New Directions and Paradigms for the Study of Greek Architecture: Interdisciplinary Dialogues in the Field [1 ed.] 9789004416659, 9789004416635

Citation preview

New Directions and Paradigms for the Study of Greek Architecture

Monumenta Graeca et Romana Editor-in-Chief John M. Fossey FRSC (McGill University & Musée des beaux-arts de Montréal) Senior Editor Troels Myrup Kristensen (Aarhus University) Associate Editor Beaudoin Caron (Université de Montréal) Assistant Editors Laure Sarah Éthier & Marilie Jacob (Université de Montréal) Advisory Board Christina Avronidaki (National Archaeological Museum, Athens) Anna Collar (University of Souhampton) Andreas Konecny (University of Graz) Massimo Osanna (Università degli Studi di Napoli “Federico II”) Duane W. Roller (The Ohio State University)

volume 25

The titles published in this series are listed at brill.com/mgr

New Directions and Paradigms for the Study of Greek Architecture Interdisciplinary Dialogues in the Field Edited by

Philip Sapirstein PhD University of Toronto David Scahill PhD American School of Classical Studies at Athens

LEIDEN | BOSTON

Cover illustration: (front) View of the western colonnade of the Temple of Apollon at Korinthos (orthographic projection generated by means of photogrammetry; P. Sapirstein); (back) Elevation of Temple B at Selinous (orthographic projection combining photogrammetry and digital modeling; M. Limoncelli). Library of Congress Cataloging-in-Publication Data Names: New Approaches and Paradigms in the Study of Greek Architecture  (Conference) (2016 : American School of Classical Studies at Athens),  author. | Sapirstein, Philip, editor. | Scahill, David, editor. Title: New directions and paradigms for the study of Greek architecture :  interdisciplinary dialogues in the field / edited by Philip Sapirstein,  David Scahill. Description: Leiden ; Boston : Brill, [2020] | Series: Monumenta Graeca et  Romana, 0169-8850 ; volume 25 | Includes bibliographical references and  index. Identifiers: LCCN 2019040238 (print) | LCCN 2019040239 (ebook) |  ISBN 9789004416635 (hardback) | ISBN 9789004416659 (ebook) Subjects: LCSH: Architecture, Greek—Congresses. Classification: LCC NA270 .N49 2016 (print) | LCC NA270 (ebook) |  DDC 720.9495—dc23 LC record available at https://lccn.loc.gov/2019040238 LC ebook record available at https://lccn.loc.gov/2019040239

Typeface for the Latin, Greek, and Cyrillic scripts: “Brill”. See and download: brill.com/brill-typeface. ISSN 0169-8850 ISBN 978-90-04-41663-5 (hardback) ISBN 978-90-04-41665-9 (e-book) Copyright 2020 by Koninklijke Brill NV, Leiden, The Netherlands. Koninklijke Brill NV incorporates the imprints Brill, Brill Hes & De Graaf, Brill Nijhoff, Brill Rodopi, Brill Sense, Hotei Publishing, mentis Verlag, Verlag Ferdinand Schöningh and Wilhelm Fink Verlag. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher. Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA. Fees are subject to change. This book is printed on acid-free paper and produced in a sustainable manner.

Contents Preface ix List of Figures and Tables xi Notes on Bibliographic Abbreviations xvii Notes on the Contributors xviii

Introduction: Recent Developments in the Study of Greek Architecture 1 Philip Sapirstein

part 1 Planning, Organization, and Methods of Ancient Greek Architects and Masons 1

The Parthenon’s North Colonnade: Comments on Its Construction 21 Lena Lambrinou

2

New Evidence for the Construction Phases of the Parthenon Peristyle: Anomalies at the Southwest Corner 39 Vasileia Manidaki

3

Ancient Blueprints: New Prospects and Interpretations in Light of Recent Discoveries 56 Jeanne Capelle

4

Three Hellenistic ‘Naïskoi’ in the Theatre Area at Aigeira: Small Buildings in the Context of an Urban Sanctuary 74 Alexandra Tanner

part 2 Life History of Greek Monuments and Sites 5

The Small Limestone Buildings on the Akropolis of Athenai 91 Nancy L. Klein

6

Early Temples Built of Wood and Stone: New Finds from Kalapódhi (Phokis) 106 Nils Hellner

7

Recent Architectural Studies at Goúrnia in East Krete: 2011–2016 123 D. Matthew Buell, John C. McEnroe, Jorge Andreas Botero Besadalombana, and Rafał Bieńkowski

vi

8

Contents

The Temple and Hestiatorion of the Sanctuary on Dhespótiko: Archaeology, Architecture, and Restoration 135 Yannos Kourayos, Kornilia Daifa, Goulielmos Orestidis, Dimitrios Egglezos, Vasilis Papavasileiou, and Eleni-Eva Toumbakari

part 3 Architecture, Cultural History, and Communication 9

Building Change: Domestic Architecture and Identity during the Bronze Age to Iron Age Transition 151 Kyle A. Jazwa

10 Greek Temple Building from an Economic Perspective: Case Studies from the Western Peloponnesos 168 András Patay-Horváth 11 Old Questions and New Approaches: The Significance of Affinities between the Tectonic Arts and the Technical Arts of Ancient Greece 178 Mark Wilson Jones 12 More than War: Symbolic Functions of Greek Fortifications 199 Silke Müth 13 Upcycling as a New Methodological Approach to Reuse in Greek Architecture 215 Sarah A. Rous 14 Looking at the Unfinished: Roughed-Out Ornamentation in Greek Architecture 229 Matthias Grawehr part 4 Simulation, Experience, and Interaction with Greek Architecture 15 Contexts for Greek Architecture: Places and People 243 Mary B. Hollinshead 16 The House of the Rhyta at Pseira: 3D Crowdsourcing in an Online Virtual Environment 258 Miriam G. Clinton with Ansel MacLaughlin

Contents

17 Comparing Greek ‘Bouleuteria’ and Roman ‘Curiae’: Two Case Studies on the Parallels and Differences in the Acoustic Reconstruction and Simulation of Roman Senate Sessions and Greek Boule Meetings 274 Christian Fron, Verena Stappmanns, Xiaoru Zhou, and Philip Leistner 18 New Architectural Work on the Akropolis of Selinous, Sicily: Towards a Digital Platform for Cultural Heritage 289 Clemente Marconi, David Scahill, and Massimo Limoncelli 19 Architectural Documentation and Visual Evocation: Choices, Iterations, and Virtual Representation in the Sanctuary of the Great Gods on Samothrake 305 Bonna D. Wescoat Index 323

vii

Preface David Scahill and Philip Sapirstein This book is the final result of a conference organized by the editors on New Approaches and Paradigms in the Study of Greek Architecture / Νέες προσεγγίσεις και θεωρητικά μοντέλα στη μελέτη της Ελληνικής αρχιτεκτονικής held at Cotsen Hall in the American School of Classical Studies in Athens (ASCSA), which brought together leading figures and emerging scholars in the field to illustrate, discuss, and debate new research and methods for the study of Greek architecture. More than 40 speakers originating from 11 countries came together in Athens for three days of stimulating discussions during November 3–5, 2016. First, we wish to warmly thank the American School for hosting the conference from which this book derives and its former director, James C. Wright, for the invitation to hold the event at the School and his enthusiastic participation throughout. We also thank the ASCSA, the Hixson-Lied College of Fine and Performing Arts, the Office of Research & Development at the University of Nebraska-Lincoln, and the 1984 Foundation for financial support without which the conference would not have been possible. We also express our deepest gratitude to the keynote speakers, Mark Wilson Jones, Manolis Korres, and Bonna Wescoat, each of whom has set high standards for the field through their impressive publications, for stimulating presentations of their recent research at the event. We also extend our thanks to the session chairs, Georg Herdt, Jari Pakkanen, Lena Lambrinou, and Chrysanthos Kanellopoulos, for provoking conversation with their commentary and questions posed to the presenters. Many others assisted us along the way; Niamh Michalopoulou and Konstantinos Tzortzinis deserve special mention for their tireless devotion during the preparations and management of the venue and its technology. While there have been past conferences held at the ASCSA on particular aspects of architecture—notably those concerning architectural terracottas and sculpture—the conference in 2016 was in fact the first since the foundation of the School 135 years beforehand that was dedicated solely to Greek architecture as a field of study. The ASCSA has a long history of fostering architects and architectural historians dedicated specifically to the study of Greek material, many of whom went on to make significant contributions in the field. A prime example of this relationship is Gorham P. Stevens, the first Fellow in Architecture at the ASCSA from 1903–1905, the director of the American Academy in Rome from 1917–1932, and director of the ASCSA from 1939–1947. It goes without saying that architectural historians have been and continue to be prominent among the community of scholars working in the field of Greek archaeology. Those of us who work on Greek architecture straddle many fields, including architecture, archaeology, and ancient history. It is always a challenge to bring the picture into focus. Architectural historians have a unique relationship to the growth of their field. As we strive to move forward and break ground as a scientific discipline by integrating new approaches, methodologies, and paradigms into our research, we must acknowledge the accomplishments and learn from previous scholars and methods. It is not by any means clear that we have surpassed scholars like Penrose in terms of accuracy, or the representations of ancient buildings created by members of the École des Beaux-Arts as exemplars of artistic presentation, even as we have entered eagerly into the digital world of architectural recording and illustration. Even in the digital era, the combination of accuracy with artistic ability and vision remains crucial to all Greek architectural historians. While many new digital methodologies are featured in the ensuing chapters, they typically combine older methods of

x representation created by hand which we might be tempted to describe as “traditional,” yet which continue to be central in contemporary research. Digital and especially 3D technologies also offer new possibilities for comprehensive documentation that were simply inconceivable in the past. What is clear is that we are in a new age of computerized documentation where it has become possible to process massive quantities of material evidence both from past excavations and new and ongoing projects throughout the Mediterranean. The challenge is to harness the new technologies in ways that will allow us to document, interpret, and present the architectural record with not only precision, but also clarity and artistry. While we understand the built environment by means of two and three-dimensional representations, we may now consider the fourth dimension of time—the movement through that recreated environment—and the spatial realities of the built environment in new and exciting ways. Since we see and experience the world in three dimensions, we are compelled to think about not just the static presence of a building, but also its broader life history, from inception and design to construction, use, reuse, reception, destruction, and—in many cases—reconstruction(s). What we witness in the following chapters emerging from the ASCSA conference is how exciting new work is taking place with the exploration and testing of new methods and paradigms. These developments are forcing us to re-evaluate our relationship with, our perspective to, and understanding of the built environment and, particularly, the legacy of ancient Greek architecture and its interconnections with other fields in the humanities and social sciences. Many outstanding questions remain, such as to what extent can and should we reconstruct ancient buildings, either virtually or physically? How do 3D representations and virtual realities change the ways in which we document and interpret buildings and sites? The ensuing papers address a wide range of such questions and raise many others of social and historical significance. It is our hope that this volume encourages discussion and debate on these exciting topics.

Preface

Figures and Tables Figures 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 1.10 1.11 1.12 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.10 2.11 2.12 2.13 2.14 2.15 3.1 3.2 3.3

North pteron and colonnade (L. Lambrinou) 21 Upper surface of capital NC14130 (L. Lambrinou) 22 Stylobate below the (left) eighth and (right) tenth columns (L. Lambrinou) 23 Stylobate under the fifth column (L. Lambrinou after illustration by K. Matala, 2004) 23 Section through the N pteron (above: L. Lambrinou), and plan (below: after Korres 1994a, fig. 38) 25 Bottom drums of the (left) sixth and (right) ninth columns (left: L. Lambrinou; right: A. Papandropoulos, ESMA archives) 27 Schematic diagram of a bottom drum on the north colonnade (L. Lambrinou) 28 Bottom of an unfinished base drum from the Old Parthenon (L. Lambrinou) 28 Measurements of column 11.9/14149 (by L. Lambrinou after drawings by K. Matala, A. Papandropoulos, ESMA Archives) 29 North elevation of the north colonnade (drawing by K. Matala, A. Papandropoulos, T. Skari, 2005; ESMA Archives) 30 Entasis on the north colonnade shafts (L. Lambrinou) 33 The effect of the inclination of the column on entasis (L. Lambrinou) 34 Parthenon southwest corner (Photo S. Gesafidis, YSMA archive) 39 Axonometric section of the southwest corner (V. Manidaki) 40 South frieze backer (Photo T. Souvlakis, YSMA archive) 41 Placement of the south frieze backer (V. Manidaki) 41 Frieze backers at the east end (a,b: 1991, YSMA archive; c,d: V. Manidaki) 42 Frieze backers at the southwest corner (Photo T. Souvlakis, YSMA archive) 44 Construction break through the whole height of the entablature (V. Manidaki) 45 Doric frieze of the southwest corner (Photogrammetric documentation, YSMA archive) 45 Southwest corner, south epistyle (V. Manidaki) 46 Last raking cornice block on the west pediment (Photogrammetric documentation, YSMA archive) 47 Southwest corner geison (Photogrammetric documentation, YSMA archive) 48 Original sima compared to the actual sima (V. Manidaki) 49 Southwest corner sima and cyma reversa moulding (Photo T. Souvlakis, YSMA archive) 50 Original tiling scheme (left) as planned, and (right) as executed (V. Manidaki) 51 Modification for the fitting of the southwest akroterion base (V. Manidaki) 52 Ephesos, Upper agora, basilike stoa (J. Capelle) 57 Chronological map of blueprints (J. Capelle) 58 The “sketch of Pytheos” at the Athena Temple of Priene (courtesy W. Koenigs) 59

xii 3.4

Blueprints from the “Mason’s room” at the Castle of Blois, France, incised in the mortar or drawn on it with charcoal or red chalk during the restoration of the 1880s–1920s (J. Capelle) 60 3.5 Known blueprints quantified according to building type (J. Capelle) 61 3.6a–c Above left: Miletos, theatre (J. Capelle). Below left: Priene, theatre (after von Gerkan 1921: pl. 9). Right: Miletos, Bouleuterion, location of blueprints (after Knackfuß 1908: fig. 53) 62 3.7 (a) RTI with I. Boyer at Miletos (courtesy A. Vacek); (b) Specular enhancement from RTI-Viewer (J. Capelle) 63 3.8 Miletos, theatre: (a) blueprint of a pediment (J. Capelle); (b) blueprint of the arches of the western retaining wall (adapted from Krauss 1973: pl. 18) 64 3.9a–e Above: Miletos, theatre, blueprint; middle left: ending pedestal of the eastern retaining wall; middle right: original roughing out surfaces; below left: proportions used in the drawing; below right: Euklidian construction (J. Capelle) 66 3.10a–d Above: Miletos, Bouleuterion, blueprint; middle left: similar profiles at Miletos; middle right: Priene, theatre; below: Miletos, Bouleuterion (J. Capelle) 68 3.11a–d Above: Priene, theatre, blueprint (J. Capelle); middle: Angista, chamber tomb, blueprint (adapted from Hoepfner 1984); below left: Delos, Hypostyle Hall (J. Capelle); below right: Agora of the Italians (J. Capelle) 69 3.12a–c Larissa, theatre: above: proskenion façade; middle: blueprint of same; below: setting lines on the stylobate (J. Capelle) 70 4.1 Plan of the theatre area (S. Gogos, T. Hagn, A. Tanner, H. Birk; courtesy ÖAW/ÖAI Athen) 75 4.2a–c Ground plans of Buildings D, E, and F (A. Tanner; courtesy ÖAW/ÖAI Athen) 76 4.3 Elevations of Buildings (a) E, (b) D, (c) D; and (d) cross sections (A. Tanner; courtesy ÖAW/ÖAI Athen) 77 4.4 South wall socle of Buildings (a) E and (b) D (A. Tanner; courtesy ÖAW/ÖAI Athen) 78 4.5 Plans of Buildings D and E (A. Tanner) 80 4.6a–b Reconstruction scenarios for Buildings D and E (A. Tanner) 82 5.1 Acr. Inv. 4510, 4503 (N.L. Klein) 93 5.2 Acr. Inv. 4503 (N.L. Klein) 94 5.3 Acr. Inv. 4409 (N.L. Klein) 95 5.4 Acr. Inv. 4503 (N.L. Klein) 95 5.5 Acr. Inv. 7390 (N.L. Klein) 96 5.6 Top: Reconstruction of Building A (Wiegand 1904: pl. XIII); bottom: same (illustration by D.G. Mullen) 96 5.7 Building A: (a) pediment corner; (b) Acr. Inv. 4415; (c) Acr. Inv. 4464 (illustration by D.G. Mullen; photos by N.L. Klein) 97 5.8 Acr. Inv. 4510 (N.L. Klein) 98 5.9 Pinakotheke foundations (N.L. Klein) 99 5.10 Acr. Inv. 9 (N.L. Klein) 102 5.11 Typology of geison blocks: Acr. Inv. 4446, 4433, 4441 (N.L. Klein) 103 6.1 Archaic South Temple at Kalapódhi (N. Hellner) 107 6.2 Northwestern corner of the Archaic South Temple (N. Hellner; courtesy DAI) 109 6.3 Western front of the Archaic South Temple (N. Hellner; courtesy DAI) 110 6.4 Detail of the François vase (N. Hellner, after Østby 2000, 254 fig. 9) 112

Figures and Tables

Figures and Tables

6.5 6.6 6.7 6.8 6.9 6.10 7.1 7.2 7.3 7.4 7.5 7.6 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 8.10 9.1 9.2 9.3 9.4 9.5 9.6 9.7 10.1 11.1 11.2 11.3 11.4

Wooden brackets (N. Hellner; left after Kawerau) 113 Corner solution of brackets (left) and a “bracket-capital” (right) (N. Hellner) 113 Reconstruction with “bracket-capitals” (N. Hellner) 114 Reconstruction with stone capitals, without triglyphs (N. Hellner) 115 Three-peaked antefixes (N. Hellner; courtesy DAI) 116 Reconstruction with triglyphs and tympanon sima (N. Hellner) 117 Plans of Goúrnia (a: Hawes 1908; b: D.M. Buell, J.C. McEnroe, J.A. Botero, and R. Bieńkowski) 123 House Ac (D.M. Buell, J.C. McEnroe, J.A. Botero, and R. Bieńkowski) 125 Masonry fabrics (photographs Janet Spiller) 127 Distribution of masonry fabrics (D.M. Buell, J.C. McEnroe, J.A. Botero, and R. Bieńkowski) 128 Photogrammetric images of Goúrnia (D.M. Buell, J.C. McEnroe, J.A. Botero, and R. Bieńkowski) 130 Town and landscape at Goúrnia (image by D.M. Buell, J.C. McEnroe, J.A. Botero, and R. Bieńkowski) 131 Dhespótiko, temenos (plan by G. Orestidis, photo byY. Kourayos) 135 Building A before restoration (Y. Kourayos) 136 Photogrammetric models (G. Orestidis) 137 East restored façade of temple and Hestiatorion (G. Orestidis) 138 Restored plan (G. Orestidis) 139 Section, north elevation in (below) original state / (above) restoration (G. Orestidis) 140 Section, south elevation: (below) original state / (above) restoration (G. Orestidis) 141 Numerical model for freestanding colonnades (D. Egglezos, V. Papavasileiou, and E.-E. Toumbakari) 145 Numerical model with intermediate wall (D. Egglezos, V. Papavasileiou, and E.-E. Toumbakari) 145 Reinforcements for NK1Σ1 and NE1 (D. Egglezos, V. Papavasileiou, and E.-E. Toumbakari) 146 Statistical correspondence analysis (K. Jazwa) 155 Distributions of (a) average stone width and (b) proportional length-to-width (K. Jazwa) 156 PG, UPGMA dendrogram (K. Jazwa) 157 Sites in the analysis (K. Jazwa) 159 Substantial changes to BQ values (K. Jazwa) 160 Correspondence analysis of PG (K. Jazwa) 163 Correspondence analysis of PG structures (K. Jazwa) 163 Map of Doric peripteral temples in the Western Peloponnesos (A. Patay-Horváth) 169 Kalathoi (a) from Eleusis (DAI Athen, neg. Elefsis 338); (b) Newcastle (UK) 182 Korinthian capitals at (a) Bassai (Bauer 1973, …); (b) Epidauros (Bryn Mawr College Library Lantern Slide Collection); (c) Korinthos (Scahill 2009); (d) Round Temple by the Tiber (M.W. Jones) 183 Korinthian capitals in Rome: (a) Round Temple by the Tiber; (b) temple of Vespasianus; (c) temple of Hadrianus (Wilson Jones 2000, fig. 7.20) 184 Archaic Doric capitals: (a) Kerkyra; (b) Tiryns; (c) Kalapódhi; (d) Syrakousai, Apollonion 186

xiii

xiv 11.5

Doricising capitals: (a) at Metropolis (Manolis Korres); (b) Paestum (Mertens 1993, Abb. 56b); (c) Delphoi (École Française, Athens, neg. 22.351); (d) Athenian Akropolis (DAI, Hege 1554) 187 11.6 Phialai: (a) from Ithake (BM inv. GR 1920.5–29.2); (b) from Karmin-Blur (Vorderasiatisches Museum, Berlin, inv. 796); (c) from the reign of Artaxerxes 1 (BM inv. 1994,0127.1) 188 11.7 Two major branches of Ionic capital: (a) Kykladic type; (b) Attic development of the Ionian type (Georg Herdt) 189 11.8 Hypothetical cruciform solution for the Kykladic type of Ionic capital (Georg Herdt) 189 11.9 Friezes that may have contributed to the formation of the Doric frieze: (a) Egyptian cornice; (b) Mykenai; (c) Tiryns; (d–e) Zagora; (f) Thasos; (g) Samos (Georg Herdt) 191 11.10 Miniature gold offerings found at Ephesos (Seipel 2008, figs. on pp. 148, 148, 149 and 146; courtesy of Ulrike Muss) 193 12.1 Messene, west wall (Jürgen Giese) 201 12.2 Eleusis, Lykourgan (S. Müth) 201 12.3 Herakleia upon Latmos, towers (S. Müth) 203 12.4 Thasos, gate of Zeus and Hera (T. Kozelj 2003; after Grandjean 2011, fig. 423; courtesy of École Française d’Athènes) 204 12.5 Messene, Arkadian Gate (S. Müth) 205 12.6 City wall of Stageira (S. Müth) 206 12.7 Amphipolis, north wall (S. Müth) 208 12.8 Fort Euryalos at Syrakousai (H.-J. Beste, after Beste & Mertens 2015, fig. 219; courtesy Reichert Verlag) 209 12.9 Consoles in fortifications at (a–b) Messene (S. Müth), (c) Eleutherai (S. Müth), (d) Siphai (Ute Schwertheim) 210 13.1 North Akropolis Wall, details (S.A. Rous) 217 13.2 Post-Herulian Wall near the Library of Pantainos (S.A. Rous) 220 13.3 Post-Herulian Wall along the Panathenaic Way (S.A. Rous) 221 13.4 North Akropolis Wall from the Agora (S.A. Rous) 222 13.5 Replica column drums at the north end of the Tomb of the Unknown Soldier in front of the Greek Parliament building, Syntagma Square (S.A. Rous) 224 14.1 Miletos, Bouleuterion (Knackfuß 1908, pl. 14) 231 14.2 Miletos, Bouleuterion, architrave (M. Grawehr) 232 14.3 Nea Paphos, Tomb 4 (© Chris Combe, CC BY 2.0) 233 14.4 Miletos, South Agora (Knackfuß 1924, fig. 40) 234 14.5 Priene, Agora (bpk/Antikensammlung, SMB) 235 14.6 Priene, South Stoa drums (M. Grawehr) 236 15.1 Korinthos, Sanctuary of Demeter and Kore (a: Bookidis and Stroud 1997, Corinth 18.3 plan 4, D. Peck; b: Bookidis and Stroud 1997, Corinth 18.3 pl. 2; both courtesy of the American School of Classical Studies at Athens, Corinth Excavations) 247 15.2 Sanctuary of Demeter and Kore (M.B. Hollinshead and R. Hutt) 248 15.3 Postures on steps according to dimensions and proportions (Hollinshead 2015, 22) 251 15.4 Labraunda, Sanctuary of Zeus (a: J. Blid, courtesy of the Labraunda Excavations; b: Hollinshead 2015, pl. 24b) 252 15.5 Plans: (a) Alipheira (Hollinshead 2015, pl. 3a); (b) Argive Heraion (Pfaff 2003, Argive Heraion 1, plan 1; courtesy of the Trustees of the American School of Classical Studies at Athens); (c) Korinthos (Williams and Fisher 1971, Hesperia

Figures and Tables

Figures and Tables

40, fig. 5; courtesy of T. Boyd and C.K. Williams II; American School of Classical Studies, Corinth Excavations) 253 16.1 State plan, Block AF (after Betancourt 2009, 7, ill. 2.1; 15, ill. 2.5) 259 16.2 State model, House of the Rhyta (M.G. Clinton) 261 16.3 3D reconstruction: (a–b) sections of the ground and upper floors; (c) overview from northeast (A. MacLaughlin and M.G. Clinton) 262 16.4 3D reconstruction viewed from the (a) south and (b) northeast (B. Wang, D. Floyd, and M.G. Clinton) 263 16.5 Position tracking data from Unity 3D Analytics: (a) X-Y-Z position tracking data; (b) “Event” tracking data recording transitions between rooms (M.G. Clinton) 268 17.1 Metroön-Bouleuterion complex in the Athenian agora (after Thompson 1937, pl. 6, by J. Travlos. Courtesy ASCSA) 277 17.2 Thompson’s suggested reconstructions of the New Bouleuterion and Metroön (after Thompson 1937, pl. 8, by J. Travlos. Courtesy ASCSA) 278 17.3 Virtual model of the New Bouleuterion after Thompson (V. Stappmanns) 279 17.4 Kuhn’s reconstruction of the New Bouleuterion (V. Stappmanns, modified from Thompson 1937, pl. 8 and Kuhn 1984, pl. XX) 280 17.5 Virtual model of the New Bouleuterion following Kuhn (V. Stappmanns) 280 17.6 Reconstruction of the Curia Iulia (V. Stappmanns, after Fatucci 2009, fig. 1) 281 17.7 Virtual model of the Curia Iulia following Fatucci 2009 (V. Stappmanns) 281 17.8a–b Comparison between the two proposed reconstructions of the New Bouleuterion showing the SPL metric (X. Zhou, acoustic models generated in ODEON) 282 17.9 Acoustic simulation in the New Bouleuterion: (a) trained speaker speaking loudly with people outside talking to each other; (b) the same speaker with silence outside; (c) the untrained “sausage seller” outside speaking loudly towards the debating boule inside; (d) the same speaker with a silent boule (X. Zhou) 284 17.10 Acoustic simulation in the Curia Iulia: (a) speaker P3 facing Listener 2; (b) speaker P2 facing Listener 1; (c) speaker P3 standing up to address his colleagues; (d) P2 standing in the aisle facing Listener 1 (X. Zhou) 285 18.1 Plan of the main urban sanctuary, Selinous (F. Pisciotta, D. Scahill, and M. Limoncelli) 290 18.2a,b Photogrammetry of Temple B (M. Limoncelli) 293 18.3 VR image of the area (C. Marconi, D. Scahill, and M. Limoncelli) 294 18.4 Temple R (M. Limoncelli) 296 18.5 Temple R cutaway (M. Limoncelli) 296 18.6 Temple R photogrammetric model (C. Marconi, D. Scahill, and M. Limoncelli) 297 18.7 South Building (D. Scahill) 298 18.8 Temple R (a) plan (D. Scahill), (b) VR (M. Limoncelli) 299 18.9 Area after construction of Temple B (M. Limoncelli) 300 18.10 Analytical elevations of Temple B (M. Limoncelli) 302 18.11 Size comparison of the area under investigation with Temple C (M. Limoncelli) 303 19.1 Reconstructed views of the Sanctuary of the Great Gods, Samothrake (Left: after Conze, Hauser, and Benndorf 1880, pl. LXXVI; right: courtesy American Excavations Samothrace) 305 19.2 Reconstructed plan of the Sanctuary of the Great Gods in the early first century CE (Courtesy American Excavations Samothrace) 307

xv

xvi

Figures and Tables

19.3

Ionic Porch attached to the west wall of the Dedication of Philippos III and Alexandros IV (courtesy American Excavations Samothrace) 308 19.4 Possible reconstructions of the Ionic Porch (courtesy American Excavations Samothrace) 309 19.5 3D model reconstruction of the entrance to the Sanctuary of the Great Gods (courtesy American Excavations Samothrace) 310 19.6 View up the Sacred Way (courtesy American Excavations Samothrace) 310 19.7 Nike Precinct (courtesy American Excavations Samothrace) 312 19.8 Plaster fragments from the Nike Monument (courtesy American Excavations Samothrace) 313 19.9 Alternate reconstructions for the Nike Monument (courtesy American Excavations Samothrace/I. Burr) 314 19.10 Alternate reconstructions for the Nike Monument (courtesy American Excavations Samothrace) 315 19.11a–c Central ravine in the area of the “Pont Génois” (plan: A. Ward and Y. Poularakis; photogrammetric model: V. Baillet; aerial view: M. Page; courtesy American Excavations Samothrace) 317 19.12 Split screen view of Unity 3D interactive scene (courtesy American Excavations Samothrace/A. Basu) 318 Tables 1.1 1.2 4.1 5.1 9.1 9.2 9.3 9.4 10.1 11.1 11.2 11.3 17.1 17.2

Measurements of the circles incised on top of the stylobate and the lower diameter of the corresponding columns on the north colonnade (from the east, measurements in cm) 24 Underside of the bottom drums: groupings based on technique, and zone dimensions (widths measured along the column radius, in cm) 27 Naïskoi D, E: Dimensions and design units 81 Architectural elements from Building B in the Propylaia and surrounding area 100 A sample of the 180 Behavioural Qualities (BQs) that were examined for each structure and assigned to a specific type of practice (after Jazwa 2016, 72–102) 153 A sample of the absence/presence and categorical data and the normalisation of these values 155 Substantial changes to BQ values from LH IIIB to LH IIIC 161 Substantial changes to BQ values from LH IIIC to PG 162 Basic measurements and cost estimates of the temples in Fig. 1 169 Evidence for and against a link between the Korinthian capital and the kalathos 185 Evidence for and against a link between Doricising capitals and the phiale 190 Evidence for and against a link between triglyphs and tripods 192 SPL and STI results in the Curia Iulia where Speaker P3 addresses Listener 2 (supplemental to Fig. 10a,c) 286 SPL and STI results in the Curia Iulia where Speaker P2 addresses Listener 1 (supplemental to Fig. 10b,d) 286

Notes on Bibliographic Abbreviations Rather than a combined bibliography for the whole volume, a list of references is located at the end of each chapter. Ancient literary sources are abbreviated according to the conventions for the Oxford Classical Dictionary, available online: https://oxfordre.com/classics/page/abbreviation-list/ Abbreviations for secondary literature follow the system for the American Journal of Archaeology: https://www.ajaonline.org/submissions/journals-series

Notes on the Contributors Jorge Andres Botero Besadalombana graduate of the Universitat de Barcelona (2013), specialises in photogrammetry and 3D documentation, cultural heritage, and archeology. He works in Spain, Greece, and Sudan, documenting the progress of ongoing and finished excavations as well as portable objects. Rafał Bieńkowski M.A. (2015), is currently employed at the Systems Research Institute, Polish Academy of Sciences. He is a Ph.D. student at the Institute of Archaeology and Ethnology, Polish Academy of Sciences. He has published articles about computer-aided tools for archaeological research and application of photogrammetry in archaeology.  D. Matthew Buell Ph.D. (2014), University at Buffalo, SUNY, is a faculty member in the Departments of Art History and Classics, Modern Languages, and Linguistics at Concordia University (Montreal). He has published articles concerned with Minoan architecture, urbanism, and settlement patterns. Jeanne Capelle Ph.D. student in archaeology at Lyon 2 University, has also taught and conducted research at Strasbourg University and the École Normale Supérieure (Paris). She has published several articles on ancient theatre and ancient blueprints, including “Les épures du théâtre de Milet: pratiques de chantiers antiques” (BCH, 2017.2: 769–820). Miriam G. Clinton Ph.D. (2013), University of Pennsylvania, is an Assistant Professor in the Department of Art and Art History at Rhodes College. Her archaeological work focuses on architectural circulation as a clue to function and on integrating digital and traditional methods of architectural analysis. Kornilia Daifa M.A., is currently a Ph.D. candidate in the University of Thessaly. She has worked on the Dhespótiko archaeological project since 2001. Her research interests focus on the topography and spatial organisation of the sanctuary and the wider role of cult in the Geometric and archaic Aigaian. She has published articles about Dhespótiko, edited three books, and participated in many conferences. Dimitris Egglezos Ph.D., is a Civil Engineer at the National Technical University of Athens. With more than 30 years in the field, he has conducted numerous studies and has been a technical consultant for many projects. His research interests include soil-structure interaction as well as geotechnical and structural earthquake engineering, with an emphasis on seismic analysis of dry masonry constructions and the seismic behaviour of monuments and other geostructures (retaining walls, dams, etc.). He is a member of the Technical Committee (TC 301) of the International Society for Soil Mechanics and Geotechnical Engineering (ISSMGE), “Preservation of Historic Sites.”

Notes on the Contributors

Christian Fron Dr. (1984), University of Heidelberg, has worked as a scientific assistant at that university. He has published many articles on the Second Sophistic as well as on ancient rhetoric in general. Matthias Grawehr Ph.D., is a Lecturer of Classical Archaeology at the universities of Basel and Zurich in Switzerland. He has excavated in Jordan and Syria and wrote his Habilitationsschrift on Unfinished Architecture in the Hellenistic and Roman Imperial Period (2018). Nils Hellner Ph.D. in Engineering (2002), Technical University of Munich (Germany), worked from 1989 as field architect on excavations in Turkey (Miletos), Greece (Samos, Trapeza Aigiou, Kalapódhi, Kerameikos) and the Sultanate of Oman (Sumhuram, Salut). From 1996–2002 he was an Assistant Professor for History of Architecture at the Technical University of Munich, later a Teaching Fellow at the University of Applied Sciences at Weihenstephan (Germany) and the University of Trieste (Italy), and from 2009–2017 the Scientific Employee for Building Archaeology at the German Archaeological Institute at Athens. He has published monographs and articles on ancient Greek architecture. Mary Hollinshead Professor Emerita of Art History at the University of Rhode Island, received a Ph.D. in Classical and Near Eastern Archaeology and Greek from Bryn Mawr College in 1979. Her work focuses on the relationship between human behaviour and Greek architecture, as in Shaping Ceremony: Monumental Steps and Greek Architecture (Wisconsin, 2016). She has also written about uses of Greek temples, Aigaian wall painting, and processes of making Roman sculpture. Kyle A. Jazwa Ph.D. (2016), Florida State University, is an Instructor of Greek Archaeology and Classical Studies at Duke University. He has published on Bronze and Early Iron architecture in the eastern Mediterranean and the application of settlement ecology models to Greece. Nancy L. Klein Ph.D. (1991), Bryn Mawr College, is an Associate Professor of Architectural History in the Department of Architecture at Texas A&M University. Her research and publications explore the relationship of architecture and society in Late Bronze Age/Early Iron Age Krete and the architecture of the Athenian Akropolis in the archaic and early classical periods. Yannos Kourayos Archaeologist responsible for the islands of Paros and Dhespótiko in the Kykladic Ephorate of Antiquities, has directed the excavation and restoration projects on Dhespótiko for the past 20 years. He has published monographs and articles about Dhespótiko and Paros, participated in many conferences, and given lectures in Greece and abroad. Lena Lambrinou Ph.D. (2015), Greek Ministry of Culture and Sports, is an architect-archaeologist for the Akropolis Monuments Restoration Service (YSMA), and a member of the Parthenon restoration team since 2000 responsible for the North Colonnade (2001–2009), while also supervising or contributing to other Parthenon projects (Pronaos, West and South Walls,

xix

xx column fluting). A past team architect for Greek and foreign excavations, she supervised archaeological documentation at new Attiko Metro stations. As an active researcher and scholar she has published extensively, including a contribution to Blackwell’s 2016 A Companion to Greek Architecture. Her professional interests include Hellenistic stoas and the development of the Doric order. Philip Leistner Prof. Dr. (1964), University of Stuttgart, is professor of acoustics and director of the Fraunhofer Institute for Building Physics. He has published many articles on architectural and technical acoustics. Massimo Limoncelli archaeologist, teaches Virtual Archeology at the ISUFI High School of the University of Salento and Digital Restoration at the Academy of Fine Arts of Lecce. He collaborates in excavation and restoration works with Italian (CNR-Ibam, University of Salento, Calabria, Bari, Naples, Venice) and foreign research institutes (Deutsches Archaologisches Institut Rom, University of Zurich, Institute of Fine Arts of New York, University of Texas); he has conducted research in Virtual Archeology in the Mediterranean, in particular in Hierapolis of Frigia (Turkey), Soknopaiou Nesos (Egypt), Leptis Magna (Libya), Nabeul (Tunisia), Rome, Selinous and Metaponto. Ansel MacLaughlin is a Ph.D. student in Northeastern University’s College of Computer and Information Science in the natural language processing programme. MacLaughlin is a graduate of Rhodes College, where he received a B.S. in Computer Science and a B.A. in Greek and Roman Studies. Vasileia Manidaki has been an Architect at the Akropolis Restoration Service since 2001, participating in the study, programming, documentation, and supervision of the works for the restoration of the Parthenon, Erektheion, Pandroseion, Arrhephorion, and Akropolis Walls. She has published articles on the design of archaic and classical Greek architecture, the spatial arrangement of the sanctuaries, the structure and architecture of the Parthenon, the Parthenon frieze, the restoration of the Akropolis monuments, and the management of the architectural sculptures. Clemente Marconi is the James R. McCredie Professor of Greek Art and Archaeology and University Professor at the Institute of Fine Arts–NYU and Professor of Classical Archaeology at the Università degli Studi di Milano. He is the Director of the IFA–NYU and UniMi excavations on the akropolis of Selinous. Among his publications, he is the author of Temple Decoration and Cultural Identity in the Archaic Greek World (Cambridge University Press, 2007) and the editor of The Oxford Handbook of Greek and Roman Art and Architecture (Oxford University Press, 2014). John McEnroe Ph.D. is the John and Anne Fischer Professor in Fine Arts at Hamilton College (Clinton, New York). He is the author of The Architecture of Minoan Crete (University of Texas Press, 2010) and other studies.

Notes on the Contributors

Notes on the Contributors

Silke Müth Ph.D. (2005), National Museum of Denmark, is a Senior Researcher at that museum. Her research on Greek urban planning has focused for more than 15 years on Greek and Roman fortifications. She co-organised the international research network “Fokus Fortifikation” and has published various articles and books on this subject. Currently she is co-directing a field project on Archaic and Classical Sikyon, Greece. Goulielmos Orestidis Architect M.Sc., freelances in the field of restoration and enhancement of ancient monuments and archaeological sites, having previously served as Architect in the Hellenic Ministry of Culture. With extensive experience supervising various projects, he has published and conducted specialised studies on these topics. Vasilis Papavasileiou Ph.D. (2002, K.U. Leuven), is Deputy Director of the Directorate for Restoration of Ancient Monuments in the Hellenic Ministry of Culture & Sports. She has more than 20 years’ experience in the analysis and supervision of projects for the protection of ancient monuments. From 2000–2007, she was the engineer supervising the anastylosis of the opisthodomos and north colonnade entablature of the Parthenon, and her current projects include Eleusis and Aigosthena. She publishes on structural analysis and materials technology applied on cultural heritage. András Patay-Horváth Ph.D. (2003) is an Assistant lecturer of Ancient History at the Eötvös Loránd University (Budapest, Hungary). His publications mainly concern the work of Pausanias, the temples of Olympia, and ancient Greek sculpture. His most recent monograph is The Origins of the Olympic Games (Budapest, 2015). Sarah A. Rous Ph.D. (2016, Harvard University), is a Lecturer in Classics at San Francisco State University. She is the author of Reset in Stone: Memory and Reuse in Ancient Athens (University of Wisconsin Press, 2019) and has also published on American founder John Adams’ relationship to the Classics. Philip Sapirstein Ph.D. (2008), Cornell University, is an Assistant Professor of Art History at the University of Toronto. He has published on Greek architecture, architectural terracottas, ancient artisans, and digital technologies for the study of antiquity. David Scahill Ph.D. (2012), University of Bath, teaches courses on ancient Greek architecture at the National and Kapodistrian University of Athens in the M.A. programme for Near Eastern and Mediterranean Archaeology and Greek archaeology at the Athens Centre. He has published on design and construction principles in Greek architecture and documentation of the built environment. He collaborates in fieldwork at many archaeological projects around the Mediterranean, including as field architect for the IFA–NYU and University of Milan excavations on the akropolis of Selinous.

xxi

xxii Verena Stappmanns Dipl.-Ing (1977) is an independent architectural historian. She has published articles on the Hellenistic Gymnasion of Pergamon, the virtual reconstruction of Pergamon, and burial monuments. Alexandra Tanner is a postdoctoral fellow at the Classical Archaeology Division at the University of Zurich. Her doctoral thesis (2019) for the University of Zurich concerns the ‘Naiskoi’ at Aigeira. She holds a diploma in architecture (ETH Zurich) and a Master’s degree in Heritage Conservation (University of Bamberg) with specialisation in historic building research and the preservation of built heritage, fields in which she has been regularly involved in archaeological research projects in Greece and Italy since 2007. Eleni-Eva Toumbakari Dr. Civil Engineering (Structural), is Head of the Section for Studies of Ancient Monuments in the Directorate for Restoration of Ancient Monuments, Hellenic Ministry of Culture & Sports. Bonna D. Wescoat PhD. (1983), Emory University, is Samuel Candler Dobbs Professor of Art History and Director of Excavations in the Sanctuary of the Great Gods on Samothrake. Her published work on ancient Greek sacred architecture includes The Temple of Athena at Assos (Oxford 2012) and Samothrace, vol. 9, Monuments of the Eastern Hill (American School of Classical Studies at Athens, 2017), as well as the edited volumes Architecture of the Sacred (Cambridge 2012, Wescoat & R. Ousterhout, eds) and Samothracian Connections; Essays in Honor of James R. McCredie (Oxbow, 2010, O. Palagia & B.D. Wescoat, eds). Mark Wilson Jones teaches at the Department of Architecture and Civil Engineering, University of Bath. An architect and architectural historian, his research and publications, including prizewinning books with Yale UP and Cambridge UP, focus on both Greek and Roman antiquity. Xiaoru Zhou Dipl.-Phys (1960), Fraunhofer Institute for Building Physics Stuttgart, is senior scientist for architectural acoustics. He graduated from the physics department of Nanjing University and worked many years in diverse areas of acoustics. He has published many articles on architectural acoustics as well as acoustical reconstruction and simulation for historical buildings.

Notes on the Contributors

Introduction

Recent Developments in the Study of Greek Architecture Philip Sapirstein 1 Introduction This book attempts to take the pulse of current research in Greek architecture. There is a sense of rapid change as researchers are confronted with a stream of information from ongoing excavations alongside an increasingly powerful and often bewildering array of new digital techniques for recording and analysis. The question of why and how we study ancient architecture should be continuously raised, especially in light of such pressures which are likely to keep the field in flux over coming years. Since no editors will ever be fully equipped to digest and assess the work fairly across the whole of the discipline, this book has instead turned for answers to the field at large. The endeavour began with soliciting ideas from scholars engaged with Greek buildings and settlements, broadly understood, expecting that the responses would reveal common themes and interests shared among many researchers. In 2014, a call for examples of new discoveries and innovative methods concerning Greek architecture was circulated publicly, and accepted proposals were presented at an international conference in Athens, New Approaches and Paradigms in the Study of Greek Architecture / Νέες προσεγγίσεις και θεωρητικά μοντέλα στη μελέτη της Ελληνικής αρχιτεκτονικής. We were pleased to receive many responses from researchers involved in Greek architecture proposing what should be considered new approaches and paradigms. As described in the Preface, more than 40 speakers participated in the conference over the course of three long but exhilarating days, on November 3–5, 2016, at Cotsen Hall in the American School of Classical Studies at Athens (ASCSA). Selected presentations have been expanded as chapters in the current volume. The method of polling the field at large has resulted in what we hope to be a representative sample of ideas, curated thematically to highlight several recent developments within the study of Greek architecture. 2

What Do We Mean by Greek Architecture?

First, we must consider the definition of the discipline. Academic institutions have been pushing scholarship

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004416659_002

away from narrow specialization, which calls into question whether the study of ancient Greek architecture in its own right is a meaningful category at all. Architectural history of the ancient Greco-Roman world is situated at the intersection of several fields concerned with classical archaeology, with the result that research can be highly interdisciplinary, cutting broadly across the structures of the modern university, as archaeologists, art and architectural historians, students of visual culture, classicists, ancient historians, practising architects, anthropologists, ethnographers, engineers, and many others have enriched the discourse with diverse intellectual perspectives. To begin, recent publications articulate the continued relevance of ancient Greek architecture as a discrete field of inquiry: notably the review article summarizing the state of the discipline published in 2011 by Barbara Barletta in the American Journal of Archaeology, and the 2016 “Companion” volume on the same subject edited by Margaret M. Miles, with 35 chapters authored by a significant swath of the anglophone community. Both of these works recognise a chronological framework for Greek architecture spanning roughly the whole of the first millennium BCE and ending with the transition to the Roman Empire. The editors of the current volume agree that substantive changes in materials, building techniques, patronage, and architectural theory recognizable by the Hellenistic period and coming to fruition under Roman rule are significant enough that Roman architectural research may legitimately be separated as a distinct albeit highly interrelated discipline (Townsend 2016). The core of our definition for ancient architectural research is its engagement with the material remains from the Mediterranean. While enriched by the literary and historical sources from antiquity, and their reception in subsequent architectural practice and theory up to the present, the discipline is fundamentally archaeological. The lengthy period of time required to acquire a thorough subject knowledge—including immersion in the remains, landscape, and modern cultural environments—means that the Greek material will continue to be a specialization among researchers in antiquity or architecture, with additional geographical sub-divisions within the field imposed by modern national and linguistic borders. It is of course important to think about more than just the

2

Sapirstein

monuments from the Greek world, but after some time engaged in fieldwork one usually finds oneself specialised in one area over others. The general handbook recently edited by Clemente Marconi (2014) on architecture and art serves as a useful comparison to the current book. Combining both Greek and Roman under a broad notion of the “Classical”, the book is particularly aimed at advanced students of art history and philology. The theoretical apparatus emphasises hermeneutics, the analysis of the surviving ancient (primarily Roman) texts concerning art and building, and the reception of Greek and Roman writings and visual culture—a battery of approaches that especially encourage viewing Greece and Rome as a continuum. While rich in interpretive tools, archaeological theory is largely absent, as are the methods of close inspection, reconstruction, and structural analysis of the physical remains of ancient architecture which are central during analysis at the site. By contrast, all of the contributors to the current volume have invested a considerable amount of their time inspecting Greek ruins firsthand during fieldwork, secondhand through immersion in the published literature, or both. Through its inclusion of the prehistoric period, however, this volume understands Greek architecture more broadly than do the overviews by Barletta, Miles, or Marconi. The widened scope arises from more than the increasing popularity of viewing the later Bronze Age, Early Iron Age, and Classical Greece as part of a general cultural continuum. Despite profound changes in the character, scale, and purpose of buildings from the third through first millennia BCE, we may also observe that ancient Greek buildings employed many techniques over the longue durée, some of which are still encountered in modern vernacular architecture. More importantly, prehistorians and Classical archaeologists raise many of the same questions about architecture and address them using similar methodologies—whereas their traditional separation is partially attributable to the formulation of academic structures over the past two centuries (Morris 2000: 6, 37–78). It is hoped that the resulting book stimulates more dialogue among specialists across periods— prehistoric, Greek, Roman, and beyond. 3

Where is the Field Now?

If we are to gauge the current directions in research on Greek architecture, we should first examine the interests shown in recent publications. Barletta reviews the history of the field, which since its origins in the 18th century has prioritised primary recording and detailed description

(Barletta 2011: 611–15). The model for the architectural monograph crystallised over the course of the 20th century, comprising detailed illustrations, reconstructions, and consideration of historical context; focused either on a particular monument, or else a type of building or component—mouldings, bases, friezes, and so on. The Greek archaeological service, foreign schools, long-term excavations, and academic presses have provided institutional support to scholars during the years or decades that might be needed to complete a polished architectural monograph; journals and conferences have fostered, respectively, shorter investigations and discourse among many authors on diverse themes (Barletta 2011: 615–20). In preparation of this introductory chapter, I compiled recent architectural publications in order to gauge current areas of interest. More than 200 significant works of scholarship published since Barletta’s 2011 review up to 2018 are listed in the bibliography, organised by topic as: (1) monumental and domestic architecture from the Bronze Age, (2) sanctuaries and cult buildings, (3) construction and design, (4) public monuments, and (5) urbanism and city planning; subtopics and cross-citations in these lists accommodate works that fit multiple categories. Monographs and conference volumes dominated by architectural themes are the primary focus, although several long articles have also been included. Inevitably some scholarship has been missed here, but the objective is to assess overall frequencies of publication rather than to provide an all-encompassing bibliography. Three major international conferences, two devoted to architecture of the third century BCE, and the other dedicated in honour of Manoles Korres, should be mentioned for their particular importance to Greek architectural studies, those edited by: des Courtils (2015); Zambas et al. (2016); and Caliò & des Courtils (2017). Because they address so many areas of research, they have not been formally categorised in the following notes, although the topics of their chapters do generally bear out the patterns observed in the following pages. Several edited volumes with a regional focus have included a significant number of chapters on architecture (Kissas & Niemeier 2013; Gebhard & Gregory 2015; Miles 2015; Patay-Horváth 2015; and Greco & Nicolucci 2016). As might be expected, the works published since 2011 have not drastically changed from those of previous decades, although several topics have markedly risen in prominence. First, we may observe that the division between prehistoric and Classical architectural history persists due to the classification of buildings by type and period over other possible groupings. Besides the explicitly typological surveys, monographs about one building usually treat it as an example of a type: the Temple of X,

INTRODUCTION: Recent Developments in the Study of Greek Architecture

Palace at Y, Walls of Z, etc. In the Bronze Age, studies continue to focus on monumental palatial architecture, tombs, or engineering works, usually treated separately from non-monumental residences, workshops, and settlements as a whole (see below, Topical bibliography 1, with: Shaw 2009; 2014; McEnroe 2010; Tartaron et al. 2011; Betancourt 2012; Shaw & Shaw 2012; Brysbaert 2014; Dimitriou 2016; Tsipopoulou 2016; Cooper & Fortenberry 2017; Fitzsimons 2017; Devolder 2018; and papers in Bonacasa, Buscemi & La Rosa 2016; Letesson & Knappett 2017; Bailey 2018). The first-millennium growth of sanctuaries radically transformed the nature of buildings and patronage, especially at the expense of palatial and funerary architecture, which are largely absent from the Greek milieu until the latter fourth century BCE (see Topical bibliography 2–3, discussed in more detail below). Hellenistic monarchs nonetheless invested heavily in public works. Agoras and stoas (see Topical bibliography 4: Scahill 2012; Mertens 2012; Emme 2013; Rocco 2013; Sielhorst 2015; and several edited volumes stemming from conferences: Giannikouri 2011; Ampolo 2012; Cavalier, Descat & des Courtils 2012; Chandrasekaran and Kouremenos 2015; Caliò, Caminneci & Liviadotti 2017), baths and water facilities (e.g., Robinson 2011; 2013; Betancourt 2012; Wassenhoven 2012; Lucore & Trümper 2013; Campisi 2015; van Tilburg 2015; Wellbrock 2016; Klingborg 2017), and theatres (among others, see: Gogos & Kampourakis 2011; Csapo et al. 2014; Frederiksen, Gebhard & Sokolicek 2015; Drougou 2017; Isler 2017; Krinzinger & Ruggendorfer 2017; Gybas 2018) represent a substantial fraction of the archaeological remains from this era, and each has also witnessed a significant uptick in publication during recent years (compare to the work in these areas from 1980–2010: Barletta 2011: 625–26). Monumental Hellenistic palaces and tombs have drawn slightly less attention in recent years (for palatial architecture, see the last subsection of Topical bibliography 4: Akamates 2013–2014; Bachmann, Radt & Schwarting 2017; and papers in Caliò & des Courtils 2017; on tombs, see Barletta 2011, 626; von Mangoldt 2012; Sporn 2013; Drougou 2016; Henry & Kelp 2016). Sanctuaries remain the primary division between prehistoric and Classical scholarship. The genesis of the Greek temple and its development from the Geometric through Late Classical era still dominates recent publications of sanctuaries (Barletta 2011: 623). In the recent handbook on Greek architecture edited by Miles (2016), 11 of the 35 chapters feature at least prominent, if not exclusive discussion of the temple; altogether, temples and sanctuaries together represent about half the content of the book (also see Topical bibliography 2: Hansen & Le Roy 2012; Wescoat 2012; Anzalone 2013; Gruben 2014; Prignitz 2014;

3

Wilson Jones 2014; Hellner & Gennatou 2015; Koenigs 2015; Mattern 2015; Theodoropoulou-Polychroniadis 2015; Cahill & Greenwalt Jr. 2016; Paga & Miles 2016; Sapirstein 2016; Barletta, Dinsmoor Jr. & Thompson 2017). Despite calls for more attention to other types of cult architecture—such as altars, tholoi, and treasuries—such studies have yet to surpass the numbers dedicated to the temple (Barletta 2011: 623–25; Schulz 2012; Hayashida, Yoshitake & Ito 2013; Arapogianni 2014; Bommelaer 2015; Nielsen 2014; Prignitz 2014; Hering 2015; Hollinshead 2015; Kienast 2016/2017; Kienast et al. 2017; Voigts 2017; recent work at Olympia includes investigations of the many sanctuary buildings besides temples: Kyrieleis 2013). Comprehensive catalogues and analysis of particular building components are less common nowadays than in the past century, with the exception of terracotta roofs, which are usually treated in isolation from the rest of a monument (Barletta 2011: 621–23, and the last subsection of Topical bibliography 3: Lulof & Rescigno 2011; Aversa 2012; Conti 2012; Sapirstein 2012; Lohmann, Kalaitzoglou & Lüdorf 2013; Kolia 2014; Lejsgaard Christensen & Bøggild Johannsen 2015; Giaccone 2015; for typologies of stone members, see, e.g., Barletta 2011: 622–63; Dirschedl 2013; and examples cited below). Thus, the traditional organization of research by chronology and typology continues to segment much of the current literature, drawing sharp and sometimes artificial dividing lines between the people and cultures who at different times inhabited the same places. In contrast, a considerable volume of new scholarship is either explicitly diachronic or methodologically responsive to developments in other academic disciplines. Concerning the sanctuary, many recent monographs and conferences adopt a more holistic view of cult activity and ritual in relation to architectural frames, space, and topography (including Proskynetopoulou 2011; Biraschi, Cipriani & Greco 2012; Emmerling 2012; Kourayos 2012; Interdonato 2013; Kalogeropoulos 2013; Verdan 2013; Agostino & Macrì 2014; Østby 2014; Romano & Voyatzis 2014; Nordquist, Voyatzis & Østby 2014; Déroche et al. 2014; Papapostolou 2014; Schaus 2014; Scott 2014; Sporn 2016/2017; Wescoat 2017; edited volumes include: Wescoat & Ousterhout 2012; Raue & Gerlach 2013; Melfi & Bobou 2016; Charalambidou & Morgan 2017; Frielinghaus & Stroszec 2017; Kristensen & Friese 2017; Mazarakis Ainian 2017). Furthermore, the increasing attention paid to building process and design, especially for monumental architecture, promises to dissolve some of the traditional distinctions between scholarly approaches to Minoan, Mykenaian, and Classical Greek construction techniques (e.g., Shaw 2009; McEnroe 2010; Senseney 2011; Pakkanen 2013a; Weber 2013; Blackwell 2014; 2018; Brysbaert 2014; Wiersma 2014; Wilson Jones

4 2014; Balık 2015; Pullen 2015; Corso 2016; Sapirstein 2016; Sassù 2016; Gounares 2017; Scahill 2017; Devolder 2018; and papers in recent conference volumes, some dedicated specifically to the topic of architectural techniques and materials: von Kienlin 2011; Svenshon, Boos & Lang 2012; Miles 2015; Patay-Horváth 2015; Greco & Nicolucci 2016; Zambas et al. 2016; Kurapkat & Wulf-Rheidt 2017). The handbook edited by Miles (2016) also includes seven of 35 chapters concerned with these matters. In contrast, Barletta (2011: 622, 628–29) lists comparatively few examples of publications on architectural design and technique. The study of architectural energetics—estimating the cost of materials and labour invested in an ancient structure—has taken off with great speed in both prehistoric and Classical research, accompanied by a renewed epigraphical interest in building contracts (including Patay-Horváth 2012; Brysbaert 2013; Devolder 2013; Pakkanen 2013b; Prignitz 2014; Harper 2016; Fitzsimons 2017; and several chapters in: Steinkeller & Hudson 2015; Tommaso & Scardozzi 2016; Barletta’s 2011 review bypasses this topic altogether, which since foundational work by Burford [1965; 1969] has been a steadily growing area of interest in both Greek and Roman architectural studies, such as: DeLaine 1997, 2001; Feyel 2006). An important interdisciplinary aspect of the study of Greek architecture is urbanism, capitalizing on the rich ancient literary evidence for city planning and management together with ever more extensively excavated settlements throughout the Mediterranean from the prehistoric to classical eras (Barletta 2011: 626–27; and Topical bibliography 5: Haggis, Mook & Fitzsimons 2011; Lolos & Gourley 2011; Tartaron et al. 2011; Gallou & Henderson 2012; Perna 2012; Banks & Reese 2013; Pozzatello 2014; Anzalone 2015; Uggeri 2015; Watrous, Buell & McEnroe 2015; Guzzo 2016; Nevett, Tsigarida & Archibald 2017; Tombrägel 2017). Architecture is the focus of several chapters in edited volumes relating to urbanism: e.g., Neudecker (2011); Mulliez (2013); Niemeier, Pilz & Kaiser (2013); Gaignerot-Driessen & Driessen (2014); Sposito (2014); Matthaei & Zimmermann (2015); Bonacasa, Buscemi & La Rosa (2016); Istituto per la Storia e l’Archeologia della Magna Grecia (2016); Caliò & des Courtils (2017); Letesson & Knappett (2017); Martin-McAuliffe & Millette (2018). Urban planning and the relationship of cities to the landscape is connected to an explosion of interest in a particular type of civic structure, the fortification, which Barletta noted had only seen limited publication by the 2000s. Since then, more than a dozen major investigations of walls and other varieties of military and naval architecture have appeared in print (compare Barletta 201, 626,

Sapirstein

with Fachard 2016 and the subsection of Topical bibliography 4; for general surveys of fortifications see Frederiksen 2011; Fachard 2012; 2016; Beck-Brandt, Ladstätter & Yener-Marksteiner 2015; Frederiksen et al. 2016; Müth et al. 2016; Maher 2017; Ballmer, Fernández-Götz & Mielke 2018; on particular sites: Grandjean, Wurch-Kozelj & Kozelj 2011; Konecny, Aravantinos & Marchese 2013; Konecny & Ruggendorfer 2014; Seifried & Parkinson 2014; Reinders 2014; Dietz & Kolonas 2016. Shipsheds, harbours: Lovén 2011; Ladstätter, Pirson & Schmidts 2014). At the domestic level, studies of houses (Barletta 2011: 625–26; see the last subsection of Topical bibliography 4: Haggis, Mook & Fitzsimons 2011; Glowacki & Vogeikoff-Brogan 2011; Day & Glowacki 2012; Haug & Steuernagel 2014; Di Castro, Hope & Parr 2015), farmsteads (Lapdula 2012; Lanza Catti & Swift 2014; McHugh 2017), workshops, and other sites of production (e.g., Esposito & Sanidas 2012; Sanidas 2013; Vogeikoff-Brogan 2014; Glazebrook & Tsakirgis 2016) have likewise flourished. Each of these developments show how scholars of Greek architecture have begun to interrogate the material record in new ways, many of which are breaking down traditional typological and chronological divisions. These directions promise to reframe the discipline around new paradigms and approaches. 4

Structure of the Volume

The contributions to this volume interrelate with one another across many axes, but four general themes unify the 19 chapters. The first two sections are rooted in the careful study and reconstruction of individual buildings. Those in Part 1 are primarily concerned with the initial design and construction process, whereas Part 2 examines modifications, destruction, and preservation of architectural remains. The third section is concerned with how buildings communicate, i.e., what they potentially tell us about meaning, group identity, and cultural or economic history. Although digital technologies play a central role in many chapters from the preceding sections, those of Part 4 explicitly address the potential of 3D visualization, virtual reconstruction, and simulation in Greek architectural research. Part 1, “Planning, Organization, and Methods of Ancient Greek Architects and Masons”, comprises four chapters investigating ancient architectural practice through close analysis and autopsy of monuments. The subjects range in scale from the small cult buildings at Aigeira to the Parthenon, but each analysis reveals how the ancient builders who planned and executed monuments

INTRODUCTION: Recent Developments in the Study of Greek Architecture

allowed for considerable variation between architectural members of the same kind, routinely made significant alterations to the plan midway during construction, or designed details like mouldings on the building site. Even at the Parthenon—one of the largest, costliest, and most precisely executed Greek monuments which nonetheless was completed within a comparatively narrow time frame—major changes in the design appear to have been made up until the installation of the roof. The specifications at the outset of a job might be altered by the architect, and the details worked out course-by-course through the process of construction. The first two chapters, by Lena Lambrinou and Vasileia Manidaki, arose from the extensive work on the Athenian Akropolis conducted by the Greek Ministry of Culture to rehabilitate the anastylosis from the early 20th century, which necessitated the disassembly and examination of standing monuments as well as the fabrication of new blocks for stabilizing the final restoration. The concomitant benefits to our understanding of the history and design of the monument cannot be overstated, as already reflected in research by Manolis Korres, Kostas Zambas, and others (see, e.g., Bouras & Eleutheriou 2013; and many papers in Zambas et al. 2016). Focusing on specific components, the two chapters in this book lead to significant new insights about the execution and planning of the Parthenon as a whole. In Chapter 1, the study of the north colonnade by Lambrinou is well informed by its thorough examination of not only the exterior surfaces of the stylobate and drums, but also the concealed joint seams inside the temporarily disassembled columns. Variant jointing methods reveal the “hands” of different teams of masons working side by side, whereas unused setting marks indicate an ad hoc change perhaps attributable to the arrival of a new architect. Variations in the degree of entasis on the shafts shed light on the methods by which the curvature throughout the building was implemented at the construction site. The collaboration with masons who fabricated new members for the anastylosis provides an important contemporary analogue, which together with epigraphic evidence supports an estimate of how much labour was invested in fluting the columns. Also concerned with anomalies in the Parthenon, Manidaki focuses on the southwest corner of the Parthenon entablature in Chapter 2. Apparent deviations from the original plan provide another window into the decision-making process of Greek architects, who chose to complete the facades of the temple before the courses of the entablature in the flanks. Manidaki also accounts

5

for alterations at the roof level from traces left on the corner geison block, and she brackets the timing of the changes within the overall construction sequence. In Chapter 3, Jeanne Capelle examines architectural practice and design from the evidence of incised architectural diagrams. The drawing for the entasis of the columns in the fourth-century temple at Didyma has dominated much scholarship in this area (Heisel 1993; Corso 2016), but Capelle broadens the inquiry to incorporate newly discovered examples. Although the diagrams are a sort of “blueprint”, the situation in antiquity differs from the finished conception expected for modern buildings—even if the as-built structure may diverge from a modern blueprint. The preserved ancient diagrams suggest comparatively incomplete preliminary planning, where details were planned out at the construction site by applying simple rules of geometry and proportion. In Chapter 4, Alexandra Tanner variously approaches the three small cult buildings near the theatre of Aigeira, which were constructed and modified during the first half of the Hellenistic period. A metrological study of the building plans identifies a regular grid used to lay out each building. An important result is the lack of the standardised foot units often assumed in Greek architecture: one building matches the unit observed in the adjacent theatre as well as on the Salamis relief, but a different unit appears in the neighbouring building from the same century. Tanner also considers the function of these cult buildings, usually identified as naïskoi, and finds evidence for a wider range of potential uses. Part 2, “Life History of Greek Monuments and Sites”, considers the lifespan of buildings beyond the period of initial design and execution. The “life history” concept is drawn from anthropological research focused on the archaeology of households, which typically examines stratigraphy and the finds associated with daily activities (e.g., Binford 1968: 21–22; Schiffer 1996: 13–15; LaMotta & Schiffer 1999). Although stratigraphic analysis is not particularly revealing at the sites considered in this section, many insights are gleaned from the material and technique of the structures, the findspots and architectural context of stone blocks and terracotta tiles, and the traces of tooling and other interventions preserved on the surfaces of the masonry. Much of the research throughout this book is preoccupied with understanding the past, but the final chapter of this section, in proposing and testing a new intervention at Dhespótiko, reminds us that the architectural history of a monument is a continuum, including the presentation and conservation of an archaeological site for posterity.

6 In Chapter 5, Nancy L. Klein articulates the life history approach as part of her ongoing study of the small archaic buildings from the Athenian Akropolis, known by limestone blocks discarded or reused after the Persian sack. On the one hand, traces of tooling on the extant blocks reveal their sequence of installation, repairs and maintenance during use, and how they were broken apart for burial after their destruction. On the other, Klein deduces from where their blocks were discovered in the foundations of the Mnesiklean Propylaia that two of the archaic buildings must have been disassembled and stockpiled on the Akropolis. She also identifies new archaic buildings with Kykladic mouldings, signaling an Aigaian influence in the early Akropolis. The next chapter by Nils Hellner reassesses the origins and demolition of the South Temple at Kalapódhi, whose appearance can be restored with unusual precision due to the burning of the sanctuary by the Persian army in 480 BCE. The stone blocks and terracotta tiles of the west pediment collapsed just outside the building, and the lack of other elements of the superstructure among the debris makes it very likely that the peristyle columns and architraves were wooden. Hellner presents different iterations of the reconstructed temple, with the capitals, if wooden, shaped as crossing brackets; or, if stone, with a more traditional flaring Doric echinos; and then integrating a series of specially truncated tiles that must have decorated the base of the pediment. The example of the South Temple cautions against assuming canonical Doric forms in less well-preserved archaic architecture. The latter two chapters of Part 2 extensively employ digital recording to examine the use and modification of ancient architecture. In Chapter 7, Buell, McEnroe, Besadolombana & Bieńkowski describe their remapping of the site and environs of Goúrnia. Their deployment of photogrammetry for 3D recording is germane to the final section of this book (on the technique, see Remondino et al. 2014; Sapirstein & Murray 2017), but their primary research questions concern the history of occupation of individual buildings and assessment of the relationship of the prehistoric settlement to the landscape. The new work has examined the phasing across the site and within individual houses, as revealed by building features, seams in wall construction, and a variety of masonry “fabrics”, where different materials and techniques correspond to different periods in the life of the settlement. Photogrammetric modeling also plays an important role in the work presented by Kourayos, Daifa, Orestidis, Egglezos, Papavasileiou & Toumbakari in Chapter 8. At the sanctuary of Apollon in Dhespótiko, the Temple-Hestiatorion complex can be virtually restored in 3D from the excavated

Sapirstein

remains, including much of the walls, colonnades, and entablature. Beyond the architectural autopsy, however, the 3D models were also used to prepare for physical reconstructions at the site. The impact of extreme earthquakes and windstorms on variant proposals for anastylosis could be simulated using the discrete element method. Not only did the structural analyses demonstrate which model was preferable, but they also recommended the future schedule for assessments. Their contribution is a valuable reminder that the life history of excavated monuments extends into the present and future (for related discussions, see Lambrinou 2016 and the papers in Masino, Mighetto & Sobrà 2012; Greco & Nicolucci 2016). Part 3, “Architecture, Cultural History, and Communication” unites papers that investigate the ways in which buildings speak to audiences in antiquity as well as contemporary scholars. The first two chapters are diachronic studies examining large numbers of monuments in order to reconstruct ancient economic history and social groups. The latter chapters investigate how monuments communicated with their intended audiences, either from the period of their construction or over time, as older monuments acquired new meanings in changing cultural environments. Each author carefully develops an appropriate theoretical apparatus for the interpretation of meaning at different points during the life and afterlife of ancient buildings. In Chapter 9, Kyle Jazwa models how buildings and technical attributes correlate to behaviour, social groups, and identity over the longue durée, presenting a case study of domestic architecture in the late Mykenaian through the Protogeometric period. While individual structures or building techniques might vary idiosyncratically, the “big data” approach adopted by his study reveals broader socio-historical patterns in each period. From hundreds of attributes and structures, Jazwa concludes that after the collapse of the palatial system, even if residences generally became smaller, building techniques in the LH IIIC period are not fundamentally changed, which is consistent with an overall social continuity. However, the transition to the Protogeometric period is distinguished by new spatial organisation and building techniques, pointing to more profound social changes which might have been caused, for example, by immigration and the integration of semi-nomadic groups into sedentary farming communities. In the next chapter András Patay-Horváth also extrapolates social information from architectural remains but in the case of Doric temples in late Archaic and Classical Triphylia. The study begins with a rough estimate of financial outlays which emphasises that the temple of Zeus

INTRODUCTION: Recent Developments in the Study of Greek Architecture

at Olympia dwarfed the cost of the others in the region. An important conclusion is that we tend to give outsized importance to structures like the temple of Apollon at Bassai, which are exceptionally well preserved, although the more recently discovered temples at Mákiston and Prasidháki appear to have been nearly as expensive to build. Next, Patay-Horváth interrogates the historical records for which groups were likely to have sponsored each project, arguing that each temple was constructed in order to reinforce political claims by Elian, Spartan, Arkadian, and local groups who were fighting for control over Triphylia over the course of the Classical era. The remaining chapters of Part 3 address the meaning of architecture, including the intentions of designers and patrons as well as perceptions by ancient communities. Mark Wilson Jones in Chapter 11 revisits several topics from his recent monograph on the origins of the Classical orders (Wilson Jones 2014), expanding on the idea that sacred architecture was intimately tied with other arts, especially votives located nearby buildings in sanctuaries. Recognizing important insights from 19th century scholarship, in particular that of Karl Bötticher and Gottfried Semper, he scrutinises several cases where Greek architectural elements have strong formal and symbolic affinities with minor arts: the kalathos of the Korinthian capital and a basket; the Doric and Ionic echinos and an offering bowl; and the triglyph and tripod. Supplemented by a series of organisational tables, the text considers carefully the thorny problem of how to reconstruct the influences and meaning of ancient architectural elements from the limited archaeological and textual evidence which has survived. The innovative methods not only shed light on the developments of the three architectural elements reviewed in this study but also should prove instructive for future studies of related problems. In Chapter 12, Silke Müth examines the symbolism of Greek fortifications, which previous scholarship has engaged primarily in terms of defensive function. Her study invokes semiotics, more specifically the symbol theory of architecture (Baumberger 2010), adapted to understand those aspects of fortifications which do not appear motivated by practical concerns. For example, walls routed through non-strategic but visually impressive positions, ornamental facades at key passageways, and the selection of varying stones, colours, and finishes for decorative effect—all commonplace in Greek fortifications—suggest deliberate communicative choices by the designers. Although we have few textual sources from which directly to restore the objectives of the patrons, Müth-Frederiksen’s diachronic survey of fortifications from Geometric through Hellenistic Greece makes a

7

strong case that broader social changes, such as the transition to monarchy, are reflected in the symbolic elements in walls from different periods. Sarah A. Rous (Chapter 13) also explores the communicative aspect of fortification walls, specifically those which reuse and display older building material. As the term spolia has increasingly been problematised in scholarship due to its pejorative connotations, she promotes “upcycling” as an alternative, a term with the advantage of recognizing the analogous processes observable in visual culture created from any medium, and opening a rich potential for diachronic analysis of upcycling as it reoccurs from antiquity up to the present (Rous 2019; also see Altekamp, Marcks-Jacobs & Seiler 2013; 2017; Frey 2016). The chapter focuses on architectural expressions of community and social memory in two Athenian fortifications integrating reused material: the North Akropolis and the Post-Herulian walls. Scholars have associated the former with the spirit of early Classical Athenai, in sharp contrast to the latter, categorised as spolia signaling the end of antiquity. However, Rous sees many signs that the Post-Herulian Wall was carefully planned and executed, and celebrated by its creators. Rather than the one project signaling a beginning and the other an end, both walls can be viewed as reinforcing similar messages of recovery after a devastating invasion. In Chapter 14, Matthias Grawehr turns to another question of symbolism in public architecture, the aesthetics of “roughed-out” masonry in Hellenistic Greece. While the discussion relates to questions of design raised in the first section of this book, Grawehr is concerned with the intentions of architects who left parts of buildings in a noticeably unfinished condition. The phenomenon is common in fully completed Greek monuments, as if a principle of design where roughness could be chosen for purely visual effect was a standard element of architectural practice. Avoiding the term “rustication” due to its specific associations with the Renaissance and Baroque periods, he compares the technique to asperitas or trachytes, referring to a rough style of Hellenistic rhetoric, also used by Vitruvius (3.3.9) to describe the harsh contrast of light and shadow in the peristyle of a pseudodipteral temple. In a manner analogous to the rhetorical technique, Hellenistic architects appear to have juxtaposed rough and polished elements for emphasis, where unfinished play against finished elements and direct the attention of the viewer toward specific positions of a building or urban space. Part 4, “Simulation, Experience, and Interaction with Greek Architecture” addresses the impact of digital technology on the analysis of ancient monuments. Many of the chapters in the previous sections integrate new imaging

8 technologies and computational techniques—notably, Manidaki’s (Chapter 2) combination of photogrammetric 3D models with renderings from CAD reconstructions of the Parthenon, Capelle’s (Chapter 3) utilisation of RTI imagery to clarify construction diagrams, the deployment by Buell et al. (Chapter 7) of photogrammetric recording at a massive scale, the combination by Kourayos et al. (Chapter 8) of photogrammetry with numerical modeling, and Jazwa’s (Chapter 9) statistical and cultural modeling based on architectural data. However, the foregoing chapters explore how 3D software, related spatial tools, and the increasing accessibility of powerful computers have opened up many new areas of investigation through interactive visualisation. Virtual reconstructions complement and in several important ways go beyond what can be imagined during an in-person visit to an archaeological site, even in the presence of a physical anastylosis of the surviving remains. In Chapter 15, Mary B. Hollinshead observes the contrasting experiences generated by Geographic Information Systems (GIS) and Virtual Reality (VR) tools. She articulates the fundamental limitations of GIS for her phenomenological approach, turning instead to VR representations with their potential for the experience of place at a human scale. Even if significantly abstracted, such models simulate perception and encourage us to move and explore pathways, as can be seen in the VR environment created for the Sanctuary of Demeter and Kore at Korinthos. Based on the qualitative experiences from this model, Hollinshead considers how the configuration of sanctuaries modulates the experience and navigation through sacred places. Next, Miriam G. Clinton (with A. MacLaughlin) investigates the quantitative potential for VR models as a means of generating new experiential data in lost structures (Chapter 16). She relates the concept of “digital heuristics” to the study of Greek architecture, where recording and reconstruction of a building in 3D software presses researchers to think beyond the 2D plan view. From a photogrammetric state model of the House of the Rhyta at Pseira, Clinton reconstructs a hypothetical 3D restoration that serves as a virtual platform for examining circulation patterns. She proposes a gaming environment, with incentives for online users to explore the interiors, and the software tracking user movements, gathering evidence through this crowdsourcing for circulation patterns within the House of the Rhyta. Fron, Stappmanns, Zhou & Leistner in Chapter 17 have also conducted experiential modeling, but in an auditory environment. Most archaeological VR simulation has relied on vision for the essential immersive experience,

Sapirstein

although the integration of sound and other senses has more recently become a goal (see, e.g., Richards-Rissetto 2017: 11–14). Movement through the simulated spaces is restricted by topography and walls, although direct tactile feedback is lacking. The authors lay the groundwork for examining the sense of hearing in council halls, where the intelligibility of speech would be essential. Designing 3D models for the Bouleuteria in Athenai and the Curia in Rome, they simulate how well both an untrained and a professional speaker could be understood by a listener in various positions. These experimental methods have bearings on social questions, for example, demonstrating how citizens standing in the court outside the Athenian Bouleuterion would not be heard well by those gathered inside the building, and that the Roman Curia has poor acoustics which would have required not only professional training but also a slow, measured style of speech in order to be understood. In Chapter 18, Marconi, Scahill & Limoncelli describe the integration of multiple digital technologies for the recording, reconstruction, and simulation of monuments on the Akropolis of Selinous, raising three challenges confronted by the concurrent excavation and architectural study ongoing since 2006. First, how does a team manage the wide range of digital information now acquired by cutting-edge technologies for excavation and survey? The “digital state plan” concept invoked at Selinous includes the walls, excavated contexts, objects, and survey points that have long been standard in archaeological fieldwork, but it also incorporates much richer spatial data—notably photogrammetric models recording in 3D down to submillimetric detail the current preservation of the remains, as well as idealised CAD models restored from these state models. Second, they consider how these new resources have impacted the architectural recording, analysis, and reconstruction of individual monuments of the akropolis. Third, the authors describe their development of a VR environment, now at an advanced stage, which permits exploration of the monuments both in their current state of preservation, and also rendered in reconstruction views. They observe the impact of this rich visual material on their thinking beyond isolated buildings about the relationships among the akropolis monuments and open-air cult spaces. The closing chapter by Bonna D. Wescoat addresses many of the questions raised in Part 4, and indeed throughout this volume, about the repercussions on architectural analysis of new visualisation techniques, through the particular case of the 3D CAD modeling of the Sanctuary of the Great Gods at Samothrake. On the one hand, the extensive and increasingly lifelike 3D renderings

INTRODUCTION: Recent Developments in the Study of Greek Architecture

and VR simulations of the sanctuary have prompted investigations of vision, perception, and navigation. On the other hand, Wescoat raises important concerns over the fidelity to damaged and incomplete ancient monuments. As visualisations and digital environments become increasingly immersive they simultaneously convey a false sense of certainty in reconstruction. Since fundamental aspects of the configuration of the Ionic Porch and the Nike Monument are unknown, an advantage of digital modeling is its ability to generate multiple alternative reconstructions, viewed side by side to convey ambiguity. In fact, managing uncertainty occupies all of the chapters in this volume, and we may observe related solutions to Wescoat’s iterative views in the previous discussions of Kalapódhi (Chapter 6), Dhespótiko (Chapter 8), and Pseira (Chapter 16). 5

Summary: New Directions in Greek Architecture

On the most basic level, the studies in this book reflect the prevailing trends in scholarship mapped out above in the bibliographic review (Part 3). However, the format of this volume, where individuals describe current work in comparatively short chapters, provides a different perspective on the state of the discipline. Monographs require a substantial period of time to craft, and thus reflect research inaugurated many years before the final date of publication. Multi-authored volumes, especially those arising from conferences, can take equally long to prepare, and, despite exceptions, as a whole favour topics that are established enough to attract numerous contributions already at an advanced stage of research. Thus, the current project has the potential to reveal subtle and more recent shifts in thinking than would a review of published books and long articles. To begin, the papers here resonate with observations about the direction of the field enumerated in previous overviews. For example, Barletta (2011: 629–30) advocated for more contextual analysis of Greek architecture, which we witness in the contributions here considering the multiplicity of functions or life-histories (Chapter 4–8), communication, and perceptions of buildings (Chapters 11–19). Reception occupies a prominent position in the general handbook on Greek architecture edited by Miles and, whether explicitly named or not, also plays a role in many of the current studies, in particular Chapters 12–15. More specifically, in the volume by Miles (2016, part IV), reception comprises 9 of the 35 total chapters, including discussions of the influence of Greek architecture

9

within the Mediterranean through the Roman era, and later receptions in the early Modern period. In contrast, the study of reception does not play as significant a role in the current volume because each chapter is dominated by analysis and interpretation of the Greek material rather than later reactions to it. Neither is the hermeneutical approach advocated in Marconi (2014) central here, probably due to the focus on material culture (see above). All of the following studies expand upon the 20thcentury paradigm of detailed recording, description, and reconstruction of a single monument, or compiled examples of one building component. While rooted in detailed observation, the chapters in Part 1 are concerned with ancient approaches to design and construction, and thus they are motivated by questions of agency, such as the choices of individual architects, the activities of teams of builders, or the allocation of funds for construction. We have already observed in the recent literature an increasing attention to such questions, in particular ancient building contracts or energetics in the context of both prehistoric and Classical building (Topical bibliography 3; Chapters 1, 2, and 10). The meticulous cataloguing of the surviving architectural elements has often led to a focus on restoring a single phase of a monument—usually immediately after its completion—while downplaying later modifications to a structure during its use and disuse. The primacy given to a single, authoritative vision of a building at the moment of its dedication is challenged by the life history concept variously exhibited by the chapters in Part 2, and also by the interest expressed in Part 1 in the construction process before the completion of a monument. Similarly, the studies in Part 4 question how Greek architecture should be represented and experienced despite fundamental uncertainties in the reconstruction. Overall, the research in this volume avoids singular explanations or universalizing models, instead advocating diversity, multiplicity, and nuance of interpretation. This is a genuine departure from most architectural research of the previous century. The chapters in the latter half of this book look beyond the traditional limits of the scholarship on Greek architecture. Part 3 makes the case that architectural historians should seek inspiration from and engage with conversations taking place in other disciplines of the humanities. On the one hand, specialists in the field gain important insights into ancient society and meaning through the application of extra-disciplinary methods. On the other hand, architecture represents an important source of information from which to reconstruct historical and cultural phenomena in ancient societies. The abundant remains of Greek structures present a vast and largely

10 underexploited data set upon which to reconstruct cultural history (see Morris 2000; 2005: 107–25; and Chapter 9 for additional literature). Due to its permanent connection to a place, the collective nature of the construction process, the usage of common materials and standardised building techniques, and other attributes unique to architecture, ancient buildings provide insight into cultural processes that might be missed in studies of other categories of material culture—such as pottery, graves, and the like. The research in Part 3 indicates many promising new avenues for developing cultural, economic, and social histories through reassessing architectural remains, either through diachronic investigation of individual monuments (Chapters 10, 13–14) or of large numbers of buildings to investigate cultural change over the longue durée (Chapters 9–11, 12). The digital technologies examined in Part 4 probably represent the most radical departure from even the recent research on Greek architecture represented by monographs, long-format articles, and edited volumes. Even as a wide array of mapping and 3D modeling techniques have become fundamental tools for architectural fieldwork and analysis over the last two decades, most discussions are limited to short papers in journals or edited volumes on subjects well outside the discipline of architectural history. The ensuing chapters reveal several important recent developments in the field. First, we may observe that digital methods are not being justified from an assumption that they are inherently valuable or necessarily an improvement over manual methods. Instead, technologies are being selectively applied to address well-defined research questions—as seen in Chapters 2–4, 7, and 8. Second, not only are digital methods harnessed to serve an intellectual agenda in Part 4, but they also address problems in a manner that would be impractical or impossible to imagine without computerisation—such as immersive exploration, acoustic modeling, or iterative visualisations of lost structures and whole sites. Such models introduce new possibilities for experimentation, and simulations generate data with the potential to shed light on cultural questions, including the nature of ancient cult or rhetorical practices (Chapters 16–19). Third, digital reconstructions created in CAD software have since their early days been criticised for their tendency to render an immaculate “originalist” view of the structure, implying an unwarranted sense of certainty and finality beyond how we perceive traditional line-drawing reconstructions (e.g., Miller & Richards 1995; Ryan 1996; Kensek, Dodd & Cipolla 2004; Frischer & Dakouri-Hild 2008; Clark 2010: 68–71). Another sign of the maturity of digital methods

Sapirstein

in Greek architecture is the cautious handling of incomplete evidence, accompanied by the exploration in Part 4 of how digital methods might convey uncertainty more explicitly than possible through past techniques. In closing, the traditional paradigms appear to leave more room for change, particularly in two areas. First, while many recent publications include digitally generated imagery, architectural historians have generally been hesitant to address the impact of the increasing reliance on digital methods for the recording, analysis, and publication of antiquity. The relative silence in the literature does not reflect the lively conversations taking place informally among researchers who have eagerly experimented with and integrated a wide range of new methods into their day-to-day work. More discourse along the lines of Part 4 is warranted, as the discipline realises the potential of digital technologies to improve existing approaches and to address new, previously unexplored questions. Second, the review of recent scholarship as well as the papers in this volume show that the typological divisions established since the 20th century remain strong. Prehistoric is separated from classical, and monuments are further subdivided by assumed function and location. The categorisations developed over the last centuries of research are necessary and useful—and indeed it would be difficult to navigate the vast literature concerning ancient architectural remains without these classifications by chronology, form, and function—but there is no inherent need to limit investigation within particular types of buildings. Many of the ensuing chapters are beginning to erode these conventional divisions, some in pursuit of a more comprehensive understanding of the urban or rural built environment, and others through examining historical changes observable through the comparison of many structures over time. New priorities that promise to dissolve typology include the study of construction technique and process (as in Part 1), function and use of buildings over time (Part 2), various socio-cultural phenomena (Part 3), and experience of and interaction with Greek architecture (Part 4). Architectural historians will only improve their ability to address questions of broad interdisciplinary relevance by engaging with multiple building types, non-architectural material culture, and written evidence. In sum, we all stand to benefit by looking beyond traditional frames. Much has already changed in recent decades. Further enriched by an engagement with scholarly perspectives outside the field, the study of architectural history appears to have been broadened and revitalised, contributing back to the many other areas of the humanities from which it has drawn new research questions and methodologies.

INTRODUCTION: Recent Developments in the Study of Greek Architecture

Archaeologists and historians have also turned to critical theory to develop more sophisticated and nuanced understandings of the purposes and functions of ancient architecture and spaces. Architectural remains are not only meaningful in and of themselves but, collectively, constitute a potentially immense mine of data that promises to enhance our narratives of ancient cultural and social history. Increasingly, we are relying on digital technologies to record, assess, and simulate ancient buildings and places in service of our investigations into ancient process, experience, and history. List of References Altekamp, S., Marcks-Jacobs, C. & Seiler, P. (edd.), 2013: Perspektiven der Spolienforschung 1: Spoliierung und Transposition. Topoi 15 (Berlin). Altekamp, S., Marcks-Jacobs, C. & Seiler, P., 2017: Perspektiven der Spolienforschung 2: Zentren und Konjunkturen der Spoliierung. Topoi 40 (Berlin). Barletta, B.A., 2011: “State of the Discipline: Greek Architecture”, AJA 115.4: 611–40. Baumberger, C., 2010: Gebaute Zeichen: Eine Symboltheorie der Architektur (Frankfurt). Binford, L.R., 1968: “Archaeological Perspectives”, in Binford, S.R. & Binford, L.R. (edd.), New Perspectives in Archaeology (Chicago) 5–32. Bouras, C. & Eleutheriou, V. (edd.), 2013: Επεμβάσεις στα μνημεία της Ακρόπολης 2000–2012: τα ολοκληρωμένα προγράμματα (Athens). Burford, A., 1965: “The Economics of Greek Temple Building”, PCPS 191: 21–34. Burford, A., 1969: The Greek Temple Builders at Epidauros: A Social and Economic Study of Building in the Asklepian Sanctuary during the Fourth and Early Third Centuries B.C. (Toronto). Caliò, L.M. & des Courtils, J. (edd.), 2017: L’architettura greca in Occidente nel III secolo a.C. Atti del convegno di studi, Pompei-Napoli 20–22 maggio 2015. Thaisos Monographs 8 (Rome). Clark, J.T., 2010: “The Fallacy of Reconstruction”, in Forte, M. (ed.), Cyber-Archaeology. BAR-IS 2177 (Oxford) 63–73. Corso, A., 2016: Drawings in Greek and Roman Architecture (Oxford). Delaine, J., 1997: The Baths of Caracalla: A Study in the Design, Construction, and Economics of Large-scale Building Projects in Imperial Rome. JRA Suppl. 25 (Portsmouth). Delaine, J., 2001: “Bricks and Mortar: Exploring the Economies of Building Techniques at Rome and Ostia”, in Mattingly,

11

D.J. & Salmon, J. (edd.), Economies Beyond Agriculture in the Classical World (London) 230–68. des Courtils, J. (ed.), 2015: L’architecture monumentale grecque au IIIe siècle a.C. (Bordeaux). Dirschedl, U., 2013: Die griechischen Säulenbasen. AF 28 (Wiesbaden). Feyel, C., 2006: Les artisans dans les sanctuaires grecs aux époques classique et hellénistique à travers la documentation financière en Grèce. BEFAR 318 (Athens). Frey, J.M., 2016: Spolia in Fortifications and the Common Builder in Late Antiquity. Mnemosyne Suppl. 389 (Leiden). Frischer, B. & Dakouri-Hild, A. (edd.), 2008: Beyond Illustration: 2D and 3D Digital Technology as Tools for Discovery in Archaeology. BAR-IS 1805 (Oxford). Gebhard, E.R. & Gregory, T.E., 2015: Bridge of the Untiring Sea: The Corinthian Isthmus From Prehistory to Late Antiquity. Hesperia Suppl. 48 (Princeton). Greco, C. & Nicolucci, V. (edd.), 2016: Selinunte: Restauri dell’antico (Selinunte, Baglio Florio, 20–23 ottobre 2011) (Rome). Heisel, J.P., 1993: Antike Bauzeichnungen (Darmstadt). Kensek, K.M., Dodd, L.S. & Cipolla, N., 2004: “Fantastic Reconstructions or Reconstructions of the Fantastic? Tracking and Presenting Ambiguity, Alternatives, and Documentation in Virtual Worlds”, Automation in Construction 13: 175–86. Kissas, K. & Niemeier, W.-D. (edd.), 2013: The Corinthia and the Northeast Peloponnese: Topography and History From Prehistoric Times Until the End of Antiquity. Proceedings of the International Conference, Organized by the Directorate of Prehistoric and Classical Antiquities, the LZ’ Ephorate of Prehistoric and Classical Antiquities and the German Archaeological Institute, Athens, held at Loutraki, March 26–29, 2009 (Munich). Lambrinou, L., 2016: “Preservation: The Parthenon’s East Porch”, in Miles 2016, 526–45. Lamotta, V.M. & Schiffer, M.B., 1999: “Formation Processes of House Floor Assemblages”, in Allison, P. (ed.), The Archaeology of Household Activities (London) 19–29. Marconi, C. (ed.), 2014: The Oxford Handbook of Greek and Roman Art and Architecture (Oxford). Masino, F., Mighetto, P. & Sobrà, G. (edd.), 2012: Restoration and Management of Ancient Theaters in Turkey: Methods, Research, Results. Proceedings of the Hierapolis International Symposium, Karahayit-Pamukkale (Denizli), Lycus River Hotel, 7th–8th of September 2007 (Galatina). Miles, M.M., 2015: Autopsy in Athens: Recent Archaeological Research on Athens and Attica (Oxford). Miles, M.M. (ed.), 2016: A Companion to Greek Architecture (Chichester). Miller, P. & Richards, J., 1995: “The Good, the Bad, and the Downright Misleading: Archaeological Adoption of Computer

12

Sapirstein

Visualization”, in Huggett, J. & Ryan, N. (edd.), Computer Applications and Quantitative Methods in Archaeology 1994. BAR-IS 600 (Oxford) 19–22. Morris, I., 2000: Archaeology as Cultural History: Words and Things in Iron Age Greece (Malden, MA). Morris, I., 2005: “Archaeology, Standards of Living, and Greek Economic History”, in Manning, J.G. & Morris, I. (edd.), Ancient Economy: Evidence and Models (Stanford) 91–126. Patay-Horváth, A. (ed.), 2015: New Approaches to the Temple of Zeus at Olympia. Proceedings of the First Olympia-Seminar 8th–10th May 2014 (Newcastle upon Tyne). Remondino, F., Spera, M.G., Nocerino, E., Menna, F. & Nex, F., 2014: “State of the Art in High Density Image Matching”, The Photogrammetric Record 29.146: 144–66. Richards-Rissetto, H., 2017: “An Iterative 3D GIS Analysis of the Role of Visibility in Ancient Maya Landscapes: A Case Study From Copan, Honduras”, Digital Scholarship in the Humanities 32.2: 195–212. Rous, S.A., 2019: Reset in Stone: Memory and Reuse in Ancient Athens (Madison). Ryan, N.S., 1996: “Computer Based Visualisation of the Past: Technical ‘Realism’ and Historical Credibility”, in Higgins, T., Main, P. & Lang, J. (edd.), Imaging the Past: Electronic Imaging and Computer Graphics in Museums and Archaeology. BMOP 144 (London) 95–108. Sapirstein, P. & Murray, S., 2017: “Establishing Best Practices for Photogrammetric Recording During Archaeological Fieldwork”, JFA 42.4: 337–50. Schiffer, M.B., 1996: Formation Processes of the Archaeological Record (Salt Lake City). Reprint. Original edition: Albuquerque, 1987. Townsend, R.F., 2016: “From Hellenistic to Roman Architecture”, in Miles 2016, 454–69. Wilson Jones, M., 2014: Origins of Classical Architecture: Temples, Orders and Gifts to the Gods in Ancient Greece (New Haven). Zambas, C., Lambrinoudakis, V., Simantoni-Bournia, E. & Ohnesorg, A., 2016: ΑΡΧΙΤΕΚΤΩΝ: Honorary Volume for Professor Manolis Korres (Athens).



Recent Bibliography on Greek Architecture, Organised by Topic Bronze Age Architecture

Bailey, D., 2018: Breaking the Surface: An Art/Archaeology of Prehistoric Architecture (Oxford). Betancourt, P.P., 2012: The Dams and Water Management Systems of Minoan Pseira (Philadelphia). Blackwell, N., 2014: “Making the Lion Gate Relief at Mycenae: Tool Marks and Foreign Influence”, AJA 118.3: 451–88. Bonacasa, N., Buscemi, F. & La Rosa, V. (edd.), 2016: Architetture del Mediterraneo: scritti in onore di Franscesco Tomasello. Thiasos Monografie 6 (Rome).

Brysbaert, A., 2014: “Talking Shop. Multicraft Workshop Materials and Architecture in Prehistoric Tiryns, Greece”, in RebaySalisbury, K., Brysbaert, A. & Foxhall, L. (edd.), Knowledge Networks and Craft Traditions in the Ancient World: Material Crossovers (New York) 37–61. Cooper, F.A. & Fortenberry, D., 2017: The Minnesota Pylos Project, 1990–98. BAR-IS 2856 (Oxford). Devolder, M., 2018: “The Functions of Masons’ Marks in the Bronze Age Palace at Malia (Crete)”, AJA 122.3: 343–65. Dimitriou, V.E., 2016: “L’Acropoli di Atene durante il neolitico finale e il Bronzo antico: lo studio ex novo dei rinvenimenti dello scavo Levi sulle pendici meridionali. Rapporto preliminare”, ASAtene 92: 15–31. Fitzsimons, R.D., 2017: “Architectural Energetics and Archaic Cretan Urbanization”, in Rupp, D.W. & Tomlinson, J.E. (edd.), From Maple to Olive. Proceedings of a Colloquium to Celebrate the 40th Anniversary of the Canadian Institute in Greece: Athens, 10–11 June 2016 (Athens) 345–83. Letesson, Q. & Knappett, C. (edd.), 2017: Minoan Architecture and Urbanism: New Perspectives on an Ancient Built Environment (Oxford). McEnroe, J.C., 2010: The Architecture of Minoan Crete: Constructing Identity in the Aegean Bronze Age (Austin). Shaw, J.W., 2009: Minoan Architecture: Materials and Techniques. Studi di Archeologia Cretese 7 (Padua). Shaw, J.W., 2014: Elite Minoan Architecture: Its Development at Knossos, Phaistos, and Malia (Philadelphia). Shaw, M.C. & Shaw, J.W. (edd.), 2012: House X at Kommos: A Minoan Mansion Near the Sea. Part 1. Architecture, Stratigraphy, and Selected Finds (Philadelphia). Tartaron, T.F., Pullen, D.J., Dunn, R.K., Tzortzopoulou-Gregory, L., Dill, A. & Boyce, J.I., 2011: “The Saronic Harbors Archaeological Research Project (SHARP): Investigations at Mycenaean Kalamianos, 2007–2009”, Hesperia 80.4: 559–634. Tsipopoulou, M., 2016: A Minoan Palatial Settlement in Eastern Crete: Excavation of Houses I.1 and I.2. Petras, Siteia I. Prehistory Monographs 53 (Philadelphia). Wiersma, C., 2014: Building the Bronze Age: Architectural and Social Change on the Greek Mainland during Early Helladic III, Middle Helladic and Late Helladic I (Oxford).

Sanctuaries Temples

Cited above: Wilson Jones 2014. Anzalone, R.M., 2013: “Una nuova area sacra di Gortina preromana: l’ ‘edificio A’ sulla collina di Armì”, ASAtene 91: 229–85. Barletta, B.A., Dinsmoor Jr., W.B. & Thompson, H.A., 2017: The Sanctuary of Athena at Sounion. Ancient Art and Architecture in Context 4 (Princeton). Cahill, N. & Greenwalt Jr., C.H., 2016: “The Sanctuary of Artemis at Sardis: Preliminary Report, 2002–2012”, AJA 120.3: 473–509.

INTRODUCTION: Recent Developments in the Study of Greek Architecture Gruben, G., 2014: Der Polykratische Tempel im Heraion von Samos. Samos 27. Edited by H.J. Kienast (Bonn). Hansen, E. & Le Roy, C., 2012: Le temple de Léto au Létoon de Xanthos: Etude architecturale. Fouilles de Xanthos 11. 2 vols. (Copenhagen). Hellner, N. & Gennatou, F., 2015: “Il tempio arcaico a Trapezà presso Eghion: ricerche e proposte di ricostruzione”, ASAtene 93: 115–33. Koenigs, W., 2015: Der Athenatempel von Priene. AF 33 (Wiesbaden). Mattern, T., 2015: Das Herakles-Heiligtum von Kleonai: Architektur und Kult im Kontext. Kleonai 1 (Wiesbaden). Paga, J. & Miles, M.M., 2016: “The Archaic Temple of Poseidon at Sounion”, Hesperia 85.4: 657–710. Prignitz, S., 2014: Bauurkunden und Bauprogramm von Epidauros (400–350): Asklepiostempel, Tholos, Kultbild, Brunnenhaus. Vestigia: Beiträge zur Alten Geschichte 67 (Munich). Sapirstein, P., 2016: “The Columns of the Heraion at Olympia: Dörpfeld and Early Doric Architecture”, AJA 120.4: 565–601. Theodoropoulou-Polychroniadis, Z., 2015: Sounion Revisited: The Sanctuaries of Poseidon and Athena at Sounion in Attica (Oxford). Wescoat, B.D., 2012: The Temple of Athena at Assos (Oxford).



Sanctuary Buildings besides Temples

Cited above: Prignitz 2014. Arapogianni, X., 2014: “O Θησαυρός από τoν ναό τoυ Aσκληπιoύ και της Υγιείας στήν αρχαία Θoυρία”, ArchEph 153: 185–96. Bommelaer, J.-F., 2015: “Nouveautés concernant l’architecture de la Tholos de Delphes”, CRAI 2015: 1733–58. Hayashida, Y., Yoshitaki, R. & Ito, J., 2013: Architectural Study of the Stoas of the Asklepieion at Ancient Messene (Fukuoka). Hering, K., 2015: Schatzhäuser in griechischen Heiligtümern (Rahden). Hollinshead, M.B., 2015: Shaping Ceremony: Monumental Steps and Greek Architecture (Madison). Kienast, H.J., 2016/2017: “Eine monumentale Terrasse im Heraion von Samos”, AM 131/132: 79–97. Kienast, H.J., Moustaka, A., Grossschmidt, K. & Kanz, F., 2017: “Das archaische Osttor des Heraion von Samos. Bericht über die Ausgrabungen der Jahre 1996 und 1998”, AA 2017/1: 125–212. Kyrieleis, H. (ed.), 2013: XIII. Bericht über die Ausgrabungen in Olympia. 2000 bis 2005 (Tübingen). Nielsen, I., 2014: Housing the Chosen: The Architectural Context of Mystery Groups and Religious Associations in the Ancient World (Turnhout). Schulz, T. (ed.), 2012: Dipteros und Pseudodipteros: Bauhistorische und archäologische Forschungen. Internationale Tagung 13.–15. November 2009 an der Hochschule Regensburg. Byzas 12 (Istanbul).

13

Voigts, C., 2017: Die Ältäre in den Stadtheiligtümern. Studien zur westgriechischen Altararchitektur im 6. und 5. Jahrhundert v.Chr. Selinus 6 (Wiesbaden).



Cult, Ritual Activity

Agostino, R. & Macrì, M.M. (edd.), 2014: Il Thesmophorion di Locri Epizefiri (Reggio Calabria). Østby, E. (ed.), 2014: Investigations in the Sanctuary of Athena Alea 1990–94 and 2004. Tegea 2 (Athens). Biraschi, A.M., Cipriani, M., Greco, G. & Taliercio Mensitieri, M., 2012: Culti greci in Occidente 3: Poseidonia-Paestum (Taranto). Charalambidou, X. & Morgan, C., 2017: Interpreting the Seventh Century B.C.: Tradition and Innovation (Oxford). Déroche, V., Pétridis, P., Badie, A., Destrooper-Georgiades, A. & Karali, L., 2014: Le secteur au sud-est du péribole. FdD 2.15 (Athens). Emmerling, T.E., 2012: Studien zu Datierung, Gestalt und Funktion der “Kultbauten” im Zeus-Heiligtum von Dodona. Antiquitates—Archäologische Forschungsergebnisse 58 (Hamburg). Frielinghaus, H. & Stroszec, J. (edd.), 2017: Kulte und Heiligtümer in Griechenland: Neue Funde und Forschungen. Beiträge zur Archäologie Griechenlands 4 (Möhnesee). Raue, D. & Gerlach, I. (edd.), 2013: Sanktuar und Ritual: Heilige Plätze im archäologischen Befund. ForschungsCluster 4.10 (Rahden). Interdonato, E., 2013: L’Asklepieion di Kos: Archeologia del culto. ArchCl Suppl. 12 (Rome). Kalogeropoulos, K., 2013: To ιερό της Aρτέμιδoς Tαυρoπόλoυ στις Aλές Aραφηνίδες (Λoύτσα). Πραγματείαι της Aκαδημίας Aθηνών 71 (Athens). Kourayos, Y., 2012: Despotiko: The Sanctuary of Apollo (Athens). Kristensen, T.M. & Friese, W. (edd.), 2017: Excavating Pilgrimage: Archaeological Approaches to Sacred Travel and Movement in the Ancient World (Oxon). Mazarakis Ainian, A. (ed.), 2017: Sanctuaires archaïques des Cyclades (Rennes). Melfi, M. & Bobou, O., 2016: Hellenistic Sanctuaries: Between Greece and Rome (Oxford). Nordquist, G.C., Voyatzis, M.E. & Østby, E., 2014: Investigations in the Temple of Athena Alea 1991–94. Tegea 1 (Athens). Papapostolou, I.A., 2014: To ιερό τoυ Θέρμoυ στην Aιτωλία: Iστoρία—μνημεία—περιήγηση τoυ χώρoυ. Bιβλιoθήκη της εν Aθήναις Aρχαιoλoγικής Eταιρείας 291 (Athens). Proskynetopoulou, R., 2011: Aρχαία Eπίδαυρoς: εικόνες μιας αργoλικής πόλης από την πρoϊστoρική επoχή έως την ύστερη αρχαιότητα. Aρχαιoλoγικά ευρήματα και ιστoρικές μαρτυρίες (Athens). Romano, D.G. & Voyatzis, M.E., 2014: “Mt. Lykaion Excavation and Survey Project, Part 1: The Upper Sanctuary”, Hesperia 83.4: 569–652.

14

Sapirstein

Schaus, G.P., 2014: Stymphalos: The Acropolis Sanctuary 1. Phoenix Suppl. 54 (Toronto). Scott, M., 2014: Delphi: A History of the Center of the Ancient World (Princeton). Sporn, K., 2016/2017: “Forschungen zur Anlage, Ausdehnung und Infrastruktur des heiligtums von Kalapodi: die Kampagnen 2014–2016”, AM 131/132: 193–278. Verdan, S., 2013: Le sanctuaire d’Apollon Daphnéphoros à l’époque géométrique. Eretria 22. 2 vols. (Gollion). Wescoat, B.D., 2017: The Monuments of the Eastern Hill. Samothrace 9. 2 vols. (Princeton). Wescoat, B.D. & Ousterhout, R.G. (edd.), 2012: Architecture of the Sacred: Space, Ritual, and Experience from Classical Greece to Byzantium (Cambridge).



Architectural Practice and Design

Cited above: Shaw 2009; Mcenroe 2010; Blackwell 2014; Brysbaert 2014; Wilson Jones 2014; Miles 2015; Patay-Horváth 2015; Corso 2016; Greco & Nicolucci 2016; Sapirstein 2016; Zambas et al. 2016; Devolder 2018. Balik, D., 2015: Deciphering Ornaments: Discourses and Thresholds in Architectural History. Phoibos Humanities Series 4 (Vienna). Blackwell, N., 2018: “Experimental Stone-cutting with the Mycenaean Pendulum Saw”, Antiquity 92 (361): 217–32. Gounares, A.P., 2017: Τεκτονική και οικοδομική κατά Πλάτωνα. Σχόλια και σημειώσεις επί των απαρχών της έννοιας της αρχιτεκτονικής, από την αμάρτυρον στον Πλάτωνα αρχιτεκτονική στην Architectura του Βιτρουβίου (Athens). Kurapkat, D. & Wulf-Rheidt, U. (edd.), 2017: Werkspuren: Materialverarbeitungen und handwerkliches Wissen im antiken Bauwesen. Internationales Kolloquium in Berlin vom 13. 16 Mai 2015 (Regensburg). Pakkanen, J., 2013a: Classical Greek Architectural Design: A Quantitative Approach (Helsinki). Pullen, D.J., 2015: “How to Build a Mycenaean Town: The Architecture of Kalamianos”, in Schallin, A.-L. & Tournavitou, I. (edd.), Mycenaeans Up to Date: The Archaeology of the Northeastern Peloponnese. Current Concepts and New Directions. ActaAth-4° 56 (Athens) 377–90. Sassù, A., 2016: Iktinos: l’architetto del Partenone (Rome). Scahill, D.R., 2017: “Craftsmen and Technologies in the Corinthia: The Development of the Doric Order”, in Hanberg, S. & Gadolou, A. (edd.), Material Koinai in the Greek Early Iron Age and Archaic Period. Acts of an International Conference at the Danish Institute at Athens, 30 January–1 February 2015 (Aarhus) 221–44. Senseney, J., 2011: Art of Building in the Classical World: Vision, Craftsmanship, and Linear Perspective in Greek and Roman Architecture (Cambridge).

Svenshon, H., Boos, M. & Lang, F. (edd.), 2012: Werkraum Antike: Beiträge zur Archäologie und antiken Baugeschichte (Darmstadt). Von Kienlin, A. (ed.), 2011: Holztragwerke der Antike. Internationale Konferenz 30. März–1. April 2007 in München. Byzas 11 (Istanbul). Weber, U., 2013: Versatzmarken im antiken griechischen Bauwesen. Philippika 58 (Wiesbaden).



Building Contracts, Energetics



Building Elements: Terracotta Roofs

Cited above: Prignitz 2014; Fitzsimons 2017. Brysbaert, A., 2013: “Set in Stone? Socio-economic Reflections on Human and Animal Resources in Monumental Architecture of Late Bronze Age Tiryns in the Argos Plain, Greece”, Arctos 47: 49–96. Devolder, M., 2013: Construire en crète minoenne: une approche énergétique de l’architecture néopalatiale. Aegaeum 35 (Leuven). Harper, C.R., 2016: Laboring with the Economics of Mycenaean Architecture: Theories, Methods, and Explorations of Mycenaean Architectural Production, (diss., Florida State University). Pakkanen, J., 2013b: “The Economics of Shipshed Complexes: Zea, a Case Study”, in Blackman, D., Rankov, B., Baika, K., Gerding, H. & Pakkanen, J. (edd.), Shipsheds of the Ancient Mediterranean (Cambridge) 55–75. Patay-Horváth, A., 2012: “Die Baukosten des Zeustempels von Olympia”, Boreas 35: 1–10. Steinkeller, P. & Hudson, M. (edd.), 2015: Labor in the Ancient World. International Scholars Conference on Ancient Near Eastern Economies 5 (Dresden). Tommaso, I. & Scardozzi, G. (edd.), 2016: Ancient Quarries and Building Sites in Asia Minor. Research on Hierapolis in Phrygia and Other Cities in South-western Anatolia: Archaeology, Archaeometry, Conservation. Bibliotheca archaeologica 45 (Bari).

Aversa, G., 2012: I tetti achei: terrecotte architettoniche di età arcaica in Magna Grecia. Tekmeria 15 (Paestum). Conti, M.C., 2012: Le terrecotte architettoniche di Selinunte: Tetti del VI e V secolo a.C. Museo civico di Castelvetrano e Parco archeologico di Selinunte. Sicilia antica 5 (Pisa). Giaccone, N., 2015: Architectural Terracottas at the Sanctuary of Punta Stilo at Kaulonia: Genesis, Problems, Developments. BAR-IS 2777 (Oxford). Kolia, E., 2014: “Archaic Terracotta Reliefs from Ancient Helike”, Hesperia 83.3: 409–45. Lejsgaard Christensen, J. & Bøggild Johannsen, K., 2015: Thorvaldsen’s Ancient Terracottas: A Catalogue of the Ancient

INTRODUCTION: Recent Developments in the Study of Greek Architecture Greek, Etruscan and Roman Terracottas in Thorvaaldsens Museum (Copenhagen). Lohmann, H., Kalaitzoglou, G. & Lüdorf, G. (edd.), 2013: Das Dach des archaischen Panionion. Forschungen in der Mykale III,2. Asia Minor Studien 70 (Bonn). Lulof, P. & Rescigno, C. (edd.), 2011: Deliciae Fictiles IV. Architectural Terracottas in Ancient Italy: Images of Gods, Monsters and heroes. Proceedings of the international conference held in Rome (Museo Nazionale Etrusco di Villa Giulia, Royal Netherlands Institute) and Syracuse (Museo Archeologico Regionale ‘Paolo Orsi’), October 21–25, 2009 (Barnsley). Sapirstein, P., 2012: “The Monumental Archaic Roof of the Temple of Hera at Mon Repos, Corfu”, Hesperia 81.1: 31–91.



Public Monuments Agora, Stoas

Ampolo, C. (ed.), 2012: Agora greca e agorai di Sicilia (Pisa). Caliò, L.M., Caminneci, V. & Liviadotti, M. (edd.), 2017: Agrigento. Nouve richerche sull’area pubblica centrale (Rome). Cavalier, L., Descat, R. & des Courtils, J. (edd.), 2012: Basiliques et agoras de Grèce et d’Asie Mineure. Mémoires 27 (Bordeaux). Chandrasekaran, S. & Kouremenos, A. (edd.), 2015: Continuity and Destruction in the Greek East: The Transformation of Monumental Space from the Hellenistic Period to Late Antiquity. BAR-IS 2765 (Oxford). Emme, B., 2013: Peristyl und Polis: Entwicklung und Funktionen öffentlicher griechischer Hofanlagen. Urban Spaces 1 (Berlin). Giannikouri, A. (ed.), 2011: The Agora in the Mediterranean from Homeric to Roman Times. International Conference Kos, 14–17 April 2011 (Athens). Mertens, D., 2012: “Die Agora von Selinunt. Der Platz und die Hallen”, RM 118: 51–178. Rocco, G., 2013: Monumenti di Kos 1: la stoà meridionale dell’agorà. Thiasos 3 (Rome). Scahill, D.R., 2012: The South Stoa at Corinth: Design, Construction and Function of the Greek Phase (diss., University of Bath). Sielhorst, B., 2015: Hellenistische Agorai: Gestaltung, Rezeption und Semantik eines urbanen Raumes. Urban Spaces 3 (Berlin).



Baths, Fountains

Cited above: Betancourt 2012. Campisi, T., 2015: Terme e bagni di Sicilia antica: caratteri di un’architettura specialistica. Trame di architettura e tecnica 1 (Palermo). Klingborg, P., 2017: Greek Cisterns: Water and Risk in Ancient Greece, 600–50 BC, (diss., Uppsala Universitet). Lucore, S.K. & Trümper, M. (edd.), 2013: Greek Baths and Bathing Culture: New Discoveries and Approaches. BABesch Suppl. 23 (Leuven).

15

Robinson, B.A., 2011: Histories of Peirene: A Corinthian Fountain in Three Millennia. Ancient Art and Architecture in Context 2 (Princeton). Robinson, B.A., 2013: “Playing in the Sun: Hydraulic Architecture and Water Displays in Imperial Corinth”, Hesperia 82.2: 341–84. van Tilburg, C., 2015: Streets and Streams: Health Conditions and City Planning in the Graeco-Roman World (Leiden). Wassenhoven, M.E., 2012: The Bath in Greece in Classical Antiquity: The Peloponnese. BAR-IS 2368 (Oxford). Wellbrock, K., 2016: Die innerstädtische Wasserbewirtschaftung im hellenistisch-römischen Pergamon. DWhG Sonderband 14 (Siegburg).

Theaters

Csapo, E., Goette, H.R., Green, J.R. & Wilson, P. (edd.), 2014: Greek Theatre in the Fourth Century B.C. (Berlin). Drougou, S., 2017: To αρχαίo θέατρo της Bεργίνας. Συμβoλή στην ιστoρία τoυ Θεάτρoυ στην αρχαία Mακεδoνία (Thessalonike). Frederiksen, R., Gebhart, E.R. & Sokolicek, A. (edd.), 2015: The Architecture of the Ancient Greek Theater. Acts of an International Conference at the Danish Institute of Athens, 27–30 January 2012 (Aarhus). Gogos, S. & Kampourakis, G., 2011: Das Theater von Epidauros. Mit einem Beitrag zur Akustik des Theaters von Georgios Kampourakis (Vienna). Gybas, M., 2018: Das Theater in der Stadt und die Stadt im Theater: Urbanistischer Kontext und Funktionen von Theatern im kaiserzeitlichen Kleinasien (Hamburg). Isler, H.P., 2017: Antike Theaterbauten: ein Handbuch. 3 vols. (Vienna). Kringzinger, F. & Ruggendorfer, P. (edd.), 2017: Das Theater von Ephesos: archäologischer Befund, Funde und Chronologie. Ephesos 2/1 (Vienna).



Fortifications and Harbours

Ballmer, A., Fernández-Götz, M. & Mielke, D.P. (edd.), 2018: Understanding Ancient Fortifications: Between Regionality and Connectivity (Oxford). Beck-Brandt, B., Ladstätter, S. & Yener-Marksteiner, B., 2015: Turm und Tor: Siedlungsstrukturen in Lykien und benachbarten Kulturlandschaften. Akten des Gedenkkolloquiums für Thomas Marksteiner in Wien. Forschungen in Limyra 7 (Vienna). Dietz, S. & Kolonas, L. (edd.), 2016: The Emporion. Fortification Systems at Aghia Triada and the Late Classical and Hellenistic Habitation in Area III. The Fortifications at Pangali. Chalkis Aitolias 3 (Aarhus). Fachard, S., 2012: La défense du territoire. Eretria 21 (Gollion). Fachard, S., 2016: “A Decade of Research on Greek Fortifications”, AR 62: 77–88.

16

Sapirstein

Frederiksen, R., Müth, S., Schneider, P.I. & Schnelle, M. (edd.), 2016: Focus on Fortifications: New Research on Fortifications in the Ancient Mediterranean and the Near East. Fokus Fortifikation 2 (Oxford). Frederiksen, R., 2011: Greek City Walls of the Archaic Period, 900– 480 BC (Oxford). Grandjean, Y., Wurch-Kozelj, M. & Kozelj, T., 2011: Le rempart de Thasos. Études Thasiennes 22 (Athens). Konecny, A., Aravantinos, V. & Marchese, R. (edd.), 2013. Plataiai: Archäologie und Geschichte einer boiotischen Polis. Sonderschr. ÖAI 48 (Vienna). Konecny, A.L. & Ruggendorfer, P., 2014: “Alinda in Karia: The Fortifications”, Hesperia 83.4: 709–46. Ladstätter, S., Pirson, F. & Schmidts, T. (edd.), 2014: Hafen und Hafenstädte im östlichen Mittelmeerraum von der Antike bis in byzantinische Zeit: Neue Entdeckungen und aktuelle Forschungsansätze. Harbors and Harbor Cities in the Eastern Mediterranean from Antiquity to the Byzantine Period: Recent Discoveries and Current Approaches. Istanbul, 30.5.–1.6.2011. Byzas 19. 2 vols. (Istanbul). Lovén, B., 2011: The Zea Shipsheds and Slipways: Architecture and Topography. The Ancient Harbours of the Piraeus I.1 (Athens). Maher, M.P., 2017: The Fortifications of Arkadian City States in the Classical and Hellenistic Periods (Oxford). Müth, S., Schneider, P.I., Schnelle, M. & De Staebler, P. (edd.), 2016: Ancient Fortifications: A Compendium of Theory and Practice. Fokus Fortifikation 1 (Oxford). Reinders, H.R., 2014: The City of New Halos and its Southeast Gate (Groningen). Seifried, R.M. & Parkinson, W.A., 2014: “The Ancient Towers of the Paximadi Peninsula, Southern Euboia”, Hesperia 83.2: 277–313.



Domestic Architecture, Farms, and Workshops

Day, L.P. & Glowacki, K.T., 2012: Kavousi IIB: The Late Minoan IIIC Settlement at Vronda. The Buildings on the Periphery (Philadelphia). Di Castro, A.A., Hope, C.A. & Parr, B.E. (edd.), 2015: Housing and Habitat in the Ancient Mediterranean: Cultural and Environmental Responses. BABesch Suppl. 26 (Leuven). Esposito, A. & Sanidas, G.M. (edd.), 2012: «Quartiers» artisanaux en Grèce ancienne: une perspective méditerranéenne (Villeneuve d’Ascq). Glazebrook, A. & Tsakirgis, B. (edd.), 2016: Houses of Ill-Repute: The Archaeology of Brothels and Taverns in the Greek World (Philadelphia). Glowacki, K.T. & Vogeikoff-brogan, N. (edd.), 2011: ΣΤΕΓΑ: The Archaeology of Houses and Households in Ancient Crete. Hesperia Suppl. 44 (Princeton).

Haggis, D.C., Mook, M.S., Fitzsimons, R.D., Scarry, C.M. & Snyder, LM., 2011: “The Excavations of Archaic Houses at Azoria in 2005–2006”, Hesperia 80.3: 431–89. Haug, A. & Steuernagel, D., 2014: Hellenistische Häuser und ihre Funktionen. Internationale Tagung Kiel, 4. bis 6. April 2013 (Bonn). Lanza Catti, E. & Swift, K., 2014: A Greek Farmhouse at Ponte Fabrizio. The Chora of Metaponto 5 (Austin). Lapadula, E., 2012: The Late Roman Farmhouse at San Biagio. The Chora of Metaponto 4 (Austin). McHugh, M., 2017: The Ancient Greek Farmstead (Oxford). Sanidas, G.M., 2013: La production artisanale en Grèce: une approche spatiale et topographique à partir des exemples de l’Attique et du Péloponnèse du VIIe au Ier siècle avant J.-C. Archéologie et d’histoire de l’art 33 (Lille). Vogeikoff-Brogan, N., 2014: The Late Hellenistic Settlement: The Beam-Press Complex. Mochlos 3. Prehistory Monographs 48 (Philadelphia).



Elite Monuments



Urbanism and Planning

Cited above: Caliò & des Courtils 2017. Akamatis, N., 2013–2014: “Mία oικία των πρώιμων ελληνιστικών χρόνων από την Πέλλα. Η σωστική ανασκαφή στo oικόπεδo Γκόγκαλη”, Makedonika 40: 1–35. Bachmann, M., Radt, W. & Schwarting, A., 2017: Die Stadtgrabung 5: Bau Z, Architektur und Wanddekor. Altertümer von Pergamon 15 (Berlin). Drougou, S., 2016: “Vergina-Aigai. The Macedonian Tomb with Ionic Façade. Observations on the Form and Function of Macedonian Tombs”, in Katsonopoulou, D. & Partida, E. (edd.), Φίλελλην: Essays Presented to Stephen G. Miller / Mελετές πρoς τιμήν τoυ Στέφανoυ Mίλλερ (Athens) 335–50. Henry, O. & Kelp, U., 2016: Tumulus as Sema: Space, Politics, Culture and Religion in the First Millennium B.C. Topoi 27 (Berlin). Sporn, K., (ed.), 2013: Griechische Grabbezirke klassischer Zeit: Normen und Regionalismen. Akten des Internationalen Kolloquium am Deutschen Archäologischen Institut, Abteilung Athen, 20.–21. November 2009 (Munich). von Mangoldt, H., 2012: Makedonische Grabarchitektur: die Makedonischen Kammergräber und ihre Vorläufer. 2 vols. (Berlin).

Cited above: Haggis, Mook & Fitzsimons 2011; Tartaron et al. 2011; Bonacasa, Buscemi & La Rosa 2016; Caliò & des Courtils 2017; Letesson & Knappett 2017. Anzalone, R.M., 2015: Città e territorio dal protogeometrico all’età classica. Gortina 7. ASAtene Monografie 22 (Athens). Banks, E.C. & Reese, D.S., 2013: Lerna. A Preclassical Site in the Argolid. Results of Excavations Conducted by the American

INTRODUCTION: Recent Developments in the Study of Greek Architecture School of Classical Studies at Athens 6: The Settlement and Architecture of Lerna (Princeton). Gaignerot-Driessen, F. & Driessen, J. (edd.), 2014: Cretan Cities: Formation and Transformation. Aegis 7 (Louvain-la-Neuve). Gallou, C. & Henderson, J., 2012: “Pavlopetri, an Early Bronze Age Harbour Town in South-east Laconia”, Pharos 18.1: 79–104. Guzzo, P.G., 2016: Le città di Magna Grecia e di Sicilia dal VI al I secolo 1: La Magna Grecia. Abitare il Mediterraneo 1 (Rome). Istituto Per La Storia e L’Archeologia Della Magna Grecia, 2016: Poleis e politeiai nella Magna Grecia arcaica e classica. Atti del Cinquantatreesimo convegno di studi sulla Magna Grecia. Taranto 26–29 settembre 2013 (Taranto). Lolos, Y. & Gourley, B., 2011: “The Town Planning of Hellenistic Sikyon”, AA 2011.1: 87–140. Martin-McAuliffe, S. & Millette, D.M., 2018: Ancient Urban Planning in the Mediterranean: New Research Directions (New York). Matthaei, A. & Zimmermann, M. (edd.), 2015: Urbane Strukturen und bügerliche Identität im Hellenismus. Die hellenistische polis als Lebensform 5 (Heidelberg). Mulliez, D. (ed.), 2013: Στα βήματα τoυ Wilhelm Vollgraff: Eκατό χρόνια αρχαιoλoγικής δραστηριότητας στo Aργoς. Πρακτικά τoυ διεθνoύς συνεδρίoυ, 25–28 σεπτεμβρίoυ 2003. Sur les pas de Wilhelm Vollgraff: Cent ans d’activités archéologiques à Argos. Actes du colloque international, 25–28 septembre 2003 (Athens). Neudecker, R. (ed.), 2011: Krise und Wandel. Süditalien im 4. und 3. Jahrhundert v. Chr. Internationaler Kongress anlässlich des

17

65. Geburtstages von Dieter Mertens, Rom 26. bis 28. Juni 2006. Palilia 23 (Wiesbaden). Nevett, L.C., Tsigarida, B., Archibald, Z.H., Sonte, D.L., Horsley, T.J., Ault, B.A., Panti, A., Lynch, K.M., Pethen, H., Stallibrass, S.M., Salminen, E., Gaffney, C., Sparrow, T.J., Taylor, S., Manousakis, J. & Zekkos, D., 2017: “Towards a Multi-scalar, Multidisciplinary Approach to the Classical Greek City: the Olynthos Project”, BSA 112: 155–206. Niemeier, W.-D., Pilz, O. & Kaiser, I. (edd.), 2013: Kreta in der geometrischen und archaischen Zeit. Akten des Internationalen Kolloquiums am Deutschen Archäologischen Institut, Abteilung Athen, 27.–29. Januar 2006 (Munich). Perna, R., 2012: L’acropoli di Gortina: la tavola «A» della carta archeologica della città di Gortina. Ichnia 6 (Macerata). Pozzatello, A., 2014: Stasis e polis: Architettura e conflitto nel progetto della città greca tra il V e il IV secolo (Florence). Sposito, A., 2014: Solunto: Paesaggio città architettura (Rome). Tombrägel, M., 2017: Studien zum spätklassischen und frühhellenistischen Städtebau in Arkadien, der Dodekanes und Makedonien (Wiesbaden). Uggeri, G., 2015: Camarina: storia e topografia di una colonia greca di Sicilia e del suo territorio. Journal of Ancient Topography Suppl. 8 (Galatina). Watrous, L.V., Buell, D.M., McEnroe, J.C., Younger, J.G., Turner, L.A., Kunkel, B.S., Glowacki, K., Gallimore, S., Smith, A., Pantou, P.A., Chapi, A. & Margaritis, E., 2015: “Excavations at Gournia, 2010–2012”, Hesperia 84.3: 397–465.

part 1 Planning, Organization, and Methods of Ancient Greek Architects and Masons

∵

chapter 1

The Parthenon’s North Colonnade: Comments on Its Construction Lena Lambrinou 1 Introduction The north colonnade of the Parthenon stood intact for more than two millennia until 1687, when the middle eight columns were blown outward by an enormous explosion, triggered by a Venetian artillery shell which was fired into the temple and ignited an ammunition dump stored there by the Ottoman Turks.1 This major injury to the building was not completely repaired until the restoration of Nikolaos Balanos in 1923–1931 (Fig. 1.1) (Balanos 1940). Mistakes in the choice of materials and in the repositioning of the ancient architectural members made during this early 20th century project led the Greek Ministry of Culture to consider a new intervention that could remedy these problems for the sake of the long-term preservation of the building. The resulting study for the restoration of the north colonnade, originally prepared in 1998 by civil engineer Kostas Zambas, was updated by the present author for the columns and architraves in 2001–2002 (Zambas 2002b; Lambrinou 2005). The new intervention began in 2001 and was completed in 2009. The architectural solution proposed in the new study involved identifying the original column positions, which was based on various criteria that were either “constructional” or “archaeological/historical”. Constructional criteria refer to the differences between architectural members dictated by the positions that they occupied in the original structure—such as the diameters of individual column drums, which vary according to the columns’ entasis along their vertical axis; or variations in height and inclination along the sides of the architraves due to the curvature of the colonnade’s entablature. Archaeological/historical criteria include differences or similarities between architectural members that have resulted from their positions on the monument and from the effects of time. An example is the distinctive traces of weathering that consistently appears on the north-facing column flutes, which is attributable to their exposure to the harsh northern winds. Another criterion which can demonstrate the location of adjacent drums is the traces left by later structures that were built against the 1  This chapter expands upon the author’s article in Greek published in electronic format in 2013: Lambrinou 2013a.

