SOMA 2004: Symposium on Mediterranean Archaeology. Proceedings of the eighth annual meeting of postgraduate researchers, School of Classics, Trinity College Dublin. 20-22 February 2004 9781841719474, 9781407329680

Table of contents :
Front Cover
Title Page
Copyright
Contents
Preface
List of Contributors
Acknowledgements
Italian Wine between the Late Republic and the First Century AD: Analysis of Production and Trade through the Principal Types of Amphora
The Necropolis of Castelluzzo at Militello in the Catania Valley (Italy)
The Maritime Villae of Antium: A Case Study in Coastal Settlement
Origins of the Romano-African House
Homesickness and Greek Architecture in a Colony: The example of Cyrene (Libya)
Making ‘Scents’ of Flowery Pots. Considering Minoan Ceramic Vessels with Botanical Relief
Beyond the Consumer City; the Case of Ancient Antioch
The Site of Iuvanum (Abruzzi, Italy) in the Classical period; Its Public and Private Buildings: A Preliminary Study of a Taverna
Central Adriatic Sea: real ‘importuosa litora’?
Understanding the Limits of Pollution: The Use of the Dead
Monkeying Around the Mediterranean: A Fresh Perspective on Ancient Primates
Special Sites in the Cretan Neolithic Landscape
From Antiquity To Feudalism At The Southern Border Of Navarra. The Design of a Study in Landscape Archaeology
The Urban Development of Leptis Magna in the Late First Century BC and Early First Century AD
Eteocretans in The Bronze Age? The Far East of Crete During The LM II-III Periods (1450-1100 BC)
Reconstructing Minoan Monumentality
Terracotta Human Figures and Figurines from the LBA Cyclades (The Cases of Ayia Irini and Phylakopi): Art, Religion, and Politics
The Myth of the Synagogue on Delos
A ‘Hall of the Muses’ at Cyrene (Libya)?
Disentangling Material Cultures:Late Roman and Sasanian Facet Cut Glassware in Late Antiquity
Cities Under Siege? A Look at Bronze Age Iconography
Symbols of Superiority: Western Greek Votive Offerings at Delphi
Sacred or Profane: Swallow-Painted Nippled Ewers from Akrotiri
Is There ONE Crusader Archaeology?
Exploring the Divine Cow in Ancient Egypt
The Uses of Bow Imagery in a Greek Context: The Enemy Within
Altars and Cult Installations of Punic Tradition in North Africa
Sun, Sea and Archaeological Sites: Archaeology and the State – A View from Cyprus
Body, Senses and Space in Late Bronze Age Central Macedonia, Greece
A New Look at Minoan Metalworking Techniques
Debates and Trends on Heritage Management and Architectural Conservation in the Mediterranean Region: The Issue of the Driving Forces Behind Conservation Planning

Citation preview

BAR S1514 2006

SOMA 2004 Symposium on Mediterranean Archaeology

DAY ET AL (Eds)

Edited by

Jo Day, Cybelle Greenlaw, Heinrich Hall, Amanda Kelly, Lidia Matassa, Killian McAleese, Emma Saunders and Deirdre Stritch

SOMA 2004

B A R

BAR International Series 1514 2006

SOMA 2004 Symposium on Mediterranean Archaeology Proceedings of the eighth annual meeting of postgraduate researchers School of Classics, Trinity College Dublin. 20-22 February 2004

Edited by

Jo Day, Cybelle Greenlaw, Heinrich Hall, Amanda Kelly, Lidia Matassa, Killian McAleese, Emma Saunders and Deirdre Stritch

BAR International Series 1514 2006

Published in 2016 by BAR Publishing, Oxford BAR International Series 1514 SOMA 2004 © The editors and contributors severally and the Publisher 2006 The authors' moral rights under the 1988 UK Copyright, Designs and Patents Act are hereby expressly asserted. All rights reserved. No part of this work may be copied, reproduced, stored, sold, distributed, scanned, saved in any form of digital format or transmitted in any form digitally, without the written permission of the Publisher.

ISBN 9781841719474 paperback ISBN 9781407329680 e-format DOI https://doi.org/10.30861/9781841719474 A catalogue record for this book is available from the British Library BAR Publishing is the trading name of British Archaeological Reports (Oxford) Ltd. British Archaeological Reports was first incorporated in 1974 to publish the BAR Series, International and British. In 1992 Hadrian Books Ltd became part of the BAR group. This volume was originally published by Archaeopress in conjunction with British Archaeological Reports (Oxford) Ltd / Hadrian Books Ltd, the Series principal publisher, in 2006. This present volume is published by BAR Publishing, 2016.

BAR PUBLISHING BAR titles are available from:

EMAIL P HONE F AX

BAR Publishing 122 Banbury Rd, Oxford, OX2 7BP, UK [email protected] +44 (0)1865 310431 +44 (0)1865 316916 www.barpublishing.com

Contents Preface

iii

List of Contributors to SOMA 2004

v

Acknowledgements

vii

Francesco Aliperti Italian Wine between the Late Republic and the First Century AD: Analysis of Production and Trade through the Principal Types of Amphora

1

Susanna Amari The Necropolis of Castelluzzo at Militello in the Catania Valley (Italy)

7

Tjark Blokzijl The Maritime Villae of Antium: A Case Study in Coastal Settlement

11

Margherita Carucci Origins of the Romano-African House

17

Luca Cherstich Homesickness and Greek Architecture in a Colony: The example of Cyrene (Libya)

25

Jo Day Making ‘Scents’ of Flowery Pots. Considering Minoan Ceramic Vessels with Botanical Relief

33

Andrea Umberto De Giorgi Beyond the Consumer City; the Case of Ancient Antioch

39

Vanessa D’Orazio The Site of Iuvanum (Abruzzi, Italy) in the Classical period; Its Public and Private Buildings: A Preliminary Study of a Taverna

45

Domenico Fossataro Central Adriatic Sea: real ‘importuosa litora’?

51

Florentia Fragkopoulou Understanding the Limits of Pollution: The Use of the Dead

57

Cybelle Greenlaw Monkeying Around the Mediterranean: A Fresh Perspective on Ancient Primates

63

Heinrich Hall Special Sites in the Cretan Neolithic Landscape

69

Jesús Lorenzo Jiménez From Antiquity to Feudalism. At the Southern Border Of Navarra. The Design of a Study in Landscape Archaeology

77

Ibrahim Kreetallah The Urban Development of Leptis Magna in the Late First Century BC and Early First Century AD

83

Charlotte Langohr Eteocretans in The Bronze Age? The Far East of Crete During The LM II-III Periods (1450-1100 BC)

87

i

Quentin Letesson Reconstructing Minoan Monumentality

95

Dionysia Marinou Terracotta Human Figures and Figurines from the LBA Cyclades (The Cases of Ayia Irini and Phylakopi): Art, Religion, and Politics

101

Lidia Matassa The Myth of the Synagogue on Delos

105

Oscar Mei A ‘Hall of the Muses’ at Cyrene (Libya)?

115

Hallie Meredith-Goymour Disentangling Material Cultures: Late Roman and Sasanian Facet Cut Glassware in Late Antiquity

123

Angelos Papadopoulos A City Under Siege? A Look at Bronze Age Iconography

131

Elvira Rocca Symbols of Superiority: Western Greek votive offerings at Delphi

139

Heather Mae Russell Sacred or Profane: Swallow-Painted Nippled Ewers from Akrotiri

147

James G. Schryver Is There ONE Crusader Archaeology?

155

Nina Shiel Exploring the Divine Cow in Ancient Egypt

161

Joe Skinner The Uses of Bow Imagery in a Greek Context: The Enemy Within

167

Federica Spagnoli Altars and Cult Installations of Punic tradition in North Africa

173

Deirdre Stritch Sun, Sea and Archaeological Sites: Archaeology and the State – A View from Cyprus

181

Vasileios Tsamis Body, Senses and Space in Late Bronze Age Central Macedonia, Greece

187

Thomas Tselios A New Look at Minoan Metalworking Techniques

193

Kalliopi Vacharopoulou Debates and Trends on Heritage Management and Architectural Conservation in the Mediterranean Region: The Issue of the Driving Forces Behind Conservation Planning

199

ii

Preface The eighth annual Symposium on Mediterranean Archaeology took place from the 20th to 22nd February 2004, hosted by the School of Classics in Trinity College Dublin. Since its inception, SOMA has sought to provide postgraduate researchers in Mediterranean archaeology with a forum in which to present their work in a friendly, collegial environment, and to give voice to new trends in Mediterranean archaeological research. This year, one of the main aims of the organisers was to make the symposium inclusive of regions and time periods not normally covered by the conference, i.e., to reach beyond the eastern Mediterranean region, and the Bronze Age and Classical periods. This aim was satisfactorily achieved, with presentations covering places as diverse as Libya, Sardinia, and Jerusalem, and ranging chronologically from Neanderthals to Crusaders! Professor David Mattingly of the University of Leicester opened the conference with a paper ‘From Mystery to History; the Garamantes of Libya’, while Dr. Hazel Dodge, the Louis Claude Purser Senior Lecturer in Classical Archaeology at Trinity concluded the symposium with a review of proceedings. She also presented some suggestions for future SOMA meetings, and raised key questions such as whether, in its new expanded form, SOMA should remain monolingual or become multilingual. Such issues are indeed challenging and stimulating, and provide matter for future SOMA committees to contemplate. After eight years, perhaps a discussion on the future direction of SOMA would be a helpful tool in ensuring the long-term usefulness of the conference to both postgraduate students, and to research on Mediterranean archaeology in general. The 2004 symposium was a great success, attracting more than 70 participants, thus making this the largest SOMA to date. The scope of the symposium was reflected in the broad range of papers and posters, both presented at the conference and included within this volume. The sessions and participants were as follows: Cultural and Social Identity: James Schryver, Dorian Borbonus, Harri Kiiskinen, Maria Cristina Mancini, Gina Martella, Charlotte Langohr, Agryro Nafplioti, Daniel Sahlén, Francesca Ramondetti. Archaeological Landscapes: Luke Sollars, Andrew Bicket, Domenico Fossataro, PierLuigi di Clemente, Jesus Lorenzo, Erin Gibson, Richard Jennings, Heinrich Hall, Carmen Soria. Religion and Ritual: Julie Candy, Vasiliki Koutrafouri, Federica Spagnoli, Camilla Briault, Martine Hogervorst, Elvira Rocca. Death and Burial: Susanna Amari, Juan Antonia Camara-Serrano, Liliana Spanedda, Florentia Fragkopolou, Yannis Galanakis. Technology: Thomas Tselios, Rachel Conroy. Cultural Heritage Management: Kalliopi Vacharopoulou, Nota Pantzou, Ed Richardson, Duane Gehlsen, Deirdre Stritch. Art and Iconography: Nina Shiel, Heather Russell, Jo Day, Angelos Papadopoulos, Dionysia Marinou, Victoria Gyori, Ali Ammar, Olympia Bobou, Catherine Cooper, Cybelle Greenlaw, Eleni Hatzivassilou, David Saunders, Helle Hochscheid, Joe Skinner. Sacred Space: Maria Dikomitou, Lidia Matassa. Architecture and Spatial Analysis: Luca Cherstich, Oscar Mei, Stella Skaltsa, Vasileios Tsamis, Margherita Carucci, Anastasia Christophilopou, Ibrahim Kreetallah, Quentin Letesson. Trade and Cultural Interaction: Rachel Fentem, Anthis Kaldelis, Hallie Meredith-Goymour, Alain Gaulon, Kristian Goransson, Amber Creighton. Settlement Patterns: Helen Dawson, Tjark Blokzijl, Andrea Umberto De Giorgi, John Harte, Cedric Devais, Ludovic Decock, Peev Preslav, Lynne Kvapil. Sources and Evidence: Valentina Asciutti, Sara Chumbley, Karen Ní Mheallaigh. Posters were presented by: Francesco Aliperti, Areti Chalkioti and Vasiliki Ivrou, Igor Cherstich, Giordano D’Addazio and Angela Cinalli, Eugenio Di Valerio, Vanessa D’Orazio, Domenico Fossataro and PierLuigi Di Clemente, Debora Lagatta, Gina Martella and Maria Cristina Mancini, Stefano Marucci, Luana Quattrocelli, Amedeo Ruggiero, Vittoria Sardo, Francesca Siciliano. We extend our best wishes to the Committee of SOMA 2005 at Chieti University, Italy. SOMA 2004 Committee

iii

List of Contributors Francesco Aliperti Facolta di Conservazione dei beni Culturali Università Della Tuscia Viterbo

Ibrahim Kreetallah School of Classics Trinity College Dublin

Susanna Amari Università di Catania – C.E.A.

Charlotte Langohr Département d’Archéologie et d’Histoire Université Catholique de Louvain, Belgique

Tjark Blokzijl Department of History (Ancient History) Rijksuniversiteit Groningen

Quentin Letesson (KMcA) Département d’Archéologie et d’Histoire Université Catholique de Louvain, Belgique

Margherita Carucci Department of Archaeology University of Nottingham

Dionysia Marinou Department of Archaeology University of Sheffield

Luca Cherstich Institute of Archaeology Brasenose College Oxford University

Lidia Matassa School of Religions Trinity College Dublin

Jo Day School of Classics Trinity College Dublin

Oscar Mei Postgraduate Specialisation School in Archaeology Università Cattolica, Milano

Andrea Umberto De Giorgi Department of Classical & Near Eastern Archaeology Bryn Mawr College

Hallie Meredith-Goymour Department of Classical Archaeology Lincoln College, Oxford University

Vanessa D’Orazio Dipartimento di Scienze dell’Antichità Università G. d’Annunzio, Chieti

Angelos Papadopoulos School of Archaeology, Classics and Egyptology University of Liverpool

Domenico Fossataro Universities of Rome and Chieti

Elvira Rocca Department of Archaeology University of Cambridge

Florentia Fragkopoulou University of Cambridge

Heather Mae Russell University of Oregon

Cybelle Greenlaw School of Classics Trinity College Dublin

James G. Schryver Medieval Studies Program Cornell University

Heinrich Hall Department of Classics University College Dublin

Nina Shiel School of Classics Trinity College Dublin

Jesús Lorenzo Jiménez Departamento de Geografia, Prehistoria y Arqueología Universidad del País Vasco – Euskal Herriko Unibertsitatea

Joe Skinner Leicester University

v

Federica Spagnoli Università degli studi di Roma “La Sapienza” Deirdre Stritch School of Classics Trinity College Dublin Vasileios Tsamis Department of Archaeology University of Southampton Thomas Tselios Department of Archaeology University of Crete Kalliopi Vacharopoulou Institute of Archaeology University College London

vi

Acknowledgements The Committee of SOMA 2004 would like to acknowledge the help of many people, without whom the conference and publication could never have taken place. Firstly, the programme for Mediterranean and Near Eastern Studies, Trinity College Dublin, whose generous financial support allowed both the conference and publication to go ahead. The School of Classics, Trinity College Dublin, also helped greatly by providing facilities, not just over the weekend of the conference, but during the preparations for it and throughout the publication process. Support was also gratefully received from Archaeological Dialogues, SOMA 2002 Glasgow, and all those who provided prizes for the fund-raising table quiz. We would like to thank Trinity College Dublin for hosting the conference, and in particular the security guards who gave up their weekend to look after us. Many thanks are due also to our key-note speaker, Prof. David Mattingly, and our closing speaker Dr. Hazel Dodge. We are especially grateful to Dr. Christine Morris, not just for being our opening speaker, but for her constant support and encouragement throughout the whole project. Without the help of our student volunteers, and Simon Spence, our technical expert, the weekend would not have run so smoothly. Laois Education Centre kindly provided extra audio-visual equipment. Thanks also to BAR for publishing the proceedings and running the book sale during the conference. Finally, we would like to express our gratitude to the anonymous reviewers who carried out the valuable and timeconsuming process of peer-reviewing all the articles. The committee of SOMA 2004 was: Jo Day, Cybelle Greenlaw, Heinrich Hall, Amanda Kelly, Ibrahim Kreetallah, Lidia Matassa, Killian McAleese, Emma Saunders, Nina Shiel and Deirdre Stritch.

vii

Italian Wine between the Late Republic and the First Century AD: Analysis of Production and Trade through the Principal Types of Amphora Francesco Aliperti

1980: 279). This introduction is necessary to clarify the aim of the study: to give an account of centres for production and distribution of wine amphorae and to speak about the product contained therein, Italian wine.

Introduction The importance of wine in Greek and Roman society is evidenced by a flourishing trade involving all Mediterranean countries, those of continental Europe and the British Isles. It is important to remember that it was very different from today’s wine. It was more syrupy and had more body than that of today; it was usual to mix it with water and to flavour it in the appropriate kraters. Sometimes socalled defrutum was added, a concentrate rich in sugar and alcohol (Tchernia 1986: 141). The study of wine in itself is very difficult as it cannot be limited to a quantitative analysis. It is necessary to stress the qualitative aspect in order to understand its economic implications. It is known, from the sources we have, that there were many varieties of wine, from the most precious, such as the well-known Falernum, to the more common and popular ones such as the wines coming from Gallia in the second century AD (Tchernia 1986: 62). Moreover the wine trade must be analysed on two levels: that of regional and international trade and that of retail trade in the numerous tabernae located in the cities of the Roman state. The importance of wine selling is evidenced by the fact that the main street of Trajan’s Market, close to Trajan’s Forum, was probably already called Via Biberatica in the Roman Age, although the first known mention of it comes from the Medieval period (Coarelli 2001: 147).

‘Graeco-Italic’ Amphorae ‘Graeco-Italic’amphora production represents the evolution of a Greek tradition on Italian soil. Though this tradition was not the only one present on the Italian peninsula (for example Etruscan amphorae were also produced) or on the major islands (Punic amphorae were also produced here), it constitutes, both directly and indirectly, the most important one as a model for the later Roman production. The production of ‘Graeco-Italic’ amphorae covers a long span of time from the fourth to the second centuries BC; only the later production is related to the late Republican period (Tchernia 1986: 42). E. Lyding Will, in an interesting study of the shape of these amphorae, names five different sub-types categorising them with the first letter of the alphabet (1982: 338-56); we will use his categorization to facilitate the explanation. The sub-group identified by the letter ‘C’ and especially that indicated by the letter ‘D’, commonly dated to the second century BC, constitute the evolutionary focal point of these amphorae (Figs. 1-2). Some amphorae produced in

A further difficulty arises from the fact that the study of wine can only be carried out through the indirect analysis of amphorae. These containers are characterised by recurrent features: a wheel-made belly, with the handles, shoulders and a rim atop the neck area, together with a conical base (Sciallano and Sibella 1991: 11-12). All these parts were made separately before being assembled for the firing process and then filled with wine, closed with a cork and sealed with the so-called operculum made of pozzolana. An important study of the techniques of ceramic production, and therefore the production of amphorae, is that by N. Cuomo di Caprio (1985: 15-150). Finally finished with graffiti and tituli picti containing the relevant information for trade, they were piled in several layers on cargo ships directed to different destinations (Gianfrotta and Pomey

FIGURE 2 FIGURE 1 ‘GRAECO-ITALIC’ AMPHORA ‘GRAECO-ITALIC’ AMPHORA All figures are taken from: Empereur and Hesnard 1987 1

SOMA 2004 the Terme del Nuotatore, at Ostia Antica (attestations of this in the second century AD are also found in Settefinestre. With reference to Settefinestre see Cambi and Volpe 1985: 72-4, 89-93. For Ostia see Panella 1970; 1973, particularly Ostia III: 492 and Panella 1989: 177). Nevertheless, it is necessary to manage these data cautiously, since in Rome the excavation of Meta Sudans does not register the presence of Dressel 1 amphorae in the strata of the second half of the second century AD (Panella 1989: 174).

FIGURE 3 DRESSEL 1 AMPHORA

Dressel 1 was produced in a number of places (Panella 1981: 66-7). The production of this important container also developed in the provinces, as demonstrated by the discovery in the excavations of Gallic and Spanish figlinae, of fragments of Dressel 1 amphorae.1 Archaeological records show that there were production centres in Astura, Albinia, and Sinuessa, to mention only some of them (Hesnard, Tchernia et al. 1989: 21-30). The analysis of the clay used for certain amphorae also suggests Pompeii as a centre with figlinae; Reggio Calabria is quoted in the inscriptions in connection with Rheginum vinum (Panella 1981: 68-9). The wines produced in these areas, and which filled these amphorae, were the Setinum, from Sezze, and the Caecubum and Fundanum from towns to the south of Rome. The famous Falernum probably came from the area of today’s Campania region.2

FIGURE 4 DRESSEL 1 AMPHORA

Spain are identified by Will with the ‘E’ group (Empereur and Hesnard 1987: 25-6). Nevertheless, the greatest production of this kind of amphora is to be found in Italy, and particularly on the Tyrrhenian side, whether in Etruria or in Latium and Campania (For Etruria see Manacorda 1981: 3-54; for Latium and Campania see instead Hesnard, Ricq, et al. 1989: 21-30). The ‘Graeco-Italic’ amphora type underwent a process of evolution of the neck and belly area (Empereur and Hesnard 1987: 29). In addition, important changes can be identified on the rim, which, originally in a horizontal position, becomes more and more inclined (Empereur and Hesnard 1987: 30). We cannot fail to notice a much wider use of the Latin language, in place of Greek, on the imprints on the amphorae; this is a clear sign of the changing political and social conditions due to the rise of Rome (Manacorda 1981: 22-5).

The analysis of the spread of amphorae in the territory reveals Dressel 1’s presence in the whole of Etruria, while the type is less frequently found in Cisalpina and on the Adriatic coast. Dressel 1 is also documented in Rome and Ostia, Gallia and the Iberian Peninsula and along the Rhine and the Thames. Its presence witnesses the important role played by waterways in the development of ancient trade. There is less evidence for the type in the East, North Africa, and Sicily, where Greek forms were still predominant (Panella 1981: 73; Peacock and Williams 1991: 87, 90, 92, 94; Sciallano and Sibella 1991: 34-6; Caravale and Toffoletti 1997: 98). Dressel 2/4 Amphorae The Dressel 2/4 amphora (Fig. 6) is the result of an evolution of the amphora from Kos (Fig. 5), a Greek production characterised by the bifid handles and the terminal flat base. Within the general category of these amphorae there is a variety of sub-categories. Its main features are bifid handles, a long cylindrical neck and a solid conical bottom

Dressel 1 Amphorae Within the general category, the amphorae named Dressel 1 show a variety of sub-groups, the study of which has not yet given definite answers to all of the questions raised by experts. The problem of chronological sub-grouping remains unsolved, and it is difficult to understand the way different amphorae, named Dressel 1a (Fig. 3), Dressel 1b and Dressel 1c (Fig. 4), are related to one another.

1. For more information about the general production of Dressel 1 see the particularly useful volumes by Peacock and Williams 1991: 86-95; Sciallano and Sibella 1991: 34-6; Caravale and Toffoletti 1997: 98-103. With reference to the figlinae that produced Dressel 1 see Laubenheimer 1989: 116-8 and Caravale and Toffoletti 1997: 98, while for the Spanish production see Peacock and Williams 1991: 91, Sciallano and Sibella 1991: 36 and Caravale and Toffoletti 1997: 98. 2. Particularly important for this proposal is the volume by Tchernia 1986. For the Falernum, see Tchernia 1986: 62; for the Fundanum, the Caecubum and the Setinum see Tchernia 1986: 160. The interesting appendices II – III – IV to the text of Tchernia 1986 refer to other wines.

It is commonly accepted that Dressel 1a was in use between 150 BC and the Augustan age, Dressel 1b between 97 BC and 13 BC, while Dressel 1c only appears later. The production of Dressel 1 amphorae, at least in Italy, is active until the second century AD, as witnessed also by some fragments found in the interiors of Traian-Adrianan strata at 2

FRANCESCO ALIPERTI attested for almost all of the second century AD (Cambi and Volpe 1985: 73-4; 89-93; Panella 1989: 174-8).

FIGURE 5 KOS AMPHORA

Lamboglia 2 Amphorae In addition to production on a massive scale, as seen in the above examples, other equally important production centres existed with a regional and interregional distribution. The Lamboglia 2 amphorae (Fig. 7), which are contemporaneous with Dressel 1, fall into precisely this category, as also happens for the sporadic findings at Ostia and Settefinestre in strata where they can only be considered residuals (Empereur and Hesnard 1987: 33, for the dating Cambi and Volpe 1985:74, 89-93; Panella 1970: 117). Experts long believed these containers to be more suitable for transporting oil rather than wine, though this was not the case. It is commonly accepted that they were produced in the area between Brindisi and Taranto, and it is therefore presumed that they transported wines from the vineyards at Brundisium, Arpi, and Canusium (Baldacci 1972: 35). The central and Adriatic coasts are, nevertheless, suspected to be production areas (Tchernia 1986: 54; Cipriano and Carre 1989: 80-3). The vessels are documented principally on the Adriatic Coast, along the course of the Po River, and in the east. But examples of a certain importance are found in the western area of the Empire (Zevi 1967: 21-31; Baldacci 1972: 20; Cipriano and Carre 1989: 84-5).

FIGURE 6 DRESSEL 2/4 AMPHORA

(Cipriano 1985: 191). The figlinae producing these amphorae were often located in the cities already mentioned regarding Dressel 1, as for example Terracina, Albinia, Sinuessa and the Vesuvian area, while in other instances they are mentioned in areas such as Salento and Istria (Panella 1981: 66-7; Peacock and Williams 1991: 105-6, Sciallano and Sibella 1991: 40; Caravale and Toffoletti 1997: 107). This information brings us to the conclusion that the quality of the wines transported in Italic Dressel 2/4 must have been the same as that of Dressel 1.

Dressel 6 Amphorae Between the Augustan era and the year 1 AD, the northern Italian market enjoyed enormous success in its local production of amphorae, that of Dressel 6 (Fig. 8) (Peacock and Williams 1991: 98-101; Sciallano and Sibella 1991: 38; Caravale and Toffoletti 1997: 108). Nevertheless, we cannot forget that the examples of Dressel 6 also come from

The Italic Dressel 2/4 enjoyed considerable success, even though it could not reach that of Dressel 1. According to the map of distribution, a consistent number is indicated, not only in Italy, but also in Narbonense and Sicily, while occasionally also beyond these borders. It is to be noted that no evidence for Dressel 2/4 has been found on the Adriatic coast or in Istria. This is unusual, especially considering the hypothesis of an Istrian production centre for this kind of amphora (Baldacci 1972: 19; Panella 1981: 76-8). The Dressel 2/4 shape was produced outside Italy, as demonstrated by the Gallic, Spanish and African atelier, and perhaps also the English atelier of Brocley Hill.3 This happened rather early on, firstly by independent Italic production centres, and later by local figlinae. This production was to reduce significantly the Italic negotiatores vinarii market and was to change Italy from a country that exported wine to one that imported it (Panella 1981: 78). Chronologically, the Dressel 2/4 amphorae start to circulate around the mid-first century BC and are 3. For the Spanish production see Tchernia and Zevi 1972: 35-67; Beltran Looris 1977: 97-132. For the French production see Zevi 1966: 207-47, Laubenheimer 1989: 118-23. For the English production see Castle 1978: 383-92. For the general production of Dressel 2/4 see Panella 1981: 5864, 74-6.

FIGURE 7 LAMBOGLIA 2 AMPHORA 3

FIGURE 8 DRESSEL 6 AMPHORA

SOMA 2004 later strata, such as the fragment from the late-Antoninian strata at Settefinestre, or the ones from the end of the second century AD at Ostia.4 This is morphologically similar in appearance to the Lamboglia 2 amphora. It is sometimes difficult to distinguish between them, especially in the absence of inscriptions on the containers.

Bibliography Baldacci, P. (1972) Importazioni cisalpine e produzione apula. IN: Recherches sur les amphores romaines, Collection École Française Rome, 10, 35. Beltran Lloris, M. (1977) Problemas de la morfologia y del concepto historico-geografico que recubre la nocion tipo- aportaciones a la tipologia de las anforas beticas. IN: Méthodes classiques et methodes formelles dans l’étude tipologique des amphores. Actes du colloque de Rome, 27-29 mai 1974, Roma, 97-131. Cambi, F. and G. Volpe (1985) Contenitori da cantina e da trasporto. IN: Ricci, A. eds.Settefinestre. Una villa schiavistica nell’Etruria romana, III, La villa e i suoi reperti, Modena, 72-93. Carandini, A. (1981) Sviluppo e crisi delle manifatture rurali e urbane. IN: Giardina, A.and A. Schiavone eds., Società romana e produzione schiavistica, II, Merci, mercati e scambi nel Mediterraneo, Roma-Bari, 250-5. Caravale A. and I.Toffoletti (1997) Anfore antiche. Conoscerle ed identificarle, Formello. Castle, S.A. (1978) Amphorae from Brockley Hill, 1975, Britannia, 9, 383-92. Cipriano, M.T. (1985) Le anfore. Alcune produzioni documentate a Roma tra repubblica basso impero. IN: Bussi, R. and V. Vandelli eds, Misurare la terra: centuriazione e coloni nel mondo romano. Città, agricoltura, commercio: materiali da Roma e dal suburbio (catal. Mostra), Modena, 190-9. Cipriano, M.T. and M.-B.Carre (1989) Production et typologie des amphores sur la côte adriatique de l’Italie. IN: Amphores romaines et histoire économique: dix ans de recherche. Actes du colloque de Sienne (22-24 mai 1986), Roma, 67-104. Coarelli, F. (2001) Guida archeologica di Roma, Roma. Cuomo Di Caprio, N. (1985) La ceramica in archeologia, Roma, 15-150. Empereur, J.Y. and A. Hesnard (1987) Les amphores hellénistiques. IN: Lévếque, P.and J.P. Morel eds., Céramiques héllenistiques et romaines, II, Paris, 9-71. Gianfrotta, P.A. and P. Pomey (1980) Archeologia subacquea, storia, tecniche, scoperte e relitti, Milano, 143-64. Hesnard, A., Ricq, M., Arthur, P., Picon, M. and A. Tchernia (1989) Aires de productin des gréco-italique et des Dr.1. IN: Amphores romaines et histoire économique: dix ans de recherche. Actes du colloque de Sienne (2224 mai 1986), Roma, 21-30. Laubenheimer, F. (1980) Ruscino I. À propos de deux amphore de Ruscino: définition d’un noveau type d’amphores, Revue Arquéologique de Narbonnaise suppl. 7. Laubenheimer, F. (1989) Les amphores gauloises sous l’Empire. Recherches nouvelles sur leur production et leur chronologie. IN: Amphores romaines et histoire économique: dix ans de recherche. Actes du colloque de Sienne (22-24 mai 1986), Roma, 105-38. Lyding Will, E.(1982) Graeco-italic amphoras, Hesperia, 51, 338-56.

The figlinae that produced this type are believed to be located in the area between Aquileia and Emilia.5 Nevertheless, a production with a diffusion little more than regional is attested in the Piceno area (Cipriano and Carre 1989: 86-8). The factories situated in today’s Triveneto must often have been owned by large families connected with the senate, such as that of the Gavii in Verona, the Plinii of Comasco, or the Sepulii in Padova, to mention but a few (Baldacci 1972: 20). Analysis of the seals emphasises the largescale participation of these rich senatorial and sometimes equestrian families belonging to the government and administrative environment and suggests that, through the right of conquest (and probably some imperial favours), their fortunes increased with a swiftness equalled only by that of their loss during the second century AD, and, in fact, there are few testimonies of activities connected to private production and selling during the reign of Hadrian (117-138 AD) (Baldacci 1972: 20). Experts believe that these amphorae were also intended for transporting fruit and garum (Cipriano 1985: 193). Conclusions Drawing to the end of this short study, we can positively state that in the second century BC, there was a big change in the wine trade, which can be identified, with some certainty, in amphora processing favouring Roman-made articles over Greek ones, as shown by the inscriptions on the amphorae written predominantly in Latin. This century also saw the affirming of a new system of production (that remained in use until the second century AD): the little-landowners system (where the work of the slaves was only subsidiary to the owner’s work) was substituted by a system based exclusively on the slaves’ work (Carandini 1981: 251-3). The advantages of this new system promoted the first massproduction of amphorae, such as the Dressel 1 and then the Dressel 2/4 amphorae. Only in the Cisalpina region did different systems survive with a certain commercial success. The political and military affirmation of Rome, which extended its dominions, also brought her economic success. For this reason, Italian wines such as Falernum, Caebucum, Setinum and Fundanum, (to mention but a few) invaded the markets of Western Provinces.

4. In accordance with this dating is the fragment found in the Vatican carpark dating to the second century AD (Cipriano and Carre 1989: 85-6). Similar dates are given in Cambi and Volpe 1985: 74, 89-93, referring to the excavation at Settefinestre and in Ostia IV: 364 as regards Ostia Antica). 5. On the existence of figlinae in north-eastern Italy and in the Istrian peninsula see Baldacci 1972: 20; Cambi and Volpe 1985: 74; Peacock and Williams 1991: 99; Sciallano and Sibella 1991: 38; Caravale and Toffoletti 1997: 94.

4

FRANCESCO ALIPERTI Manacorda, D. (1977) Anfore, IN: AA.VV., Ostia IV, Studi Miscellanei, 23, Roma, 116-266 and 359-83. Manacorda, D. (1981) Produzione agricola, produzione ceramica e proprietari nell’ager cosanus nel I a.C. IN: Giardina, A.and A. Schiavone eds., Società romana e produzione schiavistica, II, Merci, mercati e scambi nel Mediterraneo, Bari- Roma, 3-54. Panella, C. (1970) Anfore, IN: AA.VV., Ostia II, Studi Miscellanei, 16, Roma, 102-56. Panella, C. (1973) Anfore, IN: AA.VV., Ostia III, Studi Miscellanei, 21, Roma, 460-633. Panella, C. (1981) La distribuzione e i mercati. IN: Giardina, A.and A. Schiavone eds., Società romana e produzione schiavistica, II, Merci, mercati e scambi nel Mediterraneo, Bari- Roma, 55-80. Panella, C. (1989) Le anfore italiche del II secolo d.C. IN: Amphores romaines et histoire économique: dix ans de recherche. Actes du colloque de Sienne (22-24 mai 1986), Roma, 139-178.

Peacock, D.P.S. and D.F. Williams (1991), Amphorae and the roman economy, an introductory guide, LondonNew York (paperback edition). Sciallano, M. and P. Sibella (1991) Amphores: comment les identifier?, Aix-en-Provence. Tchernia, A. and F. Zevi (1972) Amphores de Tarraconaise, IN: AA.VV., IN: Recherches sur les amphores romaines, Mélanges de l’Ècole française de Rome, suppl.10, 35- 67. Tchernia, A. (1986) Le vin de l’Italie romaine, Roma, 1986. Zevi, F. (1966) Appunti sulle anfore romane, Archeologia Classica, 18, 207-47. Zevi, F. (1967) Anfore istriane ad Ostia, Atti e memorie della società istriana di archeologia e storia patria, LXVII, Venezia, 21-31.

5

The Necropolis of Castelluzzo at Militello in the Catania Valley (Italy) Susanna Amari

Castelluzzo. No traces of the dwellings are known, though the naturally defended settlement of round and quadrangular huts was probably located in a strategic position on top of the Castelluzzo plateau. The Castelluzzo hill has the form of a little castle and is a naturally well-defended location as is usual in all prehistoric settlements (Frasca 1996: 139-40). A new hypothesis suggests that these high settlements were seasonal, linked to sheep-farming and the stone industry. This hypothesis is based on the discovery at Militello of Pleistocene dry river-beds from 2,000,000 years ago with pebbles and basalts (McConnell 2003).

Introduction Militello in Val di Catania is today a little town, rich in history, art and natural landscapes. It is 400 metres above sea level, on the south-western edge of the Plain of Catania. The territory of Militello has an abundance of prehistoric necropoleis and settlements from the Chalcolithic to the Mediaeval Age, among which are those at Dosso Tamburaro (McConnell 2003: 489-98), Fildidonna (Branciforti 2000; McConnell 2003), Ossini (Branciforti 1995: 101-26) and Castelluzzo which are situated along the canyons of the local rivers in a landscape that is still mostly unspoilt. At Castelluzzo some prehistoric tombs and Byzantine grottoes had already been found in 1985 (Branciforti 1995: 10910). This paper deals with the archaeological exploration carried out by the Superintendency of Catania which, at the beginning of 1999, located an Early Bronze-Age/Iron-Age necropolis in a calcareous canyon rich in fossils and wild vegetation, hollowed out by the water of the Carcarone River (Fig. 1). In 1999, another necropolis was found at Castelluzzo with rock-cut coffin-graves, probably from the late Roman period.

