate." When identifications were uncertain as to
domestication status (e.g., Vicia sp., Papaver sp.), we excluded those results as possibly wild. In a few cases, generic
identifications were converted to specific attributions where it was clear that only a single domesticated taxon was
possible-e.g., in one instance a taxon recorded as Olea sp. could only be Olea europaea on phytogeographic grounds
and in another thousands of Iron Age finds were recorded only as Hordeum sp., while described as unequivocally
hulled domesticated barley (H vulgare) in the text. We excluded all definitively identified wild taxa (e.g., Triticum
boeoticum), regardless of quantity or evidence for cultivation. Tuber and parenchyma fragments are likewise excluded
from our database.
Raw counts were recorded for each period from each site; when multiple reports presented distinct results from the
same period, these were combined. In a few instances, only weights were reported for cereals and pulses (e.g.,
Gordion, Hacmebi, Gritille [Miller 1998, 2010; Stein et al. 1996]); these were converted to counts using site-specific
data, where available, or else approximated at a ratio of 0.0lg = l seed, following empirical data provided by Miller
(Stein et al. 1996: 255). Nutshell and fruit endocarp reporting is not standardized in the literature; we converted
nutshell fragment counts to approximate whole nut equivalents following Filipovic (2014), at a ratio of 15 fragments=
l Pistacia and 20 fragments = l Amygdalus, Prunus, or Quercus. Olea fragment counts were converted to whole
endocarps at a ratio of 4: 1, following Bouchaud et al. (2017), while we converted Olea weights to counts at 0.3 g = l
endocarp (following figures provided by Galili et al. 2021). We convert grape berries to grape seeds at a ratio of 1:4.
Given that fig (Ficus carica) "fruits" (technically syconia) are rare and inconsistently enumerated (in some cases
entire fruits, in others potentially fragments of fruit flesh), we did not convert these to "seed" (technically fruit) counts.
A challenge with using raw count values to compare sites is that sites with different sampling intensities may have
different numbers of seeds. Additionally, single caches of relatively pure crop seeds from a single site may
substantially skew the apparent importance of a tax.on during the period to which that cache dates. As a result,
quantitative comparisons of sites on a regional basis are often challenging to present accurately, with some form of
data reduction, such as multivariate analysis or relative statistical ratios, often employed to mitigate these issues
(Marston 2014; Pearsall 2015; Stevens 2014). One such method is conversion to Representativeness Index (RI)
statistics (Effenberger 2018 ; Heiss and Stika 2013; Stika and Heiss 2013), a weighted ranking system (Popper 1988)
that aims to account for different sampling intensities and different absolute quantities of finds between sites. RI
values can then be compared across a region or through time to detect shifts in crop utilization that are less obscured
by differential sampling across sites than metrics based on raw counts.
Specimens per category < 1000 specimens per category ::::: 1000 specimens per category
per site-period
RI scores
1 <10 specimens < 10 specimens
2 10-49 specimens 10-49 specimens
5 :::::soo specimens :::::soo specimens
6 25-49% of category specimens
7 :::::50% of category specimens
RI factors Per site

x2 :::::20 samples or :::::1000 L sediment
x4 :::::40 samples or :::::5000 L sediment
x5 :::::so samples
Table 26-2. Representative Index (RI) scoring matrix, after Effenberger (2018: 67) with modifications as
described in the text. "Category" refers to one of: cereals, pulses, fruits and nuts, oil and fiber crops.
We transformed count data into RI values following the approach of Effenberger (2018), modified in the three
following ways for this study. First, Effenberger (2018: 67) merged taxa differently from the method we adopted and
describe above. Second, while Effenberger (2018: 67) set a threshold of 1000 total specimens (both cultivars and wild
seeds) per site-period for assigning RI scores to taxa, that approach is misleading for Anatolian assemblages with large
wild seed assemblages and frequently substantial differences in the types of crops preserved; e.g., some sites are
dominated by cereals, others by fruit remains, and so forth. We instead use a threshold of 1000 specimens per category
of domesticates (cereals, pulses, fruits and nuts, oil and fiber crops) in place of Effenberger's single threshold. Third,
for sites with fewer than 1000 specimens in a category of domesticates, we include an additional RI category with a
score of 5 for taxa with more than 500 specimens for that site-period. The full set of calculations employed is
summarized above as Table 26-2. These RI statistics were calculated on a site-specific basis, for each period with data
reported, and RI values were summed by site, period, or region as needed for the following analyses.
_[:;~film-
-~~~r
□- Mediterranean
Nr. sites
Epipal. - PPN - Neolithic - E-Chalc. - M-Chalc. - L-Chalc. - EBA - MBA - LBA - IA - Hell. - Rom. - Med. 1- Med. II
Nr. samples
• full number of samples is not published for one site
Figure 26-3. Number of sites and number of archaeobotanical samples included in this dataset, by region and period.
Bars above the line denote number of sites; bars below the line denote minimum numbers of samples included.
Sample numbers are incompletely reported where denoted by an asterisk.
RESULTS
Distribution of Sites and Samples by Period and Region
As indicated in Table 26-1 and in Fig. 26-3, the distribution of sites eligible for inclusion in this dataset is not even
across regions or time periods. Some areas, especially central and southeastern Anatolia, include at least one site from
nearly every period, while others, such as Northern Anatolia and the Transitional Mediterranean, are spotty and thin.
Certain regions and periods include much larger numbers of samples than others, which in some cases is correlated
with the number of sites included, but not in all cases, as some sites are more intensively sampled than others. Fig. 26-
3 depicts the reason why we adopt the RI approach, as it permits comparison in the frequency of plant types across
periods and regions more effectively than an approach based on raw counts, which would be more heavily influenced
by differences in sampling intensity within the dataset.
All Taxa by Period
We present the comprehensive dataset by period in Figs. 26-4 and 26-5. These figures report two methods for
reporting frequency and intensity of use: ubiquity (the number of sites at which a taxon appears) and summed RI
across all sites from that period. We provide the total number of sites and summed RI values per period in the header
row, which provide a summary value of the quantity and quality of data for the period, with a higher total RI value
indicating a more robust database (Effenberger 2018: 66). This pairing of metrics for each taxon allows us to
distinguish dominant taxa that are frequent and numerous across many sites-e.g., barley for the Iron Age, which is
found at 17 of 19 sites reported, and totals 34% of the total RI for cereals for that period (Fig. 26-4}-from those that
are found in large quantities in singular instances for their period-e.g., opium poppy during the Epipalaeolithic,
which is found at only one site but in a large quantity and is the only oil seed attested for that period and so comprises
100% of the summed RI for that category of finds (Fig. 26-4).
Epipal PPN Neol. E-Ch. M-Ch. L-Ch . EBA MBA LBA IA Hell. Rom. Med.I Med.II
4 1186 71672 3 I 780 7 I 495 14 I 808 21 I 1243 14 I 378 16 I 1057 19 I 1398 8 I 349 7 I 345 14 I 516 10 I 285
Einkorn 4 I 28 2 I 45 4 I 29 10 I 94 16 I 108 15 120 8171 9 I 60 2112 112
Emmer 3 I 40 4 I 47 2 I ss iii 17 I 126 11 I 79 9 I 87
Bread/Macaroni Wheat 3 I 34 s I ss 3 I 40
Barley 2 I 36 6 I 27 3 I 75
Naked Barkey 2 I 1s 3 I 46 2 I 3s 2 I 20 2 14 214 213 s I 25 219 112 111
Broomcorn Mill et 111 11 2 2 14 6 I 38 2 I 19 2 15 111 41 9
Foxtail Millet 112 4 I 33 2 25 21 4 111 114
Rye 1 11 21 7 5 I 21 115 11 4 5 I 15 41 6
Oat 11 5 11 8
Asian Rice
Lentil 10 I 128 $i ~ 21 I 101 10 I 23
Pea 7 I 11 4 I 58 3 I so 3 I 49 6 I 31 10 I 61 8 I 23
Bitter Vetch 6 I 102 6 I 49 3 I 40 4 I 29 6 I 57 13 I 71 9 I 26 418 4 21 s I 14
Chickpea 3 I 28 s I 24 21 30 2 I 13 4 I 15 6 I so 3 16 SI 37 s I 20 4117 1[ 7 2 16 319
Gra ss Pea 6 In s I 23 3 I 35 21 3 7 I 34 12 I 47 s I 12 s I 47 9 I 27 11 7 111 31 25 s I 10
Broad Bean 2 I 14 115 111 11 2 s I 27 2 12 s I 44 6 I 23 114 2 I 32 113
Common Vetch 2110 111
Opium poppy
Gold -of-pl easure 115 112
-
11 5
('nrsites' I 'RI') Sesam e 111 21 6
<5% 5-10% ·3 RI -grou p% Cotton 11 2
Figure 26-4. Frequency and intensity of use of cereal, pulse, and oil and fiber taxa by period. In each cell, the first
number records the number of sites (for each period, across all of Anatolia) in which that taxon appears; the second
number is the total RI value for that taxon across all sites. Shading intensity corresponds to the relative percentage of
the total RI for all taxa within the category for the taxon within that period; darker colors are relatively more dominant
taxa by RI. Not included are rare and ambiguous taxa: "new glume wheat," spelt, undifferentiated hulled wheats, and
undifferentiated millets. Full period names and chronological extent are provided in Table 26-1.
This form of presentation distinguishes well-supported patterns, such as the relative intensity of barley use
compared to other cereals and the gradual diminishment in the use of hulled wheats following the Bronze Age, from
those that are more poorly attested, such as changing frequencies of fruit use over time. The uneven sampling of
periods adds another dimension of ambiguity, which must be acknowledged in interpreting the results of these
analyses.
Epipal PPN Neol. E-Ch. M-Ch. L-Ch. EBA MBA LBA IA Hell. Rom. Med.I Med.II
11 I1142 7 1 672 3 I 780 7 I 495 14 I 808 21 I 1243 14 I 378 16 I 1051 19 I 1398 8 I 349 7 I 345 14 I 516 10 I 285
Almond s I 73 2 I 21 11 5 21 9 2 13 11 5 1 I4 215 215
Hawthorns 11 2 4 112 3 14 11 2 11 4
Hackberry 7 I 85 1 I 35 1 I 35 1I1 11 2 212 11 3 1 15 41 11 114
Appl e/Pear 1 I 10 1 I 10 11 2 111 216 11 2 114
Pi stachio 3 125 2 120 2 I 12 419 11 8 111
Oak 3 I 15 1 I 10 1 14 1I2
Grape 3 I 17 1110 3 111 9 I 25 18 I 73
Common Fig 3 I 11 4 I 26 3 I 50 319 3 I 16 7 I 36 2 I 40 5 I 20 311
Corn elian Cherry 11 5 111 1112 11 4 114
Plums genus 2 I 14 216 417 2 16 11 2 215
Brambles 1 I 10 115 3 I 13 11 3 318 21 7 11 1
Olive 111 112 111 4 I 2s 7 11 2 217 111
Hazelnut 11 5 111 1 12 3 16 1 I8 1 14 1 14 11 4
Cherry Plum 11 4
Walnut 111 21 8 2 15
Pine 11 2 11 1 1 112 11 1
Common Plum 111 2 I 10 114
Blackthorn 111
Pom egranate 2 16 1 I4 215 2 112 111
Cucurbits 111 2 112 114 11 2
Sour/Sweet Cherry 2 19 114 114
Mulberry 11 1 11 4
('nrsites' I 'RI ') Peach 1 14
<5% 5-10% 10-30% RI-group% Apricot 114
Figure 26-5. Frequency and intensity of use of the most common fruit taxa by period. In each cell, the first number
records the number of sites (for each period, across all of Anatolia) in which that taxon appears; the second number is
the total RI value for that tax on across all sites. Shading intensity corresponds to the relative percentage of the total RI
for all taxa within the category for the taxon within that period; darker colors are relatively more dominant taxa by RI.
Not included are rare and ambiguous taxa: Cotoneaster, Sorbus, Rhus coriaria, Rosa, Sambucus, and undifferentiated
plum/almond genus. Full period names and chronological extent are provided in Table 26-1.
Cereals
Summed RI values by period, drawn from the raw data in Fig. 26-4, are displayed graphically as Fig. 26-6. Note
that we exclude wild species from our dataset, leaving the Epipalaeolithic without data, although wild cereals were, to
a variable extent, exploited during that period. Fig. 26-6 indicates more clearly several trends also evident in Fig. 26-4.
One is the decrease in hulled wheats relative to increases in both free-threshing wheats and hulled barley over time. A
second is the near disappearance of naked barley following the Middle Chalcolithic. A third is the appearance and
continued use of millet and rye, albeit as less abundant cereal taxa, beginning in the Iron Age.
It is more difficult to tease apart regional patterns in plant use from the overall diachronic trends noted above,
primarily because there are often few (or no) sites from any one region in any one period. To maximize the number of
sites and assess regional patterning, we divide the assemblages into macro-regions (combining the ecologically similar
Marmara, Aegean, Mediterranean, and Transitional Mediterranean regions into one "circum-Mediterranean" macro-
region) and macro-periods (combining the entire Chalcolithic into one period, the Hellenistic and Roman into another,
and the two Medieval periods into a third). Fig. 26-7 presents a graphical depiction of summed RI values for each
macro-region, grouped by period. Although these data are much noisier, due to the limited number of sites per region-
period, the data suggest a few regional trends. Millet appears to be more utilized in central and eastern Anatolia in
post-Iron Age contexts than in other regions. Oat is restricted to northern Anatolia, though is uncommon there as well.
Eastern Anatolia appears to give up hulled wheat agriculture earlier and more completely than other regions,
beginning in the Middle Bronze Age, although this observation must be made with caution given the relatively poor
sampling (in terms of numbers of samples) for most periods in the region. Naked barley is consistently absent from
southeastern Anatolia, though its use is temporally sporadic in other regions. Most notable, however, is the relative
consistency in wheat and barley use across macro-regions, in contrast to the notable chronological pattern visible in
Fig. 26-6. While some macro-regions are outliers during certain periods, potentially due to sample-size effects or the
sampling of atypical contexts, those trends are not temporally sustained. For example, while southeastern Anatolia has
the highest abundance of free-threshing wheats in the Late Bronze Age, both the adjacent earlier and later periods
show low abundance of free-threshing wheats when compared to other regions, a pattern driven nearly entirely by a
few large concentrations of free-threshing wheat from Tille Hoytik (Nesbitt 2016) in contrast to the low frequency of
cereals at other Late Bronze Age sites in the region.
□
Asian Rice
(Oryza sativa)
□
Oat
(Avena sativa)
■
Rye
(Secale cerea/e)
■
Millet
(P. miliaceum/5. italica)
□
Naked Barley
(H. vu/gore var. nudum)
■
Barley
(Hordeum vu/gore s.l.)
5■ Spelt
(Triticum spelta)
■ " New Glume Wheat"
4
(Triticum timopheevii?)
Domesticated glume wheat
(Triticum spp. hulled s.l.)
Einkorn
(Triticum monococcum)
Emmer
(Triticum dicoccum)
■ Free-Threshing wheat
(Triticum aestivum/durum)
Epipal. PPN Neol. E-Ch. M-Ch. L-Ch. EBA MBA LBA IA Hell. Rom. Med. I Med. II
Figure 26-6. Relative abundance of cereal taxa based on summed Rl scores by period.
Epipaleolithic PPN Neolithic Chalcolithic Early Bronze Age
liiJ Oat [g]Asian Rice

1111 Mill et ■ Rye
~ Naked Barl ey
■ Spelt ■ Barl ey
■ " New Glum e Wheat "
■ Glum e w heat
~ Emm er 3 Ein korn
■ Free-Threshin g w heat
~ ~
0
0
~0
~0
0
0
z z z z
ii;
- -
ii;
N
-d
0
ii; ii;
E
ii;
§:
-d ii;
§:
- N
-d ii; <iiC -d <iiC <iiC

"'
-d ii;
C
<f
u
C
<f
w
'"
C
<f
w
Cl>
:::;;
e
'z<f"
C C
<f
C
<f
w
C
<f
Cl>
:::;;
_g z<f
C
ii;
C
<f
<iiC
<f
Cl>
:::;;
e
<iiC
<f
C
<f
u
<f
w <f
'"
C
<f
w _g z
Cl>
:::;;
<f
u
'" '"
C
l l
C
Cl>
:::;;
e z<f
C
u w u w z Cf)
Cf)
i:3 Cf)
u Cf)
i:3 Cf)
u i:3
Middle Bronze Age Late Bronze Age Iron Age Hellenistic/Roman Medieval (I and II)
9 I
~ !'o; ~
ii;
C
<f
'"
C
<f
<iiC
<f
-d
Cl>
:::;;
<iiC <iiC
'"
C
<f
u w w
Cf)
e z<f
()
<f
u w
Figure 26-7. Relative abundance of cereal taxa based on summed RI scores by macro-region and macro-period.
The circum-Mediterranean macro-region combines the ecologically similar Marmara, Aegean,
Mediterranean, and Transitional Mediterranean regions. Numbers following macro-region names
are the number of sites with botanical assemblages for that period.
Pulses
Pulse data are displayed in Figs. 26-8 and 26-9 in the same way as cereals were above. Fig. 26-8 indicates several
temporal trends in pulse use. One is the relative consistency in the use of grass pea, pea, and chickpea over time,
although a few periods show a relative decline in those taxa in favor of vetches (Late Bronze Age and Iron Age) or
■ Grass pea genus

(Lathyrus spp.)
□ Common pea
(Pisum sp.)
□ Chickpea
(Cicer sp.)
■ Lentils
(Lens sp.)
□ Common vetch
(Vicia sativa)
□ Broad bean
(Vicia faba)
□ Bitter vetch
(Vicia ervilia)
■ Vetch indif.
(Vicia sp.)
EDiDal. PPN Neol. E-Ch. M-Ch. L-Ch. EBA MBA LBA IA Hell. Rom. Med. I Med. II
Fig. 26-8. Relative abundance of pulse taxa based on summed Rl scores by period.
Epipaleolithic PPN Neolithic Chalcolithic Early Bronze Age
1'l 1'l 1'l 1'l 1'l 1'l 1'l

□ □ □ □ □ □ □
0 0 0 0 0
z it z z z z it
0
1iiC:
- 1ii-
1iiC:
N
,:j
Q)
0
1iiC:
~
1iiC:
0
1iiC:
...
1iiC:
0
,:j
Q)
0
1iiC:
-
1iiC:
£
1iiC:
N
1iiC:
~
,:j
Q)
£
1iiC:
~
1iiC:
e e
1iiC: 1ii
§:
,:j
Q)
-
1iiC: 1iiC: 1iiC: 1iiC: ,:j
Q) 1iiC:
C: C:
1LJ.J 1 ::;; ::;; ::;; 1LJ.J 1 ::;; ::;;
1u 1 1 1LJ.J 1LJ.J 1z 1 1 1LJ.J 1z 1u 1z 1u 1LJ.J 1LJ.J 1z
LJ.J
(/)
e
0
z u e u LJ.J e LJ.J
(/)
e
0 (/)
e
0
(/)
0 (/)
0
Middle Bronze Age Late Bronze Age Iron Age Hellenistic/Roman Medieval (I and II)
§: ~ M ~ ~ ~
1iiC: 1iiC: 1iiC: 1iiC: 1iiC: 1iiC: 1iiC: 1iiC: 1iiC: 1iiC: 1iiC: 1iiC:
1u 1LJ.J 1
z
1LJ.J 1z
1u 1LJ.J 1
z
1u 1LJ.J 1z 1LJ.J
(/) (/) (/) (/)
Figure 26-9. Relative abundance of pulse taxa based on summed RI scores by macro-region and macro-period.
Mediterranean, and Transitional Mediterranean regions. Numbers following macro-region names
are the number of sites with botanical assemblages for that period.
lentils (Roman). The Roman emphasis on lentils stands out from other periods, especially those immediately preceding
and following. Broad bean becomes more prevalent beginning in the Early Bronze Age through the Medieval periods,
although its intensity of use fluctuates. Most interesting, perhaps, is the general balance between bitter vetch, lentil,
and the combination of pea, grass pea, and chickpea in nearly every period.
Regional trends are again difficult to establish with certainty, given the small number of sites represented in any
single pairing of region and period. Regions are more similar than different from one another overall, excepting
periods in which regions are represented by single sites. One pattern apparent in these data is the greater prevalence of
chickpea in eastern and southeastern Anatolia during the Hellenistic and Roman periods, in comparison with other
regions. A second is the relative emphasis on bitter vetch and lentil versus pea and grass pea in central Anatolia from
the Middle Bronze Age through the Roman period, in comparison with other regions. A third is the higher proportion
of pea, grass pea, and chickpea compared to bitter vetch and lentil in southeastern Anatolia from the Iron Age through
the Medieval period. Finally, broad bean is more prevalent in the circum-Mediterranean region, southeastern Anatolia,
and northern Anatolia than in central and eastern Anatolia, where broad bean is present only during the Bronze Age.
Oil and Fiber Crops
Oil and fiber crop taxa are few, and finds are much less numerous than those of cereal and pulse taxa, rendering
these taxa unsuitable for detailed chronological and regional analysis. The general trends in oil and fiber crop use can
be seen clearly in Fig. 26-4. Opium poppy is the earliest oil seed, with use attested only at Epipalaeolithic and Pre-
Pottery Neolithic Kortik Tepe (Rossner et al. 2018), while flixweed (Descurainia sophia, a wild mustard) was used
intensively between the Pre-Pottery Neolithic and Early Chalcolithic only at <;::atalhoyiik (Fairbairn et al. 2007). Flax is
the dominant oil and fiber crop from the Pre-Pottery Neolithic through the Medieval and in some periods is the sole oil
and fiber crop attested. Cotton may appear as early as the Hellenistic (though based on a single seed that may be
intrusive from A~van Kale; Nesbitt et al. 201 7) but becomes more prevalent than flax during the two Medieval
periods.
Fruits
The fruit assemblage is far more diverse than the cereal, pulse, and oil and fiber crop assemblages. At least 30
distinct species, and probably more, are attested in Anatolia, including both wild and domesticated taxa. While
temporal trends in rare taxa are difficult to identify, given their diversity, more numerous taxa show clear trends (Fig.
26-10). Grape, presumably wild, appears at single sites in the Epipalaeolithic and Pre-Pottery Neolithic, dropping out
l=Crataegus sp. 17=Prunus spinosa

2aMa/us/Pyrus lBaPunica granatum
3=Quercus sp. 19=Rhus coriaria
4=Rosa sp. 20=P. cerasus/avium
5=5ambucus sp. 21=Frangu/a a/nus
6=Cornus mas 22=Morus sp.
7=Prunus sp. 23=Persica vulgaris
B=Rubus sp. 24=Armeniaca vu/garis
9=Corylus sp.
lO=Prunus cerasifera
11=Cotoneaster sp.
12=Sorbus sp.
13=Cucurbitaceae s.l.
14=Juglans regia
lS=Pinus spp.
16aPrunus domestico
Hackberry
(Celtis sp)
□ Almond
(Amygdalus sp.)
□ Almond/Plum
(Amygdalus/Prunus)
■ Pistachio
(Pistacia sp)
■ Olive
(Olea europaea)
□ Commonfig
(Ficus carica)
■ Grape
(Vitis vinifero)
Epipal. PPN Neol. E-Ch. M-Ch. L-Ch. EBA MBA LBA IA Hell. Rom. Med. I Med. II
Figure 26-10. Relative abundance of fruit taxa based on summed RI scores by period.
of use in the Neolithic, and rising again through the Chalcolithic (presumably now as domesticated grape) to become
the dominant fruit from the Late Chalcolithic on. Fig is consistently the second-most prevalent fruit from the
Neolithic on. Olive is more sporadic in its presence, with spikes of intensity in the Early Bronze Age, Late Bronze
Age, Iron Age, and Roman periods.
Other major fruit taxa are ambiguously wild or cultivated. Pistachio, almond, and hackberry form important
components of the fruit assemblage from the Epipalaeolithic through the Early Chalcolithic; the Middle Chalcolithic is
dominated by pistachio alone (though found at only three of seven total sites; Fig. 26-5). All of these taxa nearly
disappear from the fruit assemblage by the Early Bronze Age; the apparent minor resurgence of pistachio in the
Hellenistic is based only on finds from two of eight total sites for the period (Fig. 26-5). Notable is the relative dearth
of evidence for the widespread cultivation of non-native fruit trees. Pomegranate appears in the Middle Bronze Age
but is attested only at two sites; cherries first appear in the Iron Age, followed by mulberry in the Hellenistic, peach in
the Roman, and apricot in the Medieval I period. The Roman and Medieval I periods include the most diverse array of
fruit taxa, although many of these taxa appear only in a single site, and often then with only a few finds, for each
period when attested.
Regional trends in fruit use can only be established for the main cultivated fruit taxa: grape, fig, and olive. The
scant data for fruit use in eastern and Northern Anatolia, with very few finds from those regions, limit regional
differentiation to central and southeastern Anatolia, as well as the circum-Mediterranean region. The limited number
of finds per site render ubiquity, the proportion of sites in which a taxon appears, a better metric for comparison than
summed RI values (Fig. 26-11). Clear regional differences exist for fig and olive use: both taxa are rarely attested in
central Anatolia, but when they do appear, they are typically present only at one site and attested by fewer than 10
specimens. Olive is more common in the circum-Mediterranean region than southeastern Anatolia, although in the
Early Bronze Age we see a substantial expansion of both olive and grape cultivation in the region for the first time.
Grape is the most ubiquitous fruit in all three regions and is often attested at nearly every site in a region from the
Middle Chalcolithic through the Medieval. The regional difference, however, is that many more sites in the circum-
Mediterranean region than in southeastern Anatolia, and in southeastern Anatolia compared to central Anatolia,
contain ten or more grape seeds. Additionally, only one site in central Anatolia (Can Hasan I; Stroud 2016) includes
evidence for grape cultivation prior to the Early Bronze Age.
Gra pe (% of sites) Common Fig (% of sites) Olive (% of sites)
Epipal.
tl
C-Anat. 1
SE-Anal.
No Data No Data No Data
PPN
C-Anat.
SE-Anal. 7
n·
Gire.Med . 2
n
Gire.Med . 1
Neolithic
C-Anat.
SE-Anal. 2
n
Gire.Med . 3
E-Chalc. ....................
C-Anat. No Data No Data No Data
SE-Anal. 0
- - -t=====================:::::;
M-Chalc.
C-Anat.
SE-Anal. 3
n
Gire.Med . 2
Gire.Med . 2
L-Chalc.
C-Anat. 2
SE-Anal. tl
Gire.Med . 5
EBA
tl
C-An at. 5
SE-Anal.
Gire.Med . 7
MBA
tl
C-Anat. 4
SE-Anal.
Gire.Med . 1
LBA
tl
C-Anat. 3
SE-Anal.
Iron Age
C-Anat.
SE-Anal. 4
n
Gire.Med . 7
Hell./Rom.
C-Anat.
SE-Anal. 4
n
Gire.Med . 7
n
Gire .Med . 4 >1 >10 finds
Medieval D■ Grape
C-Anat. D LJ Common fig
SE-Anal. 5 D■ Olive
Gire.Med . 9
20 40 60 60 100% 60 BO 100%
Figure 26-11. Ubiquity of major fruit taxa for three macro-regions where fruits are best attested.
Mediterranean, and Transitional Mediterranean regions. Numbers following macro-region
names are the number of sites with botanical assemblages for that period.
DISCUSSION
Distribution of Quantitative Archaeobotanical Data in Anatolia
This study has illuminated that despite including 145 chronologically distinct quantitative archaeobotanical
datasets from 86 sites across Turkey, sampling for any one regional or temporal slice of the Holocene of Anatolia is
likely to be poor. The best represented period is the Early Bronze Age, with 21 assemblages dating to this period
(totaling more than 727 samples), while the early Chalcolithic includes only three (but these total 480 samples); for
this reason, we present both number of sites and number of samples in Fig. 26-3. As indicated by Table 26-1, where
many cells are empty, some combinations of period and region have no published quantitative archaeobotanical
records. This limits our resolution to detect differences between regional and temporal trends in plant use across
Anatolia. Some areas are well sampled, however, especially central, eastern, and southeastern Anatolia, and the
Mediterranean region from the Early Bronze Age on. Trends identified for these regions are thus more likely to be
representations of real patterns in use, rather than sampling errors.
As shown in ESM 26-1, at least as many reports exist that were excluded from this report as those that were
included. Excluded reports most often failed to meet our criteria for inclusion because they published only qualitative
or semi-quantitative results. An alternative approach that uses only qualitative metrics should enable broader coverage,
at the expense of the quantitative metrics employed here, and potentially expose additional patterns in the data
available to date. We are fully aware that we exclude sites where ongoing research has not yet resulted in publications,
so this picture will continue to improve. We encourage future archaeological projects, especially in periods and
regions underrepresented in our survey, to conduct archaeobotanical research and ensure that full quantitative data are
promptly published.
Diachronic Trends in Plant Use
Several well-supported diachronic trends in plant use are evident in our dataset. The first is the replacement of wild
fruit and nut resources with domesticated fruit and nut trees. We identify a gradual decline in the use of wild fruit and
nut resources in the Early Holocene, with hackberry, almond, and pistachio dominating the fruit assemblage in the
Epipalaeolithic and Pre-Pottery Neolithic (Baird et al. 2018; Willcox et al. 2008) but beginning to diminish as the use
of domesticated figs increases beginning in the Neolithic and as grape increases substantially over the Chalcolithic
(Figs. 26-5 and 26-10; Fuller and Stevens 2019). This is likely domesticated grape based on its introduction to and
consistent cultivation in central Anatolia, which is outside the expected natural range of grape (Zohary et al. 2012:
123). By the Early Bronze Age, the three major domesticated fruits (grape, fig, and olive) comprise a majority of the
summed RI values (Fig. 26-10), and by the Late Bronze Age only two definitively wild fruit resources are attested
among the 16 assemblages analyzed (Fig. 26-5). The diversity of domesticated tree crops increases substantially over
the Late Holocene, from only three definitively domesticated fruit taxa in the Early Bronze Age to six in the Iron Age
and nine in the Roman era. More generously, considering genera that are ambiguously domesticated or wild (e.g.,
almond, pistachio, and hazel) as at least cultivated, if not domesticated, at least 14 fruit and nut trees were utilized
during the Roman period. This evidence points to an expansion of arboriculture and fruit-growing in the later
protohistoric and historic periods, a trend already well documented in palynological sequences (e.g., Roberts 2018)
and more recently supported also by wood charcoal analysis (Castellano 2021).
Changes in cereal agriculture over time include a gradually diminished role of hulled wheats in favor of hulled
barley (by the Middle Chalcolithic) and free-threshing wheats (by the Late Bronze Age and especially in the Roman
and Medieval periods) (Fig. 26-6; see also Nesbitt and Samuel 1996b; Ula;, and Fiorentino 2020). The rise in
"alternative" cereals, primarily millet and rye, beginning in the Iron Age, builds on early identifications of this
phenomenon in single assemblages (e.g., Miller et al. 2016; Nesbitt and Summers 1988). The pattern observed in
naked barley cultivation, with a rapid rise in use (through the Neolithic) and subsequent rapid near abandonment (by
the Late Chalcolithic; confirming the supra-regional reconstruction provided by Lister and Jones 2013), is another
notable "experiment" in cereal cultivation without the longevity of hulled barley and free-threshing wheat cultivation,
which by the Roman period form a near duopoly in cereal cultivation.
Oil and fiber crops resemble the pattern of cereal cultivation in that several crops become economically significant
at single sites and then fall out of use. Opium poppy is used only in the Epipaleolithic and Pre-Pottery Neolithic at
Kortik Hoyi.lk, and flixweed at <;atalhoyi.lk from the Pre-Pottery Neolithic to the Early Chalcolithic. Gold-of-pleasure
(Camelina saliva) appears in significant quantities at multiple sites from the Late Chalcolithic through the Iron Age,
with only singular appearances thereafter (Fig. 26-4). Both safflower and sesame appear only between the Early
Bronze Age and the Hellenistic. Flax is the only constant, found in every period from the Pre-Pottery Neolithic on,
save the Roman period, which is relatively poorly sampled in comparison to other periods, so this may be a sample-
size effect. Only in the Medieval periods does another crop rival flax in ubiquity and prevalence, with cotton
becoming the most attested oil and fiber crop. Notably, however, cotton is restricted to eastern and southeastern
Anatolian sites, save its appearance in a singular assemblage from the Medieval II period at Gordion in central
Anatolia (Marston 2017; Miller 2010), unlike the widespread cultivation of flax.
In contrast to fruit and cereal cultivation, where clear diachronic trends are evident, pulse cultivation in Anatolia is
notable for its lack of a temporal trend. While some transitions between periods result in apparently substantial swings
in pulse crop choices, such as the Roman to Medieval I (Fig. 26-8), no sustained diachronic trends are evident. The
relative balance among vetches, lentils, and some combination of pea, grass pea, and chickpea is sustained over time;
for example, the Pre-Pottery Neolithic pulse assemblage closely resembles that of the Medieval II period, in marked
contrast with the cereal and fruit assemblages of those periods (Fig. 26-8).
Our data provide a terminus ante quem for the introduction of East and South Asian crops to Anatolia. While
singular finds of such crops may have been excluded from this database, due to their reported presence in qualitative
or semi-quantitative archaeobotanical reports only, for situations with numbers of seeds sufficient to indicate
cultivation (and not potentially intrusive from later periods, such as the single Hellenistic cotton seed) our data provide
a reliable latest date for the introduction of these crops. Notably, the pattern that emerges is of gradual or periodic
introductions, rather than the coordinated arrival of agricultural "packages" (Marston 2021 ), a pattern also revealed in
recent supra-regional studies (Boivin et al. 2012; Fuller and Stevens 2019). We identify broomcorn millet as the
earliest East Asian domesticate in Anatolia, with a first attested appearance in the Middle Chalcolithic that consists
only of two seeds at one site (Mersin-Yumuktepe; Fiorentino et al. 2014), so we consider it to be of questionable
reliability, but in consistent though limited use from the Middle Bronze Age on, at the same time when AMS-dated
broomcorn millet first appears in Europe (in the North Pantie region, Filipovic et al. 2020). It is the same period when
the first South Asia domesticate, sesame, also appears in our dataset (Fig. 26-4). Foxtail millet, another East Asian
domesticate, follows by the Late Bronze Age. The pace of introductions picks up during the Hellenistic, with mulberry
(East Asian) first attested. More East Asian tree crops arrive in the Roman (peach) and Medieval I periods (apricot),
alongside definite evidence for South Asian cotton in the Medieval I period; rice appears only in the Medieval II
period (Figs. 26-4 and 26-5).
A final temporal pattern is the identification of a marked outlier: the Roman period. While seven sites document
quantitative Roman assemblages and the total sample count is 206, both lower than average numbers across the
temporal sequence in this dataset (Fig. 26-3), these span five regions of Anatolia, so are regionally diverse. The
Roman period is an outlier in every domain of agriculture. Cereal cultivation contracted substantially, with hulled
barley and free-threshing wheats comprising nearly the entire summed RI total, with a marked rise in free-threshing
wheat use in particular compared to all earlier periods (Fig. 26-6); rye and millet are low-frequency finds, and emmer
is the only other cereal attested. The pulse assemblage is diverse but dominated by lentil to a degree unprecedented in
earlier periods (Fig. 26-8). Fruit cultivation is also diverse, with grape especially ubiquitous (found at 5 of? sites; Fig.
26-5). Notably, evidence for oil seed cultivation is entirely absent from Roman assemblages, perhaps related to the rise
in olive production as a relative proportion of the summed RI values (Figs. 26-5 and 26-10). Together, these changes
paint a picture of a substantial shift in agricultural effort and attention during the Roman period in ways that differ
qualitatively from earlier periods, a phenomenon that has been previously identified in both Roman plant (Marston
and Miller 2014) and animal assemblages (<;akular and Marston 2019), albeit at the scale of the site rather than all of
Anatolia. Whether such a shift is a facet of Roman taxation policies (Marston 2012) or military provisioning systems
(Bennett 2013) is a complex question that requires additional research on this relatively understudied period, at least
from the archaeobotanical perspective.
Regional Differences in Agriculture
Our aggregated data distinguish two macro-regional trends in agricultural patterns. One is that of central and
eastern Anatolia, and the other that of the circum-Mediterranean macro-region and southeastern Anatolia. That of
central and eastern Anatolia is indicated by similar patterns in cereal cultivation through the entire time sequence, as
well as in pulse agriculture, for the fewer periods during which robust assemblages are available from both areas:
namely the Chalcolithic, Early Bronze Age, and Iron Age through Medieval. The cereal signature of this pair of
regions includes a greater importance of millet from the Iron Age through the Medieval (Fig. 26-7). Pulse cultivation
in these two regions differs from other macro-regions of Anatolia in the minimal presence of broad bean (save during
the Late Bronze Age) and a greater emphasis on bitter vetch and lentil versus pea, grass pea, and chickpea (save
during the Early Bronze Age) (Fig. 26-9). Central Anatolia additionally lacks evidence for olive or fig cultivation after
the Neolithic, save for a few chance finds from later periods (Fig. 26-11).
Southeastern Anatolia more closely resembles the circum-Mediterranean sites than those of central and eastern
Anatolia, although perhaps we can best understand these macro-regions on a continuum, with the circum-
Mediterranean at one extreme of the spectrum and southeastern Anatolia in an intermediate position. From the Late
Bronze Age on, both southeastern Anatolia and the circum-Mediterranean region indicate a greater emphasis on the
combination of pea, grass pea, and chickpea in comparison to central and eastern Anatolia (Fig. 26-9). They are even
more distinct in the fruit assemblage, with olive nearly exclusive to these two regions and a much greater frequency of
both fig and grape use than in central Anatolia (Fig. 26-11). We do, however, identify a consistent difference of degree
in fruit cultivation between the two macro-regions, with the circum-Mediterranean consistently indicating a greater
frequency of grape, fig, and olive use, as well as more evidence for substantial use of those resources in nearly every
period (Fig. 26-11). This trend is most marked for olive, which is exclusively found in circum-Mediterranean sites
during four separate periods; additionally, only during the Hellenistic/Roman period is a substantial (> 10) number of
olive endocarps present at a site (and even then, only a single site) outside of the circum-Mediterranean macro-region
(Fig. 26-11 ).
The circum-Mediterranean differs from southeastern Anatolia in one more dimension. After hulled wheats begin to
decline rapidly across all of Anatolia from the Middle Bronze Age on, the circum-Mediterranean data consistently
indicate a greater importance of hulled wheat cultivation and a greater diversity of those wheats at circum-
Mediterranean sites compared to other regions, especially southeastern Anatolia. This trend has been previously
identified as a notable feature that distinguishes Aegean and Near Eastern archaeobotanical assemblages, at least
during the Late Bronze and Iron Ages (Riehl and Nesbitt 2003).
The reason for most of the differences outlined above is likely to be climatic, at least to a large extent. The circum-
Mediterranean macro-region is unique in that it shares a Mediterranean climate, with cool but rarely freezing
temperatures in the winter and higher precipitation than areas further from the coast (Fig. 26-2). Northern Anatolia is
wet but also cold; unfortunately, it is also the most poorly represented region in our study, so it is difficult to link its
climate with agricultural preferences. Central and eastern Anatolia are the driest regions, where the emphasis on pulses
with lower moisture requirements (bitter vetch and lentil; Riehl 2009: 98) is evident, and cold, precluding olive and fig
cultivation on a wide scale (England et al. 2008). Some regional patterns, such as the near-total restriction of cotton to
southeastern Anatolia, may have more to do with local political and economic networks that encouraged such
cultivation, though the summer heat of this region may also play a role (Brite and Marston 2013; Watson 1983). A
more fine-grained analysis than possible here is needed to identify such patterns among additional crops and to trace
the forces behind them.
CONCLUSIONS
This chapter summarizes, in broad strokes, the current state of the field in the archaeobotany of Anatolia. Most
critically, we provide a full inventory and bibliography for all published archaeobotanical reports of Anatolia,
representing the first comprehensive picture of archaeobotany in Anatolia in 25 years (see ESM 26-1 for full inventory
and bibliography). By using only fully quantitative data, we ensure maximum resolution on plant-use practices, at
least as understood at the time of publication, and maximum comparability across archaeobotanical assemblages. We
integrate the data along regional boundaries tied to climate and by chronological periods as generally applied to the
archaeology of Anatolia. The Representativeness Index (RI) metric aims to enhance comparability among sites with
different sampling intensities and to more accurately represent a broad picture of agricultural practices on a regional
scale.
Through this approach we identify several temporal and regional trends that are well supported, as well as others
that we suggest more tentatively due to limitations in the number of reported samples or taxonomic resolution among
the incorporated assemblages. We describe well-known diachronic trends in the decline in wild fruit use and
replacement by domesticated fruit trees and vines, and in the replacement of hulled wheats by hulled barley and free-
threshing wheats. We also illuminate lesser-discussed facets of shorter duration or more limited regional scope,
including the rise and fall of naked barley cultivation, the pacing of the appearance of East and South Asian
domesticates, the sequence of alternative oil seed cultivation, and the degree to which olive, fig, and grape cultivation
differ on a regional scale.
Despite these contributions, the richness and diversity, but also unevenness, of the assembled information suggests
multiple avenues for further research into this database. Smaller-scale regional differences may have been overlooked
in our groupings, so we intend further investigation at that scale. The RI approach is necessarily a coarse one,
suggesting that further improvements of the method may yield additional insights. Pairing RI data with ubiquity and
raw counts may illuminate additional patterning in the data.
Nonetheless, our study demonstrates that the Anatolian agricultural system was not static over either time or space.
Following the transition to agricultural economies, these economies continued to evolve in ways that we do not yet
fully understand, which require the more systematic inclusion of archaeobotany in archaeological projects. As we
docun1ent here, these processes are not only qualitative (as might be the case for domestication, or crop introductions)
but quantitative, requiring large datasets in order to detect and understand such patterns. We then will be able to
address more effectively questions about the political, economic, and ecological processes underneath these dynamics.
Under whose agency do changes in agricultural practices take place? With what pace? To what extent do local
communities have autonomy to adopt their own agricultural trajectories? Once again, these questions point to the need
to incorporate archaeobotany in a more cohesive way within archaeological and historical research. In particular, we
hope this chapter sheds additional light on which regions and periods are the most underrepresented and encourages
the archaeobotany of those times and places in the future.
REFERENCES CITED
Atalay, ibrahim. 2014. Turkiye'nin Ekolojik Bolgeleri - Ecoregions of Turkey. izmir: Meta Press.
- . 2018. Vegetation. In The Soils of Turkey, S. Kapur, E. Ak9a, and H. Gi.inal, eds., 15-24. Cham: Springer.
Baird, Douglas, Andrew Fairbairn, Emma Jenkins, Louise Martin, Caroline Middleton, Jessica Pearson, Eleni Asouti,
Yvonne Edwards, Ceren Kabukcu, Gokhan Mustafaoglu, Nerissa Russell, Ofer Bar-Yosef, Geraldine Jacobsen,
Xiaohong Wu, Ambroise Baker, and Sarah Elliott. 2018. Agricultural Origins on the Anatolian Plateau.
Proceedings of the National Academy ofSciences 115: E3077.
Bennett, Julian. 2013. Agricultural Strategies and the Roman Military in Central Anatolia During the Early Imperial
Period. OLEA 21: 315-343.
Boivin, Nicole, Dorian Q. Fuller, and Alison Crowther. 2012. Old World Globalization and the Columbian Exchange:
Comparison and Contrast. World Archaeology 44: 452-469.
Bouchaud, Charlene, Christiane Jacquat, and Daniele Martinoli. 2017. Landscape Use and Fruit Cultivation in Petra
(Jordan) from Early Nabataean to Byzantine Times (2nd Century BC- 5th Century AD). Vegetation History and
Archaeobotany 26: 223-244.
Brite, Elizabeth Baker and John M. Marston. 2013. Environmental Change, Agricultural Innovation, and the Spread of
Cotton Agriculture in the Old World. Journal ofAnthropological Archaeology 32: 39- 53.
<;akrrlar, Canan and John M. Marston. 2019. Rural Agricultural Economies and Military Provisioning at Roman
Gordian (Central Turkey). Environmental Archaeology 24: 91- 105.
Casana, Jesse. 2008. Mediterranean Valleys Revisited: Linking Soil Erosion, Land Use and Climate Variability in the
Northern Levant. Geomorphology 101: 429-442.
Castellano, Lorenzo. 2021. A New Anthracological Sequence from Nigde-K1mk Hoyiik (Turkey): Woodland
Vegetation and Arboriculture in Southern Cappadocia from the Late Bronze Age to the Ottoman Period.
Archaeological and Anthropological Sciences 13: 49.
<;iyek, ihsan. 2003. The Statistical Analysis of Precipitation in Ankara, Turkey. Fzrat Oniversitesi Sosyal Bilimler
Dergisi 13: 1- 20.
Davis, Peter H. 1965-2000. Flora of Turkey and the East Aegean Islands. Edinburgh: Edinburgh University Press.
Effenberger, Hemike. 2018. The Plant Economy of the Northern European Bronze Age-More Diversity through
Increased Trade with Southern Regions. Vegetation History and Archaeobotany 27: 65-74.
England, Ann, Warren J. Eastwood, C. Neil Roberts, Rebecca Turner, and John F. Haldon. 2008. Historical Landscape
Change in Cappadocia (Central Turkey): A Palaeoecological Investigation of Annually Laminated Sediments from
Nar Lake. The Holocene 18: 1229- 1245.
Fairbairn, Andrew, Daniele Martinoli, Ann Butler, and Gordon Hillman. 2007. Wild Plant Seed Storage at Neolithic
<;atalhoyiik East, Turkey. Vegetation History and Archaeobotany 16: 467-479.
Fairbairn, Andrew S. 2021. Agriculture in the Chalcolithic and Bronze Age of Asia Minor. In A Companion to Ancient
Agriculture, D. Hollander and T. Howe, eds., 211- 239. Hoboken: John Wiley & Sons, Inc.
Fick, Stephen E and Robert J. Hijmans. 2017. Worldclim 2: New 1-Km Spatial Resolution Climate Surfaces for
Global Land Areas. International Journal of Climatology 3 7: 4302-4315.
Filipovic, Dragana. 2014. Early Farming in Central Anatolia: An Archaeobotanical Study of Crop Husbandry, Animal
Diet and Land Use at Neolithic <;atalhoyiik. Oxford: Archaeopress.
Filipovic, Dragana, John Meadows, Marta Dal Corso, Wiebke Kirleis, Almuth Alsleben, Orni Akeret, Felix Bittmann,
Giovanna Bosi, Beatrice Ciuta, Dagmar Dreslerova, Hemike Effenberger, Ferenc Gyulai, Andreas G. Heiss,
Monika Hellmund, Susanne Jahns, Thorsten Jakobitsch, Magda Kapcia, Stefanie Klool3, Marianne Kohler-
Schneider, Helmut Kroll, Przemyslaw Makarowicz, Elena Marinova, Tanja Markle, Aleksandar Medovic, Anna
Maria Mercuri, Aldana Mueller-Bieniek, Renato Nisbet, Galina Pashkevich, Renata Perego, Petr Pokorny, Lukasz
Pospieszny, Marcin Przybyla, Kelly Reed, Joanna Rennwanz, Hans-Peter Stika, Astrid Stobbe, Tjasa Tolar,
Krystyna Wasylikowa, Julian Wiethold, and Tanja Zerl. 2020. New Ams 14C Dates Track the Arrival and Spread
ofBroomcorn Millet Cultivation and Agricultural Change in Prehistoric Europe. Scientific Reports 10: 13698.
Fiorentino, Girolamo, Milena Primavera, and Valentina Caracuta. 2014. Archaeobotanical Analysis at Mersin-
Yumuktepe: Food Habits from Neolithic to Medieval Period. In Paleonutrition and Food Practices in the Ancient
Near East: Towards a Multidisciplinary Approach, L. Milano, ed. 85-94. Padova: Sargon Editrice e Libreria.
Fuller, Dorian Q and Chris J. Stevens. 2019. Between Domestication and Civilization: The Role of Agriculture and
Arboriculture in the Emergence of the First Urban Societies. Vegetation History and Archaeobotany 28: 263-282.
Galili, E., D. Langgut, J.P. Terral, 0. Barazani, A. Dag, L. Kolska Horwitz, I. Ogloblin Ramirez, B. Rosen, M.
Weinstein-Evron, S. Chaim, E. Kremer, S. Lev-Yadun, E. Boaretto, Z. Ben-Barak-Zelas, and A. Fishman. 2021.
Early Production of Table Olives at a Mid-7th Millennium BP Submerged Site Off the Carmel Coast (Israel).
Scientific Reports 11: 2218.
Heiss, Andreas G. and Hans-Peter Stika. 2013. Making the Incomparable Comparable? A New Attempt of the Semi-
Quantitative Evaluation of Large-Scale Archaeobotanical Data. Paper presented at the 16th Syn1posium of the
International Work Group for Palaeoethnobotany, Thessaloniki.
Kaniewski, David, E. Paulissen, V. De Laet, K. Dossche, and Marc Waelkens. 2007. A High-Resolution Late
Holocene Landscape Ecological History Inferred from an Intramontane Basin in the Western Taurus Mountains,
Turkey. Quaternary Science Reviews 26: 2201-2218.
Lister, Diane L. and Martin K. Jones. 2013. Is Naked Barley an Eastern or a Western Crop? The Combined Evidence
of Archaeobotany and Genetics. Vegetation History and Archaeobotany 22: 439-446.
Marston, John M. 2012. Agricultural Strategies and Political Economy in Ancient Anatolia. American Journal of
Archaeology 116: 377-403.
-. 2014. Ratios and Simple Statistics in Paleoethnobotanical Analysis: Data Exploration and Hypothesis Testing. In
Method and Theory in Paleoethnobotany, J.M. Marston, J. d'Alpoim Guedes, and C. Warinner, eds., 163- 179.
Boulder: University Press of Colorado.
-. 2017. Agricultural Sustainability and Environmental Change at Ancient Gordian. Philadelphia: University of
Pennsylvania Museum Press.
-. 2021. Archaeological Approaches to Agricultural Economies. Journal of Archaeological Research online before
print.
Marston, John M. and Lorenzo Castellano. 2021. Archaeobotanical Bibliography for Anatolia. https://core.tdar.org.
Accessed February 27, 2021.
Marston, John M. and Naomi F. Miller. 2014. Intensive Agriculture and Land Use at Roman Gordion, Central Turkey.
Vegetation History and Archaeobotany 23: 761- 773.
Miller, Naomi F. 1998. Patterns of Agriculture and Land Use at Medieval Gritille. In The Archaeology of the Frontier
in the Medieval Near East: Excavations at Gritille, Turkey, S. Redford, ed. 211- 252. Philadelphia: University of
Pennsylvania Museum of Archaeology and Anthropology.
-. 2010. Botanical Aspects of Environment and Economy at Gordian, Turkey. Philadelphia: University of
Pennsylvania Museum of Archaeology and Anthropology.
Miller, Naomi F., Robert N. Spengler, and Michael Frachetti. 2016. Millet Cultivation across Eurasia: Origins, Spread,
and the Influence of Seasonal Climate. The Holocene 26: 1566- 1575.
Nesbitt, Mark. 2016. The Plant Remains. In Tille Hoyuk 3.2: The Iron Age: Pottery, Objects and Conclusions, S.
Blaylock, ed., 369- 394. Ankara: The British Institute at Ankara.
Nesbitt, Mark, Jennifer Bates, Gordon Hillman, and Stephen Mitchell. 2017. The Archaeobotany of A~van:
Environment and Cultivation in Eastern Anatolia from the Cha/eolithic to the Medieval Period. London: British
Institute at Ankara.
Nesbitt, Mark and Delwen Samuel. 1996a. Archaeobotany in Turkey: A Review of Current Research. Orient-Express
1996: 91- 96.
-. 1996b. From Staple Crop to Extinction? The Archaeology and History of the Hulled Wheats. In Hulled Wheats:
Proceedings of the First International Workshop on Hulled Wheats, 21-22 July 1995, Castelvecchio Pascoli,
Tuscany, Italy, S. Padulosi, ed. 40-99. Rome: International Plant Genetic Resources Institute.
Nesbitt, Mark and Gordon D. Summers. 1988. Some Recent Discoveries of Millet (Panicum miliaceum L. and Setaria
italica (L.) P. Beauv.) at Excavations in Turkey and Iran. Anatolian Studies 38: 85- 97.
Pearsall, Deborah M. 2015. Paleoethnobotany: A Handbook ofProcedures, 3rd edition. Walnut Creek, CA: Left Coast
Press.
Popper, Virginia S. 1988. Selecting Quantitative Measurements in Paleoethnobotany. In Current Paleoethnobotany:
Analytical Methods and Cultural Interpretations ofArchaeological Plant Remains, C.A. Hastorf and V.S. Popper,
eds., 53- 71. Chicago: University of Chicago Press.
Riehl, Simone. 2009. Archaeobotanical Evidence for the Interrelationship of Agricultural Decision-Making and
Climate Change in the Ancient Near East. Quaternary International 197: 93-114.
-. 2014. Changes in Crop Production in Anatolia from the Neolithic Period until the End of the Early Bronze Age. In
Prehistoric Economies of Anatolia: Subsistence Strategies and Exchange, C. Wawruschka, ed. 59-71. Rahden,
Gennany: Verlag Marie LeidorfGmbH.
- . 2021. Archaeobotanical Database of Eastern Mediterranean and Near Eastern Sites (Ademnes).
http://www.ademnes.de/index.php. Accessed February 27, 2021.
Riehl, Simone and Mark Nesbitt. 2003. Crops and Cultivation in the Iron Age Near East: Change or Continuity? In
Identifying Changes: The Transition from Bronze Age to Iron Ages in Anatolia and Its Neighbouring Regions:
Proceedings of the International Workshop, Istanbul, November 8-9, 2002, B. Fischer, H. Genz, E. Jean, and K.
Koroglu, eds., 301- 312. istanbul: Turk Eskil;:ag Bilimleri Enstitusu.
Roberts, Neil. 2018. Revisiting the Bey~ehir Occupation Phase: Land-Cover Change and the Rural Economy in the
Eastern Mediteuanean During the First Millennium AD. Late Antique Archaeology 11: 53-68.
Roberts, Neil, Samantha L. Allcock, Fabien Arnaud, Jonathan R. Dean, Warren J. Eastwood, Matthew D. Jones,
Melanie J. Leng, Sarah E. Metcalfe, Emmanuel Malet, Jessie Woodbridge, and Hakan Yigitb~10glu. 2016. A Tale
of Two Lakes: A Multi-Proxy Comparison of Lateglacial and Holocene Environmental Change in Cappadocia,
Turkey. Journal ofQuaternary Science 31: 348-362.
Rossner, Corinna, Katleen Deckers, Marion Benz, Vecihi Ozkaya, and Simone Riehl. 2018. Subsistence Strategies and
Vegetation Development at Aceramic Neolithic Kortik Tepe, Southeastern Anatolia, Turkey. Vegetation History
and Archaeobotany 27: 15-29.
Stein, Gil J., Reinhard Bernbeck, Cheryl Coursey, Augusta McMahon, Naomi F. Miller, Adnan M1su, Jeffrey Nicola,
Holly Pittman, Susan Pollock, and Henry Wright. 1996. Uruk Colonies and Anatolian Communities: An Interim
Report on the 1992- 1993 Excavations at Hacmebi, Turkey. American Journal ofArchaeology 100(2): 205-260.
Stevens, Chris J. 2014. Intersite Variation within Archaeobotanical Charred Assemblages: A Case Study Exploring the
Social Organization of Agricultural Husbandry in Iron Age and Roman Britain. In Method and Theory in
Paleoethnobotany, J.M. Marston, J. d' Alpoim Guedes, and C. Warinner, eds., 235-255. Boulder: University Press
of Colorado.
Stika, Hans-Peter and Andreas G. Heiss. 2013. Plant Cultivation in the Bronze Age. In The Oxford Handbook of the
European Bronze Age, Harry Fokkens and Anthony Harding, eds., 348- 369. Oxford: Oxford University Press.
Stroud, Elizabeth. 2016. An Archaeobotanical Investigation into the Chalcolithic Economy and Social Organization of
Central Anatolia. Ph.D. thesis, University of Oxford.
Turke~, Murat. 2003. Spatial and Temporal Variations in Precipitation and Aridity Index Series of Turkey. In
Mediterranean Climate-Variability and Trends, H.-J. Nolle, ed., 181-213. Heidelberg: Springer Verlag.
Tururn;:oglu, Ufuk Utku, Murat Turke~, Deniz Bozkurt, Bari~ Ono!, Omer Liltfi ~en, and Hasan Nilzhet Dalfes. 2018.
Climate. In The Soils of Turkey. S. Kapur, E. Ak9a, and H. Gi.inal, eds., 25-44. Cham: Springer.
Ula~, Burhan and Girolamo Fiorentino. 2020. Recent Attestations of "New" Glume Wheat in Turkey: A Reassessment
ofits Role in the Reconstruction of Neolithic Agriculture. Vegetation History and Archaeobotany 30: 685-701.
Vennoere, Marleen, Leo Vanhecke, Marc Waelkens, and Erik Smets. 2003. Modem and Ancient Olive Stands near
Sagalassos (South-West Turkey) and Reconstruction of the Ancient Agricultural Landscape in Two Valleys.
Global Ecology and Biogeography 12: 217- 236.
Watson, Andrew M. 1983. Agricultural Innovation in the Early Islamic World: The Diffusion of Crops and Farming
Techniques, 700-1100. Cambridge: Cambridge University Press.
Willcox, George, Sandra Fomite, and Linda Herveux. 2008. Early Holocene Cultivation before Domestication in
Northern Syria. Vegetation History and Archaeobotany 17: 313-325.
Zohary, Daniel, Maria Hopf, and Ehud Weiss. 2012. Domestication of Plants in the Old World· The Origin and
Spread of Domesticated Plants in Southwest Asia, Europe, and the Mediterranean Basin, 4th edition. Oxford:
Oxford University Press.
Zohary, Michael. 1973. Geobotanical Foundations of the Middle East. Stuttgart: G. Fischer.
CHAPTER TWENTY-SEVEN
FROM BONES TO GENOMES: CURRENT RESEARCH ON THE POPULATION