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004416659_003

peristyle columns. Plant roots have also left patterns on the lower joint faces of the architraves where they rested on the capitals; this telltale indication was used by the author to identify the sequence of the architrave blocks on the north side of the Parthenon (Fig. 1.2) (Lambrinou 2005: 114–17). Likewise, manmade features on the uppersurfaces of the column capitals and under-surfaces of the architraves, which consist of scars on the marble left by metal pins or rollers that were used during construction to facilitate the placement of architrave blocks in their final position on the capitals, indicate the original associations between and positions of certain blocks (Lambrinou 2002: 201; Lambrinou 2005: 113–14). This chapter presents a series of new observations about the design and execution of the columns in the north colonnade resulting from the restoration work. Changes in the plan are evident at the stylobate level,

figure 1.1 North pteron and colonnade author

22

Lambrinou

figure 1.2 Imprints of plant roots on the upper surface of the north capital NC14130 (Tenth north column) drawing by author

while the variations in the undersides of the lowest column drums reveal techniques attributable to separate teams of masons. Next, this chapter explores the surprisingly high degree of irregularity in the diameters of the Parthenon column drums, which is best explained by the inclination of the columns and the technique for executing the entasis. The organisation of the workers finishing the columns are considered in the conclusion. 2

Design Changes in the North Colonnade at the Stylobate Level

During the new restoration, the lowest drums of the seven middle columns on the Parthenon’s north side, which had never before been moved from their original positions (with the exception of the fourth and sixth columns), were removed for structural and analytical purposes. Only six drums of the eleventh column were dismantled (i.e., down to the fifth drum). While the stylobate around the columns was given a careful final carving with a straightedged chisel (lamaki), the invisible area inside the incised circle on the stylobate under the columns alternatively exhibits a rough treatment, consistent with tooth-edged chisels (chontrodonti xoida) or more likely with toothed hammers (thrapina) of various sizes (Figs. 1.3–1.4). In those areas where the diameter of the lower drum exceeded that of the circle incised in the stylobate, the extremely fine original finish of the stylobate surface is preserved— appearing just as it would have at the time of its completion. In these areas protected from damage during the

long history of the use of the building, we see the perfectly smoothed surface that the temple would have possessed when it was finished. Furthermore, the tooth of an iron chisel was found on the stylobate beneath the tenth column, when its first drum, still in situ until the most recent intervention, was removed. Traces of the lowest drums were eroded into the stylobate by the penetration of water into recesses that had been cut around the base of the lowest drums to protect their lower edges from harm during setting (Fig. 1.3a–c). In addition, incised circles were discovered on the top surface of the stylobate. It appears that these circles were intended to mark the positions of the columns prior to their installation. These circles vary in size, however, from column to column (despite the columns having almost identical diameters). The diameter of the circles is usually bigger than, and in one case equal to, that of the recesses around the base of the lowest drum (Table 1.1). They were incised on the pavement with large compasses, both ends of which apparently had sharp metal tips. The small holes at the centre of these circles left by the compass tip were found exactly on the joints between the stylobate blocks beneath the columns (Fig. 1.3d). These holes have been invaluable in helping us to determine precisely the dimension and accuracy with which each incision describes a true circle, although the stylobate at the position of the sixth column does not preserve any trace of an incised circle. Of the six positions on the stylobate that preserve circles, the centre does not coincide with that of the recesses around the perimeter of the lowest drum or the

The Parthenon ’ s North Colonnade: Comments on Its Construction

23

figure 1.3 Stylobate exposed below the (a) eighth and (b) tenth columns; details showing (c) the incised circle, traces of the column flutes, and differential treatment below the column; and (d) compass point centered on the stylobate block joint photographs 2004, author

figure 1.4 Stylobate under the fifth column showing the difference between the actual setting of the column (dark gray) and the incised circle (light gray, outlined black) drawing by author after illustration by K. Matala, 2004

24 table 1.1

Lambrinou Measurements of the circles inscribed on top of the stylobate and the lower diameter of the corresponding columns on the north colonnade (from the east, measurements in cm)

Column (from east)

4

5

6

7

8

9

10

Diam. inscribed circle Diam. recessed zone at base W. recessed zone Offset of inscribed circle from N edge of stylobate Offset to N of inscribed center relative to recessed zone

172.6 172.0 3.5 15.85

172.0 172.0 3.5 15.5

none 173.0 3.0 _

172.8 172.45 3.2 13.6

173.0 169.8 4.6 15.8

174.3 172.5 3.2 15.25

171.4 170.0 4.5 15.7

_

0.2

2.4

1.5

impressions of the fluted columns eroded into the stylobate of the flutes. This means that at none of the six column positions where incised circles are preserved are these circles exactly concentric with the recesses and other marks left by the columns (Fig. 1.3a–c). Instead, the incised circles are consistently located to the south of the recessed zones by 1.5–2.5 cm, with a slight shift either to the east or west relative to the positioning of the columns over the underlying stylobate joint. The north edge of the incised circles are inset by 15.25–15.85 cm from the corresponding edge of the stylobate (greater on the fourth column, less on the ninth). These offsets have no relation to the position of the columns as installed, 10.25–10.50 cm from the stylobate’s north edge (as measured from the deepest point of the northern flute observable from impressions on the stylobate surface). The variation of only 2.50 mm is thus approximately constant in all eight columns as installed. The distances of the incised circles from the northern edge of the stylobate vary up to 2.3 cm (at the seventh column), while their diameters vary up to 3 cm—a considerable difference, given that the diameters of the bases of the columns were executed consistently within half a millimeter. All seven of the central columns of the north colonnade thus exhibit a slight repositioning to the north by an average distance of 2.0 cm (ca. 1 ancient dactyl) from their originally intended positions, as indicated by their underlying incised circles on the stylobate (Fig 1.4)—a dactyl is 1/16th of a foot (Dinsmoor 1923: 242–43; 1950: 161–63 [Aeginitan foot of 32.685 cm at the Parthenon]; vs. Bankel 1983: 82–89; Korres 1994b: 59 [Solonic/Attika foot of 29.37 cm]; resulting in a dactyl of 2.04 / 1.84 cm, respectively). This evidence for the final placement of the columns, during which they were moved slightly to the north in comparison to the incised “blueprint” on the stylobate, could be included with other small adjustments apparent in the building as indications of a last-minute design change

1.3

1.5

0.7

from the original plan. Dinsmoor (1923: 243) remarked: “The case is quite different with the inner building. Here we have an example of the constant revision and alteration to which most designs were subject in the course of erection; but the resulting irregularities can all be explained”. The small offset eastward (4–6 mm on the fourth, fifth, and tenth columns) or westward (4–5 mm on the seventh and ninth columns) is considered to be another kind of rearrangement connected with corrections made to the intercolumniations so as to achieve greater uniformity in these spaces. This is a rare case in which the discovery of the planning incisions on the stylobate allows us to follow the evolution in the process of construction. This particular change to the original design resulted in a small enlargement northward of the pteron, amounting to a difference of 1.5–2.7 cm between the centres of the two circles—i.e., the one incised on the stylobate, which indicates the original planned placement of the columns, and the eroded circle, which indicates the actual placement of the columns. On the opposite side of the north pteron, Manolis Korres had previously discovered a small adjustment in the second course of the cella, which he calculated had resulted in the north wall of the cella being shifted southward by as much as 7 cm (Fig. 1.5: top). Korres (1994b: 91) connects this enlargement of the pteron with an early decision to change the original plan in order to install what was likely a Doric frieze on the cella walls. Although he does not believe this deviation from the original plan was connected with a desire to align the outer side of the north cella wall with the axis of the second column on the east side, in the end this movement resulted in just such an alignment (Fig. 1.5: below). The alignment of the north cella wall with the second column on the east is in fact a general practice in most Doric temples, but oddly enough, as Korres (1994b: 89) points out, it seems that this rule was not applied in the original plan for the Parthenon. Barbara

The Parthenon ’ s North Colonnade: Comments on Its Construction

25

figure 1.5 Section through the N pteron (above), and plan (below) showing the alignment of the cella walls with the second columns of the peristyle above: author; below: after Korres 1994a, fig. 38

Barletta (2009: 554), however, has argued convincingly that the alteration of the plan was specifically aimed at aligning the cella walls with the second columns’ axes in accordance with this Doric canon, while also making the case that the Ionic frieze was not an alteration, but rather an important feature in the original plan of the building. When the shifting of the north cella wall to the south is combined with the aforementioned repositioning of the peristyle columns, the northern pteron would have been widened by a total of 9 cm. This widening on both sides

of the pteron was likely done at the same time for some still undetermined reason—perhaps, as Korres (1994b: 91) suggests, to allow a better view of a proposed frieze (at first Doric, but Ionic in the end). These two shifts may indeed have been corrections to the original plan but were implemented for different reasons. Here I would like to offer a new hypothesis: the correction to the north cella wall was made to achieve the standard Doric alignment with the second column on the east, as shown by Barletta. However, the correction

26

Lambrinou

on the north colonnade was made to achieve an exterior alignment with a newly redesigned, thicker northeast corner column that had been placed slightly north of its originally intended position in order to compensate for the significant reduction in the first intercolumniation on the east. The movement of the entire north colonnade would thus have been undertaken because of the increased thickness of the corner columns (an optical refinement) on the east facade. This particular “correction” to the north colonnade could also explain the differently sized capitals on the north side, where the capitals are larger on the first four columns (from the east) and smaller to the west. The shift of the northern columns would allow the abacus on their capitals to be diminished by 3–4 cm, because otherwise their northern edges would have projected beyond the stylobate. Most likely, the larger capitals at the east of the north colonnade had already been carved before the change in plan and had been set aside until their columns—the first three after the northeast corner column—could be erected. In general, the small corrections apparent in many areas of the building may be the result of disagreements over the original plan, perhaps, as some scholars have long argued, due to the differing views of a new architect, who wished only to make important adjustments, not to rebuild the entire temple. Carpenter (1970: 67–68, 85, 111) imagined an intermediate Parthenon designed by Kallikrates in the Kimonean era and left only halffinished in 450 BCE, which was subsequently dismantled by Iktinos in preparation for the building of the Periklean temple (also see: Bundgaard 1976: 67; Barletta 2005: 92). 3

Undersides of the Column Bases and Preliminary Planning

The restoration works also afforded a unique opportunity to examine the undersides of the seven fully dismantled columns from the north colonnade. The bottom drums lack a socket for the wooden empolion, containing a cylindrical pin, used everywhere else to guide and secure the drums into their final positions, which is a normal feature in classical Doric architecture. The bottom surfaces of the lowest drums exhibit a distinctly rough treatment unlike the joints of the other drums, and they are divided into circular zones with different treatments, separated by concentric circles incised with a compass, whose common centre can also be distinguished (Fig. 1.6). From the exterior, the zones are: (1) a recessed perimeter; (2) the main weight-bearing band,

carved with a small-toothed chisel (leptodonti xoida); (3) a rougher, recessed interior; and (4) a small, finely dressed circular or square framing a central compass hole (Fig. 1.6: right). The last was likely meant to facilitate the marking of the central point for the compass. In three of the seven cases, the weight-bearing zone (2) also contains a distinct ring with fine surface treatment. In four cases, the outer zone of excellent surface treatment is missing. Generally, all joint surfaces of the drums have horizontal anathyrosis. In addition to varying surface treatment, we can also observe differences in the width of the four zones and the appearance of the centermost areas (Table 1.2). These small differences may reflect the different “hands” or techniques of individual craftsmen, who could vary their methods even while following the same general construction rules. If individual characteristics may be identified, we can see four sets (Table 1.2): Group A (Columns 4, 5) with Zone 2 in two sub-bands similar to anathyrosis; incised lines along the axes, dividing the underside into quarters; fine treatment, and a circular Zone 4; Group A1 (Column 10) with a carved rectangular piece of marble inserted into the centre of its lower surface instead; Group B with a circular Zone 4; Group C with a square Zone 4; and Group D with a poorly executed square recess at the centre of Zone 4. In these seven drums whose lower surfaces could be examined, the marble masons—or “workshops” comprised of as many as four individuals—apparently undertook the preparation of one or two drums each. Interestingly, the drums of the same individual or group are found adjacent to each other, except in the case of the central ninth column, whose treatment is different from that of its neighbours (Table 1.2). The refinements in the building account for another peculiarity of the lowermost drums. As prescribed by Vitruvius (3.3.13) and found on the actual monument, the vertical axis along the deepest point of the southern flute is perpendicular to the stylobate, so that the bottom of the drum is at a 90º angle to the flute on its south edge (Fig. 1.7: above). However, due to the inclination of the columns, the vertical axes of every other flute stand at different angles relative to the surface of the stylobate— increasing to 95.5º at the midpoint of the drum, and up to 98.5º on the north side. Moreover, all the lowest drums of the columns of the north colonnade have a noticeably tilted zone of carving ca. 10 cm high set at a constant angle of about 84.5º in relation to the bottom of the drum—equivalent to a 95.5º angle with the stylobate. This 10 cm zone at the base of the flutes was a standardized feature carved around the lower

27

The Parthenon ’ s North Colonnade: Comments on Its Construction

figure 1.6 Underside of the bottom drums of the (left) sixth and (right) ninth columns left: author; right: A. Papandropoulos, ESMA archives table 1.2

Lower surface of the bottom drums: groupings based on technique and zone dimensions (widths measured along the column radius, in cm)

Column / Drum number

4.1

Masonry group A Zone 1: recessed exterior band 3.5 Zone 2: load-bearing face 60.5 Sub-zone 2b 13.0 Zone 3: recessed interior panel 25.5 Zone 4: raised central circle 6.0 ” : central square (L. × W.) –

5.1

6.1

7.1

8.1

9.1

A 3.5 61.5 14.0 26.0 8.0 –

B 3.0 60.5 – 26.25 8.0 –

B 3.2 60.5 – 26.25 8.0 –

C 4.6 60.5 – 24.5 3.0 2.5 × 2.5

D A1 3.0 4.5 66.0 57.1 – 21.1 20.25 27.85 – – 7.0 × 7.5 7.0 × 5.0

edge of the drum before installation and was intended as a guide for the later fluting of the rest of the column. The same method is evident on the unfinished lowest drums of the Old Parthenon (Fig. 1.8). As a result, along the southern flutes facing the cella, and immediately above the aforementioned lower guideline flutes, there is an inward angle, a kind of “elbow” (described below) in the vertical axis of the columns at the transition to the desired right angle with the stylobate (Fig. 1.7: below). The prefabricated zone apparently set the initial angle between the columns and their base, taking into consideration the taper of the shaft but not its inclination relative to the stylobate. The bidirectional curvature of the stylobate apparently made it impractical to

10.1

calculate the exact orientation of the lowest drums and their inclination on every side. Instead, these final determinations were made during the positioning of the bottom drums on the stylobate and the carving of their upper surfaces. In this way, the lowest drums served as guides for determining the final setting and carving of the upper portions of the column. These observations highlight the distinct stages in the Parthenon’s construction, which involved both a preliminary survey of features such as column settings and the bottom fluted band and secondary adjustments during construction. The architectural members were prefabricated according to the basic design of the building while also leaving some flexibility to the later construction

28

Lambrinou

figure 1.7 Schematic diagram of a bottom drum on the north colonnade. On the right, sections with exaggerated angles (above) reveal inclinations relative to the stylobate, and (below) the prefabricated “elbow” around the bottom edge at exaggerated scale. The photograph shows the perpendicular south flute of the eighth column and its “elbow” author

figure 1.8 Bottom of an unfinished base drum from the Old Parthenon, showing prefabricated flutes author

The Parthenon ’ s North Colonnade: Comments on Its Construction

stages for adapting to the particularities of their final positioning. Furthermore, such prefabrication would be expedient. The small zone of flutes initially cut at the base of the columns, with its standardized angle that deviates on the south from the final angle, indicates an acceptance by the builders of a small imperfection in the fluting for the sake of practicality. 4

Non-circular Column Drums

The new restoration disassembled sections of the north colonnade previously re-erected by Balanos, in part to study the architectural members and to replace them in the correct original positions. Initial new diameter measurements revealed that the drums are not perfectly circular, prompting the author to develop a new method for the evaluation of potential matches which revised the study of 2002 (Zambas 2002b). For the 85 drums of the middle eight columns, precise measurements of not just one diameter on each upper and lower surfaces were measured—as previously had been the procedure—but rather every flute was measured from its deepest point to the opposite flute’s deepest point, for a total of 10 measurements per drum (e.g., Fig. 1.9). Measurements of diameters that include the arrises themselves have been excluded due to poor preservation. Measurements were taken using a calibrated stainless steel measuring bar (laid directly on the horizontal surface of the column drum), in conjunction with two stainless

figure 1.9 Measurement sheet with the diameters of column 11.9/ 14149, upper side (measurements by author) after drawings by K. Matala, A. Papandropoulos, ESMA Archives

29

steel rulers (held vertically and directly against the axis of the deepest point of the flute), only during the first two hours of the work day, in the early spring of 2002. The author, working with the same two assistants (a sculptor and an experienced marble mason) throughout the diametermeasurement process, was able to distinguish measurements within a half millimeter on the calibrated bar. The ten highly accurate diameters taken on both ends of the stone drums thus provide a unique, reliable identifier for each drum, reducing the potential number of options for selecting and matching overlying or underlying drums. This more exact procedure based on multiple measurements has resulted, in many cases, in unique matches that can only reflect the original positions of the architectural members in question (Fig. 1.10). The ten diameter measurements were found to fluctuate, with deviations as high as 4 mm within a drum. The general inconsistency in diameter has been observed in the past, but explained by Dinsmoor (1927) as normal fluctuations in precision, “the ancient margin of error” (Mallouchou-Tufano 1998: Appendix 11, 319), and similarly by Zambas (2002b: 43–44) as the insignificant result either of hand-crafting or a network of internal micro-cracks. For Zambas, these internal, almost undetectable micro-cracks may have been responsible for random expansions along certain dimensions, yet it seems impossible for any hidden cracks to cause an expansion of 3 mm in any detectable direction. Notably, it was the disagreements among the measurements of earlier scholars that began to reveal the need for the more exacting method of documentation adopted here (formerly those by Dinsmoor, Schleif & Balanos, as compiled in Zambas 2002b: Table 1). For example, the upper end of column drum NC11.9 had published discrepancies in diameter of 3–4 mm (Inventory no. 14145, whose diameter to Dinsmoor was 148.1 cm; Schleif: 148.4 cm (+3 mm); Balanos: 148.45 cm (+3.5 mm); and Zambas: 148.5 cm (+4 mm), from an average of two different diameter measurements (N-S: 148.32 cm / E-W: 148.59 cm in his hand-written table of measurements in the YSMA Archives). These dimensions are all encompassed within the new per-flute measurements of 148.18–148.50 cm (Lambrinou 2005: Table II). Given the results of the new measurements, it may be concluded that these scholars likely measured different pairs of opposing flutes. This procedure represents a new approach, employing more complete data describing the changes in diameter and allowing the determination of the original positioning of column drums. After completion of the new diameter measurements, it was learned that Dinsmoor (1927, in Mallouchou-Tufano 1998: App. 11, 319), having observed the variation in diameters, had suggested—unbeknownst to the present author—that successive tests must be

30

figure 1.10

Lambrinou

North elevation of the north colonnade including restored members, and marking the height of the maximum swelling due to entasis drawing by K. Matala, A. Papandropoulos, T. Skari, 2005; ESMA Archives

made on those drums showing differences of less than 3 mm until the correct match is found: “… by rotating the drums and testing the ten diameters, such differences would disappear”. In fact, all the drums in the north colonnade have internal variation in their diameters—and, although only the disassembled drums of the middle eight columns of the north colonnade were measured for this study, it is highly probable that all of the peristyle columns exhibit such variation in the internal diameters of their column drums. On the upper and lower joint surfaces of a drum, the deepest points of the flute channels do not describe an exact circle, and no column drum in the north colonnade has been found to have a perfectly circular upper or lower end. In 1927, Wilhelm Dörpfeld and Hans Schleif came to a similar conclusion when making measurements on the north colonnade (Mallouchou-Tufano 1998: 195, note 565). When measuring the EC6 and SC12 columns to determine the entasis, Balanos (1940: 85) also found differences in diameter which he attributes to the inclination of the columns. Anastasios Orlandos (1977: 171) considered the base of the lowest drum as a perfect circle and dismissed discrepancies on the upper surface as negligible. Yet within each joint face of every drum, the ten diameter measurements differ from each other by an average of 2.8 mm. The discovery of such small yet pervasive differences in the columns’ flutes demand explanation. Is this inconsistency due to some fault in the production process or random error? Is it the result of an effort to control the light and shade effects produced by the flutes, or some other, previously unrecognized architectural refinement? Is there a correlation between shallower and narrower flutes, and the overall diameters?

To begin, small deviations in the dimensions of the architectural members of the Parthenon (at a level of millimeters) define either the level of desired perfection, or the capacity of the ancient masons to execute the ideal design. Francis Penrose (1888: 12, n. 1) conceived the latter capacity as a limit of constructional precision, which he quantified as 0.2 mm per meter, according to variations in the sizes of the ancient wooden measuring sticks that resulted from changes in atmospheric temperature and moisture (0.023 foot / 100 feet, or 8 mm / 33.33 m). The newly determined 2.8 mm average discrepancy in the diameters of the north colonnade drums is about an order of magnitude greater than Penrose’s theoretical margin of error, which would be less than half a millimeter per drum. The column diameters vary from 138–186 cm, top to bottom, in the deepest point of the flutes, and from 148.5– 190.2 cm on the arrises. It is customary to believe that the Parthenon exhibits the highest level of perfection—i.e., the highest level of control over implementation—and that “a desire for perfection” meant primarily a willingness to strive towards resolving any technical issue. However, perhaps our unavoidably incomplete present-day knowledge of ancient technical abilities leads us to misconceptions concerning the ancients’ intentions, abilities, and goals. The smaller diameters—correlated to deeper flutes— of the drums are usually found near their N-S axis, raising another possibility that such fluctuations were introduced by external environmental factors. The north side of the Parthenon is particularly affected by the stronger, more erosive northern winds. Furthermore, the lack of direct sunlight increases moisture, which leads to the development of destructive microorganisms shown to cause

The Parthenon ’ s North Colonnade: Comments on Its Construction

microcrystalline deterioration in the marble (Doganis et al. 1994: 184; Zambas 2002b: 55). The orientation of the marble’s internal veins relative to the orientation of the drums also might be a factor in how much erosion a drum has suffered through time and exposure to external conditions. It is known that the hardness/erosional resistance of the marble is combined with the direction of the stratification. In many cases, however, deeper flutes are observed beside or near much shallower flutes, or deeper flutes do not extend over the full height of the column. Moreover, a deep flute is often recorded on the SW, the most protected side of the drum facing inward towards the interior of the building. This fluctuation in depth has been observed even in the prefabricated flutes cut initially into the lowest drums (described above in Part 3), as well as in the neck (hypotrachilion) of the capitals, before the rest of the shaft was fluted later in the construction process. The irregularity in the profiles of the flutes, therefore, does not seem attributable to weathering or a general constructional rule. Understanding this irregularity requires that we take a broader look at the step-by-step construction of the monument. 5

Design and Execution of Entasis

We have yet to consider a possible explanation for the substantial variation in the drum diameters: the entasis in the Parthenon column shafts. The matter, however, is complicated by the lack of consensus about how entasis was generally planned and executed by ancient architects. According to Joseph Hoffer (1838: 373), the entasis constitutes the arc of a circle; to Francis Penrose (1888: 40, 41), a portion of a hyperbola; to Auguste Choisy (1909/1974: 147, pl. 34), a segment of two parabolas; and to Gorham Stevens (1924: 148), a conic section with different variations, parabola or hyperbola, or a combination of two conic sections, using the method of scamilli impares. In the case of the Parthenon, Dinsmoor (1950: 168) considered the entasis as a circular arc with a radius of ca. 800 m, while Lothar Haselberger (1980: 201) at the third century BCE temple of Apollon at Didyma demonstrated it to be a portion of an ellipse. Jari Pakkanen (1997: 341; 1998: 68) does not exclude any conic sections, although from his in situ measurements of temples in the Peloponnesos and at Delphoi an elliptical curve seems to best fit the evidence. John Senseney (2011: 108) considers the possibility of a Pythagorean inspiration in the diameter index incised on the Apollon temple’s cella wall at Didyma, although he

31

agrees with Haselberger that the entasis is the result of the transformation of a circular to an elliptical arc through the stretching (protraction = entasis) of the former along a vertical axis. The idea of stretching a circular arc along an axis, resulting in an ellipse, seems to be reflected in Vitruvius 1.2.1–2. The term (from the ancient Greek verb τείνω—to extend, stretch—akin to the English “tension”) seems particularly appropriate, because it captures this conversion of an arc as it is “stretched” during its application to the realisation of the shafts of the building’s columns. Thus the Greek term used by Vitruvius probably does not refer to the swelling per se of the column shaft, but to the process of producing the column’s profile curve through manipulation of the scaled work plan. “Entasis” may have been the craftsmen’s preferred term that immediately encapsulated their understanding of the process. It is noteworthy that for all the optical refinements applied to ancient temples, a special term has survived only for the swelling of columns (also see Vitr. 3.3.13). In Roman examples, Mark Wilson Jones (1999: 229) observes simpler methods to form the entasis, either using two straight lines connected together by a softened angle or “elbow”, or a combination of circular or elliptical sections with straight lines. Others have expressed skepticism. Hans Seybold (1999: 105) believes that seeking a geometric shape that accurately describes a form in an ancient construction amounts to a utopian effort, since the construction of ancient buildings was based on empirical rather than mathematical rules. Zambas (2002a: 108) also suggests that the accurate determination of the entasis curve in the columns of the Parthenon’s external colonnade is problematic, since the measurements predicted by the various mathematical models vary by only one millimeter. Nevertheless, he argues that the entasis corresponds better to an elliptical model for the north colonnade, while a circular arc more closely fits the entasis curve in the columns of the eastern pronaos (Zambas 2002a: 114). Direct evidence for the ancient method of implementing entasis remained unknown until Haselberger’s 1979 discovery of the construction diagram at Didyma (Haselberger 1985: 131; 1999a: 28). This working sketch, referred to by Korres (1991: 50; 1999: 101) as an “index of diameters” or “rays”, is a scale drawing of an entasis curve from which the diameter can be determined at every point along the column’s height (Haselberger 1985: 131). At the Parthenon, Korres and Zambas believe that an index of diameters was created as a reference when the drums were still in roughly cylindrical form (Korres 1991: 50; Zambas 2002a: 84), before their assembly and final positioning in

32 the shafts. The chart at Didyma indicates that the entasis was implemented after the cella had been finished, since it was engraved on its side walls (Haselberger 1985: 129). According to Burford (1963: 29), the cella was constructed after the peristyle, but also apparently before the columns were fluted. After the completion of a column, the entasis could have been achieved in several ways. One method, according to Stevens, was to use tightly stretched strings, from which a depth for carving could be measured at specific elevations (Stevens 1924: 124; Wilson Jones 2000: 131). Stevens does not mention whether the joints would be used as measuring points, but—as in the case of the stylobate curvature—this was supported by Choisy’s theory of the “method of odd numbers” (Choisy 1971: 146). This approach calls for the curve of the entasis to have been achieved in the same way as the stylobate curvature, with Vitruvian scamilli impares and two parabolas of different equations. The method would implement a parabolic curve with successive odd-number components, the impares, in combination with a division of the shaft into equal parts (heights and horizontal distances) (Stevens 1924: 150). Stevens noted that, even if the individual drums do not have the same height, this realisation of entasis would treat the shafts as if they were monolithic. Alternatively, according to Haselberger from his interpretation of the Didyma diagram, the index specified diameters at particular elevations drawn one dactyl apart on the diagram but scaled up to proportionately greater heights on the column shaft. While the horizontal scale was at full size, the vertical scale was effectively at 1:16 because the successive diameters were applied to the actual columns at distances one foot apart (16 dactyl = 1 foot), resulting in a shallow curve equivalent to a section of an ellipse (Haselberger 1985: 131; Wilson Jones 2000: 128). The lines on the cella walls of the temple of Apollon at Didyma are incised every 1.85 cm, equivalent to 1 dactyl. Thus, the diameters in the index could be scaled up to one foot (in this case, a Roman foot of 29.6 cm) apart on the shaft of the column, rather than setting the diameter at the joints of the drums, which are of uneven heights. In both proposals, the shaft is effectively treated as if monolithic. Wilson Jones (1999: 241), however, suggests that the desired depth of the entasis was initially marked at the joint of each drum, and not at fixed heights, by means of square depressions cut into the drums’ perimeters which served as guides for carving circumferential horizontal bands at the joints. During the recent restoration study of the north colonnade, the placement of drums in their original positions and the reconstruction of columns—with the correction

Lambrinou

of any distortions due to historical incidents or the early20th-century Balanos restoration—provided the opportunity to investigate the actual evidence for their entasis, and to make comparative observations of all the restored columns on the Parthenon’s north side. Carving new marble additions to the fragmentary ancient members as well as some entirely new drums had two results: first, these procedures confirmed the original positions of the drums determined by the new approach were accurate; and, second, they enabled taking measurements from the restored columns which shed new light on the methods for executing the entasis. The average diameters and heights of each restored drum in the north colonnade were used for calculating the entasis, following the computer-based method introduced by Zambas (2002b). According to Penrose (1888: 39), at the Parthenon the entasis curve has its greatest swelling at about the middle of the height of the columns. Zambas (2002a: 97), however, observed from the in situ third and twelfth north columns that the maximum entasis does not always occur at the same height, and furthermore does not exhibit the same curve in every column. More recently during the new restoration, it has been shown through field measurements and analysis in AutoCAD that in the third column from the east, the maximum swelling (meaning the expansion of the shaft relative to a straight taper) of 1.63 cm occurs at 5.33 m above the stylobate, whereas in the twelfth column, the maximum swelling of 1.78 cm is at a height of 5.18 m (above, Fig. 1.10). Although similar results are found for the reconstructed columns NC4– NC11, the maximum point of entasis always coincides with the upper joint of the sixth drum of the column, whose variable elevations account for the differences in the height of maximum swelling. The sixth upper joint is usually located a little above the midpoint of the column when including the capital (10.434/2 = 5.22 m). The highest maximum point is found on the central ninth column, while the elevations of maximum entasis in the adjacent columns occur at gradually lower elevations, a pattern that generally mirrors the curvature of the stylobate below. It seems that the upper joint of the sixth drum was a significant point in the formulation of the entasis, perhaps representing a middle checkpoint in the height of the column which consistently followed the curve of the stylobate. In the flutes of the northern columns, the new restoration has shown that the entasis is actually formed as a facetted line with smoothed transitions, hereafter described as “elbows”. Derived from θλάσεις—meaning a break, fracture, or rupture—this was a general observation by Korres (1991: 52). In this case, it is used to describe the points

The Parthenon ’ s North Colonnade: Comments on Its Construction

33

figure 1.11 Entasis on the north colonnade shafts. Left: photograph showing the effect of an “elbow” relative to a straightedge. Middle: the relative position of drum joints and “elbows” (e) in the entasis. Right: the semi-polygonal lines of entasis are shaded, with (e) indicating the maximum swelling of the north and south flutes (horizontal scale exaggerated) photograph and drawings by author

where the slope abruptly changes in the column profile, similar to Wilson Jones (1999: 228) who also coined the term “elbow”. As the present author observed in the north colonnade, what appears to the naked eye as a smooth curve is in reality formed of straight segments meeting at convex curves in the elbows, each ca. 5–20 cm high (Fig. 1.11: left). Each drum joint usually falls somewhere within these elbows, although in some higher drums the elbows are located roughly in the middle of the drum— as in the western flutes of the five upper drums of the seventh column, and the northern flutes of the top two drums of the eighth column. This happens particularly in the uppermost drums that make the final transition to the neck of the column capital. On these drums, the elbow starts from about 20–30 cm above the lower drum joint, spanning the middle of the drum’s elevation and exhibiting more acute angles. Moreover, evidence from the north colonnade shows that the polygonal profile of the entasis in every column is not identical on each of the twenty flutes. In the northern flutes, the elbows begin from the upper joint of the first drum on the stylobate. The change of angle in these elbows at every joint is distinctly sharp (Fig 1.11: centre).

On the other sides of the columns, however, especially on the south and occasionally in between the cardinal orientations (SW, SE, etc.), the elbows start at the upper joint of the second drum above the stylobate and occur every 1.5–2 drums (Fig. 1.11: right). In the eastern flutes on the western half of the north colonnade, and in the western flutes on the eastern half of the colonnade, the elbows also start from the second joint above the stylobate, as in columns NC7–NC9. In a few cases, the elbows start from the third joint, as in the southern flute of NC7, or the SW flutes of NC9, while the next elbow often occurs on the next drum well above the previous joint, as in the sixth column. 6

Inward Inclination of the Columns, Entasis, and Non-circular Drums

These irregularities in the entasis are best understood in light of another refinement mentioned by Vitruvius (3.5.4), the inclination of the columns towards the interior of the building. The axis of the Parthenon columns is tilted by 0.78%, or 0.45° (Fig. 1.7: above; Fig. 1.12). In other words,

34 the axis of the columns deviates horizontally by 7.5 cm towards the interior, compared to a shaft and total column height of 9.57 and 10.43 m, respectively (Zambas 2002a: 73). This returns to the previous observation in Part 3 that, at its deepest point, the south flute facing the cella stands perpendicular to the stylobate. In compensation for the stylobate slope and the taper of the column shaft, the opposite northern flute was more steeply inclined towards the interior—altogether three times more than the southern flute, while the two lateral flutes have twice the inclination of the southern (above, Fig. 1.7; Zambas 2002a: 74). Due to the different inclination of the columns, the degree of the entasis is not symmetrical around each column but rather is greatest on the northern flutes, with the highest inclination, and is at its least on the southern flutes (Fig. 1.11: right; Fig. 1.12). Because of the additional inclination towards the middle of the north colonnade, the entasis of the inner diagonal flutes is slightly smaller than that of the outer diagonal flutes. This explains why the entasis elbows were already necessary from the first drum joint about 1 m above the stylobate on the north side, while on the south, where the inclination is lowest, the first elbow is about 2 m above the stylobate, where it coincides with the upper joint of the second drum. On the ninth column, located at the centre of the north colonnade and near the highest point in the stylobate curvature, most flutes are almost perpendicular to the stylobate. There, the entasis elbows begin at the second drum joint, about 1.90 m above the stylobate, except in the case of the three northern flutes, whose initial elbow is located as usual at the first joint. Furthermore, because the entasis of the ninth column generally starts from a higher elevation, the maximum expansion is observed at a higher point on the column. In the columns of the north colonnade, the south flute stands perpendicular to the stylobate through the first two drums, up to about 2 m elevation, which means that its profile is vertical for the first two 2 m of the shaft and accordingly does not precisely match the entasis of the northernmost flute (Fig. 1.11: right). Thus, the southern profile more closely resembles the column depicted in the Didyma diagram whose lower portion also stands straight up and down. In the case of the Didyma column, however, this vertical section extends for just 1/12th the height (about 1.5 of the total 18 m), while at the Parthenon it is 1/5th of the shaft (1.80 of 9.57 m). The peristyle columns were thus executed with slightly different curves around their perimeter, varying with inclination. As observed previously, the maximum entasis in the northern flutes is always at the upper joint of the sixth drum, but in the southern flutes the maximum entasis is

Lambrinou

figure 1.12

The effect of the inclination of the column on entasis (scale exaggerated) by author

The Parthenon ’ s North Colonnade: Comments on Its Construction

actually located about one drum higher (Fig. 1.11: right). This means that the two entasis curves do not have their deepest point at the same elevation. While the north profile reaches a maximum swelling of 1.63 cm, the greatest on the south is just 1.35 cm—a difference of 2.8 mm. These asymmetries in the entasis probably account for the non-uniform depths of the flutes and diameters of each drum observed in Part 4. We may conclude that the uneven diameters are not due to imperfect craftsmanship or weathering. Instead, it is the consequence of the inclinations of the column and the subsequent asymmetrical application of the entasis. We may also assume that these differences in the entasis evident at the base of the flutes would also appear in the arrises, because in practice the entasis must have first been set for the arrises, after which the flutes would be carved in correspondence. Today, the marble masons follow such a procedure, beginning with the arrises and then carving the flute channels using metal guides. Evidently the builders did not strive for perfectly circular drums, which would not make any sense in light of the various refinements observed above. What does appear to have been intentional and in fact meticulously implemented was the correct inclination and entasis for each column. 7

Sequence and Timing of the Fluting

The final fluting of the Parthenon columns was undertaken only after the plain cylindrical drums had all been installed and the columns were complete. Leaving the flutes for last was a common procedure known from ancient building accounts (Burford 1961: 89) which can be observed from unfinished monuments, including the Doric temple at Segesta in Sicily, the temple of Nemesis at Rhamnous, and the Pre-Parthenon on the Athenian Akropolis. The relevant building inscription from the Parthenon itself is extremely fragmentary, and the timing of the fluting can only be inferred from the general context of the inscription (IG I 300–311, 327; Dinsmoor 1913; 1921). This inscription appears to state that the column fluting started as early as 444/3 BCE, in the fourth year of construction when timber for scaffolding was purchased (the construction is dated from 447/6–438 BCE). According to Dinsmoor (1913: 67), the fluting was probably completed during the sixth year, in 442/1 BCE, when the inscription refers to the payment of wages for this work. He had initially put the completion of the carving of the flutes in the seventh year, but later transferred it to the sixth (Dinsmoor 1913: 78 for 441/0 BCE; 1921: 243 for 442/1 BCE).

35

Alternatively, Alison Burford (1963: 31) dated the completion in the ninth year of construction, in 439/8 BCE. The entire fluting process therefore appears to have been completed within three to six years. Given that the Parthenon has 108 columns (equivalent to 15,800 linear-km of flutes), and the removal of the external material and fluting of the drums were achieved in approximately three years, we can calculate that on average 2.3 columns (25 drums with 439 linear-m of fluting) were completed every month.2 Of course, the difference in size between the columns of the peristyle and those of the internal colonnades would have led to significantly different times for their treatment. As discussed above (Part 3; Table 1.2), the craftsmen who worked on the peristyle columns can be distinguished as individuals or “workshops” by differences in the techniques for preparing the bases of the lowest drums. Does this evidence also indicate a similar division in the carving of the rest of the columns, in particular the process of their fluting? If so, the masons would likely have divided up the labour as they carved the joints of each drum, so that each individual or team of masons fluted two adjacent columns. In this scenario, the fluting of the 46 peristyle columns could have been executed by as many as 23 different crews. As for the crews, in the building inscription of the Erektheion, the fluting of each column (just 2.57 m perimeter at the base) is reported to have been done by a team of four to five craftsmen working simultaneously on all its sides (IG I2 474–479 for 409–405 BCE; Burford 1963: 33). At the east side of the Erektheion up to 5–7 craftsmen worked together (Stevens, Caskey & Paton 1927: 411–13; Korres 1991: 110). If every craftsman required an individual workspace of about 70–80 cm in which to work with traditional carving tools (Neufert & Neufert 2000: 15), then in the Parthenon’s north colonnade a maximum of six masons could have worked at the same level of scaffolding, crafting three flutes each. The circumference of the shaft is 4.66 m (from the 1.485 m diameter), leaving 0.78 m per worker near the top. In 2013 we actually tested such an arrangement around the top of a northern column using our trained masons. In the original process, as the flute-carving progressed downwards and the column perimeter expanded, there would have been more workspace available, allowing as many as eight masons to work 2  The peristyle has 46 columns, the porches 12 columns, the opisthodomos 4 columns, and the internal colonnade 46 columns. The 108 columns were comprised of 886 drums. The surface area of fluting is estimated for the peristyle columns, which omitting the prefabricated flutes at the base is about 9.50 m long × 20 flutes, amounting to 190 linear-m of fluting.

36 side-by-side at the largest circumference at the base (1.902 m base diameter equals a 5.97 m circumference, thus leaving 0.75 m per worker). An average of seven craftsmen per column is not unreasonable, if we accept that they were working at only one level at a time. Assuming a maximum number of eight craftsmen were employed around each column, up to 184 masons might have worked simultaneously on the fluting, or even more if teams were active at more than one level. If the masons were working on two levels, the numbers of workers could be doubled. Workers at the lower level would do the initial stages of carving, and at the upper level the finishing, in order to minimize the debris falling from above. Burford (1963: 34) proposed up to 200 total masons—probably based on the building inscription of the Erektheion, whose columns are much thinner than those of the Parthenon—and Korres also mentions this figure for the Parthenon (Korres, Panetsos & Seki 1996: 12). In all, over the course of one month, each craftsman could have carved 2.39 linear-meters of fluting, and a crew of seven masons could finish one drum per month (this is assuming 2.3 columns were finished each month, which is equivalent to 439 m of flutes / 184 workers = 2.39; on average, each drum has 17.3 m of flutes, from 190 m per shaft / 11 drums). Every day they would have carved approximately 9 cm of fluting, beginning from the drum’s cylindrical exterior surface and working down to the level of the final, fluted surface. This per diem is reasonable if we consider that the whole process was done by hand. In comparison, the present-day craftsmen, who use a combination of both mechanical and manual techniques, finish only about 5 cm of fluting per day. The average of 10 artisans carving new members in the northern colonnade completed 235 linear-m of flutes over 24 months, or 9.8 m per month (Lambrinou 2013b). Working 20 five-hour days per month, each craftsman cut close to 1 m of fluting per month. One should note that these figures describe only the final step of treatment, involving the removal of the last millimeters of debris from the flutes, as the external marble debris had already been removed using a mechanical cutter beforehand. Even the final polishing step was accelerated by the auxiliary use of small electric disk-blade cutters. In the end, however, it is clear that the modern craftsman, even working with his mechanized equipment, only produces about half the finished product of the ancient artisan. The above figures should be revised upward for today’s de facto five-hour workday, which surely was much higher in antiquity, perhaps 12 hours, and we also might assume about 300 working days every year (Burford 1963: 34). Korres (1991: 97) has calculated each stage of the building’s

Lambrinou

construction in detail, estimating that the carving of a full column with its capital would have required 9,500 hours for two to four carvers, while the carving of the shaft itself without its capital would have required 7,800 hours. He considers four workers to be the appropriate number for the available working space, which in the long term would have been the most economical (Korres 1991: 111). If indeed the workday was 12 hours and the craftsmen worked at least 300 days per year, then two to four carvers would have needed more than two years to complete the fluting of one column (without its prefabricated capital). However, if the number of carvers were increased, or even doubled, the carving of the column could have been completed in just one year in the manner shown above. In sum, according to the above calculations, the peristyle columns of the Parthenon would have been completed in less than two years, with a fluting rate of about one column per year for each crew. 8 Conclusions The recent restoration of the middle section of the Parthenon’s north peristyle has revealed previously unknown elements of its construction. It has provided the opportunity to reexamine the entasis of the columns and propose new explanations for its implementation. In addition, the substantial variations in internal diameters represent a significant criterion that has aided us in discovering the original positions and sequence of assembly for the drums in the restored columns. Α profound blend of traditional techniques, inspired innovation, and masterful technical craftsmanship came together in the creation of the Parthenon. Still, many changes during its construction have been detected, possibly as a result of disagreements over the original plan that may have arisen after changes in the architectural team overseeing the work. The discovery of the circular incisions beneath the columns may be associated with another small change of plan during construction which could be connected with the correction of the building’s optical refinements, specifically the extreme contraction of its corner intercolumniations. The carving techniques under the bases of the columns have allowed us to distinguish four workshops of marble carvers and assisted in estimating the construction time required for the peristyle columns. The long-held perception that the Parthenon exhibits the highest level of constructional perfection assumes that its architects exerted the highest possible degree of control over its realisation and were clearly willing to manage

The Parthenon ’ s North Colonnade: Comments on Its Construction

and resolve all technical issues in the very best way. One indication of this idealized approach to the organisation and control of the building project is the prefabrication of the upward tapering of the columns at the base of the lowermost drum, while omitting the inclination of their shafts due to the inability to calculate this accurately for the unfinished, unset drums. The recent intervention on the north colonnade has also shed light on the method behind the practical realisation of the entasis. It has now been shown that the entasis curve of the Parthenon’s columns consisted of long and short straight lines connected in transitional areas by curved portions, or elbows. This polygonal line, with elbows smoothing the sharp angles, was used to formulate what appeared from a distance to be a perfect arc that described the ideal, swelling outline of the column shaft. Indeed, in this way, ancient masons (and now the modern masons following them) could implement the entasis as closely as possible to a true curve. At the same time, compensating for the inclination of the columns towards the interior of the temple, the builders placed the maximum swelling of each column at a significantly higher point on its interior side than on its exterior—in accordance with the positioning of the entasis elbows, usually located at the joints between the column drums. At those opportune moments when the “hidden” constructional secrets of this revered Classical building are revealed through modern research, we have the chance to gain greater insight into the technical abilities of the Parthenon’s ancient architects, including their efforts to prioritize their constructional goals and to plan and prefabricate certain steps in the building process. Aided by such insights, we may undertake restorations with a new, greater awareness that allows us to follow more closely the spirit of the original ancient Greek builders and to proceed more respectfully when applying our own hands to ancient structures for the preservation of their magnificent marble remains. List of References Balanos, N., 1940: Η Αναστήλωση των Μνημείων Ακροπόλεως (Athens). Bankel, H., 1983: “Zum Fußmaß attischer Bauten des 5. Jahrhunderts v. Chr”, AM 98: 65–99. Barletta, B.A., 2005: “The Architecture and Architects of the Classical Parthenon”, in Neils, J. (ed.), The Parthenon: From Antiquity to the Present (Cambridge) 67–100. Barletta, B.A., 2009: “In Defense of the Ionic Frieze of the Parthenon”, AJA 113.4: 547–68.

37

Bundgaard, J.A., 1976: Parthenon and the Mycenaean City on the Heights (Copenhagen). Burford, A., 1961: “Building at Segesta”, CQ 11.1: 87–93. Burford, A., 1963: “The Builders of the Parthenon”, in Hooker, G.T.W. (ed.), Parthenos and Parthenon (Oxford) 23–35. GaR Suppl. 10. Carpenter, R., 1970: The Architects of the Parthenon (Harmondsworth). Choisy, A., 1971: Vitruve II (Paris). Original edition, 1909. Dinsmoor, W.B., 1913: “Attic Building Accounts I: The Parthenon”, AJA 17.1: 53–58. Dinsmoor, W.B., 1921: “Building Accounts V: Supplementary Notes”, AJA 25.3: 233–47. Dinsmoor, W.B., 1923: “How the Parthenon was Planned: Modern Theory and Ancient Practice”, in Architecture 21.xlvii–xlviii: 177–80, 241–44. Dinsmoor, W.B., 1950: The Architecture of Ancient Greece: An Account of its Historic Development. 3rd ed. (London). Dogani, G., Moraitou A., Papakonstantinou E., & Charalambous, D., 1994: “Παρθενώνας: Επισήμανση φθορών, Ειδικές Μελέτες, Προτάσεις Συντήρησης, Αποκατάστασης και Προστασίας”, in Korres, M., Toganidis, N., Zambas, K. & Skoulikidis, T. (edd.), Μελέτη Αποκαταστάσεως Παρθενώνα 2 (Athens) 183–86. Haselberger, L., 1980: “Werkzeichnungen am jüngeren Didymeion—Vorbericht”, IstMitt 30: 191–215. Haselberger, L., 1985: “The Construction Plans for the Temple of Apollo at Didyma”, Scientific American 253: 126–32. Haselberger, L., 1999a: “Old Issues, New Research, Latest Discoveries: Curvature and Other Classical Refinements”, in Haselberger 1999b, 1–56. Haselberger, L. (ed.), 1999b: Appearance and Essence. Refinements of Classical Architecture—Curvature (Philadelphia). Hoffer, J., 1938: “Das Parthenon zu Athen, in seinen Hauptteilen neu gemessen”, Allgemeine Bauzeitung 3: 249–51, 371–75, 379–83, 387–91. Korres, M., 1991: Συμβολή στην Οικοδομική Μελέτη των Αρχαίων Κιόνων (diss., Εθνικό Μετσόβιο Πολυτεχνείο). Korres, M., 1994a: “The Architecture of the Parthenon”, in Tournikiotis, P. (ed.), The Parthenon and its Impact in Modern Times (Athens) 55–97. Korres, M., 1994b: “Der Plan des Parthenon”, AM 109: 53–120. Korres, M., 1999: “Refinement of Refinements”, in Haselberger 1999b, 79–104. Korres, M., Panetsos, G.A. & Seki, T. (edd.), 1996: The Parthenon, Architecture and Conservation (Athens). Lambrinou, L., 2002: “Ο θριγκός της βόρειας Όψης του Παρθενώνα: ταύτιση μελών και ανατοποθέτηση κιονοκράνων και επιστυλίων”, in Πρακτικά 5ης διεθνούς Συνάντησης για την Αποκατάσταση των Μνημείων Ακροπόλεως, Αθήνα 4–6.10.2002 (Athens) 195–208. Lambrinou, L., 2005: Αναδιάταξη Μελών Βόρειας Κιονοστοιχίας: Κίονες, Κιονόκρανα, Επιστύλια (Unpublished report, Athens).

38 Lambrinou, L., 2013a: “4.3.1.2. Η Βόρεια Κιονοστοιχία του Παρθενώνα: Μελέτη Κιόνων και Επιστυλίων και Νέες Κατασκευαστικές Παρατηρήσεις”, in Boura, C. & Eleutheriou, V. (edd.), Επεμβάσεις στα μνημεία της Ακρόπολης 2000–2012: τα ολοκληρωμένα προγράμματα (Athens) 1–38. Lambrinou, L., 2013b: “Κατάξεση ραβδώσεων κιόνων Βόρειας Κιονοστοιχίας” (Unpublished report, Athens). Mallouchou-Tufano, F., 1998: Οι Αναστηλώσεις των Αρχαίων Μνημείων στη Νεώτερη Ελλάδα (Athens). Neufert, E. & Neufert, P., 2000: Architects’ Data. 3rd ed. Edited by B. Baiche & N. Walliman (Oxford). Orlandos, A., 1977: Η Αρχιτεκτονική του Παρθενώνος 1 (Athens). Pakkanen, J., 1997: “Entasis in Fourth-Century BC Doric Buildings in the Peloponnese and Delphi”, BSA 92: 323–44. Pakkanen, J., 1998: The Temple of Athena Alea at Tegea: A Reconstruction of the Peristyle Column (Helsinki). Penrose, F.C., 1888: An Investigation of the Principles of Athenian Architecture, or the Results of a Survey Conducted Chiefly

Lambrinou With Reference to the Optical Refinements Exhibited in the Construction of the Ancient Buildings at Athens (London). Senseney, J.R., 2011: The Art of Building in the Classical World: Vision, Craftsmanship, and Linear Perspective in Greek and Roman Architecture (Cambridge). Seybold, Η., 1999: “The Mathematic Basis for the Evaluation of Curvatures”, in Haselberger 1999b, 105–12. Stevens, G.P., 1924: “Entasis of Roman Columns”, MAAR 4: 121–52. Stevens, G.P., Caskey, L.D. & Paton, J.M. (edd.), 1927: The Erechtheum, Measured, Drawn and Restored (Cambridge, MA). Wilson Jones, M., 1999: “The Practicalities of Roman Entasis”, in Haselberger 1999b, 225–49. Wilson Jones, M., 2000: Principles of Roman Architecture (New Haven). Zambas, K., 2002a: Οι Εκλεπτύνσεις των κιόνων του Παρθενώνα (Athens). Wilson Jones, M., 2002b: Μελέτη Δομικής Αποκατάστασης Βόρειας Όψης Παρθενώνα (Athens).

chapter 2

New Evidence for the Construction Phases of the Parthenon Peristyle: Anomalies at the Southwest Corner Vasileia Manidaki to J.J. Coulton

∵ 1

Irregularities and Anomalies in the Structure of the Southwest Corner1

The southwest corner of the Parthenon is preserved in a reasonably good condition. The recent 2011–2015 restoration by the Akropolis Restoration Service (YSMA) where all the elements from the akroterion base to the capitals were dismantled provided an opportunity to investigate an area of the Parthenon which was largely undisturbed

since antiquity (Fig. 2.1) (Manidaki 2018). The examination of the extent and nature of the 1902 restoration by Nikolaos Balanos, which was documented in a brief and unspecific manner (Balanos 1938: 69), found that the previous dismantling has been limited to certain blocks at the corner of the pediment: the akroterion base, the sima, and two blocks of the raking geison. During the recent investigation, hitherto unknown details of the original construction came to light which reveal deviations from normal construction practice (Fig. 2.2). The anomalies are as follows: in the northwest entablature are an incision in the south frieze backer (a1) and an unusually reduced fastening of the south architrave (a2). On the top of the corner geison and the cavity for its weight reduction are: an unused bedding surface (a3), two partly

figure 2.1 Parthenon southwest corner under restoration, November 2012 Photo S. Gesafidis, YSMA archive 1  I would like to give my warm thanks to David Scahill and Philip Sapirstein for the invitation to participate in such a successful and fruitful scientific meeting, as well as for the meticulous editing of the manuscript, and to the American school in Athens for their kind hospitality. I am grateful to all members of the Akropolis Restoration Committee (ESMA) and to the YSMA director, Vasiliki Eleftheriou © Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004416659_004

  for allowing me to convey this research. I am indebted to Professor Manolis Korres for providing expertise while delving into the building techniques, to Dr. Andreas Lambropoulos and Dr. Alexander Herda for their pertinent observations on this paper, and to Professor J.J. Coulton for his generous assistance with the translation of the text from Greek and insightful comments on the manuscript.

40

Manidaki

figure 2.2 Axonometric section of the southwest corner with labelled irregularities. The numbers in the figure are prefixed by the letter “a” in the text author

abandoned levelling blocks (a4), and five abandoned bridging sockets (a5). On the end block of the raking geison are two abandoned clamp cuttings (a6), and beneath the akroterion base are five abandoned cover tile dowels (a7). These features attest to disruptions to the regular sequence of work, either omissions of standard processes, or later alterations to an already completed section. In this chapter, I show how they are not the result of mistakes but a result of the organisation of the Parthenon’s construction. 2

A Construction Break between the West and South Sides: Doric Frieze Backers and Geison

An incision at the upper surface of the south frieze backer (a1) is cut to a depth of ca. 3.5 cm (Fig. 2.2:1, 2.3:1) and a

breadth equal to the extent of the corner geison block set on top of it. As a result, there was no contact with the overlying geison, leaving a gap in its bedding. The incision can be explained as a simple constructional device which allowed the south frieze backer to be slid easily into place beneath the corner geison, at first as a trial and later during its final installation. An additional abnormality is the omission of the standard double-T clamp on the top face of the block which would normally connect it to its counterpart on the west side, as visible on its east side. Instead, a rather unusual clamp was used at the junction with the corner: a flat bar set into a lower surface of the block (Fig. 2.3:x, 2.4:x). The presence of this anomalous cutting on top of the south frieze backer (a1) together with the omission of the standard double-T clamp on its west side indicates that the south frieze backer was installed after the corner geison block had already been set in the course above. This sequence would contradict the regular order of construction in continuous horizontal courses from the ground up. In contrast to the expected order of construction, the evidence indicates that the west entablature, including the corner geison, the west frieze backer, and the first south metope—which supported the corner geison block—had been installed before the south entablature (Fig. 2.4). In this case, we may conclude that there was a secondary interlocking with the blocks at the southwest corner which ran down from the level of the geison blocks to the frieze backers and could be considered as a construction break between the west and south sides. Anomalies in the connection of the short to the long sides were also observed at the east end of the temple, but the solutions are different. At the southeast corner, the east end of the adjoining south backer (NDZ1) has a large vertical cutting (Fig. 2.5a, 2.5c). Τhe part of this block that normally would support the corner geison was chiselled off, another measure intended to avoid contact with the geison as well as the need for this block to be slid into the already built east entablature. At the northeast corner, the adjoining north frieze backer (BDZ1) is the only examples of its type divided into two layers. The lower part is a simple rectangular block (BDZ1a), and the upper (BDZ1b) is smaller (ca. 45 cm wide) with only a roughly cut narrow band (ca. 10 cm wide) extending beneath the corner geison (Fig. 2.5b, 2.5d). Obviously, it was easier to slide a small block into place here instead of a backer of full height. The builders devised these three different solutions in response to the irregular building sequence at the corners. The solutions adopted at the west end are similar in both corners: the blocks are subordinated to the form and

New Evidence for the Construction Phases of the Parthenon Peristyle

41

figure 2.3 The incision (1) in the south frieze backer at the west end of its upper face; (x) an unusual clamp Photo T. Souvlakis, YSMA archive

figure 2.4 On the left, the placement of the south frieze backer by sliding beneath the southwest corner geison with the help of the incision (1). On the right, the use of a flat bar (x) placed in a lower layer of the block instead of the typical double T clamp on its upper surface author

rhythm of the construction of this course and, in comparison with the east end, are less experimental and more regular and economical in terms of labour. In contrast, at the east end the corner frieze backers are treated differently from each other. Their interlocking is a result of an ad hoc solution, using comparatively more blocks, each one cut to irregular dimensions that deviate from the form of the other blocks of this course. All in all, the anomalies at the four corners of the building in the connection of the

frieze backers indicate the priority of the installation of the Doric frieze at the fronts over the flanks, and possibly that the east end slightly preceded the west. Another detail that is better known and extensively discussed in the previous literature, and which also supports this conclusion, is the change in the design of the thranos in the peristyle while construction was underway (thranos originates in the ancient architectural term for the blocks which supported the ceiling, sometimes called

42

Manidaki

a

c

b

d

figure 2.5 Interlocking of the frieze backers at the east end. (a,c) Southeast corner, frieze backer with a large vertical cutting (bracket) avoiding contact with the geison above. (b,d) Northeast corner, two-layered frieze backer. The upper block has only a narrower part (circled) which reaches beneath the corner geison a,b: 1991, YSMA archive; c,d: author

the crown moulding of the wall or the entablature: Korres 1994: 92; Barletta 2009: 548). At the Parthenon the interlocking of the thranos in the structure was formed differently at the two faces of the peristyle. On the inner face of the Doric entablature, the thranos is composed of narrow blocks placed along a “ledge” on the upper part of the frieze backers, while on the outer face of the cella it is comprised of regular blocks in a continuous course on the top of the Ionic frieze. Manolis Korres (1994: 92–108) demonstrated a significant change during construction in the thranos at the inner face of the Doric entablature in his article on the design of the Parthenon, which presented new observations about the east entablature following the

restoration works of 1984–1991. This discovery was based on various anomalies, the most characteristic of which are abandoned clamp cuttings and the gaping anathyrosis of the east frieze backers. After the removal of the geison, the clamps which originally connected the frieze backers to the packing stones within the entablature were found abandoned with their west end cut short (for illustrations, see: Korres 1994: 97, fig. 23; Wesenberg 2016: 155, fig. 3). In addition, after the dismantling of the Doric frieze, the eastern frieze backers were also found to be lacking the expected band of anathyrosis below and behind the thranos, whereas the anathyrosis along the long sides ran regularly around the joint faces.

New Evidence for the Construction Phases of the Parthenon Peristyle

Korres concluded that the ancient builders had decided to create a “ledge” for receiving a new thranos which had not been part of the original plan. For this, a rectangular section of stone was cut out along the east edge of the top of the frieze backers after these blocks were already in place and connected to one another. As a result, the masons chiselled the upper part of the blocks while inadvertently removing much of their clamps. However, because the frieze backers along the flanks were not yet in place, the thranos ledge there could instead be planned from the beginning. The design changed occurred after the backers of the east side were clamped in position, and before the installation of the equivalent blocks of the flanks, whose joint faces preserve their clamps and the full band of anathyrosis around every side. In 2002, Rosalia Christodoulopoulou (2002: 144) published a similar observation at the west end relating particularly to the truncation of the clamp cuttings of the frieze backers in the middle of the west entablature, extending Korres’ interpretation of these irregularities in the west frieze backers. At that time, two truncated clamps (one transverse and one longitudinal) were visible at the spot above the sixth intercolumniation where the thranos blocks were missing; the remains of transverse clamps in some other joints were partly detectable through the small gap between the thranos and geison. These observations showed that, as with the east side, the cutting for the thranos had not been anticipated on the west side, although there was no clear evidence for how far the building construction had progressed by the time of the change in design. Christodoulopoulou (2002: 144) suggested that, at the time of the design change, the geison blocks had not yet been laid on the west side, based on the assumption that the iron clamps had not been set in the clamp cuttings prepared in the course beneath. However, this area was not accessible, and it was only possible to see the exposed east edge of the cuttings. Nevertheless, a drawing of the west entablature made in 1900 but not published until 1938 by Balanos depicts clamps between the frieze backers in the area where interventions took place at that time (Balanos 1938: pl. 6; republished in Orlandos 1977–1978 vol. 1: 243). The recent restoration demonstrated that this drawing was partly hypothetical and mistook both the form of the blocks and their connectors. Since there was no distinction between the documented and hypothetical elements, the incorrect indication that there were clamps under the geison misled research into the original structure at the west end. The recent restoration of 2013–2015 shed new light on this issue through the removal of the corner blocks at the

43

west end. It was discovered that beneath the geisa were no clamps for the lengthwise connection among the west frieze backers at the corners. The recent findings from the upper face of the frieze backers indicate that not only the original lengthwise clamps were truncated when the thranos ledge was created, but also that the clamps were not replaced by new ones, even though there was enough space for them on the remaining upper surface of the frieze backers. In contrast, the standard clamps were only installed on the backers in the flanks. The obvious explanation for the omission of the clamps on the west side is that the surface required for their placement was already covered by the blocks of the next course: the geison. The evidence at the west side is much clearer than on the east side, where the frieze backers are thinner, and their upper surface is not wide enough for the lengthwise clamps. Korres (1994: 95, fig. 23, 25) argued that the geison had previously been placed because the geison dowels were positioned at the west edge of the thranos cutting. However, this placement of the dowels appears irregular mainly due to a failure of the marble caused by the inadequate thickness of the east frieze backers (just 12–14 cm), which left only 4–6 cm from the edge of the block for the dowel cutting. As a result, the appearance of dowels in this case is not a secure indication that the thranos cutting was made after the geison had been placed. Only in one case (the second dowel in the first frieze backer) was a dowel socket found exactly at the inner edge of the thranos ledge—a suggestion that the geison had already been set. This observation of the thin top of the frieze backers at the east side was verified on the much thicker backers of the west side. In addition, by dismantling the frieze backers at the two corners of the west side, it was possible to confirm that, in contrast to the adjoining backer of the west side, the westernmost frieze backer on the flanks preserves a complete band of anathyrosis (Fig. 2.6). Similar to the eastern side (described above), the western backer lacks its bands of anathyrosis below and behind the thranos. The original contact band with anathyrosis must have been the result of a later truncation when its “thranos ledge” was cut out. Their discrepancies not only confirm a revision of the design at the west end but also suggest the time and location for this revision: exactly at the construction break at the southwest corner, as was also the case in the northwest corner and the two corners on the east side. In sum, when the design revision occurred, the construction had reached a stage when the geison was in place and the entire entablature completed on both fronts of the building, whereas on the flanks, the frieze backers

44

Manidaki

figure 2.6 Anathyrosis of the frieze backers at the southwest corner. Left: regular anathyrosis band at the south end of south NDZ45. Right: truncated anathyrosis band at the west end of the west DDZ18 Photo T. Souvlakis, YSMA archive

had not yet been installed, and their joint faces had not even been shaped. This is an important conclusion for understanding the building procedure of the Parthenon—a rare testimony to a design change during the construction detected at the two opposite sides of the temple—and has provided clear evidence that the Doric entablature of the Parthenon was completed on both facades before the flanks. 3

A Construction Break between the West and South Sides: Doric Frieze and Architrave

The southernmost west frieze backer (DDZ18), which is 3.28 m long, extends into the south frieze area to serve as a backer for the metope of the south side (Fig. 2.2, 2.7–2.8), where it would have strengthened the bonding of the blocks and ensured the stability of the corner geison. Similarly, the other backers at the inner corners at the north side are exceptionally long, extending into the

structure of the transversal frieze. At the northeast corner, the northernmost frieze backer (3.44 m long) is also exceptionally wide on the part where it extends into the entablature. The block at its hidden part (its north side) is 24 cm wider than elsewhere, which is ostensibly an inefficient use of material and labour. However, the additional width increased its bearing surface, improving the seating for the corner geison and allowing that block to be dowelled at its west end (see Korres 1994: fig. 25). It is worth mentioning that the corresponding frieze backer at the opposite southeast corner is shorter, and the backing of the south metope consisted of other blocks which provided extra width for supporting the corner geison above (Korres & Bouras 1983:80). All in all, the use of exceptionally long frieze backers in the three corners as well as the special form of the south metope-backer at the southeast corner may be interpreted within the framework of the construction stages: they are reinforcements that allowed the independent erection and completion of the fronts (frieze and geison) before

New Evidence for the Construction Phases of the Parthenon Peristyle

45

figure 2.7 Construction break through the whole height of the entablature, from geison to epistyle (in white), with the long frieze backer DDZ18 indicated by a bracket author

figure 2.8 The structure of the Doric frieze of the southwest corner. The southernmost west frieze backer DDZ18 is in contact with the rear face of the first south metope Photogrammetric documentation, YSMA archive

46

Manidaki

figure 2.9 Southwest corner, the reduced fastening of the south epistyle. The actual clamps are shown in black, while white indicates the expected positions of missing clamps author

the installation of the frieze on the flanks (with the exception, of course, of the first metope). The discovery of the stages of the building process of the entablature concerns another issue related to the south metopes. At the time when the design change was made, the first metope of the south side must have already been incorporated into the southwest corner together with all the west frieze blocks and its backers, the whole being connected with clamps and partly supporting the corner geison (Fig. 2.7). None of these members have ever been restored or replaced: the state of the clamps and dowels confirms that the first south metope was already set in place and has remained undisturbed up to the present (more specifically, the clamp connecting the metope with the corner triglyph and the dowels of the corner geison were found in their original positions). This fact, identified from the absence of any scars on the marble left by an attempted removal, excludes any possibility of a replacement metope here. Consequently, we must conclude that the installation of the series of Kentauromakhia metopes on the south side had started before the decision to change the design of the thranos. This observation bears on an old debate over the relatively early style of the south metopes, which has led some scholars to argue that these metopes were initially prepared for the Parthenon porches (Wesenberg 1983: 60–69; Korres 1994: 68; Barletta 2005: 76; Barletta 2009). Korres

(1994: 96) attributes the design change of the thranos in the pteron to the introduction of the Ionic frieze around the cella instead of an original plan for a Doric frieze with triglyphs and metopes. He based this interpretation on the fact that the form of the revised thranos incorporates the mouldings of a thranos usually crowned by the Ionic frieze. However, even if the design change was due to the introduction of an Ionic frieze over the cella, which would have replaced a Doric one in the porches initially planned to incorporate metopes, these could not have included the Kentauromakhia metopes. Had there been such a change from a Doric to an Ionic frieze, the south metopes with the Kentauromakhia must have been installed beforehand. The construction stages of the entablature of the southwest corner may also be connected with an anomaly in the course beneath (a2): the reduction in fastenings of the south architrave (Fig. 2.2:2). The three blocks forming the westernmost south architrave were laid without the usual connections to the architrave of the west side and to the underlying capital. All three dowels were omitted together with two of the three longitudinal clamps and all the transverse clamps (Fig. 2.9). According to the research so far, all architrave beams of the Parthenon peristyle are dowelled and connected transversely and longitudinally, and there are no comparable omission of the connectors known elsewhere in the building. The construction process at the southwest corner may explain this important

New Evidence for the Construction Phases of the Parthenon Peristyle

omission of the fastenings of the south architrave, which can be attributed to a hasty completion of the west side. The builders had to proceed to the next course on the top, placing the corner triglyph and the frieze backer at the corner, and covering the area on the top of the architrave where the clamps would normally have been installed. In this interpretation, the lack of fastenings on these architrave blocks and the unusual form of the southernmost west frieze backer are additional evidence for a construction break at this corner, extended over three courses— epistyle, frieze, and geison—comprising the whole height of the entablature (Fig. 2.7). 4

The Replacement of the Corner Sima and the Change of the Tiling Design

In the last raking geison block of the south wing of the west pediment are two T-shaped cuttings for clamps (a6) which were abandoned and chiselled away in antiquity (Fig. 2.2:6, 2.10). Not only do these clamp cuttings lack a continuation into the two adjoining blocks, but the eastern one is located very close to the joint between the adjoining blocks. The abandoned cuttings can only be explained by an earlier phase where a different block had been clamped in this position and later replaced by the present members which are of different lengths. In order

figure 2.10

47

to determine the dating of the abandoned clamp cuttings, and whether the replacement was due to a change of plan or a repair, I consider this together with three other anomalies in the lower course. First, on the top of the corner geison we observed an unused bedding surface (a3) that is clearly framed by an unworked marble surface (Fig. 2.2:3, 2.11:3). A second anomaly was detected inside the cavity which was cut into the geison in order to reduce its weight (Penrose 1851: 46, pl. 17, below left; Orlandos 1977–1978: 1:246; Korres & Bouras 1983: 81–82). Two support blocks (a4) were inserted into the cavity, secured in place by sealing the joints with lead, and carefully trimmed down to the upper level of the cornice block (Fig. 2.2:4, 2.11:4). A third stone (Fig. 2.11: x) inserted between them has a sloping surface which breaks the continuity of the level and prevents its use as a bedding for an overlying block. Finally, there are five sockets (a5) cut through the cavity and the adjoining filling stones (Fig. 2.11:5). These small cuttings, ca. 10 cm wide and 4–6 cm deep, resemble dowels found elsewhere in the building which have been interpreted as sockets for iron bars, temporarily bridging a gap in the construction to allow blocks on rollers to pass over them (Korres & Bouras 1983: 105). However, in this instance the sockets are much larger, and their position at the edge of the geison is unsuitable for such a function. Their interpretation as bridging sockets is complicated by the fact that they are arranged

Two abandoned clamp cuttings (circled) in the last raking cornice block on the west pediment Photogrammetric documentation, YSMA archive

48

figure 2.11

Manidaki

Southwest corner geison. On the upper surface is the tooled unused bedding surface and to the right (3), the unworked apergon, with an arrow indicating their boundary. Inside the south weight-reduction cavity are (4) two partly abandoned levelling stones, and (5) five bridging sockets. The position of the sixth socket, which was never cut, is marked “?”; nearby is a masons’ mark of Σ. The numbers in the figure are prefixed by the letter “a” in the text Photogrammetric documentation, YSMA archive

along an east-west axis, indicating a direction of rolling towards the east. Thus, their relative position would not make a significant contribution to the breadth of the rolling surface. Together, these features (a3–6) represent a unique testimony to a previously unknown intermediate stage in the construction at the corners of the Parthenon. Indeed, the southwest geison block is the only one of the four corner blocks where the stones fitted inside the weight-reduction cavity are undisturbed, preserving evidence of this otherwise undetected revision of the design. A thorough examination led us to conclude that, in lieu of the corner sima with a lion head waterspout, there had previously been a longer corner sima installed from the time before the eaves pan tile to the east was set in place. The unused bedding surface on the underlying geison (a3) was formed to

receive the original sima block whose length can be restored exactly from the clear boundary between the tooled and unworked surfaces (Fig. 2.11: arrow). We arrived at this conclusion through consideration of the standard building process where the unworked, rough, protective mantle of the upper surface of each block, an apergon (άπεργον) ca. 2–3 cm thick, was removed only during the setting of the course above. As a result, the border between the dressed and rough areas reveals the placement and the dimensions of the block above. At the upper surface of the corner geison, the east end of the protective mantle falls exactly at the centre of the first regular palmette antefix, 230.5 cm from the west corner (Fig. 2.12b). The aforementioned pair of double T-clamps (a6) cut into the adjacent raking geison block and later abandoned would have originally fastened to the raking sima block (Fig. 2.12a). The

New Evidence for the Construction Phases of the Parthenon Peristyle

a

figure 2.12

49

b

The (a) original sima compared to the (b) actual sima. The east end of the original corner sima coincided with the east limit of the bedding surface on the top of the corner geison and the position of the actual first palmette antefix author

original sima must have been bedded on the two levelling stones (a4) inserted into the weight-reduction cavity of the corner geison, where it would have been prevented from slipping outwards by means of projecting bosses or perhaps metal rods on the underside of the sima anchored in the five sockets in the geison below (Fig. 2.11:5). Similar bosses found on the underside of the marble tiles of the Parthenon would have secured them to the wooden battens (Orlandos 1949; 1977–1978: 2:600–608). At some point the plan was changed, and the original, larger sima at the corner was removed. When did the changes occur, and why? It is clear that the change and the installation of the final corner sima with the lion head preceded the laying of the tiles, because the corner tile must always be the first set (Fig. 2.12b). Next, all the remaining eaves tiles of the course would have been laid, and likewise the next course running up the roof beginning with the sima blocks. The roof construction follows the positions established by these “tile borders” at the eaves and sima. Since there are no alterations evident among the adjacent roof tiles, we must be witnessing another design change rather than a later repair. Accordingly, the replacement of the sima can be securely dated between two construction phases: after the pediment was completed, and before the installation of the tiles. This design change reveals that the break in the construction of the entablature at the southwest corner (as demonstrated in the previous section) continued all the way up to the level of the eaves

tiles, allowing the entire west facade including its pediment to be erected independent of the flanks or the cella. The new corner sima block replaced the original after the hasty truncation of its two clamps, omitting any new connection to the adjacent raking geison. On the west face of the sima, the cyma reversa was carved to match the moulding of the raking geison, but not as a typical straight moulding. Its height was reduced, creating a break a short distance from the end, while preserving its original width (Fig. 2.13). The moulding is actually slightly twisted so that its profile remains identical even as its inclination increases towards the corner (Eleftheriou, Manidaki & Vrouva 2015: 51). To my knowledge, this is the only case in a straight block where the axis of a linear moulding has been twisted. Even though the actual geometry of the moulding is more complex, I suspect that its curvature was generated through a relatively simple procedure guided by a stencil with the standard sima moulding rotated around the longitudinal axis of the sima. The replacement of the corner sima block provides the first indication of a much larger change in the design of the roof tiles. The original scheme can be deduced from the different length of the corner sima and the normal sima tiles, whose east edges were not aligned with each other but rather differed by 102 cm. Consequently, the axial width of the first row of tiles must have been 102 cm, which corresponds to the angle contraction at the ends of the long sides in contrast to the typical axial spacing of ca.

50

figure 2.13

Manidaki

The southwest corner sima and cyma reversa moulding. The dotted line indicates a straight projection along the bearing of the cornice Photo T. Souvlakis, YSMA archive

107.5 cm. The original, unrealised tiling scheme was comprised of fewer larger tiles spaced at a rhythm of four pan tiles per intercolumniation. At the eaves, the line of palmette antefixes was designed to correspond to the spacing of the rows of cover tiles above (Fig. 2.14a). For a reconstruction of the width of the unrealised tiles, we should take into consideration the width of the gap in between the tiles, since the axial spacing is not identical with the width of the tiles. The actual axial spacing of the regular tiles of the Parthenon is ca. 71.5 cm (Penrose 1851: pl. 17) while the pantiles now displayed to the south of the Parthenon have a width of 61.5 cm. This would leave a gap of ca. 10 cm between pantile rows, which is unusually wide, but possible with a cover tile hollow of ca. 16.5 cm. Thus, according to these calculations of the intervening gap, the intended pantile width would have been around 97 cm, planned at four rows per intercolumniation. However, instead of four tiles, the revised tiling scheme that was actually implemented incorporated six tiles per intercolumniation. Nevertheless, the previous rhythm of the palmette antefixes on the eaves pantiles was retained in the new scheme. Of course, this required the wellknown, eccentric insertion of false palmette antefixes with no rows of cover tiles behind them, and the alternation of two narrow cover tiles with a wider one in the row aligned with the antefixes (Fig. 2.14b). In the tiling of the Parthenon, the only antefixes coinciding with vertical

rows of cover tiles were those above the triglyphs, whereas false palmette antefixes are above the metopes, combined with a false cover tile extending back only for the small distance needed to seat them (Orlandos 1949; 1977–1978: 2:604–06). The change in the tiling scheme might have been motivated by technical or aesthetic reasons but, since the original wider spacing of the antefixes was retained in the eaves pantiles, it seems more likely that the change was expedient. For instance, shorter regular pan tiles would weigh less, be more easily handled by two workmen during manufacture and setting, and allow for thinner wooden rafters (Fig. 2.14c). When the change in the tiling scheme was decided, the corner sima block, which had already been laid, needed to be modified. This does not seem to have been a problem, since the larger original sima block was simply replaced by a smaller one. However, the fact that the old block was not cut down to suit the requirements of the new scheme, but instead replaced with an entirely new block, does suggest that the form of the sima had been substantially altered. Perhaps this replacement was necessitated by a decision to add a projecting mass to the old block—i.e., the lion head at each corner of the roof. Furthermore, the change to the tiling scheme can only have been decided after the pediment was already complete, and consequently the order for tile production must have been determined at this stage in the work. Was there

New Evidence for the Construction Phases of the Parthenon Peristyle

a

b

c

d

figure 2.14

51

The original tiling scheme (left) as planned, and (right) as executed. The ceiling has been omitted from the diagrams for legibility author

time, however, for the production of 8,957 tiles (Korres & Bouras 1983: 40) after the pediments were already complete? (The following discussion does not take into account the six akroterion bases, which were the subject of a different building stage.) The roof, of course, was the last part of the structure to be installed. Because all the woodwork for the roof must have been completed before the tiles could be laid, its period of installation would also provide time for manufacturing tiles. Although as mentioned before we cannot fully restore this construction phase, the eave tiles had not yet been laid when the decision was made to replace the corner sima. Since the rafters of the roof had to be set first in alignment with the eaves tiles, we may suppose that the

woodwork of the roof (and with it the wooden ceiling of the cella) had not yet begun. According to the Parthenon accounts, timber was purchased in the year 441/40 BCE, probably for the roof (Dinsmoor 1913: 78). Work in wood—ἔργοις ξύλινοις—was carried out in the following year 440/39 BCE. The construction of the roof and ceiling would have been completed in 440/39 or the latest by the year 439/8, so that the cella would be fully protected for the erection of the cult statue, dedicated in 438/7. If this interpretation of the accounts is correct, then the decision for a change in the tiling was made no later than 441/0 BCE, and within two years (439/8) the tiles must have been in place. At this stage, most of the marble-masons would have been available for

52 hire and so enough craftsmen would have been available to work on the tiles. Their last big project was probably the marble ceiling coffers (214 blocks), whose installation should have been completed before the construction of the wooden roof (Stanier 1953). Two years would have been sufficient for producing the tiles. Unlike terracotta tiles, there are no inscriptions, to my knowledge, specifying the cost of marble tiles. Assuming five man-days for each pantile and three man-days for each cover tile, the carving of the entire number of tiles might be completed by 60 artisans in two years—an estimate based on the experience of the marble masons working on the Akropolis restorations. As a result, it is reasonable to assume that the change of the corner sima and the revision of the tiling scheme could have taken place without causing delays to the overall work progress.

Manidaki

Additional cuttings point to another revision from the very end of the tile installation. Beneath the base of the corner akroterion, to either side of the joint between the corner sima block and the first eaves pantile, are five dowel holes for fixing a cover tile in position (Fig. 2.15). Evidently the removal of the five dowels and the cover tile attests to yet another design change that occurred after the completion of the tiling: the addition of a new element, the akroterion. The same thing occurred at the east side of the temple, as shown by Korres (1995: 18). This intervention on the installed tiles was carried out with surgical precision to allow the safe fitting and fastening of an exceptionally large akroterion base to support the substantial projection of the Nike figure. The arrangements for its setting and securing were made ad hoc and thus apparently had not been envisaged from the start.

figure 2.15 Modification for the fitting of the southwest akroterion base (white). Underneath this base are five abandoned dowels holes for a cover tile author

New Evidence for the Construction Phases of the Parthenon Peristyle

5

Conclusions: Revisions of Design and Organisation of Work in Subprojects

The anomalies documented during dismantling the southwestern corner of the geison and entablature of the Parthenon reveal changes to the design during the course of work that illuminate the whole organisation of the construction. First, in the entablature, the special setting of the first south frieze backer and the first architrave on the south indicate a construction break between the western and the southern entablature. Furthermore, comparison of the special form of all frieze backers at the edges of the flanks at the four inner corners of the building has revealed that the east and west facades were built independently and preceded the installation of the flanks. Second, the newly observed anomalies in the area of the geison revealed a previously unknown replacement of the corner sima block and a significant change in the tiling scheme during the course of construction, which took place immediately after the completion of the pediments. Third, the examination of the tiling as executed also points to additional changes during the setting of the corner akroteria bases on both sides. Finally, if we add the later cutting of the thranos bedding in the frieze backers of the east and west sides, we have four areas whose completion evidently involved substantial ad hoc changes during construction. As has been shown above, these alterations were the result of fitting and interlocking new elements with the courses of already laid blocks, motivated either by design revisions or adapting to the existing masonry. A longstanding question for research on Greek architecture is the role of the architect and whether a comprehensive plan existed from the outset where every detail was foreseen in advance. This paper has examined evidence that is better understood in light of Coulton’s theory of how ancient Greek architects actually worked. Coulton (1977: 17, 51–73) describes the experimental character of architectural practice and argues that a “detailed design would take place as the building went up, in light of the established conventions of architecture, the particular problems that emerged, and the architect’s own decisions”. On the issue of the Parthenon’s irregularities, Barletta (2005: 82) criticises the assumption of a thorough revamping during the construction, preferring to see them as “simply afterthoughts on the part of the architects”. In each case, the features that she discusses attest to an earlier phase of construction and furthermore to pauses in the work after large building units such as the entablature had been completed. Assuming that these anomalies were not random, these observations encourage us to consider the perspective of construction organisation—i.e., from the

53

viewpoint of the work groups undertaking subprojects within the whole project. How might we determine the organisation of subprojects? First of all, the separate treatment of the facades of the building is absolutely justified by the general programme of the work. Examining the stages from start to finish, we see that a precondition for the start of tiling was the completion of the pediments, including the eaves pan tiles. However, the short sides require several additional construction layers than do the flanks: the orthostates, the backing walls of the tympanon, and the raking geisa. The separate construction of the fronts before that of the flanks makes sense because the different teams of builders could better coordinate the work, adhering to a stricter time schedule, with the obvious benefit that the whole project could be completed faster. Another explanation of work divided into units is that the breaks in construction are the boundaries between separate building phases. In this view, it would be convenient to organise the construction into subphases, since they could then be allocated to different teams of workers. This hypothesis is further supported by the observation of differences and variations among similar parts of the Parthenon peristyle: for instance, in the backing walls of the pediment and the filler blocks in the entablature (Korres & Bouras 1983:80). A representative example in our case is the difference in building methods used at the corners of the frieze-backer course. Since the construction of both short sides would have been carried out in parallel, it is more probable that the variation in building methods was due to the independent choices of the two or more separate workgroups simultaneously engaged at opposite ends of the building (see Coulton 1977: 57, 59; 1983 for additional discussion). Following this line of analysis, we might imagine that the reduction in fastenings at the south architrave was due to haste in finishing the west entablature. After that, in the next subphase, we may include the introduction of the south frieze backer with the necessary top cutting, and the delayed cutting of the thranos ledge. Therefore, the corner sima block must necessarily have been laid in the west pediment subphase, even if all the details concerning the exact form of the tiling had not yet been decided, because without it the pediment would have been incomplete. Accordingly, in the next stage when the important decision on the dimensions of the tiling was finally made, the sima block had to be replaced with a new one, which therefore belonged to the following subphase of roofing. In this view, the subprojects would constitute separate undertakings and might have been described in separate contracts, allocated to different work groups, and allotted

54 different completion dates. There is no clear evidence, however, for the normal practice in building projects of the fifth century BCE. Davis (1937: 111–12) has argued that in the mid-fifth century a direct relation between state and individual was more common, and that the mode of execution changed gradually into the fourth century, from daily wages (μισθώματα) to work under contracts (εργώνας, μισθωσάμενος τὸ ἔργον). We cannot reach any firm conclusions from the very fragmentary remnants of the Parthenon accounts, and piecework contracts are recorded together with daily wages in the Erektheion accounts (Randall 1953: 210). Thus, we should not exclude the possibility that both methods of assignment were used for the Parthenon (Hellmann 1999: nos. 7–16 and Feyel 2006: esp. 469–510; 2007 for a general discussion of building contracts). Furthermore, in a very tight timetable like that of the Parthenon, contracts might have been more suitable for managing costs and time. Nevertheless, separate crews working simultaneously on the building could implement different choices, according to their judgment of each case, albeit subject to certain agreed-upon limits and the approval of the master architect. The master architect would have been responsible for drawing up the contracts with a detailed description of each job. In due time as planning moved to execution, he would re-examine the state of the building, the possibilities for quarrying, the availability of workmen, and the timetable—designing accordingly the details for the following phase. Only then would the need for revisions become apparent. It is interesting to observe here that such a design and execution procedure is well known in architectural practice through the ages and up through the present. In current practice, the contractor must submit a revised set of drawings known as the “as-built drawings” after the completion of the building, reflecting all changes made in the specifications and working drawings during the construction process. Revisions would be carried out spatially at a junction between two constructional units, and chronologically between two subphases, when responsibility passed from one work group to another. The anomalies in the construction of the Parthenon are not just significant evidence for building practice but also fuel for research into details of design procedure, the execution of alterations, and the role of the architect and work groups. Of course, these all constitute an exceptional challenge for architectural research and will be subjects for future investigations.

Manidaki

List of References Balanos, Ν., 1938: Les monuments de l’acropole : prélèvement et conservation (Paris). Barletta, B., 2005: “The Architecture and Architects of the Classical Parthenon”, in Neils, J. (ed.), The Parthenon: From Antiquity to the Present (Cambridge) 67–100. Barletta, B., 2009: “In Defense of the Ionic Frieze of the Parthenon”, AJA 113.4: 547–88. Christodoulopoulou, R., 2002: “Το Δυτικό Αέτωμα του Παρθενώνος σε Σχέση με το Πρόγραμμα Αποκαταστάσεως του Οπισθονάου”, in Mallouchou-Tufano, F. (ed.), Proceedings of the Fifth International Meeting for the Restoration of the Acropolis Monuments, October 2002 (Athens) 137–48. Coulton, J.J., 1977: Ancient Greek Architects at Work: Problems of Structure and Design (Ithaca). Coulton, J.J., 1983: “Greek Architects and the Transmission of Design”, in Ecole française de Rome (ed.), Architecture et société. De l’archaïsme grec à la fin de la république. Actes du colloque international organisé par le Centre national de la recherche scientifique et l’École française de Rome (2–4 décembre 1980) (Rome) 453–70. Davis, P., 1937: “The Delian Building Contracts”, BCH 61: 109–35. Dinsmoor, W.B., 1913: “Attic Building Accounts I: The Parthenon”, AJA 17.1: 53–58. Eleftheriou, V., Manidaki, V. & Vrouva, A., 2015: Μελέτη Αποκατάστασης της Δυτικής Πλευράς του Παρθενώνα, Γενικός Προγραμματισμός του Έργου & Προτάσεις Επέμβασης της Δύο Γωνίες του Θριγκού. Μελέτη Αποκαταστάσεως Παρθενώνος 8 (Athens). Feyel, C., 2006: Les artisans dans les sanctuaires grecs aux époques classique et hellénistique à travers la documentation financière en grèce. BEFAR 318 (Athens). Feyel, C., 2007: “Le monde du travail à travers les comptes de construction des grandes sanctuaires grecs”, in Brun, P. (ed.), Economies et sociétés en Grèce classique et hellénistique. Actes du colloque de la SOPHAU, Bordeaux, 30–31 mars 2007 (Toulouse) 77–92. Hellmann, M.-C., 1999: Choix d’inscriptions architecturales grecques, traduites et commentées (Lyon). Korres, Μ., 1994: “Der Plan des Parthenon”, AM 109: 53–120. Korres, Μ., 1995: “Νέες Διαπιστώσεις στην Ανατολική Πλευρά του Παρθενώνος”, in Mallouchou-Tufano, F. (ed.), Proceedings of the Fourth International Meeting for the Restoration of the Acropolis Monuments, Athens, May 1994 (Athens) 17–24. Korres, Μ. & Bouras, C., 1983: Μελέτη Αποκαταστάσεως του Παρθενώνος 1 (Athens).

New Evidence for the Construction Phases of the Parthenon Peristyle Manidaki, V., 2018: “Η οικοδομική σύνθεση της νοτιοδυτικής γωνίας του Παρθενώνα—Μέρος Ι. Η οικοδόμηση του οριζόντιου γείσου”, in the proceedings of Διημερίδα ΕΣΜΑ-ΥΣΜΑ, Eιδικά Θέματα Έρευνας και Εφαρμογών στα Αναστηλωτικά Έργα της Aκρόπολης την Περίοδο 2010–2015, Αθήνα 18–19 Νοεμβρίου 2016 (Athens) 57–74. Orlandos, Α., 1949: “Notes on the Roof Tiles of the Parthenon”, in Commemorative Studies in Honor of Theodore Leslie Shear. Hesperia Suppl. 8 (Princeton) 259–67. Orlandos, Α., 1977–1978: Η Αρχιτεκτονική του Παρθενώνος. 2 vols. (Athens).

55

Penrose, F.C., 1851: An Investigation of the Principles of Athenian Architecture (London). Randall, R.H., Jr., 1953: “The Erechtheum Workmen”, AJA 57: 199–210. Stanier, R.S., 1953: “The Cost of the Parthenon”, JHS 73: 68–76. Wesenberg, B., 1983: “Parthenongebälk und Südmetopenpro­ blem”, JdI 98: 57–86. Wesenberg, B., 2016: “Parthenonfries und Pterondecke”, in Zampas, Κ., Lambrinoudakis, V., Simantori-Bournia, Ε. & Ohnesorg, A. (edd.), Αρχιτέκτων, Honorary volume for Professor Manolis Korres (Athens) 153–162.

chapter 3

Ancient Blueprints: New Prospects and Interpretations in Light of Recent Discoveries Jeanne Capelle 1 Introduction The use of blueprints in construction yards is a key to our understanding of Greek architecture.1 It allows us to dispel misconceptions of the building process. The French mathematician Javary epitomizes such prejudices when he wrote in 1894 his question about the origin of “blueprints” (“épures” in French) in the mathematical journal l’Intermédiaire des Mathématiciens: When can we say that we began to make blueprints, that is to say, geometric drawings that one tries to make as precise as possible? It seems that the Greeks did not create any blueprints; besides, it would have been very difficult for them; and their geometry, which was handled in a highly speculative manner, did not lead them in that direction. They must have drawn figures in the sand, on tablets, etc. to guide their reasoning and set down their ideas, but those were not blueprints. According to scholars who have studied these questions, M. Choisy for instance, they did not make any blueprint even to raise buildings; at least their strict methods of calculation seem to aim at 1  This article was made possible through the permission and hospitality of the Turkish Ministry of Culture and Tourism as well as of the Greek Ephor of Antiquities Stavroula Sdrolia and the archaeologist Athanasios Tziafalias. I benefited from the generous support of Lyon 2 University (IRAA, HiSoMA, ED 483). I owe special thanks to the directors of the Miletos Archaeological Project, Christof Berns (Bochum University) and Philipp Niewöhner (DAI) for their encouragement throughout my investigations of the blueprints as well as to Alexander Herda for his guidance at the beginning of my research in 2014. I am grateful to Axel Filges (Frankfurt University) and Hakan I. Mert who kindly organized our stay at Priene and to Wolf Koenigs for his precious advice. I would also like to thank Richard Bouchon and Bruno Helly (Lyon 2) for their warm welcome at Larisa. I am greatly indebted to my husband, Ivan Boyer, who joined me to document the blueprints in 2015 and 2016. Lastly, I would like to express my gratitude to the editors of this book, Philip Sapirstein and David Scahill, for having organized such a stimulating event. Figures and translations are my own unless otherwise noted. More colour pictures are available on the videocast of this conference (http://www .ascsa.edu.gr/index.php/News/newsDetails/videocast-ancient-blue prints-in-light-of-recent-discoveries-the-theater-at). © Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004416659_005

avoiding the necessity for blueprints (see M. Auguste Choisy’s studies in Greek architecture: the Arsenal of Piraeus, the Erechtheion, etc.) Who can be said to be the first one to have cared about the accuracy of a geometric figure? Or at least when did people begin to care about it? Javary 1894

This conviction that blueprints cannot have been used in Greek architecture could already have been refuted and would be denied by the four responses to Javary’s question in several issues of the same journal. In reality, Egyptian blueprints were known since the French campaign of Egypt (Jomard et al. 1821: vol. 4, 294–297, pl. 62, figs. 3–5). Tannery (1894) objects to Javary that blueprints should have been necessary for sundials, and Bouriant (1895; 1896) quoted some Egyptian examples used for stonecutting. Choisy (1894) claimed that blueprints were necessary for carpentry, as opposed to masonry, but he changed his mind when he published two Egyptian blueprints from Edfu (Choisy 1904), a couple of years after Borchardt’s first synthesis on the subject (Borchardt 1896). Still, Javary’s mindset must be put into context: he wrote at a time when almost nothing was known about ancient blueprints, although technical drawings promoted by descriptive geometry were still widely used for stone-cutting, and also when mathematics could at last give an answer to ancient theoretical problems like squaring the circle, trisecting the angle, or duplicating the cube. Javary could not believe in ancient blueprints because of his idealistic vision of antiquity, of ancient geometry as speculative and abstract, and of ancient architecture as highly conceptual and fully predictable. The example of Javary reminds us of the danger of retaining paradigms disconnected from reality. In this paper, a comprehensive answer to his question will be developed based on up-to-date evidence, looking not only for the origin of blueprints but also questioning what the geometry used in blueprints teaches us about Greek architecture and its design. First, what we may call ancient blueprints will be defined. Then a method for finding and studying them will be proposed. Last, an interpretation of such drawings will be attempted.

ANCIENT BLUEPRINTS: NEW PROSPECTS IN LIGHT OF RECENT DISCOVERIES

2

Ancient Blueprints: A Recent Research Field

The word blueprint has already been used by Lothar Haselberger to refer to the famous construction drawings he discovered at Didyma and Rome in the 1980s and 1990s (esp. Haselberger 1980; 1983; 1994; 1996). This modern name was a colourful way to address readers of Scientific American (Haselberger 1995). Since then I published a general, extensively referenced inventory of drawings (Capelle 2017a), and the current chapter provides a more global approach to the phenomenon. The word blueprint may refer to a “design plan or other technical drawing” in an architectural context. First used in the late 19th century, blueprint was derived “from the original process in which prints were composed of white lines on a blue ground” (adapted from the Oxford Dictionary of English 2016), i.e., the technique of cyanotypia: denoting first the multipliable copy of the architect’s original plan, it came to mean the plan itself. This describes perfectly the ancient findings in which lines appear white on a coloured ground or in few cases the reverse. Indeed, various techniques were used: at the Apollon Temple in Didyma, Haselberger found traces of a red chalk solution used to colour the wall before incising the lines—the blueprint that remains is a faint negative of this process (Didyma: Haselberger 1983: 92 [Didyma]; 1986

57

[reconstructing the process]). Wolf Koenigs (2015) argues that, in the absence of earth or mineral pigments, more transient vegetal ones (chlorophyll) might have been rubbed over the walls before drawing began at the Athena Temple in Priene, as was done in the 1960s at the Cologne cathedral (mentioned in Haselberger 1983: 92 n. 4). At Ephesos conversely, on the back wall of the Basilike Stoa in the Upper Agora, one can still see a thin layer of painting into which the lines (once studied by Wilhelm Alzinger but still unpublished) had been incised (Fig. 3.1). Blueprints could also be etched on thick mortar (Krause 1985 [Terracina, ca. 50 BCE]; Hanoune 1996 [Bulla Regia, after the second century CE; n. 9: Domus Augustea]) or simply painted on white stone, as in Hellenistic tombs of Makedonia (Hoepfner 1984 [Angista]) and Thrakia (Čičikova, Stoyanova & Stoyanov 2012 [Sveshtari]; Manetta, Stoyanova & Luglio 2016 [Ostrusha]) as well as in the Egyptian quarries of Abu Fodah (Jomard et al. 1821; Petrie 1888). In this article and line of research, we will use blueprint as a concept referring to a specific kind of architectural drawing that is always made on monumental architecture, mostly on marble, on a flat surface, either vertical or horizontal, that is, on walls or on paving. Some drawings were made on limestone—not surprisingly on marmoreal, crystalline limestone at the Ptolemaion of Limyra (Stanzl 1999) or the sanctuary of Mars Mullo at Allonnes

figure 3.1 Ephesos, Upper agora, basilike stoa, vertical and oblique lines incised in a coat of paint. The “blueprint” is visible on the surface of the marble author

58

Capelle Allonnes Saint-Boil

Pola

Sveshtari Ostrusha Rome Terracina Capua

Angista Larissa

Athens

Italica

Messene Bulla Regia

Delos

Uthina Thysdrus

Palmyra Bziza Heliopolis

ASIA MINOR

Pergamon

Gerasa

Map Key Sardis

New Kingdom of Egypt Classical Period

Ephesus

Hellenistic/Republican Period Imperial Period Number of blueprints

Didyma

Priene Miletus

>10 6-10 2-5 1

Aphrodisias

Lagina Mylasa

Abu Foda Perge Valley of the Kings

200 km Outline map : d-maps.com

Limyra

Edfu Philae

Syene Meroe

figure 3.2 Chronological map of blueprints author

in France (Brouquier-Reddé & Gruel 2003), but also in the quarries of Abu Fodah (see above) or on travertine at the Amphitheatre of Capua (De Franciscis 1983) and at the Mausoleum of Augustus (Haselberger 1994; 1996)— sandstone at Philai (Borchardt 1896) or on plaster, as stated above. This kind of drawing represents architectural elements or details—never a whole building (a partial exception is the drawing delineating the general shape of a small Meroitic pyramid; see Hinkel 1976; 1980; 1981)— in elevation, plan or section. It differs from setting lines in that it is made on an independent surface selected for drawing. It plays a role in the construction of the building on which it is made—or rarely a nearby one; some monumental exceptions include the blueprints for the Pantheon and possibly other buildings at the Mausoleum of Augustus (see above; Capelle forthcoming) and according to one interpretation a blueprint for the Temple of Artemis inside the Temple of Apollon at Didyma (Weber et al. 2015; Weber 2015). It is carefully realised with drawing tools (sharp drawing point, compass, ruler, set square) to scale, in most cases at a 1:1 scale; only about 20% of the cases are at reduced scale. Incision is the most common technique, but it is not necessary (see above). This

definition excludes setting lines; marks for clamps; axes and guiding lines for stonecutting; preparatory lines for the carving of mouldings; and architectural graffiti which share common features with blueprints but should be distinguished from them. We know today of over 120 drawings meeting this definition from which we can suggest the following chronology shown in Figure 3.2. Although it is hard to give a definite answer to Javary’s question about the origin of blueprints, one may argue that it is definitely not a Greek one. Indeed, among the many ancient Egyptian models and architectural drawings, most of which are reduced plans drawn on small, movable material, there is one that can be considered a blueprint. It dates to the New Kingdom of Egypt—more precisely to the second half of the 12th century BCE—and represents in elevation the elliptic vault of the chamber of the tomb of Ramses VI at the Valley of the Kings, painted at full scale in the passageway of this tomb; the masons who carved the chamber probably took measurements from it (Daressy 1907, with figure). Nevertheless, this case is isolated. If we come back to Greek architecture, there is a noticeable gap during the Archaic period: we know plenty

ANCIENT BLUEPRINTS: NEW PROSPECTS IN LIGHT OF RECENT DISCOVERIES

59

figure 3.3 The “sketch of Pytheos” at the Athena Temple of Priene, ca. 340 BC courtesy W. Koenigs

of setting lines to guide the construction of buildings by delineating their plan on the top of the successive courses, but none of these drawings can be considered a proper blueprint (Petronotis 1968; 1972). One of the drawings at the Old Temple of Aphaia on Aigina has been compared to a blueprint by Ernst-Ludwig Schwandner (1985: 80), but it corresponds to a reused stylobate block with setting lines for a column. As observed by Clemente Marconi (pers. comm.), the absence of evidence for the Archaic period might be explained by an insufficient preservation of the monuments. Likewise, my distribution map (Fig. 3.2) is but an incomplete picture of what was most certainly a widespread phenomenon. The only blueprint that can be dated to classical times is the so-called “sketch of Pytheos” at the temple of Athena at Priene. It represents an elevation of the entablature and the pediment of the naos (Fig. 3.3) and dates to the second half of the fourth century BCE, at the start of the Hellenistic period (Koenigs 2015: vol. 33, no. 464). Blueprints indeed develop during this period, possibly because of the greater variety of buildings and

architectural forms which required more than setting lines. From this time onward, elevations, plans, and sections are employed, sometimes combined for complex objects like columns or capitals, and most at full scale— representing a marked change for the working out of design directly in the field. Blueprints continued to multiply all over the Mediterranean during the Imperial period. In fact, the practice never really ceased from the Middle Ages to the present day, as many European cathedrals show (e.g.: Inglese 2014; Sakarovitch 1998; and Ruiz de la Rosa 1987). A recent, rarely cited case is that of the so-called “masons’ room” of the castle of Blois in France. During the restoration of the castle from the 1880s to the 1920s, the walls of this room were covered with blueprints (Fig. 3.4). Although these drawings serve a slightly different purpose since they are about rebuilding, not building, they share common features with ancient blueprints, like the superimposition of drawings; the similarity of some of the objects represented (arch, entablature, pediment); the presence of a volute from nearby Saint Vincent’s next to a pediment from the castle; and the many frontal views and

60

Capelle

figure 3.4 Blueprints from the “Mason’s room” at the Castle of Blois, France, incised in the mortar or drawn on it with charcoal or red chalk during the restoration of the 1880s–1920s author

61

ANCIENT BLUEPRINTS: NEW PROSPECTS IN LIGHT OF RECENT DISCOVERIES

are not uncommon in many other public buildings as well as in monumental tombs and quarries. Theatres are well represented: at Bulla Regia (Hanoune 1996), possibly Terracina (Krause 1985), Perge (to be published by Frank Rumscheid), Aphrodisias (Hueber 1998; Capelle 2017a), Miletos (Capelle 2017a), Priene, and Larisa (below, part 3). Of course, the ratios shown in the figure may change with new discoveries in other buildings. If we consider now the location of blueprints inside the buildings, the example of the theatre at Miletos shows that they can be situated both in a closed space—as the hyposkenion under the stage—as well as in an open one— like the western retaining wall (below, Fig. 3.8b). It is clear that the latter was drawn as close as possible to the object, rendering the arches constructed higher up on the same wall; in the hyposkenion, one pediment of the first Imperial scaenae frons is drawn on its very foundations, along with the base of the nearby terminal pedestals of the retaining walls (Fig. 3.6a). Other drawings which will be discussed below in section 3 are in a similar position: at the Bouleuterion of Miletos (Fig. 3.6c) a blueprint for the seats is located inside the corridor before the auditorium,

moulding profiles, whereas a few axonometric views contrast with the ancient drawings. We may also remark that the drawings are on a thick layer of mortar; the technique varies, combining incisions, charcoal, and red chalk; and the quality fluctuates. Lastly, there are also numbers, diameter indications, computations, notes, graffiti, hatches, and the like. Without being anachronistic, if we consider the richness of this well-preserved evidence, we may wonder if there was not much more to the ancient drawings than the scanty lines which have partially survived. 3

Methods for Finding and Recording Blueprints

We may argue that the few drawings we know today are but rare survivals of a widespread phenomenon. It is important to develop a method to look for hitherto unknown blueprints and decipher them. Since blueprints are difficult to see, the first problem is determining the right place to look for them. The type of building (Fig. 3.5) is not much help: about half of the blueprints are in temples, on the inner or outside walls of the cella, but they

Blueprint repartition according to building type 70

60

50

40

30

20

religious buildings

spectacle buildings

funerary monuments total

figure 3.5 Known blueprints quantified according to building type author

agora/forum detail

pol itical buildings

domus Augusta

prytanei on

council house

total

stoa

total

chamber tomb

mausol eum/heroon

total

amphitheater

theater

total

sanctuary

temple

total

0

basilica

10

quarries

other

figure 3.6a–c

b

a

(a) Miletos, theatre, location of some of the blueprints inside the hyposkenion; (b) Priene, theatre, location of a blueprint for the front of the proskenion inside the scene building; (c) Miletos, Bouleuterion, location of blueprints (A) AUTHOR (b) after von Gerkan 1921: pl. 9 (c) after Knackfus 1908: fig. 53

c

62 Capelle

ANCIENT BLUEPRINTS: NEW PROSPECTS IN LIGHT OF RECENT DISCOVERIES

63

a

b

figure 3.7 (a) Using RTI with I. Boyer at Miletos, here on a Christian monogram. (b) Specular enhancement, snapshot from RTI-Viewer of a horizontal line with a compass hole in its middle (a) courtesy A. Vacek (b) author

and at the theatre of Priene (Fig. 3.6b) the Doric frieze of the proskenion is drawn in the middle room of the skene. It appears that blueprints may be found in any monumental building next to their subject. There are special spots one can examine closely: a blueprint requires a smooth, easily accessible, and large surface. All these requirements are met by orthostates, which are particularly tall and set at eye level, and indeed a vast majority of blueprints or at least their lower lines are drawn on orthostates. For instance, the western retaining wall of the theatre at Miletos—including its profile (Krauss 1974)—is quite similar to the inner walls of the temple at Didyma where Haselberger found so many blueprints. Once a blueprint has been identified, however, deciphering it is still a difficult matter. Most frequently, one looks at weathered surfaces and needs to be in the right place at the right time. Oblique light from the sun or light reflected by a mirror helps to reveal the lines. However, it is often impossible to have a full view of the drawing

because many lines can only be seen from extremely close up. Some basic principles may help to see more. By putting one’s eye at the end of a line, one can see that it continues farther than when one looks at it frontally. A piece of string can help us find compass holes, but also the continuation of lines and curves which have partially vanished. It has proved helpful as well to have a comprehensive view of the six lines of the pediment drawing and their two different slopes from Miletos, as will be discussed in the following section (below, Fig. 3.8a). Regarding documentation, a squeeze worked for the extremely well-preserved, deeply carved “Pytheos sketch” at Priene (Koenigs 2015: fig. 48). But in Miletos, as in many places, the surface of the blocks is not smooth enough since it is picked, and the lines are too worn to feel their depth. For this reason we applied Reflectance Transformation Imaging (RTI) to the blueprints of Miletos (Fig. 3.7a). This technique, developed at the beginning of the 2000s by Tom Malzbender at Hewlett-Packard Labs,

64

Capelle

a

b

figure 3.8 Miletos, theatre: (a) blueprint of a pediment of the first Imperial scaenae frons; (b) blueprint of the arches of the western retaining wall adapted from Krauss 1973: pl. 18, keeping only the extant remains as drawn by Knackfus

enabled us to combine in the same image varying lighting conditions, which are produced by an autonomous flash placed at various angles. After having rendered the image with the RTI-builder software, one can visualise it with the RTI-viewer, using the specular enhancement (Fig. 3.7b) and changing the light as much as one likes. The “normals visualisation” offers an equivalent of a virtual squeeze which helps us to ensure that the visible lines have some depth. RTI proved to be able to reveal barely visible inscriptions (Capelle 2017b), especially those marking reserved seats (“topos inscriptions”) in the theatre, but the results are less spectacular for blueprints: we used it mostly to ensure that the lines we discerned with the naked eye and good lighting had a real depth, though we occasionally also found some new markings. Nevertheless, RTI can be

considered the most commonly adapted technology for these kinds of meticulous studies: it is particularly suited for deciphering hardly visible fine surface deformations and superimposed incisions.2 2  In comparison, 3D white light scanning and 3D modelling has been attempted in Didyma: it took two nights and five days of data processing to obtain the same results as Haselberger with the naked eye for the “entasis drawing” (Bankel 2013). For a convincing application of both RTI and 3D (photogrammetry) as complementary technologies, the first one being useful for deciphering details, the second one for collecting data on a larger scale, see the El Morro Project (http://culturalheritageimaging.org/What_We_Do/Projects/elmorro/ index.html). Another significant advantage of RTI is that anyone can process the data efficiently using the free Hewlett-Packard software (http://culturalheritageimaging.org/Technologies/RTI/) and affordable equipment.

ANCIENT BLUEPRINTS: NEW PROSPECTS IN LIGHT OF RECENT DISCOVERIES

4

The Uses of Blueprints: Between Theory and Practice

It is not an easy task to give an interpretation of the silent evidence we have of ancient blueprints. We cannot know for sure if they were drawn by architects, overseers of work teams, or stone masons. When one looks at, for instance, modern French treatises on architecture and stereotomy from the 18th century to present, the answer varies, but often mentioned is the special role of blueprint-maker assumed by the “appareilleur”—the “dresser” of stone or the “fitter” who puts the blocks together (from Jaubert & Didot 1773, s.v. Appareilleur to Sakarovitch 1998: 11–12). The appareilleur is a key intermediary who works out how the stone will be cut by the stonemasons to execute the architect’s design. This dialectic of theory and practice may guide us to a valid interpretation of some recently discovered ancient drawings. The earliest identified drawing from the theatre of Miletos probably represents one of the small pediments of the scaneae frons of the first century CE (Fig. 3.8a) (Capelle 2017a). This is an important piece of evidence for the reconstruction of the first Imperial scaenae frons (Altenhöfer 1986), as very few blocks from the five pediments have survived. It seems clear that the drawing contributes to the design by comparing two different slopes, which has serious implications for the general shape and dimensions of the pediment. Even if a general idea existed, it was discussed and adapted in the field. Furthermore, once one choice was made, one could still add information to the drawing. Lines, perhaps added later on, that are parallel to the steeper of the two slopes provide additional measurements for the simas, and the drawing probably served as a reference for the masons as well. If we consider the drawing of the arches of the western retaining wall (Fig. 3.8b), which dates to the second century CE, the diagram is contemporary and measures exactly as the real arches—which were built without a change in decision. It may just be the geometric representation of a well-defined design, beginning with a 30-foot-long horizontal line which is divided into parts. As for the pediment and most of the drawings, only half of the symmetrical shape is drawn, and it is protracted towards the left just enough to give the necessary lines for stonecutting: one of the springers with its special shape and the keystone to which the four other voussoirs are equal. We are at the last stage of design, just before execution. Nonetheless, the effort of drawing out the two arches would still have something to do with testing their proportions.

65

As a third example, we may examine two of the four detailed drawings which were designed in the hyposkenion at Miletos. We can find side by side the upper and lower profiles (Fig. 3.9a) of the terminal pedestals of the retaining walls, which date to the second century CE. The eastern one bore the statue of the Asiarch M. Antonios Apollodoros (Fig. 3.9b) (Krauss 1973: no. 936). The lower profile on the right once continued on the next course but is now lost. We can imagine that it was useful for the stonecutters to take measurements from the drawing and use some lines as references. One tangent of the base drawing corresponds to a roughed-out surface visible under the steps of the staircase (Fig. 3.9c). Still, these drawings are much more than just profiles for masons to copy. They are the result of geometric constructions which define the respective proportions of the subdivisions of the moulding. In the lower left corner of the geison drawing, we can see a construction angle that corresponds to a 3:4 ratio. In all probability, it served to fit a set square whose perpendicular arms were set to this 3:4 proportion, so that the same ratio could be applied at least three times to divide the moulding into parts (Fig. 3.9d). This can be achieved by a typical Euklidian construction (Fig. 3.9e), which we could describe as follows: Given two parallel lines, put a set square following the required proportion parallel to the lines. Carry the height forward onto the lowest horizontal. Then draw a vertical perpendicular to it and draw a new horizontal where it intersects with the slope of the square. You now have the desired proportion (after Euc. 6.2 Intercept theorem). A last interesting feature of this drawing is that the general design is broken down into parts, and the profile has been shifted left, right, and up for the upper part, which follows a different proportion of 2:3 (Capelle 2017a: fig. 11). This blueprint indicates a very pragmatic application of the principles of Euklidian geometry to architecture. The geometric construction is part of the design and can be read as directions by the stonecutter. Furthermore, it seems that even if a general idea of the design exists, the drawing could still be a place where the details and the inner proportions of a given shape were determined. A clear example of Hellenistic date can be found at the Milesian Bouleuterion. It represents the profile of the original seats of the auditorium, which were replaced in Imperial times (Fig. 3.10a) (Knackfuß 1908: 35,

66

Capelle

1.5

14.6

Hyposkenion pillar

1.7

9.6

1.4

5.0

1.9

2.5

6.8

29.9

10.6

7.9

2.8

4.7

7.4

a

0

10

100 cm

b

d

c

2

3

3

e

3

3 4

3

figure 3.9a–e

4

4

4

4

4

4

7

(a) Miletos, theatre, blueprint of the upper and lower profile of the terminal pedestals of the retaining walls; (b) ending pedestal of the eastern retaining wall, with the inscription of the Asiarch M. Antonios Apollodoros; (c) original roughing out surfaces preserved under the steps of the staircase; (d) proportions used in the drawing; (e) Euklidian construction with a proportion square author

ANCIENT BLUEPRINTS: NEW PROSPECTS IN LIGHT OF RECENT DISCOVERIES

fig. 8). As far as I know up to the present, the representation of seat mouldings on a blueprint is unique. Somewhat surprisingly, the drawing is of a well-known profile used not only at the Bouleuterion, but also at the stadium and the theatre of Miletos (Fig. 3.10b). It can also be observed in Priene—in its Bouleuterion and in the lower proedria of the theatre (Fig. 3.10c)—and on Samos (Gerkan 1921a: 3, fig. 2 [stadium at Miletos]; 1921b: pl. 14 [theatre at Priene]; Krauss 1973: 79–89, pl. 27–28 [theatre at Miletos]; Wiegand & Schrader 1904: 224, fig. 218 [Bouleuterion at Priene]; for Samos see the marble bench displayed today in the courtyard of the Pythagoreion museum). All dated examples go back to the beginning of the second century BCE, apart from the Imperial cavea of the theatre of Miletos, the lower part of which seems to be a reassembling of an older koilon (some clues not reviewed here suggest a reuse of the Hellenistic seats). We may wonder why such a canonical moulding was drawn with so much hesitation. Apparently, the shape required some adaptation. Each profile has its own variations, and it seems that the beginning of the ovolo was shifted down to find the desired ratio between the convex and concave parts of the moulding. One might even have used an existing template from another building, and tried to move the upper and lower lines to modify the shape. As at the theatre, this blueprint was not isolated: another drawing (Fig. 3.10d) covers the two orthostates (the right one of which was not recovered in situ) of the northern wall of the corridor next to the other blueprint. Its identification remains uncertain but the intervals between the vertical lines seem to match the various widths of the lotus and palmettes which adorn the frieze of the raking sima of the auditorium (see Knackfuß 1908: 48–49, fig. 6, pl. 9–10). Lastly, it is certain that blueprints were only one part in the chain of architectural drawings. At the theatre of Priene a drawing on the northern wall of the central room of the scene building probably works out the partition of the triglyph in the Doric frieze of the proskenion (Fig. 3.11a). The vertical axis equals the total height of the entablature, which is one third that of a column. The 3-mm discrepancy between these two intervals is not an obstacle to this interpretation. On the contrary, as observed by A. von Gerkan (1921b: 38) and J. Misiakiewicz (Schumacher & Misiakiewicz 2007: 57–58), the whole scene building of Priene, including its triglyphs, is full of such irregularities which one does not notice at first sight. It may be the sign of a continued carelessness through the chain of drawings. Indeed, the blueprint has similarities to other kinds of drawings, such as the delineation of the plan of metopes and triglyphs found on the floor of the Angista tomb by

67

W. Hoepfner (Fig. 3.11b) (Hoepfner 1984: 22; Katerina Charatzopoulou kindly informs me that she will reexamine the drawing in her chapter about the architecture and the ornamentation of the tomb in a forthcoming book under the direction of Chaido Koukouli-Chrysanthaki). Many setting marks in turn respond to that drawing, such as the lines (i.e., crosses) on the top surface of the epistyle blocks corresponding to the regulae at the Bouleuterion of Miletos. We can also find various marks which are reminiscent of our elevation drawing on the front face of frieze blocks in Delos. On one block of unknown origin at the Hypostyle Hall (Fig. 3.11c) (Poulsen & Vallois 1914: no. 40), a line marks the vertical axis, whereas a horizontal line is drawn above the glyphs on the triglyph blocks from the late Hellenistic porticoes of the Agora of the Italians (Fig. 3.11d) (Lapalus 1939). So it seems that not only could the project materialise through a blueprint, but also the blueprint could help workers when they were carving guide marks before cutting the stone. Finally, a recently discovered example summarizes many of the foregoing points while showing how one blueprint can clarify the interpretation of many others. In November 2017, we identified a blueprint incised on the western retaining wall of the theatre of Larisa, Thessalia, on smooth orthostates at eye level. If one looks casually, one can hardly see anything. But with some patience and some lengths of string, one may retrace long horizontal lines which are reminiscent of column drawings on the temple of Didyma and the theatre of Aphrodisias, which make clear that it is easier to rotate the elevation drawing horizontally (Fig. 3.12b) (Haselberger 1980: 193–202, fig. 1–3; Hueber 1998: fig. 9). A search for a column whose dimensions coincide with those of the diagram resulted in an exact match to the half-columns of the proskenion (Fig. 3.12a). There are even setting lines delimiting the lower diameter on the stylobate, the axis and the circumference of the column, and some additional marks for the flutes (Fig. 3.12c). The circle in the middle of the drawing corresponds to the plan of the lower diameter, as with the superimposed drawings of section, elevation, and plan of the column at Didyma. But contrary to the entasis drawings of Didyma and Aphrodisias, we have at Larisa a study for the narrowing of the shaft. The working out of the design appears also with the two middle vertical lines, which delimit a Thessalian foot of ca. 31.9 cm; this approximation seems to be confirmed by the 31.91 m long—most probably hekatompedon—retaining wall. The shaft follows a 1-1/4 by 7-1/2 ft. scheme for a proportion of 1:6. The drawing was close at hand for the builders. They may have also checked

68

38.8 33.5 33.4 31.8

Capelle

a

0 5 10

50

100cm

b

c

Council house

Theater, ima cavea

Stadium

Theater, summa cavea

d

36.1 cm

25.8 cm

29 cm

22.5 cm 31.5 cm

0

10

figure 3.10a–d

50

100

200 cm

(a) Miletos, Bouleuterion, blueprint of the Hellenistic bench; (b) similar profiles at Miletos; (c) Priene, theatre, lower proedria; (d) Miletos, Bouleuterion, unidentified blueprint (frieze?) author

69

a

6

0 5 10

5.7

50

27.1 29

26.4

central door

partition wall

ANCIENT BLUEPRINTS: NEW PROSPECTS IN LIGHT OF RECENT DISCOVERIES

27.3

100 cm

Axis

b

25.8 Triglyph

19.2

19.2 Metope

26.1 Triglyph

10 cm

c

figure 3.11a–d

d

(a) Priene, theatre, blueprint of the Doric frieze of the proskenion (actual thickness of the emphasized lines ca. 0.1 mm) with drawing of a triglyph block superimposed (dashed lines); (b) Angista, chamber tomb, blueprint of the plan of the Doric frieze; (c) Delos, Hypostyle Hall, triglyph with a mark for the axis; (d) Agora of the Italians, triglyph with a line above the glyphs (black lines overlaid) (A) AUTHOR (b) adapted from Hoepfner 1984 (C)–(D) AUTHOR

37,5

20

17,5

17,3

17,3

34,6

figure 3.12a–c

0 5 1

100 cm

31,9

45,8

40,1

Larisa, theatre: (a) proskenion façade in the foreground and western retaining wall in the background; (b) blueprint of the proskenion semi-columns; (c) setting lines on the stylobate of the scene building (white lines overlaid) photographs and drawing are by the author

5

40,1 239,1

19,5 19,5

b

c

19,3 20,8

a

70 Capelle

imprint of parodos door

ANCIENT BLUEPRINTS: NEW PROSPECTS IN LIGHT OF RECENT DISCOVERIES

the height of the even-numbered (the second and fourth) courses of the pseudo-isodomic masonry, which equals the interval between the two verticals to the right. Lastly, the imprint of the parodos door can be seen on a part of the drawing: this means the colonnade of the proskenion was made before the parodos door. Although this was not our point here, blueprints not only teach us a great deal about design and architectural practice, but they also provide evidence for the date and the reconstruction of ancient buildings. This multi-faceted information represents a great opportunity for the advancement of Bauforschung. 5 Conclusions We may conclude from these examples that we are now far from the long-established but misconceived notion of Greek architecture as highly normative, repetitive, and conceptual—a model which once led Javary to deny the existence of blueprints. It is today possible to give a substantial answer to this scholar’s question, not only because Haselberger’s work paved the way for many more discoveries documenting a phenomenon widespread across the Mediterranean and over the ages, but also because our increased knowledge has contributed to changing our way of thinking about ancient architecture. The evidence reveals that a great part of the design was skillfully worked out in the field, during the building, and often on the very walls of the monument being built. Drawings were used to visualise abstract shapes at a natural scale, to adapt architectural types to specific buildings, and sometimes to test several building options. They helped to determine the general measurements and proportions as well as the details, especially moulding profiles. They represent most generally the final decision, but they can also be places where discussions took place, not only about a slight variation, like the torus on the famous entasis drawing at Didyma, but also about the dimensions, shape, and proportions of architectural forms like the pediment, arches, and seat profile at Miletos. These drawings were the translation, often at a natural scale, of a general sketch, which was either purely conceptual or drawn on perishable material. They follow well-mastered, millimeter-precise Euklidian geometrical constructions. The masons could refer to such patterns in order to mark corresponding guidelines before cutting the stone. The very few blueprints that have survived to the present day offer us the missing link between theory and practice, by giving us an insight into a well-advanced stage of design and into the ways plans evolved in the construction

71

yard itself. Indeed, once the project of an architect was adopted by the assembly, the building had to be realised in a process that might last for years, decades, or even centuries. When the labour and the financial means were available, it was necessary for the people in charge of the work to give a pragmatic shape to the original ideas—possibly adapting them to the actual conditions—and to issue clear instructions for stonecutting. It is in this context that blueprints became necessary. Blueprints are the proof that “incomplete preliminary planning”, as J.J. Coulton (1985) put it, was a common practice, and maybe the rule rather than the exception, in Greek architecture. List of References Altenhöfer, E., 1986: “Das erste römische Bühnengebäude des Theaters von Milet”, in Müller-Wiener, W. (ed.), Milet 1899– 1980. IstMitt Suppl. 31 (Tübingen) 165–73. Bankel, H., 2013: “Ancient Construction Drawings and New Methods of Documentation: 3D White Light Scanning and 3D Modelling”, JRA 21.1: 383–92. Borchardt, L., 1896: “Altägyptische Werkzeichnungen”, ZÄS 34: 69–76. Bouriant, U., 1895: “(Javary) Question n°35. Sur l’origine historique des épures: Réponse 3”, L’intermédiaire des mathématiciens 2: 207–08. Bouriant, U., 1896: “(Javary) Question n°35. Sur l’origine historique des épures: Réponse 4”, L’intermédiaire des mathématiciens 3: 87. Brouquier-Reddé, V. & Gruel, K. (edd.), 2003: Le sanctuaire de Mars Mullo—Allonnes (Sarthe) (Le Mans). Capelle, J., 2017a: “Les épures du théâtre de Milet: pratiques de chantiers antiques”, BCH 141.2: 769–820. Capelle, J., 2017b: “Reflectance Transformation Imaging (RTI) et épigraphie”, RAAN: https://raan.hypotheses.org/1326. Capelle, J., forthcoming (2020): “Une épure de chapiteau corinthien gigantesque gravée sur le parvis du mausolée d’Auguste”, RA 69 (2020.1). Choisy, A., 1894: “(Javary) Question n°35. Sur l’origine historique des épures: Réponse 2”, L’intermédiaire des mathématiciens 2: 241–42. Choisy, A., 1904: “Note sur deux épures égyptiennes conservées à Edfou”, Journal of the Royal Institute of British Architects 27/08: 503–05. Čičikova, M., Stoyanova, D. & Stoyanov, T., 2012: Carskata grobnica s kariatidite kraj selo Sveŝari: 30 godini ot otkrivaneto = The Caryatids Royal Tomb near the Village of Sveshtari: 30 Years since the Discovery (Isperih). Coulton, J.J., 1985: “Incomplete Preliminary Planning in Greek Architecture: Some New Evidence”, in Bommelaer, J.-F. (ed.),

72 Le dessin d’architecture dans les sociétés antiques, actes du colloque de Strasbourg, 1984 (Leiden) 103–17. Daressy, G., 1907: “Un tracé égyptien d’une voûte elliptique”, ASAE 8: 237–41. De Franciscis, A., 1983: “Abozzi per la decorazione dell’anfiteatro campano di S. Maria Capua Vetere”, NumAntCl 12: 171–76. Hanoune, R., 1996: “Un dessin d’architecture au théâtre de « Bulla Regia » (Tunisie)”, in Khanoussi, M., Ruggeri, P. & Vismara, C. (edd.), L’Africa romana, Atti del XI Convegno (Cartagine, 1994). Africa Romana 11 (Ozieri) 911–14. Haselberger, L., 1980: “Werkzeichnungen am Jüngeren Didymeion—Vorbericht”, IstMitt 30: 191–215. Haselberger, L., 1983: “Bericht über die Arbeit am Jüngeren Apollontempel von Didyma—Zwischenbericht”, IstMitt 33: 90–123. Haselberger, L., 1986. “Werkzeichnungen des Apollontempels III: das antike Zeichenverfahren”, Didyma Wegweiser 23: 2 pp. Haselberger, L., 1994: “Ein Giebelriß der Vorhalle des Pantheon—die Werkrisse vor dem Augustus-Mausoleum”, RM 101: 279–307. Haselberger, L., 1995: “Deciphering a Roman Blueprint”, Scientific American June 1995: 56–61. Haselberger, L., 1996: “Die Fronthalle des Pantheon—Ein Werkriss des Dachstuhls?”, in Schwandner, E.-L. (ed.), Säule und Gebälk: zu Struktur und Wandlungsprozess griechischrömischer Architektur: Bauforschungskolloquium in Berlin vom 16. bis 18. Juni 1994. DiskAB 6 (Berlin) 182–89. Hinkel, F.W., 1976: “Erstmals Bauplan einer Pyramide gefunden”, Spektrum 6: 30–32. Hinkel, F.W., 1980: “Überraschende Entdeckung im Sudan: Die 2000 Jahre alte erste Zeichnung einer Pyramide”, Altertum 26: 27–33. Hinkel, F.W., 1981: “Pyramide oder Pyramidenstumpf  ? Ein Beitrag zu Fragen der Planung, konstruktiven Baudurchführung und Architektur der Pyramiden von Meroë (Teil A)”, ZÄS 108: 105–24. Hoepfner, W., 1984: “Masse—Proportionen—Zeichnungen”, in Hoepfner, W. (ed.), Bauplanung und Bautheorie in der Antike. DiskAB 4 (Berlin) 21–23. Hueber, F., 1998: “Werkrisse, Vorzeichnungen und Meßmarken am Bühnengebäude des Theaters von Aphrodisias”, AntW 29.5: 439–45. Inglese, C., 2014: Progetti sulla pietra: lo studio dei tracciati di cantiere attraverso il rilevamento (Rome). Jaubert, P. & Didot, P.-F., 1773: Dictionnaire raisonné universel des arts et métiers, contenant l’histoire, la description, la police des fabriques et manufactures de France & des pays étrangers (Paris). Javary, 1894: “Question 35. À quel moment peut-on dire que l’on a commencé à faire des épures, c’est-à-dire des tracés

Capelle géométriques que l’on s’efforce de rendre aussi exacts que possible ?”, L’Intermédiaire des mathématiciens 1: 9–10. Jomard, E.-F., Lancret, M.A., De Chabrol De Volvic, G.-J.-G., De Rozière, F.-M., De Saint-Genis, A., Jollois, J.-B.-P., De Villiers Du Terrage, É. & Costaz, L., 1821: Description de l’Égypte— Antiquités (Paris). Knackfuß, H., 1908: Milet 1.2: Das Rathaus von Milet (Berlin). Koenigs, W., 2015: Der Athenatempel von Priene. AF 33 (Wiesbaden). Krause, C., 1985: “Das Graffito in Terracina”, in Krause, C. (ed.), La prospettiva pittorica, un convegno, Rome, 1980 (Rome) 131–33. Krauss, F., 1973: Milet IV.1: Das Theater von Milet 1: das hellenistische Theater—der römische Zuschauerbau. (Berlin). Krauss, F., 1974: “Milet und Didyma: ein Vergleich der Sockelprofilierung an der westlichen Parodoswand des Theaters und den Cellawänden des Apollontempels”, in Mansel’e Armağan, TTKY VII, 60 (Ankara) 185–92. Lapalus, E., 1939: L’agora des Italiens. Délos 19 (Paris). Manetta, C., Stoyanova, D. & Luglio, G., 2016: “A New Survey in the Ostrusha Tomb Near Kazanluk. First Results Concerning Architecture and Painting”, Bulgarian E-Journal of Archaeology 7: http://be-ja.org. Petrie, W.M.F., 1888: A Season in Egypt: 1887 (London). Petronotis, A., 1968: Bauritzlinien und andere Aufschnürungen am Unterbau griechischer Bauwerke in der Archaik und Klassik (Athens). Petronotis, A., 1972: Zum Problem der Bauzeichnungen bei den Griechen (Athens). Poulsen, G. & Vallois, R., 1914: Nouvelles recherches sur la salle hypostyle. Délos 2 (Paris). Ruiz De La Rosa, J.A., 1987: Traza y simetría de la arquitectura, en la Antigüedad y Medievo (Seville). Sakarovitch, J., 1998: Épures d’architecture: de la coupe des pierres à la géométrie descriptive XVIe–XIXe siècles (Basel). Schumacher, A. & Misiakiewicz, J., 2007: Priene: die Restaurierung des Theaters 1992–1998. Edited by W. Koenigs (Mainz). Schwandner, E.-L., 1985: “Zu Entwurf, Zeichnung und Maßsystem des älteren Aphaiatempels von Aegina”, in Bommelaer, J.-Fr. (ed.), Le dessin d’architecture dans les sociétés antiques, actes du colloque de Strasbourg, 1984 (Leiden) 75–85. Stanzl, G., 1999: “The Ptolemaion at Limyra and Its Recently Discovered Curvature”, in Haselberger, L. (ed.), Appearance and Essence: Refinements of Classical Architecture—Curvature. Proceedings of the Second Williams Symposium on Classical Architecture Held at the University of Pennsylvania, Philadelphia, April 2–4, 1993 (Philadelphia) 155–72. Tannery, P., 1894: “(Javary) Question n°35. Sur l’origine historique des épures. Réponse 1”, L’Intermédiaire des mathématiciens 1: 44. von Gerkan, A., 1921a: Milet 2.1: Das Stadion (Berlin).

ANCIENT BLUEPRINTS: NEW PROSPECTS IN LIGHT OF RECENT DISCOVERIES von Gerkan, A., 1921b: Das Theater von Priene: als Einzelanlage und in seiner Bedeutung für das hellenistische Bühnenwesen (Leipzig). Weber, U., 2015: “Ein zweiter hellenistischer Naiskos im Apollonheiligtum von Didyma? (Kurzfassung)”, in Bachmann, M., Wulf-Rheidt, U., Bänkel, H. & Schwarting, A. (edd.), Bericht über die 48. Tagung für Ausgrabungswissenschaft und Bauforschung: vom 28. Mai bis 1. Juni 2014 in Erfurt (Stuttgart) 169–171.

73

Weber, U., Bumke, H., Breder, J., Kaiser, I. & Reichardt, B., 2015: “Didyma—Bericht über die Arbeiten der Jahre 2010–2013”, AA 2015.1: 109–72. Wiegand, T. & Schrader, H., 1904: Priene: Ergebnisse der Ausgrabungen und Untersuchungen in den Jahren 1895–1898 (Berlin).

chapter 4

Three Hellenistic ‘Naïskoi’ in the Theatre Area at Aigeira: Small Buildings in the Context of an Urban Sanctuary Alexandra Tanner 1 Introduction This chapter presents the first results of the architectural study of the three so-called naïskoi—Buildings D, E, and F—in the theatre area at Aigeira.1 The ancient polis of Aigeira is located on a ridge close to the southern shore of the Korinthian Gulf in the Peloponnesos. The site was provided with a harbour as well as access to the interior. The settlement preserves traces of Neolithic occupation, was important in the Bronze Age, and continued to exist throughout Archaic and Classic times. In the Hellenistic period it experienced important growth, which is best recognised by the construction of the city walls and an architectural ensemble built on a previously unoccupied site that would remain in use until late Roman times. It comprises a theatre, a large peristyle-building complex, the small freestanding “naïskoi” D to F, and other small structures A to C that are situated around a large central open area (Figs. 4.1–4.2). This arrangement of buildings does not find any direct parallels in Greek urbanism, which raises the question of its role and function. Was this one of the civic or religious centres of the polis, or did it serve both purposes? Was it planned and built in a single, coherent building programme or gradually developed over a considerable amount of time? These questions can only be addressed after a careful study of each building and its chronology. One building type that is especially prominent because of its multiple instances, is the so-called naïskos. The three small “naïskoi”, Buildings D, E, and F, are located on either side of the theatre, and two more small temple-like structures situated on a lower terrace, Buildings I and II, are now backfilled. The main focus of this project are the three Buildings D, E, and F. The study includes the examination of their 1  This study is part of my dissertation project “Three hellenistic ‘naiskoi’ in the theatre area at Aigeira” at the University of Zurich. I would like to thank W. Gauss, the director of the Aigeira excavation for supporting my research. Furthermore, I am grateful to my advisor and co-advisor Prof. Dr. C. Reusser and Prof. Dr.-Ing A. von Kienlin as well as colleagues and participants of the conference preceding this volume, especially the editors, for all discussions.

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004416659_006

building history, functions, and role within the urban context of Aigeira. The distinction of such a “naïskos” from a treasury and a building for banqueting is unclear. The naïskos, consisting of a cella and a pronaos, is the smallest type of a prestigious one-room building and is suitable for various uses rather than one specific function. In this research, the exclusive function of the “naïskoi” as temples is questioned; instead of the term “naïskos”, the more neutral “building” is used here. The problem of identifying small, temple-like buildings as cult buildings, treasuries, or dining rooms has already been raised by many scholars and recently within the framework of Hellenistic studies: apart from general discussions mainly about the Archaic period (Roux 1984, Svenson-Evers 1997, Neer 2001, and Hölscher 2001) there have been discussions of the Hellenistic period and studies on the late classical andrones at Labraunda, as well as the buildings at Dodona (Nielsen 2007; Hellström 1990; Emmerling 2012; and Mancini 2013). During the Hellenistic period, as requirements for architecture changed and a new, more holistic perception of space emerged, we may observe new experiments in architecture. At the same time, traditional architectural types retained their importance or attracted revived interest. In sacred architecture of the Hellenistic, trends towards small buildings as well as a museum-like presentation have been particularly clear (e.g. Lauter 1986: 190–96; Cain 1995; Felten 1996; Zimmermann 2015)— tendencies which shall be explored with the present case study. Furthermore, for the first time the methodologies developed in Bauforschung are applied to buildings which have been excavated and known for a long time. A close reading of stratigraphic and constructional evidence, an identification of measurement units, and a functional analysis are contributing important new insights into Buildings D, E, and F (Fig. 4.2) Due to the many open questions about the entire site regarding chronology, architecture and function, a holistic approach was necessary: the careful study of the material remains as well as the analysis of the overall design, including historical sources and the results of the entire research in the theatre area.