Features of the Necropolis The prehistoric necropolis of Castelluzzo, situated on a steep rocky slope overlooking the river, consists of rock-cut chamber tombs which date from the Early Bronze Age to the Iron Age (Fig. 2). The archaeological evidence suggests the abandonment of the necropolis at the beginning of the second phase of the Sicilian Iron Age, in the Finocchito phase (Frasca 1996: 139), when, in the last three decades of the eighth century BC, the Greek settlers arrived on various coastal sites. No Greek pottery or ornaments have been found in the tombs explored. After its abandonment the site was used as an occasional quarry. Sixty tombs have been identified at Castelluzzo cut into the limestone in rows along the natural escarpment of the canyon of the Carcarone River. The tombs have different characteristics.

FIGURE 1 VIEW (PHOTO: BRIAN MCCONNELL) Drawings and photos belong to the Superintendency of Catania The Prehistoric Settlement This analysis focuses on the architectural features of the tombs and the grave goods of the prehistoric necropolis at

FIGURE 2 TOMBS 5, 6, 7, 8, 9, 10 - FRONT (PHOTO: SUSANNA AMARI) 7

SOMA 2004 cut drains, a few centimetres wide. There is a big drain on the north rock face of Tomb 5. Tombs 5, 6, 7 and 8 are on different levels and go down towards the south. The tombs have chambers with a rectangular plan (with sides measuring 1.60/2.00 metres). In tombs 5, 6 and 7, part of the chamber is occupied by a rock bed. There is a little trench in Tomb 5. The tombs have monumental features: a large open forecourt, rarely more than 2 metres in width; the façade around the entrance of the chamber has a large ledge (Fig. 4); two benches (0.20 metres in width) run along the dromos (about 2.00 metres in width and 4.00 metres long) in front of the façade and the forecourt. The tomb entrances were closed with a rectangular slab, a monolithic door with a wedge-shaped wall. The slab was held in place with wooden posts fixed on the exterior in rock sockets cut into the walls of the forecourt (Fig. 5). Small and medium-sized stones fixed the slab to the floor. The chambers and the forecourts have flat ceilings. Tombs 9 and 10 have false pilasters cut out of the rock on the tomb façade, on either side of the entrance to the chamber (Fig. 6).

FIGURE 3 TOMBS 5, 6, 7, 8 - PLAN (DRAWING: MARIELLA PUGLISI) There are some tombs with more elliptical chambers and vaulted ceilings, while other chambers have an elliptical or a quadrangular plan and flat ceilings and coffin-graves. All of the burials are inhumations, and no signs of cremation have been found. In the coffin-graves there were multiple burials and some grave goods, which were found in situ. Features of the Final-Bronze/Early-Iron Age Tombs The subject of this paper, the four tombs of the Final-Bronze/ Early-Iron Age, from about 900-730 BC (Frasca 1996: 139), had been the focus of the 1999 investigation (Fig. 3). Rock-cut tombs numbers 5, 6, 7 and 8 are noteworthy for their shape and monumental façade with a rock-cut pathway and drains in between. The pathway starts at Tomb 5 and ends in front of Tomb FIGURE 4 8 with a large step. TOMB 6 - FRONT Along the sides of the (PHOTO: SUSANNA AMARI) pathway are two rock-

Burial Rite and External Assemblages In the four tombs studied, the burial rite documented is exclusively that of inhumation. It is impossible to say

FIGURE 6 TOMB 8 - FRONT (PHOTO: SUSANNA AMARI)

FIGURE 5 TOMB 7 - SECTION (DRAWING: MARIELLA PUGLISI) 8

SUSANNA AMARI

FIGURE 7 TOMB 6 - FUNERARY ASSEMBLAGE (PHOTO: GIUSEPPE BARBAGIOVANNI) how many bodies were placed stretched out in the chambers, perhaps multiple deposits of members of a single family. The anthropologist, Dr. Sebastiano Lisi, is presently carrying out an analysis of the skeletons for his postgraduate degree. All the tombs were disturbed and their grave goods stolen. The bones were found scattered in the inner chambers and in front of the entrances. Of exceptional interest is the discovery of two external assemblages. It is noteworthy that, in the tombs of this necropolis, funerary assemblages have been found in front of the exterior monumental façade. The assemblage of Tomb 6 consists of 3 small dippers. These dipping cups, T. 6/1, T.6/2 and T. 6/3, are coarse wares with raised or horizontal loop handles and bosses on the rim and a flat bottom. They range from 3 to 3.8 cms in height and 5.5 to 8.5 cms in diameter (Fig. 7). The dippers were found in situ on the left-hand side of the forecourt at the entrance to the tomb and were placed there after the tomb had been closed (Fig. 3). The vessels may have served for libations. The external assemblage of Tomb 6 represents an example of a kind of pottery service, which has parallels from the Early Bronze Age until the Iron Age in Sicilian funerary and domestic contexts (Maniscalco McConnell 1996: 84). The assemblage in situ in the forecourt of Tomb 8 (Fig. 3) consists of a bronze ring with a circular section (T. 8/20, diameter 1.6 cms), bronze pincers (T. 8/21, H. 3.4 cms), an undecorated jug (T. 8/19, H. max. 5.5 and bottom diameter 4.0 cms) and other small fragments of the pottery and stone industries (Fig. 8). All may have been put in place in the Early Iron Age (Pantalica South phase) because there is an oxidised iron rod inside the bronze pincers. Some pottery fragments found in the

dromoi during the exploration of the tombs 5, 6, 7 and 8 may also be attributed to the Pantalica III or South phase, about 850-730 BC (Frasca 1996: 139). Conclusion The Castelluzzo tombs with quadrangular chambers and a dromos belong to a tradition of funerary architecture in Sicily dating from the Late Bronze Age (Maniscalco 1986: 244), to the Iron Age (Frasca 1996: 141), as we have at Pantalica and Ossini (at Militello). These features—the large forecourt, the false pillars, benches, the large façade with a large ledge and slab—have been connected, however, with the funerary architecture of the Early Bronze Age in Sicily and Malta (Procelli 1981: 104; Maniscalco McConnell 1996: 87). The Castelluzzo tombs are noteworthy for their architectural features, while the discovery of two external assemblages in the exterior monumental façade is of exceptional interest. In eastern Sicily, a similar layout occurs in the Early Bronze-Age/Iron-Age necropolis of Valle del Marcellino at Villasmundo (Voza 1977: 568), where monumental rock-cut tombs of the eighth century BC and some external funerary assemblages have been found. In this case, however, the assemblages were on the benches of the dromoi. The architectural features and the external assemblages of the Castelluzzo Tombs, numbers 5, 6, 7 and 8, look as though they were a sign of the special status or role of the mighty families of the settlement and indicate an unusual kind of burial rite that opens up some new problematical issues about the cult of the dead. We know little about the ceremonial aspects of the funerary rite, although there is some evidence for ceremonies in honour of the dead, including funerary meals or libations

FIGURE 8 TOMB 8 - FUNERARY ASSEMBLAGE (PHOTO: GIUSEPPE BARBAGIOVANNI) 9

SOMA 2004

Bibliography

or offerings (Maniscalco 1986: 244-265; Frasca 1996: 141). These ceremonies point to the importance of the great ties that existed within the family between the dead man and his survivors (Maniscalco McConnell 1996: 86). These external assemblages suggest that the court and the dromos with the benches formed a monumental entrance and a visible place for the ritual activities of the family after the burial. They formed an entrance to the deposition rite of the dead in the chamber and a place for the funerary meals or libations or offerings of the family of the dead to take place. The pathway to the single monumental tomb of a family of special status was a public place. The Bronze Age and the Iron Age are marked by a high number of settlements in Sicily. At Castelluzzo the monumental tombs may be seen as an indication of the stability of the settlement, and these tombs were family tombs. The restricted number of monumental tombs in the necropolis may be seen as an indication of the power of a few families with a special role or status in the settlement and suggest marked differences in the social structure and organisation inside the settlement. The external assemblages indicate a kind of burial rite associated with the performance of ritual banquets after deposition, which opens up some new problematic issues about the familial ties, the cult of the dead in the social ideologies and the traditions of the local communities.

Branciforti, M. G. (1995) Militello nei taccuini di Paolo Orsi. Lèmbasi. Archivio Storico, I, 2, 101-26. Branciforti, M. G. (2000) Attività del Servizio Archeologico della Soprintendenza per i Beni Culturali ed Ambientali di Catania. Kokalos, XLV-XLVI, 1999-2000, II, in press. Frasca, M. (1996) Iron Age settlements and cemeteries in southeastern Sicily: an introductory survey, IN: Leighton, R. ed., Early Societies in Sicily. New developments in archaeological research Accordia Specialist Studies on Italy, 5, Accordia Research Centre. University of London, 139-45. Maniscalco, L. (1986), Tipologie funerarie nella Sicilia del tardo bronzo: Pantalica, Dessueri, Caltagirone, Archivio Storico per la Sicilia Orientale, LXXXILXXXII, 1985-1986, 241-65. Maniscalco McConnell, L. (1996) Early Bronze Age funerary ritual and architecture: monumental tombs at Santa Febronia. IN: Leighton, R. ed., Early Societies in Sicily. New developments in archaeological research, Accordia Specialist Studies on Italy, 5, Accordia Research Centre. University of London, 81-7. McConnell, B. E. (2003) Resti preistorici a Dosso Tamburaro (Militello in Val di Catania) e l’età del rame nella Sicilia orientale. IN: Fiorentini G., Caltabiano M., and A. Calderone eds., Archeologia del Mediterraneo. Studi in onore di Ernesto De Miro, Bibliotheca Archaeologica, 35, Roma, ‘L’Erma’ di Bretschneider, 489-98. Procelli, E. (1981) Il complesso tombale di contrada Paolina ed il problema dei rapporti tra Sicilia e Malta nella prima età del bronzo. Bollettino d’Arte, 9, 83110. Voza, G. (1977) L’attività della Soprintendenza alle Antichità della Sicilia Orientale.Parte II. Valle del Marcellino (Villasmundo). Kokalos, XXII-XXIII, 1976-1977, II, 1, 568-71.

Acknowledgements My thanks go to Professor Giovanni Rizza and Dr. Antonella Pautasso, archaeology researcher of the I.B.A.M.-CNR of Catania, for encouraging me to take part in SOMA 2004, and to Dr. Maria Grazia Branciforti, archaeologist and supervisor, and Dr. Laura Maniscalco, archaeologist of the Superintendency of Catania, for their precious assistance and for permission to publish the entire content of my paper including the drawings and photos belonging to the Superintendency of Catania.

10

The Maritime Villae of Antium: A Case Study in Coastal Settlement Tjark Blokzijl

had fishponds, one of them produced tegulae (rooftiles) and amphorae, and one of the villae may have exploited sulphur deposits (see Fig. 2).

Introduction Anyone walking from Torre Astura to Nettuno on a sunny afternoon will pass the visible remains of at least six Roman maritime villae. They are located near the Roman village, Astura, and the towns of Colonia Antium (modern Anzio) and Civitas Antium (modern Nettuno) (Brandizzi-Vittucci, 2000). Furthermore, they are all located within a radius of c. 60 km of Rome. There appear to be at least 25 maritime villae located on the littoral between Anzio and Nettuno, when a larger regional context is taken into account (see Fig. 1). Remains of possible commercial activities were discovered near some of these maritime villae. Six of them

There seems to be a paradox implied in the sources concerning Roman maritime villae. While the literary sources generally mention that the maritime villa and the attached fishponds only served an almost decadent, luxurious residential purpose (Varro Rust. 3.17.2, 3.17.89., Col. Rust. 8.16-17., Plin. HN. 9.170.-2., Vell. Pat. 2.33.4. and Cic. Att. 1.18.6., 1.19.6., 1.20.3. and 2.1.7.), the archaeological evidence appears to suggest otherwise.

FIGURE 1 THE VILLAE OF ANTIUM (AFTER LAFON 2001: 363) 11

SOMA 2004

FIGURE 2 COMMERCIAL ACTIVITIES (AFTER LAFON 2003: 363) Therefore, it is my hypothesis that the maritime villae of Antium in particular and Roman maritime villae in general, served both a residential and a commercial purpose.

Tiber, along the Tyrrhenian Sea. It is bounded by Monte Circeo to the east and by the Alban hills and the Lepini Mountains to the north. The Ausonian Mountains bear onto the sea. The Ager Pontino, or Pontine plain, is located at the foot of the Alban hills and Lepini Mountains. It is traversed by the river Astura.

Several modern scholars have studied the phenomenon of the maritime villa or its associated pisciculture. The main focus of these studies, however, was aimed at the luxurious aspect of these villae and the literary sources describing them (for instance: D’Arms 1970, Mielsch 1987: 23-32, Higginbotham 1997). Only recently, have archaeologists seemed to develop an increasing interest in the commercial activities of the maritime villae (Dyson 2003: 22).

The Pontine plain is formed of marine clay sediments and consists of four terraces (Kamermans 1991: 21-9, for the geology of the Ager Pontino). It was covered with aeolian sediments, mainly sand, during the last period of the Würm glacial and the beginning of the Holocene. The socalled ‘Pontine range of dunes’ came into being during the Holocene. Marine and fluvial sediments locally breached this range of dunes. Recent stratigraphic research carried out by the Groninger Institute for Archaeology (GIA), placed the geology of the littoral landscape in the general geological perspective of Latium Vetus (Attema, Nijboer et al. 2003: 108). Their findings reveal that the red sand sediments are remains of paleodunes and are of aeolic origin. This sand layer rests on tufa formations formed during the Middle Pleistocene. These tufa layers, in turn, rest on marine sediments, formed in the Lower Pleistocene. Tufa was commonly used as a building material throughout the Archaic and Roman periods in this region.

My criticism of these studies is not in any way directed at the thoughtful, detailed scholarship compiled from the literary sources. On the contrary, it is concerned with the specific methodology used by these scholars, and even more importantly, with the aspects they left out. It has been generally assumed that the maritime villae served only a luxurious residential purpose. Most of the studies expound this argument without even considering whether such villae could have served other purposes, such as economic ones. As a result of this preconceived idea, any evidence that might suggest otherwise has not received enough attention. The Landscape around the Maritime Villae The region of Antium lies in the province of Lazio, or ancient Latium (Vetus). Latium is located south of the river

Lagoon-lakes, formed by marine and fluvial breaches, came into existence immediately behind the range of 12

TJARK BLOKZIJL

FIGURE 3 FIRST PHASE OF HABITATION (AFTER LAFON 2001: 363) dunes. Large formations of marshes developed between the dunes and the Lepini mountains. Two terraces were formed by marine sediments. The area between these marine sediments and the Lepini Mountains was therefore situated on a much lower level than either of these so that water run-off from the slopes of the Lepini Mountains was caught here. Bad drainage also contributed to the development of the Pontine Marshes. They were finally drained during the agricultural reforms, the Bonifica Integrale, in the 1930s.

infrastructural developments are probably connected to a further increase in population and urbanisation during the Republican period. There was also an increase in settlement in the Pontine range of dunes at that time. Eventually, large luxurious villae were built on the coastline in the Late Republic and the Early Imperial period (Fig. 3). The Fieldwork The GIA is currently working on the Astura Project in this region. The project was started in 2001 (after a preliminary campaign in 2000) and is still on-going (see Attema, Nijboer, et al. 2003 and Attema, Nijboer, et al. 2000). I made a topographical survey of the coastline of this area during the campaigns of 2002 and 2003. Unfortunately, not all of the 25 villae could be visited during these campaigns. For instance, some remains of villae are located on the military base of Nettuno and remain thereafter inaccessible (see below).

History of Roman Settlement The Romanisation of Latium Vetus and the colonization of the Ager Pontino took place between 400 and 100 BC. According to Livy (3.1), Antium became a Roman colony in 467 BC. Although the town had originally been Volscian, Roman families were settled alongside Volscian ones when the colony was formed. Latium Vetus was opened up in this period, in part due to construction of the Via Appia at the end of the fourth century BC. There is also evidence, for example near Sezze (Lazio), that the local road network developed and did connect with the Via Appia soon after its construction (Attema, Delvigne et al. 1999), although it was to take several more centuries until the region of Antium was fully integrated into the Roman road network through the construction of the Via Severiana. These

One of the villae in the region of Anzio was located on Tor Caldara (Fig. 1: 1A). In his index, Lafon (2001: 361) mentions that this particular villa was totally destroyed by tourism. For that reason, I was unable to investigate this site during the campaigns. It should also be noted that many of the villae are difficult to find due to considerable tourism-related activities on the beaches between Tor Caldara and Anzio. 13

SOMA 2004 century BC, but still well within the Republican period. However, its walls are executed in opus vittatum, which is rarely seen before the Augustan period (Adam 1994: 135.). Therefore, its period of habitation should be extended well beyond the Augustan period.

Tor Caldara The first villa to be discussed here is located on Tor Caldara near Anzio (Fig. 1: 1B). A topographic survey published in 1984 by Quilici and Quilici-Gigli focuses particularly on this villa. Tor Caldara is a small headland with a profile some three or four metres high. The villa dates to the late first century AD. The soil profile here includes much Roman material; some remains of Roman decorative plaster are also still visible. However, the site has been designated a nature reserve and is therefore, unfortunately, inaccessible. This is regrettable since, according to Quilici and Quilici-Gigli, there is evidence for the exploitation of several sulphur deposits, possibly during Roman times (1984). Sulphur was used for a variety of purposes, eg, in the wool industry, as medicine or for cleansing rituals (Pliny HN 35.174-6.).

Only some walls in opus reticulatum still remain of villa 7. They can be dated to the first century BC. However, it cannot be stated with certainty whether these walls originated from its first building phase or from a later one. Villa 9 is one of the oldest villae present in the area; it dates to the end of the second or the beginning of the first century BC. Villae 11 and 12 are hard to trace; their periods of habitation are unknown. Lafon is unsure whether number 11 even represents a villa at all. Villa 12 should be located in the harbour of Anzio, but has probably been destroyed by tourist activities.

The second and third villae discussed here are far less spectacular than the first one. Villa 2 is represented only by a small piece of wall and some Roman material in the soil profile. Villa 3 is even more obscure. There is no sign of any remains apart from a large scatter of Roman roof tiles on the beach. This scatter continues for almost 50 metres and is concentrated around a small stream. In my opinion, it may represent the remains of a villa or another form of habitation. As the terrain at this location is lower and much less steep than at the previous ones, it would have been an excellent location for a settlement.

From Nettuno to Torre Astura Unfortunately, the villae of Nettuno (nos. 13 A/B/C) can no longer be traced. Fishponds were present at three locations during Roman times. The surface areas of these fishponds are 600 (13A), 700 (13B) and 500 (13C) metres squared respectively. However, the ponds and villae have all been destroyed by tourism. The most recent drawings of the fishponds date to 1924 (Jacono 1924). Villae 14, 15 and 16 are not actually located on the coastline. Their distance from the coastline varies between 100 metres (no. 14) and one kilometre (nos. 15 and 16). Nevertheless, they should still be taken into account here, as they are located very near the other maritime villae. There could have been economic interaction between these more inland villae and the maritime ones. However, all three of these inland villae are now located on a military base and are, therefore, inaccessible, as mentioned previously. These villae were studied for the last time by Piccarreta in 1977 (see Piccarreta 1977: nos. 54, 86 and 65.) and remain unavailable for further study. However, since the data from the Astura survey conducted by the GIA in 2003 confirm the location of most of the sites mapped by Piccarreta, the locations he provides for villae nos. 14, 15 and 16 are probably correct.

Anzio Brandizzi-Vittucci (2000) provides a broad outline of the villae in Anzio (nos. 4-12). However, some villae are described very specifically, while others are just mentioned briefly. Special attention is given to the large Imperial villa (no. 10). The emperor Caligula liked this villa so much that he reportedly adopted the plan to move the capital to Antium (Suet. Calig. 8.5.). Apparently, Antium was also Nero’s favourite place of residence (Tac. Ann. 15.23.).The first phase of habitation here dates to the Republican period. The villa was then expanded by the Julio-Claudian, the Flavian, and finally the Severan dynasties. One of its most striking features is the bathhouse, the remains of which are still visible from the beach. Equally remarkable are the artificial caves. There is discussion as to whether villa 8, which belongs to the Augustan period in its first phases and was expanded in the second century AD, was also part of this large Imperial complex.

A kiln has been found near villa 17. The large scatters of fragments found near it indicate that it must have produced amphorae and tegulae. Remains of a bathhouse are also present, as well as some fragments of beautifully painted plaster. This villa was first inhabited in the first century BC. A stamp on one of the tegulae reveals the name of a possible owner to this particular villa, which I translate as Marcus Aurelius Aginus of the tribus Pomptina (Blokzijl 2005). The tribus Pomptina makes up the region of the Ager Pontino. The stamp dates from the late second or early third century AD (see Sandys 1919, Bloch 1967). This could be a clue that tile production as well as habitation continued, perhaps with intervals, well into the third century AD. The amphorae fragments found here

According to Lafon (2001: 361), Cicero could have been the owner of villa 4. He sold a villa in Antium to Lepidus (Cic. Att. 13.47a.). However, little remains today to allow us to ascribe this particular villa to Cicero, or Lepidus for that matter. Villa 5 is particularly interesting. Its remains, located in the garden of a wealthy mansion, still include a small aqueduct, a cistern, funerary inscriptions, and fragments of mosaics. Its walls and pavements date from the first century BC. According to Lafon (2001: 361), villa 6 dates from the first 14

TJARK BLOKZIJL are of the Graeco-Italic type (second half of the second century BC), the Dressel 1A type (end of second to first quarter of the first century BC) and the Dressel 6A type (Augustan period); several fragments of unknown types also occurred (Attema, Nijboer et al. 2003: 121-5). This may indicate that the production of amphorae ceased after the Augustan period.

suburbium (see also Morely 1996: 149). The transport costs for these goods were low so that a good profit could have been made. The locally produced amphorae could have been used as transport vessels. Plausible as it appears, I have to admit that no concrete archaeological evidence for the production of salted fish or garum near Antium has yet been found. For one, the tanks described in this article are very different to known garum and salting installations in the Mediterranean (See Curtis 1991 on such installations). Nevertheless, literary and epigraphic evidence, such as tituli picti on amphorae, suggest that some such installations were indeed present. Athenaeus (6.224c) mentions the transport of fish from Antium to Rome. As untreated fish would have perished on such a long journey, it must have been salted before transport. In addition, an inscription on an amphora found in Rome reveals that the content of the vessel was probably garum produced in Antium (CIL 15.4712). Salted fish and garum were often transported in Dressel 6A amphorae (Curtis 1991: 88), a type produced, among others, by Villa 17. In short, the combined evidence makes the suggestion of large-scale production of seafood near Antium worthy of further exploration.

Some large walls and foundations are all that remain of villa 18. More striking are the fishponds of villa 19. While there remains almost nothing of the residential area, the semicircular fishponds are quite impressive. They have a surface area of 3,200 metres squared. The fishponds of villa 20 consist of two large tanks, measuring 675 metres squared, with an interior walkway. When the sea is calm, large foundations extending for at least 50 m are visible beneath the waterline to their left. They have not yet been identified as the residential part of this particular villa, but the absence of other important foundations in the vicinity supports such an interpretation. Only a small amount of material in a two-metre soil profile survives of villa 21. One of the largest and most informative maritime villae of the region is number 22, also known as Torre Astura. Its first phase of habitation dates to the first century BC, the last phase to the third century AD. An impressive system of fishponds and channels still survives. Large lozenge-shaped tanks are visible below the water level. The combined surface area of these fishponds exceeds 21,500 square metres, making this villa one of the largest Roman semi-marine constructions in Italy. The tanks of the fishponds could have been filled with seawater, fresh water or brackish water. A large cistern is located at the back of the villa. A sizeable port was attached on the left side of the villa. Although, today, this villa is located on the coastline, in Roman times a bridge of 130 metres’ length connected it to the nearby coast. Cicero may have been one of the owners of this villa (Higginbotham 1997: 151). He wrote many letters to his good friend Atticus from Astura. In one of these letters Cicero gives a description of what might possibly be this particular structure (Cic. Att. 12.19.1).

In contrast to the products considered above, the tegulae were probably made only for the local market. Transporting them to Rome would have been extremely expensive due to their weight and volume. Thus, the region of Antium was an excellent location for intensive pisciculture. The presence of ports obviously played a very important role in the transportation of these commodities. Villa 22, with its port, probably served as a central place of economic interaction for the local community. Conclusion There were at least 25 villae present in the region of Antium. Six of them had fishponds, one of them produced tegulae and amphorae, and one may have exploited sulphur deposits. Their dates clearly indicate a peak in activities between in the first centuries BC and AD, but at some of the villae, occupation continued into the second and third centuries AD. The stamp on a tegula found at villa number 17 reveals the name of a possible owner of that structure, whereas other possible owners of villae can be suggested on the basis of evidence from the epigraphic and literary sources.

Production and Trade The main focuses of this article are the possible commercial activities of the maritime villae. Their produce could be sold on the nearby market of Antium. Of course, the neighbouring villae themselves would also have generated a demand for goods. Even beyond the region of Antium, there was good demand for the commodities such villae are likely to have produced. The area is located just 30 km away from the suburbium of Rome and 60 km from the city itself.

While the sea is obviously an important part of the surrounding landscape, there is hardly any evidence that maritime villae exploited their non-marine hinterlands as well. On the other hand, there is also no evidence to the contrary. This is a new question in the study of Roman coastal habitation that still requires more research. Environmental sampling should be of great use in elucidating this issue, as should surveys of their surrounds, with a special focus on any kinds of boundary markers.

The commodities produced in the region of Antium could only have been non-perishable and slow perishing products such as sulphur and salted fish, garum (fish sauce). Such commodities could have been transported to Rome or its 15

SOMA 2004 As a final thought, let us suppose that maritime villae did exploit their hinterlands. What would this exploitation have looked like? Probably just like that of any other villa rustica. There would have been zones for arable cultivation, pasture and horticulture. If the hinterland of the maritime villa was indeed brought under cultivation, then the maritime villa, with its ability to exploit both land and sea, would have had some advantages over a common villa rustica. In any case, it is clear from this case study that, even without the exploitation of the hinterland, there is a strong possibility that more commercial activities took place on the estates of maritime villae than has generally been assumed.

Blokzijl, T. (2005) Wonen en werken: Commerciële activiteiten van Romeinse kustvillae. Paleo-aktueel 14/15, 139-42. Brandizzi-Vittucci, P. (2000) Antium. Anzio e Nettuno in epoca Romana, Rome, Bardi. Curtis, R.I. (1991) Garum and salsamenta. Production and commerce in materia medica, Leiden, Brill. Dyson, S.L. (2003) The Roman countryside, London, Duckworth. Higginbotham, J. (1997) Piscinae: artificial fishponds in Roman Italy, Chapel Hill NC., University of North Carolina Press. Jacono, L. (1924) Nettuno-Piscinae in litore constructae. Notizie degli Scavi antichita 21, 333-40. Kamermans, H. (1991) Faulted land: The geology of the Agro Pontino. IN: Voorips, A., Loving, S.H. and H. Kamermans eds., The Agro Pontino survey project, Amsterdam, Universiteit van Amsterdam, 21-9. D’Arms, J.H. (1970) Romans on the Bay of Naples, Cambridge MS, Harvard University Press. Lafon, X. (2001) Villa Maritima, Rome, École Française de Rome. Mielsch, H. (1987) Die römische villa. Architektur und lebensform, Münich, Beck. Morley, N. (1996) Metropolis and hinterland. The city of Rome and the Italian economy 200 B.C.-A.D. 200, Cambridge, Cambridge University Press. Piccarreta, F. (1977) Astura, Forma Italiae vol. XIII, Florence, Olschki. Quilici, L. and S. Quilici-Gigli (1984) Attivita estrattiva dello zolfo nella zona tra Ardea ed Anzio. Archeologia Laziale VI, 229-49. Sandys, J.E. (1919) Latin epigraphy: an introduction to the study of Latin inscriptions, Cambridge, The University Press.

Bibliography Adam, J.P. (1994) Roman building. Materials and techniques, London, Batsford. Attema, P.A.J., Nijboer, A.J. and T. de Haas (2003) The Astura project, interim report of the 2001 and 2002 campaigns of the Groninger Institute of Archaeology along the coast between Nettuno and Torre Astura (Lazio,Italy). BABesch 78, 107-40. Attema, P.A.J., Nijboer, A.J. and B.M.J. Mater (2000) Veldwerk op het strand tussen Nettuno en Torre Astura. Paleo-aktueel 12, 85-8. Attema, P.A.J., Delvigne, J. and B.-J. Haagsma (1999) Case studies from the Pontine region in Central Italy on settlement and environmental change in the first millennium. IN: Leveau, P., Trément, F., Walsh K. and G. Barker eds., Environmental reconstruction in Mediterranean landscape archaeology, Oxford, Oxbow books, 105-21. Bloch, H. (1967) The Roman brick stamps, not published in vol. XV.1 of the Corpus Inscriptionum Latinarum, including indices to the Roman brick-stamps, Rome, ‘L’Erma’ di Bretschneider.

16

Origins of the Romano-African House Margherita Carucci

In the 1960s there was a debate among Roman archaeologists in North Africa about the origins of the town house in the African provinces of the Roman Empire. In the analysis of Maison à la mosaïque de Vénus at Volubilis, Thouvenot stated that it was Roman in type, as the arrangement of vestibulum, peristyle and triclinium on the same axis resembled the axiality of fauces, atrium and tablinum of the Italic atrium house (1958: 60). Two years later Etienne published a greater number of imperial houses from the same site (1960). By focusing on the central courtyard and the main room, he had maintained that they were Greek in type. The ratio of roofed-over to exposed ground and the absence of alae in the courtyard show that it was not influenced by the Italic atrium but the Greek peristyle; the main room looked like the traditional oecus of the Greek house, as its entrance was often not on the same axis as the courtyard and its decoration made it the grandest room in the house. The arguments of both scholars were reviewed a few years later by Graham (1966). The similarities between the second century houses at Ostia with the Volubilitan houses led him to conclude that the North African house owed its origin to the direct importation of the atrium house type characteristic of Italy by colonists in the second century AD. More recently, Daniels argued that in the domestic architecture of Volubilis it is possible to detect a local tradition, drawn from a Punic heritage (1995).

relate to the same typology – the peristyle house (Rebuffat 1969/1974; Romanelli 1970; Thébert 1987; Muth 1998) – which became the dominant aristocratic house type across the Roman empire from the first century AD (MacDonald 1965; Ellis 1988; 2000; Gros 2001; Hales 2003). The peristyle house can be divided into three main architectural elements: the entrance hall (vestibulum), the colonnaded courtyard (peristyle) and the main room (triclinium). The arrangement of those spaces, which were the principal elements in the reception of visitors, affected the distribution of the other rooms (bedrooms, secondary dining rooms, kitchen and secondary courts) and outlined the design of the house (Fig. 1). The analysis of those specific spaces may clarify some issues concerning the origin of the Romano-African house. The Axial Peristyle House The statement of Thouvenot about the Roman origin of the plan of Maison à la mosaïque de Vénus (1958: 60) may be widely accepted. In fact, the arrangement of vestibulum, peristyle and triclinium on the same axis was based upon the standard plan of the Italic atrium house in Republican times: fauces, atrium, and tablinum, which were typologically related to the entrance hall, courtyard and main room of the peristyle house, were arranged along the central axis. From the second half of the first century AD the axis, as the organising principle which dominated the planning, the arrangement and the layout of domestic spaces was introduced in the peristyle house type. At Volubilis the axial plan is quite common. Further examples of axial layout in Tingitana have been identified at Banasa: Maison M1 (Rebuffat 1969: 663 n.1) and Maison M2 (Rebuffat 1969: 663 n.2); at Lixus: Maison d’Helios (Rebuffat 1969: 665 n.1); and at Thamusida: Maison du Dallage (Rebuffat 1969: 666 n.1). However, the axial plan was also adopted in the eastern part of North Africa, as demonstrated at Maison des Nymphes at Neapolis (Darmon 1980).

The debate has been focused on Volubilis, as its houses form a relatively well excavated and well published group, by far the largest number from any site. However, the houses at Volubilis cannot be considered as typical town houses of the Romano-African elite based on the availability of such a sample. It is evident that the arrangement of domestic space was influenced by the local conditions (geographic, economic and political) of Mauretania Tingitania. The subterranean rooms of the houses at Bulla Regia in Tunisia show that regional variation occurred within Africa (Beschaouch, Hanoune, et al. 1977; Hanoune 1980). Charles-Picard stated that the thousands of Romano-African houses are almost all like – except in size and in magnificence (1959: 211). In fact, all the examples of domestic architecture in North Africa

The axial peristyle type is not confined within the boundary of Roman Africa. Axial peristyle houses have been 17

SOMA 2004

FIGURE 1 FLOOR PLANS OF PERISTYLE HOUSE

(DRAWING: MARGHERITA CARUCCI) reported in other areas of the western Empire, in southern Gallia, in the Iberian Peninsula, and in Italy. In her article on the axial peristyle houses in the western Empire, Meyer states that the origins and the geographical distribution of the type are evidence for the Romanisation of the urban upper classes (1999). The architectural layout of their houses mirrors the requirements and the standard pattern of behaviour of Roman society. However, the axial plan is only one of several options for town houses in the Imperial period. The peristyle houses with a radial ground plan are equally an indicator of the adoption of a Roman lifestyle by their inhabitants (Balty 1984: 473).

are arranged. As in the peristyle house with a bayonet plan, the Punic domestic space is structured in such a way that passers-by would be unable to see into the courtyard and disturb the privacy of the family. Vestibulum Etienne stated that the two kinds of vestibulum of RomanoAfrican houses (square and oblong) draw upon two different traditions, Greek and Roman respectively (1960: 118-20). However, the examples of domestic architecture in the ancient Mediterranean society suggest different relations and influences. There are several examples of square entrances in the Roman provinces of North Africa. In these examples the vestibulum is a big room ornamented with a monumental façade, carpeted with mosaic and often provided with two doors for different uses and for different times of the day. It is evident that the African vestibulum cannot be typologically related to the small square entrance of the Classical Greek house, as Etienne argued. On the contrary, because of its lavish layout and its large dimensions, it can be typologically and functionally connected with the entrance of the peristyle house of the Imperial period. The geographical distribution and the architectural features of the square vestibulum are evidence for Romanisation of the urban upper classes and for their social functions. The lavish decoration displays the owners’ opulence, and the generous dimensions allow them to receive their clients in the morning salutatio.

The Bayonet Plan The bayonet plan is by far the most common in domestic architecture of Roman Africa. The arrangement of the vestibulum at an angle to the peristyle-triclinium axis affected the relation of the house with the outside and the personal needs of its inhabitants. In the axial plan the house is immediately presented to the visitor standing in the vestibulum, with a vista that leads directly through the central point of the peristyle and into the triclinium. In the bayonet plan house, on the contrary, a direct view from the vestibulum to the triclinium is not possible: only in the doorway from vestibulum to peristyle does the view open through the courtyard as far as the triclinium (See for example, Maison d’Icarios, Oudna in Fig. 2).

Etienne maintained that the oblong entrance type of Romano-African houses was Roman in origin, as it resembled the fauces of the Italic atrium house. The resemblance cannot be accepted, as the fauces are lacking the two main features of the African oblong vestibulum: size and off-centre location. Both those elements instead appear in the corridor-like entrance of the Punic house. It is therefore arguable that the shape of the African vestibulum was influenced by Punic traditions.