HISTORY OF PREHISTORIC ANATOLIA
DUYGU DENiZ KAzANCI, N. EZGi ALTINI$IK, AYc;A AYD0GAN,

FOSUN OZER, ELiF SURER, yiGDEM ATAKUMAN'
MEHMET S0MEL, * AND YILMAZ SELiM ERDAL **
In this chapter, we provide an overview of current knowledge of human population history in Anatolia, and more
specifically, the demographic history of the region with regards to migration and the admixture processes across time.
Although the earliest evidence of the human presence in Anatolia dates back to the Palaeolithic period (Gulei;: 1999;
T~klran 2008), the scope of our review will be limited to a time period from the Final Pleistocene to the late
Holocene, specifically focusing on the Holocene, when human population sizes and activity increased dramatically,
driven by the Neolithic transition.
For over a century, archaeological and historical studies in Anatolia have yielded rich accounts of social
transformation over the last 15,000 years. These include major shifts in subsistence, culture, and language (Sagona
and Zimansky 2015). As in many cases in human history, a persistent question has been the degree to which these
cultural changes in Anatolia were driven by human movement, instead of pure cultural interaction, i.e., sharing or
diffusion of cultures and technology. Although the two processes are not mutually exclusive, identifying their relative
contributions can significantly deepen our understanding of the dynamics of social transition.
Our review covers results collected in different periods and from different data sources: anthropometric analysis,
molecular genetic studies on modem-day Anatolians, and recent ancient DNA and archaeogenomic work. However,
this is not intended to be an exhaustive review of the full body of such work. Instead, we focus on ancient genome
analyses, given their power in resolving past population movements. We provide several exemplary cases where
ancient genomes have been used to evaluate the possible roles of human movement in cultural change in the Neolithic
and Bronze Age periods.
ANTHROPOMETRIC STUDIES ON ANATOLIAN HISTORY:

RACE, MIGRATION, AND ENVIRONMENTAL INFLUENCES
Anthropological investigations of the remains of Anatolian people began by the end of the Ottoman Empire, in the
last quarter of the 19th century. Conducted by European scholars, these studies did not aim to interpret the
osteobiographies of past peoples; rather, their aim was to identify the "race" of the people, in parallel with the
common scientific approach of this period. Cranial studies within the racial framework continued into the early 20th
century, tackling questions related to the origins of modem-day Turks, Hittites, and Indo-European groups (see
Atakuman 2008; Dstilndag and Yaz1c1oglu 2014).
In the second half of the 20th century, anthropological research in Turkey turned to studying bioarchaeological
remains from novel angles. This included Lawrence Angel's pioneering work, which reported historical associations
between the frequency of anaemia signatures, including porotic hyperostosis, with the intensity of agriculture in
marshlands and the resultant occurrence of malaria (Angel 1966). This exemplified one of the earliest
bioarchaeological studies on human-environment interaction. Meanwhile, anthropometric analyses by Angel and
others also tackled questions about population movement. These investigations used cranial types to gain insight into
population mobility. For example, studying Early Bronze Age Karata~ human remains, Machteld Mellink and Angel
suggested that these remains possessed craniofacial similarities to skeletons from Neolithic Anatolia, and to those
from the contemporaneous East Mediterranean and the Caucasus/Iran (Mellink and Angel 1970). The studies of
Muzaffer Suleyman Senyiirek must also be mentioned as standing at the beginning of the bioarchaeological studies in
Anatolia (i.e., Senyiirek 1949, 1950; Senyiirek and Tunakan 1951). Meanwhile, Mario Cappieri, who studied cranial
measurements across 14 ancient settlements from the Neolithic to the Iron Age, reported signatures of population
isolation and endogamy, and inferred an evolutionary trend towards specific cranial types (Cappieri 1970).
* Senior contributing author.

** Senior contributing author.
356 Chapter Twenty-Seven: Kazanc1 et al.
After the 1990s, the use of cranial morphology to classify populations with respect to "race" was fully abandoned.
Instead, human remains were used to reconstruct the lifestyle of the past people. Cranial and dental characteristics
were further used as proxies for genetic distance to study individual- or population-level relatedness. Indeed, dental
similarities may serve as weak but reliable proxies for population-level genetic similarities in situations where DNA
data are missing (Irish et al. 2020). For instance, Serpil Eroglu and Y1lmaz Selim Erdal (2008) studied the frequencies
of the palatine torus trait in twelve Anatolian societies from the Early Bronze Age to the 20th century and reported a
significant increase in its frequency across time (Eroglu and Erdal 2008). The researchers suggested that this signal
could be evidence for immigration from eastern Eurasia. In another study using cranial nonmetric traits, Ricaut and
Waelkens (2008) reported affinities between the southwestern Anatolian Byzantine population of Sagalassos with
African populations, and interpreted these as evidence for post-Neolithic migrations. Eroglu (2016) also used cranial
nonmetric traits to suggest limited change in Anatolia over recent millennia, and that the Anatolian population
clustered with European populations. In a study on intrasite individual-level relatedness, Marin Pilloud and Clark
Larsen used dental nonmetric traits to suggest that <;atalhoyuk subfloor burials within the same building were not
biologically more closely related than burials in different buildings (Pilloud and Larsen 2011). They also inferred a
patrilocal tradition in <;atalhoyuk, given higher dental trait diversity among the females studied.
Meanwhile, there is growing appreciation of the various sources of variation in phenotypic traits, such as
developmental stochasticity, wear, or the subject's age, that could confound interpreting phenotypic similarities. For
example, a recent study on a cranial shape and size of examples from Arslantepe suggests that the size and shape of
the crania were affected by environmental stress such as developmental anomalies, intrauterine, and postnatal
positional conditions, as well as secondary pathological conditions (Erdal and D'Amico 2020). If the skull is
frequently shaped by intentional and unintentional modification, population-level similarities in cranial measurement
and cranial typologies would not reflect genetic relationships, but rather biocultural similarities. Accordingly, most
contemporary studies of population genetic history currently rely on DNA information, when it is available.
MOLECULAR GENETIC STUDIES ON MODERN-DAY ANATOLIA:

SIGNALS OF EASTERN EURASIAN INFLUENCE
With the growing feasibility of DNA sequencing over the last decades, DNA data have increasingly been used to
infer population structure and demographic history, including those of Anatolian populations. In the late 20th and early
21st century, research relied on genotyping a single DNA region, or a few DNA regions, and usually targeted
uniparental DNA loci such as the mitochondrial DNA (mtDNA) or the Y chromosome, or short tandem repeat (STR)
regions, due to the relative ease in genotyping such loci. These preferences changed with the availability of high-
throughput sequencing (HTS) technologies, also called next generation sequencing (NGS), since the late 2000s. These
tools provide a substantially larger wealth of data, which we explain later.
Studies Using Uniparental Genetic Markers and Nuclear Repeats
Early examples of single locus genetic studies were focused on inferring past migrations between Anatolia
and Europe. Richards and colleagues analyzed mtDNA loci from Europe and SW Asia, including Anatolia
(Richards et al. 2000). The study analysed the mtDNA control region, which comprises a relatively short ( ca.
1000 bp) segment of the full mtDNA molecule, but is more variable than the rest of the molecule, and thus more
informative about relatedness. The authors inferred migrations from the Near East to Europe, and also
migrations back to the Near East, and suggested that the latter process could represent "the expansion of Greek,
Phrygian, and Armenian speakers into western Anatolia, central Anatolia, and Armenia" (Richards et al. 2000).
It is interesting to note that ancient DNA results have not yielded support for this scenario (see below).
Another question tackled by these early studies was the relationship between Anatolia and eastern Eurasia.
Cinnioglu and colleagues (2004) genotyped Y chromosomal markers across 523 modem-day Turkish
individuals. They found Y chromosome haplogroups in the Anatolian gene pool largely shared with the Balkans
and with the Caucasus, and further estimated a ca. 9% paternal contribution from Central Asia. Another study
using 10 autosomal Alu repeat loci genotyped in 202 modem-day Anatolians estimated a similar frequency, ca.
13% (Berkman et al. 2008). Both studies argued that the high population size of Anatolia when Turkic
migrations began, estimated to be 6-7 million by the end of the Byzantine period by Russell (1958), may have
diluted the effect of gene flow from Central Asia in the Anatolian gene pool, despite leaving the clear linguistic
legacy.
Genome-wide Studies on Modern-Day Anatolia
Since uniparental markers are only inherited through maternal or paternal lineages, they provide limited
information about population demographics. In tum, studies that rely on a limited number of nuclear markers, such as
microsatellites (STRs), may suffer from sampling error due to the small sample size. For instance, polymorphisms at a
From Bones to Genomes 357
dozen nuclear loci may not provide information regarding an individual's full set of genetic ancestors (Mathieson and
Scally 2020). Using genome-wide markers across thousands of single nucleotide polymorphisms (SNPs) will instead
be significantly more powerful in representing an individual's genetic ancestry.
In 2012, Hodoglugil and Mahley (2012) genotyped 64 modem-day individuals from three cities in central and
western Anatolia on Illumina SNP-chips. The dataset, covering more than 500,000 SNPs, was compared with
published Human Genome Diversity Panel genotype data, as well as SNP-chip data the authors generated from
Central Asians. Using supervised analyses to estimate ancestry components, the authors apportioned the ancestry of
their Anatolian sample to 38% European, 35% Middle Eastern, and also with lower contributions from South Asia
(18%) and Central Asia (9%). The authors also noted the homogeneity of the Anatolian sample with respect to city of
origin.
In 2014, Alkan and colleagues reported fourteen high-quality genomes from across the peninsula, generated and
sequenced using Illumina HTS technology between 32---48x coverage (Alkan et al. 2014). The authors identified ca. 3
million SNPs (non-reference alleles) per individual, close to the Eurasian average. They also measured the nucleotide
diversity (n), which they reported to be on a par with those of European genomes. Comparing their data with
populations from the 1000 Genomes dataset (Clarke et al. 2012), the authors reported that the Anatolian sample
clusters with southern Europeans (Italy and Spain) relative to northern Europeans. They also predicted a gene flow
event related to the East Asian branch (presumably Central Asia-related) into Turkey. It is worth noting that the study
used only the 1000 Genomes 2012 dataset, which did not include other SW Asian or S Asian populations, and the
resolution of the analyses was thus limited. Like Hodoglugil and Mahley (2012), Alkan and colleagues also reported
an apparent lack of genetic structure within modern-day Anatolia (2014), with no significant correlation between
geographic and genomic distances with their sample (it is important to note that, given the limited sample sizes, subtle
population structure may have been missed by these studies).
Yunusbayev and colleagues (2015), who studied population relationships across a wide stretch of Eurasia using
SNP genotyping data, also analysed data from modem-day Turkey. Their results showed that Turkic groups in SW
Asia were genetically closer to their neighbouring regions than their linguistic counterparts in Asia, again implying
limited gene flow from an Asian source.
Overall, these genetic studies on modem-day Anatolians have consistently revealed affinities to neighbouring
populations in the eastern Mediterranean and the Caucasus, with limited gene flow from Central Asia despite strong
linguistic change. Still, insights from analyses of modem-day genetic data about demographic history are bound to
remain superficial, given the high dynamism of human populations throughout history. Ancient DNA is the most
powerful tool to overcome this constraint, as recent results now clearly demonstrate.
FIRST GENERATION ANCIENT DNA STUDIES USING TARGETED SEQUENCING
Although ancient DNA (aDNA) studies go back to the mid-1980s, significant progress was achieved only after the
development of NGS technologies in the 21 st century. In this sense, it is possible to divide the history of ancient DNA
studies into two phases, before and after NGS.
Most aDNA studies prior to NGS (i.e., the first generation) were conducted by targeting mitochondrial DNA
(mtDNA) fragments. DNA breaks down over time, and most DNA molecules in tissue are lost, and therefore the high
copy number of mtDNA per cell becomes an advantage in retrieving fragments (Higuchi et al. 1984; Paabo et al.
1988; Hagelberg and Clegg 1991; Stone and Stoneking 1993). The high sequence variability at the hypervariable
(1-NR) loci of mtDNA, relative to any similar sized nuclear locus, was also recognised as helpful in studying genetic
relatedness.
However, the nascent aDNA field soon realised a significant challenge: distinguishing between authentic ancient
DNA (endogenous aDNA) and modern contaminant human DNA. Through the 1990s, there was growing
understanding that excluding modem-day DNA contamination was a ubiquitous risk when amplifying and studying
loci using targeted sequencing. Effective anti-contamination measures were developed (such as the replication of
results in multiple laboratories), but these also acted as constraints on the proliferation of aDNA studies, especially on
human remains (Hofreiter et al. 2001; Malmstrom et al. 2005). This problem, as well as the limited amount of
information available from mtDNA for inferring population history (explained earlier), significantly hindered the field
for multiple decades.
Ancient Mitochondrial DNA Studies on Anatolian History:

Evidence/or Continuity of the Maternal Lineage
The difficulties of reliably conducting first generation aDNA is reflected in the limited number of human ancient
studies in Anatolia using targeted sequencing. One such pilot study was performed on 13 skeletons from Titri~ Hoyi.ik
from southeastern Anatolia, ca. 2300-2100 BCE (Matney et al. 2012). The authors performed a double-blind study at
two separate labs using the mtDNA HVR region and reported successful results from twelve individuals. Surprisingly,
ten of these twelve revealed the Cambridge Reference Sequence haplotype, rare in modem-day Turkey. The authors
provided a number of arguments for why this result cannot represent contamination. They also noted that homogeneity
of the local maternal gene pool at Titri~ Hoyuk would be consistent with ethnographic evidence in modem-day
Turkey.
In two studies, Ottoni and colleagues studied mtDNA HVR sequences from Roman (n=24) and Middle Byzantine
(n=51) skeletons from Sagalassos in southwestern Anatolia (Ottoni et al. 2011, 2016). Sagalassos was a major urban
centre in the Roman period (Fig. 27-1), but declined in size in the Byzantine period and was abandoned by the 13 th
century. The analyses included comparison with mtDNA data from a modem-day sample collected from a
neighbouring village (Ottoni et al. 2016). The authors studied temporal change in the mtDNA gene pool over two
millennia and compared the observed patterns with population simulations of genetic drift. Their results suggested that
the region's maternal gene pool was largely continuous from Roman times until today, with some eastern Eurasian
contribution, possibly after the Byzantine period. Comparisons with simulations also suggested that the population
underwent decline either in the Late Roman or Byzantine period, although it was not possible to ascertain the exact
timing and severity with the available data.
c;::emialo S1rt 1
Mycenaeans
Karata~-SemayOk
IWezmeh Cave]
y ~~~
Ganj Dareh
Tepe Abdul Hosein
• Epi-Palaeolithic Bronze Age

Ceramic Neolithic ~ Hellenistic Era
y Aceramic Neolithic l8l Roman Era
♦ Chalcolithic
Figure 2 7-1. Geographic locations of the sites mentioned in this chapter.
A 2018 study analysed mtDNA HVR in 16 individuals from C,::emialo S1rt1 in southeastern Anatolia, occupied
between the Late Iron Age and Hellenistic period (Yaka et al. 2018). Only twelve individuals yielded mtDNA, and the
sample showed high genetic diversity. Comparing this dataset with published mtDNA haplotype data from Neolithic
and modem-day samples from northern Mesopotamia, the authors found limited differentiation (Fs1 values < 0.06).
Evaluating this result by comparing it with differentiation levels observed in genetic drift simulations, the authors
suggested that the maternal gene pool of the region was largely continuous from the Neolithic to the present day. They
also noted that this conclusion does not fully preclude admixture in the region's history, and low levels of maternal
admixture could have gone undetected with the small sample sizes used. It is also possible that the maternal gene pool
remained more stable than nuclear loci if large scale population movements involved more males than females; this
question may eventually be resolved using ancient genome data.
More recently, Chylenski and colleagues studied mtDNA from the central Anatolian Neolithic site of C,::atalhoyuk
(Chylenski et al. 2019). The authors used in-solution capture (enrichment of human molecules) using DNA baits
designed against mtDNA, and applied NGS on the enriched libraries. They could thus sequence >85% of the mtDNA
genome for ten individuals. The main goal of the study was to assess genetic relationships among intramural burials at
C,::atalhoyuk East Mound, and the results revealed that all of the ten individuals, buried in four adjacent buildings, had
different mtDNA haplotypes-i.e., none shared maternal relatedness. The authors also noted that their C,::atalhoyuk
sample was similar in haplogroup composition to Neolithic mtDNA samples from central Anatolia, western Anatolia,
and Europe (i.e., European samples from Neolithic contexts), but distinct from other Fertile Crescent groups. The
authors suggested that this could imply that Neolithic migrations into Europe were linked with central Anatolia.
ANCIENT GENOMICS USING NEXT GENERATION SEQUENCING
Although insights from targeted sequencing of ancient mtDNA have been illuminating, their limitations have
pushed the field to abandon this approach in favour of ancient genome analyses. Indeed, NGS technologies allow
obtaining orders of magnitude more information at similar costs (depending how much a genome is sequenced), and
also provide additional advantages, as we explain below. In the NGS approach, DNA is extracted and converted in
molecular libraries, and millions to billions of molecules in the pool are shotgun sequenced (as opposed to targeted
sequencing of single molecules). DNA sequence information from across the genome is thus obtained in a single
experiment.
In 2010, the very first ancient whole genome from a human, a Paleo-Inuit specimen from Greenland, was generated
by shotgun sequencing (Rasmussen et al. 2010). The same year, a draft genome representing Late Pleistocene
Neanderthals was published (Green et al. 2010), and by the end of 2010, a new hominin species, Denisovan, was
described for the first time based on its genomic data (Reich et al. 2010). These pioneering studies paved the way for
the archaeogenomic era. The early 2010s also saw significant changes in standard aDNA analyses, both in the
molecular laboratory and in the computational domains. These changes adapted NGS procedures to suit the features of
ancient DNA, such as broken short molecules of variable size and the presence of postmortem damage (mostly uracils
caused by cytosine deamination) at molecule ends (see Stoneking and Krause 2011; Pickrell and Reich 2014).
Ancient genomic procedures are currently mostly standardised, but there still remains a major distinction between
two alternative approaches: shotgun sequencing, or shotgun sequencing after targeted capture. To date, some
researchers have largely relied on capturing preselected 1.2 million SNPs to mass produce partial genomes at low cost,
while others have argued for pure shotgun sequencing (or capturing whole genomes) to avoid biases and thus increase the
reusability of the data produced. In general, it is possible to jointly analyse both types of data (shotgun and SNP capture),
although technical biases that may arise in meta-analyses involving the two data types still await systematic investigation
(Yaka et al. 2021 ). Below, we refer to genomic data produced using SNP capture as simply "captured genomes."
Final Pleistocene Through the Holocene Population Dynamics

in Anatolia in Light ofAncient Genomes
Here we describe demographic processes as inferred by published analyses of ancient genomes. This section's
account starts around 20,000 years ago, towards the end of the Last Glacial Maximum (LGM), and ends with the
appearance of fully sedentary agricultural communities around the early 7th millennium BCE in Anatolia.
Published Ancient Genome Data from Early Holocene Anatolia
Genomes published from early Holocene Anatolia since 2015 have already reached significant numbers. These
include 26 captured genomes from Neolithic Barcm and Mente~e Hoyi.ik in northwestern Anatolia (Mathieson et al.
2015), and two shotgun genomes again from Barcm Hoyi.ik (Hofmanova et al. 2016). A total of 37 genomes derive
from central Anatolian sites: four shotgun genomes from Aceramic Neolithic Boncuklu Hoyi.ik, five genomes from
Ceramic Neolithic Tepecik-yiftlik (K.1lm9 et al. 2016), five captured genomes from Boncuklu Hoyi.ik and a single
captured genome from Epipalaeolithic Pmarb~1 (Feldman et al. 2019), eight shotgun genomes from Aceramic
Neolithic A~1kh Hoy(.ik, and 14 from Ceramic Neolithic <;atalhoyi.ik (Yaka et al. 2021) (Fig. 27-1). There also exists a
recent preprint with a pair of genomes, one from Barcm and one from Aktoprakhk, both belonging to the Ceramic
Neolithic of northwestern Anatolia, although the data are not yet published (Marchi et al. 2020).
As ofmid-2021, most published genomes from early Holocene Anatolia are low coverage (<0.5x), i.e., not all of
the genome, or not all polymorphic sites are represented. Still, such partial genomes are useful for simple measures of
kinship, as well as gauging population genetic similarities. In the meantime, ancient early Holocene genomes from the
Balkans (Mathieson et al. 2018), the Caucasus (Jones et al. 2015), the Zagros (Lazaridis et al. 2016; Broushaki et al.
2016), and the Levant (Lazaridis et al. 2016; Feldman et al. 2019), have also been published, thus providing regional
context to the study of Anatolian population history in this period. We note that all hitherto published genomes from
Anatolia originate from central and northwestern Anatolia. The reason for western and southeastern Anatolia not
being represented to date may partly be attributed to chance, and partly to DNA preservation issues, while the lack of
representation of early Holocene eastern and northern Anatolia is due to the apparent absence of identified sites and/or
excavations from those regions. In the following, we provide a synthesis of what these ancient genome studies have
shown, with particular reference to the Neolithisation process in the region.
The Anatolian Final Pleistocene/Epipalaeolithic

Gene Pool as a Product of West-East Admixture
The Anatolian Epipalaeolithic gene pool is currently represented by a single individual's genome, that of the
Pmarb~1 rock shelter of central Anatolia, dated to 13,642-13,073 cal BC (Fig. 27-1). This genome, despite being low
coverage, is distinct from its contemporaries sampled from the southern Levant (Natufians) and from Europe (Fig. 27-
2A) (Feldman et al. 2019). Using the f-statistics-based qpAdm approach, Feldman and colleagues inferred the Pmarb~1
lineage to be the outcome of an admixture between (a) a population with affinity to the early Holocene Levant
A _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ B _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __
Epi-Palaeolithic & Neolithic Era Chalcolithic Era
• C Anatolia Chalcolithic
0 C Anatol ia Epi-Paleolilhic
■ N Anatolia Chalcolithlc
• C An atol ia Neolithic
♦ NW Anatolia Chalcolithic
♦ NW Anatolia Neolithic
... SE Anatolia Chalcolilhic
II Zagros Neolithic
K&'. N Levant Chalcolilhic
II Caucasus Epi-Paleol ithic
1B1 Caucasus Neolithic
♦ Levant Epi-Paleolithic
♦ Levant Neolithic
1111
1111
c ______________________ D- - - - - - - -lron
-- -~--~-------
Bronze Age •
Age & Ottoman Era
C Anatolia IA
e C Anatolia EBA
0 C Anatolia Ottoman
0
A. S Anatolia EBA
• SE Anatolia EBA
0 C Anatolia MLBA
r::J SE A natolia MLBA
0
V
•
•
Figure 27-2. Diagrams showing the principal component analysis of the ancient genomes projected
onto modem-day western Eurasian populations: a) Epi-Palaeolithic and Neolithic populations from SW Asia;
b) Chalcolithic populations from Anatolia and N Levant; c) Bronze Age Anatolian populations;
d) Iron Age and Ottoman Era populations from Anatolia.
(genetically sampled Natufians) and (b) a population with affinity to Late Palaeolithic and Mesolithic Europeans, with
roughly equal contributions from both groups. Interestingly, Marchi et al. (2020) reach a similar conclusion without
using the Pmarba~1 data. Relying instead on site frequency spectrum-based modelling on high coverage (>5x)
genomes from Europe, Anatolia, and Iran, they predict a similar pattern of 50:50 admixture involving "western" and
"eastern" sources that moulded the early Holocene Anatolian gene pool, occurring around ca. 19,000 years ago (Fig.
27-3A). The agreement between the two results, one based on a single genome, the other based on indirect inference,
. .
1s reassunng.
It still remains unclear where and when this presumed admixture happened and how long it continued, and what
role climatic warming by the end of the LGM may have played. We also lack information about Late Pleistocene gene
pools of Anatolia with the exception of the terminal Pleistocene site of Pmarba~1. Obviously, wider sampling of Upper
Palaeolithic and Epipalaeolithic human remains would significantly help here. In addition, future work with
sedimentary DNA could shed light on population dynamics of the region during the LGM and how human populations
responded to climate change (see below).
Regional Continuity from the Epipalaeolithic to the Aceramic Neolithic in Central Anatolia
Recent ancient genome analyses indicate that during the early (Pre-Pottery, or Aceramic) Neolithic, around 10,000-
7000 BCE, three gene pools can be distinguished in the area of Neolithisation in SW Asia (K1hn9 et al. 2017;
Skourtanioti et al. 2020; Yaka et al. 2021): (a) The Levant, which is currently represented by samples from the south
Levant, such as Ain Ghazal; (b) The Zagros, currently represented by sites such as Ganj Dareh, and which shows high
affinity to Caucasus Upper Palaeolithic populations; (c) Anatolia, which is currently represented by Boncuklu and
A~ikh in central Anatolia, with high genetic similarity between these two sites, the first on the Konya plain and the
latter in the Cappadocia region (Fig. 27-2A). Other major regions of early Neolithisation, the northern Levant,
northern Mesopotamia, and Cyprus, are not yet included on the genomic map. Still, it appears safe to assume that their
gene pools either overlapped with or represented mixtures of the aforementioned three groups.
The most notable pattern revealed by the available data is the regional continuity signal between pre-Neolithic and
Neolithic groups in SW Asia. This can be inferred from the high similarity between pre-Neolithic and Neolithic
context genomes in each of the three regions: Boncuklu and A~ikh to Pmarba~1, Ain Ghazal to Raqefet Cave, and
Ganj Dareh to Hotu Cave, and the general distinctiveness retained in all three gene pools during the Aceramic
Neolithic (Lazaridis et al. 2016; Feldman et al. 2019; Yaka et al. 2021).
Pre-Neolithic
Neolithic
Attar
Post-Neolithic
Homogenisation
in SW Asia
Figure 27-3. An illustration of population history in SW Asia: a) Pre-Neolithic gene pools

formed after LGM; b) Population movements during Neolithic period;
c) Post-Neolithic expansions ofYamnaya and homogenisation in SW Asia.
This result has been considered strong evidence that Neolithic development occurred mainly by cultural interaction
across these three regions, instead of mass population movements and replacement. In central Anatolia, this aligns
with what archaeology has predicted based on material culture: that despite regional interactions, such as seashells of
Levantine origin and crops of northern Mesopotamian origin, the populations of A~1kh (8350-7300 cal BC) and
Boncuklu (ca. 8300-7500 cal BC) were essentially local hunter-gatherer communities who adopted sedentism, and
experimented with animal herding and crop cultivation, while continuing with traditional foraging (Baird 2012; Baird
et al. 2018). K1hrn;: et al. (2016) had also pointed out that the relatively low genetic diversity of the Boncuklu
population reflects their hunter-gatherer-like demographic profile, akin to Upper Palaeolithic or Mesolithic groups in
the Caucasus and Europe.
Despite this strong signal of continuity during Neolithisation, we also see some degree of regional population
movement and admixture (Fig. 27-3B). For instance, Feldman et al. (2019) predict Zagros-related admixture in central
Anatolia sometime between the Epipalaeolithic ( ca. 13 kya) and the Aceramic Neolithic (ca. 10 kya). Specifically,
they estimate ca. 90% of Pmarba~1-related ancestry in their Boncuklu sample, and that ca. 10% ancestry is related to
the Neolithic period western Zagros (Feldman et al. 2019). How this admixture happened, and its exact source, remain
yet unknown, especially because Zagros and Caucasus populations were genetically similar. What can be confidently
stated, for now, is that populations across SW Asia were never fully isolated.
The Transition from Aceramic to Ceramic Neolithic:

Increasing Diversity and Changing Traditions
Regional gene flow appears to have continued, and possibly intensified, during the transition from the Aceramic
Neolithic to Ceramic Neolithic periods in SW Asia, when sedentary communities increasingly focused on farming. In
central Anatolia, multiple studies have identified admixture from the Levant, and possibly also the Zagros, although
the latter effect is less clear (K1hrn;: et al. 2017; Feldman et al. 2019; Y aka et al. 2021 ). D-test results have consistently
revealed significantly higher genetic affinity of Levant populations (pre-Neolithic and Neolithic) to Ceramic Neolithic
central Anatolians, than to Aceramic Neolithic central Anatolians (Fig. 27-3B). This pattern has been upheld with
fresh data from different Neolithic villages, including A~1kh and <;atalhoyi.ik (Fig. 27-1 ). Y aka and colleagues
estimate about 10% of ancestry at both <;atalhoyi.ik and Barcm derive from an early Holocene Levant-related source
(Yaka et al. 2021). This result indicates gene flow from the Levant into central Anatolia between the Aceramic and
Ceramic Neolithic periods. This movement does not seem to be unilateral, as in the Levant and Zagros a symmetric
signal is observed, implying gene flow from central Anatolia or its neighbouring regions (Lazaridis et al. 2016).
Inter-regional gene flow could explain another observation: the substantially higher within-population genetic
diversity in Ceramic Neolithic central Anatolia relative to Aceramic Neolithic populations (K1hrn;: et al. 2016; Y aka et
al. 2021; but see Marchi et al. 2020). In fact, increasing genetic diversity has also been reported across Eurasia
following the Neolithic transition, as traced from drastic reductions in genomic runs of homozygosity, a measure of
genetic homogeneity caused by small population size and consanguinity (Ceballos et al. 2020; Ringbauer et al. 2020).
It is important to note, however, that population growth related to agriculture and sedentism could also create an
analogous effect as admixture (Yaka et al. 2021). The contributions of the two processes, growth vs. admixture, in the
demographic response to Neolithic development thus await disentanglement.
Beyond population dynamics, archaeogenomic data have started to be used to investigate local traditions and
social structures in Anatolia, by means of studying the correspondence between burial location, genetic kinship, and
gender (Yaka et al. 2021). Yaka and colleagues reported kinship patterns consistent with biological family members
being buried together in houses in A~lkh and Boncuklu, i.e., in the Aceramic Neolithic period. Conversely, biological
relatedness among co-burials was found to be rare in two major Ceramic Neolithic sites studied: Barcm and
<;atalhoyi.ik. Although co-burials genetically studied from these two sites were dominated by subadult burials, it is
notable that the result is consistent with earlier reports based on dental data and mitochondrial DNA data collected
from adult burials (Pilloud and Larsen 2011; Chylenski et al. 2019). This outcome supports the notion that <;atalhoyi.ik
society, and perhaps some other Neolithic communities, were structured around social kinship types instead of
biological kinship (Pilloud and Larsen 2011). While these results are highly intriguing, the sample sizes used by Yaka
and colleagues were limited, with only 30 co-buried individuals across eight buildings (or building pairs) studied in
total. More data will therefore be needed to obtain a definitive picture of subfloor burial traditions in Neolithic
Anatolia.
The Elusive Origins of the European Early Neolithic Population
Ancient genome studies have described population movement to be the main driver in the Neolithisation of Europe
starting with the 7th millennium BCE (Skoglund et al. 2014; Lazaridis et al. 2014), and later studies linked this
migration process to Anatolia, as opposed to the Levant (Mathieson et al. 2015; Omrak et al. 2016; Lazaridis et al.
2016; Hofrnanova et al. 2016; K1hrn;: et al. 2016). Indeed, central Anatolian, western Anatolian, Aegean, and nearly all
genomes from Neolithic farming contexts in Europe appear highly similar, and largely distinct from Mesolithic groups
from the Balkans or elsewhere in Europe. Neolithisation of the Caucasus also involves some degree of early Holocene
Anatolia-related gene flow, but the effect is modest compared to that observed in Europe (Wang et al. 2019).
A lingering question is where the population movement that colonised Europe travelled from: central Anatolia,
western Anatolia, or from the western Aegean and the Balkans? The answer is still unclear. In fact, the role of human
movement vs. cultural interaction in the Neolithisation of western Anatolia by the 7th millennium itself remains
equivocal. This ambiguity, in tum, stems from the current lack of pre-Neolithic (Early Holocene for western Anatolia)
genomes from regions west of the original zone of Neolithisation, i.e., genomes from western Anatolia and the
Aegean. Without pre-Neolithic genomes from western Anatolia, two scenarios remain possible: (1) Ifwe assume that
pre-Neolithic western Anatolians were genetically similar to Balkan Mesolithic groups ( e.g., those from the Iron Gates
region), this would imply that there was a population replacement in western Anatolia by migrants from central
Anatolia. Archaeologists such as Mehmet Ozdogan have also been describing scenarios of human movement based on
material culture similarities (Ozdogan 2014); (2) If we assume that pre-Neolithic western Anatolians were already
genetically more similar to central Anatolians than to Balkan Mesolithic groups, then the Neolithic in western
Anatolia may have started without significant human movement. Instead, the western Anatolian Neolithic could have
been started by local hunter-gatherers simply adopting Neolithic culture by the 7th millennium. In this second scenario,
it would then be these recently Neolithised western Anatolian and/or Aegean populations who instigated the migration
process into Europe. K1hni;: et al. (2017) have raised this possibility, pointing out that Neolithic culture may have been
adopted by local hunter-gatherers. Results from Marchi et al. (2020) also partly support this notion. However, the jury
is still out on the issue.
Post-Neolithic Population Dynamics

in Light ofAncient Genomes
Post-Neolithic periods from Anatolia are mostly represented by the Chalcolithic and Bronze Age (BA), and
particularly the Early Bronze Age (EBA), with thus far limited data from the Iron Age and Ottoman eras, and no
Roman or Byzantine samples (Fig. 27-1). Ancient genomic studies on these more recent periods are fewer in number
compared to the intensity of publications on the Neolithic period. This is likely a result of an early focus on
Neolithisation, and would be expected to change over time, given the higher average preservation of more recent
skeletal samples, as well as a multitude ofrich skeletal collections. Here we present an overview of the available data
and the insights they have brought.
Published Ancient Genome Data

from Post-Neolithic Anatolia
The data published from post-Neolithic periods include: a single shotgun genome from Chalcolithic Kumtepe, in
northwestern Anatolia (Omrak et al. 2016); a single capture genome from Chalcolithic Barcm, also in northwestern
Anatolia (Lazaridis et al. 2016); three captured genomes from Bronze Age Haimanoren-Gondurle in southwestern
Anatolia (Lazaridis et al. 2017); nine shotgun genomes in total from Kaman-Kalehoyi.ik in central Anatolia from
periods ranging from the Assyrian Colony period (n=3), Old Hittite ("Middle to Late Bronze Age") (n=2), Iron
AgeJHellenistic (n=2), and Ottoman (n=2) (de Barros Damgaard et al. 2018); three shotgun genomes from EBA
Ovaoren-Topakhoyi.ik in central Anatolia (de Barros Damgaard et al. 2018); one captured genome from Early
Chalcolithic Buyi.ikkaya in north central Anatolia; 12 captured genomes from Late Chalcolithic <;amhbel Tarlas1 in
north central Anatolia; 11 captured genomes from Late Chalcolithic ikiztepe in northern Anatolia; 18 captured
genomes from Late Chalcolithic Arslantepe in southeastern Anatolia; a single captured genome from EBA Titri~
Hoyi.ik in southeastern Anatolia; and four captured genomes from EBA Arslantepe in eastern Anatolia (Skourtanioti et
al. 2020) (Fig. 27-1).
In addition, a number of publications have reported genomes from neighbouring regions, including the
Chalcolithic and BA northern Levant (Tell Kurdu, Alalakh, Ebla) (Skourtanioti et al. 2020), from the Chalcolithic and
BA Caucasus (Wang et al. 2019; Skourtanioti et al. 2020), from mainland BA Greece and Crete (i.e. Mycenaeans and
Minoans) (Lazaridis et al. 2018), from the Chalcolithic and BA Balkans (Mathieson et al. 2018), and from
Chalcolithic to Iron Age Iran (Lazaridis et al. 2016).
Eastern/Northeastern Admixture
and the Middle-Holocene Gene Pool
The clearest pattern observed, when comparing genomes from both Chalcolithic and BA Anatolia with those from
Neolithic Anatolia, is substantial admixture from early Holocene Caucasus- or Iran-related populations (Kilmi;: et al.
2016, de Barros Damgaard et al. 2018, Lazaridis et al. 2018, Skourtanioti et al. 2020) (Fig. 27-2B, C). In qpAdm
modelling by de Barros Damgaard et al. (2018) and by Skourtanioti et al. (2020), these eastern/northeastern sources
are assigned 40~50% contributions to the gene pool of BA Anatolia. In general, two Upper Palaeolithic Caucasus
genomes (so-called CHG) appear as the closest possible source of admixture. Further, Skourtanioti and colleagues use
the DATES program (Skourtanioti et al. 2020) to estimate an approximate admixture date of 6500 BCE for the
admixture event between the Neolithic Anatolian gene pool and these eastern/northeastern populations.
Although it would be appealing to associate this process with the 8.2-kiloyear event, we note that a single
admixture event is unlikely, and the eastern/northeastern gene flow was likely a protracted process. This can be
observed as a low level CHG-related admixture already by the Late Neolithic and early Chalcolithic samples from
early 6th millennium contexts, such as Tepecik-yiftlik and Bogazkoy-Buyukkaya (K1hrn;: et al. 2016; Skourtanioti et
al. 2020). Nevertheless, it is intriguing that the likely time span of this CHG-related admixture process overlaps with a
roughly 2500-year demographic decline in Anatolia estimated based on anthropogenic radiocarbon dates and site
surveys (Palmisano et al. 2021).
The exact source or sources of this gene flow are likewise obscure. It is plausible that these sources included
northern Anatolia, i.e., the Anatolian Black Sea region, or eastern Anatolia, which have not been genetically sampled
yet. In fact, there appears some variation, albeit non-significant, in the estimated admixture dates between northern
and eastern Anatolia reported by Skourtanioti et al. (2020), with earlier admixture dates estimated for ikiztepe. Future
work using a wider set of available samples and models (e.g. Kumtepe, Harmanoren-Gondurle, Kaman-Kalehoyuk)
may clarify the picture and help pinpoint the sources of gene flow. For now, we can only conclude that the gene flow
arrived from the east or northeast of the peninsula.
This middle-Holocene Anatolian gene pool, a result of admixture between Neolithic era populations and
eastern/northeastern populations, appears relatively homogeneous both spatially and temporally, from the 6th
millennium to the 2nd millennium BCE, at least based on the available data (Fig. 27-2C). Major ancestry sources of the
people at EBA Arslantepe in eastern Anatolia, at Chalcolithic Barcm in northwestern Anatolia, at Late Chalcolithic
ikiztepe in northern Anatolia, or at Old Hittite period Kaman-Kalehoyuk, are all highly similar (Skourtanioti et al.
2020). It is further interesting to note that the western Eurasian steppe expansions related to the Yamnaya culture,
which remoulded the European gene pool around 3000 BCE, do not seem to have influenced Anatolia (de Barros
Damgaard et al. 2018) (Fig. 27-3C). D-tests show that Yamnaya-related groups do not show higher affinity to Old
Hittite period central Anatolians relative to the EBA central Anatolian population.
The lack of large admixture signals between the 6th millennium to the 2nd millennium BCE is noteworthy because
this is a time span of high cultural and political dynamism documented in Anatolia. The period includes processes
such as the Transcaucasian Kura-Araxes material culture spreading from the south Caucasus into eastern and
southeastern Anatolia (Sagona 2000; Rothman 2015; Frangipane 2014; Steadman et al. 2018), the establishment of
Assyrian Trade Colonies in central Anatolia (Kulakoglu and 6zturk 2015), the expansion of Hittite culture and
language, which is suggested to have originated from outside Anatolia (Bryce 2005), and simultaneous Mycenaean
influence in western Anatolia (Kelder 2004), among others.
However, the observation that the Anatolian gene pool was relatively stable during this period does not imply that
the peninsula was insular to any human movement. Tue genomic analyses should instead be interpreted as immigrant
groups being systematically fewer in numbers relative to the resident population (e.g., as opposed to Yamnaya
migrations in Europe). Indeed, the Anatolian population may have grown to sizable numbers by intensifying
agriculture, possibly aided by the secondary products revolution during the Chalcolithic and Bronze Ages (Massa
2016). In addition, if movement involved populations with similar ancestry profiles as their contemporaneous
Anatolians ( e.g., Mycenaeans), this would not have left an easily recognizable trace in the available genomic data.
The Last 4,000 Years
Anatolia continued to witness dramatic sociocultural and political changes after the 2nd millennium BCE that
frequently involved interactions with and immigration from populations external to the peninsula, such as the
Pmygian culture, Galatian migrations, the Achaemenid expansion, the Hellenistic and Roman periods, and more
recently, Turkic expansions. Data available from these periods, to date, only include shotgun genomes from Kaman-
Kalehoyuk from central Anatolia, two reported to be Iron Age, and two from the Ottoman period (but without C14
dates). Combined with data from modem-day Anatolia (see above), we can conclude that a certain amount of eastern
Eurasia-related admixture occurred over the last 4000 years, but the tempo of the process still awaits detailed analysis
with larger sample sizes (Fig. 27-2D).
SUMMARY AND FUTURE PERSPECTIVES
This review attempted to provide an account of broad-scale population dynamics of Anatolia over the last 15,000
years. The main inflection points can be summarised as follows: following an LGM admixture process that created the
Anatolian Epipalaeolithic population during the Final Pleistocene, the Anatolian gene pool was shaped by close
interactions with areas to its south and east during Neolithisation in the Early Holocene, but with limited gene flow,
with admixture estimates not above 10%. Starting with the Late Neolithic and continuing into the Early Bronze Age,
there occurred significant gene flow from Caucasus-related populations, with possible 40~50% contributions, but
thereafter the gene pool was relatively stable for millennia. No major shift was observed in the gene pool until more
recently, when a limited amount of eastern Eurasian-related admixture appeared in the modem-day Anatolian gene
pool. Here we did not attempt to systematically align these demographic shifts with particular environmental events or
with the multitude of changes in material culture or language. Nevertheless, the main events described here indicate
that many cultural, political, and linguistic changes recorded in historical data or material culture do not necessarily
involve population replacement. From the Neolithic period to the Hittite era, fundamental transformations appear to
have happened mostly by cultural interaction. We hope that this review will provide material and inspiration for future
analyses of demography, cultural dynamics, and inferences from palaeoclimate models. Meanwhile, we keenly
anticipate more palaeogenomic data from Anatolia to be published in the coming years, including data collected from
across the Holocene. Denser sampling would surely reveal surprising heterogeneity and population connections.
However, there are vital questions that could be addressed even with the current data. Examples include the role of
population growth versus admixture in the increasing diversity of the Neolithic gene pool, the search for the origins
and timing of the post-Neolithic Caucasus-related gene flow, or estimation of eastern Eurasian admixture amounts in
the 2nd millennium.
Besides population dynamics, we expect an upcoming theme in archaeogenomic studies to be the mapping of
biological kinship among co-burials, from Neolithic intramural burials in the same house, to multiple burials in BA
pithoi and in Byzantine sarcophagi. In conjunction with anthropometric studies, such endeavours will allow for better
understanding of the social structure of past societies, including kinship structures and matrimonial traditions. For
instance, genomic data have recently been used to suggest that at least some Aceramic Neolithic populations in
Anatolia may not have adopted patrilocal traditions (Yaka et al. 2021), while dental data from <;atalhoyuk have been
interpreted as indicating a patrilocal culture (Pilloud and Larsen 2011). Although preliminary, these results
demonstrate the potential power of genomic and anthropometric approaches.
Studying natural selection on Anatolian human populations against the backdrop of lifestyle changes and
admixture (Mathieson et al. 2015; Marchi et al. 2020), the study of domestic animal population dynamics as a proxy
for human interaction and trade networks (Daly et al. 2018; Yurtman et al. 2020), and the study of ancient microbes
from tooth samples and from dental calculus, including ancient pathogens and oral microbiomes (Key et al. 2020), are
additional lines of study that we expect will attract increasing attention in the near future.
Sedimentary aDNA (Rohland et al. 2018) could further become a fundamental source of information for
understanding population dynamics in Anatolia. Genetic profiles of cores from cave sites may help overcome the
dearth of skeletal material from Late Pleistocene and Early Holocene layers, especially in western and northern
Anatolia. Genomic sampling from locations like Karain Cave could help describe the Upper Palaeolithic gene pools in
Anatolia, as well as archaic hominin populations.
Finally, recent work indicates the need for even closer collaboration between population geneticists and the
archaeological and anthropological communities, for at least two reasons. The first is the establishment of accurate
and sustainable terminology in describing population genetic entities. For instance, using "Caucasus Hunter-Gatherer
(CHG)" instead of"Upper Palaeolithic/Mesolithic S Caucasus" may be appealing for its simplicity (Jones et al. 2015),
but such brevity could cause confusion in the long term, as future data may reveal both cultural and genetic
heterogeneity in the region and periods in question. Another example may be "Anatolian Aceramic Farmer" used to
describe the central Anatolian 9th millennium populations (Feldman et al. 2019). In this period, crops had only limited
contribution to subsistence in the region, and only for some communities (Baird et al. 2018)-referring to these
communities as "farmers" is, therefore, inaccurate. More generally, it appears risky to associate population genetic
entities with material culture phenomena, as neighbouring communities not associated with those cultural phenomena
may eventually turn out to belong to the same gene pool. Thus, geneticists and social scientists should jointly continue
to seek more accurate and sustainable terminology and ensure a healthy legacy in the literature. The most reliable and
universal terminology to use may be terminology based on temporal windows (millem1ia or centuries) and
geological/climatic epochs, such as the Pleistocene, Holocene, etc., as this would allow more successful comparison of
specimens at wide scales.
A second issue relates to the need to shape archaeogenomic terminology to avoid ethnic implications of scientific
studies in the face of rampant racism. To this end, ancient DNA researchers should define gene pools so as to avoid
associating prehistoric groups with modern-day ethnic groups and with modem-day nation-states, such as Turkey,
Iran, or Greece. Gene pools instead should be described based on neutral geographic nomenclature as much as
possible.
In short, we are entering an exciting decade where quite a number of long-standing questions about human history
in Anatolia will start to be resolved using biological data. We hope that an increasing number of geneticists and
archaeologists from the region, including Turkey, will be contributing to expanding this research.
ACKNOWLEDGEMENTS
This chapter was supported by European Research Council (ERC 772390), Horizon2020 Framework Programme
(Widespread 05-2020-952317), Scientific and Technological Research Council of Turkey (TOBiTAK - 1172229),
and the National Ph.D. Scholarship Program for Social Sciences and Humanities from TOBiTAK.
REFERENCES CITED
Alkan, Can, Pmar Kavak, Mehmet Some!, Omer Gokc;:ilinen, Serkan Ugurlu, Ceren Saygi, Elif Dal, Kuyas Bugra,
Tunga Gungor, S. Cenk ~ahinalp, Nesrin Ozoren, and Cemalettin Bekpen. 2014. \Vhole Genome Sequencing of
Turkish Genomes Reveals Functional Private Alleles and hnpact of Genetic mteractions with Europe, Asia and
Africa. BMC 5;enomics 15(1 ): 1-12.
Angel, J. Lawrence. 1966. Porotic Hyperostosis, Anemias, Malarias, and Marshes in the Prehistoric Eastern
Mediterranean. Science 153(3737): 760-763.
Atakuman, <;igdem. 2008. Cradle or Crucible: Anatolia and Archaeology in the Early Years of the Turkish Republic
1923-1938. Journal ofSocial Archaeology 8: 214--235.
Baird, Douglas. 2012. The Late Epipaleolithic, Neolithic, and Chalcolithic of the Anatolian Plateau, 13,000--4000 BC.
m A Companion to the Archaeology of the Ancient Near East, vol. 1, D. Potts, ed., 431--465. Oxford: Blackwell
Publishing.
Baird, Douglas, Andrew Fairbairn, Emma Jenkins, Louise Martin, Caroline Middleton, Jessica Pearson, Eleni Asouti,
Yvonne Edwards, Ceren Kabukcu, Gokhan Mustafaoglu, Nerissa Russell, Ofer Bar-Yosef, Geraldine Jacobsen,
Xiaohong Wu, Ambroise Baker, and Sarah Elliott. 2018. Agricultural Origins on the Anatolian Plateau.
Proceedings ofthe National Academy ofSciences of the United States ofAmerica 115(14): E3077-E3086.
Berkman, Ceren Caner, Havva Cine;:, Ceran ~ekeryapan, and inci Togan. 2008. Alu Insertion Polymorphisms and an
Assessment of the Genetic Contribution of Central Asia to Anatolia with Respect to the Balkans. American
Journal ofPhysical Anthropology.· The Official Publication of the American Association ofPhysical Anthropologists
136(1): 11-18.
Broushaki, Farnaz, Mark G. Thomas, Vivian Link, Saioa Lopez, Lucy Van Dorp, Karola Kirsanow, Zuzana
Hofmanova, Yoan Diekmann, Lara M. Cassidy, David Diez-del-Molino, Athanasios Kousathanas, Christian Sell,
Harry K. Robson, Rui Martiniano, Jens Blocher, Amelie Scheu, Susanne Kreutzer, Ruth Bollongino, Dean Bobo,
Hossein Davoudi, Olivia Munoz, Mathias Currat, Kamyar Abdi, Fereidoun Biglari, Oliver E. Craig, Daniel G.
Bradley, Stephen Sherman, Krishna R. Veeramah, Marjan Mashkour, Daniel Wegmarm, Garrett Hellenthal, and
Joachim Burger. 2016. Early Neolithic Genomes from the Eastern Fertile Crescent. Science 353(6298): 499-504.
Bryce, Trevor, 2005. The Kingdom ofthe Hittites. Oxford: Oxford University Press.
Cappieri, Mario. 1970. The Anatolians of the Late Neolithic and Chalcolithic Age. Belleten 34(136): 509-555.
Ceballos, Francisco C., Kanat Giliiin, N. Ezgi Alt1m~1k, Hasan Can Gemici, Cansu Karamurat, Dilek Koptekin,
K1v1lc1m Ba~ak Vural, Elif Surer, Y1lmaz Selim Erda!, Anders Gotherstrom, Fusun Ozer, <;igdem Atakuman, and
Mehmet Some!. 2020. Human Inbreeding has Decreased in Time Through the Holocene. bioRxiv: doi:
https://doi.org/10.1101/2020.09.24.3 l 1597.
Chylenski, Maciej, Edvard Ehler, Mehmet Some!, Reyhan Yaka, Maja Krzewinska, Miroslawa Dabert, Anna Juras,
and Arkadiusz Marciniak. 2019. Ancient Mitochondrial Genomes Reveal the Absence of Maternal Kinship in the
Burials of<;atalhoyuk People and Their Genetic Affmities. Genes 10(3): 207.
Cinnioglu, Cengiz, Roy King, Toomas Kivisild, Ersi Kalfoglu, Sevil Atasoy, Gianpiero L. Cavalieri, Anita S. Lillie,
Charles C. Roseman, Alice A. Lin, Kristina Prince, Peter J. Oefner, Peidong Shen, Omelia Semino, L. Luca
Cavalli-Sforza, and Peter A. Underhill. 2004. Excavating Y-chromosome Haplotype Strata in Anatolia. Human
genetics 114(2): 127-148.
Clarke, Laura, Xiangqun Zheng-Bradley, Richard Smith, Eugene Kulesha, Chunlin Xiao, Iliana Toneva, Brendan
Vaughan, Don Preuss, Rasko Leinonen, Martin Shumway, Stephen Sherry, Paul Flicek, and The 1000 Genomes
Project Consortium. 2012. The 1000 Genomes Project: Data Management and Community Access. Nature
Methods 9(5): 459--462.
Daly, Kevin G., Pierpaolo Maisano Delser, Victoria E. Mullin, Amelie Scheu, Valeria Mattiangeli, Matthew D.
Teasdale, Andrew J. Hare, Joachim Burger, Marta Pereira Verdugo, Matthew J. Collins, Ron Kehati, Cevdet Merih
Erek, Guy Bar-Oz, Franc;:ois Pompanon, Tristan Cumer, Canan <;akular, Azadeh Fatemeh Mohaseb, Delphine
Decmyenaere, Hossein Davoudi, Ozlem <;evik, Gary Rollefson, Jean-Denis Vigne, Roya Khazaeli, Homa Fathi,
Sanaz Beizaee Doost, Roghayeh Rahimi Sorkhani, Ali Akbar Vahdati, Eberhard W. Sauer, Hossein Azizi
Kharanaghi, Sepideh Maziar, Boris Gasparian, Ron Pinhasi, Louise Martin, David Orton, Benjamin S. Arbuckle,
Norbert Benecke, Andrea Manica, Liora Kolska Horwitz, Ma1j an Mashkour, and Daniel G. Bradley. 2018. Ancient
Goat Genomes Reveal Mosaic Domestication in the Fertile Crescent. Science 361: 85-88.
de Barros Damgaard, Peter de Rui Martiniano, Jack Kamm, J. Victor Moreno-Mayar, Guus Kroonen, Michael Peyrot,
Gojko Barjamovic, Simon Rasmussen, Claus Zacho, Nurbol Baimukhanov, Victor Zaibert, Victor Merz, Arjun
Biddanda, Ilja Merz, Valeriy Loman, Valeriy Evdokimov, Emma Usmanova, Brian Hemphill, Andaine Seguin-
Orlando, Fulya Eylem Yediay, mam Ullah, Karl-Goran Sjogren, Katrine Hojholt Iversen, Jeremy Choin,
Constanza de la Fuente, Melissa Ilardo, Hannes Schroeder, Vyacheslav Moiseyev, Andrey Gromov, Andrei
Polyakov, Sachihiro Omura, Suleyman Yucel ~enyurt, Habib Ahmad, Catriona McKenzie, Ashot Margaryan,
Abdul Hameed, Abdul Samad, Nazish Gul, Muhammad Hassan Khokhar, O.I. Goriunova, Vladimir I. Bazaliiskii,
John Novembre, Andrzej W. Weber, Ludovic Orlando, Morten E. Allentoft, Rasmus Nielsen, Kristian Kristiansen,
Martin Sikora, Alan K. Outram, Richard Durbin, and Eske Willerslev. 2018. The First Horse Herders and the
hnpact of Early Bronze Age Steppe Expansions into Asia. Science 360(6396): doi: 10.1126/science.aar771 l.
Erda!, Y1lmaz Selim and Valentina D'Amico. 2020. Plano-Occipital Flattening: Intentional or Unintentional
Treatment? In Pathways through Arslantepe: Essays in Honour of Marcella Frangipane, F.B. Restelli, ed., 561-
77. Roma: Sapienza Universita di Roma.
Eroglu, Serpil. 2016. Supraorbital Foramen and Hypoglossal Canal Bridging in Ancient/Modem Anatolian
Populations: Implications for Worldwide Population Distribution. Mediterranean Archaeology and Archaeometry
16(1): 169-183.
Eroglu, Serpil and Y1lmaz Selim Erda!. 2008. Why Did the Frequency of Palatine Torus Increase in the Ancient
Anatolian Populations? Homo: Internationale Zeitschrift fur die Vergleichende Forschung am Menschen 59(5):
365-382.
Feldman, Michal, Eva Fernandez-Dominguez, Luke Reynolds, Douglas Baird, Jessica Pearson, Israel Hershkovitz,
Hila May, Nigel Goring-Morris, Marion Benz, Julia Gresky, Raffaela A. Bianco, Andrew Fairbairn, Gokhan
Mustafaoglu, Philipp W. Stockhammer, Cosimo Posth, Wolfgang Haak, Choongwon Jeong, and Johannes Krause.
2019. Late Pleistocene Human Genome Suggests a Local Origin for the First Farmers of Central Anatolia. Nature
Communications 10(1): 1-10.
Frangipane, Marcella. 2014. After Collapse: Continuity and Disruption in the Settlement by Kura-Araxes-Linked
Pastoral Groups at Arslantepe-Malatya (Turkey). New Data. Paleorient 40(2): 169-182.
Green, Richard E., Johannes Krause, Adrian W. Briggs, Tomislav Maricic, Udo Stenzel, Martin Kircher, Nick
Patterson, et al. 2010. A Draft Sequence of the Neandertal Genome. Science 328(5979): 710-722.
Gulec;:, E. 1999. New Findings on the First Inhabitants of Anatolia from the Dursunlu Site: a Preliminary Appraisal. In
<;aglar Boyunca Anadolu'da Yerle~im ve Konut. Uluslararas1 Sempozyumu Habitat II (3-14 Haziran 1996), Ali
M. Dinc;:ol, ed., 211-216. istanbul: Ege Yaymlar1.
Hagelberg, E. and J.B. Clegg. 1991. Isolation and Characterization of DNA from Archaeological Bone. Proceedings
ofthe Royal Society B: Biological Sciences 244(1309): 45-50.
Higuchi, Russell, Barbara Bowman, Mary Freiberger, Oliver A Ryder, and Allan C Wilson. 1984. DNA Sequences
from the Quagga, an Extinct Member of the Horse Family. Nature 312(5991): 282-284.
Hodoglugil, Ugur and Robert W. Mahley. 2012. Turkish Population Structure and Genetic Ancestry Reveal
Relatedness Among Eurasian Populations. Annals ofHuman Genetics 76(2): 128-141.
Hofreiter, Michael, David Serre, Hendrik N. Poinar, Melanie Kuch, and Svante Paabo. 2001. Ancient DNA. Nature
Reviews Genetics 2(5): 353-359.
Hofmanova, Zuzana, Susanne Kreutzer, Garrett Hellenthal, Christian Sell, Yoan Diekmann, David Diez del Molino,
Lucy van Dorp, Saioa Lopez, Athanasios Kousathanas, Vivian Link, Karola Kirsanow, Lara M. Cassidy, Rui
Martiniano, Melanie Strobel, Amelie Scheu, Kostas Kotsak:is, Paul Halstead, Sevi Triantaphyllou, Nina Kyparissi-
Apostolika, Dushka Urem-Kotsou, Christina Ziota, Fotini Adaktylou, Shyamalika Gopalan, Dean M. Bobo, Laura
Winkelbach, Jens Blocher, Martina Unterlander, Christoph Leuenberger, Ciler Cilingiroglu, Barbara Horejs, Fokke
Gerritsen, Stephen J. Sherman, Daniel G. Bradley, Mathias Currat, Krishna R. Veeramah, Daniel Wegmann, Mark
G. Thomas, Christina Papageorgopoulou, and Joachim Burger. 2016. Early Farmers from Across Europe Directly
Descended from Neolithic Aegeans. Proceedings of the National Academy of Sciences of the United States of
America 113(25): 6886-6891.
Irish, Joel D., Adeline Morez, Linus Girdland Flink, Emma L.W. Phillips, and G. Richard Scott. 2020. Do Dental
Nonmetric Traits Actually Work as Proxies for Neutral Genomic Data? Some Answers from Continental- and
Global-Level Analyses. American Journal ofPhysical Anthropology 172(3): 347-375.
Jones, Eppie R, Gloria Gonzalez-Fortes, Sarah Connell, Veronika Siska, Anders Eriksson, Rui Martiniano, Russell L.
McLaughlin, Marcos Gallego Llorente, Lara M. Cassidy, Cristina Gamba, Tengiz Meshveliani, Ofer Bar-Yosef,
Werner Muller, Anna Belfer-Cohen, Zinovi Matskevich, Nino Jakeli, Thomas F.G. Higham, Mathias Currat,
David Lordkipanidze, Michael Hofreiter, Andrea Manica, Ron Pinhasi, and Daniel G. Bradley. 2015. Upper
Palaeolithic Genomes Reveal Deep Roots of Modem Eurasians. Nature Communications 6(1): 8912.
Kelder, Jorrit M. 2004. Mycenaeans in Western Anatolia. Talanta 36(37): 49-85.
Key, Felix M., Cosimo Posth, Luis R. Esquivel-Gomez, Ron Hubler, Maria A. Spyrou, Gunnar U. Neumann, Anja
Furtwangler, Susanna Sabin, Marta Burri, Antje Wissgott, Aditya Kumar Lankapalli, Ashild J. Vagene, Matthias
Meyer, Sarah Nagel, Rezeda Tukhbatova, Aleksandr Khokhlov, Andrey Chizhevsky, Svend Hansen, Andrey B.
Belinsky, Alexey Kalmykov, Anatoly R. Kantorovich, Vladimir E. Maslov, Philipp W. Stock:hammer, Stefania
Vai, Monica Zavattaro, Alessandro Riga, David Caramelli, Robin Skeates, Jessica Beckett, Maria Giuseppina
Gradoli, Noah Steuri, Albert Hafner, Marianne Ramstein, Inga Siebke, Sandra Losch, Y1lmaz Selim Erda!, Nabil-
Fareed Alikhan, Zhemin Zhou, Mark Achtman, Kirsten Bos, Sabine Reinhold, Wolfgang Haak, Denise Kuhr!ert,
Alexander Herbig, and Johannes Krause. 2020. Emergence of Human-Adapted Salmonella Enterica is Linked to
the Neolithization Process. Nature Ecology & Evolution 4(3): 324-33.
K1lm9, Gul~ah Merve, Dilek Koptekin, <;igdem Atakuman, Arev Pelin Sumer, Handan Melike Donert~, Reyhan
Y aka, Cerna! Can Bilgin, Ali Metin Buyukkarakaya, Douglas Baird, Ezgi Altm1~1k, Pavel Flegontov, Anders
Gotherstrom, inci Togan, and Mehmet Some!. 2017. Archaeogenomic Analysis of the First Steps ofNeolithization
in Anatolia and the Aegean. Proceedings of the Royal Society B.· Biological Sciences 284(1867): 2017-2064.
K1lm9, Gul~ahMerve, Ayc;:a Omrak, Fusun Ozer, Torsten Gunther, Ali Metin Buyukkarakaya, Erhan B19ak91, Douglas
Baird, Handan Melike Donert~, Ayshin Ghalichi, Reyhan Yaka, Dilek Koptekin, Sinan Can Ac;:an, Poorya Parvizi,
Maj a Krzewi.nska, Evangelia A. Daskalaki, Eren Yuncu, Nihan Dil~ad Dagta~, Andrew Fairbairn, Jessica Pearson,
Gokhan Mustafaoglu, Y1lmaz Selim Erdal, Yasin Gokhan <;akan, inci Togan, Mehmet Somel, Jan Stora, Mattias
Jakobsson, and Anders Gotherstrom. 2016. The Demographic Development of the First Farmers in Anatolia.
Current Biology 26(19): 2659-2666.
Kulakoglu, Fikri and Guzel 6zturk. 2015. New Evidence for International Trade in Bronze Age Central Anatolia:
Recently Discovered Bullae at Kultepe-Kanesh. Antiquity 89: 343.
Lazaridis, Iosif, Anna Belfer-Cohen, Swapan Mallick, Nick Patterson, Olivia Cheronet, Nadin Rohland, Guy Bar-Oz,
Ofer Bar-Yosef, Nino Jakeli, Eliso Kvavadze, David Lordkipanidze, Zinovi Matzkevich, Tengiz Meshveliani,
Brendan J. Culleton, Douglas J. Kennett, Ron Pinhasi, and David Reich. 2018. Paleolithic DNA from the Caucasus
Reveals Core of West Eurasian Ancestry. BioRxiv, January, 423079: doi: https://doi.org/10.1101/423079.
Lazaridis, Iosif, Alissa Mittnik, Nick Patterson, Swapan Mallick, Nadin Rohland, Saskia Pfrengle, Anja Furtwangler,
Alexander Peltzer, Cosimo Posth, Andonis Vasilakis, P.J.P. McGeorge, Eleni Konsolaki-Yannopoulou, George
Korres, Holley Martlew, Manolis Michalodimitrakis, Mehmet 6zsait, Nesrin 6zsait, Anastasia Papathanasiou,
Michael Richards, Songul Alpaslan Roodenberg, Yannis Tzedakis, Robert Arnott, Daniel M. Fernandes, Jeffery R.
Hughey, Dimitra M. Lotakis, Patrick A. Navas, Y annis Maniatis, John A. Stamatoyannopoulos, Kristin
Stewardson, Philipp Stockhammer, Ron Pinhasi, David Reich, Johannes Krause, and George Stamatoyannopoulos.
2017. Genetic Origins of the Minoans and Mycenaeans. Nature 548(7666): 214-218.
Lazaridis, Iosif, Dani Nadel, Gary Rollefson, Deborah C. Merrett, Nadin Rohland, Swapan Mallick, Daniel Fernandes,
Mario Novak, Beatriz Gamarra, Kendra Sirak, Sarah Connell, Kristin Stewardson, Eadaoin Hamey, Qiaomei Fu,
Gloria Gonzalez-Fortes, Eppie R. Jones, Songul Alpaslan Roodenberg, Gyorgy Lengyel, Fanny Bocquentin, Boris
Gasparian, Janet M. Monge, Michael Gregg, Vered Eshed, Ahuva-Sivan Mizrahi, Christopher Meiklejohn, Fokke
Gerritsen, Luminita Bejenaru, Matthias Bluher, Archie Campbell, Gianpiero Cavalleri, David Comas, Philippe
Froguel, Edmund Gilbert, Shona M. Kerr, Peter Kovacs, Johannes Krause, Darren McGettigan, Michael Merrigan,
D. Andrew Merriwether, Seamus O'Reilly, Martin B. Richards, Omella Semino, Michel Shamoon-Pour, Gheorghe
Stefanescu, Michael Stumvoll, Anke Tonjes, Antonio Torroni, James F. Wilson, Loic Yengo, Nelli A.
Hovhannisyan, Nick Patterson, Ron Pinhasi, and David Reich. 2016. Genomic Insights into the Origin of Farming
in the Ancient Near East. Nature 536(7617): 419-424.
Lazaridis, Iosif, Nick Patterson, Alissa Mittnik, Gabriel Renaud, Swapan Mallick, Karola Kirsanow, Peter H.
Sudmant, Joshua G. Schraiber, Sergi Castellano, Mark Lipson, Bonnie Berger, Christos Economou, Ruth
Bollongino, Qiaomei Fu, Kirsten I. Bos, Susanne Nordenfelt, Heng Li, Cesare de Filippo, Kay Prufer, Susanna
Sawyer, Cosimo Posth, Wolfgang Haak, Fredrik Hallgren, Elin Pomander, Nadin Rohland, Dominique Delsate,
Michael Francken, Jean-Michel Guinet, Joachim Wahl, George Ayodo, Hamza A. Babiker, Graciela Bailliet,
Elena Balanovska, Oleg Balanovsky, Ramiro Barrantes, Gabriel Bedoya, Haim Ben-Ami, Judit Bene, Fouad
Berrada, Claudio M. Bravi, Francesca Brisighelli, George B. J. Busby, Francesco Cali, Mikhail Churnosov, David
E. C. Cole, Daniel Corach, Larissa Damba, George van Driem, Stanislav Dryomov, Jean-Michel Dugoujon,
Sardana A. Fedorova, hene Gallego Romero, Marina Gubina, Michael Hammer, Brenna M. Henn, Tor Hervig,
U gur Hodoglugil, Aashish R. Jha, Sena Karachanak-Yankova, Rita Khusainova, Elza Khusnutdinova, Rick Kittles,
Toomas Kivisild, William Klitz, Vaidutis Kucinskas, Alena Kushniarevich, Leila Laredj, Sergey Litvinov,
Theologos Loukidis, Robert W. Mahley, Bela Melegh, Ene Metspalu, Julio Molina, Joanna Mountain, Klemetti
Nakkalajarvi, Desislava Nesheva, Thomas Nyambo, Ludmila Osipova, Juri Parik, Fedor Platonov, Olga Posukh,
Valentino Romano, Francisco Rothhammer, Igor Rudan, Ruslan Ruizbakiev, Hovhannes Sahakyan, Antti
Sajantila, Antonio Salas, Elena B. Starikovskaya, Ayele Tarekegn, Draga Toncheva, Shahlo Turdikulova, Ingrida
Uktveryte, Olga Utevska, Rene Vasquez, Mercedes Villena, Mikhail Voevoda, Cheryl A. Winkler, Levon
Yepiskoposyan, Pierre Zalloua, Tatijana Zemunik, Alan Cooper, Cristian Capelli, Mark G. Thomas, Andres Ruiz-
Linares, Sarah A. Tishkoff, Lalji Singh, Kumarasamy Thangaraj, Richard Villems, David Comas, Rem Sukemik,
Mait Metspalu, Matthias Meyer, Evan E. Eichler, Joachim Burger, Montgomery Slatkin, Svante Paabo, Janet
Kelso, David Reich, and Johannes Krause. 2014. Ancient Human Genomes Suggest Three Ancestral Populations
for Present-Day Europeans. Nature 513(7518): 409-13.
Malmstrom, Helena, Jan Stora, Love Dalen, Gunilla Holmlund, and Anders Gotherstrom. 2005. Extensive Human
DNA Contamination in Extracts from Ancient Dog Bones and Teeth. Molecular Biology and Evolution 22(10):
2040-2047.
Marchi, Nina, Laura Winkelbach, Ilektra Schulz, Maxime Brami, Zuzana Hofinanova, Jens Blocher, Carlos S. Reyna-
Blanco, Yoan Diekmann, Alexandre Thiery, Adamandia Kapopoulou, Vivian Link, Valerie Piuz, Susanne
Kreutzer, Sylwia M. Figarska, Elissavet Ganiatsou, Albert Pukaj, Necmi Karul, Fokke Gerritsen, Joachim Pechtl,
Joris Peters, Andrea Zeeb-Lanz, Eva Lenneis, Maria Teschler-Nicola, Sevasti Triantaphyllou, Sofija Stefanovi6,
Christina Papageorgopoulou, Daniel Wegmann, Joachim Burger, and Laurent Excoffier. 2020. The Mixed Genetic
Origin of the First Farmers of Europe. bioRxiv: doi: https:!/doi.org/10.11 0l/2020.11.23.394502.
Massa, Michele. 2016. Networks before Empires: Cultural Transfers in West and Central Anatolia During the Early
Bronze Age. Ph.D. dissertation, University College London.
Mathieson, Iain, Iosif Lazaridis, Nadin Rohland, Swapan Mallick, Nick Patterson, Songul Alpaslan Roodenberg,
Eadaoin Harney, Kristin Stewardson, Daniel Fernandes, Mario Novak, Kendra Sirak, Cristina Gamba, Eppie R.
Jones, Bastien Llamas, Stanislav Dryomov, Joseph Pickrell, Juan Luis Arsuaga, Jose Maria Bermudez de Castro,
Eudald Carbonell, Fokke Gerritsen, Aleksandr Khokhlov, Pavel Kuznetsov, Marina Lozano, Harald Meller, Oleg
Mochalov, Vyacheslav Moiseyev, Manuel A. Rojo Guerra, Jacob Roodenberg, Iosep Maria Verges, Johannes
Krause, Alan Cooper, Kurt W. Alt, Dorcas Brown, David Anthony, Carles Lalueza-Fox, Wolfgang Haak, Ron
Pinhasi, and David Reich. 2015. Genome-Wide Patterns of Selection in 230 Ancient Eurasians. Nature 528(7583):
499-503.
Mathieson, Iain, Songill Alpaslan-Roodenberg, Cosimo Posth, Anna Szecsenyi-Nagy, Nadin Rohland, Swapan
Mallick, Inigo Olalde, Nasreen Broomandkhoshbacht, Francesca Candilio, Olivia Cheronet, Daniel Fernandes,
Matthew Ferry, Beatriz Gamarra, Gloria Gonzalez Fortes, Wolfgang Haak, Eadaoin Hamey, Eppie Jones, Denise
Keating, Ben Krause-Kyora, Isil Kucukkalipci, Megan Michel, Alissa Mittnik, Kathiin Nagele, Mario Novak,
Jonas Oppenheimer, Nick Patterson, Saskia Pfrengle, Kendra Sirak, Kristin Stewardson, Stefania Vai, Stefan
Alexandrov, Kurt W. Alt, Radian Andreescu, Dragana Antonovic, Abigail Ash, Nadezhda Atanassova, Krum
Bacvarov, Mende Balazs Gusztav, Herve Bocherens, Michael Bolus, Adina Boroneant, Yavor Boyadzhiev, Alicja
Budnik, Josip Burmaz, Stefan Chohadzhiev, Nicholas J. Conard, Richard Cottiaux, Maja Cuka, Christophe
Cupillard, Dorothee G. Drucker, Nedko Elenski, Michael Francken, Borislava Galabova, Georgi Ganetsovski,
Bernard Gely, Tamas Hajdu, Veneta Handzhyiska, Katerina Harvati, Thomas Higham, Stanislav Iliev, Ivor
Jankovic, Ivor Karavanic, Douglas J. Kennett, Darko Komso, Alexandra Kozak, Damian Labuda, Martina Lari,
Catalin Lazar, Maleen Leppek, Krassimir Leshtakov, Domenico Lo Vetro, Dzeni Los, Ivaylo Lozanov, Maria
Malina, Fabio Martini, Kath Mcsweeney, Harald Meller, Marko Mendusic, Pavel Mirea, Vyacheslav Moiseyev,
Vanya Petrova, T. Douglas Price, Angela Simalcsik, Luca Sineo, Mario Slaus, Vladimir Slavchev, Petar Stanev,
Andrej Starovic, Tamas Szeniczey, Sahra Talamo, Maria Teschler-Nicola, Corinne Thevenet, Ivan Valchev,
Frederique Valentin, Sergey Vasilyev, Fanica Veljanovska, Svetlana Venelinova, Elizaveta Veselovskaya, Bence
Viola, Cristian Virag, Josko Zaninovic, Steve Zauner, Philipp W. Stockhammer, Giulio Catalano, Raiko Kraul3,
David Caramelli, Gunita Zar~a, Bisserka Gaydarska, Malcolm Lillie, Alexey G. Nikitin, Inna Potekhina,
Anastasia Papathanasiou, Dusan Boric, Clive Bonsall, Johannes Krause, Ron Pinhasi, and David Reich. 2018. The
Genomic History of Southeastern Europe. Nature 555(7695): 197-203.
Mathieson, Iain and Aylwyn Scally. 2020. What is Ancestry? PLoS Genetics 16.3: el008624;
(https://doi.org/10.1371/joumal.pgen. l 008624).
Matney, Timothy, Guillermo Algaze, M.C. Dulik, 6mm Dilek Erda!, Y1lmaz Selim Erda!, Omer Goki;:ilinen, J.
Lorenz, and Hatice Mergen. 2012. Understanding Early Bronze Age Social Structure through Mortuary Remains:
A Pilot aDNA Study from Titri~ Hoyi.ik, Southeastern Turkey. International Journal of Osteoarchaeology 223:
338-351.
Mellink, Machteld J. and J. Lawrence Angel. 1970. Excavations at Karata~-Semayi.ik and Elmah, Lycia, 1969.
American Journal ofArchaeology 74(3): 245-259.
Omrak, Ay,;:a, Torsten Gunther, Cristina Valdiosera, Emma M. Svensson, Helena Malmstrom, Henrike Kiesewetter,
William Aylward, Jan Stora, Mattias Jakobsson, and Anders Gotherstrom. 2016. Genomic Evidence Establishes
Anatolia as the Source of the European Neolithic Gene Pool. Current Biolozy 26(2): 270-275.
Ottoni, Claudio, Frani;:ois-X Ricaut, Nancy Vanderheyden, Nicolas Brucato, Marc Waelkens, and Ronny Decorte.
2011. Mitochondrial Analysis of a Byzantine Population Reveals the Differential Impact of Multiple Historical
Events in South Anatolia. European Journal ofHuman Genetics 19(5): 571-576.
Ottoni, Claudio, Rita Rasteirn, Rinse Willet, Johan Claeys, Peter Talloen, Katrien Van de Vijver, Lounes Chikhi,
Jeroen Poblome, and Ronny Decorte. 2016. Comparing Maternal Genetic Variation Across Two Millennia Reveals
the Demographic History of An Ancient Human Population in Southwest Turkey. Royal Society open science 3.2:
doi: https:// doi.org/10. l 098/rsos.150250.
Ozdogan, Mehmet. 2014. A New Look at the Introduction of the Neolithic Way of Life in Southeastern Europe.
Changing Paradigms of the Expansion of the Neolithic Way of Life. Documenta Praehistorica 41: 33--49.
Palmisano, Alessio, Dan Lawrence, Michelle W. de Gruchy, Andrew Bevan, and Stephen Shennan. 2021. Holocene
Regional Population Dynamics and Climatic Trends in the Near East: A First Comparison Using Aichaeo-
Demographic Proxies. Quaternary Science Reviews 252: https://doi.org/10. l 016/j .quascirev.2020.106739.
Paabo, S., J.A. Gifford, and A.C. Wilson. 1988. Mitochondrial DNA Sequences from a 7000-Year Old Brain. Nucleic
Acids Research 16(20): 9775-9787.
Pickrell, Joseph K. and David Reich. 2014. Toward a New History and Geography of Human Genes Informed by
Ancient DNA. Trends in Genetics 30(9): 377-389.
Pilloud, Marin A. and Clark Spencer Larsen. 2011. "Official" and "Practical" Kin: Inferring Social and Community
Structure from Dental Phenotype at Neolithic <;atalhoyi.ik, Turkey. American Journal of Physical Anthropology
145(4): 519-530.
Rasmussen, Morten, Yingrui Li, Stinus Lindgreen, Jakob Skou Pedersen, Anders Albrechtsen, Ida Moltke, Mait
Metspalu, Toomas Kivisild, Ramneek Gupta, Marcelo Bertalan, Kasper Nielsen, M. Thomas P. Gilbert, Yong
Wang, Maanasa Raghavan, Paula F. Campos, Hanne Munkholm Kamp, Andrew S. Wilson, Andrew Gledhill,
Silvana Tridico, Michael Bunce, Eline D. Lorenzen, Jonas Binladen, Xiaosen Guo, Jing Zhao, Xiuqing Zhang,
Hao Zhang, Zhuo Li, Minfeng Chen, Ludovic Orlando, Karsten Kristiansen, Mads Bak, Niels Tommerup,
Christian Bendixen, Tracey L. Pierre, Bjarne Gronnow, Morten Meldgaard, Claus Andreasen, Sardana A.
Fedorova, Ludmila P. Osipova, Thomas F.G. Higham, Christopher Bronk Ramsey, Thomas v. 0. Hansen, Finn C.
Nielsen, Michael H. Crawford, Soren Brunak, Thomas Sicheritz-Ponten, Richard Villems, Rasmus Nielsen,
Anders Krogh, Jun Wang, and Eske Willerslev. 2010. Ancient Human Genome Sequence of an Extinct Palaeo-
Eskimo. Nature 463(7282): 757-762.
Reich, David, Richard E Green, Martin Kircher, Johannes Krause, Nick Patterson, Eric Y. Durand, Bence Viola,
Adrian W. Briggs, Udo Stenzel, Philip L.F. Johnson, Tomislav Maricic, Jeffrey M. Good, Tomas Marques-Bonet,
Can Alkan, Qiaomei Fu, Swapan Mallick, Heng Li, Matthias Meyer, Evan E. Eichler, Mark Stoneking, Michael
Richards, Sahra Talamo, Michael V. Shunkov, Anatoli P. Derevianko, Jean-Jacques Hublin, Janet Kelso,
Montgomery Slatkin, and Svante Paabo. 2010. Genetic History of an Archaic Hominin Group from Denisova Cave
in Siberia. Nature 46(7327): 1053-1060.
Ricaut, Frarn;:ois-Xavier and Marc Waelkens. 2008. Cranial Discrete Traits in a Byzantine Population and Eastern
Mediterranean Population Movements. Human Biolozy 80(5): 535-564.
Richards, Martin, Vincent Macaulay, Eileen Hickey, Emilee Vega, Bryan Sykes, Valentina Guida, Chiara Rengo,
Daniele Sellitto, Fulvio Cruciani, Toomas Kivisild, Richard Villems, Mark Thomas, Serge Rychkov, Oksana
Rychkov, Yuri Rychkov, Mukaddes Golge, Dimitar Dimitrov, Emmeline Hill, Dan Bradley, Valentino Romano,
Francesco Cali, Giuseppe Vona, Andrew Demaine, Surinder Papiha, Costas Triantaphyllidis, Gheorghe
Stefanescu, Jil'i Hatina, Michele Belledi, Anna Di Rienzo, Ariella Oppenheim, Soren Norby, Nadia Al-Zaheri,
Silvana Santachiara-Benerecetti, Rosaria Scozzari, Antonio Torroni, and Hans-Jurgen Bandelt. 2000. Tracing
European Founder Lineages in the Near Eastern mtDNA Pool. The American Journal ofHuman Genetics 67(5):
1251-1276.
Ringbauer, Harald, John Novembre, and Matthias Steinriicken. 2020. Human Parental Relatedness Through Time-
Detecting Runs ofHomozygosity in Ancient DNA. bioRxiv: doi: https://doi.org/10.1101/2020.05.3 l.126912.
Rohland, Nadin, Isabelle Glocke, Ayinuer Aximu-Petri, and Matthias Meyer. 2018. Extraction of Highly Degraded
DNA from Ancient Bones, Teeth and Sediments for High-Throughput Sequencing. Nature Protocols 13(11):
2447-2461.
Rothman, Mitchell S. 2015. Early Bronze Age Migrants and Ethnicity in the Middle Eastern Mountain Zone.
Proceedings ofthe National Academy ofSciences 112(30): 9190-9195.
Russell, Josiah Cox. 1958. Late Ancient and Medieval Population. Transactions of the American Philosophical
Society 48(3): 1-152.
Sagona, Antonio G. 2000. Sos Hoyuk and the Erzurum region in late prehistory: a provisional chronology for
northeast Anatolia In Chronologies des Pays du Caucase et de l 'Euphrate aux IVe-lIIe Millenaires: Actes du
Collogue International, Istanbul, 16-19 Decembre 1998, C. Marro and H. Hauptmann eds., 329-373. Varia
Anatolica XI. Paris: Institut Frani;:ais d'Etudes Anatoliennes d'Istanbul.
Sagona, Antonio, G. and Paul Zimansky. 2015. Ancient Turkey. London: Routledge.
Skoglund, Pontus, Helena Malmstrom, Ayi;:a Omrak, Maanasa Raghavan, Cristina Valdiosera, Torsten Giirlther, Per
Hall, Kristiina Tambets, Juri Parik, Karl-Goran Sjogren, Jan Apel, Eske Willerslev, Jan Stora, Anders
Gotherstrom, and Mattias Jakobsson. 2014. Genomic Diversity and Admixture Differs for Stone-Age
Scandinavian Foragers and Farmers. Science 344(6185): 747 LP-750.
Skourtanioti, Eirini, Y1lmaz Selim Erdal, Marcella Frangipane, Francesca Balossi Restelli, K. Ashhan Y ener, Frances
Pinnock, Paolo Matthiae, Rana Ozbal, Ulf-Dietrich Schoop, Farhad Guliyev, Tufan Akhundov, Bertille Lyonnet,
Emily L. Hammer, Selin E.Nugent, Marta Burri, Gunnar U. Neumann, Sandra Penske, Tara Ingman, Murat Akar,
Rula Shafiq, Giulio Palumbi, Stefanie Eisenmann, Marta D'Andrea, Adam B. Rohrlach, Christina Warinner,
Choongwon Jeong, Philipp W. Stockhammer, Wolfgang Haak, and Johannes Krause. 2020. Genomic History of
Neolithic to Bronze Age Anatolia, Northern Levant, and Southern Caucasus. Cell 181(5): l 158-l 175.e28.
Steadman, Sharon R., Benjamin S. Arbuckle, and Gregory McMahon. 2018. Pivoting East: <;adrr Hoyuk,
Transcaucasia, and complex connectivity in the Late Chalcolithic. Documenta Praehistorica XLV: 64-84.
Stone, Anne C. and Mark Stoneking. 1993. Ancient DNA from a Pre-Columbian Amerindian Population. American
Journal ofPhysical Anthropology 92(4): 463-471.
Stoneking, Mark and Johannes Krause. 2011. Leaming about Human Population History from Ancient and Modem
Genomes. Nature Reviews Genetics 12(9): 603-614.
~enyurek, Muzaffer Suleyman. 1949. Anadolu'nun eski sakinlerinde buyuk az1 di~lerinin a~1nmas1. (The attrition of
molars in the ancient inhabitants of Anatolia). Belleten 50: 229-244.
- . 1950. Study of the Skeleton of a Chalcolithic Age Warrior from Buyuk Gullucek. Ankara Universitesi Di! ve
Tarih-Cografya Fakultesi Dergisi 8(3): 269-310.
~enyurek, Muzaffer Suleyman and Seniha Tunakan. 1951. The Skeletons from ~eyh Hoyuk. Belleten 60: 431-445.
Ta~krran, Harun. 2008. Reflexions sur l'Acheuleen d'Anatolie -Thoughts about the Acheulean in Anatolia.
L'Antropologie 112(1): 140-152.
Ustiirldag, Handan and Goki;:e Bike Yaz1c10glu. 2014. The History of Physical Anthropology in Turkey. In
Archaeological Human Remains: Global Perspectives, B. O'Donnabhain and M. Lozada, eds., 199-211. Cham:
Springer.
Wang, Chuan-Chao, Sabine Reinhold, Alexey Kalmykov, Antje Wissgott, Guido Brandt, Choongwon Jeong, Olivia
Cheronet, Matthew Ferry, Eadaoin Hamey, Denise Keating, Swapan Mallick, Nadin Rohland, Kristin Stewardson,
Anatoly R. Kantorovich, Vladimir E. Maslov, Vladimira G. Petrenko, Vladimir R. Erlikh, Biaslan Ch. Atabiev,
Rabadan G. Magomedov, Philipp L. Kohl, Kurt W. Alt, Sandra L. Pichler, Claudia Gerling, Harald Meller, Benik
Vardanyan, Larisa Yeganyan, Alexey D. Rezepkin, Dirk Mariaschk, Natalia Berezina, Julia Gresky, Katharina
Fuchs, Corina Knipper, Stephan Schiffels, Elena Balanovska, Oleg Balanovsky, Iain Mathieson, Thomas Higham,
Yakov B. Berezin, Alexandra Buzhilova, Viktor Trifonov, Ron Pinhasi, Andrej B. Belinskij, David Reich, Svend
Hansen, Johannes Krause, and Wolfgang Haak. 2019. Ancient Human Genome-Wide Data from a 3000-Year
Interval in the Caucasus Corresponds with Eco-Geographic Regions. Nature Communications 10(1 ): 1-13.
Yaka, Reyhan, Ay~egill Birand, Yasemin Y1lmaz, Ceren Caner, Sinan Can As:an, Sidar Gundilzalp, Poorya Parvizi,
Ash Erim Ozdogan, inci Togan, and Mehmet Some!. 2018. Archaeogenetics of Late Iron Age <;emialo Sut1,
Batman: Investigating Maternal Genetic Continuity in North Mesopotamia Since the Neolithic. American Journal
ofPhysical Anthropology 166(1): 196-207.
Yaka, Reyhan, Igor Mapelli, Damla Kaptan, Ays:a Dogu, Maciej Chylenski, Omur Dilek Erda!, Dilek Koptekin,
K!Vllc1m Ba~ak Vural, Alex Bayliss, Camilla Mazzucato, Evrim Fer, Sevim Seda <;okoglu, Vendela Kempe
Lagerholm, Maja Krzewinska, Cansu Karamurat, Hasan Can Gemici, Arda Sevkar, Nihan Dil~ad Dagt~, Gi.il~ah
Merve K1Ims:, Donovan Adams, Arielle R. Munters, Ekin Saghcan, Marco Milella, Eline M.J. Schotsmans, Erins:
Yurtman, Mehmet <;etin, Sevgi Yorulmaz, N. Ezgi Altm1~1k, Ayshin Ghalichi, Arma Juras, C. Can Bilgin, Torsten
Gunther, Jan Stora, Mattias Jakobsson, Maurice de Kleijn, Gokhan Mustafaoglu, Andrew Fairbairn, Jessica
Pearson, inci Togan, Nurcan Kayacan, Arkadiusz Marciniak, Clark Spencer Larsen, Ian Hodder, <;igdem
Atakuman, Marin Pilloud, Elif Surer, Fokke Gerritsen, Rana Ozbal, Douglas Baird, Y1lmaz Selim Erda!, Gune~
Duru, Mihriban Ozba~aran, Scott D. Haddow, Christopher J. Knilsel, Anders Gotherstrom, Filsun Ozer, and
Mehmet Some!. 2021. Variable Kinship Patterns in Neolithic Anatolia Revealed by Ancient Genomes. Current
Biology 31(11): 2455-2468.
Yunusbayev, Bayazit, Mait Metspalu, Ene Metspalu, Albert Valeev, Sergei Litvinov, Ruslan Valiev, Vita Akhmetova,
Elena Balanovska, Oleg Balanovsky, Shahlo Turdikulova, Dilbar Dalimova, Pagbajabyn Nymadawa, Ardeshir
Bahmanimehr, Hovhannes Sahakyan, Kristiina Tambets, Sardana Fedorova, Nikolay Barashkov, Irina
Khidiyatova, Evelin Mihailov, Rita Khusainova, Larisa Damba, Miroslava Derenko, Boris Malyarchuk, Ludmila
Osipova, Mikhail Voevoda, Levon Yepiskoposyan, Toomas Kivisild, Elza Khusnutdinova, and Richard Villems.
2015. The Genetic Legacy of the Expansion of Turkic-Speaking Nomads Across Eurasia. PLoS Genetics 11.4:
https://doi.org/10.1371/joumal.pgen.1005068.
Yurtman, Erins:, Onur Ozer, Eren Yuncil, Nihan Dil~ad Dagta~, Dilek Koptekin, Yasin Gokhan <;akan, Mustafa
Ozkan, Ali Akbaba, Damla Kaptan, Gozde Atag, KlVllcim B~ak Vural, Can Yilmni Gundem, Louise Martin,
Gill~ah Merve K1Ims:, Ayshin Ghalichi, Sinan Can As:an, Reyhan Yaka, Ekin Saghcan, Vendela Kempe
Lagerholm, Maja Krzewinska, Evangelia Pi~kin, Milge ~evketoglu, C. Can Bilgin, <;igdem Atakuman, Y1lmaz
Selim Erda!, Elif Surer, Johannes Lenstra, Sevgi Yorulmaz, Foad Abazari, Javad Hoseinzadeh, Douglas Baird,
Erhan B1yaks:1, Ozlem <;evik, Fokke Gerritsen, Rana Ozbal, Anders Gotherstrom, Mehmet Some!, inci Togan, and
Filsun Ozer. 2020. Archaeogenetic Analysis of Neolithic Sheep from Anatolia Suggests a Complex Demographic
History since Domestication. BioRxiv: doi: https://doi.org/10.1101/2020.04.17.033415.
CHAPTER TWENTY-EIGHT
CHALCOLITHIC RELIGION AND RITUAL ON THE ANATOLIAN PLATEAU
BURCU YILDIRIM AND SHARON R. STEADMAN
The Anatolian Neolithic boasts sites that offer fascinating glimpses into ritual and religious behaviors (Baird 2012;
Boivin 2000; Lichter 2016; M. 6zdogan 2014; Yakar 2003; see 6zba~aran 2011 for a review of the Neolithic plateau).
The same cannot be said about the Chalcolithic period, particularly on the Anatolian plateau. The last decade or so has
seen a number of excellent syntheses of data regarding what we do and do not know about the Chalcolithic period
across this region (e.g., During 2011; Gemici and Atakuman 2021; Gul9ur 2012; Schoop 2005a, 2005b, 2010, 201 la).
Great strides in illuminating ceramic and architectural traditions, settlement patterns, and more closely pinpointing an
apparent habitational (and cultural?) break between the Early and Middle Chalcolithic in the early 6th millennium in
western central Anatolia (Biehl 2012; <;::evik and Erdogu 2019, 2020; Naumov and Biehl 2019; Turkcan 2017a;
Vandam et al. 2019; Vandam and Willett 2020), have advanced our understanding of the period. Attention to religious
and ritual practices in the same period, and for the same region, however, has been largely lacking (a few excellent
studies are noted below). This is due to a smaller number of excavated sites, but also to the general lack of a thorough
review of Chalcolithic religion and ritual. This chapter serves as an initial step to rectify this lacuna.
The bulk of the data presented here is drawn from sites on the Anatolian plateau, the confines of which are variable
depending on which author one consults. This study includes excavated sites resting north of the Taurus, and south of
the Pontic Mountain ranges, as far west as the Eski~ehir and Lake District regions (Burdur), and as far east as the
Sivas region. The fifteen sites comprising the database (Fig. 28-1; Table 28-1) were substantially excavated and
(sometimes) sufficiently published to allow for data-mining, treating the categories outlined in Table 28-1 .
SI
...
<;:adar
......
r Hoyiik
Giivercinkayasi A Gelveri-Yiiksekkilise
...
• Tepecik <;:iftlik
~atalhoyiik West • Ko~k Ho .
Figure 28-1. Map of Anatolian Plateau showing sites included in this chapter (courtesy of Anthony J. Lauricella).
Chalcolithic Religion and Ritual 373
Site Name Period of Occupation Specialized/ Material Ceramics: Burial