THREE HELLENISTIC ‘ Naïskoi ’ IN THE THEATRE AREA AT AIGEIRA

figure 4.1 Overall plan of the theatre area: theatre, peristyle building, Buildings A–F S. Gogos, T. Hagn, author, H. Birk; courtesy ÖAW/ÖAI Athen

75

76

Tanner

figure 4.2 Ground plans of Buildings D, E, and F author; courtesy ÖAW/ÖAI Athen

2

Previous and Current Fieldwork

The first archaeological investigations in the theatre area took place under the direction of Otto Walter in 1916 and 1925. Systematic excavations were directed by Wilhelm Alzinger in the 1970s and the 1980s, and Anton Bammer dug a few additional small trenches in 1992 and 1993. Apart from the theatre, which was described in a monograph by Savas Gogos (1992), the other buildings and all finds from the excavations were only presented in preliminary studies that were published in a few articles and summaries (esp. Walter 1919; 1932; Alzinger 1972–1975; 1985; 1990; Alzinger, Gogos & Trummer 1986; Gogos 2001; Bammer 1993; 2001; Hainzmann 2001; Schrettle 2007). Guided by Pausanias’ description of Aigeira (7.26.2–10) and fragments of sculptures found during the excavations—among them the socalled head of Zeus—the excavators Alzinger and Gogos identified Building D as a temple of Zeus, and Building E as a temple of Artemis (Alzinger, Gogos & Trummer 1986: 49–50; and Gogos 2001: 79). Pausanias (7.26.2–10) enumerated several cult statues and sanctuaries without precise location. However, as Maria Aurenhammer points

out in her most recent research (currently in press), no sculpture can be attributed with certainty to a particular deity or specific building in Aigeira (on the cult attribution, see also Osanna 1996: 261–68; 272–75; Solima 2011: 17–18). Moreover, the excavators identified the entire group of buildings as the sanctuary of Zeus mentioned by Pausanias. It is important to emphasise that very little of the area of about 50 hectares once occupied by the Hellenistic city of Aigeira, which is indicated by the circuit of the city walls, has been excavated and, apart from the buildings at the theatre area, no other sanctuary that might have still existed in Pausanias’ times has yet been excavated. Based on the dating of the first phase of the theatre building by Gogos to the mid-third century BCE, the building programme would have been initiated shortly after Aigeira joined the Akhaian League, 275/274 BCE as the excavators presumed. The transformation of the skene building from a Hellenistic proskenion to a Roman scaenae frons is dated between the time of Hadrianus and the beginning of the crisis of the third century CE, which would have left the building unfinished. Both dates are mainly based on numismatic evidence (Gogos 1992: 85,

THREE HELLENISTIC ‘ Naïskoi ’ IN THE THEATRE AREA AT AIGEIRA

119–25). As for the relative chronology of the “naïskoi”, Gogos (2001: 86) argues that Building D was the temple of Zeus, and due to its topographic situation the oldest of the three small buildings, while Building E was built after Building D and the theatre. He considered the building plot as unfavourable because of the slope and proximity to the theatre. As a consequence Building E would have been shortened. However, this interpretation is not supported by any archaeological evidence and thus is open to questioning. For all future research in the theatre area, the relative chronology of the buildings is crucial. The first step of the ongoing project has, therefore, been the identification of the building phases based on a new documentation and study of the current state of preservation. In 2011 the Austrian Archaeological Institute with Walter Gauss launched a project for the systematic study of the finds and architecture from the theatre area at Aigeira (Gauss et al. 2012; 2013; 2015a; the latest results are presented in Gauss, in press). Drawing of ground plans and elevation

77

views of the three buildings at a scale of 1:20 by the author of this paper promoted the careful study of every detail of the structures (Figs. 4.2–4.3). Together with the documentation produced during the excavation, they form the basis for the new architectural analysis presented here. Building D, which preserves a floor mosaic in excellent condition that was re-exposed during the campaign of 2013, was also recorded with a high-resolution photogrammetric 3D model (Gauss et al. 2015a: 38–41). While this model has served primarily for the purpose of public presentation, it was also used to generate orthophotos in support of the elevation drawings. In terms of the relative chronology, a detailed study was carried out on the two neighbouring Buildings D and E. Their simultaneous investigation allowed the observation of their similarities, differences, and relations better than a separate analysis of each building (see Tanner [in press] for more details on the archeological evidence). After presenting the conclusions of this examination, two reconstruction scenarios will be discussed in terms of possible correspondence of

figure 4.3 (a) South elevation of Building E in front of Building D; (b) south elevation of Building D; (c) north elevation of Building D; (d) cross section of Buildings D and E towards west author; courtesy ÖAW/ÖAI Athen

78

Tanner

the different building phases of the two “naïskoi”. The development of these two scenarios is a method for dealing with the many ambiguities concerning the building history which promotes the ensuing discussion of the more probable reconstruction. Finally, the architectural analysis enables the study of the functions and meanings of those buildings and their context. 3

Evidence and Architectural Phases

Buildings D and E are located side by side to the north of the theatre, where they are built into the sloping terrain, whereas Building F is situated south of the skene building. Like the northern Buildings A, B, and C (Walter 1932: 232), several statue bases, and the peristyle building (Hagn 2001), they face a wide, open public space (above, Fig. 4.1), where neither trenches nor geophysical analyses found evidence for structures (Rusch et al., in press). Although the group of buildings gives the impression of a symmetrical ensemble constructed on a rectangular grid, it is slightly irregular: Building F is located closer to the theatre than Building E, and the three “naïskoi” on both sides of the scene building are not quite oriented along orthogonal lines to each other. The three buildings are all prostyle. The preserved dimensions of the foundations are 8.4 × 12.5 m for Building E, 9.7 × 17.7 m for Building D, and 8.2 × 14.5 m for Building F. While their deep pronaoi, podium at the rear, and floor

mosaics make Buildings D and F typologically similar to one another, Building E has only a shallow pronaos, and a collection of bases and pedestals in the rear of the cella, which is nearly square (Fig. 4.2). The foundations of Building E were cut into the bedrock. Its socle was built with orthostates on a slightly wider dado base, both of local conglomerate, and with a dado crown of sandstone. The upper layer consists of irregular, partly reused blocks and tiles belonging to a later phase (Figs. 4.3a, 4.4a). The longer Building D is similar regarding its above-ground wall construction, except for being made entirely in the local conglomerate, but its foundations differ considerably. Rather than cutting a level platform, the foundation is built over the sloping bedrock, and so differs in elevation by about 1.50 m from the eastern to the western part (Figs. 4.3b-c, 4.4b). As a result, the interior floor of Building D is significantly, i.e. 90 cm, higher than the floor level in E (Fig. 4.3d). In the area south of the theatre building, the bedrock is at a lower elevation. Therefore, Building F is constructed on a foundation several courses deep that is founded on the bedrock. The socle is also built of orthostates in conglomerate, but no dado crown is preserved. Due to its incomplete preservation and its location away from the other two “naïskoi”, the ensuing discussion will concentrate on the other two buildings. The elaborate socle of Building E was visible all around, and rough stone only appears below the bottom of the dado base (at +345.85 m above sea level; Figs. 4.3a, d–4.4a).

figure 4.4 Detail of the south wall socle of (a) Building E and (b) Building D photos by author; courtesy ÖAW/ÖAI Athen

79

THREE HELLENISTIC ‘ Naïskoi ’ IN THE THEATRE AREA AT AIGEIRA

Although there is a joint between the cella and pronaos, the lack of a foundation on the eastern wall of the cella precludes the existence of an initial phase of the building as an oikos lacking any pronaos. Instead, the construction technique is consistent throughout the building (Fig. 4.3a). The first floor is on the level +346.0 m, directly above the trimmed bedrock, indicated by wall plaster on the interior and remains of a mortar floor in the cella and pronaos at the same level. In light of this floor elevation, in the first phase the krepis could only have been one step high. This first phase was very likely decorated with wall plaster in the masonry style, fragments of which were found in the cella (Alzinger 1986: 41–2; Leibetseder [in press]). Fills in the cella containing these fragments and the threshold that was set at a higher elevation indicate a slight rise of the interior floor and a more substantial increase in the outer ground level, including the pronaos, by about 30 cm. A second step for the krepis was necessitated, while two steps led from the pronaos down into the cella. The superstructure of the wall was most likely built from mudbrick; air-dried bricks are preserved in situ at the rear, inside of Building E. Hans Lauter (1986: 50–1) points out that the use of mud-brick became more common in the Hellenistic era even for public buildings. When the outside ground level was raised again, as revealed by the plaster on the south wall of Building D (Fig. 4.4b), Building E was rebuilt with a higher socle in order to protect its brick wall against ground moisture from the soil (Fig. 4.3a). The layer of reused blocks and roof tiles contains spolia from the skene building of the theatre dating the rebuilding to Roman times. The two fragments of semi-columns, P2 and P3, are comparable with the fragments P1 and P4 found in situ and near the proskenion (Gogos 1992: 62–66; 123–25; table 31–39). No blocks of an upper structure are preserved, apart from two Doric columns reused as bases in the cella which probably belonged to the initial pronaos. Following the natural slope of the bedrock, Building D was situated at a higher elevation than its neighbour Building E. The masonry of the pronaos and cella walls was built within one phase (Fig. 4.3c). There is only one floor level throughout the pronaos and cella, that containing the aforementioned pebble mosaic (+346.90 m; Figs. 4.2, 4.3d). The level of the threshold and the width of the pronaos foundations lead to the reconstruction of a three-step krepis. The podium in the rear part of the cella also belongs to the same building phase, as shown by the rough inner surface of the orthostates (Fig. 4.3d) (Alzinger 1988: 12). The space between the two Buildings D and E and the two facing longitudinal walls is important in terms of the

relative chronology. While the socle of Building E is elaborated all around, that of Building D follows the natural slope of the bedrock and is unworked in the western part at a higher level than the socle of Building E (+346.70 m; Figs. 4.3a, 4.4a, b). Therefore, contrary to the initial hypothesis of the excavators, Building D must have been built after E, and the space between them filled up subsequently. Due to this raising of the outside ground level, the socle of Building E was raised during a later phase, when spolia were used as building material. In summary, three main building phases could be identified, two for Building E and one for Building D. In Building D, several layers of plaster on the exterior walls and painted entablature blocks attest to further alterations and use phases, as do modifications on pedestals in Building E. As for their dating, at the moment only the style of the floor mosaic in Building D provides a direct indication of the construction date. Recent research by Veronika Scheibelreiter-Gail (in press) confirms its dating by Dieter Salzmann (1982: 33–4) to the third century BCE, while he dated the mosaic more precisely to the middle third of that century. The architectural terracottas can only be dated generally in that century. Finally, Building F is the most recent, with its floor mosaic dated by ScheibelreiterGail to the second century BCE. This building, however, reuses foundations of an earlier building—the southern rear wall and a euthynteria of 120 cm width—and therefore most probably was the krepis of a stoa facing the open space to the north. 4

Design Units and Reconstruction

As a basis for reconstructing the “naïskoi”, their design scheme and measurement units were investigated. Researchers have debated whether foot units were restricted to three principal systems (Attic, Doric-Pheidonic, Samian-Ionic) or a wider variety existed (e.g., Coulton 1975: 85–7; Müller-Wiener 1988: 31–2; Koenigs 1990; 2015: 51–53 and n. 181, with bibliography). Analysis of the ground plans revealed two possible modular systems based on differing foot units for Buildings D and E, consistent with two independent processes of design and construction. The wall thickness on each building can be assumed to correspond to one design module, because the main building dimensions are evenly divided by this multiple (Fig. 4.5). The same modules can also be observed in elevation: in Building D corresponding to the height of the orthostates, and in Building E to the combined height of the orthostates together with the euthynteria.

80

Tanner

figure 4.5 Reconstruction of the ground plans of Buildings D and E author

The probable foot units for each building have been derived from these modules in Table 4.1, where one module consists of 2.5 feet. Building E utilises a foot unit of 30.80 cm (from a module of 77.0 cm), significantly smaller than the foot unit of 31.90 at Building D (from a module of nearly 80 cm, or more precisely 79.75 cm). In contrast to the very accurate execution of Building E, the dimensions on Building D are less clear due to its less precise construction and poorer state of preservation. Building D also shows an irregularity in the width of the building, which measures 10 and 1/3 modules, that might be the result of a wider central intercolumniation. However, since the stylobate is lost, no traces of columns are preserved to confirm this assumption. The socle of Building E

measures 12.32 m long—equivalent to 16 modules (40 feet of 30.80 cm)—by 7.70 m wide—equivalent to 10 modules (25 feet of 30.80 cm). The length of the cella is 12 modules (30 feet of 30.80 cm), and the depth of the pronaos is 4 modules (10 feet of 30.80 cm), meaning that the proportion of the cella to the pronaos is 3:1. The reconstructed length of Building D is 17.56 m—equivalent to 22 modules (55 feet of 31.90 cm)—and its width of 8.29 m equals 10 1/3 modules (26 feet of 31.90 cm). The ratio of the length of the cella to the depth of the pronaos is about 15 modules (37.5 feet of 31.90 cm) to about 7 modules (17.5 feet of 31.90 cm), and thus slightly greater than 2:1. The two distinctive metrological systems reveal that foot units were not standardised at Aigeira, even in two

81

THREE HELLENISTIC ‘ Naïskoi ’ IN THE THEATRE AREA AT AIGEIRA table 4.1 Building D, E: Dimensions and design units

   

Length cella (orthostates) Length pronaos (euthyntheria) Total length of building Width cella (orthostates) Wall-thickness (orthostates) Height orthostates Height socle (orthostates + euthyntheria, min.) Height frieze + geison Lateral length of floor mosaic

Building E cm

30.8-cm foot

Building D 77-cm module

cm

31.9-cm foot

80-cm module

30.8-cm foot

32.5-cm foot

924.0 30.0 308.0 10.0

12.0 4.0

37.49 (17.55)

14.95 (7.0)

38.83 (18.18)

36.80 (17.23)

1232.0 40.0

16.0

(55.05)

(21.95)

(57.01)

(54.03)

770.0 25.0 77.0 2.5   77.0 2.5

10.0 1.0   1.0

1196.0  560.0 (reconstructed) 1756.0 (reconstructed)  829.0   80.0   80.0  

25.99 2.51 2.51  

10.36 1.0 1.0  

26.92 2.6 2.6  

25.51 2.46 2.46  

   

  64.0  415.0

2.01 13.01

0.80 5.19

2.08 13.47

1.97 12.77

   

   

neighbouring buildings. There are further examples demonstrating the existence of these separate foot units.2 An interesting fact mentioned by Hennemeyer (2013: 104) is a passage by Polybios (2.37.7–10) stating that in the second century BCE, measurement units were standardised within the Akhaian League, with an implication that various units had still coexisted in this region in the third century. Furthermore, the 30.8-cm foot in Building E is almost identical to the foot of 30.7–30.8 cm in the theatre identified by Gogos (1992: 133). This was confirmed by the new study of the layout of the ensemble which revealed that the theatre and Building E are based on a radial scheme originating at the centre of the orchestra and using circles of radii of 25, 70, and 100 feet of 30.8 cm. Thus, the contemporary planning and building of the two structures is confirmed. During excavations several architectural blocks were found in an upper destruction layer between the two buildings. These whole blocks and fragments belong to the frieze and geison courses in the Ionic or Korinthian order, the 12 2  As for the unit of 30.80 cm of Building E, on the Salamis relief Mark Wilson Jones (2000: 79–80) detected a unit of 30.60 to 30.70 cm along with the so-called Doric-Pheidonic foot. At Argos a foot unit of 30.80 cm has been derived by René Ginouvès (1956: 112–3) from the foundations of the second temple of Hera. On three stoas and other buildings at Priene, Arnd Hennemeyer (2013: 102–105) discerned feet ranging from 30.75 to 31.00 cm. Regarding the foot of 31.90 cm found on Building D, a similar 31.60-cm unit discovered on the temple of Zeus at Stratos by the excavators Courby & Picard (1924: 85) was recently confirmed by Jari Pakkanen (2013: 86–7).

such blocks adding up to a total length of 7.17 m. According to their findspots they must have fallen from the southern wall of Building D. Since the metal clamps have been removed from the blocks, they would have been moved by later interventions after their collapse. Still, this must have involved no more than turning them around their longitudinal axes, which would be the easiest way to overturn them. Accordingly, the original position of the blocks— facing south—can still be reconstructed. The lateral entablature seems to consist only of a frieze and geison that together were 64 cm high, which is close enough to 2 feet of 31.90 cm to confirm their assignment to Building D. The extensive architectural analysis of Buildings D and E demonstrates that Building E was built first. Furthermore, the alignment of the colonnades of the pronaoi indicate that both buildings were simultaneously in use for some period of time. Under these conditions, two possible scenarios of the reconstruction of the building history will be presented and discussed by means of schematic drawings (Fig. 4.6). Here Building D is restored with an Ionic order, while the upper structure of E is hypothetical, unless the reused Doric colums in the cella are attributed to the building. In addition, each building consists of a tetrastyle prostasis. In the first scenario, the first phase consists of Building E, constructed with a one-step-high krepis with four columns (Fig. 4.6: upper left). Later, next to Building E, Building D was constructed with a three-step-high krepis of similar width, with four columns in the Ionic or Korinthian order. The pronaos floor is 80–90 cm

82

figure 4.6a–b

Tanner

Reconstruction scenarios I and II of the building history of Buildings D and E (author)

higher than that of Building E. Probably the pronaos of Building E was already slightly raised by the height of one step. At some point, both buildings collapsed, and at least Building E was destroyed down to its socle (Fig. 4.6: middle left). When both buildings were rebuilt afterwards in Roman times, the socle and the floor elevation in the pronaos of Building E were again increased along with the outside ground level, and the entablature of Building D was repainted. Building D was probably in use for a longer period of time, since the entablature blocks of only this building have been discovered, whereas all the reusable blocks from Building E seem to have already been removed and reused in ancient times (Fig. 4.6: lower left). As for Scenario 2, the first phase is the same as in Scenario 1, but Building E would have collapsed early (Fig. 4.6: upper right). The building was then reconstructed at an elevation raised modestly by 20–40 cm. Simultaneously, Building D was constructed above the

natural bedrock, whose foundations thus required no labourious rock-cutting. After a second demolition, Building E was rebuilt with a higher socle and floor level in the pronaos in Roman times, the entablature of Building D was repainted, and the outside ground level raised again. The final phase and abandonment of the buildings is unchanged from the first scenario (Fig. 4.6: lower right). In Scenario 1, on the one hand, the raised ground level would have been motivated by a new architectural conception of the site that included the addition of a second building (D) and more prestigious prostyle porches on both Buildings D and E. In Scenario 2, on the other hand, the destruction of Building E would instead have led to the raising of the floor level. From the architectural point of view, Scenario 1 encounters some difficulty explaining the relatively large difference of nearly 90 cm between the two floor levels. Yet, that neighbouring buildings can have

THREE HELLENISTIC ‘ Naïskoi ’ IN THE THEATRE AREA AT AIGEIRA

differing floor levels may be observed at the two treasuries in the sanctuary of Athena Pronaia at Delphoi, where there is a discrepancy in floor level of about 30 cm (Bommelaer 2015: 80, fig. 11). Scenario 2, however, requires there to have been a second rebuilding in Roman times—and thus is a more complicated account of the remains because it must posit another destruction or decay. Still, the fact that Building F was built on a predecessor supports the assumption of the reconstruction of the initial building E already in Hellenistic times. Then, possible historic events can be considered that could have eventually provoked the demolition. Nevertheless, the few relevant literary sources—i.e., about the offshore earthquake in the region in 373 BCE (see, e.g., Lafond 1998), the sack of Aigeira by the Aitolians in 219 BCE that is described by Polybios (4.57–58), or the dissolution of the Akhaian League by the Romans in 146 BCE (Pausanias 7.16.9; Schwertfeger 1974: 18)—do not provide definitive information about any destructions at Aigeira. 5

Typological and Functional Study

Whereas all three buildings share in common a prostyle plan and krepis limited to the area of the pronaos, the ground plan of the first built Building E differs from those of Buildings D and F, which, with elongated plans, are more similar to each other (Fig. 4.2). A building with a similar ground plan to Building E is the temple of Artemis Orthia in the Asklepieion of Messene, which was built in the fourth century BCE. It consists of a comparable nearly square cella and a narrow pronaos, although its dimensions (8.42 × 5.62 m) are smaller than those of Building E (Müth 2007: 164–7, fig. 92). A parallel to Building D is found in the organisation of the interior space of the temple of Despoina at Lykosoura, from the third century BCE (Lauter-Bufe 2009: 94–6), although at 11.40 × 21.40 m the Lykosoura temple is slightly larger than Building D (cf. Lauter 1986: 189–94; Schrettle 2007; Mattern 2015: 105–15 for general observations about the typology of Hellenistic naïskoi). In terms of the use of the buildings, however, the inner organisation of the ground plan is more significant than the general building type. The typological similarities of Buildings D and F—e.g., a podium in the rear part of the cella—suggest a similar function which has perhaps been altered in comparison to the earlier Building E. The later reorganisation of the bases in the rear of Building E may also represent an attempt to create a podium. In order to investigate the question of whether these typological differences are a result of a general change in function or fashion among such “naïskoi”, and to what

83

extent the different ground plan typologies were influenced by their function, the possible uses of the small temple-like Buildings D, E, and F in Aigeira must be reconsidered here—as temple, treasury, or banquet hall, as well as the possibility of multiple functions. Often for such small structures neither the archeological evidence nor the ancient terminology is certain (Hölscher 2001: 143–44; Neer 2001: 274, n. 5 with bibliography; Leypold 2008: 12–14). For example, the function of the Hellenistic temple-like buildings at Dodona, which are comparable to the small buildings at Aigeira, is also unclear and controversial. At the very least, their most common interpretation as shrines for various deities has recently been questioned, entertaining possible functions as treasuries or dining halls (Quantin 2008: 20–9; Emmerling 2012: 201– 10; 150–55; Mancini 2013; Piccinini 2006: esp. 160), and at the time of their discovery, the excavator had suggested a function as treasuries (Evangelidis 1929: 108). As for the most obvious possible function for a naïskos as a temple, the constitutive elements are a cult image and an altar (Burkert 1988: 36). No traces of an altar were found in the area of Buildings D–F, but still the existence of an altar cannot be excluded. While Buildings D and E face East, as expected for temples, Building F instead faces north. Each building has internal bases suitable for cult statues, even if relevant inscriptions are lacking and no statues have been recovered that could be attributed to these bases, and in fact no other pedestal is known from the theatre area. The study of the figural terracottas by Rudolfine Smetana (in press) has revealed that most belonged to Building E, which supports a cult function. Upon review, the elaborate architecture of the buildings and the presence of numerous statue bases suggest cult functions especially for Building E, but also for Building D. Regarding the second possible function as a treasury, a distinction is often made between “proper” treasuries founded by cities in panhellenic sanctuaries and those used more generally as storerooms (Neer 2001: 279–81; Svenson-Evers 1997: 141). Furthermore, Georges Roux (1984: 156–57) and Richard T. Neer (2001: 279) emphasise that solidly built walls and other means to impede access to the interior are characteristics of treasuries. Those features cannot easily be attributed to the three buildings at Aigeira, which, for example, most probably had mud-brick walls. More broadly, Roux interprets a treasury building as not only a shelter for dedications but also itself as a dedication, although the distinction from a temple is not always clear (Roux 1984: 171; also see Hölscher 2001: 149). For more accurate differentiation, Roux (1984) introduces the terms “temple-trésors” and “temple-sanctuaire”.

84 According to this interpretation, we cannot rule out a function as treasury for the three buildings. The third possible identification as buildings for banqueting is even more problematic, since dining halls are typologically variable; Roux (1973: 538) states simply, “il n’existe pas d’hestiatorion type” (and see Leypold 2008: 176–85). Fortunately, some of the oikoi—called “small buildings” in Delphoi, for example—may be identified as dining buildings thanks to epigraphical and archeological evidence (Hellmann 1992: 300–4; also see Bruneau and Ducat 2005: 171). The temple-like Buildings A and B at Labraunda even have inscriptions on their architraves denominating them as andrones (Hellström 1990). Usually, however, the use of a room for banqueting is identified by interior features, primarily the floor construction. In this view, the central floor mosaic inside Building D that is surrounded by a slightly raised pebble floor—a form typical of dining rooms—is strong evidence for such an identification (Börker 1983; Leypold 2008: 142–75, esp. 148–50, 163–64). In the cella of Building D, nine symmetrically arranged klinai of 170–175 × 80 cm can be reconstructed on the floor, which is raised by 1 cm. In conclusion, the “naïskoi” at Aigeira present some evidence for use as temples as well as for dining, whereas a function as treasuries seems less likely but is not entirely excluded. Could they possibly have been multifunctional buildings? Lauter (1986: 191, 193) generally denies any additional uses for Hellenistic naïskoi besides as a temple, although he notes the similarity to treasuries: “Dem Wesen und der Konvention nach sind kleinere Prostyloi und Antenbauten Gehäuse für das in ihrer Cella Aufzubewahrende. Als Tempel sind sie bzw. ihre Cellen demnach Gehäuse für die Kultbilder und sonst nichts”. Tonio Hölscher excludes an additional function as a dining building for the Olympian treasuries due to the central position of the original cult statue (Hölscher 2001: 149). For Building D with its statue base in the rear of the cella, besides its evident cult function, at least a temporary use as a dining room appears likely. Moreover, the andrones at Labraunda, which seem to have had both sacred and dining functions, provide a parallel for the “naïskoi” at Aigeira. Pontus Hellström (1990: 252) remarks: “It thus appears that temple-like buildings can be banqueting halls even if there is a niche or a base for a statue at the back wall”. If we look back to the pre-archaic “Herdhaus”, or “temple-hestiatorion”, we encounter a multi-functional building used for cult sacrifice and social-political meetings (Drerup 1969: 123–28; see also Börker 1983: 10 and Mazarakis Ainian 1988: 116–19). Over the course of the Archaic period, according to Mazarakis Ainian (1988: 118), the different functions were separated from each other

Tanner

and attributed to separate buildings: “Little by little the three primordial functions of early Greek temples were separated from each other: the temple itself remained the house of the divinity in which the cult image was kept while distinct edifices, serving as treasuries and hestiatoria, were erected in the proximity of the temple”. Perhaps this separation of function did not occur uniformly everywhere in the Greek world, or it was no longer maintained by Hellenistic and later times. Inge Nielsen (2007: 34–6) draws a parallel between the multi-functional Hellenistic and Roman buildings—used by religious associations for meetings, banquets, and cult—through the “templehestiatorion”, and, in the Roman era, multi-functionality in temples became even more common, with uses including not only religious ceremonies, but also meetings of all kind, archives, art expositions, etc. (see, e.g., Anderson 1997: 242–7). Although of a more civic character than those buildings for private religious associates, the small buildings at Aigeira could have been similarly multi-functional in a later phase. After all, they seem to have served more purposes than solely as shrines for a museum-like presentation of cult images, or as miniature temples (Lauter 1986: 194–6; Cain 1995: 123–5). 6 Conclusions The current research has led to new results concerning the architectural design and relative chronology of Buildings D and E, and the nature of the entire architectural ensemble. The holistic approach to the study of the three so-called naïskoi D, E, and F including the layout of the entire group of buildings and results of material studies by other scholars—such as, for example, the figural terracottas—has led to new insights regarding the planning and function of the buildings. Despite the lack of abundant architectural blocks of the upper structure, we may offer possible reconstructions of the buildings and their life histories. The detailed examination of the evidence has shown that Building E was constructed before its neighbour, Building D. Building E as well as the theatre and probably a stoa preceding Building F would have been the first constructed in the area using the same measuring unit, whereas the one identified in Building D is distinct from them. Despite the chronological differences, Buildings D and E were still closely related to each other, as witnessed by their aligned prostyle porches of similar width—at least during their first period of coexistence. This joint conception of the buildings must have originated in the course of third century BCE, the time of the construction indicated

THREE HELLENISTIC ‘ Naïskoi ’ IN THE THEATRE AREA AT AIGEIRA

by the floor mosaic of Building D. Building F, dateable to the second century BCE by its floor mosaic, followed the typological model of Building D. Furthermore, the differences between the three buildings concerning building techniques, dimensional units, typology, and interior layout indicate design and construction at different points in time, and by different people or groups of people, public or private. If the relative sequence shows that the theatre and Building E are contemporary and probably earlier than the middle of the third century BCE, the accepted dating of the theatre to ca. 250 BCE should also be reconsidered (Gogos 1992: 119; see also Gauss et al. 2015b: 268–69). Furthermore, it is unclear when Aigeira joined the Second Akhaian League founded in 281/280 BCE, though this probably was only towards the middle of the third century (Löbel 2014: 42–45, 405). The population transfer from the abandoned polis Aigai to Aigeira in the second half of the fourth century might have led to urban construction projects in the theatre area (on Aigai, see Löbel 2014: 44; Rizakis 2016: 23–4) while further building activity took place after Aigeira joined the League. To conclude, the three temple-like buildings at Aigeira each appear to have had different functions, some designed as multi-functional buildings. Combined uses for cult activity, the display of dedications, and dining is at least possible for Buildings D and F. Nevertheless, due to the lack of attributed cult statues, altars or inscriptions, the deities and ritual conducted in this area remain unknown. Whereas the early Hellenistic ensemble comprising one naïskos is typical for a local sanctuary, the later arrangement of the architecture in the theatre area does not emphasise one single outstanding building that could be regarded as the main temple, as we would usually expect in a Hellenistic sanctuary—as in, for example, the Asklepieon of Kos, the Sanctuary of Zeus at Labraunda, or the sanctuary of Dodona. Nevertheless, since a temple is not a requirement for a Greek sanctuary, the group of buildings could still form a sanctuary, regardless of their particular functions. Furthermore, the presence of multiple small temple-like buildings indicates cult activity as well as dining in the framework of the enlarged sanctuary of later Hellenistic times. List of References Alzinger, W., 1972–1975: “Aigeira – Achaia”, ÖJh 50, Grabungen 1971/72: 9–16. Alzinger, W., 1985: “Aigeira 1984”, ÖJh 56, Grabungen 1984: 10–12. Alzinger, W., 1988: “Aigeira 1987”, ÖJh 58, Grabungen 1987: 11–13.

85

Alzinger, W., 1990: “Die hellenistischen Tempel von Aigeira”, in the Deutsches Archäologisches Institut (ed.), Akten des XIII. Internationalen Kongresses für Klassische Archäologie Berlin 1988 (Mainz) 549–51. Alzinger, W., Gogos, S. & Trummer, R., 1986: “Aigeira-Hyperesia und die Siedlung Phelloë in Achaia. Österreichische Ausgrabungen auf der Peloponnes 1972–1983. Teil II: Theater und Umgebung”, Klio 68.1: 5–62. Anderson, J.C., 1997: Roman Architecture and Society (Baltimore). Aurenhammer, M., (in press): “Die Skulptur aus dem öffentlichen Zentrum von Aigeira”, in Gauss in press. Bammer, A., 1993: “Aigeira”, ÖJh 62, Grabungen 1982: 33–37. Bammer, A., 2001: “Neue Heiligtümer in Aigeira”, in MitsopoulosLeon, V. (ed.), Forschungen in der Peloponnes. Akten des Symposions anlässlich der Feier „100 Jahre Österreichisches Archäologisches Institut Athen”, Athen 5.3.–7.3.1998. ÖAI Sonderschriften 38 (Athens) 95–105. Bommelaer, J.-F., 2015: Guide de Delphes: le site. 2nd ed. (Athens). Börker, C., 1983: Festbankett und griechische Architektur. Xenia. Konstanzer althistorische Vorträge und Forschungen 4 (Konstanz). Bruneau, P. & Ducat, J., 2005: Guide de Délos, 4th ed. (Athens). Burkert, W., 1988: “The Meaning and Function of the Temple in Classical Greece”, in Fox, M.V. (ed.), Temple in Society (Winona Lake) 27–47. Cain, H.-U., 1995: “Hellenistische Kultbilder. Religiöse Präsenz und museale Präsentation der Götter im Heiligtum und beim Fest”, in Wörrle, M. & Zanker, P. (edd.), Stadtbild und Bürgerbild im Hellenismus. Kolloquium, München, 24. bis 26. Juni 1993. Vestigia 47 (Munich) 115–30. Coulton, J.J., 1975: “Towards Understanding Greek Temple Design: General Considerations”, BSA 70: 59–99. Courby, F. & Picard, C., 1924: Recherches archéologiques à Stratos d’Acarnanie (Paris). Drerup, H., 1969: Griechische Baukunst in geometrische Zeit. ArchHom II:O (Göttingen). Emmerling, T.E., 2012: Studien zu Datierung, Gestalt und Funktion der ‘Kultbauten’ im Zeus-Heiligtum von Dodona. Antiquitates—Archäologische Forschungsergebnisse 58 (Hamburg). Evangelidis, D., 1929: “Η ανασκαφή της Δωδώνης”, Prakt 1929: 104–29. Felten, F., 1996: “Griechische Heiligtümer in hellenistischer Zeit”, in Blakolmer, F., Krierer, K.R., Krinzinger, F., Landskron-Dinstl, A., Szemethy, H.D. & Zhuber-Okrog, K. (edd.), Fremde Zeiten: Festschrift für Jürgen Borchhardt zum sechzigsten Geburtstag am 25. Februar 1996 dargebracht von Kollegen, Schülern und Freunden vol. 2 (Vienna) 139–58. Gauss, W., (ed.), (in press): Forschungen im Bereich des Theaters von Aigeira, 2011–2018. Forschungen in Aigeira 3.

86 Gauss, W., Smetana, R., Dorner, J., Eitzinger, P., Forstenpointner, G., Galik, A., Kurz, A., Lätzer-Lasar, A., Leibetseder, M., Regner, C., Tanner, A., Trapichler, M. & Weissengruber, G., 2012: “Aigeira 2011. Bericht über Aufarbeitung und Grabung”, ÖJh 81: 33–50. Gauss, W., Smetana, R., Rutter, J.B., Dorner, J., Eitzinger, P., Klein, C., Kurz, A., Lätzer-Lasar, A., Leibetseder, M., Regner, C., Stümpel, H., Tanner, A., Trainor, C., & Trapichler, M., 2013: “Aigeira 2012. Bericht über Aufarbeitung und Grabung”, ÖJh 82: 69–91. Gauss, W., Smetana, R., Rutter, J.B., Regern, C., Rusch, K., Stümpel, H., Rabbel, W., Ruppenstein, F., Heiden, J., Leibetseder, M., Tanner, A. & Hinker, C., 2015a: “Aigeira 2013–2014. Bericht über Aufarbeitung und Grabung”, ÖJh 84: 11–50. Gauss, W., Smetana, R., Dorner, J., Lätzer-Lasar, A., Leibetseder, M. & Trapichler, M., 2015b: “Old and New Observations from the Theatre at Aigeira”, in Frederiksen, R., Gebhard, E.R. & Sokolicek, A. (edd.), The Architecture of the Ancient Greek Theatre. Acts of an International Conference at the Danish Institute at Athens 27–30 January 2012 (Aarhus) 267–77. Ginouvès, R., 1956: “Notes sur quelques relations numériques dans la construction des fondations de temples grecs”, BCH 80: 104–17. Gogos, S., 1992: Das Theater von Aigeira. Ein Beitrag zum antiken Theaterbau. ÖAI Sonderschriften 21 (Vienna). Gogos, S., 2001: “Das Theater von Aigeira. Ein Beitrag zur Chronologie des Zeus-Heiligtums”, in Mitsopoulos-Leon, V. (ed.), Forschungen in der Peloponnes. Akten des Symposions anlässlich der Feier „100 Jahre Österreichisches Archäologisches Institut Athen”, Athen 5.3.–7.3.1998. ÖAI Sonderschriften 38 (Athens) 79–87. Hagn, T., 2001: “Das Tycheion von Aigeira und daran anschliessende Bauten”, BCH Suppl. 39: 297–311. Hainzmann, M., 2001: “Hyperesia/Aigeira—Eine historische Spurensuche”, in Mitsopoulos-Leon, V. (ed.), Forschungen in der Peloponnes. Akten des Symposions anlässlich der Feier “100 Jahre Österreichisches Archäologisches Institut Athen”, Athen 5.3.–7.3.1998. ÖAI Sonderschriften 38 (Athens) 73–78. Hellmann, M.-C., 1992: Recherches sur le vocabulaire de l’architecture grecque, d’après les inscriptions de Délos. BÉFAR 278 (Athens). Hellström, P., 1990: “Hellenistic Architecture in Light of Late Classical Labraunda”, in the Deutsches Archäologisches Institut (ed.), Akten des XIII. Internationalen Kongresses für Klassische Archäologie Berlin 1988 (Mainz) 243–52. Hennemeyer, A., 2013: Das Athenaheiligtum von Priene: die Nebenbauten—Altar, Halle und Propylon—und die bauliche Entwicklung des Heiligtums. AF 27 (Wiesbaden). Hölscher, T., 2001: “Schatzhäuser—Banketthäuser?”, in Böhm, S. & von Eickstedt, K.-V. (edd.), ITHAKE. Festschrift für Jörg

Tanner Schäfer zum 75. Geburtstag am 25. April 2001 (Würzburg) 143–52. Koenigs, W., 1990: “Masse und Proportionen in der griechischen Baukunst”, in Beck, H., Bol, P.C. & Bückling, M. (edd.), Polyklet: der Bildhauer der griechischen Klassik. Ausstellung im Liebieghaus, Museum alter Plastik, Frankfurt am Main (Mainz) 121–34. Koenigs, W., 2015: Der Athenatempel von Priene. AF 33 (Wiesbaden). Lafond, Y., 1998: “Die Katastrophe von 373 v. Chr. und das Verschwinden der Stadt Helike in Achaia”, in Olshausen, E. & Sonnabend, H. (edd.), Naturkatastrophen in der antiken Welt: Stuttgarter Kolloquium zur historischen Geographie des Altertums 6, 1996. Geographica historica 10 (Stuttgart) 118–23. Lauter, H., 1986: Die Architektur des Hellenismus (Darmstadt). Lauter-Bufe, H., 2008: Das Heiligtum des Zeus Soter in Megalopolis (Mainz). Lauter-Bufe, H., 2008: Das Heiligtum des Zeus Soter in Megalopolis (Mainz). Leibetseder, M., (in press): “Die Wandmalerei aus Naiskos E”, in Gauss in press. Leypold, C., 2008: Bankettgebäude in griechischen Heiligtümern (Wiesbaden). Löbel, Y., 2014: Die Poleis der bundesstaatlichen Gemeinwesen im antiken Griechenland. Untersuchungen zum Machtverhältnis zwischen Poleis und Zentralgewalten bis 167 v. Chr. Studi di storia greca e romana 10 (Alexandria). Mancini, L., 2013: “Templi, thesauroi, ‘temples-trésors’. Note sull’edilizia templare non periptera nei santuari dell’Epiro ellenistico”, Ocnus. Quaderni della Scuola di Specializzazione in Beni Archeologici 21: 75–99. Mattern, T., 2015: Das Herakles-Heiligtum von Kleonai: Architektur und Kult im Kontext. Kleonai 1 (Wiesbaden). Mazarakis Ainian, A.J., 1988: “Early Greek Temples: Their Origin and Function”, in Hägg, R., Marinatos, N. & Nordquist, G.C. (edd.), Early Greek Cult Practice. Proceedings of the fifth International Symposium at the Swedish Institute at Athens, 26–29 June, 1986. ActaAth 4°:38 (Stockholm) 105–119. Müller-Wiener, W., 1988: Griechisches Bauwesen in der Antike (Munich). Müth, S., 2007: Eigene Wege: Topographie und Stadtplan von Messene in spätklassisch-hellenistischer Zeit. Internationale Archäologie 99 (Rahden). Neer, R.T., 2001: “Framing the Gift: The Politics of the Siphnian Treasury at Delphi”, ClAnt 20.2: 273–344. Nielsen, I., 2007: “Vorbilder für Räumlichkeiten der religiösen Vereine hellenistischer und römischer Zeit”, in Nielsen, I. (ed.), Zwischen Kult und Gesellschaft: Kosmopolitische Zentren des antiken Mittelmeerraumes als Aktionsraum von Kultvereinen und Religionsgemeinschaften: Akten eines Symposiums

THREE HELLENISTIC ‘ Naïskoi ’ IN THE THEATRE AREA AT AIGEIRA des Archäologischen Instituts der Universität Hamburg. 12.–14. Oktober 2005. Hephaistos 24 (2006) (Augsburg) 31–46. Osanna, M., 1996: Santuari e culti dell’Acaia antica. Aucnus 5 (Naples). Pakkanen, J., 2013: Classical Greek Architectural Design: A Quantitative Approach. Papers and monographs of the Finnish Institute at Athens 18 (Helsinki). Piccinini, J., 2016: “Renaissance or Decline? The Shrine of Dodona in the Hellenistic Period”, in Melfi, M. & Bobou, O. (edd.), Hellenistic Sanctuaries: Between Greece and Rome (Oxford) 152–69. Quantin, F., 2008: “Recherches sur l’histoire et l’archéologie du sanctuaire du Dodone. Les oikoi, Zeus Naios et les Naia”, Kernos 21: 9–48. Rizakis, A.D., 2016: “Territorio e riorganizzazione delle poleis nell’ Achaia orientale da Omero a Strabone”, in Pontrandolfo, A. (ed.), Egialea. Richerche nella Valle del Krios. 1. Monografie della Scuola Archeologica di Atene e delle Missioni Italiane in Oriente 24 (Athenes) 19–30. Roux, G., 1973: “Salles de Banquets à Délos”, BCH Suppl. 1: 525–54. Roux, G., 1984: “Trésors, temples, tholos”, in Roux, G. (ed.), Temples et sanctuaires: Séminaire de recherche 1981–1983. Travaux de la Maison de l’Orient 7 (Lyon) 153–71. Rusch, K., Stümpel, H., Klein, C., Rabbel, W. & Gauss, W. (in press): “Bericht zu den geophysikalischen Untersuchungen”, in Gauss in press. Salzmann, D., 1982: Untersuchungen zu den antiken Kieselmosaiken: von den Anfängen bis zum Beginn der Tesseratechnik. AF 10 (Berlin).

87

Scheibelreiter-Gail, V., (in press): “Mosaiken in Aigeira”, in Gauss in press. Schrettle, B. 2007. Prostyle Tempel. Untersuchungen zur Herausbildung und Entwicklung des Bautyps in Griechenland (diss., Karl-Franzens-Universität Graz). Schwertfeger, T., 1974: Der Achaiische Bund von 146 bis 27 v. Chr. Vestigia 19 (Munich). Smetana, R. (in press): “Die Terrakotten aus dem Bereich des Theaters und der Naiskoi”, in Gauss in press. Solima, I., 2011: Heiligtümer der Artemis auf der Peloponnes. Studien zu antiken Heiligtümern 4 (Heidelberg). Svenson-Evers, H., 1997: “Ieros Oikos. Zum Ursprung des griechischen Tempels”, in Hoepfner, W. (ed.), Kult und Kultbauten auf der Akropolis (Berlin) 132–51. Tanner, A., (in press): “Die ‘Naiskoi’ D und E im Theaterbereich von Aigeira”, in Gauss in press. Walter, O., 1919: “Eine archäologische Voruntersuchung in Aigeira”, ÖJhBeibl 19–20: 5–42. Walter, O., 1932: “Versuchsgrabung in Aigeira”, ÖJhBeibl 27: 223–34. Wilson Jones, M., 2000: “Doric Measure and Architectural Design 1: The Evidence of the Relief from Salamis”, AJA 104.1: 73–93. Zimmermann, M., 2015: “Die hellenistische Polis in neuer Perspektive. Das DFG-Schwerpunktprogramm zur hellenistischen Polis zieht Bilanz”, in Matthaei, A. & Zimmermann, M. (edd.), Urbane Strukturen und bürgerliche Identität im Hellenismus. Die hellenistische Polis als Lebensform 5 (Heidelberg) 7–11.

part 2 Life History of Greek Monuments and Sites

∵

chapter 5

The Small Limestone Buildings on the Akropolis of Athenai Nancy L. Klein 1

Introduction: A New Investigation of the Small Limestone Buildings1

likely to have belonged to two additional buildings. Their form and decoration are Ionic, and probably reflect influence coming from the Kyklades in the Archaic and early Classical period.

The current study was undertaken in order to re-examine the architectural fragments from the small limestone buildings that came to light during the 19th century excavations on the Athenian Akropolis. The unexpected discovery of fragmentary architecture and sculpture, among other objects, in layers of fill and built into foundations, presented a challenge for scholars seeking to understand their meaning and the processes which resulted in their deposition and reuse. In his 1904 publication, Theodor Wiegand described five small buildings, designated Buildings A–E, as well as two large temples, but did not provide a complete catalogue or thorough description of the relevant architectural blocks. As a result, many questions remain about the number of individual buildings, their appearance, and their date of construction. Since the individual blocks are a primary source of information, the present study began with a close examination of the preserved fragments, their design, tool marks, and condition. This was followed by a critical evaluation of the basis for assigning blocks to a specific building, including dimensions, material, decoration, and design. The architectural blocks were also evaluated as artifacts, evidence of active processes such as discarding useless objects, while caching and recycling others—the research using excavation reports, archival documents, and other publications. The benefits of this approach are presented in three case studies. In the first, careful observation of the condition of the blocks of Building A facilitates a better understanding of the construction methods, evidence for repairs, and the final demolition of the structure. The second case study examines the reuse of architectural elements from several different buildings in the foundations of the Propylaia as a means to understanding how obsolete or damaged buildings were intentionally dismantled and their blocks cached and repurposed for use in a new construction. In the third case study, a reconsideration of Wiegand’s Buildings C and D suggests that some of the blocks he assigned to their reconstruction are more

The monuments of the classical Akropolis of Athenai, built under the leadership of Perikles and Phidias in the second half of the fifth century BCE and the focus of excavation, research, and conservation over the last two centuries, stand as symbols of the artistic and cultural achievements of ancient Greece. The architectural development of the Akropolis during the preceding sixth and early fifth centuries, however, has been eclipsed by the brilliance of the classical building program. Prior to the 19th century there was very little visible evidence of the early sanctuary. The limestone Doric entablature and unfinished marble drums in the north wall of the Akropolis gave proof of an earlier sanctuary, but the excavations carried out under the direction of Panayiotis Eustratiadis and Panayiotis Kavvadias dramatically changed our understanding of the pre-Persian Akropolis. Their excavation of the deep layers of fill to the east and south of the Parthenon foundations brought to light fragments of limestone architecture and sculpture, among other finds. In 1882, two small pediments in low relief with Herakles fighting the Hydra (Acr. Inv. 1: Brouskari 1974: 29, figs. 15–17) and the struggle between Herakles and Triton (Acr. Inv. 2 “Red Triton Pediment”: Brouskari 1974: 37, fig. 47) were found in this area (Kavvadias & Kawerau 1906: 18; Lechat 1888: 239; Stewart 2008: 393–400). Kavvadias subsequently described the stratigraphy to the east of the Parthenon as follows: “It appears … that the last [lowest] layer was the soil of the Akropolis, which in earliest times covered the bedrock. On top of this were thrown heaps of fragments of old poros buildings unknown to us, as well as decorated reliefs and sculptures, that constituted the poros layer. On top of this was another layer, in which was found mostly marble and other things” (Kavvadias 1888: 10–11; Kavvadias & Kawerau 1906).2

1  I would like to express my gratitude to the Hellenic Ministry of Culture and Sports for the permission to study the small limestone buildings from the Akropolis of Athenai.

2  “Ἐκ τούτου φαίνεται ὃτι τὸ τελεταῖον τοῦτο στρῶμα ἦτο τὸ ἔδαφος τῆς Ἀκροπόλεως, ὃπερ ἐκ παλαιτάτων χρόνων ἐκάλυπτε τὸν βράχον. Ἐπι τούτου ἐρρίφθησαν σωρηδὸν θραύσματα ἀγνώστων ἡμῖν ἀρχαιοτάτων

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004416659_007

2

History of Excavation and Scholarship

92

Klein

The presence of older, “pre-Persian [destruction]” architecture was noted by the excavator and other archaeologists and, although it is difficult to determine where individual blocks were found based upon these early reports, some elements were distinctive enough to merit special mention. Both Penrose (1887) and Gardner (1889) described pieces of a limestone column with spiral flutes, and Penrose mentioned a painted Doric cornice with white guttae “inserted like so many pegs” that was found in the excavations in the southeast corner of the Akropolis in 1887 (Penrose 1887: 271; Gardner 1889: 260). Today we know these may come from Wiegand’s Building A, which has separately inserted guttae on all of its geison blocks. In his publication, Wiegand mentioned that the two geison blocks from Building E were discovered in 1887, so it is likely that they came from this area as well (Acr. Inv. 4387, 4388: Wiegand 1904: 169–70, fig. 169). In 1888, excavation continued to the east of the Parthenon, then moved to the south in the area between its foundations and the south wall of the Akropolis. Here, the foundations of Building VI, the so-called Ergasterion were discovered and found to reuse unfinished column drums from the Older Parthenon. Drawings of the foundations made by Georg Kawerau at the time of excavation and published by J.A. Bundgaard in 1974 note the presence of additional reused Doric elements as well (Bundgaard 1974: 47, pl. 157 [Kawerau Nachlass p. 183]). Other recognisable blocks that were found in the area and drawn by Kawerau include a geison with a small half round crown moulding that is now assigned to Wiegand’s Building C (Bundgaard 1974: 53, pl. 216.1; Wiegand 1904: 162–66, figs. 156, 159, 161), and a geison with cyma reversa soffit moulding that he assigned to the raking geison of his Building D (Bundgaard 1974: 22, 48, pl. 162; Wiegand 1904: 166–68, fig. 165; see further discussion of these geison blocks below). Fragments from several limestone pediments were discovered in the area as well, including the introduction of Herakles to the Olympian gods (Acr. Inv. 9: Brouskari 1974: 34, figs. 27–29), the “Olive Tree Pediment” (Acr. Inv. 52: Brouskari 1974: 42, fig. 74), a lioness attacking a bull (Acr. Inv. 4: Brouskari 1974: 28, fig. 14), two lions attacking a bull (Acr. Inv. 3: Brouskari 1974: 46, figs. 80–81), two large, coiling snakes (Acr. Inv. 37–40: Brouskari 1974: 36, figs. 44, 46), part of “Bluebeard” (Acr. Inv. 35: Brouskari 1974: 39–40, fig. 55), and the figures

πωρίνων οἰκοδομημάτων, ἅτινα δι᾽ἀναγλύφων (σημειωτέον ὃτι ἐνταῦθα που εὑρέθησαν καὶ τὰ ἐν σελ. 17 μνημονευόμενα ἀετώματα ἐξ ἀναγλύφων) καὶ παντοίον ἀγαλμάτων θὰ ἦσαν κεκοσμημένα, καὶ οὓτος ἀπετελέσθη τὸ πώρινον στρῶμα. Ἐπὶ τοῦ πωρίνου τούτο στρὠματος ἐγἐνετο καὶ ἄλλη ἐπίχωσις, ἐν ᾗ εὑρίσκονται κυρίως μαρμάρινα καὶ ἄλλα τοιαῦτα πράγματα, οἷα τὰ μέχρι τοῦ βράχου εὑρισκόμενα ἐν τῇ κατὰ τὸ βόρειον τεῖχος τῆς Ἁκροπόλεως σκαφῇ” (Kavvadias & Kawerau 1906: 35).

of Herakles and Triton (Acr. Inv. 36: Brouskari 1974: 39–40, figs. 54, 56). In 1889, the excavation of fill inside of the Northwest Wing of the Classical Propylaia revealed that its foundations reused architectural elements from several older Doric buildings. Kavvadias (1889: 105) mentioned limestone geison and triglyph blocks in the south and east walls, as well as the foundation for an extension to the north. Some of the geison blocks were curved and preserved traces of colour. Kawerau made annotated drawings that document the types of blocks and the coursing used in the construction of the foundations (Bundgaard 1974: 33–34, pls. 1–9). Wiegand later assigned the curved blocks to an apsidal structure Building B, while other Doric geisa match the blocks with a half round crown moulding found to the south of the Parthenon and assigned to Building C. Additional reused limestone architectural blocks were also discovered in the foundations of the nearby pre-Mnesiklean cistern and the Northwest Building (Kavvadias & Kawerau 1906: 84–88). Although the methods of the 19th century excavations have left us with many questions, it is clear that valuable information can still be obtained from a close consideration of the observations of the excavators, the archaeologists who visited the Akropolis, and the surviving documentation from this period. These sources indicate that pieces of the small limestone buildings A, C, D, and E were recovered from the layers of fill to the east and southeast of the Parthenon, while the building projects in and around the Mnesiklean Propylaia reused architectural blocks from Buildings B, C, and other structures in their foundations. In 1904, Wiegand published the primary study of archaic limestone architecture from the Akropolis. He used a formal approach based on block dimensions, decorative details, and an understanding of Doric design to assemble the fragmentary remains into distinct structures. In addition to two large temples, the “alte Athenatempel in antis (Hekatompedon)” and the “‘Peisistratische’ Peripteralbau”, he recognised several smaller Doric buildings that he designated A, B, C, D, and E. Numerous other fragments of architecture and sculpture were noted as well but were not assigned to specific buildings. Wiegand’s methodology was typical of the time: he offered a qualitative analysis of the material, described the diagnostic features of each building, illustrated some of the most important elements, and provided a reconstruction of the façade elevation. A parallel study of fragments of architectural sculpture led him to reconstruct pedimental compositions and propose their association with specific buildings. In many ways, Wiegand’s publication is an essential resource for the study of the limestone architecture, but his conclusions must be re-examined in light of subsequent

THE SMALL LIMESTONE BUILDINGS ON THE AKROPOLIS OF ATHENAI

research and advances in methodology. For example, Wiegand did not publish a comprehensive catalogue of architectural members, as is now common practice, so that the evidence for his reconstructions is not always explicit. The drawings and photographs that were, for their time, an important advance in archaeological illustration and publication are not as numerous or accurate as is possible today. Wiegand also faced an immense task as the first person to sort through hundreds of architectural fragments and assign them to individual structures. While the overall division of blocks was largely correct, he appears to have proposed the association of some blocks based on tenuous evidence, which led to some dubious combinations. Finally, the richness and diversity of archaic Greek architecture had not yet been documented or fully appreciated. Wiegand had few archaic temples with which to compare the small limestone buildings and was left to make inferences based upon later, better-known, classical forms. 3

The “Life History” of Building A

This study of the small limestone buildings is based on a detailed examination of the extant architectural blocks in the Akropolis storerooms and the Akropolis Museum. A primary goal is to collect information from each block

93

to reconstruct a “biography” or “life history” of a building, beginning with methods of construction and the original appearance of the structure, indications of damage and repair, and ending with evidence related to the terminal phases of dismantling and reuse. The consideration of the archaeological contexts and depositional processes that shaped the Classical Akropolis shed additional light on the life history of the small buildings as well. Building A is the oldest of the small limestone buildings (ca. 560 BCE) and has the greatest number of blocks preserved from its entablature, so it will provide a case study for this approach. The reconstructed entablature that now stands in the Akropolis Museum uses only a few of the dozens of fragments preserved from the structure and a closer look at the blocks, by comparison with Wiegand’s publication, will demonstrate how much information can be gathered about the building (Fig. 5.1). The view of the top of the corner geison (Acr. Inv. 4503) reveals several cuttings and tool marks (Fig. 5.2). Two U-shaped channels would have held ropes used to lift or maneuver the block into position at the time of construction. At the far left is a deep cutting that is missing from Wiegand’s drawing. Its shape and depth correspond to the iron pieces used in the previous museum display, which suggests it was made for the 20th century installation and is not ancient. Along the edge, the top surface has been carefully finished and

figure 5.1 Athens, Akropolis Museum, Building A, Acr. Inv. 4510 (left) and 4503 (right) photo by author

94

Klein

Modern cutting

Holes to attach sima

U-shaped channels

figure 5.2 Athens, Akropolis Museum, Building A, Corner Geison Inv. 4503. View of top showing ancient holes to attach sima, U-shaped channels, and modern cutting at left photo by author

two small holes were drilled for small metal dowels to attach a lateral sima. Similar holes are found in other lateral and raking geison blocks, and one (Acr. Inv. 4409, Fig. 5.3) has a pair of holes, one empty, and the other with a flattened metal dowel preserved in situ, which demonstrate that a sima was secured along the flanks and facades of the building. The spacing of the holes on the geison top surface match those of a small marble sima with rosettes in low relief, fragments of which were also discovered in the layers of fill around the Akropolis (Dörpfeld 1886: 168). Although Dörpfeld does not illustrate specific examples, he mentioned seven marble simas, some of which were painted, while others were carved in low relief and painted (see Ohnesorg 1993: 14–15, pls. 69, 7–8, Plastischen Rosettensima). Several pieces are included in the museum display. Traces of claw chisel are also apparent along the lateral joint of Acr. Inv. 4409. Traces of this tool can be found on some, but not all, of the blocks from Building A. These marks show that the claw chisel was in use, but the flat chisel was far more common (Paga 2012/2013 provides a discussion of the use of the claw chisel in Athenai, especially as it relates to the location of the H-Architecture or Bluebeard temple, but without examples of its use on limestone architectural elements from the Akropolis). The corner of the building (Acr. Inv. 4503, Fig. 5.4) shows evidence of repair. A small piece was broken out of the crown moulding of the geison on the flank and was patched with a piece of limestone attached with two lead

pins, still in situ. The corner triglyph below had imperfections in the stone that may have appeared when the glyph on the left was cut. These cavities were filled with molten lead when the block was still on the ground, prior to its placement on the building. Many other blocks reveal areas that were damaged and repaired. Two other blocks show how lead pins were used to attach small pieces of stone to areas that had been damaged. The stone of lateral geison block Acr. Inv. 7390 (Fig. 5.5) has a flaw from top to bottom that probably appeared when it was being dressed. Despite the flaw, the block was finished as shown by the sculpted and painted crown moulding and traces of stucco on the face. The top of the block shows that an effort was made to bridge the crack with a sort of lead staple to prevent the block from breaking apart. The nature and location of these repairs to the preserved blocks from Building A suggest that the damage occurred during the construction or lifetime of the building and efforts were made to both fix and conceal the problems. Based on the quantity and variety of preserved architectural elements, it is possible to propose a reconstruction of the façade entablature of Building A (Fig. 5.6b). My proposal is not substantially different from Wiegand’s drawing (Fig. 5.6a), with one exception. Based upon his understanding of Doric architecture, Wiegand felt that there should have been a soffit moulding below the raking geison and proposed the inclusion of a separate block

95

THE SMALL LIMESTONE BUILDINGS ON THE AKROPOLIS OF ATHENAI

Hole for dowel to attach sima

Claw chisel along top edge of lateral joint Metal dowel to attach sima still in place figure 5.3 Athens, Akropolis, Building A, Geison Inv. 4409. Top of block with hole for dowel to attach sima, metal dowel still in situ, and detail of claw chisel along top edge of lateral joint photos by author

figure 5.4 Athens, Akropolis Museum, Building A, Corner Entablature Acr. Inv. 4503 (left). Detail of repair to crown moulding (top right), lead repair to triglyph (bottom right) photos by author

96

figure 5.5 Athens, Akropolis, Building A, Lateral Geison Inv. 7390. At left, view of face showing crack in face of block, traces of stucco on taenia and partially preserved mutule. At right, view of top surface with pair of holes to attach sima at back, one of two holes for repair along front edge, and traces of claw chisel at right photos by author

figure 5.6 Top: Reconstruction of Building A façade with four supports. (Wiegand 1904: pl. XIII); Bottom: Reconstruction of Building A façade with four supports illustration by D.G. Mullen

Klein

THE SMALL LIMESTONE BUILDINGS ON THE AKROPOLIS OF ATHENAI

97

figure 5.7 Building A: (a) detail of pediment corner with stepped and sloping soffit of horizontal geison; (b) horizontal geison, Acr. Inv. 4415. Building with Painted Pediments: (c) horizontal geison, Acr. Inv. 4464 illustration by D.G. Mullen; photos by author

with the tympanon moulding. Subsequent research has shown that early Doric buildings do not always have mouldings in this location and, more importantly, the scale and length of preserved blocks from this tympanon moulding cannot be accommodated in a building of this size. The tympanon mouldings have also been studied by Heberdey (1919: 146–54), who divided them between two small buildings, and Beyer (1974: 639–45), who assigned them to the Old Athena Temple [Hekatompedon]. A detail of the new model shows a reconstruction of the corner pediment with the raking geison abutting the corner block (Fig. 5.7a). Below is a stepped and sloping top surface of the soffit, as demonstrated in the horizontal geison block Acr. Inv. 4415 (Fig. 5.7b). A horizontal geison with sloping soffit is also seen on another limestone geison from the Akropolis belonging to a building with painted pediments (Fig. 5.7c) (Wiegand 1904: 230–31, fig. 245, pl. VI 1–3; Heberdey 1919: 125–27, figs. 167, 168), and two buildings at Olympia (for the Treasury of Syrakousai, see Mallwitz 1961: 42 n. 17, pls. 7, 9, 11; the same characteristic is seen in two geison blocks at Olympia discussed by Herrmann (1976: 326–29) and attributed to his “Seilöhr Bau”). The current condition of the extant blocks tells us about the end of the building’s functional life, when

the entablature was intentionally broken into pieces. Although rarely studied as a significant phase in a building’s life, the dismantling of a masonry structure would have required considerable effort, comparable to that of its construction. Steps must have included the assessment and management (reuse or disposal) of building materials, the last of which can be observed as depositional processes in the archaeological record. Indications of the demolition process are seen in the lateral geison in the Akropolis Museum (Acr. Inv. 4510, Fig. 5.8). A triangular scar marks the spot where a heavy tool, such as a pick, struck the face of the block in order to break it apart. Below, the top corner edges of each triglyph were broken away, perhaps to facilitate the removal of the metope slabs that were slotted into the entablature. The result of such action can be appreciated by looking at how the individual blocks of the entablature from Building A were systematically broken into pieces when the building was taken down. In fact, there are few if any complete, intact blocks preserved from Building A, and their reduced size is significant. Such comprehensive reduction of the entablature suggests a deliberate process of assessment and demolition, with the intention of using its pieces as fill, not as repurposed building blocks.

98

Klein

figure 5.8 Athens, Akropolis Museum, Building A, Inv. 4510. Arrows indicating damage photos by author

4

Buildings B, C, and the Propylaia Foundations

In 1889, excavations within the Northwest Wing of the Mnesiklean Propylaia and the unrealised Northeast Hall revealed evidence of previously undocumented older buildings whose blocks were reused in the foundations (Kavvadias 1889; Kavvadias & Kawerau 1906: 42, 58, 60; Bundgaard 1974: pl. 3). Wiegand assigned a series of wall blocks, triglyphs, and geisa to two different Doric buildings on the basis of a qualitative difference in the limestone (B is a hard gray limestone, while C is yellower) and other characteristics (Wiegand 1904: 155–61). The geison blocks from Building B have a hawksbeak crown moulding and one lateral geison block shows a transition from a straight side to a curved end, which indicates that the building had an apsidal plan. Elements from Building B were found in the foundations of the Mnesiklean Propylaia, the cover of a drain built by Mnesikles to the east of the Pinakotheke, and the west foundations of the

Northwest Building, but not in layers of fill elsewhere on the Akropolis (Tanoulas 1992: 201–02; note Korres 1997: 244, who disputes the attribution of the curved blocks in the foundation of the Northwest Building and the cover of the conduit to Building B). Blocks from Building C have slightly different dimensions than Building B, and the geison has a small half round crown moulding. Elements from Building C were found both in the Propylaia foundations and in the layers of fill between the Parthenon and south wall. Wiegand’s division of material and reconstructions of Buildings B and C are based on the evidence from the preserved blocks, but he did not explore the significance of their placement in the classical foundations and what it reveals about the reuse of architectural elements from the Akropolis in the second half of the fifth century. The careful consideration of where and when the blocks from Building B were placed in the foundations sheds light on what happened to the structure at the end of its original functional lifespan. By contrast with

THE SMALL LIMESTONE BUILDINGS ON THE AKROPOLIS OF ATHENAI

99

figure 5.9 Athenai, Akropolis, Pinakotheke east wall foundations, west face photo by author

the deliberate breaking up of the blocks from Building A, it appears that Building B was evaluated and dismantled with a view to reusing its blocks in the Mnesiklean building projects. In the western façade of the Propylaia, below the marble stylobate of its portico, there are several courses of limestone foundations, some of which employ older blocks in their construction (Dinsmoor & Dinsmoor Jr. 2004: 61–65, fig. 10.1). These older blocks can be recognised by the presence of anathyrosis, drafted margins, and cuttings on the visible faces. The different types of limestone and varying dimensions indicate that the blocks come from several different buildings. Many of these are straight and curving wall blocks that can be attributed to Building B based upon their shape and dimensions. Other foundations for the walls of Mnesikles’ project show a similar composition of reused architectural blocks. The foundations for the west wall of the (unrealised) Northeast Hall employ many blocks from Buildings B and C, including geisa, triglyphs, wall blocks, orthostates, and a stylobate or toichobate block (Dinsmoor & Dinsmoor

Jr. 2004: 314–17, figs. 18.1, 20.4 [left]). Elements from other structures include a large unfluted column drum and other ashlar blocks of indeterminate source. The largest quantity of reused blocks is found in the west face of the foundations of the Pinakotheke east wall (Fig. 5.9) (Dinsmoor & Dinsmoor Jr. 2004: 48, 316, 371, fig. 20.4). This line of foundations was built in two phases. The southern section was built first and contains ashlar limestone wall blocks from two different buildings. The smaller wall blocks and orthostates belong to Building B, and the larger wall blocks originate from another structure. To the north, the coursing of the foundations is different, and there is a much greater variety of reused blocks. In addition to wall blocks and orthostates, there are 15 straight geison blocks, one curved geison, and one transitional block from the springing of the apse of Building B. Table 5.1 shows the extraordinary quantity and variety of blocks from Building B reused in the Propylaia foundations, Northwest Buildings, and Mnesiklean drain. Not surprisingly, the rectangular shape of the ashlar wall

100 table 5.1

Klein Architectural elements from Building B in the Propylaia and surrounding area

Type of block

Quantity

Location

Stylobate or Toichobate blocks Orthostate blocks

1 2

Northeast Hall west wall foundation, east face, lowest course Pinakotheke east wall foundation, west face, southern section, second course from bottom Pinakotheke, east wall foundation, west face, northern section, fourth, seventh, and eighth courses from bottom Northeast Hall west wall foundation, west face, fourth course from bottom Pinakotheke east wall foundation, west face, southern section, first, third, and fourth courses from bottom Pinakotheke east wall foundation, west face, northern section, third course from bottom Pinakotheke south crosswall foundation, north face, lowest to uppermost courses Northeast Hall west wall foundation, east face, uppermost course Northeast Hall west wall foundation, west face, fourth and fifth courses from bottom Central Building west portico foundations, north west side Central Building west portico foundations, south west side Central Building west portico foundations, south west side Northwest Building west foundations Mnesiklean drain, cover Central Building west portico foundations, north west side Northeast Hall west wall foundation, east face, fourth course from bottom Northeast Hall west wall foundation, west face, fifth course from bottom Pinakotheke east wall foundation, west face, northern section, second, sixth, and seventh courses from bottom Northeast Hall west wall foundation, east face, sixth and seventh courses from bottom Northeast Hall west wall foundation, west face, second, sixth, seventh, and eighth courses from bottom Pinakotheke east wall foundation, west face, northern section, sixth course from bottom Northeast Hall west wall foundation, east face, seventh course from bottom Pinakotheke east wall foundation, west face, northern section, seventh course from bottom

≥9

Straight wall blocks

1 ≥7 1 ca. 35 21 ≥4

Curved wall blocks

Lintel block Triglyphs Straight geison blocks

17 7 4 2 2 1 4 2 15 3 ≥ 12

Curved geison blocks

1

Transitional geison block

≥3 1

blocks made them very useful as building material, followed by geison blocks and orthostates. Many other parts of the building are missing, such as column drums, antae and anta capitals, architraves, metopes, corner geisa, tympanon, and raking geison blocks. In addition to the quantity and variety of blocks reused in the foundations, the quality of the architectural elements from both Building B and Building C is noteworthy. By contrast with the evidence for the demolition of Building A, whose blocks were deliberately broken apart,

the blocks from Buildings B and C show little damage to their surfaces, and some have bright paint preserved on the soffit. Many retain their original dimensions, while others have been carefully trimmed or recut to fit in their current position. The quantity and condition of the blocks suggest that Building B was a well-maintained structure that had become obsolete and, in its terminal phase, a decision was made to carefully dismantle the structure with the intention of making the material available for reuse. Their presence in the foundations of the Mnesiklean buildings is

101

THE SMALL LIMESTONE BUILDINGS ON THE AKROPOLIS OF ATHENAI

evidence for a deliberate process of salvaging and reusing architectural material on the fifth century Akropolis. A final consideration of the distribution of architectural elements throughout the Propylaia foundations sheds light on both the process of stockpiling building material for future projects and the suggestion that Building B originally stood on the terrace where the Pinakotheke was later built. As can be seen in Table 1, the blocks from Building B were reused in several different foundations of the Propylaia: the east and south walls of the Pinakotheke (Northwest Wing), the west wall of the Northeast Hall, and the western portico of the Central Building. Other blocks were found nearby in the Northwest Building’s west foundations and as a cover for the Mnesiklean drain. Blocks from the same part of the superstructure, such as wall blocks, orthostates, and geisa, were often used in the same course of the foundations, as a practical means of maintaining consistent course height. But they are also found in different parts of the building and different courses of the same foundation. In their analysis of the foundations for the east wall of the Pinakotheke, the Dinsmoors pointed out the clear distinction between the north and south sections of the foundations: the courses are not aligned, and the size and variety of blocks are quite different (Dinsmoor & Dinsmoor Jr. 2004: 48–49, fig. 20.4). This evidence suggested to them that the foundations of the east wall were built in two phases that correspond to an original and revised plan for the Northwest Wing of the Propylaia. To the south, in the older section, the eight courses are built from the orthostates and wall blocks of two different buildings—Building B and a much larger structure. To the north, the foundations built in a second phase have nine courses constructed with reused wall blocks, orthostates, and geison blocks from Building B, as well as blocks from other structures. This mixed distribution of elements from Building B and other structures in the different foundations of the Propylaia has a bearing on the suggestion that Building B originally stood at the northwest corner of the Akropolis (Heberdey 1919: 156). Bundgaard strongly supported this theory and suggested that Building B was dismantled and its blocks reused immediately (Bundgaard 1957: 55). However, if Building B had been standing up to the beginning of construction on the Propylaia, Mnesikles would have been faced with a logistical challenge in dismantling the older structure, removing its blocks a suitable distance away for storage, and preparing the building site (Dinsmoor Jr. 1982: 30 n. 17; Eiteljorg 1993: 58 n. 104). If this were the case, we might expect to find a “reverse

architectural stratigraphy” with the upper parts of the building in the oldest and lowest foundations, and the lower parts (such as the orthostates and toichobate) found above. In fact, the oldest part of the foundation for the east wall of the Pinakotheke, to the south, contains blocks from the lower courses of the superstructure (orthostates and wall blocks), while the younger foundations to the north contain geison blocks from the top of the entablature as well as more wall blocks. It should also be recalled that the foundations of the western portico also have wall blocks from Building B in their lowest courses. While a detailed discussion of all of the issues related to the original location of Building B is beyond the scope of the current paper, the careful examination of the Propylaia foundations reveals that each section of foundation was built from the several different types of blocks from different buildings, indicating that the builders had a variety of older blocks from several dismantled structures to choose from during the process of construction. Thus, the quantity of blocks from Building B cannot, by itself, confirm that it stood on the same location as the Pinakotheke. Instead, Building B was one of several structures that had recently been dismantled and whose blocks were available for reuse. 5

Kykladic and Ionic Qualities in the Small Limestone Buildings

A critical analysis of Wiegand’s reconstructions of Buildings C and D provides evidence for separating the geison blocks into four, not two buildings. The two new groups of geison blocks have decorative mouldings that may reflect Kykladic and Ionic influence in Athenai during the Archaic and Early Classical periods. Wiegand’s Building C has a lateral geison with a small half round crown moulding (Shoe 1936: p. 151, pl. LXX, 9 proposed a date ca. 550 BCE), and he proposed the association of a raking geison with a much larger half round soffit moulding, carved and painted with alternating beads and disks (Wiegand 1904: 165, fig. 160, pl. VII.1). Some of the largest fragments have been partially restored and are on display with the Introduction Pediment (Fig. 5.10), but the pieces of the geison and pediment do not join, meaning that the association is by no means secure. Thirty-five additional raking geison fragments are preserved in the storerooms. The small, battered fragments were discovered in the deep fill on the east and south sides of the Akropolis and must have belonged to an archaic structure, perhaps built around the middle of the 6th century BCE.

102

figure 5.10

Klein

Athens, Akropolis Museum, Introduction Pediment (Acr. Inv. 9) and raking geison with half round moulding photo by author

Despite the unusual size and shape of the moulding, very little attention has been paid to its significance. The half round is an Ionic moulding, and the best parallels for a half round moulding used anywhere on a building come from Ionic temples and treasuries. The Siphnian Treasury at Delphoi, in particular, employs a large moulding with bead and reel at the base of the wall, and the continuous, sculpted frieze is framed by a large cyma reversa above and ovolo below, carved respectively with a Lesbian leaf and egg and dart motif (Daux & Hansen 1987: 73–79, figs. 60–61). The other Ionic treasuries at Delphoi show a similar propensity for sculpted architectural mouldings, but none have the same combination of a half round profile with alternating large beads and disks; at Delphoi, the Ionic treasuries of Knidos, Klazomenai, and Massilia, use a horizontally fluted torus as base moulding for their walls (see Martin 1965: fig. 156). The closest parallel to the Akropolis moulding instead is found at the fourth temple of Dionysos near Iria on Naxos, built between 580–550 BCE. Here, Lambrinoudakis and Gruben discovered a half round moulding decorated with beads and reels and proposed that it was used above or below the continuous frieze of the south porch (Lambrinoudakis

& Gruben 1987: 573, 597, fig. 5). Fragments from a geison with concave soffit have also been found. The evidence from both Delphoi and Naxos suggest that the appearance of the large half round moulding on the Akropolis may be due to Kykladic influence. The soffit moulding of the raking geison on the Akropolis is, however, an experimental form of half round moulding, acting as a compromise between the large scale of a torus, but with the bead and reel decoration appropriate to an astragal. Such a moulding does not appear to have been used again in the same position, either in Athenai or in the Kyklades. A second Ionic moulding has escaped notice among the blocks assigned by Wiegand to Building D. Approximately two dozen geison blocks made from a distinctively soft and powdery white limestone have cyma reversa crown and soffit mouldings framing a simple concave soffit. Wiegand originally identified twelve of these blocks and suggested that they should be associated with three fragments of lateral geison with a cyma reversa crown moulding and similar soffit projection (Wiegand 1904: 167–68, fig. 165, pl. VII.5). This ensemble, along with a single triglyph and fragment of architrave make up all that remains of Building D. Based upon the profile of the cyma reversa

THE SMALL LIMESTONE BUILDINGS ON THE AKROPOLIS OF ATHENAI

103

a. Type 1: Flat top, [front edge with crown molding not preserved], concave soffit, cyma reversa soffit molding, no lower taenia.

b. Type 2: Sloping top, [front edge with crown molding not preserved], concave soffit, cyma reversa soffit molding, lower taenia.

c. Type 3: Slightly sloping top, shallow ledge at back, cyma reversa crown molding, concave soffit, cyma reversa soffit molding, lower taenia. figure 5.11

Athenai, Akropolis, Typology of geison blocks with cyma reversa crown and soffit mouldings: Type 1 (Acr. Inv. 4446, oblique view), Type 2 (Acr. Inv. 4433 profile view), Type 3 (Acr. Inc. 4441, profile view photos by author

moulding and details of the lateral geison, Wiegand suggested that this was the youngest of the small limestone buildings, but without assigning an absolute date. While the presence of a similar cyma reversa moulding on both lateral and raking geison blocks does support an association, there is evidence to suggest that the 24 geison blocks belong to a separate building, and one that is Ionic, not Doric. While all of the blocks with a concave (non-mutular) soffit share the same cyma reversa crown and soffit mouldings, an illustration of representative profiles of these blocks demonstrates that there are in fact three distinct types (Fig. 5.11). The first has a flat top surface with cyma reversa mouldings framing a concave soffit. The second has a sloping top surface with the same crown and soffit mouldings, but there is a taenia below the soffit. The third has a sloping top surface with ledge at back and a taenia below the soffit. While the first type could have served as a raking geison block framing the tympanon, the blocks with sloping upper surface are unsuitable for this position. The form of these blocks is more appropriate for a lateral geison, where the upper surface is designed to secure timber roofing beams along the flanks of the building. .

If this observation is correct, these blocks cannot belong to the raking geison of the Doric Building D, but rather a completely separate structure. It is conceivable that a geison with simple, concave soffit was used in a Doric building. The late sixth century Megarian treasury at Olympia, for example, features a mutular geison on the façade above the triglyph-metope frieze and a geison with smooth soffit on the undecorated flanks (Mallwitz 1972: 174–75; Herrmann 1974: 85–83). The flanks of the treasury did not have a triglyph metope frieze and there is no moulding below the concave lateral geison soffit. A more closely comparable geison, however, is found in a completely Ionic building, the early fifth century Temple of Athena at Sounion. The profile of the moulding illustrated in Shoe (1936: pl. XXX, 1, 2), shows the similarity of the Sounion geison blocks to those on the Akropolis (Shoe 1936: pl. XXX, 3; more recent drawings of the Sounion blocks can now be found in the publication of the Temple of Athena by Barletta, Dinsmoor Jr. & Thompson 2017, figs. 28 [drawings of these blocks made by Staïs], 143, 144–45 [by W.B. Dinsmoor Jr.]). The building at Sounion had an Ionic colonnade on two sides, above which a geison with concave soffit and cyma reversa soffit moulding was

104

Klein

placed. This combination suggests that the blocks on the Akropolis belonged to an Ionic structure dating to the second quarter of the fifth century. Additional comparanda for the cyma reversa soffit moulding are found on Doric buildings that also display elements of Ionic design, such as the Mnesiklean Propylaia and the Temple of Apollon at Bassai (Shoe 1936: 68, pl. XXX, 12, and 168, noting that the cyma reversa becomes the standard soffit moulding in both Doric and Ionic buildings beginning with the temple of Athena at Sounion [Ionic] and the Propylaia [Doric]), while the cyma reversa crown moulding appears on the raking geison of the Parthenon. 6 Conclusions This study of the small limestone buildings on the Akropolis of Athenai draws upon the work of earlier scholars and employs new methods of research to expand our understanding of the architectural development of the sanctuary in the archaic and early classical periods. It demonstrates that careful inspection of individual blocks reveals details of construction and repair from the active life of the building, as well as signs of dismantling or demolition at the end of its original function. A critical review of scholarship suggests that there are more small buildings than previously identified, and some of them have Ionic characteristics that may reflect Kykladic influence on the Greek mainland. By evaluating the architectural blocks as archaeological material and considering their condition and context of deposition or reuse, it is also possible to distinguish the different ways the buildings were treated at the end of their original, functional existence. Some structures such as Building A were demolished, their blocks broken into smaller pieces and deposited in layers of fill. By contrast, the condition and quality of the blocks from Building B that were built into the Propylaia foundations suggest that the structure was carefully dismantled and its most usefully shaped blocks reused whole, or carefully trimmed. These approaches allow us to understand more fully the “biography” or “life history” of these structures from the archaic and early classical periods, as well as the active discard, caching, and recycling strategies that transformed the appearance of the Akropolis in the fifth century BCE. List of References Barletta, B.A., Dinsmoor Jr., W.B. & Thompson, H.A., 2017: The Sanctuary of Athena at Sounion. Ancient Art and Architecture in Context 4 (Princeton).

Beyer, I., 1974: “Die Reliefgiebel des alten Athena-Tempels der Akropolis”, AA 89: 639–51. Brouskari, M.S., 1974: The Acropolis Museum: A Descriptive Catalogue (Athens). Bundgaard, J.A., 1957: Mnesicles: A Greek Architect at Work (Copenhagen). Bundgaard, J.A., 1974: The Excavation of the Athenian Acropolis 1882–1890: The Original Drawings, Edited From the Papers of Georg Kawerau (Copenhagen). Daux, G. & Hansen, E., 1987: FdD II.23 : Le trésor de Siphnos (Paris). Dinsmoor, W.B. & Dinsmoor Jr., W.B., 2004: The Propylaia to the Athenian Akropolis Volume II. The Classical Building. Edited by A.N. Dinsmoor (Princeton). Dinsmoor Jr., W.B., 1982: “The Asymmetry of the Pinakotheke— For the Last Time?”, in Studies in Athenian Architecture, Sculpture, and Topography Presented to Homer A. Thompson. Hesperia Suppl. 20 (Princeton) 18–33. Dörpfeld, W., 1886: “Über die Ausgrabungen auf der Akropolis”, AM 11: 162–69. Eiteljorg II, H., 1993: The Entrance to the Athenian Acropolis before Mnesikles. AIA Monograph New Series 1 (Boston). Gardner, E.A., 1889: “Archaeology in Greece, 1888–1889”, JHS 10: 254–80. Heberdey, R., 1919: Altattische Porosskulptur. Ein Beitrag zur Geschichte der archaischen griechischen Kunst (Vienna). Herrmann, K., 1974: “Die Giebelrekonstruktion des Schatz­ hauses von Megara”, AM 89: 75–84. Herrmann, K., 1976: “Beobachtungen zur SchatzhausArchitektur Olympias”, in Jantzen, U. (ed.), Neue Forschungen in griechischen Heiligtümern (Tübingen) 321–50. Kavvadias, P., 1888: “Ἀνασκαφαὶ ἐν τῇ Ἀκροπόλει”, ArchDelt 1888: 10–11. Kavvadias, P., 1889: “Ἀνασκαφαὶ ἐν τῇ Ἀκροπόλει”, ArchDelt 1889: 105–06. Kavvadias, P. & Kawerau, G., 1906: Die Ausgrabung der Akropolis, 1885–1890 (Athens). Korres, M., 1997: “Some Remarks on the Structural Relations Between the Propylaea and the NW Building of the Athenian Akropolis”, in Hoepfner, W. (ed.), Kult und Kultbauten auf der Akropolis (Berlin) 244–45. Lambrinoudakis, V. & Gruben, G., 1987: “Das neuentdeckte Heiligtum von Iria auf Naxos”, AA 1987: 569–621. Lechat, H., 1888: “Variétés. Les fouilles de l’Acropole”, BCH 12: 238–41. Mallwitz, A., 1961: “Architektur eines Schatzhauses”, OlBer VII: 29–55. Mallwitz, A., 1972: Olympia und seine Bauten (Munich). Martin, R., 1965: Manuel d’architecture greque 1: matériaux et techniques (Paris). Ohnesorg, A., 1993: Inselionische Marmordächer. DAI Denkmäler antiker Architektur 18.2 (Berlin).

THE SMALL LIMESTONE BUILDINGS ON THE AKROPOLIS OF ATHENAI Paga, J., 2012/2013: “The claw-tooth chisel and the Hekatompedon problem. Issues of tool and technique in Archaic Athens”, AM 127/128: 169–203. Penrose, J., 1887: “Excavations in Greece, 1886–1887”, JHS 8: 269–77. Shoe, L.T., 1936: The Profiles of Greek Mouldings (Cambridge, MA). Stewart, A., 2008: “The Persian and Carthaginian Invasions of 480 B.C.E. and the Beginning of the Classical Style. Part 1:

105

The Stratigraphy, Chronology, and Significance of the Acropolis Deposits”, AJA 112: 377–412. Tanoulas, T., 1992: “Structural Relations between the Propylaea and the NW Building of the Athenian Acropolis”, AM 107: 199–215. Wiegand, T., 1904: Die archäische Poros-Architektur der Akropolis zu Athen (Leipzig).

chapter 6

Early Temples Built of Wood and Stone: New Finds from Kalapódhi (Phokis) Nils Hellner 1 Introduction The excavation of the Archaic sanctuary at Kalapódhi during 1973–1982 under the directorship of Rainer C.S. Felsch and during 2004–2013 under Wolf-Dietrich Niemeier brought to light two parallel building complexes.1 The chronological phases of the sanctuary start in the southern area in Late Helladic IIIA1 (Niemeier 2011: 98; 2012: 95; 2016: 7–10; Archibald et al. 2012: 20) and extend until Byzantine times, with thirteen chronological phases of buildings erected in succession (Niemeier 2013; 2016). The northern area was not as old, as it has only four building phases from Geometric times onwards. It seems that approximately every two generations a new building was built above the old. The focus of this paper is the archaic South Temple 9 (Fig. 6.1) (Hellner 2013: 51–52; 2014: 297–99; 2016: 555–56, with 556 fig. 1; Miles 2016: 106, 108.). Its predecessor, the South Temple 7 with an apsidal plan was begun at the end of the eighth century BCE and destroyed by fire about 595–570 BCE (Felsch 1987; 1991; 1998; 2001a; Whitley et al. 2006: 68; Niemeier 2006: 166–67; Wilson Jones 2014: 43; Hellner 2015: 126; Niemeier 2016: 18; Miles 2016: 86.). Construction of the archaic South Temple started in 550–540 BCE, and it was burnt in 480 BCE by the Persian army as it exacted revenge on the Phokians for their involvement in the battle of the Thermopylai (Felsch 1987: 19–21; 1996: 233; 2001b: 11; Hellner 2015: 126; 2016: 555; Niemeier 2016: 19). As Pausanias (10.35.2) mentioned, the devastated sanctuary was left in ruins to show the barbarous behaviour of the Persians. Consequently, we have the archaic ruin sealed by burnt debris, and the archaeological 1  I would like to express my deepest gratitude towards Wolf-Dietrich Niemeier, who involved me from 2006 as an architectural historian in the excavation. Also, I want to thank Barbara Niemeier and the teams of the last years, specially Soi Agelidis, Dimitris Grigoropoulos, Jan-Marc Henke, Ivonne Kaiser, Laura C. Rizotto. I am indebted to Hans Birk for the digital enhancement of my handmade drawings, and Julie Thomas for correcting my English. And last but not least I want to thank Rainer Felsch for all the important information about his excavations and the advice he so generously shared with me. Reports: Felsch & Kienast 1975; Felsch, Kienast & Schuler 1980; Felsch 1987; 1988; 1991; 1998; 2001a; Kienast 1988; Niemeier 2006; 2009; 2011; 2012; 2013; 2016.

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004416659_008

good fortune that the temple foundations are almost completely preserved; only at a part of the southern side had modern mechanical ploughing destroyed the stone foundations. Excavation revealed that the uppermost layer consisted primarily of roof tiles, which sealed the building. This layer also contained pieces of carbonised wood. Under and within this layer not even small stone fragments of the superstructure were found: no columns, capitals, or architraves. Under the burnt debris, the whole stylobate and toichobate of the archaic South Temple came to light. Traces of the columns on the stylobate reveal a peripteral plan with 6 by 11 columns and a cella lacking opisthodomos and pronaos (Fig. 6.1) (Felsch 1987: figs. 3, 22). Exceptional details that should be noted include the deep pteromata, with a space of almost two intercolumnations at the short ends (Hellner 2013: 51 fig. 6; 2014: 297–99, with 298 fig. 9). Most of the Archaic temples have front halls with a depth of 1.5 intercolumnations (see the temple plans at Dinsmoor Jr. 1973: 79 fig. 28; Østby 1992/1993: 75 III.3; Wilson Jones 2014: 51 fig. 2.25). The few exceptions include the temples of Apollon at Cirò Marina (Krimisa)/Calabria (Mertens 2006: 97–98) and the Ionic temple D at Metaponto/ Basilicata (Mertens 2006: 295–302) with almost the same position of the cella as at Kalapódhi. Also notable is a pair of ramps at the eastern and western facade (Hellner 2013: 52; 2014: 297–98). The temple plan with 6 by 11 columns was categorised by Heiner Knell and Wolfgang Wurster as a “Kurztempel”, a short-temple (Wurster 1973; 1974: 115–18; Knell 1975; 1983); the other temples in this series are those of Apollon at Aigina (520–510 BCE) (Wurster 1974) and Kyrene (about 550 BCE) (Pernier 1931: pl. 4.8; Dinsmoor 1950: 86.220; Wurster 1974: 115); of Poseidon at Kalauria (520 BCE) (Welter 1941: 43–45 pl. 31; Wide and Kjellberg 1895: 268–70; Wurster 1974: 115 [possibly 6 by 11 columns]); and of Athena at Karthaia (about 500 BCE) (Graindor 1905: 337–39; Østby 1980: 203 fig. 16a [reconstruction of the temple plan with dimensions]; Wurster 1974: 115 [restoring possibly 6 by 11 columns]; Simantoni-Bournia et al. 2009: 113–22, 133 fig. 82). These temples were planned simply with 5 by 10 intercolumnial spaces of equal dimensions, without refinements such as corner contraction; at Kalapódhi, the intercolumnial space is about 2.52–2.53 m,

EARLY TEMPLES BUILT OF WOOD AND STONE: NEW FINDS FROM Kalapódhi ( PHOKIS )

107

figure 6.1 Ground plan of the Archaic South Temple at Kalapódhi. Darkened areas belong to the sixth-century phase of the sanctuary. author; courtesy DAI

and only the central intercolumnium was extended to 2.60 m at the east and 2.68 m at the west. 2

Wooden Colonnades

The column positions at Kalapódhi could be identified by circular traces on the stylobate blocks. Heavily burnt calcite surrounds these circular markings. Additionally, most of the preserved column traces preserve typical crescent-shaped slots. Of the 16 slots, four have a crescent shape, while six are closer to trapezoidal (similar to those at the temple of Athena Alea at Tegea: Dörpfeld 1935: 182; Østby 1986: 77–78. 84–85 figs. 10–17; Felsch 2001: 7; and Mattern 2016: 626), and six others are almost rectangular. These slots have been widely interpreted as used for lifting wooden column shafts since Wilhelm Dörpfeld discovered them at the Heraion at Olympia. The typical slots are illustrated by Dörpfeld (1892: 33 fig. 14), and described by Hans Schleif in Dörpfeld (1935: 180, fig. 47; also discussed by von Gerkan 1948/1949: 1; Dinsmoor 1950: 54; Amandry 1952: 223–224; Eckhart 1953; Orlandos 1955–1958: Martin 1965: 11–15; Mallwitz 1966: 319–322; 1972: 142; Wesenberg 1971: 51–52, 57–60; Beyer 1972: 210; Herrmann 1972: 338 n. 375;

Kalpaxis 1975: 83; 1976: 53 n. 252; de Waele 1982: 33; Østby 1986: 84 f. n. 34, fig. 16.17 and 97; 2000: 254; 2005: 498; Sinn 2001: 63, questioning whether they exist at all; Barletta 2001: 32–39; 2011: 621–22; Hellmann 2002: 130–36; 2006: 35–49; Donderer 2005: 7; Fehr 2015: 66–67; in comparison with Kalapódhi: Felsch 2001b: 6–15; also Whitley et al. 2007: 41; Mattern 2016). In the opisthodomos, where Pausanias (5.16.1) observed a wooden column, the blocks from the stylobate level are missing; on the peristyle stylobate today only one (W5) is exposed (Mattern 2016: 624), while on the remaining positions the area where a slot would be expected is obscured by capitals (S5, S6, S14) or column drums (W4) moved onto the stylobate. These crescentshaped slots may originally have been present under many or all column settings (Mattern 2016: 625; Sapirstein 2016: 582). Philip Sapirstein (2016) compares the evidence at the Heraion to Kyrene, which has crescent-shaped slots and stone columns, and concludes the slots could be used to erect monolithic stone columns and so there may never have been wooden precedessors in the peristyle. Similar crescent-shaped slots are known elsewhere from buildings dating exclusively to the Archaic period—such as at the fourth temple of Apollon at Delphoi, where the blocks of the stylobate are rebuilt in the fountain building on

108 the terrace to the south of the temple (Dörpfeld 1935: 182; Mallwitz 1966: 321; Kalpaxis 1976: 56–58; Østby 2000: 242, 244 fig. 4; Felsch 2001b: 6–7; Mattern 2016: 626); the temple of Apollon at Kyrene (Felsch 2001b: 6–7; Mattern 2016: 626); the temple of Hera at Mykenai (Kalpaxis 1976: 88; Felsch 2001b: 11–15; Mattern 2016: 626); and the older temple of Athena Alea at Tegea (Dörpfeld 1935: 182; Østby 1986: 77–78, 84–85, figs. 10–17; Felsch 2001b: 7; Mattern 2016: 626). Some of these slots may have been used for installing wooden columns, whereas Sapirstein (2016: 586–587) proposes stone columns are possible in some cases. Many species of trees in Greece are suitable for monumental wooden architecture. In antiquity, rich forests grew in Arkadia, Euboia, Phokis (Parnassos), and Thessalia, and long trunks were imported from the Black Sea region, Phrygia, and Lebanon. Orlandos (1955–1958: 18) mentions as local wood apart from δρύος (oak) and ἐλάτης (fir) also ἀρίας, a hardwood, which might have been either the sorb or service tree (Sorbus domestica L.), the rowan (Sorbus aucuparia L.) or the elm (Ulmus). In the coastal regions of Greece coastal pine is also found (Pinus pinaster). In middle elevations (500–1200 m) the sessile oak (Quercus petraea) reaches heights up to 35 m and is found primarily in northern Greece together with maple (Acer), elm (Ulmus), cypress (Cupressus), pine (Pinus L.), and Syrian juniper ( Juniperus drupacea Labill.), alongside scarcely used building wood like fig (Ficus) and walnut ( Juglans regia). The most suitable wood grows in high to alpine elevations (2000 m), like holm oak (Quercus ilex), maple (Acer), and fir (in Greece mostly Abies cephalonica), which grows even higher (Miles 2016: 63). Oak is hardwood, which made it difficult to work according to Theophrastos (Hist. pl. 5.5.1; see Orlandos 1955–1958: 21), while fir is softwood almost free of resin and has a better degree of shrinkage than oak. Therefore Theophrastos (Hist. pl. 5.1.7; 5.6.2; 5.7.5) named fir and cypress (according Meiggs 1983: 351 ordered explicitly from Korinthos for the temple of Asklepeios in Epidauros) as preferred for roofs (see Orlandos 1955–1958: 25; Meiggs 1983: 200). According to Ernst-Ludwig Schwandner and Jörg Denkinger, the wooden shafts would have been moved while lying horizontally on hardwood rollers until the lower edge reached the intended position on the stylobate (Felsch 2001b: 10, fig. 6 [by E.-L. Schwandner and J. Denkinger]). The lifting gear at that time would have consisted of tripods: Coulton (1974: 7) proposed that the crane was used not before 515 BCE, but Hemans (2015: 49) has argued according to the evidence of post holes at Isthmia (diameter 35 cm and ca. 30–40 cm depth under the level of the cella foundations) for the use of a crane as early as 690 BCE. In whatever position the lifting gear

Hellner

was installed, the lifting of the wooden shaft would concentrate the force of its weight at its bottom edge. To prevent the dangerous sliding of the shaft, which would have caused, in most cases, a collapse of the tripod, the crescent-shaped slots were designed to secure the footing of the shaft. Once successfully lifted to the vertical position, the thrust at its foot was minimised, and the shaft could be placed with a small repositioning of the tripod exactly on the intended position on the stylobate. The northwestern corner of the stylobate shows perfectly how the columns were erected: one can observe that the stone beneath the circular traces of the column remained unharmed, even preserving tool marks whereas the surrounding surface was completely burnt and the edges of the stone split and flaked off by the heat. At the northwestern corner under the column position N11/W1, four crescent-shaped flat irons with a thickness of about 1–2 mm each were found laid in pairs, one above the other (KAL06.12515–18; length from tip to tip = 33.5 cm, width = 5.2 cm; the exact original thickness could not be determined, as the irons were almost melted onto the stone and the remains were heavily corroded; note that the positions of the columns are numbered from north towards south and east towards west, such that position N5 refers to the fifth column on the northern stylobate from the eastern corner, N1/E1). Most probably these irons were hammered under the shaft to sustain its vertical position (Fig. 6.2). Two more of these crescent-shaped flat irons were found at the adjacent position W2 (KAL06.12513–14), and another at N8. The explanation for these irons might have been the difficulty of cutting a wooden trunk with a diameter of around 75 cm exactly to the horizontal with an adze, especially as the section had to be perpendicular to the direction of the wood fibers. Bronze straightedge saws operated by two workers for cutting both stone and wood are attested since the second millennium in Hattusa-Boğazköy (Neve 1989: 399–406) and in Late Minoan Krete (Deshayes 1960: 1:359–61; 2:153), but those in Greece were designed to be drawn only in one direction (Goodman 1964: 115), and carpenters operated them primarily on small-dimensioned timbers (see Barletta 2009: 157). It is doubtful whether these irons could have been inserted under stone shafts without destroying their delicate edges. Of the 21 positions preserving traces of columns, 16 have lifting slots, and from their position it is possible to reconstruct the working direction for the installation of the shafts. Normally the wooden shafts were raised from the perpendicular to the stylobate. However, the close proximity of the northern temple prevented this approach on the north peristyle, where the shafts were erected from

EARLY TEMPLES BUILT OF WOOD AND STONE: NEW FINDS FROM Kalapódhi ( PHOKIS )

109

figure 6.2 Northwestern corner of the Archaic South Temple author; courtesy DAI

the interior of the southern temple. Due to the obstacle of the cella foundations to the south, they were erected from an oblique orientation. The maximum height of a wooden shaft can therefore be reconstructed as at most 4.90 m (this information was generously provided by Felsch 2006). 3

Restoring the Capitals and Entablature

As mentioned before, the temple was destroyed by the Persians, and an impressive burnt layer sealed the building. In the course of excavation in 2009, directly in front of the western end of the temple above the ramp, a series of splintered and burnt slabs with a thickness of 25.9 cm were unearthed (Fig. 6.3). Further to the west, a second series of slabs 29.4 cm thick came to light, which clearly showed the shape of a tympanon with an angle of 11° and its peak almost exactly on the middle axis of the building (Niemeier 2009: 109, fig. 3; Morgan, Pitt & Whitelaw 2009: 44, fig. 71). The limestone was so heavily burnt that it was calicined, which suggests temperature of burning in excess of 800°C for enough time for the transformation to be completed (probably several tens of minutes or even hours). Unfortunately, it is not possible to estimate the duration and the exact temperature of burning (unpublished 2009 report by Panagiotis Karkanas, geologist and director

of the Malcolm H. Wiener Laboratory for Archaeological Science at the American School of Classical Studies in Athens). Further west was excavated a series of badly broken slabs with a thickness of only 20 cm, in front of which lay the broken terracotta sima and the fragments of a ridge akroterion shaped like a horse of roughly life size (Moustaka 2010, 72 fig. 4 a-e, 73 fig. 5), which marked the ridge of the temple (Fig. 6.3). At the end of the excavated area, the whole western pediment was found preserved in its collapsed position in front of the temple. All parts of a normal pediment were attested: horizontal geison, gable orthostates, and the raking geison, but in a still rudimentary, simple form without any moldings or even a simple profile (Fig. 6.3). The stone pediment collapsed from a height of about 5 m onto the burning wooden columns in front of the temple and, heated by the fire, split into thousands of pieces. Given the total lack of any other stone elements under the sealed destruction layer, we have clear evidence for a stone stylobate and a wooden superstructure supporting a stone pediment and terracotta roof. As has been stressed in previous research, the Kalapódhi temples represent an intermediate stage in the process of petrification, combining wooden columns and architraves with a stone pediment (Felsch 2001a; Hellner 2010: 159; 2011: 258; 2013: 50; 2016: 555). The various theories about the genesis of the

110

figure 6.3 Western front of the Archaic South Temple with collapsed pediment author; courtesy DAI

peristyle and the process of petrification are reviewed in Barletta (2001: 32–39; 2011: 621–22), Hellmann (2002: 130– 36; 2006: 35–49), and Fehr (2015: 66–67). For Mark Wilson Jones (2014: 48–54), the first temples preserving a peristyle are the Artemision at Áno Mazaráki, the earliest temples at

Hellner

the Argive Heraion and Ephesos, and possibly the temple of Poseidon at Isthmia, whereas Burkhard Fehr (2015: 67) remarks that the buildings at Levkandí, Áno Mazaráki, and Ephesos constitute only a building with a shed roof supported on posts, which does not meet the requirements— according to his theoretical interpretation—of a raised, seemingly hovering (“schwebend”) construction on posts roofed by a heavy layer of tiles, which are fulfilled only by the Poseidon temple at Isthmia. A building with wooden posts, horizontal ceiling beams, and rafters is not such a difficult construction to imagine, but how the stone elements were technically integrated into the wooden roof construction can only be hypothesised. One must imagine some iron joints in the roof to fix the stone members to the wooden beams and probably dowels to fix the capitals to the architraves. In fact, the excavations recovered iron clamps, elongated U-shaped bars, a kind of iron grid (Moustaka 2010: 72) and nails of various dimensions which await publication. The crescent-shaped slots are absent from the surviving foundations of earlier buildings dated to the seventh century BCE. Because foundations of that period were typically built without a continuous stone stylobate, the wooden posts were set on isolated stone bases, as can be seen from the preceeding temple at Kalapódhi (Hellner 2010; 2013: 47–50; 2014: 295–97) and in other examples like the temple of Artemis at Áno Mazaráki (Petropoulos 1987: 81–96; 1992–1993: 141–57; 1996: 220–25; 2001: 39–45; 2002: 143–64; cf. Gruben 2001: 33); the temple of Poseidon (?) at Nikoléïka (Kolia & Gadolou 2007; Vordos & Kolia 2008: 53–55, fig. 24; cf. Evely et al. 2008: 44–45; Kolia & Gadolou 2011: 193, fig. 3; Kolia 2011: 205–06, 228–231, figs. 4, 6; Wilson Jones 2014: 43, 44, fig. 2.14); the temple of Artemis Orthia at Sparta (Dawkins 1929; Boardman 1963: 1–7 [dating the temple to the early seventh century BCE]; Drerup 1969: 19, fig. 16; Kalpaxis 1976: 77; Fagerström 1988: 31–32, esp. n. 120 with critics at Dawkins’ chronology); the temples of Apollon at Halieis (Jameson 1963; 1966; 1967; 1968; 1969a; 1969b; 1971; 1974: 116, 118; 1982: 367; with proposed dates from the roof tiles ranging between ca. 640 BCE and the first quarter of the sixth century: Jameson 1982: 365; Berquist 1990: 24; Cooper 1990: 77; Winter 1993: 160–62; cf. Mazarakis Ainian 1997: 162–63, 163 fig. 245) and at Eretria (Auberson 1968: 13–14, fig. 14, plan IV; 1974: 60, 66; Auberson & Schefold 1972: 188–89; Mazarakis Ainian 1997: 102–04, 243; Ducrey, Simon & Verdan 2002: 131 n. 14; Ducrey et al. 2004: 226–37; Lippolis, Livadiotti & Rocco 2007: 684; Wilson Jones 2014: table 38 and 39, nn. 3.4); and the earliest phase of the temple of Artemis at Ephesos (Mazarakis Ainian 1997: 205–07 fig. 423–24 [with Anton Bammer’s inaccurate chronology]; see Weißl 2002: 321–27

EARLY TEMPLES BUILT OF WOOD AND STONE: NEW FINDS FROM Kalapódhi ( PHOKIS )

[dating the earliest temple in the first quarter of the seventh century BCE]; Kerschner & Prochaska 2011: 77–82; Weißl 2010: 134–39, for criticism of the assumption of a peristyle without real archaeological data). The temple at Nikoléïka (Akhaia) had at its eastern front an apsidal stylobate supporting supposedly wooden columns (Kolia 2011: 229 fig. 45 is mirrored twice in comparison to 206 fig. 5; Kolia & Gadolou 2011: 192, 202 fig. 2), and is thus the earliest continuous stone foundation for columns or posts—i.e., a stylobate. The Argive Heraion is the second earliest example in mainland Greece with a continuous stone stylobate (Amandry 1952: 224, fig. 1), though its chronology is debated (Drerup (1969: 57 [second half of the seventh century BCE]; Mallwitz 1981: 634 [late seventh century BCE]; Wright 1982: 191 [third quarter of the seventh century BCE]). Pierre Amandry (1952: 224 fig. 1) observed that the stylobate width of 1.026–1.050 m is adequate to accommodate the bases but that that the 90-cm diameter of the three column bases is too big for the four columns settings (diameter 78–80 cm) on the stylobate of the first temple of Hera, although he did not exclude them from the temple (Amandry 1952: 225 n. 14; Hellner 2004). In the Archaic period, the column diameters could vary within one building (e.g., the hekatompedos at Orkhomenos or the Heraion at Olympia), meaning that the stone bases could have supported columns elsewhere in the temple. The Heraion still has what Burkhardt Wesenberg described as “drum-like column-bases”—a kind of cylindrical plinth on the stylobate reminiscent of the single-foundation technique of earlier buildings (Wesenberg 1996: 7, fig. 8). The archaeological evidence at Kalapódhi clearly shows a development of single wooden posts paired with a continuous stone stylobate and wooden column shafts. But an important additional question in the long-lasting discussion of the petrification process remains: was there a previous form of wooden capital before the first stone capitals, which are attested around 600 BCE? The theory of wooden forerunners to stone elements was extensively discussed by Durm (1892) and Dörpfeld (1935: 179–82), and the use of the crescent-shaped slots for wooden columns have since been discussed frequently (see above, p. 106). As discussed below, I believe that the Doric capital might have had small wooden forerunners in furniture or small votive columns but was developed in monumental architecture from the beginning in soft stone. Joseph Durm drew one of the most frequently published images of a Greek temple in its wooden form, including a wooden capital with an abacus (Durm 1892: 114, fig. 87; Orlandos 1955–1958: 10, fig. 4; Beyer 1972: 203, fig. 6; Hellmann 2002: 134, fig. 172; Wesenberg 2008: fig. 5, pl. 46; Barletta 2009: 155, fig. 18.1; von Kienlin 2011: XIV, fig. 1;

111

Wilson Jones 2014: 64, fig. 3.3). Depictions of columns with capitals on vase paintings from the early sixth century BCE are found on the François vase (Museo archeologico nazionale di Firenze Inv.-Nr. 4209: Beazley 1986: 24–34, pl. 24.1, 25.1), showing a fountain building with a normal Doric freeze, a very low architrave, and two very slender fluted columns, which are placed on flat bases and have therefore been interpreted as wooden (Fig. 6.4) (Vallois 1908: 365–69 [capitals on black figured vases], 374–76 [capitals on red figured vases], fig. 1–13, 23; Eckhart 1953; Orlandos 1955–1958: 9, fig. 3 [wooden], 10, fig. 4, 11, fig. 6; Wesenberg 1971: 51–52, 59–60, figs. 111.112 with capitals on vases from Vári and Perakhóra; Beyer 1972: 210, fig. 13; Østby 2000: 254, fig. 9; additional depictions are in Marconi 2009: 5–9). Such bases attested in early architecture (e.g., the Argive Heraion, described above) have been explained as protection for the base of the wooden post or shaft from moisture, which would have caused rapid decay. The capitals on the vase painting appear in the canonical form, and Durm and later Mallwitz (1981: 639, 640, fig. 33) draw them like this—but did wooden capitals shaped like stone capitals really exist at this time? Or, if we ask the question the other way round, did other forms of capitals exist before the first examples in stone? Turning profiled wooden shapes on a lathe was quite common from Mykenaian times onwards—mostly for legs of furniture as stools, couches, and beds (e.g., Richter 1966: 39–43, figs. 200–21, 407, stool [δίφρος] with four legs). This technique seems possible for the small echinos of a votive column, and probably also for columns and capitals of the scale shown in vase paintings. Technically an echinos with a height of about 15–17 cm, an upper diameter of approximately 90 cm, and a diminished lower diameter of about 55 cm could not be realised in wood (Kawerau 1909: 228). Carpenters reject the possibility of shaping an echinos in oak or fir using ordinary tools, and one specialist in lathe-working even pointed out that normal wood will be torn along its fibers. Modern High Density Fiberboard (HDF) could get around this problem, with several dozens of sheets laminated to reach the necessary thickness of about 95 cm. Furthermore, it is doubtful that carpenters could shape 32 almost equal pieces for one building, considering that wood always suffers from drying shrinkage and the dimensions would vary noticeably and unpredictably. Finally, a fundamental rule in carpentry would have to be ignored: never put wood with its end-grain on another end-grain (Kawerau 1909: 229). One technical detail in Durm’s drawing is also worth mentioning because it seems to be deduced clearly and very formally from the later stone construction: the architrave is constructed from two beams. The turning of profiled stone elements of huge

112

Hellner

figure 6.4 Detail of the François vase author, after Østby 2000, 254 fig. 9

diameters is attested on Samos from the early sixth century BCE onwards (for turning huge stone drums on the tornos, or lathe, see Hendrich 2007: 67–86; Hellner 2009: 157–66, with additional literature). The cutting of soft stone was very likely made with the σκέπαρνον, or adze, used by both carpenters and masons (Korres & Ohnesorg 2017: 17 [on its use]; on the terminology: Liddell & Scott 1996: 1606 s.v. σκέπαρνον; Orlandos 1955–1958: 1:30–33, fig. 13–17; Ginouvès & Martin 1985: 68–70; Hellmann 1992: 329–30; Korres 1995: 77 fig. 10; cf. Blümner 1879: 203–05; Hellner 2011: 229–31). On a technical level, a simple wooden construction resembling the architecture depicted on vases would consist of a corbel or bracket. In similar wooden constructions in vernacular architecture around the world, brackets are used to sustain a horizontal beam where it rests on a post. The Ionic capital is thought to be derived from a wooden bracket as a possible forerunner (von Gerkan 1946/1947: 18; Gruben 1996: 65, fig. 5). A bracket seems more technically feasible in the type of temple in antis with an architrave running straight between the two anta walls shown in the François vase (Fig. 6.4), but not in a peristyle where the architrave turning around the corners creates technical problems. If a carpenter had to handle the problem of architrave beams with expanded width in monumental architecture, or even an architrave composed of several beams, a broader wooden element in the lateral direction needs to mediate the weight to the narrower wooden bracket (Fig. 6.5). This element is drawn as a flat rectangular plank—but was it manufactured from one piece of wood like an abacus? This kind of building element does not derive from any carpentry tradition, taking into consideration that woodworkers of that time worked huge

beams with the adze. First, the shaping of such a rectangular plank would have required an immense trunk with an abundance of wasted material, and second this flat wooden element would most probably have split and cracked very quickly due to the normal shrinkage process during drying. The same production process for a round echinos cannot be imagined. A rarely noticed paper by Georg Kawerau (1909) published after his death appeals to a less formal and a more technical approach towards reconstruction in architectural history. Considering the problem created when architecture became increasingly monumental, thereby increasing the width of the architrave, he proposed that, instead of just the normal lateral bracket on slender wooden posts, a set of transverse beams would be added to support the much larger width of the architraves. Because no drawings were published, it is worth sketching out Kawerau’s hypothesis (Fig. 6.5, left). The abacus could have consisted of a series of wooden beams set perpendicular to the bracket and the architrave, and if they were capped by planks, they would have appeared from the outside as one continuous rectangular element (Fig. 6.5, middle). Another hypothesis is that the echinos element and the abacus were assembled from several wooden pieces attached with dowels and glue. Phrygian woodworking techniques were highly developed even in the ninth century BCE and could have been imported to Greece. Samuel Holzman drew my attention to the evidence from Gordion, where thin wooden plates laminated from several pieces of boxwood and yew with a diameter of about 33–36 cm (Simpson 1999: 783, pl. 75a; Simpson & Spirydowicz 1999: 70–71, fig. 97–98) and a bigger plate of a tripod tray table with a diameter of about 76

113

EARLY TEMPLES BUILT OF WOOD AND STONE: NEW FINDS FROM Kalapódhi ( PHOKIS )

figure 6.5 Wooden brackets author

figure 6.6 Corner solution of brackets (left) and a “bracket-capital” (right) author

cm were recovered from the later tumulus P (Simpson & Spirydowicz 1999: 57–58, fig. 71; the analysis of the glue is still pending). Another relevant comparison is the highly delicate yet resilient aspis—the Greek infantry shield— which was made of thin pieces of wood and glue (I thank the anonymous reviewer for this observation). Even if the diameter of the available wood is insufficient for the loadbearing bracket of a capital, it is at least possible that the formal problem of shaping the echinos might have been solved through joining several smaller pieces together with glue (Fig. 6.5, right).

Nevertheless, a bracket is a far more realistic solution. Applying the principal at the corners results in solutions similar to the Ionic corner capital, a cross-shaped “bracket-capital” (Gruben 1963: 162–63, figs. 41, 42; for the earliest Ionic marble capitals see Gruben 1989: 168–72). This construction would have been—from the classical point of view—heretical (Fig. 6.6), although a hypothetical solution has been drawn in order to show that, in an elevation, such “bracket-capitals” would have resembled the schematic ones depicted on the early vases (Fig. 6.7). When we instead consider the aforementioned evidence of early Archaic temples with columns supported on

114

Hellner

figure 6.7 Reconstructed perspective of the Archaic South Temple with “bracket-capitals” author

“drum-like column-bases” derived from older single-post foundations, they stand in contradiction to the developed Doric order, where the columns have no plinths. Yet such “drum-like column-bases” were depicted on contemporary vase paintings. Because of the lack of stone elements from the superstructure, a wooden colonnade and entablature has been proposed for the temples at the Argive Heraion (above), Thermon (Kuhn 1993: 41, 45, proposing an archaic temple C and a Hellenistic repair with “drumlike column-bases”, some with shallow flutes), Kallion (Themelis 1983: 237–38, photo 239 fig. 38, 240 fig. 39, pl. 2; 14, with “drumlike column-bases” [“stone-footings”] in situ, diameter 55–57 cm, height 33 cm; but cf. Felten 1996: 141) and Glanitsa (Metzger 1940/1941: 15 fig. 8, 16; cf. Østby 1995: 328 n. 550 dividing these “drum-like column-bases” into two groups: first, smooth bases with the beginning of the shaft and a height of 47 cm; second, smooth bases with a taper; both groups with an upper diameter of 44.0–46.5 cm, and some preserving conical dowel-holes on their upper surfaces). At the Archaic temple of Orkhomenos are also eight of these “drum-like column-bases” associated with stone capitals (Blum and Plassart 1914: 82, 83 fig. 11, seven in situ, one seemingly moved from its original position; but cf. Østby 1995: fig. 188; 2000: 253; 2005: 498), and the lack of monoliths or even fragments of columns drums obliged the French excavators to reconstruct wooden column shafts. The capitals from Orkhomenos have an unusual conical dowel hole on their lower sides, a type

which has been identified as the earliest form of dowel. Such conical holes are also known from four early stone capitals, all of them reconstructed on wooden shafts: capitals D (Amandry 1952: 234 fig. 5, pl. 62e,f) and E (Amandry 1952: 244 fig. 9, pl. 66a,b) from the Argive Heraion; the capital from Tiryns (Schwandner 1985: 116, fig. 73); and the “altar-capital” from Kalapódhi (Hellner 2015). I am convinced that the element of the echinos had pre­decessors in wooden furniture and small columns, but the monumental form could only have been developed in soft stone (Barletta 2001: 80; Kienast 2002: 66; also cf. Kyrieleis 2008: 205). Small votive stone columns from the first quarter of the sixth century BCE could be regarded as the paradigms, namely the Xenvares capital at Kerkyra (Schleif 1940: 77 fig. 60; Wesenberg 1971, 51, n. 251, figs. 94– 95; Barletta 2001, 60–62; Wilson Jones 2014, 78–80, fig. 3.20), Capital B from Delphoi (de la Coste-Messelière 1942/1943: 36–38, 73 fig. 6), and the examples at Tiryns (Dörpfeld 1886: 334–36, fig. 122; Frickenhaus 1912: 7–9; Sulze 1936: 14–36; Wesenberg 1971: 51, fig. 104; Barletta 2001: 54–62, 173 fig. 25; Schwandner 1985: 116, fig. 73 [a new drawing]; Wilson Jones 2014: 78–80, fig. 3.21), Aigina (Mus. Inv. 2375: Dörpfeld 1886: 293–94; Frickenhaus 1912: 7–9; Sulze 1936: 14–36; Welter 1938: 16–18 fig. 8.9; Schleif 1940: 90–91, fig. 70; Wesenberg 1971: 54 no. 24, fig. 108, 51 n. 256; Schwandner 1985: 115–16; Hoffelner & Kerschner 1996: 18 [dated to 580 BCE]; cf. Hoffelner 1999: 18 fig. 4; Wilson Jones 2014: 78–80, fig. 3.20), and Akragas (Wesenberg 1971: 51, pointing out

EARLY TEMPLES BUILT OF WOOD AND STONE: NEW FINDS FROM Kalapódhi ( PHOKIS )

115

figure 6.8 Reconstructed perspective of the Archaic South Temple with stone capitals and without a triglyph-metope zone author

the chronological problem of compressed echinos-profile, which would indicate it is among the oldest preserved, but can be dated only after the colonial foundation in 580 BCE; cf. Kalpaxis 1976: 43–44, n. 187; Mertens 2006: 103, fig. 151, suggesting it was brought from Gela) and Selinous (Mertens 2006: 104, figs. 153–54). The Kalapódhi South Temple would probably have looked like the reconstruction in Figure 6.8, where it has been imagined with stone capitals between wooden shafts and entablature. Yet it should be emphasised that no identifiable fragment of a stone capital has been found in the sealed destruction layer at Kalapódhi. 4

Antefix Tiles from the Tympanon

During restoration work in 2016, a surprise came when one of the fragmented slabs of the tympanon orthostates was lifted (Hellner 2016/2017). Just beneath it, in a heavily burnt layer were eaves tiles and two fragmented threepeaked “horn” antefixes found 58 cm apart—the length of the eaves tiles (KAL16.138.004 and 005, Fig. 6.9)—similar to those which had already been excavated at the southeastern front and at the south side by Felsch and which clearly belonged to the temple (Hübner 1987; 1990; 1995; 1997; cf. in general Winter 1993: 188–203). At the east facade in the early 1980s, Felsch found one fragment of an antefix of the same type whose full depth is only 68 cm,

significantly less than that of the antefixes at the flanks measuring 76–77 cm (Hübner 1997: 133–34; 144 fig. 1; the normal tiles dimensions are 56–58 by 76–77 cm for pans, 18–20 by 76–77 cm for covers, and both are about 3 cm thick). However, he did not dare to reconstruct this special tile on the building without further evidence at the front, although he hypothesised that it was somehow used in the pediment (Felsch: pers. comm. 2016). The idea was first hinted at by Hübner (1997: 132 n. 4), who discarded the hypothesis because of the uncertain correspondence of the antefix findspot at the east front of the temple. The new discovery has demonstrated that shortened eaves and antefix tiles must have been placed on top of the horizontal stone geison. The antefixes were set on “geison” tiles 7 cm thick at the front with a soffit projecting 12 cm, as indicated by a red-painted band along the bottom edge (Hübner: 1990, 170; 1995: 154; 1997: 144 fig. 1 [tile A]; the tiles are 76–77 by 46–48 cm). The western pediment has another solution omitting geison-tiles, with sima tiles (68.5 by 46–48 cm, and 14.5 cm high at the front) laid directly on the stone pediment (Hübner 1980: 113 fig. 101). Fragments from this sima were found in fillings in the pronaos of the Classical North Temple and could not be attributed to a building at that time (Felsch & Kienast 1975: 21, figs. 28–29; cf. Winter 1993: 194, pl. 80). In the 1980s, Gerhild Hübner reconstructed nine tiles on the raking geison and exactly twice this number, 18, of the specially shortened type of antefixes, which could have been seated on the horizontal

116

figure 6.9 Reconstructed three-peaked antefixes (KAL16.138.004 and KAL16.138.005) author; courtesy DAI

Hellner

EARLY TEMPLES BUILT OF WOOD AND STONE: NEW FINDS FROM Kalapódhi ( PHOKIS )

figure 6.10

117

Reconstructed perspective of the Archaic South Temple with an additional triglyph-metope zone and a sima on the horizontal geison in the tympanon (compare to Fig. 6.8) author

geison in the tympanon, with the ninth antefix exactly on the middle axis of the temple under the ridge line (Hellner 2016/2017). In conclusion, the last drawing shows that reconstructions of Archaic architecture must be handled very carefully (Fig. 6.10). Our architectural reconstructions should only be based on clear evidence and not the assumption that a building adheres to canonical forms. We should regard all reconstructions with a healthy dose of skepticism, or at least with an open mind, before the next discovery points to another different solution that departs from our expectations for normal or “canonical” forms. References Amandry, P., 1952: “Observations sur les Monuments de l’Héraion d’Argos”, Hesperia 21: 229–74. Archibald, Z., C. Morgan, I.S. Lemos, R. Pitt, R. Sweetman, D. Stewart, J. Bennet, M. Stamatopoulou & A. Dunn., 2012: “Archaeology in Greece. Archaeological Reports for 2011–12”, ARepLondon 58: 1–121. Auberson, P., 1968: Temple d’Apollon Daphnephoros. Eretria 1 (Bern). Auberson, P., 1974: “La reconstitution du Daphnéphoréion d’Erétrie”, AK 17: 60–68.

Auberson, P. & Schefold, K., 1972: Führer durch Eretria (Bern). Barletta, B.A., 2001: The Origins of the Greek Architectural Orders (Cambridge). Barletta, B.A., 2009: “The Greek Entablature and Wooden Antecedents”, in Counts, D.B. & Tuck, A.S. (edd.), Koine: Mediterranean studies in honor of R. Ross Holloway (Oakville) 154–66. Barletta, B.A., 2011: “State of the Discipline: Greek Architecture”, AJA 115.4: 611–40. Beazley, J.D., 1986: The Development of Attic Black-Figure. Second revised edition by D. von Bothmer & M.B. Moore (Berkeley). Bergquist, B., 1990: “Primary or Secondary Temple Function: The Case of Halieis”, OpAth 18: 23–37. Beyer, I., 1972: “Der Triglyphenfries von Thermos C”, AA 1972: 197–226. Blum, G. & Plassart, A., 1914: “Orchomène d’Arcadie: fouilles de 1913”, BCH 38: 71–88. Blümner, H., 1879: Technologie und Terminologie der Gewerbe und Künste bei Griechen und Römern (Leipzig). Boardman, J., 1963: “Artemis Orthia and Chronology”, BSA 58: 1–7. Cooper, N.K., 1990: “Archaic Architectural Terracottas from Halieis and Bassai”, Hesperia 59: 65–93. Coulton, J.J., 1974: “Lifting in Early Greek Architecture”, JHS 94: 1–17. Dawkins, R.M., 1929: The Sanctuary of Artemis Orthia at Sparta. Excavated and Described by Members of the British School at Athens 1906–10 (London).

118 de La Coste-Messelière, P., 1942/1943: “Chapiteaux dorique de Delphes”, BCH 66/67: 22–67. de Waele, J.A.K.E., 1982: “Der Entwurf des Heraion von Olympia”, BABesch 57: 26–37. Deshayes, J., 1960: Les outils de bronze, de l’Indus au Danube (IVe au IIe millénaire). Bibliothèque Archéologique et Historique (Beyrouth) 71 (Paris). Dinsmoor, W.B., 1950: The Architecture of Ancient Greece: An Account of its Historic Development. 3rd ed. (London). Dinsmoor, W.B., Jr., 1973: “The Kardaki Temple Re-examined”, AM 88: 165–74. Donderer, M., 2005: “Das Heraion in Olympia und sein Säulenkranz”, BABesch 80: 7–20. Dörpfeld, W., 1886: “Die Bauwerke von Tiryns”, in Schliemann, H. (ed.), Tiryns: Der prähistorische Palast der Könige von Tiryns. Ergebnisse der neuesten Ausgrabungen (Leipzig) 200–352. Dörpfeld, W., 1892: “Das Heraion”, in Curtius, E. & Adler, F. (edd.), Die Baudenkmäler von Olympia (Berlin) 27–36. Dörpfeld, W., 1935: Alt-Olympia. Untersuchungen und Ausgrabungen zur Geschichte des ältesten Heiligtums von Olympia und der älteren griechischen Kunst (Berlin). Drerup, H., 1969: Griechische Baukunst in geometrische Zeit. ArchHom II:O (Göttingen). Ducrey, P., Simon, P. & Verdan, S., 2002: “Les activitès de l’Ecole suisse d’archéologie en Grèce 2001”, AK 45: 124–32. Ducrey, P., Knoepfler, D., Rendall, S. & Badinou, P., 2004: Eretria: A Guide to the Ancient City (Gollion). Durm, J., 1892: Baukunst der Griechen. 2nd ed. (Leipzig). Eckhart, L., 1953: “Bemerkungen zu dorischen Säulen auf archaischen griechischen Vasen”, ÖJh 40: 60–72. Evely, D., Hall, H., Morgan, C. & Pitt, R.K., 2008: “Archaeology in Greece”, ARepLondon 54: 1–113. Fagerström, K., 1988: Greek Iron Age Architecture: Developments Through Changing Times (Göteborg). Fehr, B., 2015: “Das Mirakel der schwebenden Last. Zur Peristais des griechischen Tempels”, in Faust, S., Seifert, M. & Ziemer, L. (edd.), Antike. Architektur. Geschichte: Festschrift für Inge Nielsen zum 65. Geburtstag (Aachen) 63–80. Felsch, R.C.S., 1987: “Kalapodi. Bericht über die Grabungen im Heiligtum der Artemis Elaphebolos und des Apollon von Hyampolis 1978–1982”, AA 1987: 1–99. Felsch, R.C.S., 1988: “Das Heiligtum von Kalapodi in Vor- und Frühklassischer Zeit”, in Πρακτικά του XII Διεθνούς Συνεδρίου Κλασσικής Αρχαιολογίας, Αθήνα 4–10 Σεπτεμβρίου 1983, (Athens): 61–64. Felsch, R.C.S., 1991: “Tempel und Altäre im Heiligtum der Artemis Elaphebolos von Hyampolis bei Kalapodi”, in Étienne, R. & Le Dinahet, M.-T. (edd.), L’espace sacrificiel dans les civilisations méditerranées de l’antiquité (Paris) 85–91.

Hellner Felsch, R.C.S. (ed.), 1996: Ergebnisse der Ausgrabungen im Heiligtum der Artemis und des Apollon von Hyampolis in der antiken Phokis. Kalapodi 1 (Mainz). Felsch, R.C.S., 1998: “Kalapodi und Delphi—Zur Frühzeit des Apollonkultes in Mittelgriechenland”, in Rolle, R. & Schmidt, K. (edd.), Archäologische Studien in Kontaktzonen der Antiken Welt (Göttingen) 219–36. Felsch, R.C.S., 2001a: “Opferhandlungen des Alltagslebens im Heiligtum der Artemis Elaphebolos von Hyampolis in den Phasen SH IIIC—spätgeometrisch”, in Laffineur, R. & Hägg, R. (edd.), Potnia. Deities and Religion in the Aegean Bronze Age. Aegaeum 22 (Liège) 193–99. Felsch, R.C.S., 2001b: “Drei frühe Phasen des Dorischen Tempels: Delphi-Kalapodi-Mykene”, JdI 116: 1–15. Felsch, R.C.S. (ed.), 2007: Ergebnisse der Ausgrabungen im Heiligtum der Artemis und des Apollon von Hyampolis in der antiken Phokis. Kalapodi 2 (Mainz). Felsch, R.C.S. & Kienast, H.J., 1975: “Ein Heiligtum in der Phokis”, AAA 8: 1–24. Felsch, R.C.S., Kienast, H.J. & Schuler, H., 1980: “Apollon und Artemis oder Artemis und Apollon? Bericht von den Grabungen im neu entdeckten Heiligtum bei Kalapodi 1973– 1977”, AA 1980: 38–123. Felten, F., 1996: “Griechische Heiligtümer in hellenistischer Zeit”, in Blakolmer, F., Krierer, K.R., Krinzinger, F., Landskron-Dinstl, A., Szemethy, H.D. & Zhuber-Okrog, K. (edd.), Fremde Zeiten: Festschrift für Jürgen Borchhardt zum sechzigsten Geburtstag am 25. Februar 1996 dargebracht von Kollegen, Schülern und Freunden (Vienna) 139–58. Frickenhaus, A.H, 1912: Die Hera von Tiryns. Tiryns 1.1 (Berlin). Ginouvès, R. & Martin, R., 1985: Dictionnaire méthodique de l’architecture grecque et romaine (Athens). Goodman, W.L., 1964: The History of Woodworking Tools (London). Graindor, P., 1905: “Fouilles de Karthaia (Isle de Kéos)”, BCH 29: 329–61. Gruben, G., 1963: “Das archaische Didymaion. Mit einem Exkurs: Die Säulenhöhe des Naxier-Oikos in Delos”, JdI 78: 78–182. Gruben, G., 1989: “Das älteste marmorne Volutenkapitell”, IstMit 39: 161–72. Gruben, G., 1996: “Griechische Un-Ordnungen”, in Schwandner, E.-L. (ed.), Säule und Gebälk: Zu Struktur und Wandlungsprozess griechisch-römischer Architektur. Bauforschungskolloquium in Berlin vom 16.–18. Juni 1994. DiskAB 6 (Mainz) 61–77. Gruben, G., 2001: Griechische Tempel und Heiligtümer. 5th ed. (München). Hellmann, M.-C., 1992: Recherches sur le vocabulaire de l’architecture Grecque, d’après les inscriptions de Délos (Athens).

EARLY TEMPLES BUILT OF WOOD AND STONE: NEW FINDS FROM Kalapódhi ( PHOKIS ) Hellmann, M.-C., 2002: L’architecture grecque 1: les principes de la construction (Paris). Hellmann, M.-C., 2006: L’architecture grecque 2: architecture religieuse et funéraire (Paris). Hellner, N., 2004: “Drehspuren am Säulenbauteil des archaischen Heraion von Argos?”, RA 37: 69–78. Hellner, N., 2009: Die Säulenbasen des Dipteros II im Heraion von Samos. Grundlage für die Rekonstruktion des Tempels in seinen Bauphasen. Samos 26 (Bonn). Hellner, N., 2010: “Die Anfänge des griechischen Tempelbaus und die Gestaltungsidee der Säulenkannelur in Kalapodi/ Phokis (GR)”, in Sack, D. & Schulz, T. (edd.), Bericht der 45. Tagung der Koldewey-Gesellschaft in Regensburg 2008 (Dresden) 153–59. Hellner, N., 2011: “Überlegungen zu achteckigen Stützen in der antiken griechischen Architektur”, RA 52: 227–62. Hellner, N., 2013: “Kalapodi. Neue Kriterien einer Typologie der dorischen Architektur”, in Raue, D. & Gerlach, I. (edd.), Sanktuar und Ritual: Heilige Plätze im archäologischen Befund. ForschungsCluster 4.10 (Rahden) 43–52. Hellner, N., 2014: “Räumliche Führung am Beispiel der spätgeometrischen und archaischen Süd-Tempel von Abai/ Kalapodi”, in Kurapkat, D., Schneider, P.I. & Wulf-Rheidt, U. (edd.), Architektur des Weges. Gestaltete Bewegung im Gebauten Raum. Bauforschungskolloquium in Berlin vom 6. bis 11. Februar 2012 veranstaltet vom Architekturreferat des DAI. DiskAB 11 (Regensburg) 289–307. Hellner, N., 2015: “Ein stratigraphisch datiertes, hocharchaisches Kapitell aus Abai/Kalapodi”, in Kaiser, I., Kouka, O. & Panagiotopoulos, D. (edd.), Ein Minoer im Exil. Festschrift zum 65. Geburtstag von Wolf-Dietrich Niemeier. Universitätsforschungen zur Prähistorischen Archäologie 188 (Bonn) 125–45. Hellner, N., 2016: “Gab es in der Frühphase des archaischen Tempelbaus kannelierte Holzsäulen?”, in Zambas, C., Lambrinoudakis, V., Simantoni-Bournia, E. & Ohnesorg, A. (edd.), ΑΡΧΙΤΕΚΤΩΝ: τιμητικός τόμος για τον καθηγητή Μανόλη Κορρέ. Honorary volume for Professor Manolis Korres (Athens) 555–68. Hellner, N., 2016/2017: “Sondage 2016 am Westgiebel—Die Dachausbildung und der Westgiebel des Südtempels von Kalapodi”, AM 131/132: 206–11. Hemans, F.P., 2015: “The Archaic Temple of Poseidon: Problems of Design and Invention”, in Gebhard, E.R. & Gregory, T.E. (edd.), The Bridge of the Untiring Sea: The Isthmus of Corinth from Prehistory to Late Antiquity. Hesperia Suppl. 48 (Princeton) 39–63. Hendrich, C., 2007: Die Säulenordnung des ersten Dipteros von Samos. Samos 25 (Bonn).