This kind of layout seems to draw upon older local traditions. The Punic houses excavated at Kerkouane, which are dated from the sixth to the third centuries BC (Fantar 1987), and on the slopes of the Byrsa hill at Carthage, which are dated to the first half of the second century BC (Lancel, Robine, et al. 1980; Lancel 1983), relate to a standard plan: a lateral corridor leads into a central inner courtyard around which a number of rooms 18

MARGHERITA CARUCCI In some Romano-African houses the vestibulum does not lead directly onto the peristyle, as a room or corridor lies between the entrance and the courtyard. As far as archaeological evidence shows, the intervening room/ corridor does not have parallels with the contemporary peristyle houses in the western Empire. Here the entrance may be made of two unequally sized elements, which communicate with each other: the one is a small space in front of the domestic building and the other is the vestibulum. However, the relation of those spaces is functionally different from that in the African house; as in the houses of the western provinces, the front room is a transitional space between the street and the reception room of clients (vestibulum). In the North African house the two elements are in reverse order, as the vestibulum is the front room close to the street and the adjoining room/ corridor is a passageway.

separated them from the main reception spaces (entrance, courtyard, and main room). In the Romano-African houses the intervening room/corridor seems to have been arranged according to the same concept of seclusion, as it enhanced the privacy of the family and protected it from the view of outsiders and of the clients received in the vestibulum. However, the transitional space between vestibulum and peristyle may have fulfilled a further function in relation to the guests. It may have enhanced the lavish layout of the entrance and made the house look bigger to the eyes of guests walking through the vestibulum, room/corridor and peristyle as far as the triclinium. The organisation of those spaces has precedents in the Hellenistic palaces, such as in the Palace at Larisa (Nielsen 1994: 66-70 cat. 9; Kutbay 1998: 4-6), in the Palace at Vergina (Nielsen 1994: 81-4 cat. 10; Kutbay 1998: 18-28), and in the Palazzo delle Colonne at Ptolemais (Nielsen 1994: 146-52; Kutbay 1998: 55-7). Although the entrances of the Hellenistic palaces are axial in relation to the central court, their layout shows a striking similarity with the arrangement of the entrance rooms in the Romano-African house. A similar entrance layout has also been found in the House of Dionysos at Paphos (Kondoleon 1995), which is dated to the late

The organisation of the intervening room/corridor may have been influenced by a local tradition, drawn from a Punic heritage. In the Punic house, space was structured according to the principle of seclusion: the corridor-like entrance to the side of the house removed the view of the domestic core from outside, and the most secluded rooms were located on either side of an intervening room that

FIGURE 2 MAISON D’ICARIOS, OUDNA (AFTER MUTH 1998, FIG. 25) 19

SOMA 2004 second century AD. Here the proper vestibulum was entered from an antechamber and followed by a room, which, in axis with the western porticus, served as transition between the entrance area and the peristyletriclinium complex. It is interesting to note that the plan and the mosaic decoration of the House of Dionysos reveal the influence of North-African models.

into new markets in the eastern Mediterranean from which the Africans imported marble and architectural design of public buildings (Pensabene 1986). The trade intensified the relations that already existed between the two communities: epigraphical evidence testifies to an increasing array of people of eastern Mediterranean (and especially of Syrian) origin in North Africa from the first century AD (Lassère 1977: 397-405; Lassère 1982). In virtue of those relations a similar arrangement of triclinia was adopted in the domestic buildings of the two provincial elites. Unfortunately, as the Antiochene and African houses are contemporary and archaeological evidence in the Roman east is scanty, it is difficult to ascertain who influenced whom. Africans might have exported the architectural design of their triclinia in their trade route to Ostia and Antioch; on the other hand, they might have adopted the fashions set by Antiochene elites, which in turn were copied by the wealthy of Ostia.

Triclinium The triclinium of the North-African house shares the architectural features of the dining room of the Roman peristyle house in Imperial times. The importance of the reception room was emphasised by its location off the peristyle, its size, its monumental furnishings and its lavish floor mosaic. The monumental façade towards the peristyle is one more common feature of all triclinia: the hall usually opened onto the courtyard through three doors, the middle one of which was wider and was framed by two piers. The façade type of decoration in triclinia recalls the main halls of Hellenistic houses on Delos (Charmonard 1922; Bruneau and Vatin 1970; Hoepfner and Schwandner 1994: 293-7) and the oblong halls of Dura Europos (Hoepfner and Schwandner 1994: 274-87) where the middle door was framed by two side windows.

The lavish and monumental setting of the triclinium may have been enhanced by an interior colonnade along three sides, as demonstrated at Maison de Neptune at Acholla (Gozlan 1992), Maison des Masques at Sousse (Foucher 1965) and Maison au Grand Oecus at Utica (Duliere 1974:118). Triclinia of this kind can be found in the first century BC to the first century AD houses at Pompeii, such as in Casa del Labirinto (Richardson 1988: 164-7; Gros 2001: 64), in Casa delle Nozze d’Argento (Richardson 1988: 1559; Gros 2001: 63), and in Casa del Meleagro (Richardson 1988: 318-22; Gros 2001: 64.). There is no doubt that the arrangement of the colonnaded triclinia were modelled on the great halls of the Hellenistic palaces (Nielsen 1998), such as the Palazzo delle Colonne at Ptolemais and the First (Nielsen 1994: 193-6 cat. 28) and Third (Nielsen 1994: 196-200 cat. 30) Winter Palaces at Jericho. The idea that this layout might have taken over from the Hellenistic palaces of the East seems to be supported by Vitruvius (De architectura 6.3.9) who describes the reception rooms of this kind as aegyptii and cyziceni.

The special importance of the triclinium as space for the presentation of the owner’s self-image was also underlined by fountains and pools in the opposite courtyard. Their magnificent décor, which emphasised the natural aspect of the peristyle, was intended for the eyes of guests entertained in the triclinium. The pools of the peristyle houses show different forms, which reflect the individual tastes of the owners: in domestic architecture of North Africa, the semicircular basin opposite the main entrance of the triclinium is the most widely occurring form. The architectural layout of the main dining room shows a strong similarity with the second century to fourth century houses of the capital of the Roman province of Syria, Antioch (Kondoleon 2000). Here one feature appears to have been standard: a semicircular nymphaeum set into a colonnaded court facing the triclinium (Stillwell 1961; Dobbins 2000). Unfortunately, because only the cores of the houses were excavated owing to the great haste of the expedition and the extensive pillaging of their walls, their full plans are unknown.

Peristyle The courtyard, as an open space allowing air and light into the surrounding rooms, is a common feature in the domestic architecture of the ancient Mediterranean. In the traditional Italic house the main courtyard (atrium), off the vestibulum, fulfilled a utilitarian function with its large opening in the roof to admit light and to channel rainwater into the impluvium below. But it had a public function, too, as it was the space where clients were received, owing to its location off the tablinum.

A similar arrangement can be found in the third and fourth century houses at Ostia (Becatti 1948; Meiggs 1977: 235-62; Pavolini 1986), where the designs are attributed to the influence of Syrian merchants. However, as a similar association of triclinia with semicircular basins is documented in the earlier houses of North Africa, it is arguable that Ostian domestic architecture was influenced by North African models. In fact, as the port of Ostia was Rome’s harbour, where the African products (grain, olive oil, wine, wild beasts and pottery artefacts) were traded, the contacts between Ostian and African people were frequent and became intense in the third century. In the late second century the African trade expanded

The traditional Italic house with an atrium is unknown in North Africa. The core of the Romano-African house is the peristyle as complementary space to the rooms in which guests were received. The absence of atria in North Africa proves that African domestic architecture was not a mere by-product of Italic architecture. Moreover, the arrangement of the peristyle in African houses was not modelled upon the Italic house, as the first examples 20

MARGHERITA CARUCCI of a peristyle house in Africa have been documented at Kerkouane, a Punic city that was abandoned in the third century BC. There is no doubt that African elites brought the colonnaded courtyard into their houses directly from Greek architecture without waiting for the Roman conquest.

and Millett 1990; Millett 1990; Wood and Queiroga 1992; Freeman 1993; Metzler, Millett, et al. 1995). In the present work the term Romanisation is used with some reluctance simply to draw out the links between the elements that emerged in African domestic architecture and that of other areas of the Roman Empire.

The first examples of peristyle domestic buildings are the fourth century BC houses at Olynthos and Eretria (Hoepfner and Schwander 1994: 68-113; Nevett 1999: 53126). When, in imperial times, the architectural layout of the peristyle placed at the centre of the private building spread throughout Italy (Ostia) and the Roman provinces, African élites developed a feature that they had sporadically introduced in their houses three centuries earlier. The model for the new arrangement of the peristyle does not seem to have been the Classical and Hellenistic house of Greece. In fact, the pastas (the deepest and longest porticus off the main living rooms), the asymmetrical position and the size of the peristyle (often smaller that the largest room in the house) are architectural elements that do not appear in the colonnaded courtyard of the Roman house. This, on the contrary, seems to have been modelled upon Hellenistic palaces, where the peristyle was the central and largest unit in the building.

The analysis of African houses shows that a local identity constantly emerged alongside the general trend towards a uniform Imperial culture (Bénabou 1976; Mattingly and Hitchner 1995: 204-09): in the arrangement of the main reception rooms Punic and Hellenistic traditions continued to be asserted. The peristyle houses at the Punic sites of Kerkouane and Carthage and their form of urban life show that Africans had taken part in the Mediterranean culture dominated by Greece before Roman conquest. When the African population was integrated into the Roman world, that pre-existing cultural relationship intensified: Punic elements and Greek ideas continued to affect the Roman prototypes and created a very distinctive regional identity under Roman rule.

Bibliography Balti, J. ed. (1984) Apamée de Syrie, Bruxelles, Diffusion de Boccard. Becatti, G. (1948) Case ostiensi del tardo impero, Roma, Libreria dello Stato. Bénabou, M. (1976) La résistance africaine à la romanisation, Paris, F. Maspero. Beschaouch, A., Hanoune, R. and Y. Thébert (1977) Les ruines de Bulla Regia, Roma, École française de Rome. Blagg, T. and M. Millett eds. (1990) The Early Roman Empire in the West, Oxford, Oxbow. Bruneau, P. and C. Vatin (1970) Délos, vol. 27, L’ilot de la Maison des Comédiens, Paris, Éditions de Boccard. Charles-Picard, G. (1959) La civilisation de l’Afrique romaine, Paris, Plon. Charmonard, J. (1922) Délos, vol. 8, Le Quartier du Théâtre, Paris, Éditions de Boccard. Daniels, R. (1995) Punic Influence in the Domestic Architecture of Roman Volubilis (Morocco). Oxford Journal of Archaeology, 14.1, 79-95. Darmon, J.P. (1980) Nymfarum domus. Les pavements de la maison des Nymphes à Néapolis (Nabeul, Tunisie) et leur lecture, Leiden, E.J. Brill. Dobbins, J.J. (2000) The Houses at Antioch. IN: Kondoleon, C. ed. Antioch: The Lost Ancient City, Princeton, Princeton University Press in association with the Worcester Art Museum, 51-61. Duliere, C. (1974). Utique. Les mosaïques in situ en dehors des insulae I-II-III, Corpus des Mosaïques de Tunisie, I. 2, Tunis, Institut National d’Archéologie et d’Art. Ellis, S. (1988) The End of the Roman House. American Journal of Archaeology, 92, 565-76. Ellis, S. (2000) Roman housing, London, Duckworth. Etienne, R. (1960) Le quartier nord-est de Volubilis, Paris, Éditions de Boccard.

Conclusions The African peristyle house was the product of the interplay of Roman and African cultures. The arrangement of the peristyle and the lavish layout of the triclinium are common features of domestic buildings of all provinces of the Imperial period. Moreover, the axial plan, the square vestibulum, and the triclinium with interior colonnade are architectural elements that African houses share with other domestic buildings in the Roman empire. Thébert explains the international features of Roman domestic architecture as the product of theoretical reflection (1987: 326-47). In making plans and proposals, both architect and client referred to a highly elaborate architectural theory, which was neither influenced by climate nor by availability of building materials or other forms of local limitations. The result was a remarkable unity of Mediterranean domestic architecture, which was, in turn, the consequence of the social and cultural homogeneity of the urban upper classes. Throughout the empire the ruling classes designed their homes to facilitate living in the Roman manner and to symbolise their participation in the government of the Empire. Scholars have defined that process whereby elites from all parts of the Empire expressed their romanitas through a shared vocabulary of art, architecture, and even social habits as ‘Romanisation’. The word is a loaded term, as it implies a one-way process. Recent studies of Romanisation, which explain the processes of cultural development in individual Roman provinces from different perspectives (political, economic, social, and ideological) show that opinions vary about the mechanism and the outcome of those processes (Blagg 21

SOMA 2004 Fantar, M. (1987) Kerkouane: une cité punique au cap Bon, Tunis, Maison tunisienne de l’édition. Foucher, L. (1965) La Maison des Masques à Sousse, Tunis, Imp. du Secrétariat d’Etat aux affaires culturelles. Freeman, P.W.M. (1993) “Romanisation” and material culture. Journal of Roman Archaeology, 6, 438-45. Gozlan, S. (1992) La maison du triomphe de Neptune à Acholla (Botria, Tunisie), I, Les mosaïques, Roma, École française de Rome. Graham, W.J. (1966) Origins and Interrelations of the Greek House and the Roman House. Phoenix, 20, 332. Gros, P. (2001) L’architecture romaine du début du IIIe siècle av. J.-C. à la fin du Haut-Empire. 2: Maisons, palais, villas et tombeaux, Paris, Picard. Hales, S. (2003) The Roman House and Social Identity, Cambridge, Cambridge University Press. Hanoune, R. (1980) Recherches archéologiques francotunisiennes à Bulla Regia. 4: Les mosaïques, Roma, École française de Rome . Hoepfner, W. and E.L. Schwander, (1994) Haus and Stadt im klassischen Griechland, München, Deutscher Kunstverlag. Kondoleon, C. (1995) Domestic and Divine. Roman Mosaics in the House of Dionysos, Ithaca-London, Cornell University Press. Kondoleon, C. (2000). Antioch: The Lost Ancient City, Princeton, Princeton University Press in association with the Worcester Art Museum. Kutbay, B.L. (1998) Palaces and Large Residences of the Hellenistic Age, Lewiston-Queenston-Lampeter, Edwin Mellen Press. Lancel, S. Robine, G. and J.P. Thuillier (1980) Town Planning and Domestic Architecture of the Early Second Century B.C. on the Byrsa, Carthage”. IN: Pedley J.G. ed. New Light of Ancient Carthage: papers of a symposium sponsored by the Kelsey Museum of Archaeology, the University of Michigan, marking the fiftieth anniversary of the Museum, Ann Arbor, University of Michigan Press, 13-27. Lancel, S. (1983) La Colline de Byrsa à l’époque punique, Paris, Éditions Recherche sur les civilisations. Lassère, J-M. (1977) Ubique populus: peuplement et mouvements de population dans l’Afrique romaine de la chute de Carthage à la fin de la dynastie des Sévères (146 a.C.-235 p.C.), Paris, Éditions du Centre national de la recherche scientifique. Lassère, J.-M. (1982) L’organisation des contacts de population dans l’Afrique romaine, sous la République et au Haut-Empire. In: Temporini, H. ed., Aufstieg und Niederang der Römischen Welt II. 10. 2: Politische Geschichte, Berlin-New York, Walter de Gruyter, 397426. MacDonald, W. (1965) The Architecture of the Roman Empire I. An introductory study, New Haven and London, Yale University Press. Mattingly, D.J. and R.B. Hitchner (1995) Roman Africa: an Archaeological Review. Journal of Roman Studies, 85, 165-213.

Meiggs, R. (1977) Roman Ostia, 3rd ed. Oxford, Clarendon Press. Metzler, J., Millett, M., Roymans, N. and J. Slofstra eds. (1995). Integration in the early Roman west: the role of culture and ideology, Luxembourg, Musée national d’histoire et d’art. Meyer, K. E. (1999) Axial peristyle houses in the western empire. Journal of Roman Archaeology, 12, 101-21. Millett, M. (1990) The Romanization of Britain. An Essay in Archaeological Interpretation, Cambridge, Cambridge University Press. Muth, S. (1998) Erleben von Raum – Leben im Raum: zur Funktion mythologischer Mosaikbilder in der römischkaiserzeitlichen Wohnarchitektur, Heidelberg, Verlag Archäologie und Geschichte. Nevett, L. (1999) House and Society in the Greek World. Cambridge, Cambridge University Press. Nielsen, I. (1994) Hellenistic Palaces: Tradition and Renewal, Aarhus, Aarhus University Press. Nielsen, I. (1998) Royal Banquets: The Development of Royal Banquets and Banqueting Halls from Alexander to the Tetrarchs. IN: Nielsen, I. and H. S. Nielsen eds. Meals in a Social Context. Aspects of the Communal Meal in the Hellenistic and Roman World, Aarhus, Aarhus University Press, 102-33. Pavolini, C. (1986) L’edilizia commerciale e l’edilizia abitativa nel contesto di Ostia tardoantica. IN: Giardina, A. ed. Società romana e impero tardoantico. II: Roma: politica, economia, paesaggio urbano, Bari, Laterza, 239-97. Pensabene, P. (1986) La decorazione architettonica nell’Africa romana. Studio preliminare sui capitelli. IN: Giardina, A. ed. Società romana e impero tardoantico. II: Roma: politica, economia, paesaggio urbano, vol. I, II o III, Laterza, Bari, 358-429. Rebuffat, R. (1969) Maisons à péristyle d’Afrique du Nord: répertoire de plans publiés I. Mélanges de l’Ecole française de Rome. Antiquite, 81, 659-724. Rebuffat, R. (1974) Maisons à péristyle d’Afrique du Nord: répertoire de plans publiés II. Mélanges de l’Ecole française de Rome. Antiquite, 86, 445-99. Richardson, L. (1988) Pompeii: an Architectural History, Baltimore-London, Johns Hopkins University Press. Romanelli, P. (1970). Topografia e archeologia dell’Africa Romana, Torino, Società editrice internazionale. Stillwell, R. (1961) Houses at Antioch. Dumbarton Oaks Papers, 15, 45-57. Thébert, Y. (1987) Private Life and Domestic Architecture in Roman Africa. IN: Veyne, P. ed. A History of Private Life. I: From Pagan Rome to Byzantium, CambridgeLondon, Belknap Press of Harvard University Press, 313-409. Thouvenot, R. (1958) Maisons de Volubilis: le palais dit de Gordien et la maison à la mosaïque de Vénus, Rabat, Service des antiquités du Maroc. Wood, M. and F. Queiroga eds. (1992) Current Research on the Romanization of the Western Provinces, Oxford, BAR Publishing.

22

MARGHERITA CARUCCI Vitruvius (Translated by Frank Granger 1931) On architecture, London, Loeb Classical Library

Ancient Sources Tacitus (Translated by Anthony R. Birley 1999) Agricola and Germany, Oxford, Oxford University Press.

23

Homesickness and Greek Architecture in a Colony: The example of Cyrene (Libya) Luca Cherstich

This paper is a by-product of a doctoral thesis on the necropolis of Cyrene, a Greek colony in Libya. Attention is focused here on one of the most underdeveloped sides of the interpretation of Cyrenean architecture: its link with the desire to display group affiliation and cultural identity.

creation, the orders kept their significance and probably a certain link with their regional origins, at least in temple architecture. In any case, the picture of Greek architecture in the Mediterranean was incredibly more complicated and dynamic, especially in colonial settings like Sicily and Southern Italy (Barletta 1983, 1990, 2000, 2001). When, during Archaic times, colonists arrived in the new lands the orders were still in formation: no fixed architectural tradition was brought with the newcomers. In any case, links throughout the Greek world were strong. Trade, migrations of artisans and meetings in panhellenic sanctuaries made influences between styles possible, and it is quite clear that colonies played an important role in the formation of the orders during Archaic times. Colonies had their own versions of Greek architecture, in which the orders were often blended more than in the homelands. It is quite clear that the typical temples of Magna Graecia were Doric but of a kind of Doric that was alien to the Greeks of the mainland.

Before starting, a methodological note is required: the architectural orders will be used here as interpretative tools. Few general considerations will be made. For more information the reader is referred to Barletta (2001) and, to a lesser degree, to Lawrence (1996), and Coulton (1977). In Greek times the geographical distribution of the Doric and Ionic orders did not reflect any ethnic division between Ionians and Dorians. Various examples testify to this. The Dorian inhabitants of Knidos preferred Ionic architecture while Athens used the Doric order for her major temples, in spite of her ethnic links with the Ionians. The division between the two orders was regional, not ethnic: Doric was the order of the mainland and Ionic the order of Eastern Greeks (Barletta 1983: 3-4; 2001: 3, 153-6). Influences between the systems are well known, especially during the Archaic (Barletta 2001) and Hellenistic periods (Lauter 1999: 235-7). Although the orders were never as fixed as was once thought, the division still contains some truth. For example, in Hellenistic Ionia the Doric order was widely used in civic architecture (numerous examples are cited in Coulton 1976) while the major temples were still Ionic (for example, the second century BC temple of Artemis Leukophryene, at Magnesia, in Politt 1986: 246), the traditional style of Eastern Greeks. The reverse is true in the mainland where no major Ionic Greek temple is known.

This dynamism of the orders during Archaic times reflects the birth of local visual models in architecture. The reasons why different styles were chosen in the creation of such models are worth investigating, especially where colonists came from various different ethnic backgrounds. Different groups with different tastes, ideas or origins could favour different visual models. One must also consider the different purpose of different buildings. Public architecture represents the choices of the whole polis or, at least, of the government. On the other hand, private architecture represents the choices of an individual or of a minor group choosing to agree or disagree with the model proposed by the city.

The decision to keep traditional visual models for temples, the most important buildings on which a polis could spend money, means that even during the great experimentations of Hellenistic times, such models were still felt to be the most proper ones to use for a god. Even centuries after their

All of these considerations will be examined here in a colony during Archaic times, Cyrene in Libya, in an attempt to speculate about the possible historical meanings of the different visual models in architecture which were chosen.

25

SOMA 2004 Stucchi stressed the prevalence of the Doric order in Cyrene, limiting the importance of the Ionic order, for the most part, to later times (1975: 82-3, 189). This seems to be confirmed by the images of the Archaic Apollonion (Stucchi 1961; 1975: 16-19, Fig. 10) or the late-Archaic temple of Zeus (Bacchielli 1998; Stucchi 1975: 20, 23-9). These were the most important public temples of Cyrene, showing the way in which Cyrene, as a polis, appeared; mighty Doric temples reminiscent, in some respects, of what can be seen in the Peloponnese. This is hardly a surprise. Cyrene was founded by colonists from Thera who claimed descent from Sparta. Furthermore, there is also literary and archaeological evidence for the relevant presence of Laconians, or at

least Peloponnesians, in Archaic Cyrene (Schaus 1985a; 1985b: 98-102). However, if certain rock-cut tombs (Fig. 1) are considered, a different story is told. Tomb N8 (Thorn forthcoming: II.4, VII.1.3, 2.3, 3.3; Stucchi 1975: 39, 41, figs. 27-8, 32d) has a distyle façade with Ionic capitals. Furthermore, the mouldings on the pediment and internal door (Fig. 2) are the Lesbian leaf, rendered as a kyma reversa, and the bead-andreel, as a half-round, as was typical in Ionia. Particularly interesting are the corner-volute akroteria, an Ionic feature with comparisons in Larisa, Miletus, Lesbos and Phrygia (Winter 1993: 247, 258). The general appearance of this façade (especially the position of corner-volutes), finds comparisons with a sarcophagus from Samos (Barletta 2001: Fig. 63). The presence of such a style in Cyrenean tombs implies that some families wanted their dead to be commemorated in a purely Ionic way, differing from the Doric model displayed by the city in her major temples. This datum suggests the need for a reconsideration of the Ionic style in Archaic Cyrene in the hope of assessing the degree to which local architecture accepted models and elements not completely respondent to the ‘mainland Doric’ style. We will run quickly through such elements. One can notice, for example, decorative features like the little Aeolic capitals on the sides of an altar in the Agorà (Purcaro 2001: 37-45) or the Gorgoneion with lyre-volutes that acted as the central akroterion for the first Apollo temple (Pernier 1935: 56-8, figs. 52, 54). Although the Gorgon could be linked with Laconia (Fabbricotti 2000: 188) the source of this central volute

FIGURE 1 FAÇADE OF TOMB N8 (AFTER THORN FORTHCOMING)

FIGURE 2 MOULDING ON THE INTERNAL DOOR OF TOMB N8 (AUTHOR’S OWN PHOTO) 26

LUCA CHERSTICH Particularly interesting is an extra-mural sanctuary (White 1971; Stucchi 1975: 21), demolished probably after the coming of the Persians in 515 BC. There is a sphinx on an Ionic column (Fig. 3) (recently restored by the University of Chieti) dated to around mid-sixth century BC. The lack of eyes in the volutes finds comparison with the sphinx columns of Delos and Delphi (probably both erected by the Naxians) and with the capitals of the first temple of Apollo at Didyma (Barletta 2001: figs. 57-8, 71). However, the double series of egg and dart in the echinus is unique and the only decent comparison is with tomb N8 which probably imitated it, according to Prof. White’s interpretation (1971: 51, figs. 5-7, pl. 10). Although the capital seems to indicate island or East Aegean workshops, one must not forget that the real shape of this sanctuary has not yet been identified, so it is impossible to say how ‘Eastern Greek’ this architecture was. Further indication could be found in another medium, the sculpture found in the sanctuary, which seems to indicate island or Eastern Greek artists (Pedley 1971). However, one must not forget possible links with Attic workshops, especially for the body of the sphinx. Furthermore, the presence of relief bronze panels points to the Peloponnese, especially for the Gorgon relief (from personal conversations with Prof. E. Fabbricotti and Dr. O. Menozzi). It is impossible to say how Ionic the architecture of this little sanctuary was. Even if it was completely Ionic, one must not forget that it could have been a private structure reflecting the tastes of a few rich families. In summary, the evidence for the use of the Ionic style in public architecture is not so clear in Archaic Cyrene. For the most part, the Ionic order seems to be a minor

FIGURE 3 THE CYRENE SPHINX (ARCHIVE OF THE ITALIAN ARCHAEOLOGICAL MISSION OF CHIETI UNIVERSITY IN CYRENAICA) akroterion is linked with the corner akroteria of tomb N8. It was an element which appeared in various East Greek sites before the end of the sixth century (Barletta 1983: 155 note 83; Kjelberg 1940: 133-4, Fig. 40, pls. 69, 70.2-3; Akerstrom 1966: 30, pl. 11.1-2; von Gerkan 1925: 21-3, figs. 13-15). In Archaic Cyrene, among the non-purely mainland Doric elements, it is possible to count some more dubious ones like the first altar of Apollo, which is stepped. Its style is often called ‘Ionic’ (Parisi-Presicce 1991: 159-61, pl. XLIX.b-c) as the possible origin of the type. Also uncertain are the probable prostyle plans without peristyles of the first temples of Apollo and Zeus (Stucchi 1975: 16-8, 19-20), which could be tentatively linked with Cycladic plans (Barletta 2001: 120; Gruben 1972: 27-8).

FIGURE 4 TOMB S80 (AFTER THORN FORTHCOMING) 27

SOMA 2004

FIGURE 5 TOMB W28 (PHOTO BY EUGENIO DI VALERIO) component in major Doric settings. More interesting is the Ionic component in tombs; their nature as private monuments helps in defining visual models that differ from the public one. There are two kinds of Archaic tombs: tumuli and rock-cut tombs.

S80 suggests the first decade of the fifth century BC. The change between old and new tumuli could be linked with the modification of Battus’ heroon (Stucchi 1965: 11115), testifying that minor tumuli were still linked with the image of Battus’ grave.

Tumuli The first known tumulus is that of Battus, the founder of Cyrene (Stucchi 1965: 58-65). Pottery dating to the end of seventh BC century offers a chronology coherent with the literary evidence for that king. Battus’ tomb was sited in the Agora, the centre of the city, where it received a cult. Because of its position and the nature of the dead, the tomb certainly had a particular significance for the Cyrenaeans, probably acting as a model. No other tumulus of this date is known in Cyrene (and even this datum could be significant, in a society in which complex funerary display was probably limited to a small elite), but a slightly later one was found in Messa, in the Cyrenean chora (Stucchi 1964).

Battus guided the first group of colonists coming from the Dorian island of Thera; probably the use of the model set by Battus expressed links with the original colonists. This also seems to be supported by the use of cremation in both tumulus types (as testified by the excavations of Stucchi and Beschi). Inhumation was later the predominant burial rite in Cyrene, but the use of cremation in Archaic tumuli may have had a particular meaning if one considers that in Iron Age Thera cremation was the only rite used (Boardman 1999: 25). Rock-Cut Tombs On the other hand, rock-cut tombs are completely exotic symbols: Cyrene was probably the first Greek polis to use them. In Archaic times, there is no comparison in Greek lands (where they started to be diffused in Hellenistic times) but only in the East (Fedak 1990: 46-56).

A later type is more widespread in Cyrene. A good example is tomb S80 (Fig. 4) (Thorn forthcoming: II.12; Rowe 1956: 6-7; Beschi 1972: 150-68, figs. 10-23), where the mound is enclosed by a round wall of orthostats and minor debris. Tumuli like this can be dated to the end of the sixth-beginning of the fifth century BC; their building technique recalls the Archaic sanctuary of Demeter in the Agora (Bacchielli 1981: 27-39) and the pottery in

Persians used the model for their kings, carving façades with false porticoes. They ruled Egypt during the second part of the sixth century BC, and Cyrene possibly adopted the custom from them (Boardman 1999: 156-7). In any case, 28

LUCA CHERSTICH rock-cut tombs with real porticoes are attested in Egypt already from the second millennium BC (Dodson 1991: 11), and Cyrenaeans could have visited the relatively close land of Egypt even before the Persian conquest. Alternatively, it is possible that these are the graves of colonists who came from areas closer to Asia Minor where pre-Hellenistic tombs with rock-cut façades (but not porticoes) existed, for example in Phrygia (Haspels 1971: 134-8).

display messages for travellers in the vicinity of Cyrene. Probably no more than half a century later, the portico motif began to be rendered by half-columns alone (Fig. 5). Although Stucchi does not admit the Archaic date of this second type of rock-cut tomb (1975: 172-3), I agree with the archaic features noticed by both Beschi (1972: 146-7) and Thorn (personal conversation, and in Thorn forthcoming: IV.1.2). These tombs must be Late Archaic at the very least.

Whatever their origin (which will probably remain a mystery), with or without the portico, the Greeks used the model, adapting it through more familiar shapes: the portico and the classical orders. The Cyrenean tombs were shaped according to Greek architectural vocabulary, but they are still exotic kinds of burials, showing the deliberate choice of foreign visual models. They certainly were powerful symbols: they were all laid on the most important routes just outside city-walls, so their façades were meant to

Furthermore, if one compares the concentration of the first type of tomb (Fig. 6) in two north-western areas with that of the second type, it is clear that they represent two different phases of the same burial groups. On the other hand, tumuli are separated by their sparse position in southern cemeteries. This could be due to the nature of the tombs: tumuli cannot be built in the steep northern cliffs used for rock-cut tombs.

FIGURE 6 MAP OF THE ARCHAIC TOMBS (DRAWN BY THE AUTHOR BASED ON THE MAP IN DENT 1985) 29

SOMA 2004 Rock-cut tombs were an alternative to tumuli, and this is proved by their contemporary chronology. Stucchi (1975: 41) and Boardman (1999: 157) date portico tombs to the last quarter of the sixth century BC, which is confirmed by the possible imitation of the sphinx capital in tomb N8, a date roughly contemporary with the tumuli of secondtype. In the same time period different groups used rockcut tombs and tumuli; they represent alternative models and different display strategies.

1) Therans and the Perioikoi 2) Cretans and Peloponnesians 3) Islanders These islanders were possibly linked with rock-cut tombs, nearly all of which have Ionic or even Cycladic-Ionic styles. If the model of the rock-cut tombs is eastern, an origin from the islands near Ionia is plausible, due to the relationships that Eastern Greeks had with the East, not only in Asia Minor but also through ports like Al Mina in Syria and, more significantly for Cyrene, Naucratis in Egypt (Boardman 1999: 35-159).

While nearly all tumuli were meant for single graves, portico-tombs were designed for a larger number of graves (from three to six). While in any tumulus the identity of a single deceased is displayed, rock-cut tombs are the communal representation (in a single façade) of all the various dead buried inside. This sense of ‘community’ is further stressed by the spatial organisation of rock-cut tombs in contiguous rows of seven or eight tombs, while tumuli are usually alone or organised in couples suggesting more individual models. Furthermore, if one counts all the Archaic tombs, hypothesising one dead per tumulus and about four per rock-cut tomb, it is clear that in the fifty or one hundred years between the sixth and fifth centuries we have about 35 dead buried in tumuli and about 90 in rockcut tombs (considering only portico-tombs) and 200 if we add those buried in the second type of rock-cut tombs. These ridiculous numbers clearly show that the greater part of the Cyrenean population was buried in archaeologically invisible ways roughly similar to those suggested by Morris (1987) for Iron Age Athens. However, it is significant that the visible elite buried in rock-cut tombs are three to seven times more numerous than in tumuli.

Which islands are treated here is more difficult to say. Literary evidence is (obviously) unclear. Herodotus (IV, 163) describes a later arrival of mercenaries from Samos after the promise of land. There are also generic relationships with Rhodes, Knidos and Cyprus (Herodotus IV, 162-4) but no indication of any link with the migrants mentioned in Herodotus IV, 161. We know that the Cyrenaican dialect, in later times, was strongly Dorian, but these islanders need not be Ionian as they could have come from Dorian islands using the Ionic order. The Ionic features in tombs show generic links with the Cyclades and Ionia but we cannot go further in identifying the exact sources. The presence of Aeolic capitals linked with pilasters (and not columns) (Stucchi 1975: Fig. 32c; Rowe 1959: pl. 17b) recalls examples from the Halicarnassos peninsula (Betancourt 1977: 32-3) and perhaps the nearby Dodecanese areas, where a Doric dialect was spoken. However, because the volutes do not have parallels, except for the Temple of Apollo’s Gorgoneion (Pernier 1935: Fig. 52, 54), this type was only Cyrenean. Tomb W48 shows a Lesbian-leaf capital (Stucchi 1975: Fig. 155 a, b, wrongly dated as Hellenistic) which is comparable with examples from Thasos and Larisa (Betancourt 1977: pls. 37, 42 wrongly reconstructed with a volute-block) both datable to the second half of the sixth century. Furthermore, possible links with the Cyclades cannot be excluded (Barletta 2001: 110).

No actual body has been excavated in any Archaic rockcut tomb, but such tombs have features which could suggest inhumation: rock-cut klinai (funerary beds), rockcut sarcophagi and rectangular pits whose dimensions fit those of an adult man. This difference of rite (inhumation and cremation) and the fact that a different percentage of the family is archaeologically visible (one in a tumulus, about four in a rock-cut tomb) suggests that we are dealing with two groups with different funerary ideals and, possibly, different familial structures and ethnic groups. The different visual models (exotic or traditional) suggest that these groups have different origins, or at least desire to show different things in funeral rites. The prevalence of the Doric style in public architecture and of the ‘Volute’ style (Ionic or Aeolic) in rock-cut tombs points out that the second, untraditional group was, in some sense, a minority who chose different visual models to represent themselves. But who were they?

In any case, things are not so straightforward. Not all rock-cut tombs are Ionic and four of the 22 portico-tombs show Doric (not Ionic) columns (Stucchi 1975: Fig. 28; Thorn forthcoming: II.4). They are all set together forming a coherent group that seems to be earlier than the Ionic tombs because the Ionic ones cut the space of three Doric tombs (Thorn II: 2.4.4). However, even if Doric, these tombs were still an alternative to the tumuli. Furthermore, for crafting an Ionic (or Doric) style it is more important to have a skilled craftsman than a wealthy purchaser. Doric tombs could be from a time when Doriccrafting artisans were the only ones available. All later portico tombs show Ionic styles suggesting that when the Ionic style was available, this group deliberately choose it, maybe in the last quarter of the sixth century BC, when the influx of Ionic craftsmen and objects (like the sphinx capital) was perhaps stronger. Although their old tombs

From Herodotus (IV, 159) we know that during the reign of Battus II new colonists arrived from many parts of the Greek world. The complexity of ethnic composition of the Cyrenean colonists seems to be confirmed by finds of pottery (Schaus 1985b: 96-105). Later, (Herodotus IV, 161) during the reign of Battus III (mid-sixth century), Damonax of Mantinea divided the Cyreneans into 3 groups: 30

LUCA CHERSTICH were Doric, the ‘strange colonists’ were ready to accept the new styles to distinguish themselves. In this sense the ‘homesickness’ in my title does not intend to suggest that they brought with them the Ionic style (still in formation in the sixth century) but that they kept links with the homelands where the style was developing and may have participated in its formation. Consequently, they chose such a style for a new, un-Greek kind of monument: the rock-cut tombs which helped them in defining themselves from the other colonists.