Ritual Culture Symbolic (?) Practices
Structures (Ritual/ Decoration
Symbolic)
Alacahoyillc Late Chalcolithic ✓
Ali~r Hoyillc Late Chalcolithic ✓ ✓
<;adrr Hoyillc Late Chalcolithic ✓ ✓ ✓
<;amhbel Tarlas1 Late Chalcolithic ✓ ✓ ✓ ✓
Biiyillc Giilliicek Late Chalcolithic ✓ ✓ ✓
<;atalhoyillc West Early Chalcolithic ✓ ✓
Orman Fidanhg1 Early-Late Chalcolithic ✓ ✓ ✓
Canhasan Early-Middle Chalcolithic ✓ ✓ ✓
Beycesultan Late Chalcolithic ✓ ✓
Hac1lar Early-Middle Chalcolithic ✓ ✓ ✓
Kurrn;ay Hoyillc Late Chalcolithic ✓ ✓ ✓
Ko~k Hoyillc Early Chalcolithic ✓ ✓ ✓
Gelveri- Middle Chalcolithic ✓ ✓
Yiiksekkilise
Giivercinkayas1 Middle Chalcolithic ✓ ✓ ✓
Tepecik <;iftlik Early Chalcolithic ✓ ✓ ✓
Table 28-1. List of sites included in the study and indication of which categories of data
discussed in the chapter are represented at the site.
The Chalcolithic spans a three-millennium period. As is always the case in archaeology, scholars are not united on
the precise beginning, internal divisions within, and ending of this technological phase. The present study primarily
relies on the periodization presented by Schoop (201 la; see Table 28-2).
Early Chalcolithic ca. 6100- 5500 BCE
Middle Chalcolithic ca. 5500-4250 BCE
Late Chalcolithic ca. 4250- 3000 BCE
Table 28-2. Chalcolithic periodization used in the study (based on Schoop 201 la).
The following sections offer a brief, and consequently incomplete, overview of research on Neolithic and
Chalcolithic religion and ritual. Data from 15 sites are then presented in three categories: examples of extra-household
buildings or areas that may have been dedicated to religious/ritualized behaviors; exan1ples of material culture that
may have been dedicated to symbolic meanings or religious/ritualized behaviors; and evidence from burial practices.
Interpretations of these data, both those presented by excavators as well as by the chapter authors, will be included in
each section.
A BRIEF SUMMARY OF NEOLITHIC RELIGION AND RITUAL
Compared to the Chalcolithic, Neolithic symbolic belief and ritual appears far more rich and complex (e.g., Baird
et al. 2011; Croucher 2012; Gebel et al. 2002; Hodder 2010; Hodder and Meskell 2011 ; Kuijt 2000a; Tiirkcan 2007;
Verhoeven 2002a; Watkins 2015). The evidence for ritual practice during the PPNB (8600-7000 BCE) in the Levant
and southeastern Anatolia is exemplified by large ritual buildings, plastered human skulls, stone masks, stelae
decorated with animal reliefs, and human statuary. Buildings and objects suggest the existence of a complex ritual
system. Added to these is the element of mortuary practices spanning at least a millennium at sites across the Levant
and Anatolia (Kuijt 2000b, 2008; Verhoeven 2002b), with a focus, at the very least, on ancestor veneration.
Some of the most evocative Neolithic discoveries are the cached or individual skulls and headless skeletons
retrieved from early Levantine sites, and from Anatolian sites dating from 7000 BCE to the second half of the 011
millennium BCE. Some skulls were modeled with plaster, marl, or animal collagen, while some were decorated with
paint, red ochre, or shell (Bonogofsky 2003, 2006). The archaeological contexts of plastered and undecorated skulls
vary, and they belong to all age groups and sexes; those in burials are occasionally accompanied by offerings. The
skulls have been found both in groups or alone, sometimes accompanied by funerary offerings (Bonogofsky 2003,
2006). The widespread nature of these practices has led scholars to suggest that a Neolithic "skull cult" existed across
the region (Bienert 1991; Gresky et al. 2017; Kuijt 2008; Lichter 2016; Verhoeven 2011). Other Neolithic burial
customs also provide evidence of symbolic beliefs and ritual practices (e.g., Bonogofsky 2005; Brami 2014; During
2003; Haddow et al. 2016; Ozbek 2009; Russell and During 2006). For instance, the intramural burial of adult
374 Chapter Twenty-Eight: Y1ldmm and Steadman
individuals under house floors was a common praxis in central Anatolia, as well as at sites in southeastern Anatolia,
and the Levant.
The existence of monumental architecture, T-shaped pillars, and large stone sculptures, found at a number of sites
across 9th millennium BCE Anatolia (Atakuman 2014; Aurenche 1981; B1s;aks;1 2003; <;::elik 2004; Hauptmann 1999;
Kami 2011; A. Ozdogan 2011; Rosenberg 201 la, 2011b; Schirmer 1990; Schmidt 2011; Tatsumi 2020; and see Okse,
this volume) are much-discussed examples of Neolithic religion and ritual activity, often described as communal in
nature. Extensive scholarship has been devoted to the meaning of these constructions ( e.g., Atakuman 2014; Banning
2011; Dietrich et al. 2012; Dietrich and Notroff 2015; Peters and Schmidt 2004; Rosenberg and Erim-Ozdogan 2011;
Sweatman and Tsikritsis 2017; and see Erdogu et al. this volume). Undoubtedly, the shared systems of symbols at
these locations, including accompanying figurines, amulets, and seals, served as a type of social contract that regulated
changing societal needs (Rappaport 1999; Atakuman 2014).
As the Neolithic progressed into the Pottery Neolithic phase, ritual behavior veered away from communal practice
to center more on domestic architecture, perhaps representing a shift in social relations from community to household
(Watkins 1990, 2004; Flannery 2002). The focus moved to the residence, including burials under floors, the presence
of human and animal figurines, and the decoration of house interiors and ceramics with lively and complex symbols
(e.g., Baird et al. 2017; Croucher and Belcher 2017; During and Marciniak 2006; Marciniak and Czerniak 2007;
Pilloud et al. 2020; Voigt 2007; Y akar 2003). While some of these practices do continue into the Early Chalcolithic
period, as discussed below, there is yet another shift in ritual behavior not long after the Chalcolithic commences.
RELIGION AND RITUAL IN THE CHALCOLITHIC: PREVIOUS RESEARCH
There is a small but important collection of studies that have addressed topics regarding specific elements of
religion, religious symbolism, and ritual behavior in the Chalcolithic. Several scholars have suggested that ritual
activity, more commonly practiced in partially or wholly public buildings in the Neolithic, was transferred to activities
involving portable objects in the Chalcolithic, including figurines, ceramics, and other more unique items, located in
domestic or non-public areas (Erdogu 2009a; Hodder 2006; Last 1998). The data below suggest that there is validity
in this observation, which may help to explain why identifying Chalcolithic ritual behavior has been a far more elusive
endeavor.
A number of studies have noted the importance of the color red, which is derived from an ochre-based paint which
was used to paint images on the walls of Neolithic structures; it also appears frequently in painted decoration on
Chalcolithic ceramics and in a few other contexts. Researchers have noted that color symbolism in general, and the
color red in particular, likely carried sacred meaning in ancient Anatolia (Duru and Ozba~aran 2005; Erdogu and
Ulubey 2011; Hodder 2006; Mellaart 1967; Ozb~aran 2012; and see Gage 1999; Verhoeven 2002b).
A number of studies have noted the rich symbolism expressed on decorated ceramics, by figurines, and in other
media (e.g., Atakuman 2015; Bischoff 2002; Hodder 2006; Kolankaya-Bostanci 2014; Naumov and Biehl 2019;
Yakar 2003; and see Biehl 1996; Renfrew 1998). While attempting to identify the exact meanings such images may
have held for those living thousands of years ago, the theories put forth by scholars are based on contextualization of
what was critical to survival and maintenance of daily life in prehistoric times in their relative settings.
RELIGIOUS/RITUALIZED NON-DOMESTIC BUILDINGS AND AREAS
There are few non-domestic Chalcolithic-period buildings that can be considered associated with ritual behavior.
There may be a variety ofreasons for the dearth of such structures; as noted above, several scholars have argued that
the Neolithic was the "age" of religious buildings and that ritual symbolism in the Chalcolithic shifted to portable
objects (Erdogu 2009a; Hodder 2006; Last 1998). Another explanation is rooted in the "luck of the draw" nature of
archaeology. The fewer number of extensively excavated Chalcolithic sites on the plateau may simply have failed to
reveal such buildings as yet. The following section includes descriptions of structures ranging from the Early to Late
Chalcolithic, some of which are identified by their excavators as buildings with ritual elements and others suggested to
be so by the present authors.
Hacilar
The Hac1lar IIA level was assigned to the Early Chalcolithic by the excavator (Mellaart 1970), though these
stratigraphic ascriptions have been questioned (see During 2011: 171; Umurtak 2011). The Hac1lar IIA community
features two structures identified as "shrines" by the excavator (Mellaart 1970). One is a three-roomed structure in the
southwestern comer of the walled settlement (Fig. 28-2a), which rests at the highest point of the Hac1lar IIA
community. Within the three rooms, furniture included a bread oven and additional ovens/hearths, a pebbled surface
(Fig. 28-2a: see #1 on plan), a storage bin, and a wall niche containing a stone slab (Mellaart 1970: 29-30). Contents
in this structure included fragments of painted wall plaster, numerous examples of painted pottery, painted figurines, a
clay seal, and vessels identified as of ritual type, such as a portion of a cup in the form of a human head, and vessels
decorated with humans and animals (Mellaart 1970: 30). While the painted plaster, figurines (see below), and vessels
do set this structure apart from others in Hacilar IIA, it is equally possible that it was domestic in nature, perhaps
belonging to an important village personage (Steadman 2000: 184).
The second "shrine" can be found in the opposite northeastern comer of the Hacilar IIA settlement (Fig. 28-2a).
This large structure includes a complex of spaces, including a well in the eastern open area. Like the structure in the
southwestern comer of Hacilar IIA, it contained fire installations, clay bins, and wall niches (Mellaart 1970: 35). A
small room in the northwest comer contained a 24 x 6 cm limestone slab within a niche, standing upright (Fig. 28-2a:
see #2 on plan); in front of this were two clay-lined "oval holes," a bowl with a pouring spout contained in one; the
fl:J
J.._ ....., ~
X
... Well
• +,, -
Li
a I
A
-=-=-
0 Sm
Hearth/Oven
□ Limestone Slab
■ Mudbrick Architecture
■ Possible Specialized
Structure
D Bench
• Posthole
X Human Remains/Burial
b
Figure 28-2. a) Plan ofHacilar IIA with "shrines" highlighted in red (after Mellaart 1970: 26-27, fig. 20); b) Plan of
Canhasan I, Layer 2B, with possible specialized Structure 3 highlighted in red (after French 1998: 28, fig. 11).
excavator identifies this room as an area serving cult purposes (Mellaart 1970: 35-36). A sudden fire that destroyed
this quarter of the settlement appears to have caught a person tending a hearth in the center of the structure unawares;
there were, however, several intentional burials under the floor of the central area of this structure, including a double
grave containing a female and a child, and a single burial, both with grave goods. These constitute the only intramural
burials in the Hacilar IIA settlement (Mellaart 1970: 36, 88). The unusual northeastern room with the stone slab, the
burials, and position adjacent to the village water source, suggest that the excavator may be accurate in associating a
ritual identity to this structure (but see Umurtak 2011: 4).
376 Chapter Twenty-Eight: Yildmm and Steadman
Canhasan
Canhasan I, Layer 2B (Early/Middle Chalcolithic, but see During 2011: 142) presents an interesting case. There
are ten structures in Layer 2B, consistently termed either houses or structures by the excavator; most or all feature two
stories (French 1962, 1963, 1966, 1998). Structure 3 (Fig. 28-2b), is two-roomed and somewhat larger than most of
the other structures. Structure 3 had ground-floor benches, architectural features that are somewhat uncommon at
Canhasan (French 1998: 68). In addition to the benches, another architectural feature setting Structure 3 apart was the
presence of painted plaster walls, likely on the upper story (French 2010: 159). Structure 3 also contained a significant
number of in situ human figurines (see below; French 2010: 3), as well as other unusual items such as a copper/bronze
mace-head, a bone-bead necklace, and a human skeleton wearing a copper/bronze bracelet (French 1962: 33, 1963:
36). Finally, a large open dish, with painted design apparently matching some of the wall painting patterns (see
below), was also discovered (French 1963: 36). The excavator does not identify Structure 3 as non-domestic but does
consistently remark on the unusual finds within this building. The wall treatments and artifacts that are largely unique
to Structure 3 suggest that it could have served more of a ritual than domestic purpose at Canhasan.
Kuru<;ay
The Lake District site of Kuru9ay Hoyuk features both Neolithic and Chalcolithic levels, the latter consisting of
levels 6A-3 and dated by the excavator to the Late Chalcolithic. Structure 8 in the center of the 6A settlement was
identified as "something other than a residence," designated as a shrine by the excavator (Duru 1996: 115). Though
the 5 x 7 m architectural shell resembles domestic structures, interior contents persuaded the excavator that this was a
specialized building (Fig. 28-3). Two wooden columns stood in a largely furniture-free room. A fire installation,
c:;
Mudbrick
Architecture
Specialized
-
Structure
□
Fire
Installations
Clay Table & Pit

--- A
Sm
• Posthole
Figure 28-3. Plan ofKuru9ay Level 6A, with

Structure 8 highlighted in red (after Duru 1996: pls. 32, 43).
described as "an altar for burnt sacrifice" (Duru 1996: 115) was set in the center of the space next to a column. Also
contained in the room was an oven and trough in the southwest comer, a hearth, and a four-sided clay table with a
plaster-lined pit next to it. The excavator identified this furniture as a "sacrificial table," and the pit was "probably
intended to receive or collect liquid, a libation shaft" (Duru 1996: 118). Near the end of its life this room experienced
some alterations: the main door was bricked up and a storage room was built against it; the excavator notes "the shrine
had been suddenly cut off from use and its contents sealed up untouched" (Duru 1996: 116). Shortly after this
alteration the 6A settlement met with destruction.
Structure 56 in the later level 3 was also identified as having a shrine (Duru 1996: 117). The largest of the three
Structure 56 rooms contained a hearth the excavator termed a "fire altar for burnt sacrifice" (Duru 1996: 118). This
fire installation was 1.23 x 1. 7 m in size, resting 10 cm above the floor; it also had a mudbrick "stele" behind it (Fig.
28-4a). Also located on the east wall of Structure 56 were two clay features, creating what would seem to be "troughs"
but open at one end; these protrude from the wall. The excavator suggested that they had religious significance, but
their purpose is not identified (Duru 1996: 118).
r D Hearth/troughs Mud brick "stele"
---
0
A
Sm
I
Mudbrick ■ Specialized a
■ Architecture
0=-2m
Structure
X Burial
• Posthole/Poststand
D Hearth/Oven/Furniture
D Bin, Pot Emplacement
Figure 28-4. a) Plan ofKuru9ay Level 3, showing Structure 56 highlighted in red (after Duru 1996: pl. 40);
b) Plan of W estem Compound ("Burnt House Phase) at C,::ad1r Hoyiik with specialized structure highlighted in red.
<;ad1r Hoyiik
The north central plateau Late Chalcolithic community at (:ad1r Hoyuk includes a structure excavators named the
"Non-Domestic Building" (Steadman et al. 2015; Steadman and McMahon 2015), which they believe served ritual
functions. The substantial walls enclose a 5 x 5 m space, the largest room so far exposed in the Late Chalcolithic.
Within the structure were two infant burials beneath circular depressions in the floor that functioned as pot
emplacements (Fig. 28-4b ). These two, combined with a third pot emplacement, formed a triangle in front of a unique
mudbrick feature in the center of the building (Yildmm et al. 2018). This feature is a half-circle in form, and as of the
writing of this chapter, stands ca. 1 min height (the lowest level of the feature has not yet been exposed). A posthole
in the top of the half-circle and one just to its west, in the floor and between two rocks, fall in line with the sunrise. To
the sides of the floor-level posthole were deposits of wild and domestic grains. Also discovered within this structure
was a bronze axe (the finest example of metalwork from the site), a crystal amulet, a tiny unbaked clay human
figurine, ceramic "fruitstands," and a portion of an unusual ceramic vessel intended for pouring purposes (Steadman et
al. 2017, 2019b ). As excavations proceeded at the site, it became clear that the mudbrick half-circle feature once stood
in an open area, rather than in an enclosed building. This could correspond with the de-emphasis on ritual buildings
after the Neolithic, and the reorientation to more object-based ritual activities, but may also represent a persistence of
areas designated as "ritualized" in the community (Hackley et al. 2018, 2021). The enclosing of the mudbrick feature
appears to have taken place toward the final centuries of the Late Chalcolithic phase on the Plateau.
<;aml1bel Tarlas1
Not far from the Hittite center of Bogazkoy-ijattusa lies the Late Chalcolithic site of <;amlibel Tarlas1, which
features four major phases of habitation interspersed with more ephemeral occupational periods (Schoop 2011b).
<;amhbel Tarlas1 III saw the construction of a large building built of stone and pise, with two pillars supporting the
roof and a well-made lime plaster floor, which was then covered by another, the latter painted red. Embedded in the
floor of this structure were two cattle bones (Schoop 2009: 60-61). A large plastered courtyard with a stone wall
border was associated with this structure; fire installations including an oven and a furnace were found in the
courtyard (Schoop 2009: 59, 61). In addition to the size and construction of this building, its contents also set it apart.
One unique discovery was a clay mold for making "ring idols," common in southeastern Europe in the 4 th millennium
BCE (Schoop 2009: 65). Also discovered in the building was an anthropomorphic figurine and an animal figurine
head, likely of a bull (Schoop 2015: 55), not dissimilar from one found at C,::adu Hoyilk: (see below). The structure was
eventually largely cleaned out and then (intentionally?) burnt to the ground and left in a heap of ashes (Schoop 2009:
59-61). This entire assemblage encouraged the excavator to suggest that it is "likely that this building served a special
purpose with a ritual component" (Schoop 2015: 55).
Summary
It is perhaps notable that 33% of the sites surveyed in this study yielded possible non-domestic ritualized structures
or spaces. The percentage might have been even higher if the site of Ko~k Hoyilk: had been included, given that the
original excavator identified a building in Early Chalcolithic Level l as "ritualized," due to the presence of two
figurines and bull horns (Silistreli 1989: 95-96, 1990). More recent excavations, however, identify this structure as a
simple residence (Oztan 2003: 70, and see Oztan 2010).
Though 33% of the sample might seem significant, these occurrences are scattered across 3000 years of
occupation, and several, at Hac1lar and Canhasan for instance, can perhaps just as easily have been described as
"houses with special purpose." Further, the unusual furniture and artifacts used to identify these structures as
"ritualized" are not consistent across the sites or periods. At each, inhabitants appear to have found symbolic
resonance in varying types of internal features and objects, perhaps relevant to their own region and life experiences.
Evidence presented here suggests that specifically designated spaces for public or specialist-led ritual activity were not
a priority for Chalcolithic inhabitants on the Anatolian plateau.
MATERIAL CULTURE WITH (POSSIBLE) RITUAL OR SYMBOLIC CONNOTATION
This section was by far the hardest for the authors to compile. Many decisions had to be made, including, most
importantly: "what material culture can be considered to have symbolic value?" Figurines, unusual items and deposits
(such as animal horns), and furniture or buildirlg features of inexplicable function, are included as these are often
noted by their excavators as havirig ritual or symbolic meaning. Other items, such as personal ornaments, were left to
others to discuss (as recently ably done, see Baysal 2017). When possible, the context of included items is noted,
given that this may be key to understanding the performance of household or perhaps non-household-based ritual.
Another critical discussion was the organization of the section: should it be chronological or regional? Both have their
merits. In keeping with the organization of the other sections, a chronological format was followed. The section
concludes with a discussion of unusual ceramic styles that emerged in the Chalcolithic.
Early and Middle Cha/eolithic Sites
<;atalhoyuk West
The largely Early Chalcolithic <;atalhoyillc West mound (late ?ili to mid-6th millennia BCE; Orton et al. 2018)
contains far fewer examples of pairited walls than its earlier sibling <;atalhoyilk: East. The largest structure in the West
Mound, known as B98, which offered evidence ofred-painted walls in what was likely an upper story (Erdogu 2009b,
2010), suggests "a possibility that symbolic elaboration was present" (Anvari et al. 2017: 13). Fragments of red plaster
in building fills suggest that other structures might also have had some wall painting (Biehl et al. 2012: 56).
Also, unlike the Neolithic East mound, the West Mound provides far fewer examples of spaces and items that
might be termed "ritualized." Excavators identified a cache of "clay balls" from Trench 5, located between floors irI
structure B98 (Anvari et al. 2017) as possibly having symbolic value. Excavators also noted the presence of caprine
frontlets (sets of horns) in opposite comers of a room in structure Bl06 (Russell et al. 2013: 64---65), in some ways
echoing the extensive emplacement ofbucrania in the Neolithic East Mound structures.
A total of 86 (clay) figurines were found at c;::atalhoyiik West, but most were from fill contexts or from within
mudbricks; only a few were found within walls (Anvari et al. 2017: 16). Of the 38 animal figurines, the majority of
recognizable animals were cattle (14); one was clearly a goat, nicely constructed, and perhaps representing an
importance of caprids in the Early Chalcolithic. Of the recognizably human figurines (34), many (11) could not be
sexed; the rest were female. The excavators note that these figurines do not feature emphasized sexual characteristics
that were more common on figurines from the Neolithic East Mound (Anvari et al. 2017: 18), perhaps signaling a
different symbolic value for human representation in the Chalcolithic period at the settlement.
Haczlar
Hacilar is well known for its Neolithic period

figurines, which makes it somewhat surprising that
so few were recovered from the Chalcolithic. Only
one intact figurine, and approximately 35 fragments,
were retrieved from the Early/Middle Chalcolithic
levels (V-II) (Mellaart 1970: 177). A review of the
fragments reveals the norm to be standing figures,
with the few head fragments featuring elongated
faces with almond eyes (Fig. 28-5a). Figurines from
Hacilar V are not painted; it is in Hacilar III and II
that painted decoration becomes more common
(Mellaart 1970: 177). The painted decoration
suggests clothing but could be body paint (Fig. 28-
5b). Unlike examples from the Neolithic period at
this site, breasts are far less accentuated on
Chalcolithic figurines; the excavator suggests that
this might reflect the depiction of younger rather
than older women (Mellaart 1970: 178), but the
present authors are not convinced by this explanation.
The intact figurine (Fig. 28-5c) was discovered in
the Hacilar IIB granary, but the find spots of the
fragments are not reported. The lack of find location
for the figurine fragments is unfortunate, since the
preponderance of such could indicate intentional
breakage and disposal, but such suppositions must
e
2cm 2cm remain speculative given the available data.
Figure 28-5. a) Sketch of figurine head from Hacilar V