119

Herrmann, H.-V., 1972: Olympia. Heiligtum und Wettkampfstätte (München). Hoffelner, K., 1999: Das Apollon-Heiligtum: Tempel. Altäre. Temenosmauer. Thearion. Alt-Agina I 3 (Mainz). Hoffelner, K. & Kerschner, M., 1996: Die Sphinxsäule: Votivträger, Altäre, Steingeräte. Perirrhanterien und Becken. Alt-Ägina II 4 (Mainz). Hübner, G., 1980: “Die Dachterrakotten”, AA 1980: 112–15. Hübner, G., 1987: “Dach und Decke”, AA 1987: 76–82. Hübner, G., 1990: “Die Dachterrakotten der archaischen Tempel von Kalapodi (Phokis)”, in Winter, N.A. (ed.), Proceedings of the First International Conference on Archaic Greek Architectural Terracottas, December 2–4, 1988. Hesperia 59 (Princeton) 167–74. Hübner, G., 1995: “Zur Forschungsgeschichte griechischer Dachziegel aus gebranntem Ton. Leitvorstellungen und Annäherungen”, ArchEph : 115–61. Hübner, G., 1997: “Die archaischen Dachterrakotten von Kalapodi. Entwurf, Herstellung und Versatz”, in Muller, A. (ed.), Le Moulage en Terre Cuite dans l’Antiquité. Création et Production Dérivée, Fabrication et Diffusion. Actes du XVIIIe Colloque du Centre de Recherches Archéologiques—Lille III (7–8 déc. 1995) (Lille) 131–52. Jameson, M.H., 1963: “Excavations at Porto Cheli”, AD 18 (Chr.1): 73–74. Jameson, M.H., 1966: “Excavations at Porto Cheli”, AD 21 (Chr.1): 148–51. Jameson, M.H., 1967: “Excavations at Porto Cheli”, AD 22 (Chr.1): 195–96. Jameson, M.H., 1968: “Excavations at Porto Cheli”, AD 23 (Chr.1): 144–47. Jameson, M.H., 1969a: “Excavations at Porto Cheli”, AD 24 (Chr.1): 124–26. Jameson, M.H., 1969b: “Excavations at Porto Cheli and Vicinity, Preliminary Report, I: Halieis, 1962–68”, Hesperia 38.3: 311–42. Jameson, M.H., 1971: “Excavations at Porto Cheli”, AD 26 (Chr.1): 114–17. Jameson, M.H., 1974: “The Excavation of a Drowned Greek Temple”, Scientific American 231.4: 110–11. Jameson, M.H., 1982: “The Submerged Sanctuary of Apollon at Halieis in the Argolid of Greece”, National Geographic Society, Research Reports 14 (1973 Projects): 363–67. Kalpaxis, A.E., 1975: “Bemerkungen zu den Innensäulen des Heraion von Olympia”, AM 90: 83–96. Kalpaxis, A.E., 1976: Früharchaische Baukunst in Griechenland und Kleinasien (Athens). Kawerau, G., 1909: “Holzsäulen im dorischen Bau”, Zeitschrift für Geschichte der Architektur 2 (10/11): 227–33.

120 Kerschner, M. & Prochaska, W., 2011: “Die Tempel und Altäre der Artemis in Ephesos und ihre Baumaterialien. Naturwissenschafltliche Untersuchungen zur Herkunft der Marmore und ihr archäologischer Kontext”, ÖJh 80: 73–153. Kienast, H.J., 1988: “Die klassischen Tempel im Heiligtum von Kalapodi”, in Πρακτικά του XII Διεθνούς Συνεδρίου Κλασσικής Αρχαιολογίας, Αθήνα 4–10 Σεπτεμβρίου 1983, 61–64 (Athens) 100–03. Kienast, H.J., 2002: “Zum dorischen Triglyphenfries”, AM 117: 53–68. Knell, H., 1975: “Dorische Peripteraltempel mit gedrungenem Grundriss”, AA 1975: 10–13. Knell, H., 1983: “Dorische Ringhallentempel in spät- und nach­ klassischer Zeit”, JdI 98: 203–33. Kolia, E.I., 2011: “Α Sanctuary of the Geometric Period in Ancient Helike, Achaea”, BSA 106: 201–46. Kolia, E.I. & Gadolou, A., 2007: “Ναός Γεωμετρικών Χρόνων στην περιοχή της Αρχαίας Ελίκης”, Αρχαιολογία και Τέxνες 104: 71–73. Kolia, E.I. & Gadolou, A., 2011: “Ναός γεωμετρικών χρόνων στα Νικολαίϊκα Αχαίας: πρώτη παρουσίαση της ανασκαφής”, in Mazarakis Ainian, A. (ed.), The ‘Dark Ages’ Revisited. Acts of an International Symposium in Memory of William D.E. Coulson, Volos, 14–17 June 2007 (Volos) 191–209. Korres, M., 1995: Από την Πεντέλη στον Παρθενώνα: Kατάλογος της έκθεσης. Tα αρχαία λατομεία και η ιστορία ενός ημιτελούς δωρικού κιονοκράνου του πρώτου μαρμάρινου Παρθενώνος (Athens). Korres, M. & Ohnesorg, A., 2017: “Werkspuren im antiken Griechenland—Technik und Terminologie, mit einem Annex zu speziellen Werkspuren am Parthenon”, in Kurapkat, D. & Wulf-Rheidt, U. (edd.), Werkspuren: Materialverarbeitung und handwerkliches Wissen im antiken Bauwesen. Internationales Kolloquium in Berlin vom 13–16. Mai 2015, veranstaltet vom Architekturreferat des DAI im Henry-Ford-Bau der Freien Universität Berlin. DiskAB 12 (Regensburg) 11–32. Kuhn, G., 1993: “Bau B und Tempel C in Thermos”, AM 108: 29–47. Kyrieleis, H., 2008: “Der dorische Triglyphenfries und die mykenische Architektur”, in Junker, K., Stähli, A. & Kunze, C. (edd.), Original und Kopie: Formen der Konzepte der Nachahmung in der antiken Kunst. Akten eines Kolloquiums in Berlin 17.–19. Februar 2005 (Wiesbaden) 199–212. Liddell, H.G. & Scott, R., 1996: Greek-English Lexicon With a Revised supplement (Oxford). Lippolis, E., Livadiotti, M. & Rocco, G., 2007: Architettura greca: storia e monumenti del mondo della polis dalle origini al V secolo (Milan). Mallwitz, A., 1966: “Das Heraion von Olympia und seine Vorgänger”, JdI 81: 310–76. Mallwitz, A., 1972: Olympia und seine Bauten (Munich). Mallwitz, A., 1981: “Kritisches zur Architektur Griechenlands im 8. und 7. Jh”, AA 1981: 599–642.

Hellner Marconi, C., 2009: “Early Greek Architectural Decoration in Function”, in Counts, D.B. & Tuck, A.S. (edd.), Koine: Mediterranean Studies in Honor of R. Ross Holloway (Oakville) 4–17. Martin, R., 1965: Manuel d’architecture greque 1: matériaux et techniques (Paris). Mattern, T., 2016: “»Wer der Baumeister war, daran erinnert man sich nicht mehr.« (Paus. 5, 16, 1)—Zur Holzphase des Heraions von Olympia”, in Schwarzer, H. & Nieswandt, H.H. (edd.), »Man kann es sich nicht prächtig genug vorstellen!« Festschrift für Dieter Salzmann zum 65. Geburtstag (Münster) 621–35. Mazarakis Ainian, A., 1997: From Rulers’ Dwellings to Temples: Architecture, Religion and Society in Early Iron Age Greece (1100–700 B.C.). SIMA 121 (Jonsered). Meiggs, R., 1983: Trees and Timber in the Ancient Mediterranean World (Oxford). Mertens, D., 2006: Städte und Bauten der Westgriechen: von der Kolonisationszeit bis zur Krise um 400 vor Christus (München). Metzger, H., 1940/1941: “Le sanctuaire de Glanitsa (Gortynie)”, BCH 64/65: 5–33. Miles, M.M. (ed.), 2016: A Companion to Greek Architecture (Chichester). Morgan, C.A., Pitt, R.K. & Whitelaw, T., 2009: “Archaeology in Greece. Archaeological Reports for 2008–2009”, ARepLondon 55: 1–101. Moustaka, A., 2010: “Considerazioni sugli acroteri in forma di cavallo”, in Lulof, P. & Rescigno, C. (edd.), Deliciae Fictiles IV: Architectural Terracottas in Ancient Italy Images of Gods, Monsters and Heroes. Proceedings of the International Conference held in Rome (Museo Nazionale Etrusco di Villa Giulia, Royal Netherlands Institute) and Syracuse (Museo Archeologico Regionale ‘Paolo Orsi’) October 21–25, 2009 (Oxford) 69–73. Neve, P., 1989: “Eine hethitische Bronzesäge aus HattusaBoğazköy”, IstMitt 39: 399–406. Niemeier, W.-D., 2006: “Deutsches Archäologisches Institut, Jahresbericht 2005. Abteilung Athen, Kalapodi”, AA 2006: 166–68. Niemeier, W.-D., 2009: “Deutsches Archäologisches Institut, Jahresbericht 2008. Abteilung Athen, Kalapodi”, AA 2009: 107–10. Niemeier, W.-D., 2011: “Deutsches Archäologisches Institut, Jahresbericht 2010. Abteilung Athen, Kalapodi”, AA 2011: 97–99. Niemeier, W.-D., 2012: “Deutsches Archäologisches Institut, Jahresbericht 2011. Abteilung Athen, Kalapodi”, AA 2012: 94–96. Niemeier, W.-D., 2013: “Kultkontinuität von der Bronzezeit bis zur römischen Kaiserzeit im Orakel-Heiligtum des Apollon

EARLY TEMPLES BUILT OF WOOD AND STONE: NEW FINDS FROM Kalapódhi ( PHOKIS ) von Abai (Kalapodi)”, in Raue, D. & Gerlach, I. (edd.), Sanktuar und Ritual: Heilige Plätze im archäologischen Befund. ForschungsCluster 4.10 (Rahden) 33–42. Niemeier, W.-D., 2016: “Das Orakelheiligtum des Apollon von Abai/Kalapodi”, Trierer Winckelmannsprogramm 25 (2013): 1–46. Orlandos, A.K., 1955–1958: Τα ύλικά δομής των αρχαίων Ελλήνων κατά τους συγγραφείς, τας επιγραφάς και τα μνημεία (Athens). Østby, E., 1980: “The Athenaion of Karthaia”, OpAth 13: 189–223. Østby, E., 1986: “The Archaic Temple of Athena Alea at Tegea”, OpAth 16: 75–102. Østby, E., 1992/1993: “The Temples of Pallantion and the Archaic Temple Architecture of Arcadia”, in Πρακτικά τοῦ Δ΄ Διεθνοῦς Συνεδρίου Πελοποννησιακῶν Σπουδῶν, Κόρινθος 9–16 Σεπτεμβρίου 1990. Peloponnesiaka Suppl. 19 (Athens) 65–75. Østby, E., 1995: “Scavi di Pallantion. Templi di Pallantion e dell’Arcadia. Confronti e sviluppi”, ASAtene 68 (1990–91): 285–391. Østby, E., 2000: “Delphi and the Archaic Doric Architecture in the Peloponnese”, in Jacquemin, A. (ed.), Delphes cent ans après la grande fouilles. BCH Suppl. 36 (Paris) 239–61. Østby, E. (ed.), 2005: “Ancient Arcadia”, in Østby, E. (ed.), Papers from the Third International Seminar on Ancient Arcadia, held at the Norwegian Institute at Athens, 7–10 May 2002 (Athens) 493–506. Pernier, L., 1931: L’Artemision di Cyrene. Africa Italiana 4 (Bergamo). Petropoulos, M., 1987: “Τρίτη ἀνασκαφικὴ περίοδος στὸ Ἄνω Μαζαράκι (Ρακίτα) Ἀχαΐας”, in Πρακτικά τοῦ Γ’ Διεθνοῦς Συνεδρίου Πελοποννησιακῶν Σπουδῶν, Καλαμάτα 8–15 Σεπτεμβρίου 1985. Peloponnesiaka Suppl. 13 (Athens) 81–96. Petropoulos, M., 1992–1993: “Περίπτερος ἁψιδωτὸς γεωμετρικὸς ναὸς στὸ Ἄνω Μαζαράκι (Ρακίτα) Πατρῶν”, in Πρακτικά τοῦ Δ’ Διεθνοῦς Συνεδρίου Πελοποννησιακῶν Σπουδῶν, Κόρινθος 9–16 Σεπτεμβρίου 1990. Peloponnesiaka Suppl. 19 (Athens) 141–157. Petropoulos, M., 1996: “Άνω Μαζαράκι (Ρακίτα)”, AD 50: 220–25. Petropoulos, M., 2001: “Γεωμετρικός ναός Ρακίτας. Λατρευóμενη θεóτητα”, in Mitsopoulos-Leon, V. (ed.), Forschungen in der Peloponnes. Akten des Symposions anlässlich der Feier “100 Jahre Österreichisches Archäologisches Institut Athen”, Athen 5.3.–7.3.1998, (Athens) 39–45. Petropoulos, M., 2002: “The Geometric Temple at Ano Mazaraki (Rakita) in Achaia”, in Greco, E. (ed.), Gli Achei e L’identità etnica degli Achei d’Occidente, Atti del Convegno di Studi. Tekmiria 3 (Athens) 143–64. Richter, G.M.A., 1966: The Furniture of the Greeks, Etruscans and Romans (London). Sapirstein, P., 2016: “The Columns of the Heraion at Olympia: Dörpfeld and Early Doric Architecture”, AJA 120.4: 565–601.

121

Schleif, H., 1940: “Der Artemistempel. Architektur, Dachterrakotten, Inschriften”, in Rodenwaldt, G. (ed.), Korkyra I (Berlin) 15–96. Schwandner, E.-L., 1985: Der ältere Porostempel der Aphaia auf Ägina. DAA 16 (Berlin). Simantoni-Bournia, E., Mendoni, L.G. & Panagou, T., 2009: Καρθαία ... ελαχύνωτον στέρνον χθονός (Athens). Simpson, E., 1999: “Early Evidence for the Use of the Lathe in Antiquity”, in Betancourt, P.P., Karageorghis, V., Laffineur, R. & Niemeier, W.-D. (edd.), Meletemata: Studies in Aegean Archaeology Presented to Malcolm H. Wiener as he enters his 65th Year. Aegaeum 20 (Liège) 781–86. Simpson, E. & Spirydowicz, K., 1999: Gordion Wooden Furniture (Ankara). Sinn, U., 2001: “Die Stellung des Hera-Tempels im Kultbetrieb von Olympia”, in Bietak, M. (ed.), Archaische Griechische Tempel und Altägypten (Vienna) 63–70. Sulze, H., 1936: “Das dorische Kapitell der Burg von Tiryns”, AA 51: 14–36. Themelis, P.G., 1983: “Δελφoί και περιoχή τoν 8o και 7o π.Χ. αιώνα. Φωκίδα, δυτική Λoκρίδα”, ASAtene 45: 213–55. Vallois, R., 1908: “Études sur les formes architecturales dans la peinture des vases grecs”, RA 11: 359–90. von Gerkan, A., 1946/1947: “Betrachtungen zum ionischen Gebälk”, AM 61/62: 17–29. von Gerkan, A., 1948/1949: “Die Herkunft des dorischen Gebälks”, JdI 63/64: 1–13. von Kienlin, A. (ed.), 2011: Holztragwerke der Antike. Internationale Konferenz 30. März–1. April 2007 in München. Byzas 11 (Istanbul). Vordos, A.G. & Kolia, E.I., 2008: ΑΙΓΙΑΛΕΙΑ. Αρχαίες πόλεις και μνημεία (Patras). Weißl, M., 2002: “Grundzüge der Bau- und Schichtenfolge im Artemision von Ephesos”, ÖJh 71: 313–46. Weißl, M., 2010: “Tempelarchitektur zwischen Sardes und Sais”, in Rollinger, R., Gufler, B., Lang, M. & Madreiter, I. (edd.), Interkulturalität in der Alten Welt: Vorderasien, Hellas, Ägypten und die vielfältigen Ebenen des Kontakts. Phillipika 34 (Wiesbaden) 125–49. Welter, F.G., 1938: “Aegenetica I–XII”, AA 53: 1–33. Welter, F.G., 1941: Troizen und Kalauria (Berlin). Wesenberg, B., 1971: Kapitelle und Basen: Beobachtungen zur Entstehung der Griechischen Säulenform. BJb Beiheft 32 (Düsseldorf). Wesenberg, B., 1996: “Die Entstehung der griechischen Säulen- und Gebälkform in der literarischen Überlieferung der Antike”, in Schwandner, E.-L. (ed.), Säule und Gebälk: zu Struktur und Wandlungsprozess griechisch-römischer Architektur. Bauforschungskolloquium in Berlin vom 16–18. Juni 1994. DiskAB 6 (Mainz) 1–15.

122 Wesenberg, B., 2008: “Pro Vitruvio—iterum: Zur mimetischen Formengenese in der griechischen Architektur”, in Junker, K., Stähli, A. & Kunze, C. (edd.), Original und Kopie: Formen der Konzepte der Nachahmung in der antiken Kunst. Akten eines Kolloquiums in Berlin 17–19. Februar 2005 (Wiesbaden) 185–97. Whitley, J., Germanidou, S., Urem-Kotsou, D., Dimoula, A., Nikolakopoulou, I., Karnav, A. & Hatzaki, E., 2006: “Archaeology in Greece 2005–2006”, ARepLondon 52: 1–112. Whitley, J., Germanidou, S., Urem-Kotsou, D., Dimoula, A., Nikolakopoulou, I., Karnav, A. & Evely, D., 2007: “Archaeology in Greece 2006–2007”, ARepLondon 53: 1–121.

Hellner Wide, S.K.A. & Kjellberg, L., 1895: “Ausgrabungen auf Kalauria. Topographie und Architektur”, AM 20: 267–87. Wilson Jones, M., 2014: Origins of Classical Architecture: Temples, Orders and Gifts to the Gods in Ancient Greece (New Haven). Winter, N.A., 1993: Greek Architectural Terracottas: From the Prehistoric to the End of the Archaic Period (Oxford). Wurster, W.W., 1973: “Dorische Peripteraltempel mit gedrungenem Grundriß”, AA 1973: 200–11. Wurster, W.W., 1974: Der Apollontempel. Alt-Ägina I.1 (Mainz).

chapter 7

Recent Architectural Studies at Goúrnia in East Krete: 2011–2016 D. Matthew Buell, John C. McEnroe, Jorge Andreas Botero Besadalombana, and Rafał Bieńkowski 1

Earlier Architectural Studies at Goúrnia

Shortly after the first season of excavations at Goúrnia in 1901, the British archaeologist David Hogarth wrote an enthusiastic article in The Times, reporting that “[Goúrnia] is the most perfect example yet discovered of a small ‘Mycenaean’ town, uncontaminated with later remains, and at this moment, after the two great palaces, it is the ‘sight’ best worth visiting in Krete” (Quoted in Hawes et al. 1908: v). Excavated in three seasons (1901, 1903, and 1904) under the direction of Harriet Boyd (later Hawes), Goúrnia has remained the archetypal example of a small city in the Bronze Age Aigaian for more than a century (Boyd 1904; 1904–1905; No author reported 1906; Hawes et al. 1908; Fotou and Brown 2004). Narrow streets follow the contours of the low hill, or scale the steep ascent to the broad,

public court on the saddle beneath the summit (Fig. 7.1a). A small palace overlooks the public court. Both are surrounded by the houses of potters, metal workers, carpenters, and farmers. The picturesque town provides a sense of intimacy that perfectly complements the grandeur that Arthur Evans was revealing at Knossos during the same years. Goúrnia provided an important window onto life in the Bronze Age. As Hawes put it, “The chief archaeological value of Goúrnia is that it has given us a remarkably clear picture of the everyday circumstances, occupations and ideals of the Aigaian folk at the height of their true prosperity” (Hawes et al. 1908: 27). The plan that Hawes published in 1908 has been so frequently reproduced that it too has come to acquire an iconic status (Fig. 7.1a). Hawes did much of the drawing herself over the course of five years. During the first season in 1901, Hawes measured and made annotated sketches of

figure 7.1 Plans of the Town at Goúrnia: (a) 1908 plan, Hawes 1908; (b) current plan of Goúrnia, by the authors

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004416659_009

124 each of the buildings. At the end of the excavation season, she called on Theodore Fyfe, the architect working with Arthur Evans at Knossos, to do an instrument survey of the site and to fit her sketches to a gridded plan (Boyd 1904: 33). This formed the basis of the plan that Hawes published in 1904. The final plan of the site was drawn during the study season in 1905. That year the Austrian architect Wassily Sejk worked with Boyd Hawes at Goúrnia for 16 days. He spent a month during the following winter combining regularised tracings of Hawes’ field sketches with Fyfe’s gridded plan (No author reported 1906). Since its initial publication in 1908, this plan has served as the basic analytical tool for understanding the site. Over the years, other scholars have made important contributions to several aspects of the architecture at Goúrnia. In 1910 Edith Hall returned to the site to excavate the cemetery on the adjacent hill of Sphoungarás (Hall 1912). There she continued the excavation of the early Prepalatial rock shelter burials and inhumations, and she also excavated more than 150 pithos burials that belonged to a vast cemetery associated with the later Middle Minoan III–Late Minoan I (MM III–LM I) town. In 1971 and 1972 Costis Davaras excavated two more house tombs in the North Cemetery (Davaras 1973, 1974). In 1979 and 1992 Jeffrey Soles published supplementary studies of the Protopalatial town at Goúrnia and the tombs in the North Cemetery (Soles 1979, 1992). In 1980 Soles began an architectural study of the palace, which included the creation of a new and relatively accurate state plan, using a builder’s transit (Soles 1991). In 1992 Wilson and Eleanor Myers published a low-altitude balloon photograph that literally provided us with a new perspective on the town (Myers, Myers & Cadogan 1992). Vasso Fotou’s 1993 publication included an annotated version of Hawes’s plan and provided access to previously unpublished documents, including several of Hawes’s original field drawings (Fotou 1993). These gave us new insight into Hawes’ excavation and recording methods, and brought us much closer to the groundbreaking early years of the excavation. More recently in the mid-1990’s, David Romano did the first total station survey work at the site, attempting to establish a GIS framework that might be used to orthorectify Soles’ plan of the palace and the Myers’ balloon photo (Romano 2003). At about the same time, L. Vance Watrous directed an archaeological survey of the Goúrnia area and linked the results with those of other recent surveys in the adjacent Mirabello region (Watrous et al. 2012; 2015). In 2008 and 2009 Watrous reinvestigated the buildings that Hawes had partly excavated along the shore, north of the town site (Watrous 2012). Watrous directed a five-year excavation project at the site (2010–2014) concerned with documenting the

BUELL ET AL.

settlement’s early history. As part of the Gournia Excavation Project, in 2011 we began a study of the settlement’s architectural features. Over the course of four seasons of fieldwork and two study seasons, this project has produced a new and accurate map of the settlement, as well as detailed studies of architectural phasing and masonry styles. In order to assist us in these studies, we implemented a program of photogrammetric modelling at both high- and low-resolution scales. Finally, we have also begun to study Goúrnia’s landscape in an effort to unravel the complex relationship between it and the town’s built environment. Our studies reveal that the history of Goúrnia was vastly more complex and nuanced than previously thought. 2

Recent Architectural Studies of the Gournia Excavation Project

While scholars focused on various components of the plan, the palace, the cemeteries, the Shore Buildings, etc., no one has addressed the obvious issue: the overall site plan. This was probably due, in part, to the daunting scale of the project. To measure and draw each wall of more than 60 houses would require a significant investment of time and labour. In addition, over time Hawes’s plan had acquired a certain authority of its own (Fig. 7.1a). The plan’s level of detail gave it believability, and numerous reproductions made it familiar to later generations, giving it a sense of permanence, fixity, and inevitability. The plan proclaimed: “This is what a Bronze Age town looked like”. The problem, of course, is that the Bronze Age town never looked like this meticulously constructed drawing. Despite the two sections provided at the top, the plan, unlike the site, is flat. In addition, the plan freezes time. It creates a conceptual moment in the life of a town that continually changed over the course of two millennia. Like any site plan, it shows us buildings that, until the end of the excavations in 1904, had never actually coexisted. When we began the Gournia Excavation Project in 2010, it did not occur to us to consider the theoretical implications of Hawes’ plan. We naively assumed that we would be able to continue in the comfortable tradition of working around the edges, adding to, rather than replacing the 1908 representation. We made detailed state plans of the newly excavated buildings in the northern part of the site, took appropriate photos, and hoped that we could simply add our work to the original plan. By the end of the first season, we realised that would not be feasible, since it was not possible to coordinate the locations of newly found features with the existing drawing. For these purely practical reasons, we realised that we would have to create an entirely new plan.

Recent Architectural Studies at Goúrnia in East Krete: 2011–2016

Early on, we decided not to draw a stone-by-stone state plan of the site because, over the years, the top courses of many of the walls been altered during various consolidation projects, and we thought that any stone-by-stone plan would imply a misleading illusion of precision. We also tried to keep something of the look of the classic 1908 plan (cf. Figs. 7.1a, 7.1b). We based our new plan on the same Greek military 1:5000 topographical map and geodetic markers that Romano had used, so that the two surveys would be consistent. From 2011 to 2013 we shot more than 90,000 points on features from both the new and the old excavations. This spatial data was then entered into a database and uploaded into GIS software (Esri’s ArcMap) to create shapefiles of walls and other built and unbuilt features. During the slow process of drawing the site we made several significant discoveries. We came across walls, numerous rooms, and in some blocks, even entire buildings that had not been previously recorded (cf. Figs. 7.1a, 7.1b). In our plan the site began to take on a new appearance. In contrast to the 1908 plan, for example, the walls in our plan are not uniform in thickness. They have few straight lines, and few right-angle corners. While to modern eyes, these irregularities might seem be the result of simple carelessness on the part of the builders, they actually reflect the fascinating complexity of the Minoan design process. As Clairy Palyvou observed in 1990, designing a Minoan house was not a simple matter of drawing lines on a two dimensional surface and then replicating the pattern in stone and mudbrick (Fig. 7.2a) (Palyvou 1990). Minoan domestic architecture belonged to the broad category of vernacular architecture in which everyone in the community shared a common set of expectations about what a house should look like, how it was to be used, and how it would be constructed (Rudofsky 1965; Rapoport 1969: 4; McEnroe 1990: 195; Palyvou 1990: 54–55). The job of the builder was to adjust that conceptual model to fit the specific requirements of the available building plot. This meant that each wall had to be individually conceived to fit the slope of the building plot and had to adapt to existing architectural features. This process of design necessarily required a special set of skills. Builders had to be able to think three-dimensionally. While working on the tiny ground-story rooms, the builder had to be able anticipate the intended layout of the more important upper-story spaces. In addition, the builder needed the flexibility to make continual adjustments as the building took shape. In contrast to modern practice in which design and construction are two distinct processes, often carried out by two different parties, in pre-industrial architecture they were part of the same experiential process. The design evolved during the construction project (Palyvou 1990). A

125

2a

2b

figure 7.2 Plan of House Ac at Goúrnia: (a) construction sequence; (b) diagram of wall junctions. Circles between joints indicate that the walls were bonded, while spaces indicate that they are abutted both plans by the authors

Minoan house was not a static object, but part of an ongoing process. That process often continued long after the completion of the initial construction phase, since houses continually deteriorated and underwent repairs, and they expanded and contracted. Things accumulated and were cleared out. Sometimes houses were abandoned and later reoccupied, taking on a new life. Over time, each house acquired its own unique structural biography (Weiberg 2007; Driessen 2010: 43–45). House Ac provides a good example of these processes (Fig. 7.2a). Rather than simply assigning the building to the Neopalatial period, we can recognise four major building phases spread across the approximately eight generations of people who lived in the house over the course of two centuries. Only the east wall remains

126 from the first phase. During the second phase, House Ac actually belonged to a larger building that included what is now the south wall of House Ab to the north. In phase 3, a narrow alley divided Houses Ab and Ac into separate buildings, and House Ac became much smaller than its neighbour to the north. The thin coursed rubble west wall was the latest repair to the house shortly before its destruction by fire at the end of the LM IB period. This tidy but fragile facade wall probably indicates that by this phase the western part of the house had been reduced to a single story. Each of the four phases of House Ac could be further divided into sub-phases. Those sub-phases can, in turn, also be more closely examined, and in some cases it is possible to reconstruct mentally the sequence in which the stones in a particular wall were laid. The overall picture of House Ac that emerges from these exercises is one of ongoing ad hoc changes rather than a series of distinct, periodic construction phases. House Ac continually changed over time. These architectural changes can be interpreted in different ways. For example, Jan Driessen regards them as relatively minor modifications to a fundamentally continuous structure. He writes, “The architectural container represents stability even if it is modified, embellished, enlarged or even moved from one spot to another. Humans die, the House does not—but at the same time it offers its inhabitants a sense of immortality” (Driessen 2010: 38–40). This, in turn, leads Driessen to argue “for a slow evolution and a high degree of conservatism in Minoan society between the Neolithic and the Late Bronze Age in which continuity of place played a major role” (Driessen 2010: 57). Yet there is another way to read these architectural alterations. Rather than reflecting intergenerational continuity, ongoing architectural modifications of the sort we have seen in House Ac may be indications of correspondingly dynamic changes among the members of the continually shifting community. As Michael E. Smith has recently reminded us, demographic change and population mobility at the local level are much more characteristic of preindustrial towns than we have traditionally assumed. People routinely migrated from the surrounding countryside or from distant places in pursuit of safety, economic security, or advantages of economic scale. Too often we have simply failed to recognise that mobility in the archaeological record (Smith 2014; Buell & McEnroe 2016), which can affect our understanding of the complexity of ancient communities. Susan Alcock warns that choosing to emphasise continuity rather than change means running the risk of failing to appreciate the cultural diversity within and among the ancient communities, and

BUELL ET AL.

minimising the extent to which circumstances changed over time (Alcock 2002: 101). One simple way to record some of these changes involves the examination of butted and bonded joints (Fig. 7.2b). At each point where two walls met, we recorded whether one wall was added against the other or whether they appeared to have been bonded together as integrated parts of a single construction (Kevin Glowacki suggested how to represent these wall junctions schematically). While some abutting walls were, no doubt, constructed during the building’s initial construction phase or soon after, others were added over time, as the structure changed. By observing bonded and abutting walls and comparing other wall features—including masonry fabrics (explained below), dimensions, relative levels, associated features, and the like—one can develop a sequence of construction. Following this method was not only helpful in understanding the construction sequence of individual buildings, such as House Ac, but it is a technique that we have applied across the entire site. We now have a reasonably clear understanding of how the town evolved as a whole, almost like a living organism, over the course of its long life. Nearly every building at the site has provided evidence of this complex sub-phasing. In order to keep track of these complicated changes, we established a database of all the architectural features (walls, windows, doors, thresholds, column bases, gournes [stone basins], kernoi, etc.) and their key attributes and features (dimensions, materials, associated features, etc.). The database allows us to compare quickly categories of features—thresholds, for example—across the site to determine whether there are meaningful patterns of distribution. One important attribute that we have studied across the site is masonry fabric. Over the years there have been a number of attempts to categorise different types of masonry in Bronze Age architecture (Walsh & McDonald 1986; McEnroe 2001: 36–38; Shaw 2009). The problem with developing a comprehensive set of categories for all of Bronze Age Krete is that the specific materials and techniques used in rubble masonry, the most common form of Minoan construction, are site specific. That is, over time, each site arrives at its own specific style of masonry (McEnroe 1990). At Goúrnia, we have been able to define several distinct masonry fabrics (Figs. 7.3a–f). In our terms, a “masonry fabric” is not simply a building style. Similar to a ceramic fabric, it involves three interdependent components: the materials, the masonry techniques, and the sources of the materials. We have learned that several of these masonry fabrics were chronologically specific.

Recent Architectural Studies at Goúrnia in East Krete: 2011–2016

127

figure 7.3 Masonry fabrics at Goúrnia: (a) monumental white crystalline limestone; (b) monumental grey tripolitsa and conglomerate; (c) coursed ashlar; (d) half-timber; (e) mud brick on stone socle; (f) large coursed rubble photographs by Janet Spiller

We have, for example, identified two distinct monumental rubble fabrics incorporating stones larger than 0.75 m in at least one dimension. The earlier MM IB–MM II fabric consists of massive boulders of unworked white crystalline limestone laid irregularly (Fig. 7.3a). The later MM III–LM I monumental rubble fabric is of water-worn tripolitsa and conglomerate boulders that had been hammer dressed (Fig. 7.3b). The stones chosen for these two monumental styles came from different sources. First, the white crystalline limestone forms the bedrock ridge that runs through the town from north to south. Natural cleavages make it easily available in the outcrop just south of the town. Second, the water-worn tripolitsa and

conglomerate boulders of the later monumental walls were collected from the riverbed north of the settlement. The process of collecting the stones and transporting them to the building site before the actual construction began allowed each wall to be constructed in a consistent style. Other masonry fabrics include coursed ashlar and coursed rubble masonry (both dating to LM IB), mud brick on a rubble socle, and half-timber construction (Figs. 7.3c–f). Plotting the distribution of these masonry fabrics across the site has provided some interesting information (Fig. 7.4). For example, previously scholars had followed Hawes in dating the earliest palace to the later Neopalatial

128

figure 7.4 Distribution of the masonry fabrics in Goúrnia plan by authors

BUELL ET AL.

Recent Architectural Studies at Goúrnia in East Krete: 2011–2016

period, specifically to the LM IA period (Hawes et al. 1908: 24; Soles 1991: 21–26; Driessen & Macdonald 1997: 213; Gomrée 2013: 128–83). Combining our identification of masonry fabrics with the examination of construction sequencing and strategically placed stratigraphic soundings has allowed us to date the earliest construction of the palace to the Protopalatial (MM IB–MM II) period, extending its history back by roughly two centuries. This is one of the most important discoveries that the Gournia Excavation Project has made, because it significantly changes our understanding of the historical and social development of the town. The early palace was not an isolated monument. We have also been able to document that it was part of a much broader, sweeping transformation of the town in the Protopalatial period. This massive transformation of the town also involved the construction of the first monumental houses at the site, and the construction of the first extensive system of paved streets that, along with the broad plateias, knitted the monumental buildings with the smaller buildings into a cohesive community of interdependent parts (Buell & McEnroe 2015a, 2015b, 2016). 3

Photogrammetric Modelling

During the course of making the new plan, our understanding of the history and the organisation of the town fundamentally changed. Yet, we also came to realise that our plans, like those of Hawes, are schematic, twodimensional, conceptual images or constructs. They do not provide a sense of the buildings as three-dimensional structures built around the contours of an irregularly sloping hill. To address this shortcoming, Jorge Botero began a programme of terrestrially based high-resolution photography and photogrammetric recording in 2015 (Figs. 7.5a, b, d). In just two seasons (ca. 13 weeks) Botero took more than 27,000 photos, using the total station to geo-reference the work of each day. We used the semiautomated PhotoScan software (AgiSoft LLC) to extract a three-dimensional point cloud and to render a mesh from which we can compute various other models (Verhoeven 2011; Remondino et al. 2014). From these, it is possible to generate architectural sections and Digital Elevation Models (DEMs). By highlighting walls we can create interpretive hybrids of orthophotos and plans (Fig. 7.2a). Even in a simplified form, these plans are much more detailed, more complete, and more subtle than traditional drawings. These three-dimensional models allow us to look at the buildings from every point of view and to move virtually through the town. There are

129

several other advantages to the use of photogrammetry. These include: 1) Photogrammetry provides more detailed plans than conventional drawing; 2) it is faster than traditional drawing (provided that an area is well-cleaned, we estimate that this method, after data capture and post-processing, takes roughly half the time than does traditional drawing); 3) it gives us the ability to generate sections “on-the-fly”; 4) we are able to do modelling on top of the mesh for study and for three-dimensional reconstructions; 5) it preserves sufficient data for other scholars to check our interpretations, providing for greater accountability; and 6) it will also prove to be an important tool for public interpretation, conservation and historic preservation. One disadvantage of using high resolution images for photogrammetric recording is the sheer size of the model. The full photogrammetric project encompasses 725 GB of data, with the photoset for the processing encompassing 202 GB, the DEMs 5.34 GB, and the rest divided into 59 high resolution three-dimensional models, some as large as 1.5 GB, and 59 low-resolution orthophotos (at resolutions of both 1 mm and 5 mm per pixel). How big can or should a plan be? Is there a point when a plan becomes too detailed? Jorge Luis Borges alluded to a similar dilemma in a short parable about a guild of cartographers who were obsessed with making a perfect map of the empire. They eventually “struck a map whose size was that of the Empire and which coincided point for point with it”. It was a perfect simulacrum of the actual empire. Later generations who were less interested in cartography for the sake of cartography itself realised that the vast map was useless (Borges 1998). With modern technology Borges’ fictional map has become a virtual/digital possibility. While our high-resolution photogrammetric model is not useless, it is large, and it requires time and significant computing power to manipulate. Due to the sheer size of the photogrammetric project, we face two major challenges. The first is storage and the potential for data loss, which we combat by storing all information on a project server, as well as two external hard drives. The second challenge concerns the usability of the stored data. We are currently in the process of designing a database that allows the consultant to visualise only the information required and the resolution needed for a specific task. The general organisation of the database follows a scheme: block, display of grid, selection of a specific area, and selection of the information required (e.g., high resolution three-dimensional model, DEM, or orthophotos).

130

figure 7.5 Photogrammetric images: (a) Digital Elevation Model; (b) orthorectified photograph; (c) Digital Elevation Model of the Goúrnia landscape; (d) photogrammetric image of the Town at Goúrnia as if seen from the southwest all images by authors

BUELL ET AL.

Recent Architectural Studies at Goúrnia in East Krete: 2011–2016

4

Landscape Studies

In the Gournia Excavation Project, we are interested not only in the architecture within the town, but also in human interventions in the surrounding area (Fig. 7.6). We are concerned with the landscape not as a neutral spatial setting, but as a meaningful component in people’s lives, shaped by (and shaping) their experiences, memories, and interactions (Tilley 1997; Ashmore & Knapp 1999; Alcock 2002; Herva 2006a; 2006b; Hitchcock 2007; Thomas 2001). At the north edge of the town, two conspicuous cemeteries mark the places where the living townspeople interacted with their ancestors, continually reminding them of both their origins and final destinations (Varvouranakis 2006; Legarra Herrero 2014). Buildings on the shore are also clearly visible from much of the town. Together with the line of the ancient road, the shore buildings signaled the connections between the town and the outside world (Watrous 2012). On the surrounding hills, terraces reshaped and stabilised the natural slopes, serving as witnesses to farmers’ lifelong labours and of the sources of the townspeople’s daily sustenance. We not only located

figure 7.6 Town and landscape at Goúrnia, viewed from the south image by authors

131

these key components of the landscape, but we have also begun to document how relationships among them changed over the course of the town’s history. We learned that often those changes in the local environment corresponded with significant changes within the town (Buell & McEnroe 2016). In order to record and investigate these issues, Rafał Bieńkowski began a programme of topographical mapping and low-altitude drone-based photography, videos, and photogrammetry that would expand our investigations into the surrounding micro-region (Fig. 7.5c). For the topographical map, due to the variable topography and the overgrowth of vegetation in some areas, we did not follow a strict pattern for data collecting. Instead, we took spatial measurements using a total station every 10 meters or so on average in accordance with prominent topographical features, including bottomland, and the base, breaking slope, and summit of significant rises in the local topography. When the topography was irregular, we collected our spatial data at closer intervals (ca. 1–2 m), whereas in flat bottom-land we used larger intervals (ca. 10 m). This data was then uploaded to the Goúrnia spatial database, and processed with GIS software to produce a DEM, from

132 which we generated a contour map with extrapolated 1-meter contours. The resolution of the intermediate DEM is such that one pixel has an area of 9 m². This new and up-to-date map, when further refined, will replace the lower resolution 1:5000 Greek army map that served as the basis for all earlier regional studies. This DEM will be a valuable source of information in the future (e.g., intervisibility, slope, watershed, least-cost, etc.) and as a starting point for more detailed analysis of the town and its hinterlands. As part of this programme, in 2016 we took ca. 3,000 aerial photos, expanding our programme of photogrammetric recording to cover some 54 hectares (Fig. 7.5c). To limit the data to a manageable size, we decided to take photos and videos at two different altitudes. Within the settlement, we captured photos at 10 meters above ground level, while those of the surrounding micro-region were taken from 20 meters above ground level. All ca. 3,000 photos were aligned in PhotoScan and orthorectified using the spatial data from ground control points captured by the total station. The model was then generated automatically with some points having to be adjusted manually. Photorealistic textures taken from the photos were overlaid on the model and an orthoimage was created. During the 2017 season we expanded the area to include the entirety of the Goúrnia valley and the surrounding hilltops. As with our ground-based, high-resolution photo­grammetry, perhaps the most distinctive feature of our drone-based photogrammetry is the sheer scale of the project. The lower resolution data gathered with the drone are also providing an opportunity to experiment with 3D modelling. Working from the DEM and our orthophotos, we used Photoscan to generate a point cloud. The point cloud was then used to create a 3D model, to which we added realistic textures extracted from the orthophotos. Eventually it will be possible to extend this modelling into the entire surrounding landscape. The advantages of creating a low resolution model are numerous. First, it is smaller and, therefore, easier to transfer between computers, servers, and users. Second, with the use of aerial photos it is possible to generate a model for the whole site at once, without the need of dividing it into smaller “chunks” that may be processed within the memory limits of the system. Third, it is easier to create a 3D model of the terrain, which is adequate for topographical studies, in part because the time required for capturing photos was much reduced—in this case, seven days compared to three seasons of fieldwork. The

BUELL ET AL.

obvious disadvantage of the low resolution model when compared to the high resolution one is the loss of detail. Still, the low resolution model serves as a perfect tool for preliminary research and survey documentation. 5

New Technologies and New Scholarly Goals

Much of this paper will probably seem to have been about technological changes in methods of recording architecture from Boyd Hawes’s hand-drawn field sketches to drone-based terrestrial and aerial photogrammetry. That was not our initial goal when we joined the Gournia Excavation Project, nor is it the main thing we have learned over the course of our six seasons at the site. Our interest is not in technology for the sake of technology, but in how our understanding of the ancient community fundamentally changes as we move from a fixed, two-dimensional schematic drawing to dynamic, three-dimensional models. At this point we have not yet focused on using our three-dimensional modelling as a presentation tool—although that will come—rather, we are interested in using these new technologies as investigative tools to help us understand the complexities and the nuances of this ancient community. We have learned much over the past seven seasons. We have become more aware that each building at the site has a complex, individual biography (Herva 2005; 2006a; 2006b; Weiberg 2007; Driessen 2010). We have also come to appreciate that the same is true for the community as a whole, and for the broader ecosystem of which the houses and town were part. These too had dynamic life histories. In other words, our approach to architecture is changing, and new technology is enabling us to explore our evolving questions. We are increasingly interested in investigations at different scalar levels, from the micro-studies of mud bricks or the masonry of individual walls, to the interactions among houses within the town, between the town and the surrounding landscape, within the Mirabello region, and the wider Mediterranean (Knappett 2009). We understand that the term “community” is a social and cultural concept as well as a spatial unit (Yaeger & Canuto 2000). We are coming to a better understanding of Goúrnia as not a two- or three-, but a four-dimensional site. We understand that the term “building” is a verb as well as a noun—a process as well as an object. Most importantly, we are beginning to realise that our study is not primarily about buildings; it is about people and how they interacted in space and through time.

Recent Architectural Studies at Goúrnia in East Krete: 2011–2016

List of References Alcock, S.E., 2002: Archaeologies of the Greek Past: Landscape, Monuments, and Memories (Cambridge). Ashmore, W. & Knapp, A.B. (edd.), 1999: Archaeologies of Landscape. Contemporary Perspectives (Malden). Borges, J.L., 1998: “Museum, on Exactitude in Science”, in A. Hurley (ed. and trans.), Collected Fictions (New York) 325–27. Boyd, H., 1904: “Gournia: Report of the American Exploration Society’s Excavations at Gournia, Crete, 1901–1903”, Transactions of the Department of Archaeology, Free Museum of Science and Art, University of Pennsylvania 1.i: 7–44. Boyd, H., 1904–1905: “Gournia, Report of the American Exploration Society’s Excavations at Gournia, Crete in 1904”, Transactions of the Department of Archaeology, Free Museum of Science and Art, University of Pennsylvania 1.iii: 177–90. Buell, D.M. & McEnroe, J.C., 2015a: “Community Buildings/ Building Community at Gournia”, Paper presented at the conference From Static Data to Dynamic Processes: New Perspectives on Minoan Architecture and Urbanism, University of Toronto, January 5–7, 2015. Buell, D.M. & McEnroe, J.C., 2015b: “Where was the Protopalatial Palace at Gournia?” Paper presented at the Archaeological Institute of America 116th Annual Meeting, Chicago, Illinois, January 8–11, 2015. Buell, D.M. & McEnroe, J.C., 2016: “Gournia: A Case Study of Population Mobility in the Bronze Age Town and the Mirabello Region.” Paper presented at the 12th International Congress of Cretan Studies, Heraklion, Crete, September 22, 2016. Davaras, K., 1973: “Αρχαιότητες και Μνημεία Ανατολικής Κρήτης. Γουρνιά”, ArchDelt 28 (B’2): 588–89. Davaras, K., 1974: “Αρχαιότητες και Μνημεία Ανατολικής Κρήτης 1972. Γουρνιά”, Αμάλθεια 5: 49. Driessen, J. & Macdonald, C., 1997: The Troubled Island: Minoan Crete Before and After the Santorini Destruction. Aegaeum 17 (Liège). Driessen, J., 2010: “Spirit of Place. Minoan Houses as Major Actors”, in Pullen, D.J. (ed.), Political Economies of the Aegean Bronze Age (Oxford) 35–65. Fotou, V. & Brown, A., 2004: “Harriet Boyd Hawes (1871–1945)”, in Cohen, G.M. & Joukowsky, M.S. (edd.), Breaking Ground. Pioneering Women Archaeologists (Ann Arbor) 198–274. Fotou, V., 1993: New Light on Gournia: Unknown Documents of the Excavation at Gournia and Other Sites on the Isthmus of Ierapetra by Harriet Ann Boyd. Aegaeum 9 (Liège). Gomrée, T., 2013: La Voirie des villes minoennes en Crète orientale et á Cnossos (Minoen Moyen I—Minoen Récent I) (diss., Université Lumiére Lyon 2).

133

Hawes, H.B., Williams, B.E., Seager, R.B. & Hall, E.H., 1908: Gournia, Vasilike and Other Prehistoric Sites on the Isthmus of Hierapetra, Crete (Philadelphia). Herva, V.-P., 2005: “The Life of Buildings: Minoan Building Deposits in an Ecological Perspective”, OJA 24: 215–27. Herva, V.-P., 2006a: “Flower lovers, after all? Rethinking religion and human-environment relations in Minoan Crete”, WorldArch 38: 586–98. Herva, V.-P., 2006b: “Marvels of the system. Art, perception and engagement with the environment in Minoan Crete”, Archaeological Dialogues 13: 221–40. Hitchcock, L., 2007: “Naturalizing the cultural: architectonicized landscape as ideology in Minoan Crete”, in Westgate, R., Fisher, N. & Whitley, J. (edd.), Building Communities: House Settlement and Society in the Aegean and Beyond (BSA Studies 15) (London) 91–98. Knappett, C., 2009: “Scaling Up: from Household to State in Bronze Age Crete”, in Owen, S. & Preston, L. (edd.), Inside the City in the Greek Word: Studies in Urbanism from the Bronze Age to the Hellenistic Period (Oxford) 14–26. Legarra Herrero, B., 2014: Mortuary Behavior and Social Trajectories in Pre- and Protopalatial Crete (Philadelphia). McEnroe, J.C., 1990: “The Significance of Local Styles in Minoan Vernacular Architecture”, in Treuil, R. & Darcque, P. (edd.), L’Habitat Égéen préhistorique (Paris) 195–202. McEnroe, J.C., 2001: Pseira V: The Architecture of Pseira. Edited by P.P. Betancourt and C. Davaras (Philadelphia). Myers, J.W., Myers, E. & Cadogan, G., 1992: The Aerial Atlas of Ancient Crete (Berkeley, California). No Author Reported, 1906: “Annual report of the Provost to the Board of Trustees, University of Pennsylvania, for the Year ending August 31, 1905”, University Bulletins, Sixth series, no. 3, Part 4. Palyvou, C., 1990: “Architectural Design at Late Cycladic Akrotiri”, in Hardy, D., Doumas, C.G., Sakellarakis, J.A. & Warren, P.M. (edd.), Thera and the Aegean world III. Proceedings of the third international congress, Santorini, Greece, 3–9 September 1989 (London) 44–56. Rapoport, A., 1969: House Form and Culture (Englewood Cliffs). Remondino, F., Spera, M.G., Nocerino, E., Menna, F. & Nex, F., 2014: “State of the Art in High Density Image Matching”, The Photogrammetric Record 29.146: 144–66. Romano, D.G., 2003: “Minoan Surveyors and Town Planning at Gournia”, in Foster, K.P. & Laffineur, R. (edd.), Metron: Measuring the Aegean Bronze Aegean. Aegaeum 24 (Liège) 247–56. Rudofsky, B., 1965: Architecture without Architects: A Short Introduction to Non-Pedigreed Architecture (New York). Shaw, J.W., 2009: Minoan Architecture: Materials and Techniques. Studi di Archeologia Cretese 7 (Padua).

134 Smith, M.E., 2014: “Peasant Mobility, Local Migration, and Premodern Urbanization,” World Archaeology 46: 516–33. Soles, J.S., 1979: “The Early Gournia Town”, AJA 83: 149–67. Soles, J.S., 1991: “The Gournia Palace”, AJA 95: 17–78. Soles, J.S., 1992: The Prepalatial Cemeteries at Mochlos and Gournia and the House Tombs of Bronze Age Crete. Hesperia Suppl. 24 (Princeton). Thomas, J., 2001: “Archaeologies of Place and Landscape”, in Hodder, I. (ed.), Archaeological Theory Today (Cambridge) 165–86. Tilley, C., 1997: A Phenomenology of Landscape: Places, Paths and Monuments (Oxford). Vavouranakis, G., 2006: “Burials and the Landscapes of Gournia, Crete, in the Bronze Age”, in Robertson, E.C., Siebert, J.D., Fernandez, D.C. & Zender, M.U. (edd.), Space and Spatial Analysis (Calgary) 233–42. Verhoeven, G., 2011: “Taking Computer Vision Aloft— Archaeological Three-dimensional Reconstructions from Aerial Photographs with Photoscan”, Archaeological Prospection 18: 67–73.

BUELL ET AL. Walsh, V. & McDonald, W., 1986: “Greek Bronze Age Domestic Architecture: Toward a Typology of Stone Masonry”, JFA 13: 493–99. Watrous, L.V., 2012: “The Harbor Complex at Gournia”, AJA 116: 521–42. Watrous, L.V., Buell, D.M., McEnroe, J.C., Younger, J.G., Turner, L.A., Kunkel, B.S., Glowacki, K., Gallimore, S., Smith, A., Pantou, P.A., Chapi, A. & Margaritis, E., 2015: “Excavations at Gournia, 2010–2012”, Hesperia 84.3: 397–465. Watrous, L.V., Haggis, D., Novicki, K., Vogeikoff-Broga, N. & Schultz, M., 2012: An Archaeological Survey of the Gournia Landscape (Philadelphia). Weiberg, E., 2007: Thinking the Bronze Age. Life and Death in Early Helladic Greece. Boreas 29 (Uppsala). Yaeger, J. & Canuto, M., 2000: “Introducing an Archaeology of Communities”, in Canuto, M. & Yaeger, J. (edd.), The Archaeology of Communities: A New World Perspective (London) 1–15.

chapter 8

The Temple and Hestiatorion of the Sanctuary on Dhespótiko: Archaeology, Architecture, and Restoration Yannos Kourayos, Kornilia Daifa, Goulielmos Orestidis, Dimitrios Egglezos, Vasilis Papavasileiou, and Eleni-Eva Toumbakari 1

Introduction to the Sanctuary

Y. Kourayos, K. Daifa Dhespótiko is a small uninhabited island west of Antiparos located only a few miles away from Paros and Siphnos at a strategic position in the middle of the Aigaian Sea. The shallow channel between the small islet Tsimindíri and Dhespótiko was above water in antiquity, forming a low isthmus connecting these two islands. A Classical altar found in front of the main temple is inscribed ΕΣΤΙΑΣ ΙΣΘΜΙΑΣ, “of Isthmian Hestia”, clearly in reference to the sanctuary’s location on this isthmus (Kourayos et al. 2012, 122–23, citing earlier publications). It is most probable that Dhespótiko was also connected to Antiparos through Tsimindíri until the Roman era (Draganits 2009). The bay formed by the west coast of Antiparos, the east coast of Dhespótiko, and the islet of Tsimindíri at the north offers protection from the prevailing north winds—which must have been one of the main reasons for the establishment of a large sanctuary at the site of Mándra, which is located on a plateau at the northern entrance of the bay. Archaeological research at this site, which has been ongoing since 1997, has brought to light a large sanctuary

figure 8.1 Plan and aerial photo of the temenos G. Orestidis, Y. Kourayos

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004416659_010

dedicated to Apollon (Fig. 8.1). This new sanctuary, for which there is no known epigraphic or literary testimony, has significantly changed our understanding of the sacred landscape of the Kyklades, since its size, wealth, and spatial organisation are unparalleled by any other archaic Kykladic sanctuary besides Delos (Kourayos & Burns 2004–2005; Kourayos 2005; 2012; 2015: 88–93; Kourayos & Daifa 2017 [for a general overview]; Kourayos et al. 2012 [for a detailed report and bibliography]). The absence of any settlement on the island of Dhespótiko favours our hypothesis that the sanctuary on Dhespótiko was an extraordinary example of an extra-urban sanctuary founded by the polis of Paros. It seems that the Parians executed an ambitious plan in the second half of the 6th century BCE: the establishment of an extra-urban sanctuary of equal importance and reputation to Delos, which would be under Parian protection and influence (Kourayos & Daifa 2017). The earliest cult activity at Mándra dates to the Early Iron Age (Kourayos et al. 2017; Alexandridou, forthcoming a; b), but the acme of the sanctuary can be placed in the second half of the 6th century BCE, when a temenos protected by two gates was developed gradually. The most prominent edifice is Building A, which has been identified

136 as the main cult structure of the sanctuary (Fig. 8.1). It comprises the Temple, which is identified with the north part of the building (Rooms A1–A2), and the Hestiatorion (Fig. 8.2), which is identified with the south part of the building (Rooms A3, A4, A5). The main altar stood right at the centre of the temenos, directly across from the Temple (Kourayos et al. 2012: 148–50), and two marble bothroi serving cult purposes have also been found. The earliest of them was found in the prostoon of the Hestiatorion, which dates to 560–550 BCE, and the other right outside the temple. The latter dates to classical times and was dedicated to Hestia Isthmia. At the north side of the temenos was Building Δ, a temple-shaped edifice related to cult activities (Fig. 8.1) (Kourayos et al. 2012: 133–39; 2017: 345– 50). Building Ε and the so-called “Connecting Building” stood outside but almost in contact with the peribolos (Kourayos et al. 2012: 146–47). Only a few meters south of the temenos is the South Complex, which was built above an earlier structure from the late 8th–7th century BCE (Fig. 8.1). It included building units Θ and Ι (Kourayos et al. 2012: 157–61). Building Θ comprised the so-called Loutron, a room of great interest in terms of shape and function, and tentatively identified as a bath. Evidence indicates that the South Complex did

figure 8.2 Aerial photo of cult Building A before the restoration Y. Kourayos

KOURAYOS ET AL.

not serve a cultic purpose, but rather the everyday needs of the priests and visitors. Buildings M, N and Π stood east of the South Complex. Their excavation is still in progress. The area that extended outside the temenos down to the coast was occupied by at least six buildings—Β, Γ, Ζ, Η, Κ, and Λ—and a tower (Kourayos et al. 2012: 161–63). The sanctuary operated through the late Hellenistic period until the early 2nd century BCE, although it had lost its earlier popularity and wealth. During Late Antiquity the site was exclusively domestic. Finds indicate its use until the 6th century CE. After a long period of abandonment, it was re-inhabited during the late and post Byzantine period, until 1657 when it was looted by pirates and eventually deserted (Kourayos 2009: 135–37). The late antique complex reused the walls of the archaic and classical structures, while marble elements from the Temple and the Hestiatorion colonnades were used as building material. Despite the damage suffered by the archaic buildings, the recycling of their material resulted in its preservation on site. Thus, the architectural documentation of both the Temple and the Hestiatorion could be accomplished successfully by A. Ohnesorg and her collaborators, K. Papajanni and M. Lambertz (Kourayos et al. 2010; 2011; 2012; forthcoming a; b).

THE TEMPLE AND HESTIATORION OF THE SANCTUARY ON Dhespótiko

Dhespótiko has been declared an archaeological site, and no kind of construction is allowed on the island. The idyllic combination of a minimally disturbed historical and archaeological environment with the virgin nature of an uninhabited Kykladic island makes Dhespótiko an ideal case for the establishment of an archaeological park, in which the site of the Apollon sanctuary will hold a central position. Since both the excavation of the Temple and the Hestiatorion and the architectural documentation have been concluded, the Dhespótiko team was motivated to initiate the restoration of these monuments because of the potential technical, educational, and aesthetic benefits of such a project. 2

Architectural Study and Restoration of the Temple and Hestiatorion

G. Orestidis The following discussion presents the architectural study that culminated in a proposal for the partial restoration of the adjacent Temple and Hestiatorion. It presents new architectural documentation of the monument; solutions for specific issues relating to its form, function, and construction techniques; and the final decisions for the

figure 8.3 3D photogrammetric models of architectural members G. Orestidis

137

anastylosis. The project began with an architectural autopsy and virtual reconstruction (presented in this section), followed by engineering analyses of the stability of the planned anastylosis (detailed below in Section 3). The planned restoration shown in the following figures is currently being executed on the site. The definitive restoration study was preceded by the organisation and digital recording of the architectural members, which integrates and updates the previous documentation (the basic documentation of building A is presented in Kourayos et al. 2012). Ohnesorg had previously inventoried the architectural members during the architectural survey of the building remains through 2012 (some of the drawings from this survey have not yet been published). In order to obtain the evidence required to complete the final study, G. Orestidis and his team created new surveys of individual parts of the monument, when all the architectural members belonging to the building were identified, catalogued, and systematically documented. Each architectural member included in the study was modelled in 3D at high precision and photorealistic quality by means of photogrammetry. The geometry of the building components was thoroughly analysed by taking a series of sections through the 3D models (Fig. 8.3).

138

KOURAYOS ET AL.

figure 8.4 Proposed restoration: East façade of the temple and the Hestiatorion G. Orestidis

The new interpretations and observations made from the digital models and during fieldwork led to an update of the theoretical reconstruction—i.e., the proposal for the original appearance and construction of the building. The problems of form, function, and structure were each carefully assessed. The proposed restoration was based on the theoretical reconstruction, together with a general requirement that, in order to incorporate the maximum possible quantity of ancient material, the new stone would be limited to less than about 50% of the masonry in the final anastylosis (Fig. 8.4). The analytical documentation and exhaustive study of the collected ancient material has constituted the most important element of the project for, on the one hand, the confirmation and consolidation of the theoretical reconstruction and, on the other hand, the finalisation of the proposed restoration. The parts of the monument to be restored were intended to constitute a spatially balanced whole that would be both morphologically coherent and structurally stable. Therefore, as far as their documentation and proposed restoration, the Temple and the adjacent Hestiatorion were assessed as a single complex, because, although they constitute two separate buildings chronologically and functionally, they are still inextricably structurally linked. The Temple architecture was altered over time, with evidence for an earlier first phase below a second phase that in some places reused the earlier walls and in others substantially modified the building plan. The latter phase, whose remains are more substantial, is the focus of the anastylosis.

Reconstructions and Proposed Anastylosis of the Stylobate and South Anta The stylobate on the eastern sides of the two buildings needed to be supplemented with displaced ancient blocks and new material in order to provide a stable foundation for columns that would eventually be returned to their original settings (see Section 2.4, below). At the lowest level, a row of marble slabs seated on bedrock rise partly above the original ground level and support the course of stylobate blocks (Figs. 8.4–8.5). The relatively wellpreserved marble foundations in the northern part of the facade revealed that the top level of the stylobate of the Temple was continuous with that of the Hestiatorion, from which the height of the stylobate blocks could be determined as ca. 0.25 m. The restoration of the stylobate of the Temple primarily involved the filling of its foundations through reintegrating ancient material to its original position. Four displaced marble blocks identifiable as originating from the foundations could be included in the reconstruction because they joined one another, thus revealing their original ordering in the course. From the additional observation that the height of the stylobate foundations blocks gradually lowers towards the middle of the building, these four plinths could be located very close to their original position, according to their heights. This peculiarity of the construction is attributable to static reasons, since the lower portions of the marble blocks in the foundation would not be visible in the finished building after the final infilling of soil in front of the stylobate. 2.1

THE TEMPLE AND HESTIATORION OF THE SANCTUARY ON Dhespótiko

139

figure 8.5 Proposed restoration: Plan G. Orestidis

In order to complete the stylobate of the Hestiatorion, four large slabs made of gneiss would also need to be replaced. Their heights follow the same pattern as the in situ foundations (Figs. 8.4–8.5). Their position and jointing was determined by orienting the only vertical face dressed to a plane along the southern part of the stylobate, which is the system observed from all the in situ examples of the type. Between the colonnades, the south anta of the Temple, which also serves as a party wall between the Temple and Hestiatorion, may be restored up to its full height along the facade. The ancient stone blocks were integrated with a relatively high proportion of new material (Fig. 8.6). The decision was justified for several reasons. The first is historical, in that the connection of the two buildings by a common wall constitutes an important architectural trait of the two monuments as an integrated whole (Fig. 8.7). Likewise, by completing the form of the anastylosis, such a reconstruction may be justified on aesthetic as well as didactic grounds. Finally, the pilaster adds significant rigidity to the colonnade, whose reconstruction is discussed below, improving its stability (see Section 3, below). In the final study, upon the request of the Directorate of Restoration of Ancient Monuments and Dr. Toumbakari,

the colonnade of the temple was also connected to the southern pilaster by an architrave, a choice upon which the entire scientific team agreed. M. Korres, who had already proposed this restoration since the preliminary study, confirmed the solution. During the reconstruction of the southern anta, it was determined that the two different construction phases of the Temple should be restored distinctly. To the east, corresponding to the second phase, the masonry is isodomic, with large ashlar blocks dressed to a smooth plane (Fig. 8.6). The western portion of the anta wall preserves some of the first phase of construction, whose pseudoisodomic masonry is composed of smaller blocks. On the north face, corresponding to the interior of the Temple, the ashlars have drafted margins, and the interior is dressed using a fine point, while on the southern exterior face the blocks are less regular and slightly curved. The second phase of the anta consists of thirteen courses, the uppermost incorporating the anta capital. Lacking any identifiable fragments, the capital is instead modelled after the example of the temple of Delian Apollon in Paros, which, in addition to its chronological and typological proximity, shares many architectural features in common with the temple of Dhespótiko (Schuller 1991: 88–119).

140

KOURAYOS ET AL.

figure 8.6 Section through the temple and north elevation of south pilaster: (below) original state and (above) proposed restoration G. Orestidis

2.2 Restoring the Walls, Threshold Blocks, and Floors As for the earlier architecture, another wall located just inside and parallel to the restored stylobate—corresponding chronologically to the smaller pseudo-isodomic masonry on the interior of the south anta—belongs to the first temple (Figs. 8.5–8.6). The earlier wall is built from roughly carved slabs set in a double-faced construction that was partly disrupted during the construction of the later stylobate. Additional distinctive blocks recovered from the area were stored in a separate stone pile due to their resemblance to the in situ slabs of this wall. The author believes it was the foundation for a stylobate and colonnade in the first phase, primarily because of its similarity to and alignment with the stylobate of the Hestiatorion to the south, and also due to its connection with the earlier anta walls and the floor of the porch.

The floor level, according to the architectural restoration study, appears not to have been raised during the second phase but rather maintained at its initial level substantially below the elevation of the stylobate (Fig. 8.6). For this reason, the stylobate in the first phase of the Temple was most likely located at the same elevation as that of the Hestiatorion. The architectural remains also point to such a conclusion. It has been possible to restore the elevations of the floor in not only the porch, but also in Rooms A1 and A2, which follow the natural topography down to even lower levels. The general objective for the anastylosis has been to restore the flooring in each room at the foundation level of the ancient paving. The blocks in the east wall of both Rooms A1 and A2 were deliberately cut short for the widening of their doors

THE TEMPLE AND HESTIATORION OF THE SANCTUARY ON Dhespótiko

141

figure 8.7 Section through the Hestiatorion and south elevation of the south pilaster of the temple: (below) original state and (above) proposed restoration G. Orestidis

during the second phase (Fig. 8.5). The ancient threshold block leading into Room A2 from this phase has survived and will be raised to its original level, which can be restored by reference to the rough finish along its lower side faces beginning at the level of the original paving. The threshold of Room A1 may be reconstructed from a new block of the same dimensions, placed in the surviving opening at the same level and in alignment with the block for Room A2. Likewise, it has also been recommended to fabricate and install new threshold blocks for Rooms A3 and A4 in the Hestiatorion, integrating the large fragments which survived from one of them, while the threshold for Room A5 is largely intact. The vertical slabs for the door jambs will also be restored from a combination of original and new material (Figs. 8.4–8.5).

As for the rest of the building, smaller or larger supplements to the perimeter and interior walls are proposed for the protection and stabilisation of the preserved masonry, and also to encase the floors inside the rooms. A largerscale reconstruction combining ancient and new blocks is also proposed for the northern wall of room A1 (Fig. 8.5). Beyond, the ancient external ground level may be accurately restored around the northwestern corner of the Temple. Likewise, the areas around the monument will be returned wherever possible to the ancient ground level, following the installation of a drainage system. 2.3 Restoring the Colonnade and Entablature The recent study has assigned 29 fragments of column drums to the colonnade of the Temple; 15 of these are new

142 identifications. The original elevation for 21 of these fragments may be established due to the taper of the columns, whose diameter can be measured with relatively high precision. Two drums were found to belong to the same column. The high precision of the spatial recording of the drums indicates the taper of the columns, and their total height could also be estimated based on the upper and lower diameters measured from fragments from the top and bottom of the shaft. The comprehensive study also determined that the column shafts were originally constructed from five column drums of variable height. The remains of the Temple capitals were representative enough to restore them accurately, but because of their fragmentary state they could not be reintegrated into the actual anastylosis. Instead three restored capitals must be manufactured from entirely new material. The restoration of the entablature of the Temple includes the setting of three architraves on three of the fully restored columns as well as the south anta wall adjoining the Hestiatorion. This reconstruction integrates seven fragments identifiable from three different ancient architrave blocks, as well as three antithemata— rear architraves or backers—created from new marble (Figs. 8.4–8.5). Thereafter, a separate course for the taenia supplements ancient fragments with new members, above which can be installed the original fragments of five triglyphs, and a combination of ancient and new metopes. The reconstruction includes an antithema wall behind the triglyphs and metopes that support three ancient geison blocks (Fig. 8.6). This backer wall is constructed out of slightly curved ancient stone slabs identifiable by their size and treatment. The proposed width of the entablature could be verified after a thorough study of the back side of the architraves and triglyphs, as well as of the joint faces of the cornice blocks. At the Hestiatorion 15 fragments of column drums have been assigned to the colonnade, of which ten have been newly identified. Eight of these could be reintegrated at the correct elevation within the original shafts, which are comprised of two drums. Furthermore, three ancient capitals were determined to be stable enough to be reintegrated into the anastylosis, supporting three largely intact ancient architraves and their three antithemata (Figs. 8.4–8.5). Two of the latter are new, while one backer identified during the cataloguing and restudy is ancient, though restored by incorporating a new marble section. A raised panel on top of the abacus of the capitals was oriented perpendicular to the architraves, in order to accommodate the depth of the architrave blocks together with the narrower, ca. 0.20-m-thick antithemata (Fig. 8.7). These backers constitute a new element in the restored

KOURAYOS ET AL.

entablature of the Hestiatorion which solve the problem of how to support the relatively deep geison blocks that can be assigned to the building with certainty. The restoration has been proposed only for the parts of the monument that are well-represented by surviving fragments and have been precisely documented. Accordingly, despite the discovery of two corner akroteria and possibly one central akroterion, as well as a triangular block from the wall of a pediment, the restoration does not continue above the horizontal cornice of the Temple. The nine catalogued geison blocks, even including a recently identified corner geison, do not preserve enough of their upper surfaces to determine the existence and exact position of a stone pediment. 3

Numerical Modelling of the Dynamic Response of the Restored Masonry: Theoretical Basis

D. Egglezos, V. Papavasileiou, E. Toumbakari In support of the anastylosis project, a numerical model was subsequently developed in order to examine the response of the structure to horizontal forces, primarily earthquakes and wind, and the ideal locations and dimensions for clamps between the block joints (Orestidis 2016; Egglezos & Toumbakari 2016; Papavasileiou 2016). The study demonstrates the beneficial effect of the restored anta wall between the Temple and the Hestiatorion proposed by the Directorate for Restoration of Ancient Monuments. The structural analysis of a monument, building, engineering project (e.g., dam or tunnel), or even an archaeological ruin with substantial standing elements aims at studying its reaction under expected forces from earthquakes, interaction with soil, water pressure, and the like. Such an analysis may point to the need for additional supportive elements—like walls, beams, columns, or other strengthening interventions—or for an entirely different structure. In the case of a standing monument, an analysis aims to assess the stability under various external forces, the conditions under which it may be badly damaged or even collapse, and the interventions necessary to prevent such an outcome. Archaeological research supplemented by literary sources concerning the behaviour of the original monument is not only appropriate, but also provides a valuable perspective on, among other things, the causes of its gradual collapse. For a thorough structural analysis, which should go beyond empirical approaches limited to the simple consolidation of the individual building blocks, the use of reliable models is necessary.

THE TEMPLE AND HESTIATORION OF THE SANCTUARY ON Dhespótiko

Engineering analysis has always been carried out using mathematical models that, thanks to the advent of increasingly powerful computational techniques, have gradually become capable of simulating complex configurations that would have been impossible in the past. Archaeological ruins, in particular those built from individual blocks without mortar in the joints, may now be studied under the forces of earthquakes. It is beyond the scope of this chapter to present the relevant literature on this field in depth. Very briefly, however, it should be noted that: During the seismic response of discontinuous block assemblages, the deformation and failure is dominated by the movement between individual blocks. The resulting continuum models, based primarily on the finite element method, may not be appropriate numerical tools for identifying key features of the response or efficiently handling significant sliding along joints. Instead, discontinuous modelling via the discrete element method tends to function better in that role (Dasiou, Psycharis & Vayas 2008; Psycharis, Drougas & Dasiou 2011). The discrete element method, initiated by the pioneering work of P.A. Cundall at the University of Minnesota (Cundall 1988), has since been widely used in geotechnical and structural engineering (Itasca Consulting Group 1998). Research using this method and focusing on classical monuments was mainly initiated at the Faculty of Civil Engineering of the National Technical University of Athens. The results of various numerical and experimental investigations into the main features of the behaviour of archaeological ruins can be summarised as follows (Psycharis, Papastamatiou & Alexandris 2000; Papantonopoulos et al. 2002; Psycharis 2004): (a) Ancient structures do not possess natural modes of vibration like those of monolithic structures in the classical sense, and the period of their free vibrations is dependent on the amplitude of excitation. In other words, during a strong earthquake, the structural response shifts among different “modes” of vibration, each one being governed by a different set of equations of motion. As a result, the response is highly nonlinear. An example of nonlinearity is a column that may collapse under certain earthquake movements, yet remain stable when the level of excitation is magnified by a factor greater than one. (b) The response is very sensitive to even trivial changes in the parameters of the structural system or the

143

excitation. Another effect of the response sensitivity is the significant out-of-plane displacements observed even during purely planar excitations. (c) The vulnerability of ancient monuments to earthquakes depends on the size of the structure and is much affected by the predominant period of the ground motion. Low-frequency earthquakes are much more harmful than high-frequency ones. In the first case, the response is characterised by intense rocking, while in the latter, rocking is usually restricted to small values, although significant sliding of the drums will occur especially in the upper parts of the structure. Assuming the same ratio of dimensions, bulkier structures are much more stable than smaller ones. (d) Ancient masonry is not generally vulnerable to normal earthquake actions as long as it is intact (Psycharis 2004; 2007). Collapse will occur much more easily if there are imperfections, such as damaged blocks that have lost material (especially on the lower and upper surfaces), dislocated drums, or rotated foundations—all of which are common in ancient monuments. These features define the theoretical framework applied to the study of the dynamic response of the restoration proposed at Dhespótiko, which among other things concerns the dimensions of the connections between adjacent blocks and their internal reinforcements. Evaluation of the Proposed Anastylosis on the Basis of Numerical Modelling The archaeological (Kourayos 2012) and architectural documentation (Kourayos et al., forthcoming a; Orestidis 2016) provided the fundamental understanding of the Temple and Hestiatorion. The interpretation of the archaeological remains in combination with relevant architectural comparanda resulted in the reconstruction drawings of the original monuments detailed in the previous sections. A reconstruction project, of course, starts from these parameters, but the variety of options for an anastylosis is also highly dependent on the number and position of the surviving building blocks. The initial architectural proposal for the anastylosis on Dhespótiko included the reconstruction of parts of the two free-standing colonnades (Figs. 8.4–8.5). However, previous research conducted by the Directorate for the Restoration of Ancient Monuments has demonstrated that such solutions might not survive severe earthquakes, whereas the introduction of a rigid element, such as a wall, could be very beneficial to the stability of the system (Psycharis & Toumbakari 2009; 2011; Toumbakari, 3.1

144 Psycharis & Schmid 2014). Taking this into consideration, the Directorate submitted to the Central Archaeological Council two proposed measures, both of which were approved: first, to accept the construction of part of the intermediate anta wall, even if this would mean the addition of a high proportion of new material (Figs. 8.6–8.7); and, second, to conduct a structural analysis of both options in order to support the selection of the final anastylosis proposal. The following sections summarise the preparations for numerical modelling of the two proposals approved by the council. Geological and Geotechnical Background, and Historical Records of Earthquakes The island of Dhespótiko belongs to the Attic-Kykladic geological zone and is mainly composed of metamorphic rocks—namely gneiss, marble, shale, and amphibolite (Draganits 2009). The archaeological site, including the monumental complex, is founded on gneiss bedrock. The monument itself is constructed either directly on this gneiss or else on a well compacted sand-gravel layer, whose thickness is sufficiently limited to be considered negligible in the model, which assumes instead that the whole structure is founded directly on bedrock. This geotechnical information provides the boundary conditions for the model (i.e., category A non-deformable substrate), which is of importance for the earthquake analysis as well. Relevant seismic inputs—i.e., earthquake records— are paramount to the validity of the model. Two sets of data were considered in order to provide criteria for the selection of the earthquake records incorporated into the model: (a) Antiseismic codes, whose application is compulsory in all seismic analyses (Hellenic Antiseismic CodeEAK 2000 and Eurocode 8 (EC-8); both codes are gradually converging). According to the codes, the monument is located in the low-seismicity Zone I, for which the designated effective acceleration (agR) is 0.16 g (where “g” expresses the gravity acceleration). In order to study seismic events with different return periods (and therefore different intensities) applying both Hellenic and European codes (EAK 2000, EC-8) coincides with the following provisions regarding the value of the peak ground acceleration (PGA). For a return period of 72 years, the PGA is 0.12 g; for 475 years, it is 0.24 g; and for 2,500 years, 0.41 g (the return period expresses how often an event is expected to occur; the stronger the earthquake, the more infrequent it is). 3.2

KOURAYOS ET AL.

(b) Historic seismic data: the most important seismic events of the Dhespótiko area have already been compiled (Papazachos & Papazachou 2002), and the corresponding ground acceleration was estimated on the basis of empirical functions (Skarlatoudis et al. 2003) that correlate the distribution of seismic acceleration with the distance R from the epicentre and the magnitude M of the earthquake, the type of seismic fault, and the prevailing geotechnical conditions over a period of 2,500 years. The application with variables set such that 6.0 < Μ < 7.5, 24 km < R < 90 km, a time span between 550 BCE–1956 CE, a normal fault, and soil Category A results in a PGA of 0.14 g, which corresponds to effective acceleration (aeff) of 0.10 g. This value is significantly lower than the one provided by the codes. On this basis, research was carried out in order to compile the relevant historical records of earthquakes with the following criteria: similar distance to the epicentre and magnitude (provided by the historic seismic data), peak ground acceleration of around 0.24 g over the return period of 475 years stipulated by the antiseismic codes, and similar geotechnical conditions. The main analyses were carried out with a 475 year return period; however, a return period of 2,475 years was analysed as well. Finally, three earthquake records from Japan were selected for application to the model (Egglezos & Toumbakari 2016). Those records have been modified in order to resemble typical earthquake events, as required by the European antiseismic regulations (EC-8). The “typical” events characterise earthquakes by three categories based on the return period: “frequent”, with a return period of 72 years; “rare”, of 475 years; and “extremely rare”, of 2475 years. 3.3 Parameters and Methodology in the Analysis The geometry of the two proposed models with and without the intermediate anta wall originated with the architectural study. The geological substrate was modelled as a continuous base with elastic properties. The stylobate, toichobate, and foundations were modelled as monolithic elements because they are expected to follow the motion of the ground. The column drums, architrave blocks, and frieze blocks were modelled as distinct elements, with specific material properties at the joints interfaces. Moreover, each block was assigned elastic properties so as to be able to study the developing internal stresses. Finally, the intermediate wall was considered monolithic, and accordingly provisions were made—mainly the inclusion of clamps connecting the block joints—to ensure that the construction would effectively behave as a rigid body.

THE TEMPLE AND HESTIATORION OF THE SANCTUARY ON Dhespótiko

145

figure 8.8 Numerical model of the freestanding colonnades coloured by the results (permanent displacements of the blocks) from an extreme earthquake D. Egglezos, V. Papavasileiou, E. Toumbakari

figure 8.9 Numerical model of the colonnades with the intermediate wall coloured by the results (permanent displacements of the blocks) from an extreme earthquake, showing a safe situation with only minor local displacements D. Egglezos, V. Papavasileiou, E. Toumbakari

A total of eight variations in the two models were studied, differing by the presence or omission of the intermediate wall (Figs. 8.8–8.9), of connectors at certain locations, and changes in the material properties. To increase the accuracy of the model, different characteristics were assigned to the gneiss substrate, to the gneiss and marble blocks, to the building foundations, and to the joints between the blocks. The properties were derived from the generalised Hoek-Brown criterion, in which the macroscopic fracture is also considered (Hoek, Carranza-Torres & Corkum 2002; Hoek & Diederichs 2006). Finally, the geostatic constants for the mechanical description of the joint behaviour were based on widely accepted calculation methods (Barton 1972). The study of the structural behaviour of the proposed anastylosis comprised: (a) static analysis of the structure

under its own weight; (b) static analyses against wind pressure, to which two different values have been assigned, following Eurocode-1; and (c) dynamic, time-history analysis against earthquakes. For each earthquake analysis, targets were set in terms of acceptable damage, including partial collapse but excluding total collapse. The results of each analysis were then compared to the corresponding targets. In all, 32 dynamic analyses were carried out in order to define the conditions that meet the predetermined targets for the stability of the monumental complex. Towards this objective, the choice of the location of the connectors between the blocks is crucial. Bearing in mind the non-linearity of the response, the requirement that each block be connected to its neighbours must be cautiously considered. Indeed, more clamps might under certain circumstances lead to greater block collapses than

146

figure 8.10

KOURAYOS ET AL.

Design drawings of the internal reinforcement for drum NK1Σ1 and architrave NE1 D. Egglezos, V. Papavasileiou, E. Toumbakari

might be expected with fewer connections, because they modify the vibration mode of the structure. This could provoke an extensive collapse due to the monolithic behaviour of the connected blocks. In the end, it was decided to connect the architraves longitudinally with Π-shaped clamps. The architraves adjacent to the intermediate wall were connected to the latter with dowels (Fig. 8.10). Connections were designed between the triglyphs and the cornice as well. Furthermore, the fragmentary ancient blocks are assumed to be supplemented with new marble and reinforced so as to render the new block a monolithic whole. The results of an earthquake intensity (Μ) of 7.1 with a repeating period of 475 years was invoked for the dynamic analyses. For the most crucial blocks of the anastylosis, further analysis with the use of the finite element method was carried out in order to ensure a more thorough understanding of the forces developing at the joints and thus to double-check their proposed dimensions. The connectors and reinforcement are manufactured from pure Grade 2 commercial Titanium, with diameters ranging from 8–16 mm and a length between 0.5–2.0 m (Fig. 8.10). 3.4 Conclusions of the Structural Analysis Numerical modelling is an important tool for the study of the structural behaviour of ancient buildings or

their ruins, provided that the pertinent architectural, geological-geotechnical, earthquake, and materials data are available. In the case of the anastylosis of the Dhespótiko complex, numerical modelling—or more precisely, the distinct element method—was used to assess its stability against the action of wind and earthquakes. On the basis of the results, the anastylosis project was determined to meet safety criteria when it was demonstrated that the fundamental condition for stability was the construction of the anta wall between the Temple and the Hestiatorion, which drastically changes the response of the colonnades (cf. Figs. 8.8–8.9). Furthermore, the study demonstrated the need for proper connection between the wall blocks in order to ensure stiffness, as well as between the architraves and the anta wall. Finally, the numerical analysis of the structure provided data for the design of the reinforcement of the blocks. As far as safety is concerned, the lifetime for the anastylosis could be calculated within a 10% probability as (a) 130 years before collapse due to the action wind according to EC-1; (b) 250 years before partial collapse caused by a “rare” earthquake according to EC-8; (c) at least 50 years before any limited, although not total, damage by the same “rare” earthquake; and (d) a minimum of 10 years with no likely damage from a “frequent” earthquake. It is thus advisable to proceed with a structural assessment after 50 years to document

147

THE TEMPLE AND HESTIATORION OF THE SANCTUARY ON Dhespótiko

the state of preservation and assess its lifetime according to the antiseismic codes and the state of the art in earthquake engineering then prevailing. In general, the contemporary theories and tools of structural and geotechnical engineering allow for the study of complex systems such as ancient masonry constructed with stacked, unbonded blocks. It would thus be possible to study not only the ruin for the purpose of the anastylosis project, but also the original building or significant parts of it. As far as concerns the stabilising of the architectural remains, numerical modelling permits the realistic assessment of its behaviour in terms of calculated forces and displacements, by which the design of the internal reinforcements and connections may be optimised, resulting in fewer interventions on the ancient material. In this context, numerical modelling is an important tool for making decisions concerning the safety and extent of an anastylosis. As far as the study of the original monument, numerical modelling carried out with fully justified parameters and the appropriate engineering judgement could provide valuable information on a variety of interdisciplinary research questions, such as the sequence of construction, or the correlation between the long-term vulnerability of the building (including soil-building-foundation interaction) to failure and the actual failures observed from the remains. A systematic study of monuments whose original appearance can be restored with high confidence and with securely-dated construction phases might shed light on the developments over time, if any, of the principles that guided the ancient builders. 4

General Conclusions

The Temple and Hestiatorion at Dhespótiko not only present unique characteristics in terms of architectural layout and construction, but also in terms of function, since they constitute the only case currently known in the Kyklades of an archaic Hestiatorion that is adjacent to an archaic temple. Therefore, their restoration will be deeply instructive and educational both for scholars of Greek architecture and the general public. Moreover, the anastylosis project is based rationally not only on the whole of the available architectural and archaeological remains, but also on an advanced knowledge of the advantages and limitations in terms of safety of each of the proposed options for a reconstruction. The completion of the excavation and restoration project combined with the highly protected status of Dhespótiko will support the management of an

exceptional archaeological park that promises to revive the study of the culture of the ancient Kyklades. List of References Alexandridou, A., (forthcoming a): “Geometric Despotiko: on the borderline between sacred and profane”, in Tsingarida, A. & Lemos, I. (edd.), Ritual Practices in Early Greece. The Conference Proceedings, Etudes d’Archéologie (Brussels). Alexandridou, A., (forthcoming b): “Γεωμετρικό Δεσποτικό: οι απαρχές ενός ιερού ή μήπως όχι;”, in Πρακτικάτου Γ’ Κυκλαδολογικού Συνεδρίου, Σύρος 25–29 Μαϊου 2016 (Syros). Barton, N.R., 1972: “A Model Study of Rock-joint Deformation”, International Journal of Rock Mechanics and Mining Science 9: 579–602. Cundall, P.A., 1988: “Formulation of a Three-dimensional Distinct Element Model. Part I: a Scheme to Detect and Represent Contacts in a System Composed of Many Polyhedral Blocks”, International Journal of Rock Mechanics and Mining Sciences 25: 107–16. Dasiou, M.-E., Psycharis, I.N. & Vayas, I., 2008: “Numerical Analysis of the Seismic Behavior of Columns and Sub-assemblages of Ancient Temples”, in Third National Conference on Earthquake Engineering & Technical Seismology, Athens. Paper no. 1832: http://library.tee.gr/digital/m2368/m2368_dasiou.pdf. Draganits, E., 2009: “The Archaic Sanctuary on Despotiko Island (Cyclades): Geological Outline and Lithological Characterization of the Building Stones, with Their Possible Provenance”, Austrian Journal of Earth Sciences 102: 91–101. Egglezos, D. & Toumbakari, E.-E., 2016: “Study of the Dynamic Response of the Despotiko Monumental Complex” (Unpublished Report submitted to the Hellenic Ministry of Culture & Sports, Athens). Hoek, E. & Diederichs, M.S., 2006: “Empirical Estimation of Rock Mass Modulus”, International Journal of Rock Mechanics and Mining Sciences 43: 203–15. Hoek, E., Carranza-Torres, C. & Corkum, B., 2002: “Hoek-Brown criterion—2002 edition”, in Hammah, R. & the Tunnelling Association of Canada (edd.), Mining and Tunnelling Innovation and Opportunity. Proceedings of the 5th North American Rock Mechanics Symposium and the 17th Tunnelling Association of Canada Conference (Toronto) NARMS-TAC vol. 1: 267–73. Itasca Consulting Group, Inc., 1998: 3-dimensional Distinct Element Code (3DEC), Theory and Background. Minneapolis: ITASCA Consulting Group: http://www.itascacg.com/ software/3dec. Kourayos Y., A. Alexandridou, Papajanni, K. & Draganitis, E., 2017: “Ritual Dining at the sanctuary of Apollo on Despotiko:

148 the evidence from Building ∆”, in Mazarakis Ainian, A. (ed.), Les sanctuaires archaïques des Cyclades. Recherches récentes (Rennes) 345–66. Kourayos, Y. & Burns, B., 2004–2005: “Exploration of the Archaic Sanctuary at Mandra on Despotiko”, BCH 128–129: 133–74. Kourayos, Y. & Daifa, K., 2017: “Politics, territory, and religion in the Cyclades during the Archaic period. The case of Paros and the sanctuary on Despotiko”, in Mazarakis Ainian, A. (ed.), Les sanctuaires archaïques des Cyclades. Recherches récentes (Rennes) 307–26. Kourayos, Y., 2005: “Despotiko Mandra. A Sanctuary Dedicated to Apollo”, in Yeroulanou, M. & Stamatopoulou, M. (edd.), Architecture and Archaeology in the Cyclades: Papers in Honour of J.J. Coulton. BAR-IS 1455 (Oxford) 105–33. Kourayos, Y., Daifa, K., Ohnesorg, A. & Papajanni, K., 2010: “Δεσποτικό. Σημαντικά πορίσματα για την αρχιτεκτονική του αρχαικού ιερού και τα αποτελέσματα των φετινών ερευνών”, Παριανά 119: 472–88. Kourayos, Y., Daifa, K., Ohnesorg, A. & Papajanni, K., 2012: “The sanctuary of Despotiko in the Cyclades. Excavations 2001– 2012”, AA 2012.2: 93–174. Kourayos, Y., Lambertz, M., Ohnesorg, A. & Papajanni, K., 2011: “Apollon unterm Ziegenstall. Das neugefundene Heiligtum auf Despotiko”, AntW 2011: 47–56. Kourayos, Y., Ohnesorg, A., Papajanni, K. & Lambertz, M., (forthcoming a): “The Architecture of Building A at the Despotiko Sanctuary close to Antiparos”, AAA. Kourayos, Y., Sutton, R.F. & Daifa, K., 2018: “Miltiades on Paros. New evidence from Despotiko”, in Angliker, E. & Tully, J. (edd.), Cycladic Archaeology: New Approaches and Discoveries (Oxford) 113–34. Kourayos, Y. 2009: Δεσποτικό. Το ιερό του Απόλλωνα (Athens). Kourayos, Y. 2012. Despotiko: The Sanctuary of Apollo (Athens). Kourayos, Y. 2015. Πάρος, Αντίπαρος, Δεσποτικό. Από την προϊστορία στα νεώτερα χρόνια (Paros). Orestidis, G., 2016: “Architectural Study for the Restoration of the Main Temple and the Hestiatorion (Building A) of the Archaic Sanctuary of Despotiko, Antiparos” (Unpublished Report submitted to the Hellenic Ministry of Culture & Sports, Athens). Papantonopoulos, C., Psycharis, I.N., Papastamatiou, D.Y., Lemos, J.V. & Mouzakis, H., 2002: “Numerical Prediction of the Earthquake Response of Classical Columns Using the Distinct Element Method”, Earthquake Engineering and Structural Dynamics 31: 1699–717. Papavasileiou, V., 2016: “Study of the Dynamic Response of the Despotiko Monumental Complex—Design of the Connectors and Reinforcement of Ancient and New Building

KOURAYOS ET AL. Blocks” (Unpublished Report submitted to the Hellenic Ministry of Culture & Sports, Athens). Papazachos, V. & Papazachou, K., 2002: The Earthquakes in Greece (Thessaloniki). Psycharis I.N., Drougas, A.E. & Dasiou, M.-E., 2011: “Seismic Behaviour of the Walls of the Parthenon—A Numerical Study”, in Papadrakakis, M., Fragiadakis, M. & Lagaors, N.D. (edd.), Computational Methods in Earthquake Engineering (Dordrecht) 265–83. Psycharis, I.N. & Toumbakari, E.-E., 2009: “Aphrodite Temple in Amathus, Cyprus: Seismic Response of the Eastern Colonnade and Design of Structural Interventions” (Joint Hellenic Ministry of Culture and NTUA Report submitted to the French Archaeological School of Athens). Psycharis, I.N. & Toumbakari, E.-E., 2011: “Parametric Investigation of the Seismic Response of the Columns of the Aphrodite Temple in Amathus, Cyprus”, in the Macedonian Association for Earthquake Engineering (ed.), Proceedings of the 14th European Conference on Earthquake Engineering 2010. Ohrid, Republic of Macedonia, 30 August–3 September 2010 (Red Hook, NY) 2142–49. Psycharis, I.N., 2004: “Dynamic Response of a Part of the Temple of Olympios Zeus (Olympieion) in Athens, Greece to Harmonic and Earthquake Excitations”, in Flesch, H.I.R. & Krommer, M. (edd.), Proceedings of the Third European Conference on Structural Control (3ECSC) (Vienna). Psycharis, I.N., 2007: “A Probe into the Seismic History of Athens, Greece from the Current State of a Classical Monument”, Earthquake Spectra 23: 393–415. Psycharis, I.N., Papastamatiou, D.Y. & Alexandris, A., 2000: “Parametric Investigation of the Stability of Classical Columns under Harmonic and Earthquake Excitations”, Earthquake Engineering and Structural Dynamics 29: 1093–1109. Schuller, Μ., 1991: Der Artemistempel im Delion auf Paros. Denkmäler antiker Architektur 18.1 (Berlin). Skarlatoudis, A.A., Papazachos, C.B., Margaris, B.N., Theodulidis, N., Papaioannou, C., Kalogeras, I., Scordili, E.M. & Karakostas, V., 2003: “Empirical Peak Ground-Motion Predictive Relations for Shallow Earthquakes in Greece”, Bulletin of the Seismological Society of America 93: 2591–603. Toumbakari, E.-E., Psycharis, I. & Schmid, M., 2014: “Aphrodite Temple in Amathus, Cyprus: Seismic Response of the Eastern Colonnade, Design of Structural Interventions and Pilot Worksite Applications as Tools for Earthquake Risk Mitigation”, in Korka, E. (ed.), The Future of Protection Management for Archaeological Heritage in Times of Economic Crisis (Newcastle upon Tyne) 180–91.

part 3 Architecture, Cultural History, and Communication

∵

chapter 9

Building Change: Domestic Architecture and Identity during the Bronze Age to Iron Age Transition Kyle A. Jazwa 1 Introduction In this paper, I introduce a new methodology for reconstructing past social identities with an examination of domestic architecture.1 By evaluating the statistical correspondence of individual human practices, or behaviours, related to the construction and use of domestic architecture, I identify groups of individuals who likely expressed a shared identity in their daily living environment. The individuals (i.e., inhabitants) who acted most similarly in this context are members of a network of interaction through which regular and expected actions became embodied and habituated (Bentley 1987; Jones 1997: 90–91). For the application of this methodology, the identification and analysis of these practices largely follow spatial syntax analysis and studies of the chaîne opératoire (Hillier & Hanson 1984; Lemmonier 1993; Gosselain 2000; Roux 2003; Nagle 2015). However, the approach is unique because it evaluates the overall correspondence of these qualities among contemporarily built and occupied structures to reveal broader patterns of interaction and identity. This paper primarily serves as an introduction to this new methodological approach. After providing a brief description of the theoretical background, I detail the methodology and its potential for understanding past identities. For a case study, I apply the methodology to the domestic architecture of mainland Greece that was built during the Late Helladic IIIB (LH IIIB, ca. 1300–1190 BCE), Late Helladic IIIC/Submykenaian (LH IIIC, ca. 1190–1015 BCE), and Protogeometric (PG, ca. 1015–900 BCE) periods—a time of significant social, political, and economic transition in mainland Greece. Contemporary

1  This paper presents a condensed discussion of the theoretical framework and methodology in my Ph.D. dissertation: Jazwa 2016. I would like to thank Kimberley van den Berg for reading earlier versions of this paper, as well as Daniel Pullen, Christopher Pfaff, Andrea de Giorgi, and Jack Freiberg for their helpful comments on the research and analysis. I am also grateful for the support from Florida State University, its Department of Classics, and the American School of Classical Studies in Athens that made the research and data collection possible.

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004416659_011

to this, many established politico-economic structures in mainland Greece collapsed, and the Late Bronze Age (LBA, ca. 1600–1000 BCE) gave way to the Early Iron Age (EIA, ca. 1000–700 BCE). Consequently, an examination of the social identities that existed during this period can augment our understanding of society, the processes of collapse, and the potential agents of change. With this case study, I will demonstrate that established mainland identities were maintained from the LH IIIB to LH IIIC periods despite the collapse of the palatial administration. The behavioural correspondence analysis reveals only a significant disruption in PG. At that time, there were ostensible changes to social identity that were accompanied by fundamentally different building traditions, suggesting a new modus vivendi by several groups in mainland Greece. I argue that the LH IIIC period of population movement within, to, and from Greece had detrimental effects on established networks of interaction and the established mainland identities. 2

Social Groups

The term “social group” refers to a collection of individuals who are united by some type of common bond (for more detail about the theoretical framework, see Jazwa 2016: 25–55). This common bond is manifested and recognised by the performance of shared behaviours, or practices, in specific contexts of interaction (Brass 1980; Jones 1997: 88–90; Knappett 2011: 106). Such practices are adopted through frequent interactions among individuals and become habituated by the regular repetition of those practices in the same contexts (Jones 1997; Dobres 2000). The subsequent recognition of these shared ways of acting helps to engender notions of belonging among the actors in that context (Bentley 1987; Jones 1997: 90– 91). The characteristic practices of each group, however, do not remain static over time. They are constantly reassessed as new behaviours are witnessed, performed, and disseminated by individuals (Bourdieu 1977: 168–71; Jazwa 2016: 36–37). The members’ diverse responses to the new practices can have profound effects on individuals’ status

152

Jazwa

within a group and the overall organisation of the social network itself (Graves-Brown 1996; Thomas 1996: 19–20; Wenger 1998). Traditionally, archaeologists have largely been concerned with identifying specific varieties of social groups: ethnicities, tribes, or, more generally, “people”. These identities integrate individuals beyond the scale of household or site and are often expressed with notions of shared ancestry, geographic place of living, mythological origins, or other incorporating explanations. Because such groups are constituted in the context of daily living, rather than for specialised activities like craft production, I apply the term daily social group to describe that broad collection of identities. Although many of these groups distinguish themselves from others by an innate sense of “belonging” or common ancestry (Barth 1969; Hall 2002: 17), such notions of shared origins belie many commonly-held group practices that were inculcated during quotidian activities and regular interactions with other members of the social network since birth (Barth 1969; MacSweeney 2009: 105). Individuals are not members of only one social group (daily or otherwise) (Jones and Graves-Brown 1996: 7; Dobres 2000). Each person’s collection of identities at any time constitutes his/her “social persona” that is entirely individualised (Goodenough 1965). This unique quality of each person’s social persona influences his/her performed group behaviours and contributes to the variability of practices by members of each group because distinct social groups may have conflicting group practices in a single context of interaction (Cohen 1978; Thomas 1996: 19; Wenger 1998). The individual must negotiate the multiple identities by making conscious or subconscious preferences with their practices. Individuals can also pursue their creative agency to consciously act differently relative to the group norms. Such variation of practices within a group means that only a majority correspondence of group behaviours among members is expected, rather than a complete adherence to any list of group practices. For this reason, a simple list of behaviours cannot define or identify a social group. 3

Domestic Architecture

Domestic architecture is an ideal artifact to study when identifying daily social groups, because nearly everyone regularly interacts with domestic structures as the setting of most of their daily activities (Hillier & Hanson 1984). Domestic structures are typically designed to facilitate the expected actions that occur within, and individuals who interact regularly during the context of their daily

living tend to organise their structures similarly (Hillier & Hanson 1984; Rapoport 1990: 12–13). Thus, a great correspondence among individual qualities of spatial organisation indicates common notions of appropriately designed space and the practices expected to occur inside. Some variation in house design within a group, however, is still expected due to the unique social personae of the inhabitants and/or their creative human agency (Hillier & Hanson 1984; Jazwa 2016: 53–54). A careful examination of building construction can also reveal practices that are characteristic of daily social groups (Sanders 1990: 45). Anthropological studies of house construction in pre-modern societies have shown that domestic structures are typically built by the inhabitants and their close relations (e.g., Kus & Raharijaona 1990; Yakar 2000: 139). Each construction event, therefore, serves as a performance for disseminating commonly held technical knowledge among producers (Lemmonier 1993; Graves-Brown 1996: 91; Gosselain 2000). Although specialists sometimes help with the construction of individual elements of the superstructure, such as roofs, the general structural work of prehistoric Greek buildings can be built by any lay-person. Many recent studies have explored the social aspect to material production. Models of the chaîne opératoire, applied most frequently by archaeologists to ceramic and lithic analysis (e.g., Lemonnier 1993; Gosselain 2000; Roux 2003), demonstrate that individual steps in the production sequence can be altered with little or no impact on the final form of the object or its functional fitness. In other words, there is no single “right way” to build an object and production choices are products of the interpersonal relationships within networks of interaction through which knowledge and technique are transmitted. Even if production choices are altered, the remaining sequence of steps remains. In prehistoric Greece, the networks through which construction knowledge was disseminated are identical to those networks in which notions about proper organisation were maintained because the inhabitants of domestic architecture were the individuals responsible for construction. Although practices associated with construction are limited by technological or material constraints, they remain largely the product of established tradition and past building experiences. Most aspects of the construction, in fact, have minimal impact on the final form of the structure and, thus, are unlikely to be changed by aesthetic or functional influences. In short, individual components of the spatial organisation and construction all directly or indirectly represent past behaviours in a single context of interaction: the daily living environment. Consequently, a significant

DOMESTIC ARCHITECTURE & IDENTITY DURING THE BRONZE AGE TO IRON AGE TRANSITION

correspondence of many of these attributes will reveal groups of individuals who interacted frequently and acted similarly in that context. Any results from such a correspondence analysis of domestic architecture (or a constituent part) should be confirmed with the behavioural examination of other artifact types used in the context of daily interaction; however, the use of additional artifact types together with domestic architecture in an analysis would require careful consideration of chronology, patterns of use, function, and object history. By examining the correspondence of individual behaviours rather than the general form or appearance of architectural plans, this study differs from traditional approaches to prehistoric Greek architecture (e.g., Hiesel 1990; Darcque 2005; Wiersma 2014) which tend to evaluate general attributes of architecture that are easily imitated, malleable, or purpose-driven, rather than learning traditions and group knowledge. In prehistoric mainland Greece, the overwhelming number of excavated and published structures have stone socles and mud brick superstructures. The evaluated behaviours in this study are associated with these durable stone foundations and can be identified relatively easily with an inspection of these remains. The consistent placement of a larger stone at the intersection of two walls, for instance, indicates that the builders intentionally chose to employ this technique. Qualities of the spatial organisation of domestic architecture are also indicated by the arrangement of the stone walls, but generally do not represent individual behaviours of the past inhabitants.

153

Instead, they guide a suite of possible behaviours occurring within that built space (Hillier & Hanson 1984). Despite the difficulty in isolating specific behaviours of the inhabitants with this category, a significant overlap of behaviour-guiding principles is expected among members in the same daily social group. The analysed aspects of architectural construction and use are described in this study as “Behavioural Qualities” (BQ), because they represent both direct and indirect evidence for individual behaviours (in Jazwa 2016, BQs are described as Behavioural Aspects, or BAs; the abbreviation was changed here to avoid confusion with the abbreviation often used for the “Bronze Age”). I have identified a significant quantity of these BQs, 180, for potential correspondence among contemporary structures (see Jazwa 2016: 72–102 for the full list and description of each). By considering many BQs, I can mitigate potential behavioural overlap among distinct groups and diminish the statistical “weight” of each behaviour, so that potentially insignificant or incidentally correlated BQs do not distort the correspondence analysis. In fact, the precise BQs are less important than the quantity of them, so long as they all represent potential practices performed in the context of the inhabitant’s daily living environment. The analysed BQs comprise three general categories that are associated with the buildings’ construction and use: spatial organisation (use), construction techniques (construction), and scale (construction) of the building and its constituent parts (absolute and proportional) (Table 9.1). Spatial organisation represents the building’s

table 9.1 A sample of the 180 Behavioural Qualities (BQs) that were examined for each structure and assigned to a specific type of practice (after Jazwa 2016, 72–102)

Behavioural qualities Spatial organization

Construction techniques

Scale

Building Type Proportional (prop.) length : width, building Prop. building area by building type Building axiality Building symmetry Open/closed organization Average depth, building Prop. main room : building area Avg. relative asymmetry (RA) value Hearth axiality to door Prop. length : width, main room

Construction on earlier structures? Same building plan as earlier structure? Reuse of earlier walls? Chinking, exterior (ext.) wall presence Exterior-to-interior (int.) wall bond? Built platforms? Built benches? Ext. wall width consistency Int. wall width consistency Difference of the smallest to largest ext. wall widths Stone shape in the core of ext. walls

Prop. main room : porch area

Prop. surface area, average stone : anchor stone

Building Area Courtyard area Main room area Maximum room width Porch area Surface area, avg. stone Surface area, anchor stone Surface area, corner stone Avg. exterior wall width Prop. doorway width : wall width Prop. width, ext. wall : largest room width Surface area, 1st course stone

154 capacity to frame and organise the activities that are expected to occur inside. The relative placement of hearths within rooms, the location of doorways, and the size of main rooms all influence the use of the space by the inhabitants and visitors. Construction techniques, such as the use of corner stones, anchor stones, large stones for bottom courses, etc., are individual steps in the chaîne opératoire and disseminated during moments of teaching and other construction experiences. Building scale evaluates the specific material choices (in terms of size), the available space for daily living, and the supporting abilities of the architectural components. Although the scale of the constituent parts of architecture is influenced by the building material, it is not absolutely dictated by the local geology, because significant variation exists in the size of building materials used locally over time. At Asine, for instance, there is a 52% change in the average stone size for construction from EH III to PG (Jazwa 2016: 247–49). Additionally, although relative building scale has been shown to be influenced by social complexity and can be actively manipulated (Westgate 2015), it is still influenced by social factors. When considered with changes in building techniques and spatial organisation, therefore, it can reveal the purpose for such changes (i.e., active manipulation vs. broader social changes). With typical group behaviours varying diachronically, the correspondence of all these BQs must be analysed only for contemporarily built structures. The “contemporary” quality to the analysis incorporates all structures that are built during an extended period, rather than at a single moment. A study that includes overly narrow periodic distributions will not account for the long processes of dissemination and adoption within the social networks of pre-modern societies. In contrast, periods that are too long can produce results that potentially equate diachronic changes of practice within a group to real social differences. For this case study, I consider individual sub-phases LH IIIB, LH IIIC, and PG for the periods of contemporaneity (Jazwa 2016: 63–64). These periods are each between 110 to 175 years in duration, accounting for four to six generations. The analysed BQs are not distinct to any sub-phase during this period and are generally applicable to all stone-and-mud brick architecture built in mainland Greek or elsewhere until the modern period. This might aid in the application of the methodology to other periods and cultures for similar aims or, even, potentially for identifying Greeks (or others) living abroad. The BQs can also be altered to correspond more closely to peculiar environmental or cultural traditions in other regions or periods of the world. Consequently, the individual BQs

Jazwa

themselves are less important than considering many that are broadly identifiable within the study region and represent past human practices within a single context of interaction. This means that the overlapping group behaviours identified in this way do not define the social group; they are simply a product of the group and the members’ intra-group interactions. Although MacSweeney (2009: 206) argues that the practices associated with group identity must necessarily promote the ideology of the group’s social rationale, implying an act of definition, this assumes that identity is always actively constructed and that all actors are equally conscious of the group function. Bourdieu (1977) and others (e.g., Jones 1997), in fact, have shown that practices are subconsciously inculcated during the performance and recognition of activities. In the domestic realm, especially, practices are often brought about with little reflection or thought about their purpose (Ortner 1984: 150). 4 Methodology For the acquisition of the analytical data, I consulted published excavation reports and, when possible, visited the extant architectural remains for personal examination (see Jazwa 2016 for brief architectural descriptions of all buildings covered in the study). A number of synthetic monographs that include comprehensive collections of prehistoric architecture were particularly valuable for the analysis (Shear 1968; Sinos 1971; Hiesel 1990; Mazarakis Ainian 1997; Lemos 2002; Darcque 2005). All the retrieved data were then entered in a project-specific database, and the mean values of each BQ were calculated for each structure. Although I aimed for a complete identification of BQs, this was not always possible for each building due to the preservation and state of publication. Not every BQ, however, needs to be present for each structure to achieve a successful correspondence analysis. To determine correspondence among structures, a multi-step statistical analysis is required for each period (Fig. 9.1). This process begins with a normalisation of all BQ values in preparation for evaluation. During data collection, all BQs are recorded as one of the following data types: absence/presence variables, group/categorical data, or quantitative values. Because the resulting values of all BQs are not directly equivalent, they must each be transformed—in this case, into positive integers that represent separate groups of closely related values. The unique nature of each data type, however, requires different methods of normalisation. The normalisation of qualitative data, for instance, is rather simple (Table 9.2).

DOMESTIC ARCHITECTURE & IDENTITY DURING THE BRONZE AGE TO IRON AGE TRANSITION

155

figure 9.1 An outline of the statistical correspondence analysis author table 9.2 A sample of the absence/presence and categorical data and the normalization of these values

Primary data

B.1 B.2 B.3 B.4

Normalized data

Building type

Corner Corner stone stone, presence type

Corner Regularity Regularity Building Corner of wall stone type stone, of wall courses presence type courses

Apsidal Megaroid Megaroid Corridor

No Yes No Yes

Group 1

Single

Stacked Group 2

1 2 2 3

For the absence/presence data type, all absence values are assigned a “1” integer; the presence values are represented with a “2”. Categorical data receive positive integer values beginning at “1” and continue until each category is assigned a separate integer values. When multiple classifications are present in a single structure, e.g., the presence of corner stones in some exterior walls but not all, I include this as a “presence” variable so long as the technique appears in two or more walls of the same category (exterior/ interior). “Consistency” of construction techniques within a single structure is also a BQ. Most structures, however, demonstrate surprising consistency among all the walls of the same category. For BQs that could not be identified, the BQ value was left blank and did not contribute to the correspondence analysis. BQs related to the spatial organisation and scale typically record quantitative values and require a more complicated method of normalisation. Although these all represent the same data type, the values are not directly comparable due to the different scales of measurement for each. For instance, the difference in overall room size,

1 2 1 2

1 2

1

2

e.g., 2 m2 to 50 m2, is quantitatively more significant than the difference in the proportional length-to-width room measurements, e.g., 0.20 to 1.0. Consequently, I attempt to identify clusters of related values from the dataset and assign each cluster a positive integer that represents an individual group of close correspondence for that BQ. The specific method used for identifying clusters of related quantitative data, however, can yield biases concerning the qualifications for identifying relatedness. To mitigate such biases, I employ a dual strategy that incorporates both statistical impartiality and subjective inspection of the dataset for evident clusters of data. This produces parallel datasets for each list of contemporary structures that I analyse separately and reconcile later. For the first method of normalisation, I calculate the standard deviation of all data for each BQ and assign group numbers that correspond to each structure’s distribution relative to that standard deviation. All values that are less than the standard deviation are considered as a single group and receive a “1” value. Those that are found within the standard deviation are given “2”; those

156

Jazwa

a

b

Standard Deviation

Groups

Standard Deviation

Groups

figure 9.2 (a) The distribution of the average stone width values for a sample of structures. The continuously distributed data do not easily reveal significant clusters of related values. (b) The distribution of the proportional length-to-width values of complete buildings for a sample of structures. The standard deviation does not adequately account for identifiable clusters of data author

greater than the standard deviation are assigned to group “3”. The standard deviation provides a statistical metric with which to classify all data and is particularly useful for continuously distributed data for which no or few clusters of structures can be identified in the distribution graph (Fig. 9.2a). Such a method of normalisation, however, can fail to identify natural clusters in the dataset. Figure 9.2b offers a sample dataset in which the standard deviation does not recognise relevant groupings of related structures. For the second method of normalisation, I attempt to reconcile the shortcomings of the standard deviation method with a subjective measurement of relatedness. With this “group normalisation”, I visually examine the distribution of the quantitative data and assign each structure to an apparent cluster of correspondence for each BQ. Although this method is more subjective than the standard deviation normalisation, it will account for obvious clusters in the data. This method, however, is scale dependent and can offer different numbers of clusters depending on the scale at which the dataset is visually examined. In Figure 9.2b, for instance, 2, 3, or 4 groups can be identified depending on the viewer’s preference for distinguishing relatedness. For each period, the two normalised datasets are individually evaluated for total correspondence of BQs. Because the degree of correspondence is a relative metric, no single statistical test can be applied to the data to produce a list of “greatly related” structures. Consequently, I first carry out an exploratory analysis to determine the nature of relatedness before assigning a measure of significant correspondence. For this, I apply an unweighted pair-group average (UPGMA) hierarchical cluster analysis. The UPGMA cluster analysis is a multi-variate analysis that creates a dendrogram of relatedness among structures

based on the average distance among members. Structures that have a closer correspondence of BQs are connected at lower “distances” in the dendrogram relative to lessrelated structures (Fig. 9.3). These distance values reflect the degree of integration and can provide information about the relatedness of recognisable groups. The absolute values, however, are not necessarily comparable with those of other datasets, because the distance in each dendrogram varies according to the number of structures and BQs available for analysis. With this exploratory analysis, several distinct groupings are typically evident at different scales in the dendrogram. I take note of the number of groups that are clearly distinguished in the dendrogram and the distance at which each is joined together. In Figure 9.3, for instance, three groups are evident at a distance of 13, two appear at a distance of 16, and the entire system is integrated at a distance of 18.5. In the LH IIIB-PG case study, groups are identifiable at two or three distances in every dendrogram and are recorded with the distance at which the groups are integrated. These data are used for interpreting the degree of correspondence of all structures. The final step of the correspondence analysis reconciles the distinct group assignments for each normalised dataset. First, I apply two to three k-means tests to each normalised dataset (four to six total for each period). The specific number of k-means tests depends on the distances at which clearly identifiable groupings are present in the dendrogram. In most periods, only four k-means tests are required (for an overview of the k-means test and its derivation in the statistical software, see Hammer 2013: 96). The k-means test randomly assigns each structure to one of the pre-determined number of groups, and the mean values of each are calculated. For this analysis, the pre-determined number of groups are the significant groupings evident

DOMESTIC ARCHITECTURE & IDENTITY DURING THE BRONZE AGE TO IRON AGE TRANSITION

157

figure 9.3 PG’s UPGMA dendrogram of BQ correspondence (group normalisation) author

during the UPGMA analysis of each normalised dataset. Using an iterative procedure, entries are moved to the groups that share the closest mean value until entries can no longer be moved. This produces lists of structures that share a close correspondence of all BQs. Although the initial random distribution produces some variation in the distributions during additional tests, very well-integrated groups of structures tend to cluster consistently in subsequent runs of the test. Thus, the degree of relatedness among structures and the entire system can be evaluated by considering the group assignments of the k-means tests. Overall, this methodology offers a relatively conservative metric for identifying significant behavioural correspondences among the inhabitants. Most importantly, the correspondence analysis does not rely on a strict checklist of features or artefacts for inclusion in the group. Instead, it allows for the variability and dynamism inherent to any social network (see, for instance, Graves-Brown 1996). With this variation, members of each group are not treated as passive recipients of ideas or techniques, but are afforded some degree of creative agency and/or negotiation arising from the conflicts inherent to an individual’s social persona (Goodenough 1965). 5

The Case Study

The LH IIIB-PG case study corresponds to a transitional period in mainland Greece. During the initial phase of analysis, LH IIIB, the development of the Mykenaian palatial system reached its acme (Shelmerdine & Bennet 2008; Shelton 2010). Large polities that were centred on a “palace” or primary site rose to prominence to control a regional hinterland (Taylour 1964; Mylonas 1966;

Shelmerdine 1997; Shelmerdine & Bennet 2008; Shelton 2010). Linear B tablets indicate that these palaces were headed by a single king, or wanax, who oversaw complex administrative and economic structures. Each maintained a system of taxation, limited redistribution of staple and wealth items to officials, priests, and others, and acquisition and production of prestige items (Kilian 1988; Palaima 1995; 2006; Shelmerdine & Bennet 2008; Shelton 2010). Mykenaians also actively contributed to the trade and politics of the eastern Mediterranean, and maintained contact with the Hittites, Egyptians, and other cultures (Cline 1994; Parkinson 2007; Burns 2010; Schon 2010; Murray 2017). These developments and the increased interaction within Greece likely contributed to the Mykenaian material koiné of the period that was characterised by outwardly similar pottery, architecture, figurines, jewelry, wall painting styles, etc. (Shelton 2010: 145). Broadly speaking, this reflects an ostensible unity in the material record of mainland Greece by the end of the Mykenaian period, but regional differences in politics and economic strategies are still evident. At the end of the LH IIIB period, a widespread collapse of political structures is seen throughout mainland Greece. Palaces burned, the palatial administration ceased to function as it had before, trade was disrupted or reoriented, and there was widespread population movement within, from, and into mainland Greece. Unfortunately, the precise causes of these changes have not yet been determined (for an overview, see Dickinson 2006: 46–56; Cline 2014). Various models attribute agency to environmental conditions (Carpenter 1966), internal unrest, the invasion of outside groups, etc. (Desborough 1964; Rutter 1975; 1990; Winter 1977; Deger-Jalkotzy 1977; 1983). More likely, a combination of several of these factors produced

158 unrest within the tightly connected networks of the eastern Mediterranean. Despite these changes to the established political and economic systems, ongoing research into the subsequent LH IIIC period indicates at least some continuation of the Mykenaian material culture and trade networks (Dickinson 2006; Knodell 2013; Murray 2017; van den Berg 2018) with LH IIIC often treated more as an extension of the Mykenaian period, rather than the beginning of a new chapter in mainland Greece (Dickinson 2006). In many respects, there is more discontinuity in mainland Greece at the onset of the EIA. In PG, there is a distinct contraction in the overall number and size of settlements throughout mainland Greece (Mazarakis Ainian 1997; Lemos 2002), suggesting significant depopulation (Snodgrass 1971; Desborough 1972; Dickinson 2006: 63; Lemos 2002; Murray 2017). Excavations at several sites, including Levkandí (Popham, Calligas & Sackett 1993), Mitrou (Van de Moortel 2007; 2009; Van de Moortel & Zahou 2011), and Nikhoria (McDonald, Coulson & Rosser 1983), indicate the widespread appearance of large apsidal structures—a building type that had not been used for large, primary constructions at settlements for centuries. Although Lemos (2002: 149–50) writes that the apsidal building form had disappeared from Greece and was “reintroduced” in the PG period, the archaeological evidence indicates that it was still used in some areas on a much more limited scale (Jazwa 2016: 207–08). Still, the increase in the number of these apsidal buildings is dramatic enough to clearly indicate a change. With changes to other aspects of material culture, as well, “Mykenaian” culture had, on its face, largely ceased to continue. Consequently, this period has traditionally been interpreted as a starting point that eventually led to the developments of the Classical Greek culture. However, this is not an entirely accurate assessment because there is no complete break from the past with demonstrable continuity of certain elements of LBA tradition, such as myths, language, religion, and elements of artifact style, continuing into the Iron Age. To evaluate the potential cause(s) and long-term effects of societal change during LH IIIB-PG, a precise understanding of the identity of those living in Greece at that time is required. Since the initial interest in this period, the social identity of the inhabitants has contributed to interpretations about the apparent cultural change. Even Classical Greek authors, such as Thoukydides, attributed the presence of new population groups as an explanatory model, with the LBA “Age of Heroes” ending with the “Coming of the Dorians” and the “Return of the

Jazwa

Herakleidai” (Thouk. 1.12; 5.112; Diod. Sik. 4.57–58; Apollod. 2.8). Although this precise narrative has been rejected by scholars, some interpretive models still suggest population movement or invasion as a factor or result of the palatial collapse (Desborough 1964; Rutter 1975; 1990; Winter 1977; Deger-Jalkotzy 1977; 1983; see Dickinson 2006 for a summary). With this case study, I can contribute to this discussion by demonstrating continuity or disruption in the social networks during the LBA-EIA transition. To this end, I examined 184 structures that were built from LH IIIB to PG, assessing each by the 180 BQs (Fig. 9.4) (for a description, chronology, and bibliography of each structure, see Jazwa 2016, where I include all potentially domestic structures that do not display specialised construction techniques, e.g., cut stone masonry). The list of structures for the case study is by no means a complete list of all LH IIIB, LH IIIC, and PG architecture that has been excavated in mainland Greece and Aegina, but includes only structures that are published with detailed plans and have at least two connecting walls preserved. This case study represents just a small component of a much larger application of the methodology to 458 structures built between Early Helladic III and the PG periods. The results from some of these earlier periods will also be used for the interpretation of this case study. Each structure is assigned to the chronological period to which it was attributed by the original excavators unless recent reconsiderations have led to a consensus for a different date. Structures that have imprecise dates of construction, such as LH IIIB/C, are evaluated twice, once for each period. In one respect, this imprecision was a boon to my study as an internal test to the methodology. All imprecisely dated structures that share correspondence in one phase proved to share correspondence with the same imprecisely dated structures in the subsequent phase. Thus, the correspondence analysis demonstrates consistency in its groupings according to the provided BQ values. In all, thirty-eight structures included in the analysis have imprecise dates of construction (LH IIIA-B: 21; LH IIIB-C: 16; LH IIIC-PG: 1). Although the included structures are discontinuously distributed throughout the landscape of mainland Greece and Aegina for each period, this also will not impact the correspondence analysis. A benefit of the behavioural correspondence analysis is that geographic distribution is not an explanatory or interpretive agent for the reconstruction of identity (Barth 1969; Graves-Brown 1996: 83; Relaki 2004). Interaction among individuals is more likely to occur in closely distributed settlements, but geographic proximity does not determine identity.

DOMESTIC ARCHITECTURE & IDENTITY DURING THE BRONZE AGE TO IRON AGE TRANSITION

figure 9.4 The sites with structures included in this analysis author

159

160 6 Results The application of the methodology to the collection of LH IIIB architecture demonstrates the existence of a single, extended network of interaction with all structures grouping together in at least one of the five k-means tests. Within this overall correspondence, there are collections of structures that group together more consistently (for a contextual interpretation of this, see Jazwa 2016). The system-wide integration of all structures, however, occurs at a very low average maximum distance of integration of 29.5. This value represents the average distance on the dendrogram at which all 93 structures are joined in both normalised-data dendrograms. When corrected to account for the number of structures relative to all LH periods, this value, 20.24, is the lowest for the entire LBA. Such a low value represents greater correspondence among the entire system, and demonstrates that the network of interaction in mainland Greece was at its most integrated. As such, it is likely that there was a single social group among the analysed population (Jazwa 2016: 359). This unified social network is a continuation of a pattern that is unique to the LBA (Jazwa 2016 argues for the existence of a single Mykenaian social group). At the same time, there is significant stability in the average BQ values with nearly no substantial changes to significant BQ values from LH I to LH IIIB; such “significant BQs” represent prominent construction techniques or intentional construction decisions, rather than passive, potentially subconscious practices. Evident changes are largely limited to a steady growth in the scale of the buildings and its constituent parts (Fig. 9.5). The trend towards larger

Jazwa

buildings likely is due to the increasing number of elite structures built during the Mykenaian period and not to changes in social organisation. During the transition from LH IIIB to LH IIIC, the Mykenaian palaces collapse and an ostensible crisis spreads throughout mainland Greece. At first glance, an examination of the mean BQ values for that period suggests complementary and substantial changes to building traditions (Fig. 9.5). Closer inspection, however, indicates that these changes are largely decreases in scale, with a decline in the mean stone size, wall size, room size, and building size (Table 9.3). Such BQs likely correlate positively with the overall building size. Unfortunately, the available sample size for LH IIIC is too small to ascertain this conclusively for that period (a correlation analysis for these BQs from EH III to PG does indicate a positive correlation over the long-term). At the same time, the construction of walls becomes more inconsistent and special stones, such as enlarged corner stones, are employed less frequently. Because the predominant changes are to building scale and do not accompany significant changes in building techniques or spatial organisation, the alterations to the building traditions in LH IIIC likely do not indicate an altered social identity. It seems that the evident changes in scale can be attributed to the lack of monumental domestic structures that were constructed in LH IIIC and less investment in the stability and durability of any new structures. Whereas the LH IIIB period witnessed the construction of significant elite residences, such as Building Mu at Mykenai or 7-I Kalamianos (Korfos) (for Building Mu see Shear 1968: 235–49; Hiesel 1990: 52–54, 147–49, and for Kalamianos

figure 9.5 Substantial changes to significant BQ values in the LBA and PG periods author

DOMESTIC ARCHITECTURE & IDENTITY DURING THE BRONZE AGE TO IRON AGE TRANSITION

161

table 9.3 Substantial changes to BQ values from LH IIIB to LH IIIC

Significant changes to average BQs, LH IIIB to LH IIIC Average building surface area (m2) LH IIIB 215.98 LH IIIC 75.26

Average main room size (m2) LH IIIB 31.49 LH IIIC 29.00

Average Interior Wall Width (cm) LH IIIB 62.46 LH IIIC 53.25

Average Exterior Wall Width (cm2) LH IIIB 64.56 LH IIIC 58.70

Average Stone Size, L×W (cm2) LH IIIB 1364.08 LH IIIC 808.56

Average Stone Size, L×H (cm2) LH IIIB 958.42 LH IIIC 496.20

[Korfos] see Tartaron et al. 2011), there are no comparable building complexes in LH IIIC. Perhaps the widespread internal population movement and general uncertainty after the palatial collapse influenced the inhabitants of mainland Greece to build structures more hastily and with less of a labour investment. Additionally, there was widespread population movement within Greece (Dickinson 2006: 62–67), which may have occasioned the construction of less robust domestic buildings. These changes to significant BQs, however, represent only the mean building traditions of all 77 structures in LH IIIC, and not the overall correspondence within each period. The applied methodology demonstrates considerable continuity with an internal coherence of building traditions from LH IIIB to LH IIIC. Again, all structures share a group with all other structures in at least one kmeans test. The relative distance of integration for the system, 21.75, is also nearly identical to that of LH IIIB when corrected for the available BQs and the 18 fewer structures (non-corrected distance, 25). This indicates the persistence of established identities and networks of interaction through the palatial collapse. It is quite likely that the persistence of earlier social identities facilitated the transition of political power to lower level officials or local leaders after the collapse of the Mykenaian palatial administration. Although Crielaard (2011) has criticised the traditional wanax-basileus model as long-term explanation of EIA leadership terminology and structures, there certainly was continuity of power in some manner from LH IIIB to LH IIIC. As regional collections fractured, some local/sub-regional leaders or elites gained power without the palatial oversight of the

previous era. Certainly, different strategies for consolidating power were employed throughout the mainland, but at least some leaders made appeals to the past, such as with the construction of Building T at Tiryns (Maran 2000; 2001; Dickinson 2006: 61). Building T is a smaller megaron that was constructed within the still-extant ruins of the palatial megaron at Tiryns and recognised the earlier placement of the hearth, altar, and throne. Maran (2001) convincingly argues that the new leader(s) at Tiryns was respecting the older palatial authority at that site for present legitimation. The persistence of a common identity likely helped to make these appeals meaningful to their audience. Additionally, the continuity of identity into LH IIIC likely contributed to the successful population movement within Greece and the resettlement of individuals from disparate locations at new centres, like at Levkandí and Pérati. Generally speaking, population movement is quite frequently aimed at locations in which there were common identities or shared kinship groups (Steele 2013; Jazwa 2016: 370–73). The long-standing continuity in building traditions, however, dramatically changed at the onset of the EIA. There are more substantial alterations to significant BQs during the transition from LH IIIC to PG compared to the LH IIIB-LH IIIC transition, and these changes are largely of a different nature (Fig. 9.5). Rather than indicating alterations to building scale, the mean BQ values in PG demonstrate fundamentally different qualities of spatial organisation. Structures are narrower, more axially arranged, and more symmetrical (Table 9.4). There is also evidence for an increased emphasis on the main living space of the PG structures; the main room surface

162

Jazwa

table 9.4 Substantial changes to BQ values from LH IIIB to LH IIIC

Significant changes to average BQs, LH IIIC to PG Ratio, width : length, building LH IIIC 0.58 PG 0.45

Average main room size (m2) LH IIIC 29.00 PG 59.68

Ratio, Area Main Room : Building LH IIIC 0.42 PG 0.65

Maximum Covered Span (m) LH IIIC 4.27 PG 4.92

Average Doorway Length (m) LH IIIC 1.07 PG 1.40

Presence of Built Thresholds LH IIIC 71% PG 50%

area, largest covered span (all spans—supported and unsupported—were measured together to gain insight into the area of defined space; see Jazwa 2016: 220–21), and proportional size of the main room relative to the entire building all increase dramatically. The greater emphasis on the main room suggests that there were more activities in this room than had previously occurred in specialised and segmented spaces characteristic of LH domestic architecture. Besides these changes, the exterior space likely became a more prominent extension of the household activity areas. Built thresholds for doorways were less commonly employed and the mean doorway size increased. This made the interior more visible from the outside and blurred the distinction between these two spaces. Altogether, these new arrangements indicate different functional and symbolic uses of domestic space and strongly hint at a different organisation of the activities that regularly occurred inside by the inhabitants. At the same time, the formerly unified network in mainland Greece becomes fractured and dispersed during the PG period. Although only a small number of structures (18) were available for analysis, few appear in the same k-means groups of correspondence. The collection of structures is also very loosely integrated, with the corrected average distance of integration for the distance at very high, 94. Admittedly, this value is likely inflated, since the very small number of structures (18) relative to LH IIIB and IIIC (95 and 77, respectively) distorts the correction. This dispersion, however, cannot be attributed simply to geographic isolation during this so-called “Dark Age”, because all structures that share several k-means groupings are not found within discrete geographic boundaries or regional divisions. In fact, structures built in the southern

mainland quite frequently share a group with at least one structure in the northwest (Figs. 9.3–9.4). When the PG structures from multi-period sites are compared with earlier structures from the same sites, the local discontinuity is even more evident (Fig. 9.6). Except at Kynos, there is little correspondence of structures at the same sites from LH IIIC to the PG period. The correspondence of the southern and central PG structures with all LH IIIC architecture further confirms discontinuities in established building traditions (Fig. 9.7). The included structures from Asine, Mitrou, and Levkandí, for instance, only integrate with other LH IIIC structures at a distance in which all LH IIIC buildings already share correspondence with several contemporary structures. Though somewhat closer to LH IIIC building traditions, Thermos, and Nikhoria’s domestic structures still do not closely correspond with any other LH IIIC buildings. Altogether, there appears to be a fundamentally different organisation of the social groups in mainland Greece and a new suite of behaviours practiced in the context of daily living. Such changes in spatial organisation and regular daily activities cannot be attributed simply to a regressive economy, depopulation, isolation, or political changes. Instead, profound alterations to daily living must represent new patterns of interaction that resulted from a period of a crisis. In fact, the changes from LH IIIC to PG correspond well with Driessen’s (1995) definition for “crisis architecture”. The close correspondence between northern and southern structures in PG and the adoption of the apsidal building form at sites such as Asine, Mitrou, and Levkandí may shed light on the cause of at least some of these changes. Although the apsidal building form had never ceased to be built, the frequency of apsidal domestic structures in

DOMESTIC ARCHITECTURE & IDENTITY DURING THE BRONZE AGE TO IRON AGE TRANSITION

163

figure 9.6 A correspondence analysis (group normalisation data) of PG (grey background) with formally similar structures at the same site from the preceding period with architecture author

figure 9.7 A correspondence analysis (group normalisation data) of southern and central PG structures with evidence for local discontinuities and all LH IIIC structures author

the PG period is only ever matched in EH III (Jazwa 2016: 207). Because the apsidal building form is a highly visible BQ that is shared by all three structures, it is possible that the building form was an emblemic signal of a common background that belied subtler and less visible similarities in daily practices (i.e., other BQs). I would argue that the sudden proliferation of the apsidal building form at this time and the unique building tradition that it represents suggest a small-scale settlement of individuals who left few significant archaeological and architectural remnants for detection—a semi-nomadic group(s) who previously practiced pastoralism (or mixed subsistence) and lived in the mountainous regions of Greece. Such a small-scale settlement may not necessarily correspond to any named/known ethnicity, nor was it necessarily a direct cause of any political or economic upheaval. It may have simply been yet another symptom

of the crisis occurring in the eastern Mediterranean at the end of the LBA. Although the mere presence of apsidal architecture does not indicate formerly nomadic groups on its own, the sudden introduction of new attributes of spatial organisation along with the building form and other indirect evidence for such a settlement(s) is suggestive. For instance, rounded structures are much more commonly built by mobile (or formerly mobile) populations throughout the world because they are easier and quicker to construct with ephemeral materials (twigs, sticks, skins, etc.) (Whiting & Ayres 1968; Flannery & Marcus 2012: 128). The kin-based organisation of PG society, lesser segmentation of domestic space, and the increased importance of “outdoor living” that become more prominent in mainland Greece are also common characteristics of mobile populations throughout the world. This kin-based social

164 organisation is well-accepted for the PG period and is suggested by the clustering of graves around central domestic structures (see Morris 1987; Whitley 1991; Mazarakis Ainian 1997; Lemos 2002). Additionally, certain classes of ceramics appear in Greece during the Postpalatial period that suggest the reproduction of semi-nomadic vessel forms that were constructed originally with organic materials, such as the “Leather-bag Ware” (Lis 2009: 159). Although this material has not been found at sites with excavated apsidal architecture, the appearance of such items suggests interaction with the producers, who likely lived nearby. A nearby point of origin of these incoming groups, such as the mountainous regions of central and northwestern Greece, may also explain the lack of new construction techniques in PG. Close or previous contact between the established sedentary Mykenaian individuals and these indigenous mobile groups may have facilitated their settlement at established sites. As the semi-nomadic individuals settled at agricultural communities, inter-married, and received communal help, the haphazard collection of new settlers may have been taught by the remaining Mykenaians the “proper” stone-building techniques as they transformed their ephemeral structures into permanent buildings (see Flannery & Marcus 2012: 128). While maintaining the traditional form of their previous buildings in the short term, they rendered the structures in more permanent materials. Unlike the apsidal buildings, the domestic structures from Kynos, Nikhoria, and Thermos demonstrate a closer building tradition to that of LH IIIC structures and a different social identity from their PG counterparts (Fig. 9.7). Among these PG structures, Kynos is unique because it demonstrates apparent continuation of local building practices (Fig. 9.6). The structures from Nikhoria and Thermos, in contrast, are not very closely related to the local building traditions, but are distantly related to LH IIIC structures at other mainland sites. Although these sites demonstrate greater discontinuity in building traditions compared to earlier periods (Jazwa 2016), the building traditions are more similar than is seen with the apsidal structures. The evident changes among the rectilinear structures, I suggest, may have been due to several factors, including the internal population movements and the small-scale immigration of individuals from Italy, Kypros, and the Balkans in LH IIIC. Even when new populations were not present, the widespread instability may have contributed to a greater receptivity to adopting outside practices in an ad hoc manner. It seems, therefore, that the crisis during the Post­ palatial period did irreparable harm to a social group that

Jazwa

had been stable for many centuries. New identities formed and old ones dissolved at the onset of the EIA. Despite the presence of these new identities, the old population had not simply been replaced; in many cases, descendants of the Mykenaians remained in mainland Greece and simply reoriented their networks of daily interactions. This can account for the maintenance of economic ties abroad, continuity in some aspects of material culture, and the sustained use of the Greek language, despite the formation of new identities. Indeed, such attributes do not necessarily represent emic qualities of identity and can be shared among several different daily social groups at any time. Moreover, the diverse new identities do not need to correspond to later Greek social history. Hall (2002), in fact, suggests that the shared Greek identity only emerged much later in the Iron Age in mainland Greece. 7 Conclusion The results of the study demonstrate continuity in identity and social group organisation from LH IIIB to LH IIIC during and after the palatial collapse. Despite the average scale of structures and the constituent parts decreasing substantially, a unified identity among the inhabitants of the mainland persisted. By the PG period, however, the social networks became fragmented and dispersed. Building traditions were substantially altered and new identities emerged. Although the precise causes cannot be determined, the population movements within, out of, and into mainland Greece likely played a role by reorienting networks of regular, daily interactions. More broadly, the application of the proposed methodology to the case study has shown the success of this approach to archaeological analysis. Future applications of this approach, however, are not limited to the case study, region, daily social group type, or domestic architecture. A benefit of the methodology is that the essential framework can be adapted for the identification of other groups because it was built on theory rather than the archaeological evidence of a specific culture. Application requires a deep understanding of the context of action for the past social groups and the behaviours that represent such groups. Different artifact types must be examined for unique lists of BQs so that the reconstructed practices correspond with the context of interaction for the social group in question. Still, the approach offers a new line of analysis for Greek architecture and archaeological material that can contribute to established economic, political, environmental, and technological narratives of the past.