Hyblaea (Barletta 1983: 154, 162) in Sicily where, like in Cyrene, it appears first in a private monument (a funerary naiskos), probably due to private choices made differently from the public ones. Barletta demonstrated that the Ionic style arrived at Dorian Sicilian colonies from the eastern coast in the mid-sixth century BC in pure Ionic form, and then, during the Late Archaic period, it was mixed with the Doric style creating the new regional styles, unheard-of in the original homelands of the colonists (Barletta 1983: 340-50).

However, even the Doric style in these early rock-cut tombs is different from the official Cyrenean Doric of the temples. They lack any entablature and tombs N5 and N9 have unfluted columns, a typically Cycladic feature (Barletta 2001: 62). Furthermore, N7 has fluted columns but sited on bases: a typical Ionic and not Doric feature. Moreover, these bases are plain, not moulded, suggesting more the Cycladic Ionic style of the temples of Yria and Sangrì in Naxos (Barletta 2001: 84-98) than the moulded bases of the Asia Minor Ionic style. In any case we are still in a time period when the orders were fluid and still in formation all over the Greek World.

A similar situation seems to have happened in Cyrene. The identification of a group of colonists from the islands, their desire to appear ‘different’ at the beginning and the later mixture with local styles are noteworthy in this sense. The archaeological value of architecture is not its beauty but its purpose which is linked with the ways in which the ancients wanted to appear. It helps the archaeologist to formulate hypotheses that have value for social history. In Cyrene the two orders were more integrated after the mid-sixth century and this, for me, is proof of the rise of a ‘Cyrenean, communal visual model’ and, therefore, the decline of the particular visual models of singular families in tombs.

The presence of different colonists in Cyrene could have favoured, during Late Archaic times, a mixed style. This is not so clear in Late Archaic public monuments designed to represent the official (Doric) image of the polis, but it is clear in private tombs like W18 (Stucchi 1975: 170 wrongly dated as Hellenistic) in which there is a combination of Doric frieze, Ionic capitals and the pilasters used until then in Cyrene only for the ‘Aeolic’ tombs.

Bibliography Akerstrom, A. (1966) Die Architektonischen Terrakotten Kleinasiens, Lund, Gleerup. Bacchielli, L. (1981) L’Agorà di Cirene II,1. L’Area Settentrionale del Lato Ovest della Platea Inferiore, Rome, L’Erma di Breitschneider. Bacchielli, L. (1998) Il Tempio di Zeus Olimpio a Cirene. Storia e Programmi di Scavi e Restauri. IN: Catani E. and S.M. Marengo eds., La Cirenaica in Età antica, Rome-Pisa, 23-34. Barletta, B.A. (1983) Ionic Influence in Archaic Sicily. The Monumental Arts, Cambridge, P. Åströms förlag. Barletta, B.A. (1990) An Ionian Sea Style in Archaic Doric Architecture. American Journal of Archaeology 94, 45-72. Barletta, B.A. (2000) Ionic Influence in Western Greek Architecture: towards a Definition and Explanation. IN: Krizinger F. Die Agais und Wechselwirkugen 8. bis 5. Jh. v. Chr., Archaologische Forschungen, 4, 203-16. Barletta, B.A. (2001) The origin of Greek Architectural orders, Gothenburg, Cambridge University Press. Beschi, L. (1972) Divinità Funerarie Cirenaiche. Annali della Scuola Archeologica Italiana di Atene e delle Missioni in Oriente, XLVII-XLVIII, n.s. XXI-XXXII (1969-70), 133-341. Betancourt, P.P. (1977) The Aeolic style in architecture, Princeton, Princeton University Press. Boardman, J. (1999) The Greeks Overseas, Fourth Edition, London, Thames & Hudson. Coulton, J.J. (1976) The Architectural Development of the Greek Stoa, Oxford, Oxford University Press.

Private architecture in some senses formed the basis of the Cyrenean regional style. The local Doric architectural tradition was not pure mainland Doric, and this is clear if one looks at later monuments. Fourth century BC temples and tombs show a Doric style mixed with Ionian mouldings. Stucchi explains them as ‘Ionic influences’ (1975: 83) but he did not notice that the exact same mouldings (especially bead-and-reel and Lesbian-leaf) were already present in Archaic tombs (Fig. 2). Furthermore, when describing elements like bases under Hellenistic Doric columns, he defined the Ionic style as ‘finora rimasto escluso’ (Stucchi 1975: 189), in this way contradicting himself and not noticing that such bases were already to be found in the Archaic tomb N7. The situation presented here is much more complicated and dynamic than the simple ‘Doric orthodoxy’ usually stressed for Cyrene, and the acknowledgement of this relies on the new dynamic view of the orders that we have today. Similar dynamic situations can be seen in other colonial settings like Magna Graecia and Sicily, where colonies were more numerous than in Cyrenaica and ethnic origins more complicated. There, during the sixth century, Ionic and Doric features were brought together in unique, regional styles. It is surprising to find mixtures of Doric and Ionic styles in a strongly Dorian colony like Megara 31

SOMA 2004 Coulton, J.J. (1977) Greek Architects at Work: problems of structure and design, Ithaca NY, Cornell University Press. Dent J.S. (1985) Burial practices in Cyrenaica. IN: Barker G., Lloyd J. and Reynolds J. eds., Cyrenaica in Antiquity, Oxford, British Archaeological Reports, International Series, 327-36. Dodson, A. (1991) Egyptian Rock-cut tombs, Haverfordwest, Princes Risborough, Shire. Fabbricotti, E. (2000) Le arti decorative. IN: Bonacasa N. and S. Ensoli eds., Cirene, Milan, Electa, 187-92. Fedak, J. (1990) Monumental tombs of the Hellenistic Age: a study of Selected Tombs from the Pre-classical to the Early Imperial Era, Toronto, Toronto University Press. Gruben, G. (1972) Kykladische Architektur. Münchener Jahrbuch der bildenden Kunst, 23, 7-36. Haspels, C.H.E. (1971) The Highlands of Phrygia, Princeton, Princeton University Press. Kjelberg, L. (1940) Larisa am Hemros II: Die Architektonischen Terrakotten, Stockholm, Kungl. Vitterhets Historie och Antikvitets Akademien. Lauter, H. (1999) L’architettura dell’Ellenismo, Italian translation of “Die Architektur des Hellenismus”, Milano, Longanesi. Lawrence, A.W. (1996) Greek Architecture, 5th edition revised by R.A. Tomlinson, New Haven and London, Yale University Press. Morris, I. (1987) Burial and Ancient Society. The Rise of the Greek city-state, Cambridge, Cambridge University Press. Parisi-Presicce, C. (1991) Cirene: gli Altari del Santuario di Apollo. IN: Etienne R. and M.-T. Le Dinahet eds., L’Espace Sacrificiel dans les civilisations Méditerranèennes de l’Antiquité, Paris, 159-65. Pedley, J.G. (1971) The Archaic Favissa at Cyrene. American Journal of Archaeology 75, 39-46. Pernier, L. (1935) Il Tempio e l’Altare di Apollo a Cirene. Scavi e Studi dal 1925 al 1934, Bergamo, Istituto Italiano d’Arti Grafiche. Politt, J.J. (1986) Art in the Hellenistic Age, Cambridge, Cambridge University Press.

Purcaro, V. (2001) L’agorà di Cirene II,3. L’Area Meridionale del Lato Ovest dell’Agorà, Rome, L’Erma di Breitschneider. Rowe, A. (1956) Cyrenaican Expedition of the University of Manchester 1952, Manchester, Manchester University Press. Rowe, A. (1959) Cyrenaican Expeditions of the University of Manchester 1955, 1956, 1957, Manchester, Manchester University Press. Schaus, P.G. (1985a) Evidence for Laconians in Cyrenaica in the Archaic Period. IN: Barker G., Lloyd J. and Reynolds J. eds., Cyrenaica in Antiquity, Oxford, British Archaeological Reports International Series, 395-403. Schaus, P.G. (1985b) The extramural sanctuary of Demeter and Persephone at Cyrene, Libya. Final reports. Volume II. The East Greek, Island and Laconian pottery, Philadelphia, University of Philadelphia Museum Press. Stucchi, S. (1961) Le Fasi Costruttive dell’Apollonion di Cirene. Quaderni di Archeologia della Libia 4, 55-81. Stucchi, S. (1964) La Tomba a Tumulo presso Messa in Cirenaica. Libya Antiqua I, 127-31. Stucchi, S. (1965) L’Agorà di Cirene I: i Lati Nord ed Est della Platea Inferiore, Rome, L’Erma di Breitschneider. Stucchi, S. (1975) Architettura Cirenaica, Rome, L’Erma di Breitschneider. Thorn, J.C. (forthcoming) The Necropolis of Cyrene. Two Hundreds Years of Exploration, Rome, L’Erma di Breitschneider. von Gerkan, A. (1925) Milet I.8: Kalabaktepe, Athenatempel und Umgebung, Berlin, Verlag von Schoetz und Parrhysius. White, D. (1971) The Cyrene Sphinx, its Capital and its Column. American Journal of Archaeology 75, 47-55. Winter, N.A. (1993) Greek Architectural Terracottas. From the Prehistoric to the End of the Archaic Period, Oxford, Oxford University Press.

32

Making ‘Scents’ of Flowery Pots. Considering Minoan Ceramic Vessels with Botanical Relief1 Jo Day

Ceramics from the Bronze Age of Crete are perhaps one of the better studied bodies of archaeological material in the world. This may be due in no small part to the intriguing decoration which is found on much of the fine-ware vessels, especially those from the Minoan palaces. A large proportion of this material is decorated with elements from the plant world, from simple rosettes, to the polychrome Kamares style pottery, renowned for its complex organic motifs, and typical of the Middle Minoan period (c. 20001600 B.C., Dickinson 1994: 19), to the increasingly stylised floral motifs of the late Minoan period.

the body and two on each side between the handles (Fig. 1). Painted decoration consists of what has been traditionally interpreted as red ‘coral’ on the stem and upper bowl, and a black and white chequerboard pattern runs around the lower part of the bowl. The interior seems to have been left plain. Also around the base of the vessel, further white relief-work of schematic petals can be seen, which have the same form as those of the whole flowers—i.e. white, pointed, and with incised lines running up the middle.

Amidst this abundance of vegetation, there are a small number of vessels, which are particularly intriguing and the focus of this paper. These vessels feature relief elements derived from the world of plants—petals, rosettes and even whole flowers appear on certain pots. These seem to be mainly restricted to the period MM II-MM III, i.e. c.19001600 BC, although there are some examples of relief flowers of a different nature from outside this time span (Foster 1982: Chart 14). In this paper, these vessels will be examined more closely, and with particular attention being paid to their possible functions and the reasons for this elaborate decoration.2 Probably the most famous example of such a vessel is the ‘Phaistos krater’ (Levi 1976: Pl. XXVII). This pedestalled krater, (F1031),3 is 45.5cm high, and its bowl has a 38cm diameter, making it one of the largest pieces in this category. The excavator, Levi, dated it to his 1b period at Phaistos, equivalent to MM IIb. Large white flowers with orangey centres are attached to this vessel: three on the stem, four on 1. I am grateful to the Irish Research Council for the Humanities and Social Sciences for funding this research. 2. Karen Polinger Foster (1982) has catalogued these as part of her work on Minoan ceramic relief although this book is now over twenty years old and so omits more recent finds. 3. The excavation or publication catalogue numbers for each vessel are given: Phaistos numbers from Levi (1976), with the exception of HM5820, a number from Heraklion Museum; Ayia Triada from La Rosa (1994-96); Kommos from Betancourt (1990); Qubbet El-Hawa from Kemp and Merrillees (1980).

FIGURE 1 THE PHAISTOS KRATER (F. 1031), AFTER POSTCARD PRINTED BY EDITIONS D. HAITALIS, AVAILABLE AT HERAKLION ARCHAEOLOGICAL MUSEUM. (DRAWING BY AUTHOR) 33

SOMA 2004 While this vessel is perhaps the most well-known of those with flowers attached, there are several more decorated with similar white blossoms. A globular vase, (F3718), again from Phaistos phase 1b, and painted in the Kamares style, exhibits similar flowers, also white with red centres (Levi 1976: Pl. XLII). This vessel was found in many fragments, but there were at least four plastic flowers attached to the vessel’s shoulder. A more recently discovered piece, a pitharaki (HTR2286), comes from Ayia Triada. Pierced ribs run vertically up the vessel, and at the top of each is a flower, white with a red centre. It is suggested that there was a possible maximum of twelve flowers on this piece, although only six survive (La Rosa, 1994-96: 329, and Pl. 10a and b).

examples of the Dutch genre of still-life, but indicate the important role this bulb played for a short period of time. There the tulip became the ultimate status symbol, with bulbs exchanging hands for the equivalent of several years’ average wage. If you could not afford the real thing, a picture of it would at least show you to be a person of taste, and at a fraction of the cost (Pavord 1999: 145). Artists always have to make choices about what to depict, and on Crete, an island renowned in botanical circles for the richness of its flora, there are many hundreds of species to choose from. It must be asked then why did this white flower become a popular relief motif? One factor worth considering is that plastic decoration can also be an indicator of the contents of a vessel. Again turning to ethnography, the Pueblo Indians of America have a range of vessels which take the form of the spiky fruit capsule of the datura plant (Litzinger 1981). This plant has been used by a wide range of societies in both medicine and in magico-religious rites. Closer to home, one only need think of cream jugs in the shape of cows, or egg-holders in the shape of hens to realise that this is not an unusual phenomenon. As the Bronze Age vessels noted above are suited to fluids, it could be suggested that the beverages contained within had some floral extracts added, signposted by the relief flowers. Many societies add plant extracts, which are pharmacologically known as admixtures, to other beverages, which may then increase the potency of the brew. Indeed, it can even be read on the label of a bottle of the popular Bombay Sapphire gin that ten botanicals are added, each one bringing a unique benefit to the drink. Evidence for what exactly was drunk in the Bronze Age has been coming from ever-more detailed chemical analysis of ceramics. For example, an LM I tripod pot from Chania showed evidence of herbs like laurel, lavender or sage having been added to wine at this time, and a later pot from Mycenae had traces of wine with rue (Tzedakis and Martlew 1999: 164). Although not all scholars seem convinced by the scientific basis of this work (e.g. Hamilakis 2002a: 19), perhaps it is only a matter of time before these findings are accepted, and further plant alkaloids are recognised, which may, in turn, guide us towards finding the identity of the relief flowers.

These flowers are very similar to those found on a jug, again from the palace of Phaistos (HM5820). This time the blooms, white with red centres, sit atop two pedestals rising beside the mouth of the jug, which also has an animal head-shaped spout (Pernier 1935: Pl. XXIX). A very similar flower comes from Kommos, (C1083), dated to MM IIb. This example is only a fragment of a rim/handle join of an open vessel. However, the incised lines on the eight, white, pointed petals can be clearly seen, leading the excavators to compare it with those on the Phaistos krater, but adding that ‘the treatment here is much more sensitive’ (Betancourt 1990: 162, and Pl. 71, no. 1341). A rather exuberant version of a vessel with these ceramic flowers has been discovered in Egypt at Qubbet El-Hawa (JdE92304); this example is an Egyptian-made copy of a Minoan vessel (Kemp and Merrillees 1980: 215, Fig. 60). The jug’s mouth has been transformed into a flower with five petals, and sprouting from the shoulders of the vessel are another five flowers, all of which are white with incised petals. The painted decoration is again of the Kamares type. These vessels cannot be studied without questioning the point of decorating vessels with flowers in this manner. To do so does not add to their functional value—for example, no matter how many flowers are attached, it still remains a jug for pouring liquids. Indeed the more flowers that are attached, the harder it may be to actually use the vessel without breaking bits off. When the term ‘decoration’ is used now, it often loaded with the connotation of being meaningless, of merely painting for the sake of it, or to avoid the horror of blank space. Yet it is documented ethnographically that most traditional societies decorate ceramics, baskets, textiles, walls and even their bodies in a meaningful fashion (e.g. Barley 1994; David, Sterner and Gavua 1988). Plants form part of the artistic repertoire of many societies, and almost invariably those chosen for depiction are important to that society. For example, the Huichols of Mexico frequently portray the peyote cactus in their art. This hallucinogen is a key element in their culture, traditionally necessitating an annual pilgrimage to gather the cactus for the upcoming year (Myerhof 1974). Paintings of tulips from 17th century Holland are not just

If the flower chosen to be depicted was important to the Minoans and could be an indicator of the contents of the vessel, it is obviously essential to examine all ideas about its identity. Since the time of Arthur Evans, scholars have been playing at guessing the identities of the plants in Minoan art, with varying degrees of success. Levi, for example, sometimes called them ‘marguerites’, a name covering many species of daisy-like flowers (e.g. Levi 1976: 571). Yet over time, it has become part of Minoan lore that these flowers are lilies, based mainly on the fact that white Madonna lilies can be identified in wallpaintings, such as those from Amnisos (Higgins 1997: Fig. 105). However, this is an interpretation with which few, if any, botanists would agree. Something that may be worth considering here is whether the ‘coral’ found on the 34

JO DAY Phaistos krater and jug set, as well as on several Kamares cups, perhaps indicates thorny branches. If so, it may be a helpful clue as to the identity of the accompanying flower.

Kamares painted decoration, inside and outside, as well as having a rim of petals, again with incised markings (Levi 1976: Pl. XLI). An unusual shape known as a ‘pedestal’ in the literature exhibits not one but three petal bands—one on the rim, one at the base, and a third at the bottom of the upper section (Levi 1976: Pl. LXXIV). The function of this vessel remains uncertain—it is hollow all the way down, forming a sort of tube. This vessel too comes from Phaistos (F480), although it is dated to the next phase, Levi’s II, or MM IIIa. There are various other similar examples from Phaistos, Kommos, and Knossos, as well as a slightly different rhyton from Phaistos (F1905) which has pointed petals with relief spirals around its mouth (Levi 1976: Pl. XLV).

Lilies always have six petals and do not show the lines in the middle of the petals which are so carefully incised on all the examples. Also the ceramic flowers simply do not have the form of a lily, being much less trumpet-shaped. I would suggest that if the Minoan potters had wished to depict a lily in a recognisable fashion, they could have made a far better effort at doing so, even allowing for the limitations of ceramic technology. Certainly in the examples introduced above, there is a consistency in the way the flowers are formed—they are white, they have red or orange centres, their petals tend to be incised and pointed, and each bloom usually has eight petals. While use of this colour scheme may be dictated by working within the Kamares style of decoration, where white, black, and red or orange are the pigments commonly employed, there is such consistency between the examples that it is possible that there is another reason for the consistent choice of white for the petals. However, as outlined above, either the plant is not meant to be a lily, but a different species, as yet unidentified by archaeologists, or else it is not meant to be a recognisable species at all. Perhaps, therefore, it is not the visual aspect of the flower that was the most important one to the potters—a key point which will be developed further.

In all these examples, the petals are added to the rims of the vessels. This has the effect of symbolically turning the vessel itself into a flower. Other vessels which do not have plastic petals attached may also be participating in this symbolic transformation of vessels into flowers—for example, the Phaistos krater has painted white triangles on the rim, something we find on many other Kamares pieces (e.g. Levi 1976: Pl. 184e). Is this perhaps a less extravagant way of symbolising petals on vessels? Some cups have what appear to be painted petals spreading out from their bases (e.g. Levi 1976: Pl. 125h; Pl. XLIb). This does not look particularly impressive when the vessel is resting, but once the cup is raised to a mouth, the full blossom would be visible, creating the impression of drinking out of an actual flower. An interesting comparison can be made here with the ‘blossom bowls’ of the LM 1 period. These serpentine vessels also deliberately take the form of flowers, with long, slim, petals carved around their bodies (Warren 1969: 14-15). The petals again have the incised line up the middle which is so obvious on the ceramic versions. More overt examples of vessels in the form of flowers can be found in contemporary Egypt, where lotiform chalices, i.e. in the form of a lotus or waterlily (Nymphaea), are found from the 18th Dynasty onwards (Tait 1963).

A closely-related group of vessels may throw some fresh light on the matter—those that do not have whole flowers attached, but instead bands of relief petals, like the one noted on the base of the Phaistos krater. From the same room as the krater (LV), comes a so-called fruit-stand (F1053)—again a large vessel, with a diameter of 54 cm and Kamares-style painting. Around the rim runs a band of white petals with a line incised up the middle, similar to the flowers already mentioned (Levi 1976: Pl. XXVII). Also from the same period at Phaistos, Ib or MM IIb, is a small spouted basin (F4640) (Fig. 2). It exhibits typical

There seems to be little doubt that all the vessels discussed above have links to the elite, as prestige items. These elaborate and showy Kamares vessels, which seem to be mainly connected with the serving and consuming of liquids,4 were all about display (Day and Wilson 1998: 356). Levi wondered if the pedestalled krater performed a similar function to those of later Greek antiquity—namely to hold wine which was then distributed to the banqueters via a jug (Levi 1976: 101). Indeed a matching jug, (F1032), was found for this krater. It displays the same chequerboard pattern and ‘coral’ decoration as on the krater (Levi 1976: Pl. XXVII). There is also a hint that this vessel too may 4. While most of the vessels could have been used for liquids (various types of jugs and a krater), a few exceptions, such as the fruit stand from Phaistos (F1053) do not seem suited to such a use. None the less, these too had a role to play in prestige displays and may have held non-liquid substances used at communal events, perhaps foodstuffs for consuming, perhaps for other items of display.

FIGURE 2 A SPOUTED BASIN WITH PETAL RIM (F. 4640), AFTER LEVI (1976) PL. XLIB. (DRAWING BY AUTHOR) 35

SOMA 2004 have had a relief flower, as there are remains of something previously attached to the shoulder.

hallucinogenic flowers such as morning glory (Ipomoea violacea) (Schultes and Hofmann 1992: 62).

It has been suggested by various scholars that ceremonial feasting and drinking played an important part in legitimating the power of elites on Crete during the Bronze Age (Day and Wilson 1998: 356; Hamilakis 2002b: 130; Hamilakis 2002c: 186; Hamilakis 1999: 48-9). As part of what Hamilakis terms ‘material culture wars’, the vessels themselves would have been as important as their contents (Hamilakis 2002c: 186). Apart from the material consumed at such events however, the memory created at them is also essential in binding people together into social groups. This bodily memory of eating, drinking, or consuming psychoactive substances can be triggered again at later dates by the same taste or smell, thus bringing to mind a whole range of associations of people and actions from the original event (Hamilakis 2002a: 20). This, of course, is what Proust was writing about when his narrator raised to his lips a spoonful of tea in which he had soaked a morsel of madeleine, which then brought so many memories flooding back. Such sensory experiences are a powerful part of human life, and often when nothing material remains of the actual experience, ‘the smell and taste of things remain poised for a long time, like souls, ready to remind us, waiting and hoping for their moment, amid the ruins of all the rest.’ (Proust 1957:58).

Most interestingly, with reference to the Minoan vessels introduced earlier, is the strong synaesthesia experienced by these Mesoamerican societies. For example, something which is seen by the eye becomes a sign for the presence of other senses normally only picked up by the nose or ears (Houston and Taube 2000: 263). So, as Houston and Taube point out in their study of sensory perception in Mesoamerica, on Mayan vessels we often find the actual aroma of flowers depicted by a specific sign of a pair of symmetrical outward-curving elements (Houston and Taube 2000: 270; fig. 5). This same sign appears in art over the mouths of vessels containing alcohol, again signalling the odour of the beverage within (Houston and Taube 2000: 270). It is fascinating to face a symbol in art for something which cannot be perceived by the eye in real life, and obviously complicates any attempts to interpret such images. On the other hand though, it could be argued that this is not such a foreign concept to us; one only has to think of cartoons where items such as stinky feet are indicated by wavy lines of stench rising above them to realise there are signs for smell in use today. If these elaborate Minoan vessels were intended for use at feasting events, their impact could have been much more subtle and all-pervasive than has so far been considered. The idea of relief flowers as markers of contents was discussed above, and is indeed a possibility to bear in mind, although as yet there is no satisfactory candidate for which flower could be represented on these vessels. Certainly these vessels would have had a strong visual impact, especially when compared to some of the plainer wares like conical cups that may have been used by some at such feasts. When the idea of synaesthesia is considered however, the flowers and petals become sensory links, stimulating not just the eye but also the nose, and possibility the closely related sense of taste too. In this case, it is perhaps not so important whether a real species is depicted or not, as long as the idea of ‘flower’ and the associated olfactory stimulus is portrayed. Certainly, there is no shortage of texts from later classical times telling of banquets where perfumes in the form of garlands for diners’ heads, incense burning, and unguents for the skin played an important part in the whole event.

This quote underlines something that tends to be overshadowed in Western studies on perception and the senses—that taste and smell, and indeed touch and sound, can be equally as powerful as sight, the sense that is usually privileged in such works and indeed in archaeology. Studies on how objects feel in the hand, or the way sound might have carried in an ancient structure, for example, have remained relatively marginalised until recently, especially in relation to the Aegean.5 Yet it is a fact that different societies have different culturally-conditioned conceptions of the senses (Classen 1997: 401). For example, in ancient Mesoamerica, smell was extremely important. The Aztec upper classes surrounded themselves with both real flowers and vessels decorated with floral motifs, which triggered pleasant associations with smell in the minds of the users (Houston and Taube 2000: 265). Buildings had flowery emblems carved on to them, which were considered to exude an exquisite aroma (Houston and Taube 2000: 265). The gods and the dead were believed to be nourished with scents, of cooked food, of incense and of course of offerings of flowers, a practice still carried out today in cemeteries and churches. The Aztecs even had two gods of flowers, Xochiquetzal, and her twin brother Xochipilli. A stone representation of Xochipilli from Tlamanalco depicts him carved with flowers all over his body—these are all

An encounter with a vessel such as the Phaistos krater would have been a powerful multi-sensory event. The colourful decoration, the three-dimensionality of the relief flowers, the aromas of liquid or smoke wafting out of these containers as real flowers release scents all would have contributed to this effect. Add to this the effect of whatever was consumed from the vessel, with potential for altering states of consciousness, and I suggest that such vessels with flowers or petals attached were potent mnemonic links, both with the act of consuming and of the accompanying social

5. Some examples of works on an ‘archaeology of the senses’ include MacGregor 1999; Jones and Hayden 1998; Hamilakis 2002b.

36

JO DAY interaction at the event. The placing of ceramic flowers on these vessels brings expectations and associations of scent to the gathering and may indicate the contents of the vessels, while recollections of the feast would be later stimulated by smelling similar aromas or indeed visually encountering this flower. Like Proust’s Marcel, who was aware that his feelings were connected with the mouthful of tea and cake but at the same time infinitely transcended those flavours, we should perhaps look beyond just visually examining Minoan art and consider its appeal to the other senses and the long-term effect of such experiences.

Hamilakis, Y. (2002c) Too Many Chiefs? Factional Competition in Neopalatial Crete. IN: Driessen, J., I. Schoep and R. Laffineur eds. Monuments of Minos – Rethinking the Minoan Palaces. Aegaeum 23. Liège, University of Liège, 179-99. Higgins, R. (1997) Minoan and Mycenaean Art. New revised edition. London, Thames and Hudson. Houston, S. and K. Taube (2000) An Archaeology of the Senses: Perception and Cultural Expression in Ancient Mesoamerica. Cambridge Archaeological Journal, 10 (2), 261-94. Jones, C. and C. Hayden eds. (1998) The Archaeology of Perception and the Senses. Archaeological Review from Cambridge, 15 (1). Kemp, B.J. and R.S. Merrillees (1980) Additional Remarks on the Qubbet El-Hawa Vase and its Context. IN: Kemp, B.J. and R.S. Merrillees, Minoan Pottery in Second Millenium Egypt. Mainz am Rhein, P. von Zabern, 215-19. La Rosa, V. (1994-96) Ayia Triada. Archaeological Reports 1992-94, Iraklion Nomos. Kritiki Estia, 5, 326-32. Levi, D. (1976) Festós e la Civiltá Minoica. Incunabula Graeca 40. Rome, Edizioni dell’Ateneo. Litzinger, W.J. (1981) Ceramic Evidence for Prehistoric Datura Use in North America. Journal of Ethnopharmacology, 4 (1), 57-74. MacGregor, G. (1999) Making Sense of the Past in the Present: a Sensory Analysis of Carved Stone Balls. World Archaeology, 31 (2), 258-71. Myerhoff, B. (1974) Peyote Hunt: The Sacred Journey of the Huichol Indians. Ithaca, Cornell University Press. Pavord, A. (1999) The Tulip. London, Bloomsbury. Pernier, L. (1935) Il Palazzo Minoico di Festós I. Rome, Libreria dello Stato. Proust, M. (1957) [1913] Swann’s Way. Harmondsworth, Penguin. Schultes, R. E. and A. Hofmann (1992) Plants of the Gods. Rochester, Healing Arts Press. Tait, G. A. D. (1963) The Egyptian Relief Chalice. Journal of Egyptian Archaeology, 49, 93-139. Tzedakis Y. and H. Martlew eds. (1999) Minoans and Mycenaeans: Flavours of their Time. Catalogue of Exhibition in National Archaeological Museum, Athens, 12 July – 27 November 1999. Athens, Kapon Editions. Warren, P. (1969) Minoan Stone Vases. Cambridge, Cambridge University Press.

Bibliography Barley, N. (1994) Smashing Pots: Feats of Clay from Africa. London, British Museum Press. Betancourt, P. (1990) Kommos II. The final Neolithic through Middle Minoan III Pottery. Princeton, Oxford, Princeton University Press. Classen, C. (1997) Foundations for an Anthropology of the Senses. International Social Science Journal, 153, 401-12. David, N., J. Sterner and K. Gavua (1988) Why Pots are Decorated. Current Anthropology, 29 (3), 365-89. Day, P.M. and D.E. Wilson (1998) Consuming Power: Kamares Ware in Protopalatial Knossos. Antiquity, 72, 350-8. Dickinson, O. (1994) The Aegean Bronze Age. Cambridge, Cambridge University Press. Foster, Polinger K. (1982) Minoan Ceramic Relief. Studies in Mediterranean Archaeology 64. Göteborg, Paul Åström. Hamilakis, Y. (1999) Food Technologies/Technologies of the Body: the Social Context of Wine and Oil Production and Consumption in Bronze Age Crete. World Archaeology, 31 (1), 38-54. Hamilakis, Y. (2002a) What Future for the Minoan Past? Re-thinking Minoan Archaeology. IN: Hamilakis, Y. ed. Labyrinth Revisited: Rethinking Minoan Archaeology. Oxford, Oxbow Books, 2-28. Hamilakis, Y. (2002b) The Past as Oral History – Towards an Archaeology of the Senses. IN: Hamilakis, Y., M. Pluciennik and S. Tarlow eds. Thinking Through the Body: Archaeologies of Corporality. London, Kluwer Academic/Plenum, 121-36.

37

Beyond the Consumer City; the Case of Ancient Antioch Andrea Umberto De Giorgi

At the beginning of the twenty-first century, the discussion over the nature of the economic structures produced by the Roman city still raises significant interest, with the views of Sombart (1916-27) and Weber (1958) prominent in debate. By extension, the Consumer City scheme as advanced by Moses Finley, according to whom the flow of goods and resources into the urban centre posits a parasitic relationship between town and country (1981: 123), represents for many historians an expedient solution (Parkins 1997: 83-111) that also accords with the literary evidence from the ruling classes on their economic outlook and urban-cantered life-style (Wallace-Hadrill 1991: 24172). More than thirty years after Finley’s seminal Sather lectures (1999), however, there are signs of impatience with this solution.

behaviour are the decisive factors behind the economic growth that extends over the valley for the first two centuries CE. The territory of Antioch is by no means different from any other provincial landscape, similarly moulded by complex processes of negotiation and coercion (Mattingly 1997: 7-24). Nevertheless, the picture derived is one of a mutually beneficial, symbiotic relationship between town and country, in which the latter attains autarchy well before the time of Libanius in the fourth century (Libanius, Ant., 11, 230). Located in the northern Levant, the fertile Amuq Valley has been host to extensive archaeological campaigns since the 1930s (Yener, Edens, et al. 2000: 163-5). Physical diversity is the signature of the valley; the territory in fact encompasses the plain, dominated by the Amiq Gölü, the great lake drained in the 1960s, the Amanus mountain range to the west, and a series of extremely eroded limestone hills to the south, east and south-west. The extraordinary number of sites featuring rich material remains in this region confirms its importance throughout prehistory and well into the period when the successive Assyrian, Babylonian, Achaemenid, Seleucid and Roman empires exploited its wealth. The valley’s vast array of sites has offered ideal conditions for conducting this survey-based research on non-urban land use. In particular, archaeologists have made great advances in understanding settlement patterns from the Neolithic to the first millennium BCE (Yener Forthcoming). The heavy ‘Romanisation’ of this landscape, however, propelled by the city of Antioch, has remained neglected.

This model arguably conforms to the contours of the Roman city, especially the megalopoleis like Alexandria, Carthage and Antioch, urban milieus that were not the products of market forces, but were rather sustained by their sociopolitical roles (Kirsten 1956). This said the configuration fails to illustrate the range of economic variables in play at the local level, thus reducing more complex town and country relations to a single, centripetal country to town transaction. Instead of dwelling on the limits of the scheme however, one can positively advance it by examining how the urban centre stimulates market forces and production incentives through the logistics of its supply system, and intensive archaeological survey has proved itself an effective means of investigating this process (Mattingly, Stone, et al. 2001: 66-7). In this paper, I will follow these methodological lines to delineate the economic layout of the hinterlands of Roman Antioch during the first two centuries CE.

Since 1995, the AVRP, directed by A. Yener and T. Wilkinson of the Oriental Institute at the University of Chicago, has operated in this area. This multidimensional investigation, which includes excavations and field survey, examines in a holistic manner the economic and environmental background of settlement in the valley (Yener Forthcoming). Furthermore, the project is geared toward the elimination of the ‘big divide’ of Roman/Late Roman settlement vs. mounded sites, so as to provide a historical continuum of human occupation in the region.

The Amuq Valley Regional Project (henceforth AVRP) has succeeded in demonstrating that the rise of the city as provincial capital accounts for the intensification of settlement in the Amuq Valley (Yener, Edens, et al. 2000: 163-220). Moreover, the ability to seize the opportunities offered by a fertile landscape and rational economic 39

SOMA 2004

FIGURE 1 THE AMUQ VALLEY REGIONAL PROJECT AND SITES MAPPED IN THE SEASONS 1995-2002 The Amuq Valley loosely corresponds to the administrative territory of Antioch, modern day Antakya and capital of the province of Syria since 64 BCE, and thence residence of emperors, home of one of the Syrian legions and scene of numerous fierce political and ethnic riots (Downey 1961: 163-271; Butcher 2003:19-78). On a more positive note, the city’s pedigree can boast some of the most illustrious schools of rhetoric and philosophy in antiquity (Kondoleon 2000). Today, virtually nothing of the city’s glorious past is visible, with the exception of a small stretch of the aqueduct of Hadrian, the early fifth century CE fortification walls and three poorly

preserved pottery workshops and baths along the slopes of Mt. Silpius. The textual sources largely agree upon Antioch being one of the largest cities of the Empire; among the various appraisals, Strabo’s assessment of Augustan Antioch inhabited by 500,000 (Strabo, Geogr. 16.2.5), can be deemed as the most persuasive. Be that as it may, one fact is inescapable: the city had to sustain and feed a large contingent of military and administrative urbanites, and the burden of doing so lay on the communities of the hinterland. The AVRP survey data have been instrumental 40

ANDREA UMBERTO DE GIORGI in testing the validity of this hypothesis and provides the underpinnings for the economic framework presented in this paper.

for which, ultimately, Antioch was praised by the sources (Heichelheim 1938). It plausibly occupied the immediate hinterlands of the city within a 5 to 10 km radius, where market returns were always attainable. Distance constraints thus account for the dense concentration of sites, with values of 1-2 km between settlements. Farmsteads, small villages and sporadic villae are the settlement units in this sector.