Canhasan
with almond eyes and incised hair (after Mellaart 1970:
fig. 238.1) (no dimensions available); b) Sketch of
standing painted figurine from Hacilar II (after Mellaart At Early/Middle Chalcolithic Canhasan, the
1970: fig. 242) (no dimensions available); c) Sketch of excavator identifies all structures as houses; the
complete figurine from Hacilar II granary, with orange- unusual nature of Structure 3 and the specific items
brown paint (after Mellaart 1970: fig. 243); d) Sketch of from Structure 3 in layer 2B were described above.
figurine head from Canhasan Layer 2B (after French 2010: These included the bone bead necklace, the human
9, fig. 3-1); e: Sketch of the square dish with paint from figurines with elongated heads (see Figure 28-5d),
Canhasan Layer 2B, Structure 3 (after French 2010: 46, fig the mace head, and the open dish with paint echoing
.32); f: Sketch of painted squatting figurine from Canhasan the plastered wall designs (Fig. 28-5e). In addition to
Layer 2B, Structure 3 (after French 2010: 8, fig. 2). these, a selection of human and animal figurines
were recovered from fill in layers.
Seven baked clay figurines came from stratified contexts in Layer 2B (French 2010: 3); most can be described as
sitting or squatting and feature stylized heads (Fig. 28-5f). Some figurines had painted patterns on the body, and all
featured a head-dress or hair style. The painted patterns could represent make-up, painting, or tattooing on the body, or
decorations on the garments (French 2010: 3). It is notable that two structures in Layer 2B had unusual figurines:
Structure 7 contained a seated or squatting female figurine of stone (French 2010: 3), and a squat, standing, painted
female figurine came from Structure 10 (French 1965: 29, 1968a: 89). There were numerous other figurines recovered
at Canhasan, but unfortunately from unsecured contexts such as fills and the soil dump. Many of these are termed the
"chess-pawn" type by the excavator (French 2010: 3) with heads showing simplified facial features; these were
recovered in Layers 5--4, 2B and 2A, and Layer 1. Three human figurines identified as male (French 2010: 4) also
came from fill contexts. A final set of figurines consists of five elongated heads, perhaps depicting females, featuring
dots and lines that may represent painting or tattooing (French 2010: 4). The elongated heads could represent hair
styles, a headdress, or possibly skull deformation. Unfortunately, these were also all recovered from fill and dump
contexts. Animal figurines, many of them quite crude and some unbaked, were plentiful in Layer 1 (Late Chalcolithic
and later periods) which may indicate that they date to periods after the fourth millennium.
Another somewhat unusual item found both in Structure 3 and elsewhere in the Middle and Late Chalcolithic
layers were what the excavator identified as pot stands/andirons, some in vaguely human forms (French 1964: 126,
1965: 89, 1966: 115).
Similar to fragments from Structure 3, discussed above in the Ritualized Buildings section, some red or red-on-
white patterned plaster fragments were recovered from several structures in Layer 2B, including Structures 7 and 10
(French 1998: 37, 40). It should be noted that other than Structure 3, Structures 7 and 10 appear to feature the most
items that can be identified as possibly having symbolic value, including red-painted plaster on the walls, figurines,
and andirons.
Orman Fidanhgi, Ko~k Hoyuk, Tepecik 9iftlik, and Gelveri-Yuksekkilise
Only one zoomorphic figurine was recovered from Orman Fidanhg1. An additional 17 anthropomorphic figurines
were also discovered. These figurines are all apparently female in a standing position, with arms folded, or without
arms depicted. These human figurines came primarily from Early Chalcolithic levels (Phases I and IV; Efe 2001: 128-
129). Unfortunately, the majority of figurines were recovered from unstratified contexts.
At Early Chalcolithic (level 1) Ko~k HoYlilc, several human and animal figurines, made of stone or baked clay,
were discovered. The majority were recovered from inside or near a structure that was likely domestic in nature
(Oztan and Faydah 2003: 46-48). Some of these figurines appear to have ochre-based paint on them; animals
represented by the figurines include bulls and goats. The excavator believes the human figurines may represent deities,
including a mother goddess (Oztan 2003: 73; Oztan and Faydah 2003: 48).
At Tepecik <;iftlik, four figurines were recovered from the Early Chalcolithic layer 2. Three female figurines had
heads detached from their body and left arms making a motion toward the chest; their necks were decorated with
incised lines and dots. A male figurine, made of basalt, was found inside a structure. This figurine appears to wear
clothing around the waist (Ozbudak 2016: 39).
Several soundings carried out at Middle Chalcolithic Gelveri-Ytiksekkilise revealed a limited view of what appear
to be several residential structures (Ozbudak 2010). Only a few figurines, all baked clay, were recovered. One of these
depicts the head of a female with an oval face and braided hair shown at the back of the head (Ozbudak 2010: 41).
Another fragment of a figurine depicts a foot, and a third appears to be an animal figurine. The context of these is in or
near the stone architecture of the residential structures.
Middle and Late Cha/eolithic Sites
Guvercinkayasi
Giivercinkayas1 is a Middle- Late Chalcolithic (5200-4750 BCE) site in northwestern Cappadocia (Gillyur and
F1rat 2005). The settlement, resting on a rocky outcrop, consists of a citadel and a lower habitation area with numerous
houses, surrounded by a fortification wall with two towers (Gillyur 2012). Each residence has an oven near the
entrance, and a round hearth in the center of the room, both of which serve various functions (<;ayh 2009: 60). The
careful protection of ovens and fireplaces during building renovations could be considered a type of symbolic act in
the residences (<;ayh et al. 2020). Within several of these residences, three bull-headed grinding benches were
discovered; in addition, deer and cattle horns were found at the base of ovens and under the floors of residences. The
intentional placement of these has been interpreted as having had symbolic meaning in the settlement (<;ayh et al.
2020; Demirt~ 2019: 88).
In addition to animal horns at Giivercinkayas1, tools produced from in situ animal horns were also recovered. There
were numerous animal horns found in houses 26 and 27 (Gillyur et al. 2015), the majority of which belong to bovines.
A few were found on or near the round hearth in residence 26, and the rest were on the floor of residence 27,
accompanied by ground stone; it is possible that these houses contairied bone and horn tool workshops (Gillyur et al.
2015: 552-553). Besides their utilitarian function (Giilyur and Endogru 1999: 65; Giilyur et al. 2002: 105), the
excavators recently suggested the possible function of these horns as totems or symbols of protection (<;ayh et al.
2020: 49).
A pit with clay plaster, located in a residence near the western side of the settlement (in Trench 6F) contained
semi-worked and untreated horns, tools made of horn and bone, and animal skulls. Additionally, the presence of a
large animal figurine made of unbaked clay indicates that the pit may have had a symbolic aspect (Giilyur et al. 2014).
Near the southern area of the fortification wall (Trench 6G), in the lower settlement, a rather large, long, and
narrow residence was found. The location of the oven and the round hearth inside this dwelling are not consistent with
the general architectural order in the houses. Further, a marble figurine from this unusual space was recovered which
the excavator interpreted as a fat, swaddled baby with a pointed head and round face (Fig. 28-6a), with blunted arms
and legs (Gul9ur et al. 2011: 309, fig. 6). Thin grooves around the waist and across the torso of the figurine are
interpreted as clothing, or perhaps a blanket (Gul9ur et al. 2011: 297, 298).
One of the towers in the fortification system revealed a trash deposit containing numerous grinding stones, along
with an andiron, animal bones, and horns, situated within an ashy pit. At the top of this deposit was a forked antler,
identified as belonging to a very large adult deer; beneath this was the horn of a very large adult wild cow, and finally,
at the bottom was a third horn from a domestic goat (Gul9ur and Kiper 2007: 115). One possible interpretation of this
assemblage is that it served as a foundation deposit for the tower.
At least twenty figurines (Fig. 28-6b---c) have been recovered from the Middle Chalcolithic layers at
Giivercinkayas1, 11 of which are in human form, and 9 in animal form. These were generally found within the
residences, in storage or cooking areas. All but three of the human figurines can be identified as female, and are seated
(Eser 2018). Their possible representation ofa mother goddess has recently been discussed (C,::ayh et al. 2020). Human
figurines are generally made of baked clay, with the exception of the marble infant mentioned above. Animal figurines
represent bulls, pigs, and perhaps a goat with inlaid obisidian eyes (Gii19ur and C,::ayh 2010: 366). The majority of the
animal figurines (6 of the 9) represent bulls (Eser 2018) .
.)
b C
2cm
2cm
Figure 28-6. a) Sketch of marble "swaddled infant" from Giivercinkayas1 (after Eser 2018: 129, Kat. 18);
b: Sketch of human figurine head from Giivercinkayas1 (after Eser 2018: 122, kat. 11);
c: Sketch of animal figurine from Giivercinkayas1 (after Eser 2018: 119, kat. 8);
d: Sketch of seated woman from Kuru9ay level 6A, House 14 (after Duru 1996: pl. 145).
Kuruc;,ay
At Kuru9ay Hoyiik, excavators associated architectural features in the Late Chalcolithic 6A level residential
quarters and the so-called shrines with "sacral functions" which, like that in the Structure 8 "shrine" described above
(see Fig. 28-3a), were said to be "altars for burnt offerings" (Duru 1996: 118). These fire installations were located in
the center of houses, consisting of a basin edged with clay extending 5-6 cm above the floor, backed by a mudbrick
"stele" coated with plaster (Duru 1996: 118). The mudbrick backing to the basin was typically 60-70 cm high with a
depth of 18-21 cm. Such installations were usually located near an oven. These features are not found in every house;
it appears that one other structure in the 6A "shrine" level has a basin/stele feature, with more appearing in scattered
residences in subsequent level 6 layers, as well as in Late Chalcolithic level 3. Somewhat similar fire installations,
with the mudbrick backing, were found in Late Chalcolithic houses at Beycesultan, but were not described as anything
other than normal (Lloyd and Mellaart 1962: 22-23).
Human clay figurines were found in Early Chalcolithic Level 7 contexts at Kuru9ay, and these appear to be
consistent with those dating to earlier Neolithic layers (Duru 1996: 60, 105). These are broken portions of what are, in
many cases, female figurines, and include the type of "elongated" heads and necks with stylized almond eyes found at
Canhasan and Hacilar (see Fig. 28-5). While animal figurines were few at Early Chalcolithic Kuru9ay (Duru 1996:
105), plastic application of animal heads on ceramics and spoons was somewhat common, and on vessel bodies as
well (similar to those found at Ko~k Hoyuk, see below).
The only human figurine from the Late Chalcolithic levels came from the floor of House 14 in building level 6A.
The figurine is in the form of a seated woman with hands on breasts (Fig. 28-6d). A painted (red geometric patterns)
clay figure, possibly that of a "swaddled infant" was recovered from Late Chalcolithic levels, though not in situ (Duru
1996: 125). Note that a marble "infant" figurine was discovered at Early Chalcolithic Guvercinkayas1 (see above).
<;adzr Hoyuk
The unusual Non-Domestic Building and its contents (amulet, axe, figurine, fruitstands , and pouring vessel) were
described above. There are other items from the Late Chalcolithic contexts that can be included in this section,
described herein, but they are quite few in number. A curiosity is that most of these discoveries come from non-
domestic contexts. Exceedingly few figurines were recovered at <;::ad1r, and with the exception of the human figurine
in the Non-Domestic Building, they depict animals (one may be a bull, the other is unclear). The larger, likely bull,
figurine (Fig. 28-7a) was retrieved from the "Omphalos Building," a public structure involved in ceramic production
and distribution; the other came from an area near the public byway separating two compounds. A third heavily-worn
object, this one of stone, was unfortunately retrieved from a context mixed with slope wash (Fig. 28-7b ); it could be a
human-like figurine and is perhaps reminiscent of the "swaddled infants" described above, given that eroded carving
suggests clothing or a blanket, but it could also just as easily be a weight or other utilitarian object. Several more
animal figurines were retrieved from a kiln belonging to the earliest phase of the Early Bronze Age.
a b C
Figure 28-7. a) Animal figurine from Late Chalcolithic <;::ad1r Hoyuk (from Omphalos Building);
b) Stone object, possibly meant to represent a human figurine, from <;::ad1r Hoyuk;
c) Incised "cube" that may have been part of a foundation deposit at <;::ad1r Hoyuk.
Perhaps serving as a foundation deposit in the building of a ramp to the Upper Town in the settlement was a
hollow clay cube-shaped object with incised decoration and holes at the comers, perhaps for suspension purposes (Fig.
28-7c). Its partial status precludes functional interpretation, but it is unique among the <;::ad1r discoveries. Foundation
deposits at <;::ad1r are quite normal. In one case a courtyard was left littered with ceramic, bone, lithic, and animal horn
(mainly caprid) debris; a wall was pushed over on top of this deposit in preparation for an architectural overhaul of the
area. Smaller foundation deposits have been discovered under walls and sometimes furniture; deposits contain bits of
ceramic, bone, and sometimes lithics or ochre (Hackley et al. 2021; Steadman et al. 2019a, 2019c; Yildmm et al.
2018). Infant burials are another notable marker of symbolic behavior at <;::ad1r, described more fully below. Several
burning events in the domestic quarter at <;::ad1r were followed by architectural reconstructions, leading to the
possibility that there was an intentionality to the burning in an effort to "kill" the previous structure, allowing a new
one to emerge (Hackley et al. 2018).
The most remarkable "non-utilitarian" discovery at <;::ad1r came from a context under an Omphalos Building floor
(Steadman et al. 2017). This object is likely an andiron; it is heavily decorated and features an animal head, likely a
bull, on one of the preserved comers (Fig. 28-8). As noted above, a discovery at <;::amhbel Tarlas1 offers a similar
stylized head. This object does not appear to have been employed as an andiron (it is absent smudging, soot, ash, or
scratches), and its burial under the floor of an important building would suggest that it functioned as an important
foundation deposit. Present evidence from <;::ad1r Hoyuk suggests that ritual activities may have been more commonly
practiced in the designated "Non-Domestic Building" and at specific events (such as building destruction and
construction) rather than generally in household contexts.
0
-- 5cm
Figure 28-8. Drawing of the andiron from the Omphalos Building at

<;::adir Hoyuk with bull-head and incised decoration (drawing by L.D. Hackley).
Ali.Jar Hoyuk, Beycesultan, and Buyuk Gullucek
There were several human figurines recovered from the Late Chalcolithic levels of Ali~ar Hoyuk, which the
excavator identifies as having "pronounced noses" (von der Osten 1937: 78); two of these have a flat clay circle
arrayed around the head which is perforated so that perhaps ornamentation could be inserted, such as earrings or other
jewelry. One is a very crudely-made flat vaguely human shape with globular head. A handful of animal figurines, one
of which was painted, were also recovered; these likely represent either cattle or caprids. The exact contexts of these
finds is not apparent and thus nothing can be said of their possible function.
The excavators of Beycesultan assert that a marble body of a human (likely female), and clay elongated head,
almost certainly date to the Late Chalcolithic, kept as "antiques" as they were found in a Middle Bronze context
(Lloyd and Mellaart 1962: 265-266). The general form of these figurines is similar to those found at other
Chalcolithic sites, but their context is, of course, problematic. The small area of Late Chalcolithic exposure yielded
few other objects that might be identified as ritually symbolic; only two small animal figurines were discovered
(caprid or bovine) from Level XXII (Lloyd and Mellaart 1962: 269). It is worth noting that many of the Late
Chalcolithic houses at Beycesultan were burnt (Lloyd and Mellaart 1962: 17-26), raising the possibility of
intentionally "killing" the house before a new construction, as is argued for architectural change at <;::adir Hoyuk (see
above).
The Late Chalcolithic site of Buyuk Gullucek is better known for its pottery (see below), but did produce several
figurines. One is an animal figurine, and two represent humans (Ko~ay and Akok 1957: 15). One of the
anthropomorphic figurines is in the form of a violin, through the head is broken; a raised circle is drawn around the
genitalia (likely vagina). The second is more schematic and the gender is unclear; breasts are visible, in button-shape,
but are not necessarily attributable to either sex. Unfortunately, no context for these discoveries was noted.
Ritual (?) Symbolism and Chalcolithic Ceramics
The fantastic designs found on painted Neolithic ceramics on the plateau do not make their way very far into the
Chalcolithic as decorative techniques. Potentially ritualized, symbolic, or simply decorative treatment of Chalcolithic
ceramics can be grouped into three categories: incised, impressed, and painted decoration; applique; and ceramic
forms with animal or human characteristics. The first category, incised, impressed, and painted ceramics, has not
generally been treated by scholars as decorative techniques conveying symbolic meaning. More often these
decorations, which appear at most of the sites reviewed in this study, are analyzed as to their similarities with ceramics
in southeastern Europe (e.g., Efe 1990, 2000; Esin 1993; Ozbudak 2012, 2016; M. Ozdogan 1993, 2011; Steadman
1995; Thissen 1993). It is, therefore, the second two categories that form the basis of the discussion here.
Applique/ReliefDecoration
Applique decoration is found on ceramics at central plateau and Cappadocian sites, including Giivercinkayas1
(Demirta~ 2019), Biiyiik Giilliicek (Ko~ay and Akok 1957), Gelveri-Yuksekkilise (Ozbudak 2010: 66; Esin 1993),
Tepecik <;::iftlik (B1yakc1 et al. 2007; 6zbudak 2016: 62, 63), <;::amhbel Tarlas1 (Schoop 2015: 60), and Ko~k Hoyiik
(Oztan 2003, 2007; 6ztan and Faydah 2003). Perhaps the widest variety of applique images on ceramics is found at
Giivercinkayas1, which include mainly animals, with a few depictions of plants, humans, and geometric designs.
Applique usually occurs on the body of the vessels or on and around the handles (Demirta~ 2019) (Fig. 28-9a-b).
Humans are represented on several Ko~k Hoyiik ceramic examples by placing eyes and eyebrows next to a handle that
functions as the nose (Oztan and Faydah 2003: 50-51) (Fig. 28-9c), also found on vessels at Orman Fidanhg1 (Efe
2001: 43-48).
b t i,"" -f~
~ Jij
a
2cm
b
=~
~
~__al:J
5 cm
a
C
-=i
2cm
Figure 28-9. a) Sketch of animal face on handle of vessel b