DOMESTIC ARCHITECTURE & IDENTITY DURING THE BRONZE AGE TO IRON AGE TRANSITION

List of References Barth, F., 1969: “Introduction”, in Barth, F. (ed.), Ethnic Groups and Boundaries: The Social Organization of Culture Difference (Boston) 9–38. Bentley, G.C., 1987: “Ethnicity and practice”, In Comparative Studies in Society and History 29: 24–55. Bourdieu, P., 1977: Outline of a Theory of Practice (Cambridge). Brass, P.R., 1980: “Ethnic groups and nationalities: the formation, persistence, and transformation of ethnic identities”, in Sugar, P. (ed.), Ethnic Diversity and Conflict in Eastern Europe. Joint Committee on Eastern Europe Publication Series 8 (Santa Barbara) 1–68. Burns, B.E., 2010: Greece, Mediterranean Commerce, and the Formation of Mycenaean Identity (Cambridge). Carpenter, R., 1966: Discontinuity in Greek Civilization (Cambridge). Cline, E., 1994: Sailing the Wine-Dark Sea: International Trade and the late-Bronze Age Aegean. BAR-IS 591 (Oxford). Cline, E. 2014: 1177 B.C.: The Year Civilization Collapsed. Turning Points in Ancient History 1 (Princeton). Cohen, R., 1978: “Ethnicity: Problem and Focus in Anthropology”, Annual Review of Anthropology 7: 379–403. Crielaard, J.P., 2011: “The ‘Wanax to Basileus Model’ Reconsidered: Authority and Ideology after the Collapse of the Mycenaean Palaces”, in Mazarakis Ainian, A. (ed.), The ‘Dark Ages’ Revisited vol. 1: Acts of an International Symposium in Memory of William D.E. Coulson. University of Thessaly, Volos, 14–17 June 2007 (Volos) 84–111. Darcque, P., 2005: L’habitat Mycénien: formes et fonctions de l’espace bâti en Grèce continentale à la fin du IIe millénaire avant J.-C. (Athens). Deger-Jalkotzy, S., 1977: Fremde Zuwanderer im spätmykenischen Griechenland: zu einer Gruppe handgemachter Keramik aus den Myk. III C Siedlungsschichten von Aigeira (Vienna). Deger-Jalkotzy, S., 1983: “Das Problem der ‘Handmade Burnished Ware’ von Myk. IIIC”, in Deger-Jalkotzy, S. (ed.), Griechenland, die Ägäis und die Levante während der “Dark Ages” vom 12. bis zum 9.Jh. v. Chr. Akten des Symposions von Stift Zettl (NÖ), 11.–14. Oktober 1980 (Vienna) 161–78. Desborough, V.R.D’A., 1964: The Last Myceneans and their Successors: An Archaeological Survey ca. 1200–1000 B.C. (Oxford). Desborough, V.R.D’A., 1972: The Greek Dark Ages (London). Dickinson, O.T.P.K., 2006: The Aegean from Bronze Age to Iron Age: Continuity and Change Between the Twelfth and Eighth centuries B.C. (London). Dobres, M.-A., 2000: “Social Agency and Practice: The Heart and Soul of Technology”, in Dobres, M.-A. (ed.), Technology and Social Agency: Outlining a Practice Framework for Archaeology (Hoboken) 127–63.

165

Driessen, J., 1995: “‘Crisis Architecture?’ Some Observations on Architectural Adaptations as Immediate Responses to Changing Socio-Cultural Conditions”, Topoi 5.1: 63–88. Flannery, K. & Marcus, J., 2012: The Creation of Inequality: How Our Prehistoric Ancestors Set the Stage for Monarchy, Slavery, and Empire (Boston). Goodenough, W.H., 1965: “Rethinking ‘Status’ and ‘Role.’ Toward a General Model of the Cultural Organization of Social Relationships”, in Banton, M. (ed.), The Relevance of Models for Social Anthropology (London) 1–24. Gosselain, O.P., 2000: “Materializing Identities: An African Perspective”, Journal of Archaeological Method and Theory 7.3: 187–217. Graves-Brown, P., 1996: “All Things Bright and Beautiful? Species, Ethnicity and Cultural Dynamics”, in Graves-Brown, P., Jones, S. & Gamble, C. (edd.), Cultural Identity and Archaeology. The Construction of European Communities (New York) 81–95. Hall, J., 2002: Hellenicity: Between Ethnicity and Culture (Chicago). Hammer, Ø., 2013: PAST. PAleontological STatistics. Reference Manual. Version 3.0: http://folk.uio.no/ohammer/past/past 3manual.pdf. Hiesel, G., 1990: Späthelladische Hausarchitektur: Studien zur Architekturgeschichte des Griechischen Festlandes in der Späten Bronzezeit (Mainz). Hillier, B. & Hanson, J., 1984: The Social Logic of Space (Cambridge). Jazwa, K.A., 2016: Building Mycenaean Identity: A Systematic Analysis of Early Helladic III to Protogeometric Domestic Architecture in Mainland Greece for Evidence of Social Groups, (diss., Florida State University). Jones, S. and P. Graves-Brown., 1996: “Introduction: Archaeology and Cultural Identity in Europe”, in Graves-Brown, P., Jones, S. & Gamble, C. (edd.), Cultural Identity and Archaeology. The Construction of European Communities (New York) 1–24. Jones, S., 1997: The Archaeology of Ethnicity. Constructing Identities in the Past and Present (New York). Kilian, K., 1988: “The Emergence of Wanax Ideology in the Mycenaean Palaces”, OJA 7.2: 291–302. Knappett, C., 2011: An Archaeology of Interaction. Network Perspectives on Material Culture and Society (Oxford). Knodell, A., 2013: Small-World Networks and Mediterranean Dynamics in the Euboean Gulf: An Archaeology of Complexity in Late Bronze Age and Early Iron Age Greece (diss., Brown University). Kus, S. and Raharijaona V., 1990: “Domestic Space and the Tenacity of Tradition among Some Betsileo of Madagascar”, in Kent, S. (ed.), Domestic Architecture and the Use of Space. An Interdisciplinary Cross-Cultural Study (Cambridge) 21–33. Lemonnier, P., 1993: Elements for an Anthropology of Technology. Anthropological Papers, Museum, of Anthropology, University of Michigan No. 88 (Ann Arbor).

166 Lemos, I., 2002: The Protogeometric Aegean: The Archaeology of the Late Eleventh and Tenth Centuries B.C. Oxford Monographs on Classical Archaeology (Oxford). Lis, B., 2009: “Handmade and Burnished Pottery in the Eastern Mediterranean at the End of the Bronze Age: Towards an Explanation for its Diversity and Geographical Distribution”, in Bachhuber, C. & Roberts, G. (edd.), Forces of Transformation: The End of the Bronze Age in the Mediterranean (Oxford) 150–61. MacSweeney, N., 2009: “Beyond Ethnicity: the Overlooked Diversity of Group Identities”, JMA 22.1: 101–26. Maran, J., 2000: “Das Megaron im Megaron. Zur Datierung und Funktion des Antenbaus im mykenischen Palast von Tiryns”, AA: 1–16. Maran, J., 2001: “Political and Religious Aspects of Architectural Change on the Upper Citadel of Tiryns: The Case of Building T”, in Laffineur, R. & Hägg, R. (edd.), POTNIA: Deities and Religion in the Aegean Bronze Age. Aegaeum 22 (Liège) 113–121. Mazarakis Ainian, A., 1997: From Rulers’ Dwellings to Temples: Architecture, Religion and Society in Early Iron Age Greece (1100–700 B.C.). SIMA 121 (Jonsered). McDonald, W.A., Coulson, W.D.E. & Rosser, J. (edd.), 1983: Dark Age and Byzantine Occupation. Nichoria 3 (Minneapolis). Morris, I., 1987: Burial and Ancient Society: The Rise of the Greek City-state (Cambridge). Murray, S., 2017: The Collapse of the Mycenaean Economy: Imports, Trade, and Institutions (Cambridge). Mylonas, G.E., 1966: Mycenae and the Mycenaean Age (Princeton). Nagle, D.M., 2015: Principles of Spatial and Social Organization in Mycenaean Architecture and Settlements (diss., Florida State University). Ortner, S.B., 1984: “Theory in Anthropology since the Sixties”, Comparative Studies in Society and History 26.1: 126–66. Palaima, T., 1995: “The Nature of the Mycenaean Wanax: Non-Indo-European Origins and Priestly Functions”, in Laffineur, R. & Niemeier, W.-D. (edd.), The Role of the Ruler in the Prehistoric Aegean. Proceedings of a Panel Discussion presented at the Annual Meeting of the Archaeological Institute of America, New Orleans, Louisiana 28 December 1992. Aegaeum 11 (Liège) 119–39. Palaima, T., 2006: “Wanaks and Related Power Terms in Mycenaean and Later Greek”, in Deger-Jalkotzy, S. & Lemos, I.S. (edd.), Ancient Greece: From the Mycenaean Palaces to the Age of Homer. Edinburgh Leventis Studies 3 (Edinburgh) 53–71. Parkinson, W.A., 2007: “Chipping Away at Mycenaean Economy: Obsidian Exchange, Linear B, and ‘Palatial Control’ in Late Bronze Age Messenia”, in Galaty, M.L. & Parkinson, W.A. (edd.), Rethinking Mycenaean Palaces II. Cotsen Institute of Archaeology at UCLA Monograph 60 (Los Angeles) 87–101.

Jazwa Popham, M.R., Calligas, P.G. & Sackett, L.H., 1993: Lefkandi II: The Protogeometric Building at Tomba. Part 2: The Excavation, Architecture, and Finds (Athens). Rapoport, A., 1990: “Systems of Activities and Systems of Settings”, in Kent, S. (ed.), Domestic Architecture and the Use of Space (Cambridge) 9–20. Relaki, M., 2004: “Constructing a Region: The Contested Landscapes of Prepalatial Mesara”, in Barrett, J.C. & Halstead, P. (edd.), The Emergence of Civilization Revisited (Oxford) 170–88. Roux, V., 2003: “A Dynamic Systems Framework for Studying Technological Change: Application to the Emergence of the Potter’s Wheel in the Southern Levant”, Journal of Archaeological Method and Theory 10.1: 1–30. Rutter, J.B., 1975: “Ceramic Evidence for Northern Intruders in Southern Greece at the Beginning of the Late Helladic IIIC Period”, AJA 79.1: 17–32. Rutter, J.B., 1990: “Some Comments on Interpreting the Dark-surfaced Handmade Burnished Pottery of the 13th and 12th Century Aegean”, JMA 3: 29–49. Sanders, D., 1990: “Behavioral Conventions and Archaeology: Methods for the Analysis of Ancient Architecture”, in Kent, S. (ed.), Domestic Architecture and the Use of Space. An Interdisciplinary Cross-cultural Study (Cambridge) 43–72. Schon, R., 2010: “Think Locally, Act Globally. Mycenaean Elites and the Late Bronze Age World-System”, in Parkinson, W.A. & Galaty, M.L. (edd.), Archaic State Interaction: The Eastern Mediterranean in the Bronze Age (Santa Fe) 213–36. Shear, I.M., 1968: Mycenaean Domestic Architecture. 3 vols. (diss., Bryn Mawr College). Shelmerdine, C.W. and J. Bennet., 2008: “12. Mycenaean States”, in Shelmerdine, C.W. (ed.), The Cambridge Companion to the Aegean Bronze Age (Cambridge) 289–308. Shelmerdine, C.W., 1997: “Review of Aegean Prehistory VI: The Palatial Bronze Age of the Southern and Central Greek Mainland”, AJA 101.3: 537–85. Shelton, K., 2010: “Ch. 10. Mainland Greece”, in Cline, E.H. (ed.), The Oxford Handbook of the Bronze Age Aegean (Oxford) 139–48. Sinos, S., 1971: Die vorklassischen Hausformen in der Ägäis (Mainz). Snodgrass, A.M., 1971: The Dark Age of Greece: An Archaeological Survey of the Eleventh to the Eighth Centuries BC (Edinburgh). Steele, L., 2013: Materiality and Consumption in the Bronze Age Mediterranean (New York). Tartaron, T.F., Pullen, D.J., Dunn, R.K., Tzortzopoulou-Gregory, L., Dill, A. & Boyce, J.I., 2011: “The Saronic Harbors Archaeological Research Project (SHARP): Investigations at Mycenaean Kalamianos, 2007–2009”, Hesperia 80.4: 559–634. Taylour, L.W., 1964: The Mycenaeans (London). Thomas, J., 1996: Time, Culture, and Identity: An Interpretive Archaeology (London).

DOMESTIC ARCHITECTURE & IDENTITY DURING THE BRONZE AGE TO IRON AGE TRANSITION Van De Moortel, A. & Zahou, E., 2011: “The Bronze Age-Iron Age Transition at Mitrou in East Lokris: Evidence for Continuity and Discontinuity”, in Mazarakis Ainian, A. (ed.), The ‘Dark Ages’ Revisited vol. 1: Acts of an International Symposium in Memory of William D.E. Coulson. University of Thessaly, Volos, 14–17 June 2007 (Volos) 313–29. Van De Moortel, A., 2007: “The Site of Mitrou and East Lokris in Homeric Times”, in Morris, S. & Laffineur, R. (edd.), EPOS. Reconsidering Greek Epic and Aegean Bronze Age Archaeology. Proceedings of the 11th International Aegean Conference, Los Angeles 2006 (Liège) 243–54. Van De Moortel, A., 2009: “The Late Helladic III C— Protogeometric Transition at Mitrou, East Lokris”, in Deger-Jalkotzy, S. & Bächle, A.E. (edd.), LH III C Chronology and Synchronisms III: LH III C Late and the Transition to the Early Iron Age. Proceedings of the international workshop held at the Austrian Academy of Sciences at Vienna, February 23rd and 24th, 2007 (Vienna) 369–72. Van Den Berg, K.A.M., 2018: Keeping in touch in a changing world. Network dynamics and the connections between the

167

Aegean and Italy during the Bronze Age–Iron Age transition (ca. 1250–1000 BC) (diss., Vrije Universiteit Amsterdam). Wenger, E., 1998: Communities of Practice: Learning, Meaning and Identity (Cambridge). Westgate, R., 2015: “Space and Social Complexity in Greece from the Early Iron Age to the Classical Period”, Hesperia 84.1: 47–95. Whiting, J.W.M. & Ayres, B., 1968: “Inferences from the Shape of Dwellings”, in Chang, K.C. (ed.), Settlement Archaeology (Palo Alto) 117–33. Whitley, J., 1991: Style and Society in Dark Age Greece. The Changing Face of a Pre-literate Society 1100–700 BC (Cambridge). Wiersma, C., 2014: Building the Bronze Age: Architectural and Social Change on the Greek Mainland during Early Helladic III, Middle Helladic and Late Helladic I (Oxford). Winter, F.A., 1977: “An Historically Derived Model for the Dorian Invasion”, in Davis, E. (ed.), Symposium on the Dark Ages in Greece (New York) 60–76. Yakar, J., 2000: Ethnoarchaeology of Anatolia: Rural SocioEconomy in the Bronze Age and Iron Ages (Tel Aviv).

chapter 10

Greek Temple Building from an Economic Perspective: Case Studies from the Western Peloponnesos András Patay-Horváth 1 Introduction Greek temples were neither indispensable nor really necessary for the cult of the gods. Although they are clear manifestations of Greek architecture and religion (most of them can be regarded as monumental votive offerings or gifts to the gods: see most recently Wilson Jones 2014, but also Burkert 1988; 1996; Fehr 1996), their construction was not primarily a religious phenomenon, nor simply a purely artistic phenomenon—i.e., a series of technical procedures to create an aesthetically pleasing but practically useless object to be dedicated in a sanctuary. Temple constructions were communal projects par excellence and—given the large scale of material and human resources involved and also the considerable time needed to complete them—they depended on various political and socio-economic factors. Their analysis cannot, therefore, be reduced to the architectural and cultic perspective but must include the consideration of historical and economic aspects as well. But which community commissioned or built a certain temple, why the decision was made to construct one, and how it was financed have not usually been investigated in detail, unless ancient sources provide some explicit information for these questions. Even in these cases, it has only sporadically been recognised that the available written sources are often demonstrably misleading or inaccurate. Based on Pausanias 5.10.2, for example, it is usually assumed as a matter of fact that the temple of Zeus at Olympia was built by Elis. R. Taraporewalla (2011) and H. Kyrieleis (2012/2013) discuss at length why Elis could have built and decorated the temple of Zeus at Olympia, but they do not seem to consider whether Pausanias was right in attributing the entire project to Elis. A genuine debate concerning the reliability of relevant literary texts has evolved only around the Parthenon (Ameling 1985; Giovannini 1997). More importantly, it has not been realised that, even if there is no surviving written evidence about the commissioning community or the occasion of the temple building, it is still possible to attempt an answer to these basic questions (Woodward 2012 seems to be the only one to attempt this) because the size, materials,

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004416659_012

and decoration of the buildings are in most cases clearly discernible and provide a basis for calculating the approximate costs of the construction (Table 10.1)—in the words of J.J. Coulton (1974: 74), “it was upon the size that the cost would largely depend”. Based on this and other pieces of relevant archaeological or architectural information, in certain cases one can tentatively reconstruct the historical circumstances of a temple building. This method will be demonstrated on a set of monuments, the Doric peripteral temples of Olympia and neighbouring Triphylia (Fig. 10.1), which can be regarded as a representative sample for several reasons: temples in this region are heterogenous in their date (sixth to fourth century BCE), size (including the largest and the smallest ones of the Peloponnesos), decoration (some were lavishly adorned with architectural sculpture, others were not) and setting—i.e., they are located either in remote rural areas, in polis centres, or in the panhellenic sanctuary of Olympia. Some of them are relatively well-documented in ancient sources, while others are barely mentioned. Admittedly, the level of their preservation and publication is uneven as well, but this is equally true for other regions in Greece (on the temples of Triphylia in general: Lippolis, Livadiotti & Rocco 2007: 650, 661–64; Xen. An. 5.3.4–11). 2

Olympia, Temple of Zeus

The temple of Zeus at Olympia is a very well-documented case and can clearly demonstrate both the difficulties and the possibilities of this approach at the same time. Pausanias (5.10.2) tells explicitly when and by whom the temple was built and implies the financial resource for the project—i.e., the booty taken by Elis from defeated Pisa. Some scholars have expressed doubts and made short comments to this effect, notably Herrmann (1972: 128), Gauer (1993: 178), Philipp (1994: 90), Nafissi (2001: 302), Roy (2010: 297), Baitinger (2011: 149), and Hennemeyer (2012: 121). This narrative has usually been accepted at face value, although it is demonstrably a local myth and has nothing to do with real history. If we accept the ancient testimonia—including Pausanias (6.22.3–4) on the date

169

GREEK TEMPLE BUILDING FROM AN ECONOMIC PERSPECTIVE: THE WESTERN Peloponnesos table 10.1 Basic measurements and cost estimates of the temples in Fig. 10.1

Temple

Date BC

Stylobate Approx. column Special features dimensions (m) height (m)

Epidauros, Asklepios Olympia, Zeus Bassai, Apollon Lepreon, Demeter Olympia, Metroön Mákiston, Athena Prasidháki, Athena Alipheira, Athena

380–370 475–455 430–400 ca. 400 ca. 400 ca. 500 ca. 450 500–480

11.76 × 23.06 27.68 × 64.12 14.47 × 38.24 10.45 × 20.23 10.62 × 20.67 14.18 × 32.94 14.7 × 33.3 10.37 × 29.3

figure 10.1

5.5 10 6 5 5 6 6 5

Cost est. (Attic talents)

Pentelic marble pediments and akroteria  23 Parian marble roof, metopes, pediments 300–400 marble frieze, metopes, roof, coffers, capitals 45–50 no interior colonnade 8–10 —— 10–12 marble pediments added ca. 400 BC 30–40 —— 30–40 no interior colonnade 15–20

Map of Doric peripteral temples in the Western Peloponnesos

of this war between Pisa and Elis—as historically correct, the war must have occurred during the first half of the sixth century and thus it would antedate the temple construction by ca. 100 years. It is, however, almost impossible to suppose that Elis safeguarded the booty for a century and waited such a long time to celebrate the victory. The connection between this war and the construction of the temple is therefore obviously a mistake or mere fiction. In addition, not only the war but even

the early existence of a political entity named Pisa has been convincingly questioned recently, and so the booty from such a war may be ruled out entirely (Möller 2004; Luraghi 2008: 79; Gianguilio 2009). It has also been suggested that the war mentioned by Pausanias was identical with the so-called Triphylian war, mentioned in passing by Herodotos (4.148) and thought to have occurred at the time of the synoikism reported by Diodoros (11.54). Actually, the phrasing is not conclusive enough to suggest

170 a single war or any connection with the synoikism. Even if this were the case, the temple construction cannot have resulted from the hypothetical booty taken from these cities, simply because it would surely have been insufficient for financing the temple alone or the temple and the chryselephantine statue, as stated by Pausanias. Since the arguments are fully published (Patay-Horváth 2012; 2014; 2015), here I only give a short summary of these problems. The costs of building a Doric temple are attested in some cases both in the literary and epigraphic record, and therefore it is possible to make an estimate of the cost of comparable monuments. Herodotos (2.180) gives an estimate of 300 talents for the late archaic temple of Apollon at Delphoi. The temple of Asklepios at Epidauros cost, according to the preserved building accounts, ca. 16 Aeginetan (equivalent to ca. 23 Attic) talents during the first quarter of the fourth century BCE (Burford 1969; Prignitz 2014). Based on this evidence, one can approximately calculate the costs of other Doric temples as done, e.g., by Stanier (1953) for the Parthenon and by Martin (1972: 189) for the temples of Poseidonia, Selinous, and Akragas (note that De Angelis 2003: 163–69 arrives at slightly different results concerning the temples of Selinous). The calculation of the stone parts is based on the dimensions of the stylobate and the approximate column height. The volume of the resulting cuboid is taken as an indicator of the size of the temples, the costs of which can be thus set in proportion to each other without calculating the exact volume and costs of every single constructional part. This simplified calculation method (seemingly used by Burford [1965: 25] for the Hephaisteion and other temples as well) was tested on the temple of Hera II at Poseidonia and yielded approximately the same result (206 talents) as the more complicated one used by Martin (1972: 189, calculating 228 talents) and can thus be assumed to work grosso modo on every other Doric peripteral temple made of limestone with two interior colonnades. The temple of Zeus is thus ca. 12 times bigger than the temple of Asklepios at Epidauros and its limestone parts would therefore cost around 100 talents, and the entire building at least ca. 300. Since there are many unknown variables affecting the overall expense, it is estimated at three times that of the stone elements (Martin 1972: 189). The cost of the marble roof and sculptural decoration must have added considerably to this sum, and estimates for the quarrying, working, and transporting the marble components from Paros to Olympia have already been detailed (Younger & Rehak 2009: 50–51). One can therefore confidently state that the total expense of construction for this temple would have been within the range of 300–400 talents. The cost of the chryselephantine statue cannot be estimated but must

Patay-Horváth

have been also considerable. Comparing the huge amount of money needed to build the temple with figures indicating the financial capacities of contemporary Elis (e.g., Xen. Hell. 3.2.30; 6.5.19) is revealing, and this simple observation strongly suggests that this polis was not likely to be able to finance the whole project on its own. On the other hand, the suggestion made by Pausanias that the construction was financed from war booty is more plausible, and it is only the war which he seems to have mistaken. The pieces of information concerning the scale of the plunder derived from individual battles or military campaigns in the Greek world are sufficient to conclude that the booty from which this temple was financed must have been exceptional, implying a very prosperous enemy. Since a tenth was usually offered to the gods, the costs of the temple therefore imply a booty well above one thousand if not several thousands of talents (cf. Pritchett 1991: 505–41, on the size of booties). Booties above 100 talents were rarely realised in wars among Greeks, and Polybios 2.62 explicitly tells that even a prosperous city like Maniteneia cannot be expected to have yielded more than 300 talents. Booties taken from Persia, on the other hand, amount to several thousand talents, even if the figures given by the sources are exaggerated. The date of the temple construction is sufficiently secure (ca. 475–455 BCE; [Dinsmoor 1950: 151]). Accordingly, it is evident that the war from which the necessary booty derived can only be sought in the great campaign of Xerxes, and especially in the battle of Plataiai. This conclusion is admittedly hypothetical, but is actually consistent with all the available written, archaeological, and numismatic evidence and seems to be far more probable than any alternative (see Patay-Horváth 2015 for more details, including the numismatic evidence). 3

Bassai, Temple of Apollon Epikourios

The second case study concerns the temple of Apollon Epikourios at Bassai, where our starting point and source of information is again Pausanias (8.41.7–9). He connected the temple building with the great plague at the beginning of the Peloponnesian war and also reported as a fact that the Phigaleians employed Iktinos, the architect of the Parthenon, for this project. The involvement of the Athenian (or Elean, according to Dinsmoor 1950: 154) architect Iktinos is often considered as likely (e.g., Cooper 1996: 369–79), but there are strong arguments against accepting this piece of information. Korres (2001: 342–43) assumes—seemingly a priori—that Iktinos was the architect and simply discusses which parts of the

GREEK TEMPLE BUILDING FROM AN ECONOMIC PERSPECTIVE: THE WESTERN Peloponnesos

building could be attributed to him, but he clearly acknowledges the inherent problems of this exercise. In fact, certainty cannot be reached and there is no way to prove or disprove whether Iktinos was involved or not, but this question is not of primary importance for the present subject—i.e., the historical setting of the temple construction. The connection with the plague, however, is demonstrably erroneous. (Eckstein 1960; Roux 1961: 55– 56; Knell 1968; Svenson-Evers 1996: 160–66). Thoukydides (2.54.5) explicitly informs us that the plague did not reach the Peloponnesos, and the sanctuary does not show any traces of Apollon as a healing deity (Jost 1985: 486–88). It is therefore hardly surprising that modern scholars have offered alternative explanations. The most elaborate reconstruction was put forward by F.A. Cooper (1978: 10–28; 1996: 75), who argued that it was not the city of Phigaleia, but soldiers from Arkadia in general who were saved from the plague while fighting as mercenaries for Athenai. Admittedly, the archaeological remains (especially the miniature weapons) show that the cult and the epithet of Apollon were surely connected to his warlike character, but the connection with Arkadian mercenaries supposed by Cooper (but not stated by Pausanias) is not evident linguistically and rather improbable historically (Svenson-Evers 1996: 164–65). The military connotations, as evidenced by the practice of dedicating miniature armour in the sanctuary, are commonly accepted but occasionally questioned (e.g., Baitinger 2011, 160 with additional references). The alternative interpretation vaguely connects them with some kind of initiation, which is not entirely satisfactory. Cooper postulated that the temple was built by Arkadian mercenaries during periods when Spartan supremacy is not attested (429–421; 415–401 BCE), and he assumed that work was interrupted during those years when Phigaleia was, according to his interpretation, occupied by Sparta. The entire monument is thus interpreted as a manifestation of anti-Spartan resistance. This reconstruction of events is not contradicted by archaeological and architectural evidence but relies heavily on late sources that are not entirely compatible with the testimony of Thoukydides.1 1  To explain, it is hardly an accident that Phigaleia was never mentioned by Thoukydides as a target of Spartan aggression during the years 421–418 BCE, which he documents fairly thoroughly. It is very risky to assert that Phigaleia was captured because such an episode in Polyainos (6.27.2) can tentatively be connected to this period, and because Thoukydides reports some military activities in the neighbourhood. Even if the story related by Polyainos is historical, it is not certain that every detail is reliable and, therefore, the date is dubious. The silence of Thoukydides on the matter most probably

171

More importantly, it is left unexplained how mercenaries from various places in Arkadia could have afforded the costs of temple construction, which must have been considerable to judge by its relatively large size and lavish decoration. Since the temple of Apollon was lavishly decorated with marble sculpture, like the one at Epidauros, the building costs can be estimated in proportion according to the stylobate and column dimensions. The classical temple of Apollon Epikourios must have cost at least around 45–50 talents, and, considering the difficult terrain, transportation expenses are likely to have been higher (above, Table 10.1). In fact, Cooper himself listed the evidence for Arkadian soldiers in the Archaic and Classical periods, but there is nothing in this material which would show that any of them had dedicated anything in common. Indeed, it is inconceivable that poor mercenaries of disparate origins could have initiated, afforded, and organised the construction of any temple, let alone such a remarkable one as at Bassai. The temple of Artemis Ephesia at Skillous, dedicated by Xenophon and built according to himself (An. 5.3.4–7) from the tithe of a mercenary expedition, is most probably not an exception either, since the story given by Xenophon can be doubted with good reasons as far as it concerns the funding of the temple (Tuplin 2004). Furthermore, if one follows this line of reasoning, it is quite puzzling that after the collapse of Spartan rule around the middle of the fourth century, the cult did not begin to flourish. Instead, a huge statue of Apollon Epikourios was transferred to Megalopolis (Paus. 8.30.3–4), and the number of dedications dropped drastically and shortly afterwards entirely disappeared from the sanctuary (Cooper 1978: 69; 1996: 80: “the temple and the twelvefoot bronze Apollon Epikourios were among the last to be dedicated”). Admittedly, the date of the bronze statue is totally unknown, similarly to the date of its transfer to Megalopolis. Cooper undermines his own conclusion in indicates that it cannot have been an important action. The proliferation of similar stories would rather suggest that it is fictional (Pol. 1.20.1; 3.9.52; 5.44.5; 7.11.6; see Krentz & Wheeler 1994). Among the five similar stories reported by Polyainos, only two are attested in other sources (3.9.52 at Frontinus Str. 2.1.6; 5.44.5 at Diodoros Sikeliotes 17.7). The others, including the episode concerning Phigaleia, are otherwise unattested. If they are historical, they are likely to be dated to the fourth century (or even to later periods), but in the case of Phigaleia it is equally possible that the episode was somehow connected with the pseudo-historical narratives concerning the Messenian wars (see Parker 1991; Luraghi 2008). The circumstances of the compilation (Polyainos produced six books in no more than nine months during 161–62 CE) are also a strong warning against attaching much weight to this episode or its details. The passing mention in a late rhetorical exercise (Pseudo-Herodes, Peri politeias 28–29) is even more unreliable as a historical source (see Albini 1968).

172 another passage (1996: 66), where he dates the disappearance of dedications to the second half of the fourth century. However, it seems to be true that the cult of Apollon Epikourios declined markedly during the fourth century BCE and recovered only during Hellenistic and Roman times. Contrasting the rich archaic material already discovered at the beginning of the 20th century, there are very few ceramic or other finds from the fourth and third centuries BCE even from more recent excavations. Cooper (1996: 66), referring to the excavations conducted by Yalouris, states: “During the second half of the fourth century dedications dropped off sharply”. Yalouris (1965: 155) gives just a brief summary on the finds—mentioning fragments of Megarian bowls and Elean ceramics dating from the Classical-Hellenistic period—and reproduces (on pl. 135) only a selection of them. Higher precision in dating the small fragments is perhaps not to be expected, but it can probably be concluded that the ceramic evidence for the fourth and third centuries BCE is meager compared to the Archaic finds. Inscriptions are more clearly datable. After the bronze manumission tablet (dated by Cooper [1978: 29–44, 173–81] to the early fourth century) the next epigraphic document dates from ca. 240 BCE (IG v.ii.419; Cooper 1996: 387–88). In sharp contrast to this chronological gap in Bassai, numerous votives and inscriptions dating from the fourth century BCE were discovered in the central temple of Zeus and Athena at Phigaleia (Arapogianni 1996: 44–45). Comparing buildings from the surrounding area, the unusually large size of the Bassai temple is apparent: it is considerably larger than the Metroön at Olympia or the temple of Demeter at Lepreon (both slightly later in date) and even surpasses the temples at Prasidháki and Mákiston (above, Table 10.1) (both from the first half of the fifth century BCΕ; for measurements see Lippolis, Livadiotti & Rocco 2007: 653, 661–62, 669). The construction of a costly and unusually large temple and the apparent abandoning of the sanctuary soon after its completion makes much more sense if we connect it with people who were present (or prosperous, or inclined to worship the deity) only during the period of its construction and usage. I find this to be much more realistic than the explanation given by Cooper (1978: 44): “Perhaps in a gesture of patriotism, the Arkadian mercenaries may have moved the centre of their cult to the new capital, Megalopolis”. Supposing that the polis of Phigaleia decided and completed the temple construction, it would be quite surprising for the same community to abandon (or at least to neglect) the cult place a few decades afterwards. Some kind of major political reorientation could have effected

Patay-Horváth

such a significant change in the cult practice. Diodoros (15.40) in fact reports such a reform—the expulsion of the pro-Spartan regime of Phigaleia in 375 BCE—but even in this case the reason for starting the temple building would remain rather obscure. The date of the episode is controversial: whereas Diodoros places it in 375/374, this date was variously rejected and is defended only by Roy (1973) and Stylianou (1998, 330–32), with different arguments. Therefore, it is well worth asking if there were any other community besides the Phigaleians and the loosely defined Arkadian mercenaries which could have commissioned or built the temple. I think our sources can provide us with a reasonable suggestion, but in order to attempt an answer, the date of the temple construction has to be discussed briefly. The architectural and archaeological evidence does not provide incontrovertible evidence for dating. In particular, the pottery found in an ancient fill packed against the west side of the temple foundation seems to date the initial construction phase to the period between 425–420 BCE, and there are no compelling reasons for an earlier building phase starting at about the middle of the century, even if this has been assumed quite often. Cooper (1996: 163) takes the pottery as a secure terminus post quem, although its validity is not beyond doubt. An earlier date for the beginning of the temple building around 450 BCE was first and foremost advocated by Dinsmoor (1950: 154–57) and accepted with slight modification to ca. 440 BCE by Svenson-Evers (1996: 207–10). This early date is based on the comparison of column and capital profiles with those of the Athenian Hephaisteion and the Parthenon, but such a comparison is also not decisive because the proportions of Attic columns are not necessarily relevant for a building in the Peloponnesos, which is clearly following local Peloponnesian traditions and has no certain connections with the architect of the Parthenon. There is only one feature connecting the building with Athenian temples—the depth of the ptera before the pronaos and opisthodomos—which offers a terminus post quem (or ante quem non) of 440/430 BCE (Mallwitz 1975: 40). With the date for the completion placed uncontroversially around the end of the fifth or beginning of the fourth century BCE (Svenson-Evers 1996: 206–07), the temple would have been built during a period in which Spartan influence was dominant, and dedications in the sanctuary ceased at approximately the time when Spartan hegemony collapsed. It is then a reasonable inference, I think, that the temple construction was most probably due to a group of people dependent upon Sparta, first of all the neodamodeis, who could build the temple as a thank-offering for their newly acquired status and relative

GREEK TEMPLE BUILDING FROM AN ECONOMIC PERSPECTIVE: THE WESTERN Peloponnesos

prosperity. The cult of a warlike Apollon is characteristic for Sparta (even if it was by no means restricted to it), as shown by the Karneia (Ath. 4.141e–f) and by the statues of Apollon at Amyklai and at Thornax (Paus. 3.19.1–2 and 3.10.8). Fields (1994) discusses this topic at length while also accepting the connection of the Bassai temple with Arkadian mercenaries. Even Cooper (1968: 111 n. 71) envisages close connections between Apollon Epikourios at Bassai and Apollon Karneios at Sparta, based on similarities in the cult practice. As Thoukydides (5.34.1) reports, Lakedaimonian neodamodeis (helots freed after military service: Cartledge 2002: 215) were settled at Lepreon in 421 BCE and, even if their number is not exactly known, there must have been several hundred who would also have been accompanied by a similarly large contingent of Brasidaioi. At this point no numbers are given; in another passage (Thouk. 5.49.1) one thousand hoplites are implied. The exact number of the Brasidaioi is not quite clear: fewer than 700 according to Thoukydides 4.80.5; ca. 1000 according to Diodoros Sikeliotes (12.76.1). Welwei (1974: 146 n. 27) thus calculates ca. 500 Brasideioi and 500 neodamodeis; Falkner (1999: 391), however, gives 1000 neodamodeis. Paradiso & Roy (2008) argue for sending 1000 hoplites only on a single occasion, referred to at three different points in Thoukydides (5.31.4; 34.1; 49.1); Hornblower (2008: 126) is inclined to accept three successive Spartan mobilisations. At any rate, the reason for settling these former soldiers in Triphylia is explicitly stated by Thoukydides (5.34.1): Sparta was at this time hostile toward Elis. Thus, the new settlers served to reinforce and secure the border. Their loyalty to Lakedaimon was beyond doubt, since they were rewarded with freedom and with sufficient landholdings to support themselves. This does not apply, perhaps, to the Brasideioi, but most certainly to the neodamodeis (Cartledge 2002: 215), who had been probably stationed in this region already for a few years prior to 421 BCE (Thouk. 5.31.4). But even if they were sent to Lepreon only in 421 BCE along with the Brasideioi, they cannot have been garrisoned as mercenaries there for a long period, because this would have involved large amounts of payments or supplying provisions for them (cf. Ruzé & Christien 2007: 235). Moreover, an active military confrontation was most probably not expected, but rather securing the border against hostile forces over an extended period of time. Afterwards, the neodamodeis actively took part in the battle of Mantineia in 418 BCE, and their numbers grew steadily during the following decades (Welwei 1974: 149– 55; Ducat 1990: 160; and Ruzé & Christien 2007: 255); they are mentioned for the last time in 369 BCE in the battle of Oion (Xen. Hell. 6.5.24). Accordingly, they appear to have

173

fulfilled the expectations attached to them. They were in possession of the surrounding region only during the period of construction, and after the eclipse of Spartan hegemony, they disappeared from the area. That they were expelled is strongly suggested by Diodoros Sikeliotes (15.40.2–3): the text mentions only that aristocrats inclined towards Sparta were expelled by democrats, but both groups can be seen as the descendants of the neodamodeis who were settled here some fifty years earlier. Or perhaps they simply left, and afterwards the sanctuary fell out of use for a considerable time as indicated by the lack of finds already discussed above. Spartan interest in the region as a secure base against an unreliable Elis is further demonstrated by the 1000 hoplites sent to Lepreon in 420 BCE who attacked a fortress named Phyrkos near Lepreon. This led to a serious conflict between Elis and Sparta, which even tolerated exclusion from the Olympic Games without abandoning this position (Thouk. 5.49–50). As noted above, Paradiso & Roy (2008) consider this movement to be identical with the other one described by Thoukydides (5.34.1; cf. the discussion by Roy 1998; and Paradiso & Roy 2008 on the location of the fort Phyrkos). That it was not only the neodamodeis at Lepreon who were supporting Spartan interests in Triphylia is clearly shown by the settlement of Xenophon at Skillous. His famous account emphasising that his dedication resulted from the booty taken during the expedition of the 10,000 is not necessarily correct concerning the funding of the temple, and it should by no means obscure the fact that the temple construction was only made possible by the Spartan grant of an estate. Other settlers receiving smaller plots of land were presumably similarly eager to express their gratitude to the gods by erecting altars and/ or temples. This attitude is amply demonstrated by Greek colonists almost everywhere in Sicily and Magna Graecia. Malkin (1987: 135–86) surveys both the literary and the archaeological evidence concerning the foundation of sanctuaries in Greek colonies. As most colonies were founded before the invention of Greek peripteral temples, the colonists usually dedicated just a precinct and an altar to the gods. Later on, during the sixth century BCE, however, a temple was also erected almost immediately after the foundation, as implied by the story of Phalaris supervising the construction of a temple for Zeus (Polyainos 5.1.1). And since there were several hundreds of new colonists in this region subsequently known as Triphylia, it is tempting to connect them with the lavish temple building at Bassai. That the neodamodeis were according to the sources settled in Lepreon does not necessarily mean that they were far away from the sanctuary. To the contrary, the territory

174

Patay-Horváth

of Phigaleia might have belonged to Lepreon in this period and in this way, the landholdings of the newcomers might actually have been quite close to the sanctuary. This Arkadian polis is attested already at the beginning of the fifth century BCE (Hdt. 6.83), but afterwards the next recorded episode of its history dates from ca. 375 BCE (Diod. Sik. 15.40). It was assumed (Cooper 1996: 51–55) that, because the grave of the eponymous hero Lepreos was found in Phigaleia (Paus. 5.5.4), the two cities actually originated from an archaic unity “Lepreon-Phigalia”. However, it is equally possible or perhaps even more likely that the grave of Lepreos—which was, according to Pausanias, not acknowledged by the Phigalians—was only used by Lepreon to justify the conquest of this territory, an event which might have happened shortly after the Persian wars (for more, see Patay-Horváth 2016). One can safely conclude that the classical temple of Bassai was constructed by a sufficiently large group of people who had both an appropriate economic basis and a shared motivation to construct such a large and costly temple. This cannot have been otherwise, and because our only ancient source does not provide reliable information on who the temple builders were and why they decided to build it, an attempt has been made here to suggest a more plausible scenario, even if it is based in part on circumstantial evidence and conjecture. It is argued that the temple builders of Apollon Epikourios were not local residents who, having accumulated the necessary means for the temple during a relatively long period of time, had simply decided to build a new temple in order to replace an older, presumably ruinous structure. Instead, the builders were newcomers who had settled in the neighbouring area and thus acquired the necessary means (not in the form of plunder but in landholdings) in a short period of time, and who undertook and completed the construction of this relatively large temple in order to assert their dominance in this region. According to this hypothesis (Patay-Horváth 2018a), the settlement occurred during the Peloponnesian war, at a moment when this war seemed to have been concluded in such a way that they could expect to remain there permanently. 4

Temples in Triphylia and the Metroön at Olympia

The smaller temples should be investigated in a similar way to see if their measurements and construction costs can reveal anything about the historical circumstances and underlying motivations of those who erected them. There were at least two small and two medium-size peripteral

temples in this region, and we have no explicit written evidence concerning the circumstances of their construction. (The temple of Artemis built by Xenophon in Skillous, where the sponsor himself provided detailed information on the building project, may be an exception, but the temple is unfortunately not located and it is not certain that it was peripteral.) Their geographical setting and chronology is, however, quite clear, and their measurements are so similar to each other that they can be regarded as “twin” buildings (above, Table 10.1). Temples with similar overall design (i.e., similar plan dimensions and elevation) erected in close proximity to each other such as the “Theseion” and the temple of Poseidon in Sounion, or the temples of Victory at Himera and of Athena at Syrakousai, were traditionally ascribed to the same architect. Woodward (2012: 295), on the other hand, concluded that “whilst there is no indication that these temples shared an architect, there is evidence to suggest that these temples had the same client”. As the discussion of the Metroön and the temple of Demeter at Lepreon (which is not investigated in detail by Woodward) will show, this is not necessarily true either, and the two alternatives are of course not mutually exclusive. Similarities may have resulted from the intentions of the same client naturally employing the same architect and relying on the same workshop tradition, but different clients who were in direct rivalry could compete for the same workmen. First, the temple at Prasidháki (to the south of Lepreon; see Arapogianni 2002a and 2002b) and the temple of Athena in Mákiston (in the northern part of Triphylia (Nakassis 2004, with a detailed description of the similarities between the two buildings and their relative chronology) both date from the late Archaic or early Classical period and are surprisingly similar in their dimensions: Nakassis (2004: 227) was certainly right in labeling the two temples at Mákiston and Prasidháki as “twin” buildings. They were built of limestone with an interior colonnade similar to the temples discussed up to this point, so their construction costs can be estimated to reach ca. 30–40 talents (above, Table 10.1). The Metroön in Olympia and the temple of Demeter at Lepreon, on the other hand, were both among the smallest peripteral temples on the Peloponnesos and did not have interior colonnades or architectural sculpture. Their construction costs cannot therefore have exceeded 10 talents altogether. The elevations are not well preserved, but a number of key measurements indicate that the temples were almost identical and both built around 400 BCE (Knell 1983; Woodward 2012: 254: “Although the capitals belong to two different groups, the indication that the Temple of Demeter at Lepreon had 11 flank columns with lower diameters of 0.83 m, whilst the Metroön also had 11 flank columns which measured 0.85 m on the

GREEK TEMPLE BUILDING FROM AN ECONOMIC PERSPECTIVE: THE WESTERN Peloponnesos

lower diameter, suggests that their overall appearance was very similar”). The historical situation of Lepreon and Olympia is clear around 400 BCE. Lepreon was conquered by Elis during the final years of the Peloponnesian war, but Sparta liberated it during the war against Elis in 400/399 BCE, strongly suggesting that Lepreon built the temple to celebrate its newly acquired independence from Elis. Although Elis was decisively defeated, Olympia was left under Elean control, and we can safely assume that the Metroön, with its slightly larger dimensions, was built by the Elians shortly afterwards to demonstrate that the sanctuary remained firmly under their jurisdiction. The two temples thus would have visually represented the long rivalry and the contemporary political situation of the two poleis (Patay-Horváth 2013a). We cannot determine whether Prasidháki and Mákiston were also rivals or enemies during the first half of the fifth century, and it is also a matter of debate if Prasidháki was an independent polis or belonged in some fashion to Lepreon; the latter suggestion, however, seems to be far more probable (Nielsen 2004: 542–43). What we do know is that both buildings were erected outside the city centre during a time—roughly the first half of the fifth century BCE—when Elis was aggressively expanding in the region (Hdt. 4.148). It was, therefore, correct to assume that Mákiston was conquered by Elis at some point during the early fifth century, but the next assumption that this conquest occurred sometime after the completion of the temple is not necessarily true. The additional hypothesis (Nakassis 2004: 227) that afterwards, the defeated “Triphylians”, who partly retreated to the south, would have built yet another temple at Prasidháki—in the free part of Triphylia—to replace their old one which had come under Elean control is not particularly plausible either, because a substantial temple like that at Prasidháki is unlikely to have been built by a defeated community retreating from enemy attacks. It is more probable, as in the case of the temples built by the Greek colonists overseas (and possibly in Bassai as well), that both temples were built by a successful group—in this case the victorious Eleans, who had conquered these territories, sent colonists there, and offered a due share of their newly acquired wealth to the gods. The reason for erecting both temples and their similarities could be explained in this way as a competition between various groups of Elean settlers, who could naturally rely on similar financial resources and obviously used workmen and architect(s) from the same region and tradition. At the same time, this hypothesis would accord with the information provided by Thoukydides (5.49.1), that around 420 BCE, there was

175

an Elean stronghold to the south of Lepreon, i.e., in the immediate vicinity of Prasidháki (Paradiso & Roy 2008: 30). 5 Conclusion In sum, during the Classical period both the larger and the smaller peripteral temples in this region seem to be related to some military conflict. The temple of Zeus at Olympia was most probably erected by the victorious Greeks from the booty taken from the Persians at Plataiai, and the temple of Apollon Epikourios by the neodamodeis settled by Sparta on the border of Elis after the peace of Nikias. Later on, Xenophon certainly acted in a similar way, when he was settled at Skillous and built a temple for Artemis there. In the aftermath of the Spartan-Elean war in 400–399 BCE, the temple of Demeter was erected to demonstrate the newly acquired independence of Lepreon, and, in the Altis, the Metroön celebrated at the same time the diplomatic success of Elis—however modest this actually was compared to the military defeat this polis suffered from Sparta in this war. The temples at Mákiston and Prasidháki, on the other hand, resulted from veritable Elean victories and conquests in Triphylia during the first half of the fifth century BCE and were presumably erected by successful Elean settlers who established themselves in this region. Out of the eight peripteral temples known from the Archaic and Classical periods in the region, I have discussed six here. These seem to have been erected as a result of some military conflict, either a local or a supraregional one. A similar case may be made for the Heraion at Olympia (Patay-Horváth 2013b). There is certainly no positive evidence which would point to some different context, either in the cases discussed here, or in the remaining, poorly documented temple of Alipheira. One cannot, therefore, escape the conclusion that at least in this particular region “without wars, few of the temples of ancient Greece would have been built” (Pritchett 1971: 100; also see Burkert 1996: 25). Some may of course doubt whether this was true for all the temples discussed here, and there were certainly other motivations and other financial resources which eventually might have led to the construction of Greek peripteral temples in this region and elsewhere. However, I think this explanation should be considered seriously in other cases as well (Patay-Horváth 2018b), since it seems to fit the available and quite diverse evidence (literary, historical, numismatic, archaeological and architectural) very well, irrespective of the widely varying size, date, and geographical setting of the temples involved.

176

Patay-Horváth

List of References Albini, U. (ed.), 1968: [Erode Attico] ΠΕΡΙ ΠΟΛΙΤΕΙΑΣ. Introduzione, testo critico e commento (Florence). Ameling, W., 1985: “Plutarch, Perikles 12–14”, Historia 34: 47–63. Arapogianne, X., 1996: “ΦΙΓΑΛΕΙΑ”, Ergon 43: 41–47. Arapogianne, X., 2002a: “Neue archäologische Entdeckungen in der weiteren Umgebung von Olympia”, in Kyrieleis, H. (ed.), Olympia 1875–2000. 125 Jahre deutsche Ausgrabungen (Mainz) 317–29. Arapogianne, X., 2002b: “Doric Temple of Athena at Prasidaki”, in Stamatopoulou, M. & Yeroulanou, M. (edd.), Excavating Classical Culture. Recent Archaeological Discoveries in Greece (Oxford) 225–28. Baitinger, H., 2011: Waffenweihungen in griechischen Heiligtümern (Mainz). Burford, A., 1965: “The Economics of Greek Temple Building”, PCPS 191: 21–34. Burford, A., 1969: The Greek Temple Builders at Epidauros: A Social and Economic Study of Building in the Asklepian Sanctuary during the Fourth and Early Third Centuries B.C. (Liverpool). Burkert, W., 1988: “The Meaning and Function of the Temple in Classical Greece”, in Fox, M.V. (ed.), Temple in Society (Winona Lake) 27–47. Burkert, W., 1996: “Greek Temple-builders: Who, Where and Why?”, in Hägg, R. (ed.), The Role of Religion in the Early Greek Polis (Stockholm) 21–29. Cartledge, P., 2002: Sparta and Lakonia (London). Cooper, F.A., 1968: “The Temple of Apollo at Bassae. Observations on its Plan and Orientation”, AJA 72.2: 103–11. Cooper, F.A., 1978: The Temple of Apollo at Bassai (Ph.D. diss., University of Michigan). Cooper, F.A., 1996: The Temple of Apollo Bassitas I: The Architecture (Princeton). Coulton, J.J., 1977: Ancient Greek Architects at Work: Problems of Structure and Design (Ithaca). De Angelis, F., 2003: Megara Hyblaia and Selinous: The development of two Greek city-states in archaic Sicily (Oxford). Dinsmoor, W.B., 1950: The Architecture of Ancient Greece (London). Ducat, J., 1990: Les Hilotes. BCH Suppl. 20 (Athens). Eckstein, F., 1960: “Iktinos, der Baumeister des Apollontempels von Phigalia-Bassai”, in Eckstein, F. (ed.), Theoria. Festschrift für W.H. Schuchhardt (Baden-Baden) 55–62. Falkner, C., 1999: “Sparta and Lepreon in the Archidamian War (Thuc. 5.31,2–5)”, Historia 48: 385–94. Fehr, B., 1996: “The Greek Temple in the Early Archaic Period. Meaning, Use and Social context”, Hephaistos 14: 165–91. Fields, N., 1994: “Apollo: God of War and Protector of Mercenaries”, in Sheedy, K.A. (ed.), Archaeology in the Peloponnese: New Excavations and Research (Oxford) 95–113.

Gauer, W., 1993: “Die Perserkriege und die klassische Kunst der Griechen”, in Pöhlmann, E. & Gauer, W. (edd.), Griechische Klassik (Nürnberg) 167–87. Giangiulio, M., 2009: “The Emergence of Pisatis”, in Funke, P. & Luraghi, N. (edd.), The Politics of Ethnicity and the Crisis of the Peloponnesian League (Washington) 65–86. Giovannini, A., 1997: “La participation des allies au financement du Parthenon: ‘Aparche’ ou tribut?”, Historia 46: 145–57. Hennemeyer, A., 2012: “Der Zeus-Tempel von Olympia”, in Gehrke, H.-J. & Heilmeyer, W.-D. (edd.), Mythos Olympia. Kult und Spiele. Ausstellungskatalog Berlin, Martin Gropius Bau (München) 121–25. Herrmann, H.-V., 1972: Olympia: Heiligtum und Wettkampfstätte (München). Hornblower, S., 2008: A Commentary on Thucydides: Books 5.25– 8.109 (Oxford). Jost, M., 1985: Sanctuaires et cultes d’Arcadie (Paris). Knell, H., 1968: “Iktinos, Baumeister des Parthenon und des Apollontempels von Phigaleia-Bassae?”, JdI 83: 100–17. Knell, H., 1983: “Lepreon—der Tempel der Demeter”, AM 98: 113–47. Korres, M., 2001: “Iktinos”, in Vollkommer, R. (ed.), Künstlerlexikon der Antike I (München) 338–45. Krentz, P. & Wheeler, E.L., 1994: Polyaenus. Stratagems of War (Chicago). Kyrieleis, H., 2012/2013: “Pelops, Herakles, Theseus. Zur Interpretation der Skulpturen des Zeustempels von Olympia”, JdI 127/128: 51–124. Lippolis, E., Livadiotti, M. & Rocco, G., 2007: Architettura greca: storia e monumenti del mondo della polis dalle origini al V secolo (Milan). Luraghi, N., 2008: The Ancient Messenians (Cambridge). Malkin, I., 1987: Religion and Colonization in Ancient Greece (Leiden). Mallwitz, A., 1975: “Zur Architektur des Apollon-Tempels in Bassai-Phigalia”, in Hofkes-Brukker, C. & Mallwitz, A. (edd.), Der Bassai-Fries in der ursprünglich geplanten Anordnung (München) 9–44. Martin, R., 1972: “Aspects financiers et sociaux des programmes de construction dans les villes Grecques de Grande Grèce et de Sicile”, AttiMemMagnaGr 12: 185–205. Möller, A., 2004: “Elis, Olympia und das Jahr 580 v. Chr”, in Rollinger, R. & Ulf, C. (edd.), Griechische Archaik. Interne Entwicklungen—externe Impulse (Berlin) 249–270. Nafissi, M., 2001: “Pausania e la storia dell’Elide: la questione pisate”, in Knoepfler, D. & Piérart, M. (edd.), Éditer, traduire, commenter Pausanias en l’an 2000 (Genève) 301–21. Nakassis, A., 2004: O ναός της Aθηνάς Mακίστoυ (Athens). Nielsen, T.H., 2004: “Triphylia”, in Hansen, M.H. & Nielsen, T.H. (edd.), An Inventory of Archaic and Classical Greek Poleis (Oxford) 540–546.

GREEK TEMPLE BUILDING FROM AN ECONOMIC PERSPECTIVE: THE WESTERN Peloponnesos Paradiso, A. & Roy, J., 2008: “Lepreon and Phyrkos in 421–420”, Klio 90: 27–35. Parker, V., 1991: “The Dates of the Messenian Wars”, Chiron 21: 25–47. Patay-Horváth, A., 2012: “Die Bauherren des Zeustempels von Olympia”, Hephaistos 29: 35–54. Patay-Horváth, A., 2013a: “Das Metroon von Olympia als Stiftung von Elis”, BJb 213: 17–26. Patay-Horváth, A., 2013b: “Hera in Olympia: Tempel, Kult und Münzprägung”, Thetis 20: 81–99. Patay-Horváth, A., 2014: “Die Baukosten des Zeustempels von Olympia”, Boreas 35: 1–10. Patay-Horváth, A., 2015: “Adopting a New Approach to the Temple and its Sculptural Decoration”, in Patay-Horváth, A. (ed.), New Approaches to the Temple of Zeus at Olympia (Newcastle upon Tyne) 1–15. Patay-Horváth, A., 2016: “Lepreon during the 5th century BC”, in Szabó, Á. (ed.), From Polites to Magos. Studia György Németh sexagenario dedicata (Budapest) 243–54. Patay-Horváth, A., 2018a: “The Temple Builders of Apollo Epikourios at Bassai”, in Tamás, A.B., Bollók, Á. & Vida, T. (edd.), Across the Mediterranean—Along the Nile. Studies in Egyptology, Nubiology and Late Antiquity dedicated to László Török on the Occasion of His 75th Birthday (Budapest) 109–122. Patay-Horváth, A., 2018b: “Doric Temples in Southern Arcadia. Who built them and why?”, in Grüll, T. (ed.), Transfer and Mobility. Proceedings of the Fourth Conference on the Ancient Economy at Pécs (Budapest) 59–73. Philipp, H., 1994: “Olympia, die Peloponnes und die Westgrie­ chen”, JdI 109: 77–92. Prignitz, S., 2014: Bauurkunden und Bauprogramm von Epidauros (400–350): Asklepiostempel, Tholos, Kultbild, Brunnenhaus. Vestigia: Beiträge zur Alten Geschichte 67 (Munich). Pritchett, W.K., 1971: Ancient Greek Military Practices. Part I (Berkeley). Pritchett, W.K., 1991. The Greek State at War, Part V. (Berkeley).

177

Roux, G., 1961: L’architecture de l’Argolide aux IVe et IIIe siècles avant J.-C. (Paris). Roy, J., 1973: “Diodorus Siculus XV 40—The Peloponnesian Revolutions of 374 B.C.”, Klio 55: 135–39. Roy, J., 1998: “Thucydides 5.49.1–50.4: The Quarrel Between Elis and Sparta in 420 BC, and Elis’ Exploitation of Olympia”, Klio 80: 360–68. Roy, J., 2010: “The Nature and Extent of Elean Power in the Western Peloponnese”, in Lombardo, M. (ed.), Forme sovrapoleiche e interpoleiche di organizzazione nel mondo greco antico (Lecce) 293–302. Ruzé, F. & Christien, J., 2007: Sparte: Géographie, mythes et histoire (Paris). Stanier, R.S., 1953: “The Cost of the Parthenon”, JHS 73: 68–76. Stylianou, P.J., 1998: A Historical Commentary on Diodorus Siculus Book 15 (Oxford). Svenson-Evers, H., 1996: Die griechischen Architekten archaischer und klassischer Zeit. Archäologische Studien 11 (Frankfurt). Taraporewalla, R., 2011: “Size Matters: Competitive Emulation and the Statue of Zeus at Olympia”, in McWilliam, J., Puttock, S., Stevenson, T. & Taraporewalla, R. (edd.), The Statue of Zeus at Olympia: New Approaches (Newcastle upon Tyne) 33–50. Tuplin, CH., 2004: “Xenophon, Artemis and Scillus”, in Figueira, Th. (ed.), Spartan Society (Swansea) 251–81. Welwei, K.W., 1974: Unfreie im antiken Kriegsdienst 1: Athen und Sparta (Wiesbaden). Wilson Jones, M., 2014: Origins of Classical Architecture: Temples, Orders and Gifts to the Gods in Ancient Greece (New Haven). Woodward, R.J., 2012: An Architectural Investigation into the Relationship between Doric Temple Architecture and Identity in the Archaic and Classical Periods (diss., University of Sheffield). Yalouris, N., 1965: “Δoκιμαστική ανασκαφή εις τoν Nαόν τoυ Eπικoυρίoυ Aπόλλωνoς Bασσών”, Prakt 1959: 155–59. Younger, J.G. & Rehak, P., 2009: “Technical Observations on the Sculptures from the Temple of Zeus at Olympia”, Hesperia 78.1: 41–105.

chapter 11

Old Questions and New Approaches: The Significance of Affinities between the Tectonic Arts and the Technical Arts of Ancient Greece Mark Wilson Jones 1 Introduction Contemporary research on ancient Greek architecture benefits from a host of discoveries and advances achieved over the twentieth century, and that continue apace in this one. By comparison with investigators either side of the time of Greek independence we may count ourselves fortunate, in as much that we take for granted knowledge and understanding that eluded them. There have been too many subsequent advances to mention more than one of the most fundamental that has shaped understanding of the Bronze Age, the discovery by Schliemann and others of the culture we have come to call Mykenaian. Our grasp of the architecture of the Archaic period is also much more sound and articulated; it is only relatively recently that we have acquired a proper appreciation of non-canonical regional developments, as manifest, for example, in Kykladic architecture. The stock of knowledge continues to expand for all periods. Meanwhile analytical capability has progressed thanks to stratigraphical excavation methods and various laboratory techniques such as carbon dating and isotopic analysis. Laser scanning and digital technologies have taken documentation and visualisation of ancient environments to new levels. Scholarly interpretation has become more rigorous, systematic, comprehensive and detailed. Specialisms have formed and expertise has developed with regard to style, technique, and construction, all of which adds to our capacity for precision as regards both relative and absolute chronology. Rather as the Biblical account of creation yielded to scientific method exemplified by Darwin’s work, longstanding theories that do not square with the evidence have come to lose their allure: proportional systems based on the golden section, for example, or visions of ‘primitive’ phases of development based on an overly literal linear interpretation of technological evolution. A case in point is Viollet-le-Duc’s contention—in line with his penchant for rational explanation—that early Greek stone column shafts took cylindrical form so as to facilitate being rolled from the quarries (for discussion, see Wilson Jones 2014: 64–69; cf. Viollet-le-Duc 1858).

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004416659_013

Following the divide between architectural practice and academic research that widened over the course of the 20th century, the studies and conclusions of earlier periods now appear not only outdated for the reasons mentioned but also much compromised by the free interpenetration of these two domains. Renaissance architects would sometimes ‘correct’ aspects of ancient buildings they surveyed and sketched. 18th- and 19th-century architects often introduced elements of their own design into ‘copies’ of ancient monuments, which is similarly problematic should anyone seek to extract information about the originals. Yet we nonetheless owe a profound debt to those who pioneered modern research into Greek architecture. As Barbara Barletta (2011: 611) noted in her “State of the Discipline” article on Greek Architecture in the American Journal of Archaeology: The study of Greek architecture grew out of the meticulous recording of buildings and their components by 18th- and 19th-century investigators. Although the aims have changed, with an increasing emphasis on historical and social context, the basic methods of documentation remain the same. Barletta highlights the contribution to recording and documentation, a point underlined by the reputation of Francis Penrose’s survey of the Parthenon as more reliable than anything subsequent that is comparable in scope (Mertens 1984; personal communications with Manolis Korres and Vasileia Manidaki). But my emphasis here is on another aspect: the value of certain early investigators’ insights into the nature of ancient Greek design. It is inevitable that old interpretations come to be superseded, yet the loss will be ours if we put these out of mind on account of too great a faith in the march of progress. This is not so much because of access to material that has since been damaged, lost, or destroyed. More important, I would argue, are three characteristics that underpin the quality of insight:

AFFINITIES BETWEEN THE TECTONIC ARTS & THE TECHNICAL ARTS OF ANCIENT GREECE

i)

Early interpreters of ancient architecture tended to be practicing architects, concerned with the act of design and with the practical matters of making things (which, before the rupture of modernity with tradition, sometimes involved materials and techniques not entirely dissimilar to ancient equivalents). By contrast, contemporary researchers into ancient architecture often have limited experience of architectural design or engineering. ii) They were interested in the big picture as well as details, while understanding that design sometimes involves conflicts, contradictions, and the exercise of artistic judgement. Contemporary research typically pursues narrower issues in the pursuit of definite, logical, and verifiable answers. iii) They made connections between architecture and other kinds of production that today we tend to treat as separate realms, while specialisation can result in architectural elements such as friezes, capitals, akroteria, and roof tiles being studied in isolation (Tomlinson 1976: 16–7; Marconi 2004: 212; 2007: 2; Barletta 2011: esp. 621–26). The tension between wider questions and scholarly focus is important, for prospective students and a more general audience may be alienated by approaches that are perceived as overly limited, myopic, or arid. Just as some scholars of ancient philosophy participate in contemporary debate over the old question of how to live a good life, perhaps enhancing their positions in the process, so too there is value in scholars of ancient architecture engaging with longstanding avenues of inquiry that remain unresolved. Of course, past answers may now appear erroneous or overly speculative, but those originating from the better-informed and sharpest earlier interpreters may yet offer lessons of value. An illustration of this point is provided by the case of the Pantheon in Rome and certain ideas of Antonio da Sangallo the younger, a leading architect in that city in the first half of the 16th century—an exceptionally fertile period for architectural practice and the study of ancient monuments, both of which he excelled in. One of his drawings contained a plan of the portico, a plan and part elevation of the rotunda, along with annotations and commentary (Uffizi A 874v, see Bartoli 1914–22: III, Pl. 237, fig. 414; VI, 76; Wilson Jones 2000: fig. 9.19). Sangallo’s observations were driven by puzzlement—shared by many before and after—over aspects of the building’s composition. In his sketch of the portico he crossed out the actual arrangement of pilasters in the vestibule between the portico and the main door, adding what seemed to him a

179

better solution. After centuries of controversy, it now appears that Sangallo’s preferred arrangement was indeed intended as part of the original design of the Pantheon before a compromise had to be adopted (from the jointing of the capitals it is evident that the pilasters which perplexed Sangallo were erected after the antae that they butted up against: Wilson Jones 2000: ch. 10, esp. 203). On the other hand, his critique of the internal elevation of the rotunda and its apparent lack of coherence remains untenable (Wilson Jones 2000: ch. 9), yet, like the theories of Michelangelo and others, it has the virtue of highlighting issues that merit investigation using information and tools that they did not possess. As for methods for producing the entasis of column shafts, those advocated long ago in the Renaissance by Serlio and Alberti have turned out to be closer than most subsequent proposals to the spirit of ancient practice, which was finally revealed by the ancient drawings/templates discovered at Didyma and Aphrodisias in the 1980s and 1990s, respectively (Haselberger 1980; 1983; Hueber 1998; Wilson Jones 1999: 127–32; cf. Chapter 3 in this volume). 2

Affinities between the Tectonic Arts and Technical Arts

Turning to Greek architecture and the focus of this paper, like so many before I have found myself pondering fundamental questions such as why the classical columnar styles took the form they did, why some became canonic while others were discarded, and what they meant to their audience. Clearly there are no single or simple answers, to judge from both the volume of literature and the multiplicity of factors involved. The Doric, Ionic, and Korinthian vocabularies varied extensively (the notion of a canon being only that), responding as they did to disparate considerations, including structure, construction, materials, culture, symbolism, and influence from preceding civilisations (Wilson Jones 2014; for reviews, see: Barletta 2014; Allies 2015; de Figueiredo 2015; Hellmann 2015; Klein 2015; Patay-Horváth 2015 [cf. my reply 2016b]; Rojas 2015; Barresi 2016; Sapirstein 2016; Gros 2018: 176–78). Insufficient attention over the past several decades has been given to another kind of potential influence, that from non-architectural sources. After all, various affinities connect architecture to different branches of art and production, as observable most straightforwardly in shared types of ornament such as rosettes, the guilloche, the meander, stylised leaf tongues, friezes with alternating fields and dividers (e.g., metopes and triglyphs), spirals

180 and volutes, the running wave, palmettes, linked chains of lotus buds and/or palmettes, and acanthus in a variety of guises. Moreover, kindred motifs, patterns, and attitudes to composition recur across a range of media, be it ceramics, ivories, metalwork, stonework, textiles, or woodwork. It is important to give due weight to the fact that in Archaic Greece a career in architecture alone did not exist; the designers of buildings were the same individuals as artists and craftsmen (Bötticher 1874 [1844], 4; Coulton 1977: 23, n. 51; cf. Svenson-Evers 1996). For example, Theodoros, architect of the Samian dipteros, was famed as an artist, sculptor, metalworker, miniaturist, and jeweller (it was he who made Polykrates’s famous golden ring: cf. Hdt. 3.41). These multifarious activities help us to appreciate a comment made by Vitruvius in his discussion of the rise of building, when he envisaged early builders being inspired by ideas “born from the variety of their crafts” (ex varietate artium natis, Vitr. 2.1.7). There are several reasons why this statement and the possibility of architecture responding to influence from other domains has been overlooked of late. Vitruvius himself famously related elements of the Doric entablature to roof construction, a principle much espoused and expanded ever since the Renaissance (Wilson Jones 2014: ch. 3; 2016a). By the beginning of the twentieth century this principle had mutated into a dogma, with Otto Wagner declaring, with upper case for emphasis, that “EVERY ARCHITECTURAL FORM HAS ARISEN IN CONSTRUCTION AND HAS SUCCESSIVELY BECOME AN ART-FORM” (Wagner 1988 [1902]: 92). Subsequent modern architectural theory reinforced this view with claims for the autonomy of the discipline and doctrines such as ‘form follows function’ or starting from a ‘tabula rasa’—effectively putting history and wider culture to one side. There were those who did recognise commonality between the ancient arts, but more likely than not they subscribed to the first of the propositions opening Owen Jones’s much-reproduced Grammar of Ornament: “The Decorative Arts arise from, and should properly be attendant upon, Architecture” (Jones 1865: 5 [with a Preface dated 1856], and multiple later editions). This consolidated the hierarchy of the arts established by Vasari, which envisages influence flowing from the major arts of painting, sculpture, and architecture to the minor arts, and from things that are large to things that are small. The appellation ‘minor’ can be contested, however (see Roaf 1996; Hourihane 2012; Wilson Jones 2014: 159– 60; Wilson Jones 2017). The existence of affinities with other domains logically raises the potential for influence on architecture as well as from architecture. Indeed the

Wilson Jones

evidence from Greece suggests that the direction of flow went in different directions at different times. Altars and vases of the Classical period with Doric and/or Ionic features borrow them no doubt from architecture, yet the same cannot safely be presumed at the time when the forms emerged that came to define Doric and Ionic. In fact, as a general rule ornaments such as those mentioned earlier (rosettes, the guilloche, the meander and so on) occur first on non-architectural objects and later in architecture. I have already elsewhere argued the case with regards to a form of ornament not mentioned, a pattern of repeating arcs reminiscent of the scaly coverings of fish, reptiles, and imaginary monsters, as well as bird plumage (Wilson Jones 2014: 163–64). Fish-scale passed from representational art and miscellaneous kinds of surface ornament in Egyptian, Minoan, and Near Eastern contexts before appearing in Greece on protomes, sphinxes, and suchlike, then on akroteria (both sphinx and disc), and eventually on to building elements such as capitals, metopes, door grilles, and balustrading. It is also important to remember that in Greece in the eighth century there was no substantial indigenous tradition of building. When in the second half of the seventh century monumental building arose, relatively suddenly, architect-artists could naturally have transferred aspects of their regular creative work to this new and episodic problem. The most prestigious art-objects on view at the time would have been especially instructive: top-flight votive gifts, whether imported or made locally, on view in the premier Greek sanctuaries such as the Argive Heraion, Delphoi, Ephesos, Olympia, and Samos. These were the same places where the most influential temples were built (Wilson Jones 2014: 19–21, 24–27, 161–63). There is, then, reason to take seriously Vitruvius’s words quoted earlier and to entertain the possibility that early builders adopted or were inspired by forms that had been pioneered in other media. In this regard it is well to recall certain ideas of two 19thcentury architect-scholars: Karl Bötticher (1806–1889) and especially Gottfried Semper (1803–1879). This is in spite of the fact that Wagner drew on aspects of Semper’s work in arriving at the statement quoted earlier—which was only possible by selectively ignoring precisely the other aspects that interest us here. Both Bötticher and Semper were acute commentators on ancient design who championed links between architecture and other kinds of production—that is to say, between the tectonic arts and the technical arts to use the terminology announced by the titles of their main publications: Bötticher’s Tektonik der Hellenen (Architectonics of the Greeks, in two volumes

AFFINITIES BETWEEN THE TECTONIC ARTS & THE TECHNICAL ARTS OF ANCIENT GREECE

of 1844 and 1852), and Semper’s Der Stil in den technischen und tektonischen Künsten … (Style in the technical and tectonic arts, published in two volumes of 1860 and 1863) (Bötticher 1844–1852; Semper 2007 [1860–63]). Both, especially Bötticher, were much indebted to Karl Friedrich Schinkel (1781–1841), a veritable 19th-century Theodorus whose oeuvre spanned painting, fabrics, vases, tableware, fittings, furniture, set-design, and of course architecture. Bötticher, a student of Schinkel, saw him as the preeminent interpreter of Greek antiquity (Bötticher 1846). Near the beginning of the introduction to Der Stil— after trying the reader’s patience with an extensive prolegomenon covering a range of generalities and topics as varied as snowflakes and comets—Semper addressed the purpose of his book and its basic tenets. “It is not possible to understand architecture”, he stated, “without considering (the) influence of the technical arts on the emergence of its traditional forms and types” (Semper 2007 [1860– 63]: 106). He then anticipated the results of such consideration: “It will be shown that the fundamental principles of style in the technical arts are identical with those governing architecture, that the simplest and clearest expression of these principles are to be found in the technical arts where they were first established and developed” (Semper 2007 [1860–63]: 107). Semper’s main case hardly seems in doubt as regards architectural surface ornament given, as mentioned, that so much of it is found on non-architectural objects, often in eastern Mediterranean contexts that predate the emergence of monumental building in Greece. On the other hand many of Semper’s ideas appear too tangential or elusive to warrant the attention of the modern specialist. He explored, for example, the interaction between materials, techniques, and functions that guided the making of bands, seams, knots, and other such operations that he argued first arose in the textile arts, with consequences for the Greeks’ emergent aesthetic predilections. Less contentiously perhaps, both Bötticher and Semper expanded on analogies with natural forms that feature in some Greek texts and Vitruvius’s treatise, showing how vegetal, zoomorphic, and anthropomorphic traits affected architectural and non-architectural supports alike (Vitr. 1.2.5; 3.1; 4.1.3–8; cf. Rykwert 1996; Wilson Jones 2014: 141–46). Doric columns excepted, these typically divide into three parts: a base or foot, a shaft or trunk/body, and a capital or head/crown. Both men held that form distilled aesthetic responses to structural, constructional, and material realities, hence the flare to spread load where a shaft meets a base, as occurs in trees or where an animal or human leg widens at the ankle. Just such a flare, often terminating in

181

a fillet, recurs on pottery, on metal stands such as candelabra, on furniture supports, and on architectural columns with bases. This is not the place to test such theories, but I would rather shift the focus to the potential influence on the architectural orders of certain non-architectural objects not properly considered by Bötticher and Semper. Of the themes addressed in my research into the formation of the orders, the importance of this one was the least to be expected on the basis of modern scholarship. It is intriguing that the triglyph, the echinos of some Doricising and Ionic capitals, and the core of the Korinthian capital all display affinities with objects, that is to say, the tripodcauldron, the phiale, and the kalathos, respectively. Only the last features in past discussions, thanks to Vitruvius’s passage concerning the creation of the Korinthian capital. Whereas many of the explanations for specific architectural elements published by Bötticher, Semper, and other 19th-century commentators are as subjective as they are improvable, my goal here is to set out evidence that meets modern standards of scrutiny. In short, I hope to demonstrate how old paradigms may be reconciled with current scholarship so as to produce new advances. To this end I will work backwards from the Korinthian capital to the Doric triglyph, beginning with the more straightforward issue thanks to the passage just mentioned by Vitruvius. 3

The Core of the Korinthian Capital and the kalathos

In his account of the origin of the Korinthian capital, Vitruvius tells that its core originated from a kind of basket known as a kalathos (in Greek) or calathus (in Latin) (Vitr. 4.1.9–10). According to his version of events, the sculptor and craftsman extraordinaire Kallimakhos, passing through the cemetery at Korinthos, was attracted by the play of forms produced by the growth of an acanthus plant up and around a basket capped with a tile and containing things treasured by a recently deceased young woman. Inspired by the combination he went on to create the new form of capital (for recent treatments of the origins of the Korinthian capital see Rykwert 1996, 317–349]; Wesenberg 1999; Corso 2005; Scahill 2009; Wilson Jones 2014: 146–56). Vitruvius’s account has been dismissed swiftly by Semper, Riegl, and others, but recent scholars tend to give it more credence since he seems to have derived it from the treatise he cites by Arkesios, de symmetriis

182 corinthiis (Vitr. 4.3.1; 7 Praef. 12; Corso & Romano 1997: I: 426, n. 49, 459–60, n. 119; Gros 2015: 233, n. 26; for context see Gros 1993 and Giuliani 1994: 29–63; Corso & Romano 1997: I:427, n. 50; 433, n. 64). This work probably dated to the end of the first half of the fourth century BCE, a time when Korinthian design was still developing. Yet Vitruvius’s passage can hardly be taken at face value given that it condenses into a moment what were in reality decades of experimentation. Indeed, it seems ironic that neither a basket nor acanthus formed part of miscellaneous early capitals with lyre-shaped volute compositions that would lead in the direction of what would later become Korinthian (Wilson Jones 2014: 151, figs. 6.12, 6.10, left, 6.20). Nonetheless, at some point in the second half of the fifth century capitals appeared with a core in the form of a truncated cone flaring out towards a rim at the top, as is characteristic of physical kalathoi (Fig. 11.1). Descending it seems from Mykenaian antecedents that are comparable in some but not all respects, ceramic vessels of the kalathos type are known from the 7th century (Fig. 11.1a) through the Classical period and on to Hellenistic and Roman times (Williams 1961: 27; Wilson Jones 2014: 151–53; Waite 2016). Such objects evidently imitated lightweight basketwork examples that appear in figural vase paintings, including one on a red-figure kalathos in the Shefton Museum, Newcastle (Fig. 11.1b). This has been dated to the early second half of the fifth century BCE (Stackelberg 1835: pl. 33; Williams 1961: 29; Waite 2016, who confirms a date of 440), by which time Kallimakhos’ career was developing stature. On occasions the kalathos shape

a

Wilson Jones

was reproduced in stone for funeral monuments (Athens NM 1052; Lissarrague 1995: 96, fig. 7; Wesenberg 1996: 4–5, Abb. 6). The core of the famous lost capital that once stood at the head of the Temple of Apollon at Bassai had a comparable shape (Fig. 11.2a). A kalathos-like core must have recommended itself when designs earlier pioneered for square or rectangular supports were adapted to fit circular columns. The capital at Bassai also presaged later developments in having the helices and volutes rising out behind the acanthus foliage. The lyre-shaped configuration was still used however for the interior half-columns of the tholos of Athena Pronaia at Delphoi (Wilson Jones 2014: 148, fig. 6.11). Hypothetically the Parthenon could earlier have had Korinthian / Korinthianising capitals capping the four columns in the rear room (Pedersen 1989), or alternatively a single prototype could have supported the outstretched arm of Athena Parthenos (Leipen 1971; Stewart 1990: 157–59, figs. 361–63; for doubts: Lapatin 2001: 86–88). The profile employed at Bassai was consolidated at the Tholos of Epidauros in the first surviving coherently resolved set of Korinthian capitals to display most of the formal characteristics of later capitals, including acanthus foliage that was naturalistic albeit stylised (Fig. 11.2b). As at Bassai, the core presented itself as a geometrically coherent form behind the leaves that surrounded it. That the core could be regarded as a distinct entity seems to be confirmed by a capital of likely fourth-century BCE date recently discovered at Korinthos (Fig. 11.2c) (Scahill 2009: 45–50). At the lower gymnasium at Priene there is a

b

figure 11.1 Two kalathoi from different periods: (a) kalathos from Eleusis (probably 7th century) decorated with geometric designs that recall basketwork; (b) red-figure kalathos (a) DAI Athen, neg. Elefsis 338 (b) Newcastle upon Tyne, Shefton Collection: NEWGM 853

183

AFFINITIES BETWEEN THE TECTONIC ARTS & THE TECHNICAL ARTS OF ANCIENT GREECE

a

c

b

figure 11.2

d

Korinthian and related capitals involving kalathos-like characteristics: (a) temple of Apollon at Bassai; (b) tholos in the Sanctuary of Athena Asklepios, Epidauros, interior; (c) kalathos capital from Korinthos; (d) Round Temple by the Tiber (a) Bauer 1973 (b) Bryn Mawr College Library Lantern Slide Collection (c) Scahill 2009 (d) author

broadly comparable capital attached to a shaft, except that unlike the cylindrical form of the capital from Korinthos the body has a reverse taper like kalathoi, yet no fillet or rim at the top (Dirschedl 2013: pl. 46.2). With its holes for fixing metal attachments, the kalathos-capital from Korinthos would seem to herald Roman versions that belonged to the temple of Bel at Palmyra (Amy, Seyrig & Will 1975: fig. 42). Pliny (HN 34.13[7]) reported bronze Korinthian capitals in Rome belonging to Agrippa’s Pantheon and the porticus of Octavia (for a pair of bronze Korinthian capitals in the transept of the Lateran basilica in Rome, see: Liverani 1995; cf. Rossignani 1969). It is reasonable to assume, as does David Scahill, that at Korinthos these attachments

took the form of acanthus leaves, helices, or palmettes, or possibly all of these combined (Scahill 2009: 45–50; for the role of metal in early Korinthian design, see: Chipiez 1876: 306–21, figs. 149–50; Choisy 1899: I:371, fig. 2; Bauer 1973: 11–12; von Normann 1996: 113–20; Wilson Jones 2014: 151–54, with further references; for the use of bronze in the Roman period in this context, see Rossignani 1969; Liverani 1995). The nature of the connection between the core of the Korinthian capital and the kalathos is open to three main contending possibilities: i) Initially having nothing to do with the kalathos, the shape of the core developed as it did for purely

184

Wilson Jones

formal reasons, the profile flaring rather as does the bottom of a column shaft and inclining so as to better relate to the epistyle. Thus it happened to resemble a kalathos, in response to which later someone such as Arcesios could have conjured up the Kallimakhos story. ii) The core at some stage became consciously modelled on the kalathos due to the suitability of its shape. iii) The core was modelled on the kalathos both for formal reasons (as above) and by virtue of content, that is to say, the feminine and funerary associations implicit in Vitruvius’s account. The more we put our trust in Arkesios’s source and Vitruvius’s reprise, the more we would favour the last possibility. The kalathos basket had associations that were both funerary and feminine, two aspects central to Vitruvius’s text. It was used for working with wool, a feminine activity (Lissarrague 1995: 95; cf. cat. 46; Oakley 2004: 62, 173, figs. 15, 134–35, col. fig. 1a; cf. Saglio & Daremberg 1877: I:812–14, s.v. ‘calathus’; Hilgers 1969: 42–44). Like the acanthus, kalathoi feature prominently on scenes painted on white lekythoi showing women visiting graves and honouring the dead (Homolle 1916; Rykwert 1996, 321; Wilson Jones 2000: 137; 2014: 151–54). Such lekythoi were popular by the middle of the second half of the fifth century BCE (Beazley 1938; Kunze-Götte 1984; Kurtz 1975; Oakley 2004), which is more or less consistent with the floruit of Kallimakhos (Kallimakhos created the famous ‘lamp’ in the Erektheion, begun 421 BCE: Korres ref. in Lexicon). The existence of female karyatids that carried vessels, including baskets, further aligns with the feminine theme in Vitruvius’s account of Korinthian origins (on karyatids: Homolle 1917; Schmidt-Colinet 1977; Schmidt 1982; Rykwert 1996: 133–38; Hellmann 2002: 201–11). The Roman writer also likened this type of column to a standing virgin (Vitr. 4.1). However, it is unclear how the feminine theme applied to a range of Korinthian structures from the Classical and Hellenistic periods that involved male divinities/

a figure 11.3

b

protagonists: the Temple of Apollon at Bassae, the Tholos in Asklepios’s sanctuary at Epidauros, the Asklepieion in Athenai, the Monument of Lysikrates, the temple of Zeus at Nemea, and the temple of Olympian Zeus in Athenai. And there were legion Korinthian temples of the Roman period dedicated to male deities, Apollon, Jupiter, Mars, and so on. Indeed, the gender scheme Doric/male, Ionic/ female, Korinthian/virginal, bears the hallmarks of an intellectual construct of the Hellenistic period (Gros 1993; 1995; Wilson Jones 2014: 142). The symbolism of acanthus can be tracked from the funerary sphere with overtones of triumph over death, and then on to triumph, vigor, and regeneration in more general terms. This would have made it attractive to Hellenistic builders and ultimately Augustus and suggests that the Korinthian capital was favoured primarily by virtue of its acanthus along with other formal/aesthetic considerations, in which case the first of the two possibilities enumerated above seem more likely for the rationale for the kalathos. It is worth briefly reviewing aspects of the later development of the Korinthian capital. Despite deriving ultimately from Greek models such as those of the Tholos of Epidauros, Italic capitals of the Republican period presented little or no sign of the calathus. This type came to be eclipsed by the ‘Normalkapitell’ based on Hellenistic prototypes (Heilmeyer 1970). This had an understated core flaring slightly towards a delicate rim in the shadow of the abacus, as occurs at the tholos by the Tiber in the Forum Boarium (Fig. 11.2d and 11.3a). The capitals of the Round Temple by the Tiber (the first phase of which, made with Pentelic marble, dates to the mid-to-late second century BCE) are representative in this respect (Rakob and Heilmeyer 1973: pl. 26, 1–2). In the lead up to the Augustan period the flare became stronger, while the rim protruded beyond the bottom of the abacus. This detail, unknown in Greek practice, went on to become the norm for imperial practice (Fig. 11.3c). An extreme case incorporating an overfall was employed for the Temple of Vespasianus (Fig. 11.3b). The newly accentuated flare

c

Kross-sections of Korinthian capitals in Rome: (a) Round Temple by the Tiber; (b) temple of Vespasianus; (c) temple of Hadrianus Wilson Jones 2000, fig. 7.20

AFFINITIES BETWEEN THE TECTONIC ARTS & THE TECHNICAL ARTS OF ANCIENT GREECE

185

table 11.1 Evidence for and against a link between the Korinthian capital and the kalathos

Connection between core of the Korinthian capitals and a basket

Corroborating indication

not pertinent at inception morphological resemblance tell-tale details documentary and other evidence

absence in earliest experiments generally good in mature examples inclined profile, flare, and rim Vitruvius’s text funeral scenes with kalathoi Kallimakhos a plausible protagonist Kalathoi exist that are earlier than any Korinthian capital artistic milieu in Korinthos and Athenai kalathoi associated with funerals and the female realm in line with passages from Vitruvius accentuation of the calathus in capitals around the time Vitruvius wrote

anteriority of object relative to architectural manifestation mechanism for transmission symbolic valency

later echoes

and rim presumably took hold since it conferred greater depth and modelling. Furthermore, these changes subtly drew attention to the core and rendered it recognisable as a calathus. This suggests that the story of the basket and Kallimakhos promulgated by Vitruvius’s De Architectura found an echo in practice (see Wilson Jones 2017: 93 for a brief mention of this idea, one that I will investigate further in a separate publication). As a way of summarising this discussion, Table 11.1 sets out the points for and against the possibility that some designers of Korinthian capitals had the kalathos in mind. 4

The Echinos and the Phiale

Links with vessels may also be detected in certain variants of Doric and Ionic capitals with an echinos that bears comparison with a shallow bowl. There is no question, however, of the concept of a bowl being pertinent when the echinos originated in either of these contexts. The principle of a capital with a square abacus over a circular neck had earlier been established in Egypt, to which a bulbous transitional element seems to have been introduced in Minoan and Mykenaian times. The motive was presumably formal/aesthetic, although inspiration from some unknown object of symbolic resonance cannot be excluded. Minoan and Mykenaian precedents can reasonably be characterised as ‘proto-Doric,’ in the sense that they anticipate Greek Doric capitals proper dating to around 600 BCE. The Treasury of Atreus is likely to have

Negative qualification

absence in most Italic capitals vertical profile on capital from Korinthos acanthus develops not only on capitals but also on roof decoration and other contexts

Korinthian used for tholoi and temples for male deities in contrast to Vitruvius’s gender scheme

furnished key models, and other monuments at or near Mykenai might have been visible in the Archaic period, providing sources of inspiration for designers during the period when the Doric order took form. Bronze Age style could also have been transmitted via miniature ivory ‘proto-Doric’ Mykenaian columns alluded to earlier, such as those deposited as foundation offerings under the Archaic temple of Artemis on Delos (Gallet de Santerre & Tréheux 1947–1948: 193–97, pl. 34; Wesenberg 1971: 62. Numerous ivory ‘proto-Doric’ colonnettes have been recovered from Mykenai: see Poursat 1977: pls. 8 [71/7429 and 72/7430, both with a foliate necking and a lobed echinos], 13, 22, 27). In early Doric capitals the echinos could range from a flattish ‘pancake’ shape (Fig. 11.4a) to a simple quarter round (Fig. 11.4c) This suggests that the echinos was not equated to a specific object or idea beyond the formal model acquired from the Myceneans. As architects experimented in the Archaic period, the shape of capitals flexed according to a combination of aesthetic and tectonic considerations, including varying spans and the properties of timber and stone. Nonetheless, in between the possible extremes the shape of the echinos can call to mind a shallow bowl. Particularly relevant it seems was the phiale, a libation bowl that was also a favoured high-class offering in Greece (Luschey 1939; cf. Sciacca 2005; for further references: Wilson Jones 2014: 170). While imitations existed in ceramic, prestigious examples were made of bronze, and on occasions silver or gold. Certain phialai evidently inspired the detailed articulation of some Doricising

186

Wilson Jones

a

b

c

d

figure 11.4

Half-elevations and half-sections of Archaic Doric capitals: (a) Kerkyra, Xenvares column; (b) Tiryns; (c) Kalapódhi; (d) Syrakousai, Apollonion

capitals in the Archaic and Classical periods. Those of the mid-sixth century BCE temple at Metropolis in Thessalia present an echinos that resembles a kind of phiale with an Egyptianising lotus pattern (Fig. 11.5a) (Wilson Jones 2014: 167, fig. 7.11). Lotus motifs alternating with convex egg or almond-like protrusions in relief were the equivalent in stone of patterns beaten out in metal phialai (Fig. 11.6a), as in the case of a bronze example associated with the chryselephantine statue of Apollon on display in the Archaeological Museum at Delphoi that is thought also to date to around 550 BCE (for the phiale see Aruz 2014: 124, fig. 3.11; for the statue see Lapatin 2001: 57–60). It is likely that the phiale was held in the god’s right hand, consistent with images on vase painting (e.g., Wilson Jones 2014:

fig. 10.2). There are two broadly comparable phialai in the Rogozen treasure (Marazov et al. 1989: see especially cat. no. 2). The lotus motifs on the capitals of the roughly contemporary temple of Hera at Paestum have a looser distribution but nonetheless may have been borrowed from a similar source (Fig. 11.5b) (Mertens 1993: 18–28, figs. 54, 64; Wilson Jones 2014: fig. 3.23). Two capitals dating to around 500 now in Tbilisi, Georgia, present another kind of lotus ornament that was common on Achaemenid phialai around that time (Shefton 2000). It is not possible to connect these various capitals with specific phialai found at the same sites, but ones with comparable patterns were not uncommon in Greece, having descended from the eastern Mediterranean regions from where they ultimately originated (Luschey 1939). Meanwhile, the Doricising capitals borne by the karyatids of the Knidian Treasury at Delphoi (Fig. 11.5c) have an echinos treated like a different type of phiale articulated by convex ribs or flutes (Fig. 11.6b) (Wilson Jones 2014: 168, figs. 7.12 and 7.13; cf. Schmidt 1982: pl. 11). In this case excavations at Delphoi have yielded up just such a phiale, as it happens only a few metres from the treasury, in a context that is likely to anticipate its construction (Luce 2008: II, pl. 55). Over a century later the more famous karyatids of the Erektheion terminate in a hybrid capital, Doric in shape but Ionicising in the egg-and-dart decorating the echinos (Fig. 11.5d). It is possible that the egg-and-dart ornament was transferred from the ovolo moulding of the echinos of Ionic capitals—the building has two sets of Ionic columns. Yet this does not exclude the designer of the Erektheion from borrowing from another branch of phialai with variations on the theme of lobes, sometimes with labia or rims around egg- or almond-shaped lobes, sometimes with leaf-points or ‘darts’ in between the lobes (Fig. 11.6c) (see Wilson Jones 2014: fig. 9.6 for a comparable silver phiale in the British Museum [inv. 1994,0127.1], datable by its inscription to the reign of Artaxerxes 1, 464– 424—that is to say prior to the building of the Erektheion). As others have noted, the treatment of the egg-and-dart on the Erektheion capital has a metallic quality (Drerup 1952: 28–30), which reinforces the connection with phialai (see Luschey 1939; Pfrommer 1987; cf. Vickers 2014). There should be no need to belabour the case given the affinity already mentioned between karyatids and human figures bearing vessels or baskets. What is more, the specific relevance for the karyatids of the Erektheion is underscored by the fact that in their right hands they originally carried phialai (These belonged to another variant of the lobed principle, with smaller lobes treated as acorns: Wilson Jones 2014: 169, 192–95, fig. 9.7; for other karyatids

AFFINITIES BETWEEN THE TECTONIC ARTS & THE TECHNICAL ARTS OF ANCIENT GREECE

187

a

b

c

d figure 11.5

Non-canonic Doricising capitals with ornamentation: (a) temple at Metropolis; (b) temple of Hera I, Paestum; (c) karyatid of the Knidian Treasury, sanctuary of Apollon, Delphoi; (d) karyatid of the south porch of the Erektheion, Athenian Akropolis (a) Manolis Korres (b) Mertens 1993, Abb. 56b (c) École Française, Athens, neg. 22.351 (d) DAI, Hege 1554

188

a figure 11.6

Wilson Jones

b

c

Three common types of phiale (libation bowl and offering): (a) silver phiale with rings of lotus ornament, first half fourth century, from Ithake, now in the British Museum; (b) ribbed bronze phiale, seventh–early sixth century, Karmin-Blur; (c) lobed silver phiale from the reign of Artaxerxes 1 (464–24 BCE) (a) BM inv. GR 1920.5–29.2 (b) Vorderasiatisches Museum, Berlin, inv. 796 (c) BM inv. 1994,0127.1

from Limyra that also held a phiale in the right hand, see: Schmidt 1982: pl. 12–16; cf. Borchhardt 1976). As for the echinos of the Ionic capital itself, this seems to have evolved from foliate and lobed decoration on capitals and crowns belonging to miscellaneous kinds of supports, from Mykenaian ivory colonnettes to metalwork from around the eastern end of the Mediterranean (Wilson Jones 2014: 92, fig. 4.8), not forgetting Aeolic columns (Wilson Jones 2014: ch. 4), nor radial arrays of lobes crowning circular altars (Koenigs 1996; Wilson Jones 2014: 193). The development of the Ionic capital is in fact particularly complex, there being a great many strands of development, sometimes distinct, and sometimes interrelated (Theodorescu 1980). As with the Doric, the shape of Ionic capitals flexed according to a combination of aesthetic and tectonic considerations. Indeed, the appearance of the egg-and-dart appears to respond to proportional and geometrical issues affecting a particular stage of development, the stage when the earlier Kykladic form of capital gave way to the Ionian-Attic form. In order to understand these issues it is necessary to bear in mind that the echinos of Kykladic capitals was typically large, taking up a complete circle and profiled with an overfall in a manner of some leaf/petal crowns, and decorated with flattish lobes having borders running roughly parallel to each other (Fig. 11.7a) (Wilson Jones 2014: fig. 5.18, and 120–21 Table 5 for compiled references). Just as observed with early manifestations of the Doric echinos, there is nothing cup-like at this stage. It is also important to appreciate that the way in which the volutes sat over the crown-like element made the capital as a whole quite elongated. This elongation made the Kykladic type completely unsuited to corner capitals. At any rate, Kykladic corner capitals are unknown, presumably

because they would have had to be very top-heavy if they were configured like later corner capitals with two fronts meeting on the external diagonal. Hypothetically, a cruciform version of capital could have been considered, as illustrated in Figure 11.8. However, given its awkwardness and foreign-looking appearance (some Persian bracket capitals are cruciform) it is easy to appreciate why this option was not put into practice. So as to adapt itself to the corner—this being essential for the creation of peripteral temples with consistent columnar style—the capital had to have a front:side ratio nearer to a square, which is the same as saying it needed to be significantly more compact. This demanded a substantial reduction in the size of the echinos, which was achieved in Ionian capitals thanks to eliminating the overfall (Wilson Jones 2014: 129–31). The overfall was already absent on the capitals of the Artemision at Ephesos, although it is not certain that this temple incorporated corner capitals comparable with later canonic examples. Over time the echinos was reduced to the point of resembling a bowl or cup facing upwards. As seen in Ionian and then Attic capitals, the radial geometry of the bowl-shaped profile favoured egg-shaped lobes (Fig. 11.7b). A similar development can also be observed on round altars (Koenigs 1996; Berti & Masturzo 2000: 223). It is possible that such purely formal considerations led to egg-and-dart. Yet the multifaceted nature of architectural design together with the ideas of Semper and Bötticher warrant consideration of the possibility of a borrowing from lobed forms of phialai. It is clear from what we have seen of both Doric and Ionic capitals that the introduction of phiale-like characteristics represents by no means a total explanation but rather a secondary and partial development—though not for that unimportant. Even if it does not support any

AFFINITIES BETWEEN THE TECTONIC ARTS & THE TECHNICAL ARTS OF ANCIENT GREECE

189

a

b

figure 11.7

figure 11.8

Digital reconstructions of two major branches of Ionic capital: (a) Naxian sphinx column, Delphoi, Kykladic type, middle of the first half of the sixth century; (b) Propylaia, Athenian Akropolis, an Attic development of the Ionian type, middle of the second half of the fifth century Georg Herdt

Digital reconstruction of a hypothetical cruciform solution for the Kykladic type of Ionic capital. Because of the elongated form of the normal Kykladic capital it would have been impossible to create a corner variant using a similar strategy to that which later proved successful for the more compact Ionian-Attic type Georg Herdt

clear-cut thesis, the use of the term echinos does not contradict this reading. Echinos was used for a variety of natural and man-made objects with a rounded form eluding geometric definition. It could, however, indicate something vessel-like in the sense of containing a liquid, as for example in the case of a sea-urchin. When dried and with the sharp-pointed spines characteristic of this creature in life removed, the remaining shell could be used as a cup or container, and it is in this sense that Hippokrates used the term (Hippokrates, De Foem. St. 3.24; cf. Rykwert 1996: 178–79, nn. 30–31, and pp. 460–61). Vitruvius used the term twice for Doric (4.3.4) and Tuscan (4.7.3) capitals without commentary (cf. Corso & Romano 1997: I: 436, n. 136). There are then three types of treatment for the architectural echinos (lotus, ribbed, and lobed), each of which mirror types of phialai. To repeat, this does not mean that the echinos of Doric and Ionic capitals originated in bowls, since early examples had none of the later correspondences just noted. The existence of such correspondences

190

Wilson Jones

table 11.2 Evidence for and against a link between Doricising capitals and the phiale

Connection between the Doric and Ionic echinos and the phiale

Corroborating indication

not pertinent at inception morphological resemblance

absence in earliest forms of both Doric and Ionic capitals good, in some cases

tell-tale details

lotus, ribbed, and lobed patterns

documentary and other evidence

phialai known from relevant sites (e.g. Delphoi and Athenai) phialai exist that are earlier than capitals with comparable patterns artistic milieu in key sanctuaries phialai associated with offering, and as such appropriate for temples, for these too were in a sense offerings

anteriority of object relative to architectural manifestation mechanism for transmission symbolic valency

later echoes

testifies to design development, to fluid exchanges across different media. While formal considerations drove this development, considerations of content were probably a factor. It may be appreciated just how appropriate for sacred architecture were the associations of the phiale with sacrifice and offering. Temples were houses for the gods aligned with altars and surrounded by offerings, besides being in themselves offerings (Wilson Jones 2014: xii, 24–27, 208–09). Be this as it may, before long any association with phialai was weakened by the direction taken by evolution in the design of capitals. The steeply sloping, essentially conical profiles of Doric capitals of the Classical period, as exemplified by those of the Parthenon, look nothing like phialai. The Ionic echinos also became too compact and too subservient to the volutes to convey much sense of a bowl. Just as before with the kalathos and the Korinthian capital, Table 11.2 summarises the points for and against the possibility that some Greek architects had a phiale in mind when creating the echinos of variants of Doric or Ionic capitals. 5

The Tripod and the Triglyph

Turning to the Doric frieze, a connection with another vessel can be summarised swiftly given the case has been argued on several occasions (Wilson Jones 2002; 2014: ch. 8; 2016a; for criticism of the 2002 article, see Kyrieleis 2008; Wesenberg 2008; reviews of the 2014 book tend to view

Negative qualification

only in some 6th-century Doric and 5th-to-4th-century Ionic absence on the majority of capitals, except for canonic Ionic egg-and-dart no connection made in textual sources

no connection made in textual sources

none known

the triglyph-tripod connection with more skepticism than other ideas, including Hellmann 2015: 45, 47; Klein 2015: 366; Patay-Horváth 2015, although Barresi [2016: esp. 543] and Sapirstein [2017: 216] are more positive). There are affinities between triglyphs and tripod-cauldrons, given similarities in shape and the way tripods could be set up in rows and/or on high (including on columns and as akroteria), and the way representations of them were used to punctuate friezes. Yet the Doric frieze did not represent a row of tripods nor did it originate in them, any more than the Korinthian capital straightforwardly represented an acanthus-shrouded basket. Instead the origins of the Doric frieze can be traced to field-and-divider friezes that were ubiquitous in diverse contexts, including in Egypt (Fig. 11.9a), in the Mykenaian period (Fig. 11.9b–c) and later the Geometric (Fig. 11.9d–e). The tri-partite dividers of the Mykenaian split-rosette frieze would have been especially relevant for the renaissance of monumental character (Fig. 11.9b–c), and it may be conjectured that Greek architects deployed proto-triglyphs along these lines by the middle of the seventh century, possibly to decorate the ends of beams rather as Vitruvius discusses (Vitr. 4.2.2; for arguments for and against, see Wilson Jones 2014: 68–75, 199–200; 2016a). At some point the proliferation of tripods and tripod motifs and their use as frieze dividers on non-architectural objects, whether made from ceramic, metalwork and perhaps rectangular wooden chests such as that of Kypselos which was described in such detail by Pausanias (5.17.5– 19; cf. Jones 1894), may have led to proto-triglyphs being enhanced with tripod-like features. A salient ceramic

AFFINITIES BETWEEN THE TECTONIC ARTS & THE TECHNICAL ARTS OF ANCIENT GREECE

191

a

b

c

d

e

f

e

figure 11.9 Friezes that may have contributed to the formation of the Doric frieze compared with a non-canonic variant: (a) Egyptian cornice; (b) split-rosette frieze, Treasury of Atreus, Mykenai; (c) split-rosette inlaid frieze, Tiryns; (d) frieze from a pot from Zagora; (e) frieze from a pot from Zagora; (f) frieze from a seventh-century tripod-kothon from Thasos; (g) Doricising third-century architectural frieze from Samos Georg Herdt

192

Wilson Jones

example is a tripod kothon from Thasos from the second half of the 7th century (Fig. 11.9f) (Haspels 1946). There were after all multiple associations that recommended the tripod as a suitable temple adornment, it being the Greeks’ prestige votive offering par excellence—one redolent of divinity and victory from the time of heroes. It is true that triglyphs lack the ring-handles that characterised tripods. Presumably ring-handles were omitted because they clashed with the linear principle of field and divider friezes, or because they added unnecessary height to the elevation, or because they would have looked unsuited to visually supporting the roof. Ring handles were often omitted from tripods made of clay or stone as they would have been too fragile. If triglyphs were once made of wood, ring handles would have been problematic in this material too. By reason of this omission the resemblance between these two classes of object is not so striking as some of the capital/phiale parallels just discussed. Yet the characteristics of tripods account for features of triglyphs unexplained by descent from Mykenaian precedent or indeed any other theory: the absence of a horizontal band at the bottom that on Mykenaian equivalents mirrored the band at the top, and the presence of the little arches connecting the uprights underneath the capping band. In the Archaic period the form of these arches—semi-circular, arched, and ogival—echoed equivalent shapes on representations of tripods. The uprights of non-canonic triglyphs like ones from Metapontion have little ribs of a kind anticipated on bronze tripod legs (Wilson Jones 2014:

compare fig. 8.9, left, and fig. 8.10). The tripod for which Apollon and Herakles struggle on the pediment of the Siphnian treasury at Delphoi has straight legs that are chamfered similarly to the uprights of triglyphs. Perhaps the connection between triglyphs and tripods still held currency when this building was erected, although amnesia is suggested on the part of designers who on rare occasions disposed tetraglyphs, and even in one case pentaglyphs. Examples of ‘tetraglyphs’ include terracotta ones from Kroton/Crotone (Mertens 1993: fig. 74), and an anta in the Temple of Hera (E) at Selinous. ‘Pentaglyphs’ are known from the temple at the Casa Marafioti site at Locri Epizephiroi (Dinsmoor 1950: 98; Østby 1978; Costamagna & Sabbione 1990: fig. 230). Further abstraction in the Classical period saw the arches between the uprights reduced to horizontal cuts, which may suggest the link with tripods was generally lost. This accords with the principle observed at the end of the 19th century by William Robertson Smith and Alois Riegl whereby a symbol or motif can be repeated to the point of becoming a mere automatic ingredient, simply the done thing (Smith 1889: 19; Riegl 1992 [1893]: 43–44). Yet the connection may have persisted in the minds of some, or have been noted in a treatise now lost to us, which might account for a singular Hellenistic frieze of tripod-triglyphs on Samos (Fig. 11.9g) (Tölle-Kastenbein 1974: fig. 75–76; Rumscheid 1994: pl. 55). The case for influence from a non-architectural object on an architectural form can once again be summarised in tabular form (Table 11.3).

table 11.3 Evidence for and against a link between triglyphs and tripods

Connection between the Doric triglyph and the tripod

Corroborating Indication

not pertinent at inception

signs that ‘proto-Doric’ friezes revived Mykenaian antecedents some affinities, including in terms of size, but not immediately obvious capping band but no lower band shape of arches under the capping band chamfers on uprights/legs detailing of triglyphs from Metapontion tripods known from relevant sites, some set up on high and/or in rows abundant archaeological evidence tripod friezes also earlier than Doric friezes artistic milieu in key sanctuaries tripods associated with Homeric heroes, Apollon, victory, and offering, the latter being appropriate for all temples tripod-triglyph frieze from Samos

morphological resemblance tell-tale details

documentary and other evidence anteriority of object relative to architectural manifestation mechanism for transmission symbolic valency

later echoes

Contra-indication or Qualification

absence of ring-handles no bowing to suggest bottom of cauldron

no connection made in texts

no connection made in texts

AFFINITIES BETWEEN THE TECTONIC ARTS & THE TECHNICAL ARTS OF ANCIENT GREECE

193

figure 11.10 Miniature gold offerings found at the sanctuary of Artemis at Ephesos, seventh century. The motifs used on such objects anticipate comparable motifs that are known to have been employed on architecture Seipel 2008, figs. on pp. 148, 148, 149 and 146; courtesy of Ulrike Muss

6

Related Observations on Volutes

Finally, brief mention may be made of disparate nonarchitectural devices whose individual contribution is impossible to pin down but that in aggregate are likely to have affected the formation of the orders. The case has been made for the constructional basis of the Ionic capital, being like a bracket which by spreading out laterally reduced the effective span of the architrave beam (Wilson Jones 2014: 85–86, fig. 3.28). The logic seems flawed—for in many Ionic capitals the uppermost surface at the junction with the architrave is detailed in such a way as to avoid the transmission of load through the volutes—but in any event when decorating the brackets architects chose volutes as decorative devices with a long prior history. It is also possible that Ionic volutes were created by modifying rising or ‘Aeolic’ configurations. These potentially included columns seen by travellers to territories east of the Aigaian, although this is hypothetical. Perhaps traces could be seen of Mykenaian buildings with rising-volute capitals, for representations of such have come down to us from both the Minoan and Mykenaian periods along with kindred pieces of furniture and jewellery (Wilson Jones 2014: 109–10). But what is not in doubt is the abundant circulation in Greece of volute devices on small-scale material of both Greek and foreign origin, as testified by excavation finds from key sanctuaries such as Olympia and Ephesos (Wilson Jones 2014: 96–98, 161–65). The latter, the birthplace of Ionic according to Vitruvius, has yielded ivory, horn, and gold artefacts from the seventh century

BCE with various linked volute devices, with or without palmettes, rosettes, and the little ‘drops’ that can occur on architectural capitals (Fig. 11.10) (Wilson Jones 2014: 161–65). Of particular note are ‘ambiguous’ double volutes that can look Aeolic or Ionic depending on the direction of view. The principle of interchangeability goes back to 10th-century bronze Cypriot tripod-stands with colonnettes capped sometimes by rising volutes, sometimes by horizontally linked volutes (Wilson Jones 2014: fig. 5.13). So when architects ‘invented’ the Ionic column in stone, they likely adapted formal solutions already employed in other spheres of art and production. The same is surely true of friezes in general, and perhaps also for the architrave with fascia and column bases. Indeed, rather than ask which parts of the architectural orders were indebted to non-architectural sources, we should really ask which parts were not? To my mind only the mutular geison is a clear case in point, derived as it would seem from the projecting ends of roofing timbers just as Vitruvius explained. 7 Conclusions The material presented confirms the commonality between the tectonic and technical arts sustained by Bötticher and Semper, and the latter’s claim that the principles of Greek architectural style were first established in the technical arts. The value of their overarching insights is not, however, matched in their treatment of specifics. Their proposals relating to the design of the triglyph, the

194 echinos, and the kalathos are disappointing, to say the least. Semper offered an origin for the triglyph in the crenelated borders of textiles (Semper 2007 [1860–63]: 125). He called the Korinthian capital a bell capital and thought it related to ceramics and the “Doric kettle-shaped echinus” [!], ignoring Vitruvius’s link with the kalathos and so missing a connection with baskets, despite his interests in skeuomorphism and his comments regarding such objects. This is all the more baffling given that Semper (2007 [1860–63]: 491) correctly observed that “[b]askets used in working wool and other objects connected with female occupations are frequently depicted on vases”, and included an illustration featuring such a basket, i.e., a kalathos. Bötticher’s static-aesthetic theories for moulding profiles led only to confusion when he discussed the echinos (Rykwert 1996: 460–61, n. 31). Though their standing as architectural theorists has been on the rise in recent years (see Semper 2007 [1860–63]; Oeschlin 2003; Foged & Hvejsel 2018), the selective, speculative, and poetic nature of the writings of both men has disqualified them as serious commentators on Greek architecture from an archaeological standpoint. The present interpretations differ in finding evidential support, as summarised in the preceding tables charting architectural echoes of the kalathos, the phiale, and the tripod-cauldron. The morphological affinities noted are in some cases quite striking. This applies to a lesser extent to the tripod-triglyph connection given the latter’s lack of ring-handles, but the similarity in the height of these two classes of objects offers a potential explanation for the otherwise unexplained magnitude of triglyphs relative to the size of early Doric buildings. There is plenty of evidence for the cultural resonance of the three objects of interest, with tripods and phialai in sanctuary environments, and kalathoi in funerary contexts. The anteriority of all three over the architectural elements they influenced is also clear (as is also the case for volutes in general and motifs such as fish-scale). There is no shortage of potential mechanisms for the qualities of such objects to ‘contaminate’ the thinking of architects. Sacred tripods and phialai were treasured and set up on display in and around temples at the foremost sanctuaries, while the most prestigious examples are likely to have been made by artists who collaborated with architects, if not operating themselves as architects. Although such connections cannot be proven, excavation finds can be suggestive—a case in point being the already mentioned ribbed bronze phiale uncovered at Delphoi just a few metres from the Knidian treasury and its karyatids bearing a similarly ribbed echinos. It seems striking how many intimations of the Ionic style are anticipated in the precious votive offerings recovered from

Wilson Jones

seventh-century layers within or adjacent to the Ephesian Artemision. As for the kalathos-Korinthian connection, the workshops of Athenai and Korinthos would have provided no shortage of opportunities for exchanges between those who made or painted vases and sculptor-inventors such as Kallimakhos. Some of the lessons concerning the formation of the Greek architectural orders may now be summarised: – Characteristics originating in non-architectural domains could cross over into architecture. This affected attitudes towards composition, the employment of surface ornament (rosettes, guilloche, meander, etc.), and affinities between certain artefacts and threedimensional components of the orders. – The impetus for borrowing lay, it seems, in the wish to improve the appearance of architectural elements as yet not fully resolved; architects looking for inspiration would have been drawn to material offering solutions to similar formal problems. – Shared approaches between those working in the tectonic arts and the technical arts could give rise to a convergence between two separate lines of development. Yet once architectural elements chanced to call to mind familiar objects, this may have triggered conscious acts of borrowing. – Architects appear to have gravitated towards objects with an appropriate semantic charge for the buildings they were designing. This added resonance to the resulting creations, but any meanings that were thus accrued were, by dint of habit, likely to become diluted and then generally forgotten, even if on occasion revived by later designers. – None of the objects recruited as sources of inspiration were relevant at the inception of the architectural elements that came to resemble them. The affinities noted did NOT arise because these were from the outset intended to imitate or represent their counterpart objects. The artistic process can be likened to processes that occur in the cultivation of trees and plants, as when a variety appreciated for its fruit is grafted onto an established rootstock. The resulting designs are in effect hybrid in nature, even if unity of conception is conveyed by the fashioning of them out of single blocks of stone, and the cumulative effect of aesthetic improvements made over time. Over the course of such development the formal characteristics once borrowed from objects could be ironed out to the extent of no longer being recognisable. Presenting the case for a change of paradigm risks overstating the case and creating a lack of balance, so some re-balancing is called for. By highlighting artistic,

AFFINITIES BETWEEN THE TECTONIC ARTS & THE TECHNICAL ARTS OF ANCIENT GREECE

ornamental, associational, and cultural aspects of design, the present approach admits eclecticism and enlarges the frame of debate beyond traditional interpretations of the orders in terms of materials, construction, structure, and technological progress. But this is not at all to negate their importance. Technical considerations undoubtedly impacted on the development of Greek architecture; explanations that involve them only become problematic when pursued in an overly literal manner and at the expense of other aspects of design. The key to understanding how disparate aspects can each play a role lies in the appreciation of the multifaceted nature of architectural design in general, applicable to the Greek period as in any other. Structure and construction, influences whether from abroad or from objects, considerations of appearance, symbolism, and craft could all have an effect (on the craft aspect see Scahill 2017). Different modalities of design could also overlap and interpenetrate, whether continuity with tradition, revival or invention (Østby 2006; cf. Wilson Jones 2014: 211). Furthermore architecture, like art and culture more generally, can accommodate a quotient of ambiguity and contradiction. The logic of mutual exclusion has little place in understanding such phenomena. Rather we may interpret elements such as triglyphs as petrified devices of a composite character. They were at one and the same time decorative, symbolic, and tectonic in nature. Tripods conflated aspects inspired by Mykenaian frieze dividers on the one hand and tripods on the other, but this does not exclude them or any precursors being used to cover the ends of timber beams, or becoming associated with beams by virtue of their disposition (Wilson Jones 2016a). The considerations leading to the varied manifestations of Doric, Ionic, and Korinthian capitals were no less complex. There will always be difficulties in reconciling details with the big picture, but it is a necessary effort if we are to understand any artistic movement as vigorous as was the architecture of ancient Greece, progenitor of that of Rome, the Renaissance, and the Greek Revival. The ancient monuments inspired architects of the calibre of Stuart, Schinkel, Semper, and Hansen, and we should heed certain ideas of theirs and their contemporaries as we process material unknown to them with techniques unknown to them. Personally I found laser scanning and digital modelling invaluable for reconstructing Ionic capitals and appreciating the mutation from the crown-like to the cup-like echinos described earlier (Herdt & Wilson Jones 2008; 2010). My hope, then, for future research on Greek architecture lies in the reconciliation of insights born of understanding design with findings tested with scientific rigor. I dream of what might be achieved if we

195

could harness the most worthwhile earlier ideas to modern methodologies, or in other words create a fusion of the architectural and the archaeological. List of References Allies, B., 2015: Review of Wilson Jones 2014, Architecture Today Oct. 2015: 10–11. Amy, R., Seyrig, H. & Will, E., 1975: Le temple du Bel a Palmyre (Paris). Aruz, J., 2014, “Art and Networks of Interaction Across the Mediterranean”, in Aruz, J. Graff, S.B. & Rakic, Y. (edd.), Assyria to Iberia at the Dawn of the Classical Age (New York) 112–24. Barletta, B.A., 2001: The Origins of the Greek Architectural Orders (Cambridge). Barletta, B.A., 2011: “State of the Discipline: Greek Architecture”, AJA 115: 611–40. Barletta, B.A., 2014: Review of Wilson Jones 2014, The Art Newspaper 260 (Sept.): 66. Barresi, P., 2016: Review of Wilson Jones 2014, Archeologia Classica 67: 521–47. Bartoli, A. 1914–22. I monumenti antichi di Roma nei disegni degli Uffizi di Firenze (Rome). Bauer, H., 1973: Korinthische Kapitelle des 4. und 3. Jahrhunderts v. Chr. AM Beiheft 3 (Berlin). Bauer, H., 1974: “Das Kapitell des Apollo Palatinus-Tempels”, RM 76: 183–204. Beazley, J.D., 1938: Attic White Lekythoi (Oxford). Berti, F. & Masturzo, N., 2000: “Aree di culto ed elementi architettonici di periodo arcaico a Iasos (Caria)”, in F. Krinzinger (ed.), Die Ägäis und das westliche Mittelmeer. Beziehungen und Wechselwirkungen 8. bis 5. Jh. v. Chr. (Vienna) 217–29. Berve, H. & Gruben, G., 1963: Greek Temples, Theatres, and Shrines. Trans. by R. Waterhouse (New York). Borchhardt, J., 1976: Die Bauskulptur des Heroons von Limyra. Das Grabmal des lykischen Königs Perikles. Istanbuler Forschungen 32 (Berlin). Bötticher, K., 1846: “Das Prinzip der hellenischen und germanischen Bauweise hinsichtlich der Übertragung in die Bauweise unserer Tage”, Allgemeine Bauzeitung 11: 111–25 [translated in Foged and Hvejsel 2018: 1–19]. Bötticher, K., 1874: Die Tektonik der Hellenen I: die Lehre der tektonischen Kunstformen. Dorische, ionische und korintische Bauweise. 1st ed. 1844–52 (Berlin). Chipiez, C., 1876: Histoire critique des origines et de la formation des ordres grecs (Paris). Choisy, A., 1899: Histoire de l’architecture. 2 vols. (Paris). Corso, A. & Romano, E., 1997: Vitruvio. De Architectura, trans. with commentary. Edited by Pierre Gros. 2 vols. (Turin).

196 Corso, A., 2005: “De Virginum poculis. Considerazioni su Vitruvio 4,1,9”, Materiali e discussioni per l’analisi dei testi classici 55: 193–203. Costamagna, L. & Sabbione, C., 1990: Una città in Magna Grecia. Locri Epizefiri: guida archeologica (Reggio Calabria). Coulton, J.J., 1977: Ancient Greek Architects at Work: Problems of Structure and Design (Ithaca, NY). De Figueiredo, P., 2015: Review of Wilson Jones 2014, Context 141 (Sept.): 56. Dinsmoor, W.B., 1950: The Architecture of Ancient Greece. 3rd ed. (London). Dirschedl, U., 2013: Die griechischen Säulenbasen. AF 28 (Wiesbaden). Drerup, H., 1952: “Architektur und Toreutik in der griechischen Frühzeit”, AM 5: 7–38. Foged, I.W. & Hvejsel, M.F. (edd.), 2018: Reader on Tectonics in Architecture (Aalborg). Gallet De Santerre, H. & Tréheux, J., 1947–48: “Rapport sur le dépôt êgéen et géométrique de l’Artémesion à Délos”, BCH 71–2: 148–254. Giuliano, A., 1994: “Vitruvio e l’acanto”, Palladio 14: 29–36. Gros, P., 1993: “Situation stylistique et chronologique du chapiteau de Vitruve”, in L’acanthe dans la sculpture monumentale de l’Antiquité à la Renaissance. Mémoires de la section d’Archéologie et d’histoire de l’Art 4 (Paris) 27–37. Gros, P., 2018: Review of Wilson Jones 2014, Annali di Architettura 29: 176–78. Gros, P., Callebat, L., Cam, M.-T., Fleury, P. Jacquemard, C., Liou, B., Saliou, C., Soubiran, J. & Zuinghedau, M., 2015: De l’architecture = De architectura (Editio minor) (Paris). Haselberger, L., 1980: “Werkzeichnungen am Jüngeren Didymeion—Vorbericht”, IstMitt 30: 191–215. Haselberger, L., 1983: “The Construction Plans for the Temple of Apollo at Didyma”, Scientific American 253.6: 126–32. Heilmeyer, W.-D., 1970: Korinthische Normalkapitelle. Studien zur Geschichte der römischen Architekturdekoration. RM Supp. 16 (Heidelberg). Hellmann, M.-C., 2002: L’architecture grecque vol. 1: les principes de la construction (Paris). Hellmann, M.-C., 2006: L’architecture grecque vol. 2: architecture religieuse et funéraire (Paris). Hellmann, M.-C., 2015: Review of Wilson Jones 2014, RA 1 (n° 59): 43–8. Herdt, G. & Wilson Jones, M., 2008: “Scanning ancient Building Components”, Photogrammetrie Fernerkundung Geoinformation 4: 245–51. Herdt, G. & Wilson Jones, M., 2010: “Neue Techniken zur Bauaufnahme”, AntW 1: 78–83. Hölbl, G., 1984: “Ägyptischer Einfluß in der griechischen Architektur”, ÖJh 55: 1–18.

Wilson Jones Homolle, T., 1916: “L’origine du chapiteau corinthien”, RA 1916: 17–60. Homolle, T., 1917: “L’origine des caryatides”, RA 1917: 1–67. Hourihane, C. (ed.), 2012: From Minor to Major: The Minor Arts in Medieval Art History (Princeton). Hueber, F., 1998: “Werkrisse, Vorzeichnungen und Meβmarken am Bühnengebäude des Theaters von Aphrodisias”, AntW 29: 439–45. Jones, O., 1865: The Grammar of Ornament (London). Klein, N., 2015: Review of Wilson Jones 2014, JSAH 2015 (Sept.): 365–67. Koenigs, W., 1996: “Rundaltäre aus Milet”, IstMitt 46: 141–46. Kunze-Götte, E., 1984: “Akanthussäule und Grabmonument in der Darstellung eines Lekythenmalers”, AM 99: 185–97. Kurtz, D.C., 1975: Athenian White Lekythoi: Patterns and Painters (Oxford). Lapatin, K., 2001: Chryselephantine Statuary in the Ancient Mediterranean World (Oxford). Lauter-Bufe, H., 1987: Die Geschichte des sikeliotisch-korintischen Kapitells. Der sogenannte italisch-republikanische Typus (Mainz am Rhein). Lissarrague, F., 1995: “Women, Boxes, Containers: Some Signs and Metaphors”, in Reeder, E.D. (ed.), Pandora: Women in Classical Greece, Baltimore (Princeton) 91–101. Liverani, P., 1995: “Le colonne e il capitello in bronzo d’età romana dell’altare del SS. Sacramento in Laterano. Analisi archeologica e problematica storica”, RendPontAcc LXV (anno accademico 1992–1993): 75–99. Luce, J.-M., 2008: L’aire du pilier des Rhodiens ( fouille 1990–1992) à la frontière du profane et du sacré. FdD II.13 (Paris). Luschey, H., 1939: Die Phiale (Bleicherode am Harz). Marazov, I., 1989: The Rogozen treasure (Sofia). Marconi, C., 2004: “Kosmos. The imagery of the Archaic Greek Temple”, RES 45: 209–24. Mertens, D., 1984: “Zum Entwurf des Parthenon”, in Berger, E. (ed.), Parthenon-Kongreß Basel: Referate und Berichte, 4. Bis 8. April 1982, vol. 1 (Mainz am Rhein) 55–67. Mertens, D., 1993: Der alte Heratempel in Paestum und die archaische Baukunst in Unteritalien (Mainz am Rhein). Oakley, J.H., 2004: Picturing Death in Classical Athens: The Evidence of the White Lekythoi (Cambridge). Oeschlin, W. & Nerdinger, W., 2003: Gottfried Semper 1803–1879. Architektur und Wissenshchaft (Zurich). Østby, E., 1978: “The Temple of Casa Marafioti at Locri and Some Related Buildings”, ActaAArtH 8: 25–47. Østby, E., 2006: “Continuatio, Renovatio and Innovatio: The Birth of the Doric Temple”, Acta ad Archaeologiam et Artium Historiam Pertinentia 20: 9–38. Patay-Horváth, A. 2015: Review of Wilson Jones 2014, BMCR 2015.09.56.

AFFINITIES BETWEEN THE TECTONIC ARTS & THE TECHNICAL ARTS OF ANCIENT GREECE Penrose, F.C., 1888: An Investigation of the Principles of Athenian Architecture, or the Results of a Survey Conducted Chiefly With Reference to the Optical Refinements Exhibited in the Construction of the Ancient Buildings at Athens (London). Pfrommer, M., 1987: Studien zu alexandrinischer und großgriechischer Toreutik frühhellenistischer Zeit (Berlin). Poursat, J.-C., 1977: Catalogue des ivoires mycéniens du musée national d’Athènes (Paris). Rakob, F. & Heilmeyer, W.-D., 1973: Der Rundtempel am Tiber in Rom (Mainz am Rhein). Riegl, A., 1992 [1893]: Problems of Style. Foundations for a History of Ornament, trans. E. Kain (with annotations and introduction by D. Castriota). 1st German ed. 1893 (Princeton). Roaf, M., 1996: “Architecture and Furniture”, in Herrmann, G. (ed.), Furniture of Western Asia. Ancient and Traditional (Mainz am Rhein) 21–28. Rojas, F., 2015: Review of Wilson Jones 2014, JHS 135: 269–70. Rossignani, M.P., 1969: “La decorazione architettonica in bronzo nel mondo romano”, in Contributi dell’Istituto di archeologia (edd.), Università Cattolica del Sacro Cuore II (Milan) 44–98. Rumscheid, F., 1994: Untersuchungen zur kleinasiatischen Bauornamentik des Hellenismus. Beiträge zur Erschliessung hellenistischer und kaiserzeitlicher Skulptur und Architektur 14 (Mainz am Rhein). Rykwert, J., 1996: The Dancing Column. On Order in Architecture (Cambridge, MA). Saglio, E. & Daremberg, V., 1877: Dictionnaire des Antiquités grecques et romaines (Paris). Sapirstein, P., 2016: Review of Wilson Jones 2014, CR 67.1: 216–18. Scahill, D., 2009: “The Origins of the Corinthian capital”, in Schultz, P. & von den Hoff, R. (edd.), Structure, Image, Ornament: Architectural Sculpture in the Greek World (Oxford) 40–53. Scahill, D., 2017: “Craftsmen and Technologies in the Corinthia: The Development of the Doric Order”, in Handberg, S. & Gadolou, A. (edd.), Material Koinai in the Greek Early Iron Age and Archaic Period (Aarhus) 221–24. Schmidt, E., 1982: Geschichte der Karyatide. Funktion und Bedeutung der menschlichen Träger und Stützfigur in der Baukunst (Würzburg). Schmidt-Colinet, A., 1977: Antike Stützfiguren. Untersuchungen zu Typus und Bedeutung der menschengestaltigen Architekturstütze in der griechischen und römischen Kunst (Frankfurt). Sciacca, F., 2005: Patere baccellate in bronzo: Oriente, Grecia, Italia in età orientalizzante (Rome). Semper, G., 2007 [1860–63]: Style in the Technical and Tectonic Arts, trans. with commentary by Harry F. Mallgrave of Der Stil in den technischen und tektonischen Künsten. 1st ed. 1861–63 (Los Angeles).

197

Shefton, B., 2000: “The ‘Philistine’ Graves at Gezer and the White Lotus Ornament”, in Tsetskhladze, G.R., Prag, A.J.N.W. & Snodgrass, A.M. (edd.), Periplous. Papers on Classical Art and Archaeology presented to Sir John Boardman (London) 276–83. Smith, W.R., 1889: Lectures on the Religion of the Semites (London). Stackelberg, O.M., 1835: Die Gräber der Griechen in Bildwerken und Vasengemälden (Berlin). Svensen-Evers, H., 1996: Die griechischen Architekten archaischer und klassischer Zeit (Frankfurt am Main). Theodorescu, D., 1980: Le chapiteau ionique grec. Essai monographique (Geneva). Tölle-Kastenbein, R., 1974: Das Kastro Tigani. Die Bauten und Funde griechischer, römischer und byzantinischer Zeit. Samos XIV (Bonn). Tomlinson, R.A., 1976: Greek Sanctuaries (London). Vickers, M.J., 1994: Artful crafts: ancient Greek silverware and pottery (Oxford). Vickers, M.J., 1999: Skeumorphismus oder die Kunst, aus wenig viel zu machen (Mainz am Rhein). Vickers, M.J., 2014: “The Caryatids on the Erechtheum at Athens. Questions of chronology and symbolism”, Miscellanea Anthropologica et Sociologica 15.3: 119–133. Viollet-Le-Duc, E.E., 1858: Entretiens sur l’architecture. 2 vols (Paris). Von Hesberg, H., 1981: “Lo sviluppo del ordine corinzio in età tardo-repubblicana”, in L’art decoratif à Rome à la fin de la Repúblique et au debut du Principat. CollEFR 55 (Rome) 19–33. Von Normann, A., 1996: Architekturtoreutik in der Antike. Quellen und Forschungen zur antiken Welt 25 (Munich). Wagner, O., 1988 [1902]: Modern Architecture. 1st German ed. 1902 (Santa Monica). Waite, S., 2016: “An Attic Red-Figure Kalathos in the Shefton Collection”, in Boardman, J., Parkin, A. & Waite, S. (edd.), On the Fascination of Objects (Oxford) 31–62. Wesenberg, B., 1971: Kapitelle und Basen: Beobachtungen zur Entstehung der griechischen Säulenform. BJb Beihefte 32 (Düsseldorf). Wesenberg, B., 1996: “Die Entstehung der griechischen Säulenund Gebälkformen in der literarischen Überlieferung der Antike”, in Schwandner, E.-L. (ed.), Säule und Gebälk, zu Struktur und Wandlungsprozeß griechisch-römischer Architektur (Mainz am Rhein) 1–15. Wesenberg, B., 1999: “Virginis peculia. Zu Vitruvs Aitiologie des korinthischen Kapitells”, AA 1999: 313–15. Williams, R.T., 1961: “An Attic red-figure Kalathos”, AntK 1: 27–29. Wilson Jones, M., 1991: “Designing the Roman Corinthian capital”, PBSR 59: 89–150.