Our survey over the years 1995-2002 essentially sampled discrete areas in three distinct sub-regions; the plain of Antioch, the Jebel-al-Akra relief and, in addition, a cursory inspection of the Amanus Mountains. For the sake of clarity, sites on the plain will be treated separately from those in the highlands.

2.

Continuity of settlement is one of the salient trends concerning the patterning of sites; 71% of the sites inhabited in the Late Hellenistic period are occupied in the Early Roman phase, thus reinforcing the notion of passive Roman administration (Millar 1993: 27-79). The fiscal grid laid down by the governor of Syria, Quirinus, in 6 CE (Dabrowa 1998: 27-30) thus did not seemingly alter the status quo, and actually encouraged rural settlement, as the number of sites created ex novo at the beginning of the century testifies. The incentives and benefits that a census typically carried were factors that indeed encouraged the investment of capital in the country and the exploitation of land hitherto left abandoned or which was simply used for grazing (Corbier 1991: 227). The sites on the plain that can be attributed to the High Roman Empire occupation phases are typically small and hardly ever measure more than 1 hectare. They are low mounds and consist predominantly of scatters of tiles, pottery and stone. The navigable river Orontes and the main west-east artery leading to Aleppo moreover ensured connectivity among these sites as well as the frequent movement of people and goods between town and country.

All in all, this system comprising sites on the plain, secured supply of grain to Antioch. Its spatial extent and productivity, however, hinged on the distance from the city and on the ease of access to the markets. There was clearly a finite limit at which the market price of crops would not cover the cost of land-rent, production and transport to the city. Beyond a threshold plausibly situated 20 km from Antioch, other urban foci of lower rank entered the picture, and although lacking civic charters, they clearly functioned as markets. Ancient Imma, AS 345, modern Yeni Sehir, is definitely one of them. This small town accommodated one of the most celebrated markets in the province (De Ligt 1993:73) and was supplied by its own hinterland of sites, some of which display evidence of large capital investments for productivity maximisation.

It needs to be stressed, however, that the development of modern agriculture all over the region, and especially on the uplands, has profoundly altered the configuration of the landscape, and, consequently, only high intensity survey strategies can detect the presence of Roman settlement on the ground (Casana Forthcoming). Braidwood, in the 1930s, recorded only 17 Roman sites, missing the hundreds of small settlements that dot the landscape (1937). The assumption that a city like Antioch triggered economic growth through its supply system is indeed justified by the distribution of these sites, their spatial configuration and, when it can be ascertained, their economic basis. One may represent the scheme as Von Thünen did to illustrate land use complexity in German cities of the 1800s (1826). The point of his concentric scheme is such that any real city can be analyzed against the background of his isolated State, thereby extracting the patterns at stake. In the case of Roman Antioch, the outcome is that urban demand stimulates two of the forms of land use comprised by the model: 1.

A second form of land use is suggested by the outer villages situated beyond the Orontes, but still lying within an arbitrary easy market-return radius of 20 km and packed with lesser density. They primarily grew extensive crops of cereals. By and large the picture of the settlement consists of very small farmsteads, rare densely nucleated villages and several modest hamlets. The incidence of the latter is of particular interest, as it plausibly entailed temporary settlement, essentially small huts for equipment and temporary residence. By this rationale, it can be suggested that the majority of farmers, whatever their tenancy status, were likely to be city dwellers, just like Max Weber’s Ackerburger, farmers commuting to and from the city. By extension, they comprised some of those who clearly did not live on the surplus produced by the rural population and conversely fed the markets. In their case, therefore, the ‘consumer’ label can be discounted.

Several multiple sets of water mills, dams and canals demonstrate the introduction and use of water technologies in the Amuq Valley as early as the first century CE. Among the better-preserved sites is the complex at site AS 202, adjacent to the town of Imma, where a network of canals tapping water from the uplands led water to reservoirs and overshot water mills. What is evident is that the capital cost of this infrastructure must have been significant if one considers the magnitude of the undertaking and the additional expenditures for the mill apparatus. The capital investment, however, must have been rapidly paid off by the scale of production and increased output rates, because

The first form is represented by horticulture, floriculture and the production of perishables, 41

SOMA 2004 of the substantial advantages that water mills had over animal mills in these terms.

in times of shortages and also guaranteed a substantial provision of oil for public doles.

The key issue is thus the significant extent to which landowners in the Amuq Valley invested in improving their estates to maximise financial returns. Evidently, not all the farmers believed in hoarding surplus (Wilson 2002: 1-32).

An entirely different set of problems has been raised by site AS 232 in the vicinity of the Kisecik area, in the Amanus Mountains, where substantial traces of the quarrying industry have been recovered. The site does not constitute a unicum, as do other settlements in the Amanus area, such as AS 245, which can be convincingly interpreted as quarries with associated small living quarters. Nevertheless, the relatively small size of the infrastructures monitored suggests that management and quarrying operations were contracted out by the city to local communities and entrepreneurs. The possibility that the scale of this industry was significant, however, should not be discounted, considering that the extraction of local good limestone and its transportation down to the valley floor were crucial for the accomplishment of the several building programs carried out by emperors in the city during the High Empire. The information we have gathered on the extractive industry is still piecemeal; however, the upcoming research focused on these very questions will be able to provide insights on the magnitude, duration and benefits of an activity that indeed contributed to the formation of a distinct Roman landscape.

The settlement pattern in the highlands also merits attention. A survey of selected areas in the low hills of the Jebel al-Aqra to the south and in the Amanus Mountains to the west revealed no occupation prior to the Seleucid period. Towns, farmsteads, villages and a few villae are the typical features of the Roman phase. The Jebel alAqra intensive survey yielded the highest density of sites recorded, 3.50 per square km, as opposed to the 0.40 in the plain. This said, the sample areas surveyed by and large display occupational trends towards the foothills and lower elevations. Olive oil production was seemingly the economic mainstay of the highland settlement, as ubiquitous fragments of mills and pressing apparatuses testify. Furthermore, only the 11% of the sites exceed the 600 m elevation threshold, the limit above which the cultivation of the olive tree is not feasible for Mediterranean climates. Given the site density and considering that approximately only 30% of the sub-region has been surveyed, one can translate this index into numerical terms and, relying on figures from contemporary contexts in Baetica and Libya (Mattingly 1988: 33-56), hypothesise the presence of a minimum of 350 presses over the 800 square km, surveyed and non-surveyed sector of the Jebel. The yearly output of olive oil at the average rate of 5000 litres per press would thus be around 1,750,000 litres from the Jebel alone, which would constitute arbitrarily half of the total production required by a city the size of Antioch. Several texts corroborate both the importance and the volume of olive oil production in Antioch. In addition to underscoring the profitability and the cost of the land for oil production (Philostratus, Apoll., 6, 39), they also refer to the annual city dole of oil to the large Jewish community, which deeply upset the non-Jewish Antiochenes under the governorship of Mucianus in 69 CE (Josephus, AJ , 12, 120).

To conclude, this rapid overview of economic diversity in the territory of ancient Antioch suggests that the city siphoned only a relatively small portion of resources from the landscape in terms of taxes and rural rents. The city, qualifying as ‘swollen administrative capital’ (Finley 1985: 30) was, however, a milieu for commercial occupations and entrepreneurial activities linked to the trade and distribution of commodities like Eastern Sigillata A pottery, textiles and precious metals (Bowersock 1989: 68-9) and thus provided sources of income for non-passive urbanites. Urban markets continuously fed by ‘commuting farmers’, stimulated the circulation of currency for commercial transactions and offered services for the purchasing power of the rural communities as well. Thus the city, by acting as the service-provider, reversed the basic tenet of unilateralism characteristic of Finley’s consumer city. On the other side, after paying their dues to the metropolis, the communities in the hinterlands seized the opportunity to invest some of their profits in technology and more intensive production, thereby acquiring the ability to select markets and crops that were profitable. The self-sufficiency Libanius boasts of for those communities in 356 CE was plausibly attained far earlier in the Empire (Libanius, Ant., 11, 230).

The event seems to point to the existence of specific norms pertaining to the commerce of oil, perhaps giving the city the ability to buy its own supply at a lower, fixed cost. Hadrian plausibly followed these very lines when, as nomothetes in Athens, he aimed to secure a continuous supply of oil to the polis, thereby limiting export and free commerce of the harvest. To that end, he promulgated the famous oil law that required all farmers to forcibly hand in one third to one eighth of their oil production to the Eleonai, the public oil-buyers (Oliver 1989: 232-8). The aim in the enacting of this measure was twofold, in that it granted the polis the ability to protect the community

In conclusion, for the time of the pax romana, the picture thus derived is of a symbiotic relationship between town and country. The economic growth experienced by the landscape during the first two centuries CE is chronologically consonant with the grandeur reached by the city at the time of Hadrian. This mutual flourishing of town and country was, of course, fragile. The famine 42

ANDREA UMBERTO DE GIORGI of 362 CE that struck Antioch was resolved by Julian through substantial imports of grain from short distance sites lying 50 to 100 km away. Over-exploitation of the land and unfavourable environmental conditions can surely be blamed for the debacle. The truth is, however, that the inability to foresee catastrophic events and thereby generate long-term strategies to guard against them is one of the endemic shortcomings of the ancient city. Here one can hardly argue with Finley, and the problem can be positively re-addressed through the use of intensive survey data inserted in a robust interpretative framework.

Kirsten, E. (1956) Die Griechische Polis als historischgeographisches Problem des Mittelmeerraumes, Bonn, In Kommission bei F. Dümmler. Kondoleon, C. ed. (2000) Antioch. The Lost Ancient City, Princeton, Princeton University Press in association with the Worcester Art Museum. Mattingly, D., Stone, D. and L. Stirling (2001) Leptiminus (Tunisia). A ‘Producer City’.IN: Mattingly D. and J. Salmon eds., Economies Beyond Agriculture in the Classical World Leicester-Nottingham Studies in Ancient Society, Vol. 9, London and New York, Routledge, 66-89. Mattingly, D. (1997) Dialogues of Power and Experience in the Roman Empire. IN: Mattingly, D. ed., Dialogues in Roman Imperialism. Power, Discourse and Discrepant Experience in the Roman Empire, Portsmouth, Journal of Roman Archaeology 7-24. Mattingly, D. (1988) Oil for export? A Comparison of Libyan, Spanish and Tunisian Oil Production in the Roman Empire, Journal of Roman Archaeology, 1, 3356. Millar, F. (1993) The Roman Near East. 31BC-AD337, Cambridge, Harvard University Press. Oliver, J.H. (1989) Greek Constitutions of the Early Roman Emperors from Inscriptions and Papyri, MAPS, 92, 232-8. Parkins, H. (1997) The ‘Consumer City’ Domesticated? The Roman City in the Elite Economic Strategies. IN: Parkins, H. ed., Roman Urbanism. Beyond the Consumer City, London and New York, Routledge, 83-111. Sombart, W. (1916-27) Der Moderne Kapitalismus, 3 vol., Munich and Leipzig, Duncker & Humblot. Von Thünen, J.H. (1826) Der Isolierte Staat in Beziehung auf Landwirtschaft und National-ökonomie, Jena, Neud. Wallace-Hadrill, A. (1991) Elites and Trade in the Roman Town. IN: Rich, J. and A.Wallace-Hadrill, eds., City and Country in the Ancient World, Leicester-Nottingham Studies in Ancient Societies, 2, London and New York, Routledge, 241-72. Weber, M. (1958) The City, Glencoe, Free Press. Wilson, A. (2002) Machines, Power and the Ancient Economy, Journal of Roman Studies, 92, 1-32. Yener, A.K., Edens, C., Harrison, T.P., Verstraete, J. and T.J. Wilkinson (2000) The Amuq Valley Regional Project, American Journal of Archaeology, 104.2, 163-221. Yener, A.K. (Forthcoming) The Amuq Valley Regional Project, 1995-2002, Chicago, The University of Chicago Press.

Bibliography Braidwood, R.J. (1937) Mounds in the Plain of Antioch, an Archaeological Survey, Chicago, The University of Chicago Press. Butcher, K. (2003) Roman Syria and the Near East, Los Angeles, J. Paul Getty Museum/Getty Publications. Bowersock, G.W. (1989) Social and Economic History of Syria under the Roman Empire. IN: Dentzer, J.M. and W. Orthmann eds., Archeologie et Historie de la Syrie, 2, Saarbrucken, Saarbrücker Druckerei und Verlag, 6381. Casana, J. (Forthcoming) From Alalakh to Antioch: Settlement, Land Use, and Environmental Change in the Amuq Valley of Southern Turkey. Corbier, M. (1991) City, Territory and Taxation. IN: Rich J. and A. Wallace-Hadrill eds., City and Country in the Ancient World, Leicester-Nottingham Studies in Ancient Societies, 2, London and New York, Routledge, 211-39. Dabrowa, E. (1998) The Governors of Roman Syria from Augustus to Septimius Severus, Bonn, Habelt. De Ligt, L. (1993) Fairs and Markets in the Roman Empire. Economic and Social Aspects of Periodic Trade in a Pre-Industrial Society, Amsterdam, Gieben. Downey, G. (1961) A History of Antioch in Syria: From Seleucus to the Arab Conquest, Princeton, Princeton University Press. Finley, M. (1999) The Ancient Economy; with a Foreword by I. Morris. Berkeley, University of California Press. Finley, M. (1981) The Ancient City: from Foustel de Coulange to Max Weber and Beyond. IN: Shaw, B.D. and R. Saller eds., Economy and Society in Ancient Greece, New York, Viking Press, 1-23. Heichelheim, F. M. (1938) Roman Syria. IN: Frank, T. ed., An Economic Survey of Ancient Rome, 4, Baltimore, The Johns Hopkins Press, 121-257.

43

The Site of Iuvanum (Abruzzi, Italy) in the Classical period; Its Public and Private Buildings: A Preliminary Study of a Taverna Vanessa D’Orazio

Caricini, who controlled the satellite ‘vicii’ of Trebula (Lapenna 1996a: 65-74; Lapenna 1996b: 687) and Cluviae. The earliest structures on the site date from the end of the third and early second century BC and include the sanctuary, comprised of two temples on the hill of the acropolis, and the theatre, located on the slopes of this hill. The Roman city, with its fora, tavernae, basilicae and other public buildings, is arranged on a level area formally used as a marketplace.

Introduction The Roman municipium of Iuvanum, whose status as a colony or municipium is confirmed by Pliny the Elder, has been located at Santa Maria di Palazzo (Paci 1990: 51-75). It is situated 1,000 metres above sea-level in the region of Montenerodomo in the Province of Chieti, Abruzzo, Italy (Fig. 1). Iuvanum was founded by Rome after the Social War and was an important centre within this inner zone of Samnium, inhabited by Samnites

FIGURE 1 IUVANUM: LOCALISATION OF THE SITE 45

SOMA 2004

FIGURE 2 IUVANUM: GENERAL MAP OF SITE on the acropolis hill which includes two parallel temples (Fabbricotti 1990: 9-18; Lapenna 1997: 64-6), and the theatre resting on the slope. These monuments are the earliest structures on the summit of the acropolis, which subsequently formed the focus of the Roman city.

The Site of Iuvanum1 The city of Iuvanum is cited in the ancient sources as Iobanus (Buonocore-Firpo 1991: 429-54) and its inhabitants as Lanuenses (Pliny: Naturalis Historia III, 12,106; BuonocoreFirpo 1991: 429-54), also referred to as Iuvanenses (Madonna 1930: 23-64; Buonocore 1984; Buonocore and Firpo 1991: 429-54). The city has been identified and located at Santa Maria di Palazzo di Montenerodomo: a location which is supported by the associated epigraphical evidence.

The earliest buildings on the site date from the end of the third century BC and include the sanctuary, surrounded by a wall executed in polygonal stonework, the temenos

One of the two Italic temples is situated at the centre of the elevation (Temple A, tetrastila map). It is founded on a base of porous stone measuring 14.6m in length, while the wall of the cella measures 10.4m. This feature is not found in the subsequent buildings which might suggest that Temple A represents the earliest structure and possibly dates to the start of the second century BC. The structure was almost completely destroyed by the construction of the monastery of S. Maria di Palazzo, located in the southern sector of the acropolis in the twelfth century AD. This subsequent construction caused serious damage to the pre-existent structures of the acropolis as it reused stone material from these earlier structures as well as from the wider city of Iuvanum and its necropolis, as is attested by several funeral stelae inscriptions visibly inserted into the monastic construction.

1. The site of Iuvanum has been written about by a number of authors: Fabbricotti (1992: 77-82; 1996: 21-36; 1997: 67; Firpo (1990: 35-49); Mazzitti (2000: 107); Ruscitti (2000: 8-11) and Staffilani (1992: 83-91; 2000: 12-15; 2001: 40-7).

The second temple (Temple B) is comprised of a podium, with a stairway measuring 17m x 8.3m wide. The two temples lie on a parallel axis at a distance of 2.95m apart. Temple B is simple in plan comprising a cella

The site of Iuvanum (Fig. 2) was founded in the Samnite period as a cult focus and market place. It became a Roman municipium in 87 BC after the end of the Social War, becoming the seat of political and administrative power, and representing Roman influence in this inner zone of Samnium, an area which had been continually inhabited since the fourth century BC by the ethnic group of the Samnites Carecini (Lloyd 1996: 11-20).

46

VANESSA D’ORAZIO and pronaos. It is possible to envisage the presence of a frontal tetrastila prostila with a frontal stairway between parapets dating to the second half of the second century BC. Polygonal walls, dating to the beginning of the third century BC, encircle the cult area but precede the erection of the temple. It is possible that the preceding cult was associated with some form of water worship due to the proximity of the nearby spring. It remains obscure as to why this place was chosen as a sacred temenos, given its location at about 1,000 meters above sea-level; the altitude would not be a problem by itself, but combined with the icy temperatures, would have hindered communication with, and access to, the site during the cold months, as is the case today.

The theatre, located on the eastern side of the acropolis, is not perfectly aligned with the axes of the temples. The cavea rests on the slope while the frons scenae contains three niches. The stonework is precise while holes in the steps suggest some form of cover construction. The thermae of the city were located near the theatre (Fig. 2), where they are partially conserved in the foundations of a small house. Several basalt-paved roads cross the city. The Via del Foro descends from the acropolis and enters the forum from the south. The Via Orientale starts at the spring and runs parallel to the public square. It leaves the city, but is difficult to trace to the north beyond the basilica. At the intersection of the two roads, the Via Orientale and the Via di Bacco, there are three buildings with doorways opening onto the Via Orientale. These structures were part of the residential area which was organised in blocks. One of the structures has been named the ‘House of Bacchus’ due to the discovery of a small herm. The structure opens onto the main road with a room shaped like a taverna, most probably a thermopolium, as can be deduced from the remains of a counter in the back wall characteristic of the sale of beverages. Wine cellars were located in the basement which was below street level. The structure is located at a corner of the road, as is also seen also at Ostia, where taverns and thermopolia were used by tired and thirsty travellers entering the city (Pavolini 1986: 22738).

The pre-Roman finds in this zone are few and consist of various fragments of black-glazed pottery and some small bronze figurines of Heracles in onslaught that are probably associated with the run of the sheep-tracks (Di Niro 1977: 10). Heracles was the incarnation of courage and physical strength; he was adopted as a hero by the Samnites who identified in him their own distinctive traits, and he was ultimately assimilated into the worship of a local divinity. Such a divinity was venerated in sanctuaries or rest places as the god of water sources and healthy waters, of the flocks and the shepherds (Papi 1997: 142-3). Twenty five fragments of pottery dating to the Late Bronze Age were discovered during the excavations at the northeast corner of the Temple A. Their discovery confirms that the site was frequented prior to the construction of the temple probably as early as the tenth century BC, as is also testified by the chance discovery of a bronze knife.

A sarcophagus hewn in local stone, 2.25m long and 0.7m high, and with a gabled lid 1.26m wide, was discovered to the north of the forum beyond the apsidal building (Fabbricotti 1990: 142-4). The sarcophagus bears a funerary inscription of nine lines of text, dedicated to a boy, M. Titatius Celer, by his parents M. Titatius Rufus and Verania Severa, on the event of his premature death at the age of sixteen. The discovery of the sarcophagus beyond the basilica, and the presence of many funerary inscriptions reused in the monastery, suggests that this zone represents the necropolis of the city.

Recent excavations also revealed a ‘litra’ in bronze from the mint of Isernia, datable to the third century BC, while coins from the mints of Napoli, Isernia, Alba and Cales were discovered in previous excavations which testify to pre-Roman contacts, of commerce and transhumance, with Puglia and Campania. The sanctuary, in its first phase, was not directly connected with the city but was the focus of the surrounding territory, catering for the scattered small farms accessed by the road leading to the spring, which was used by both travellers and shepherds. It is possible that the associated inhabited centre was in Montenerodomo, inside the townwalls, fortified with polygonal walls and datable to the fourth century BC. It is equally likely that this fortified site was primarily defensive in nature and that the scattered population would seek refuge there in the event of an attack (Fabbricotti 1990: 13). The southern territory of ‘Carecini’ was defended by fortresses such as those at ‘Monti Pizzi, Monte di Maio, Monte Pidocchio and Montenerodomo’. The fortified sites of Montenerodomo, Monte Pidocchio and Monte di Maio, located in the highlands at altitudes over 1,000m above sea level, form a triangle over the zone of Iuvanum, and acted in defence and control of the valley.

The Forum and the Shops The establishment of the municipality after the Social War modified the structural order of Samnium, resulting in urban development that supported citizens whose activity closely depended on city life, such as public administrators, priests, craftsmen, traders, agriculturists and breeders whose properties were close to the city. The open area to the north of the acropolis, used in previous centuries for periodic fairs and markets, was, in the Augustan period, enriched with monumental buildings: fora, tavernae, basilicae, and other private and public buildings, and it became the centre around which the Roman city was structured. The forum was the main public square placed at the centre of the Roman city; it was a market place, but above all it was the fulcrum of the social, political and religious 47

SOMA 2004 life of the people. The forum in Iuvanum is rectangular and surrounded on all sides by porticos, as attested by 40 column bases found in situ, 8 on the short sides and 14 on the long ones. We can deduce that the longer sides have dimensions nearly double that of the shorter sides (64m x 27.5m). The northern portico is wider than the others, similar to a basilical hall, while offices and workshops occupy the remaining sides (Fig. 2). Statues dedicated to divinities, members of the imperial family and eminent citizens adorn the public square; small bases for these statues are also preserved. This public square is placed at a slightly lower level than that of porticos and it slopes towards the south-east to facilitate the drainage of rain water towards the small channel at the corner between the south and the east sides of the forum.

travellers could eat and sleep. All of the rooms excavated at Iuvanum which appear behind the porticos are tavernae. One exception is an apsidal structure on the western side, close to the basilica, which seems to have been a municipal office. Many tavernae were divided into two different rooms in order to create a living room for the owner at the back of the shop. The paved street, the Via del Foro, connects the sanctuary with the forum, which it accessed on the southern side where today a taverna is exposed, but originally it might have been partially closed with a wooden panel. Taverna H Tavernae aligned along a south-north axis on the west side, called G, H and R, delimit the eastern zone of the forum, starting from the B (Di Luca 1996: 37-42) corner and continuing towards the north. The study of Taverna H (Candeloro 1990: 97) (Fig. 3-4), discovered during five excavation campaigns in 1981-1994-1995-1996-1997, can be analyzed more closely. Area H faces the greater portico of the eastern side and has a hall as wide as its total width: again this might originally have been closed by a wooden panel. It measures 4.7m wide by 6.40 long. The walls have a height of about 0.5-0.6m with a wall thickness of 0.5m. They are constructed of roughly-cut

The location of the basilica (Fig. 2), enclosing the forum along its northern extent, is comparable to that of the basilicae at Alba Fucens and Sepino. The portico, which functions as a passage between two monuments, has a rectangular shape (36m x 10m) with a unique nave and front with spaces on the short sides. The back wall of the basilica, with 12 half-columns, has an apsidal quadrangular hall in the centre which might have been used as a court and was the focal point of the cult of the imperial family. This construction has been dated to the Augustan period with restorations in the following centuries. At Iuvanum, as is usual in fora designed in the first century AD, the monumentalisation of the square is completed by stone paving. An inscription is located in the centre of the forum, laid out in three rows of stone slabs within the floor paving (Iaculli 1990: 81-9). The lettering is inlaid in bronze and secured by grips. Missing letters are indicated by the holes where grips are located. The inscription contains the name of the author of the work: ‘C. Herennius Capito’. This is a typical feature of the Augustan period, and is also found at Sepinum. Herennius Capito, who paid for the paving, never saw its execution; the work was finished by his heir Herennia Proiecta. Herennius Capito was already well-known in Chieti, as indicated by an inscription dating to 41 AD. This information supplies a terminus post quem for the new city plan, now constructed and planned with a unifying design. The date of the inscription and the paving of the forum (i.e. first century AD), coincides with comparable inscriptions on the paved floor of fora in Rome, Pompeii, Terracina, Atena Lucana, Velia, Sepino and Assisi.

FIGURE 3 IUVANUM: MAP OF TAVERNA H

In the Roman world, the area of the forum incorporated public and cult buildings, tavernae, shops selling food and agricultural products and handicraft, glass objects, ceramics, metal instruments for agriculture and wool manufacturing. In addition to such shops or stalls there were often workshops where such objects were produced, while craftsmen and small traders had their homes in the back o these shops. Sometimes, as can be observed at both Ostia and Pompeii, workshops alternated with thermopolia for warm drinks and taverns or inns where

FIGURE 4 IUVANUM: PHOTO OF TAVERNA H 48

VANESSA D’ORAZIO medium-sized stones bonded together with mortar: USM 102 and USM 137 (with alignment E/W), and USM 160 (with alignment N/S) (Fig. 3). Two large rectangular bases of white local stone, respectively called USM 135 and USM 138, are visible opposite walls USM 102 and USM 137 in front of the two columns of the portico (Fig. 4). In the first stratigraphic layer within the shop six bronze and one silver coin was found, together with fragments of pottery. The coins represent a small hoard dating from between the second and the third century AD.

terra sigillata, amphorae, coarse wares and notably the discovery of 20 coins (in particular a quinario of the first century BC and three denari – one from the first century BC, and others from the second and the third century AD – all found in Taverna H), we can conclude that on the site of Iuvanum there were four great historical phases. These phases are the Republican period, when buildings close to the Acropolis were constructed; the Julio-Claudian period, when the northern part of the city was built as attested by the inscriptions in the forum; the Hadrianic period, when an earthquake or a fire destruction has been identified in the stratigraphic record; and finally, the Late Antique period, when the city falls into decline as is characteristic of an overall decline throughout the empire, with a consequent economic and demographic decline. This period ended in the middle of the fourth century AD when a sudden programme of restoration of public buildings and roads is evident, probably following the earthquake which affected the region of Samnium in 346 AD.

The group of tavernae within the portico seem to be homogenous and contemporaneous, like those in Alba Fucens and Sepino. At Alba Fucens the tavernae are placed along two parallel roads, the Via dei Pilastri and the Via del Miliario indicating that this area was specifically for the sale of drinks and goods (Coarelli and La Regina 1984: 82; Campanelli 2001; 2002). The associated workshops are symmetrical and contemporaneous, including a portico at their first conception, and belonging to the original urban project of the Late-Republican period, and subsequently restored many times. The deep wheel ruts visible in the road indicate the intensity of traffic at the time.

Acknowledgements I would like to thank Prof. Emanuela Fabbricotti and Dr. Oliva Menozzi for their advice and corrections, and my colleagues Claudio and Veronica Agostini and Lida Cefaratti for their collaboration and assistance for the last translation before printing. Thanks also to the photographer Dr. Vienna Tordone for letting me use her photos. Moreover, I would like to thank also Dr. Luca Cherstich for his advice.

At Sepino, tavernae are identified along the Decumanus, between Porta Boiano and the forum (Coarelli and La Regina 1984: 219-20), on the right side of the road where a series of porticos are associated with buildings and workshops to the rear. The road that crosses the city from the north-west to the south-east, between the Gates of Boiano and Benevento, corresponds to the ancient cattle path that always maintained the non-orthogonal crossing of the ancient Cardo and Decumanus, which had determined the planning of the walls and the location of the gates.

Bibliography Buonocore, M. (1984) Le iscrizioni latine di Iuvanum e Cluviae Carricinorum conservate nel Museo Nazionale di Chieti, Samnium 57,1-2, 43-68. Buonocore, M. and G. Firpo, eds. (1991) Carricini. IN: Fonti latine e greche per la Storia dell’Abruzzo, Deputazione abruzzese di storia patria I, Padova 1991, 429-39. Buonocore, M. and G. Firpo, eds. (1991) Iuvanum. IN: Fonti latine e greche per la storia dell’Abruzzo, Deputazione abruzzese di storia patria I, Padova 1991, 440-54. Candelero, A. (1990) Alcuni aspetti della zona Sud del Foro. IN: Iuvanum. Atti del Convegno di Studi. Chieti Maggio 1983, 95-9. Campanelli, A. (2001) Il tesoro del lago. L’archeologia del fucino e la collezione Torlonia. Ascoli Piceno. Campanelli, A. (2002) Alba Fucens, rivive la piccola Roma d’Abruzzo, Ascoli Piceno. Coarelli, F. and A. La Regina eds. (1984) Alba Fucens. Abruzzo Molise. IN: Guide Archeologiche Laterza, Roma, 62-98. Coarelli, F. and A. La Regina eds. (1984) Iuvanum. Abruzzo Molise. IN: Guide Archeologiche Laterza, Roma, 312-6.

Another series of tavernae built during the first development of the forum of Sepino, again situated at the crossing of the two roads, is placed on the northern long side of the public square, corresponding to the alignment of the Decumanus road. Conclusions The construction and the monumentalisation of the forum of Iuvanum, with its porticos and tavernae, seem to have had several phases, as can be confirmed by the stratigraphic evidence. The vast amount of data and material evidence (Della Valle 1996: 165-237; Mancini 1996: 75-80; 1996: 99-106; Martella 1996a: 83-98; Martella 1996b: 119-24; Menozzi 1996: 107-18) arising from these excavations, which constitutes an on-going study, demand particular attention. The process of cataloguing, restoring, illustrating and analysing these materials is almost complete and my future investigations will provide complete analyses of the urban fabric. From the study of the wall construction, the density of ceramic spreads including black-glazed pottery, italic 49

SOMA 2004 Coarelli, F. and A. La Regina eds. (1984) Saepinum. Abruzzo Molise. IN: Guide Archeologiche Laterza, Roma, 209-28. Della Valle, D. (1996) I vetri. IN: Fabbricotti, E., Iuvanum. Atti del II Convegno di Studi. Chieti, Marzo-Aprile 1992. Università degli Studi ‘G. d’Annunzio’ Chieti, Dipartimento di Scienze dell’Antichità, Chieti, 165-237. Di Luca, D. (1996) Il vano B. IN: Fabbricotti, E., Iuvanum. Atti del II Convegno di Studi, Chieti Marzo-Aprile 1992. Università degli Studi ‘G. d’Annunzio’ Chieti, Dipartimento di Scienze dell’Antichità, Chieti, 37-42. Di Niro, A. (1977) Il culto di Ercole tra Sanniti Pentri e Frentani. IN: Documenti Antichità Italiche e Romane IX, 1977, 10. Fabbricotti, E. (1990) Il sito di Iuvanum. IN: Fabbricotti, E., Iuvanum. Atti del Convegno di Studi. Chieti, Maggio 1983, Chieti, 9-18. Fabbricotti, E. (1990) Un sarcofago di Iuvanum. IN: Fabbricotti, E., Iuvanum. Atti del Convegno di Studi. Chieti Maggio 1983, Chieti, 142-4. Fabbricotti, E. (1992) Cambiamenti di vita a Iuvanum. Papers of the Fourth Conference in Italian Archaeology, London, 77-82. Fabbricotti, E. (1996) Dieci anni di scavi a Iuvanum. IN: Fabbricotti, E., Iuvanum. Atti del II Convegno di Studi. Chieti Marzo-Aprile 1992. Università degli Studi ‘G. d’Annunzio’ Chieti, Dipartimento di Scienze dell’Antichità, Chieti, 21-36. Fabbricotti, E. (1997) Iuvanum. Sottostrutture della città di Iuvanum. IN: Campanelli, A. and A. Faustoferri eds., I luoghi degli Dei. Sacro e natura nell’Abruzzo italico. Chieti, maggio-agosto 1997, 67. Firpo, G. (1990) Linee di storia iuvanense tra IV e I secolo a.C.. IN: Fabbricotti, E., Iuvanum. Atti del I Convegno di Studi. Chieti Maggio 1983. Chieti, 35-49. Iaculli, G. (1990) L’iscrizione pavimentale del Foro. IN: Fabbricotti, E., Iuvanum. Atti del Convegno di Studi. Chieti Maggio 1983. Chieti, 81-9. Lapenna, S. (1996a) Quadri e il territorio di Iuvanum. IN: Fabbricotti, E., Iuvanum. Atti del II Convegno di Studi. Chieti Marzo-Aprile 1992. Università degli Studi ‘G. d’Annunzio’ Chieti, Dipartimento di Scienze dell’Antichità, Chieti, 65-74. Lapenna, S. (1996b) Quadri. IN: Enciclopedia dell’Atre Antica Classica e Orientale, II° supplemento IV, 687. Lapenna, S. (1997) Iuvanum. Il santuario italico. IN: Campanelli, A. and A. Faustoferri eds., I luoghi degli Dei. Sacro e natura nell’Abruzzo italico. Chieti, maggio-agosto 1997, 64-6. Lloyd, J. A. (1996) Il Sannio nella tarda età imperiale. IN: Fabbricotti, E., Iuvanum. Atti del II Convegno di Studi. Chieti Marzo-Aprile 1992. Università degli Studi ‘G. d’Annunzio’ Chieti, Dipartimento di Scienze dell’Antichità, Chieti, 11-20.

Madonna, A. (1930) Iuvanum. Santa Maria del Palazzo. Torricella, Fallascoso, Montenerodomo, Roma, 2364. Mancini, M. C. (1996) La ceramica a vernice nera. IN: Fabbricotti, E., Iuvanum. Atti del II Convegno di Studi. Chieti Marzo-Aprile 1992. Università degli Studi ‘G. d’Annunzio’ Chieti, Dipartimento di Scienze dell’Antichità, Chieti, 75-80. Mancini, M. C. (1996) Le pareti sottili acrome. IN: Fabbricotti, E., Iuvanum. Atti del II Convegno di Studi. Chieti Marzo-Aprile 1992. Università degli Studi ‘G. d’Annunzio’ Chieti, Dipartimento di Scienze dell’Antichità, Chieti, 99-106. Martella, G. (1996a) La terra sigillata italica. IN: Fabbricotti, E., Iuvanum. Atti del II Convegno di Studi. Chieti Marzo-Aprile 1992. Università degli Studi ‘G. d’Annunzio’ Chieti, Dipartimento di Scienze dell’Antichità, Chieti, 83-98. Martella, G. (1996b) Le fibule. IN: Fabbricotti, E., Iuvanum. Atti del II Convegno di Studi. Chieti MarzoAprile 1992. Università degli Studi ‘G. d’Annunzio’ Chieti, Dipartimento di Scienze dell’Antichità, Chieti, 119-24. Mazzitti, W. (2000) Iuvanum, città dei Sanniti Carecini. IN: Mazzitti, W. and G. Lattanzi eds., Abruzzo, una storia da scoprire. Pescara, 107. Menozzi, O. (1996) La ceramica a pareti sottili grigie. IN: Fabbricotti, E., Iuvanum. Atti del II Convegno di Studi. Chieti Marzo-Aprile 1992. Università degli Studi ‘G. d’Annunzio’ Chieti, Dipartimento di Scienze dell’Antichità, Chieti, 107-18. Paci, G. (1990) Il materiale epigrafico iuvanense e il suo contributo alla storia del municipio. IN: Fabbricotti, E., Iuvanum. Atti del Convegno di Studi. Chieti Maggio 1983. Chieti, 51-75. Papi, R.(1997) La rappresentazione di Ercole nella produzione votiva di area sabellica. IN: Campanelli, A. and A. Faustoferri eds., I luoghi degli Dei. Sacro e natura nell’Abruzzo italico. Chieti, maggio-agosto 1997, 142-3. Pavolini, C. (1986) La vita quotidiana a Ostia, Roma. Pliny The Elder (1982) Storia Naturale di Gaio Plinio Secondo, voll. I-IV. Torino, Einaudi. Ruscitti, G. (2000) Il risveglio di Iuvanum. Abruzzo Beni Culturali, 15, III, 8-11. Staffilani, P. (1992) Cambiamenti di vita a Iuvanum II. Papers of the Fourth Conference in Italian Archaeology. London 1990, London, 83-91. Staffilani, P. (2000) Il risveglio di Iuvanum. Abruzzo Beni Culturali, 15, III, 12-15. Staffilani, P. (2001) Iuvanum. Una cittadina di montagna. Archeo, 5, 40-7.