from Giivercinkayas1 (after <;::ayh et al. 2020: 52, fig. 4d); b)
Sketch of animal face applique on sherd of bowl from
Giivercinkayas1 (after Gii19ur et al. 2018: 48, abb. 8) (no
dimensions available); c) Sketch of human applique face on Figure 28-10. a) Sketch of human and animal
vessel from Ko~k Hoyiik (after 6ztan and Faydah 2003: fig. applique at Giivercinkayas1 (after Demirta~ and
11) (no dimensions available); d) Sketch of animal applique Gii19ur 2017: 68, fig. 3); b) Sketch of human
on sherd from Tepecik <;::ifthk (after B1yak91 et al. 2007: 249, and animal applique at Tepecik <;::ifthk (after
fig. 35a). B1yak91 et al. 2007: 249, fig. 35d).
Animals depicted include bulls and bull horns/heads, goats, deer, dog, and snakes (Demirta~ 2019). Many of these
animals, with the addition of birds, bears, foxes , horses, and sheep, were also found on vessels at Tepecik Ciftlik (Fig.
28-9d); representations of parts of animals such as the horn, head, and testicles were also applied to the body of
vessels (Ozbudak 2016: 62, 63). The examples at Gelveri-Yuksekkilise and Ko~k Hoyiik display a smaller repertoire
of animals (deer and cow at the former, bull and tortoise at the latter).
Far fewer examples of full-bodied humans are found as applique decoration. A vessel at Giivercinkayas1 depicts
what appears to be humans engaged in a deer hunt (Demirta~ 2019: 91-92) (Figure 28-l0a). At Tepecik Ciftlik
humans also appear with animals (Figure 28-l0b); figures appear to be male, dressed in what may be animal fur. They
hold hunting weapons such as a bow and arrow or a boomerang-like tool (Ozbudak 2016: 62, 63).
Anthropomorphic and Zoomorphic Vessels
At Ko~k Hoyiik entire vessels were rendered in the form of bulls or tortoises; in one case a tortoise had obsidian
insets in the eyes (Oztan 2003: 59; 6ztan and Faydah 2003) (Figure 28-11). Vessels rendered in human form are also
found, though only few in number. At Tepecik Ciftlik anthropomorphic vessels feature human faces on the necks,
with breasts on the necks and jars; arms and hands appear to be wrapped around the body (Ozbudak 2016: 62-63).
Some examples of painted sherds showing portions of a human face protruding from what may be the neck or body of
the vessel were recovered at <;::atalhoyiik West (Erdogu 2010: 50-51). The excavator at Hacilar identifies several
vessels as "effigy pots," recovered from the level I fortress (Mellaart 1970: 180-182), though the context of these is
not entirely secure.
Figure 28-11. Sketch of vessel in the form of a tortoise from Ko~k Hoyuk
(after bztan and Faydah 2003: fig. 12-13) (no dimensions available).
Summary
The artifacts included in this section detailing those possibly carrying ritual or symbolic meaning fall into two
categories: figurines, and a second "catch-all" category of various items. On the subject of the figurines, numerous
studies (e.g., Belcher and Croucher 2016; Cauvin 2000; Hansen 2014; Hodder 2006; Lesure 2002, 2011; Meskell
1998; Voigt 2007), mostly focusing on the plentiful examples of Anatolian Neolithic figurines, profile the many
possible interpretations of these, including as goddesses, serving fertility needs, for initiation purposes, as well as for
various magico-ritual events. There is a reason why these studies focus on figurines from the Neolithic: there are far
fewer of them in the Chalcolithic. Even more notable is that for the Chalcolithic the majority occur at Early
Chalcolithic sites (e.g., Hac1lar, Kuru9ay 7, (:atalhoyuk West, and Orman Fidanhg1); by the time we are in secure Late
Chalcolithic contexts (e.g., Kuru9ay 6A, (:amhbel Tarlas1, and (:adir Hoyuk), the number of figurines present
diminishes substantially.
The discovery locations of figurines are not particularly helpful. A significant number derived from fills, and many
were broken. The rest were indeed found in residences, sometimes in walls (e.g., (:atalhoyuk West). Perhaps it is only
safe to suggest that whatever the purpose of Early and Middle Chalcolithic figurines, whether as toys (see (:ayh et al.
2020), ritual items, or other, they were most commonly employed in the household and then likely broken and tossed
away. Whatever purpose(es) they served in the sixth and fifth millennia became far less important in the Late
Chalcolithic fourth millennium. One other note on figurines is perhaps worth mentioning: animal figurines appear to
have become more common by the Late Chalcolithic. This is somewhat difficult to argue given that the overall
occurrence of figurines drops significantly. However, when a handful of figurines are recovered from a Late
Chalcolithic site such as (:adtr Hoyuk, Buyuk Gullucek, or Ali~ar Hoyuk, animals (usually bovine and goat/caprid) are
among the representations, or even dominate the collection.
The second "catch all" category is quite sparse, though this is likely due to the present authors' (and sometimes the
excavators') inability to recognize ritualized Chalcolithic objects. With the exception of architecture (painted walls)
and furniture such as the basins at Kuru9ay, and benches at Guvercinkayas1, items can be separated into three general
groups: animal horns, andirons, and "unusual items." The animal horns are primarily from cattle and caprids, though
deer antlers are found at Guvercinkayas1; contexts include inside rooms, often near hearths, and in foundation
deposits. The importance of cattle may stem from Neolithic times, but deer (particularly at Guvercinkayas1 where
there is also the hunting scene on a ceramic vessel) and goat (also rendered in figurines) may suggest a new important
focus in Chalcolithic economy and thus perhaps ritual. A review of the faunal assemblages at Chalcolithic sites may
tum up an increasing importance of goat (which is the case at (:adtr Hoyuk: Steadman et al. 2019b; and see von
Baeyer et al. 2021), while occurrence of deer, perhaps becoming harder to hunt due to scarcity, might be interwoven
with the apparent popularity of these animals in the Chalcolithic.
Andirons fall into that tricky category of "utilitarian" while also appearing unusual to their excavators. Certainly,
the one discovered at (:adtr Hoyuk (see Fig. 28-8) and some of the others, given their contexts, suggest that these
items may have had important symbolic value. Their role in food production and serving ties them to the animal horns
in the first category: a focus on the stages of subsistence that nourish a community (see (:ayh et al. 2020 for
discussion).
The most that can be said about the motley collection of "unusual items" such as the clay balls at (:atalhoyuk West
and the incised baked clay cube at (:adtr Hoyuk, is that given their contexts, they may well have been foundation
deposits. Such is also a likely interpretation for many of the animal horn discoveries at sites such as Guvercinkayas1
and (:adtr Hoyuk. It is also possible that animal horns may have functioned as foundation deposits at other sites, but
were reported as part of the faunal collection rather than highlighted for their context or non-subsistence function.
Foundation deposits, as discussed more fully below, may have had a deep symbolic meaning at Chalcolithic sites as
residents renewed or revamped the architecture of their communities, which could have mirrored changing
socioeconomic, sociopolitical, or religious trends over the generations.
Interpreting meaning of anthropomorphic and zoomorphic images on or by ceramic renderings is, if anything, even
more challenging. The "hunting scenes" featuring weapon-bearing humans and animals could simply represent daily
life, or they could have much deeper symbolic meaning, portraying rituals such as wild/herd animal maintenance
ceremonies (Hodder and Meskell 2011; Stanner 1998; and see a vast literature on the ritual symbolism of feasting),
shamanistic activities (e.g., Lewis-Williams and Pearce 2005; Price 2001; Whitehouse and Hodder 2010; but note that
such discussions usually focus on rock/wall art, not ceramics), or representation of totemic beliefs (e.g., Hodder and
Meskell 2011; Peters and Schmidt 2004; Russell 2018, 2019; Yakar 2012), among many other possibilities. Additional
discoveries of such vessels, with careful analysis of the locations within settlements, as well as residue analysis of
vessel contents may, at some future date, offer additional data upon which more secure inferences can be based.
BURIAL PRACTICES
Chalcolithic burial continues the trend of under-the-floor interment practiced in the Neolithic, but that is essentially
where the similarities end. By far the Chalcolithic "norm" is for infant and child burial, often within jars or pots,
tucked under floors or within walls of houses, or sometimes in areas just outside the domestic area. The following
summarizes these types of burials, which occur at almost every site included in this study; there are also occasional
"outlier" adult burials.
Neonate, infant, and child burial in houses seems to have been either a regular practice at a settlement, or rather
uncommon (though the latter could be a result of excavation coverage and placement). At Beycesultan an infant jar
burial was found just outside of a house wall, and one was under the house floor in Late Chalcolithic level XXIX
(Lloyd and Mellaart 1962: 23-24). At Early Chalcolithic Buyiikkaya (not otherwise included in this study), near
<;::amhbel Tarlas1, one infant burial (6-12 months in age) in a flexed position, was buried beneath a house floor
(Schoop et al. 2012a: 116, 2012b). Three other sites produced very few burials, sometimes in non-residential
locations. At <;::atalhoyiik West two neonate burials were found in Trench 5, in midden layers (Anvari et al. 2017: 14).
Similarly, at Orman Fidanhg1, a single child burial, recovered in phase VII, was found north of a trash pit (Efe 2001:
15). Two infant jar burials were recovered from Layer III at Guvercinkayas1, these within a domestic structure. One jar
burial contained a newborn baby, apparently accompanied by plant seeds, deposited in the wall of a storage space; the
second jar burial was tucked into a wall of Layer II and included a bead necklace as a funerary gift (Gul9ur and <;::ayh
2010: 365; Gul9ur et al. 2014: 553).
A few burials were recovered from Early Chalcolithic levels at Tepecik Ciftlik. They were generally in the hocker
position, though some appear to have been rearranged so that the skulls rested separately from the rest of the remains;
these burials contained individuals older than infants and were accompanied by funerary gifts such as obsidian tools,
horns, stone axes, and ceramics (B1yak91 et al. 2007: 239). The context of these burials is not clear. A larger number of
burials were recovered from Early/Middle Chalcolithic Layer 2B at Canhasan. A dozen burials in Structure 10
belonged to infants and children; these were not primary burials and appear to have been accompanied by pottery and
animal bones (French 1968a: 50, 1998: 25, 42). One of the skulls had ochre paint on it (French 1968a: 50); note that a
figurine and painted plaster were also found in this structure (see above). In Middle Chalcolithic Layer 2A, two infant
burials were found north of Structure 5 (French 1998: 49), and another was in the house wall. The burial of an adult
with an infant also likely dates to this period (French 1968b: 90).
At Ko~k Hoyiik the Early Chalcolithic level I
produced several burials. The majority of these were
infant and child burials; only two adults were
recovered (Oztan 2005, 2010). The infant and child
burials were in jars or in simple pits, generally under
house floors close to the walls; only a very few were
found outside houses (Oztan 2003: 71, 2005: 5).
Most of these burials were accompanied by grave
goods including pottery, shells, obsidian, burnishing
tools, jewelry, bone tools, and ochre (Oztan 2003:
71 ). Two adult burials in hocker position were found
in exterior contexts (Oztan 2005: 5).
Three sites, Kuru9ay, <;::amhbel Tarlas1, and <;::ad1r
Hoyuk, offer a significant number of burials, some
of which were in non-residential locations. Late
Figure 28-12. Photo of<;::amhbel Tarlas1 II, Grave 16. Chalcolithic levels 6A and 6 at Kuru9ay Hoyiik
Burial of a ca 8-15 month old girl in a grave vessel (upper provided 50 intramural infant and child burials,
part of vessel removed) (courtesy ofU.-D. Schoop). from neonatal to 8 years of age; the majority were
buried in jars or pits beneath house floors (Duru
1996: 85-88). A few of the burials were outside the houses either in spaces (streets) next to a house or in a courtyard.
Grave goods were not included in the burials; in one case, however, a preserved scrap of cloth suggests that infants
were shrouded prior to interment (Duru 1996: 120).
Late Chalcolithic <;::amhbel Tarlas1 produced as many as 28 individuals, 17 or more of them primarily in phase II
(some in phase III), mainly children and infants under 10 years of age (Schoop 2015: 58; Thomas 2011). The majority
of these were jar burials (Fig. 28-12); evidence suggests that the infants were placed head first into the vessels, and the
jar openings were then covered with a bowl or other large sherd, while children were interred in pits in the hocker
position (Schoop 2015: 58-59). Interestingly, jars of infant burials appear to have had their bases smashed, creating a
hole, while in situ, presumably after the infant and jar had been placed (Schoop 2015: 58). Placement of these burials
was nearly exclusively under house floors or adjacent to/within house walls (Thomas 2017: 83-87; Irvine et al. 2014:
21). Red staining on some of the bones suggests that ochre may have been introduced into the burial; at least some of
the burials had elongated skulls, indicating the presence of skull deformation practices (Thomas 2011 ).
I •
Figure 28-13. a) Photo of a Late Chalcolithic infant jar burial at (:ad1r Hoyuk, with most of the vessel covering the
infant (top removed), resting upside down; b) Photo of (:ad1r Hoyuk infant burial laid in broken jar covered by another
body sherd of broken jar; c) Photo of poorly-preserved jar burial of young child placed in wall of (:ad1r Hoyuk
housing compound showing metal hair slide or jewelry placed in burial; d) close-up of metal jewelry in burial.
Many of the practices found at (:amhbel were also present at (:ad1r Hoyuk. As of the writing of this chapter, a total
of 24 infant and child burials have been recovered from Late Chalcolithic contexts. The burials in the Non-Domestic
Building were described above. As was the case at other sites, infant jar burials are found in the domestic area,
sometimes inserted into walls or in courtyards (Steadman et al. 2015; Yildmm et al. 2018). Occasionally infants and
children, generally aged 0-5 years old, are in complete vessels (Fig. 28-13a), but more often they rest in the convex
space of a broken storage vessel, covered with another (Fig. 28-13b ). Infant and child burials were also found in non-
domestic areas and seem to be associated with architectural destruction, change, or renewal (Yildmm et al. 2018).
Burials are occasionally accompanied by gifts, mainly metal jewelry (Fig. 28-13c---d), though this is rare. Phytoliths
present in some of the graves indicate that the infants may have been wrapped in cloth at the time of burial, and the
presence of lime in several burials suggests that some may have been preserved for a period prior to burial. As was
found at Camhbel, head-shaping practices were identified on some of the skulls (Erdal 2019).
Though fewer in number, burials found at Ali~ar Hoyuk display practices similar to those at (:ad1r Hoyuk. Nine
infant and child burials were discovered below the floor of a house in level 13. Burials were either in pots or in pits,
with two described as being buried within boxes, one of stone and one wooden (von der Osten 1937: 42--43). A few
burial gifts accompanied these, primarily metal rings/hair slides similar to some found at (:ad1r Hoyuk (von der Osten
1937: 45, fig. 52; Steadman et al. 2019d). Two adult burials were discovered outside the domestic structure; one of
these was possibly accompanied by an infant and was in a cist tomb (von der Osten 1937: 44). In the lower level 14 a
child in a pot was discovered in a refuse pit; a rather rich array of grave goods accompanied this burial (von der Osten
1937: 40). The excavator notes that evidence suggests that these individuals were wrapped in cloth, reed matting, or
animal skins at the time of interment (von der Osten 1937: 44).
The "outlier" burials of adults or adults and children are found at a few Chalcolithic sites. At Alacahoyiik at least
six burials were recovered; none were in jars. One juvenile/young adult was recovered from a context near the edge of
the settlement at <;adrr Hoyiik. An adult was buried with a child in a stone box (Schoop 2005a: 41) reminiscent of the
one found at Ali~ar Hoyiik; other burials had funerary gifts such as ceramics and metal jewelry, and one was
accompanied by the skeleton of a dog (Schoop 2005a: 41). Jar burials are missing at Hac1lar, and burials in general are
limited. The mother and child burial within the "shrine" is noted above. Other interments include eight burials in
Hac1lar IV under a courtyard floor, and several adults in Hac1lar VI/V that may have been victims of the VI (late
Neolithic) fire in the settlement (Mellaart 1970: 24, 88). Some of these burials held funerary gifts such as ceramics or
jewelry, and most were placed in a contracted position (Mellaart 1970: 89). Early Chalcolithic Kanhta~ (otherwise not
included in this study) produced a single adult burial cut into a mudbrick pavement, accompanied by a flint blade and
a shell (Tiirkcan 2017b; and see Tiirkcan and Ertemin, this volume). Similarly, an adult male earthen burial was
discovered at Biiyiik Giilliicek, in a room belonging to the Chalcolithic period (though it is possible the burial was
inserted later); it contained a copper lance point lying beneath the skull (~enyurek 1950, 1957).
Summary
In the Chalcolithic, adult burials seem to have largely been moved into (as yet undiscovered) extramural
cemeteries; such are far better attested for the Early Bronze Age (Bachhuber 2015; Bertram and Bertram 2021; Massa
2014). This shift toward burial of adults in extramural cemeteries has been interpreted as an early indication of some
degree of political centralimtion (Laneri 2007; Selover and Durgun 2019); however, the data remain extremely scarce.
Infants and small children, however, continued to be buried within Chalcolithic houses. There may have been multiple
symbolic reasons for the Chalcolithic intramural burial of infants and children: keeping the child "close" to the family;
ensuring fertility for future children; house/building renewal rituals; kinship/lineage maintenance; and myriad other
interpretations that more data will bring to light. What the burial data attest to is the focus on the burial of infants and
children and the close association with architecture in Chalcolithic Anatolian settlements.
SUMMARY: CHALCOLITHIC RELIGION ON THE PLATEAU
Though not as rich as that in the Neolithic, architecture and material culture retrieved at Chalcolithic sites offer
some insights into ritual practices in these millennia. The construction of specific buildings dedicated to ritual
activities is somewhat rare at Chalcolithic sites. This correlates with the suggestion that post-Neolithic ritual practice
shifted to a focus on portable objects. Structures discussed here lead to several observations, the first being that the
practice of building specialized structures for ritual activities may have become more common in the Late
Chalcolithic. The structures identified as possibly serving ritual purposes at the Early Chalcolithic sites (Hac1lar and
Canhasan) might well have simply been (larger) domestic residences with unusual contents. The same might be true at
Late Chalcolithic Kuruyay (level 6A), but the contents of Structure 8 at this site, so different from the other houses,
would seem to support the excavator's identification of it as specialized. The two Late Chalcolithic structures at <;adu
Hoyiik and <;amhbel Tarlas1 are almost certainly not domestic in nature, and their contents point to unusual, non-
domestic, activities. The second observation is the uniqueness of these structures. Each community seems to have had
their own ideas of what their specialized buildings should look like, and what should happen within it: at Kuruyay
libations may have predominated; perhaps measuring the movement of the sun was critical at <;adu Hoyiik; the
production of idols was a focus at <;amhbel Tarlas1. As more Late Chalcolithic settlements are excavated across the
plateau, the possible florescence ofritualized structures at this time, and their purpose, may become clearer.
The material culture offers some (very) tentative directions for interpretation regarding how religion/ritual was
conceptualized through the Chalcolithic period. Color, especially red, remains a focus. Whatever the possibly
symbolic function of human/female figurines in the Neolithic and Early Chalcolithic, which may well have included
destruction and discarding of the figurines after use (given their consistent presence in middens and fills), the focus on
animate forms seems to have shifted to animals by the Late Chalcolithic. While bovine representations remain a
constant from the Neolithic, goat and deer (the latter only on ceramics and represented by antlers) seem to be of
interest by the Late Chalcolithic. Added to these are the presence of andirons, sometimes in concert with animal horns.
Decorative techniques on ceramics also focus a lens on a variety of animals, both domesticated and wild. This may
suggest a shifting focus of priorities, especially regarding subsistence activities, including an increasing reliance on
crop and herd management strategies and even food processing and production. The hunting of wild animals may have
gained significance due to the role wild meat might have played as a possible exotic/luxury addition to subsistence, to
a waning presence of wild animals, or to other as yet undiscemed explanations.
Finally, foundation deposits in the form of burials and material culture attest to the importance of construction and
architectural renewal in the Chalcolithic. Infant and child burials become nearly the sole form of intramural burial in
Chalcolithic communities, inserted in walls and under floors, and are often coincident with architectural changes. A
shift from the Neolithic "household" focus of often adult burials (perhaps recognizing some element of ancestor
veneration) broadens to "structures" in the Chalcolithic, whether domestic or communal/public, of almost exclusively
sub-adult burials. Material culture deposits under walls and other constructions appear to function in the same way,
inserted at times of architectural renewal, renovation, and even destruction. This shift may correlate with larger social
changes taking place in the Chalcolithic millennia including population movements, long-distance trade relations,
intensified subsistence practices, increases in organized labor, and changes in sociopolitical structures, to name just a
few.
At the very least this study has demonstrated how much there is to learn about Chalcolithic religion and ritual on
the Anatolian plateau, and yet how rich it must have been. As always, only continued careful and well-published
future excavations can fill in the enormous gaps in our knowledge.
REFERENCES CITED
Anvari, Jana, Facob Brady, Ingmar Franz, Goce Naumov, David Orton, Sonia Ostaptchouk, Elizabeth Stroud, Patrick
T. Willett, Eva Rosenstock, and Peter F. Biehl. 2017. Continuous Change: Venturing into the Early Chalcolithic at
<;atalhoyiik. In The Archaeology ofAnatolia, Volume 11: Recent Discoveries (2015- 2016), S.R. Steadman and G.
McMahon, eds., 6-39. Newcastle-upon-Tyne: Cambridge Scholars Publishing.
Atakuman, <;igdem. 2014. Architectural Discourse and Social Transformation During the Early Neolithic of southeast
Anatolia. Journal of World Prehistory 27: 1-42.
- . 2015. From Monuments to Miniatures: Emergence of Stamps and Related Image-Bearing Objects during the
Neolithic. Cambridge Archaeological Journal 24(5): 759- 788.
Aurenche, Olivier. 1981. La maison orientale: L 'architecture du Proche-Orient ancien des origines au milieu du
quatrieme millenaire. Paris: Librairie Orientaliste Paul Geuthner.
Bachhuber, Christoph. 2015. Citadel and Cemetery in Early Bronze Age Anatolia. Sheffield: Equinox Publishing.
Baird, Douglas. 2012. The Late Epipaleolithic, Neolithic, and Chalcolithic of the Anatolian Plateau, 13,000-4000 BC.
In A Companion to the Archaeology of the Ancient Near East, Volume I, D.T. Potts, ed., 431-465. West Sussex:
Wiley-Blackwell.
Baird, Douglas, D Carruthers, A Fairbairn, and J. Pearson. 2011. Ritual in the Landscape: Evidence from Pmarb~1 in
the Seventh-Millennium Cal BC Konya Plain. Antiquity 85: 3 80-394. doi: 10.1017/S0003598X0006782X.
Baird, Douglas, Andrew Fairbairn, and Louise Martin. 2017. The Animate House, the Institutionalization of the
Household in Neolithic Central Anatolia. World Archaeology 49(5): 753- 776.
Banning, Edward B. 2011. So Fair a House: Gobekli Tepe and the Identification of Temples in the Pre-Pottery
Neolithic of the Near East. Current Anthropology 52(5): 619-660.
Baysal, Emma. 2017. Reflections of Faraway Places: The Chalcolithic Personal Ornaments ofCanhasan I. Anatolian
Studies 67: 29-49.
Belcher, Ellen H. and Karina Croucher. 2016. Exchanges of Identity in Prehistoric Anatolian Figurines. Proceedings,
9'h ICAANE, Basel 2014 l: 43- 56.
Bertram, Jan-Dryzysztof and Gi.ilvin ilgezdi Bertram. 2021. The Late Cha/eolithic and Early Bronze Age in Central
Anatolia: Introduction-Research History- Chronological Concepts-Sites, their Characteristics and Stratigraphies.
B1vakc1, Erhan. 2003. Observations on the Early Pre-Pottery Neolithic Architecture in the Near East: New Building
Materials and Techniques. In From Villages to Towns: Studies Presented to Ufuk Esin, M. Ozdogan, H.
Hauptmann, and N. Ba~gelen, eds., 385-413. istanbul: Arkeoloji ve Sanat Yaymlar1.
B1vakvi, Erhan, <;iler Altmbilek Algi.il, Semra Balct, and Martin Godon. 2007. Tepecik-<;ifllik. In Tiirkiye'de Neolitik
Donem. Anadolu'da Uygarhgm Dogu~u ve Avrupa'ya Yayzhmz. Yeni Kazzlar- Yeni Bulgular, M. Ozdogan and N.
B~gelen, eds., 237- 253. istanbul: Arkeoloji ve Sanat Yaymlar1.
Biehl, Peter. 1996. Symbolic Communication Systems, Symbol and Anthropomorphic Figurines of the Neolithic and
Chalcolithic from South-Eastern Europe. Journal ofEuropean Archaeology 4: 153-176.
- . 2012. Rapid Change versus Long-Term Social Change during the Neolithic-Chalcolithic Transition in Central
Anatolia. Interdisciplinaria Archaeologica: Natural Sciences in Archaeology 1: 75- 83.
Biehl, Peter F., Ingmar Franz, Sonia Ostaptchouk, David Orton, Jana Rogasch, and Eva Rosenstock. 2012. One
Community and Two Tells: The Phenomenon of Relocating the Tell Settlements at the Turn of the 7th and 6th
Millennia in Central Anatolia. In Tells: Social and Environmental Space, R. Hofrnann, F.-K. Moetz, and J. Muller,
eds., 53-65. Bonn: Verlag Dr. RudolfHabelt GmbH.
Bienert, H.D. 1991. Skull Cult in the Prehistoric Near East. Journal ofPrehistoric Religion 5: 9-23.
Bischoff. Damien. 2002. Symbolic Worlds of Central and Southeast Anatolia in the Neolithic. In The Neolithic of
Central Anatolia. Internal Developments and External Relations during the 9'h_ (j1h millennia, F. Gerard and L.
Thissen eds., 237-251. istanbul: Ege Yaymlar1.
Boivin, Nicole. 2000: Life Rhythms and Floor Sequences: Excavating Time in Rural Rajasthan and Neolithic <;atal
Hoyiik. World Archaeology 31(3): 367-388.
Bonogofsky, Michelle. 2003. Neolithic Plastered Skulls and Railroading Epistemologies. Bulletin of the American
Schools of Oriental Research 331: 1-10.
- . 2005. Anatolian Plastered Skulls in Context: New Discoveries and Interpretations. Arkeometri Sonur;:larz
Toplantzsz 20: 13- 26.
- . 2006. Complexity in Context: Plain, Painted and Modeled Skulls from the Neolithic Middle East. In Skull
Collection, Modification and Decoration, M. Bonogofsky, ed., 15-28. British Archaeological Reports
International Series 1539. Oxford: Archaeopress.
Brami, Maxine. 2014. House-Related Practices as Markers of the Neolithic Expansion from Anatolia to the Balkans.
Bulgarian e-Journal ofArchaeology 4: 161-177.
Cauvin, Jacques. 2000. The Birth of the Gods and the Origins of Agriculture. Cambridge: Cambridge University
Press.
<;ayh, Pmar. 2009. Giivercinkayasz 14 Numaralz Evin <;anak <;omlek Verilerinden Yo/a <;zkarak Yerl~me Diizeni ve
Toplumsal Yapzlanmanzn Degerlendirilmesi. Unpublished Master's thesis, istanbul Universitesi Sosyal Bilimler
Enstitusu.
<;ayh, Pmar, I!;,Il Demirt~, Bari!;, Eser, Sevil Gulr;:ur, and Varhk indere. 2020. Guvercinkayas1'nda Sembolizma. In
Metallurgica Anatolica: Festschrift fiir Onsal Talr;:zn anli.isslich seines 65. Geburtstags, H.G. Yalr;:m and 0.
Stegemeier, eds., 45-58. istanbul: Ege Yaymlan.
c;elik, Bahattin. 2004. A New Early Neolithic Settlement in Southeastern Turkey: Hamzan Tepe. Neo-Lithics 2: 3- 5.
c;evik, Ozlem and Bun;:in Erdogu. 2019. Multiple Faces of Changes in 5600/5500 CAL, BC Anatolia and Thrace.
Anatolica 45: 1- 16.
- . 2020. Absolute Chronology of Cultural Continuity, Change and Break in Western Anatolia Between 6850-5460
CAL. BC: The Ulucak Hoyuk Case. Mediterranean Archaeology and Archaeometry 20(1): 77-92.
Croucher, Karina. 2012. Death and Dying in the Neolithic Near East. Oxford: Oxford University Press.
Croucher, Karina and Ellen Belcher. 2017. Prehistoric Figurines in Anatolia (Turkey). In The Oxford Handbook of
Prehistoric Figurines, T. Insoll, ed., 443-468. Oxford: Oxford University Press.
Demirta!;,, I~Il and Sevil Giil<;ur. 2017. Guvercinkayas1 Kabartma Bezekli Bir Av Sahnesi. In Sams at 'tan Acemhoyiik'e
Eski Uygarlzklarzn lzinde. Aliye Oztan'a Armagan, S. Ozkan, H. Hury1lmaz, and A. Turker, eds., 61- 70. izmir:
Ege Oniversitesi Bas1mevi.
Demirta~, I~1l. 2019. Guvercinkayas1'nm Kabartma Betimli <;anak <;omleginde Sembolizma. In Kapadokya Hafiza,
Kimlik ve Kiiltiirel Miras, M. Hakman, ed., 83-116. lstanbul: Arkeoloji ve Sanat Yaymlar1.
Dietrich, Oliver and Jens Notroff. 2015. A Sanctuary, or So Fair a House? In Defense of an Archaeology of Cult at
Pre-Pottery Neolithic Gobekli Tepe. In Defining the Sacred: Approaches to the Archaeology of Religion in the
Near East, N. Laneri, ed., 75-89. Oxford: Oxbow Books.
Dietrich, Oliver, Manfred Heun, Jens Notroff, Klaus Schmidt, and Martin Zarnkow. 2012. The Role of Cult and
Feasting in the Emergence of Neolithic Communities. New Evidence from Gobekli Tepe, South-Eastern Turkey.
Antiquity 88(333): 674-695.
Duru, Refik. 1996. Kurur;:ay Hoyiik II: Results of the Excavations 1978- 1988, The Late Cha/eolithic and Early Bronze
Settlements. Ankara: Turk Tarih Kurumu Bas1mevi.
Duru, Gune~ and Mihriban Ozb~aran. 2005: A 'Non-Domestic' Site in Central Anatolia. Anatolia Antiqua XIII: 15-
28.
During, Bleda. 2003. Burials in Context: The 1960s Inhumations of<;atalhoyuk East. Anatolian Studies 53: 1-15.
- . 2011. Millennia in the Middle? Reconsidering the Chalcolithic of Asia Minor. In The Oxford Handbook ofAncient
Anatolia, S.R. Steadman and G. McMahon, eds., 796-812. Oxford: Oxford University Press.
During, Bleda and Arkadiusz Marciniak. 2006. Households and Communities in the Central Anatolian Neolithic.
Archaeological Dialogues 12(2): 165-187.
Efe, Turan. 1990. An Inland Anatolian Site with Pre-Vin<;a Elements: Orman Fidanhg1, Eski~hir. Germania 68: 67-
113.
- . 2000. Recent Investigations in Inland Northwestern Anatolia and its Contribution to Early Balkan-Anatolian
Connections. InKaranovo. Beitri.ige zum Neolithikum in Siidosteuropa, vol 3, S. Hiller and V. Nikolov, eds., 173-
183. Wien: Phoibos Verlag.
- . 2001. The Salvage Excavations at Orman Fidanlzgi: A Cha/eolithic Site in Inland Northwestern Anatolia. istanbul:
TASK Vakf1 Yaymlan 3.
Erda!, Y1lmaz. 2019. Interpreting Subadult Burials and Headshaping at <;adrr Hoyuk. Journal of Eastern
Mediterranean Archaeology and Heritage Studies 7(3): 379-385.
Erdogu, Bur<;in. 2009a. Ritual Symbolism in the Early Chalcolithic Period of Central Anatolia. Journal for
Interdisciplinary Research on Religion and Science 5: 129- 151.
- . 2009b. West Mound Trench 8. <;atalhoyiik Archive Report 2009: 50-59.
http://www.catalhoyuk.com/sites/default/files/media/pdfi'Archive_Report_2009.pdf
- . 2010. West Mound Trench 8. <;atalhoyiikArchiveReport2010: 50-51.
http://www.catalhoyuk.com/sites/default/files/media/pdfi'Archive_Report_ 201 0.pdf
Erdogu, Bur<;in and Aydm Ulubey. 2011. Colour Symbolism in the Prehistoric Architecture of Central Anatolia and
the Raman Spectroscopic Investigation of Red Ochre in Chalcolithic <;atalhoyuk. Oxford Journal of Archaeology
30(1): 1-11.
Eser, Ban~. 2018. Kalkolitik Donem 'de Giivercinkayasz Figiirinleri. Unpublished master's thesis. Edirne: Trakya
Oniversitesi Sosyal Bilimler Enstitusu, Arkeoloji Anabilim dah.
Esin, U. 1993. Gelveri-Ein Beispiel fur die kulturellen Beziehungen zwischen Zentralanatolien und Sudosteuropa
wahrend des Chalcolithikums. Anatolica 19: 47- 56.
Flannery, Kent. 2002. The Origins of the Village Revisited: From Nuclear to Extended Households. American
Antiquity 67(3): 417--434.
French, David H. 1962. Excavations at Can Hasan: First preliminary report, 1961. Anatolian Studies 12: 27--40.
- . 1963. Excavations at Can Hasan: Second Preliminary Report, 1962. Anatolian Studies 13: 29--42.
- . 1964. Excavations at Can Hasan: Third Preliminary Report, 1963. Anatolian Studies 14: 125-131.
- . 1965. Excavations at Can Hasan: Fourth Preliminary Report, 1964. Anatolian Studies 15: 87- 94.
- . 1966. Excavations at Can Hasan: Fifth Preliminary Report, 1965. Anatolian Studies 16: 113-121.
- . 1968a. Excavations at Can Hasan 1967: Seventh Preliminary Report. Anatolian Studies 18: 45-53.
- . 1968b. Canhasan-Karaman 1967.TilrkArkeoloji Dergisi 16(1): 89- 94.
- . 1998. Canhasan Sites I. Canhasan I: Stratigraphy and Structures. British Institute of Archaeology at Ankara,
Monograph No. 23. Ankara: British Institute of Archaeology.
- . 2010. Canhasan Sites 3, Canhasan I: Small Finds. British Institute of Archaeology at Ankara, Monograph No. 45.
Ankara: British Institute of Archaeology.
Gage, John. 1999. What Meaning had Colour in Early Societies? Cambridge Archaeological Journal 9(1): 109- 126.
Gebel, H.G.K., B.D. Hermansen, and C.H. Jensen, eds. 2002. Magic Practices and Ritual in the Near Eastern
Neolithic. Berlin: Ex Oriente.
Gemici, Hasan C. and <;igdem Atakuman. 2021. The World of Figurines in the Neolithic and Cha/eolithic Aegean:
The Case of Ugurlu Hoyilk on Goh;eada (Imbros). BAR International Series no. 3021. Oxford: British
Archaeological Reports.
Gresky, Julia, Julian Haelm, and Lee Clare. 2017. Modified Human Crania from Gobekli Tepe Provide Evidence for a
New Form of Neolithic Skull Cult. Science Advances 3(6): el 700564
Gul9ur, Sevil. 2012. The Chalcolithic Period in Central Anatolia Aksaray-Nigde Region. Origini XXIV (Nuova Serie
V): 213-227.
Gul9ur, Sevil and Pmar <;ayh. 2010. Guvercinkayas1 2008 Y1h Kaz1s1. Kazi Sonur;larz Toplantzsi 31(3): 357-378.
Gul9ur, Sevil, Pmar <;ayh, and 1$11 Demi~. 2011. Gtivercinkayas1 2009 Y1h Kaz1s1. Kazi Sonur;lari Toplantzsz 32(1):
294-312.
Gtil9ur, Sevil, Pmar <;ayh, 1$11 Demi~, Bari$ Eser, and Varhk indere. 2018. Guvercinkayas1 und fruhe
Urbanisierung in Anatolien. In Anatolian Metal VIII: Eliten- Handwerk- Prestigegilter, 0. Yal9in, ed., 43- 56.
Rahden: Verlag Marie Leidorf.
Gul9ur, Sevil and Muhsin Endogru. 1999. Gtivercinkayas1 1997 Kaz1s1. Kazz Sonur;larz Toplantm 20(1): 77-100.
Gtil9ur, Sevil, Muhsin Endogru, and Fariz Demir. 2002. Gtivercinkayas1 2000 Kaz1s1 ve <;evre Ara$hrmalar1. Kazz
Sonur;larz Toplantzsz 23(2): 101-114.
Gtil9ur, Sevil and Celine Frrat. 2005. Spatial Analysis of Gtivercinkayas1, A Middle Chalcolithic Hilltop Settlement in
Northwestern Cappadocia: A Preliminary Report. Anatolia Antiqua 13: 41- 52.
Gtil9ur, Sevil and Yucel Kiper. 2007. Gtivercinkayas1 2005 Y1h Kazis1 On Raporu. Kazz Sonur;larz Toplantzsz 28(2):
111- 124.
Gtil9ur, Sevil, Yasemin Yilmaz, Pmar <;ayh, and 1$11 Demirt~. 2014. Guvercinkayas1 2012 Kazis1 On Rapor. Kazz
Sonur;larz Toplantzsz 35(2): 443--455.
- . 2015. Guvercinkayas1 2013 Kaz1s1 OnRapor. Kazi Sonur;larz Toplantzsi 36(2): 547- 562.
Hackley, Laurel D., Stephanie L. Selover, and Sharon R. Steadman. 2018. The Persistence of Social and Spatial
Memory at Prehistoric <;adu Hoyuk. International Journal of the Constructed Environment 9(4): 1-20.
Hackley, Laurel D., Burcu Y1ldmm, and Sharon R. Steadman. 2021. Not Seeing is Believing: Ritual practice and
Architecture at Chalcolithic <;adu Hoytik in Anatolia. Religions 12(8): https://doi.org/10.3390/rel12080665.
Haddow, Scott D., Joshua W. Sadvari, Christopher J. Kntisel, and Remi Hadad. 2016. A Tale of Two Platforms:
Commingled Remains and the Life-Course of Houses at Neolithic <;atalhoytik. In Theoretical Approaches to
Analysis and Interpretation of Commingled Human Remains, A.I. Osterholtz, ed., 5-29. New York: Springer.
Hansen, Svend. 2014. Neolithic Figurines in Anatolia. In The Neolithic in Turkey, M. Ozdogan, N. B~gelen, and P.
Kuniholm, eds., 265-292. istanbul: Archaeology & Art Publications.
Hauptmann, Harald. 1999. The Urfa Region. In Neolithic in Turkey: The Cradle of Civilisation, M. Ozdogan and N.
B~gelen, eds., 65-86. istanbul: Arkeoloji ve Sanat Yaymlan.
Hodder, Ian. 2006. The Leopard's Tale: Revealing the Mysteries of Turkey 's Ancient "Town." London: Thames and
Hudson.
- . 2010. Religion in the Emergence of Civilization: <;atalhoyilk as a Case Study. Cambridge: Cambridge University
Press.
Hodder, Ian and Lynn Meskell. 2011. A 'Curious and Sometimes a Trifle Macabre Artistry.' Current Anthropology
52(2): 235- 263.
Irvine, Benjamin, Jayne-Leigh Thomas, and Ulf-Dietrich Schoop. 2014. A Macroscopic Analysis of Human Dentition
at Late Chalcolithic <;amhbel Tarlas1, North Central Anatolia, with Special Reference to Dietary and Non-
masticatory Habits. Interdisciplinaria Archaeologica V(l): 19- 30.
Karul, Necmi. 2011. Gusir Hoytik. In The Neolithic in Turkey: The Tigris Basin, M. Ozdogan, N. B~gelen, and I.
Kuniholm, eds., pp. 1-17. istanbul: Arkeoloji ve Sanat Yaymlan.
Kolankaya-Bostanc1, Neyir. 2014. The Evidence of Shamanism Rituals in Early Prehistoric Periods of Europe and
Anatolia. In Colloquium Anatolicum: Anadolu Sohbetleri XIII, M. Alparslan, B. Htirmuzlu-Kortholt, N. Karul, and
E. Kortanoglu, eds., 185- 204. istanbul: Turk Eski<;ag Bilimleri Enstitusu.
Ko~y, Hamit Zubeyr and Mahmut Akok. 1957. Tiirk Tarih Kurumu Tarafindan Yap1lan Biiyiik Giilliicek KazlSl: 1947
ve 1949'daki <;alzJmalar Hakkmda ilk Rapor = Ausgrabungen von Biiyiik Giilliicek: Ausgefiihrt durch die
Tiirkische Historische Gesellschaft Vorbericht iiber die Arbeiten von 1947 und 1949. Turk Tarih Kurumu
Yaymlarmdan, V. Dizi; Sa. 16. Ankara: Turk Tarih Kurumu BasIIDevi.
Kuijt, Ian, ed. 2000a. Life in Neolithic Farming Commwzities: Social Organization, Identity, and Differentiation. New
York: Kluwer Academic/Plenum Publishers.
- . 2000b. Keeping the Peace: Ritual, Skull Caching, and Community Integration in the Levantine Neolithic. In Life
in Neolithic Farming Communities: Social Organization, Identity, and Differentiation, I. Kuijt, ed., 137- 164.
London: Kluwer Academic/Plenum Publishers.
- . 2008. The Regeneration of Life: Neolithic Structures of Remembering and Forgetting. Current Anthropology
49(2): 171-197.DOI: 10.1086/526097.
Laneri, Nicola. 2007. Burial Practices at Titri~ Hoyuk, Turkey: An Interpretation. Journal of Near Eastern Studies
66(4): 241- 266.
Last, Jonathan. 1998. A Design for Life: Interpreting the Art of <;atal Hoyuk. Journal of Material Culture 3(3): 355-
378.
Lesure, Richard G. 2002. The Goddess Diffracted: Thinking About the Figurines of Early Villages. Current
Anthropology 43(4): 587- 607.
- . 2011. Interpreting Ancient Figurines: Context, Comparison, and Prehistoric Art. Cambridge: Cambridge
University Press.
Lewis-Williams, David, and David Pearce. 2005. Inside the Neolithic Mind. London: Thames and Hudson.
Lichter, Clemens. 2016. Burial Customs of the Neolithic in Anatolia-An Overview. In Anatolian Metal VII: Anatolia
and Neighbours 10.000 Years Ago, -0. Yalc;in, ed., 71- 84. Bochum: Fritz Thyssen Stiftung.
Lloyd, Seton and James Mellaart. 1962. Beycesultan Volume I. London: British Institute of Archaeology at Ankara.
Marciniak, Arkadiusz and Lech Czerniak. 2007. Social Transformation in the Late Neolithic and the Early
Chalcolithic Periods in Central Anatolia. Anatolian Studies 57: 115-130.
Massa, Michele. 2014. Early Bronze Age Burial Customs on the Central Anatolian Plateau: A View from
Demircihoyuk-Sanket. Anatolian Studies 64: 73- 93.
Mellaart, James. 1967. <;atal Hiiyiik: A Neolithic Town in Anatolia. London: Thames and Hudson.
- . 1970. Excavations at Hacilar. Vols. I-II. Edinburgh: Edinburgh University Press.
Meskell, Lynn. 1998. Twin Peaks: The Archaeologies of <;atalhoyuk. In Ancient Goddess, Lucy Goodison and
Christine Morris, eds., 46-62. Madison: University of Wisconsin Press.
Naumov, Goce and Peter F. Biehl. 2019. Chalcolithic Human Representations at <;atalhoyuk. In Human Iconography
and Symbolic Meaning in Near Eastern Prehistory, J. Becker, C. Beuger, and B. Muller-Neuhof, eds., 213-224.
Vienna: Austrian Academy of Sciences Press.
Orton, David, Jana Anvari, Catriona Gibson, Jonathan Last, Amy Bogaard, Eva Rosenstock, and Peter F. Biehl. 2018.
A Tale of Two Tells: Dating the <;atalhoyuk West Mound. Antiquity 92.363: 620-639.
Ozb~aran, Mihriban. 2011. The Neolithic on the Plateau. In The Oxford Handbook of Ancient Anatolia, S.R.
Steadman and G. McMahon, eds., 99- 124. Oxford: Oxford University Press.
- . 2012. A~1kh. In The Neolithic in Turkey III, M. Ozdogan, N. B~gelen, and P. Kuniholm, eds., 135-158. istanbul:
Archaeology and Art Publications.
Ozbek, Metin. 2009. Remodeled Human Skulls in Ko~k Hoyuk (Neolithic Age, Anatolia): A New Appraisal in View
ofRecentDiscoveries. Journal ofArchaeological Science 36: 379-386.
Ozbudak, Mehmet Ozan. 2010. Gelveri Kiiltiirii. Unpublished Master's thesis. istanbul: istanbul University.
- . 2012. Kapadokya MO 6. Biny1l Kronolojisinin Kay1p Halkas1: Gelveri <;anak <;omlegi -Ozerine En Son
<;ah~malar. Colloquium Antolicum 11: 267-297.
- . 2016. Koyu Yiizlii <;izi-Kaz1 Bezemeli <;anak <;omlek IJ1gmda Orta Anadolu'da Orta Kalkolitik'e Ger;iJ Sorwzu.
Unpublished Ph.D. thesis. istanbul: istanbul University.
Ozdogan, Ash. 2011. <;ayonu. In The Neolithic in Turkey: The Tigris Basin, M. Ozdogan, N. B~gelen, and I.
Kuniholm, eds., 185-269. istanbul: Arkeoloji ve Sanat Yaymlan.
Ozdogan, Mehmet. 1993. Vinca and Anatolia: A New Look at a Very Old Problem. Anatolica 19: 173-193.
- . 2011. Eastern Thrace: The Contact Zone Between Anatolia and the Balkans. In The Oxford Handbook of Ancient
- . 2014. Anatolia: From the Pre-Pottery Neolithic to the End of the Early Bronze Age (10,500- 2000 BCE). In The
Cambridge World Prehistory, C. Renfrew and P. Bahn, eds., 1508-1544. Cambridge: Cambridge University Press.
Oztan, Aliye. 2003. A Neolithic and Chalcolithic Settlement in Anatolia: Ko~k Hoyuk. In Colloquium Anatolicum II,
M. Alparslan and M. Alparslan, eds., 69- 86. istanbul: Turk Eskic;ag Bilin1leri Enstitusu.
- . 2005. I. Donem Ko~k Hoyuk-Nigde Kazi Buluntularmm Degerlendirilmesi ve Stratigrafik Kontrolu. Bilimsel
AraJhrma Projesi Kesin Raporu. Ankara: Ankara -Oniversitesi.
-. 2007. Nigde-Bor Ovas1 'nda Bir Neolitik Y erle~im. In Tur/dye' de Neolitik Donem. Anadolu'da Uygarlzgm Dog~u
ve Avrupa'ya Yayilzmz. Yeni Kazzlar - Yeni Bulgular, M. Ozdogan and N. B~gelen, eds., 223- 235. istanbul:
Arkeoloj i ve Sanat Y aymlan.
-. 2010. Archaeological Investigations at Ko~k Hoyilk, Nigde. In Geo-Archaeological Activities in Southern
Cappadocia, Turkey. L. d' Alfonso, M.E. Balza, and C. Mora, eds., 83-95. Pavia: Italian University Press.
Oztan, Aliye and Erol Faydah. 2003. An Early Chalcolithic Building From Ko$k Hoyilk. Belleten 67(248): 45-76.
Peters, Joris and Klaus Schmidt. 2004. Animals in the Symbolic World of Pre-Pottery Neolithic Gobekli Tepe, South-
eastern Turkey: A Preliminary Assessment. Anthropozoologica 39(1): 179- 218.
Pilloud, Marin A., Scott D. Haddow, Christopher J. Knilsel, Clark S. Larsen, and Mehmet Somel. 2020. Social
Memory and Mortuary practices in Neolithic Anatolia. In The Poetics ofProcessing: Memory Formation, Identity,
and the Handling ofthe Dead, A.I. Osterholtz, ed., 145- 165. Boulder: University Press of Colorado.
Price, Neil S. 2001. An Archaeology of Altered States: Shamanism and Material Culture Studies. In The Archaeology
ofShamanism, N. Price, ed., 3- 16. London: Routledge.
Rappaport, Roy A. 1999. Ritual and Religion in the Maldng of Humanity. Cambridge Studies in Social and Cultural
Anthropology 110. New York: Cambridge University Press.
Renfrew, Colin. 1998. Mind and Matter: Cognitive Archaeology and External Symbolic Storage. In Cognition and
Material Culture: The Archaeology of Symbolic Storage, C. Renfrew and C. Scarre, eds., 1-6. Cambridge:
Rosenberg, M. 201 la. Hallan <;emi. In The Neolithic in Turkey.· The Tigris Basin, M. Ozdogan, N. Ba~gelen, and I.
Kuniholm, eds., 61- 78. istanbul: Arkeoloji ve SanatYaymlan.
-. 201 lb. Demirkoy. In The Neolithic in Turkey: The Tigris Basin, M. Ozdogan, N. B~gelen, and I. Kuniholm, eds.,
79-87. istanbul: Arkeoloji ve Sanat Yaymlar1.
Rosenberg, Michael and Ash Erim-Ozdogan. 2011. The Neolithic in South-Eastern Anatolia. In The Oxford
Handbook ofAncient Anatolia, S.R. Steadman and G. McMahon, eds, 125-149. Oxford: Oxford University Press.
Russell, Nerissa. 2018. Neolithic Taboos in Anatolia and Southeast Europe. In Social Dimensions of Food in the
Prehistoric Balkans, M. Ivanova, B. Athanassov, V. Petrova, D. Takorova, and P.W. Stockhammer, eds., 14- 30.
Oxford: Oxbow Books.
-. 2019. Spirit Birds at Neolithic <;atalhoyilk. Environmental Archaeology: The Journal of Human Palaeoecology
24: 377-386.
Russell, Nerissa and Bleda S. During. 2006. Worthy is the Lamb: A Double Burial at Neolithic <;atalhoyilk (Turkey).
Paleorient 32(1): 73- 84.
Russell, Nerissa, Katheryn C. Twiss, David C. Orton, and G. Arzu Demirergi. 2013. Changing Animal Use at
Neolithic <;atalhoyilk, Turkey. In Archaeozoology of the Near East X Proceedings of the Tenth International
Symposium on the Archaeozoology ofSouth-Western Asia and Adjacent Areas, B. De Cupere, V. Linseele, and S.
Hamilton-Dyer, eds., 45- 68. Leuven: Peeters.
Schirmer, Wulf. 1990. Some Aspects of the Building in the "Aceramic Neolithic" at <;ayonti Tepesi. World
Archaeology 21(3): 363-387.
Schmidt, Klaus. 2011. Gobekli Tepe: A Neolithic Site in Southeastern Anatolia. In The Oxford Handbook of Ancient
Schoop, Ulf-Dietrich. 2005a. Das anatolische Chalkolithikum. Eine chronologische Untersuchung zur vorbronzezeitlichen
Kultursequenz im nordlichen Zentralanatolien und den angrenzenden Gebieten. Urgeschichtliche Studien 1.
Grenzach-Wyhlen: Verlag Bernhard Albert Greiner.
-. 2005b. Early Chalcolithic in North-Central Anatolia: The Evidence from Bogazkoy-Buytikkaya. TUBA-AR 8: 15-
37.
-. 2009. Ausgrabungen in <;amhbel Tarlas12008. Archi:iologischer Anzeiger 2009/1: 56-66.
- . 2010. Some Thoughts on Social and Economic Development in Western Anatolia During the Fourth and Third
Millennia BC. In Archaeological Research in Western Central Anatolia, A.N. Bilgen, R. von den Hoff, S.
Sandalc1, and S. Silek, eds., 29-45. Kutahya: -09martPress.
- . 201 la. The Chalcolithic on the Plateau. The Oxford Handbook of Ancient Anatolia, S.R. Steadman and G.
McMahon, eds., 150-173. Oxford: Oxford University Press.
- . 201 lb. <;amhbel Tarlas1, ein metallverarbeitender Fundplatz des vierten Jahrtausends v. Chr. im nordlichen
Zentralanatolien. In Anatolian Metal V, -0. Yal<;m, ed., 53-68. Bochum: Deutsches Bergbau-MuseumBochum.
-. 2015. <;amhbel Tarlas1: Late Chalcolithic Settlement and Economy in the Budakozti Valley (North-Central
Anatolia). In The Archaeology of Anatolia: Recent Discoveries (2011- 2014), Volume I, S.R. Steadman and G.
McMahon, eds., 46-67. Newcastle-upon-Tyne: Cambridge Scholars Publishing.
Schoop, Ulf-Dietrich, Catriona Pickard, and Clive Bonsall. 2012a. Radiocarbon Dating Chalcolithic Buytikkaya.
Archi:iologischer Anzeiger 2012/1: 115-120.
Schoop, Ulf-Dietrich, Catriona Pickard, Clive Bonsall, Jayne-Leigh Thomas, Inrini Papadopoulou, and Amy Bogaard.
2012b. Arbeiten zum Material der Ausgrabungen in den chalkolithischen Siedlungen auf dem Bogazkoy-
Buyukkaya und in <;amhbel Tarlas1. Archi:iologischer Anzeiger 2012/1: 115.
Selover, Stephanie and Pmar Durgun. 2019. Reexamining Burials and Cemeteries in Early Bronze Age Anatolia. In
The Archaeology of Anatolia, Volume III: Recent Discoveries (2017-2018) , S.R. Steadman and G. McMahon,
eds., 271-283. Newcastle Upon Tyne: Cambridge Scholars Publishing.
Senyurek, Muzaffer. 1950. Buyi.ik Gullucek'te Bulunan Kalkolitik <;::aga Ait Bir Muharibin tskeletinin Tetkiki. Ankara
Oniversitesi Di! ve Tarih-Cografya Fakiiltesi Dergisi 8(3): 269- 310.
- . 1957. Buyi.ik Gullucek'te Bulunan insan 1skeleti. In Turk Tarih Kurumu Tarafindan Yapzlan Biiyiik Giilliicek
Kazzsz: 1947 ve 1949'daki <;ah~malar Hakkmda ilk Rapor = Ausgrabungen von Biiyiik Giilliicek: Ausgefiihrt
durch die Tiirkische Historische Gesellschafi Vorbericht iiber die Arbeiten von 1947 und 1949, H.Z. Ko§ay Zubeyr
and M. Akok, eds., 50. Turk Tarih Kurumu Yaymlanndan, V. Dizi; Sa. 16. Ankara: Turk Tarih Kurumu Bas1mevi.
Silistreli, Ugur. 1989. Kti§k Hoyi.ik'te Bulunan Kabartma insan ve Hayvan Figurleriyle Bezeli Vazolar. Belleten
53(206): 362- 374.
- . 1990. 1989 Kti§kHtiyi.ikKaz1s1. Kazi Sonu9lan Toplanlls1 12(1): 95-104.
Starmer, W.E.H. 1998. Some Aspects of Aboriginal Religion. In Religious Business. Essays on Australian Aboriginal
Spirituality, M. Charlesworth, ed., 1- 23. Cambridge: Cambridge University Press.
Steadman, Sharon R. 1995. Prehistoric Interregional Interaction in Anatolia and the Balkans: An Overview. Bulletin of
the American Schools of Oriental Research 299/300: 13- 32.
- . 2000. Spatial Patterning and Social Complexity on Prehistoric Near Eastern 'Tell' Sites: Models for Mounds.
Journal ofAnthropological Archaeology 19: 16499.
Steadman, Sharon R. and Gregory McMahon. 2015. Recent Work (2013- 2014) at <;::adrr Hoyi.ik on the North Central
Anatolian Plateau. In The Archaeology of Anatolia: Recent Discoveries (2011-2014). Volume I. S.R. Steadman
and G. McMahon, eds., 69-97. Newcastle Upon Tyne: Cambridge Scholars Publishing.
Steadman, Sharon R., Laurel D. Hackley, Stephanie Selover, Burcu Y1ldmm, Madelynn von Baeyer, Benjamin
Arbuckle, Ryan Robinson, and Alexia Smith. 2019a. Early Lives: The Late Chalcolithic and Early Bronze Age at
yadir Hoyi.ik. Journal ofEastern Mediterranean Archaeology and Heritage Studies 7(3): 271-98.
Steadman, Sharon R., Gregory McMahon, Benjamin S. Arbuckle, Madelynn von Baeyer, Alexia Smith, Burcu
Yildmm, Laurel D. Hackley, Stephanie Selover, and Stefano Spagni. 2019b. Stability and Change at yadir Hoyi.ik
in Central Anatolia: A Case of Late Chalcolithic Globalisation? Anatolian Studies 69: 21-57.
Steadman, Sharon R., Gregory McMahon, and Jennifer C. Ross. 2019c. Chalcolithic, Iron Age, and Byzantine
Investigations at yadir Hoyi.ik: The 2017 and 2018 Seasons. In The Archaeology of Anatolia: Recent Discoveries
(2017-2018). Volume Ill, S.R. Steadman and G. McMahon, eds., 32-52. Newcastle Upon Tyne: Cambridge
Scholars Publishing.
Steadman, Sharon R., Gregory McMahon, Jennifer C. Ross, Marica Cassis, T. Emre Serifoglu, Benjamin S. Arbuckle,
Sarah E. Adcock, Songul Alpaslan Roodenberg, Madelynn von Baeyer, and Anthony J. Lauricella. 2015. The 2013
and 2014 Seasons of Excavation at yadir Hoyi.ik on the Anatolian North Central Plateau. Anatolica 41: 87-124.
Steadman, Sharon R., Gregory McMahon, T. Emre Serifoglu, Marica Cassis, Anthony J. Lauricella, Laurel D.
Hackley, Stephanie Selover, Burcu Y1ldmm, Benjamin S. Arbuckle, Madelynn von Baeyer, Yagmur Heffron,
Katie Tardio, Sarah Adcock, Emrah Din<;, Gonca Ozger, Bengi Selvi, Stephanie Offutt, and Alicia Hartley. 2019d.
The 2017- 2018 Seasons at <;::adrr Hoyi.ik on the North Central Plateau. Anatolica 45: 77- 119.
Steadman, Sharon R., T. Emre Serifoglu, Stephanie Selover, Laurel D. Hackley, Burcu Y1ldmm, Anthony J.
Lauricella, Benjamin S. Arbuckle, Sarah E. Adcock, Katie Tardio, Emrah Dins;, Gregory McMahon, and Marica
Cassis. 2017. Recent Discoveries (2015-2016) at <;::adrr Hoyi.ik on the Anatolian North Central Plateau. Anatolica
43: 203-50.
Sweatman, Martin B. and Dimitrios Tsikritsis. 2017. Decoding Gtibekli Tepe with Archaeoastronomy: What Does the
Fox Say? Mediterranean Archaeology and Archaeometry 17(1): 233-250.
Tatsumi, Yuki. 2020. A Neolithic Sedentary Hunter-Gatherer Settlement with Densely Arranged Buildings: Results of
Geophysical Prospection at Hasankeyf Hoyi.ik in South-eastern Anatolia. Archaeological Prospection.
https://doi.org/10.1002/arp. l 777.
Thissen, Laurens. 1993. New Insights in Balkan-Anatolian Connections in the Late Chalcolithic: Old Evidence from
the Turkish Black Sea Littoral. Anatolian Studies 43: 207- 237.
Thomas, Jayne-Leigh. 2011. Preliminary Observations on the Human Skeletal Remains from <;::amhbel Tarlas1.
Archaologischer Anzeiger: 73-76.
- . 2017. Late Chalcolithic Skeletal Remains and Associated Mortuary Practices from <;::amlibel Tarlas1 in Central
Anatolia. In Children, Death, and Burial: Archaeological Discourses, E. Murphy and M. Le Roy, eds., 77-90.
Oxford: Oxbow Books.
Turkcan, Ali U. 2007. Is it Goddess or Bear? The Role of <;::atalhtiyi.ik Animal Seals in Neolithic Symbolism.
Documenta Praehistorica XXXIV: 257-266.
- . 2017a. At the Crossroads: Changing Chalcolithic Settlement Patterns in Phrygia. In Communities in Transition:
The Circum-Aegean Area During the 5th and 4th Millennia BC, S. Dietz, F. Mavridis, Z. Tankosic, and T.
Takaoglu, eds., 556-566. Oxford: Oxbow Books.
- . 2017b. Kanht~ Hoyi.ik (Eski§ehir/inonu) 2017 Y1h Kazi <;::ah§malan Raporu.
http://www.kanlitas.com/assets/files/2017.pdf
Umurtak, Gulsun. 2011. Understanding the Hac1lar II Settlement. Ada/ya XIV: 1- 16.
Vandam, Ralf and Patrick T. Willett. 2020. The Importance of Intensive Surveys in Late Prehistoric Research: A
Pisidian Case Study. In Production, Trade and Economy in Pisidia and its Surrounding Areas, A. Morel, G.
Ka~ka, H.N. Koker, M.T. Ka~ka, M. Frrat, and S.O. Akdtinul, eds., 70-83. Isparta: Suleyman Demirel Dniversitesi
Yaymlar1.
Vandam, Ralf, Patrick T. Willett, and Jeroen Poblome. 2019. The Results of the 2017 Derekoy Archaeological Survey
by the Sagalassos Project. In The Archaeology of Anatolia, Volume III- Recent Discoveries (2017- 2018) , S.R.
Verhoeven, Marc. 2002a. Transformations of Society: The Changing Role of Ritual and Symbolism in the Pre-Pottery
Neolithic Band Pottery Neolithic Periods in the Levant and South-east Anatolia. Paleorient 28(1): 5-14.
- . 2002b. Ritual and Ideology in the Pre-Pottery Neolithic B of the Levant and Southeast Anatolia. Cambridge
Archaeological Journal 12(2): 233-258.
- . 2011. Retrieving the Supernatural: Ritual and Religion in the Prehistoric Levant. In The Oxford Handbook of The
Archaeology ofRitual and Religion, T. Insoll, ed., 795-810. Oxford: Oxford University Press.
Voigt, Mary. 2007. The Splendour of Women: Late Neolithic Images from Central Anatolia. In Image and Imagination:
A Global Prehistory of Figurative Representation, C. Renfrew and I. Morley, eds., 157- 175. Cambridge:
von Baeyer, Madelynn, Alexia Smith, and Sharon R. Steadman. 2021. Expanding the Plain: Using Archaeobotany to
Examine Adaptation to the 5.2 kya Climate Change Event During the Anatolian Late Chalcolithic at C,::adn Hoyillc.
Journal ofArchaeological Science: Reports 36: 102806.
von der Osten, Hans Henning. 1937. The Alishar Huyiik. Seasons of 1930- 32 (Part 1). University of Chicago Oriental
Institute Publications 28, Researches in Anatolia 7. Chicago: University of Chicago Press.
Watkins, Trevor. 1990. The origins of house and home? World Archaeology 21: 336- 347;
https://doi.org/10.1080/00438243.1990.9980112
- . 2004. Building houses, framing concepts, constructing worlds. Paleorient 30: 5-24.
- . 2015. Religion as Practice in Neolithic Societies. In Defining the Sacred· Approaches to the Archaeology of
Religion in the Near East, N. Laneri, ed., 153-160. Oxford: Oxbow Books.
Whitehouse, Harvey and Ian Hodder. 2010. Modes of religiosity at C,::atalhoyillc. In Religion in the Emergence of
Civilization: <;atalhoyiik as a Case Study, I. Hodder, ed., 122-145. Cambridge: Cambridge University Press ..
Yakar, Jak. 2003. The Language of Symbols in Prehistoric Anatolia. Documenta Praehistorica 32: 111- 121.
- . 2012. The Nature of Symbolism in the Prehistoric Art of Anatolia. In All the Wisdom of the East. Studies in Near
Eastern Archaeology and History in Honor ofEliezer D. Oren, M. Gruber, S. Al).ituv, G. Lehmann, and Z. Talshir,
eds., 431-451. Gottingen: Academic Press Fribourg.
Y1ldmm, Burcu, Laurel D. Hackley, and Sharon R. Steadman. 2018. Sanctifying the House: Child Burial in
Prehistoric Anatolia. Near Eastern Archaeology 81(3): 164- 173.
CONTRIBUTORS
Murat Akar, Associate Professor of Archaeology, Hatay Mustafa Kemal University; Director of Excavations at Tell
Atchana, Alalakh and the Amuq Valley Regional Project. Recent publications include "The Formation of Collective,
Political and Cultural Memory in the Middle Bronze Age: Foundation and Termination Rituals at Toprakhisar
Hoyilk," with D. Kara (Anatolian Studies 70, 2020); Tell Atchana, Alalakh Volume 2: The Late Bronze II City. 2006-
2010 Excavations, K.A Yener, M. Akar, & M.T. Horowitz, eds., (Istanbul, 2019). Current research includes: social
memory, center-periphery dynamics and landscape archaeology, and the role of climate change in understanding
shifting population dynamics and cross-cultural encounters in the Eastern Mediterranean, funded by T0B1TAK 1001:
"Geological and Archaeological Traces of Climatic Changes in Amuq Valley during the Holocene."
N. Ezgi Altmt~lk, Researcher, Department of Anthropology, Hacettepe University. Recent publications include
"Palaeo-Eskimo genetic ancestry and the peopling of Chukotka and North America," with Flegontov et al. (Nature
570, 2019) ; "Circumpolar peoples and their languages: lexical and genomic data suggest ancient Chukotko-
Kan1chatkan- Nivkh and Yukaghir-Samoyedic connections," with Starostin et al. (preprint, 2021) and "Variable
kinship patterns in Neolithic Anatolia revealed by ancient genomes," with Yaka et al. ( Current Biology, 2021 ).
Oguz Aras graduated from Erzurum Atatork University, Department of Archaeology, in 2013 before commencing
postgraduate studies in the Protohistory and Near Eastern Archaeology Department at the Social Science Institute in
2015, where he is currently a Ph.D. student. He completed his doctorate in 2021. He has participated in many
excavations and surface survey prqjects in eastern and southeastern Anatolia, including working at the Van Ayanis
Castle excavations since 2011.
Nurettin Arslan, Prof. Dr., Department of Archaeology, Faculty of Arts and Science, c;anakkale Onsekiz Mart
University. Recent publications include Neue Forschungsergebnisse zur Baugeschichte und Archiiologie der
siidlichen Troas, with E.M. Mohr & K. Rheidt, eds. (Asia Minor Studien 78, 2016); "Cityscape and Places of Memory
in Assos," with E.M. Mohr & K. Rheidt, in Cityscapes and Monuments of Western Asia Minor, E. Mortensen & B.
Poulsen, eds. (Oxford, 2017); Urbanism and Architecture in Ancient Aiolis. Proceedings of the International
Conference from 7h_gth April 2017 in c;anakkale, with E.M. Mohr & K. Rheidt (Asia Minor Studien 95, 2020).
Notable projects include Assos excavations in c;anakkale since 2006 and surface surveys in the south Troad (Gargara,
Lamponia) since 2009.
<;igdem Atakuman, Associate Professor, Middle East Technical University of Ankara; Deputy Head of the Ugurlu
excavations; director of the Bozburun Survey Project; team member of ERC Project, METU Ancient DNA. Notable
publications include "Cradle or crucible: Anatolia and archaeology in the early years of the Turkish Republic (1923-
1938)" (Journal of Social Archaeology 8, 2008); "Architectural discourse and social transformation during the early
Neolithic of southeast Anatolia" (Journal of World Prehistory 27, 2014); "From monuments to miniatures: emergence
of stamps and related image-bearing objects during the Neolithic" (Cambridge Archaeological Journal 25, 2015).
Ay\'.a Aydogan, Ph.D. candidate, Department of Anthropology, Hacettepe University. Recent studies include genetic
comparison between Neolithic hunter-gatherers from mainland Sweden and the other Neolithic and Mesolithic
Scandinavian groups. Recent related projects involved include "Archaeogenomic Analysis of Genetic and Cultural
Interactions in Neolithic Societies-NEOGENE" (ERC 772390).
Caner Bakan, Research Assistant in the Archaeology Department at c;anakkale Onsekiz Mart University. Recent
publications include "The Cide Region During the Hellenistic Period," with T.E. ~erifoglu, in Kinetic Landscapes.
Cide Archaeological Project 2009- 2011: Surveying the Western Black Sea Region, B.S. During and C. Glatz, eds.
(Berlin, 2015).
Claire Barat, Associate Professor in Ancient History and Classical Archaeology, Hauts-de-France Polytechnical
University (Valenciennes, France), CRISS Research Laboratory, Associate Researcher, French Institute of Anatolian
Studies (Istanbul, Turkey). Recent publications include "L'Halys, un fleuve structurant en Asie Mineure" in Etudes
desjleuves d 'Asie Mineure dans l'Antiquite (Arras, 2018); "La Colonia Julia Felix Sinope: d'une ironie tragique a une
realite" in L 'heritage grec des colonies romaines d'Orient. Interactions culturelles dans les provinces hellenophones
de !'empire romain (Paris, 2017); "Salaisons de la mer Noire: commerce et gastronomie antiques" in Se/ et societe.
Tome 1: Techniques- usages- langage (Villeneuve d'Asq, 2017).
Ay~e Batman, M.A., Graduate Student at the Graduate Program in Settlement Archaeology, Middle East Technical
University. Recent and notable projects include Kinet Hoyilk Excavations, Dortyol-Hatay, Turkey (2001-2005),
Komana Archaeological Research Project (2018-2021).
The Archaeology of Anatolia, Volume IV: Recent Discoveries (2018-2020) 397
Alexander Bauer, Associate Professor of Anthropology, Queens College and The Graduate Center, City University
of New York. Notable work includes "Itinerant Objects" (Annual Review ofAnthropology 48, 2019), Co-Editor of The
Oxford Companion to Archaeology, 2nd ed., 3 vols. (Oxford University Press, 2012), and Social Archaeologies of
Trade and Exchange (Routledge, 2010); and service as Editor in Chief of the International Journal of Cultural
Property (Cambridge University Press).
Michael Blomer, Professor of Classical Archaeology, University of Munster; Co-director of the excavations at
Doliche, South-East Turkey. Recent publications include "The Diversity of Funerary Commemoration in Roman
Commagene and Cyrrhestice" in Funerary Portraiture in Greater Roman Syria, M. Blomer & R. Raja, eds.
(Turnhout, 2019); "Observations on Cities and their Biographies in Hellenistic North Syria," (Journal of Urban
Archaeology 2, 2020). His research revolves around Asia Minor and the Near East in the Hellenistic and Roman
periods. He has worked on urbanism, sculpture, religious iconography, and ancient religion.
Scott Branting, Associate Professor in the Department of Anthropology at the University of Central Florida; Director
of the Kerkenes Project. He has worked with the project since 1995 and has been a director since 2003. He received
his Ph.D. in Anthropology from the University at Buffalo, an M.A. in Near Eastern Languages and Civilizations from
the University of Chicago, and an M.A. in Geography from the University at Buffalo.
Guillaume Bruniaux, Ph.D. in geophysics applied to archaeology, guest researcher at LIENSs laboratory, University
of La Rochelle (France). Recent notable publication: "Data Processing Chain to High Spatial Resolution Magnetic
Survey: Application on the Neolithic Site of Le Pontet (Charente-Maritime, France)," with V. Mathe, F. Leveque, A.
Camus, & V. Ard (Archaeological Prospection 24, 2017).
Mirco Brunner, SNSF Early Post-Doctoral Mobility Fellow at the Institute for Ecosystem Research, Geoarchaeology
and Polar Ecology, Kiel University, and associated researcher at Oeschger Centre for Climate Change Research
(OCCR), University of Bern. Notable publications include "Central European Early Bronze Age chronology revisited:
A Bayesian examination of large-scale radiocarbon dating," with J. von Felten, M. Hinz, & A. Hafner (PLoS ONE 15,
2020).
Milge Bulu, received her Ph.D. from Ko9 University, Department of Archaeology and History of Art. She is the
assistant director of ceramic studies at Tell Atchana, Alalakh Excavations. Notable publications include "A Syro-
Cilician Pitcher from a Middle Bronze Age Kitchen at Tell Atchana, Alalakh" in Overturning Certainties in Near
Eastern Archaeology. A Festschrift in Honor of K. Aslzhan Yener, <;. Maner, M.T. Horowitz, & A.S. Gilbert, eds.
(Leiden, 2017); "Basins/tubs, drainage pipes, portable hobs, and miscellaneous objects" in Tell Atchana, Alalakh.
Volume 2: The Late Bronze II City, the 2006-2010 Excavation Seasons, K.A. Yener, M. Akar, & M.T. Horowitz, eds.
(Istanbul, 2019).
W. Christopher Carleton, Postdoctoral researcher and Archaeologist/Palaeo-Data Scientist, Independent Extreme