198 Wilson Jones, M., 1999: “The practicalities of Roman entasis”, in Haselberger, L. (ed.), Appearance and Essence: Refinements of Classical Architecture—Curvature: Proceedings of the Second Williams Symposium on Classical Architecture (Philadelphia) 225–249. Wilson Jones, M., 2000: Principles of Roman Architecture (New Haven and London). Wilson Jones, M., 2002: “Tripods, Triglyphs, and the Origin of the Doric Frieze”, AJA 106.3: 353–90. Wilson Jones, M., 2014: Origins of Classical Architecture. Temples, Orders and Gifts to the Gods in Ancient Greece (New Haven). Wilson Jones, M., 2016a: “Reflections on Interpretation: the origins of the Doric frieze”, in Zambas, C., Lambrinoudakis, V.,

Wilson Jones Simantoni-Bournia, E. & Ohnesorg, A. (edd.), ΑΡΧΙΤΕΚΤΩΝ: Honorary Volume for Professor Manolis Korres (Melissa) 645–58. Wilson Jones, M., 2016b: “‘Models of interpreting visual culture’: Wilson Jones on Patay-Horváth on Wilson Jones”, BMCR 2016.01.17. Wilson Jones, M., 2017: “Non-architectural sources for Greek and Roman architectural form”, in Pensabene, P., Milella, M. & Caprioli, F. (edd.), Decor. Decorazione e architettura nel mondo romano. Atti del Convegno Internazionale, Roma, 21–24 maggio 2014. Thiasos Monografie 9 (Rome) 87–102.

chapter 12

More than War: Symbolic Functions of Greek Fortifications Silke Müth In recent years, more and more scholars have started to take a second glance at the functions of fortifications, noting that with some examples there might be something beyond their defensive purposes. The semantic values of fortifications thus have come more into focus and are considered as important but mostly secondary functions of some of these monuments. Nevertheless, the symbolic meaning of fortifications has never before been studied on a broader basis using a representative number of monuments with an analysis and comparison of forms of expression; historical, political, and cultural backgrounds; and the senders, addressees, and the possible content of the symbolic messages as well as the development of these features over time. As I will present in the following pages, however, such an approach may show that symbolic aspects of fortifications were manifold and versatile, and represented a much more comprehensive phenomenon than is generally assumed.1 My focus is on the original meaning that builders or patrons bestowed on the monuments while constructing, restoring, or adding to them rather than on the possible symbolic functions the buildings acquired or developed in the course of their every-day use. As a result, I concentrate not so much on architectural agency but rather on the various aspects of symbolising through architecture.

1  This article is based on research on symbolic functions of Greek and Roman fortifications carried out within a three-years’ fellowship of the Deutsche Forschungsgemeinschaft (German Research Association; reference-no. of the project: MU 2992/2-1) at the German Archaeological Institute in Athens from 2012 to 2015, and a six-months’ fellowship from the same fund at Humboldt University in Berlin from 2015 to 2016. The publication of the results as a book are currently under preparation. The subject has its original roots in a work group of the international research network Fokus Fortifikation. Ancient fortifications in the Eastern Mediterranean, see Müth, Laufer & Brasse 2016; Müth 2016a. Research on the city wall of Messene was conducted in the years 2004–2008 by the author together with Jürgen Giese, Ute Schwertheim, and Jean-Claude Bessac in a project of the Free University of Berlin in cooperation with the excavations of the Archaeological Society of Greece under Petros Themelis. The project was generously funded by the Gerda Henkel foundation (Düsseldorf, Germany), a starting campaign in 2004 by the Free University of Berlin. The final publication is currently under preparation: Müth & Bessac, forthcoming.

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004416659_014

1 Terminology It is necessary to start with a few comments on terminology, as the meaning of symbol is not always simple and unambiguous. When I speak of a symbol, I am using the term in a comprehensive way, in the sense of any sort of sign. My aim is to find out in which cases ancient fortifications assumed the function of signs and of communicating messages, how precisely this was achieved, and what content such messages carried. The founder of the theory of signs, Charles Sanders Peirce, in particular identified three forms of signs (1983: 64–67 and 1986: 375, 409–30): (a) the icon, which exhibits a formal resemblance of the object it denotes and thus is a sort of effigy or image; (b) the index, which has a causal connection to its object; and (c) the symbol, which is connected to its object through a mere social or cultural convention (cf. Burmeister 2009: 76–80 and Berger 2009: 44; on the use of the term “Icon” by Charles William Morris cf. Morris 1972: 15–88; Schneider, Fehr & Meyer 1979: 11–13). If we consider these classes in the case of fortifications, it is obvious that fortifications did not primarily function as icons—at least if we are looking at the fortifications themselves and do not include pictorial representations of them. A fortification may, however, function as an index: e.g., a city wall has always a causal connection to a certain economic capacity of its builder, of which it may act as a sign; or it may indicate its builder through certain characteristic methods of construction or features. As a symbol in the sense of Peirce’s definition, a fortification may be used in many different ways which we will see later. However, further scrutiny reveals that the categories icon and index are also dependent on social or cultural conventions, because it is only a matter of one’s perspective whether a sign shows a likeness or a causal connection to the object it denotes. Between two arbitrary objects, it is always possible to find some similarities or causal connections if one really makes an effort. Thus, icon and index represent sub-categories of symbols, which do not generate a relevant theoretical difference, as Christoph Baumberger has pointed out in his symbol theory for architecture— to which I will return below (Baumberger 2010: 26–28; cf. Schneider, Fehr & Meyer 1979: 11–13). In this way it is legitimate to call any of these signs symbols, as I will do from

200 now on. In any case, I am focusing on cases where fortifications are, on the basis of social or cultural conventions, deliberately employed as signs by their builders, patrons or by someone later applying changes, whether they used likenesses or causal connections or not; thus, the icon and index functions are not a central part of my study. Moreover, the term representative needs some consideration: I am using it as a kind of synonym for symbolic, because to represent something in its original meaning is to stand for something else, i.e., to be a sign or a symbol for something—a meaning corresponding to Baumberger‘s “zweistellige Verwendung” (binary use) of the term (Baumberger 2010: 416–19). In our linguistic conventions, representative is often used for qualifying an object without meaning any reference of it to something else, in order to denominate some impressive effect of the object, some “showing-off”, which in its deeper sense also means to symbolise something, to communicate something beyond the original object—equivalent to Baumberger’s “einstellige Verwendung” (monadic use) of the term (Baumberger 2010: 416–19). 2 Methodology After this small excursion into terminology, some methodological matters need to be considered. The most important of these questions is how symbolic functions may be determined and distilled from all of the practical functions of a fortification. As the original purpose of a fortification is defence, its defensive efficacy must be studied first (see Jansen 2016 on the evaluation of the defensive attributes of fortifications). Whatever serves a defensive purpose has a practical function and cannot in itself belong to the symbolic level. For an evaluation of the defensive qualities of a fortification, its wider context needs to be considered. For instance, it is paramount to know the status quo of siege and fortification technology in the period the monument was constructed. Moreover, it is important to find out against whom or what it was supposed to protect, as it makes a huge difference whether it was built to defend against wild animals, natural forces, bandits, or organised armies. Apart from defence, a fortification may also have additional practical functions, for instance of an urban nature (Müth 2016b) in the case of a city wall or a diateichisma (Sokolicek 2009). It may constitute a border between a settlement and its impure areas, like nekropoleis or certain industries; it may separate different quarters of a town from each other; it may canalise and regulate the traffic between a settlement and its surroundings; and it may have economic functions, such as a boundary for

Müth

collecting tolls or other duties—although this does not seem to have been a common practice in Greek times (Hansen 1997: 52 and 2006: 104; the only evidence is Hsch. s.v. διαπύλιον). Although such urban functions might have been secondary, they should not be ignored. It may seem too simplistic to state that everything remaining after the deduction of the practical purposes of a fortification has a potential symbolic value. I believe, however, that this is indeed the best way to determine whether a fortification could have carried symbolic meaning— with the restriction, however, that one cannot claim it would be possible immediately to grasp the symbolic functions through this approach. One can only collect a number of features in this way that may indicate symbolic functions, which nevertheless needs to be cross-checked with additional evidence. Furthermore, the evaluation of a fortification or one of its features as symbolic is not enough, but it is necessary to find out more exactly about what kind of symbolic message was sent through the fortification, by whom this was done, and who could have been addressed—i.e., to gain more information about the semantics, the senders, and the possible addressees. Moreover, there are many overlaps between the practical and symbolic functions of fortifications to be taken into consideration, and it is mostly not possible to draw a clear line between the two. If, for instance, a defensive building is erected in a particularly strong fashion to resist or frighten off an enemy, it automatically evokes the impression of monumentality and representativeness with the everyday beholder, as was certainly the case with the powerful defences of Rhodos (Filimonos-Tsopotou 2004) at the beginning of the third century BCE (for more examples, see Müth, Laufer & Brasse 2016: 152 [B1]). If a fortification follows the crests and summits of hills due to defensive reasons, it is by default visible from afar and creates an impressive image for anyone approaching it, and this might well be interpreted as purposeful staging. An example of this effect is the extant northwestern part of the late-Classical city wall of Messene (Fig. 12.1) (Müth 2010a; Giese 2010; Giese & Müth 2016; Müth & Bessac, forthcoming; for more examples, see Müth, Laufer & Brasse 2016: 150 [A1]). Masonry forms regarded as particularly stable were also often executed in a very regular and precise manner. As such, they also display an aesthetic and decorative effect, such as the particularly accurate Lykourgan walls of Eleusis (Fig. 12.2) (Winter 1971: 78–80; Adam 1982: 197). And finally, a gate built in the axis of a main thoroughfare of a town not only has a very practical function for urban traffic, but at the same time offers together with the street and buildings an impressive effect as an architectural ensemble, both from the outside and from the inside—as, for example, the Dipylon and the

MORE THAN WAR: SYMBOLIC FUNCTIONS OF GREEK FORTIFICATIONS

figure 12.1

Messene: Northern part of west wall seen from the west photo by Jürgen Giese

figure 12.2

Eleusis: Lykourgan tower of the fourth century BCE with adjacent curtain photo by author

201

202 Sacred Gates in Athenai certainly did (Gruben & Müller 2018; Kuhn 2017; Stroszeck 2014: 69–81; Knigge 1988: 9 fig. 1, 56–73, 157–65). In such cases, one needs to ponder very carefully whether symbolic functions are probable—i.e., if there is evidence that the properties inherent to the fortification because of its defensive or urban function are demonstrably emphasised. Categorising symbolic functions is possible in different ways. One may, for example, categorise the evidence of symbolic functions at a fortification or the ways of symbolising, which I will present below. Evidence for symbolic functions can be discovered on various levels. As a first possibility, it might appear in the general strategic concept, the dimensions, and the course of the wall. There may be over- or under-scaling: the walls of Herakleia upon Latmos from the late fourth or early third century BCE run high up into the mountainous terrain beyond any defensive sense and, moreover, are set in scenographic but impractical way over each rocky outcrop (see Krischen 1922, on the city wall of Herakleia upon Latmos; also Müth, Laufer & Brasse 2016: 142–43 with fig. 9 for an illustration; Müth 2016a: 188–89 with fig. 5). Obviously the founder of the city—whether it was Pleistarchos, Demetrios, or some other Hellenistic ruler—wanted to set up a monument reflecting his power and potential rather than a purely practical defensive construction (for a discussion who founded the city and built the city wall, see Hülden 2000). This concept soon proved unsuccessful, because the walls were too long to be defended and already by the third century BCE needed to be shortened considerably by a diateichisma. Likewise, under-scaling may point to a symbolic meaning, especially if a wall could not fulfill a full defensive purpose. The wall of Pednelissos from the second or first century BCE with its very narrow curtains and wall-walks would not have been able to withstand an army equipped with up-to-date artillery. In the quality of execution, however, it shows a high aesthetic pretense, leading Eric Laufer (2010: 180–93) to interpret it as a symbol of urban wealth, independence, and modern quality of life. It is also on the level of the general strategic concept if towers are concentrated at only certain points of a wall without any defensive reasons. In the city wall of Akarnanian Torybaia, dated to the late fifth or early fourth century BCE, Judith Ley has identified towers only in the southern section, which was highly visible from the most important approach to the city, but less vulnerable to attack than positions in the north or east. The reason for this disposition must have been to impress approaching people with the wall’s apparent strength (Ley 2009: 296).

Müth

At the level of the single components of a fortification, the form of towers, curtains or gates can point to symbolic functions. Again at Herakleia upon Latmos one may observe huge artillery towers rising in a rocky terrain where no enemy would ever have been able to transport heavy artillery or siege machinery (Fig. 12.3) (Krischen 1922). Already around 300 BCE, the gate of Zeus and Hera at Thasos was equipped with a façade on its city-side by the wealthy citizen Pythippos (Fig. 12.4) (Grandjean 2011: 265–66, figs. 392–93, 423). The rich architectural decoration of this gate is unparalleled at the time and constitutes an instance of grandiose private representation at a city wall. An example of a curtain with representative aspects is found at Torybaia in Akarnania: close to a main gate, a decorative geison band has been inserted at the height of the wall walk (Ley 2009: 296). Additional evidence for possible symbolic functions may be identified at the level of construction details or the finishing of blocks. Larisa upon Hermos provides an example of the purposeful choice of material: already in the late Archaic city wall, multi-coloured blocks are combined in a highly effective way (Saner, Sağ & Denktaş 2016; Müth 2016a: 188–89 with fig. 9; Müth, Laufer & Brasse 2016: 139–40 with fig. 6; Frederiksen 2011: 72; Winter 1971: 78–79; and Boehlau & Schefold 1940: 48 pl. 33). Also masonry forms can be used to create a representative effect. We can see this at the West Gate of Eretria, where in the early fifth century BCE a polygonal indented cutting was made with considerable labour between the lowermost course of orthostates and the course above (Müth, Laufer & Brasse 2016: 139–40 with fig. 7; Fachard 2004: 99, pl. 12, 4; and Krause 1972: pl. 1–2, 125–26). The block finishing can be done in a decorative manner as well, as is the case, for instance, with a tower of the fourth century BCE in Eleusis (Adam 1982: 197) and at the Arkadian Gate in Messene, where we can find the same decoration with offset rows of vertical grooves, a smooth flat band below, and carefully dressed orthostates (Fig. 12.5) (Giese 2010: 86, fig. 1 and Müth 2010a: 82). All these aesthetic effects reveal that these monuments were not mere defensive constructions, but also were intended to show the wealth and pride of the cities which built them. If we want to categorise the ways of symbolising through fortifications, we need to review the field of symbol theory, especially in architecture. Christoph Baumberger, who published his PhD in Zurich with the title “Gebaute Zeichen” (Constructed Signs) on this subject in 2010, developed a symbol theory of architecture on the basis of Nelson Goodman’s general framework (Baumberger 2010). With some minor limitations, this may well be applied

MORE THAN WAR: SYMBOLIC FUNCTIONS OF GREEK FORTIFICATIONS

figure 12.3

203

Herakleia upon Latmos: Artillery towers high up in the rocky terrain photo by author

to ancient fortifications. Baumberger discerns four different ways of symbolising: “Denotation” (denotation), “Exemplifikation” (exemplification), “Ausdruck” (expression) and “Anspielung” (allusion). Denotation means the reference by a building to an object, monument, or event (Baumberger 2010: 69–160). Attachments at a building might denote something: in the case of fortifications, such attachments could be inscriptions, reliefs, or sculptures that refer to particular objects or persons. An example for an inscription can be found on the architrave of the new façade of the gate of Zeus and Hera at Thasos, where Pythippos is named as the donor, emphasising his role as urban euergetes (Grandjean 2011: 265–66, fig. 423, and Grandjean, Kozelj & Salviat 2004– 2005: 232). On the same wall we can find examples of reliefs, since many of its gates have carvings denoting the most important gods of the city (Grandjean 2011: 214–6; Holtzmann 1994; Picard 1962). As for statues, an instructive example is at the South Gate of the fort at Kalýva,

northwest of Xanthi in Thrake: herms of Herakles and Hermes were mounted in a niche in the wall of the gate courtyard, where they obviously referred to their role as the patrons and protectors of the soldiers stationed there, as Michael Weißl (2012: 235–36) has observed. Also parts of the building or the building itself may denote something. This is for instance the case if a city wall imitates another wall or parts or it, if it represents a geometric form, or if a wall or one of its parts functions as an image of the whole town. The city wall would thus denote the original city wall or its parts, the geometric form, or the town itself. One example of imitation could be the Hellenistic tower gates in city walls of Pamphylia and southern Pisidia: in Sillyon and Kremna, this building type was probably developed because of a very narrow road, as Eric Laufer assumes. The type was, however, imitated in other towns like Pednelissos, and in this way the original monument was denoted (Laufer 2010: 179–80). The city wall of Olbia in Provence from the late fourth century BCE is very close

204

figure 12.4

Müth

Thasos: Reconstruction drawing of the city-side façade of the gate of Zeus and Hera T. Kozelj 2003; image after Grandjean 2011, fig. 423; courtesy of École Française d’Athènes

to the geometric form of a rectangle (see Coupry 1986), as are many other Hellenistic through later Roman walls, although Olbia is one of the very earliest examples of this kind. A city wall may also serve as a symbol of the city itself, if its image represents the city as in the late-Classical Lykian city-reliefs (see Frederiksen, Laufer & Müth 2016: 187, figs. 4–5; Marksteiner 1993; 1997: 159–62; Childs 1978:

91–106; and Wurster 1977) or if the sculptural personification of a city wears a mural crown as a symbol of it, as with the famous example from Antiokheia on the Orontes in the early third century BCE (see Meyer 2006 [for the personification of the city of Antiokheia]; Messerschmidt 2003: 81–85 [for Hellenistic and Roman coins with the Tyche of cities]).

MORE THAN WAR: SYMBOLIC FUNCTIONS OF GREEK FORTIFICATIONS

figure 12.5

205

Messene: Eastern side of the inner courtyard of the Arkadian Gate photo by author

The second way of symbolising, exemplification, means that some inherent aspects of the building itself are emphasised (Baumberger 2010: 161–262). The protective function of a fortification may be exemplified by a particularly strong appearance, which was certainly the case with the wall of Rhodos with its strong proteichisma and even stronger main wall, densely fitted with colossal towers (on the walls of Rhodos, see Filimonos-Tsopotou 2004). A city wall, however, also functioned as kosmos, the adornment of its city, as stated by Aristoteles (Pol. 7, 1331a), and this function could be emphasised through ornamental execution. This may in particular be observed on the late Archaic or early Classical wall of Aristoteles’ home town of Stageira, on the eastern coast of Khalkidike, which stands out through its multi-coloured material and precise and mannered forms of polygonal and ladder-pattern masonry, combined with an excellent stone finishing (Fig. 12.6). Ouellet (2016) points out that the ladder-pattern masonry, which elsewhere is used very often because of practical and economic reasons, here on the contrary shows a clearly representative purpose through its careful finishing of the thin slabs (on the walls of Stageira, see also Sismanidis 2003: 24–39). Such elaborate aesthetics are unparalleled in other fortifications of this time. It cannot

be pure coincidence that Aristoteles, who grew up within such aesthetically finished walls, later remarked that a wall should be an ornament of its city. Furthermore, the entrance function of a gate may be exemplified by a special emphasis on its accessibility, which can for instance be achieved by over-dimensioning or multiplying the entrances. This happened at the Arkadian and the South Gates of Messene, where double entrances much larger than the amount of traffic would have required were erected, which constituted, however, a weak point in terms of defence (Schwertheim 2010). Baumberger’s third and fourth ways of symbolising are more abstract than the first two ways: expression is defined as a metaphorical exemplification, which means that an aspect of a building is exemplified that the latter does not actually possess (Baumberger 2010: 263–78). This may be the case if a building expresses a feeling like gaiety or sadness—which it of course cannot possess in reality—or if a stone building, which is inherently heavy due to its material, conveys a feeling of lightness. Because the first instances of ancient fortifications to illustrate this form of symbolic expression belong to the much more elaborate fortification architecture of the Roman era, this point need not be further explicated here.

206

figure 12.6

Müth

The city wall of Stageira: (a) ladder-pattern masonry; (b) polygonal masonry photos by author

Allusion is defined as an indirect reference: this means that a symbol refers to something through a chain of references (Baumberger 2010: 379–452). In general, this way of symbolising does not yet occur in Greek fortifications either, but for one of its special forms, the so-called

“tatsächliches architektonisches Zitat”, the actual architectural citation, which means that a building actually contains a part or a copy of the object it denotes. This is the case if spolia are used as a purposeful reference to their original building context. A famous example for this

MORE THAN WAR: SYMBOLIC FUNCTIONS OF GREEK FORTIFICATIONS

is the Themistoklean north wall of the Akropolis, where architectural members of the Pre-Parthenon and the Old Athena Temple were used in a clearly recognisable way. These members were visible from afar and reminded people of the devastation wrought by the Persians but also the final victory of the Greeks, including the central role played by Athenai. This deliberate use of the old sacred architecture transformed the Akropolis wall into a historical memorial (Di Cesare 2004; Lindenlauf 2003; Bäbler 2001; Stähler 1993: 17–24; and Chapter 13 in this volume). 3

The Development of Symbolic Functions of Greek Fortifications over Time

I will now proceed to a short chronological summary of the symbolic functions of Greek fortifications. In this context I will also draw attention to the senders and addressees of such messages as well as to their content—which means I will discuss the semantic aspects of Greek fortifications. Already in Geometric times we can identify symbolic aspects of fortifications. They are expressed mainly by monumentalised dimensions, as can be observed with the exceptionally massive city walls of Old Smyrna inaugurated in the ninth century BCE, and also on walls like that of Paphos on Kypros from the 8th century BCE. Curtain walls reached widths of 4–5 m or, in the case of Old Smyrna, 10 m, and enormous platforms or towers were added, which were at least in the case of Old Smyrna far larger than necessary to defend against the offensive weaponry of the time (Frederiksen 2011: 70–73, 177–79, 188–90, table 1–2, map 1). There are also, however, aesthetic aspects that can be observed already in this period: both at Old Smyrna and Paphos, the inner sides of the walls facing the settlements are more smoothly dressed than the outer sides, “which can hardly be explained as anything but aesthetic” (Frederiksen 2011: 72). There are no examples for such aesthetic aspects known from the Greek mainland in such early times. This may be understood in connection to the generally more progressive character of fortification walls in the eastern part of the Greek world during the Geometric period, probably due to the proximity to the strong eastern powers of Assyria and Lydia (Frederiksen 2011: 90, 94). In Archaic times, the precision and aesthetics of block finishing are more strongly accentuated. This is, for instance, reflected in Herodotos’ praise of the walls at Phokaia (1.163) as “made of great stones well fitted together” (Frederiksen 2011: 182–83; trans. after Godley 1920). At the transition to the Classical era, we encounter chromatic effects and decorative masonry forms at Larisa

207

upon Hermos and Stageira. Through all these means— monumental scales, aesthetically motivated wall finishes, and chromatic variations—power and wealth were advertised to external viewers, whereas viewers from the interior, the population living within the walls, were confronted with a suggestion of enhanced security within an aesthetically pleasing environment. To both sides, the perfection of craftsmanship was underlined, in the sense of the so-called Könnens-Bewusstsein, or “Consciousness of Ability”, a term adopted by Christian Meier (1978; 1990). In Stageira, which was a colony of Andros, there is probably also another motive to be found for the particularly beautiful application and finishing of the ladder-pattern masonry, consisting of thin stone slabs between larger blocks. This masonry form was also applied in a very elaborate manner to the walls of other Andrian colonies in the northern Aigaian, for instance Akanthos and Argilos, where it was used predominantly on the most visible sections and therefore obviously applied there for aesthetic or symbolic purposes, as a proper masonry style. Keven Ouellet (2016) interprets this as a deliberate reference to the building traditions of the metropolis of these colonies, Andros, to which it created a highly visible connection. Symbolic messages on a religious level are first attested on a massive scale at the early-fifth-century-BCE gates of Thasos with their singular reliefs depicting urban deities, which refer emblematically to the city gods and symbolise their protection for wall and town, as already emphasised by Olivier Picard (1962), Bernard Holtzmann (1994), and Yves Grandjean (2011: 214–6). In the early Classical era, the purposeful use of spolia emerges, inspired by the Persian destructions, a phenomenon which in the case of the northern Akropolis wall we can link to the memorial function of the fortification (see above). The employment of monumentally scaled city walls as signs of colonial power is also for the first time attested no later than the fifth century BCE. This can be observed well in the magnificent walls of Amphipolis in Thrake, built under Athenian influence in the 420s BCE (Fig. 12.7) (Lazaridis 2003). Stone was employed up to a considerable height, with excellently finished, partly pseudo-isodomic masonry that stood out clearly from that of other contemporary fortifications in its monumentality, regularity, and fine dressing. Athenai obviously wanted to signal its influence in this economically critical region. In Akarnania and Aitolia, we can observe first that towers are sited close to one another near highly frequented streets despite the lack of a defensive need—such as in the mid-fifth-century walls of Stratos at the main approach from Aitolia, or at the southern facade of the city wall of Torybaia from the turn of the fourth century BCE.

208

Müth

figure 12.7 Amphipolis: Pseudo-isodomic masonry at the north wall photo by author

It is also in these regions that curtains are particularly accentuated near gates, for instance through the geison course at the wall walk of Torybaia. Both phenomena must be understood as signals of power and wealth to the passersby, as correctly observed by Judith Ley (2009: 296). In the first third of the fourth century BCE, the fort of Euryalos at Syrakousai with its enormous five-tower battery (Beste 2016 and Beste & Mertens 2015: 127–240, 264– 69) foreshadows an otherwise Hellenistic phenomenon: the installation of monumental artillery batteries, which through their excessive scale served to frighten off and at the same time impress friend and foe. It is not by chance that this occurred under Dionysios I in Syrakousai, where at the same time the earliest catapult artillery was invented and soon further developed—a process that also inspired revolutionary changes in the techniques of fortification (Fig. 12.8) (Rihll 2007, Marsden 1969, and Garlan 1974). The five-tower battery was even embellished with monumental lion-head waterspouts on its upper part (Beste & Mertens 2015: 268–69, with 180 fig. 219). Through this, the mighty defensive structure was endowed with an aesthetic effect unique to this kind of building at the time in question and conveyed not only excessive military strength, but also the self-assurance and pride of the builder to erect a

monument with architectural merit. Towards the middle of the fourth century BCE, various forms of representation are attested at the city wall of Messene, one of them being consoles which are found under the lintels of all water outlets at the southern part of the fortifications as well as at the city-side facade of the Arkadian Gate (Fig. 12.9a–b). While the consoles at the water outlets also had a structural in addition to a decorative function, the consoles at the Arkadian Gate were purely decorative: on their upper surface, there is no indication for anything (e.g., statuettes, herms or the like) having been mounted on them, so the consoles themselves obviously served as decorative features (also see Müth 2010a: 82). The beautiful finish of the masonry at the Arkadian Gate as well as the monumental double doors at several gates in Messene, which nevertheless were disadvantageous for defensive purposes, have already been mentioned. Other representative aspects at this city wall are the door lintels of towers, which are decorated with the same offset rows of vertical grooves as the masonry blocks of the Arkadian Gate. Such decorated lintels are still in situ at tower no. 11 in the west wall; the same decoration can be found on blocks on the ground next to the scanty remains of the flanking structures no. 4, 22, and 23, and it is very likely from their form and length

MORE THAN WAR: SYMBOLIC FUNCTIONS OF GREEK FORTIFICATIONS

figure 12.8

209

The five-tower battery of the fort Euryalos at Syrakousai: (a) view from the field side; (b) reconstruction drawing of the field side H.-J. Beste, after Beste & Mertens 2015, fig. 219; Courtesy Reichert Verlag

that they also served as door lintels in these structures (see also Müth 2010a; and Giese & Müth 2016: 285, fig. 2 for a plan of the city wall of Messene with numbers of towers and flanking structures). Also representative is the fact that the two highest structures of the Ithome wall and the lower city wall, respectively, were built of precisely cut ashlar masonry in contrast to the trapezoidal masonry employed in the surrounding wall sections (Giese & Müth 2016: 285 fig. 2: Structures nos. 7 and 38). Messene was founded in 369 BCE by the Thebans under Epameinondas after they had liberated the Messenians from the Spartan yoke, and the representative city wall can be understood as a signal, directed in particular towards Sparta, of its newly acquired freedom and independence (see Müth

2014 for the historical circumstances and implications of the city wall of Messene). At the same time it was a strong symbol of Messenian identity which held together and strengthened its new population of diverse origins (see Müth 2010b for a discussion of the different groups that formed the population of the new city of Messene). The decorative consoles are paralleled on other fortifications erected under Theban influence during this period, such as those at Eleutherai and Siphai (Fig. 12.9c–d) (cf. Cooper 2000: 157–63 with fig. 6.3; Adam 1982: 70–71, figs. 35, 48, ill. 112; on Siphai: Schwandner 1977: 528–37; Cooper 2000: 179–83), and may with some caution be interpreted as a kind of marker of Theban power. Fachard (2013: 89), in discussing Eleutherai, points out rightly that the practice of

210

Müth

a

c

figure 12.9

b

d

Consoles in fortification architecture of the fourth century BCE: (a) Messene: western part of the city-side facade of the Arkadian Gate; (b) Messene: water outlet in the south wall; (c) Eleutherai: northern (interior) side of the South-East Gate; (d) Siphai: outer entrance of the Sea Gate, seen from the inside of the tower chamber all photos by author except 9b, by Ute Schwertheim

MORE THAN WAR: SYMBOLIC FUNCTIONS OF GREEK FORTIFICATIONS

decorating gates with consoles is not exclusively Boiotian, as this feature also appears at Euboian Dystos. This might, however, also be attributable to some Theban influence. At the start of the Hellenistic period, the development of torsion artillery prompted additional increases in the dimensions of defensive architecture, as may be observed in the city walls of Herakleia upon Latmos (Krischen 1922) and Rhodos (Filimonos-Tsopotou 2004). Even more enormous is the offensive battery complex erected in the late fourth century BCE on the north side of the akropolis of Selinous, whose dimensions were greatly enlarged in comparison to defensive works of earlier periods (K. Mathieu in Mertens 2003: 128–77). This powerful system clearly overshot the defensive needs of the site, and the desire to impress was obviously responsible for the excess. It is, however, important to point out in the context of exceedingly powerful bastions or walls that it also contributed to the defensive purpose of a fortification, to impress enemies with its strength and monumentality, which would have frightened off possible attackers from the start. In the architectural decoration of city walls, new symbolic forms were introduced with the private donation of a decorative facade for the gate of Zeus and Hera at Thasos around 300 BCE (see above). This kind of large-scale private representation at fortifications as well as the use of architectural or sculptural decoration became much more frequent in later Hellenistic and Roman times. Although the theory of “pompous Hellenistic gates” formulated by Hans Lauter (1986: 73–74), which is chiefly based on the gates of the Pamphylian cities Perge and Side, cannot be fully supported after recent proposals to downdate these two gates to the Roman period (Martini 2016; Lohner-Urban & Scherrer 2016), architectural decoration is nonetheless well attested on other Hellenistic gates in addition to the example at Thasos. For example, a number of city walls in Pamphylia and Pisidia probably dating from the second or first century BCE are not particularly strong defences, but on the contrary maintain a high aesthetic standard, which according to Eric Laufer implies a role as symbols of urban prosperity and independence. The main gate at Sillyon, for instance, was decorated with a Doric frieze, and the gates of Selge and Sagalassos included friezes depicting weapons. Furthermore, the towers and curtains of some of these Pamphylian and Pisidian cities incorporate elements which are obviously decorative (see Laufer 2010: esp. 175–76; and, for a short overview on weapon friezes at Pamphylian, Pisidian and Isaurian gates, Erol 2004–2005). Also the fortification of Isaura included a weapon frieze at a gate and string courses at towers, but it is not clear if this fortification dates to late Hellenistic or to early Augustan times (Winter 1971: 190–1; Swoboda, Keil & Knoll 1935:

211

119–25. The reservation of the insecure date must, however, also be expressed for the Pamphylian and Pisidian walls [Laufer 2010: 166]). An elaborately staged form of representation through fortifications by a Hellenistic ruler can be found in the palace-fortress on Mt. Karasis in Kilikia. Crowning this high, rocky, and very remote hill, the fortress shows clearly distinguished stages of representation from the outer wall and the Lower Castle up to the neatly finished masonry of the Upper Castle with its palace complex. A perfect example of architectural staging in an extreme landscape, the complex on Mt. Karasis is a mighty symbol of the sovereign’s power to any beholder (Radt 2010; 2016; Hoffmann 2011). 4 Conclusion In review, we should emphasise that the senders of symbolic messages are normally those who initiated the construction or reconstruction of the fortifications. We are only rarely able to grasp their identity from inscriptions or historical sources, and in most cases we are instead obliged to base our conclusions on the broader historical context. It is even more difficult to identify the recipients of these messages, who most often would have been the general public which perceived the monument. Only in rare cases can we deduct particular target groups through special circumstances, such as Sparta in the case of the wall of Messene. The different forms for expressing the symbolic functions in fortifications are dependent on the political and social contexts as well as on regional traditions of construction. In general, symbolic functions in Greek fortifications are more often conveyed through scale and the quality of execution than through abundant decoration, which is more characteristic of Roman fortifications. Symbolic messages are primarily directed towards the exterior, although in some cases the population of the fortified settlement itself is the receiver as well. Their content can partly be derived from their forms of expression, but we can understand them in their whole extent only by studying the specific contexts. We often cannot get beyond general assumptions and must content ourselves with the interpretation of these symbolic aspects as signals of force, power, influence, independence, wealth, or excellent craftsmanship, and as symbols of an aesthetic environment, of political, social or cultural status, or of divine protection. Only in some cases does the context help us to understand a fortification more explicitly as a sign of ethnic identity; regained independence or colonial

212

Müth

influence; private, public, or royal benefaction or representation; or as a historic monument or reflection of the horizon of urban deities. Symbolic and semantic aspects of Greek fortifications are various and manifold and cannot be seen as an exception, but rather as a much more consistent phenomenon than is generally assumed. In fact, most fortifications bear evidence for something beyond their practical functions—something more than war. Our perceptions of ancient fortifications should thus be expanded from their hitherto often limited interpretation as military defensive buildings to a wider consideration of fortifications as active parts of an ancient process of communication, as deliberately conceived elements of built space, and as mirrors of social self-conception. List of References Adam, J.-P., 1982: L’architecture militaire grecque (Paris). Bäbler, B., 2001: “Die archaischen attischen Grabstelen in der themistokleischen Stadtmauer: Grabschändung oder Apotropaion?”, Philologus 145.1: 3–15. Baumberger, C., 2010: Gebaute Zeichen: Eine Symboltheorie der Architektur (Frankfurt). Berger, A.A., 2009: What Objects Mean (Walnut Creek). Beste, H.-J., 2016: “The Castle Euryalos of Syracuse”, in Frederiksen et al. 2016, 193–206. Beste, H.-J. & Mertens, D., 2015: Die Mauern von Syrakus: Das Kastell Euryalos und die Befestigung der Epipolai (Wiesbaden). Boehlau, J. & Schefold, K. (edd.), 1940: Larisa am Hermos I: Die Bauten (Berlin). Burmeister, S., 2009: “‘Codierung/Decodierung’: Semiotik und die archäologische Untersuchung von Statussymbolen und Prestigegütern”, in Hildebrandt, B. & Veit, C. (edd.), Der Wert der Dinge—Güter im Prestigediskurs. Müncher Studien zur alten Welt 6 (München) 73–102. Childs, W.A.P., 1978: The City-Reliefs of Lycia (Princeton). Cooper, F.A., 2000: “The Fortifications of Epaminondas and the Rise of the Monumental Greek City”, in Tracy, J.D. (ed.), City Walls: The Urban Enceinte in Global Perspective (Cambridge) 155–91. Coupry, J., 1986: “Les fortifications d’Olbia de Ligurie : Propositions, questions”, in Leriche, P. & Tréziny, H. (edd.), La fortification dans l’histoire du monde grec. Actes du Colloque International: La fortification et sa place dans l’histoire politique, culturelle et sociale du Monde Grec, Valbonne, décembre 1982 (Paris) 389–99. Di Cesare, R., 2004: “La storia murata: note sul significato del riutilizzo di materiali architettonici nel muro di cinta dell’acropoli di Atene”, NumAntCl 33: 99–134.

Erol, A.F., 2004–2005: “An Analysis on Illustrations of War Materials on City-gates: Pamphilian, Pisidian and Isaurian Regions”, Anodos 4–5: 69–77. Fachard, S., 2004: “L’enceinte urbaine d’Erétrie: Un état de la question”, AntK 47: 91–108. Fachard, S., 2013: “Eleutherai as the Gates to Boeotia”, in Brélaz, C. & Fachard, S. (edd.), Pratiques militaires et Art de la Guerre dans le Monde Grec Antique (Paris) 81–106. Filimonos-Tsopotou, M., 2004: Η ελληνιστική oχύρωση της Ρóδoυ. Ρóδoς 1 (Athens). Frederiksen, R., 2011: Greek City Walls of the Archaic Period, 900– 480 BC (Oxford). Frederiksen, R., Laufer, E. & Müth, S., 2016: “Source Criticism: Fortifications in Written Sources and the Visual Arts”, in Müth et al. 2016, 173–95. Frederiksen, R., Müth, S., Schneider, P.I. & Schnelle, M. (edd.), 2016: Focus on Fortifications: New Research on Fortifications in the Ancient Mediterranean and the Near East. Papers of the Conference on the Research of Ancient Fortifications, Athens 6–9 December 2012. Fokus Fortifikation 2 (Oxford). Garlan, Y., 1974: Recherches de poliorcétique grecque (Paris). Giese, J. & Müth, S., 2016: “Messene”, in Müth et al. 2016, 278–85. Giese, J., 2010: “Bautechnische Beobachtungen am nördlichen und nordwestlichen Mauerabschnitt in Messene” in Lorentzen et al. 2010, 85–95. Godley, A.D., trans., 1920: Herodotus in Four Volumes 1. Loeb 117 (London). Grandjean, Y., 2011: Le rempart de Thasos. Avec la collaboration de Manuela Wurch-Kozelj et la participation de Tony Koželj. Études Thasiennes 22 (Athens). Grandjean, Y., Koželj, T. & Salviat, F., 2004–2005: “La porte de Zeus à Thasos”, BCH 128–129: 175–268. Gruben, G. & Müller, K., 2018: Das Dipylon. Kerameikos 22 (Wiesbaden). Hansen, M.H., 1997: “The Polis as an urban centre. The literary and epigraphical evidence”, in Hansen, M.H. (ed.), The Polis as an urban centre and as a political community. Symposium August, 29–31, 1996 (Copenhagen) 9–86. Hansen, M.H., 2006: Polis: An Introduction to the Ancient Greek City-State (Oxford). Hoffmann, A., 2011: “Warum in Kilikien? Der Karasis Residenz und Festung”, in Posamentir, R., Hoffman, A. & Sayar, M.H. (edd.), Hellenismus in der Kilikia Pedias (Istanbul) 63–86. Holtzmann, B., 1994: La sculpture de Thasos, Corpus des Reliefs. Études Thasiennes 15 (Athens). Hülden, O., 2000: “Pleistarchos und die Befestigungsanlagen von Herakleia am Latmos”, Klio 82: 382–408. Jansen, B., 2016: “Defensive Funktionen”, in Müth et al. 2016, 101–25. Knigge, U., 1988: Der Kerameikos von Athen: Führung durch Ausgrabungen und Geschichte (Athens).

MORE THAN WAR: SYMBOLIC FUNCTIONS OF GREEK FORTIFICATIONS Krause, C., 1972: Das Westtor. Eretria 4 (Bern). Krischen, F., 1922: Die Befestigungen von Herakleia am Latmos. Milet 3.2 (Berlin). Kuhn, G., 2017: Das heilige Tor. Kerameikos 22 (Wiesbaden). Laufer, E., 2010: “Pednelissos, Sillyon, Adada. ›Römische‹ Stadtmauern und kilikische Piraten?”, in Lorentzen et al. 2010, 165–93. Lauter, H., 1986: Die Architektur des Hellenismus (Darmstadt). Lazaridis, D., 2003: Amphipolis (Athens). Ley, J., 2009: Stadtbefestigungen in Akarnanien: ein bauhistorischer Beitrag zur urbanen Entwicklungsgeschichte einer antiken Landschaft (Berlin). Lindenlauf, A., 2003: “Constructing the Meaning of the Persian Wars in Athens”, in Brysbaert, A., de Bruijn, N., Gibson, E., Michael, A. & Monaghan, M. (edd.), SOMA 2002: Symposium on Mediterranean Archaeology: Proceedings of the Sixth Annual Meeting of Postgraduate Researchers, University of Glasgow, Department of Archaeology, 15–17 February, 2002. BAR-IS 1142 (Oxford) 53–62. Lohner-Urban, U. & Scherrer, P., 2016: “Hellenistische Prunktore—ein wissenschaftlicher Irrtum? Vorläufige Grabungsergebnisse vom Osttor von Side aus der Kampagne 2012”, in Frederiksen et al. 2016, 232–43. Lorentzen, J., Pirson, F., Schneider, P.I. & Wulf-Rheidt, U. (edd.), 2010: Aktuelle Forschungen zur Konstruktion, Funktion und Semantik antiker Stadtbefestigungen. Kolloquium 9./10. Februar 2007 in Istanbul (Istanbul). Marksteiner, T., 1993: “Stadtdarstellungen und lykische Städte”, in Borchhardt, J. & Dobesch, G., Akten des II. Internationalen Lykien-Symposions, Wien 6.–12. Mai 1990 (Vienna) 31–38. Marksteiner, T., 1997: Die befestigte Siedlung von Limyra. Studien zur vorrömischen Wehrarchitektur und Siedlungsentwicklung in Lykien unter besonderer Berücksichtigung der klassischen Periode. Forschungen in Limyra 1 (Vienna). Marsden, E.W., 1969: Greek and Roman Artillery (Oxford). Martini, W., 2016: “Form, Funktion und Bedeutung der Stadtmauern von Perge in Pamphylien”, in Frederiksen et al. 2016, 220–31. Meier, C., 1978: “Ein antikes Äquivalent des Fortschrittsgedankens: Das ‚Könnens-Bewußtsein’ des 5. Jahrhunderts v. Chr”, Historische Zeitschrift 226: 265–316. Meier, C., 1990: “An Ancient Equivalent of the Concept of Progress: the Fifth-Century Consciousness of Ability”, in McLintock, D. (trans.), The Greek Discovery of Politics (Cambridge, MA) 191–204. Mertens, D., 2003: Die Stadt und ihre Mauern. Selinus 1 (Mainz). Messerschmidt, W., 2003: Prosopopoiia: Personifikationen politischen Charakters in spätklassischer und hellenistischer Kunst (Cologne). Meyer, M., 2006: Die Personifikation der Stadt Antiocheia (Berlin).

213

Morris, C.W., 1972: Grundlagen der Zeichentheorie: Ästhetik und Zeichentheorie (Munich). Müth, S., 2010a: “Fortifikationskunst und Repräsentation an der Stadtmauer von Messene”, in Lorentzen et al. 2010, 57–83. Müth, S., 2010b: “Messène. Fondation et développement d’une ville de populations hétérogènes”, in Rouillard, P. (ed.), Portraits de migrants, Portraits de colons II. Colloque inter­ national Paris, 10, 11 et 12 juin 2009 (Paris) 135–46. Müth, S., 2014: “The Historical Context of the City Wall of Messene: Preconditions, written sources, success balance, and societal impacts”, Proceedings of the Danish Institute at Athens 7: 105–122. Müth, S., 2016a: “Functions and Semantics of Fortifications: An Introduction”, in Frederiksen et al. 2016, 183–92. Müth, S., 2016b: “Urbanistic Functions and Aspects”, in Müth et al. 2016, 159–72. Müth, S. & Bessac, J.-C. (forthcoming). Die Stadtmauer von Messene 1: Topographie und Beschaffenheit—Materialien und Steinbearbeitung—außenliegende Befestigungsanlagen— Geschichte und Funktion. Müth, S., Laufer, E. & Brasse, C., 2016: “Symbolische Funktionen”, in Müth et al. 2016, 126–58. Müth, S., Schneider, P.I., Schnelle, M. & De Staebler, P. (edd.), 2016: Ancient Fortifications: A Compendium of Theory and Practice. Fokus Fortifikation 1 (Oxford). Ouellet, K., 2016: “The City Walls of the Andrian Colonies: Tradition and Regionalism in Military Architecture”, in Frederiksen et al. 2016, 535–46. Peirce, C.S., 1983: “Ikon, Index und Symbol”, in Peirce, C.S. (ed.), Phänomen und Logik der Zeichen (Frankfurt) 64–67. Peirce, C.S., 1986: Semiotische Schriften 1. Edited and translated by C. Kloesel and H. Pape (Frankfurt/Main). Picard, C., 1962: Les murailles 1: les portes sculptées à images divines. Études Thasiennes 8 (Paris). Radt, T., 2010: “Fliehburg, Festung, Residenz?—Die Ruinen auf dem Karasis”, in Lorentzen et al. 2010, 195–218. Radt, T., 2016: “Fortified Palaces and Residences in Hellenistic Times: The Upper Castle on Mount Karasis and Other Examples”, in Frederiksen et al. 2016, 263–76. Rihll, T.E., 2007: The Catapult: A History (Yardley). Saner, T., Sağ, K. & Denktaş, E., 2016: “The Fortifications of Larisa (Buruncuk) Reconsidered”, in Frederiksen et al. 2016, 159–70. Schneider, L., Fehr, B. & Meyer, K.-H., 1979: “Zeichen— Kommunikation—Interaktion. Zur Bedeutung von Zeichen-, Kommunikations- und Interaktionstheorien in der Klassischen Archäologie”, Hephaistos 1: 8–41. Schwandner, E.-L., 1977: “Die Böotische Hafenstadt Siphai”, AA 92: 513–51. Schwertheim, U., 2010: “Monumentale Hoftore in Messene”, in Lorentzen et al. 2010, 97–106.

214 Sismanidis, K., 2003: Das antike Stagira: Heimat des Aristoteles (Athens). Sokolicek, A., 2009: Diateichismata: zu dem Phänomen innerer Befestigungsmauern im griechischen Städtebau (Vienna). Stähler, K., 1993: Form und Funktion: Kunstwerke als politisches Ausdrucksmittel (Münster). Stroszeck, J., 2014: Der Kerameikos in Athen: Geschichte, Bauten und Denkmäler im archäologischen Park (Möhnesee).

Müth Swoboda, H., Keil, J. & Knoll, F., 1935: Denkmäler aus Lykaonien, Pamphylien und Isaurien (Brünn). Weissl, M., 2012: Torgottheiten: Studien zum sakralen und magischen Schutz von griechischen Stadt- und Burgtoren unter Einbeziehung der benachbarten Kulturen (Ph.D. diss., University of Vienna: http://othes.univie.ac.at/17605/). Winter, F.E., 1971: Greek Fortifications (London). Wurster, W.W., 1977: “Stadtdarstellungen auf lykischen Reliefs”, Architectura 1977: 117–51.

chapter 13

Upcycling as a New Methodological Approach to Reuse in Greek Architecture Sarah A. Rous 1 Introduction The reuse of ancient architectural material in the Late Antique and Medieval Mediterranean has long been the focus of “spolia” studies within architectural and art historical scholarship. Increasingly, the study of reuse has expanded beyond this traditional typological, chronological, and geographical scope, and is now a point of inquiry in many fields (for detailed historiographies of spolia studies, see Kinney 2006; Frey 2016: 9–22). This broadened view, however, has been accompanied by great anxiety over the inappropriateness of the term spolia. Lacking an agreed-upon definition, the term is inextricably linked with the connotations of forcibly captured booty lent by its Latin etymology and Roman usage, and with the resoundingly negative views of depredation, degeneracy, and decline held by the 16th-century artist-antiquarians who resurrected the word (on the history of the term, see esp. Kinney 1995: 53–56; for growing anxiety, see esp. Alchermes 1994: 178; Kinney 1997: 119–22; 2006: 233–34; Greenhalgh 2011: 78–81). Attempts to define and classify different types of spolia—for example, R. Brilliant’s spolia in re vs. spolia in se (Brilliant 1982), or M. Greenhalgh’s 17-category pyramid (Greenhalgh 2011: 82–88; for the additions of spolia in spe see Cutler 1999: 1064–66 and spolia in re see Liverani 2011: 45–48)—or attempts simply to replace the term spolia with the very neutral term “reuse”, have not generally resulted in the desired diachronic or cross-cultural insights. It is sometimes admitted that within the ubiquitous practice of pragmatic and everyday reuse, what scholars actually want to address are only the remarkable cases where, as D. Kinney put it, “reuse emerges as value-laden” (Kinney 2011: 2; also cf. Kiilerich 2006: 135; Esch 2011: 22: “It is not accidental use, dictated by the occasion and by what lay at hand, that is the proper object of spolia studies, but rather conscious, targeted choice”). I have developed the concept of upcycling—most basically defined as intentionally meaningful reuse—and suggest that it can provide the conceptual reframing necessary to enable the examination of precisely these value-laden instances of reuse, architectural and otherwise (Rous 2019: 3–30). Upcycling is the choice to reuse a particular object in a particular new context or for a particular new function based not simply on its physical suitability, but

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004416659_015

instead (or also) on some aspect of its previous life history. It is reuse undertaken with some sense of purpose and consequence by the artist, patron, or conceiver, and meant to have implications for its viewer, reader, or audience. It achieves this through the perpetuation of some degree of visibility of or knowledge about the past function or experiences of the reused material. Some scholars have recently attempted to replace “spolia” with the general term “reuse” or its equivalents Wiederverwendung, reimpiego, remploi (see, e.g., Greenhalgh 2011: 79; Altekamp 2013: 168). “Recycling” is also occasionally used in spolia scholarship, usually uncritically and without definition as an apparent synonym of “spolia” or “reuse” (e.g., Kinney 1997: 118, 134; Cutler 1999: 1073; Greenhalgh 2012: passim). When carefully considered, though, recycling is clearly distinct from reuse of an object per se, as Kinney (2011: 3) has made explicit: in recycling, “form and function are obliterated, and the object is reduced to its material” (see also the thoughtful treatment of Munro 2011 and 2012). Similarly, Schiffer’s “life history” model of archaeological formation processes defined recycling as an object’s return to the manufacturing stage (Schiffer 1972: 158). Thus, whereas “reuse” is best maintained as a neutral umbrella term, and “recycling” connotes a return to the essential material, “upcycling” refers more specifically to self-conscious reuse that makes the prior life of the object and the act of reuse itself meaningful within the new context. “Upcycling” is broad enough to refer to diverse cases such as columns built into walls, the maintenance of ruins, recut inscriptions, and the relocation of buildings, while at the same time distinguishing such cases from and within the wider phenomenon of reuse driven by economics and pragmatism that was pervasive in the ancient world (cf. Kinney 1997: 122; 2011: 2). “Upcycling” has gained currency in modern popular culture and in contemporary art and art criticism, though it has not previously entered academic discourse. Some commercial enterprises in upcycling emphasise the transformation of useless or environmentally hazardous waste into socially valued goods. For example, a Lebanese jewelry design team combines used non-biodegradable plastic bags with Swarovski crystals (see Cigainero, J. “Upcycling Coins, Plastic Bags and Keys for Social Change”, New York

216 Times, 17 Nov. 2014).1 In my usage of the term, however, there is no value claim; the material being reused need not have been discarded or degraded, and it need not be transformed into a high status object. Rather, it is the knowledge or visibility of some aspect of the material’s previous function that is crucial. In contemporary culture, we see upcycling at work in items like handbags made of Coca-Cola cans, skirts made from neckties, and library desks composed from books. The cachet of these objects lies not in their materials per se, but in what those materials used to be; it is essential to the viewer’s understanding of them that he or she realises that they were once Coke cans, neckties, and books, and not merely aluminium, silk, and paper. I see upcycling as a nearly universal phenomenon, occurring across a wide range of materials, cultures, and time periods. Because it makes explicit reference to the past in the present, serving as a nexus between the two, a monument or object containing upcycled material often impacts the social memory of a community—that is, the “expression of collective experience” that “identifies a group, giving it a sense of its past and defining its aspirations for the future”, in J. Fentress and C. Wickham’s classic definition (Fentress & Wickham 1992: 25). In examining instances of upcycling, we have the opportunity to address both the intrusion of the past into the present and the role of the present in the construction of the past. The knowledge of an object’s past life preserved through upcycling serves as the stimulus to the communication that makes individual memory collective, and vice versa. My larger project has examined upcycling in Athenai from the Classical to Late Roman period, with a focus on buildings and objects made of marble—a durable and high-status stone that lent itself to thoughtful and intentional reuse. Exploring questions of visibility, agency, intention, and effect has brought innovative interpretations into focus and demonstrated that upcycling was one crucial component within the “work of memory” through which Athenians materially produced mnemonic effects (Rous 2019; on the “work of memory”, see: Hamilakis 2010: 194; Hamilakis & Labanyi 2008; cf. Mills & Walker 2008: 3–10 on “memory work”; amongst the vast scholarship on social memory, insightful work especially relevant to the study of Greece and of Greek architecture includes Buchert 2000, Alcock 2002, Lindenlauf 2003, Hurwit 2004, Davis 2007, Kousser 2009, Hartmann 2010, Shear 2011, Price 2012, Steinbock 2013, and Arrington 2015). In a city so rich in marble, monuments, and memory, upcycling proved a 1  Accessed 18 Feb. 2015.

Rous

powerful and flexible tool for anchoring or reorienting the present in relation to the past. In this chapter, I will focus on the North Akropolis Wall and the Post-Herulian Wall as two examples of how contextualised examinations of architectural reuse as upcycling can advance the limits of our interpretations and tie the study of architecture into wider trends and practices within the material environment. 2

The North Akropolis Wall

A prime example of Athenian upcycling is the construction of the North Akropolis Wall not long after the Persian sack of the Akropolis in 480 BCE. Both the Akropolis and the lower city of Athenai suffered severe damage at the hands of Persian armies under Xerxes and Mardonios; archaeological evidence from throughout the city largely corroborates Herodotos’ (8.53.2, 9.13) and Thoukydides’ (1.89.2–3) accounts of extensive destruction (see esp. Shear Jr. 1993 and Lynch 2011). In the aftermath of the decisive Greek victories at Salamis and Plataiai, Athenai’s initial priority was the construction of an outer fortification wall. According to Thoukydides, the Athenians—even women and children—hastily built the Themistoklean city wall before the Spartans could discover their activities. His account emphasises the overwhelmingly pragmatic concerns motivating the reuse of grave stelai and other squared stones, many trimmed down to fit, in the foundation and lowest courses of this wall (Thouk. 1.90– 91). Themistokles’ diplomatic trickery kept the construction activity secret from the Spartans until the wall had reached a defensible height. Built in such extreme haste, the stones for the wall were chosen based on proximity and suitability, and modified with an eye towards the ultimate structural stability of the mudbrick superstructure above all else (for the wall and its construction, see esp. Theocharaki 2011; and cf. the Kerameikos excavation reports, esp. Noack 1907, 129; Ohly 1965; Knigge 1991: 49–73; Löringhoff 1995; Kuhn 1995). The reuse in the Themistoklean city wall was visible to an observant viewer, since no time was wasted in trying to hide it, but was not visually accentuated. Earlier interpretations that assigned ideological motivations like magical apotropaism or anti-aristocratic sacrilege to the reuse of grave stelai in particular are now seen as misguided and unsubstantiated (for ideological interpretations, see esp. Stähler 1993: 13–24; Schneider & Höcker 1996: 123–24; cf. Keesling 1999: 515–18; for critiques, see esp. Bäbler 2001: 6–11; Kousser 2009: 266–67; cf. Kousser 2017: 82). Later literary traditions suggest that the wall held significance as a symbol of collective effort and of Themistokles’ guileful

UPCYCLING AS A NEW METHODOLOGICAL APPROACH TO REUSE IN GREEK ARCHITECTURE

figure 13.1

217

The North Akropolis Wall as preserved today; view from the Agora. (a) Reused column drums and temple entablature indicated by arrows to the left and right, respectively, of the later Erektheion building at centre; (b) detail of entablature blocks reused from the Temple of Athena Polias; (c) detail of unfinished column drums reused from the Older Parthenon all photos by the author

diplomacy, but not as an active memorial of the Persian Wars (see, e.g., Diod. Sik. 11.40.1–2; Plut. Bios Them. 19.1; Dem. Lept. 20.73–74; Nep. Them. 6–7). Because this reuse was undertaken with little regard for the past life of the material, it should not be considered upcycling. The North Akropolis Wall, by contrast, is an instance of accentuated reuse that has been among the most obvious and visible in the Greek world, to both ancient and

modern audiences (Fig. 13.1). This massive retaining and fortification wall was also constructed in the initial aftermath of the Persian Wars, as M. Korres has established (on the architectural history of the North Akropolis Wall, including the establishment of a Themistoklean date, see esp. Korres 2002; cf. Dinsmoor Jr. 1980: 63–64 for a summary of earlier dating debates). In it, the Athenians made extensive use of architectural material from the

218 two most important temples on the Akropolis at the time, the Temple of Athena Polias and the unfinished Older Parthenon, both of which had been heavily damaged by the Persians; reused elements in the North Akropolis Wall included ashlars, entablature blocks (marble metopes and limestone triglyphs and architrave blocks), and at least 16 column capitals from the Temple of Athena Polias, and at least 29 marble column drums from the Older Parthenon (Hill 1912: 557–58; Tschira 1940; Hurwit 1999: 142; Lambrinoudakis 1999: 551–53; Korres 2002; Di Cesare 2004: 110–15). While this reuse involved a measure of practicality as an effective way to deal with sacred temple material that was too structurally unsound to be useful in reconstructing the temples themselves (see Ridgway 1999: 205–06 and Miles 2011: 670–73 for a discussion of architectural blocks as divine property), an intentionality beyond hasty pragmatism is clear when we examine how and where the temple materials were placed in the new wall. The arrangement of the blocks in the outer face of the wall is strikingly deliberate: the triglyphs, metopes, and architrave blocks from the Temple of Athena Polias were put together to reconstitute the temple’s entablature within the wall (Fig. 13.1b); column drums of the Older Parthenon were stacked at least two high and arranged in a long row, evoking a colonnade or peristyle (Fig. 13.1c). The segment with the entablature blocks was the first section of the wall to be constructed, and thus central to the overall plan of the project. It was placed almost directly below the entablature’s original position on the temple that had previously dominated the visual landscape of the north side of the Akropolis (Korres 2002: 181, 184). These reused blocks are thus prominently visible from the city below, especially from the Agora, and their previous function as temple architecture is immediately recognisable (Korres 1994, 42; cf. Martin-McAuliffe & Papadopoulos 2012: 348– 52). Any Athenian, accustomed to seeing the north side of the Temple of Athena Polias when looking up to the Akropolis from below, would readily have identified that same entablature within the new wall. Monumental column drums, likewise, would have looked completely out of place in a fortification wall, and clearly understood as belonging to a temple. Overall, the material chosen for reuse in the Akropolis wall was not that which was best for wall building, as in the city wall, but that which was “most distinctively templelike”, in R. Kousser’s words, and it was used in such a way as to make its former life as temple architecture most plainly discernible (Kousser 2009: 271; contra Steskal 2004: 211). The Athenians clearly gave more careful consideration to the appearance and visibility of the reused temple materials in the North Akropolis Wall than they had to the

Rous

reused materials in the Themistoklean city wall. While the stelai in the bottom courses of the city wall were theoretically visible but not on display, the temple parts on the Akropolis were boldly shown off to the city below. We can usefully view the recognition of the former temple architecture within the north wall as nascent social memory; seeing the wall prompted viewers to remember and discuss the Temple of Athena and the grand new Parthenon project, their destruction by the Persians, and the Athenians’ own recovery of their city. By displaying the reused stone so conspicuously—preserving and magnifying both its prior context and the ravages it suffered in that context—the North Akropolis Wall functioned as an overt war memorial. It stood as a reminder of the impiety of the Persian barbarians and as a symbol of the Athenians’ ultimate triumph. That the upcycling in the North Akropolis Wall played a role in the formulation of Athenian social memory about the Persian Wars is seen in the increasing emphasis on sacrilege, asebeia, as the defining crime of the Persians and as a rallying point for the Greeks in the mid-fifth century BCE. During the wars themselves, the Athenians were far from innocent of such offences, having participated in the burning of Sardes during the Ionian Revolt in 498 BCE, for which Dareios famously vowed revenge (Hdt. 5.105). But by the 460s BCE the ideological dichotomy of the noble, temperate Greek and the violent, hubristic Asiatic was a well-developed explanation for many of the causes and outcomes of the Persian Wars, influenced by the constant display of the evidence of Persian sacrilege above the city, “a looming reminder of the impiety of the barbarians”, in J. Hurwit’s summation (Hurwit 1999: 142; also see Lindenlauf 2003: 55). The impiety of the Persian barbarian would become a prominent topos in the Periklean ideological scheme (as in, e.g., the Parthenon metopes; see: Ferrari 2000; the Oath of Plataiai, or the later conception of such an oath, is of course relevant here; see Rous 2019: 99–102 for the Oath as it relates to the reuse of the Temple of Athena Polias). Comparing the two great Themistoklean refortification projects shows how reuse that can be considered intentional upcycling can work to actively shape collective memory within a community, whereas reuse that does not consider the visibility of the past life of the material being reused is less likely to do so. Though the North Akropolis Wall and the Themistoklean city wall were part of the same general cleanup and rebuilding effort, they had different goals and messages (on the differing commemoration strategies on the Akropolis vs. the Agora and Kerameikos, see esp. Lindenlauf 2003: 56–57). The city wall was an urgent defensive necessity and a strong statement directed

UPCYCLING AS A NEW METHODOLOGICAL APPROACH TO REUSE IN GREEK ARCHITECTURE

at Sparta. With its completion, the Athenians turned their attention to the Akropolis with less urgency and with more consideration for the memorial outcomes of their actions, ultimately constructing a remarkable reminder of the tribulations and triumphs of the Persian Wars. 3 The Post-Herulian Wall Given its striking visibility within the Athenian landscape even today, and the relatively firm historical evidence of the events preceding its construction, this interpretation of the North Akropolis Wall as a war memorial has found wide acceptance (see, e.g., Rhodes 1995: 32–33; Lindenlauf 2003: 55; Di Cesare 2004: 117; Martin-McAuliffe & Papadopoulos 2012: 346–48; Miles 2014: 132). The much later Post-Herulian Wall, while it shares several fundamental characteristics with the North Akropolis Wall, has generally not been interpreted as a memorial of any sort. Re-examining this wall in light of the mnemonic effect of its predecessor, however, can help make sense of the visible accentuation of reused material throughout the wall, which, I argue, did indeed have memorial implications. Let us fast-forward 750 years to the third century CE. Although Greece had been under Roman provincial control for centuries, Athenai retained a significant degree of civic autonomy due to the wide esteem for its ancient cultural heritage, its prestige as the empire’s oldest university town, and the fact that elite Athenian families maintained their traditional roles in the political life of the city outside of the Roman aristocratic structures (on this phenomenon among Athenian elites, in contrast to powerful local families elsewhere in the empire, see Millar 1969: 21). The Panathenaic Festival with its great procession through the old Agora and up to the Akropolis continued to be celebrated into the fourth century CE (Himer. Or. 47.12–14; see Frantz 1988: 23–24). Despite economic decline in the first half of the third century, traditional civic institutions like the Kleisthenic deme and tribe system, the Areios Pagos council, and the office of eponymous archon remained intact and important alongside Roman magistracies like the proconsulship (Frantz 1988: 12). In 267 CE Athenai was ravaged in a raid by a Germanic tribe from the Black Sea region—the so-called Herulian sack (on the historiography of the Heruli and their invasion of Greece, see Millar 1969: 26–27; Frantz 1988: 1–2; Brown 2011: 82–88; for more detailed analyses of the written sources, see Alföldi in Cook 1939: 721–23; Straub 1952: 40–74; on the origins of the Heruli, see esp. Ellegård 1987). After inflicting considerable damage in the lower city, especially in the old Agora, the Heruli were ultimately driven

219

from the city by a band of two thousand Athenians led by Publius Herennius Dexippos, an historian who later wrote an account of the liberation of the city that included his own rousing speech (on the archaeological evidence for the sack, see Thompson & Wycherley 1972: 208–9; Frantz 1988: 3–5; and Camp 2001: 223; on the contentious question of Herulian damage on the Akropolis, see esp. Travlos 1973; Frantz 1979; the foundational treatment of Dexippos and his milieu by Millar 1969 is now joined by Brandt 1999 and the editions and translations of his surviving fragments and testimonia by Martin 2006 and Mecella 2013). In the following decade, facing intermittent but continued threats from the north, the Athenians constructed a new city wall around a much smaller core of the urban area, including the Akropolis but excluding the Agora, utilising much of the architectural and sculptural material damaged in the sack. Most of the primary investigation of the Post-Herulian Wall, including the dismantling of several sections, towers, and gates, took place in the mid-19th century and in the earlier years of the American excavations in the Ancient Agora (see reports in Pittakis, Charamis & Eustratiades 1852: 7–8; Koumanoudes 1861: 14–18, 37; Shear 1935; 1937; 1938; 1940). J. Travlos’ comprehensive work on the wall (in Frantz 1988) is joined by A.M. Theocharaki’s valuable compilation of excavation reports relevant to each documented segment (Theocharaki 2011: 151; see also Di Cesare & Marchiandi in Greco 2014: 1138–39). That the wall extended around the Akropolis to include the important sanctuaries and monuments of its south slope is based on the discoveries of Korres in the Stoa of Eumenes (Korres 1988: 18–19; cf. Korres 2015: 38 and Tanoulas 1997: 255), and has come to be generally accepted over the old view (found in Frantz 1988, inter alios) that the circuit included only the area north of the Akropolis. In several areas of the circuit, the wall was laid directly on the foundations or floors of ruined buildings. The thickness varies from 2.5 to 3.5 m, with two solidly built faces filled with a mortared core of fragments of broken architecture, sculpture, and inscriptions, and unworked stone rubble (Fig. 13.2). The highest preserved section is 7.5 m high, though J. Travlos estimated the original height to have been up to 11.5 m (in Frantz 1988: 126–27). The Post-Herulian Wall has long been seen as a line in the sand of history marking the end of the ancient city and the beginning of the medieval village. Frantz’s summation is emblematic of this idea: the sack in 267 CE “defines clearly the end of the ancient city and its transition to the status of a minor provincial town, a character which it retained all through the Middle Ages, with life disrupted to such an extent that the old pattern could never

220

figure 13.2

Rous

Section view of the inner and outer faces of the Post-Herulian Wall as preserved today in the area of the Library of Pantainos in the Agora photo by the author

be resumed” (Frantz 1988: 3; cf. Wilson 1971: 122–23 and more recently Oikonomidis 2013: 339–41; Miles 2014: 138). Unlike the North Akropolis Wall, the wall built after this later sack—dated at the beginning of Late Antiquity— has been discussed explicitly as a “spolia construction”. Dismissed as, for example, “a topographical monstrosity”, interpretation of the wall has largely stopped with pragmatic and economic explanations (for the traditional view of the Post-Herulian Wall as a spolia construction most akin to later and understudied fortification projects, see, e.g., Millar 1969, 29; Gregory 1982; Frantz 1988, 5; Wilkes 1989, 190–92, quote at 190; Tsoniotis 2008, 55; but see now Frey 2016 for an illuminating re-evaluation of some of these Late Antique spolia-filled walls at Aigina, Sparta, and Isthmia). By contrast, the material in the North Akropolis Wall is only very rarely discussed as spolia (e.g., Dally 2009: 48; Frey 2016: 32; Shear Jr. 2016: 8). Most scholarly interest has focused on identifying and reconstructing the original earlier structures from which the materials reused in the Post-Herulian Wall came (see, e.g., Thompson 1960: 350–59). From the 1960s onward rescue excavations by the Greek Archaeological Service have

brought to light small sections of the wall, mostly in the Plaka district (reports gathered in Bouras 2010: 32 n. 98; for more synthetic treatments of the recent discoveries, see Tsoniotis 2008; Bouras 2010; Sourlas 2013; 2014; note that the wall is usually referred to in Greek scholarship as the υστερορρωμαϊκό τείχος, “Late Roman Wall”). This is useful and important work, of course, but by examining the reused materials more holistically within the new context of the wall, we can also think more explicitly about the third-century Athenians who actually built the wall, and their choices and motivations. Let us look first at a large section still standing along the Panathenaic Way in the Agora (Fig. 13.3). In this segment, materials from several Agora buildings have been identified, including the Southwest Temple, the Southeast Stoa, the Library of Pantainos, the Metroön, and the Odeion of Agrippa. The beginning of construction has been dated to the reign of the emperor Probus (276–82 CE) based on numismatic evidence found within a dismantled section (Shear 1938: 332; Thompson 1959: 65; Frantz 1988: 6, 130–31; Camp 2010: 135–36; and now confirmed by Tsoniotis 2016; contra Di Branco 2006: 67–72 and Greco 2009: 217–20). As

UPCYCLING AS A NEW METHODOLOGICAL APPROACH TO REUSE IN GREEK ARCHITECTURE

with most reuse, there is surely an element of pragmatism at work here. It indeed made practical and economic sense to reuse the ruined material scattered nearby. But the wall was not hastily or carelessly built; instead, careful attention was paid to its aesthetic appearance in both faces and to the visibility of the act of reuse—that is, to preserving the knowledge that the architectural elements had once been the familiar monuments of the Agora before their destruction at the hands of the Heruli. Brown (2011: 85) is one of few scholars to note the attention paid to the appearance of the wall, observing that it “shows no evidence of emergency construction, having been built with great care and attention to detail in its course and external appearance”. The construction of the wall is irregular enough for the viewer to grasp readily that the materials are in secondary use rather than freshly quarried. Yet, also apparent is a deliberate effort to create striking visual patterns through the arrangement of differently coloured stones in various sections and courses (e.g., the string courses of dark

figure 13.3

221

Hymettian marble visible in Figs. 13.3–13.4). This use of colour contrast is one of the key strategies by which the reused material was accentuated and shown off within the wall, and is evident in several extant segments including a recently excavated section in the Diogeneion at the northeast angle of the wall (see, e.g., Tsoniotis 2008: fig. 16). Rather than using all the stones from one building in one discrete section for ease of transport and construction, the builders took measures like spreading the conspicuous Hymettian marble through multiple sections at different elevations, going beyond practical concerns like maintaining consistent course heights (Fig. 13.3a). It is also significant that not all the material reused in the wall came from very nearby sources. At least fifteen inscribed fragments of public funerary monuments were found within the wall or in its immediate vicinity (Aliferi 1992–98: 198–99). These were originally set up in the area of the demosion sema northwest of the Kerameikos, and were presumably also victims of the Heruli, who may have entered the city from that direction (for the theory

Outer face of the Post-Herulian Wall along the Panathenaic Way in the Agora, as preserved today: (a) Panathenaic Way in foreground; (b) detail of wall segment in background of 3a, illustrating colour contrast in different courses both photos by the author

222

figure 13.4

Rous

View of the North Akropolis Wall, indicated by arrows, from the Panathenaic Way alongside the Post-Herulian Wall in the Agora, as preserved today photo by the author

that the Heruli entered the city in two bands, see Frantz 1988: 4). In addition to the willingness to transport material from outside the previous city walls for use in the new wall, the several-year gap between the sack itself and the construction of the wall indicates that the level of haste seen in the erection of the Themistoklean city wall, for instance, was not such a factor in the Post-Herulian Wall. Tsoniotis (2008: 68) points to the elaborate foundations for the wall on the west side of the Library of Hadrianus as another indication that its construction was less rushed than is often assumed. The deliberate highlighting of the reused material through patterns and contrast suggests that we may have upcycling at work. Did the wall builders deliberately seek to accentuate the reuse in order to have the past life and experiences of the ruined material lend some added significance to the new wall, beyond practical defence? Two dedicatory inscriptions found in association with the wall give us a crucial window into the cultural context and ideological motivations surrounding its construction (the basic editions and most thorough treatment of both inscriptions are Sironen 1994: 21–22; 1997: 98–102; also see Frantz 1988: 9–11). Both inscriptions are in verse, written

in a similar and traditional script, and were found at the location of known or presumed gates in the wall (Sironen 1994: 60). The first, IG ii/iii2 ed alt. 13289 (=IG ii2 5199), now only partially preserved, was found on the east side of the wall, and proclaims in a typically Athenian epigram: Ἀμφίων Μούσαις κιθάρης ἔστησεν Θήβης τείχεα· νῦν δ’ ἐπ’ ἐμᾶς πατρίδος Ἰλλυριὸς ἁδύλογον Μοῦσαν μεθέπων· τῷ καὶ δοκέουσι ἀκμῆτες ῥέζειν πείρατα πάντα τέχνας. Amphion put up the walls of Thebes by the music of his cithara; now Illyrius (put up the walls) in my home city, following the sweet-voiced muse. Thus, untiringly (the workmen) seem to achieve all of the limits of their craft. trans. E. Sironen 1997: 98–100 no. 30

This Illyrios was Claudius Leonticus Illyrios, a prominent Athenian citizen, member of the Areios Pagos council and the proconsul of Akhaia, with similarly accomplished ancestors (on Illyrios and his career, see Groag 1939: 94–95; Christol 1986: 177–82; Frantz 1988: 9–10; Byrne

UPCYCLING AS A NEW METHODOLOGICAL APPROACH TO REUSE IN GREEK ARCHITECTURE

2003: 197–98 no. 424). Importantly, the inscription compares Illyrios to the legendary hero Amphion, famous in Greek mythology for building the great walls of Thebes by charming the stones with his music, and an important figure within the Neoplatonic circles at Athenai and Eleusis to which Dexippos certainly and Illyrios probably belonged (Di Branco 2006: 78–81 discusses the implications of the references to Amphion and the Kyklopes in these dedicatory epigrams). The second inscription, IG ii/ iii2 ed alt. 13290 (=IG ii2 5200), consisting of fragments of two epigrams found near the west flank of the wall in the Agora, boasts that neither Amphion’s music nor the hands of the Kyklopes were necessary for the construction of this wall, the latter plausibly a reference to the Bronze Age Mykenaian walls of the Akropolis, revered by later Athenians (this epigram finds a close parallel in the Greek Anthology [Anth. Pal. 7.379], where the city of Dikaiarkhia and the sea converse about sea walls seemingly built by Κυκλώπων χεῖρες): A: οὐ τάδε θελξιμελὴς Ἀμφιονὶς ἤρα[ρε φόρμιγξ] οὐδὲ Κυκλωπείας χειρὸς ἔδ[ειμε βία]. B: [ – – – – – ] πειθοῦς [ – – – – – ἀ]ρετᾶ[ς]. This (wall) was not put together by Amphion’s sweetsounding lyre. Neither did the powerful hands of the Kyklopes build (this wall). [ – – – ] of obedience. [ – – – ] of virtue (?) trans. E. Sironen 1997: 101–02 no. 31

The similarities in content, form, script, and topographical context make it clear that both these inscriptions belong to the Post-Herulian Wall, but since only the first names Claudius Illyrios, there remains debate about whether he was solely responsible for the entire project. Regardless, through these inscriptions Illyrios and his potential collaborators established an overt connection between contemporary Athenai and the all-important Greek past, and proclaimed the ability of modern Athenians to equal the achievements of their legendary predecessors. This is strong evidence, alongside the visual accentuation of reuse, that the wall was a deliberate monument and not simply a stop-gap defensive measure. Illyrios and other elites like Dexippos were justifiably proud of their actions in eventually overcoming the Heruli and recovering their city (the Athenians seem to have received no help from the Roman army, see Wilkes 1989: 189; Tsoniotis 2008: 56). Both Illyrios (IG ii/iii2 ed. alt. 13263, 13264 = IG ii2 3689, 3690) and Dexippos (IG ii/ iii2 ed. alt. 13262 = IG ii2 3669) received honourific statues

223

on the Akropolis with inscriptions that emphasised their services to the city and their descent from established and traditional Athenian families. Dexippos was praised specifically for his historical writing, showing that memorialisation of the sack and defence was seen as a key part of recovery in its aftermath (Sironen 1994: 18–20; 1997: 55–63; cf. Puech 2002: 210–25). These honours both participated in and promulgated the traditional honourific system of benefactions and rewards. Later, in the fourth and fifth centuries CE, Athenai would be unique within the empire in continuing to award statue honours to local elites regularly, often in return for contributions towards the city’s defence on the model of Illyrios and Dexippos (see esp. Gehn 2016: 195). Overall, the epigraphic evidence makes clear that the construction of the Post-Herulian Wall was something in which Athenian elites took great pride, and which they exploited to place themselves, their actions, and the recent experiences of their city within the line of its remembered past. Physically, too, the wall should be interpreted within the cultural continuum of the built memorial space of Athenai. With the North Akropolis Wall as a chronologically distant but physically conspicuous precedent, the Post-Herulian Wall can be seen as another intentional memorial of a devastating sack and subsequent triumphant recovery. As in the North Akropolis Wall, the careful arrangement of the reused material again suggests not merely pragmatism but indeed a further motivation behind the choice of which blocks to use where. Significantly, the Post-Herulian Wall is also a direct physical echo of the North Akropolis Wall. The visual connection is striking, especially from the Agora, where one sees the temple entablature within the Akropolis Wall looming above the later city wall (Fig. 13.4). The relationship of the two walls would have been particularly affecting during the great Panathenaic procession, as Athenians walked directly along the new wall on the Panathenaic Way up to the Akropolis. The Panathenaic Festival and other civic institutions continued well beyond 267 CE, and it has now become clear that daily life in Athenai returned fairly quickly to normal after the Herulian sack, forcing a reassessment of the bleak view of life in the Late Antique city (see, e.g., Millar 1969; Kapetanopoulos 1983; Castrén 1989; 1994; 1999; Martin 2006: 15–24). Whereas the contracted city wall has traditionally been seen as marking the end of antiquity in Athenai, rethinking questions of visibility, agency, intention, and effect— and thereby recognising the reuse as upcycling—helps us to focus on the larger cultural context and to see the wall instead as an attempt to create continuity with the great Athenian past. With the North Akropolis Wall as an

224

Rous

imposing physical and ideological backdrop, the much later Athenian wall-builders took pride in reusing ruined architectural material to create a useful collective memorial of destruction and revitalisation. 4 Conclusion An intriguing modern echo of the North Akropolis Wall underlines the point that in spite of (or perhaps because of) its age, its memorial message could still resonate and be influential in the third century CE. The long and austere Tomb of the Unknown Soldier, constructed in front of the Greek Parliament building in Athens’ Syntagma Square in 1932, presents several citations of Classical antiquity, including rows of roughly finished fictive column drums set in the wall at both ends, deliberately evoking the unfinished and burnt column drums of the Older Parthenon immured in the North Akropolis Wall (Fig. 13.5). Once more the idea of temple architecture in a wall, this time juxtaposed with quotes from Perikles’ funeral oration, works to commemorate battles of Greeks vs.

figure 13.5

barbarians, as the monument was conceived (Davis 2007: 240–45; see also Rous 2019: 213–18). The investigation undertaken here of the Post-Herulian Wall in tandem with the North Akropolis Wall has demonstrated the efficacy of examining reuse and social memory hand-in-hand, and of interpreting chronologically disparate instances of meaningful reuse in relation to each other within their common societal milieu. The other cases I have investigated in my larger project have borne out the value of this new approach to the study of intentional reuse as upcycling (Rous 2019). With the two fortification walls discussed here as chronological bookends, this project examines social memory and the reuse of marble in Athenai over many centuries, illuminating iterations of upcycling that worked to create, preserve, or alter social memory within the community. Building on the memory work of the post–Persian Wars era, including the North Akropolis Wall, later Athenians upcycled existing monuments and materials in order to reorient the present in relation to the past, particularly in periods of profound social or political transformation marked by uncertainty about the future. I have productively

Replica column drums, echoing those of the North Akropolis Wall, at the north end of the Tomb of the Unknown Soldier in front of the Greek Parliament building, Syntagma Square, Athens. Others visible at the south end, on far right of the image photo by the author

UPCYCLING AS A NEW METHODOLOGICAL APPROACH TO REUSE IN GREEK ARCHITECTURE

re-investigated other instances of architectural reuse in Athenai that few would think of as involving spolia, like the transplantation of the Temple of Ares, the treatment of the ruins of the Temple of Athena Polias, and the embellishment of the Mykenaian Bastion at the entrance to the Akropolis, alongside non-architectural examples like the remodelling of the Monument of the Eponymous Heroes and the reuse and reinscription of portrait statues on the Akropolis. Since upcycling is not limited to architectural reuse, a major advantage of the approach is that the necessary focus on cultural and physical context knits the study of architecture into a more holistic investigation of the work of memory in the material realm. Identifying intentionally meaningful reuse, architectural and otherwise, within the archaeological record and examining it as upcycling can help us develop more insightful interpretations of material culture. Significantly, investigating upcycling in conjunction with social memory allows us to take a more emic view, foregrounding the self-understanding of a community and the cultural mindset behind its choices and their outcomes. It is difficult, from the perspective of a modern classicist, not to see decline over the longue durée from Classical to Byzantine or Medieval Athenai, and to attempt to mark its beginning or its turning point. But an Athenian contemplating his city after the Herulian sack, particularly a civic or intellectual leader like Illyrios or Dexippos, would have focused on how to rebuild, how to commemorate, how to move forward, inspired by the exempla-rich past of his city; he would not have thought that the Middle Ages had begun. By illuminating human motivations, intentions, and reactions, this new approach to reuse and memory has significant potential to further our understanding of how a community can craft a conception of its collective past and of the importance of that past to its present and future—literally building its history into its city. List of References Alchermes, J., 1994: “Spolia in Roman Cities of the Late Empire: Legislative Rationales and Architectural Reuse”, DOP 48: 167–78. Alcock, S.E., 2002: Archaeologies of the Greek Past: Landscape, Monuments, and Memories (Cambridge). Aliferi, S., 1992–1998: “Τὰ διεσπαρμένα μνημεῖα ὡς πηγὲς γιὰ τὴν καταστροφὴ τοῦ Δημοσίου Σήματος”, Horos 10–12: 183–203. Altekamp, S., 2013: “Architectural Re-Use Processes in Late Antique North Africa: Prolegomena”, in Altekamp, S., MarcksJacobs, C. & Seiler, P. (edd.), Perspektiven der Spolienforschung I. Spoliierung und Transposition. Topoi 15 (Berlin) 159–205.

225

Arrington, N.T., 2015: Ashes, Images, and Memories: The Presence of the War Dead in Fifth-Century Athens (Oxford). Bäbler, B., 2001: “Die archaischen attischen Grabstelen in der themistokleischen Stadtmauer: Grabschändung oder Apotropaion?”, Philologus 145.1: 3–15. Bouras, C., 2010: Βυζαντινή Αθήνα, 10ος—12ος αι (Athens). Brandt, H., 1999: “Dexipp und die Geschichtsschreibung des 3. Jh. n.Chr”, in Zimmermann, M. (ed.), Geschichtsschreibung und politischer Wandel im 3. Jh. n.Chr. Historia Einzelschriften 127 (Stuttgart) 169–81. Brilliant, R., 1982: “I piedistalli del giardino di Boboli: spolia in se, spolia in re”, Prospettiva 31: 2–17. Brown, A.R., 2011: “Banditry or Catastrophy?: History, Archaeology, and Barbarian Raids on Roman Greece”, in Mathisen, R.W. & Shanzer, D. (edd.), Romans, Barbarians, and the Transformation of the Roman World: Cultural Interaction and the Creation of Identity in Late Antiquity (Burlington) 79–96. Buchert, U., 2000: Denkmalpflege im antiken Griechenland: Massnahmen zur Bewahrung historischer Bausubstanz (New York). Byrne, S.G., 2003: Roman Citizens of Athens. Studia Hellenistica 40 (Dudley). Camp II, J.M., 2001: The Archaeology of Athens (New Haven). Camp II, J.M., 2010: The Athenian Agora: Site Guide. 5th ed. (Princeton). Castrén, P., 1989: “The Post-Herulian Revival of Athens”, in Walker, S. & Cameron, A. (edd.), The Greek Renaissance in the Roman Empire: Papers from the Tenth British Museum Classical Colloquium. BICS Suppl. 55 (London) 45–49. Castrén, P., 1994: “General Aspects of Life in Post-Herulian Athens”, in Castrén, P. (ed.), Post-Herulian Athens: Aspects of Life and Culture in Athens, A.D. 267–529. Papers and Monographs of the Finnish Institute at Athens 1 (Helsinki) 1–14. Castrén, P., 1999: “Paganism and Christianity in Athens and Vicinity during the Fourth to Sixth Centuries A.D.”, in Brogiolo, G.P. & Ward-Perkins, B. (edd.), The Idea and Ideal of the Town between Late Antiquity and the Early Middle Ages (Leiden) 211–23. Christol, M., 1986: Essai sur l’évolution des carrières sénatoriales dans la seconde moitié du IIIe siècle ap. J.C. (Paris). Cook, S.A., 1939: The Imperial Crisis and Recovery, A.D. 193–324. 1st ed. CAH 12 (Cambridge). Cutler, A., 1999: “Use or Reuse? Theoretical and Practical Attitudes toward Objects in the Early Middle Ages”, in Ideologie e pratiche del reimpiego nell’alto Medioevo: 16–21 aprile 1998. Settimane di studio del Centro italiano di studi sull’alto Medioevo 46 vol. 2 (Spoleto) 1055–83. Dally, O., 2009: “Spolien in Italien und dem östlichen Mittelmeerraum vor dem Beginn der römischen Kaiserzeit”, in Schattner, T.G. & Valdés Fernández, F. (edd.), Spolien im

226 Umkreis der Macht = Spolia en el entorno del poder. Akten der Tagung in Toledo vom 21. bis 22. September 2006. Iberia Archaeologica 12 (Mainz am Rhein) 45–57. Davis, J.L., 2007: “Memory Groups and the State. Erasing the Past and Inscribing the Present in the Landscapes of the Mediterranean and Near East”, in Yoffee, N. (ed.), Negotiating the Past in the Past: Identity, Memory, and Landscape in Archaeological Research (Tucson) 227–56. Di Branco, M., 2006: La città dei filosofi: Storia di Atene da Marco Aurelio a Giustiniano (Florence). Di Cesare, R., 2004: “La storia murata: note sul significato del riutilizzo di materiali architettonici nel muro di cinta dell’acropoli di Atene”, NumAntCl 33: 99–134. Dinsmoor Jr., W.B., 1980: The Propylaia to the Athenian Akropolis, vol 1: The Predecessors (Princeton). Ellegård, A., 1987: “Who Were the Eruli?”, Scandia 53.1: 5–34. Esch, A., 2011: “On the Reuse of Antiquity: The Perspectives of the Archaeologist and of the Historian”, in Brilliant, R. & Kinney, D. (edd.), Reuse Value: Spolia and Appropriation in Art and Architecture from Constantine to Sherrie Levine (Farnham) 13–31. Fentress, J. & Wickham, C., 1992: Social Memory (Oxford). Ferrari, G., 2000: “The Ilioupersis in Athens”, HSCP 100: 119–50. Frantz, A., 1979: “Did Julian the Apostate Rebuild the Parthenon?”, AJA 83.4: 395–401. Frantz, A., 1988: Late Antiquity, A.D. 267–700. Agora 24 (Princeton). Frey, J.M., 2016: Spolia in Fortifications and the Common Builder in Late Antiquity. Mnemosyne Suppl. 389 (Leiden). Gehn, U., 2016: “Athens”, in Smith, R.R.R. & Ward-Perkins, B. (edd.), The Last Statues of Antiquity (Oxford) 190–99. Greco, E., 2009: “Su alcuni studi di topografia Ateniese alla SAIA: Vecchie ipotesi e nuove prospettive”, ASAtene 87 (ser. III 9.1): 217–33. Greco, E., 2014: Topografia di Atene. Sviluppo urbano e monumenti dalle origini al III secolo d.C. Tomo 3. SATAA 1 (Athens). Greenhalgh, M., 2011: “Spolia: A Definition in Ruins”, in Brilliant, R. & Kinney, D. (edd.), Reuse Value: Spolia and Appropriation in Art and Architecture from Constantine to Sherrie Levine (Farnham) 75–95. Greenhalgh, M., 2012: Constantinople to Córdoba: Dismantling Ancient Architecture in the East, North Africa and Islamic Spain (Leiden). Gregory, T., 1982: “The Fortified Cities of Byzantine Greece”, Archaeology 35.1: 14–21. Groag, E., 1939: Die römischen Reichsbeamten von Achaia bis auf Diokletian (Vienna). Hamilakis, Y., & Labanyi, J., 2008: “Introduction: Time, Materiality, and the Work of Memory”, History and Memory 20.2: 5–17. Hamilakis, Y., 2010: “Re-collecting the Fragments: Archaeology as Mnemonic Practice”, in Hamilakis, Y., Lillios, K.T. &

Rous Tsamis, V. (edd.), Material Mnemonics: Everyday Memory in Prehistoric Europe (Oxford) 188–99. Hartmann, A., 2010: Zwischen Relikt und Reliquie: Objektbezogene Erinnerungspraktiken in antiken Gesellschaften (Berlin). Hill, B.H., 1912: “The Older Parthenon”, AJA 16.4: 535–58. Hurwit, J.M., 1999: The Athenian Acropolis: History, Mythology, and Archaeology from the Neolithic Era to the Present (Cambridge). Hurwit, J.M., 2004: The Acropolis in the Age of Pericles (Cambridge). Kapetanopoulos, E., 1983: “Some Remarks on Athens of about 270”, AAA 16.1–2: 51–57. Keesling, C.M., 1999: “Endoios’s Painting from the Themistoklean Wall: A Reconstruction”, Hesperia 68.4: 509–48. Kiilerich, B., 2006: “Antiquus et modernus: Spolia in Medieval Art—Western, Byzantine and Islamic”, in Quintavalle, A.C. (ed.), Medioevo: Il tempo degli antichi. Atti del Convegno internazionale di studi, Parma, 24–28 Settembre 2003 (Milan) 135–45. Kinney, D., 1995: “Rape or Restitution of the Past? Interpreting Spolia”, Papers in Art History from the Pennsylvania State University 9 (The Art of Interpreting): 53–67. Kinney, D., 1997: “Spolia. Damnatio and renovatio memoriae”, MAAR 42: 117–48. Kinney, D., 2006: “The Concept of Spolia”, in Rudolph, C. (ed.), A Companion to Medieval Art: Romanesque and Gothic in Northern Europe (Oxford) 233–52. Kinney, D., 2011: “Introduction”, in Brilliant, R. & Kinney, D. (edd.), Reuse Value: Spolia and Appropriation in Art and Architecture from Constantine to Sherrie Levine (Farnham) 1–11. Knigge, U., 1991: The Athenian Kerameikos: History, Monuments, Excavations (Athens). Korres, M., 1988: “Ἐργασίες στα μνημεία”, ArchDelt 35 (Β’1): 9–21. Korres, M., 1994: “The History of the Acropolis Monuments”, in Economakis, R. (ed.), Acropolis Restoration: The CCAM Interventions (London) 34–51. Korres, M., 2002: “On the North Acropolis Wall”, in Stamatopoulou, M. & Yeroulanou, M. (edd.), Excavating Classical Culture: Recent Archaeological Discoveries in Greece (Oxford) 179–86. Korres, M., 2015: The Odeion Roof of Herodes Atticus and Other Giant Spans (Athens). Koumanoudes, S., 1861: Γενική συνέλευσις τών μελών της εν Αθήναις Αρχαιολογικής Εταιρεας (Athens). Kousser, R.M., 2009: “Destruction and Memory on the Athenian Acropolis”, ArtB 91.3: 263–82. Kousser, R.M., 2017: The Afterlives of Greek Sculpture: Interaction, Transformation, and Destruction. (Cambridge). Kuhn, G., 1995: “Das themistokleische Heilige Tor”, AA 110: 649–59. Lambrinoudakis, V., 1999: “Le mur de l’enceinte classique de l’Acropole d’Athènes et son rôle de péribole”, Comptes rendus

UPCYCLING AS A NEW METHODOLOGICAL APPROACH TO REUSE IN GREEK ARCHITECTURE des séances de l’Académie des Inscriptions et Belles-Lettres 143.2: 551–61. Lindenlauf, A., 2003: “Constructing the Meaning of the Persian Wars in Athens”, in Brysbaert, A., de Bruijn, N., Gibson, E., Michael, A. & Monaghan, M. (edd.), SOMA 2002: Symposium on Mediterranean Archaeology: Proceedings of the Sixth Annual Meeting of Postgraduate Researchers, University of Glasgow, Department of Archaeology, 15–17 February, 2002. BAR-IS 1142 (Oxford) 53–62. Liverani, P., 2011: “Reading Spolia in Late Antiquity and Contemporary Perception”, in Brilliant, R. & Kinney, D. (edd.), Reuse Value: Spolia and Appropriation in Art and Architecture from Constantine to Sherrie Levine (Farnham) 33–51. Löringhoff, B. Von Freytag, 1995: “Untersuchungen in Turm N”, AA 110: 639–49. Lynch, K., 2011: “Depositional Patterns and Behavior in the Athenian Agora: When Disaster Strikes”, in Lawall, M.L. & Lund, J. (edd.), Pottery in the Archaeological Record: Greece and Beyond. Acts of the International Colloquium Held at the Danish and Canadian Institutes in Athens, June 20–22, 2008 (Aarhus) 67–74. Martin, G., 2006: Dexipp von Athen: Edition, Ubersetzung und begleitende Studien. Classica Monacensia: Münchener Studien zur Klassischen Philologie 32 (Tübingen). Martin-McAuliffe, S. & Papadopoulos, J.K., 2012: “Framing Victory: Salamis, the Athenian Acropolis, and the Agora”, JSAH 71.3: 332–61. Mecella, L., 2013: Dexippo di Atene: Testimonianze e frammenti. I frammenti degli storici greci 6 (Tivoli). Miles, M.M., 2011: “The Lapis Primus and the Older Parthenon”, Hesperia 80.4: 657–75. Miles, M.M., 2014: “Burnt Temples in the Landscape of the Past”, in Ker, J. & Pieper, C. (edd.), Valuing the Past in the Greco-Roman World. Proceedings from the Penn-Leiden Colloquia on Ancient Values VII. Mnemosyne Suppl. 369 (Leiden) 111–45. Millar, F., 1969: “P. Herennius Dexippus: The Greek World and the Third-Century Invasions”, JRS 59: 12–29. Mills, B.J. & Walker, W.H., 2008: “Introduction: Memory, Materiality, and Depositional Practice”, in Mills, B.J. & Walker, W.H. (edd.), Memory Work: Archaeologies of Material Practice (Santa Fe) 3–23. Munro, B., 2011: “Approaching Architectural Recycling in Roman and Late Roman Villas”, in Mladenović, D. & Russell, B. (edd.), TRAC 2010: Proceedings of the Twentieth Annual Theoretical Roman Archaeology Conference (Oxford) 76–88. Munro, B., 2012: “Recycling, Demand for Materials, and Land Ownership at Villas in Italy and the Western Provinces in Late Antiquity (5th–6th Century AD)”, JRA 25: 351–70. Noack, F., 1907: “Die Mauern Athens. Ausgrabungen und Untersuchungen”, AM 32: 123–60. Ohly, D., 1965: “Kerameikos-Grabung Tätigkeitsbericht 1956– 1961”, AA 1965: 277–375.

227

Oikonomidis, S., 2013: “Spolia and Spoilage of the Archaeological Environment. Construction—Destruction—Reconstruction: The Case of the Historical Center of Athens”, in Driessen, J. (ed.), Destruction: Archaeological, Philological and Historical Perspectives (Louvain-la-Neuve) 337–54. Pittakis, K., Charamis, D. & Eustratiades, P., 1852: Επιγραφαί ανεκδότοι, fasc. 2 (Athens). Price, S., 2012: “Memory and Ancient Greece”, in Dignas, B. & Smith, R.R.R. (edd.), Historical and Religious Memory in the Ancient World (Oxford) 15–36. Puech, B., 2002: Orateurs et sophistes grecs dans les inscriptions d’époque impériale. Textes et traditions 4 (Paris). Rhodes, R.F., 1995: Architecture and Meaning on the Athenian Acropolis (Cambridge). Ridgway, B.S., 1999: Prayers in Stone: Greek Architectural Sculpture ca. 600–100 B.C.E. (Berkeley). Rous, S.A., 2019: Reset in Stone: Memory and Reuse in Ancient Athens (Madison). Schiffer, M.B., 1972: “Archaeological Context and Systemic Context”, AmerAnt 37.2: 156–65. Schneider, L. & Höcker, C., 1996: Griechisches Festland: Antike und Byzanz, Islam und Klassizismus zwischen Korinthischem Golf und nordgriechischen Bergland (Cologne). Shear Jr., T.L., 1993: “The Persian Destruction of Athens: Evidence from Agora Deposits”, Hesperia 62.4: 383–482. Shear Jr., T.L., 2016: Trophies of Victory: Public Building in Periklean Athens (Princeton). Shear, J.L., 2011: Polis and Revolution: Responding to Oligarchy in Classical Athens (Cambridge). Shear, T.L., 1935: “The Campaign of 1933”, Hesperia 4.3: 311–39. Shear, T.L., 1937: “The Campaign of 1936”, Hesperia 6.3: 333–81. Shear, T.L., 1938: “The Campaign of 1937”, Hesperia 7.3: 311–62. Shear, T.L., 1940: “The Campaign of 1939”, Hesperia 9.3: 261–308. Sironen, E., 1994: “Life and Administration of Late Roman Attica in the Light of Public Inscriptions”, in Castrén, P. (ed.), Post-Herulian Athens: Aspects of Life and Culture in Athens, A.D. 267–529. Papers and Monographs of the Finnish Institute at Athens 1 (Helsinki) 15–62. Sironen, E., 1997: The Late Roman and Early Byzantine Inscriptions of Athens and Attica (Helsinki). Sourlas, D., 2013: “Πρόσφατες αρχαιολογικές έρευνες στη Ρωμαϊκή Αγορά, στη Βιβλιοθήκη του Αδριανού και τις όμορες περιοχές”, in Oikonomou, S. & Donga-Toli, M. (edd.), Αρχιολογικές συμβολές 2 (Athens) 149–68. Sourlas, D., 2014: “The Monument within the Monument: Preliminary Results of the Excavation at the ‘Aiolos’ Hotel in Plaka, Athens”, in Caliò, L.M., Lippolis, E. & Parisi, V. (edd.), Gli Ateniesi e il loro modello di città: seminari di storia e archeologia greca I. Roma, 25–26 giugno 2012. Thiasos Monografie 5 (Rome) 299–306. Stähler, K., 1993: Form und Funktion: Kunstwerke als politisches Ausdrucksmittel (Münster).

228 Steinbock, B., 2013: Social Memory in Athenian Public Discourse: Uses and Meanings of the Past (Ann Arbor). Steskal, M., 2004: Der Zerstörungsbefund 480/79 der Athener Akropolis. Eine Fallstudie zum etablierten Chronologiegerüst. Antiquitates 30 (Hamburg). Straub, J., 1952: Studien zur Historia Augusta (Bernae). Tanoulas, T., 1997: Τα Προπύλαια της αθηναϊκής Ακρόπολης κατά τον Μεσαίωνα (Athens). Theocharaki, A.M., 2011: “The Ancient Circuit Wall of Athens: Its Changing Course and the Phases of Construction”, Hesperia 80.1: 71–156. Thompson, H.A. & Wycherley, R.E., 1972: The Agora of Athens: The History, Shape, and Uses of an Ancient City Center. Agora 14 (Princeton). Thompson, H.A., 1959: “Athenian Twilight: A.D. 267–600”, JRS 49: 61–72. Thompson, H.A., 1960: “Activities in the Athenian Agora: 1959”, Hesperia 29.4: 327–68. Travlos, J., 1973: “Ἡ πυρπόλησις τοῦ παρθενῶνος ὑπὸ τῶν Ἐρούλων καὶ ἡ ἐπισκευή του κατὰ τοὺς χρόνους τοῦ Αὐτοκράτορος Ἰουλιανοῦ”, ArchEph 1973: 218–36.

Rous Tschira, A., 1940: “Die unfertigen Säulentrommeln auf der Akropolis von Athen”, JdI 55: 242–61. Tsoniotis, N., 2008: “Νέα στοιχεία για το υστερορωμαϊκό τείχος της Αθήνας”, in Vlizos, S. (ed.), Η Αθήνα κατά τη ρωμαϊκή εποχή: Πρόσφατες ανακαλύψεις, νέες έρευνες = Athens during the Roman Period: Recent Discoveries, New Evidence. Benaki Museum Suppl. 4 (Athens) 55–74. Tsoniotis, N., 2016: “The Benizeli Mansion Excavation: Latest Evidence on the Post-Herulian Fortification Wall in Athens”, in Frederiksen, R., Müth, S., Schneider, P.I. & Schnelle, M. (edd.), Focus on Fortifications: New Research on Fortifications in the Ancient Mediterranean and the Near East. Fokus Fortifikation Studies 2, Monographs of the Danish Institute at Athens 18 (Oxford) 712–24. Wilkes, J.J., 1989: “Civil Defence in Third-Century Achaia”, in Walker, S. & Cameron, A. (edd.), The Greek Renaissance in the Roman Empire: Papers from the Tenth British Museum Classical Colloquium. BICS Suppl. 55 (London) 187–92. Wilson, G.E., 1971: The Herulian Sack of Athens, A.D. 267 (diss., University of Washington).

chapter 14

Looking at the Unfinished: Roughed-Out Ornamentation in Greek Architecture Matthias Grawehr Each period in the history of art brings forth its own styles and aesthetics, changes previous attitudes towards the world, and calls in new perspectives, ones hitherto neglected. In the same way, each period in the historiography of art perceives its subject in a different light. Certain aspects are emphasised and others marginalised, and, despite their possible value, many interpretations seem awkward or forced in hindsight. In the Greek architectural tradition, workmen prepared stones on the ground of their construction site with a rough finish. Only after the last stone had been moved into position would the visible surfaces be smoothed and the ornamentation cut in order to prevent breakage. On Doric columns, the flutes were finished only to a few cm in height on the bottom drum and on the necking of the capital. They were completed from top to bottom only after the column was erected entirely to guarantee perfectly vertical lines. In this contribution, I will draw attention to surfaces and ornaments in Greek architecture that were intentionally left in a roughed-out state. In modern usage, this effect is known as rustication, a term coined by Italian Renaissance architects to designate a desired rustic appearance of wall surfaces. But unlike for the Renaissance and Baroque periods, for antiquity we do not possess any written sources to explain the connotations and intended effects of this peculiar style, and therefore I will avoid the term “rustication”. Neoclassicism, in which the discipline of Classical Archaeology is rooted, looked down on the excesses of the Late Baroque and in general denied the possibility of the intentional use of unfinished elements in ancient Greek architecture. Archaeology has, therefore, long neglected the subject. As we ourselves live in an age that displays a keen interest in the aesthetics of industrial-style living and in the charms of roughed-out architecture, I will take a fresh approach to the examination of “unfinished” Greek architecture and offer new explanations and insights. 1

Histories of Unfinished Architecture

To the scholars of the late 18th and the 19th century who laid out many of the cornerstones of Classical Archaeology © Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004416659_016

upon which we are still building today, it would probably have seemed a rather vain effort to study the “unfinished” parts of buildings to trace the principles of Greek architecture. It would have run counter to their aesthetic convictions, which praised Greek art as the ideal, as perfection in stone. When Nicholas Revett (1720–1804) drew his sober and painstakingly measured plans and views of ancient buildings, he simply omitted the many bosses or other “imperfections” that he encountered but that did not conform to the Neoclassical ideal. One could certainly name other Neoclassicists like Julien-David Le Roy (1724–1803) or Louis-François Cassas (1756–1827) who display a similar attitude towards such “imperfections”. This attitude can be seen, for example, in Revett’s drawings of the propylon in the Roman agora at Athenai, where the lifting bosses on the architrave are present in the veduta provided by his co-author James Stuart but do not figure in the more idealised elevation drawings by his own hand (Stuart & Revett 1762: chap. 1 pl. 1, 3; also see Le Roy 1758, 1: pl. 19, 2: pl. 14). Another example is his documentation of the Thrasyllos monument, where the prominent lifting bosses on the steps and the pillars are omitted (Stuart & Revett 1787: chap. 4 pl. 3, although I am not sure how much of the steps could be seen at that time; cf. Le Roy 1758, 1: pl. 8). It was only in the wake of the positivist approach to science in the late 19th century that new standards were established to document ancient architecture in its actual surviving state down to the smallest detail. It was mainly the second generation of the “Deutsche Bauforschung” with scholars like Hubert Knackfuß (1866–1948), Daniel Krencker (1874–1941) or Armin von Gerkan (1884–1969) that paid attention to unfinished elements. Notably the Genevan draughtsman Paul Schazmann (1871–1941), who had no formal training as a classical archaeologist, set high standards in documenting architecture. But still, the question of why parts of buildings were left in a roughened state was hardly ever addressed. The question was clearly formulated for the first time by Knackfuß (1908: 47): “An verschiedenen Blöcken der anderen Seiten sind die Hebebossen oder richtiger Kantenschutzbossen stehen geblieben, ob als Zeichen der Nichtvollendung oder in dekorativer Absicht dürfte fraglich sein”. Most scholars dismissed rough finishing by either referring in a general

230

Grawehr

way to a deplorable negligence on finish by the “idle easterner”, or citing the equally deplorable and ever-present dearth of funds, or more specifically postulating an unexpected death for the commissioner of a monument. Often, to postulate the unexpected death of a commissioner offered a welcome historical anchor for the dating of the building (Fabricius 1881: 15; Kleiner 1968: 73; Hoepfner 1971: 84; Mattern 2008). That a rough finish was intended seemed out of the question. It was only from the late 1960s onward that scholars like Trevor Hodge, Richard Tomlinson, Thanassis Kalpaxis, and Hans Lauter acknowledged roughed-out architecture as intentional. For them it was a style characteristic of a certain period of time, although there was no consensus about which period.1 The scholarly community seemed confused and hesitant to enter the discussion (for example, Gruben 1997: 376). Indeed, I think, all attempts to understand “Bossenstil” as a short-lived fancy for the decorative effect proved to be a dead end (Lauter 1983: 297; Müller-Wiener 1988: 77; von Hesberg 1994b: 256–57; Frejman 2014; and Fragaki 2015: 294). Like the modern “industrial style”, whose unplastered brick walls and exposed steel beams seem to imitate the improvised living space of young urbanites, and therefore lay bare the semantics of style (as being socially determined in this case), my approach to roughed-out architecture is to look for the factors that conditioned its use in the Greek and Roman period (for a similar approach see von Hesberg 1994a: 84–88; Liljenstolpe 2000–2001). I aim to elucidate some of the principles that guided ancient architects in their design decisions. 2

One Question, Many Answers

One of the shortcomings of earlier attempts in the scholarship of “unfinished architecture” was that it tried to collect too much evidence under the same heading. There is 1  Kalpaxis (1986, 126) stated: “Gegen die Mitte des 5. Jhs. münden diese, in Ansätzen schon seit dem 6. Jh. erkennbaren Tendenzen im Entstehen eines neuen … Stils in der griechischen Baukunst [i.e. the ‘Bossenstil’]”; Hodge & Tomlinson (1969, 192): “The architects of the fourth century were more ready than their predecessors to apply variegated and rusticated effects to their buildings”; Hoepfner (1990: 277): “Die Vorliebe für das Unfertige auch in der Architektur entwickelte sich gerade in dieser Zeit [i.e., the early 2nd century BCE]”; Lauter (1983: 296): “Vielmehr darf man wohl mit aller Vorsicht vermuten, daß dieses Motiv in der zweiten Hälfte des 2. und der ersten Hälfte des 1. Jahrhunderts v. Chr. seine größte Wirkung entfaltete bzw. dieser Periode eigentlich zugehört”; and Kleiner (2010, 114): “This rusticated style is unique to Claudian architecture”; cf. Ward-Perkins & Boëthius 1970, 209.

no single explanation that fits all instances of “unfinished architecture”. First of all, we have to decide whether or not the present state of a building was indeed intentional. Was it ever used? Does the state of finish represent a natural stage in the building process or not? Of course there are many building projects that for one reason or another were never finished and used, such as the temples of Zeus at Stratos or at Lebadeia. The unfinished state of the temple of Zeus in Stratos is sometimes erroneously considered intentional (for example by Kalpaxis 1986: 156– 64; the crucial evidence is collected in Schwandner and Kolonas 1996: 187–90, and many more unfinished building blocks that have rolled down from the hill where the temple stands remain unpublished)—but the temple was a building project of the Akarnanian league and was most probably abandoned when the federation was dissolved in the mid-3rd century BCE (for the temple in Lebadeia see: Paus. 9.39.4; Turner 1994; Gadolou 2008; Partida 2015). In other instances, like the temple of Apollon in Didyma, it was expected from the beginning that the project would take a long time to complete, and priority instead was given to the functional core of the building (Knackfuß 1941; Borg 2001; Borg & Borg 2002). But even where the choice for a rough finish was deliberate, many different factors may have played a part. The use of roughly finished masonry could be dictated by the type of a building alone, whether a city wall or a supporting wall, or by the position of the wall within a building, whether an exterior facade or a basement story (Koenigs 1980: 30; Lauter 1983: 304; von Hesberg 1994b: 256; Liljenstolpe 2000–2001; Thomas 2015: 281; Grawehr 2015: 486; and Grawehr 2017: 103). Otherwise, it could be a way to adapt to the properties of available stone qualities, thereby creating a typical vernacular architectural style (Grawehr 2017: 108–10). Leaving all these aspects aside, in the following pages I will concentrate on only one other issue, namely that a raw finish could provide a contrasting effect between rough and smooth surfaces and therefore channel the viewer’s attention towards the more elaborate parts of a building. 3

Directing the Viewer’s Attention

The Bouleuterion of Miletos (Fig. 14.1) (Knackfuß 1908; Krischen 1941: 7–12, pl. 1–11; Schaaf 1992: 37–61; Bringmann and von Steuben 1995: 514–15; Kockel 1995: 31, 34; and Emme 2013: 109–13, 345) was built by two Milesian brothers in Seleucid service to honour Antiokhos IV (175–164 BCE) (RE 8, 1 (1912) 465–68 s. v. Herakleides 32; RE 6A, 1 (1936) 1237–38 s. v. Timarkhos 5; Herrmann 1987; and Chrubasik 2016: 127–35.). Timarkhos, satrap of Media, later usurped

LOOKING AT THE UNFINISHED: ROUGHED-OUT ORNAMENTATION IN GREEK ARCHITECTURE

231

figure 14.1 Miletos, Bouleuterion, view from the northeast, reconstruction drawing Knackfus 1908, pl. 14

the Seleucid throne against Demetrios I with Roman consent. Herakleides, the younger of the brothers, was the treasurer of Antiokhos IV and later, after his brother’s failure, supported Alexandros Balas against Demetrios. Both brothers may have been raised with Antiokhos in Rome and returned there several times on diplomatic missions. The patronage of the council house befitted their high aspirations, and its architecture must be regarded as avant-garde for its time. The design of the building responded in a most innovative way to its urban setting and its function. When standing in front of it, the Bouleuterion could be enjoyed up close by looking at its splendid propylon with an unusually early instance of the Korinthian order. It is interesting to note the similarities between these Korinthian capitals and the ones from the Athenian Olympieion, another building project of Antiokhos IV realised by the Roman architect D. Cossutius (Rawson 1975). But as an assembly hall for the city’s boule, its architecture also had to establish a close link to the agora as the town’s

political centre, despite the fact that the allotted space was separated from the square by the third-century porticoes that surrounded the agora. The architect therefore divided the building’s facade artificially into two stories. The upper story with an engaged Doric order was placed over a plain basement story. Seen from the agora, the hall turned its gable towards the viewer and hovered like an akropolis temple just above the portico’s flat roof. It was not possible to enjoy this elaborate southern view of the building up close, as only a small path separated the facade from the agora’s back wall. This is where the rough finish comes into play. To highlight the intended orientation towards the agora and to define the northern facade as the back of the building, a raw finish on its northern side was used in contrast to a fine polish on the southern side (Knackfuß 1908: 26–30 fig. 2–5). This can be observed on the socle where the surface of the north and west wall was not smoothed after the stones had been moved into place but left in a raw

232

Grawehr

figure 14.2 Miletos, Bouleuterion, wall architrave photo by author

state. In a similar way, many blocks of the wall architrave retain lifting bosses that have not been chiseled away after placing the blocks (Fig. 14.2) (Knackfuß 1908: 47 fig. 29). Their exact position within the building is not known, for in the publication the findspot of the architraves is not commented upon in detail, but it is stated that the lifting bosses are retained on some architrave blocks from all sides other than the east (Knackfuß 1908: 47). Today fragments of at least 4 architrave blocks are stored at the south side of the building. At least one of them must have been found there, because it can be discerned in a photograph taken during the excavation (Knackfuß 1908: 30 fig. 4). The idea, however, that they stem from the east side where the dedicatory inscription was placed can be ruled out. The raw finish on the north and west side would clearly communicate to anyone who approached from the harbour that this was the back side of the building, and that ideologically the Bouleuterion belonged to the civic centre, the agora. In a similar manner, a rough finish could be used in a courtyard or peristyle court to define marginal spaces and to provide a contrasting effect for the main facade. Such an emphasis on one facade in the court reflects Hellenistic aesthetics and can be perceived in an exemplary manner in a literary description of palace architecture by Apollonios of Rhodos, who worked at Alexandria towards the middle of the 3rd century BCE, and described in his Argonautica the palace of king Aeëtes at Kolkhis (Ap. Rhod. Argon. 3.215–48). The author modelled this

passage after the description of Alkinous’ palace in the Odysseia (Hom. Od. 7.83–97) but updated the design of the imagined architecture. From a position in the royal courtyard, he describes the lofty apartments of the king’s son and his daughters opposite each other on the left and right side of the peristyle. On the main face of the courtyard, Apollonios places the tallest dwelling of all, which is home to the king himself and his consort. The royal suite clearly is the focus of all attention. On the opposite end, the wing towards the entrance, the “rear façade” of the peristyle, is not even mentioned. One way to direct the viewer’s attention to the main vista in a peristyle courtyard is described by Vitruvius (6.7.3) and known through several examples: this is the socalled “Rhodian peristyle”, where the side opposite the entrance is equipped with columns higher than on the other wings. From there the largest and most sumptuous reception spaces for the entertainment of guests open off. A paradigmatic example from Delos is the House of the Masks (Chamonard 1933: 115–21 and Trümper 1998: 255– 57). In this 2nd-century BCE house, the Rhodian peristyle displays an intriguing distribution of building materials: the higher columns of the northern main wing are made from granite, the two lateral wings have conglomerate shafts, and on the side towards the entrance two obviously reused calcarenite columns are employed (for the building stones on Delos cf. Hadjidakis, Matarangas, and Varti-Matarangas 2009). Although a thick layer of stucco was applied to give all columns a uniform look, the choice

LOOKING AT THE UNFINISHED: ROUGHED-OUT ORNAMENTATION IN GREEK ARCHITECTURE

figure 14.3

233

Nea Paphos, “Tombs of the Kings”, Tomb 4 © Chris Combe, BY CC 2.0

of material does nevertheless reflect the hierarchy of the colonnades. While in the House of the Masks the importance of the different wings is only demonstrated by the “Rhodian” arrangement and by the value of the different building materials, in other instances it is expressed by the use of different orders and of different stages of finish. In two very similar court buildings of undetermined function in Miletos, an Ionic order with fluted columns was chosen for the main facade of the Rhodian peristyle, whereas the other colonnades are Doric. Both buildings, the so-called Hellenistic gymnasium (von Gerkan 1928: 1–21 and Emme 2013: 113–18, 345–46) and the courthouse by the temple of Athena on the Kalabaktepe (von Gerkan 1925: 88–91 and Emme 2013: 199–204, 348–49), are tentatively dated to the 2nd century BCE. On Kalabaktepe the only surviving lower drum of the Doric order is not fluted, and in the so-called Hellenistic gymnasium, where more blocks survived, unfluted shafts are combined with capitals that show flutes on the necking. If it had ever been intended to flute the columns entirely, one would expect another strip of flutes on the lower end of the bottom drums to act as a guideline, but such a band is not present. Nevertheless,

the columns could be perceived as having been left in a raw state of finish through the contrast of the flutes on the necking and the protective mantle on the shaft. Clearly we have to speak of an intended unfinished state, which in this case helped to underline the importance of the main facade. A similar design can be observed in one of the rockcut tombs in the so-called Tombs of the Kings cemetery at Nea Paphos in Kypros (for the nekropolis in general, see Hadjisavvas 1985). The tombs in this extensive northern nekropolis of Paphos are not well dated archaeologically or by inscriptions. Nevertheless, by the apparent similarities to the tombs of Alexandria, they can be readily identified as burial sites for rich families during the Hellenistic period, when Kypros belonged to the Ptolemies. The tombs of Paphos, houses for the dead in the literal sense, mirror the basic layout of urban dwellings, and the burial chamber opens off the peristyle’s most important wing opposite the entrance. This blueprint is perfectly reproduced in Tomb 4 of the nekropolis, where a dromos-like stairway leads down from the west to the central peristyle courtyard and the burial chamber opens off the east wing (Fig. 14.3) (Ross 1851: 325–26 pl. 28; Jeffery 1915: 168–69

234

figure 14.4

Grawehr

Miletos, South Agora Knackfus 1924, fig. 40

fig. 8; Greve 2014: 235). In this tomb, there is no Rhodian peristyle, but the hierarchy between the colonnades is made clear enough by a rough finish on the western side. Whereas on the main facade and the two lateral wings Doric columns have been carved out of the living rock, the four supports on the entrance side never received this final shape (Fig. 14.3). Instead, they were left as raw, hulking pillars. That the pillars indeed represent the initial stage of carving is indicated by the fact that they protrude slightly from underneath the architrave in exactly the same way as the abacus of the carved capitals on the other sides, unlike other tombs in Paphos that use only pillars. In my final examples the same tactics are used, but the monumental scale highlights the most important vista in a town square. The south agora of Miletos (Fig. 14.4) was surrounded on all four sides by long colonnades (Knackfuß 1924: 3–47 and Sielhorst 2015: 128–29). On its north, west, and south sides, the square was framed by a pi-like arrangement of two L-shaped porticoes of uniform height, whereas on the east, where the agora was traversed

by the Sacred Way, it was bordered by a long portico with three rows of rooms behind it. The east stoa was built first as a donation of Antiokhos I at the beginning of the 3rd century BCE. The revenues of the attached shops contributed to the construction of the temple in Didyma as indicated by the scarce remains of the dedicatory inscription and a much better preserved honourary inscription for Antiokhos (OGIS 213; Knackfuß 1924: 44. 281–82 no. 193a; Rehm 1958: 282–84 no. 479; Bringmann & von Steuben 1995: 338–41; Herrmann 1997: 199 no. 193a, pl. 13.2; and Meier 2012: 382–87). The traces on the surviving stylobate show that the Doric columns of this east stoa were fluted. With an axial width of 2.39 m and a lower diameter of 0.75 m they must have risen to a greater height than the Doric columns on the other sides of the square, whose axial width is 2.23 m and lower diameter 0.675–0.695 m. Only parts of their shafts have survived, but there is enough to demonstrate that they were left unfluted for their entire height. In the agora of Priene (Fig. 14.5) the building history is more complicated (Wiegand & Schrader 1904: 189–203; von Kienlin 2004; Filges & Kreuz 2013; and Sielhorst 2015:

LOOKING AT THE UNFINISHED: ROUGHED-OUT ORNAMENTATION IN GREEK ARCHITECTURE

figure 14.5

235

Priene, Agora, view from the Southeast, model once in the Antikensammlung, Staatliche Museen zu Berlin bpk/Antikensammlung, SMB

108–15). In an early stage the square was enclosed on its western, eastern, and southern sides by rows of rooms planned to be supplemented with porticoes in front of them. Whereas the stylobate was already laid out on all three sides, only the colonnades on the east and west wings had been erected by the middle of the 3rd century BCE (von Kienlin 2004, 39–40). Their Doric columns had a fluted necking below the capital with the unusual number of 24 flutes, but entirely unfluted shafts. In the second half of the 3rd century, priority was given to the erection of a separate stoa in the north, which is located on a slightly elevated terrace behind the street that passed along the northern side of the square (Wiegand & Schrader 1904: 215–17 and von Kienlin 2004: 45–53). Kienlin argues that the Doric frieze of the older North Stoa was later reused in the South Stoa, but he denies that the architrave was reused, because it was too short (2.4 m instead of the required 2.44 m). If this argument is valid, it would have to apply also for the frieze, where a correspondence between the axis of the columns and the triglyph is required. In the 130s this stoa was replaced by a more elaborate version donated by a Kappadokian king (Wiegand & Schrader 1904: 192–203, 214–216; Rumscheid 1994, 1:46; Bringmann & von Steuben 1995: 429–31; and von Kienlin 2004: 54–85). Because only very little of the dedicatory inscription has survived, the exact dating of the stoa between ca. 155–129 BCE is notoriously contested and the donation assigned to either Orophernes, Ariarathes V, or Ariarathes VI. Its Doric order boasted a rich ornamentation, a mixed Doric-Ionic entablature, and column shafts with the usual number of 20 flutes but of Ionic form with flat arrises. At about the same time the square’s south colonnade was finally built (Wiegand & Schrader 1904: 191–92; Rumscheid 1994, 2: 75 no. 306; Rumscheid 1998: 80–82; and von Kienlin 2004: 12– 42, 157–80). The rooms behind the colonnade are largely destroyed today, and there has been some confusion about

their arrangement (von Kienlin 2004: 34–35). It has been ascertained that between a series of rooms at both ends, the middle section was given to a large two-aisled hall, fenced off from the square by a screen wall between the columns of the colonnade. Of this screen wall and its columns, a considerable number of elements have survived showing diverse states of finish. To judge by the present location of the fragments, the first columns at both ends of the colonnade had 12 flutes executed on the face towards the square, while the inner half was smooth (Fig. 14.6a). In the portico in between, the lower part of the supports are articulated as half columns attached to the screen wall, with full columns emerging above. There the fluting was not finished but rather only an intermediary step in their preparation was executed: vertical strips had been dressed to indicate the later placement of the arrises (Fig. 14.6b). Again this treatment was only applied to the outer face of the columns, while the inner face of the top drum above the screen wall was left smooth. In the middle third of the colonnade even this preparatory step was omitted and the half columns are entirely enveloped by their thick protective mantle (Fig. 14.6c). At the lower end of each half column, a band of about 0.10 m height was smoothed. Finally, the middle pair of half columns that stood beside a central doorway in the screen wall have at their bottom a guiding band about 0.06 m in height where the fluting was executed (Fig. 14.6d). Obviously there is no way to explain the systematic arrangement of a wide array of different finishing stages as the consequence of an unforeseen interruption of a normal building process. Seen from inside, the colonnade, partly screened off by a blank wall, would have appeared to have entirely smooth columns. Seen from the outside, the colonnade looked quite different. At its extreme ends the first columns had fully executed fluting, which separated the south colonnade of the agora visually from the

236

Grawehr

a

c

b

figure 14.6

d

Priene, Agora, South Stoa, column drums photo by author

eastern and western colonnades with smooth shafts. What followed was a demonstration of the different working stages in the execution of a column’s fluting in reversed order symmetrically leading from both corners towards the middle of the colonnade. A subtle emphasis was given to the middle doorway where a guiding band of executed

flutings could be glimpsed at the foot of the columns. In a similar manner, the ashlars of the screen wall also showed a drafted margin around the raised and stippled panel. In my understanding this sophisticated and deliberate employment of a rough finish must be evaluated in the larger context of the whole square: it was used to highlight the

LOOKING AT THE UNFINISHED: ROUGHED-OUT ORNAMENTATION IN GREEK ARCHITECTURE

splendid new northern stoa as the focus of attention on the opposite side of the agora. 4

Principles at Work

The principle at work in all these examples is an attempt to apply a weighting factor to emphasise certain elements in the design of a building: the most important and richly decorated parts stick out from a markedly plain background, giving more contrast to the building’s visual appearance. If rich ornamentation can be regarded as especially labour intensive, it seems a logical choice to characterise the plainest parts within this design by the omission of the last working stages. The resulting effect can be described as a visual one: the viewer’s attention is guided by bringing the important elements into focus while blurring the comparatively featureless background. Finding the blossoming of this technique in the 3rd and 2nd century BCE fits neatly into our general understanding of Hellenistic architecture, where the pictorial effect often overruled the strict tectonic logic of the art of building (Lauter 1983: 297, 306–09; Bergmann 1988; and von Hesberg 1994a: 53–114). If we are looking more specifically for a theoretical framework within the limited corpus of Hellenistic and Roman art criticism that could motivate the choice for a raw finish, we may want to look to the concept of asperitas. This Latin equivalent of the Greek τραχύτης, meaning “rough- or ruggedness”, was used by Vitruvius to characterise the work of the Hellenistic architect Hermogenes. Vitruvius borrowed it, like most of his art theoretical terms, from the field of rhetoric, where it described a Hellenistic style of rugged, emphatic speech that was favoured in Asia Minor and frowned upon by the Roman Atticists (for asperitas as a style of speech, see, e.g.: Hermog. Peri ideon 254–60; on the topic in general: Martin 1974: 341–45; Pernot 2000: 112–15). According to Vitruvius, the architect Hermogenes introduced the pseudodipteral order to enhance the visual impact “through the harsh contrast of the intercolumnations” (propter asperitatem intercolumniorum, Vitr. De arch. 3.3.9). As this is our only source, the exact meaning of asperitas in the context of architectural theory is hard to pin down. Recently Lothar Haselberger and Samuel Holzman summarised the state of affairs and tested the current interpretations against the visual evidence provided by a new virtual model of Hermogenes’ pseudodipteral temple of Artemis at Magnesia (Haselberger & Holzman 2015). As a result, they convincingly described asperitas as the use of strong contrasts between light and shadow to make more

237

pronounced the three-dimensional qualities of colonnaded architecture. However, in a more comprehensive sense there have been other meanings of the term. Vitruvius himself used it in his famous polemic against anticlassicist movements in Early Augustan art (Vitr. 7.5.5–6). He describes how a certain Apatourios of Alabanda created a painting that first made a great impression on the audience by its asperitas, expressly explained as the bizarre combination of disparate architectural elements (and not as a chiaroscuro effect), but then this had to be altered because it rebelled against any structural logic (von Hesberg 1994a: 60 and Grüner 2004: 240–47). We may reasonably assume that Vitruvius imported the rhetorical term impromptu into his architectural theory. But if it had a longer history in the theoretical writings of Greek architects, could asperitas also encompass the contrasting effects between diligently ornamented parts on the one hand and only roughed-out areas on the other? The case may never be proven, but it is worthwhile to mention that it is exactly Hermogenes’ temple of Artemis at Magnesia that boasts at least eleven different variants of bolster ornamentation on the Ionic capitals (Humann, Kohte & Watzinger 1904: 56; Lethaby 1915: pl. 4–6; Rumscheid 1994, 1:204–06, 2:38; and Bingöl 1996: 150; 2007; 2012: 117), ranging from the most elaborate versions in the middle of the main facade and at the corners, to much plainer versions in the pronaos and opisthodomos, and finally to a version with the ornament cut only on the inner half of the capital in the cella, where also some bases and columns have been entirely left roughed-out and unfluted. To sum up, “unfinished” elements do not deviate from the ideals of ancient architecture, as they have been understood by the promoters of the neoclassical tradition (who therefore suppressed them in their drawings). Instead, paying attention to them can lead the way to a better understanding of fundamental design principles in Greek architecture. Employing a rough finish on some parts of a building to direct the viewer’s attention to other parts is one of these principles. List of References Bergmann, M., 1988: “Perspektivische Malerei in Stein: Einige alexandrinische Architekturmotive”, in Büsing, H.H. & Hiller, F. (edd.), Bathron: Beiträge zur Architektur und verwandten Künsten für Heinrich Drerup zu seinem 80. Geburtstag von seinen Schülern und Freunden (Saarbrücken) 59–77. Bingöl, O., 1996: “Zu Säule und Gebälk bei Hermogenes”, in Schwandner, E.-L. (ed.), Säule und Gebälk. DiskAB 6 (Mainz) 148–52.

238 Bingöl, O., 2007: “Die Westseite des Artemistempels in Magnesia”, in Öztepe, E. & Kadıoğlu, M. (edd.), Coşkun Özgünel’e 65. Yaş Armağanı (Istanbul) 113–18. Bingöl, O., 2012: “Neue Erkenntnisse am Tempel der Artemis Leukophryene in Magnesia”, in Schulz, T. (ed.), Dipteros und Pseudodipteros: Bauhistorische und archäologische Forschungen. Byzas 12 (Istanbul) 113–21. Boëthius, A. & Ward-Perkins, J.B., 1970: Etruscan and Roman Architecture (Harmondsworth). Borg, B.E. & Borg, G., 2002: “The History of Apollo’s Temple at Didyma, as told by Marble Analyses and Historical Sources”, in Lazzarini, L., Herz, N. & Newman, R. (edd.), ASMOSIA VI: Interdisciplinary Studies on Ancient Stone; Proceedings of the Sixth International Conference of the Association for the Study of Marble and Other Stones in Antiquity, Venice, June 15–18, 2000 (Padova) 271–78. Borg, B.E., 2001: “Marmor fur Apoll. Ein Beitrag zur Baugeschichte des jüngeren Didymaion und der Geschichte Milets und seines Territoriums”, in Bergemann, J. (ed.), Wissenschaft mit Enthusiasmus. Beiträge zu antiken Bildnissen und zur historischen Landeskunde, Klaus Fittschen gewidmet (Rahden) 79–102. Bringmann, K. & von Steuben, H., 1995: Schenkungen hellenistischer Herrscher an griechische Städte und Heiligtümer 1: Zeugnisse und Kommentare (Berlin). Chamonard, J., 1933: “Fouilles de Délos (Juillet–Septembre 1930)”, BCH 57: 98–169. Chrubasik, B., 2016: Kings and Usurpers in the Seleukid Empire: The Men Who Would Be King (Oxford). Emme, B., 2013: Peristyl und Polis: Entwicklung und Funktionen öffentlicher griechischer Hofanlagen. Urban Spaces 1 (Berlin). Fabricius, E., 1881: De Architectura graeca commentatio epigraphica prior (Berolini). Filges, A. & Kreuz, P.-A., 2013: “Der lange Weg zur Musterstadt: Zu Ausbau und Vervollständigung von Städten am Beispiel von Bononia (Bologna) und Priene”, in Haensch, R. & Wulf-Rheidt, U. (edd.), Dialoge über politische Räume in vormodernen Kulturen. Menschen—Kulturen—Traditionen 13 (Rahden) 55–78. Fragaki, H., 2015: “L’architecture alexandrine du IIIe s. a.C.: caractéristiques et tendances”, in des Courtils, J. (ed.), L’architecture monumentale grecque au IIIe siècle a.C. (Bordeaux) 283–304. Frejman, A., 2014: “Marginally drafted masonry as an aesthetic element”, in Karlsson, L., Carlsson, S. & Kullberg, J.B. (edd.), Labrys: Studies presented to Pontus Hellström (Uppsala) 237–50. Gadolou, A., 2008: “Η πρόσφατη αρχαιολογική έρευνα στο ναό του Διός Βασιλέως στη Λιβαδειά”, in Arvantinos, V. (ed.), Ἐπετηρὶς τῆς Ἑταιρείας Βοιωτικῶν Μελετῶν 4 (Athens) 547–65. Grawehr, M., 2015: “Römische Bossenkapitelle aus Travertin und lunensischem Marmor”, RM 121: 483–506.