50

Central Adriatic Sea: real ‘importuosa litora’? Domenico Fossataro

This paper represents a preliminary result of a project begun in 2002 and carried out by the author (who focused on the archaeological evidence), in collaboration with Pierluigi Di Clemente (who analysed the literary sources and presented a related poster at the conference in Dublin) and graduate and undergraduate students of Chieti University. However, the project has since developed significantly and is currently a five-year project with a wider team of researchers, including the local Soprintendenza and various scholars from Rome, Chieti and other universities. The aim of the project has been, from the beginning, to find evidence attesting to the presence of ancient harbours in the Abruzzo region through a cross-study of literary and archaeological evidence.

Pliny (Nat. Hist., III, 106) portrays a different situation from that of Livy, referring to the region as ‘portuosus’, which is in direct, and probably deliberate, contrast to the Livian ‘importuositas’. This apparent dichotomy may be the result of a discontinued use of the coasts, rather than any negative feature within the region, and consequently, the ancient sources represent the different situations within different time periods, as well as reflecting their own specific sources and the political situation of the area. With a view to the geographical conditions of the area there are no adverse streams which prohibit effective navigation of the Adriatic (see also Castagnino Berlinghieri 2003: 1726, 157-62). In fact there are three kinds of currents: a very deep current, running from Greece to Italy through Venice; another, following the same direction but with stronger intensity which causes some problems for navigation for certain brief periods of the year; and a third stream, and the most important for our study, which is directly influenced by the winds. Using information from the Aureonautica Militare Italiana (Agenda Nautica 1993-2000: 212-15) it was possible to present a combination of these streams which resulted in a cyclical stream which changes direction from season to season, favouring navigation from Greece to Italy through Venice during the summer and vice versa during the winter. From an analysis of this model, it becomes clear that there is an alternative navigation route also, indeed offering better conditions, which could be used simultaneously. This alternative route connects the western coasts of the Adriatic directly to the eastern coasts during the winter and vice versa during the summer.

An analysis of the ancient sources, although still in a preliminary stage, has identified several authors, both Greek and Latin, who refer to ports in this area. However, these sources often provide contradictory evidence, presenting two opposing images of the coasts of this region. Livy (X, 2-4) claims that the coasts are unsuitable for ports, while Strabo (5.4.2), refers to the existence of ‘epineia’, the ancient Greek term for ports of non-maritime centres which were generally not located far apart (between 5 and 10 km). The term ‘epineia’, however, does not necessarily imply that such harbours were small outlets, considering that ports such as Piraeus and Ostia were considered the epineia of Athens and Rome respectively; rather the term refers specifically to the distance between the main town from the harbour settlement (to which we can probably attribute the status of ‘non-town’). Despite the fact that the literary study is still in its preliminary phase it is already clear that the ‘Livian Tradition’, which attests the absolute lack of ports, contrasts with the Greek sources which mention some epineia in Abruzzo, in particular near ancient Hordona in the area called ‘Scalo’. The toponymn ‘Scalo’ is particularly relevant as in Italian it refers to a ‘maritime outlet’ which supports the commercial significance of the area and suggests the possibility of navigation of these waters. However, there is not even agreement among the Latin sources, in fact

Regarding the main focus of the project, i.e. the identification of physical archaeological remains of ancient ports, the data is already promising, especially when viewed in combination with the literary evidence and the general historical background. The first phase of the practical investigation concerned the remote sensing of Abruzzo’s coasts using two kinds of images: a Zulu mosaic with a definition of 15 metres per pixel, with 7 chromatic bands, and then using a landsat of 51

SOMA 2004 the same area with the same chromatic bands but with a different definition. Subsequently, the terrain was analysed with the aid of aerial photographs taken in 2000 by VoloItalia (an Italian company specialising in these types of images). These were scanned and geo-referred to obtain high-definition images that were easily interpretable and topographically suitable and reliable. The second phase involved fieldwork whereby specific methodologies were adapted in relation to the diverse topographical conditions associated with the archaeological remains. The archaeological evidence fell into two main categories of topographical features: the first category includes submerged ports, while the second consists of silted-up remains. The latter was investigated using the standard intensive survey methods. The areas were surveyed in cross transects with walkers spaced at intervals of 10m, and consequently, a sampling strategy covering 40% of the surface area was attained. In cases which yielded dense and positive results, selected areas were marked out for re-survey using a more intense strategy. The intensive survey consists of walking cross transects at intervals of 5m and thereby yielding a higher surface coverage of 80%. Currently, the survey area has only been walked using the lower sampling strategy, while the second and more intensive phase is still in progress.

FIGURE 1 GIS OF THE MAIN PORTS: WITH STARS FOR FLUVIAL PORTS AND SQUARES FOR SEA PORTS (DOMENICO FOSSATARO) In the investigations focusing on Northern Abruzzo the first site is Castrum Truentinum, modern Martinsicuro (no.1 in Fig. 1). The location of the settlement which related to the fluvial port structure has been established through limited investigations (Staffa 2001: 348). The site has been located at the mouth of the Tronto River, a river which was certainly large and partially-navigable in antiquity. It was probably used in pre-Roman times by local inhabitants to move goods from the coasts to the countryside and vice versa. According to the archaeologist working on this site, the port was in continuous use from the fourth century BC to the seventh century AD (Staffa 2001: 348). Moreover, as we know from the ancient sources, the site was well-known in Roman times: it is mentioned in the Peutinger Table (Segm. IV, 5); and by both Strabo (V, 4,2) and Pliny (Nat. Hist, 110) who refers to the site as a Roman Municipium. The ancient sources provide important information regarding the possible navigability of this river and the strong urban relationship between the settlement and the port structure. Moreover, the important strategic position of the village at the crossroads of the Salaria and the coastal road network must also be considered. Also, the ancient sources provide information regarding the mythical foundation of the city by the Illyrian tribe of the Liburni, with whom Castrum Truentinum is thought to have maintained trade contact (Pliny, Nat. Hist., III, 13, 110).

For the remains of ancient ports situated in submerged levels, specific systematic immersions are applied according to the diverse water conditions in order to attain the highest possible visibility. Once the most promising areas were identified, we applied a specific methodology in order to engage in a more intensive survey which avoided any disproportionate sampling strategies which may arise from difficulties involved in sea surveys. Therefore, we initially conducted the underwater survey by boat using an eco-sonar which traced parallel lines at intervals of 10m along a north-south, and then a west-east axis. This produced a grid marking the locations of the main finds using a 40% sampling strategy which corresponds to our field surveys. Once the most promising underwater densities were established, we engaged in more intensive immersions in these areas which again corresponded with the similar strategy conducted in the intensive field survey (although weather conditions and currents do not allow for the precise level of identification which is associated with the sampling methodology of a ‘normal’ field survey). The preliminary results of the project, including survey and analysis of the sources, is presented in the GIS map (Fig. 1), where the ancient ports, relating to the two topographical types identified in this research, have been distinguished. In fact, there are two different kinds of ancient ports in the region: fluvial ports (marked by stars in Fig. 1), along rivers which were once navigable, as in the case of Ostia Aterni and the Sangro rivers; and seaports (marked by squares in Fig. 1) which were built directly on the coast, in certain specific areas, which related directly with particular towns.

The site of Castrum Novum, modern Giuglianova, is located to the south of the region (no.2 in Fig. 1). The original settlement became a Latin colony after 290 BC, but the archaeological record provides evidence for an earlier settlement. This location was probably chosen for the colony due to the presence of, or potential for, a fluvial port associated with the Tordino River; a river which was certainly used for navigation, as is evident from its topography (traces of the ancient beach have been identified). 52

DOMENICO FOSSATARO In locating the actual structural remains of the ancient port it must be acknowledged that the river deviated, and its mouth was relocated some 640m away (Staffa 2001: 351352), where it is said ‘si può trovare riscontro nel Rilievo della Costa Adriatica eseguito tra il 1830 e 1835 dal Regio Ufficio Topografico di Napoli.’ However, the area of the ancient port can be easily located near the church of S. Maria a Mare (a most appropriate toponymn), where some structures, probably dockyards, have been identified as the ancient port.

early settlement seems to have been regularly organised following the topography of the hill. As is the case for the other colony within the region, Castrum Novum, Hatria was chosen for a Latin colony due to its location and position whereby it could act as a stronghold in the heart of the strategic area controlling the coasts of the central Adriatic. Its position also effectively monitored the activities of the Samnite League and the territories of the Sabini and Piceni, thereby physically blocking any possible alliance between them (Menozzi forthcoming).

A similar identification of a fluvial port can also probably be applied to the mouth of the Vomano River (no.3, Fig. 1). Unfortunately, the intensive urbanisation of this area does not allow for a clear interpretation or precise location of the structure. While research on this site is still in progress, the possibility that this river was navigable, using small boats to transport local agricultural produce to the coasts was probable at least in medieval times.

After the foundation of the Latin colony, the settlement went through different phases of urbanisation; the first phase, dating to the Augustan period, involved a reorganisation of the town plan. This was followed a by programme of monumentalisation in the Julio-Claudian period superseded by a major embellishment of the town, including the construction or restoration of monumental complexes such as baths and the theatre, in the Hadrianic period (Azzena 1987).

Moving further south, one of the sites more firmly documented as an ancient port is the ancient port of Hatria, modern Atri, (Menozzi 1995: 217-39) (no. 4, Fig. 1), located c. 6/7 km from the town in the area now known as Cerrano (no. 5 in Fig. 1). In pre-Roman times Hatria was a Pretutio-Picene settlement, inhabited therefore by Italic peoples who shared a homogeneous culture and a common social organisation with their neighbouring tribes. This has long been established by various scholars in the past who have applied a variety of terms to the group, such as ‘South Picenes’, ‘Eastern Italics’, ‘Sabellians’ and so on (Menozzi forthcoming). The term Sifin is popular in modern scholarship as this is the term attested in the epigraphic record, specifically in the famous ‘Stele Paleosabelliche’ from Penna S. Andrea (Prosdocimi 1982: 35-55). Hatria was a flourishing settlement when a Latin colony was established in 289 BC; however, evidence from two cemeteries discovered in the area confirms that the settlement had been thriving since the end of the sixth or beginning of the fifth century BC (Martella 1995: 223-229). Recent extensive excavations in the town have revealed two phases relating to the indigenous settlement, dating to the Archaic and a pre-colonial period. The

In the aerial image of Cerrano, Hatria’s port, the remote sensing and the eco-sonar data have been effectively mapped onto an aerial photograph which was preliminary horto-rectified. In this image two areas presenting interesting ‘anomalies’ attesting the presence of structures are discernable. In area A, a northern L-shaped pier is evident. The long side of the ‘L’ runs perpendicular to the coastline and has already been the subject of several past investigations (Data and Angeletti 1983: 117-23; Menozzi 1995: 217-39). The second area, B, presents further anomalies which are more difficult to interpret and could either represent archaeological remains or could reflect the natural undulations of the seabed. Consequently, they will be the focus of intense surveying next season. The preliminary maritime surveys of the remains along the waterfront at Cerrano, at depths between 4m and 11m, have yielded interesting data regarding the structure of this port, data which effectively complements the vague image offered by the ancient sources. The L-shaped northern pier represents the main structure. It is constructed of large squared stone blocks (2m x 4m x 4m, or 2m x 6m x 4m), the majority of which remain in their original position but some have been moved slightly by the maritime streams (Data and Angeletti 1983: 117-23, figs 1-6). At about 100 metres from the coast, numerous structures have been discovered which probably belong to storerooms and commercial buildings related to the life of the port. Among them, different types of walls have been found: constructed in opus quadratum (with rectangular blocks laid down regularly) as well as in opus latericium (brick facing) and also in opus vittatum (alternating courses of rectangular blocks and bricks). Numerous architectonic elements have also been found, such as moulded frameworks, a fragment of a fluted column or pilaster, and the remains of stone drainage channels. One structure among the architectural

FIGURE 2 REMOTE SENSING OF CERRANO’S AREA IN WHICH TWO MAIN AREAS OF REMAINS ARE IDENTIFIED (DOMENICO FOSSATARO) 53

SOMA 2004 Remote sensing indicated two main areas in Hortona, modern Ortona (n.8, Fig. 1): area A clearly represents the site of the Roman port, while area B, enclosed within the structures of the modern harbour, represents the medieval port, part of which is still visible, located just below the ‘Castello Aragonese’ which dates to the post-Medieval period. The result of remote sensing in area A, this time without ecosonar, but with the results of thermo-photo interpretation, established the position of the ancient pier, thereby confirming the accounts of the ancient sources which refer to the harbour as an ‘epineion’. However, if Hortona is the ‘port of a non-maritime town’, as ‘epineion’ has been interpreted, as mentioned earlier, then where is its main town? This apparent dichotomy is probably the result of anachronistic detail within the ancient texts. The original settlement of Hortona was probably related to the harbour and served some internal and more important town, possibly Teate, modern Chieti, located at a distance of 18 km (no.9 in Fig. 1). However, subsequently, Hortona became an important developed urban centre in and of itself, gaining the role of town (albeit due to its important location on the coast) and discarding that of port-settlement.

FIGURE 3 CERRANO: PROBABLE BASEMENT OF A MONUMENT OR A BEACON (AFTER MENOZZI 1995) remains is of particular interest (Fig. 3): it is a type of large circular depression limited by three concentric steps which become gradually smaller. This could be interpreted as the foundation for a monument or a beacon. Ostia Aterni, modern Pescara, is the next highly significant archaeological site (no.7 in Fig. 1). The settlement became an important town after its municipalisation. The triangular shape of the Roman town plan (Fig. 4) is due to its location between the river and the main ancient road approaching the site. Recent excavation along the river has brought to light the remains of a long bench belonging to the port (Staffa 2001). The importance of the port is clearly attested in an inscription mentioning the reconstruction of the port by Dioclezianus who used this harbour to reach Salona on the Illyrian coast. As is clear from the schematic plan (Fig. 4) the bench runs along the riverside parallel to the remains of a paved road. Moreover, the discovery of several piers of a Roman bridge not far from the bench of the port is also deeply significant. The archaeological evidence clearly represents a fluvial port at the mouth of the Aternus river from which its name Ostia Aterni is derived, literally the ‘mouth of the Aternus’. Moreover, the importance of this port is heightened by fact that the Via Claudia-Valeria, connecting Rome with the Adriatic coast, passed through here.

In area A squared stones, possible representing structures, were discovered during a preliminary underwater survey despite poor visibility. These now also need a systematic survey season. Numerous finds came from the area of the port attesting maritime navigation, such as several anchors, amphorae and amphora stoppers complete with stamps and decoration. These stamps probably relate to the content of the pots or the pottery-production centre. Among the stamps, one in particular is extremely important because it belongs to a Rhodian amphora, attesting trade with Greek areas. Before concluding, it must be mentioned that there are three other interesting areas where the presence of ports is also highly probable but which have not, as yet, been the focus of our research. The first site is the port of Angulum, modern Città S. Angelo (n.6 in Fig. 1) between Hatria and Ostia Aterni, which was probably a seaport. This port may have been closely related to the salt trade, as is attested by the ancient toponymn ‘Saline’, still used locally and also attested in the Peutinger Table as ‘Statio ad Salinas’ (Segm. V,1). Another important area, relating to the fluvial harbour type, is the Sangro river (n.10 in Fig. 1). This river is significant due to its navigability and should yield multiple internal ports mainly to service local trade and transport and, consequently, the area will be the subject of a more detailed study in future seasons. Lastly, the area of Histonium, modern Vasto (n.11 in Fig. 1), is extremely interesting, but has not been discussed here as the survey is still in progress. Also it had two ancient ports used in Republican and Imperial times, which are still very highly debated and require further research focusing on both literary and archaeological sources. However, these two harbours must be mentioned among the seaports which are not closely related to rivers.

FIGURE 4 THE AREA OF PESCARA, IN WHICH THE ANCIENT ATERNO AND THE REMAINS OF THE FLUVIAL PORT ARE SCHEMATICALLY REPRESENTED

(DOMENICO FOSSATARO)

54

DOMENICO FOSSATARO discussed above, it was necessary to establish specific, although consistent/corresponding, sampling strategies for their intensive survey. Currently, the project has developed into a wider and longterm research investigation and future seasons will engage in the continued survey of the submerged ports, focusing on the intensive field-survey of the silted-up ports, including the smaller outlets points. Bibliography Agenda Nautica (1993-2000) Istituto Idrografico dell Aeronautica Militare Italiana. Azzena G. (1987) Atri, Roma. Castagnino Berlinghieri, E.F. (2003), The Aeolian Islands: crossroads of Mediterranean Maritime Routes, BAR International Series 1181, Oxford. Data, P. and G. Angeletti, (1983) Cerrano ieri e oggi, S.Atto-Teramo. Martella, G. (1995) L’Ager Hatrianus: sfruttamento economico del territorio, contatti e scambi commerciali, IN: J.Swaddling, S.Walker, P. Roberts eds., Italy in Europe: economic relations 700 BC-AD 50. Occasional Papers of the British Museum 97, London, 223-39. Menozzi, O. (1995) L’antica Hatria Picena e l’uso del suo porto nell’antichità. IN: Swaddling, J. Walker, S. And P. Roberts eds., Italy in Europe: economic relations 700 BC-AD 50. Occasional Papers of the British Museum 97, London, 217-39. Menozzi, O. (forthcoming) Cultural Change and the problem of Romanization (PhD Thesis, University of Oxford). Prosdocimi, A. (1982) I Safini delle iscrizioni Sud-Picene, Preistoria, Storia e Civiltà dei Sabini Rieti, 35-55. Staffa, A. (2001) Abruzzo: strutture portuali e assetto del litorale fra antichità ed alto medioevo, Strutture portuali e rotte marittime nell’adriatico in età romana, Trieste-Roma.

FIGURE 5 REMOTE SENSING OF ORTONA’S AREA, WITH THE MAIN AREA OF INVESTIGATION (DOMENICO FOSSATARO) In conclusion, the view of the central Adriatic coasts as ‘importuosa litora’, a label so popular among the scholars for far too long, can be seriously questioned in view of both the literary sources and the recent archaeological discoveries presented in this paper. The archaeological investigations, although only in the early stages of the project, indicate impressive port structures and large harbour-related settlements. Consequently, the ‘importuositas’ attested in the sources has to be confined to certain periods or attributed to political circumstances such as piracy or an exaggerated pro-Roman view of Roman sources. The typologies of the ports as catalogued seem to fall into two groups, i.e. fluvial and seaports, the latter probably representing the main harbours. However, there must have been even smaller outlets which remain difficult to identify, and which could be identified in the future as a third type. The topography of these ports is geographically distinctive, and they can be sub-divided into the two following forms: submerged and silted-up remains. As

55

Understanding the Limits of Pollution: The Use of the Dead Florentia Fragkopoulou

Introduction Death is both a natural event, ending the existence of all beings on earth, and, in the case of human beings in particular, a social event and process. Death prevents the human body as an organism from fulfilling all its physical functions; at the same time the human being stops existing as a social being – and the community is deprived of one of its members. Hence the event of death – as opposed to the moment of a person’s passing – is a process. What follows a death is clearly and strictly defined by the community of the living, making the process ultimately a social one.

About Death We have already stressed the fact that death is both a social event and a process. Because of this, societies employ symbols to speak about death in the context of rituals and these rituals involve actions and rules. Death in particular is a rite of passage. It denotes a change, both in the status of the person who has just passed away – his soul has left this world to integrate into the spirit world – and in the life of the society which has just lost one of its members and thus has to regain its coherence without him/her (Garland 1985). This change, stressed by death as a ‘rite of passage’, is signified by the uncertainties and disorder death involves (Douglas 1966: 91). This disorder concerns both metaphysical and social aspects of death: the living relatives have to secure, on the one hand, the successful integration of the deceased’s soul into the spirit world and on the other hand their own return to social life. Their aim is to re-establish the order which was disturbed by death. In order to achieve this, the living have to follow a particular death ritual – in other words they have to act in a specific way, obeying specific rules.

Yet no one feels very comfortable talking about death (Tarlow 1999: 6; Metcalf and Huntington 1991: xi). It is therefore difficult to understand the exact meaning of the actions or the beliefs of the living in relation to death, especially when we are dealing with societies of the past – in our case society in Greece during the Classical period. As Morris puts it, people use symbols in order to ‘make explicit the social structure’ of the community in which they live (1992a: 1). We should thus approach our topic with the understanding that societies employ symbols to cope with events like death. These symbols are employed in the context of rituals, and the rituals offer a way of interpreting everyday activities within a society. This was the case in Greece during the Classical period on which we will focus.

But what do ‘order’ and ‘return to normal social life’ mean? Once death occurs the body of the person who has just passed away becomes a source of ‘corruption’ in both physical-literal and social-symbolic terms. The corpse rots, thus becoming a danger to public health. What is more, everyone around it becomes polluted by it in a symbolic way: the people polluted by the corpse are deprived for a certain period of time – until they are purified – of their right to participate in the social life of the community as they did before the pollution incurred by the death (Garland 1985; Kurtz and Boardman 1971). The problem of the appearance of pollution in social terms (what the ancient Greeks defined as miasma) becomes more complex when one considers that the deceased’s body is a source of pollution, yet at the same time, a consecrated object. So we deal with a paradox: the corpse is a pure source of impurity (Vernant 1982: 116). The result of this impurity, of miasma, is the social isolation of the polluted living people, as they are regarded as a source of pollution for the community, threatening its well-being (Adkins 1960:

In particular, this paper is interested in the precise meaning of the ‘pollution’ (miasma) that accompanies death in the classical Greek world and asks what exactly this ‘pollution’ is in connection with the ‘use of the dead’. And why have I chosen the phrase ‘the use of the dead’? I will try to answer these questions by examining how the living reacted to the dead, and what their actions suggest about the way society faced death in classical Greece, since the implications of their actions and beliefs about death can reveal, in part, the way in which the contemporary community functioned on an everyday level. 57

SOMA 2004 88). However there are many cases in everyday life, as we shall see, when this pollution is ignored. So the questions are, how is miasma defined and how is it treated by the community?

the funeral, and, in consequence, the display of wealth and prestige. Going through the text we find several forms of the verb ‘miaino’4 – ‘pollute’. The defiled women are characterised as ‘miainomenas’ (A, 26); the infinitive ‘miainesthai’ is employed to describe who the contaminated relatives are (A, 26). It is adequate for the polluted people to wash themselves in order to become purified (A, 30-31, B 6, 10). The house of the deceased is similarly to be cleaned, ‘diarrainein’, by a free man, ‘eleutheron’ (A, 15; Arnaoutoglou 1998: 142).

A lot can be revealed about the organization of social life in ancient Greece during the Classical period through an examination of how the normal conditions of social life in relation to death and all that surrounds it are achieved. Therefore, in order to see how the Classical Greek world approached the pollution provoked by death, I will examine the available epigraphic and archaeological evidence. However, due to the nature of the material, at our disposal,1 I will focus primarily on the available epigraphic evidence. I will also refer very briefly to the archaeological evidence.2

The defiled women are, in a way, privileged as they are the only women allowed to enter the deceased’s house.5 The law defines these women as the mother, the wife, the daughters and the sisters of the deceased; a maximum of five other people who belong to the close kin of the dead person are considered to be defiled as well (A 25-30). It seems strange that people polluted by the deceased are defined by the degree of kinship; in theory, at least, anyone who comes into contact with the corpse, or any living relative who is away from the corpse, even those who have not touched it, should be polluted by it, by virtue of their relationship to it (Garland 1985; Parker 1983: 39; Adkins 1960: 88). We might, therefore, expect no limitations upon the number of people who may become contaminated, whether they are relatives of the dead person or not. Hence, the question of what leads the community to consider someone polluted remains somehow obscure. As Garland says, it is a kind of ‘right and privilege which needed to be restricted to a few’ (1985: 41). However this does not explain this ‘limitation’ of contamination. Parker compares this group of people who are supposed to be contaminated to the Attic kinship (anchisteia), which, according to Solon’s legislation, granted the right to participate in the mourning (1983: 40; Alexiou 1974: 10). According to the provisions of the Iulis law, the women have to return to the house before the men after the end of the funeral (A 19-20). Thus pollution appears to be a kind of contamination which has a special and more severe implication for women, a fact which we also find in the Delphi Law, as well as in Solon’s legislation. Finally the relatives are permitted to commemorate the third day as well as the one-year anniversary from the death but not the thirty days after death.6 In contrast to the other three laws considered, we observe here the absence of any kind of sanction concerning the offenders of the law.

The epigraphic evidence All the texts examined here come from the Greek world during the sixth to the fourth centuries BC. They are: the funerary law from Keos, the so-called ‘Iulis law’, (fifth century) the legislation of Solon in Athens (sixth century), the funerary law of the Labyadai (fifth century) and the purification law from Cyrene (late fourth century). The reason for choosing these four laws is the fact that they constitute the prescription of the community as well as a testimony of the contemporary social organization, referring to death to a certain extent at least. So the information they offer can help to clarify the nature of miasma. The funerary law from Keos (Iulis), dated to the late fifth century BC (Michel 1900: 398, A-B), exclusively concerns death and the concomitant need for purification. By contrast, the law of the Labyadai and the legislation of Solon both refer to death but do not stress at all the fact that death provokes pollution, while the sacred law from Cyrene refers to the need for purification, but not due to the pollution caused by death as we shall see. In the law from Keos we find restrictions concerning the procession, and the expenses of, the funeral (A 6-14, 46) as well as the behaviour of the mourners (A 11-12, 19). The law also speaks about the defilement brought by the corpse and the way this can be dealt with (LSCG 97; Arnaoutoglou 1998: 142). Many details concern the amount of money spent on the funeral. The body has to be wrapped in three garments; a further wrapping is allowed, but its value cannot exceed 100 drachmas (A 2-6). Furthermore only three choes (pouring out of liquid, drink-offering, especially made to the dead or over their graves)3 of wine and one of oil are permitted to be brought to the grave (A 9-10). Overall there is an attempt to limit the expenses of

Pollution and purification are central elements in the sacred law coming from Cyrene (LSS 115). In this text we find the word ‘miai’ (A 40, 41), which according to Parker means ‘pollution incurs’ (1983: 340). A man who had intercourse

1. Stears refers to it as ‘[A] composite picture has to be extrapolated from literary sources as the fifth-century tragedians, fourth century orators and the Byzantine lexicographers, together with archaeological evidence from painted pottery and burials’ (1988: 113). 2. The archaeological data demands further analysis due to its complicated character which time and space do not permit here. 3. cf. LSJ , s.v. χοή.

4. In terms of the ritual defilement in funerary ritual, cf. LSJ s.v. μιαίνω. 5. I will refer to the way women are treated by legislation in more detail in what will follow. 6. The commemoration of the dead on the third, ninth, thirtieth or oneyear anniversary after the day of death is common practice in Classical Greece. Offerings were made to the dead so that his/her souls would be appeased (Kurtz and Boardman 1971:99 ff).

58

FLORENTIA FRAGKOPOULOU constraint that the women are permitted to carry food or drink worth no more than one obol. In addition the dead should be buried with no more than three changes of raiment. No ox is allowed to be sacrificed at the grave (XXI.5). Solon also forbade laceration of the flesh, the use of set lamentations as well as the bewailing. If one does not follow these provisos then one is to be punished by the board of censors. Extravagance in mourning is considered to be anandros (cowardly)9 and gynaikodes (womanish)10 (XXI.5). The last services and salutions to the dead are to be performed in any manner desired, but only in private.

by day and wants to sacrifice needs purification (A 11-15), a woman in childbed pollutes the house (A 16-20), for example. Purification is also mentioned in the last lines of the A column where a significant part of it is missing (A 7382). However, contamination by death is not mentioned. The funerary law from Delphi (Michel 1900: 995; C 2051) focuses on the expense of the funeral, as well as on the definitions of lamenting and mourning: one should spent no more than 35 drachmas on the burial articles. Just one mattress and one pillow are to be used for the bier; otherwise, there will be a fine of 50 drachmas (C 19-28). There seems to be an attempt to control the whole cost of the funeral and the funerary display (Nielsen, Bjertrop, et al. 1989: 413; Morris 1992b: 45) in the same way seen in the Keos law. Moreover, the deceased must be carried in silence to the grave. Lamenting and wailing are not allowed outside the house before reaching the grave, and the lamenting must refer only to the one who has just passed away. Lamenting is forbidden on the second and the tenth day of commemoration as well as on the annual anniversary after the death (C 39-46). And as it appears there was some kind of punishment for the offenders – the text is missing, but one can read ‘if anyone transgresses any of these written rules …’ (C 50-52).

All these funerary laws show how the definition of the behaviour of people relating to the deceased is central to the city legislation. They all aim primarily at two things: they impose restraints, firstly upon the behaviour of the mourners, in particular of women (Stears 1988:113-28),11 which relates to the general social position of women, and secondly upon the amount spent on the funeral. The importance of purification due to pollution by the deceased is stressed only in the Iulis law. Death seems to be regarded more as a social event directly connected to wealth and prestige. But who decides all this? The only law that deems itself sacred is the law from Cyrene, which presents itself as an oracular response (A 1-3) (Parker 1983: 333). In the Keos law all the provisions are decided by ‘the Boule and the people’ (B 5). Since this law applies to a primarily metaphysical aspect of human life, one might not expect these decisions to be of clearly political character. Moreover, in the Delphi law the burial regulations are decided by the Labyades group, a phratry.12 Thus the fact that the restraints are voted by a hereditary group stresses the important role of this brotherhood in all aspects of life at Delphi. According to Vernant, religion has no intention of imposing severe punishments or restrictions in the case of defilement, as purity according to religion is ‘not an end itself’ (1982: 112). Therefore miasma is not a problem that can be faced effectively in clearly and strictly religious terms. The secular element, that is to say the group of people in political control as well as the community that follows its rules, plays an important role too.

Finally, there is the reform in Athenian legislation made by Solon during the sixth century (Ruschenbusch: 1966). Solon established specific rules concerning the appearance of mourning women (XXI.2.4). At the same time he forbade the visiting of tombs other than the one that belonged to the family (XXI.1.4-5). In Solon’s legislation, as handed down to us by Plutarch and Sositheos’ quotations (Broadbent 1968: 119-120), no word such as ‘miasma’ or ‘miaros’7 is used. It is striking that funerary legislation is once again particularly strict regarding mourning women, an element already seen in the Iulis law. We find no reference to the behaviour of men or to the way they are expected to dress during the mourning period. It is women who are considered to be a source of disorder and license. This is a common perception from an anthropological point of view.8 However, it is not the purpose of this paper to explore why the female plays such a role in the funerary process. What interests me here is the way miasma is manipulated and used by the living: the living defines who is polluted by the corpse, and which of these are more polluted than others. Thus the concept of miasma is defined according to the way the community of the living functions. They are permitted to wear no more than three garments (XXI.2.4). During the procession the men are the ones who proceed while the women have to follow. No woman is permitted to enter the place where the corpse lies, or to accompany it when carried to the tomb unless she is over sixty or within five degrees of kin (Broadbent 1968: 119). No reference is made to the expenses of the funeral itself except for the

Moreover the constant reference of every religious attitude was precisely the city (Zaidman and Pantel 1992: 101) in the sense that the relation of religion to politics was such that the former affected the latter in a way that each one ‘intruded’ into the matters of the other and vice versa (Garland 1989: 7). No strict separation of these two elements can be detected in social life, or in the case of death. 9. cf. LSJ s.v. ανανδρος. 10. cf. LSJ s.v. γυναικωδες. 11. For further discussion about the role of women in Classical Greece see Humphreys 1993; Farnell 1904; Bloch 1982. 12. Φρατρα, brotherhood, cf. LSJ.

7. Someone who gets defiled by the corpse, cf. LSJ s.v. μιαρός. 8. Stears 1998; Farnell 1904.

59

SOMA 2004 Thus the pollution provoked by death cannot be approached strictly in terms of religion or politics on their own, and nor does it appear to be defined in any concrete way. It appears to be absent from any official prescription of communal life, that is to say the laws. It appears to exist in everyday life, as a ‘common secret’ among people.

family burial peribolos. The sixth century Themistoclean wall was built so that it cut through the cemetery at Dipylon, including a part of it within the city boundaries. In addition there are burials throughout the sixth century in other areas later to be included within these walls (Papachatzis 1974: 146, 148, 168-74).

We may now turn to the archaeological data and ask if it gives any more precise information about the nature and content of miasma?

Thus, in Athens we see a city, where at different times, different parts of the circuit of the city walls were erected incorporating into the city places which had been previously used as burial grounds before the existence of any walls, or which had been used simultaneously both as burial and as habitation grounds.

The archaeological evidence During the Classical period, cemeteries were supposed to be located outside the city walls, away from the area occupied by the inhabitants of the city (Kurtz and Boardman, 1971: 178). This practice was related to the fear of miasma produced by the dead (Robinson 1942: 131).13

In the Agora there are also intramural burials of heroes (Camp 2003). The practice of burying the heroes within the city can be traced back to the eighth century as a gesture aimed at honouring the dead man who had offered his services to the city (Farnell 1921).

In order to see what the archaeological evidence suggests about the practice of burying the dead outside the city walls, and the existence of miasma, I will examine briefly the data from Olynthus and the Athenian Agora, since they are two well-investigated areas of Classical Greece.

In both Olynthus and the Agora, we can observe two different modes of interment: cremation and inhumation. In the case of cremation we find both primary and secondary cremations (Papachatzis 1974: 146-78; Young 1948: 377-8; 1951: 67-134).14 Another important element is that in Olynthus there are ‘unprotected’ graves. There was no particular form of protection for the corpse: it was covered with a shroud and another cloth was thrown over it to protect it from the earth. In most cases the unprotected graves contained child burials (Robinson 1942: 125-68).

Olynthus and the Athenian Agora: two case studies In Olynthus the three cemeteries are located outside the city walls. These are the Riverside cemetery at the south end of the North Hill, the North cemetery beyond the valley to the north, and the East cemetery along the road to Mecyberna. A few child burials are also found within the city boundaries (Robinson 1942: 125-68).

According to Seaford, the hero-cult is a form of cult which promotes social cohesion among citizens of different kinship and locality. The hero-cult is a form of death ritual and, as such, it becomes a duty of the kin. Since the hero has given his life for the whole community, his cult extends to a larger group than merely his immediate kin. Thus a group of people is united, not by actual kinship, but by the hero-cult, which provides a shared point of reference for the city: the hero’s tomb. In addition, the hero-cult was largely determined by the desire of the aristocratic clans to retain their prestige at a time when the demos was in control and the power of the individual was seen as unacceptable (Seaford 1995: 109-13).15 There was thus an acceptable form of ‘transgression’ of the rule that the deceased had to be buried outside the city walls.