Events Research Group, Max Planck Institute for Chemical Ecology, the Science of Human History. Recent
publications include: "Climate change, not human population growth, correlates with the Late Pleistocene megafauna
extinctions in North America," with M. Stewart & H. Groucutt (Nature Communications, 2021 [DOI:10.1038/s41467-
021-21201-8]); "Evaluating Bayesian radiocarbon-dated event-count (REC) models for the study of long-term human
and environmental processes" (Journal of Quaternary Science, 2020 (DOI: 10.1002/jqs.3256]); "Sum things are not
what they seem: Problems with point-wise interpretations and quantitative analyses of proxies based on aggregated
radiocarbon dates," with H. Groucutt (The Holocene, 2020 [DOI: 10.1177/0959683620981700]). Research interests
include uncertainty quantification, time series analysis, Bayesian methods, and GIS.
Lorenzo Castellano, Ph.D. candidate at the Institute for the Study of the Ancient World, New York University.
Recent and notable publications include "A new anthracological sequence from Nigde-Km1k Hoyillc (Turkey):
woodland vegetation and arboriculture in southern Cappadocia from the Late Bronze Age to the Ottoman Period"
(Anthropological and Archaeological Sciences 13, 2021); "Staple economies and storage in Post-Hittite Anatolia.
New data from Nigde Km1k Hoyillc (Southern Cappadocia)" (Journal of Eastern Mediterranean Archaeology and
Heritage Studies 6, 2018).
Ali (if\:i, Assistant Professor of Ancient History, Marmara University; Assistant Director of the Elbistan Karahoyiik
and Hamzatepe excavations. He has directed the Kahramanmar~ Survey Project since 2019. Notable publications
include: The Socio-Economic Organisation of the Urartian Kingdom (Brill, 2017), "The Economy of the Urartian
Kingdom," in Archaeology and History of Urartu (Biainili) , G.R. Tsetskhladze, ed. (Peeters, 2021); "Urartian Studies
during the Foundation of Modem Nation States: Archaeological Traditions, Ideologies and Perceptions" (Ancient West
& East 19, 2020); "Urartian Irrigation Systems" (Ancient Near Eastern Studies 50, 2013); Elbistan Karahoyiik
Kaz1s1," with B. Uysal (Tarih fncelemeleri Dergisi 41, 2019).
Ashhan <;ivilidag, Graduate Student at the Archaeology Department, istanbul University. Recent projects include:
Komana Archaeological Research Project (2014-2021).
398 Contributors
Anacleto D 'Agostino, Associate Professor of Archaeology and Art History of the Ancient Near East at the University
of Pisa. He has participated in excavations and surveys in Syria, Turkey, and Iraq. Recent publications include
'Tracing fire events and destructions of Late Bronze Age date: the end of the Hittite Building on the citadel of U ~akh
Hoyuk," in Anatolia between the 13th and the 12th century BCE, de Martino & Devecchi eds. (Firenze, 2020); "A
mosaic floor from the Late Bronze Age Building II ofU~akh Hoyuk, central Turkey" (Antiquity 93/372, 2019); L 'Alto
Tigri nelle eta de/ Bronzo Antico e Media. Siti, sequenze e ceramiche rosso-brune tra fine III e inizio II mill. a. C
(Firenze, 2016).
Owen Doonan, Professor of Art History, CSU Northridge; Director, Sinop Regional Archaeological Project (SRAP)
(1996-2012) and Sinop Kalesi Excavations (2015-) under the auspices of the Sinop Museum. Doonan has authored
50+ articles and a monograph (Sinop Landscapes, UPenn Museum Press 2004). He has served as PI on numerous
grants in support of the work at Sinop, including two three-year grants from the National Endowment for the
Humanities, grants from the National Geographic Society, ARIT, and other sources. He was recognized by the AIA as
the G.M.A. Hanfinann Lecturer (2016-2017) and has received fellowships from the NEH, the Getty Villa, and the
Alexander von Humboldt Foundation.
Miige Durusu-Tanritiver, Assistant Professor in the Department of Urban Design and Landscape Architecture,
Bilkent University. Her research focuses on manifestations of ancient Near Eastern empires in their borderlands,
particularly in Anatolia and northern Mesopotamia. Since 2019, Durusu-Tannover has been directing the Polath
Landscape Archaeology and Survey Project.
D. Burcu Erciyas, Professor at the Graduate Program in Settlement Archaeology, Middle East Technical University.
Recent publications include Komana Small Finds, Settlement Archaeology Series 7, with M. Acara Eser (istanbul,
2019); "Evaluation of the Recent Finds at Komana from the Early and Middle Byzantine Period," with M.N. Tatbul,
in Settlements and Necropoleis of the Black Sea and its Hinterland in Antiquity, G.R. Tsetskhladze and S. Atasoy, eds.
(Oxford, 2019), and "A Preliminary Study on Roman Komana" with M.N. Tatbul, in The Black Sea in Light of Nf;W
Archaeological Data and Theoretical Approaches, M. Manoledakis, ed. (Oxford, 2016).
Ytlmaz Selim Erdal, Professor, Department of Anthropology, Hacettepe University, Ankara. Recent publications
include "Bone or Flesh: Defleshing and Post-Depositional Treatments at Kortik Tepe (Southeastern Anatolia, PPNA
Period)" (European Journal of Archaeology 18, 2015); "Modified bodies: an interpretation of social identity
embedded into bones," with V. D'Amico, in The Sacred Body: Materializing the Divine through Human Remains in
Antiquity, 115-131, N. Laneri, ed. (Oxbow, 2021); "Variable kinship patterns in Neolithic Anatolia revealed by
ancient genomes," with Yaka et al. (Current Biology 31, 2021). Recent related projects involved include
"Archaeogenomic Analysis of Genetic and Cultural Interactions in Neolithic Societies-NEOGENE" (ERC 772390).
Bur-;in Erdogu. Professor, Akdeniz University of Antalya; Director of excavations at Ugurlu on the island of
Gok<;eada. Notable excavation project sites include <;atalhoyuk West Mound and Girmeler Cave; prehistoric surveys
in Turkish Thrace and the Salt Lake region of Central Anatolia. Notable publications include "Visualizing Neolithic
Landscape: Archaeological Theory in Aegean Islands," (BJ'ZAS 2, 2005); "Whither the Aegean Neolithic," in Going
West? A. Reingruber, Z. Tsirtsoni, & P. Nedelcheva, eds. (Taylor & Francis 2017); "An Entanglement Approach to
the Neolithic of the Aegean Islands," in Concluding the N eolithic, A. Marciniak, ed. (Lockwood, 2019).
Duygu Ertemin, Ph.D. student in the Anthropology department at McMaster University, Canada. She also works in
the department as a teaching assistant. She has worked with the Kanhta~ Excavation Project since 2013 and directed
the field operations in 2018. She was a member of the <;atalhoyuk Research Project, directed by Professor Ian Hodder,
as a field archaeologist between 2014 and 2017. She works in the Ontario CRM industry and currently is a member of
the Konya <;atalhoyuk Excavation and Research Project, directed by Associate Professor Ali Umut Ti.irkcan. Recent
work includes a digital archaeology project at the Sherman Centerfor Digital Scholarship (https://scds.ca/2021-
sherman-graduate-residentsD .
Biilent Gen-;, Associate Professor of Archaeology, Mardin Artuklu University. Recent publications include
"Archaeology of Destruction: Toprakkale" (Iraq 80, 2018); "First Steps in the Archaeology of Assyria: Botta's Letters
and the 'Excavation House' at Khorsabad." (Iraq 81, 2019); "From Khazane Kapoussi/Hazine Kap1s1 to Anahklz:
Rethinking a Place at Tuspa Citadel," in Over the Mountains and Far Away: Studies in Near Eastern History and
Archaeology Presented to Mirjo Salvini on the Occasion of his 80th Birthday, P.S. Avetisyan, R. Dan, & Y.H.
Grekyan, eds. (Oxford, 2019).
Andrew L. Goldman, Professor, Department of History, Gonzaga University; formerly the Arnold Distinguished
Professor (2015-2018). His research has focused on material culture in the Roman world, in particular within the
Roman East and its Anatolian provinces. In addition to his work as field director at Sinop, he is currently publishing
the Roman excavations at Gordion (Turkey) and Roman Republican military equipment recovered by the Aegates
Islands nautical survey project (Sicily).
The Archaeology of Anatolia, Volume IV: Recent Discoveries (20 I 8 2020) 399
Emine Koker Gok-;e, Dr. Lecturer of Classical Archeology, Hac1 Bektal] Veli University. Recent publications include
"Loom weights in Isaura" (Arkeoloji ve Sanat 156, 2017); "Roman Ceramics with Gray and Red Slip Stamped and
Rouletted Decoration in Isaura Ancient City and Its Surrounding" (Arkeoloji ve Sana! 159, 2018); "Glass Finds in
Porsuk /Zeyve Hoyi.ik" (Colloquium Anatolicum 19, 2020).
Sarah R. Graff, Principal Lecturer and Honors Faculty Fellow at Barrett, Tue Honors College of Arizona State
University. She is researching the production, use, and exchange of ceramic containers at Kerkenes and the broader
economy of the city. Recent publications include "Production requires water: Material remains of the hydrosocial
cycle in an ancient Anatolian city," with S.A. Branting and J.M. Marston (Economic Anthropology 6, 2019).
Amanda T. Groff, Senior Lecturer of Anthropology, University of Central Florida. Dr. Groff's primary area of
research utilizes stable isotopes to determine migration and mobility of ancient individuals. She has been associated
with the Caracol Archaeological Project in Belize (2002-2010), the Pietramarina Project in Italy (2012-2014), the
Dakhleh Oasis Project in Egypt (2007-present), and the Kerkenes Project (2019-present). Dr. Groff is also involved in
CRM work in the state of Florida.
Sinem Hac10smanoglu, Doctoral candidate, Institute for Archaeological Sciences - Geoscience, Eberhard Karls
University of Tubingen. Notable publications include "Geochemical and Petrographic Analysis of Late Bronze Age
Cypriot Ceramics (White Slip I and II and Monochrome) from Tell Atchana/Alalakh (Hatay) in the Amuq Valley,"
with M. Kibaroglu, G. Sunal, E. Koza!, & P. Gutsuz (Archaeometry 60, 2017); "Geochemical characterization of clay
deposits in the Amuq Valley (Southern Turkey) and the implications for archaeometric study of ancient ceramics,"
with P. Gutsuz, M. Kibaroglu, & G. Sunal (Applied Clay Science 141, 2017).
Omiir Harman~ah, Associate Professor of Art History, University of Illinois at Chicago. Notable publications
include Cities and the Shaping ofMemory in the Ancient Near East (Cambridge, 2013); Place, Memory, and Healing:
An Archaeology of Anatolian Rock Monuments (Routledge, 2015). Since 2010, Harman~ah has directed Yalburt
Yaylas1 Archaeological Landscape Research Project. His current research focuses on the history of landscapes in the
Middle East and the politics of ecology, place, and heritage in the age of the Anthropocene.
Tara Ingman, Post-doctoral fellow, Kos; -Oniversitesi Anadolu Medeniyetleri Ara~t1rma Merkezi/Ko9 University
Research Center for Anatolian Civilizations. She is the assistant director of publications at Tell Atchana, Alalakh
Excavations. Notable publications include "M011uary Practices and GIS Modeling at Tell Atchana, Alalakh," in
Alalakh and its Neighbors: Proceedings of the 15th Anniversary Symposium at the New Hatay Archaeology Museum,
June 10-12, 2015, K.A. Yener & T. Ingman, eds. (Leiden, 2020); "Laid to Rest: LB II Burials" with R. Shafiq, in Tell
Atchana, Alalakh. Volume 2: The Late Bronze II City, the 2006-2010 Excavation Seasons, K.A. Yener, M. Akar, &
M.T. Horowitz, eds. (Istanbul, 2019).
Kenan l~tk, Independent Researcher, Van City, Turkey. Recent publications include "The Recently Found Upper Part
of the Inscribed Arin Stele ofUrartian King Argisti I: About a Royal Vineyard and Orchard," with B. Gens; (Nouvelles
Assyriologiques Breves et Utilitaires n° 1 mars, 2021); "A new Urartian temple in Korzut fortress, Turkey: a report on
the rescue excavation of 2016 and new approaches on the origin of Urartian square temple architecture," with R.
Kuvan9 & B. Gens; (Annenian Journal of Near Eastern Studies XIV/1-2, 2020); "The Irrigation Canal Stele of the
Urartian King Argisti I Recently Discovered in Erci~/Salmanaga, North of Lake Van" (Zeitschrift fur Assyriologie und
Vorderasiatische Archaologie 109, 2019).
Mehmet l~tkh, Professor of Archaeology, Ataturk University. Director, Ayanis excavations since 2013. Interests
include South Caucasian and East Anatolian Archaeology. Notable publications include Dogu Anadolu Erken
Transkajkasya Kulturu: <;ok Bile~enli Geli~kin Bir Kulturun Analizi (Arkeoloji ve Sanat Yaymlan, 2011); editor of
International Symposium of Eastern Anatolia and Southern Caucasus Cultures, with B. Can (Newcastle upon Tyne,
2015); "The Kura-Araxes Culture in the Erzurum Region: The Process of its Development" (TUBA-AR 18, 2015). He
has published an extensive number of articles, papers, presentations, projects, and books regarding the archaeology of
this remarkable and still mysterious region of eastern Anatolia and the southern Caucasus.
Peri Johnson, adjunct Assistant Professor, Department of Art History at the University of Illinois at Chicago. Notable
publications include "How did the landscape of Pompeiopolis become Roman?" in Landscape Dynamics and
Settlement Patterns in Northern Anatolia during the Roman and Byzantine Period, K. Winther-Jacobsen & L.
Summerer, eds. (Stuttgart 2015). She specializes in the material and visual culture of the Iron Age on the Anatolian
Peninsula, particularly the Achaemenid period.
Tuna Kalayc1, Assistant Professor of Computational Archaeology, Leiden University. He is specializing in remote
sensing and GIS applications in archaeology. Previously he was a Marie Sklodowska-Curie, Individual Fellow at the
Consiglio delle Ricerche Nazionale (CNR, IBAM), Italy and Durham University, the UK. He also worked as a post-
doctoral scholar at IMS-FORTH, Greece, with a focus on geophysics.
400 Contributors
Salih Kaymak-;1, Associate Professor, Erzincan Binali Y1ldmm Universitesi. Recent publications include "Dogu
Karadeniz Bolgesi Ges; Demir <;ag1 Kultilii.ine Ait Yeni Arkeolojik Bulgular ve Karadikmen Kalesi" (Karadeniz
Uluslararas1 Bilimsel Dergi. 46, 2020); "Dogu Pontos Bolgesi Kaleleri ve Basamakh Tunellerine Bir Omek-Kaledere
Kalesi-<;amoluk Yeni Bulgular l~@nda Arkeolojik Bir Degerlendirme (Karadeniz jncelemeleri Der7:i,si 28, 2020); "Giresun
and its Vicinity in the Greek Colonisation Period, in The Greeks and Romans in the Black Sea and the Importance of the
Pantie Region in the Graeco-Roman World, G.R. Tsetskhladze, A. Avram, and J. Hargrave, eds. (Oxford, 2021).
Duygu Deniz Kazanc1, Postdoctoral researcher, Hacettepe University Molecular Anthropology Group (Human_G)
Hacettepe University and Ancient DNA Laboratory, Middle East Technical University. Recent related projects
involved include "Archaeogenomic Analysis of Genetic and Cultural Interactions in Neolithic Societies-NEOGENE,"
(ERC 772390).
Erkan Konyar, Associate Professor of Archaeology, istanbul University. Recent publications include "Urartian and
Post-Urartian Periods at Van Fortress Mound in Light of New Excavations" in Urartians: A Civilization in the Eastern
Anatolia. The Proceedings of the 1st International Symposium held at jstanbul in 13-15 October, 2014, A.
<;ilingiroglu, K. Koroglu, Z. <;ulha, & G. Oncu, eds. (istanbul, 2018); "A New Rock-Cut Tomb in Van
Fortress/Tushpa," in Over the Mountains and Far Away: Studies in Near Eastern History and Archaeology Presented
to Mirjo Salvini on the Occasion ofhis 80th Birthday, P.S. Avetisyan, R. Dan, & Y.H. Grekyan, eds. (Oxford, 2019).
Ekin Kozal, Professor, Onsekiz Mart University. Notable publications include Fremdes in Anatolien. Importguter aus
dem Ostmittelmeerraum und Mesopotamien als Indikator fii.r spatbronzezeitliche Handels- und Kulturkontakte
(Schriften zur Vorderasiatischen Archaologie 11) (Harrassowitz, 2017); "The Dualistic Nature of a Red Lustrous
Wheel-made Bowl from Bogazkoy with a Depiction of a Victorious Armed Warrior" (Anatolian Studies 69, 2019).
Ulrike Krotscheck, Associate Professor of Classical Studies and Archaeology, The Evergreen State College. She has
excavated and consulted at numerous sites including Emporio, Lattara, Isthmia, and Miletus. Her scholarly
publications focus primarily on Greek ceramics in the diaspora, the cultural value of ceramics, and trade and exchange
on the fringes of the Mediterranean.
Sabina Kulemann-Ossen, Associated Researcher, Institute of Archaeological Sciences, Ancient Near Eastern
Archaeology, University of Bern. Recent publications include "Hybridity of Style: Iron Age Potte1y from Sirkeli
Hoyuk" with H. Monninghoff (Studi Micenei ed Egeo-Anatolici, Nuova Serie 5, 2019); "Materielle Verbindungen
zwischen der ljabfu-Region und Kilikien im 1. Jt. v. Chr., in Zwischen Ausgrabung und Ausstellung. Beitrage zur
Archaologie Vorderasiens. Festschrift Lutz Martin, with N. Cholidis & E. Katzy, eds. (Marru 9, 2020).
Dominique Kriiger, Research Assistant, Bogazkoy/ljattusa; Member of the DFG Research Group 2757 »Local Self-
Governance in the Context of Weak Statehood in Antiquity and the Modem Era«, Julius-Maximilians-Universitat
Wurzburg. Recent publications include "Securing a New Province. Activities of the Roman Military in 1st Century AD
Galatia," in Understanding Transformations. Exploring the Middle Black Sea Region and Northern Central Anatolia
in Antiquity (c. 4th/3rd Century BCE - 4th/5 th Century CE), E. Sokmen and A. Schachner, eds. (Istanbul, 2021);
"Strukturen und Befunde der romischen Kaiserzeit," with S. Kuhn and A. Schachner, in "Die Ausgrabungen in
Bogazkoy/ljattusa 2019" (Archaologischer Anzeiger 2020/1).
Sven Kiihn, Research Assistant, Bogazkoy/ljattusa; Ph.D. student at Eberhard Karls Universitat Tubingen. Recent
publications include "Consolidation of Power? -A 2nd Century Representative Villa Complex in Hattusa," in
Understanding Transformations. Exploring the Middle Black Sea Region and Northern Central Anatolia in Antiquity
(c. 4th/3 rd Century BCE - 4th/5 th Century CE), E. Sokmen and A. Schachner, eds. (Istanbul, 2021).
Dominique Langis-Barsetti, Ph.D. candidate at the University of Toronto in the department of Near and Middle
Eastern Civilizations. She has worked with the Kerkenes Project since 2010 and currently serves as an Assistant
Director. She has been a member of the Computational Research on the Ancient Near East (CRANE) Project since
2012, working on a variety of mapping and outreach projects, and joined the Cape Gelidonya shipwreck research team
in 2017 as a photogrammetry and 3D recording specialist. Her research focuses on Anatolian archaeology, geophysics,
ancient urbanism, 3D modeling and simulation, and remote sensing.
Joseph Lehner, Australian Research Council Discovery Early Career Award Fellow at The University of Sydney
Department of Archaeology; Co-director of The Kerkenes Project. As a specialist in ancient technologies, political
economy, and archaeological materials science, Lehner is also co-director of the Cape Gelidonya Shipwreck Project
and co-director of the Archaeological Water Histories Project in Oman.
<;igdem Maner, Assistant Professor, Kos; University. Recent publications include "Eine bedeutende Salzquelle in
Karapmar Konya - Meke Golu (Meke Maar) und ein Gleichsetzungsvorschlag fur liki ,,Salzlecke" im Staatsvertrag
des Kuruntija und TutlJ.alija IV" (Altorientalische Forschungen 48/2, 2021); "Archaologische Forschungen am
Karacadag und eine hieroglyphenluwische Inschrift aus Karaoren," with M. Weeden & M. Alparslan (Altorientalische
Forschungen 48/2, 2021); "Form and Function of Ashlar in the Middle and Late Bronze Age Anatolia," in Ashlar.
Exploring the Materiality of Cut-Stone Masonry in the Eastern Mediterranean Bronze Age, M. Devolder & I.
Kreimerman, eds. (Presses Universitaires de Louvain, 2020).
Paolo Maranzana, Assistant Professor in the Department of History at Bogazic;i University. His research focuses on
urban systems in the Roman and Late Roman periods in Anatolia and Italy. He has excavated at numerous sites in
Italy and Turkey including Gabii, Ostia, Sardis, Pessinus, and Sinop Kalesi. He has received grants from Bogazic;i
University, TOBiTAK, Dumbarton Oaks, and the Andrew W. Mellon Foundation. His current work considers the
impact of urbanism on local economies and the natural environment.
Ben Marsh, Professor of Geography and Environmental Studies, Bucknell University. Notable publications include
"Scales of Impact: Settlement History and Landscape Change in the Gordian Region, Central Anatolia," with L.
Kealhofer (The Holocene 24, 2014); "Ceramics, Trade, Provenience and Geology: Cyprus in the Late Bronze Age,"
with P. Grave, L. Kealhofer, U.-D. Schoop, J. Seeher, J.W. Bennett, and A. Stopic (Antiquity 88, 2014).
John M. Marston, Associate Professor of Anthropology and Director, Program in Archaeology, Boston University.
Recent publications include "Archaeological approaches to agricultural economies" (Journal of Archaeological
Research 29: 2021); "Agricultural practices at Bronze Age Kaymak91, western Anatolia" with N. Shin, C. Luke, S.
Riehl, and C.H. Roosevelt (Journal of Archaeological Science: Reports 36, 2021); Agricultural Sustainability and
Environmental Change at Ancient Gordian (University of Pennsylvania Museum Press, 2017).
Shannon Martino, faculty member in the Humanities Department of Morton College; member of the Yalburt Yaylas1
Archaeological Landscape Research Project since 2017. Her work focuses on the Chalcolithic and Early Bronze Age
ceramics and clay figurines from Turkey and southeastern Europe and their interconnections. She is also currently
examining the role that video games play in teaching about culture.
Vivien Mathe, Lecturer in geophysics applied to archaeology, University of La Rochelle (UMR 7266 LIENSs).
Recent publication: "Palaeoenvironmental and archaeological records for the reconstruction of the ancient landscape
of the Roman harbour of Narbonne (Aude, France)", with C. FaYsse, G. Bruniaux, J. Labussiere, J. Cavern, M.-P.
Jezegou, D. Lefevre, & C. Sanchez (Quaternary International 463, 2018).
Stefania Mazzoni, Full Professor of Archaeology and Art History of the Ancient Near East at the University of
Florence. Recent publications include: "Crisis and Fall: A View from U~akh Hoyuk and its Territory," in Anatolia
Between 13th and 12th Centuries BCE, de Martino & Devecchi eds. (Firenze, 2020); "Gable Seals from Tell Afis: a
group of Chalcolithic stamps across Syria and Anatolia," Pathways through Arslantepe. Essays in Honour ofMarcella
Frangipane, F. Balassi Restelli et al. eds. (Roma, 2020); "Northern Levant in EB III-IV: Economic Wealth and the
International Landscape of 'Second Urbanization,"' New Horizons in the Study of the Early Bronze III and Early
Bronze IV of the Levant, S. Richard, ed. (University Park, PA 2020).
Gregory McMahon, Professor of Classics and Ancient History, University of New Hampshire; Director of the <;ad1r
Hoyuk Archaeological Project (2010- 2019). Notable publications include The Hittite State Cult ofthe Tutelary Deities
(Chicago, 1991); The Oxford Handbook of Ancient Anatolia, ed. with S.R. Steadman (Oxford, 2011); "Agency and
Identity among the Hittites," in Agency and Identity in the Ancient Near East (Equinox, 2010); and "Stability and
Change at <;ad1r Hoyuk in Central Anatolia: A Case of Late Chalcolithic Globalisation?" with S.R. Steadman, B.
Arbuckle, M. von Baeyer, A. Smith, B. Y1ldmm, L.D. Hackley, Stephanie Selover, & S. Spagni (Anatolian Studies
69, 2019).
Hannah Monninghoff, Postdoctoral researcher, Bern University. Notable publications include Der Palas! in Nuzi
(Wiesbaden, 2020); "The Development of the Southeast Lower Town of Sirkeli Hoyuk," with A.E. Sollee, E. Kozal,
D. Karakaya, J. Heim, & S. Giir (Altorientalische Forschungen 47, 2020); "Hybridity of Styles: Iron Age Pottery from
Sirkeli Hoyuk" with S. Kulemann-Ossen (SN!EA NS 5, 2019).
Mirko Novak, Professor of Near Eastern Archaeology, Institute of Archaeological Sciences, University of Bern.
Notable publications include Sirkeli Hoyiik. Vorbericht der schweizerisch-tiirkischen Ausgrabungen 2006-2015. Ein
urbanes Zentrum am Puruna-Pyramos im Ebenen Kilikien, with E. Kozal & D. Y~in, eds. (Schriften zur
Vorderasiatischen Archiiologie 13, 2019); Urban Cultures of Central Asia from the Bronze Age to the Karakhanids
(2500 BCE-1200 CE). Learnings and conclusions from new archaeological investigations and discoveries, with C.
Baumer, eds. (Schriften zur Vorderasiatischen Archiiologie 12, 2019).
Valentina Orsi, Adjunct Professor of Settlements and Landscapes Archaeology at the University of Siena. She
worked in Upper and Central Mesopotamia, Central and Southeastern Anatolia. Notable publications include "The
Transition from the Bronze to the Iron Age at U~akh Hoyuk: The Ceramic Sequence," in Anatolia between the 13th
and the 12th century BCE, de Martino & Devecchi, eds. (Firenze, 2020); "Reading the Late Bronze Age Ceramic
Evidence at U~akh Hoyuk (Central Turkey). The Pottery from the Area A Test Sounding" (Anatolica 44, 2018); Crisi
e Rigenerazione nella valle dell'Alto Khabur (Siria) - La produzione ceramica nel passaggio dal Bronzo Antico al
Bronzo Media (Firenze, 2011).
402 Contributors
A. Tuba Okse, Professor, Kocaeli University. Notable publications include "The Middle Bronze Age Farming
Economy of the Upper Tigris Region: A Reconstruction Basing on Archaeological, Archaeobotanical and Ethno-
archaeological Analysis," with E.O. Donmez, A. Gormu~, M. Gurbuz, and M. Ozdeger (Akkadica 133, 2012); "The
Tradition of Burning the Corpse in the Upper Tigris Region: A Case Study on Zeviya Tivilki in the Upper Tigris
Region," with S. Eroglu (Akkadica 134, 2013); "Eastern Anatolian 'Early Iron Age' Tribes in the Upper Tigris
Region," in At the Northern Frontier ofNear Eastern Archaeology: Recent Research on Caucasia and Anatolia in the
Bronze Age, E. Rova, ed. (Turnhout, 2017); "Reflection on the Sunrise Positions in Early and Middle Bronze Age
Extramural Cemeteries in Anatolia" (Adalya 23, 2020).
Fiisun Ozer, Assistant Professor, Department of Anthropology, Hacettepe University, Ankara. Recent publications
include "Ancient genomics in Neolithic Central Anatolia and <;atalhoyiik" with Y aka, et al. in Peoplin:; the Landscape
of <;atalhoyiik: Reports from the 2009-2017 Seasons, I. Hodder, ed. (Oxbow, 2021); and "Variable kinship patterns in
Neolithic Anatolia revealed by ancient genomes," with Yaka et al. (Current Biology 31.11, 2021). Recent related
projects involved include "Archaeogenomic Analysis of Genetic and Cultural Interactions in Neolithic Societies-
NEOGENE" (ERC 772390).
Paige Paulsen, Ph.D. student in Near Eastern Studies at Johns Hopkins University. Her research interests focus on
Iron Age landscape archaeology in Turkey and Oman.
Jean-Fram;ois Pichonneau, Archaeologist, Regional Service of Archaeology, Nouvelle Aquitaine Regional