Grawehr Grawehr, M., 2017: “Bossenstil und Baumaterial”, in Kurapkat, D. & Wulf-Rheidt, U. (edd.), Werkspuren: Materialverarbeitung und handwerkliches Wissen im antiken Bauwesen; Internationales Kolloquium in Berlin vom 13.–16. Mai 2015. DiskAB 12 (Regenburg) 103–18. Greve, A., 2014: Sepulkrale Hofarchitekturen im Hellenismus: Alexandria—Nea Paphos—Kyrene (Turnhout). Gruben, G., 1997: “Naxos und Delos: Studien zur archaischen Architektur der Kykladen”, JdI 112: 261–416. Grüner, A., 2004: Venus Ordinis: der Wandel von Malerei und Literatur im Zeitalter der römischen Bürgerkriege (Paderborn). Hadjidakis, P., Matarangas, D. & Varti-Matarangas, M., 2009: “Ancient Quarries in Delos, Greece”, in Maniatis, Y. (ed.), ASMOSIA VII. BCH Suppl. 51 (Paris) 273–88. Hadjisavvas, S., 1985: “Excavations at the Tombs of the Kings, Kato Paphos”, in Karageorghis, V. (ed.), Archaeology in Cyprus: 1960–1985 (Nikosia) 262–68. Haselberger, L. & Holzman, S., 2015: “Visualizing asperitas: Vitruvius (3.3.9) and the ‘asperity’ of Hermogenes’ pseudodipteral temple”, JRA 28: 371–91. Herrmann, P., 1987: “Milesier am Seleukidenhof”, Chiron 17: 171–92. Herrmann, P., 1997: “Teil 1: A. Inschriften n. 187–406”, in Rehm, A. (ed.), Inschriften von Milet: Nachträge und Übersetzungen zu den Inschriften n. 1–406. Milet 6.1 (Berlin) 153–217. Hodge, A.T. & Tomlinson, R.A., 1969: “Some Notes on the Temple of Nemesis at Rhamnous”, AJA 73.2: 185–92. Hoepfner, W., 1971: Zwei Ptolemaierbauten: Das Ptolemaierweihgeschenk in Olympia und ein Bauvorhaben in Alexandria. AM-BH 1 (Berlin). Hoepfner, W., 1990: “Von Alexandria über Pergamon nach Nikopolis. Städtebau und Stadtbilder hellenistischer Zeit”, in Akten des 13. Internationalen Kongresses für Klassische Archäologie (Mainz) 275–85. Humann, C., Kohte, J. & Watzinger, C., 1904: Magnesia am Maeander: Bericht über die Ergebnisse der Ausgrabungen der Jahre 1891–1893 (Berlin). Jeffery, G., 1915: “Rock-cutting and Tomb-architecture in Cyprus during the Graeco-Roman Occupation”, Archaeologia 66: 159–78. Kalpaxis, T.E., 1986: Hemiteles: Akzidentelle Unfertigkeit und „Bossen-Stil” in der griechischen Baukunst (Mainz). Kleiner, F.S., 2010: A History of Roman Art (Boston). Kleiner, G., 1968: Die Ruinen von Milet (Berlin). Knackfuß, H., 1908: Das Rathaus von Milet. Milet 1.2 (Berlin). Knackfuß, H., 1924: Der Südmarkt und die benachbarten Bauanlagen. Milet 1.7 (Berlin). Knackfuß, H., 1941: Die Baubeschreibung in drei Bänden. Didyma 1 (Berlin). Kockel, V., 1995: “Bouleuteria: Architektonische Form und urbanistischer Kontext”, in Wörrle, M. & Zanker, P. (edd.), Stadtbild

LOOKING AT THE UNFINISHED: ROUGHED-OUT ORNAMENTATION IN GREEK ARCHITECTURE und Bürgerbild im Hellenismus: Kolloquium, München, 24. bis 26. Juni 1993. Vestigia 47 (Munich) 29–40. Koenigs, W., 1980: “Ein archaischer Rundbau im Kerameikos”, in Koenigs, W., Knigge, U., Mallwitz, A. & Bohnen, B. (edd.), Rundbauten im Kerameikos. Kerameikos 12 (Berlin) 57–94. Krischen, F., 1941: Antike Rathäuser (Berlin). Lauter, H., 1983: “Künstliche Unfertigkeit: Hellenistische Bossensäulen”, JdI 98: 287–310. Le Roy, D., 1758: Les ruines des plus beaux monuments de la Grèce (Paris). Lethaby, W.R. (ed.), 1915: Antiquities of Ionia Published by the Society of Dilettanti: Part the Fifth Being a Supplement to Part III (London). Liljenstolpe, P., 2000–2001: “Rustication and Decor in Roman Architecture. Their Reflection in the Architecture of the 16th Century With Special Attention to Their Use in the Classical Orders”, OpRom 25–26: 45–72. Martin, J., 1974: Antike Rhetorik: Technik und Methode (Munich). Mattern, T., 2008: “Ein Vermächtnis Alexanders des Grossen? Antiochos IV und drei monumentale hellenistische Tempel”, in Franek, C., Lamm, S., Neuhauser, T., Porod, B. & Zöhrer, K. (edd.), Thiasos: Festschrift für Erwin Pochmarski zum 65. Geburtstag (Vienna) 617–25. Meier, L., 2012: Die Finanzierung öffentlicher Bauten in der hellenistischen Polis. Die hellenistische Polis als Lebensform 3 (Mainz). Müller-Wiener, W., 1988: Griechisches Bauwesen in der Antike (Munich). Partida, E., 2015: “Architectural elements and historic circumstances that shaped the sanctuary of Delphi during the so-called ‘Age of the Warriors’”, in Jacques des Courtils (ed.), L’architecture monumentale grecque au IIIe siècle a.C. (Bordeaux) 29–50. Pernot, L., 2000: La Rhétorique dans l’Antiquité (Paris). Rawson, E., 1975: “Architecture and Sculpture: The Activities of the Cossutii”, PBSR 43: 36–47. Rehm, A., 1958: Die Inschriften. Didyma 2 (Berlin). Ross, L., 1851: “Phönicische Gräber auf Cypern”, AZ 9: 321–28.

239

Rumscheid, F., 1994: Untersuchungen zur kleinasiatischen Bauornamentik des Hellenismus. 2 vols. (Mainz). Rumscheid, F., 1998: Priene: Führer durch das „Pompeji Kleinasiens” (Istanbul). Schaaf, H., 1992: Untersuchungen zu Gebäudestiftungen in hellenistischer Zeit (Köln). Schwandner, E.-L. & Kolonas, L., 1996: “Beobachtungen am Zeusheiligtum von Stratos”, IstMitt 46: 187–96. Sielhorst, B., 2015: Hellenistische Agorai: Gestaltung, Rezeption und Semantik eines urbanen Raumes. Urban Spaces 3 (Berlin). Stuart, J. & Revett, N., 1762: Antiquities of Athens 1 (London). Stuart, J. & Revett, N., 1787: Antiquities of Athens 2 (London). Thomas, E., 2015: “The Beauties of Architecture”, in Destrée, P. & Murray, P. (edd.), A Companion to Ancient Aesthetics (Chichester) 274–90. Trümper, M., 1998: Wohnen in Delos: eine baugeschichtliche Untersuchung zum Wandel der Wohnkultur in hellenistischer Zeit (Rahden). Turner, L.A., 1994: The History, Monuments and Topography of Ancient Lebadeia in Boeotia, Greece (diss., University of Pennsylvania). von Gerkan, A. & Krischen, F., 1928: Thermen und Palästren. Milet 1.9 (Berlin). von Gerkan, A., 1925: Kalabaktepe, Athena-Tempel und Umgebung. Milet 1.8 (Berlin). von Hesberg, H., 1994a: Formen privater Repräsentation in der Baukunst des 2. und 1. Jahrhunderts v. Chr (Cologne). von Hesberg, H., 1994b: “Bogenmonumente und Stadttore in claudischer Zeit”, in Strocka, V.M. (ed.), Die Regierungszeit des Kaisers Claudius (41–54 n. Chr.): Umbruch oder Episode? Internationales interdisziplinäres Symposion aus Anlass des hundertjährigen Jubiläums des Archäologischen Instituts der Universität Freiburg i. Br., 16.–18. Februar 1991 (Mainz) 245–60. von Kienlin, A., 2004 (March): Die Agora von Priene, https:// mediatum.ub.tum.de/doc/601008/601008.pdf. Wiegand, T., & Schrader, H. (1904): Priene: Ergebnisse der Ausgrabungen und Untersuchungen in den Jahren 1895–1898 (Berlin).

part 4 Simulation, Experience, and Interaction with Greek Architecture

∵

chapter 15

Contexts for Greek Architecture: Places and People Mary B. Hollinshead 1 Introduction Context is essential to archaeological analysis. To most archaeologists, contextual information for architecture comes from artefacts and their distribution, and from inscriptions and texts, when such information is available. However, context for architecture also encompasses a more comprehensive view of terrain, pathways, and the placement of structures in relation to the lay of the land and to each other, the built environment. Since monumental architecture is intrinsically expressive of social, economic, and political priorities, our ability to understand it requires maximising information about context, both for the architecture we study and for those of us who study architecture. We aim to explain not only buildings but also the practices and attitudes of those who made and used them, as shaped by the built environment. The rich spatial information afforded by tools such as Geographic Information Systems (GIS) and Virtual Reality (VR) in combination with site-specific evidence of movement greatly enhances our ability to synthesise and interpret archaeological sites. Manipulating and simulating the forms and effects of sites, spaces, and structures can provide evidence for behaviour so as to understand better social practices, patterns, and even values. This study addresses the foundational concepts of GIS and VR to offer a more socially oriented mode of analysing the built environment of ancient Greece. Assessing premises and interpretive processes of GIS and VR—data production, selection, classification, representation—permits us to recognise our own conceptual frameworks for research. Attempting to reproduce always-incomplete information in four dimensions (that is, three-dimensional space plus time) is further complicated by varied scales of evidence. Archaeological analysis and interpretation must be commensurate with the scope and scale of data from material studied. Wheatley and Gillings note the problem of “how to explain large-scale spatial and temporal patterns and processes without divorcing them from the smaller scale actions that created them” (Wheatley & Gillings 2000: 8; cf. Lock & Harris 2000 [xviii–xxi]), while Lock & Molyneaux (2006: xi, and cf. 1–11) comment, “archaeological analysis has two general referents: the culture of

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004416659_017

production and the culture of interpretation. The challenge of the archaeologist is to understand the dynamics of scale that entered into production and to account for these in interpretation”. GIS, for example, tends to express information that extends over a larger area and time span than the smaller, more immediately human scale of VR, a difference that needs to be represented in the archaeologist’s interpretive framework. This study focuses primarily on the human scale, so as to look at evidence of behaviour in antiquity. GIS and VR are mighty tools, instruments of methodology with approaches that have been considered at odds with each other, attributable to fundamentally different epistemologies of archaeology. Beyond basic descriptions of these tools, I shall address the premises that underlie them and their use in order to explore a rapprochement— how to think about the metaphorical and literal middle ground between the macro focus and analytical approach of GIS, and the more individual scale, phenomenologically oriented approach of VR (on rapprochement, see Gillings & Goodrick 1996: 1.1–1.5; Witcher 1999; Lock & Harris 2002: xxi–xxiii; Chadwick 2004: 21; Cripps 2008: 150–51; Wheatley 2014: 124; also: Lake 2007; Lake & Ortega 2013: 213–15). From there we can consider how people interacted with their built environment. GIS and VR can be seen to embody the age-old issue of defining space and place, as well as more recent professional divides between positivist and post-processual archaeologists. With its quantifiable data and verifiable relationships among geographic factors, GIS has been seen to represent an empiricist, analytical approach to studying environments. In contrast, VR can be seen to offer a more overtly subjective and interpretive framework for explaining environments and their use (see Lake & Woodman 2003; Chrysanthi, Murieta Flores & Papadopoulos 2012: 8; Wheatley 2014). How we characterise space and people’s interaction with it is reflected in archaeological practices and interpretation. Space—when considered as undifferentiated and abstract in the Euklidean, or Cartesian approach—is a universal, measurable backdrop for human activities (Tilley 1994: 7; Witcher 1999: 14; Wheatley 2004: 1.1). GIS is intended to generate the measurable aspects of that backdrop in complex relationships. Place, considered

244

Hollinshead

defined and particular, is often described (though not by all) as a subset of space generated by human perception based on lived experience—that is, the phenomenological approach (Tilley 1994; 2004; Brück 2005). Casey (2001: 404– 05) envisions place as conceptually different from space). Place is characterised by its relational properties, in contrast to the absolute properties of space, and by the presence of human agents (Newsome 2011: 21; Preucel & Meskell 2004: 215). Although societal processes and historical circumstances have a pervasive effect on the perception, cognition, and actions defining place, proponents of this latter approach emphasise the direct human experience of place (on the historical contingency of perception, see Pred 1984; Zedeño 2000). Ian Gregory distinguishes between absolute space, the Euklidean measure of distance on a flat plane; and relative space, that “uses topology to show where different features are in relation to each other using connectivity” (Gregory 2008: 136). Combining the Euklidean space of GIS with the phenomenological place of VR, together with archaeological evidence of people’s mobility provides for enriched interpretation of Greek architecture in context, including both physical and social contexts. 2

Geographic Information Systems (GIS)

GIS, now widely used by field archaeologists, expresses essential topographical information as well as a variety of data selected for a given project—such as geology, hydrology, paths and roads, built features, political boundaries, population density, historical circumstances, and so on. Text-based information can be incorporated into an image as long as it lends itself to spatial representation. Each category of information is stored electronically as a separate (conceptual) layer. Originally developed to analyse natural resources across vast territories, GIS is now used for countless studies ranging in scale from continents to single monuments. Data is organised in either a vector-based format or encoded as raster data. Vectors represent features by means of points, lines, or polygonal areas, and so are best suited for displaying precisely defined elements. Rasters represent information as a grid of pixels or cells, allowing them to express gradations of data over the pictorial space. Using either approach, discrete units of information from electronic devices (such as vector points measured with a Total Station) can be incorporated into a GIS and then organised as layers into a relational database. The stored information may be manipulated so as to demonstrate spatial relationships, topology, and much more. Fundamental to most users, especially archaeologists, is

the capacity to establish a Digital Elevation Model (sometime called a Digital Terrain Model), a raster representing some topographical configuration, a reconstruction of the “lay of the land”, and a basis upon which other data can be related in various combinations so as to facilitate analysis and modelling (Marble 1990; Wheatley & Gillings 2002; Wheatley 2004; Conolly & Lake 2001; Ott & Swiaczny 2001: 148). Tomlin (2013: 27) notes “raster structures generally tend to be better suited to the interpretation of where, while vector structures tend to be better suited to the interpretation of what”. Most Digital Terrain Models incorporate limited elevation data (based on single points) and so are not fully three-dimensional renderings, even though they are presented so as to appear 3D; they are conventionally called 2.5D images. The capabilities of GIS have allowed its users to identify what is meaningful in the spatial arrangement of archaeological remains, including both the natural and built environments, what David Wheatley has called “an archaeology of place” (Wheatley 2004: 1.1). Associating quantifiable, tangible variables in all manner of combinations and relationships is like bringing maps to life, as seen, for example, by models at Messene or Korinthos, where one can construct a GIS model with customised features accessible through the site’s website (for Messene: Tokmakidis, Kalyvioti & Nanakou 2004; for Korinthos: https://www.ascsa.edu.gr/excavations/ancient-corinth/ digital-corinth/maps-gis-data-and-archaeological-data -for-corinth-and-greece). This is a powerful resource, intellectually and visually. However, the enthusiastic reception of GIS by archaeologists has raised an array of criticisms. At the most basic level of data entry, there is no straightforward means to express ambiguity or uncertainty, requiring a degree of interpretation from the beginning of the process. In addition, terrain is typically represented, but vegetation—a key component of ancient lives and a key factor in visibility— cannot be mapped so precisely, especially in the vertical dimension (Wheatley & Gillings 2000: 5–6; Llobera 2007a; Gregory 2008: 139–46; Cripps 2008: 152–53). In GIS, time is treated as a separate variable from space (although sequential points have been used to simulate motion), limiting effective representation of movement, which is how people actually experience space—as will be considered below (Witcher 1999: 14; Cripps 2008: 152; Gregory 2008). Separate layers of data can express the same location at different phases of time, reflecting spans of time on a larger scale than individual experience. Layers representing different phases can be overlaid so as to demonstrate change—but not motion. Space is presented as absolute,

CONTEXTS FOR GREEK ARCHITECTURE: PLACES AND PEOPLE

neutral, objective, a setting for human activities. Data and measurement are central. Consequently, landforms and structures are the primary subjects, and human behaviour secondary. To humanise GIS, researchers have extended its applicability to social behaviours, such as by identifying “least cost paths” and developing visibility studies. On the one hand, assessing least cost paths has been criticised as ineffective in studying ancient sites, undoubtedly because of our ignorance of nuances of local mores and priorities in antiquity (Conolly & Lake 2001: 234–62; Déderix 2015; Gibson 2007: 81–82; Herzog 2013). On the other hand, incorporating “viewshed” and “intervisibility” analyses into terrain models of landscape archaeology and “isovists” in architectural studies uses the three-dimensional capacity of GIS to re-create at the human scale what people could see in various settings (for dynamic viewsheds see Cripps 2008: 152; Wheatley & Gillings 2000: 6–7; Llobera 2003, 2007b; Wheatley 2014). Studies such as Eleftheria Paliou’s re-creation of fields of vision at Bronze Age Akrotíri, Thera and medieval San Vitale at Ravenna bring the power of GIS modelling to the level of human experience by documenting what could be observed from specific viewpoints (Paliou 2011; 2013; 2014b). Nevertheless, a literal worldview is not a metaphorical worldview, full of meanings, attitudes, and values. Indeed, Paliou acknowledged that “human experience cannot be solely reduced to a set of measurable spatial properties” (Paliou 2014a: 8). Gary Lock and Trevor Harris observed that despite these and other efforts, there are limits to using “quantitative software to incorporate qualitative experiences” (Lock & Harris 2000: xiv). Lock added the comment that “we are trying to incorporate social and cultural information into the landscape itself, whereas it actually resides within people” (Lock 2000: 62). Designed and used primarily to address issues of large scale, GIS describes space effectively but is less successful in describing place, at the level of the individual. 3

Virtual Reality (VR)

Virtual Reality, the use of digital video technology to generate images that simulate lived experiences, aspires to present place. Archaeological data is expressed by means of a digital abstraction that is a spatial representation of volumes, voids, and distances, with varying degrees of detail. Some VR, but far from all, includes human figures. With rapid advances in technology, the scope of computerised reconstruction has expanded dramatically,

245 including Augmented Reality, combining photographic images (single or mosaic) with computer-generated reconstructions, and Immersive Reality, a 360 degree, three-dimensional arrangement that situates the observer within a specific re-created setting (Barceló 2000: 26; Argyriou & Papadopoulos 2015). Goodrick and Gillings noted that aside from technical differences, the key distinction among the various forms of Virtual Reality lies in “the desired degree of collapse between user and representation” (Goodrick & Gillings 2000: 46). The process of making a VR model is so complex (and expensive) and the outcome so compelling that generating an expression of a site in VR can be seen as a terminal goal, and the model itself as authentic. Assembled from a fragmentary archaeological record, the VR model makes a site whole, providing it with an appealing sense of coherence. Moreover, because their pictorial nature makes them more accessible than maps and plans, VR renderings are often seen as more useful for public information, and their value as research tools is underestimated (Favro 2006; Clark 2010). In studying ancient architecture, re-creating context through VR permits the viewer an approximation of topographic and geographic features, of buildings and spaces, and their relationships at a variety of scales—including the human scale— offering a more directly experiential replication of how individuals, groups and populations might have perceived aspects of an ancient site (Goodrick and Gillings 2000). “Visualisation changes the user’s role from passive to active” (Ott & Swiaczny 2001: 144). As with reconstruction drawings and physical models, the digital models of VR combine abstract distillation and simplification of known data with conjectural information to complete the picture, both literally and figuratively. VR representation has been criticised as fundamentally reductionist, simplifying what exists in the archaeological record and obscuring what does not—yet that can be said of any reconstruction created by any means (Gillings 2005; Smiles & Moser 2005; Haselberger & Humphrey 2006: 7–11, 335–37; Favro 2006; Vouzaxakis 2006; Clark 2010; Chrysanthi, Murieta Flores & Papadopoulos 2012). Total verisimilitude is not only impossible, but undesirable, as the purpose of archaeological VR is to represent essential information relevant to the archaeologist’s agenda, not to re-create antiquity. Herein lies much of the ambivalence scholars feel about VR. The design of any VR model needs to suit its audience. For research purposes, details can be schematic; depicting how people in antiquity might have moved around a particular site requires less attention to faces or clothing for example. On the other hand, VR

246 models for didactic purposes often look more complete, as they present a visual “record” that summarises a site for the non-expert. A venture into creating a VR model of the sanctuary of Demeter and Kore on the slopes of Akrokorinthos that began as part of my research on ancient monumental steps has demonstrated the interpretive demands and explanatory potential of Virtual Reality modelling. The design work, carried out in 2008 by Ron Hutt, my colleague at the U. of Rhode Island, was generated primarily with Maya software, together with Unity for the game-like processes. This well-published site, with several phases of use as a centre for ritual and dining, is rich in contextual information from artefacts and patterns of deposition (Fig. 15.1). A series of dining chambers arrayed on both sides of a broad stairway led up to a larger, apparently ceremonial building and open-air platform with rock-cut stepped seating facing onto it from above (Bookidis & Stroud 1997; Hollinshead 2015: 46–47; Sanders et al. 2017). Few of the many walls are preserved above one meter in height. The plans and photographs of the site report are clear and precise, but they convey little of the sloping terrain or the relationship between spaces and structures in the sanctuary complex. Representing the site around the end of the 5th century BCE, our video aims to reenact the experience of visitors to the sanctuary, as they might have moved from the roadway at the lower, north end, past a seated woman and girl holding terracotta votives, past three musicians at the foot of the stairway, then looking and turning uphill onto the staircase with landings leading to the dining rooms, and finally ascending to the ritual location at the upper, southeast sector of the sanctuary (Fig. 15.2). Our original plan was to create a model of the site, whose hillside location would have affected people’s movement and patterns of use. Realising that we could reconstruct ritual behaviours and re-create people’s perception of their surroundings more precisely using game-based software, we added human figures and put them in motion in the model. Favro (2013: 168) notes that “[e]ffective psychoenvironmental analyses are impossible without human subjects”. Eventually we started to add voice-overs to describe what the figures were doing and why. Archaeologist and designer together envisioned an outcome that would be an activated, accurate model of information published in the site reports while also presenting an aesthetically pleasing, potentially interactive video that would engage any visitor to the actual or virtual site. Over time we recognised the complexity of that dual goal. Details of the seated figures and their accoutrements, and the musicians (based on artefacts from the site and textual information from elsewhere) enhance the visual

Hollinshead

effect and suggest social context, but are time-consuming to render, and in fact are relatively inconsequential for the archaeologist’s research agenda, to depict how the spaces and structures of the sanctuary were perceived and used by worshippers. Those details do contribute substantially to the narrative appeal and its explanatory function. The project remains unfinished due to a shortage of resources (time, personnel) to achieve both didactic and research purposes; nevertheless, it demonstrates that key spatial concepts can be conveyed by an incomplete model. VR as a research tool serves to address defined questions and need not require the more photo-realistic representation of a model for the general public (Goodrick & Gillings 2000: 44–45; Clark 2010). Favro (2013: 162) distinguishes between purposes of knowledge transfer associated with immersive reconstructions versus knowledge production for academic research, while acknowledging the value of both. The building of any VR model requires many choices and great detail. Unlike GIS, the archaeologist serves in a capacity analogous to the Total Station, providing quantities of primary data by selecting information from site reports, photographs, maps, and elsewhere, involving even more interpretive judgments than GIS. However, that role is already diminishing with the recent introduction of software (such as REVEAL and half a dozen others) that promises to reconstruct a vision of features even as their data are recorded and photographed on handheld devices during the process of excavation (Sanders 2014: 42–46; http://www.vizin.org/projects/reveal/project.html). VR offers a huge range of scales, from drone-like flyovers of surroundings and site to that of an individual person. At the human scale (with humans represented) VR simulates the experience of terrain, buildings, open areas in topological relationships and attempts to approximate the perception of place in antiquity. VR thus is more conducive to the phenomenological approach to archaeological explanation, as most fully advocated by Christopher Tilley (Tilley 1994; 2004). Tilley and others have argued that people understand and engage with the world through embodied experience, through the situational and sensory effects of their physical selves in a given setting (Bender 1993; Thomas 1993). People’s engagement with the world emerges from landscapes, monuments, and artefacts, as well as from social interaction. For Tilley and colleagues, archaeologists understand ancient sites by themselves walking the terrain, employing body and sense as the medium of interpretation, always grounded in extant material remains (Tilley 2004: 219). Although vision is the dominant sense in VR, as in GIS, the aim is larger, to reproduce how a site was perceived.

CONTEXTS FOR GREEK ARCHITECTURE: PLACES AND PEOPLE

figure 15.1

Korinthos, Sanctuary of Demeter and Kore: (a) Plan of site, ca. 400 BCE; (b) Photograph of site from northwest, 1972 BOOKIDIS & STROUD 1997 PLAN 4 AND PLATE 2, COURTESY OF THE AMERICAN SCHOOL OF CLASSICAL STUDIES AT ATHENS, CORINTH EXCAVATIONS

247

248

figure 15.2

Hollinshead

Sanctuary of Demeter and Kore: (a) general view of sanctuary from north; (b) woman and girl at entrance to sanctuary still images from video by Hollinshead and Hutt

CONTEXTS FOR GREEK ARCHITECTURE: PLACES AND PEOPLE

4 Perception Perception is more than vision, and, indeed, it is more than sensory. Many scholars have noted the limitations of what Goodrick and Gillings called “ocularcentric” interpretation (Gillings & Goodrick 1996: 1.4; Goodrick & Gillings 2000: 44; Hamilakis 2002: 122; Chadwick 2004: 2; Gillings 2005: 229–30, 233; Chrysanthi, Murieta Flores & Papadopoulos 2012: 10; discussed more fully by Frieman & Gillings 2007; Wheatley 2014: 121–24; for a defence of prioritising visual information see Llobera 2007b: 51–54). It includes stimuli from various senses, and also how people make use of that input in light of their own cognitive array, such as memory, previous experiences, training, and expectations. It has been described as a “mediator between the world and the mind” (Witcher 1999: 17). Any environment, but especially the built environment, furnishes information encoded with meanings that are rooted in cultural and historical circumstances. Joanna Brück has noted “the act of perception is also an act of interpretation”, as humans prioritise among the integrated sensory and cultural messages when they perceive a situation (Brück 2005: 56; Frieman & Gillings 2007). Edward Casey commented, “Even the most primordial level of perceiving is inlaid with cultural categories in the form of differential patterns of recognition, ways of organising the perceptual field and acting in it, and manners of designating and naming items in this field” (Casey 1996: 34; see also Ashmore & Knapp 1999: 8). A major critique of phenomenology in archaeology is that it projects modern constructs of past societies into the explanation of how ancient people viewed their world. Critics ask whether there is indeed such an entity as a universal human body—just who is doing the perceiving? Is it a white Western male who is embodied in phenomenological explanations? (Goodrick & Gillings 2000: 54; Hamilakis, Pluciennik & Tarlow 2002: 5–10; Chadwick 2004: 22; Brück 2005: 54–55, 58–59; Gillings 2005: 233; Frieman & Gillings 2007: 6) Another criticism, that human agency can be understood primarily through the physical world, may indicate why archaeologists look favourably on phenomenology—especially since they are the ones doing the perceiving. That said, VR with human figures can portray relationships between people, among groups, with landforms and the built environment, and how people moved among them. 5 Mobility Virtual Reality simulates perception by integrating time with space. People engage in placemaking through their activities in a location. Activities, whether habitual or

249 exceptional, involve action that occurs over time. Moving is how humans experience, perceive, and interpret terrain and vegetation, structures and spaces—the forms and textures of where they are—in addition to other people, animals, and objects. Tilley commented “perception is a fundamentally ambient activity” (Tilley 2004: 26; also see Wheatley & Gillings 2000: 4, 6–7). For three-dimensional buildings and their environs, mobility is inherent in the process of perception. Mobility has become a popular topic in recent years, a logical outcome of the so-called “spatial turn” since the 1980’s in the social sciences. Expanding the scope of research in geography, sociology, anthropology, and archaeology from static concepts of space to its activation through human use, in 2006 Sheller & Urry defined the “new mobilities paradigm” that has been applied to communities from prehistoric Britain to international airports, on scales from personal pathways to global migrations (Sheller & Urry 2006; also see, among many, Urry 2007; Adey 2010; Creswell 2010; Beaudry & Parno 2013). Tilley, Ingold, and other proponents of phenomenological interpretation had previously noted the importance of time and motion in people’s experience of terrain and the built environment (Ingold 1993; 2011; Tilley 1994: 28–9; 2004: 26). The “mobility turn” can be seen as an intellectual extension of the “spatial turn” together with phenomenology that concentrates on the somatic experience of surroundings. One can only speculate as to the degree of correlation between the advanced mapping capability of GIS and the “spatial turn”, followed by the widespread proliferation of computer games with sophisticated digital modelling and lively video, fostering interest in the “mobility turn” across many fields of endeavor. Roman scholars have recognised the value of studying mobility and associated perceptions since at least the 1990’s. Diane Favro’s “walk through Augustan Rome” is well known (Favro 1996: 24–41, 252–80). Laurence and Newsome’s Rome, Ostia, and Pompeii: movement and space and subsequently Ida Östenberg, Simon Malmberg, and Jonas Bjørnebye’s The Moving City: Processions, Passages, and Promenades in Ancient Rome have continued and extended the theme (Laurence & Newsome 2011; Östenberg, Malmberg & Bjørnebye 2015). VR projects pioneered by Bernard Frischer representing Rome and the Villa of Hadrianus have inspired many others https:// www.romereborn.org/; http://vwhl.soic.indiana.edu/villa/ index.php; also see Haselberger 2006: 8; and the papers in that volume). In the Greek realm and beyond, Bonna Wescoat and Robert Ousterhout promoted greater recognition of movement and its meaning in Architecture of the Sacred. Space, Ritual, and Experience from Classical Greece to Byzantium (Wescoat & Ousterhout 2012). Wescoat’s

250

Hollinshead

VR model of a visit to the sanctuary of the Great Gods on Samothrake is distinguished by its consistent concentration on the human-scale procession through the topography and the structures of the site, offering a reenactment of a visitor’s experience (https://samothrace .emory.edu/visualizing-the-sanctuary/; also see Chapter 19 below). Joan Breton Connelly’s study of ritual kinesis cited processions, dances, and footraces as examples of meaningful movement in Greek sanctuaries (Connelly 2011). Eftychia Stavrianopoulou recognised “the dynamic character of processions, particularly with respect to the formation of space through movement”, and other works on processions include or imply themes present in mobility studies, whether or not they are articulated as such (Stavrianopoulou 2013: 352; Kavoulaki 1999; Mylonopoulos 2006: 103–09; Hollinshead 2015: 29–32). 6

Pathways and Steps

In studying monumental steps, primarily in Greek sanctuaries, I have encountered many physical provisions for movement. Steps were built in sanctuaries to support processions to sacrifice and promote group viewing of rituals, actions that in turn were made permanent in the stone features of the steps. They are pathways and destinations, created and animated by their users, whose perception of their worlds would have been profoundly affected by where they walked and with whom. Tim Cresswell (2010: 20) noted “human mobility is … enacted and experienced through the body”. By using the physical characteristics of the human body, we find that many steps reveal their ancient roles by their absolute and relative dimensions: in height, steps for walking are ideally about 0.18 m (7 in), while those over 0.30 m (ca. 18 in) are for sitting (Fig. 15.3). Sitting requires greater depth (ca. 0.35–0.40 m) There are also arrangements with steps shallow enough that they must be for standing, and some steps clearly served multiple purposes (Hollinshead 2015: 19–24). Such practical details of posture reveal how people behaved in specific places, and so provide evidence of social patterns. Monumental pathways, including both steps and ramps, to and within Greek sanctuaries demonstrate with their breadth and their permanent stone construction where important processions (pompai) took place, sometimes offering tangible confirmation of textual and iconographic testimonia that describe colourful, festive crowds gathered from far and near in processions to sacrifice. A visitor to any of these sites on a quiet day would envision the scope of processions when viewing the constructed route. During festivals, the worshippers in parade would

constitute a performance, generated by the energy from the co-presence of fellow participants moving together towards the altar-temple complex, the source of meaning for all this movement. As early as the sixth century BCE, a broad ramp ca. 11 m wide led up to the Athenian Akropolis, to accommodate the well-documented Panathenaic procession; by the 5th century BCE the breadth was nearly doubled, to over 21 m wide (Neils 1996; Shear 2001; Stavrianopoulou 2013: 352–53). Sixth century BCE ramps over 7 m wide led up to the temple of Apollon at Korinthos (Fowler & Stillwell 1932, 204, 219–20; Sanders et al. 2017) and approached the sanctuary of Zeus on Aigina (Goette 2001, 345–48), evidently providing for large groups moving towards ritual locations. Access up and down slopes and erosion control may have made such structures necessary, but their grand form, enduring material, and careful construction would have announced the scale and significance of ceremonies even when empty. In addition to expressing communal celebration, built pathways articulated—and controlled—how participants were intended to experience a site: how persons or groups perceived monuments and structures, and in what sequence (Tilley 1994: 29–31; Gibson 2007; Snead, Erickson & Darling 2009; Lorimer 2010; Stavrianopoulou 2013). At the sanctuary of Zeus at Labraunda in Karia two sets of stairs over 10 m wide led to terraces linked by additional broad steps (Fig. 15.4). Although most of the buildings in the sanctuary were constructed of local gneiss, two propylaea at the start (bottom) of the processional path were made of marble, and facades of the monumental andrones (dining structures) that dominate each of two major terraces were of marble, as well as the facades of at least three additional structures facing the pathway up to the temple terrace, thus presenting an elegant front to worshippers ascending in procession (www.labraunda .org; Hollinshead 2015: 57–59). Here, the architectural enhancement is directly reflective of social practice animating the spaces of the sanctuary. Our mobility turn in interpretation means greater attention to spaces as well as structures as meaningful social locations. What looks like a comparable architectural form at the sanctuary of Athena at Alipheira in the Peloponnesos, the Argive Heraion, and the sanctuary of Apollon at Korinthos—each with broad ascending routes of access arriving at the same location, between temple and altar—takes on a richer significance if we envision in each case a communal procession with worshippers, musicians, and sacrificial animals pouring into the sacred space for the drama of sacrifice and subsequent feasting (Fig. 15.5) (Alipheira: Orlandos 1967–1968: 9, 29, 43–45; Argive Heraion: Pfaff 2003: 5–8; Korinthos: Sanders et al.

CONTEXTS FOR GREEK ARCHITECTURE: PLACES AND PEOPLE

figure 15.3

Postures on steps according to dimensions and proportions Hollinshead 2015, 22

251

252

figure 15.4

Hollinshead

Labraunda, Sanctuary of Zeus: (a) plan of sanctuary; (b) photograph of lower steps and great stairway from south (a) COURTESY OF THE LABRAUNDA EXCAVATIONS (b) Hollinshead 2015, pl. 24b

CONTEXTS FOR GREEK ARCHITECTURE: PLACES AND PEOPLE

a

figure 15.5 Plans of three sanctuaries: (a) Alipheira, of Athena; (b) Argive Heraion; (c) Korinthos, of Apollon and central area (a) HOLLINSHEAD 2015, pl.3a (b) PFAFF 2003, PLAN 1, COURTESY OF THE TRUSTEES OF THE AMERICAN SCHOOL OF CLASSICAL STUDIES AT ATHENS (c) WILLIAMS & FISHER 1971 FIG. 5, COURTESY OF AMERICAN SCHOOL OF CLASSICAL STUDIES AT ATHENS

b

c

253

254 2017; Hollinshead 2015: 44). Ranks of steps adjacent to altars at Perakhóra, on the Aspis of Argos, and recently identified on the south edge of the akropolis at Selinous present stepped grandstand-like facilities indicating that crowds assembled to view events, presumably sacrificial, enacted in the area in front of the temple (Perakhóra; Becker 2003: 221–24; Argos: Vollgraff 1956: 43–49; Becker 2003: 42–47; for Selinous, see Chapter 18 in this volume). The close proximity of people (and for that matter, animals en route to sacrifice) intensified shared experience. The converging of a community, even a contingent community, creates social bonds and solidarity, personal and group affirmation. Each festival was a performance, with variables of timing, movement, and spatial arrangement. Recurrent reenactment created cultural memory, promoting group identity among participants, all associated with the sensory experience of place. 7 Conclusions Establishing context—topography, space, volume— provides more complete understanding of how worshippers experienced the built environment. Buildings gain authority from how they are perceived, from their intrinsic traits such as material, form, and size, as well as relational traits such as placement, restriction, and access. John Barrett commented, “context …provides a mechanism that moves us from observation to interpretation” (Barrett 2006: 194). By expanding the study of Greek architecture to include terrain, places, and structures that may not be buildings (such as paved paths and steps), and by expanding our investigations to consider a more dynamic, mobile view of ancient behaviour, we can see better the function and meaning of ancient buildings and sites. Technology supports such an extended scope of architectural research. The study of Greek architecture has been closely tied to empirical approaches and multiple measurements—of blocks, profiles, details of construction—as scholars have focused on reconstructing ancient buildings, either physically or on paper (see Vouzaxakis 2006 and Clark 2010 on reconstructions of many kinds). Even M. Korres’ meticulous rebuilding of the Parthenon is a kind of model. Evans’ reconstituted palatial court complex at Knossos is an extreme example of restoration as interpretation. GIS expands this tradition with its capacity to supply precise measurements, not only of individual structures and the built environment, but also the natural environment and human interventions such as roads or political boundaries. The relational database of GIS offers opportunities to combine, correlate, and compare information on a range

Hollinshead

of scales so as to depict an archaeological site—or area, or region—with a rich, comprehensive expression of its physical setting. Hewing to a Cartesian notion of space, GIS is not designed to represent human behaviour and movement. Including time and motion to create an interpretive narrative, VR can incorporate the visual results generated using GIS into video models that simulate human perception and mobility in antiquity, the activities that constitute place making. Located in a larger, GIS-generated landscape (space), human figures in VR move towards, around, in and out of architectural structures, demonstrating how people experience place. Their perceptions of and interactions with individual buildings result from site-specific archaeological information, derived directly, such as from steps and paths, or indirectly, from the juxtaposition of spaces and structures. GIS and VR encourage us to look beyond buildings. Contextual analysis of ancient Greek architecture can use modern technological tools together with current theoretical constructs to interpret the experiences, social practices and even the values expressed by archaeological remains of the built environment in Greek antiquity. The end result not only puts Greek architecture in context, but also brings awareness of our own context as we study Greek architecture. List of References Adey, P., 2010: Mobility (London). Argyriou, L. & Papadopoulos, N.S., 2015: “Mixed Reality Applications, Innovative 3d Reconstruction Techniques and GIS Data Integration for Cultural Heritage”, in Sarris, A. (ed.), Best Practices of GeoInformatic Technologies for the Mapping of Archaeolandscapes (Oxford) 149–58. Ashmore, W. & Knapp, A.B. (edd.), 1999: Archaeologies of Landscape: Contemporary Perspectives (Malden). Barceló, J.A., 2000: “Visualizing What Might Be: An Introduction to Virtual Reality Techniques in Archaeology”, in Barceló, J.A., Forte, M. & Sanders, D.H. (edd.), Virtual Reality in Archaeology (Oxford) 9–35. Barrett, J.C., 2006: “Archaeology as the Investigation of the Contexts of Humanity”, in Papaconstantinou, D. (ed.), Deconstructing Context: A Critical Approach to Archaeological Practice (Oxford) 194–211. Beaudry, M.C. & Parno, T.G. (edd.), 2013: “Mobilities in Contemporary and Historical Archaeology”, in Beaudry, M.C. & Parno, T.G. (edd.), Archaeologies of Mobility and Movement (New York) 1–14. Becker, T., 2003: Griechische Stufenanlagen: Untersuchungen zur Architektur, Entwicklungsgeschichte, Funktion und Rapräsen­ tation (Munster).

CONTEXTS FOR GREEK ARCHITECTURE: PLACES AND PEOPLE Bender, B., 1993: “Introduction: Landscape—Meaning and Action”, in Bender, B. (ed.), Landscape: Politics and Perspectives (Providence) 1–18. Bookidis, N. & Stroud, R.S., 1997: The Sanctuary of Demeter and Kore. Topography and Architecture. Corinth 18.3 (Princeton). Brück, J., 2005: “Experiencing the Past? The development of a phenomenological archaeology in British prehistory”, Archae­ ological Dialogues 12: 45–72: http://journals.cambridge .org/ARD. Casey, E.S., 1996: “How to get from Space to Place in a Fairly Short Stretch of Time: Phenomenological Prolegomena”, in Feld, S. & Basso, K. (edd.), Senses of Place (Seattle) 13–52. Casey, E.S., 2001: “Body, Self, and Landscape: A Geophilosophical Inquiry into the Place-World”, in Adams, P.C., Hoelscher, S. & Till, K.E. (edd.), Textures of Place: Exploring Humanist Geographies (Minneapolis) 403–25. Chadwick, A.M., 2004: Stories from the Landscape: Archaeologies of Inhabitation (Oxford). Chrysanthi, A., Flores, P.M. & Papadopoulos, C. (edd.), 2012: “Introduction. Archaeological Computing: Towards Prothesis or Amputation?”, in Chrysanthi, A., Flores, P.M. & Papadopoulos, C. (edd.), Thinking Beyond the Tool: Archaeo­ logical Computing and the Interpretive Process. BAR-IS 2344 (Oxford) 7–13. Clark, J.T., 2010: “The Fallacy of Reconstruction”, in Forte, M. (ed.), Cyber-Archaeology. BAR-IS 2177 (Oxford) 63–73. Connelly, J.B., 2011: “Ritual Movement through Greek Sacred Space: Towards an Archaeology of Performance”, in Chaniotis, A. (ed.), Ritual Dynamics in the Ancient Mediterranean. Agency, Emotion, Gender, Representation (Stuttgart) 313–46. Conolly, J. & Lake, M., 2001: Geographical Information Systems in Archaeology (Cambridge). Cresswell, T., 2010: “Towards a Politics of Mobility”, Society and Space 28: 17–31. Cripps, P., 2008: “Spatial Technologies in Archaeology in the Twenty-first Century”, in Greengrass, M. & Hughes, L. (edd.), The Virtual Representation of the Past (Burlington) 147–56. Déderix, S., 2015: “More Than Line of Sight and Least Cost Path. An Application of GIS to the Study of the Circular Tombs of South-Central Crete”, in Sarris, A. (ed.), Best Practices of GeoInformatic Technologies for the Mapping of Archaeolandscapes (Oxford) 137–47. Favro, D., 1996: The Urban Image of Augustan Rome (New York). Favro, D., 2006: “In the Eyes of the Beholder: Virtual Reality Re-creations and Academia”, in Haselberger, L. & Humphrey, J. (edd.), Imaging Ancient Rome: Documentation— Visualization—Imagination. JRA Suppl. 61 (Portsmouth) 321–34. Favro, D., 2013: “To Be or Not To Be in Past Spaces: Thoughts on Roman Immersive Reconstructions”, in Bonde, S. & Houston,

255 S. (edd.), Re-Presenting the Past. Archaeology through Text and Image (Oxford) 151–203. Fowler, H.N. & Stillwell, R., 1932: Corinth 1: Introduction, Topography, Architecture (Cambridge, MA). Frieman, C. & Gillings, M., 2007: “Seeing is Perceiving?”, WorldArch 39.1: 4–16. Gibson, E., 2007: “The Archaeology of Movement in a Mediterranean Landscape”, JMA 20.1: 61–87. Gillings, M. & Goodrick, G.T., 1996: “Sensuous and reflexive GIS: exploring visualisation and VRML”, Intarch 1: 1.1–3.3. Gillings, M., 2005: “The Real, the Virtually Real, and the Hyperreal: The Role of VR in Archaeology”, in Smiles, S. & Moser, S. (edd.), Envisioning the Past. Archaeology and the Image (Oxford) 223–39. Goette, H.-R., 2001: Athens, Attica and the Megarid: An Archaeological Guide. Rev. English ed. (London). Goodrick, G. & Gillings, M., 2000: “Constructs, Simulations and Hyperreal Worlds: The Role of Virtual Reality (VR) in Archaeological Research”, in Lock, G.R. & Brown, K. (edd.), On the Theory and Practice of Archaeological Computing. University of Oxford Committee for Committee for Archaeology Monograph 51 (Oxford) 41–58. Gregory, I., 2008: “Using Geographical Information Systems to Explore Space and Time in the Humanities”, in Greengrass, M. & Hughes, L. (edd.), The Virtual Representation of the Past (Aldershot) 135–46. Hamilakis, Y., 2002: “The Past as Oral History: Towards an Archaeology of the Senses”, in Hamilakis, Pluciennik, and Tarlow 2002: 121–36. Hamilakis, Y., Pluciennik, M., & Tarlow, S. (edd.), 2002. Thinking through the Body: archaeologies of corporeality (New York). Haselberger, L. & Humphrey, J., 2006: Imaging Ancient Rome: Documentation—Visualization—Imagination. JRA Suppl. 61 (Portsmouth). Herzog, I., 2013: “The Potential and Limits of Optimal Path Analysis”, in Bevan, A. & Lake, M. (edd.), Computational Approaches to Archaeological Spaces (Walnut Creek) 179–212. Hollinshead, M.B., 2015: Shaping Ceremony: Monumental Steps and Greek Architecture (Madison). Ingold, T., 1993: “The Temporality of the Landscape”, WorldArch 25.2: 152–74. Ingold, T., 2011: “Against Space. Place, Movement, Knowledge”, in Ingold, T. (ed.), Being Alive (Oxford) 145–55. Kavoulaki, A., 1999: “Processional Performance and the Democratic Polis”, in Goldhill, S. & Osborne, R. (edd.), Performance Culture and Athenian Democracy (Cambridge) 293–320. Lake, M. & Ortega, D., 2013: “Computer-Intensive GIS Visibility Analysis of the Settings of Prehistoric Stone Circles”, in Bevan, A. & Lake, M. (edd.), Computational Approaches to Archaeological Spaces (Walnut Creek) 213–41.

256 Lake, M. & Woodman, P., 2003. “Visibility Studies in Archaeology: a Review and Case Study”, Environment and Planning B: Planning and Design 30: 689–707. Lake, M., 2007: “Viewing Space”, WorldArch 39.1: 1–3. Laurence, R. & Newsome, D.J., 2011: Rome, Ostia, Pompeii (Oxford). Llobera, M., 2003: “Extending GIS-based Visual Analysis: The Concept of Visualscapes”, International Journal of Geographi­ cal Information Science 17: 25–48. Llobera, M., 2007a: “Modeling Visibility through Vegetation”, International Journal of Geographical Information Science 21.7: 799–810. Llobera, M., 2007b: “Reconstructing Visual Landscapes”, WorldArch 39.1: 51–69. Lock, G. & Molyneaux, B.L. (edd.), 2006: Confronting Scale in Archaeology. Issues of Theory and Practice (New York). Lock, G. (ed.), 2000: Beyond the Map: Archaeology and Spatial Technology (Amsterdam). Lock, G., & Harris, T., 2000: “Introduction: Return to Ravello”, in Lock, G. (ed.), Beyond the Map: Archaeology and Spatial Technology (Amsterdam) xiii–xxiv. Lorimer, H., 2010: “Walking: New Forms and Spaces for Studies of Pedestrianism”, in Cresswell, T. & Merriman, P. (edd.), Geographies of Mobilities: Practices, Spaces, Subjects (Farnham) 19–33. Marble, D.F., 1990: “The Potential Methodological Impact of Geographic Information Systems on the Social Sciences”, in Allen, K.M.S., Green, S.W. & Zubrow, E.B.W. (edd.), Interpreting Space: GIS and Archaeology (London) 9–21. Mylonopoulos, I., 2006: “Greek Sanctuaries as Places of Communication through Rituals: An Archaeological Perspective”, in Stavrianopoulou, E. (ed.), Ritual and Communication in the Graeco-Roman World. Kernos Suppl. 16 (Liège) 69–110. Kernos Suppl. 16. Neils, J. (ed.), 1996: Worshipping Athena. Panathenaia and Parthenon (Madison). Newsome, D.J., 2011: “Introduction: Making Movement Meaningful”, in Laurence, R. & Newsome, D.J. (edd.), Rome, Ostia, Pompeii (Oxford) 1–54. Orlandos, A.K., 1967–1968: Ἡ Ἀρκαδικὴ Ἀλιφείρα καὶ τὰ μνημεῖα τῆς (Athens). Östenberg, I., Malmberg, S. & Bjørnebye, J., 2015: The Moving City. Processions, Passages, and Promenades in Ancient Rome (London). Ott, T. & Swiaczny, F., 2001: Time-Integrative Geographic Informa­ tion Systems. Management and Analysis of Spatio-Temporal Data (New York). Paliou, E., 2011: “Three-dimensional Visibility Analysis of Architectural Spaces: Iconography and Visibility of the Wall Paintings of Xeste 3 (Late Bronze Age Akrotiri)”, JAS 38.2: 375–86.

Hollinshead Paliou, E., 2013: “Reconsidering Concepts of Visualscape: Recent Advances in Three-Dimensional Visibility Analysis”, in Bevan, A. & Lake, M. (edd.), Computational Approaches to Archaeological Spaces (Walnut Creek) 243–63. Paliou, E., 2014a: “Introduction”, in Paliou, E., Lieberwirth, U. & Polla, S. (edd.), Spatial Analysis and Social Spaces. Interdisciplinary Approaches to the Interpretation of Prehistoric and Historic Built Environments. Topoi 18 (Berlin) 1–17. Paliou, E., 2014b: “Visibility Analysis in 3D Built Spaces: A New Dimension to the Understanding of Social Space”, in Paliou, E., Lieberwirth, U. & Polla, S. (edd.), Spatial Analysis and Social Spaces: Interdisciplinary Approaches to the Interpretation of Prehistoric and Historic Built Environments. Topoi 18 (Berlin) 91–113. Pfaff, C.A., 2003: The Argive Heraion 1: The Architecture of the Classical Temple of Hera (Princeton). Pred, A., 1984: “Place as Historically Contingent Process: Structuration and the Time-geography of Becoming Place”, Annals of the Association of American Geographers 74.2: 279–97. Preucel, R.W. & Meskell, L. (2004): “Places”, in Meskell, L. & Preucel, R.W. (edd.), A Companion to Social Archaeology (Malden) 215–29. Sanders, D.H., 2014: “Virtual Heritage. Researching and Visualizing the Past in 3D”, Journal of Eastern Mediterranean Archaeology and Heritage Studies 2.1: 30–47. Sanders, G.D.R., Palinkas, J., Tzonou-Herbst, I. & Herbst, J., 2017: Ancient Corinth: Site Guide. 7th ed. (Grand Forks). Shear, J.L., 2001: Polis and Panathenaia: The History and Development of Athena’s Festival, (diss., University of Pennsylvania). Sheller, M. & Urry, J., 2006: “The New Mobilities Paradigm”, Environment and Planning A 38: 207–26. Smiles, S. & Moser, S., 2005: “Introduction: The Image in Question”, in Smiles, S. & Moser, S. (edd.), Envisioning the Past: Archaeology and the Image (Oxford) 1–12. Snead, J.E., Erickson, C.L. & Darling, J.A., 2009: “Making Human Space: The Archaeology of Trails, Paths, and Roads”, in Snead, J.E., Erickson, C.L. & Darling, J.A. (edd.), Landscapes of Movement. Trails, Paths, and Roads in Anthropological Perspective (Philadelphia) 1–19. Stavrianopoulou, E., 2013: “The Archaeology of Processions”, in Raja, R. & Rüpke, J. (edd.), A Companion to the Archaeology of Religion in the Ancient World (Chichester) 349–61. Thomas, J., 1993: “The Politics of Vision and the Archaeologies of Landscape”, in Bender, B. (ed.), Landscape: Politics and Perspectives (Providence) 19–48. Tilley, C., 1994: A Phenomenology of Landscape. Places, Paths, and Monuments (Oxford).

CONTEXTS FOR GREEK ARCHITECTURE: PLACES AND PEOPLE Tilley, C., 2004: The Materiality of Stone: Explorations in Landscape Phenomenology (Oxford). Tokmakidis, K., Kalyvioti, M.-E. & Nanakou, P., 2004: “Geographic Information System Applied in Archaeological Site (WSA3.2)” in FIG Working Week 2004 in Athens. Workshop—Archaeological Surveys: https://www.fig.net/ resources/proceedings/fig_proceedings/athens/papers/ wsa3/WSA3_2_Tokmakidis_et_al.pdf Tomlin, C.D., 2013: GIS and Cartographic Modeling (Redlands). Urry, J., 2007: Mobilities (Oxford). Vollgraff, C.W., 1956: Le Sanctuaire d’Apollon Pythéen à Argos (Paris). Vouzaxakis, K., 2006: “Reconstructions: Materialized Narratives”, in Papaconstantinou, D. (ed.), Deconstructing Context: A Critical Approach to Archaeological Practice (Oxford) 176–93. Wescoat, B.D. & Ousterhout, R.G. (edd.), 2012: Architecture of the Sacred: Space, Ritual, and Experience from Classical Greece to Byzantium (Cambridge). Wheatley, D. & Gillings, M., 2000: “Vision, Perception, and GIS: Developing Enriched Approaches to the Study of Archaeological Visibility”, in Lock, G. (ed.), Beyond the Map: Archaeology and Spatial Technology (Amsterdam) 1–27.

257 Wheatley, D. & Gillings, M., 2002: Spatial Technology and Archaeology: the Archaeological Applications of GIS (London). Wheatley, D., 2004: “Making space for an archaeology of place”, IntArch 15: https://doi.org/10.11141/ia.15.10. Wheatley, D., 2014: “Connecting Landscapes with Built Environments: Visibility Analysis, Scale, and the Senses”, in Paliou, E., Lieberwirth, U. & Polla, S. (edd.), Spatial Analysis and Social Space: Interdisciplinary Approaches to the Interpretation of Prehistoric and Historic Built Environments. Topoi 18 (Berlin) 115–34. Williams II, C.K. & Fisher, J.E., 1971: “Corinth 1970: Forum Area”, Hesperia 40.1: 1–51. Witcher, R.E, 1999: “GIS and Landscapes of Perception”, in Gillings, M., Mattingly, D.J. & van Dalen, J. (edd.), Geographical Information Systems and Landscape Archaeology. The Archaeology of Mediterranean Landscapes 3 (Oxford) 13–22. Zedeño, M.N., 2000: “On What People Make of Places: A Behavioral Cartography”, in Schiffer, M.B. (ed.), Social Theory in Archaeology (Salt Lake City) 97–111.

chapter 16

The House of the Rhyta at Pseira: 3D Crowdsourcing in an Online Virtual Environment Miriam G. Clinton with Ansel MacLaughlin 1 Introduction In the last few years, 3D modelling in archaeology has attained a new and extraordinarily popular status.1 Whereas in 2014, when this project began, the use of photogrammetry was limited in Minoan archaeology, now there is scarcely an excavation in Greece that does not employ some form of photogrammetry to generate 3D models of its architecture and stratigraphy. While widespread and uncritical acceptance of any technological advance will, of course, bring its own problems, archaeologists should not be too wary of utilising a technology that has already led to substantial progress in other fields, including medicine, education, history, sports, social work, and military applications (Champion 2006a: 13; 2011: 1–2; López, Cuenca & Cáceres 2010: 1336). In fact, archaeologists should not be afraid to use the capabilities of 3D modelling even more than they have already done, as long as they do so with proper training and theoretical underpinning. Too often, models are relegated to the traditional role of illustration and even presented as static images, so that the full capabilities of the third dimension for aiding human understanding of space and even social context are rarely exploited (Bentkowska-Kafel 2012: 258; Champion 2011: 2–3; Favro 2012: 276; Forte 2008: 23; Forte & Pietroni 2009: 65; Frischer 2008: vi–viii; Gill 2009: 317; Goodrick & Gillings 2000: 42–43). 3D modelling should be an integral part of hypothesis creation, evidence gathering, and argumentation. Bernard Frischer has called this function digital heuristics, meaning the use of models as a form of direct experimentation (Frischer 2008: xiii). In the case of archi1  I would like to thank Rhodes College, the American School of Classical Studies at Athens, the Institute for Aegean Prehistory, the University of Pennsylvania Museum of Archaeology and Anthropology, and the National Endowment for the Humanities for their support of my research. This work would not have been possible without the mentorship of Philip P. Betancourt and the efforts of my student research assistants Kevin Ennis, Ansel MacLaughlin, and Kathryn Boehm. As always, my appreciation goes out to my husband Charles Umiker. I am especially grateful to Ansel MacLaughlin for his attention to detail in building the model, which pushed me to question my own previously disregarded assumptions about the ancient architecture.

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004416659_018

tecture, 3D state models can allow multiple architects to study structures, mine the data, and critique each other’s theories with a significant reduction in fieldwork, and at a substantial savings in time, money, and political capital in obtaining permits. Reconstructed 3D models can make the architects’ theories manifest in a format that is both more visual and interactive, as well as clearer than text alone. Thus, digital heuristics in architecture create a new level of scholarly accessibility that has never before been enjoyed. This paper addresses the Minoan case study of the House of the Rhyta at Pseira, Krete, and discusses one application of 3D modelling as both a visualisation and a scientific tool. A reconstructed model of the House of the Rhyta provides an easily understandable visualisation of one theory concerning the original construction and use of an enigmatic building. The model can also, however, be used to test additional theories about the structure, how people interact with it, and more generally the use of space in Minoan architecture. Specifically, a reconstructed 3D model of the House of the Rhyta will be used as the basis for an online interactive process—a game—that will provide the basic data for a crowdsourced research study on access and circulation patterns. 2

The House of the Rhyta at Pseira

The House of the Rhyta, or Building AF North (Fig. 16.1), is found on Pseira, a small island in the Gulf of Mirabello just over 3 km from the northeastern coast of Krete (Betancourt 2001; 2009; Betancourt, Banou & Floyd 1997; Betancourt & Davaras 1993; Dierckx 1995; Floyd 1995; 1997; Floyd et al. 1995; Pariente 1992). In the Minoan period, the settlement was an important seaport. The island was occupied from the Neolithic through the Byzantine periods (Betancourt 2009: 3). The Minoan town of Pseira was sited on the Katsoúni peninsula on the southeastern coast of the island, facing Krete. Block AF, at the southern tip of the peninsula, was excavated in 1990–1991 by a team from Temple University under the direction of Philip P. Betancourt and Costis Davaras (Betancourt 2009: xix–xx).

THE HOUSE OF THE RHYTA AT PSEIRA: 3D CROWDSOURCING

figure 16.1

State plan after original Block AF excavation after Betancourt 2009, 7, ill. 2.1; 15, ill. 2.5

259

260 The block includes two overlapping structures of different dates: AF South, the earlier building, and AF North, the later one. Two rooms, AF 5A and 5B, were used in both phases (Betancourt 2009: 3–4). The buildings of Block AF are especially important, not only because they preserve one of the longest occupation sequences on the island, but also because in at least one phase, called the House of the Rhyta (AF North), they combined both cult and domestic activities (Betancourt 2009: xvii). AF North, also known as the House of the Rhyta, dates to LM IB (ca. 1500–1450 BCE) and was both a cult building and a domestic structure (Betancourt 2009: 170). It falls into McEnroe Type 2b (McEnroe 1982: 7–9). This combination of domestic and ritual functions is unusual on Pseira, which had few ritual structures (Betancourt & Davaras 1998: 1–77; Davaras 2001: 86–88; Floyd 1998: 208–9). For this reason, the building has the potential to inform conclusions about Pseira as a whole, about the architecture of mixed use in something other than the most elite Minoan contexts, and about Minoan ritual—particularly domestic ritual, which despite years of study remains enigmatic. The House of the Rhyta included six rooms in three zones: an upper terrace, a lower terrace, and a set of southern rooms (I use here the conventions for directions in Betancourt 2009, so “north” conventionally refers to approximately 315°). As preserved, no doorways connect the ground floor zones at all. Rooms AF 8 and 9, which together constitute the upper terrace, were at the northwest. The lower terrace, at the northeastern corner of the building, included AF 6 and 7. The southern rooms, the third zone of the building, were AF 5A and 5B. All three zones were on different terrace levels from one another, with the structure sloping sharply downwards towards the south and less steeply towards the east. Finally, the collapsed remains in rooms AF 5A, 5B, 6, 8, and 9 prove that an upper floor existed, although no architecture from the upper story remains (Betancourt 2009: 160; 168–69). Despite the unusual lack of circulation between the ground floor zones, the House of the Rhyta nonetheless can be determined to have functioned as a single structure based on the finds, which include similar object types and even joining sherds among the various rooms of both the upper and lower floors (Betancourt 2009: 166). The upper story appears to have been more elaborate, while the domestic functions seem to have been concentrated on the lower floor, a pattern that is familiar, for example, from the houses of Akrotíri (Palyvou 2005: 51). The cult space was located in the upper story over AF 5A, 5B, 6, 8, and 9. The ritual finds include pithoi with Linear A inscriptions, rhyta, bull-shaped vessels, a marble chalice, a triton shell, and wall plaster (Betancourt 2009: 168–70). The ground

Clinton with MacLaughlin

floor included primarily storage, food preparation, and serving pottery as well as loom weights (Betancourt 2009: 166–68). 3

The Minoan Modelling Project at Pseira

In 2014, members of the Minoan Modelling Project restudied the architecture in the field with dGPS, GIS, photogrammetry (Koch & Kaehler 2010: 2–3; Lo Brutto & Meli 2012; Olson et al. 2013: 247–60; Pollefeys et al. 2000; Santagati, Inzerillo & Di Paola 2013; Wulf, Sedlazeck & Koch 2012), and traditional drawing and measurement. The first step was to produce a state model (Fig. 16.2) from the photogrammetry on the site—a high resolution model built from more than 840 photos. Measurement was accomplished with a Topcon Hiper Lite Green Label Differential GPS (dGPS), with the resulting measurements imported into ESRI ArcMap for analysis. This photogrammetric state model is the basis of a new hypothetical reconstruction of the building.2 This reconstructed model (Figs. 16.3–16.4) was built directly on the basis of the state model in SketchUp, importing both the ArcGIS-generated block plan as a JPG and the state model in the OBJ format. The reconstructed walls can then be positioned directly over the photogrammetric state model, using the ground plan as necessary for clarity and adjusting the floor heights based on excavation records. In SketchUp, only a decimated version of the state model could be imported due to the software’s relatively limited capabilities. The use of the decimated state model reduces the final accuracy of the reconstructed model. SketchUp’s texturing and rendering capacity, too, is insufficient to create a fully photorealistic virtual reality reconstruction. For this reason, the preliminary version was reconstructed a second time on the basis of the SketchUp original, this time using AutoDesk Maya—a more robust software package that also allows a full scale version of the state model to be imported as an FBX file. The initial reconstruction in SketchUp, however, is sufficient as a proof of concept and to test the online tracking capabilities in Unity (see Section 5).

2  The theory behind the reconstruction was first published in Clinton 2016. The live model can be viewed at https://umacesdweb2.campus .ads.umass.edu/Unity/index.html. As far as the photogrammetry on site, the photos were shot using a 36.3 megapixel Nikon D800 with a 50 mm fixed focal length lens. The photographs were taken at an average distance of 3 m with approximately 53% overlap and processed in Agisoft PhotoScan Professional Edition. The theoretical accuracy of each photograph is 0.3 mm.

THE HOUSE OF THE RHYTA AT PSEIRA: 3D CROWDSOURCING

figure 16.2

3D state model of the House of the Rhyta from south model and image by M.G. Clinton

261

262

figure 16.3

Clinton with MacLaughlin

3D reconstruction in SketchUp of the House of the Rhyta: (a) section of the ground floor from north; (b) section of the upper floor from northwest; (c) 3D reconstruction from northeast model and images by A. MacLaughlin and M.G. Clinton

THE HOUSE OF THE RHYTA AT PSEIRA: 3D CROWDSOURCING

figure 16.4

3D reconstruction in Autodesk Maya of the House of the Rhyta, viewed from the (a) south and (b) northeast model and images by B. Wang, D. Floyd, and M.G. Clinton

Because this research project is still in a preliminary stage, some images in this article are neither textured nor rendered to imitate materials and simulate photorealism. In some situations, publishing 3D models that avoid the appearance of perfect realism can be preferable and function better for research purposes. In other words,

263 photorealism is not always necessary; many suggest that it becomes a distraction from the real purpose of some models (Bentkowska-Kafel 2012: 251–52; Champion 2011: 3; Eiteljorg 1998; Favro 2012: 273–74; Goodrick & Gillings 2000: 45–46; Johanson 2009: 413–14; Mosaker 2001: 21). In this case, the purpose of this article is to discuss the capabilities of the 3D gaming technology, rather than to prove a hypothesis concerning the materials from which the House of the Rhyta was constructed. For this reason, an untextured version is displayed as Fig. 16.3 and used in discussions of the architectural reconstruction, while Fig. 16.4 represents a textured and illuminated version of the model as an example of its final appearance. The preliminary version of the reconstructed model includes both the ground floor and the upper story of the House of the Rhyta, but it omits surrounding structures that are not excavated or for which remains no longer exist—even though such structures shared walls with the House of the Rhyta (Betancourt 2009: 17–18). The reconstructed model shows architecture that can no longer be seen in the archaeological record, but its presence can be deduced from contextual clues in the existing architecture and based on comparanda, especially the houses of Akrotíri, Thera. These houses provide the best comparisons possible due to their preservation, but also because many of them share distinct similarities with the House of the Rhyta, including the stepped or terraced nature of the ground floor (Palyvou 2005: 46). Some are also mixed-use domestic and ritual structures, such as Xeste 3 (Palyvou 2005: 57–61), and most of them appear to be private structures, rather than official administrative structures, like the so-called “villas” (Palyvou 2005: 45–46). The hypothetical reconstruction suggests that there were three access points between the ground floor and the upper story of the House of the Rhyta: two stairways in AF 7/8 (Fig. 16.3a, bottom right) and AF 6/9 (Fig. 16.3a, centre), and a ladder in AF 5A (Fig. 16.3a, top centre) (Clinton 2016: 73–76). The three zones were not inter-accessible on the ground floor, except that the first flight of the staircase in AF 6/9 may have provided indirect access between the two northern terraces. There were also three entrances to the structure, one of which only accessed room AF 5B (Fig. 16.3a, top). A second entrance led into AF 6, one of the service areas of the house, with access to AF 7 and likely to a staircase that led up to the room above AF 8/9. The more public entrance was through AF 8, marked by an L-shaped bench; the ordinary path for a ritual participant coming in that door would be to proceed up the staircase in a bent axis pattern to reach the cult space above AF 6 (Fig. 16.3b, centre). The multitude of access points and staircases offered varying degrees of participation

264 in the ritual, depending on which room one was allowed to enter. A participant could be stopped in AF 8 before ascending the staircase, stopped again at the head of the stairs before entering the cult space above AF 6, shunted onto the landing above AF 9, or allowed to enter the room above AF 6. The stairways in AF 6/9 were more private, used to access the service areas. They did not provide access to the cult space directly. These conclusions about relative privacy of the rooms in the House of the Rhyta arise from a qualitative typology of access and circulation patterns and focus particularly on privacy from visitors, rather than from other inhabitants (Clinton 2013: 44–63). The House of the Rhyta appears to have been designed primarily around a matrix pattern of circulation (Clinton 2013: 51–52), in which each of the three terraces forms its own matrix with internal circulation patterns. Within each of the two northern matrices, the circulation on the ground floor is a simple room-toroom (Clinton 2013: 53–54) pattern with bent axis (Clinton 2013: 55–56). The two southern rooms, AF 5A and AF 5B, are different, as a single doorless room (Clinton 2013: 225) and a room with exterior circulation (Clinton 2013: 54–55), respectively (Fig. 16.3a, top). In fact, as a doorless space accessed only by ladder from the upper story, AF 5A is the most private room in the structure. Thus, AF 5B would be considered a public room, the most public in the structure, followed first by AF 8 and then AF 6. Though access and circulation alone would mean that AF 8 and AF 6 were equally public, the bench marks AF 8 as a formal entrance and therefore increases its public status. The more extreme bent axis pattern between AF 6 and AF 7 (Fig. 16.3a, bottom left) than between AF 8 and AF 9 (Fig. 16.3a, bottom right), similarly, marks the matrix of AF 8/9 as more public. The upper floor, too, should be considered more private than the lower floors because of the extreme bent axis and room-to-room patterns required to reach them, with the room above AF 5A (Fig. 16.3b, right) being the most private on the upper story, followed by the room above AF 6 (Fig. 16.3b, centre). Since the room above AF 6 was the cult space, and since AF 5A was likely a storage room for the cult materials, this pattern of increasingly private spaces associated both with ritual and storage of valuable materials accords well with the pattern seen in other Minoan structures. Thus, both the access and circulation pattern typology and the finds from the structure indicate a function as a building with mixed ritual and domestic functions but without substantial architectural distinction, important as a specialised structure within its community but relatively small and not elaborate when compared to buildings with similar functions in other

Clinton with MacLaughlin

Minoan towns. In other words, it is an excellent case study to test the access and circulation pattern typology. 4

Digital Heuristics and 3D Modelling

The above hypotheses regarding the private and public nature of the spaces in the House of the Rhyta are based on already established theories about how people interact with architecture, for example that certain features, like the bench, will catch a person’s attention and signal a public or even a ritual space (Davaras 2001: 84–86; McEnroe 2001: 53; Poursat 1992: 31; Rutkowski 1986: 152, cat. VI). The wide and visible staircase in AF 8, for example, might be more attractive to people entering the doorway than the narrow and offset door leading down a step to AF 9, because the Minoans routinely manipulated sight lines and used visual boundaries like steps to demarcate zones within structures, both in domestic and ritual spaces (Betancourt 2007: 81–82; Letesson 2009: 358–59; Marinatos & Hägg 1986: 72; Preziosi 1983: 159–60; Preziosi & Hitchcock 1999: 77; Sanders 1990: 59–61, 65). Though these arguments have been debated and refined over the years by architectural scholars in the world of archaeology, they are qualitative theories that are not susceptible to direct quantitative proof (see, e.g., Chermayeff & Alexander 1965; Floyd 1998: 193–99; Kent 1990a; 1990b; Palyvou 1987; 2004; Sanders 1990; Wallace-Hadrill 1988; 1994). For this reason, archaeologists have sometimes turned to graphical analysis, also known as Gamma Analysis, in an attempt to find quantitative support for their qualitative theories about the use of space (Blanton 1994; Hanson 1998; Hillier 1996; Hillier & Hanson 1984; Letesson 2009). Gamma Analysis, however, developed as a method of interpreting modern architecture and may not be appropriate for every ancient context. Even when it is, the method discards a great deal of information about the ancient structure, including the plan, in an attempt to provide quantitative data (Clinton 2013: 34–35). In short, though graphical analysis can provide quantitative data, it cannot offer direct proof. A 3D model, however, can be used as a tool for digital heuristics to gain something closer to direct quantitative proof for architectural theories, because it can provide direct data of how humans interact with the structure itself. Digital heuristics is a proven approach for the use of 3D modelling and has been used effectively in Roman archaeology (Favro 2012: 276; Frischer 2008: xiii; Frischer & Fillwalk 2013: 342; Goodrick & Gillings 2000: 52; Johanson 2009: 406–07; McCarty 2004: 255; Snyder 2012: 396–97). One of the most common applications is to verify the

THE HOUSE OF THE RHYTA AT PSEIRA: 3D CROWDSOURCING

265

interpretation of archaeological architecture, but it is also used to test complicated hypotheses about social questions that rely in part upon interaction with architecture, such as how public were early gladiatorial contests in the Roman forum, how to assemble wall painting fragments, the purposes of structures at the Villa of Hadrianus at Tivoli, the relationship between the Ara Pacis and the sun’s progression, and—in the Minoan world—the public nature of the frescoes from the West House at Akrotíri, Thera and the degree of participation in rituals in the lustral basin in Xeste 3 at Akrotíri, Thera (Clarke 2014a; 2014b; 2015; Clarke & Muntasser 2014; Favro 2012: 273; Forte & Siliotti 1997: 12–13; Frischer 2014; Frischer & Fillwalk 2012; 2013; Johanson 2009; Paliou 2011a; 2011b; RomeLab: http:// hvwc.etc.ucla.edu/projects). Such an application is one way to create and test hypotheses using a 3D model as a scientific tool, especially in architectural studies. In the case of the House of the Rhyta at Pseira, digital heuristics have already been employed in several ways. For example, the hypothesis behind the reconstruction had already been published without creating a model to test it (Clinton 2016), but building the model allowed the hypothesis to be refined substantially (Favro 2012: 273). Modelling the whole of the structure—not just what had previously been seen as the most problematic—at a high level of detail forced the team to reexamine its assumptions and question the reasoning behind every reconstruction decision. Thus, the very process of creating the model led the team to think carefully about every aspect of the building. For example, even though there was a suggested reconstruction for the ground floors before the model was begun, such details as the rise of the staircases and even more basic necessities of reconstruction like the height of the doorways had not previously been taken into account. Designing the reconstruction in plan does not require anyone to think of the full three dimensionality of a structure. Not surprisingly, therefore, creating a reconstructed model of any sort, even a drawing, is a time-honoured tool of the architectural trade from at least the Renaissance onwards, creating a feedback loop that leads to the final product being more complete (Alberti 1988: 33–34; 313). In fact, building a reconstruction that as a model meets the standards of the digital modelling community in archaeology, as laid out in the London Charter, while also recording and publishing the modelling procedures themselves requires greater attention to detail and recording than any 2D reconstruction would demand (Champion 2017; Denard 2013; Koller, Frischer & Humphreys 2009: 7, 10–11; Snyder 2013: 2; London Charter: http://www.london charter.org/). Both metadata (the technical details of the

modelling) and paradata (records of the choices made in creating the reconstruction) must be recorded and published (Bentkowska-Kafel 2012; Boeykens & Bogdani 2008). In and of itself, that reconstruction becomes an important way both to build and test a hypothesis. Another commonly acknowledged way of using 3D models as evidence in scholarly arguments is to present the model itself—or rather, stills of the model—as illustration of a hypothesis about the architecture, just as any other illustration, such as a drawing, might be used (Barceló 2007: 452–53; e.g., Betancourt 2012: fig. 1c; Clinton 2016: figs. 1, 5, 7–8; Forte & Siliotti 1997: 7; Marlowe 2006: figs. 15–16; McEnroe 2010: fig. 2.10; Palyvou 2005: pl. 2–4). The still images of the model in this article can be used in that way, and they will be in publications focused primarily on the reconstruction itself. The model illustrates an argument concerning the most likely reconstruction of the House of the Rhyta. In such a case, as with the House of the Rhyta, the model is built to match the reconstruction hypothesis, using the research and comparanda cited in the accompanying argument; the text of the publications constructs the argument for those reconstruction choices and provides sources to support each decision. In other words, the model becomes an illustration of an argument. A new way of using models is also gaining popularity, where permits allow: digital surrogacy. As a “digital surrogate”, the state model stands in for the original in scholarly research, especially when—like Pseira—the site is difficult to study in person (Cameron 2007: 55; Gilliland 2008: 16; Knell 2013: 443). In other words, a digital version of the structure could become a surrogate for visiting the site, allowing scholars to study the remains independently. One reason digital open-access publications are becoming increasingly popular is that they open up new opportunities for scholars to improve scholarly critique by accessing an archaeological site’s raw data directly, rather than relying on the information that the author selects for presentation (Barceló 2007: 452; Hazan 2007: 136; Kansa & Kansa 2013: 90–92; Molloy 2011). In the case of a project including 3D modelling, that raw data will include the state model (Fig. 16.2) in addition to the reconstruction—allowing for easy comparison and for other scholars both to interrogate the conclusions about the structure and to form their own hypotheses. The goal is to make the reconstructed model itself an object of scholarly critique, to spur debate about reconstruction choices in this house and about Minoan structures in general. Since a reconstruction is necessarily an interpretation of the existing remains, the best way to accomplish that debate is to provide the raw data, the state model, alongside the reconstructed model.

266 Whether one is publishing a state model, a reconstructed model, or both, a live model with which people can interact encourages debate better than a static 2D view or series of views of this model (Kuzminsky & Gardiner 2012: 2749; Lerma et al. 2010: 506; Niven et al. 2009: 2022). Static views represent an author’s choice of images to support the arguments, while publishing a live 3D model takes away authorial control of the building’s representation, thereby leading to enhanced debate and critique. In other words, it becomes far less possible to cherry-pick certain images that best support the author’s hypotheses. Others may find aspects of the structure that disprove the author’s theories. Some may find a portion of the reconstruction that was simply overlooked by the author and can be improved. Others may use the model to hypothesise about completely different topics. In any case, like publishing a catalog or a database, publishing a state model puts raw data into the hands of other scholars, and publishing a reconstructed model opens all aspects of the hypothesis up for critique. Both allow for better and more informed debate. Open access publication allows the understanding of architecture to evolve with new information and new interpretations, rather than remaining fixed and stagnant. In addition to these intuitive functions of modelling that many architects currently use or could use, the House of the Rhyta reconstruction is also set to function in a very different way as a scientific tool. The heuristics above focus primarily on the physical properties of the structure, such as the materials, measurements, and room arrangement. It is possible, however, to test hypotheses not just on physical but also on social questions about human interaction with and within space. Although in this particular study the House of the Rhyta will be used to test questions of access, circulation, and privacy, using 3D models heuristically could potentially provide new data on a whole range of architectural questions—about phenomenology, access and circulation patterns, identity, and many other social questions, some of which may not yet even be in the scholarly debate. In the case of the House of the Rhyta, online gaming is the key to making the model useful for testing access and circulation pattern theories. Once the model is perfected, it will be released online as an educational video game that will be suitable for classroom use, but hopefully it will also garner a wider audience. Cultural heritage games, like virtual reality and 3D models in cultural heritage more generally, have most frequently been used as learning tools and as public outreach by museums or other sites (Champion 2003: 273; Mortara et al. 2014; Neto et al. 2011). The House of the Rhyta game will, in fact, be made freely accessible as an online learning tool, creating

Clinton with MacLaughlin

new opportunities for everyone to experience the world of the Minoans in new ways. It will serve as an interactive portal to a site that is not generally accessible. However, outreach is not the primary goal of the model and game; rather it is an integrated part of the scientific research that is the model’s real function. The questions of how people moved through ancient structures have, in the past, not been susceptible to much independent verification. Unless clear use-wear patterns on the floors or lower portions of the walls can be established, all that theorists have to help them answer questions, like where people went first and most frequently in a structure, is a logical interpretation of the plan. Methods such as Gamma Analysis may be a step forward in creating a quantitative methodology for the study of access and circulation patterns, but still none of the results can be directly tested on ancient structures. In other words, there is no opportunity to observe people moving around within the intact ancient structures to verify the theories. In the end, the validity of a method is simply decided by whether the logic of the interpretation convinces the reader. Of course, substantiating evidence for theories is always sought in the finds, but, in the Minoan world, structures and rooms were always multifunctional, so that artifacts provide conflicting evidence for how people were likely to have used each room, let alone how they moved through it. Thus, the reconstructive and interactive capabilities of 3D modelling provide an important new opportunity to test how real people interact with the ancient architecture. A reconstruction in a virtual world allows people to be immersed in a structure that appears to be complete, with walls that reach above their heads, realistic floors, windows, doors, and even ceilings and a roof. It is as close as one can come to reliving the original building without physically rebuilding it. Therefore, those people will act and move more similarly to how they would in a real structure, because they will be constrained by the architecture in ways that ruins simply cannot accomplish. In other words, it is impossible to step over a ruined wall in a reconstruction, but tourists do just that every day at archaeological sites. Therefore, their access and circulation patterns through a reconstructed model are more likely to be similar to those within the original structure. 3D modelling facilitates increasingly realistic interaction in ways that simply were impossible in the past, which makes a reconstructed model an essential component of testing access and circulation hypotheses empirically. For example, I suggest, based on my assessment of access and circulation patterns, that the lower floor of the House of the Rhyta is more public than the upper and that AF 7 is less

THE HOUSE OF THE RHYTA AT PSEIRA: 3D CROWDSOURCING

267

public than AF 9 (see Section 3). If I am correct, more people should enter the lower floor of the model than they do the upper, which is only logical given that the entrances are on the ground floor. On the other hand, my hypothesis concerning the relative privacy of AF 7 and 9 on the lower floor is less certain. Only data on how real people move through the structure can test that hypothesis.

movements are tracked to provide data about their access and circulation patterns. To allow for this new crowdsourced research study, the reconstructed model has been imported into Unity 3D, an online gaming engine often used by game designers to build and test new games but increasingly coming into use for cultural heritage and other academic virtual reality applications (Merlo, Dalcò & Fantini 2012: 628; Mortara et al. 2014: 323; Neto et al. 2011). The model can be built in any modelling software, and no custom code is required to import it into Unity or to allow users to interact with the environment (some custom code can be used to facilitate certain types of interactions, such as climbing ladders, but even that is readily available via a simple Internet search). In fact, even the recording of access and circulation pattern data is readily available. Built into the Unity engine is a suite of analytics, including a position tracker that in essence generates a tabular record of each user’s trail (Fig. 16.5a). By creating custom “events”, also a built-in feature of the analytics, it is also possible to record which room each user enters, how long they spend in that room, and which room they next move into (Fig. 16.5b). A similar functionality has been incorporated into a video game available for download that premiered in October 2016, with a focus on the sea voyages of the Argonauts in relation to the cult of the Great Gods at Samothrake (Blakely 2015; https://scholarblogs.emory.edu/samothraciannetworks/). This functionality, however, relies on custom code built by the research team and is, therefore, not as widely available as the tracker in Unity 3D. For those looking for something ready-made, the analytics functionality is already embedded in the software because of its main purpose, game development; it is important for video game developers to know what portions of their games are popular and how much time players spend in their games so that they can maximise their revenue. The Unity software, however, tracks according to an x-y-z position relative to the start point. In order to track movement according to room in addition to the x-y-z data, it is necessary to build a new invisible but uniquely numbered object to represent each room in Unity; those objects can be the basis for the custom events, the transitions between rooms. These hidden object identifiers allow the model to appear as close to the original Minoan structure as the ingenuity of digital reconstructors can make it. In other words, no thresholds, doorways, or other demarcations of space except those visible in the archaeological record need be added to the model to indicate to the computer which room is which. The room transitions, as well as how long each player spent in each room, can be exported to a spreadsheet. The x-y-z grid data is also

5

Crowdsourcing the House of the Rhyta

In order for such empirical data about access and circulation patterns to rise beyond the anecdotal level, enough people need to interact with the structure to allow for statistical analysis. The use of the House of the Rhyta model as a primary statistical research tool relies on the widespread popularity of virtual worlds, and especially online virtual worlds. Even ten years ago, tens of millions of people played online games set in virtual worlds built based on the past, present, or future (Messinger, Stroulia & Lyons 2008: 2). The number has only grown since then, into the hundreds of millions (Faiola et al. 2013: 1115; Spence 2008: 2). This level of use, in combination with the immersive potential of 3D reconstruction, allows a huge online gaming community to become, essentially, participants in a research study, similar to one that might be conducted in other social science disciplines, such as psychology or sociology. In the past, archaeologists have not had access to such experimentation, but photorealistic reconstructions have changed the landscape of research. In fact, the popularity of virtual worlds even opens the possibility of so-called “crowdsourcing”, a “web-based business model that harnesses the creative solutions of a distributed network of individuals through what amounts to an open call” (Brabham 2008: 76). Although the term originated as a business model (Howe 2006a), it has since been applied to a number of academic pursuits, from transcribing and editing the collected works of Jeremy Bentham to solving computational biology problems, and to aerial image analysis in archaeology (Causer & Terras 2014; Causer, Tonwa & Wallace 2012; Causer & Wallace 2012; Lakhani et al. 2013; Lin et al. 2014). Thus, crowdsourcing now more generally signifies a reliance on a large number of unaffiliated agents to resolve a problem, such as how people interact with architecture. The open call to a large network of people is the sine qua non (Howe 2006b). In this case, the crowd is not called upon to generate content, edit, or perform any deliberate task; rather, the agents become the data. Online users in a virtual environment can interact with and circulate through the various rooms in the reconstruction of this complex structure, while their

268

Clinton with MacLaughlin

figure 16.5 Position tracking data from Unity 3D Analytics: (a) X-Y-Z position tracking data; (b) “Event” tracking data recording transitions between rooms IMAGE BY M.G. CLINTON

available, and it too could be used for analysis. Each visitor is assigned a number, and each number is individually tracked through the spaces. This data produces a linear and temporal version of access and circulation patterns which could be mapped onto the access and circulation pattern typology to check its results. Unity can even automatically generate heatmaps, maps aggregating the tracking information such that more popular routes show up in “warmer” colours (yellow-red) and less in cooler (bluegreen). This ease of data collection means that the online game has the potential to provide big data susceptible to statistical analysis. In addition, the data could easily be used to test additional theories on access and circulation, such as Gamma Analysis, which is based on a linear map of a structure in the order in which each room is entered, like that generated by the tracker. The crowdsourced tracking information is the basic data that can be used to verify theories concerning the Minoan arrangement of access and circulation patterns to create more public and more private spaces. Though the game is not yet live, a preliminary version of the model (without the texture and illumination) is already available on a basic website. A limited amount of tracking data has been collected from casual hits and from fellow scholars who have been invited to visit the site. So far, approximately

40 people have visited the model for varying lengths of time; some of those visits include the research team’s tests to check that the model is working. Those few hits have, so far, shown mixed results in the analytics data. For example, though a knowledge of Minoan architecture leads one to conclude that most Minoans would prefer to enter at AF 8, due to the presence of the bench, approximately half of visitors so far have chosen instead to enter through what should have been the service entrance, AF 6 (above, Figs. 16.1, 16.3). Very few have chosen to continue down the street AF 1 or entered AF 5B, no doubt in part because the stark white and heavily shadowed street is not visually appealing (the final version should avoid that problem). Once they enter the structure at AF 6, however, the users who continued to explore the structure tend to follow the predicted patterns of circulation (see Section 3), continuing up the stairs to the space above AF 9 and on to the space above AF 6. Nonetheless, it is clear that the stark white and unlit model does not inspire people to explore the entire structure: on average visitors entered only two rooms and stayed less than three minutes in the house—a less than ideal result, when the goal is to learn how visitors move through the entire structure of eleven rooms. While these initial hits are not enough to provide the quantitative data the project requires, they have served

THE HOUSE OF THE RHYTA AT PSEIRA: 3D CROWDSOURCING

269

their function of indicating how the model can better facilitate access and circulation pattern research. They reveal some of the reasons that a game, rather than a simple interactive model, is necessary (as well as, of course, the importance of texturing, lighting, and rendering in making a model inviting). In today’s climate of constant stimulation and multitasking online, casual visitors to the website will likely expect to be entertained—rather than being there for the pure joy of learning about an ancient building. Online consumers have frequently been shown to have a short attention span, and people tend to use the web primarily for goal-directed, rather than purely experiential purposes (Koufaris 2002: 209; Novak, Hoffman & Yung 2000: 15). The best way to encourage users to spend time on a website is to require them to pay more focused attention (Novak, Hoffman & Yung 2000: 22). A game with task-based interaction is more engaging than a simple model, however well-constructed the model may be. Visitors will be more inspired to spend time within the structure and to explore the entire house if they have a reason to do so, such as a goal within a game (Champion 2006a: 90–98; 2011: 83–89; Tan & Ramahan 2009: 147–48). In the game of the House of the Rhyta, it is important to choose tasks that do not precondition users’ movement through the space. Visitors cannot be blocked from or guided to rooms by the tasks that they perform. For this reason and after testing 30 online games of different types, a team of student researchers has concluded that the tasks replicating the Minoan use of the space will also require interaction with replicas of the artifacts found in the rooms, which would ideally balance the need for stimulating tasks with the requirements of the research into access and circulation patterns. The game will include multiple levels. In the first level, the tasks will be presented but optional, and visitors will be told that they are being given the chance to explore the house in preparation for higher levels of the game. In additional levels, visitors will be tasked to find every room in the house within a certain timeframe and tasked to find a certain object. Some levels will start the visitors at different access points to the House of the Rhyta, such as outside AF 5B, so that the data is not skewed towards one of the house’s entrances. The more difficult hurdle is the issue of cultural immersion. While a digital model can reproduce existing or reconstructed remains, without some sort of annotations it cannot easily recreate the contextual experience of living as a Minoan (Champion 2006a: 67–69; 2011: 69–72; Tan & Ramahan 2009: 146–47). This flaw is inherent in any modern model of ancient objects, regardless of its format, as well as, of course, any modern interpretation of

archaeological data. This limitation will never be entirely overcome. However, a game can build in some degree of that ancient knowledge through its format with the goal of making the modern visitor behave more closely, even if never identically, to an ancient visitor (Champion 2004; 2006a: 103–5; Tan & Ramahan 2009: 152–53). One way is through explanatory texts, first preparing the gamer before he/she even enters the virtual space—using “hints” at the beginning of each level for the House of the Rhyta— and later informing him/her of the significance when he/she performs certain actions, such as approaching a bench, via optional “more information” panels in the House of the Rhyta. The other is through visual or aural clues that attempt to recreate some of the context, such as lighting or street sounds (Champion 2006a: 80–88; 2011: 203; Champion & Dave 2007: 335–36). Both will be used in the House of the Rhyta. In this way, the interactive experience of a game is more suited to creating a contextualised user than a model alone (Aldrich 2004; Champion 2003; 2006a: 23–32; 45–46; 58; 69–79; 2006b: 52–53; 2011: 1–2; 10; 37–38; 50–51; 70–74; Champion & Dave 2007: 338–42; Laird 2001; Manninen 2004; Prensky 2001; Schroeder 1996). Even if that were not so, however, the data generated by tracking users through the model would still be valuable, because currently there is no way of testing how anyone, ancient or modern, interacts with now-destroyed ancient architecture. The initial results are not discouraging, even though they do not precisely match the predictive theories on access and circulation patterns discussed in Section 3. Likewise, 3D modelling technology itself initially disappointed the creators of early models that failed to match a perfect photorealistic virtual reality, but those early models were still valuable and produced useful information (Champion 2011: 17; Goodrick & Gillings 2000: 47–48; Snyder 2012: 418). So, too, the techniques of social research through 3D modelling and gaming are still in development, but, even in an initial stage, they are yielding important and useful data against which theories can be tested. Though the technique is necessarily imperfect, it can only improve as new and more sophisticated models and games are developed. Already, the initial tests of the House of the Rhyta have led the research team to improve both the model and the game. No doubt future researchers will take game development even further. 6 Conclusion The model of the House of the Rhyta is just one attempt to use 3D modelling in a new way that goes beyond

270

Clinton with MacLaughlin

illustration and reconstruction. Starting with the possibility of making the raw data of archaeological sites more accessible to experts and amateurs alike, it also highlights the possibility of using 3D models for direct experimentation. Specifically, a new online game provides basic position tracking to test the hypothesis of a dialogic social relationship between the structure and its inhabitants. While position tracking may not be the answer to every question about ancient architecture, it is important for archaeologists to take full advantage of every possible source of data to test their arguments. 3D models represent a powerful new tool for open access data that is still in its infancy but has the potential to become a mainstay of the field once the tool is refined, especially because of the capabilities for building and testing social hypotheses. There is no reason to limit the application of 3D modelling to solely illustrative applications or even to purely physical architectural questions like reconstruction. 3D modelling can and should be an integral part of testing any architectural theory, as long as the experimenters are guided by valid parameters and the visualisation is based on sound data and modelling practices. The enigmas of the House of the Rhyta at Pseira are fruitful ground for just such a project. List of References Alberti, L.B., 1988: On the Art of Building in Ten Books. Translated by J. Rykwert, N. Leach, and R. Tavernor (Cambridge, MA). Aldrich, C., 2004: Simulations and the Future of Learning: an Innovative (and Perhaps Revolutionary) Approach to E-learning (San Francisco). Barceló, J.A., 2007: “Automatic Archaeology: Bridging the Gap between Virtual Archaeologuy, Artificial Intelligence, and Archaeology”, in Cameron, F. & Kenderdine, S. (edd.), Theorizing Digital Cultural Heritage: A Critical Discourse (Cambridge, MA) 437–56. Bentkowska-Kafel, A., 2012: “Processual Scholia: The Importance of Paradata in Heritage Visualization”, in Bentkowska-Kafel, A., Denard, H. & Baker, D. (edd.), Paradata and Transparency in Virtual Heritage (Farnham, UK) 245–59. Betancourt, P.P, 2001: “The Household Shrine in the House of the Rhyta at Pseira”, in Laffineur, R. & Hägg, R. (edd.), POTNIA: Deities and Religion in the Aegean Bronze Age. Aegaeum 22 (Liège) 145–49. Betancourt, P.P, 2007: Introduction to Aegean Art (Philadelphia). Betancourt, P.P, 2009: Pseira X: The Excavation of Block AF. Prehistory Monographs 28 (Philadelphia). Betancourt, P.P, 2012: “The Architecture of the House Tombs at Petras”, in Tsipopoulou, M. (ed.), Petras, Siteia: 25 Years of Excavations and Studies (Aarhus) 107–16.

Betancourt, P.P. & Davaras, C. (edd.), 1998: Pseira II: Building AC (the “Shrine”) and Other Buildings in Area A. University Museum Monograph 94 (Philadelphia). Betancourt, P.P. & Davaras, C., 1993: “Excavations at Pseira, Crete, in 1991–92”, AJA 97.2: 301–02. Betancourt, P.P., Banou, E.S. & Floyd, C.R., 1997: “Provincial LM III at Pseira, Crete”, in Hallager, E. & Hallager, B. (edd.), Late Minoan III Pottery: Chronology and Terminology. Acts of a Meeting Held at the Danish Institute at Athens, August 12–14, 1994 (Aarhus) 57–75. Blakely, S., 2015: “Human Geography, GIS Technology, and Ancient Mysteries: A Case Study from the Island of Samothrace”, Getty Research Journal 7: 133–41. Blanton, R.E., 1994: Houses and Households: A Comparative Study (New York). Boeykens, S., & Bogani, E., 2008: “Metadata for 3D Models: How to Search in 3D Model Repositories?”, in International Association of Technology, Education and Development (ed.), ICERI 2008. Proceedings: International Conference of Education, Research and Innovation: Madrid 17th–19th of November, 2008, 11 (Madrid) 17–19. Brabham, D.C., 2008: “Crowdsourcing as a Model for Problem Solving: An Introduction and Cases”, Convergence 14.1: 75–90. Cameron, F., 2007: “Beyond the Cult of the Replicant: Museums and Historical Digital Objects—Traditional Concerns, New Discourses”, in Cameron, F. & Kenderdine, S. (edd.), Theorizing Digital Cultural Heritage: A Critical Discourse (Cambridge, MA) 49–75. Causer, T. & Terras, M., 2014: “Crowdsourcing Bentham: Beyond the Traditional Boundaries of Academic History”, International Journal of Humanities and Arts Computing 8.1: 46–64. Causer, T. & Wallace, V., 2012: “Building a Volunteer Community: Results and Findings from Transcribe Bentham”, Digital Humanities Quarterly 6.2: http://www.digitalhumanities.org/ dhq/vol/6/2/000125/000125.html. Causer, T., Tonra, J. & Wallace, V., 2012: “Transcription Maximized; Expense Minimized? Crowdsourcing and Editing the Collected Works of Jeremy Bentham”, Literary and Linguistic Computing 27.2: 119–37. Champion, E.M. & Dave, B., 2007: “Dialing Up the Past”, in Cameron, F. & Kenderdine, S. (edd.), Theorizing Digital Cultural Heritage: A Critical Discourse (Cambridge, MA) 333–48. Champion, E.M., 2003: “Applying Game Design Theory to Virtual Heritage Environments”, in Spencer, S. (ed.), Graphite2003, Proceedings of the 1st International Conference on Computer Graphics and Interactive Techniques in Australasia and South East Asia (Melbourne) 273–74. Champion, E.M., 2004: “Heritage Role Playing: History as an Interactive Digital Game”, in Pisan, Y. (ed.), IE2004 Australian Workshop on Interactive Entertainment (Sydney) 41–48.

THE HOUSE OF THE RHYTA AT PSEIRA: 3D CROWDSOURCING

271

Champion, E.M., 2006a: Evaluating Cultural Learning in Virtual Environments (diss., University of Melbourne). Champion, E.M., 2006b: “Playing With A Career in Ruins: Game Design and Virtual Heritage”, Treballs d’Arqueologia 12: 45–61. Champion, E.M., 2011: Playing with the Past (London). Champion, E.M., 2017: “The Role of 3D Models in Virtual Heritage Infrastructures”, in Benardou, A., Champion, E., Dallas, C. & Hughes, L. (edd.), Cultural Heritage Digital Tools and Infrastructures (London) 15–35. Chermayeff, S. & Alexander, C., 1965: Community and Privacy: Toward a New Architecture of Humanism (Garden City, NY). Clarke, J.R, 2014A: “The Villa of Oplontis: A ‘born digital’ project”, in Delve, J. & Anderson, D. (edd.), Preserving Complex Digital Objects (London) 259–72. Clarke, J.R, 2014b: “A Virtual Villa”, Apollo 179.617: 48–53. Clarke, J.R, 2015: “Retrieving the Decorative Program of Villa A (‘of Poppaea’) at Oplontis (Torre Annunziata, Italy): How Orphaned Fragments Find a Home in the Virtual-Reality 3D Model”, in Lepinski, S. & McFadden, S. (edd.), Beyond Iconography: Materials, Methods, and Meaning in Ancient Surface Decoration (Boston) 97–108. Clarke, J.R. & Muntasser, N.K., 2014: Oplontis. Villa A (“of Poppaea”) at Torre Annunziata, Italy 1: The Ancient Setting and Modern Rediscovery (New York). Clinton, M.G., 2013: Access and Circulation Pattern Analysis in Neopalatial Architecture on Crete: A Methodology for Identifying Private Spaces (diss., University of Pennsylvania). Clinton, M.G., 2016: “The Power of the Future: Re-analyzing the House of the Rhyta at Pseira”, in Chalikias, K., Beeler, M., Pearce, A. & Renette, S. (edd.), The Future of the Past: From Amphipolis to Mosul (Boston) 69–79. Davaras, C., 2001: “Comments on the Plateia Building”, in Betancourt, P.P. & Davaras, C. (edd.), Pseira V: The Architecture of Pseira (Philadelphia) 79–88. Denard, H., 2013: “Implementing Best Practice in Cultural Heritage Visualisation: The London Charter”, in Corsi, C. (ed.), Good Practice in Archaeological Diagnostics (Cham) 255–68. Diercx, H.M.C., 1995: “Aspects of Economy and Technology at the Late Bronze Age Site of Pseira, Crete. Based on the Stone Tool Assemblage”, Ex Scientia 1: 57–72. Eiteljorg II, H., 1998: “Photorealistic Visualizations May Be Too Good”, CSA Newsletter 11.2: http://www.csanet.org/news letter/fall98/nlf9804.html. Faiola, A., Newlon, C., Pfaff, M. & Smyslova, O., 2013: “Correlating the Effects of Flow and Telepresence in Virtual Worlds: Enhancing Our Understanding of User Behavior in Game-Based Learning”, Computers in Human Behavior 29.3: 1113–21. Favro, D., 2012: “Se non è vero, è ben trovato (If Not True, It is Well Conceived): Digital Immersive Reconstructions of Historical

Environments”, JSAH 71.3 (Special Issue on Architectural Representations): 273–77. Floyd, C.R., 1995: “Fragments from Two Pithoi with Linear A Inscriptions from Pseira”, Kadmos 34: 39–48. Floyd, C.R., 1997: “LM IIIC Pottery on Pseira: Pirates, Traders, or Fishermen?”, AJA 101.2: 364. Floyd, C.R., 1998: Pseira III: The Plateia Building, edited by P.P. Betancourt and C. Davaras. University Museum Monograph 102 (Philadelphia). Floyd, C.R., Onyshkevych, L., Beaudoin, J. & Mitrakis, G., 1995: “Excavations at Pseira, 1993–1994”, AJA 99.2: 313. Forte, M. & Pietroni, E., 2009: “3D Collaborative Environments in Archaeology: Experiencing the Reconstruction of the Past”, International Journal of Architectural Computing 7.1: 57–76. Forte, M. & Siliotti, A. (edd.), 1997: Virtual Archaeology: Re-Creating Ancient Worlds (New York). Forte, M., 2008: “Virtual Archaeology: Communication in 3D and Ecological Thinking”, in Frischer, B. & Dakouri-Hild, A. (edd.), Beyond Illustration: 2D and 3D Digital Technology as Tools for Discovery in Archaeology. BAR-IS 1805 (Oxford) 20–34. Frischer, B. & Fillwalk, J., 2012: “The Digital Hadrian’s Villa Project: Using Virtual Worlds to Control Suspected Solar Alignments”, in Guidi, G. (ed.), Proceedings of 2012: 18th International Conference on Virtual Systems and Multimedia (VSMM 2012) (Piscataway) 49–56. Frischer, B. & Fillwalk, J., 2013: “A Computer Simulation to Test the Buchner Thesis: The Relationship of the Ara Pacis and the Meridian in the Campus Martius, Rome”, in Institute of Electrical and Electronics Engineers (ed.), Proceedings of the 2013 Digital Heritage International Conference (DigitalHeritage) vol. 1 (Piscataway) 341–45. Frischer, B., 2008: “From Digital Illustration to Digital Heuristics”, in Frischer, B. & Dakouri-Hild, A. (edd.), Beyond Illustration: 2D and 3D Digital Technology as Tools for Discovery in Archaeology. BAR–IS 1805 (Oxford) v–xxiv. Frischer, B., 2014: “New Digital Simulation Studies on the Obelisk, Meridian, and Ara Pacis of Augustus”, in Haselberger, L. (ed.), The Horologium of Augustus: Debate and Context. JRA Suppl. 99 (Portsmouth) 77–90. Gill, A.A., 2009: “Digitizing the Past: Charting New Courses in the Modeling of Virtual Landscapes”, in Flaten, A.R. & Gill, A.A. (edd.), Digital Crossroads: New Directions in 3D Architectural Modeling in the Humanities. Visual Resources: An International Journal of Documentation 25.4: 313–32. Gilliland, A.J., 2008: “Setting the stage”, in Introduction to Metadata 3.0 (Los Angeles): http://www.getty.edu/research/ publications/electronic_publications/intrometadata/ setting.pdf. Goodrick, G. & Gillings, M., 2000: “Constructs, Simulations and Hyperreal Worlds: The Role of Virtual Reality (VR) in Archaeological Research”, in Lock, G.R. & Brown, K. (edd.),

272 On the Theory and Practice of Archaeological Computing. University of Oxford Committee for Committee for Archaeology Monograph 51 (Oxford) 41–58. Hanson, J., 1998: Decoding Homes and Houses (Cambridge). Hazan, S., 2007: “A Crisis of Authority: New Lamps for Old”, in Cameron, F. & Kenderdine, S. (edd.), Theorizing Digital Cultural Heritage: A Critical Discourse (Cambridge, MA) 133–48. Hillier, B. & Hanson, J., 1984: The Social Logic of Space (Cambridge). Hillier, B., 1996: Space is the Machine: A Configurational Theory of Architecture (Cambridge). Howe, J., 2006a: “Crowdsourcing: A Definition”, in Crowdsourcing: Tracking the Rise of the Amateur blog, June 2: http://crowd sourcing.typepad.com/cs/2006/06/crowdsourcing_a.html. Howe, J., 2006b: “The Rise of Crowdsourcing”, Wired magazine 14.6: 1–4. Johanson, C., 2009: “Visualizing History: Modeling in the Eternal City”, Visual Resources 25.4: 403–18. Kansa, E.C. & Kansa, S.W., 2013: “We All Know That a 14 Is a Sheep: Data Publication and Professionalism in Archaeological Communication”, Journal of Eastern Mediterranean Archaeology and Heritage Studies 1.1: 88–97. Kent, S., 1990a: “A Cross-cultural Study of Segmentation, Architecture, and the Use of Space”, in Kent, S. (ed.), Domestic Architecture and the Use of Space: An Interdisciplinary Cross-cultural Study (Cambridge) 127–51. Kent, S., 1990b: “Activity Areas and Architecture: An Interdisciplinary View of the Relationship between Use of Space and Domestic Built Environments”, in Kent, S. (ed.), Domestic Architecture and the Use of Space: An Interdisciplinary Cross-cultural Study (Cambridge) 1–8. Knell, S.J., 2013: “The Shape of Things to Come: Museums in the Technological Landscape”, in Parry, R. (ed.), Museums in a Digital Age (Hoboken) 435–53. Koch, M. & Kaehler, M., 2010: “Combining 3D Laser-Scanning and Close-Range Photogrammetry—An Approach to Exploit the Strength of Both Methods”, in Frischer, B. (ed.), Making History Interactive: Computer Applications and Quantitative Methods in Archaeology (CAA). Proceedings of the 37th International Conference, Williamsburg, Virginia, United States of America, March 22–26, 2009. BAR-IS 2079 (Oxford) 1–7. Koller, D., Frischer, B. & Humphreys, G., 2009: “Research Challenges for Digital Archives of 3D Cultural Heritage Models”, ACM Journal on Computing and Cultural Heritage 2.3: 7.1–17. Koufaris, M., 2002: “Applying the Technology Acceptance Model and Flow Theory to Online Consumer Behavior”, Information Systems Research 13.2: 205–23. Kuzminsky, S.C. & Gardiner, M.S., 2012: “Three-Dimensional Laser Scanning: Potential Uses for Museum Conservation and Scientific Research”, JAS 39.8: 2744–51.

Clinton with MacLaughlin Laird, J.E., 2001: “Using Computer Game to Develop Advanced AI”, Computer 34.7: 70–75. Lakhani, K.R., Boudreau, K.J., Loh, P.-R., Backstrom, L., Baldwin, C., Lonstein, E., Lydon, M., Maccormack, A., Arnaou, R.A. & Guinan, E.C., 2013: “Prize-Based Contests Can Provide Solutions to Computational Biology Problems”, Nature Biotechnology 31.2: 108–11. Lerma, J.L., Navarro, S., Cabrelles, M. & Villaverde, V., 2010: “Terrestrial Laser Scanning and Close Range Photogrammetry for 3D Archaeological Documentation: the Upper Palaeolithic Cave of Parpalló as a Case Study”, JAS 37.3: 499–507. Letesson, Q., 2009: Du phenotype au genotype: analyse de la syntaxe spatiale en architecture minoenne (MMIIIB—MRIB) (Louvain-La-Neuve). Lin, A.Y.-M., Huynh, A., Lanckriet, G. & Barrington, L., 2014: “Crowdsourcing the Unknown: The Satellite Search for Genghis Khan”, PloS one 9.12: e114046. Lo Brutto, M. & Meli, P., 2012: “Computer Vision Tools for 3D Modelling in Archaeology”, in Ioannides, M. (ed.), Progress in Cultural Heritage Preservation: 4th International Conference, EuroMed 2012: Limassol, Cyprus, October 29–November 3, 2012 (New York) 1–6. International Journal of Heritage in the Digital Era Suppl. 1. López, J., Cuenca, M. & Martín Cáceres, M.J., 2010: “Virtual Games in Social Science Education”, Computers & Education 55.3: 1336–45. Manninen, T., 2004: Rich Interaction Model for Game and Virtual Environment Design (diss., University of Oulu). Marinatos, N. & Hägg, R., 1986: “On the Ceremonial Function of the Minoan Polythyron”, OpAth 16.6: 57–73. Marlowe, E., 2006: “Framing the Sun: The Arch of Constantine and the Roman Cityscape”, The Art Bulletin 88.2: 223–42. McCarty, W., 2004: “Modeling: A Study in Words and Meanings”, in Schreibman, S., Seimens, R. & Unsworth, J. (edd.), A Companion to Digital Humanities. Blackwell Companions to Literature and Culture 26 (Oxford) 254–70. McEnroe, J.C., 1982: “A Typology of Minoan Neopalatial Houses”, AJA 86.1: 3–19. McEnroe, J.C., 2001: Pseira V: The Architecture of Pseira. Edited by P.P. Betancourt and C. Davaras (Philadelphia). McEnroe, J.C., 2010: The Architecture of Minoan Crete: Constructing Identity in the Aegean Bronze Age (Austin). Merlo, A., Dalcò, L. & Fantini, F., 2012: “Game Engine for Cultural Heritage: New Opportunities in the Relation between Simplified models and Database”, in Guidi, G. (ed.), Proceedings of 2012: 18th International Conference on Virtual Systems and Multimedia (VSMM 2012) (Piscataway) 623–28. Messinger, P.R., Stroulia, E. & Lyons, K., 2008: “A Typology of Virtual Worlds: Historical Overview and Future Directions”, Journal of Virtual Worlds Research 1.1: 1–18.

THE HOUSE OF THE RHYTA AT PSEIRA: 3D CROWDSOURCING

273

Molloy, J.C., 2011: “The Open Knowledge Foundation: Open Data Means Better Science”, PLoS Biol 9.12: e1001195. http://dx.doi .org/10.1371/journal.pbio.1001195. Mortara, M., Catalano, C.E., Bellotti, F., Fiucci, G., HouryPanchetti, M. & Petridis, P., 2014: “Learning Cultural Heritage by Serious Games”, Journal of Cultural Heritage 15.3: 318–25. Mosaker, L., 2001: “Visualizing Historical Knowledge Using Virtual Reality Technology”, Digital Creativity 12.1: 15–25. Neto, J.N., Silva, R., Neto, J.P., Madeiras Pereira, J. & Fernandes, J., 2011: “Solis’ Curse-A Cultural Heritage Game Using Voice Interaction with a Virtual Agent”, in Institute of Electrical and Electronics Engineers (ed.), Games and Virtual Worlds for Serious Applications: VS-GAMES 2011 (Piscataway) 164–67. Niven, L., Steele, T.E., Finke, H., Gernat, T. & Hublin, J.-J., 2009: “Virtual Skeletons: Using a Structured Light Scanner to Create a 3D Faunal Comparative Collection”, JAS 36: 2018–23. Novak, T.P., Hoffman, D.L. & Yung, Y.-F., 2000: “Measuring the Customer Experience in Online Environments: A Structural Modeling Approach”, Marketing Science 19.1: 22–44. Olson, B.R., Placchetti, R.A., Quartermaine, J. & Killebrew, A.E., 2013: “The Tel Akko Total Archaeology Project (Akko, Israel): Assessing the Suitability of Multi-Scale 3D Field Recording in Archaeology”, JFA 38.3: 244–62. Paliou, E., 2011a: “The Communicative Potential of Theran Murals in Late Bronze Age Akrotiri: Applying Viewshed Analysis in 3D Townscapes”, OJA 30.3: 247–72. Paliou, E., 2011b: “Three-dimensional Visibility Analysis of Architectural Spaces: Iconography and Visibility of the Wall Paintings of Xeste 3 (Late Bronze Age Akrotiri)”, JAS 38.2: 375–86. Palyvou, C., 1987: “Circulatory Patterns in Minoan Architecture”, in Hägg, R. & Marinatos, N. (edd.), The Function of the Minoan Palaces: Proceedings of the Fourth International Symposium at the Swedish Institute in Athens, 10–16 June 1984 (Stockholm) 195–202. Palyvou, C., 2004: “Outdoor Space in Minoan Architecture: ‘Community and Privacy’”, in Cadogan, G., Hatzaki, E. & Vasilakis, A. (edd.), Knossos: Palace, City, State. Proceedings of the Conference in Herakleion Organised by the British School at Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities of Herakleion, in November 2000, for the Centenary of Sir Arthur Evans’s Excavations at Knossos. BSA Studies 12 (London) 207–17. Palyvou, C., 2005: Akrotiri, Thera: An Architecture of Affluence 3,500 Years Old. Prehistory Monographs 15 (Philadelphia). Pariente, A., 1992: “Chronique des Fouilles en 1991, Pseira”, BCH 116: 934–35.

Pollefeys, M., Koch, R., Vergauwen, M. & Van Gool, L., 2000: “Automated Reconstruction of 3D Scenes from Sequences of Images”, ISPRS Journal of Photogrammetry and Remote Sensing 55.4: 251–67. Poursat, J.-C., 1992: Guide de Malia au temps des premiers palais. Le Quartier Mu. École Francaise d’Athènes VIII (Paris). Prensky, M., 2001: Digital Game-Based Learning (New York). Preziosi, D. & Hitchcock, L.A., 1999: Aegean Art and Architecture (Oxford). Preziosi, D., 1983: Minoan Architectural Design: Formation and Signification (Berlin). Rutkowski, B., 1986: The Cult Places of the Aegean (New Haven). Sanders, D., 1990: “Behavioral Conventions and Archaeology: Methods for the Analysis of Ancient Architecture”, in Kent, S. (ed.), Domestic Architecture and the Use of Space: An Interdisciplinary Cross-cultural Study (Cambridge) 43–72. Santagati, C., Inzerillo, L. & Di Paola, F., 2013: “Image-Based Modeling Techniques for Architectural Heritage 3D Digitalization: Limits and Potentialities”, International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XL-5/W2: 555–60. Schroeder, R., 1996: Possible Worlds: The Social Dynamic of Virtual Reality Technology (London). Snyder, L.M, 2012: “Virtual Reality for Humanities Scholarship”, in Nelson, B. & Terras, M.M. (edd.), Digitizing Medieval and Early Modern Material Culture. Medieval and Renaissance Texts and Studies 426 (Tempe) 395–428. Snyder, L.M, 2013: “Software Interface for Real-Time Exploration and Educational Use of Three-Dimensional Computer Models of Historic Urban Environments (VSim)”, White Paper—National Endowment for the Humanities, September 16, 2013 https://hcommons.org/deposits/download/hc:12000/ CONTENT/hd-50958-10.pdf/. Spence, J., 2008: “Demographics of Virtual Worlds”, Journal of Virtual Worlds Research 1.2: https://doi.org/10.4101/jvwr .v1i2.360. Tan, B.-K. & Rahaman, H., 2009: “Virtual Heritage: Reality and Criticism”, in Tidafi, T. & Dorta, T. (edd.), Joining Languages Cultures and Visions: CAAD Futures 2009: Montreal, June 17– 18–19 (Montréal) 143–56. Wallace-Hadrill, A., 1988: “The Social Structure of the Roman House”, PBSR 56: 43–97. Wallace-Hadrill, A., 1994: Houses and Society in Pompeii and Herculaneum (Princeton, NJ). Wulf, R., Sedlazeck, A. & Koch, R., 2012: “3D Reconstruction of Archaeological Trenches from Photographs”, in Bock, H.G., Jäger, W., & Winckler, M.J. (edd.), Scientific Computing and Cultural Heritage (Berlin) 273–81.

chapter 17

Comparing Greek ‘Bouleuteria’ and Roman ‘Curiae’: Two Case Studies on the Parallels and Differences in the Acoustic Reconstruction and Simulation of Roman Senate Sessions and Greek Boule Meetings Christian Fron, Verena Stappmanns, Xiaoru Zhou, and Philip Leistner Council meetings were of great significance in the everyday life of ancient societies. Just how different the political practices in Greece and Roma were is reflected in the development of antithetical architectural concepts of Greek Bouleuteria and Roman Curiae. The design of the Greek Bouleuterion varies with time and location, whereas the Curiae found in Italic cities and Roman provinces generally resemble the Curia in Roma. Previous research has focused, on the one hand, on the material remains, their architectural design, and the development of the building type; and, on the other hand, on the written sources of the council meetings, and the political procedures. This essay cannot provide a full account of the research history of studies on Bouleuteria or Curiae, but rather focuses on the key publications related to the research question (the best general overviews are still Krischen 1941 and McDonald 1943; more recent studies discuss certain aspects of the buildings, such as Kockel 1995, and regarding further information about the history and development of the Curia, see: http://www.digitales-forum-romanum.de/). A new research project at the University of Stuttgart offers a new perspective and aims to unite these until now rather isolated fields of research. New technologies render possible a way of exploration that compounds the two previously used discrete approaches into one virtual, acoustic reconstruction. 1

Context of the Research

This essay provides an initial insight into a new research project at the University of Stuttgart (Germany), which raises new questions about Greek as well as Roman architecture. Since the research project is still at an early stage, the results—even though convincing—are still preliminary and will be substantiated during further research. The project is part of a multidisciplinary research group on “speeches without a microphone”, which focuses on the interdependency of spatial acoustics, architectural design and oratory from the origins of the culture of rhetoric

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004416659_019

and public debate—before microphones became common in public speeches in the early 20th century. The interdisciplinary collaboration at the University of Stuttgart includes historians and architectural historians, specialists in visualisation and in the history of technology, members of the department of building physics and acoustics from the Fraunhofer Institute at Stuttgart, and rhetoricians from the Eberhard-Karls-University of Tübingen. The main objectives are to gain a better understanding of the interrelations between speaker, audience, and orating space as well as establishing a new research method. Modern sound systems make it possible for a single speaker to be understood by a large audience. Although we take this for granted, the technology is only about one hundred years old, and ancient speakers were obliged to depend on their vocal power alone. The rhetorical and vocal education of professional speakers is well documented by written sources. Much information on rhetorical training is given by Cicero and Quintilianus as well as other rhetorical writings such Aristoteles Peri rhetorike; a fruitful introduction to ancient rhetoric is given by Worthington (2007; also note the papers by Corbeill, Dominik, and Morgan; and Schulz 2014 on vocal training and the impact of the voice in ancient rhetorics). However, only digital simulations give us insight into how these acquired skills were applied in public and political practice, and how important the vocal abilities of the orator would be. The architectural setting was of great significance for the speaker as well as for the audience, and its acoustic properties give us an indication of how much interaction was possible between the two parties. Was the gathering a mere ritual, or was the audience part of a decision-making process in which the intelligibility of the speech was crucial? In itself, ancient architecture demonstrates the effort put into creating ideal conditions for speakers and audiences. Like no other building type, Greek theatres represent the invention and development of an acoustic space, dedicated to the entertainment of large audiences. Their acoustic properties are well documented and the subject of numerous studies. Vitruvius (5.3–7) gives a detailed

275

COMPARING GREEK ‘ BOULEUTERIA ’ AND ROMAN ‘ CURIAE ’

description on the acoustic properties of Greek theatres and therefore is an important source for the acoustical knowledge of ancient architects, and some of the wellpreserved Greek and Roman theaters are used in modern times and have been the subjects of several studies on acoustics (e.g., the contribution by G. Kamourakis in Gogos 2008; or the 2011 conference, The Acoustics of the Ancient Theater, held at Patras: http://www.ancientacous tics2011.upatras.gr/). However, the influence of spatial acoustics on the genesis of the Bouleuterion remains largely unexplored. In their early days, different building types are attested, particularly the roofed theatre that emerged in Hellenistic times. A detailed comparison with Roman Curiae is still lacking, as well as a general positioning within ancient rhetoric spaces or their connection to related building types such as odeon or theatre. The long-term objective of our research project is therefore a comprehensive study of various buildings for political assemblies, not only Bouleuteria and Curiae, but also Ekklesiasteria, theatres, and Roman temples. These meeting places can be understood as the architectural testimonies of a rhetorically active society and culture. The assumption is that different architectural layouts and their specific acoustical properties reflect distinctive social requirements on the correct way to speak and the expected skills of the speaker, and they provide crucial information about the procedure of a meeting. Because public speeches at the city councils were the core of the political practice, their comprehension was essential to the council members for making decisions. What acoustic framework does the architecture provide as an orating space? To find answers to these questions, the research group developed a working process, which consists of three stages: reconstruction, simulation, and analysis. All data and acoustic samples used and analysed in this essay can be found on https://www.hi.uni-stuttgart .de/ag/forschung/rom/. 2 Reconstruction Over the past twenty years, 3D reconstructions have been established in the research of ancient architecture as a useful tool for the analysis of the visual qualities of singular monuments or even entire cityscapes. Virtual reconstructions make it possible to simulate the perspective of an ancient viewer and evaluate a city and its monuments not only from a bird’s eye view, but also from within. The research project at the University of Stuttgart adds a new dimension to this approach: the acoustic properties

of interior spaces, primarily those designated for public speaking. The multidisciplinary approach of the research project presented us with the challenge of finding buildings upon whose reconstruction and acoustic simulation all parties could agree. Written sources about oratory practice and the preservation of the archaeological remains often differ in quality. Interiors typically included ephemeral materials such as wood or fabric that are important for the acoustic properties but are seldom preserved and thus can only be reconstructed through contemporary written sources or analogies. Therefore, some aspects in the reconstructions must remain hypothetical, although giving us an opportunity to experiment with the acoustic properties by changing parameters, such as the height of the ceiling or the material of the wall paneling. The involvement of acousticians in the reconstruction process proved to be important, since they could point out which parameters are significant for the acoustic simulation or can be neglected. The detailed virtual models are a helpful tool to communicate and visualise the reconstruction within the research group as well as to the public (see below, Figs. 17.3, 17.5, 17.7). For the simulation process, architectural elements had to be simplified after the specifications of the acousticians, in order to reduce computing time. 3 Simulation The acoustic simulations have been generated at the Fraunhofer-Institute of Building Physics and Acoustics using ODEON, a program designed for acoustic simulations of modern buildings, such as opera halls (http:// www.odeon.dk/). The program allows the application of acoustic impulses into the geometry of the virtual model, simulating the expansions and reflection of these signals. Since the appropriate acoustic properties of the different building materials are assigned to the surfaces of the model, the simulations are close to reality. Moreover, different positions of speakers and listeners can be simulated as well as analysed separately and compared to each other. The acoustic setting of the program can take certain parameters into account such as a full audience, background noise, and so on. For a good speech intelligibility, two components have to be taken into account. First, the Sound Pressure Level (SPL), measured in decibel (dB), shows the volume of the orator’s speech for different positions within the audience. Second, the Speech Transmission Index (STI)

276

FRON ET AL.

indicates how clearly the acoustic signal can be understood.1 It takes into account that the sound is altered by architectural features and natural or man-made disturbances. For interior spaces an important parameter is the reverberation of a room, which is not to be confused with a simple echo. The acoustic output signal as well as these numerical values (i.e., SPL, STI) give us crucial information about the characteristics of the oratorical setting and what the audience could actually hear during a speech. For the reconstructions discussed in this paper, a full audience with every seat taken is simulated for the council halls. For future simulations the arrangement of the audience and its size will be varied, since at least senate sessions were not always attended by all its members. This must take into consideration that there were only 465 seats for 600 members in the Curia Iulia, which roughly correlates with the 437 seats for 630 members at the British House of Parliament (Taylor & Scott 1969: esp. 529–30, 532 with records of attendance in the Curia, 536–48). The Athenians, on the other hand, seemed to be more eager to guarantee space for a full capacity of 500 bouleutai, and every member had a substitute (Aiskhines 3.62). Nevertheless, at least some council members did not attend the meetings (see, e.g., Dem. 22.36). The input signals were recorded in an anechoic chamber, a room with maximal sound absorption to avoid reverberation and ensure good sound isolation (a box in a box) for keeping background noise low, which guarantees an adequate simulation of the acoustic properties of the analysed room (For example, open audio sample 1 [linked above], an extract of the Prooemium of Lysias Logoi 16 [transl. Lamb 1930]). After initial recordings from students and staff, we came to the conclusion that only an opera singer and professional speaker could come close to the vocal volume of highly trained ancient orators. Pascal Zureck, who made the recordings, is professionally skilled in both fields, having the volume capacity of an opera singer and the distinct intonation of a professional speaker. His recordings were implemented into the acoustic simulation and subsequent auralisation. 4

Analysis and Reflection

The results of the acoustic reconstructions have been reviewed by all members of the research group using the 1  A scale on the Speech Transmission Index is available at http:// www.fohonline.com/images/stories/15/02/current-foh/techfeature/ intelligibility%20scale.jpg. A small collection of different Sound Pressures is available at https://www.welt.de/print-welt/article 334313/Vom-Ticken-der-Uhr-bis-zum-Presslufthammer.html, and also see http://www.sengpielaudio.com/Rechner-pegelaenderung.htm.

aforementioned procedures. What is their validity for our knowledge of political practices and public gathering? Are these results congruent with the written sources and their reflection of the rhetorical setting? The procedures and customs of public speeches mentioned in ancient literacy texts are put into context with the acoustic space provided by the reconstructed architecture. Any contradictions between our expectations from the written and archaeological sources and our results must be analysed, potentially leading to a reconsideration of the parameters selected for the reconstruction and simulation. In other cases, literary texts may have distorted the reality, such as the great speeches of generals before battle, which in no way could be heard by every soldier. The objective cannot be the clarification of every exception or even a proof of concept for a particular reconstruction, but rather to reassess our perspective on ancient acoustic spaces through simulations. We have applied these methods to the two most prominent examples of ancient council halls: the New Bouleuterion of Athenai, and the Curia Iulia in Roma. 5 The Bouleuteria of Athenai: Sources and Reconstruction Since Athenai was the ancient centre of rhetorical culture and political debate, we chose it to be the starting point for the comparison of Greek Bouleuteria and Roman Curiae. The historic significance and the numerous written sources stand in contrast to the rather poor preservation of the archaeological remains (Rhodes 1972). Diverging interpretation by numerous scholars of the remains led to contradictory reconstructions over the past years. Miller (1995) questions Thompson’s reconstructions of the architectural remains underneath the Hellenistic Metroön as the old Bouleuterion and instead reconstructs the remains as the predecessor of the Metroön. The arguments he presents are comprehensible, but Shear (1995) contradicts him with equally plausible evidence. Without new archaeological evidence a verification or falsification of these hypotheses is impossible and therefore not the objective of our research project. Instead we want to analyse the acoustical properties of these diverging reconstructions and put them in context to the written sources. Excavations in the 1930s on the west side of the Athenian Agora uncovered a group of building structures identified as the Bouleuterion (Fig. 17.1). Homer A. Thompson published a comprehensive report in 1937, including a reconstruction of the so called the Metroön-Bouleuterion Complex (Thompson 1937: 115–217). Set on the foot of the

figure 17.1

Metroön-Bouleuterion complex on the west side of the Athenian agora, actual state after the excavations in the 1930s after Thompson 1937, pl. 6, by J. Travlos Courtesy ASCSA

COMPARING GREEK ‘ BOULEUTERIA ’ AND ROMAN ‘ CURIAE ’

277

278

figure 17.2

FRON ET AL.

Thompson’s suggestions for the reconstruction of the New Bouleuterion and the Metroön after Thompson 1937, pl. 8, by J. Travlos Courtesy ASCSA

Kolonos Agoraios at the edge of the open space of the Agora between the temple of Apollon Patros in the north and the tholos, the seat of the Prytaneion, the building complex consist of the so-called Old Bouleuterion, the Hellenistic Metroön and the New Bouleuterion. The excavations spanned the area up to the natural rock formation and uncovered building remains of several phases: Dörpfeld (1896: 107–08) started in excavations on the West side of the Agora in 1895 with the intension to uncover the Stoa Basileios, and in the 1930s H.A. Thompson excavated the whole area on the west side of the Agora. The earliest remains date as early as the seventh century BCE but are too incomplete to comment on their utilisation. Thompson dated the first substantially preserved building to the late sixth or early fifth century BCE and reconstructs a rectangular hall with interior columns

(Fig. 17.2). The front faces south, not towards the Agora square but towards the Tholos. An interior wall divides the approximately square building into a vestibule and the main council hall (Thompson 1937: 127–34). As no foundations were uncovered inside the hall which could be interpreted as tiers of stone seating, any seating would have been made from wood and therefore not arranged in a semicircle, but rather arranged along three of the four walls of the room (Kuhn 1984: 17–26). After approximately 100 years, the Old Bouleuterion was replaced with a new building directly to the west. The neighbouring building site had to be carved out from the bedrock of the Kolonos Agoraios. To reduce labour, the New Bouleuterion was erected ca. 1.50 m above the level of the predecessor and made accessible though stairs. The building features of the New Bouleuterion are even more

279

COMPARING GREEK ‘ BOULEUTERIA ’ AND ROMAN ‘ CURIAE ’

figure 17.3

Virtual model of Thompson’s reconstruction of the New Bouleuterion V. Stappmanns

poorly preserved, consisting mainly of bedding trenches and little of the corresponding masonry (Fig. 17.1). They allow a reconstruction of the general outline of a building 20 × 16 m in plan. The remaining interior structures only consist of four bedding trenches in the centre of the building and two diagonally oriented structures close to the northern wall. No ashlar blocks or architectural members can be assigned to the Bouleuterion with certainty (Thompson 1937: 140–50; Kuhn 1984: 22). Thompson’s reconstruction of the interior of the structure follows the layout of the Bouleuterion of Miletos (Knackfuß 1908), orienting the koilon to the east and leaving only a small gap between the old Bouleuterion/ Metroön and the main entrance (Figs. 17.2–17.3). In contrast the slightly better preserved predecessor, the Old Bouleuterion, is facing south towards a small square and the Prytaneion. Thompson’s reconstruction shows a closed façade with two entrances. Due to the poor preservation of the architectural remains, Thompson’s reconstruction is based largely on the comparison with other Bouleuteria, such as the Bouleuterion of Miletos, which had been published shortly before. The written sources were not taken directly into account. Of relevance is that they state that it was possible to observe events taking place in the square and vice

versa from within the Bouleuterion. They also indicate that the meetings of the Boule could be heard by people on the open square outside, for instance the famous “sausage seller” story of Aristophanes. However, since the erection of the New Bouleuterion can only roughly be dated around 415 BCE, it is uncertain whether the ancient sources refer to the Old or the New Bouleuterion. In a new reconstruction by Roux and Kuhn, the building opens to the south towards the square with the Prytaneion (Figs. 17.4–17.5). Roux (1976) started a new debate on the reconstruction of the Bouleuteria at Athenai in consideration of the written sources, and Kuhn (1984) pursued these ideas with an additional review of the archaeological remains. Their reconstruction has been generally accepted, but we still wanted to compare the two versions for the simulation since they represent the two different architectural layouts attested in other Bouleuteria. Furthermore, we wanted to see the impact of the orientation of the tiers and the façade design to help us to identify the main acoustic elements. We have already concluded that both versions have almost equal acoustic properties, as we demonstrated in our acoustic simulation. Still, the modelling and simulation process is by no means complete, and in the future we intend to consider additional parameters.