In the area where the Athenian Agora was set out from the sixth century onwards, tombs from the Mycenaean period have been found. During this same period dwellings were also erected within the limits of the Agora (Kurtz and Boardman 1971: 177). It is obvious, therefore, that the area which was used as a burial place was used at the same time as a place of habitation from the sixth century onwards. During the sub-Mycenaean period, the Dipylon began to be used as a burial ground and this continued until the eighth century. This was a period when many dwellings began to appear in the area of the Agora. Nevertheless, burials within the Agora continued even during the archaic period. This phenomenon relates to the existence of family burial periboloi (Kurtz and Boardman 1971: 100); wealthy and politically important families maintained their family’s tradition of being buried in the specific area defined as their

What about the different modes of burial? Is there any symbolic difference between inhumation and cremation? No adequate explanation of the reasons behind cremation has yet been set forth (Clairmont 1983). It has been suggested that all corpses were burnt at least lightly for symbolic

13. There is no particular law concerning this particular practice. However, the letter Servius sent to Cicero about M. Marcellus, a friend of Cicero, who was murdered near Athens in 45 BC, suggests that the practice of burying the dead outside of the city-walls was in effect. Servius undertook the arrangements of the funeral of Marcellus, but he was denied a place of burial within the city because of religious usage - Ab Atheniensibus locum sepultrae intra urbem ut darent, impetrare non potui, quod religione se impediri dicerent; neque tamen id antea cuiquam concesserant… Cicero, Ad familiares, IV, 12, 3. Moreover, Plutarch refers to the fact that Lycourgus permitted the burial within the boundaries of the city, neglecting the fear of the dead and the pollution provoked by them, Plutarch, 27, 1-2.

14. Primary cremation means that as soon as the pyre built in a trench burned down the trench was filled with earth without moving the bones. Secondary cremation means that the remains of the corpse after having been burnt, are put in a vessel before being buried (Kurtz and Boardman 1971: 90 ff). 15. For further discussion on the matter of hero cults see Farnell 1921; Houby-Nielsen 1996; Kearns 1989; Whitley 1988; Whitley 1994.

60

FLORENTIA FRAGKOPOULOU reasons. We may find both cremations and inhumations within the same burial area. Cremation, however, is much more expensive than inhumation (Kurtz and Boardman 1971). For this reason, many see it is a practice used only by the higher strata of the society (Robinson 1942: 149-51). Another suggestion is that it was a matter of individual choice according to the region, the opportunity and the economic circumstances of the deceased. For sure, the decision to inhume may have been influenced both by the expense and by the inconvenience; the smoke and the odour from the burning pyre must certainly have been unpleasant. On the other hand, cremation was a means of preventing violence to the body – no animals would disturb the burials (Robinson 1942: 150). But was there any symbolic reason for the use of cremation? Fire is supposed to have a purificatory power (Parker 1983: 329). One might, therefore, expect all corpses to have been burnt so that the fear of pollution provoked by the corpse could be avoided. At the same time, many cases of inhumation have been observed where katadesmoi (curse-tablets) have been placed in the grave. In no case of cremation do we find curse-tablets (Gager 1992; Faraone and Obbink 1991). This might suggest that only the souls of inhumed corpses could be expected to carry out the task of delivering the messages of the curse-tablets to the chthonic deities. A symbolic implication in the use of the mode of inhumation or cremation seems an inevitable conclusion.

Moreover, the nature of the archaeological data does not seem to help us define miasma in a more concrete way. The rules that are followed in each case vary according to the place and time, making the definition of the concept of miasma even more complicated. What we are able to say about miasma is that there is not just one universal specific and strict definition of the concept of miasma. It acquires particular features according to the particular needs of the different communities over time. In this sense, miasma has to be treated and studied as a concept that exists but is dealt with by the contemporary community on a case by case basis. Bibliography Adkins, A.W.H. (1960) Merit and responsibility: a study in Greek values, Oxford, Clarendon Press. Alexiou, M. (1974) The ritual lament in Ancient Greece, Cambridge, Cambridge University Press. Arnaoutoglou, I. (1998) Ancient Greek Laws: a sourcebook, London, Routledge. Broadbent, M. (1968) Studies in Greek Genealogies, Leiden, Brill. Camp, J.M. (2003) Gods and heroes in the Athenian Agora, Athens, American school of Classical Studies at Athens, Princeton. Clairmont, C.W. (1983). Patrios Nomos: public burial in Athens during the fifth and fourth centuries BC: the archaeological, epigraphic-literary and historical evidence, British Archaeological Reports, International series, 161, Oxford, BAR Publishing. Douglas, M. (1966). Purity and danger. An analysis of the concepts of pollution and taboo, New York, Ark. Faraone, C. and Obbink, D. (1991) Magika hiera: ancient Greek magic and religion, Oxford, Oxford University Press. Farnell, L R, 1904, Sociological hypotheses concerning the position of women in ancient religion, Archiv fur religionswissenschaft 7, 70-94. Farnell, L.R. (1921) Greek hero cults and ideas of immortality, Oxford, Clarendon Press. Gager, J.G. (1992) Curse tablets and binding spells from antiquity and the ancient world, Oxford, Oxford University Press. Garland, R. (1985) The Greek way of Death, London, Duckworth. Garland, R. 1989. The well ordered corpse, Bulletin of Classical Studies, 36, 1-15. Houby-Nielsen, S. (1996) The archaeology of ideology in the Kerameikos: new interpretations of the ‘Opferrinnen’, in The role of religion in the early Greek polis, Hägg, R. ed., Stockholm, Jonsered. Kearns, E. (1989) The heroes of Attica, London, University of London, Institute of Classics. Kurtz, D. and Boardman, J. (1971) Greek burial customs, London, Thames and Hudson.

So burials exist not only in cemeteries outside the city walls, but also within the boundaries of the city’s habitation areas. There is also variation in the way the corpses are buried, even within the same burial place; there are inhumations and cremations, primary as well as secondary. No specific rule or norm seems to be followed concerning either the place or the mode of the burial. Conclusion Death provokes pollution. However, miasma as a concept seems to lack any concrete definition. It exists, but no one speaks about it officially. It is a taboo, ‘a very badly kept secret’ (Tarlow 1999: 6), and taboos are much stronger than the state prohibitions involved in laws. It appears to be highly complex, dangerous and non-rational. The only sure thing is that miasma exists and that it is not a simple notion. Vernant describes it as a ‘stain’, a ‘thing that flies’, a ‘sickness’, a ‘trouble’, a ‘wound’, and these are just some of the forms that it may take (1982: 113). It is a veil covering all social actions relating to death, beginning at the very moment of death, and ending as the period of mourning finishes; it then re-appears at every instance of commemoration of the death (Kurtz and Boardman: 1971). Everyone knows it exists; it is an everyday feature. But at the same time, no one seems to admit its existence. There is no specific definition of its content. Miasma, therefore, appears on the one hand to exist naturally, in the sense that the body of the deceased rots, forming a danger for public health. On the other hand, it exists culturally and is used and manipulated by the community in a way which appears random, lacking any particular rules or norms. 61

SOMA 2004 Metcalf, P. and Huntington, R. (1991) Celebrations of death, Cambridge, Cambridge University Press. Michel, Ch. (1900) Receuil d’inscriptions Grecques, Bruxelles, H. Lamentin. Morris, I. (1992a) Death-Ritual and Social structure in Classical antiquity, Cambridge, Cambridge University Press. Morris, I. (1992b) Law, Culture and Funerary Art, Hephaistos, 11-12, 35-51. Nielsen, T.H., Bjertrup, L., Hansen, M.H., Rubinstein, L., Vestergaard T. (1989) ‘Athenian grave monuments and social class’, in Greek, Roman and Byzantine Studies, 30, 411-420. Papachatzis, D. (1974) Pausaniou Hellados peri-eg-esis. 1, Attika / eisag-og-e sto ergo tpu Pausania kai sta Attika apokatastas-e tou archaiou keimenou, metaphras-e kai s-emei-oseis istorikes, archaiologikes, mythologikes, Athens, Ekdotike Athenon. Parker, R. (1983) Miasma, Pollution and Purification in early Greek Religion, Oxford, Clarendon Press. Robinson, D.M. (1942) Necrolynthia: a study in Greek burial customs and anthropology, Baltimore, The John Hopkins Press.

Ruschenbusch, E. (1966) Solonos nomoi, Wiesbaden: Historia Einzelschrift, Volume 9. Seaford, R. (1995) Reciprocity and Ritual: Homer and tragedy in the developing city-state, Oxford, Clarendon Press. Stears, K. (1998) Death becomes her: gender and Athenian death ritual in The sacred and the feminine in ancient Greece, 1998, New York, Routledge, 113-28. Tarlow, S. (1999) Bereavement and commemoration : an archaeology of mortality, Oxford, Blackwell. Whitley, J. (1988) ‘Early states and hero-cults: a reappraisal’, Journal of Hellenic Studies 108, 173-182. Whitley, J. (1994) ‘The monuments that stood before Marathon: tomb cult and hero cult in Archaic Attica’, in American Journal of Archaeology XCVIII, 213-230. Young, R.S. (1948) ‘Excavation in the Agora’, American Journal of Archaeology LII, 377-378. Vernant, J.P. (1982) Myth and Society in Ancient Greece, London, Methuen. Zaidman, L.B., and Pantel, P.S. (1992) Religion in the ancient Greek city, Cambridge, Cambridge University Press.

62

Monkeying Around the Mediterranean: A Fresh Perspective on Ancient Primates Cybelle Greenlaw

People often ask me, ‘Why monkeys?’ Then they say, ‘I didn’t know there were monkeys in the Mediterranean.’ My interest in monkeys began in 1984, when my mother and I decided to convert our guesthouse into a primate sanctuary for baboons and macaques. After more than a decade of building cages, cleaning cages and digging drainage ditches, I decided to leave the monkeying around to my mother and do a Bachelor’s degree in French literature with a minor in Classics. All was going well until my third year as an undergraduate when I saw the blue monkeys of Thera—I knew there would never be any way out of the monkey business.

elsewhere. Pre-existing cultural attitudes also contributed to the wide range of roles and ‘personalities’ attributed to the species and the ways in which they were represented as they were introduced to other Mediterranean peoples. Part of McDermott’s difficulty in identifying the various species depicted in Mediterranean art may, in fact, be linked to the title of his book: The Ape in Antiquity. He explains his use of the term in this way: … the word “ape” is used as a general term and as a term where the reference is to the tail-less animal; the word “monkey” has been used in general for apes with tails; more restricted words, such as “langur,” “gibbon,” “guenon” and so forth, are used only when speaking in particular places where references to apes in ancient times seems to indicate these generic names (1938: 102).

It is hardly surprising, though, that so many people are unaware of the presence and importance of monkeys in the ancient Mediterranean. No major work has been written on the subject since William Coffman McDermott’s The Ape in Antiquity, published in 1938. McDermott is still the most commonly cited authority on ancient monkeys, and his book remains an excellent source for references to non-human primates in Greek and Latin literature. It even provides a useful catalogue of simian iconography in a variety of media. However, many of McDermott’s comments on primate behaviour and attempts to identify the species are outdated and highly inaccurate. Unfortunately, the majority of more recent books and articles have failed to question his knowledge of these areas, and a few scholars have even created new and more outlandish inaccuracies. This paper aims to explain and amend some of these problems and provide a general survey of the diverse cultural attitudes toward monkeys through an examination of the iconographical, physical and textual evidence from several Mediterranean cultures, specifically those of Egypt, Mesopotamia, the Bronze Age Aegean, the Levant and the later Greco-Roman world. The survey, then, covers a time span of over 3,000 years and begins in Predynastic Egypt, where monkeys were first introduced to the Mediterranean. The surviving evidence demonstrates that the Egyptians had more first-hand knowledge of the animals than any other Mediterranean culture, and this, no doubt, contributed to the fact that their depictions of monkeys are often much more realistic than those produced

Now, of course, the word ape is used specifically to describe a small number of species from the order of primates, namely the great apes, gorillas, chimpanzees, bonobos and orang-utans and the lesser apes, gibbons and siamangs. It is highly unlikely that any of these animals were portrayed in ancient Mediterranean art, since gibbons, siamangs and orang-utans are native to Southeast Asia and the others to central and western Africa. In fact, only monkeys were known in the Mediterranean, and these can be named with some degree of certainty as hamadryas and anubis baboons, vervets, patas monkeys, and Barbary and Rhesus macaques. The first three were imported to Egypt, even from Predynastic times, from the southern lands of Nubia, modern day Ethiopia, and Punt, which may have been located in Erythrea. The only monkey that had a range extending to the Upper Nile Valley was the patas monkey. Another species that became popular around the Mediterranean is the Barbary macaque, the only African macaque, which most people recognise, inaccurately, as the ape of Gibraltar. Although these monkeys could have been introduced to the Rock in ancient times from North 63

SOMA 2004 Africa, they may not have been introduced until 711 CE. Indeed, the only certain records of their presence appear after 1740, when British troops first introduced them to the island (Fa and Lind 1996: 235). The last species is the Rhesus macaque, whose westernmost habitat is Afghanistan (Nowak 1999: 140).

Monkeys were also associated with vessels—as many as fifteen pithemorphic vessels have survived from the Old Kingdom. Most are made of alabaster, and a few bear the names of Sixth Dynasty monarchs. Like many Old Kingdom monkey amulets, the vessels are usually in the form of a female vervet clasping an infant to her chest. They were generally used as perfume containers—an appropriate use, since both myrrh and vervets were imported from Punt (Fischer 1993: 1-9).

Since most of the African species would have entered the Mediterranean through Egypt, this seems the best place to start. The appearance and significance of simian imagery in other areas of the Mediterranean can then be traced in chronological order, and the inaccuracies of modern scholars corrected along the way. In Egyptian art, monkeys are often very realistically portrayed, and specific roles are assigned to the various species. Baboons had great religious significance even from Early Dynastic times. Glazed-composition figurines found at Hierakonpolis, Abydos and Elephantine attest to the existence of early baboon cults and are believed to represent the baboon deity Hedj-Wer, or the Great White. A large, alabaster baboon in Berlin (22607), which bears the name of Narmer, may also represent this god, although some scholars have suggested that early kings may have had themselves portrayed as baboons (Friedman 1995: 25). Since Hedj-Wer merged with the lunar god, Thoth, in the Old Kingdom, it was believed that the baboon depicted in the figurines was the hamadryas, with which Thoth was inextricably linked. However, baboon skulls found in the Royal Cemetery of Hierakonpolis, which dates from the late Predynastic period, have now been identified as belonging to anubis baboons (Linseele 2003).

Now let’s briefly swing over to the Mesopotamian city of Ur. Terracotta plaques, believed to have been produced in the third millennium, depict a man in a long robe holding one or two monkeys by a leash. In some examples, one of the monkeys sits on his shoulder. The man, and sometimes one of the monkeys, is holding or playing the flute. Mendleson suggests these are travelling showmen from India with trained rhesus macaques, the species employed by monkey Wallahs in modern times (1983: 81-3). The identification of the species is probably correct, since lapis lazuli was also imported to Mesopotamia from the macaque’s westernmost habitat of Afghanistan. By the First Intermediate period or EM II-III in the Aegean, the monkey also made its way over to Crete in the form of ivory stamp seals. One of the earliest examples comes from Trapeza-Lasithi and depicts a monkey sitting in a human pose, using its long tail as a stool. Clearly, then, the Minoans were warming to the idea of Egyptian monkeys, but the image would not become popular until much later. In Middle Kingdom Egypt, monkeys continued to be associated with vessels—this time blue marble or anhydrite toilet jars, decorated with monkeys or baboons in relief (Terrace 1966: 57-63; Bourriou 1988: 142). After a brief absence during the First Intermediate period, they also reappear in tomb chapel scenes. Many of the Old Kingdom themes disappeared, but one scene from the tomb of Khnumhotep at Beni-Hasan is worthy of attention. It depicts three baboons, probably anubis because of their

By the Old Kingdom, baboons were used to represent a number of deities. Thoth, as previously mentioned, was associated with the hamadryas, and the bau of the east, who greet the sun at its rising, were similarly depicted. The animal’s negative characteristics, however, are also present in the thieving baboon god, Babi, addressed in Utterance 549 of the Pyramid Texts: Get back, Babi, red of ear and purple of hindquarters! You have taken the thigh joint of your goddess to your mouth! (Faulkner, trans. 1998) The smaller vervet and patas monkeys were not portrayed as gods but often appeared, along with baboons, in a variety of light-hearted tomb chapel scenes, sitting under the chair of the tomb owner or his wife—a theme that was repeated almost continuously from the Old to New Kingdoms. This is certainly not the only theme, though. Monkeys are often shown with their keepers or engaged in a variety of human activities, especially related to music and dance. They are often depicted wearing belts, bracelets, anklets and collars.1

FIGURE 1 DETAIL OF WALL PAINTING FROM THE TOMB OF KHNUMHOTEP AT BENI-HASAN IN LINE DRAWING (AFTER HOULIHAN 1997: 35, FIG. 2; DRAWING: CYBELLE GREENLAW)

1. Vandier d’Abbadie’s series of articles, ‘Les singes familiers dans l’Ancienne Egypte,’ which appeared in the Revue d’Egyptologie from 1964-6, remain the best source for monkeys in Egyptian daily life.

64

CYBELLE GREENLAW brown fur and lack of mane, eating figs in a tree, while two men on either side of the tree collect fruit in a basket (Fig. 1). This is the first example of a theme used extensively in the New Kingdom, the exact nature of which is unclear. Many scholars argue that the monkeys in scenes such as this are helping to pick the fruit and that the baboons were trained specifically for this purpose. The major opposition to this notion comes from Patrick Houlihan. Although he acknowledges that pig-tail macaques have long been employed as assistants in coconut harvests in parts of Southeast Asia, he makes the assertion that monkeys do not assist with the harvest of soft fruits (1997: 32-3). However, macaques are, in fact, being trained to harvest tamarinds and mangoes, which involves climbing trees the height of a three-storey house and shaking the branches. While the monkeys are in the trees, they are allowed to eat as much fruit as they like (CNN.com 2000).

a chariot pulled by another monkey is believed to be an unfavourable representation of the heretic king, Akhenaten, who regularly rode through the city with his family. Another unfinished statuette that seems to depict a pair of monkeys engaged in grooming activities may have been created to satirise scenes of the king hugging and kissing his daughters (Frankfort and Pendlebury 1933: 99). The baboon’s religious significance also grew tremendously during the New Kingdom. Thoth was commonly depicted in his lunar aspect as a blue faience baboon wearing the lunar disk, as was another lunar god, Khonsu, whose form was conflated with that of Thoth. Because of his associations with language and writing, Thoth was also considered the special patron and protector of scribes. Numerous small monuments from the New Kingdom portray scribes sitting with Thoth in his baboon manifestation to demonstrate their closeness and importance to the god.

From the Eighteenth Dynasty on, monkeys began to appear in a much wider variety of non-religious art, which was largely the result of the active colonization that took place after the departure of the Hyksos rulers. Numerous scenes of tribute bearers bringing baboons, vervets and other exotic animals to the Pharaoh attest to Egypt’s newfound control of the lands to the south. One of the most charming scenes comes from the tomb of Rekh-mi-Re, who held the office of vizier in Thebes under Thutmosis III and Amenhotep II (Houlihan 1996: 203, pl. 15). In a Nubian procession, a small vervet climbs the neck of a giraffe (Fig. 2).

FIGURE 3 FRAGMENTS IN LINE DRAWING SHOWING THE BODY OF THE SAFFRON GATHERER (ADAPTED FROM PAINTING AND DRAWING IN HOOD 1978, FIGS. 27-8; DRAWING: CYBELLE GREENLAW Obviously, the Egyptians genuinely loved non-human primates and were not speciesist, to use a modern animal rights term. Although they recognised the monkey’s ability to mimic human actions, they did not consider these animals to be poor imitations of human beings. On the contrary, baboons, in particular, were believed to possess qualities worth imitating on the human level. By the Eighteenth Dynasty, the period roughly corresponding to LM IA, this love of monkeys had spread to the Aegean islands of Crete and Thera. Six wall paintings depict blue monkeys engaged in a variety of activities, some completely simian, some human, and some perhaps divine.2 Two paintings come from Crete. One is the famous Saffron Gatherer, which Evans originally restored as a blue boy, apparently fooled by the armbands and belt (Hood 1978: 48-9; Fig. 3). The mistake

FIGURE 2 DETAIL FROM THE TOMB OF RHEKMIRE IN LINE DRAWING (AFTER VANDIER D’ABBADIE 1966: 153, FIG. 12; DRAWING: CYBELLE GREENLAW) In addition to their value as exotics, monkeys began to be enjoyed for their entertainment value once again. Many New Kingdom ostraca depict monkeys engaged in a variety of human activities, particularly in association with music and dance. Political satire may also be found in limestone toys from Amarna. A statuette depicting a monkey driving

2. The Theran frescoes are beautifully illustrated in Doumas (1992) The Wall Paintings of Thera.

65

SOMA 2004 is understandable, but as previously mentioned, such adornments were the common attributes of pet monkeys in Egyptian art. In this painting, a monkey is reaching out to pick a crocus growing in a pot. Because of the saffron connection, it can be compared with a scene from Xeste 3, room 3a. There a baboon is handing saffron to a seated female figure, probably a goddess. The offering seems to be made on behalf of the young women and girls who pick saffron in the surrounding rocky landscape.

Discovery of the First Known Victim of Thera’s Bronze Age Eruption,’ which appeared in Archaeology in 1972. An object, found in 1966 by a supporter of Galanopoulos’ belief that Thera was the lost Atlantis, was identified as the skull of a monkey and classified ‘… among the subfamily Colobinae of the Cercopithecidae family’ (1972: 230). The animal’s death was then described in graphic detail. The poor monkey was said to have been struck by flying debris and trapped in flowing lava, where it slowly cooked to death. Fortunately, the ‘skull’ is only a piece of lava, and no monkey died in its formation (Doumas 2000: 735). Indeed, even the X-rays published in Poulianos’ article do not resemble a skull.

The House of Frescoes scene from Knossos shows monkeys in a naturalistic, but not realistic, landscape. The scene contains a mixture of elements from marshland and rocky areas with the Egyptian element of papyrus plants. No humans are present in the scene, and according to Cameron’s widely accepted reconstruction, the monkeys are engaged in an egg hunt (1968: 19-22). A similar scene from Beta 6 on Thera depicts monkeys jumping and swinging through a rocky landscape. The stylistic differences, though, are striking. These monkeys look like human acrobats with long tails and vervet faces.

After the destruction of Thera, monkeys disappeared from Aegean iconography. The Mycenaeans did not adopt the image, and it would not reappear until the late Geometric period. Nevertheless, monkeys continued to be represented in Egyptian art, and several Middle and Neo-Assyrian kings mention receiving tribute in the form of exotic animals from the west, including monkeys (Dunham 1985: 236-40). Around 750-700 BCE, monkey-like, seated figurines were beginning to appear in a number of sites throughout the Mediterranean, including Elis, Laconia, Arcadia, Euboea, Rhodes and Central Italy (Langdon 1990: 407). Langdon suggests that these are a motif intended for votive offering, since all those which come from a recorded context were found in sanctuaries (1990: 413).

In Xeste 3 room 4, there is fragment of a monkey playing the lyre and another of a monkey or monkeys engaged in a sword fight or dance. Although Marinatos has suggested that because room 4 is connected to 3a by pier and door partition, the monkeys may be holding a ritual contest in honour of the goddess (1984: 113), it seems more likely that the Therans simply adopted the iconography of monkeys as entertainers from Egypt.

Further eastern influences can be found in the Phoenician silver bowls from Kourion and Praeneste, both dating to the late eight century and depicting the same story of an ‘ape’ hunt—McDermott identified the primate in question as a gorilla (1938: 235). This, of course, is highly unlikely. Nine separate episodes are present in the scene, and some scholars believe it reflects a lost Phoenician fable or epic (Markoe 1985: 67-8). This seems plausible enough, considering the long tradition of fables in Egypt and the East, and it is around this time that references to the better known Aesopic fables were beginning to appear in the writings of Greek authors. Archilochus, writing in the early to mid seventh century BCE, was the first to make reference to a monkey fable (McDermott 1938: 112; Holzberg 2002: 12). The first extant collection of fables, however, was composed by Phaedrus, a freedman of Augustus, in the early first century CE. Sometime later, Babrius composed a second collection of fables in Greek. In these fables, the monkey is often a braggart, physically ugly and foolish. From time to time, though, he displays a hint of wisdom.

The last Theran scene comes from Sector A in the North Magazines and depicts a monkey facing an architectural structure, usually identified as an altar or shrine because of the papyrus capital and horns of consecration. The monkey has his arms upraised in the manner of the Egyptian solar baboons. It would seem this is another religious association borrowed from Egypt and modified to suit Aegean needs. The element of sun worship disappears, but the monkey continues its association with this gesture of respect. Before moving on, a couple of comical errors related to Aegean monkeys are worth noting here. I have found that the monkey is never portrayed as ithyphallic in Aegean art and that the monkey’s gender is never identifiable, so I was surprised to see Nanno Marinatos’ reconstruction of a Minoan seal stone with ithyphallic baboons (1987: 125, fig. 7.1). After taking a long, hard look at a photo of the original seal, however, I concluded that there was nothing long and hard to be seen. Also, it has long been suggested that the Minoans may have imported monkeys to Crete and Thera, and many scholars, including Galanopoulos (1969: 1534), Vanschoonwinkel (1990: 36), Hood (1971: 90), Marinatos (1984: 112), and Immerwahr (1990: 42), cite the discovery of a petrified monkey skull on Thera as evidence of this. The alleged skull was first brought to light in a rather fanciful article by Aris Poulianos, ‘The

Lissarague has suggested that some terracotta representations of monkeys may have been inspired by fables, since other animals from fables, such as the fox, are often depicted with the monkey in these figurines (2000: 140-2) Some terracottas, however, would definitely not be linked to fable, particularly those demonstrating a connection between monkeys and ithyphallic satyrs. Both creatures were thought of as hyper-sexual and always 66

CYBELLE GREENLAW chasing after human women. This idea of non-human primates has persisted into our own times and is reflected in a number of films, the most famous of which is King Kong.

Holzberg, N. (2002) The Ancient Fable: An Introduction, C. Jackson-Holzberg, trans., Bloomington, Indiana University Press. Hood, S. (1978) The Arts in Prehistoric Greece, New Haven, Yale University Press. Hood, S. (1971) The Minoans: Crete in the Bronze Age, London, Thames and Hudson. Houlihan, P. (1996) The Animal World of the Pharaohs, Cairo, American University Press. Houlihan, P. (1997) Harvesters or Monkey Business? Göttinger Miszellen, 157, 31-47. Immerwahr, S. (1990) Aegean Painting in the Bronze Age, Philadelphia, Pennsylvania State University Press. Langdon, S. (1990) From Monkey to Man: The Evolution of a Geometric Sculptural Type. American Journal of Archaeology, 94, 407-24. Linseele, V. (2003) Weird Animals from the Elite Cemetery. Archaeology’s Interactive Dig. Available from: http:// www.archaeology.org/interactive/hierakonpolis/ animals.html> [Accessed 6 October 2003] Lissarague, F. (2000) Aesop, Between Man and Beast: Ancient Portraits and Illustrations. IN: Cohen, B. ed., Not the Classical Ideal: Athens and the Construction of the Other in Greek Art, Leiden, Brill, 132-49. McDermott, W.C. (1938) The Ape in Antiquity, Baltimore, Johns Hopkins Press. Marinatos, N. (1984) Art and Religion in Thera: Reconstructing a Bronze Age Society, Athens, Mathioulakis. Marinatos, N. (1987) An Offering of Saffron to the Minoan Goddess of Nature. Boreas, 15, 123-32. Markoe, G. (1985) Phoenician Bronze and Silver Bowls from Cyprus and the Mediterranean, Classical Studies, 26, Berkeley, University of California Press. Mendleson, C. (1983) More Monkey Business. Anatolian Studies, 33, 81-3. Nowak, R.M. (1999) Walker’s Primates of the World, Baltimore, Johns Hopkins University Press. Poulianos, A. (1972) The Discovery of the First Known Victim of Thera’s Bronze Age Eruption. Archaeology, 25, 229-30. Terrace, E.L.B. (1966) “Blue Marble” Plastic Vessels and Other Figures. Journal of the American Research Center in Egypt, 5, 57-63. Vandier d’Abbadie, J. (1964) Les singes familiers dans l’Ancienne Egypte: l’Ancien Empire. Revue d’Egyptologie, 16, 147-77. Vandier d’Abbadie, J. (1965) Les singes familiers dans l’Ancienne Egypte: le Moyen Empire. Revue d’Egyptologie, 17, 177-88. Vandier d’Abbadie, J. (1966) Les singes familiers dans l’Ancienne Egypte: le Nouvel Empire. Revue d’Egyptologie, 18, 143-201. Vanschoonwinkel, J. (1990) Animal Representations in Theran and Other Aegean Arts. IN: Hardy, D.A. et al. eds., Thera and the Aegean World III, vol. 1, London, Thera Foundation.

It seems, then, that the monkey’s journey through the Mediterranean must end on a sad note—these noble animals that left Egypt as manifestations of the divine and symbols of wealth circled the Mediterranean only to become objects of ridicule. Clearly, the Greeks and Romans did not share the earlier Egyptian and Aegean view of monkeys. These were most definitely speciesist cultures! Hopefully, though, this paper has helped to reintroduce the importance of an animal whose presence in the ancient Mediterranean has often been ignored or misunderstood and provided a further response to the question I am most frequently asked. Why monkeys? Because the human reaction to this exotic, almost-human creature reveals so much about how people viewed themselves and their place in the world. Bibliography Bourriou, J. (1988) Pharaohs and Mortals: Egyptian Art in the Middle Kingdom, Cambridge, Cambridge University Press. Cameron, M.A.S. (1968) Unpublished Paintings from the ‘House of Frescoes’ at Knossos. British School at Athens, 63, 1-30. CNN.com (2000) There’s no business (like monkey business). Available from: [Accessed 26 November 2002] Doumas, Ch. (2000) Discussion. IN: Sherratt, S. ed., The Wall Paintings of Thera: Proceedings from the First International Symposium, Athens, Thera Foundation, 735. Doumas, Ch. (1992) The Wall Paintings of Thera, Athens, Thera Foundation. Dunham, S. (1985) The Monkey in the Middle. Zeitschrift für Assyriologie und Vorderasiatische Archäologie, 75: 2, 234-64. Fa, J.E. and R. Lind (1996) Population management and the viability of the Gibraltar Barbary macaques. IN: Fa, J.E and D.G. Lindburg eds., The Ecology and Evolution of Macaque Societies, Cambridge, Cambridge University Press, 235-62. Faulkner, R.O. trans. (1998) The Ancient Egyptian Pyramid Texts, Special ed. Oxford, Oxford University Press. Fischer, H.G. (1993) Another Pithemorphic Vessel of the Sixth Dynasty. Journal of the American Research Center in Egypt, 30, 1-9. Frankfort, H. and J.D.S. Pendlebury (1933) The City of Akhenaten, 2, Oxford, Oxford University Press. Friedman, F.D. (1995) The Underground Relief Panels of King Djoser. Journal of the American Research Center in Egypt, 32, 1-42. Galanopoulos, A.G. and E. Bacon (1969) Atlantis: The Truth Behind the Legend, London, Nelson. 67

Special Sites in the Cretan Neolithic Landscape Heinrich Hall

This paper attempts to do two things: firstly, to make some suggestions regarding site classification in Neolithic Crete, especially in regard to hilltop sites and caves. Secondly, to propose a starting point for a Cretan application of interpretative landscape archaeology, i.e. a landscape archaeology that goes beyond environmental archaeology in its economic and ecological aspects, so as to include conceptual issues.

to over 200 for the c. 500 years in question. There can be no doubt that this indicates a real and rapid settling not only of all parts of the island, but also of virtually all its types of terrain or landscape, from coastal and lowland to upland and mountain. It may still be too early to develop any detailed models of the processes that led to such a development (for a first published approach, Nowicki 2003: 65-72), but it is clear that we are looking at a period of immense changes, not merely in quantity, but also in the way people lived and exploited their environment. Such changes must have been accompanied by shifts in social and political organisation, in contacts and exchange networks. Furthermore, modern anthropology and archaeology agree that humans do not settle landscapes merely in terms of subsistence and economy, but also appropriate them conceptually. They associate their environment with the way they see themselves and the world; they give the landscape ideological, cosmological and mythological meaning or meanings (Bender 1993; Green 1995; Ucko and Layton 1999). These meanings, too, would play a role, and undergo changes in a phase of settlement expansion.

Early and Middle Neolithic Crete It is generally accepted that Crete was settled by farming communities at some point in the seventh or sixth millennium BC, when a settlement was established on Kefala hill, underneath what is now the Minoan palace of Knossos. Unfortunately, our knowledge of the first twoto three thousand years (Early and Middle Neolithic) of human activity on the island remains very limited, since the only site to undergo substantial excavation and publication is Knossos itself. That site certainly presents a picture of remarkable uniformity and considerable continuity, and perhaps also of emerging complexity, throughout the Early and Middle Neolithic and beyond (Evans 1971; 1990; Broodbank 1992). However, we are unable so far to place it in a wider Cretan context, since literally only a handful of other sites are known from those millennia (see gazetteer in Strasser 1992: 193-282). It has been argued elsewhere that this remarkable scarcity of settlement is at least partially due to taphonomic factors (Hall 1999: 62-4; Tomkins and Day 2001: 260).

Lack of Categories In such a situation, we can expect considerable variety in activities and their locations, not only between different parts of the island, but also within regions. A representation of sites as simple dots on a distribution map (as Papathanassopoulos 1996: 207-8; Strasser 1992) is not a sufficient tool to represent such variety. There is a call for some differentiation between sites, i.e. for classification or categorisation, as a first step to understanding the fullness of activities at the time. It may be fashionable to dismiss categorisation as the creation of mental templates that prevent a full understanding of prehistoric activity in all its ambiguities and varieties, but it can also be argued that categories are a necessary intellectual stepping stone on the way to such understanding, like a ladder that can be safely discarded only after it has served to climb to a different plane.

Late and Final Neolithic In contrast to that, there can be no doubt that the few centuries in the fourth millennium known as Late Neolithic (LN) and Final Neolithic (FN), represent a period of profound changes throughout Crete (Vagnetti & Belli 1978; Tzedakis 1984; Vagnetti 1996; Nowicki 2003). Although Cretan material culture retains many earlier traits during this time, there are a number of striking innovations, including new pottery shapes and evidence for increasing contacts. The most outstanding development, however, concerns the settlement pattern. The total number of recognised sites increases from under ten for the previous three millennia 69

SOMA 2004 Although, so far, few FN and LN sites in the island have been excavated, and even fewer have been published, some basic differentiations have already been made, explicitly or implicitly (Table 1). It should be noted that these differentiations, even where they touch upon site function, are mostly morphological or topographic in character, and that most interpretations rely on individual sites alone rather than on their place within a regional framework.

category as a small, dark one like Trapeza (Pendlebury, Pendlebury, et al. 1935-36). Of course, we should keep in mind the likelihood that Cretan life at the time may have been subject to rapid changes and that at certain times and in certain places societies may have adopted temporary, perhaps even semi-nomadic settlement strategies (Watrous 1994: 700-1; for a differing view Moody 1987: 290). In fact, that possibility calls for a deeper engagement with the known sites rather than for generalisation. In any case, where multiple sites indicate clusters of activities of variable intensity in a variety of locations, we should not be satisfied to regard them as an amorphous blur.