Directorate of Cultural Affairs, Ministry of Culture and Communication. Notable projects sites include Porsuk-Zeyve
Hoyiik, Kodapa, Kitanaura, and Hyllarima (Turkey), Bordeaux, Perigueux, and Brion-Saint-Germain d'Esteuil
(France). Notable publications include "Porsuk - Zeyve Hoyiik: Rapport preliminaire des campagnes 2018 et 2019,"
with C. Barat, E. Koker Goki:;e, V. Mathe, G. Bruniaux, & R. StoraY (Anatolia Antigua 28, 2020); "Prospections
geophysiques multi-methodes du site gallo-romain de Brion (Gironde, France): une agglomeration secondaire en zone
humide," with V. Mathe, A. Camus, M. Martineau, D. Barraud, & F. Tassaux (ArcheoSciences, Revue d'archeometrie
36, 2012).
Jeroen Poblome, Professor of Archaeology, KU Leuven; Director, ICRATES Project, the Sagalassos Archaeological
Research Project, and the KU Leuven Institute for Cultural Heritage. Recent publications include: "Meanwhile in the
Mountains. Sagalassos," with E. Torun, P. Talloen, and M. Waelkens, eds. (Istanbul, 2019); "Demeter in support of
agriculture, and as indicator of systemic economic processes?" with P. Talloen & R. Vandam, in Detur Dignissimo.
Studies in Honour of Johan van Heesch (Travaux du Cercle d'Etudes Numismatiques 21, 2020); "Intramuros:
Investigating relations between cross-industry practices and networks through sixth century A.O. Sagalassos," with E.
Murphy (Journal of Urban Archaeology, 3, 2021); "A framework of thought for Pisidian economies," in Pisidian
Studies II. Production, trade and economy in Pisidia (Open Access, 2020).
Annalisa Polosa, Associate Professor of Ancient Numismatics, Sapienza University of Rome. Recent publications
include "Coin production and coin circulation in Cilicia Tracheia," in Identity and cultural exchange in Ancient
Cilicia. N8-V results and future perspectives, A. Cortese, ed. (Mitteilungen zur Spiitantiken Archiiologie und
Byzantinischen Kunstgeschichte 7, 2020); "Elaiussa Sebaste: note di topografia," with M. Braini, in Studies in honour
ofEugenia Equini Schneider, A. Polosa, A. KlZllarslanoglu, & M. Oral, eds. (Istanbul 2019).
Lucas Proctor, Postdoctoral Fellow in the Institute for Archaeological Science at J.W. Goethe University; he works
on the BMBF-funded "UmWeltWandel" Joint Research Project in Central Oman. As a specialist in environmental
archaeology and archaeobotany, his research explores the social and environmental dynamics of early fuel and
agropastoral economies. He has been a member of the Kerkenes Project since 2011, and works as an archaeobotanist
on several collaborative research projects in Turkey, Iraqi Kurdistan, Oman, and Azerbaijan.
Jane Rempel is a Lecturer in Classical Archaeology, University of Sheffield. She specialises in the archaeology of the
Greeks in the Black Sea region, having conducted fieldwork in Crimea, Armenia, and currently Sinop Kalesi, Turkey.
Notable publications include Living Through the Dead: Burial and commemoration in the Classical world (co-edited
with M. Carroll, Oxbow 2011); and with 0. Doonan, "Rural hinterlands in the Black Sea during the fourth century
BCE: Expansion, intensification and new connections (Anatolian Studies 70, 2020).
Jessica Robkin is a Ph.D. candidate at the University of Central Florida in the depai1ment of Anthropology. Her
research combines traditional archaeological methodology with cutting edge technology and modeling and simulation
to further understand social organization and heritage preservation. She has worked on the Kerkenes Project since
2019.
Nicole Rose received her Ph.D. from The Graduate Center, City University of New York in 2021 for her dissertation,
"Mobile Practices and the Production and Curation of Pottery: A View from the Ancient Southern Russian Steppe
Using Portable Methods of Investigation." Her research focuses on the use of ceramic technology and composition to
study mobility and the development of communities of practice.
Susanne Rutishauser, postdoctoral researcher in archaeology at the University of Bern. Notable publications include
Siedlungskammer Kilikien: Studien zur Kultur und Landschaftsgeschichte des Ebenen Kilikien (Harrassowitz, 2020).
Mona Saba received her B.A. and M.A. in Iran (Tahran) and her Ph.D. in Protohistory and Near Eastern Archaeology
from Istanbul University. She began working as a research assistant in the Faculty of Letters at Istanbul University in
the Department of Protohistory and Near Eastern Archaeology in 2017; before that she was involved in excavations in
northwestern and central Iran. In 2018 she was appointed as Lecturer in the Department of Fine Arts at Istanbul
University. She is a member of the Oluz Hoylik excavation team. Her research mainly focuses on the Iron Age
cultures of Central Anatolia, Zoroastrianism, and the Achaemenid period in Anatolia.
Grazia Semeraro, Professor of Classical Archaeology at the University of Salento. Notable publications include:
"Archaeology of the cult in the sanctuary of Apollo in Hierapolis," in Fra ii Meandro e ii Lico. Archeologia e storia in
un paesaggio anatolico, F. Guizzi, ed. (Scienze dell'Antichita 20.2, 2014); "Archeologia dei fenomeni di interazione e
scambio in un contesto pluristratificato: il caso di Hierapolis di Frigia - <;ok katmanh bir kontekstte etkile~im ve
degi~im olgusunun arkeolojisi: Phrygia Hierapolisi ornegi" (Arkeoloji ve Sanat 160, 2019).
E. Susan Sherratt, Reader Emerita in Mediterranean Archaeology, University of Sheffield. Recent publications
include Archaeology and Homeric Epic, S. Sherratt and J. Bennet, eds. (Oxbow, 2017); "From 'institutional' to
'private': traders, routes and commerce from the Late Bronze Age to the Iron Age," in Dynamics ofProduction in the
Ancient Near East 1300-500 BC, J.C. Moreno Garcia, ed. (Oxbow, 2016); "A globalizing Bronze and Iron Age
Mediterranean," in The Routledge Handbook of Archaeology and Globalization, T. Hodos, ed. (Routledge, 2017).
Alexander E. Sollee, Research Associate to the Memories for Life Project, University of Cambridge. Notable
publications include: Bergesgleich baute ich hoch. Untersuchungen zur Architektur, Funktion und Bedeutung
neuassyrischer Befestigungsanlagen (Schriften zur Vorderasiatischen Archiiologie 17, 2020); "Tell Halaf before
Kapara. Reconsidering the Structure and Development of the Early Iron Age Settlement," in Formation, Organisation
and Development ofIron Age Societies. Proceedings of the Workshop Held at the I 0th ICAANE in Vienna, April 20 I 6.
A Comparative View (Oriental and European Archaeology, 2020); "The Development of the Southeast Lower Town
of Sirkeli Hoylik. A Preliminary Assessment based on the 2013-2019 Campaigns," with H. Monninghoff, E. Kozal,
D. Karakaya, J. Heim, & S. Gur (Altorientalische Forschungen 47, 2020).
Mehmet Somel, Professor, Department of Biological Sciences, Middle East Technical University. Recent publications
include "Million-year-old DNA sheds light on the genomic history of mammoths," with van der Valk, et al. (Nature,
2021); "The demographic development of the first farmers in Anatolia," with K1h111; et al. (Current Biolozy 2016).
Recent related projects involved include "Archaeogenomic Analysis of Genetic and Cultural Interactions in Neolithic
Societies-NEOGENE" (ERC 772390) and "Mapping the Neolithic Expansion in the Mediterranean-NEOMATRIX"
(Widespread 05-2020-952317).
Emine Sokmen, Assistant Professor of Anthropology, Social Sciences University of Ankara; Director of the Orukaya
Dam project (<;orum), Assistant Director of the Sinop Kalesi Excavations, and previously served as Field Director of
the Komana excavations. Her publications examine settlement in Hellenistic and Roman northern Anatolia, most
recently the role of hydraulic infrastructure in the establishment of the legitimacy of Roman administration in the
region.
Sharon R. Steadman, SUNY Distinguished Professor of Anthropology, State University of New York, Cortland; Co-
director of the <;adir Hoylik Archaeological Project. Notable publications include "Pivoting East: <;adir Hoylik,
Transcaucasia, and Complex Connectivity," with B.S. Arbuckle & G. McMahon (Documenta Praehistorica 45, 2018);
"Sanctifying the House: Child Burial in Prehistoric Anatolia," with L.D. Hackley & B. Y1ldmm (Near Eastern
Archaeology 81, 2018); "Not Seeing is Believing: Ritual Practice and Architecture at Chalcolithic <;adrr Hoylik in
Anatolia" with L.D. Hackley & B. Y1ldmm (Religions 12, 2021); Archaeology of Architecture and the Human Use of
Space (Routledge, 2015).
Elif Siirer, Assistant Professor, Department of Modeling and Simulation, Graduate School of Informatics, Middle
East Technical University. Recent publications include "Ancient genomics in Neolithic Central Anatolia and
<;atalhoylik" in Peopling the Landscape of <;atalhoyuk: Reports from the 2009-2017 Seasons, with Yaka, et al.
(Oxbow, 2021); "Variable kinship patterns in Neolithic Anatolia revealed by ancient genomes," with Yaka, et al.
(Current Biology, 2021). Recent related projects involved include "Mapping the Neolithic Expansion in the
Mediterranean-NEOMATRIX" (Widespread 05-2020-952317) and "Archaeogenomic Analysis of Genetic and
Cultural Interactions in Neolithic Societies-NEOGENE" (ERC 772390).
Sonke Szidat, Associate Professor of Chemistry and Head of the Laboratory for the Analysis of Radiocarbon with
AMS (LARA), University of Bern. Notable publications include " 14C analysis and sample preparation at the new Bern
Laboratory for the Analysis of Radiocarbon with AMS (LARA)," with G.A. Salazar, E. Vogel, M. Battaglia, L.
Wacker, H.-A. Synal, and A. Turler (Radiocarbon 56, 2014).
404 Contributors
Alper ~engiil, Istanbul University - Cerrahpasa, Institute of Graduate Studies. He has studied alignment geology,
geological mapping, structural geology and active tectonics (earthquakes, active faults and mass movements). He has
taken part in different projects regarding engineering geology, alignment geology, geoarchaeology and geotechnical
studies. He is especially experienced in field geology and he has conducted numerous academic and research studies
throughout Turkey and African Countries. Research interests include geo-archeology, geological mapping, structural
geology, active tectonics, earthquake geology, active faults and mass movements.
Turan Takaoglu, Professor, <;anakkale Onsekiz Mart University; Director of the Tenedos (Bozcaada) Excavation
Project. Notable Notable project sites include Gulpmar and Co~kuntepe in north-western Anatolia. Notable
publications include Communities in Transition. The Circum-Aegean Area During the 5th and 4th Millennia BC, with
S. Dietz, F. Mavridis, & Z. Tankosic (Oxbow, 2018); Ethnoarchaeolof;ical Investif;ations in Rural Anatolia, volumes
I-IV (Istanbul, 2004-2007); "Melian Obsidian in NW Turkey: Evidence for early Neolithic Trade" with C. Perles and
B. Gratuze (Journal of Field Archaeology 36, 2011); and "The Late Neolithic in the Eastern Aegean: Excavations at
Gulpmar in the Troad" (Hesperia 75, 2006).
Armagan Tan, Research Assistant, Istanbul University. Recent publications include "Establishment of the Urartu
Kingdom." in Urartians: A Civilization in the Eastern Anatolia. The Proceedings of the 1st International Symposium
held at jstanbul in 13-15 October, 2014, A. <;ilingiroglu, K. Koroglu, Z. <;ulha, & G. Oncu, eds. (Istanbul, 2018);
"Urm.1u Kralhg1'nda Tehcir Uygulama!ar1 ve Toplumsal Yap1ya Etkileri," in Eski Yakmdogu 'da Siirgiinler, t.
Albayrak, ed. (Ankara, 2020).
Sevil Baltah Tlrpan, Assistant Professor, Department of Humanities and Social Sciences, Istanbul Technical
University. Trained both in archaeology and cultural anthropology, her recent research explores the topics of politics
of the past, local perceptions of the past, impact of archaeological praxis on local people, decolonizing methods in
archaeology, space/place, and memory. She has been the associate director of the Kerkenes Project since 2015. She
has been conducting traditional and digital ethnographic field work on the local perceptions of the past and
archaeology.
Ali Umut Tiirkcan, Associate Professor (since 2011), Department of Archaeology, Anadolu University. He has
participated in numerous archaeological research projects since 1992, including at Sagalassos (Burdur), Kyme
(Izmir), and the Lower Meander Area (izmir) Survey Project. His main work, between 1993-2010, was as
archaeologist and figurines and stamp seals specialist at <;atalhoyuk. He has conducted archaeological survey research
projects in the Eski~ehir area in northwestern Anatolia, Balkayas1 (Sivrihisar), and at Kanht~ Mound and its environs
in the inonu Valley (2008, 2009, 2012). From 2013-2019 he directed the Kanht~ Mound excavations
(www.kanlitas.com). Since 2020 he has directed the <;atalhoyuk Excavation and Research Project. His numerous
publications can be found here: https://aliumutturkcan.academia.edu/.
Bora Uysal, Associate Professor, Department of Archaeology, Sivas Cumhuriyet University; Director of the Elbistan
Karahoyuk and Hamzatepe Excavations. Notable publications include "Elbistan Karahoyuk'te Yapllan Yeni Kazllar
Hakkmda Gene! Bir Degerlendirme" in From Samosata to Acemhoyiik. Trailing the Ancient Civilizations, Studies
Presented to Honour ofAliye Oztan (izmir, 2017); "Gok9eada Y enibademli Hoyuk 'ten Ele Ge9en Erken Tun<;: <;ag1 'na
Ait Bir Grup Hayvan Figurini" in In Memoriam Ugur Silistreli: Studies on Anatolian Archaeology, and articles in
Belleten and Zeitschriftfur Assyriologie und vorderasiatische Archiiologie.
Ralf Vandam, Part-time senior Postdoctoral Researcher, Research Foundation Flanders, KU Leuven; part-time
Professor of Archaeology, Department of Art Studies and Archaeology, Vrije Universiteit Brussel. Notable
publications include: "Contextualizing Kuru9ay Hoyuk: Assessing the unexplored Late Chalcolithic Landscape near
the beginning of early social complexity in SW Turkey," with B. Music & I. Medaric (Journal of Field Archaeology
44, 2019); Editor of the special issue: On the margins? Thinking through mm.·ginality in the Holocene Mediterrm.1ean
(Journal ofEastern Mediterranean Archaeology and Heritage Studies, 2019).
Hiiseyin Vural, Director of the Sinop Museum since 2013. A specialist in Bronze Age and underwater archaeology,
he trained at Ege University before taking up positions in the <;e~me and Bergama Museums. Under his leadership the
Sinop Museum has undertaken an ambitious program of cultural heritage management and archaeological excavations,
including the Sinop Kalesi Excavations.
Patrick T. Willett, Research Fellow, Sagalassos Project, KU Leuven. Recent publications include: "Modeling
Archaeological Potentials in SW Anatolia: A Tool for Planning Sustainable Futures at Ancient Sagalassos," with
multiple co-authors, in Critical Archaeology in the Digital Age, K. Garstki, ed. (Los Angeles, 2021 ); "The Aftermath
of the 8.2 Event: Cultural and Environmental Effects in the Anatolian Late Neolithic and Early Chalcolithic," with
multiple co-authors, in Climate and Cultural Change in Prehistoric Europe and the Near East, P.F. Biehl & 0.
Nieuwenhuyse, eds. (Albany, 2016); "Continuous Change: Venturing into the Early Cha!colithic at <;atalhoyuk," with
multiple co-authors, in The Archaeology of Anatolia Volume II: Recent Discoveries (2015-2016), (Newcastle upon
Tyne, 2017).
Deniz Ya~in, Research Associate, Institute of Archaeological Sciences, University of Bern. Notable publications
include Sirkeli Hoyiik. Vorbericht der schweizerisch-tiirkischen Ausgrabungen 2006-2015. Ein urbanes Zentrum am
Puruna-Pyramos im Ebenen Kilikien, with M. Novak & E. Koza!, eds. (Leipzig, 2019); "Islamic Period Ceramics
found in the excavations at Adana/Tepebag Hoyi.ik," with N. Ozkul, in The 11 th Congress on the Archaeology of the
Ancient Near East, Munich (Leipzig, 2020); "Excavations of Tepebag Hoyi.ik 2018," with N. Dervi~oglu, in The 41 th
International Symposium ofExcavations, Surveys and Archaeometry (2020).
K. Ashhan Yener, Research Affiliate, Institute for the Study of the Ancient World, New York University, (ISAW);
Professor Emerita of Archaeology, University of Chicago Oriental Institute, Near Eastern Languages and Civilizations
Department; Professor Emerita of Archaeology, Kos; University; Director Emerita of the Tell Atchana, Alalakh and
the Amuq Valley Regional Projects. Recent publications include Goltepe Excavations: Tin Production at an Early
Bronze Age Mining Town in the Central Taurus Mountains, Turkey (Philadelphia, 2021); "Some Thoughts about
Middle Bronze Age Alalakh and Ugarit: Reassessing an Alalakh Wall Painting with Archival Data," in V. Matoian,
ed., Ougarit, 90 ans apres, (Paris, 2021). Research interests include local expressions of style at Alalakh; use of lead
isotope and tin isotope analyses in sourcing metals.
Burcu Y1ld1rnn, Ph.D. candidate, Middle East Technical University, Department of Settlement Archaeology;
Assistant Director of the <;adu Hoyi.ik Archaeological Project. Recent publications include "Not Seeing is Believing:
Ritual Practice and Architecture at Chalcolithic <;ad!f Hoyi.ik in Anatolia," with L.D. Hackley and S.R. Steadman
(Religions 12, 2021); "Stability and Change at <;ad!f Hoyiik in Central Anatolia: A Case of Late Chalcolithic
Globalisation?" with S.R. Steadman, G. McMahon, B. Arbuckle, M. von Baeyer, A. Smith, L.D. Hackley, S. Selover,
& S. Spagni (Anatolian Studies 69, 2019); "Sanctifying the House: Child Burial in Prehistoric Anatolia," with L.D.
Hackley & S.R. Steadman (Near Eastern Archaeology 81.3, 2018).
Niliifer Baturayoglu Yoney, Professor of architectural preservation at Mustafa Kemal University Faculty of
Architecture since 2020. She previously worked at Istanbul Technical University and Abdullah Gu! University. She
studies, teaches, and publishes on architectural preservation history and theory, heritage management, architectural
urban and archaeological survey, documentation and preservation methods, history, characterization and conservation
of building materials and technologies, history and theory of architecture, and preservation of modern and industrial
heritage. She has been involved in the Kerkenes Project since 1993, carrying out various documentation projects,
including the city wall, as well as the documentation, strengthening, and restoration project for the Cappadocia Gate.
She is a member of ICOMOS.
Nejat Yiicel. Ph.D. candidate, Istanbul University; Second Deputy Head of the Ugurlu excavations. Worked as a
professional excavator and photographer on many excavation projects, including ikiztepe, Toptepe, <;atalhoyi.ik, and
Girmeler. Interests include prehistoric salt trade and early colonization of the Aegean islands.
INDEX
3D sacrifice, 175, 198, 208, 216

model, 82, 151 - shaped vessel, 86
photography, 225 Annex(es), 15, 253, 271
scan, 151-152 Antioch, 252, 265
Agora, 71, 210, 211, 212, 249, 273, 334 Aphrodisias, 53
North, 210, 223, 226 Apollo, 210, 212,218, 223
'Arubaini Apollo Kareios, 216
cult, 175, 177, 183-184 Sanctuary of, 210, 211-212, 214, 218, 223, 224,
goddess, 170, 175, 176-177, 179,183 225
Abandonment, 17, 35, 40, 41, 59, 61, 68, 75, 125, 147, Archaeobotany, 339- 351
212, 216, 250, 265, 288, 350 Archaeogenomic, 355, 359, 362, 365
Achaemenidperiod, 70, 238, 241, 313, 316-317 Archaeological
Admixture, 355, 358, 359-360, 362, 363-365 park, 267
Aegean potential, 312-314, 316,317
coast, 1, 45 predictive modeling, 312
island(s), 21, 22, 45, 46, 48 Archaic
Neolithic, 30 Age/period, 160, 167, 313, 324, 329, 333
region, 21, 23, 30, 33, 71, 80, 87, 88, 162, 241, Cypro-, 105, 120
338,340 Late, 160
Sea, 1, 327 Archaeometric analysis, 54, 87, 90, 120- 121, 224
Aeolic capital(s), 324, 327-328, 333-334 Architecture
Aerial photography, 150, 189, 263 arches, 196, 219, 269-270, 271
Agate, 22, 30, 176, 183, 203 domestic, 30, 61, 374
Agriculture, 46, 229, 278, 294, 304, 309, 333, 338- Medieval, 199, 200-202
339, 340,344. 349-351, 355, 362, 364 monumental, 25, 30, 70, 77, 79, 122, 374
Ahurcuk,277-278,282,284 pise, 15, 16, 17, 18, 22, 40-42, 126, 378
Akhisar Plain, 45, 52 sacral/sacred, 70, 71, 246, 374- 378
AkkayaMevkii, 289, 295, 296, 297- 298 Urartian, 170, 186, 197, 202
Aktoprakhk, 33, 359 Vernacular, 139, 256, 332
Alacahoyilk, 236, 373, 388 Arslantepe, 356, 363, 364
Ali~r Hoyilk, 67-68, 126, 130, 147, 160, 236, Asia Minor, 210,218, 226, 238, 256, 339
373, 383,385, 387- 388 Research Centre, 254
Amasya, 126-127, 226, 229, 236,237,239 Assos, 322- 324, 327,328, 333
Ambar A~1kh, 359, 360, 361-362
yay, 4, 18 Augmented reality, 152
Dam, 1,4 Augustan
Hoyi.ik, 5-6, 18 and Julio-Claudian Age, 211, 217
Amorites, 88 period, 267
Amphora(e), 160, 162, 167, 235, 237, 267, 271, 294, Ayanis
317,330,333 Castle, 170, 174, 177-178, 179, 181, 183, 185,
Amuq Valley, 76, 77, 88, 90 184-187, 188, 191
Andiron, 207, 380, 381, 382-383, 385, 388 Fortress, 174,175,170,177, 178, 179, 184
Animal(s) Temple Complex, 171, 175- 176, 186
bones, 10, 13, 17, 23, 25, 26, 37, 38, 39. 54,
60, 88, 115, 161, 167, 187, 234, 240, 350, 381, Balk1, 287, 290, 291, 294, 295, 298
386 -Battal River, 291, 293, 294- 295, 298
domestic, 339, 365 Banquet hall (room), 258, 262, 264
depictions, 64, 327, 374, 382,383, 384, 385 Barcm, 30, 359, 362, 363, 364
figurines, 15, 38, 39, 101, 374, 378, 379-380, Barley, 206, 343
381-382, 383 hulled, 341,344, 349, 350, 351
homs,22,378, 380,382,385,388 naked, 344, 349, 351
management/husbandry, 23, 30, 234, 294, 307, Basilica, 212, 246, 250, 253-254
309,333,362,385,388 complex at Elaiussa Sebaste, 2, 267, 271-273
motifs, 86, 89 Baths, 267, 271-272
reliefs, 373 Roman,212,246, 247-248,249,250, 253, 258,
263-264, 265
sacred, 175
Beads, 9, 12, 15, 16, 25, 37, 42, 83, 104-105, 181, Bilyiik Gilllilcek, 373, 383-384, 385, 388
184, 200--203,204,206,279,386 Bilyilkkaya,64, 125, 126,363, 364,386
bone,376,379 Byzantine
glass, 152, 297 architecture, 157, 159, 269, 293, 295, 305
Goddess of, 176--177 Early, 155,213, 232-234, 240, 250,254
shell, 22, 24, 29 fortifications/kale, 149, 157, 158
stone, 22, 23 Middle, 212, 230, 232, 233- 234, 358
Bedrock,6,33, 35,37, 39, 41,46,50, 103- 104, 116, period, 1, 5, 9, 166, 216, 218, 226,267, 273,
185, 186, 187, 195-198, 231, 233,249, 292, 285,293, 313, 315-319, 333,356, 358
294,295,324,327, 331 settlement (town), 210, 295, 307
Berna, 232, 251-252
Be~ik:-Sivitepe, 52 Canhasan, 373, 375, 376, 378, 379, 382, 386, 388
Beycesultan, 42, 53, 373, 381, 383, 386 Cappadocia, 33, 56, 132, 143, 146, 148
Black Sea, 1, 2, 147, 155, 160, 165, 167, 229, 236, Gate, 146--148, 151-152
237- 238,240,241276--278,282- 284,340, 364 Cattle, 13, 23, 30, 54, 206,378,379, 380, 383,
Bogazki:iy, 2, 64, 67, 70, 118, 125- 126, 236- 237, 385
256-265, 364,378 Caucasus, 203, 355, 356, 357, 359, 360, 362-363,
Bolkar 364- 365
Daglar1, 132, 136 Cedar, 58, 287
Mountains, 300, 307, 309 Cereal, 17, 23, 290-291, 294, 341-342, 343, 344-345,
Boncuklu, 4, ll,30, 300, 359, 360-362 346, 347, 349- 350
Boz Mountain Range, 289-29, 294, 295,297, 298 Ceramics, see "Pottery"
Bronze Ceyhan River, 97, 108, 109
cauldron, 100- 101, 176 Chalcolithic
fibula, 201, 234, 235, 237 Early, 2, 33-36, 39, 40, 43, 292, 297, 339, 347,
Bronze Age 348, 349, 363
Early, 34, 35, 39, 40, 42, 46, 53, 63, 69, 76, Late, 52,232,234, 235, 284, 317, 348,349,
88,96, 126, 160- 161, 165, 167, 198,200--201, 363, 364
206- 207,232,277- 279,281,282,284,294, Middle,15-16, 46, 48, 52-54, 294, 295, 344, 348,
302, 303, 307, 308, 309,317,347-348, 349, 349, 350
350,355,356,363,364,382,388 period, 1, 2, 35, 40, 45, 89, 315, 344, 349, 350,
Late, 56-57, 61, 62, 63, 68, 78, 82, 89, 96, 360, 363, 364, 372-374
97- 98, 100- 101, 102, 106, 116, 121, 139, 298, Chipped stone, 9- 10, 12, 17, 23, 30, 34, 37, 38, 42-43,
303,305- 307,338, 339,344- 346,348,349, 52, 104, 295
350, 351, 363 Church, 212, 232,233, 234, 246,252, 269, 271,
Middle, 1, 5, 75- 76, 96, 101, 102, 295,308, 273, 319
339, 344,347, 350,351,383 Terrace (at Doliche), 250--251, 253
Building Cilicia(n)
apsidal, 78, 87, 88 Gates, 132, 143
buttressed, 11, 12, 17 Plain, 2, 108, 121, 122
cell-planned, 11, 12 Rough, 2, 268
communal, 26- 27, 30 Syro-, 83, 86, 89, 118
grill-plan building, 10, 11, 17 Citadel, 57, 59, 63, 110, 121, 122, 240, 380
oval, 9- 10, 12, 18 atAyanis, 170-171, 175, 177, 181, 184, 185,
pise, see "Architecture" 187, 189
public,57,61,206, 208,250, 374 atTu~pa, 194- 197,205, 206, 207- 208
quadrangular, 9, 10-11, 12, 269 at Gargara, 323, 325-326, 327, 329, 331, 333
Bulcuk, 291-292, 294-295,298 Inner, 109, 114-115
Bullae, 204, 206, 208, 248 Outer, 109, 113-114
of'Arubaini, see"'Arubaini" City
ofAyanis, 177- 181, 184, 187 Fortifications, see "Fortifications"
Burial gate, 109, 115,116, 117, 146,148, 149-151
adult (sub-), 363, 386, 387, 388, 389 lower, 56--57, 109, 110, 206, 256, 258, 264
at Tu~pa Citadel, 195-196 upper, 70, 118, 125
child (infant), 17, 377, 382, 386-387, 388 wall, 79, 83, 109, 110, 122, 148, 151, 155,
cremation, 104, 220 167, 186, 194, 269
gift(s), 203, 387 Climate change/event, 75, 88, 341, 360
hocker, 15, 202, 203, 208 Coins, 7, 15, 161, 200, 231, 232, 234, 235, 237,
intramural, 88, 106, 358, 365, 373, 374 240- 241, 250, 265, 316,332, 333- 334
jar (pot, urn), 83, 102, 386, 387, 388 Colonialism/colonial periods, 155, 160, 165, 166,
pit, 26- 27, 30, 83, 199, 200 167, 287, 298
pithos, 8, 203 Commagene, 244, 267
Button seal, 101, 105 complementary landscapes, 289-290, 298
408 Index
Consolidation, 138-139, 162,271 (casemate) wall, 82-83, 115-116, 134, 137, 162-
Conversion, 52, 140, 341 162- 164, 186- 187, 191
Cooking, city, 77, 78, 88
activity/area, 13, 79, 98, 142, 381 (fort) remains, 297, 305
pots, 13, 15, 54, 83, 85, 86, 87, 118,216, 222, Hellenistic, see "Hellenistic"
223,235 Iron Age, 138, 147, 151
Copper, Foundation
artifacts, 11, 25, 52, 53- 54, 83, 150, 376,388 of city/settlement, 83, 244, 295, 110, 146,
ore (and mines), 291, 308 147, 171, 184, 191
slag, 53, 54 ritual (deposit), 11, 382, 385, 388
Cotton, 347, 349, 350, 351
<;;adrr Hoyiik:, 67, 126, 373, 378-379, 382-383, Gaziantep, 106, 244, 246
385, 386, 387, 388 Gelveri-Yillcsekkilise, 373, 380, 384
<;;amhbel Tarlas1, 363, 373,378, 382, 384,385, Gene pool, 356, 358, 359-361, 363-365
386, 388 G1can Kalesi, 301, 305-306
<;;amoluk, 277, 282, 283, 285 Glacis, 22, 57, 59, 61, 67, 114, 159, 160, 167
<;;atalhoyiik:, 236, 297, 300, 309, 347, 349, 356, Glass, 7-8, 81, 89, 152, 200, 202, 203, 250, 265,
358,359,362,365 280,297
West, 33, 373, 378-379, 384, 385, 386 Gokgoz Tepesi, 301, 302-303
<;;ine-Tepecik, 48, 53 Golcillc, 277-279, 282, 284
Croesus, 147, 210 Gordion, 67, 68, 70, 126, 147, 236, 237, 238, 240,
<;;ukurii:;i, 30, 53 241, 341, 349
GPS, 136, 148, 162
Dagoren,301,306-308,309 Granite, 62, 146, 150
Dark Age, 125-127, 316 boulders, 57-58, 61
Depot Grape, 294,307,341, 347- 348, 349, 350, 351
archaeological, 134-137, 149, 151-152 Greek colonies, 147, 155, 178, 210, 333-334
storage, 41, 300 Ground (stone)
artifact, 9-10, 12, 15, 16, 37, 38, 42, 43, 53
DNA, ancient, 2, 355, 356-357, 358, 365 slab (grinding stone), 10, 12, 13, 15, 17, 26,
mtDNA, 356, 357- 358, 362 28, 29,37, 42, 52,99, 204, 330, 333, 380
Gillpmar, 48, 53-54
Ecology, 288,289,294,298,333 Guvercinkayas1, 373, 380-381, 382,384, 385,
Elbistan Plain, 2, 97, 106 386
Emirgazi, 300, 301, 304, 305,307, 309 Gypsum, 89,133,134,135,137, 141-143
Epipalaeolithic, 338, 343, 344, 347, 348, 349
359- 362,364 Hac1lar, 33,316,373, 374-375, 378, 379,382,
Eregli (Konya), 300, 301, 304, 305, 307, 308, 309 384,385, 388
eroded piedmont, 289, 293, 294-295, 298 Ijaldi, 171, 172, 175, 176, 177, 178, 183- 184, 198,
Eski Belkaya, 301, 304 206
Eskiyapar, 125, 126, 236, 239 E.BARA, 174-175, 176,183
Temple, 170,176, 183
Fault line/plane, 187, 189, 218-220, 295 Ijattufa, 70, 71, 122, 125, 148,256, 258, 260,378
Fig, 341 , 348,349, 350,351 Hamamtepe, 230, 231, 232, 235
Figurine(s), 25, 42, 160- 161, 167,241, 379,386 Hellenistic
animal, 15, 38, 39, 378, 380,381, 382, 383 architecture, 142, 143, 233, 234, 259, 269
human,26, 88, 177, 377,378, 379,380- 381, city, 210, 248, 295
382 fortifications, 155-158, 161, 167
Kilia, 45, 46, 48, 51-53 necropolis, 116- 117, 258, 295
marble, 45, 46, 48, 49, 52, 54, 380, 382 period, 63, 70, 96, 97, 98, 110, 125, 126, 130,
terracotta, 9, 11, 12, 16, 17, 18, 114, 234 143, 155, 157, 166, 210, 214,216, 219, 224,
Fire 229,230-234,240,244, 246,257-258, 277,
destroyed by, 23, 58, 59, 115, 139, 147, 185, 278,279, 292, 295,298, 281,282, 315, 316-
248, 249, 253, 375 317,339,344,347,350,358,363, 364
event, 40, 187, 220, 388 pottery, 68-69, 116, 162, 167, 235, 236-240,
installation/place, 25, 36-37, 39, 40, 41-42, 280,294,284
60, 61, 142, 216,375,376, 377, 378,381 Hieroglyphic, 100, 101, 118, 133, 289, 305, 307
Flint, 12, 17, 23, 52 Hilltop settlements, 33, 34, 35, 40, 109, 284
artifacts, 11, 43, 160, 388 Hittite
knapping, 22, 23, 41 architecture, 58, 61, 70, 71, 134,258,259,
Fortification, 260,263
Byzantine, see "Byzantine" Empire (kingdom, state), 56, 70, 97, 98, 100, 109,
126,147, 148,258, 295, 298
fortification, 139, 258-259, 306 Silva, 198, 202

king, 108, 258, 283, 305 Van, 170, 184, 188, 194
Neo-, 97, 98, 106, 116, 122, 133, 147 Landscape, 108, 146, 167,210, 212, 214,226,
period, 63, 70, 96, 100-101, 106, 137-138, 230, 248, 284, 287-295, 298, 309, 312-313,
229, 256,289,309,363,364 317,340
pottery, 63, 101-102, 126, 133 medium-scale, 289-290, 298
texts, 56, 88, 133 of alterity, 287- 288, 298
Holocene, 30, 293, 298, 313, 315, 317, 341, 349, urban, 70, 110, 122, 212, 226,248
355, 359, 360, 362, 363, 364,365 Laodikeia, 210, 223
Hunter-gatherer, 362, 363, 365, laser scanner, 151-152, 162
Last Glacial Maximum, 22, 359
ikiztepe, 167, 363, 364 Late Antique, 5- 6, 8, 246, 250, 252- 253, 307, 313,
Ilgm Plain, 287, 288- 289, 290- 293, 294, 295, 297, 315- 319
298 Late Antiquity, 218, 244, 250, 252, 269
incised valleys, 289- 290, 291-294, 295, 297, 298 Lentil, 39, 347, 350-351
Intensive survey, see "Survey" Lewis holes, 304--305
Iron Age, 18, 56-57, 76, 96, 111, 121, 147, Liman Tepe, 48, 53- 54
220,235,315,338- 339, 341,343, 344, Limestone, 6, 9- 10, 12, 15, 46, 50, 62-63, 70, 80,
346-347,348, 349, 350, 351,355,360, 363, 89, 116, 136, 187,210, 243, 248-249, 252, 269,
364 271, 290, 292, 294, 317, 324-325, 327, 329,
Early,59,63, 98,99, 139, 205,206- 207, 208 332-333, 375
Late, 57, 59, 68, 98, 110, 199, 200-203, 232, 244, Linear ecologies, 287, 289-290, 294-295, 298
278,295,298,358 Lithic(s), 9, 11, 12, 18, 244, 315, 318,382
Middle, 59, 60, 98, 99, 146, 159, 203, 223, 298 Flint, 11- 12, 17, 22- 23, 41, 43, 52, 160, 388
settlements, 2, 57, 97, 110, 122, 148-149, 277,282, Nemrikpoint, 10- 12
284, 300, 302,309,313,317 Microlith, 11- 12
Islamic burials/cemetery, 199-200, 304 Pressure technique, 23, 30
Ivory, 79, 150-151 Luwian, 97, 107, 118, 305, 307
Lydia, Lydian, Lydianizing, 67- 68, 147, 160, 211
Jupiter Dolichenus, 244, 248 Lykos Valley, 210, 212, 223,
276--277
Kabaktepe,277, 279-280, 282, 284
Kahramanmarfili, 96-97, 106-107 Mahmuthisar, 289- 290, 295, 298
Kaleta~, 277, 282 Marble, 47, 50, 52- 54, 133, 136, 172, 174, 202,
Kalythies, 48 211, 215-218, 224, 247, 263,269,271,
Kaman Kalehoytik, 64, 67, 126 380-383
Kanhta~, 2, 33-35, 40, 42-44, 236, 363-364, 388 bracelet, 34, 37, 41-43
Karacadag, 305- 309 columns, 271
Karain Cave, 52, 365 figurine, 46, 48, 53- 54, 380
Karaoren, 301, 305-307, 309 workshop, see "Workshops"
Karapmar, 300- 309 Mari, 88-89
KelkitBasin (Lykos), 276- 285 Masonry
Kempos, 291-292, 294-295, 298 ashlar, 117, 159, 186, 249, 252, 258,260,
KEYAR, 300, 308-309 263
Kilisetfili, 277, 280-282, 284 cyclopean, 304-305, 323
Kilia figurine, 45-46, 48, 51-53 dry, 29, 138, 233, 264, 323
Kinship, 359, 362, 365, 388 orthogonal, 58, 333
Kiztlrrmak,33, 125-126, 130, 147, 236--237, 239,241, polygonal, 269-270, 323-325, 327, 329, 331,
256, 277, 309 333- 334
Kocadag,322- 325,331,333 Ma~atHoytik,67, 11-12,68, 70, 125, 126, 130
Konya (Plain), 33, 288-290, 298,301,305, 307, Medieval, 1, 15, 155, 199- 201, 208, 223, 232,
309,360 240, 246, 291-292, 294, 296, 338, 344, 347-
Ko~k Hoytik, 33, 373, 378, 380, 384- 386 350
Kumtepe,53-54, 277, 282,363-364 Mediterranean, 17, 70, 71 fn. 23, 79-80, 87-89,
106, 120, 160, 167, 233, 300, 308, 340,
Kurbantepe,277,280,282
343-345, 347-351, 355, 357
Kuruc;:ay Hoytik, 33, 42, 373, 376- 377, 381- 382,
coast, 105, 340
385-386,388
Transitional, 340, 343- 346, 348
Melos, Melian, 23, 50, 52- 54
Lake
Middle Age, 6, 9, 13, 15, 18,246
Bey~ehir, 289, 298, 309
M1hraphkaya, 258, 260-261, 263-265
Burdur,316- 317,372
Military, 70, 116, 170, 186, 256--260, 263- 264,
District, 33, 340, 372, 376
283,285, 290, 305, 309, 350
410 Index
Minua Fountains, 207 Palaeolithic, 34, 244, 312, 317, 318, 355, 360, 362,
Mithradates VI, 229,277, 282 363,365
Mittani, 108 Phiygian, 106, 147, 167
Monumental, 25, 30, 88, 184, 194, 196, 204, 212, Pmarb3-?1, 309, 359, 360, 361, 362
215-216, 249,267,271,325, 329, 374 Plateia, 210, 211, 226
Mosaic, 62, 70-71, 212, 247, 251-253, 263- Platform, 23-24,28, 63, 70, 103,142, 150, 164, 206,
264, 269, 270- 271 252
Motifs (Triangle), 64, 67, 71, 81, 126- 127, 130, packed mud, 36, 37, 41
202, 235 stone, 9, 29, 38, 50, 52, 61, 197, 198
Mudbrick, 10, 13, 34, 36, 41--42, 57-59, 61, 77, terra cotta, 17, 104-105
82,98-100, 114,126, 138-142, 158-160, 173- Pleistocene (Late), 30, 355, 359-360, 364, 365
174, 188- 189, 199- 202, 204, 206-207, 298, Ploutonion, 210, 211, 212, 218, 219, 224- 225
305,377- 379,381,388 Pontus, 229, 234, 235, 241
(super)structure, construction, 15, 98-99, 114, Kingdom of, 143, 276-277
126, 137, 159, 167, 184, 186-187 Mountains, 146, 340
wall(s), 17, 37, 40, 42, 58, 79, 88, 102, 114, 137, Population, 48, 88, 250, 253, 333, 355-356, 357, 358-
186, 188- 189, 200, 202, 206 365, 389
Mukis, 75, 89 Porsuk,33, 43, 133,139, 143
Muwattalli II, 108-110, 117 Pottery
Ali~ar IV style, 60, 64, 67, 125, 126, 130
Necropolis, 116, 202, 208, 210, 212- 214, 220- anthropomorphic, 26, 385
221, 223, 226, 231, 258, 267-269, 271, 277, applique, 101, 206, 383, 384
281,282, 334 Bronze Age
Neo-Hittite, see "Hittite" Early, 42, 126, 160, 198, 200, 201, 278,
Neolithic 279, 281,284, 303,307, 317
buildings, 9, 13, 374 Late, 63, 101, 303, 305- 306, 307
ceramics, 9, 15, 292, 297 Dark Faced Burnished Ware, 16, 42
Pre-Pottery (Aceramic) Neolithic, 6, 9, 12, 22- (Delicate) Banded Ware, 236-237, 238, 240,
23, 30, 89, 339,347,349, 350, 360-361, 241
362,365 "eared pots," 26, 27
Pottery (Ceramic) Neolithic, 5, 6, 13, 16, 17, festoon, 5, 64, 126, 130, 202
22,25, 359, 362, 374, 383 Galatian Ware, 235-236, 239
settlements (sites), 4, 11, 21, 22, 30, 53, 292, geometric motifs, 17, 42, 63, 64, 67, 68, 70-
297, 358, 359, 362 71, 86, 100, 118, 125, 126, 130, 222, 223,
transition, 13, 30, 355, 362 238,247, 251- 252,297, 384
Nephrite, 22-23, 29, 30 glazed, 9, 15, 157, 200, 201, 215, 231-232,
Nodalar Hoyi.ik, 290, 292, 294, 295 234, 293,333
Nymphaeum,212, 215, 216-220, 224,225-226 Gray Burnished Ware, 83, 86
handmade,5,63-64, 101, 118, 126,160, 165,
Obelisk, 25, 30 167,284
Obsidian, 5, 11, 12, 16, 17, 23, 30, 50, 52-53, 54, 206, Hassuna, 9, 17
297, 384, 386 impressed, 16, 17, 101, 118,165, 383
Ochre, 30,42, 81-82,373,374,380, 382,386, 387 Iron Age, 1, 5, 64- 68, 69, 100, 105, 106, 108,
Oil, 224, 342, 343, 347, 349 125-130, 166, 167, 235-237, 238, 240, 281 ,
lamp, 15, 157 284,303- 306, 307,317
olive (seed), 77, 244, 343, 347, 350, 351 Kamares Ware, 79, 87
Olive, 238, 328, 340, 348, 349, 350-351 monochrome,46,67, 68, 100,118,235
Opus (sectile), 271 Red Lustrous Wheelmade, 101, 118
caementicium, 258, 260, 264 Red Slipped (Burnished Ware), 24, 42, 63, 67,
reticulatum, 263, 264 86, 101, 118, 120, 240, 317
Orman Fidanl1g1, 33, 34, 373, 380, 384, 385, 386 relief(decoration), 17, 47, 101- 102, 165, 240,
Ottoman, 9, 34, 155, 156, 162,200, 230, 231-232, 384
240,294,295,300, 312,313, 315- 318,319, polychrome, 67, 68, 81, 235-236, 237, 238,
355, 360, 363, 364 241
Predictive Modelling, 1, 312, 313-314, 315, 317,
Palace, 70, 71, 76, 77, 78, 79, 80, 81, 88, 179, 194, 318, 320
206 Pulse, 341, 343, 346, 347, 350
Level VII, 76-77, 79-80, 83, 87, 89 pXRF analyses, 81-82, 165-166, 167
Painted
plaster, 23, 70, 71, 262, 374, 376, 378. 380 Qaqani, 179, 183
pottery, see "pottery," multiple entries Qatna, 71, 87, 89
wall, 80-81, 174, 213, 378, 385, 386
Rakka Hoyi.ik, 301, 307
Residential geophysical, 61, 109, 110, 115, 116, 137,

area(quarter), 9, 30, 79, 115- 116, 212,213, 329- 147,148, 149, 152, 218, 226, 246, 247,
330, 331- 332, 333, 381 249,250, 258,309
structures, 210, 264, 267, 329, 380 intensive, 34, 39,246,291,315, 318
Restoration and (reconstruction) conservation, 138, Magnetometer (magnetometry), 35, 115-116,
139, 219, 224 136, 149, 269
Ritual Remote sensing, 110, 148, 246, 291
activities (behavior), 9, 30, 70, 88, 217, 218,
220, 224, 252, 372, 373-374,378,382, Taurus Mountains, 5, 132, 139, 300, 307, 309,
388 340, 372
building (structure), 9, 30, 88, 373, 377-378 Temple, 198, 230, 267, 324
material culture, 78, 175, 385 at Gargara, 327, 328, 329, 332, 333- 334
pit, 26, 28, 78 atHierapolis, 215- 217, 219, 223
sacrifice, 78, 175, 216 at Tell Atchana, 76-78, 88
Rock Hittite, 70-71, 125, 258
-cut (spaces), 34, 40, 41--42, 116, 196, 198, susi(Ayanis), 170, 172, 173-175, 177, 183, 184
232, 261,278, 279, 280, 295 Terrace, 41, 57, 61,63, 70,77, 79, 102,143, 156,
relief, 109- 110, 114, 117 160, 198, 210, 212, 214, 216- 217, 226, 230,
tomb, 88, 116-117, 194--196,207,230, 267, 267,268, 269, 279, 331
269, 271, 278, 295 Church, 246, 250-253
Roman North, 97, 98, 99- 102, 106
buildings/structures, 133- 134, 143,210, 212, wall, 79, 142, 143, 157, 323, 324, 329
246,247- 248,257- 258,263,264, 305 Terrazzo, 10, 23- 24
Emprre, 244,248,263-264,265,283,313, Tomb, 83, 88, 116, 199-200,210, 213, 214, 245,
315,316,317, 318 267, 269,271, 281, 387
Hellenistic and/or, 96, 98, 116, 117, 167, 210, chamber, 194- 196
248, 277, 278, 281, 284, 292,294, 344 rock-cut, see "rock"
military, 257, 258- 260 royal, 195- 196, 203, 207
necropolis, 116, 258, 281 Town
period, 18, 53, 63, 68, 96-97, 137-138, 156, lower, 70, 76, 110, 115-116, 117, 121
224, 229, 231,232, 241, 244, 247, 257, 267, upper, 116, 121-122, 382
278-279, 281,282-283, 289,291,292-294, Trade routes (networks), 1, 53, 54, 76, 146,339,
295, 297, 298, 315,318- 319, 339, 347,348, 365
349-350, 358,364
Rough Cilicia, see "Cilicia" Upper Tigris, 4, 5, 11, 16, 18, 30
Royal Precinct at Alalakh, 77-78, 79, 83, 89 Urartu, 147, 175, 179, 181, 183, 184, 185
Rusa(king), 170-171, 174,175,178,181, 184-- Ulucak,23, 30, 33, 53
185, 186, 190
Workshop, 9, 37, 40, 41, 43, 64, 77, 79, 110, 111,
St. Philip, 210, 212 115, 121, 130,142, 204, 232,238,265, 267,
Spolia, 231, 232, 294, 295, 309 305, 380
Spondylus, 25, 26, 29 kitchen, 79, 82, 98
Survey marble, 1, 45--46, 48--49, 51- 52, 54
electrical resistance, 148, 149 mudbrick, 138- 140
extensive, 230, 313, 315
(geo)magnetic, 1, 97, 110, 135- 138

Sharon R. Steadman - John Gregory McMahon - Archaeology of Anatolia, Volume IV - Recent Discoveries (2018 - 2020) (2021)

Uploaded by

Copyright:

Available Formats

Sharon R. Steadman - John Gregory McMahon - Archaeology of Anatolia, Volume IV - Recent Discoveries (2018 - 2020) (2021)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Sharon R. Steadman - John Gregory McMahon - Archaeology of Anatolia, Volume IV - Recent Discoveries (2018 - 2020) (2021)

Uploaded by

Copyright:

Available Formats

The Archaeology of

Recent Discoveries (2018-2020)

Sharon R. Steadman and Gregory McMahon

Edited by Sharon R. Steadman and Gregory McMahon

This book first published 2021

Cambridge Scholars Publishing

Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK

British Library Cataloguing in Publication Data

Copyright@ 2021 by Sharon R. Steadman, Gregory McMahon and contributors

ISBN (10): 1-5275-7601-9

List of Figures ...................................................................................................................................................... viii

List of Tables ........................................................................................................................................................ xiv

Chapter One ............................................................................................................................................................. 1

Chapter Two ............................................................................................................................................................ 4

Chapter Three ........................................................................................................................................................ 21

Chapter Five .......................................................................................................................................................... 45

Chapter Six ............................................................................................................................................................ 56

Chapter Seven ....................................................................................................................................................... 75

Chapter Eight. ........................................................................................................................................................ 96

Chapter Nine ....................................................................................................................................................... 108

Chapter Ten ......................................................................................................................................................... 125

Chapter Eleven .................................................................................................................................................... 132

Chapter Twelve ................................................................................................................................................... 146

Chapter Thirteen .................................................................................................................................................. 155

Chapter Fourteen ................................................................................................................................................. 170

Chapter Fifteen .................................................................................................................................................... 194

Chapter Sixteen ................................................................................................................................................... 210

Chapter Seventeen ............................................................................................................................................... 229

Chapter Eighteen ................................................................................................................................................. 244

Chapter Nineteen ................................................................................................................................................. 256

Chapter Twenty ................................................................................................................................................... 267

Part II: Surveys

Chapter Twenty-One ........................................................................................................................................... 276

Chapter Twenty-Two .......................................................................................................................................... 287

Chapter Twenty-Three ........................................................................................................................................ 300

Chapter Twenty-Four .......................................................................................................................................... 312

Chapter Twenty-Five ........................................................................................................................................... 322

Part III: The State of the Field

Chapter Twenty-Six ............................................................................................................................................ 338

Chapter Twenty-Seven ........................................................................................................................................ 355

Chapter Twenty-Eight ......................................................................................................................................... 372

Contributors to the Volume ................................................................................................................................. 396

Figure 2-1. Location of Ambar Dam.

Figure 6-4. Area D: structures dating to the Middle Iron Age.

Figure 10-12. Potsherd, Middle Iron Age, Oluz Hoyuk.

Figure 15-7. Tuspa Mound during excavation.

Figure 20-1. General plan of Elaiussa.

Figure 25-12. Ruins of a house.

Table 3-1. Stratigraphy and Dates of Ugurlu

INTRODUCTION TO THE ARCHAEOLOGY OF ANATOLIA: VOLUME JV

(Photo courtesy of the <;::adir Hoyuk Project)

AMBAR DAM SALVAGE EXCAVATIONS 2018-2020:

Gre Filla I: Late Antique Graveyard

Gre Filla II-III: Middle Age and Pottery Neolithic Period

Gre Filla IV: Pre-Pottery Neolithic B

Gre Filla V: Pre-Pottery Neolithic A

Kendale Hecala I: Middle Ages

Kendale Hecala I Kendale Hecala II

Kendale Hecala IHA Kendale Hecala IIIB-upper levels