280

FRON ET AL.

figure 17.4 Kuhn’s reconstruction of the New Bouleuterion V. Stappmanns, modified from Thompson 1937, pl. 8 and Kuhn 1984, pl. XX

figure 17.5

Virtual model following Kuhn’s reconstruction of the New Bouleuterion V. Stappmanns

281

COMPARING GREEK ‘ BOULEUTERIA ’ AND ROMAN ‘ CURIAE ’

6 The Curia Iulia at the Forum Romanum: Sources and Reconstruction In contrast to the Greek Bouleuteria, the Roman Curiae never developed a distinctive building type. Inscriptions state their existence in several Italic cities as well as in the Roman provinces. Often it is impossible to identify any remains with certainty, because the Curiae mainly consist of a large hall with or without an apse. The archetypal form, however, was the Curia Iulia in the Forum Romanum. The Curia Iulia was constructed during the period of many upheavals at the gradual transition from the late Roman Republic up to the early Roman Empire, when the topography of the Forum Romanum was being transformed. Simultaneously the general layout and design of the building itself also represents a continuity with the glorious past (see, e.g., http://www.digitales-forum-romanum .de/gebaeude/curia/; and Bartoli 1963; von Gerkan 1941 for the reconstruction of the building itself; also see the critical review by Lauter [1991] of the monograph Gneisz 1990). Furthermore, the Curia remained as the regular place of assembly for the senators in Roma (Bonnefond-Coudry 1989: 47 [on the Republican era]; Talbert 1984: 113–30 [on the Imperial era]). The layout of the Curia Iulia is based on that of its predecessors—the Curia Hostilia and Curia Cornelia—except for its increased size and probably the seat distribution. But since the Church Santi Luca e Martina was erected in the seventh century CE at the presumed location of the Republican Curiae, no archaeological evidence can presently substantiate this hypothesis. The construction undertaken for the new Curia was connected to the monumental project of the Forum Iulium, only finished in 29 BCE by Caesar’s successor Augustus. In 94 and 283 CE, the building was destroyed by devastating fires and reconstructed without major alterations in the plan (Fatucci 2009: 117). The interior space of the Curia is 25.63 m in length, 17.75 m in width and 23.4 m in height (Fig. 17.6) (Bartoli 1963: 38). These measures nearly match the proportions given by Vitruvius, who as a contemporary of Augustus living in Roma surely had this particular structure in mind. Vitruvius (5.2.1) gives the following advice of the ground plan of the Curia: “if it [the Curia] be oblong, let the length and breadth be added together and let half of the total amount be given to the height under the ceiling” (sin autem oblonga fuerit, longitudo et latitude componatur, et summae compositae eius dimidia pars sub lacunaris altitudini detur, trans. Granger 1955). According to this recommendation, the height should have been 21.69 m (17.75 × 25.63 m; see the commentary in Saliou 2009: 164–68). Even though the construction of the Church of Sant’Adriano within the Curia’s walls preserved

figure 17.6

Reconstruction of the Curia Iulia V. Stappmanns, after Fatucci 2009, fig. 1

figure 17.7

Virtual model of the Curia Iulia following Fatucci 2009 V. Stappmanns

the building structure over the centuries, most of the interior features were destroyed, and for the reconstruction of wall paneling or a cornice we depend on written sources or comparisons to other buildings (Fig. 17.7). In his recommendations for the assembly hall, Vitruvius suggests: The walls, moreover, at half their height, are to have cornices run round them of wood or plaster. For if

282

FRON ET AL.

such be not provided, the voices of the disputants meeting with no check in their ascent, will not be intelligible to the audience. But when the walls are encircled round with cornices, the voice, being thereby impeded, will reach the ear before its ascent and dissipation in the air (Vitr. 5.2.2: praeterea praecingendi sunt parietes medii coronis ex intestino opere aut albario ad dimidiam partem altitudinis. quae si non erunt, vox ibi disputantium elata in altitudinem intellectui non poterit esse audientibus. cum autem coronis praecincti parietes erunt, vox ab imis, morata priusquam in aere elata dissipabitur, auribus erit intellecta). The floor with the level stairs on the longitudinal sides of the Curia and the podium opposite to the main entrance are well preserved (Bartoli 1963: 51–60). A full audience of senators, including furniture, was assumed for the simulation. 7

The Acoustic Reconstruction of Meetings of the Athenian Boule in Classical Times: The New Bouleuterion

Returning to Athenai, we tested the acoustic properties of the two different reconstructions. Both reconstructions shown here provide good acoustic properties which facilitated listening to speeches even by an incompletely trained but loud speaker (Fig. 17.8). Concerning reverberation time, both models are quite similar, although the average distance between speaker and audience is a little lower in the first reconstruction (after Thompson, Fig. 17.8b). In contrast, the second reconstruction (Roux and Kuhn) allows the inclusion of a forecourt in the acoustic analysis,

providing an opportunity to analyse whether an indirect audience, such as the citizens of Athenai which currently were not members of the Boule, could also listen to the debate from this exterior space. Hereafter, we will therefore focus on the acoustic properties of the Bouleuterion of this reconstruction (Fig. 17.8a). Based on our reconstruction and simulation model, the New Bouleuterion of Athenai provides excellent acoustics for meetings of the Boule, which is also be demonstrated by our audio samples (listen to samples 2–3 [linked above], with a professional versus a shy, untrained speaker, respectively, reading an extract of the Prooemium of Lysias Logoi 16 [transl. Lamb 1930]). These results are not surprising, since the general design of the building closely resembles the general layout of Greek theatres, which are exceptionally good orating spaces (Gogos 2008, which includes a detailed essay about the acoustic properties of the theatre itself; Vitr. 5.3.7). Furthermore, since the Boule played an important role in ancient Athenian democracy as the place for preparing and organising the large assemblies of the demos, an appropriate setting is to be expected (Aristot. Pol. 1299b30–38). The 500 members of the Boule were determined by lot and changed every year. Only a few citizens were members of the Boule more than once in their lives (Rhodes 1972: 242–43; 1980; 1981; 1984; Develin 1989). In fourthcentury BCE Athenai the Boule was generally open to every citizen over the age of 30. Although it might be the case that the proportion of aristocratic men was higher than in the Ekklesia, there must still have been a substantial number of Bouleutai who did not belong to the Athenian elite (see Ober 1989: 140–41, n. 93 on the dispute about the percentage of citizens, which were bouleutai at least once in their life). An active participation in those meetings was at least expected, if not demanded (Dem.

figure 17.8a–b Comparison between the two proposed reconstructions of the New Bouleuterion showing the SPL metric. In both cases, even an untrained speaker can easily be heard by the audience as long as he speaks loudly X. Zhou, acoustic models generated in ODEON

COMPARING GREEK ‘ BOULEUTERIA ’ AND ROMAN ‘ CURIAE ’

22.36; on speeches in front of the Bouleuterion, see Ober 1989: esp. 138–41). Moreover, every member of the Boule was very likely to once become proedros of the prytany or epistates (Ath. Pol. 44.2–3; Develin 1989: 22–23). Every member of the council had a good chance of speaking in front of his colleagues, regardless of his social status, education, or whether he was trained in public speaking or not. In some cases, even non-members of the Council could be invited to speak in front of the Boule, irrespective of the speaker’s social standing (Rhodes 1972: 3–4). Since a specialised education by one of the sophists or rhetorical trainers such as Isokrates was expensive and demanded a significant amount of spare time, most of the speakers in the council would presumably have been rhetorical amateurs (Marrou 1950, 60–138; 1957, 61–140; Papillon 2007; the Athenian education in Classical times and social gap in education is described by Pritchard 2015: esp. 114–15). As a result, the orational space needed to accommodate the voices of many individuals lacking formal training in public speaking, while making it possible for the assembly to discern every spoken word. Any disturbing background noise or other form of reverberation needed to be reduced to the minimum (it was the duty of the proedroi to establish discipline and order during the session: Ath. Pol. 44.3; cf. Aesch. 1.35; Fisher 2001, 164; Carey 2003: 36, no. 41). Additionally, the Boule was slightly set aside from the Agora and its manifold activities and noises, which would allow a good working environment. This location also provided some privacy, since the Boule was entitled to meet in secret when demanded by the circumstances (Rhodes 1972: 40–42; Thompson & Wycherley 1972: 33–34 provides some additional detail especially on the fences and barriers around the Bouleuterion). While speaking to his colleagues, the councilor stood on the bema, big enough for at least two people, and could directly face and address the whole auditorium (e.g., Antiph. 6.40; more sources on speeches in the Boule are collected in Thompson & Wycherley 1972: 33–34). Thereby he could represent the democratic ideal of isegoria (ἰσηγορία) even without any special rhetorical training—a link between the practice of isegoria and the Council of 500 which had already been stressed by A.G. Woodhead (1967). Indeed, while every citizen had the right to express his opinion in the Ekklesia, he was more or less ordered to do so as councilor (βουλευτής; on the development of an age clause, with a preference of the old before the young, see Kapparis 1998). If the meetings were not held in secret, every citizen must have been allowed to listen to the words spoken in the Boule from a safe distance. Although the account of the sausage seller in Aristophanes’ Hippeis (esp. 625–680) is a parody of one of those secret meetings during the

283 Peloponnesian War, the episode nevertheless might convey essential information about the way in which this would have been possible. P.J. Rhodes (1972: 41–42) gave two possible areas where those normal citizens (ἰδιῶται) could have listened to the debate in the council meetings: in the aisle on the south side of the chamber, or outside the Bouleuterion in an open area to the south and beyond the fence (κιγκλίς) (mentioning the east side of the New Bouleuterion based on its old reconstruction). Rhodes much preferred the first arrangement, while observing that the Bouleuterion would already have been crowded: “If the ἰδιῶται were kept beyond the κιγκλίς, they will have been outside the chamber altogether, able to hear the proceedings only with difficulty, and if the βῆμα is correctly located in the ‛orchestra’ perhaps unable to see the speakers” (Rhodes 1972: 41). Accordingly, we simulated the second case to test Rhodes’ assumption and to analyse whether or not Aristophanes’ sausage seller could have heard any words spoken by the “Paphlagonian” from the outside. Our initial results show that it was in fact possible to hear the speeches from the outside (listen to audio sample 4 [linked above], simulating what the sausage seller might have heard if the speaker spoke loudly but did not have the volume of a professional speaker). Moreover, the limited view provided through the κιγκλίς for these bystanders might have given the necessary exclusivity and spatial distance for the ordinary members of the council to keep their focus on the debate and to hinder the nonmembers of the Boule from intervening. Thanks to this new research technique, it seems highly likely that the ἰδιῶται indeed gathered in front of the Bouleuterion and need not have crowded inside the building. Our reconstruction not only casts light on whether an audience in the forecourt could listen to the debate inside the Bouleuterion, but also whether this outside audience could address the members of the Boule. The simulations in Figure 17.9 reveal that it was only possible for the public outside the building to hear speech from within if they were completely silent, and vice-versa. In Fig. 17.9a, speech clear within the Bouleuterion is difficult to discern outside when the public is also speaking even at a soft volume but is intelligible when it is silent outside (Fig. 17.9b). Similarly, in Fig. 17.9c loud speech outside the Bouleuterion is obscured by quieter discussion within but can be followed if the Boule falls silent (Fig. 17.9d). Since it can reasonably be assumed that the members of the Boule were continuously discussing the matters at hand, it would thus have been impossible for the audience standing outside to directly address and potentially disturb the members of the Boule inside of the council hall. The acoustic analysis thus provides an additional reason for

284

FRON ET AL.

figure 17.9

Model of the acoustic simulation with ODEON in the New Bouleuterion: (a) trained speaker speaking loudly (82 dBA) to the audience in front of him, with people in front of the fences outside talking to each other (45 dBA); (b) the same speaker (82 dBA) with silence outside; (c) the untrained “sausage seller” outside speaking loudly (75 dBA) towards the boule inside, which is currently debating (45 dBA); (d) the “sausage seller” (75 dBA) with a silent boule (an STI above 0.5 indicates audible speech, 0.3–0.5 is difficult to understand, and below 0.3 is unintelligible) X. Zhou

why the sausage seller first had to “forcibly” enter the interior in order to address and capture the attention of the members of the Boule. 8

Analysis of the Curia Iulia

In contrast to the New Bouleuterion of Athenai, the Curia Iulia provides poor acoustics judged by modern standards. Due to the high ceiling of the room and the marble facing on the walls, speeches in the Curia frequently suffer from a long reverberation time that hinders their intelligibility. Different problems were examined while working on the Curia Iulia. First of all, Vitruvius (5.2.2) in his discussion of the ideal forms of Curiae writes about the acoustic properties of these council halls, especially stressing the importance of cornices. Since in modern reconstructions of the interior of the Curia Iulia there are no depictions of

such cornices, we questioned in which way restored cornices could have had an impact on the acoustic properties of the Curia. Since there are no traces of any strong bonding of cornices to the inner walls, any restored cornice could not have exceeded 1 m in width. Indeed, the cornices could facilitate the listening of speeches at least for people sitting below them on the back benches. Nevertheless, the STI metric in most other areas remains between 0.3 and 0.5, a level at which it becomes difficult for the audience to listen to the orator due to high reverberation (Fig. 17.10, Tables 17.1–17.2). Another question that we raised is the optimal position for the speaker. Did a speaker step into the aisle to address the audience, and especially the emperor, or did he merely stand up from his bench to engage the audience? In the acoustic analysis, it proved to be possible for a speaker located at position P3 (red in Figs. 17.10a, c) to address everyone in the Curia and even be heard by citizens

285

COMPARING GREEK ‘ BOULEUTERIA ’ AND ROMAN ‘ CURIAE ’

figure 17.10 Model of the acoustic simulation with ODEON in the Curia Iulia: (a) speaker P3 facing the direction of Listener 2; and (b) speaker P2 facing Listener 1 (the emperor). With interior cornices, the STI is shown for: (c) speaker P3 standing up from his bench to address his colleagues; and (d) P2 standing in the aisle speaking directly to the emperor (conventions as in Fig. 17.9). X. Zhou

standing outside of the council hall, while a speaker at position P2 addressing the emperor would not be as clearly heard but still be intelligible to everyone inside the Curia (Figs. 17.10b, d). Thus it seems quite plausible that a speaker, such as Plinius the Younger, stood up from his bench and spoke from there instead of walking to the aisle in the middle of the Curia. To better understand the specific style in which speeches in the Curia had to be performed by the senators, a short digression to the order of senators and their

rhetorical professionalism is necessary. Every senator was a member of the aristocracy, belonging to the uppermost ranks of the elite in Roman imperial society, and in most cases each of them was a member of the senate until death or retirement (e.g., Plin. Ep. 4.23; on the retirement age, also see Talbert 1984: 152–54). Membership in the senatorial order remained hereditary in the Imperial era and was passed from each senator to his legitimate sons. Since the age of six, a young aristocrat would have received an excellent education as part of his training to

286

FRON ET AL.

table 17.1 SPL and STI results in the Curia Iulia where Speaker P3 addresses Listener 2 (supplemental to Fig. 17.10a,c)

Position

SPL dB (A) (82dBA) Ist / OG / TO

STI (Q: 82dBA) Ist / OG / TO

M1 M2 M3 M4 M8 M5 (1m) M6 (4m) M7 (8m) Average

59.2 / 59.3 / 59.2 59.3 / 59.1 / 59.2 60.2 / 60.1 / 59.9 61.6 / 61.6 / 61.4 58.2 / 58.3 / 58.1 — / — / 54.2 — / — / 50.9 — / — / 48.4 59.7 / 59.7 / 59.6

0.43 / 0.43 / 0.46 0.50 / 0.50 / 0.50 0.52 / 0.52 / 0.54 0.60 / 0.60 / 0.61 0.46 / 0.45 / 0.46 — / — / 0.42 — / — / 0.45 — / — / 0.51 0.50 / 0.50 / 0.51

Ist = Doors are closed, some cornices are included in the reconstruction. OG = Doors are closed, cornices omitted. TO = Doors are open, cornices included. table 17.2 SPL and STI results in the Curia Iulia where Speaker P2 addresses Listener 1 (supplemental to Fig. 17.10b,d)

Position

SPL dB (A) (82dBA) Ist / OG / TO

STI (Q: 82dBA) Ist / OG / TO

M1 M2 M3 M4 M8 M5 (1m) M6 (4m) M7 (8m) Average

61.3 / 61.6 / 61.4 64.1 / 64.0 / 64.1 58.0 / 57.9 / 57.7 60.0 / 59.7 / 59.7 57.4 / 57.4 / 57.1 — / — / 53.7 — / — / 49.9 — / — / 46.1 60.2 / 60.1 / 60

0.57 / 0.55 / 0.57 0.67 / 0.68 / 0.68 0.40 / 0.38 / 0.41 0.45 / 0.46 / 0.46 0.43 / 0.41 / 0.44 — / — / 0.39 — / — / 0.40 — / — / 0.40 0.50 / 0.50 / 0.51

Ist = Doors are closed, cornices are included in the reconstruction. OG = Doors are closed, cornices omitted. TO = Doors are open, cornices included.

become a senator (Bonner 1977; Scholz 2011; Wolff 2015; and Neuhauser 1958, who discusses the strong interrelation between the terminus orator and the senate). Already in Republican times during the tirocinium fori, the young aristocrat listened to the senators speak in the Curia and learned to adopt their style of speech (Tac. Dial. 34.1–7; Plin. Ep. 8.14.4–8). To listen to sessions as a youth before joining the Senate as one of its future members seems to have continued in modified form through the Imperial era (Plin. Ep. 8.14.8). In the early stages of his cursus honorum, the young magistrate generally kept silent and could learn

useful rhetorical techniques from established members of the council. Still, some spoke even as junior senators (Talbert 1984: 240–52). To speak in front of this auditorium was of great importance, especially for ambitious members of high rank. Not only the speakers but also the audience of senate sessions in most cases would have been trained in the art of rhetoric.2 To cope with this special acoustic setting, the speaker had to adapt his style and to significantly slow down the speed of his speech (listen to audio samples 5–6 [linked above] with speech in Latin addressing the emperor read by Andreas Hansen; the first, an extract of Cic. Leg. Man. 3, simulates a semi-professional senator speaking too quickly; the second, from the Prooemium of Plin. Pan., simulates a well-trained senator [transl. Melmoth and Hutchinson 1915]). Any deviation from these techniques would impede communication due to the special acoustics of a high ceiling and the low absorption of the side walls. A poorly trained young senator could easily have disgraced himself as a speaker. To prove himself a true speaker in the senate, the council member required a considerable amount of practice, a strong, steady voice, and some prior knowledge of the special acoustic limitations of the Curia (see Schulz 2014 for extensive research on the importance of the voice in ancient rhetoric). The setting might be an additional reason why, according to Roman rhetorical theory, a speech in the senate should be characterised by minimal ornamentation and should possess or be given a sense of weight, i.e., gravitas (Ramsey 2007: esp. 123–25; a specific view of the very different public speeches and the great impact of the inhomogeneous nature of the auditorium to them is exemplarily delivered by Leovant-Cirefice 2000). Due to the long reverberation and the deliberately slow delivery, every word was naturally provided with some gravitas. At least Loukianos (Peri tou oikou 3) states that for a good speaker, sometimes an echo would not be a 2  An instructive introduction on the importance of eloquence for imperial senators is given by Salomies (2005: 229–62), who rightly states that not all senators were excellent speakers (2005: 235–36). Nevertheless, rhetorical training was an essential part of education for every member of the imperial elite. This means, that at least all sons of senators were commonly well trained in rhetoric and were quite experienced in evaluating the quality of a speech. However, according to ancient theory on rhetoric, there was more than just education and practice to become a good speaker. A good speaker also needed a fitting ingenium/physis, or talent, as reflected in Cic. De or. 1.113–33; Fronto, Ep. 1.9.3 (ad M. Caes.); Plout. Ethika 2A; Tac. Dial. 16.1; see also Syson (2009). Moreover, this criticism should be viewed in contrast to the high ground level that surely was expected of any honourary member of the senate, especially by those who more than met these expectations and were actively literate. Finally, some of these collected examples of senators were homines novi.

COMPARING GREEK ‘ BOULEUTERIA ’ AND ROMAN ‘ CURIAE ’

hindrance, but a pillar of his speech. Vitruvius (5.2.2) also did not express any concern about echoes, but instead wished to avoid the voice being scattered and leaving the room. 9 Conclusion We hope to have illustrated the new prospects of an acoustic reconstruction of ancient council halls, which raises new questions about the interdependency between architecture and different kinds of ancient cultural praxis. On the one hand, the contribution to the research on ancient architecture lies in the identification of those building components which are important for the acoustic properties of a structure, such as the general interior setting, the building material, the ceiling, or the dimensions of the room. On the other hand, thanks to the combination of different scientific disciplines, the correlation of space and speech can be re-established. This, for instance, leads to a better understanding of the different vocal requirements of a speaker and the context of a speech in the senate or the Bouleuterion of Athenai. Questions for further research include the speed of speeches or the reception of rhetorical style depending on the acoustic properties of the space. Furthermore, we must reconsider whether our modern attitudes towards what constitute “good” acoustics are relevant to ancient architecture and society. List of References Bartoli, A., 1963: Curia Senatus (Rome). Bonnefond-Coudry, M., 1989: Le senat de la république romaine de la guerre d’Hannibal à Auguste. Pratiques délibératives et prise de decision. BÉFAR 273 (Rome). Bonner, S.F., 1977: Education in Ancient Rome: From the Elder Cato to the Younger Pliny (London). Carey, C., 2003: Aeschines. 2nd ed. (Austin). Develin, R., 1989: Athenian Officials 684–321 B.C. (Cambridge). Dörpfeld, W., 1896: “Funde”, AM 21: 103–20. Fatucci, G., 2009: “La Curia Iulia: una proposta di ricostruzione”, Workshop di Archeologia Classica 6: 113–21. Fisher, N., 2001: Aeschines. Against Timarchos: Introduction, Translation, and Commentary (Oxford). Gneisz, D., 1990: Das antike Rathaus: das griechische Bouleuterion und die frührömische Curia. Dissertationen der Universität Wien 205 (Vienna). Gogos, S., 2008: Das Dionysostheater von Athen: architektonische Gestalt und Funktion (Vienna).

287 Granger, F. (ed.), 1955: Vitruvius. On Architecture. 2 vols. Loeb 251 (Cambridge, MA). Kapparis, K., 1998: “The Law on the Age of the Speakers in the Athenian Assembly”, RhM 141: 255–59. Knackfuß, H., 1908: Das Rathaus von Milet. Milet 1.2 (Berlin). Kockel, V., 1995: “Bouleuteria: Architektonische Form und urbanistischer Kontext”, in Wörrle, M. & Zanker, P. (edd.), Stadtbild und Bürgerbild im Hellenismus: Kolloquium, München, 24. bis 26. Juni 1993. Vestigia 47 (Munich) 29–40. Krischen, F., 1941: Antike Rathäuser (Berlin). Kuhn, G., 1984: “Das neue Bouleuterion von Athen”, AA 1984: 17–26. Lamb, W.R.M. (ed.), 1930: Lysias. Loeb 251 (Cambridge, MA). Lauter, H., 1991: “Review of Das antike Rathaus: das griechische Bouleuterion und die frührömische Curia, by D. Gneisz”, Gnomon 63: 745–46. Leovant-Cirefice, V., 2000: “Le rôle de l’apostrophe aux Quirites dans les discours de Cicéron adressés au peuple”, in Achard, G. & Ledentu, M. (edd.), Orateur, auditeurs, lecteurs (Paris) 43–55. Marrou, H.I., 1950: Histoire de l’education dans l’antiquité. 2nd ed. (Paris). Marrou, H.I., 1957: Geschichte der Erziehung im klassischen Altertum (Freiburg). McDonald, W.A., 1943: The Political Meeting Places of the Greeks (Baltimore). Melmoth, W. & Hutchinson, W.M.L. (edd.), 1915: The letters of Pliny 2. Loeb 59 (London). Miller, S.G., 1995: “Old Metroon and Old Bouleuterion in the Classical Agora of Athens”, in Hansen, M.H. & Raaflaub, K.A. (edd.), Studies in the Ancient Greek Polis. Historia 95 (Stuttgart) 151–90. Neuhauser, W., 1958: Patronus und Orator: eine Geschichte der Begriffe von ihren Anfängen bis in die Augusteische Zeit (Innsbruck). Ober, J., 1989: Mass and Elite in Democratic Athens. Rhetoric, Ideology, and the Power of the People (Princeton). Papillon, T.L., 2007: “Isocrates”, in Worthington, I. (ed.), A Companion to Greek Rhetoric (Oxford) 58–74. Pritchard, D.M., 2015: “Athens”, in Bloomer, W.M. (ed.), A Companion to Ancient Education (Malden) 112–22. Ramsey, J.T., 2007: “Roman Senatorial Oratory”, in Dominik, W.J. & Hall, J. (edd.), A Companion to Roman Rhetoric (Malden) 122–35. Rhodes, P.J., 1972: The Athenian Boule (Oxford). Rhodes, P.J., 1980: “Ephebi, Bouleutae and the population of Athens”, ZPE 38: 191–201. Rhodes, P.J., 1981: “More Members Serving Twice in the Athenian Boule”, ZPE 41: 101–02. Rhodes, P.J., 1984: “Members Serving Twice in the Athenian Boule and the Population of Athens Again”, ZPE 57: 200–02.

288 Roux, G., 1976: “Aristophane, Xénophon, le Pseudo-Démosthène et l’architecture du bouleutérion d’Athènes”, BCH 100: 475–83. Saliou, C., 2009: Vitruve de l’architecture, livre V: Texte établi, traduit et commenté (Paris). Salomies, O., 2005: “Redner und Senatoren. Eloquenz als Standeskultur (1.–3. Jh. n. Chr.)”, in Eck, W. & Heil, M. (edd.), Senatores populi Romani. Realität und mediale Präsentation einer Führungsschicht. Heidelberger althistorische Beiträge und epigraphische Studien 40 (Stuttgart) 229–62. Scholz, P., 2011: Den Vätern folgen. Sozialisation und Erziehung der republikanischen Senatsaristokratie. Studien zur Alten Geschichte 13 (Berlin). Schulz, V., 2014: Die Stimme in der antiken Rhetorik. Hypomnemata 194 (Göttingen). Shear, T.L., 1995: “Bouleuterion, Metroon and the Archives at Athens”, in Hansen, M.H. & Raaflaub, K.A. (edd.), Studies in the Ancient Greek Polis. Historia 95 (Stuttgart) 151–90. Syson, A., 2009: “Born to Speak: Ingenium and Natura in Tacitus’s Dialogue on Orators”, Arethusa 42: 45–75.

FRON ET AL. Talbert, R.J.A., 1984: The Senate of Imperial Rome (Princeton). Taylor, L.R. & Scott, R.T., 1969: “Seating Space in the Roman Senate and the Senatores Pedarii”, TAPA 100: 529–82. Thompson, H.A. & Wycherley, R.E., 1972: The Agora of Athens: The History, Shape, and Uses of an Ancient City Center. Agora 14 (Princeton). Thompson, H.A., 1937: “Buildings on the West Side of the Agora. The American Excavations in the Athenian Agora (Eleventh Report)”, Hesperia 6.1: 1–226. Von Gerkan, A., 1941: “Die römische Curia”, in Krischen, F. (ed.), Antike Rathäuser (Berlin) 34–44. Wolff, C., 2015: L’éducation dans le monde romain: du début de la République à la mort de Commode. Antiquité synthèses 16 (Paris). Woodhead, A.G., 1967: “ΙΣΗΓΟΡΙΑ and the Council of 500”, Historia 16: 129–40. Worthington, I. (ed.), 2007: A Companion to Greek Rhetoric (Oxford).

chapter 18

New Architectural Work on the Akropolis of Selinous, Sicily: Towards a Digital Platform for Cultural Heritage Clemente Marconi, David Scahill, and Massimo Limoncelli The archaeological remains on the so-called akropolis of Selinous, particularly the urban sanctuaries in its southern portion, present a difficult challenge for architectural documentation and investigation. The sheer size and amount of architectural material make the site daunting to work at. Since the nineteenth century, scholars have tackled aspects of the buildings in an attempt to present a clear picture of this part of the city, but until now very little was known about the chronology of the buildings other than what could be gleaned from the stylistic details of the architecture, due to the difficult nature of documenting and studying such intimidating material, including the sheer number of architectural blocks—either in situ, in various states of destruction, or scattered—that had to be accounted for (for earlier work see esp. Hittorff & Zanth 1870; Serradifalco 1834; Koldewey & Puchstein 1899, 90–115; Gàbrici 1929; Gàbrici 1933; Gàbrici & Tusa 1956; Miles 1998–99; Mertens 2003, 46–49; Mertens 2006, 102– 103, 115–125, 184–188, 228–231; Amici 2009). The Institute of Fine Arts–NYU project began an ambitious programme of recording and documentation that has incorporated several cutting-edge methods that have aided our progress, including the latest in digital technologies. In this paper, we first give a general overview of the area under archaeological investigation with some of our initial results and conclusions regarding the phases of development in the main urban sanctuary on the akropolis. We then briefly discuss the main methods we are using for data capture and presentation, including the Virtual Reality (VR) platform, which brings together all of the documentation and research into a presentation platform for educational purposes. Finally, we present some of the architectural research we have undertaken, highlighting the methodologies we have used for documentation and presentation. We conclude with some remarks on plans for continued fieldwork and what we have learned and areas that we see could be improved regarding these new methods used for documentation and presentation.

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004416659_020

1

IFA–NYU Archaeological Investigations on the Akropolis of Selinous

In 2006, the Institute of Fine Arts–NYU, in collaboration first with the Superintendency of Trapani and then with the newly created Parco Archeologico di Selinunte, started a research project in the area of the main urban sanctuary on the akropolis (Fig. 18.1).1 In the first ten years, our work has focused on the southern sector of the temenos, particularly the area delimited to the north by Temple C, to the south by the peribolos wall, to the east by the socalled Sale A, and to the west by Temple R. Our interdisciplinary project aims at a new topographical, architectural, and archaeological analysis of the sanctuary, including a block-by-block study of the buildings and a detailed stratigraphic excavation. Although our part of the sanctuary was repeatedly investigated since the early nineteenth century, and all the way up to the early twentieth century, our work is the first to provide a detailed reconstruction of the complex sequence of human occupation in this area. This sequence is of particular significance for our understanding of life in Selinous from the Mesolithic all the way down to the Early Hellenistic Period, and provides the proper context for Temples R and B. In particular, moving from the bedrock all the way up to the present ground level, we were able to identify ten phases, as follows: (1) a first occupation dating back to Prehistory, including the Mesolithic period and the Bronze Age, the 1  We give special thanks to the successive directors of the Parco Archeologico di Selinunte (Dr. Caterina Greco, Dr. Leto Barone, and Architect Enrico Caruso) and to the Superintendency of Trapani (in particular the former Superintendent Architect Giuseppe Gini and the Director of the Archaeological Unit, Dr. Rossella Giglio). We are also grateful to the institutions that have supported our work at Selinous, including the Malcom Hewitt Wiener Foundation, the 1984 Foundation, the Kress Foundation and the Samuel I. Newhouse Foundation.

figure 18.1

Plan showing the part of the main urban sanctuary under investigation (2006–2017) by the IFA–NYU mission, including Temple R, Temple B and its altar, the South Building, and the south side of Temple C F. Pisciotta, D. Scahill, and M. Limoncelli

290 MARCONI, SCAHILL, AND LIMONCELLI

NEW ARCHITECTURAL WORK ON THE Akropolis of Selinous, SICILY

latter dating from the early to the late second millennium BCE (Cultraro & Marconi 2017); (2) a subsequent phase of abandonment between the Early Iron Age and the third quarter of the seventh century BCE; (3) the use of the area for cult practice in the open air in association with the foundation of the Greek polis of Selinous; (4) the construction of structures made of perishable materials, related to cult activity and datable at the latest to ca. 610 BCE; (5) the dismantling of these early structures to make way for the construction of Temple R, ca. 590–580 BCE; (6) a general restructuring of the entire area ca. 500 BCE—within the general renovation of the main urban sanctuary started around the middle of the sixth century—including the renovation of Temple R after damage caused by fire and the construction of the “South Building”, the latter more likely identified with a theatral structure; (7) the damage of Temple R by fire towards the end of the fifth century, presumably on the occasion of the Carthaginian sack of Selinous in 409 BCE; (8) the subsequent reconstruction of Temple R only a few years later, before the end of the fifth century BCE and thus presumably on the occasion of the reconstruction of Selinous under Hermokrates; (9) the use of Temple R during the course of the fourth century, when it is unclear for how long the building retained its original cultic function; and finally (10) the substantial transformation of the whole area in ca. 300 BCE, including the realisation of a large fill, which raised the ground level up to one meter, including the area inside the cella of Temple R, and which served as the foundation for the construction of Temple B. Within the last phase, Temple R was drastically reshaped, including adding three pilasters on the axis of the cella to support a new roof. Unfortunately, our reconstruction of the development of this area of the sanctuary ends here, given that previous excavators have eliminated most of the deposit above the fill of ca. 300 BCE. We thus have a limited understanding of later history: including the first half of the third century, before Selinous was abandoned within the context of the First Punic War, and during the Roman, Late Antique, and Medieval periods, which are nonetheless attested by a few sporadic ceramic finds. Turning to the investigation of the main buildings in our area of operation, in the first phase, in 2006–2010, research has concentrated on Temple B, which is best known

291

for its traces of polychromy (Marconi 2008; Marconi 2012). Thanks to a new block-by-block study, we can now suggest a reconstruction of the building as a prostyle tetrastyle temple standing on a podium. Now dated archaeologically to ca. 300 BCE, Temple B belongs to the phase of Punic control of Selinous, and with its Greek forms it represents an important indication of the mixed Graeco-Punic character of the settlement at the time (Helas 2012; Chiarenza 2018–19). Unfortunately, the phases of use of Temple B were almost entirely removed in the excavations carried out in the area between the nineteenth and twentieth centuries and we can only speculate about cult practice, mainly based on the architecture and the location of the building and its altar. Both have undisputable Greek forms, to which we may add, on the one hand, the location of Temple B immediately to the east of Temple R and, on the other hand, the possibility that the altar of Temple B had an earlier, Archaic phase associated with Temple R. This would suggest that Temple B was the Hellenistic successor of Temple R, a possibility already considered in earlier literature. As for Temple R itself, we have already alluded to its complex history in describing the various phases of occupation of our area (Marconi 2014a; 2014b; Marconi et al. 2017; Marconi 2018a). The temple is now archaeologically dated to ca. 590–580 BCE, based on the presence of Early to Middle Korinthian pottery in its exceptionally well preserved foundation deposit. In its original phase, the temple had an oikos plan with a long naos and relatively deep adyton. When, after a fire, the building was renovated in ca. 500 BCE, the most characteristic feature of this new intervention was the raising of the floor in the naos, to the level in the adyton: unlike the original clay floor, this new floor was made of white plaster, a sort of upgrade comparable to the floors of the large peripteral temples at the site. Temple R was damaged by fire a second time towards the end of the fifth century, and we have already mentioned the probable context for this event, as the subsequent renovation of the building, its use over the course of the fourth century, and possibly its end as a cult structure by ca. 300 BCE. At this time, the large fill covering the rest of the area included the inner cella of our building (Marconi 2017). This fill, mainly consisting of transport amphorae at the bottom, earth in the middle, and architectural terracottas at the top, has completely sealed Temple R in its Archaic and Classical phases, which, combined with our detailed excavation and documentation, allows us to gain a clear understanding of the biography of the building and the cult practice associated with it, which is the first time for a temple in Selinous.

292 2

MARCONI, SCAHILL, AND LIMONCELLI

Architectural Recording: Refining and Combining New Methodologies

Concurrent with the excavations conducted by the IFA– NYU team in the main urban sanctuary on the akropolis, we are working to document and study all of the architecture in the temenos area, first abutting the south side of temple C and eventually spreading out to include Temple C and the wider environs. Of preeminent importance was the creation of a new state plan of the area under investigation. We are working on creating a completely new digital state plan of the akropolis, concentrating first on the area where we are excavating (Fig. 18.1). The new digital site plan is, in fact, a digital archive and database capable of storing vast amounts of information related to the architecture, from survey and millimetric measurements to layers of phases, state plans and reconstruction drawings—which in turn serve as a basis for modelling in 3D and within a VR environment. Its drawings are inherently more accurate than any other previous illustration production method because the database maintains millimeter accuracy within a vector environment that can be reproduced in a myriad of ways at any scale necessary. As an archival system, it contains information from survey coordinate data to detailed block information—dimensions, catalogue numbering, etc. As of 2012, we have been using new dGPS (differential Global Positioning System) points for surveying, and as of 2014, new geodetic points established on the akropolis under the auspices of the Parco Archeologico di Selinunte. As our work includes new surveying, we have discovered and corrected some discrepancies regarding coordinates as well as the North South axis used in previous surveys. The new survey supersedes all previous surveys in terms of accuracy and full integration with the larger geodetic grid of Sicily (for measurement, we use a Leica Total Station and download directly into AutoCAD for coordination with the digital site plan and photogrammetry; see below). Adding to the digital site plan work, we are also now using photogrammetry to supplement documentation of all of the architecture, which is then imported into the vectored state plan and digitised. We are using Agisoft PhotoScan (recently renamed MetaShape), a powerful photogrammetry program that has become popular within the field of archaeology. The photogrammetry streamlines the process of recording the architecture because it allows for very accurate 3D modelling relatively quickly, but it also dictates the procedures of the recording by virtue of the photography required on site and the processing time of the software afterwards. We then process orthographic images of plans, elevations and sections of all of the architecture under study, as well as photogrammetric

documentation of the excavation trenches as they proceed (Fig. 18.2a,b) (Limoncelli 2017). The 3D data that we generate has a vast potential for architectural study and presentation. Besides serving as a primary dataset of 3D information or orthographic images for presentation, it also serves as a base for making accurate 2D drawings of architecture and for use in modelling applications. One of the most exciting possible uses of 3D photogrammetry modelling is the ability to import modelled architecture into 3D reconstructions (e.g., object files can be imported into 3D models depending on software capacity). In essence, we can do virtual reconstruction using what exists and what we recreate anew in the virtual environment. This has enormous potential for aiding heritage management projects in real world applications of preservation and reconstruction and in online applications for education and research (Limoncelli 2012; Scahill 2016). The process of documenting and recording the architecture requires constant adjustment and reassessment as we move forward in the digital world. Because computer methodologies evolve so quickly, it is always necessary to be thinking of new ways to proceed and new software to employ, but ultimately balancing this exploration with what is needed from a conceptual standpoint of the excavation, documentation, and presentation of the material. The continual technological development of our work runs in tandem to the fieldwork, while our goal is to maximise the highest quality output of information while minimising wasted time in experimentation. This is a balance that always needs to be weighed in an excavation where time is precious. We also keep in mind that while we pursue these digital methods other more traditional forms of documentation, sometimes out of necessity, can still supplement our research goals and sometimes serve as a check on accuracy. Overall, our goal is to document and present the architecture in the best way possible and to do this we use every means possible, from hand drawings to 2D AutoCAD line drawings to photogrammetry, to 3D digital reconstruction models (Scahill 2016). 3

Virtual Reality (VR) Platform

From a methodological point of view, the first step of setting up a VR platform—a screenshot of which is shown in Figure 18.3—is the documentation of the current state of the building and the gathering of all the information available by direct analysis of the architectural work being conducted, ranging from data derived from archaeological excavation to analysis of individual architectural elements of the building to the direct (with instruments)

figure 18.2a,b

Process of photogrammetry for Temple B M. Limoncelli

NEW ARCHITECTURAL WORK ON THE Akropolis of Selinous, SICILY

293

294

figure 18.3

MARCONI, SCAHILL, AND LIMONCELLI

VR image of the area with Temple R, Temple B and Temple C, from South Building steps authors

or indirect (without instruments) relief, namely the creation of floor plans, elevations, sections and 3D surfaces according to different methodologies, from laser theodolite surveying to the precise photogrammetry techniques, photo-modelling and laser scanners. The techniques and specifications for data capture are dependent on the state of preservation and conservation, the location and topographical situation, and the size and shape of the buildings studied. Once all the information is gathered we proceed to the second step, which is the transition from plans to 3D reconstructions through the development of one or more three dimensional virtual models, ones made with the technique of “hand-made” modelling—i.e., by means of user-created solid modelling, polygonal, NURBS, or subdivision 3D surfaces—of the individual historical phases of the buildings (Limoncelli 2012). In this particular phase of work attention is paid to the texturing of the models— i.e., the application of the materials for the composition of the model objects. In fact, for every single building a photographic survey including targets is performed of all the materials used in buildings, which is designed to give back to the buildings not only their volumetric-spatial aspect, but also their original appearance, regarding the relationship between the material, the colour, and decoration,

which allows for a greater understanding of the architectural composition. The Virtual Tour (VT) is the end result of a methodological and scientific path of study, so that the details and information of a single monument, as designed and reconstructed, delivers and communicates a high level of complexity of an ancient city, characterised by a long continuity of life. The research data is collected in a multimedia platform that can function both online and offline. Based on the evaluation of their communication techniques, we have chosen to use a representation system based on Real Time technology QTVR (QuickTime VR) (www.apple .com/quicktime) where navigation inside the VT is done through images of a “panoramic” cylindrical or spherical type. Panoramic images (with stereoscopic vision and without) are obtained directly within the virtual 3D models from the modelling software. The tour begins with an interactive satellite map of the area of Selinous that incorporates the main archaeological sites in the area. Each monument can have the addition of panoramic images linked together so as to ensure continuity of the virtual tour; and the movement takes place with the aid of soft transitions that allow a more gradual and realistic transition from one monument to the next. The user can query

NEW ARCHITECTURAL WORK ON THE Akropolis of Selinous, SICILY

each panorama using clickable hotspots at points of interest that offer information about what is being viewed through pop-up cards. Regarding the complexity of the transactions carried out for each building, different viewing levels are provided and these are detectable within views by switches indicated with coloured icons. The use of the platform is structured on the combined action of a series of files that cooperate among themselves for the creation of a single project within the Web environment of digital or DVD media, comprised of an HTML file, various files in XML format, a large number of JPEG images and MP3 audio files and MP4 video. The contents of the platform are designed to be usable on all portable devices such as Smartphones and Tablets compatible with Flash and / or HTML5 / CSS3, including Apple devices (iPad, iPhone, etc.). 4

Architectural Investigation and Presentation

Turning to what we can say about the architectural development of the area, discussion of ritual space and especially processional space in Greek architecture often is difficult to pin down owing to problems of understanding not only the rituals involved but also the architecture of the space in question. Fortunately, in the temenos space on the akropolis at Selinous we have a rare opportunity to understand such activity with some clarity, thanks to the findings from our excavation, already discussed above, which are refining our understanding of the chronological phasing of the architecture. The area of the akropolis to the south of temple C presents an interesting case study of sanctuary space consisting of Temple R, from the early sixth century BCE; a monumental stepped platform for viewing (the South Building), built in the late sixth–early fifth century BCE; Temple B, added in the late fourth century BCE; and an altar east of the temple B at the northeast end of the South Building. As we now understand it, prior to the construction of Temple B, the space makes sense as an architectural ensemble for ritual or festival activity related to cult at the earliest structure, Temple R. 5

Temple R

As noted in the introduction, Temple R was constructed in the early sixth century BCE (Fig. 18.4). The temple is, in fact, among the first attempts at Selinous to build a monumental temple on the akropolis. Similar temples at Malophoros and Triolo Nord and at least one other possible

295

small temple in the same main urban sanctuary (Temple S) complement our structure. These buildings are of particular interest for the study of the early development of monumental Greek architecture. The construction of non-peripteral temples by a Greek city at a time in which temples in Syrakousai (Apollon), Korkyra (Artemis) and the mainland (Olympia) were already being constructed on a much grander scale with peripteral colonnades is instructive regarding the transmission of building practices at Selinous. If our chronology here and elsewhere for early temples is correct, Selinous was slower to adopt the new trends in Greek architecture; yet, soon after our city was inclined to build an ensemble of structures for cult practices and expand construction on an enormous scale within a short period of time in the sixth century. Although relatively small in comparison to the later temples at Selinous, Temple R is constructed in a manner consistent with early Greek monumental buildings, with regular cut blocks of limestone on a rectangular plan with a cella and backroom, or adyton, and additional backroom, or opisthodomos, which was added early in the lifetime of the building (Fig. 18.5). The total dimensions of the building before the addition of the opisthodomos were 15.50 × 5.64 m, and with the addition of the opisthodomos the total length became 17.53 m. The building had a terracotta roof with local Sicilian type pan and cover tiles—a modified version of Early Korinthian tiles—as well as a decorated geison revetment and sima and possibly horse-rider akroteria above both facades (Marconi 2018b). The presentation of all of these details is possible through combination of the methods for documentation and are best illustrated by the reconstructions in Cinema 4D. Using all of methods we are able to present the building in a number of different ways to highlight various aspects of the building, including everything from line drawings, plans, and elevations to photogrammetry and 3D modelling. All of these methods are essential to present the architecture in the best analytical and artistic way possible and lead to a greater understanding of the character and impact of the architecture on the viewer/ participant. This kind of comprehensive approach is still typically followed in the field of modern architecture for analysis and presentation, but is not always followed for ancient architecture. An aspect of using photogrammetry particularly useful for architectural recording is to help document and understand the settling and displacement of walls in buildings. This is especially relevant to our study of Temple R, whose walls are coursed with a great deal of unevenness (Fig. 18.6). Although it appears to be due to earthquake damage, the breakage and settling cannot be completely

296

MARCONI, SCAHILL, AND LIMONCELLI

figure 18.4

Temple R model M. Limoncelli

figure 18.5

Temple R cutaway showing interior, Archaic Phase M. Limoncelli

297

NEW ARCHITECTURAL WORK ON THE Akropolis of Selinous, SICILY

figure 18.6

Temple R, photogrammetric model of back wall of adyton showing deformation of the wall authors

explained in this way. The west wall is clearly lower at the northern end, gradually descending in all of its courses due to breakage in the lower courses. The breakage would appear to be due to shearing: vertical breaks from blocks and/or earth settling under them, however the north wall continues to follow the courses of the west wall without breakage, and in fact there is a slight rise from the west end to the east end of the building on this side without apparent warping in the courses. The nature of the north wall remains to be examined in further seasons when the lower courses of the north face of the north wall are revealed. The South wall of Temple R slopes downward gradually from the west end to the east end. The east end is nearly equal in height at both north and south corners. This suggests that the building settled on all sides except the southwest corner, which perhaps did not settle due to more stable bedrock below this corner. The documentation of the walls of Temple R with the photogrammetry modelling also allows for extremely accurate measurements in all dimensions. The ability to study the building three dimensionally with the accuracy

of the photogrammetry modelling is enormously beneficial towards understanding the deformations and destruction of the building over time. The data also help us understand the processes of site formation regarding the stratigraphy, which is affected by the settling and damage within the building. In addition, the photogrammetry has served as a dataset for us to assess damage to the structural integrity of the building as we excavate the levels around and inside the walls. Finally, the conservation team has been using the datasets from the photogrammetry to assess the treatment of the limestone blocks of the building and their stabilisation (Walthew et al. 2017 provides an indepth look at collaboration on this problem between the architecture and conservation teams). 6

The South Building

Roughly on the same axis of orientation as Temple R and just to the southeast is the so-called South Building, consisting of a large rectangular platform for a monumental

298 stepped structure facing north, which we interpret (following earlier suggestions by Thomas Becker, Dieter Mertens, and Clemens Voigts) as a stepped viewing platform, or rectilinear theatron (Fig. 18.7). The structure abuts the temenos wall. Earlier interpretations of its foundations suggested they were for a monumental altar, but this is problematic given the location, orientation and construction. Because our full discussion of the South Building has been published elsewhere (Marconi & Scahill 2015: 279–92), here we will focus on the analytical side of our research pertaining to new documentation and modelling that allows us to analyse and more fully develop an understanding of the spatial characteristics and relationship that the South Building has with the other built structures in the area. The South Building is critically important for understanding the ritual activity in this area related to Temple R. Dividing the total 23-m length of the South Building by an average of 0.50 m per occupant leaves room for 46 people per row. Multiplied by 11 rows, there would be room for 506 standing or seated people. There is room for discussion as to whether viewers stood or were seated. With eleven rows of steps, our building would have had a total height of ca. 2.70 m from the lower front krepis to the top step. The back peribolos wall most likely rose at least another metre to metre and a half with a capping geison course on top, comparable to that found in association with the peribolos wall of the Sanctuary of Malophoros (Gàbrici 1927: 16–21, fig. 6). Along both the east and west sides of the steps it is most likely that there were balustrades, with capping courses rising perhaps in a stepped fashion (like at Malophoros). For the dating of this original phase, the relationship of the South Building to the peribolos wall offers a terminus post quem after the construction of the latter wall, or after ca. 550 BCE. Excavation in front and in layers that extend underneath the foundations and stylistic analysis of the step construction indicate that most likely the South Building was built at the same time as the general restructuring of the sanctuary in the last decades of the sixth century or beginning of the fifth century BCE. The use of the South Building as a monumental viewing platform/theatron makes sense when we consider the temporal and spatial relationships of the architecture within this part of the temenos (Fig. 18.8a) (Hollinshead 2015). The distance from Temple R to the end of the temenos is 25.74 m, while the width is 14.30 m to the edge of the krepidoma of Temple C. The platform of the South Building thus would provide a perfect vantage point from which to view ritual performances in front of Temple R (Fig. 18.8a,b). The space would seem more than adequate

MARCONI, SCAHILL, AND LIMONCELLI

figure 18.7

South Building state plan and elevation, first phase D. Scahill

for processions or dancing accompanied by musical entertainment, such as would be expected for ritual activities connected to the cult of the temple. It is worth noting that the area itself was secluded from the road and the southern sector of the sanctuary by the high temenos

NEW ARCHITECTURAL WORK ON THE Akropolis of Selinous, SICILY

a

b

figure 18.8

(a) Plan showing open space in front of Temple R and the South Building (Scahill); (b) VR view of the area in front of the South Building looking west to Temple R, showing alternate view with altar directly in front of Temple R M. Limoncelli

299

300

figure 18.9

MARCONI, SCAHILL, AND LIMONCELLI

View of area after construction of Temple B M. Limoncelli

wall. Even more important for the consideration of musical performance, the temenos wall, 3.50 to perhaps 4.0 m high, would help to enclose the sound, acting to some extent like a theatre cavea. It is only when thinking about the space three dimensionally that we can truly appreciate the effect of the enclosed area as a theatre-like experience that would include Temple R to the west, the altar to the east and Temple C to the North with its own altar to the east. The grouping of buildings then takes on what appears to be the deliberative characteristics of an interactive spatial environment for performance. The 3D models and VR platform convey this most clearly. After Temple B was constructed, the space within the temenos was altered significantly and its character changed, while the South Building continued to be used in some way (Fig. 18.9). Two later phases of the South Building are identified possibly with an addition of a staircase casement at the east end, which does not bond with the steps, dating to perhaps the fifth century, and a fourthcentury phase in which the ground level was raised and an additional course of steps added on the front—during or after which Temple B was built by the end of the century. Cuttings that held stelai of some kind and are evident in the front steps belong most probably to a Hellenistic shift

in use of the steps. That Temple B was built at the end of the fourth century attests to a “Greek” presence of building style still present at Selinous and the question of continuation of cult practice through the fourth century after the Carthaginian sack in 409 BCE must be addressed. If our reconstruction is correct, the spatial relationship between Temple R and the South Building are indisputably of key importance to the ritual activities at the temple. The ability to visualise these aspects more fully with 3D modelling helps us to better understand that relationship and explore aspects of it that might not have been so apparent otherwise. The construction of the South Building against the temenos wall and the fact that the wall extends and frames the space on the South side of the temenos can best be understood, based on this analysis, as theatral in quality. 7

The Altar of Temple B

Adding to the complexity of the area, the altar for Temple B came under scrutiny once we had analysed the spatial relationship of Temple R and the South Building, and we began to wonder if this altar may have possibly had an

301

NEW ARCHITECTURAL WORK ON THE Akropolis of Selinous, SICILY

earlier phase coinciding with the theatral space, due to its position framing the east end of the area. We can now offer some tentative assessments that may have a bearing on the ritual activities related to Temple R and the South Building. To the east of Temple B are the foundations for an altar (Fig. 18.1). There are only a few fragments that can be tentatively associated with a superstructure, comprising several small fragments of triglyphs, found immediately to the west of the altar, in what appears to have been the fill of the foundation trench for the altar itself. Three joining fragments and several small non-joining fragments belong to a triglyph with a width of 0.25–0.27 m—the same width as the triglyph for Temple B. However, one fragment preserves the undercut of the glyph, which is more angular than that of the triglyphs of Temple B. This type of undercut should be earlier than the fourth century and is more like the undercut of triglyphs from the sixth and fifth centuries BCE. If the triglyph fragments found in excavation in front of the altar foundations are from a phase of the altar, and if they can be dated to an earlier phase rather than in the late fourth century BCE when Temple B was constructed, then we are dealing with an older altar on the same site as a later replacement (see Voigts 2017 for comparisons, the most useful of which are the large triglyph altar near Temple D and the small triglyph altar). Aside from the triglyph fragments, one piece of an upper section of an anta pier for a plinth and one fragment of a base molding perhaps belong to the late fourth century incarnation of the altar. Following the type of triglyph altars we see at Selinous, we may reconstruct a set of seven stairs leading up to a platform with an altar table rising up at the back (Fig. 18.8a). These dimensions for the reconstruction are based on the overall dimensions of the foundations and on the dimensions of the triglyph as reconstructed from the fragments preserved. A frieze utilising these triglyph fragments, having triglyphs of 0.27 m and metopes 0.41 m in width, would fit exactly the base dimensions of the altar with six triglyphs and five metopes (Fig. 18.9). The depth of such an altar likewise works within the space of the extant foundations. We have many examples of triglyph altars at Selinous to serve as parallels, and this type is attested from the Archaic period to the late fourth century BCE. An altar in this location in the Archaic period would likely have been connected to Temple R, and its alignment with Temple R supports this hypothesis (Fig. 18.8a). This would connect Temple R, the South Building, and the altar together as an architectural unit for ritual activities taking

place between Temple R and the altar in an open area later occupied by Temple B. 8

Temple B

Finally, concluding this brief discussion of the buildings within the temenos is Temple B (Fig. 18.2a,b, 18.9, 18.10). Temple B was a small Doric temple built sometime around the end of the fourth century BCE. Placed right in front of the South Building steps, nearly abutting it at the west end, the temple was elevated on a platform accessed by a staircase on the east side. The elements that belong to it are decidedly Doric and stylistically belong to the late fourth century BCE, although it was initially published by Jacob Ignaz Hittorff as Ionic (Marconi 2008; Marconi 2012; a comprehensive publication of Temple B is in preparation). We have reconstructed the temple as tetrastyle prostyle based on the evidence of the preserved antae, which indicate a prostyle arrangement. Temple B provides a good example of the hybrid documentation and reconstruction efforts using CAD, photogrammetry, and 3D modelling in combination to produce a reconstruction with a rare example of polychromy still preserved on several fragments of the superstructure which are in the Archaeological Museum in Palermo (Fig. 18.9, 18.10). On site, stucco is still preserved on the antae and some lower wall blocks, as well as the krepidoma of the building, and the foundations themselves exhibit a colourful distinction according to the coursing. The presentation of the building, particularly through the 3D modelling, best illustrates the unusual features of the stepped platform, especially the colours and textures of the different stone used, as well as the polychromy of the superstructure, preserved and reproduced here. The models for Temple B perhaps best illustrate how we strive to create realistic rendering that is not pristine, but reflects the reality of weathering that buildings do encounter in the environment, in part through complex texturing based on the photogrammetric imagery. 9 Conclusions In conclusion, we might pause to consider that the area under study would be dwarfed by comparison with Temple C after its construction in the middle of the sixth century, but cult practice in this area no doubt took place alongside Temple C, as the phasing for our three structures, Temple R, the South Building and the altar, continues down into

302

MARCONI, SCAHILL, AND LIMONCELLI

figure 18.10 Analytical elevations of Temple B combining photogrammetry and digital modelling M. Limoncelli

the fourth century BCE (Fig. 18.11). The formal, theatral arrangement of the space was eventually altered by the imposition of Temple B at the end of the fourth century BCE. Looking at the bigger picture of the temenos itself, the main entrance to the space would likely have been from the east edge of the South Building and its back wall would have blocked the area from the road. In a wider context we can now begin to situate our buildings in relation to the rest of the sanctuary and other areas of the akropolis with more clarity. Already we can discuss an architectural and technological development in terms of infrastructure capability by the mid-sixth century at Selinous, when we see Temple C constructed, that expands in the fifth century into areas especially north and south of our area. The next stage of architectural work for the IFA–NYU project includes documentation of Temple C’s remains, including the fallen columns and other parts of the superstructure. This is a difficult task, but with all of the methods outlined above we feel we now have the means to accomplish it. Regarding immediate and long-term issues that need to be addressed concerning methodologies employed for architectural study, we might single out a few in the interest of transparency. Hand-drafted drawings, especially large state plans of foundations, walls and the like, can be difficult to incorporate into a digital environment. For instance, there is no easy solution for scanning and digitising hand drawings into CAD, where millimeter accuracy in the software exposes the inherent inaccuracies of hand

drawing, the process of scanning physical drawings, and the digital tracing of the linework. We have hand drawings at 1:10 scale of temple foundations in our area of investigations, which still require scanning and digitisation, and we are currently weighing the best way and the right time to accomplish this. This is a lengthy process in which the overall accuracy of the scan must be carefully maintained in a CAD environment. With photogrammetry, the potential applications are vast. Both extreme accuracy of measurement in 3D and modelling of detail are now achievable. The variability of results and levels of refinement involved are the major hurdles that must be overcome for 3D photogrammetry to be truly useful in the long term for architectural data capture. Survey point data input, high camera resolution and reduction of “noise” in the source photographs are all measures that can increase accuracy but must be applied in a rigorous way (see Sapirstein & Murray 2017). The usefulness of photogrammetry for architectural historians and archaeologists is only limited by the data capture methods and results achieved in the field by the methods spelled out above. The technique is immediately useful for everything from 2D orthographic projections and elevations, plan views of small or large areas of architecture, 3D modelling, sections, photo-realistic modelling with texture, virtual reconstruction and an endless supply of data and presentation views that can be extracted at any time. In theory, the 3D reconstructions and VR platform utilise the data captured and the subsequent research outlined

303

NEW ARCHITECTURAL WORK ON THE Akropolis of Selinous, SICILY

figure 18.11

Size comparison of the area under investigation (2016–2017) with Temple C M. Limoncelli

above, which means the reconstructions are held to a rigorous level of accuracy but are still subject to our interpretation of the evidence. The reconstructions are, however, adaptable and can be continually modified and updated for accuracy as new information is gathered and new conclusions are drawn. Another issue of concern is the virtual reconstruction environment itself. At the moment, maneuvering within the virtual world created by current software, one is immediately aware of a skewed perspective. In fact, the perspective views that the software produces may not register as realistic, or “architecturally correct” in the sense of axonometric or isometric views or point perspective, but rather appear as distorted perspectives that the software establishes as the view changes. This environment will no doubt improve over time. Our goal is to achieve a scientific, architectural presentation suitable for archaeological and educational purposes. To end on a positive note, digital data capture and VR software are powerful tools for documenting, presenting and understanding the ancient structures we seek to study and the spaces they inhabit. As researchers, we are always consciously looking at ways to improve documentation and presentation. The ability to improve is dependent on funding and flexibility to evolve as the hardware and software become available. Ultimately, the aim is for a holistic approach in which all forms of data capture and documentation, from traditional hand drawings and measurements to photogrammetry and virtual reconstructions, can be used interactively and, in an ideal world, seamlessly for architectural analysis and research, allowing for accurate and illuminating presentations of

the built environment for academic and public purposes alike. An immediate outcome of the VR/VT platform at Selinous is its use by the Parco Archeologico di Selinunte for educational exhibits aimed at the public. We hope to continue to expand and improve on this and other platforms for study and education. References Amici, C. 2009: “Selinunte, Tempio C: analisi tecnica per la ricostruzione”, Palladio 44: 11–30. Chiarenza, N., 2018–2019: “Selinunte tra la seconda metà del IV e il III secolo a.C. Un insediamento dell’eparchia cartaginese al centro del Mediterraneo”, Karthago 31: 27–63. Cultraro, M. & Marconi, C., 2017: “Mycenaeans and Others along Western Sicily: A View from Selinunte”, Aegaeum 41: 131–37. Gàbrici, E. & Tusa, V., 1956. “Studi archeologici selinuntini”, MonAnt 43: 205–408. Gàbrici, E., 1927: “Il Santuario della Malophoros a Selinunte”, MonAnt 32: cols. 1–419. Gàbrici, E., 1929: “Acropoli di Selinunte: scavi e topografia”, MonAnt 33: 61–112. Gàbrici, E., 1933: “Per la storia dell’architettura dorica in Sicilia”, MonAnt 35: 137–250. Helas, S., 2012: Selinus II. Die punische Stadt auf der Akropolis (Mainz). Hittorff, J.I. & Zanth, K.L.W., 1870: Architecture antique de la Sicile. Recueil des monuments de Ségeste et de Sélinonte (Paris). Hollinshead, M., 2015: Shaping Ceremony, Monumental Steps and Greek Architecture (Madison).

304 Koldewey, R. & Puchstein, O., 1899: Die griechischen Tempel in Unteritalien und Sicilien (Berlin). Limoncelli, M. 2012: Il Restauro Virtuale in Archeologia (Rome). Limoncelli, M., 2017: Virtual Restoration. Paintings and Mosaics (Rome). Lo Faso Pietrasanta, D., Duca Di Serradifalco, 1834: Le antichità della Sicilia esposte ed illustrate. Vol. II: Selinunte (Palermo). Marconi, C. & Scahill, D., 2015: “The ‘South Building’ in the Main Urban Sanctuary of Selinunte: A Theatral Structure?”, in Frederiksen, R., Gebhard, E.R. & Sokolicek, A. (edd.), The Architecture of the Ancient Greek Theatre (Aarhus) 279–92. Marconi, C., 2008: “Il tempio B di Selinunte: Hittorff, Serradifalco e la disputa sulla policromia dell’architettura greca nell’Ottocento”, Sicilia Antiqua 4: 59–91. Marconi, C., 2012: “Le attività dell’Institute of Fine Arts—NYU sull’Acropoli di Selinunte (2006–2010)”, in Ampolo II, C. (ed.), Atti delle settime giornate internazionali di studi sull’area elima e la Sicilia occidentale nel contesto mediterraneo. Erice, 12–15 ottobre 2009 (Pisa) 279–86. Marconi, C., 2014a: “Nuovi dati sui culti del settore meridionale del grande santuario urbano di Selinunte”, in Istituti editoriali e poligrafici internazionali (ed.), Κατὰ κορυφῆν φάος. Studi in onore di Graziella Fiorentini (Pisa and Rome) 263–71. Marconi, C., 2014b: “Two New Aulos Fragments from Selinunte: Cult, Music and Spectacle in the Main Urban Sanctuary of a Greek Colony in the West”, in Bellia, A. (ed.), Musica, culti e riti nell’Occidente greco (Pisa): 105–16. Marconi, C., 2017: “Un busto in terracotta dalla fronte del Tempio R di Selinunte”, Sicilia Antiqua 14: 191–97. Marconi, C., 2018a: “La dea del Tempio R”, in Antonetti, C. (ed.), Gli esametri Getty e Selinunte: Testo e contesto (Alessandria) 179–200.

MARCONI, SCAHILL, AND LIMONCELLI Marconi, C., 2018b: “Un acroterio equestre da Selinunte?”, In Nizzo, V. & Pizzo, A. (edd.), Antico e non antico: Scritti multidisciplinari offerti a Giuseppe Pucci (Milano and Udine) 377–84. Marconi, C., Kiene, M. & Lazzarini, L., 2017: “Sicile Ancienne”. Hittorff and the Architecture of Classical Sicily (Cologne). Marconi, C., Micciché, R. & Ward, A., 2017: “Contextualizing an Animal Sacrifice in the Foundations of Temple R: A Preliminary Report of the Institute of Fine Arts—NYU Excavations on the Acropolis of Selinunte (2013–2015 Campaigns)”, Mare Internum 9: 71–88. Mertens, D., 2003: Selinus I. Die Stadt und ihre Mauern (Mainz). Mertens, D., 2006: Städte und Bauten der Westgriechen (Munich). Miles, M.M., 1998–99: “Interior Staircases in Western Greek Temples”, MemAmAc 43–44: 1–26. Sapirstein, P. & Murray, S., 2017: “Establishing Best Practices for Photogrammetry in Archaeology”, JFA 42: 337–50. Scahill, D. 2016: “Architectural Reconstruction at Ancient Corinth, Old and New: The South Stoa” in C. Greco (ed.), Selinunte: Restauri dell’antico, Atti del Convegno Selinunte, Baglio Florio, 20–23 Ottobre 2011 (Rome): 287–96. Voigts, C., 2017: Selinus 6. Die Altäre in den Stadtheiligtümern: Studien zur westgriechischen Altararchitektur im 6. und 5. Jahrhundert v. Chr (Wiesbaden). Walthew, J., Mayberger, E., Serotta, A., Scahill, D. & Hight, A., 2017: “Collaboratively Thinking Forward: Three-dimensional (3D) Data in Conservation and Archaeology”, in Owczarek, N., Gleeson, M. & Grant, L. (edd.), Engaging Conservation: Collaboration across Disciplines (London): 266–78.

chapter 19

Architectural Documentation and Visual Evocation: Choices, Iterations, and Virtual Representation in the Sanctuary of the Great Gods on Samothrake Bonna D. Wescoat As archaeologists and architectural historians, we work within a fragmentary world, and it remains a central challenge to communicate structure, space, and environment, while simultaneously—and equally effectively—acknowledging the extent of the evidence we draw upon to do so.1 How do we bridge the gap between architectural documentation and visual evocation, particularly in the context of whole-site reconstructions and virtual environments? While lacunae and ambiguous evidence can be straightforwardly described in text, they are more challenging to convey visually. Drawings and models—our most powerfully communicative devices—tend to take on their own reality as they leave our hands. Archaeological explanation, metadata, and paradata detach as an image is broadly disseminated. How might we use models themselves to expose and communicate uncertainty without confusing or frustrating the viewer? I raise these issues to open discussion on the nature, purpose, and (dis)

figure 19.1

advantages of 3D virtual architectural reconstruction, which is our 21st-century way of reinventing the highly evocative 18th and the 19th century site reconstructions (Fig. 19.1). But we enter the hypervisual 21st century with a 20th-century commitment to archaeological precision, and with the desire to foreground the subtle differentiation (rather than generic iteration) that defines ancient Greek architecture. The Sanctuary of the Great Gods on the island of Samothrake provides an excellent case study for examining the needs, opportunities, and limitations of digital applications for highly individual Greek buildings within the equally idiosyncratic topography of a whole-site reconstruction. I do not claim to offer cutting-edge frontiers or review best practices in digital archaeological visualisation (see, e.g., “London Charter” 2.1 2009; “International Principles of Virtual Archaeology; The Seville Principles” 2010; Denard 2012; and extensive work conducted on

Reconstructed views of the Sanctuary of the Great Gods, Samothrake, 1880 and 2017 Left: after Conze, Hauser, and Benndorf 1880, pl. LXXVI; right: courtesy American Excavations Samothrace

1  My deepest thanks to the Samothrake team, as well as Domna Terzopoulou of the Ephorate of Antiquities of Evros, and Chryssa Karadima of the Ephorate of Antiquities of Rhodopi, for their support of our research.

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004416659_021

306

Wescoat

procedural modeling in archaeology, see below). Rather, I hope to share our evolving experience working digitally with a site that demands architectural precision, gains meaning through visual evocation, and generates new insights through virtual kinesthetic investigation. 1

Sanctuary of the Great Gods on Samothrake

The Sanctuary of the Great Gods was famed in antiquity for its mystery cult, whose secret rites promised initiates protection at sea and the opportunity for moral improvement (Diod. Sik. 5.48.4–50.1) (for the cult generally, see Hemberg 1950; Cole 1984; Lehmann 1998; Bowden 2010; Bremmer 2014, with earlier bibliography). Initiates kept silent trust concerning the rites they endured, and the archaeological record, rich in its own regard, has complicated rather than clarified many questions regarding the experience of initiation (Lehmann 1998, 29–45; K. Lehmann in P.W. Lehmann 1969, Text 2:3–50; see critique and further investigation in Cole 1984; Burkert 1993; Clinton 2003; Blakely 2007; 2012; Dimitrova 2008; Wescoat 2012; 2017a; 2017b; Cruccas 2014; Nielsen 2014). The architecture in the Sanctuary of the Great Gods, on the other hand, is uncontestably brilliant; its buildings are among the most highly inventive designs of the Hellenistic period. For the most part, the dozen or so sophisticated buildings constructed from the mid-4th to early-2nd century BCE are obligingly well preserved even if no longer standing (Fig. 19.2). Understandably, they became the focus of intellectual energy in the second half of the 20th century, first under the direction of Karl Lehmann and then James R. McCredie. Each publication of record centred on the reconstruction and historical explication of individual monuments (see the series, Samothrace: Excavations Conducted by the Institute of Fine Arts of New York University). What dropped out of such an approach was the essential spaces between the buildings, the terrain within which they are placed, and the way in which they interact. We turned to 3D digital modeling and computer graphics because such platforms allowed us to explore a fundamental aspect of the site that heretofore had been hiding in plain sight—the physical environment and the embodied actions and perceptions of pilgrims moving through it. 3D modeling has offered an effective way of interrogating the nexus of terrain, buildings, and movement that shaped the experience of initiation. Understanding how and why buildings and pathways were deployed across the three converging ravines and two plateaus, and exploring—in

an immersive environment—how pilgrims moved through the terrain cannot lead to the recreation of secret ritual actions, but it can shed light on how the passage of the pilgrim was effectively crafted to heighten the experience. And it has immeasurably improved our ability to communicate such ideas through rendering and animation. For the first three generations of our digital model, we worked in a static (quasi-dynamic) environment (both Lightwave and 3dsMax), from which we produced animations and renderings. We privileged control of the pilgrim’s passage over interactive real-time exploration. We have now moved the virtual landscape into a game engine platform, Unity3D, as well, because of its greater flexibility as an exploratory tool with real-time interactivity (https:// samothrace.emory.edu/visualizing-the-sanctuary/; early discussion, Wescoat 2014a; the first generation of the nonlinear model was created by J.M. Harrington; the second by K. Thayer; the third by M. Musker, V. Baillet, and I. Burr. The Unity 3D model is the work of A. Basu). From the outset, 3D digital modelling has been a particularly effective forensic tool for highlighting shortcomings in our understanding of elevation that were easily elided in the 2D plan. It has caused us to pay closer attention to the governing configuration of the landscape and the profound importance of the natural torrents that frame, bisect, and shape the temenos. Additionally, 3D reconstruction modelling has forced us to confront uncertainty within the archaeological record. Although many of the Samothrakian buildings are well preserved, there are important instances of structures that play an essential role in the sanctuary but in whose reconstructed elevation we are not confident. The most striking example is the building that framed the famous statue of the Winged Victory (Nike). Another is the strategically-placed Ionic Porch attached to the Dedication of Philippos III and Alexandros IV. A third involves the fugitive bridges over which the pilgrims traversed the central ravine that divides the temenos. In offering a whole site reconstruction, how should we handle these buildings, for which the evidence is sufficiently well preserved or the structure is of such importance that work towards a full reconstruction is warranted, but for which key components have not survived? How might we visualise uncertainty and choice in these instances, and to what extent we are obliged to do so? How can we use 3D modelling to reveal rather than mask ambiguity? In this paper, I discuss how we have chosen to approach these different challenges. The work presented spans the last 20 years and therefore bridges our transition from hand-drawn, ink-on-mylar reconstruction drawings to digital media. The decision to express uncertainty

DOCUMENTATION & EVOCATION: THE SANCTUARY OF THE GREAT GODS ON SAMOTHRAKE

307

figure 19.2 Reconstructed plan of the Sanctuary of the Great Gods in the early first century CE. Nos. 1, 2, 3: Unidentified Late Hellenistic buildings; 4: Unfinished Early Hellenistic building (Building A); 6: Milesian Dedication; 7, 8, 10: Dining rooms; 9: Archaistic niche; 11: Stoa; 12: Nike Monument; 13: Theatre; 14: Altar Court; 15: Hieron; 16: Hall of Votive Gifts; 17: Hall of Choral Dancers; 18: Sacred Way; 20: Rotunda of Arsinoe II; 22: Sacristy; 23: Anaktoron; 24: Dedication of Philippos III and Alexandros IV; 25: Theatral Circle; 26: Propylon of Ptolemaios II; 27: South Nekropolis: 28: Doric Rotunda; 29: Neorion; 30: Stepped Retaining Wall; 31: Ionic Porch; 32: Hestiatorion Courtesy American Excavations Samothrace

through hand-drawn options was a considerably greater commitment than it is today, when we have such a vast opportunity to replicate that the challenge now centres on how to control proliferation and focus iteration. 2

Architectural Uncertainty in the Age of Hand Drawing: The Ionic Porch

The Ionic Porch was part of an entrance complex on the Eastern Hill of the sanctuary, which is unequaled in ancient Greek architecture for its position, orientation, and configuration (Wescoat 2017b; earlier: McCredie 1965: 122–24; 1968: 216–34; 1979: 6–8; Wescoat 2012; 2017a). Over its half millennium of use, the region was progressively developed from a simple sunken orchestra to a complex framed by marble buildings dedicated by Hellenistic kings and surrounded by a host of bronze statues that serve as both welcome committee and permanent witness to the rites conducted in this space. It is clearly an important station within the temenos, but—given its exposure— not one in which any of the secret rites may have been conducted; that was reserved for the several buildings sequestered in the valley below. The Ionic Porch, which was added to the back of the Dedication of Philippos III and Alexandros IV in the late 3rd or early 2nd century BCE,

took pride of place on the Sacred Way just as it turned westward to make its steep descent into the heart of the sanctuary. When the region was destroyed by earthquake in the early 2nd century CE, the Samothrakians decided not to rebuild. Instead they salvaged the bronze, pushed the broken blocks into the centre of the theatral space, and covered it all with earth that remained undisturbed until the excavations of James R. McCredie in 1964–1968. The fine Doric hexastyle prostyle pavilion dedicated by Philippos III and Alexandros IV, which had collapsed spectacularly into the Theatral Circle, practically reconstructed itself. However, the small Ionic Porch attached to the back of that building did not fare as well. Its exposed position on the pathway left it vulnerable to post-antique predation. Many marble blocks were broken into small fragments; others were burned for lime. We have the position of the foundations, and very fragmentary remains of an Ionic elevation, including the anta base, column base, capital, two drums from the column shaft, epistlye, dentil-geison block, sima block, antefixes, and, most remarkably, bits of ceiling coffers and splendid floral coffer lids. The remains are clearly worthy of a reconstruction, but several fundamental aspects of the design, including the plan, the height of the elevation, and the shape of the roof, remain uncertain. The lack of any raking elements could

308

figure 19.3

Wescoat

Ionic Porch attached to the west wall of the Dedication of Philippos III and Alexandros IV; elements that do not survive are indicated in grey courtesy American Excavations Samothrace

indicate a hip roof, but we have no positive evidence for it. Lack of identifiable wall blocks suggests a prostyle structure, but again the evidence is negative rather than positive (Fig. 19.3) (for the surviving architectural evidence, see Wescoat & Koch in Wescoat 2017b: 185–254; earlier: McCredie 1968: 229–30; 1979: 6–7). In offering a full reconstruction, we were concerned to be transparent about the state of our knowledge, not only to represent appropriately the partiality of evidence, but also because different choices of architectural form telegraph different associations of function. A shed or hip roof suggests a stoa-like structure with the functions of gathering and shelter; a pitched roof is more suggestive of a naïskos for the display of an image. We created three viable options showing variations of the roof design, elevation, and plan (Fig. 19.4) (Wescoat 2017b: pls. LXXV– LXXX). At the time we produced the finished drawings in 2004, it took days of careful inking for each side and front elevation. The investment of time and energy was considered by some team members to be an extravagance to demonstrate less rather than more knowledge. And

the variations did not entirely satisfy. One of the reviewers for the manuscript of Samothrace 9, The Monuments of the Eastern Hill, accepted the concept of multiple reconstructions, but preferred a different combination of components than the ones we offered—i.e., a pitched roof with prostyle colonnade, which of course is another viable iteration. Working digitally with Rhinoceros or Adobe Illustrator, such renders of this and other viable reconstructions could be generated in a matter of hours rather than painstakingly over days. We debated but rejected the idea of offering a fourth option, not because it was less viable, but because we had met our goal of acknowledging uncertainty and suggesting a range of options with the architectural ideas that they communicate. If we were beginning work on the Ionic Porch now, we likely would consider a computational strategy for exploring and communicating uncertainty, one that would be more comprehensive in its parameters and variations, even if the basic conclusions would not materially differ. Computer-generated iterations, controlled through parametric shape grammars in procedural modelling allow

DOCUMENTATION & EVOCATION: THE SANCTUARY OF THE GREAT GODS ON SAMOTHRAKE

figure 19.4

309

Possible reconstructions of the Ionic Porch courtesy American Excavations Samothrace

for a more intensive and deliberate exploration of uncertainty leading to a fuller range of viable reconstructions (Mamoli 2013; 2015; Mamoli & Knight 2013, with further references). The approach harnesses the power of computation for a rule-based system of design, in generating a series of viable variations within the parameters of the underlying language of architectural shapes. In generating the widest possible range of iterations within the defined grammatical/shape conventions in a functionally neutral environment, shape grammars have the potential to offer new insights by producing variations that we might not have been predisposed to consider. This kind of procedural modelling proves particularly effective for generating variations on a generic class of building with standardised components (e.g., the house, a portico, a city block) that may be used to fill in large areas of an urban model in which generalised architectural hypotheses (ranging from “the archaeological evidence suggests a house” to “houses may have been here”) match the extent of certainty. Procedural modelling has been effectively deployed for building out ancient cityscapes (see Müller et al. 2005; 2006a; 2006b; 2007; Dylla et al. 2008; Watson et al. 2008; Haegler, Müller & Van Gool 2009; Saladana & Johanson 2013; Piccoli 2014; Smelik et al. 2014; Richards-Rissetto & Plessing 2015; Schäfer et al. 2015; Jesus, Coelho & Sousa 2016). More appropriately defined as visualisations rather than reconstructions, these generations create a richer visual image while suggesting variety and possibility. But is it an effective approach to reconstructing a specific building for which we have some components of

both plan and elevation? On the one hand, Greek architecture favours such an approach because its orders are composed of defined elements of certain shape, proportion, and relationship. We tacitly engage in parametric shape grammars at every point in an architectural reconstruction (e.g., the Hellenistic Doric column may be as short as 5.5 or as tall as 7.7 lower-diameters high). However, the computational investment required to generate the fully diverse range of possible reconstructions for a single, partially preserved and fairly idiosyncratic structure such as the Ionic Porch may not currently be time effective (see Konečný, Syllaiou & Liarokapis 2016, for procedural modelling of an Ionic column shaft, where the goal is a product for the gaming industry that will look “realistic”, but which falls short of the more stringent needs of architectural historians). But the tools for this approach exist and likely will shape the way we tackle this kind of problem in the future. For our 3D model, we chose to reconstruct the prostyle design with hip roof for three basic reasons: the proportions of the porch seem more stoa-like than naïskos-like, no statue bases were found in the vicinity, and the hipped roof reconstruction required the least number of missing parts (Figs. 19.5–19.6). When reconstructing the Ionic Porch on paper and thus outside its topographic environment (as in Fig. 19.4), we assumed the roof design had powerful semantic value in understanding the building. However, once we built the Ionic Porch in a 3D environment and modeled the viewscape of a pilgrim moving past the building along the Sacred Way, we realised that

310

figure 19.5

Wescoat

3D model reconstruction of the entrance to the Sanctuary of the Great Gods, from the north, with Propylon of Ptolemaios II on the left, and the Dedication of Philippos III and Alexandros IV and the Ionic Porch on the right courtesy American Excavations Samothrace

figure 19.6 View up the Sacred Way, eastward towards the Ionic Porch courtesy American Excavations Samothrace

DOCUMENTATION & EVOCATION: THE SANCTUARY OF THE GREAT GODS ON SAMOTHRAKE

the roof choice was not as visually significant as we imagined. There was, in fact, no view along the Sacred Way or in the central sanctuary that allowed visitors to see the roof frontally. The only place from which it would be seen in full (although obliquely) was from the western city gate, and possibly from out at sea. A primary goal in developing the 3D model was to understand how and when buildings came within the experience of the viewer during the rites of initiation, which involved entering the sanctuary at night and leaving it during the day. If the Porch was designed as a gathering and viewing place, it would function equally well in both situations, for viewers had a prime place along the processional way. However, as an area of display, there is no question that its greater visual strength of the Ionic Porch came when viewers retraced their steps going up the Sacred Way as they left the sanctuary. From this vantage, the Ionic Porch stood at the summit of the passage, and its splendid ceiling could be enjoyed as the viewer approached (Fig. 19.6). 3

Toggling Options: Design and Visibility of the Nike Monument

As a second case, I take our experience working with the most famous monument in the sanctuary, the Winged Victory or Nike. Despite its importance, the Nike Precinct remains essentially unpublished because of the challenges it poses. The statue and the marble prow on which it stood are strikingly preserved in essentials that capture the drama and beauty of the monument. The surrounding architecture, however, is severely degraded and curiously sub-par by Samothrakian standards (Fig. 19.7) (for the precinct, see Conze, Hauser & Benndorf 1880: 52–53, pl. LX, fig. 1; Kern 1893: 339–43; Lehmann-Hartleben 1940: 352; Lehmann 1973: 182–90; Lehmann 1998: 102–103; Wescoat 2014b; for the discovery, construction, and restoration of the statue, see principally Hamiaux 1998; 2001; 2014a; 2014b). We would expect a statue of such brilliance to be housed in an equally magnificent architectural frame, but it was not. Instead of marble, the structure is composed of a weakly bonded, pebbly, calcareous sandstone not used elsewhere in the sanctuary (the classification of local rocks has been established by William Size, Department of Environmental Sciences, Emory University). Most Samothrakian buildings are quite well preserved, but the Nike Monument is not. The boulder retaining walls framing the precinct are clearly later than and even chinked with small fragments of the building, but no traces of the earlier walls were uncovered in the excavations of the

311

1980s—both the Lehmann and McCredie excavations produced Roman thin-walled ware behind the retaining walls, indicating a terminus post quem of the early 2nd century CE. Nineteenth-century predecessors dug extensively in the region looking for sculpture but made only schematic plans of the relationship of the statue’s base to its architectural frame or the retaining walls that shape the niche (Deville & Coquart 1867: 277 and plan; Conze, Hauser & Benndorf 1880: pl. LX, fig. 1; Champoiseau 1880: 16; Kern 1893: 340; for sketch by Champoiseau, likely in 1863, see Hamiaux 2001: 160, fig. 3; Hamiaux 2014a: fig. 51). By the time the architects for Karl Lehmann, Stuart Shaw and Alec Daykin, completed a detailed, scaled drawing of the precinct in 1952, only part of the krepis survived, to the level of the second step at the back and to the lowest course of the foundation in the front. The blocks supporting the socle for the base of the statue, complete in 1875, had disappeared (the Daykin-Shaw plan [N030] forms the basis for the state plan drawn by John Kurtich [N030AK]: Wescoat 2010: fig. 3.27). The Lehmann team put serious attention into understanding the design of the precinct. Observing channels cut by water in the front of the monument, Karl Lehmann was persuaded that the structure was a fountain; a waterpipe found imbedded in the hillside above seemed to offer confirmation (Lehmann 1973: 180–90, fig. 5). He reconstructed a structure composed of an upper platform supporting the statue and a lower basin of water with great boulders forming rocky shoals. The idea brilliantly conjured an environmental setting for a statue that itself engages the forces of nature in its windswept drapery. Unfortunately, the fountain option quite literally does not hold water. The apparent water channels are in fact postantique water damage. The basin Lehmann envisioned is not lined with hydraulic cement, and the pipe likely supplies a water system directed towards the stoa (for the water channels leading to the Stoa, see McCredie 1965: 113–14, pl. 33). If not a fountain, what was the design of this structure? Was the Nike set in an open peribolos (a revision of the fountain concept), or was she enclosed within a covered naïskos-like structure? The choice has a profound impact on how we may appreciate the statue. Some evidence can be brought to bear on this question, which may, as our investigation progresses, coalesce more strongly around one of the options. Its summary here chiefly demonstrates the fragility of any reconstruction. We can start by eliminating the enormous boulder in the front section of the structure, which played an important role in Lehmann’s fountain, because it lies within the foundations of the krepis and thus need not be taken as a visible feature of

312

figure 19.7

Wescoat

Nike Precinct, from the north courtesy American Excavations Samothrace

the design (see Wescoat 2014b: fig. 160; Kern 1893: 342 suggested the boulders in this region had been dislodged from the west “cyclopean” retaining wall). The poor preservation of the building also finds explanation. Over 500 fragments of the Nike Monument’s soft and easily broken blocks were repurposed to construct a Byzantine building over the Neorion (noted briefly in Wescoat 2013; J. Paga, Z. Dommach, and H. Smagh inventoried the remains in the Byzantine building). Despite the quantity, not one has a distinctive architectural feature associated with a particular architectural order. The lack of any articulated blocks marginally favours a simple peribolos or enclosure wall for the monument. Also in favour of an open monument is the iconography of Nike, who is displayed with her wings still unfurled in flight as she alights on the prow. As an airborne herald, Nike most often appears in open display in sanctuaries, whether as an akroterion or part of a monument (LIMC VI, 1992: 850–904, nos. 16–35, 129–47, 381–82, 580–87, s.v. Nike [A. Moustaka, A. GoulakiVoutira, and U. Grote], esp. 137 [Nike of Paionios] or nos. 381–82 [akroterial nikai from the Samothrakian Hieron]; see Palagia 2010, 162–63, fig. 10.16, who follows Homer Thompson’s suggestion that the figure restored to the centre of the Hieron pediment is also an akroterial Nike, and

also the Nike corner akroterion restored to the Parthenon by Korres 1995: 113, fig. 36). Evidence suggesting the statue stood within a roofed structure centres on its surface condition, which shows very little weathering, particularly in comparison to the akroterial nikai from the Hieron (for the surface condition of the statue, see generally Hamiaux, Laugier & Martinez 2014). The type of marble, length of exposure, micro-climate within the sanctuary, and deposition would affect the degree of surface erosion, but the differences between the two are striking. In addition, a layer of roof tiles found in the 1980 excavation at the southwest corner of the precinct suggests a roof. However, the tile fragments differ in fabric, paint design, and thickness, and thus would be difficult to fit on one roof (work on the tiles has been conducted by A. Jiang, K. Lee, and M. Glennon). The situation raises the possibility that only some of the tiles belong to the Nike Monument; that the tiles belong to structures south and west of the Nike Precinct; or that they belong to a dump of some sort created by our predecessors. A third body of evidence consists of fragments of architectural plaster found within the precinct (Fig. 19.8). Champoiseau, Deville and Coquart, and Kern found remains of red, blue/black and white plaster within the region

DOCUMENTATION & EVOCATION: THE SANCTUARY OF THE GREAT GODS ON SAMOTHRAKE

figure 19.8

313

Plaster fragments from the Nike Monument courtesy American Excavations Samothrace

(see Hamiaux 2001: 205, for the report of Champoiseau’s shipment to the Louvre in 1891, which contained about 60 fragments of blue, red, and white plaster), and coloured plaster was also observed at the site by Deville & Coquart (1867: 277) and by Kern (1893: 342). Similar material was found in the 1950s and 1980s excavations, and fragments continue to appear on the slope of the Theatre in the area of the French dump (this material has been studied by M. Glennon, D. Majarwitz, and L. Neiman). Among the fragments are an ovolo (50.495) and a cavetto (50.395), the latter still attached to the stone it covered. These mouldings could be accommodated either as part of a building, or the crown of a peribolos wall. Potentially more significant are the small coloured fragments that preserve drafted edges. Similar coloured plaster panels imitating drafted margin masonry are indicative of Samothrakian interiors, including the Hieron (Lehmann 1969: Text I, 138–42, 204–12, pls. LVII, XCIII, XCIV, CV, CVI), Stoa (McCredie 1965: 108–10, pl. 31), Neorion (McCredie in Catling 1986–1987, 50–51), and Dining Room A (McCredie 1979: 12–13). The earliest surviving example with plaster interior in imitation of drafted masonry, the Fieldstone Building, has beveled

white panels with a thin painted red frame (see Wescoat 2017b: 66–70, 72–81). One further object, constructed in plaster and imitating a lion’s head water spout (or spouts), remains an enigma (Fig. 19.8). The fragments consist of a lower jaw and tongue with three short framing locks of the mane (one today unglued). Several additional locks may belong to this or another lion’s head (jaw and two locks: 39.1080, a-e; locks: 50.424, 54.104, 54.201, 54.202, 84.120, 2017. 3, 2017.4; this object has studied by M. Glennon and L. Neiman). The back of the jaw has an uneven surface and likely was inset into a socket or hole in the object to which it was attached. The locks, on the other hand, have a flat back surface; they likely were glued in place. Karl Lehmann preliminarily restored the object as one of four ornamental fountainheads (Lehmann 1973: fig. 5). However, although the object approximates a water spout, it clearly never functioned as such: the broad, flat tongue is not shaped to direct water and shows no evidence of water wear, and the material is poorly suited to the task. The restored scale of the head is large in comparison to the marble and terracotta lion’s head waterspouts that decorate Samothrakian

314

figure 19.9

Wescoat

Alternate reconstructions for the Nike Monument: open peribolos or roofed naïskos. 3D model view from the central valley, to the south courtesy American Excavations Samothrace/I. Burr

simas of much larger buildings, even in the position of a false waterspout at the corner of the sima.2 If we abandon the idea of a fountain spout, and the head is overly large for a false corner waterspout on the sima, it is not immediately evident (at least to me) what the lion’s head originally decorated. While new data or strategies could tip the balance, the evidence as we currently understand it is not sufficiently strong to decide between the open precinct or roofed naïskos. And yet in the whole-site model, we cannot leave the Victory stranded. We have again opted to express uncertainty in three-dimensions by offering choices. In this instance, because there is so little evidence, two general options for the design of the Nike Monument suffice: an open structure framed on three sides by a peribolos wall with cavetto crown, and a fully roofed naïskos-like tetrastyle prostyle structure of an Ionic order. For the latter, we worked with the repertoire of Ionic architecture on Samothrake, and basic elements from the Ionic Porch when possible, because it is closest to the Nike Monument in both scale and date (for Samothrakian Ionic, see 2  The approximate width of the plaster lion’s head across the lower locks is ca. 0.19 m; presumably the upper locks gave the head even greater dimension, on a par with the heads from the Rotunda of Arsinoe, Hieron, and Milesian Dedication. By comparison, the restored lion’s head spout on the Ionic Porch is only ca. 0.12 m wide. For large scale Hellenistic architectural elements constructed in plaster, see especially Delos (Marcadé 1952: 109–11, figs. 9–11, with discussion of Egyptian origin for the technique; Ling 1972: 16–18). Lion’s heads imitating waterspouts belong mostly to miniature decorative entablatures, e.g., from the House of Hermes on Delos (Marcadé 1953: 500, fig. 4b, no. B7455; width of the head, 0.065 m).

Wescoat 2017b: 231–42; the model of the Nike Precinct was begun by A. Myer, developed by C. Jordan, and revised by I. Burr). We scaled the proportions to fit the restored width of the stylobate, ca. 7.98 m, and the height of the statue, over 5.57 m (Figs. 19.9–19.10). The Nike stands on the highest ground within the sanctuary, angled towards the major cult buildings in the valley below. Yet the statue is set at the back of a niche, itself cut deeply back into the hillside on lines angled slightly towards the Stoa. A passionately debated issue remains her visibility, not least if the statue had been enclosed within a building. In the 3D model, we can toggle between the open and roofed options from various vantages around the sanctuary. 3D animation is particularly useful for tracing lines of sight while walking towards the monument from the central valley. From this vantage, the statue is clearly visible in an open peribolos, but it is also fairly legible even when set within a roofed naïskos. The statue, open or enclosed, drops out of sight when mounting the steps of the theatre, but then emerges dramatically on the top steps. The 3D visualisation also modified our thinking about the visibility of the statue from the Stoa plateau. The naïskos frame does significantly block the statue from within the Stoa, but, from the terrace, the orientation of the building actually helps focus attention on the statue, particularly from the northern half of the terrace (Fig. 19.10). The building (peribolos or naïskos) subtly draws in viewers from the Stoa terrace, while the statue redirects attention towards the heart of the Sanctuary. We could perhaps have imagined in our mind’s eye the visibility of the statue, but the 3D model secures at least this aspect of the statue within its setting.

DOCUMENTATION & EVOCATION: THE SANCTUARY OF THE GREAT GODS ON SAMOTHRAKE

315

figure 19.10 Alternate reconstructions for the Nike Monument: open peribolos or roofed naïskos. 3D model view from the Stoa terrace, to the south courtesy American Excavations Samothrace

4

Real-time Kinesthetics: Rethinking the Central Ravine

The third and final example involves the use of real-time modelling to understand passage through the temenos (we began our study of the walls in 2016; team members including A. Ward, V. Baillet, and A. Basu are presently engaged in a full study of the central ravine and its relationship to the sanctuary; Y. Poularakis produced the first set of drawings, and J.C. Houseman assisted in recording the remains). The sanctuary is set in rugged terrain spanning two plateaus and a deep central valley. From its inception, the heart of the cult was on the valley floor east of the seasonal torrent that runs through the wooded glen. Early evidence of cult includes the deposit of G2-3 ware in the west chamber of the Hall of Choral Dancers (Lehmann 1982: 267–71, 317–74), and the so-called black pit beneath

the Altar Court (Lehmann & Spittle 1964: 11, 109–10, 168– 89, 225–28). By at least the 4th century BCE, architectural development began on the western bank, and with it the need to control and provide access across the channel of the torrent. For example, retaining walls and foundations that suggest a suite of dining rooms preceding the Hellenistic series labeled L, M, N (no. 7 on Fig. 19.2) and some walls on the Stoa terrace that precede the Stoa may be earlier than the 4th century BCE, but the excavators did not find material to date them. Extensive visible remains of two types of ancient wall, one composed of basalt and trachyte boulders, and the other built of Roman concrete using a wide range of rubble and occasionally recycled blocks, attest to major interventions in the channel, likely in the Hellenistic and Roman periods. All visitors to the sanctuary are tacitly aware of how the central ravine shapes their kinesthetic experience; rather than consider

316 the torrent as a natural hazard to be controlled and essentially eliminated, we aim to understand its role in the construction of cult and sacred experience. Where and how it was bridged, therefore, stands as a key question. Two-dimensional approaches to the issue produced compromised results. In their 1866 plan, Deville and Coquart identified a bridge (their “Pont Génois”) in the remains of retaining walls west of the rotunda of Arsinoe. Deville & Coquart (1867) identified the feature as a bridge, labeled “Pont Génois” on their plan, and the Austrian team (Conze, Hauser & Niemann 1875: 31–32) suggested that, while the natural terrain favoured a bridge at this point, the remains were Late Roman—not Medieval— walls, and that there was no evidence for a bridge. Karl Lehmann focused his research on reconstructing a second crossing joining the Altar Court to the theatre (Lehmann & Spittle 1964: 25–26, frontispiece, figs. 24, 26, pls. I, XL; for documentation of the challenges the excavators experienced in handling the elevations and connections between these monuments, see the different plans offered in Samothrace: A Guide to the Excavations and the Museum in comparison to the site plans in Lehmann 1955; 1960; 1966). James McCredie incorporated both crossings in his 1978 restored plan of the sanctuary, which remained current even through the discovery in the mid 1990s that the Hall of Choral Dancers spanned the entire width of the valley (compare plan IV in Lehmann 1983 to Lehmann 1998). However, our new understanding of the building forced a full reconsideration of how pilgrims moved through the sanctuary. In privileging passage rather than structure, it readily became apparent that in the plan as reconstructed, it was impossible to enter the theatre or reach the Hieron, Altar Court, and Hall of Votive Gifts. Our first reaction was simply to cover over the entire channel, thus creating a broader central valley and easily traversable torrent (see, e.g., Wescoat 2010: fig. 3.3). This solution, however, generates its own problems. First, it assumes that the central torrent has no value to the cult. Second, as made especially clear in 3D modelling, in the descending topography, it is physically impractical if not impossible to convert the entire channelised course into a subterranean channel.3 In 2016, we initiated a program to investigate the central torrent and its retaining walls as a dynamic natural 3  The physical remains of the central channel extend at least ca. 164 m. The ground drops ca. 10 m in elevation from the region south of the Hieron (near the preserved evidence of the boulder retaining wall) to the level of the Anaktoron. Where measurable, the channel is between ca. 1.67 and 3.80 m wide. There is no evidence that the channel was covered by a vault. Moreover, the flooding during torrential storms (witnessed in the present era and demonstrated by the repeated rebuilding of the retaining walls in antiquity) strongly suggest that a fully covered conduit would not be sustainable.

Wescoat

feature, a potential sacred signifier, an architectural monument, and a place of passage. Although known since the 19th century, the extensive ancient (and modern) walls had never been fully recorded. Using photogrammetry, we created 3D digital models of 20 segments of the ravine, which we stitched in Agisoft and orthorectified in 3dsMax. Projecting the retaining walls in plan and elevation, we were able to draw the remains in Adobe Illustrator, and colour-code them according to phases of construction and destruction (Fig. 19.11). We were able to complete the architectural documentation in both two and three dimensions within one season. Working both with a 3D model and in the more traditional drafted medium allowed for a better understanding of the relationship of the banks to each other and the adjacent terrain. (Andrew Ward conducted the archaeological research, Vincent Baillet made the photogrammetric models, Yiannis Poularakis made the architectural drawings, and J. Cody Housman assisted in documentation). Points of vulnerability and change, where the Greek, Roman, and modern walls diverge, were identified. Mapping to the base of the walls revealed the extraordinary measures Samothrakian builders took to grade the steep terrain in the region of the Rotunda and Hall of Choral Dancers. However, despite the fact that sections of the preserved retaining walls rise over 3 m, they do not appear to reach their full height or demonstrate the precise method or position of bridging. At this initial stage in our research, we have not been able to explore the full range of evidence for bridging, which may have differed both over time and across various sections of the channel. The variable width of the channel suggests that some elasticity of construction is likely. Given Samothrake’s ancient forests, a traditional Samothrakian wooden technique for bridging, which consists of planks over cross-beams that rest on a boulder retaining wall, provides one likely form of construction, which is flexible and can be deployed opportunistically (compare, e.g., the 7.5 m long wooden bridge at Kerasia, which consists of planks over crossbeams that rest on a boulder retaining wall). Our new understanding of the topography and monuments in the central valley indicates that pilgrims could only access the key areas of sacred space south of the Hall of Choral Dancers by crossing back and forth over the central ravine. Multiple bridges would have been necessary, but physical evidence for their location is fugitive. However, the dynamics of human movement can also count as a form of evidence against which to test the hypothesis of multiple bridges. Working in the real-time, interactive gaming environment of Unity 3D, we explored the most opportune position for bridges by simulating the kinaesthetics of movement through the virtual site

DOCUMENTATION & EVOCATION: THE SANCTUARY OF THE GREAT GODS ON SAMOTHRAKE

a

317

b

c

figure 19.11a–c Central ravine in the area of the “Pont Génois”: (a) east elevation and plan with phases coloured ; (b) aerial view; (c) photogrammetric model plan: A. Ward and Y. Poularakis; photogrammetric model: V. Baillet; aerial view: M. Page; courtesy American Excavations Samothrace

(Fig. 19.12) (this work is being conducted by Aryabrata Basu). Using a split screen, with the plan above and camera at eye level below, we modelled possible paths from the known base of the Sacred Way to the clusters of monuments in the central valley. We privileged natural movement, convenience, and lines of sight in moving between the small open areas within the dense fabric of the sanctuary. We first looked for the most opportune crossing between the triangular open space west of the Hall of Choral Dancers and Rotunda leading to the small area that opens both to the intermediate terrace with the Neorion and to the Milesian Dedication, which closely matched the area where the most extensive remains of retaining walls survive (the so-called Pont Génois). We then made a series of passages that followed the physically most intuitive points of crossing to reach the Hieron, the Hall of Votive Gifts, and the Altar Court. At a minimum, three crossings are required. However, four to six are necessary to eliminate back-tracking and allow direct access to each building. The scale of the bridges would depend on the traffic (pedestrian, animal, carts). Not all of the crossings would need to be the same width or level of sturdiness. Working

in Unity 3D has the advantage that multiple hypotheses can be tested rapidly through real-time movement. This type of exploration also helps identify where potentially we should focus our field investigation as we develop this project. 5

Concluding Thoughts

In sum, what has 3D digital modelling done for us? Perspective views and physical models have offered ways to build and explore whole site environments for centuries. A digital model’s added value comes with the addition of atmosphere, light, texture, and animation: the ability to move through a model in animation or real time offers an immersive experience that makes a powerfully forensic and communicative tool. Building the model of the Sanctuary of the Great Gods has forced us to confront and embrace the Z (elevation) value as a powerful governing factor in the construction of the sanctuary and its mystery cult. Working in three dimensions has dramatically altered our perceptions of topographic manipulation

318

Wescoat

figure 19.12 Split screen view of Unity 3D interactive scene. Above, the pink line traces in plan the movement of viewer exploring possible points of crossing over the central ravine in the scene below courtesy American Excavations Samothrace/A. Basu

and architectural hierarchy. In plan, the Hall of Choral Dancers covers the greatest area, but in scale the Rotunda of Arsinoe is the more dominant building, communicating at two elevations: its articulated gallery communicating on the higher level of the Stoa and Eastern Hill, and its closed drum working on the valley floor. We have always admired how the great basalt and trachyte boulder retaining walls sculpt the terrain with its own materials. Mapping those textures onto walls reconstructed to full height dramatically emphasises the way the sanctuary cuts into the earth, heightening the bold chthonic overtones that shape the experience of entering the sanctuary. Concerning the human factor, modelling has helped us explore how people can move around, what they could see, and when they could see it. One can more immediately appreciate the density of construction in the Sanctuary of the Great Gods, the complexities of moving from one micro-zone to another, and the physical dominance of the central torrent. If we can move to something so subjective, putting our feet in ancient shoes has allowed us to consider how the spaces within the sanctuary play on our perceptions and emotions. While it is not impossible to

probe all these factors through conventional drawings, 3D digital modelling has made them more palpable. There are liabilities, however, both in communication and effort. To achieve the powerfully evocative effects of 3D immersive modelling, we must—as architects did in the 17th through 19th centuries—complete the scene across the uneven archaeological record. Such reconstructions gloss over rather than distinguish levels of confidence. Rather than make a weakly informed single choice (e.g., the Nike Monument), sketch in missing parts with different opacity or texture (e.g., Ionic Porch), or ignore the problem altogether (e.g., bridging the central ravine), we have chosen iteration, with the notion that in repeating the scene with a range of variations, we may more closely approximate viable solution(s). We cannot control whether viewers or users maintain our commitment to options as an expression of uncertainty or choose to promote one reconstruction over another. But offering the options opens the debate. With regard to effort, we are in our third generation of the 3D digital model of the Sanctuary of the Great Gods. Each earlier version looks primitive by comparison. Our

DOCUMENTATION & EVOCATION: THE SANCTUARY OF THE GREAT GODS ON SAMOTHRAKE

21st century Pixar expectations are foiled by shoestring budgets and small teams. Scholarly efforts at 3D animation for research purposes will inevitably appear amateurish. The learning curve for 3D graphics programs is high; currently, mastery is best suited to scholars whose major research trajectory is the interface of architecture and digital modelling. Communication between the several software platforms preferred by our different specialists (ArcGIS, AutoCAD, Rhinoceros, 3dsMax, and Unity 3D) is possible in theory but by no means seamless in practice. We invest significant energy in making certain that nothing is lost in translation. As a focusing strategy, optimisation—achieving the best solution within the lightest framework—works both computationally and archaeologically (my thanks to Aryabrata Basu for discussions of optimisation). To be time- and research-effective, we need to stay focused on the specific set of research questions for which 3D modelling is the most effective research medium. However, with the widespread and effective use of photogrammetry to record remains three-dimensionally, it is inevitable that 3D digital whole-site reconstruction models will become the desired and expected norm (Remondino & Campana 2014; Sapirstein & Murray 2017). Best strategies to signal the limitations of certainty may take other forms than ours, which is cumbersome in its duplication, if honest in its ambiguity. But while we might agonise over whether an open or roofed structure framed the Nike, it is better that we conjure some suggestion of context than none at all. As architectural archaeologists, we are uniquely positioned to communicate—through our drawings, renderings, and animations—the brilliance of the ancient Mediterranean world to a wide and interested public across the globe. Images are efficient and evocative, and so we must continue to invest our time, energy, and creativity into making them meet the visual expectations of the 21st century, while embracing the fragmentary nature of the evidence and our uncertainties of interpretation. List of References Blakely, S., 2007: “Kadmos, Jason, and the Great Gods of Samothrace: Initiation as Mediation in a Northern Aegean Context”, Electronic Antiquity 11.1: 67–95. Blakely, S., 2012: “Toward an Archaeology of Secrecy: Power, Paradox, and the Great Gods of Samothrace”, Archeological Papers of the American Anthropological Association 21: 49–71. Bowden, H., 2010: Mystery Cults of the Ancient World (Princeton). Bremmer, J.N., 2014: Initiation into the Mysteries of the Ancient World. Münchner Vorlesungen Zu Antiken Welten 1 (Berlin).

319

Burkert, W., 1993: “Concordia Discours: The Literary and the Archaeological Evidence on the Sanctuary of Samothrace”, in Marinatos, N. & Hägg, R. (edd.), Greek Sanctuaries, New Approaches (London) 178–91. Catling, H.W., 1986–1987: “Archaeology in Greece 1986–1987”, AR 33: 3–61. Champoiseau, C., 1880: “La Victoire de Samothrace”, RA 39: 11–17. Clinton, K., 2003: “Stages of Initiation in the Eleusinian and Samothracian Mysteries”, in Cosmopoulos, M. (ed.), Greek Mysteries: The Archaeology of Ancient Greek Secret Cults (London) 50–78. Cole, S.G., 1984: Theoi Megaloi: The Cult of the Great Gods at Samothrace (Leiden). Conze, A., Hauser, A. & Benndorf, O., 1880: Neue archäologische Untersuchungen auf Samothrake (Vienna). Conze, A., Hauser, A. & Niemann, G., 1875: Archäologische Untersuchungen auf Samothrake (Vienna). Cruccas, E., 2014: Gli dei senza nome: sincretismi, ritualita e iconografia dei Cabiri e dei Grandi Dei tra Grecia e Asia minore. Tübinger Archäologische Forschungen 13 (Rahden). Denard, H., 2012: “A New Introduction to the London Charter”, in Bentkowska-Kafel, A., Baker, D. & Denard, H. (edd.), Paradata and Transparency in Virtual Heritage Digital Research in the Arts and Humanities Series (Farnham) 57–71. Deville, G. & Coquart, E., 1867: “Rapport sur une Mission dans l’île de Samothrace”, Archives des Missions Scientifiques et Littéraires, 2nd series, vol. 4: 254–78. Dimitrova, N., 2008: Theoroi and Initiates in Samothrace. The Epigraphic Evidence. Hesperia Suppl. 37 (Princeton). Dylla, K., Frischer, B., Müller, P., Ulmer, A. & Haegler, S., 2008: “Rome Reborn 2.0: A Case Study of Virtual City Reconstruction Using Procedural Modeling Techniques”, Computer Graphics World 16.6: 62–66. Haegler, S., Müller, P. & Van Gool, L., 2009: “Procedural Modeling for Digital Cultural Heritage”, EURASIP Journal on Image and Video Processing 2009.1: 852392. Hamiaux, M., 1998: “La Victoire de Samothrace. Mode d’assemblage de la Statue”, CRAI 1998: 365–76. Hamiaux, M., 2001: “La Victoire de Samothrace. Découverte et Restauration”, JSav 2001: 152–223. Hamiaux, M., 2014a: “The Description and Construction of the Monument”, in Hamiaux et al. 2014, 142–63. Hamiaux, M., 2014b: “The Winged Victory of Samothrace: Discovery and Restoration”, in Hamiaux et al. 2014, 72–89. Hamiaux, M., Laugier, L. & Martinez, J.-L. (edd.), 2014: The Victory of Samothrace: Rediscovering a Masterpiece (Paris). Hemberg, B., 1950: Die Kabiren (Uppsala). Innova/Virtual Archaeology Spanish Society (ed.), 2010: “International Principles of Virtual Archaeology. The Seville Principles/Virtual Archaeology International Network.

320 Innova/Virtual Archaeology Spanish Society”, accessed March 7, 2018 (http://smartheritage.com/actuality). Jesus, D., Coelho, A. & Sousa, A.A., 2016: “Layered Shape Grammars for Procedural Modelling of Buildings”, The Visual Computer: International Journal of Computer Graphics 32.6– 8: 933–43. Kern, O., 1893: “Aus Samothrake”, AM 18: 337–84. Konečný, R., Syllaiou, S. & Liarokapis, F., 2016: “Procedural Modeling in Archaeology: Approximating Ionic Style Columns for Games”, in Evans, A., Liarokapis, F. & Pivec, M. (edd.), 2016 8th International Conference on Games and Virtual Worlds for Serious Applications (VS-GAMES ), 8 pp. Korres, M., 1995: From Pentelicon to the Parthenon: The Ancient Quarries and the Story of a Half-Worked Column Capital of the First Marble Parthenon (Athens). Lehmann, K., 1955: Samothrace: A Guide to the Excavations and the Museum (New York). Lehmann, K., 1960: Samothrace: A Guide to the Excavations and the Museum. 2nd rev. ed (Locust Valley). Lehmann, K., 1966: Samothrace: A Guide to the Excavations and the Museum. 3rd rev. ed (Locust Valley). Lehmann, K., 1973: “The Ship-Fountain from the Victory of Samothrace to the Galera”, in Lehmann, P.W. & Lehmann, K. (edd.), Samothracian Reflections. Aspects of the Revival of the Antique (Princeton) 181–259. Lehmann, K., 1983: Samothrace: A Guide to the Excavations and the Museum. 5th rev. ed. (New York). Lehmann, K., 1998: Samothrace: A Guide to the Excavations and the Museum. 6th ed. revised and enlarged by J.R. McCredie (Thessaloniki). Lehmann, K. & Spittle, D., 1964: The Altar Court. Samothrace 4.II (New York). Lehmann, P.W., 1969: The Hieron. Samothrace 3 (Princeton). Lehmann, P.W., 1982: The Temenos. Samothrace 5 (Princeton). Lehmann-Hartleben, K., 1940: “Preliminary Report on the Second Campaign of Excavation in Samothrace”, AJA 44.4: 328–58. Ling, R., 1972: “Stucco Decoration in Pre-Augustan Italy”, BSR 40: 11–57. London Charter (ed.), 2009: “London Charter for the ComputerBased Visualisation of Cultural Heritage, 2.1”, accessed March 7, 2018 (http://www.londoncharter.org/). Mamoli, M. & Knight, T., 2013: “Reconstructing Fragments: Shape Grammars and Archaeological Research”, in Earl, G., Sly, T., Chrysanthi, A., Murrieta-Flores, P., Papadopoulos, C., Romanowska, I. & Wheatley, D. (edd.), Archaeology in the Digital Era. e-Papers from the 40th Conference on Computer Applications and Quantitative Methods in Archaeology: Southampton, 26–30 March 2012. CAA 2012 (Amsterdam) 888–96. Mamoli, M., 2013: Towards of a Theory of Reconstructing Ancient Libraries (diss., Georgia Institute of Technology).

Wescoat Mamoli, M., 2015: “Library Grammar: A Shape Grammar for the Reconstruction of Fragmentary Ancient Greek and Roman Libraries”, in Martens, B., Wurzer, G., Grasl, T., Lorenz, W.E. & Schaffranek, R. (edd.), ECAADe 2015 Real Time—Extending the Reach of Computation. Proceedings of the 33rd International Conference on Education and Research in Computer Aided Architectural Design in Europe, 1 (Vienna) 463–470. Marcadé, J., 1952: “A Propos des Statuettes Hellénistique d’en Aragonite du Musée de Délos”, BCH 76: 95–135. Marcadé, J., 1953: “Les trouvailles de la maison dite de l’Hèrmes, à Délos”, BCH 77: 497–615. McCredie, J.R., 1965: “Samothrace: Preliminary Report on the Campaigns of 1962–1964”, Hesperia 34.2: 100–24. McCredie, J.R., 1968: “Samothrace: Preliminary Report on the Campaigns of 1965–1967”, Hesperia 37.2: 200–34. McCredie, J.R., 1979: “Samothrace: Supplementary Investigations, 1968–1977”, Hesperia 48.1: 1–44. Müller, P., Vereenooghe, T., Ulmer, A. & Van Gool, L., 2005: “Automatic Reconstruction of Roman Housing Architecture”, in Baltsavias, M., Guen, A., van Gool, L. & Pateraki, M. (edd.), Recording, Modeling and Visualization of Cultural Heritage: Proceedings of the International Workshop, Centro Stefano Franscini, Monte Verita, Ascona, Switzerland, May 22–27, 2005 (London) 287–97. Müller, P., Vereenooghe, T., Wonka, P., Paap, I. & Van Gool, L.J., 2006a: “Procedural 3D Reconstruction of Puuc Buildings in Xkipché”, in Ioannides, M., Arnold, D., Niccolucci, F. & Mania, K. (edd.), The 7th International Symposium on Virtual Reality, Archaeology and Cultural Heritage. VAST 2006 (Geneva) 139–46. Müller, P., Wonka, P., Haegler, S., Ulmer, A. & Van Gool, L., 2006b: “Procedural Modeling of Buildings”, ACM Transactions on Graphics 25.3: 614–23. Müller, P., Zeng, G., Wonka, P., & Van Gool, L. 2007. “ImageBased Procedural Modeling of Facades”, ACM Transactions on Graphics 26.3: article no. 85, 9 pp. Nielsen, I., 2014: Housing the Chosen: The Architectural Context of Mystery Groups and Religious Associations in the Ancient World (Turnhout). Palagia, O., 2010: “The Victory of Samothrace and the Aftermath of the Battle of Pydna”, in Palagia, O. & Wescoat, B.D. (edd.), Samothracian Connections: Essays in Honor of James R. McCredie (Oxford) 154–64. Piccoli, C., 2014: “3D Reconstruction Techniques as Research Tools in Archaeology: The Case Study of Koroneia, Greece”, Tijdschrift Voor Mediterrane Archeologie 52: 1–6. Remondino, F. & Campana, S., 2014: 3D Recording and Modelling in Archaeology and Cultural Heritage: Theory and Best Practices. BAR-IS 2598 (Oxford). Richards-Rissetto, H. & Plessing, R., 2015: “Procedural Modeling for Ancient Maya Cityscapes: Initial Methodological Challenges and Solutions”, in the Institute of Electrical and

DOCUMENTATION & EVOCATION: THE SANCTUARY OF THE GREAT GODS ON SAMOTHRAKE Electronic Engineers (ed.), 2015 Digital Heritage: Proceedings of a Meeting held 28 September–2 October 2015, Granada, Spain (Red Hook) 85–8. Saldana, M. & Johanson, C., 2013: “Procedural Modeling for Rapid-Prototyping of Multiple Building Phases”, ISPRS— International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences 5 (Feb.): 205–10. Sapirstein, P. & Murray, S., 2017: “Establishing Best Practices for Photogrammetric Recording during Archaeological Fieldwork”, JFA 42.4: 337–50. Schäfer, A., Bock, H., Sanday, J. & Leitte, H., 2015: “Virtually Reassembling Angkor-Style Khmer Temples”, Digital Applications in Archaeology and Cultural Heritage 2 (Jan.): 2–11. Smelik, R.M., Tutenel, T., Bidarra, R. & Benes, B., 2014: “A Survey on Procedural Modelling for Virtual Worlds”, Computer Graphics Forum 30.6: 31–50. Watson, B., Müller, P., Veryovka, O., Fuller, A., Wonka, P. & Sexton, C., 2008: “Procedural Urban Modeling in Practice”, IEEE Computer Graphics and Applications 28.3: 18–26. Wescoat, B.D., 2010: “James R. McCredie and Samothracian Architecture”, in Palagia, O. & Wescoat, B.D. (edd.), Samo­ thracian Connections: Essays in Honor of James R. McCredie (Oxford) 5–32.

321

Wescoat, B.D., 2012: “Coming and Going in the Sanctuary of the Great Gods, Samothrace”, in Wescoat, B.D. & Ousterhout, R.G. (edd.), Architecture of the Sacred; Ritual, Space, and Experience from Classical Greece to Byzantium (Cambridge) 66–113. Wescoat, B.D., 2013: “Samothrace, Greece”, IFA Archaeology Journal 2: 4. Wescoat, B.D., 2014a: “Digital Modeling in the Sanctuary of the Great Gods on Samothrace”, in Michela Rosso (ed.), Investigating and Writing Architectural History: Subjects, Methodologies and Frontiers Papers from the Third EAHN International Meeting (Turin) 607–13. Wescoat, B.D., 2014b: “Thoughts on the Design of the Nike Precinct”, in Hamiaux, M., Laugier, L. & Martinez, J.-L. (edd.), The Victory of Samothrace: Rediscovering a Masterpiece (Paris) 174–79. Wescoat, B.D., 2017a: “The Pilgrim’s Passage into the Sanctuary of the Great Gods, Samothrace”, in Kristensen, T.M. & Friese, W. (edd.), Excavating Pilgrimage: Archaeological Approaches to Sacred Travel and Movement in the Ancient World (London) 67–86. Wescoat, B.D., 2017b: Monuments of the Eastern Hill. Samothrace 9. 2 vols. (Princeton).

Index Ancient literary and epigraphical sources, commentary on Apollonios of Rhodos 232 Aristophanes 279, 283 Aristoteles 205, 274, 282 Arkesios 181–2, 184 Building accounts, Athenai 35 Building accounts, Epidauros 170 Building epigrams, Post-Herulian wall 222–3 Diodoros Sikeliotes 172–3 Herodotos 170, 175, 207 Pausanias 76, 83, 106, 107, 168–70 Polyainos 171 n. 1 Polybios 81, 83 Rhetoric, writings about 274, 276, 283, 285–7 Theophrastos 108 Thoukydides 158, 171, 173, 216 Xenophon 171, 173–4 Vitruvius 26, 31, 33, 180–2, 184–5, 189, 193–4, 232, 237, 275, 281–2, 284, 287 Builders, architects, patronage 9, 36–7, 125, 152, 175 Architect, role of 31, 36, 53–4, 170–1, 174, 180, 231, 237 Financing and patronage 9, 35, 168–75 As a result of military conflict, booty 170–5 Patrons, communicative intentions 200, 207–8, 211–2, 220, 222–4 Individual architects Cossutius 231 Hermogenes 237 Iktinos 26, 170–1 Individuation by technique 26 Organisation of crews, work 35–6, 51–4 Construction techniques, execution 9, 21 anathyrosis 26, 43–4 as Behavioural Qualities 153–4 Carpentry 111–3 chaîne opératoire 151–2, 154 Entasis, implementation of 31–3 Finish, degree of 139, 202, 205, 207–8, 211 apergon (protective mantle) 48 Unfinished workmanship 229, 231–7 Foundations, preparation of 78–9, 138 Handling and lifting devices Boss 49, 229–32 Crescent-shaped slot 107–9 U-shaped channel 93 Irregularities, interpretation of 22–4, 27–31, 32–4, 39–41, 46–7, 67, 80, 229 Joinery Clamp cuttings, metal fasteners 40, 46, 108, 110 Effect on stability of an ancient construction 144–6 empolion, dowels 26 Lifting techniques, cranes 108–9 Lathe or tornos 112 Masonry “fabric” 200, 202, 205–9 classification of 126–8, Fig. 7.3 Materials

Glue (woodwork) 112–3 Mud-brick 79, 153 Plaster 312–4 Wood 106–15 Polychromy 207, 301 Precision of ancient artisans 30 Design, planning of ancient buildings Acoustics 274–6, 282–7 Design 4, 9, 202, 264 Alignments in plan, rules of 24–5 Changes during construction 24–6, 36, 41–4, 47–53, 71 Direction of view, perception 231, 237, 264, 266–70 Setting lines 58–9 Units, spacing of modular elements 24, 32, 49–51, 79–81, 106, 115–7 Drawings and other building specifications 54, 56–71 As blueprints 56–8 Early modern renderings of ancient monuments 179 Egyptian 58 Materials used, pigments 57, 61 Geometry / mathematics of design 31–2, 56, 65, 71 Proportion 65, 67, 80, 106 Refinements 26, 31–7, 202 Style and change Apsidal plans, frequency in Early Iron Age 106, 158, 162–4 Cross-pollination among different media 179–95 Doric style, origins of 109–15, 185–6, 190–2 External influences, stylistic exchange 101–2 Ionic capital, influences on 186–90 Korinthian capital, origins of 181–5 Non-peripteral plans 295 Prostyle plan 78, 83 Transmission from “minor” arts 180, 194–5 Elements of buildings Bench, profile of 65–7 Brackets 112–3 Colonnades 81, 139 Peristyle 106 Rhodian peristyle 232–3 Stylobate 22–4, 138–40 Capital 26, 111–5, 181–90 Cruciform, hypothetical 188–9 Echinos 111–4, 184–90 Volute 193 Column Base 110–1, 114 Drum 22–6, 141–2 Fluting, “elbows” 32–3, 35, 234 Inclination of 33–5 Shaft 108–9 Taper 142 Entablature Architrave 46–7, 142, 232 Backers (interior blocks) 40–4, 142 Geison 40, 47–8, 81, 93–7, 101–3, 109, 142, 194 thranos 41–4, 46

324 Elements of buildings (cont.) Floor Mosaic 78–9, 84 Frieze 40–7, 81, 190–2 Field-and-divider 190 Ionic (Parthenon) 25, 46 Kentauromakhia (Parthenon) 46 Metope 46 Triglyph 142, 190–4, 301 Foundations 78, 144 Lintel 208 Non-architectural comparanda kalathos 181–5 lekythos 184 phiale 184–90, 194 Tripod 190–4 Ornament Console 208–11 Karyatid 186–8 Lion head 208–9, 313–4 Mouldings 101–4, 313 Lotus motif 186 Pediment 39, 49–51, 53, 59, 61, 65, 109–10, 115–7 Roof Akroterion 52, 312 Antefix or eaves pan tile 50, 115–7, Fig. 6.9 Arch, vault 65 Coffer 52 Protomes 109 Sima, raking 48–50, 94 Threshold 140–1 Tiles, regular 49–52, 312 Tympanon 109–10 Woodwork 51–2 Wall Anta capital 139 Party walls 139 Retaining 67–8 Socle 78–9 Methodology of and approaches to the analysis of Greek architecture Aesthetics (of masonry) 200, 202, 205–9, 211, 220–1, 229 asperitas, relationship to rhetorical effect 237 Rustication 229 Analogy to early modern or contemporary practice 59–61, 65, 111–2, 163–4, 224, 269 Archaeological experiments, replication 35–6 Cultural and social history (architecture as) 151–64 Behavioural Qualities 153–4 Economics of temple building 168–75 Social group, identity 151–4, 160–4 Energetics, determining labour investment 4, 35–6, 52, 168–70, Table 10.1 Formal analysis 111–15, 181–5, 190–2, 194 Functional analysis, multi-functionality 10, 83–5, 199–200, 211, 259–60, 263, 266, 298 Oratory and building acoustics 274–87 Interpretation, meaning of 184–93, 199–212, 230–2, 243 Social memory, memorialisation 216, 218–9, 223–5 “Upcycling”, meaningful reuse 206–7, 215–25

Index Life history of architecture 5–6, 81–4, 91–3, 97, 104, 125, 132, 215, 295–7 Maintenance, repairs, renovation 94, 125–6, 138, 215, 225 Rebuilding 79, 82 Recycling 215 Removal, disassembly 97–9, 109 Reuse (also see Methodology–Interpretation–Upcycling) 98– 101, 215, 217–25 “Reverse architectural stratigraphy” 101 spolia, commentary on term 215, 220 New directions in research on Greek architecture 2–4, 9–11, 178–81, 194–5, 199, 211–2 Petrification, theory of 109–15 Phasing, determination of 78–9, 81–3, 128–9, 172 Relative sequence of construction 35–6, 40–4, 44–54, 125–7, 138, 140–1, 229, 300 Phenomenology via simulation 243–4, 246, 249 Humanisation of digital reconstructions 245, 249–50, 258 Perception of simulations, viewshed analysis 249, 274–87, 295, 309–11 Processional architecture, performance 249–54, 298–300 “Spatial turn”, “mobility turn” 249 Practising architects, perspective on Greek architecture 179 da Sangallo the younger, Antonio 179 Reconstruction of architecture 3D/CAD modelling 258, 260–7, 270, 275, 279, 292, 309–11, 314–9, Figs. 2.2, 2.4–5, 2.7–8, 2.12, 2.14, 2.15, 8.4, 8.6–10, 11.7–8, 15.2, 16.3–4, 17.3, 17.5, 17.7, 18.3–5, 18.8–11, 19.1, 19.5–6, 19.9–10 4D navigation of digital models, interaction 129, 246, 249–54, 264–70, 315–9 anastylosis, physical reconstruction, restoration 32, 35–6, 137–42 Crowdsourcing, gaming as research tools 266–9 Digital heuristics 264, 275–6, 305 Digital surrogate 265 Hypothesis testing, as a method for 265–6, 270 Iterative illustrations, in light of uncertainty 113–7, 265–6, 306, 308–9, 314, 318, Fig. 6.7–8, 6.10, 19.3–4, 19.9 Non-photorealistic rendering 263 Procedural modeling, shape grammar 308–9 Scale, experience of 243, 245–6 Unity 3D platform, utilization of 267–9, 306, 317–9 VR (Virtual Reality), simulations 8, 138, 142–47, 243, 245–8, 276–82, 287, 292–5, 302–3, 306 Recording, illustration methods Data management, presentation 129, 132, 243–5, 294–5 Efficiency of traditional and digital methods 8, 129, 132, 308 GIS (Geographic Information Systems) 124–5, 243–5, 260 Identifying faint markings 63–4 Line drawings, stone-by-stone illustration 77, 123–5, 179, 229, 292, 308 Measuring tools, techniques 29, 77, 260, 292, 294, 302 Photogrammetric recording 77, 129, 137, 258, 260, 292, 302, 316, Figs. 2.3, 2.8, 2.10–11, 7.5, 8.3, 16.2, 18.2, 18.6, 19.11c Plans, theoretical implications of 124 RTI (Reflectance Transformation Imaging) 63–4 Scope/meaning of Greek architecture 1, 36–7 Site management 146–7, 297 Circulation patterns 260, 263–4 Contextual analysis 243, 245–6, 254 Gamma Analysis 264, 266, 268

325

Index Landscape and architecture 129, 131–2, 200, 202, 245–6, 254, 306, 315–7 Space, spatial organization 153–4 Urban development, change over time 74, 85, 124, 126–9, 200 Statistical methods 160–3 Correspondence analysis 153–7 Dendrogram, pair-group hierarchical cluster analysis 156 k-means tests 156–7 Normalisation 153–5 Structural analysis, numerical modeling 142–47, Figs. 8.8–9, 17.8–10 Acoustic modeling techniques 275–6, 282–7 Discrete element method, for seismic analysis 143 Historical earthquake records 144 Insight on ancient architectural practice 145–7 Surface markings, analysis of (also see Building practice—Drawings) Ground settling, preservation 295–7 Setting lines 22 Tools and toolmarks 22, 94 Weathering patterns 21–2, 30–1 Symbol theory, semiotics 199–200 allusion, meaning 206–7 Application to architecture 200–7 Baumberger, Christoph 199, 202–7 denotation, meaning 203–4 exemplification, meaning 205 Peirce, Charles Sanders 199 representative, definition 200 Taphonomy, architectural preservation 106, 109–11 Effect of burning, on limestone 109 Theorists of Greek architecture (early modern) Bötticher, Karl 180–1, 188, 193–4 Dörpfeld, Wilhelm 30, 107, 111 Durm, Josef 111 Jones, Owen 180 Kawerau, Georg 112 Semper, Gottfried 180–1, 188, 193–4 Schinkel, Karl Friedrich 181 Wagner, Otto 180 Typological study 2–3, 10, 61, 83–4, 179, 199–213 Units, ancient feet and other standards 24, 79–81, Fig. 4.5 Periods Archaic 93–104, 106–17, 135–42, 185–93, 207, 278, 291, 295–300 Classical 21–37, 39–54, 59, 98–104, 168–75, 182–5, 202, 207–11, 216–9, 246, 276–9, 291 Early Iron Age, Protogeometric 151, 158, 161–4 Nomadic groups, impact on architecture 163–4 Hellenistic 59, 65–6, 74–9, 83–5, 202–3, 211, 230–7, 291, 301, 307–17 Prehistoric 2–3, 289, 291 Late Helladic IIIB-C 151, 157–8, 160–2 Late Minoan I 124, 127 Late Minoan IA 129 Late Minoan IB 126–7, 260 Middle Minoan II 127, 129  Middle Minoan III 124, 127 Minoan 123–32, 185, 193, 258–70 Mykenaian 160, 182, 185–7, 192–3

Neopalatial 125, 127 Protopalatial 124, 129 Roman 65, 82–4, 183–4, 219–24, 280–2 Sites and monuments Aigeira Buildings D–F (“Naïskoi”) 74–85 Relationship with the Akhaian League 76, 81, 85 Theatre 76, 81, 84–5 Amphipolis 207–8 Áno Mazaráki 110 Asine 162–3 Athenai, Akropolis Erektheion 35, 187 History of excavation 91–2 Ionic structure 101–4 North Akropolis wall 206–7, 216–20, 223–4, Figs. 13.1, 13.4 Parthenon, Classical 21–37, 39–54 Parthenon, Older 218, 224, Fig. 13.1 Persian destruction 216–9 Propylaia (Mnesiklean) 98–9, 101 Small limestone buildings 91–104 Temple of Athena Polias 218, 225, Fig. 13.1 Athenai, City Agora, Old Bouleuterion 278 Agora, New Bouleuterion 278–80, 282–4 Dipylon 200  Herulian sack 219–23, 225 Post-Herulian wall 219–25, Figs. 13.2–4 Sacred Gate 202 Themistoklean wall 206–7, 216–9 Tomb of the Unknown Soldier 224, Fig. 13.5 Bassai, Temple of Apollon 170–74, 182 Delos, House of the Masks 232–3 Delphoi, Knidian Treasury 186 Dhespótiko, Mándra 135 History of the sanctuary 135–7 Temple-Hestiatorion complex 137–47 Eleusis 200–2 Eleutherai 209–10 Epidauros, Tholos 182 Eretria 202 Goúrnia 123–32 Cemeteries 124, 131 Palace 123–4, 127, 129  Herakleia upon Latmos 202–3, 211 Kalapódhi South Temple (Late Geometric) 106 Persian destruction 106, 109 South Temple (Archaic) 106–17 Kalýva 203 Karasis, Mt. 211  Kolkhis, Palace of king Aeëtes (imagined) 232 Korinthos, Sanctuary of Demeter and Kore 246–8 Kynos 162–4 Labraunda, Sanctuary of Zeus 250–2 Laris upon Hermos, city wall 202, 207 Larisa, Theatre 67 Lepreon Temple of Demeter 169, 174–5 relationship with Phigaleia 173–4

326 Sites and monuments (cont.) Levkandí 162–3 Magnesia, Temple of Artemis 237 Mákiston, Temple of Athena 169, 174 Messene, Fortifications 200–2, 205, 208–9 Miletos Bouleuterion 65–6, 230–2 Gymnasium 233 Kalabaktepe, Courthouse by the temple of Athena 233 South Agora 234 Theatre 65 Mitrou 162–3 Nea Paphos, “Tombs of the Kings” 233–4 Nikhoria 162–4 Olbia (Provence) 203–4 Old Smyrna 207 Olympia Elis, patronage by 168–9, 175 Metroön 169, 174–5 Temple of Hera 107–8 Temple of Zeus 168–70 Paphos 207 Pednelissos 202–3 Phigalia 171–4 Prasidháki, Temple of Athena 169, 174 Priene Agora 234–7 Theatre 67 Pseira, House of the Rhyta 258–70 Rhodos 200, 205, 211

Index Roma Curia Iulia 281–2, 284–7 Pantheon 179 Samothrake, Sanctuary of the Great Gods 305–19 Central Ravine 315–8 Dedication of Philippos III and Alexandros IV, Ionic Porch 307–11 Nike Monument 311–4 Selinous, Acropolis 289–303 Battery complex at north side 211 South Building 297–300 Temple B, triglyph altar 300–1 Temple C 301–2 Temple R 295–7 Sillyon 203, 211 Stageira, City wall 205–7 Syrakousai, Fort of Euryalos 208–9 Thasos, Gate of Zeus and Hera 202–4, 211 Thermos 162–4 Tiryns 161 Torybaia 202, 207 Types of buildings Dining hall 84 Domestic architecture 125–6, 151–64 Combined with cult function 259–60 Fortification 4, 199–212, 216–25 Meeting hall, Bouleuterion 274–87 Naïskos 74, 83–5 Steps, monumental 250–4 Temple 83, 106 Treasury 83–4