The currently prevailing undifferentiated classification of virtually all open-air sites as settlements would, for example, place Nerokourou, a lowland site in the Chania Plain (Vagnetti 1989), in the same category as Azoria in the Vrokastro uplands (Haggis 2003) or any of the sites around the Lasithi Plateau (Watrous 1982: 9-11). Similarly, the lack of differentiation among caves without clear burial evidence confines a wide, open and well-lit cave like Pelekita (Davaras 1982: 387-8) to the same

MORPHOLOGICAL/ TOPOGRAPHIC DIFFERENTIATIONS based on location Caves

Caves eg, Pendlebury 1939: 44

Classification so Far: General Differentiations It is not the object of this paper to produce a new detailed classification of sites. Such an undertaking will have to be based on the in-depth study of excavated material from as many places as possible, and of intra- and inter-regional

FUNCTIONAL DIFFERENTIATIONS usually based on location, occasionally on finds Burial/Ritual usually based on finds eg, Dickinson 1994: 42; Hall 1999: 97-144 Other eg, Watrous 1982: 9-10 (Refuge) eg, Faure 1964: 30

Settlement (Rock Shelters) see Hall 1999: 111

(Temporary Shelter) eg, Watrous 1994: 700-1

Generic Settlement & ancillary functions [applies to over 70% of site total] eg, Papathanassopoulos 1996: 207-8

Open-Air Sites (Any Location)

Hilltop/Upland Sites (eg, Nowicki 1995: 65-7) Refuge/Defensive [not universally accepted] eg, Nowicki 2003

Ritual based on finds and location [not universally accepted] eg, Peatfield 1992: 71; Morris and Batten 2000 TABLE 1 ‘TRADITIONAL’ CLASSIFICATION OF LN/FN SITES IN CRETE (TYPES IN BRACKETS HAVE LARGELY BEEN DISTINGUISHED IMPLICITLY, I.E. THEIR DISTINCT FUNCTIONS AS OUTLINED HERE HAVE BEEN SUGGESTED IN THE LITERATURE HITHERTO PUBLISHED, BUT NOT IN THE CONTEXT OF A DELIBERATE OR CONSCIOUS ATTEMPT AT SITE CLASSIFICATION). (AUTHOR’S WORK) 70

HEINRICH HALL crags, ridges etc., but most importantly their relative inaccessibility. His interpretation of them as defensive sites, linked with the arrival and establishment of intrusive groups in Crete (Nowicki 2003: 65-9) is supported by other authors on the basis of individual sites or groups (Watrous 1994: 704; Votokopoulos 2000; Schlager 2002) and by Nowicki’s own excavation of the gorge-side site at Katalimata, arguably the type site for the group (Nowicki 2003: 16-20).

relationships between them. Still, a number of suggestions regarding site distinctions can already be made. As regards topography, there is an obvious distinction between coastal sites, like Pelekita, and inland ones, like those in the Lasithi (see above). It is also reasonable to distinguish between sites at different elevations, from coastal plains, which provide a distinctive and relatively rare form of terrain in Crete (eg. Nerokourou), via hills and slopes (eg. Phaistos, Vagnetti 1972-73), to mountain plateaux and mountains (eg. Atsipadhes Korakias, see below). To some extent, these differentiations could represent different subsistence regimes – possibly at different times – and varied stages in the gradual fillingin of Crete with human settlement. As regards caves, the range between the extreme differences outlined previously is likely to represent distinct roles of individual sites within wider networks.

Some guarded criticisms of that view can be made. Generally, the question of the existence and character of warfare in so early a period is contentious (for conflicting views, Vencl 1999; Peter-Röcher 2002). The defensive interpretation would account for all or most open-air sites known from the period in some areas, eg, in the far east of the island, but not in others, like the Chania/Akrotiri area, where FN/EM is known from open sites in a variety of locations, as well as from Nerokourou (Moody 1987). This would imply that the population of eastern Crete was focused exclusively on defensive matters when choosing settlement sites, while other parts of the island saw either a co-existence or succession of different issues influencing choices or even the contemporary existence of both. In some individual cases, the close proximity of several such sites in areas of relatively poor economic potential could also be seen as surprising. An example of this is the small upland basin of Lamnoni (east Crete), with two such sites immediately to its west and east (the western site being both larger and less accessible then the eastern one), only 100m apart from each other (Branigan 2000: fig. 38; Nowicki 2003: 25-6). It is difficult to see why the inhabitants of the area, or new arrivals, should have chosen to settle both sites, but even more so how two competing groups could have coexisted here.

Caves: Specific Functions The lack of published excavated material still poses a major obstacle to a more informed, less general approach to site function. Distinctions between different economic activities can only be suggested in individual cases. For example, among the caves, Yerani provides evidence for two specific features, seasonal occupation and stone tool production, which do not apply to all sites (Godart & Tzedakis 1992: 76-9). Within settlement patterns, there may also be a distinction between central and peripheral sites. The presence of burial and some evidence for collective symbolic/ritual activity in caves is currently gaining acceptance (Strasser 1992: 61, 74; Dickinson 1994: 42; Hall 1999: 97-144), as is the idea that some caves were involved in the development of long-distance exchange of exotic materials and metal (Hall 1999: 206-8; outside Crete: Nakou 1995). Additionally, the location and accessibility, both internal and external, of caves, as well as the location of Neolithic material within them, may shed further light on their different roles.

The final, and perhaps most pertinent objection, to the defensive explanation is the fact that the high ratio of such sites against contemporary material from plains or valleys may at least partially be due to taphonomic reasons. The geomorphology and climate of Crete appear to lead to occasional catastrophic erosional events in the shape of flash floods and landslides in many areas and to a general pattern of sediments being washed from hillsides, potentially leading to the covering of sites in valley bottoms and plains and the disappearance of those located on slopes, but favouring the survival of hill- or ridge-top (and cave) material (Rackham and Moody 1996: 18-24; Brown 1997: 237-48; contra Nowicki 2003: 71). The fact that the important FN site of Nerokourou, located in the coastal plain east of Chania, apparently had no visible FN material on the surface and was discovered only by coincidence in the context of the exploration of a Minoan Villa (Vagnetti 1989), should serve as a reminder of the fact that the visible settlement distribution may be extremely biased against settlements of its type.

Open-Air Sites: Generic Settlements? The functions of open-air sites remain more contentious. Apart from Knossos and Phaistos, still the only known long-term settlements with a considerable degree of organisation, and almost certainly regional centres, other sites have generally been described simply as settlements until recently. However, in the 1990s the discovery of numerous LN and FN hilltop sites gave rise to two distinct, but conflicting, functional interpretations. Hilltop Sites: A Defensive Function? Nowicki, the discoverer of many such sites, presents and interprets them in a series of articles (Nowicki 1995: 65-7, 1999, 2000, 2003). He identifies them as a new site type, united by several common features, including their (mostly FN) dates, certain pottery fabrics, remoteness from the main tracts of cultivable land, a preference for hilltops,

In spite of these points, the defensive interpretation seems impossible to dismiss for some or most of the sites in 71

SOMA 2004 question. The possible existence of defensive architecture of FN date at some sites additionally supports it (Schlager 2002: 192-202).

the deposition of pottery should be considered for this site (Morris and Peatfield 1995: 644). Parallels with similar activity on exposed sites have been suggested for Traostalos (Morris and Batten 2000: n.12) and Lamnoni, Site 1 (Branigan 2000: 57-8, 81). Nowicki (2001: 32-5; 2003: 47) does not accept the ritual interpretation of these sites, preferring to place them within his framework of defensive settlements.

Hilltop Open-Air Sites: A Ritual Function? The second function suggested for such places is mainly based on the excavation of Atsipadhes Korakias in West Crete. During the excavation of a peak sanctuary on this prominent exposed hilltop site, a considerable amount of FN ceramics was found, mostly disturbed and mixed with Bronze Age material (Peatfield 1992; Morris and Peatfield 1995; Morris and Batten 2000). The increasing frequency of these finds towards the bottom of the peak sanctuary deposit clearly indicates considerable activity in that period. Remarkably, the excavators noted a complete absence of two features present at most excavated Neolithic settlement sites: dark organic soils and stone tools. Nowicki’s differing view of the material (Nowicki 2001: 34) is not accepted by the excavators (Peatfield pers. comm.).

Thus, exposed upland sites, the only group of open-air sites subject to specific functional interpretation, are disputed between a profane (defence) and a ritual (ceremonial deposition) approach. It should be noted that the first view is based mainly on topography and tends to embrace all sites in such locations, while the second one is based on a specific assemblage and has been applied to a maximum of three sites so far. Can this contradiction be overcome? Digression: A View From Western Europe As stated above, human societies perceive their landscape in a symbolic way, as a three-dimensional image of their cosmology. This aspect of the humanised landscape is likely to play a particular role during phases of major changes in relationships with the landscape. (eg. Cooney 2000: 41-5). In western Europe, there is good evidence for this side of human impact on the environment, in the form of monuments, especially tomb structures and enclosures, with multiple and changing metaphorical meanings

Such an absence on a fully excavated site is exceptional (Hall 1999: 102-4). In the Akrotiri, sites that contained FN or Early Minoan pottery featured stone tools visible in the surface collection in all cases (Moody 1987). Thus, the unusual character of the site, as well as the unusual assemblage, have led the excavators to suggest that some form of ritual or ceremonial activity involving

FIGURE 1 THE SITE OF ATSIPADHES KORAKIAS AND THE AYIOS VASILIOS VALLEY SEEN FROM THE SOUTH (PHOTOGRAPH: HEINRICH HALL) 72

HEINRICH HALL (eg. Bradley 1998; Thomas 1999: 34-61; Cooney 2000: 127-73). While these monuments have been studied and interpreted for a long time, some recent developments are of interest in the Cretan context. It appears, increasingly, that the distinction between artificial man-made structures and natural formations, albeit relatively obvious within a modern, western ‘scientific’ framework, should not simply be assumed to have existed in prehistory. The location, structural and material detail, even the very form of monuments can refer to or imitate natural features, which are likely to have played a similar symbolic role to the monuments themselves. Although we can never access the finer detail of their meanings, it is a fact that similar meanings, mythical or ancestral stories, magical properties etc., could be associated with monuments as well as with natural locations (Bradley 1998, 2000: 97-113).

as a Minoan peak sanctuary). In other words, it is the most monumental summit. This visual prominence, undoubtedly deeply connected with characteristics that also improve a site’s defensive capacity (like steepness, long-distance visibility, etc.), is nevertheless a striking feature of many of the defensive sites. The best-known among them, the FN refuge site of Katalimata (Fig. 2), is placed in one of the most breathtaking and evocative locations in its area (conferring a special status to its excavation in the minds of many archaeologists, including the present author). In fact, Nowicki’s gazetteer of defensive sites abounds with descriptions like ‘impressive ridge’, ‘highest mountain’, ‘most dominant’, ‘characteristic’, and ‘distinctive’ (Nowicki 2003: 20-52). The monumental location of these and other sites does not present a contradiction to their possible defensive function. The distinction between profane and sacred, or – to put it more neutrally – symbolic, functions has been criticised frequently. For example, Bradley (2000: 81-99) has shown that the sites of Neolithic axe production in Britain are often chosen for reasons of visual distinctiveness and difficulty of access rather than efficiency. More specifically, in parts of western Europe, at roughly the same time as the Cretan FN (during the fourth millennium BC), there is increasing evidence for an overlap, an ambiguity between defensive and ceremonial sites. In particular, sites that served as regional or local centres may have possessed defensive features or functions as well as symbolic ones (eg, Petrasch 1990; Geschwinde and Raetzel-Fabian 1998; Vencl 1999: 70).

Natural Monuments: A New Aspect to FN Sites in Crete How does this seemingly remote perspective affect our potential for understanding LN-FN Crete? It can be suggested that some of the island’s LN and FN sites are connected by an important characteristic that applies independently of their individual functions. This feature is what I would like to describe as natural monumentality. The following paragraphs try to explain that concept, first in reference to the hilltop sites discussed above, and then to some of the caves. For example, the site of Atsipadhes Korakias is located on only one of several marginal summits that overlook the Ayios Vasilios Valley (Fig. 1). In economic or practical (defensive) terms, it is hard to see major differences between the summits that would necessitate Atsipadhes Korakias being chosen over the others. An additional feature of the site, however, is that it is certainly the most visually prominent among those summits as seen from the valley (which is, of course, the reason for its undisputed use

Thus, whatever the exact character of the suggested FN defensive sites in Crete may be, many of them are located in places that share the features of being visually distinctive, widely visible and locally dominant. If we bear in mind that there must have been an active choice involved in the establishment of such sites, that inaccessible places are

FIGURE 2 THE ENTRANCE TO CHA GORGE FROM THE EAST. KATALIMATA IS LOCATED ON THE STEEP ROCK FACE JUST TO THE LEFT OF THE GORGE PROPER (PHOTOGRAPH:

HEINRICH HALL)

73

SOMA 2004

FIGURE 3 THE ENTRANCE TO THE IDAEAN CAVE (PHOTOGRAPH: HEINRICH HALL) common in Crete (and that less visible locations would arguably be preferable if the object in settlement there was protection from attackers), then the monumental character of these sites may well have played a role in such choices. Perhaps such sites were favoured because their visual importance and the meanings evoked by or ascribed to them would be associated with the groups or communities settling them.

compares well with, for example, Katalimata, where the settlement itself, albeit not widely visible, would always have been associated with the locally dominant gorge it clings to. The activities that took place in Stravomyti and the groups that undertook them may have aimed to be associated with Youktas itself. The last site to be pointed out here is the mother of all Cretan caves, the Idaean Cave itself (Fig.3), well known as a cult site in the Bronze Age and afterwards, but which also contained considerable LN deposits (Sakellarakis 1986, 1989; Faure 1996: 92-8, Hall 1999: 120-2). Situated on the upper northern slopes of Crete’s central mountain, it is the highest Neolithic site known on the island. It contained large amounts of Neolithic pottery and tools, including much imported Melian obsidian, which appears to have been worked, and perhaps also stored, in this cave which was habitable only for a few months each summer and as far from the sea as possible in Crete (Sakellarakis 1983: 417-23; Catling 1986-87: 57-8). It is difficult to explain the choice of such a site for apparently long-term use (albeit seasonal) on purely economic terms, and it must have been connected with more continuous activity in the lowlands of Central Crete. Idheon Antron is also remarkable not only for the truly monumental size of its mouth (a feature shared with some other caves containing prominent LN-FN material, like Pelekita near Zakros),

A related case can certainly, perhaps even more clearly, be made for some of the LN-FN caves (see Hall 1999: 2056). For example, the cave of Stravomyti, located a few kilometres north of Knossos, contained a certain amount of LN/FN activity, including a child burial (Marinatos 1950). The site is not a rock shelter, but a proper cave, with a labyrinth of tortuous and dark passages. For this reason, several authors deem it unlikely to have been occupied (Faure 1964: 72; Zois 1973: 176-7; Tyree 1975: 34-6) and it has been considered in a context of ritual and burial activities in LN/FN (Hall 1999: 126-8). While the cave itself may be relatively unimpressive, a further aspect to it is its location in the sheer cliffs that dominate the western face of Mount Youktas, a central feature of the Pedhiadha region south of Knossos, and a site of considerable importance in Bronze Age ritual. Stravomyti itself may not be widely visible, but it is part of one of the most prominent visual features of Central Crete. This 74

HEINRICH HALL but for its location within the central mountain of the island, which makes it indirectly visible throughout much of the island. Those who chose to inhabit, use, or at least frequent it, did so in a spot that could be roughly pointed out from half of Crete.

Evans, J. (1971) Neolithic Knossos; the Growth of a Settlement. Proceedings of the Prehistoric Society, 37, 95-117. Evans, J. (1990) The Early Millennia: Continuity and Change in a Farming Settlement. IN: Evely D., Hughes-Brook H. and N. Momigliano eds., Knossos, a Labyrinth of History, Papers presented in Honour of Sinclair Hood, Athens: British School at Athens, 1-20. Faure P. (1964) Fonctions des Cavernes Crétoises. Paris: de Bocard. Faure, P. (1996) Ιέρα Σπήλεια της Κρήτης. Iraklio: Βικελία Βιβλιοθήκη. Geschwinde, M. and D. Raetzel-Fabian (1998) MonumentalArchitektur aus Holz und Erde; Archäologische Erdwerksforschung in Südniedersachsen. Archäologie in Niedersachsen, 1, Oldenburg: Idensee, 34-7. Godart, L. and Y. Tzedakis (1992) Témoignages Archéologiques et Épigraphiques en Crète Occidentale du Néolithique au Minoen récent IIIB. Incunabula Graeca, 93, Roma: Gruppo Editoriale Internazionale. Green, N. (1995) Looking at the Landscape: Class Formation and the Individual. IN: Hirsch, E. and M. O’Hanlon eds., The Anthropology of Landscape: Perspectives on Place and Space, Oxford University Press, 31-42. Haggis D. (2003) Azoria Project, 2003 Season [Internet]. Azoria Project. Available from: < http://www.unc. edu/~dchaggis/2003_report.html > [Accessed April 2nd 2004]. Hall H. (1999) Ritual in Neolithic Crete. Unpublished M.Litt thesis. University College Dublin. Marinatos S. (1950) Το Μέγαρον Βαθυπέτρου. Πράκτικα τής Αρχαιολόγικης Εταιρέιας, 242-57. Moody, J. (1987) The Environmental and Cultural Prehistory of the Khania region of West Crete: Neolithic through Late Minoan III. Unpublished PhD thesis, University of Minnesota. Nakou, G. (1995) The Cutting Edge: A New Look at Early Aegean Metallurgy. Journal of Mediterranean Archaeology, 8:2, 1-32. Morris, C. and V. Batten (2000) Final Neolithic Pottery from the Atsipadhes Peak Sanctuary. IN: Πεπραγμένα Η’ Διεθνούς Κρητολογικού Συνεδρίου, A2, Iraklio: Εταιρία Κρητικών Ιστορικών Μελετών, 372-82. Morris C. and A. Peatfield (1995) Pottery from the Peak Sanctuary of Atsipadhes Korakias, Ay. Vasiliou, Rethymnon. IN: Πεπράγμενα τόυ διέθνους κρητολόγικου συνέδριου, A2, Rethymnon: Νέα Χριστιανική Κρήτη, 643-7. Nowicki, K. (1995) Report on Investigations in Greece, 9, Studies in 1993 and 1994. Archaeologia (Poland), 46, 63-70. Nowicki, K. (1999) Final Neolithic Refugees or Early Bronze Age Newcomers? The Problem of Defensible Sites in Crete in the Late Fourth Millennium BC IN: Betancourt, P., Karageorghis, V., Laffineur, R. and W.-D. Niemeier eds., Meletemata; Studies in Aegean Archaeology presented to Malcolm Wiener, Université de Liège, 575-82.

Conclusions It is sometimes claimed that one of the aims of modern archaeology should be to overcome the traditional western dichotomy between form and function. If this applies to art, to artefacts, and to settlement organisation, it should also be relevant to the choice of locations for activities within a landscape. Study of two prominent groups among Crete’s LN and FN sites, defensive settlements and caves suggests that there is a consistent, if irregular, pattern of deliberate choice of naturally monumental, locally or supra-locally visible or locatable sites for varied activities in LN-FN Crete. It is further suggested that aspects of this choice are symbolic, linking the occupants with meanings associated with such locations at the time. The exact content of such symbolism cannot be determined but may include notions of power, dominance, permanence and liminality. It should be stressed, once more, that what is proposed here is not in itself a new site type and should not necessarily be seen to contradict or support interpretations of individual site functions. Rather, the suggestion of monumental natural locations is meant to add a new, additional, layer to our understanding of what happened in LN-FN Crete. Bibliography Bender, B. ed. (1993) Landscape: Politics and Perspectives. Providence & Oxford: Berg. Bradley, R. (1998) The Significance of Monuments. London: Routledge. Bradley, R. (2000) An Archaeology of Natural Places. London: Routledge. Branigan, K. (2000) Prehistoric and Early Historic Settlement in the Ziros Region, Eastern Crete. Annual of the British School at Athens, 93, 32-90. Broodbank, C. (1992) A labyrinthine conundrum – social change in EN II and MN? Journal of Mediterranean Archaeology, 5/1, 39-75. Brown A. (1997) Alluvial geoarchaeology – Floodplain archaeology and environmental change. Cambridge University Press. Catling H. (1986-87) Archaeological Reports from Greece. Archaeological Reports, 33. Cooney, G. (2000) Landscapes of Neolithic Ireland. London: Routledge. Davaras, C. (1982) Ανασκάφικες Εργασίες. Αρχαιολόγικον Δελτίον Χρόνικα, 37 B2, 387-8. Dickinson, O. (1994) The Aegean Bronze Age. Cambridge University Press. 75

SOMA 2004 Nowicki, K. (2000) Defensible Sites in Crete c. 1200-800 BC (LM IIIB/IIIC through Early Geometric). Aegaeum, 21, Université de Liège. Nowicki, K. (2001) Minoan Peak Sanctuaries: Reassessing their Origins. IN: Laffineur, R. and R. Hägg eds., Potnia – Deities and Religion in the Aegean Bronze Age, Aegaeum, 22, 31-7. Nowicki, K. (2003) The End of the Neolithic in Crete. Aegean Archaeology, 6, 7-72. Papathanassopoulos, G. (1996) Neolithic Culture in Greece. Athens: Goulandris. Peatfield, A. (1992) Rural Ritual in Bronze Age Crete: The Peak Sanctuary at Atsipadhes. Cambridge Archaeological Journal, 2 (1), 59-87. Pendlebury, J. (1939) The Archaeology of Crete, an Introduction. London: Methuen. Pendlebury, J., Pendlebury, H. and M. Money-Coutts (1935-36): Excavations in The plain of Lasithi, 1, the cave of Trapeza. Annual of the British School at Athens, 36, 13-133. Peter-Röcher, H. (2002) Krieg und Gewalt: Zu den Kopfespositionen in der Großen Ofnet und der Diskussion um kriegerische Konflikte in prähistorischer Zeit. Prähistorische Zeitschrift, 77, 1-28. Petrasch, J. (1990) Überlegungen zur Funktion neolithischer Erdwerke anhand mittelneolithischer Grabenanlagen aus Südostbayern. Jahresschrift für Mitteldeutsche Vorgeschichte, 73, 369-87. Rackham, O. and J. Moody (1996) The Making of the Cretan Landscape. Manchester & New York: Manchester University Press. Sakellarakis, I. (1983) Ανασκαφή Ιδαίου Άντρου. Praktika, 415-500. Sakellarakis, I. (1986) Η Νέα Ευρήματα στο Ιδάιο Άντρο (1982-4). Αρχαιολογία, 15, 4-21. Sakellarakis, I. (1989) Hundert Jahre Erforschung der Ida-Höhle. Hellenika, Jahrbuch für die Freunde Griechenlands, 25, Bochum: Neugriechische Studien, 83-97. Schlager, N. (2002) Pleistozäne, neolithische, bronzezeitliche und rezente Befunde und Ruinen im fernen Osten Kretas. Dokumentation 2000. Jahreshefte des Österreichischen Archäologischen Instituts in Wien, 70, 157-220. Strasser, T. (1992) Neolithic Settlement and Land-use on Crete. Unpublished PhD. thesis, Indiana University.

Thomas, J. (1999) Understanding the Neolithic. London: Routledge. Tomkins, P. and P. Day (2001) Production and exchange of the earliest ceramic vessels in the Aegean: a view from Early Neolithic Knossos, Crete. Antiquity, 75, 259-60. Tyree, L. (1975) Cretan Sacred Caves: Archaeological Evidence. Unpublished PhD thesis, Columbia, University of Missouri. Tzedakis, I. (1984) Le Passage au Minoen Ancien en Crète Occidentale, IN: Aux Origines de l’Hellenisme, la Crète et la Grèce, Hommage à Henri van Effenterre, Paris: Centre G. Glotz, 3-7. Ucko, P. and R. Layton eds. (1999) The Archaeology and Anthropology of Landscape – Shaping your Landscape. London and New York: Routledge. Vagnetti, L. (1972-3) L’insediamento neolitico di Festòs. AsAtene, 50-51, 7-138. Vagnetti L. (1989) Saggi negli strati neolitici. IN: Vagnetti, L., Tzedakis, I. and A.Christopoulou, Scavi a Nerokourou, Kydonias, 1, Ricerche greco-italiane in Creta occidentale 1, Incunabula Graeca 91, Rome: Ateneo, 11-97. Vagnetti, L. (1996) The Final Neolithic: Crete enters the wider world. IN: Bakker, W. and R. Willets eds., Cretan Studies, 5, Amsterdam: Hakkert, 29-39. Vagnetti, L. and P. Belli (1978) Characters and Problems of the Final Neolithic in Crete. Studi Micenei ed EgeoAnatolici, 19, 125-63. Vencl, S. (1999) Stone Age Warfare. IN: Carman, J. and A. Harding eds., Ancient Warfare – Archaeological Perspectives, Thrupp: Sutton, 57-72. Votokopoulos, L. (2000) Οχυρές πρωτομινοικές θέσεις στην περιοχή Ζάκρου. IN: Πεπραγμένα Η’ Διεθνούς Κρητολογικού Συνεδρίου, A1, Iraklio: Εταιρία Κρητικών Ιστορικών Μελετών, 129-46. Watrous, L. (1982) Lasithi, a history of settlement on a highland plain in Crete. Hesperia supplement 18. Watrous, L. (1994) Review of Aegean Prehistory III: Crete from Earliest Prehistory through the Protopalatial Period. American Journal of Archaeology, 98, 695753. Zois, A. (1973) Κρήτη – Εποχή τού Λίθου. Ancient Greek Cities, 18, Athens: Athens Technological Organisation, Athens Center of Ekistics.

76

From Antiquity To Feudalism At The Southern Border Of Navarra. The Design of a Study in Landscape Archaeology1 Jesús Lorenzo Jiménez2

The present paper is part of a landscape archaeological project in a specific area of Navarra, near the Ebro river in the North of Spain. Basically, I shall present an outline of the work done so far, namely the compilation of the written and archaeological documentation preceding the present field work, as well as the results of the very first field campaign, which already reflect some of the general trends in the history of the area’s population. However, it should be stressed that, as of now, there are more questions than answers. Currently, we can identify and describe only a few facts and isolated structures in the whole area, but we expect the surface survey to provide us with knowledge of how human societies used the different landscapes throughout history.

specific area in the south of Navarre. Why should we study this area? Firstly, because there was a considerable Roman presence in the region (Pérez Laborda 1984; Tudanca Casero 1997); and secondly because the Kingdom of Navarre was one of the first peninsular states to come into existence after the Visigoth Kingdom fell to the Arab and Berber contingents in AD 711 (Manzano Moreno 1991; Larrea 1998). Our case study is an area situated by the confluence of the Arga and Aragon rivers which form an alluvium extending over nearly eight by twelve kilometres. Today, it consists of irrigated agricultural land and holds three villages: Marcilla, Falces and Peralta. The study area also includes sites situated on the rocky terrain that limits the plain to the west because we think that there is a link between the use of spaces in the plain and that of ones in the outlying hills. In total, the study area covers about 100 km² of irrigated and unirrigated lands, as well as barren areas at altitudes of up to eighty metres.

The Project We organised the project in the following way: • First, we compiled all available written documentation about the period in question. • Second, we did the same with all the published information about any archaeological explorations in the area and the places classified according to the Navarre Archaeological Inventory. • Third, we began a fieldwork project in the study area. To do so, we compiled the appropriate maps, both in paper and in digital format. To start with, we registered all the elements visible to the naked eye, which are mostly buildings. We continued, more recently, with surface survey in the form of field-walking. The first campaigns took place in 2003, and work continued in 2004. At the same time, all the information we obtained was recorded and managed in a GIS, utilising ArcView 3.3.

The Sources Apart from some isolated details, the written documentation does not provide much useful information for the reconstruction of the period; we could only use a few references in Ibn Hayyān’s account of the campaigns of ‘Abd al-Rahmān III in 920 (Ibn Hayyān 1981). Through this text, we know about the existence of some husūn or shelters/castles in a place called Falsahān, which must be present-day Falces. There are also some references to caves in another place known as Bitra Alta, probably Peralta, which served as a refuge when the troops of the Emir ‘Abd al-Rahmān evicted and killed Christians in the area. We do not find any other mention of these places until the eleventh century when the Tenencias of Arlas, Falces and Peralta appeared in this recently conquered territory (Martin Duque 1983). We must take into account that those three small Tenencias were situated in the plain, that is, in a productive area.

The first key question we are trying to approach concerns an old discussion about the origins of feudalism (Francovich and Hodges 2003). The place we chose investigate is a 1. This work can be undertaken thanks to the financial assistance of the Spanish Department of Education, Culture and Sport. 2. Grant holder of the Department of Geography, Prehistory and Archaeology at the University of the Basque Country.

The archaeological material available prior to our project is not very abundant and consists solely of the 77

SOMA 2004 1970s exploration of two sites of agricultural character dating to the first centuries of Roman presence (first and second century AD). These two sites, Los Villares and San Esteban, have been considered as villae (Mezquíriz Irujo 1996). The materials collected here are sherds of Terra Sigillata Hispanica and dolium type pottery. Some remains of an oil or wine press are also clearly visible. Finally, the Navarre Archaeological contained some information about Roman material that came from San Pedro de Arlas, in the same area.

of Roman dates and the medieval ones. Due to the lack of local excavations, we still have no knowledge of the diachronic development of occupation at each individual site. Nevertheless, from the point of view of a spatial perspective, the contiguity of the sites leaves no doubt about the connection between the Roman and Medieval elements. In this situation, our first step was to relate all the elements to each other while taking into account the geographical context. To do so we had to study the whole area extensively, through a surface survey.

In spite of this none-too-promising situation, the area bore a pleasant surprise, in the form of a horseshoe arch from the emiral ages (ninth century), which is preserved in the ruins of a building. That structure was a hermitage until it became a stable at an undetermined date. It contains not only the aforementioned arch, but also some brick and ashlar columns, probably from the High Middle Ages. The excavation of the site in 1998 revealed three silos, supposedly from the Early Middle Ages.

The Survey – Work Strategy In order to determine the areas to survey, we took two issues into account. On the one hand, we worked around the emplacements classified as sites in the archaeological inventory; on the other hand, to prepare the first campaign, we designed two large transects with the aim of including all the different ecological areas as well as some of the classified sites.

These are the only archaeological interventions we can consider until now. We should highlight, nonetheless, the existence of some more elements in the area, like the irrigation system itself and its relationship with the known sites and with the ones we expect to appear in the course of the project. It will be of particular interest to investigate the spatial relationship between the sites and the artificial watercourses, which are currently undergoing investigating.

In spite of these principles, the physical reality of the study area itself obliged us to design the survey strategy according to its topographical characteristics. The territory presents two very different landscapes: irrigated land and unirrigated land, separated broadly by the 300 m contour. The irrigated area is under intensive and frequently rotating cultivation. Thus, it is available for survey for short periods only, and all the fields are never available at the same time. The situation is very different in the second, unirrigated, area. As an effect of Community Policies, which favour large agricultural holdings over smaller ones, it is largely abandoned and dominated by rocks, weeds, etc, resulting practically in zero visibility, with minimum chances of successful survey.

The existence of two castles situated in higher locations also deserves consideration. The first one, Peralta, does not appear to have any particularly interesting features. The plank mould technique used in its walls and some of the structures excavated in the rocky terrain suggests an Early Medieval date (Cabañero Subiza 1996). The castle of Falces, on the other hand, presents some remains of ashlar-built walls dating from the ninth century. It may have been an atalaya, ie, a small tower, measuring no more than 60 m², traditionally considered typical of the Islamic period.

Taking into account these limitations and the stipulation that the project was to be completed in two years, in order to optimise resources, we decided to prepare short campaigns lasting no more than two or three weeks each. The first, and so far (i.e. by spring 2004) only campaign, took place in October 2003; the rest will follow in 2004. Our technique consists of a team of four people, in lines ten steps apart, walking through each field, collecting all the sherds and other artefacts visible at their feet.

There is also a second small tower (48 m²) whose wall is built of ashlar in its lower part. Lacking an entrance on ground level, it may be a structure of the kind described above, but in contrast to Falces, this tower is located in the middle of an agriculturally productive area on the plain.

Inevitably, the surveyed area was small (Fig. 1): hardly 10% of the total territory we planned to survey could be covered. Nevertheless, the study of the collected material has already provided some information and, more importantly, has marked some trends to consider in preparing the next campaigns.

We have recorded yet another site at an altitude of 45 m above the Arga river: the caves of Peralta. Although they are difficult to access, they are clearly recognizable even from a distance as artificial constructions. Could they be the caves that Ibn Hayyān referred to in AD 920?

The Classification of the Ceramic Material The material found is now undergoing study and some of the sherds are being compared with finds from excavations in the surrounding areas. We have completed a first classification and can now distinguish three groups:

Returning to the issue of the transition from the Roman to the Feudal period in this area, it is important to consider the relationship between the different sites and structures 78

FIGURE 1 THE AREA IN NAVARRA (WORK FRAME). DRAWING: JESÚS LORENZO

FIGURE 2 DOLIA TYPE CERAMIC CONCENTRATION. GIS : JESÚS LORENZO

79

FIGURE 3 DOLIA TYPE CERAMIC CONCENTRATION. GIS : JESÚS LORENZO

FIGURE 4 LITHIC MATERIAL: PLACES OF SIGNIFICANT PRESENCE. GIS : JESÚS LORENZO

80

JESÚS LORENZO JIMÉNEZ • • •

Building material: bricks and tiles. Lithic material: fragments of work material as well as tools. Pottery: a first classification according to the fabrics and surfaces has been made.

this field, indicating even at first glance that the scatter extends far beyond its limits into the neighbouring fields, which will be surveyed in future campaigns. As in the case of the previous site, the concentration of dolia sherds here suggests a similar function. In addition, there is some glazed pottery in this area. Could it indicate the Medieval settlement emplacement of Arlas?

The GIS The details of the assembled information were compiled in a GIS, on ArcView 3.3 software. Thus, we were able to display different areas according to the relative densities of the different ceramic groups derived from the abovementioned classification. The results of this first campaign are indicated in Figs. 2-4. The classification used indicates the ratio of sherd weight to surface area.

4. We have not yet been able to verify the presence of Early Medieval material. We are currently studying ceramic assemblages from excavated contexts securely dated to that period, in order to determine whether any parallels occur in the study area.

Provisional Conclusions It is too early to publish any detailed result at this point in the project. However, with the first stages of fieldwork already mapped, we can certainly point out some trends:

These are the preliminary conclusions drawn from a very small sample of the envisaged study area. In the further course of the project, systematic fieldwork will be used in conjunction with the study of different materials of the surrounding areas, as well as test excavations, to clarify the structure and development of the area at the transition from Antiquity to Feudalism.

1. We perceive a bigger accumulation or concentration of ceramics below the 300m contour, a trend marked even more clearly if we exclude glazed Medieval pottery. On the other hand, there is a remarkable density of lithic material on the periphery of the irrigated land, up to this same 300 m contour line, in areas that contain scarcely any Roman material.

Bibliography Cabañero Subiza, B. (1996) Los castillos catalanes del siglo X. Circunstancias históricas y cuestiones arquitectónicas, Zaragoza, Institución Fernando el Católico. Francovich, R and Hodges, R. (2003) Villa to Village, London, Duckworth. Ibn Hayyān (died 1076) (1981) Crónica del califa ‘Abd al-Rahmān III an Nāşir entre los años 912 y 942 (almuqtabis V). Traducción, notas e índices por M.J. Viguera y F. Corriente, Zaragoza, Anubar Ediciones. Larrea, J.J. (1998) La Navarre du IV au XII siècle. Peiplement et société, Paris-Bruxelles, De Boeck & Larcier. Manzano Moreno, E. (1991) La frontera de al-Andalus en época de los omeyas, Madrid,Centro Superior de Investigaciones Científicas. Martín Duque, A.J. (1983) Documentación medieval de Leire (siglos IX-XII), Pamplona, Diputación Foral de Navarra. Mezquíriz Irujo, M.Á. (1996) La producción de vino en época romana a través de los hallazgos en territorio navarro, Trabajos de Arqieología Navarra, 12, 63-90 Pérez De Laborda, A. (1984) Una calzada romana a lo largo del valle del Arga, Trabajos de Arqueología Navarra, 4, 145-56. Tudanca Casero, J.M. (1997) Evolución Socioeconómica del alto y Medio Valle del Ebro en Época Bajoimperial Romana, Logroño, Instituto de Estudios Riojanos.

2. The large amount of Roman material (sigillatae – mostly Hispanic-, dolia, and more fragments of pottery with no surface treatment, as well as fragments of tegulae) leaves no doubt as to the intensive occupation of the area in the first centuries AD. 3. These data already indicate the existence of three sites: (a) San Pedro de Arlas (field 7): A considerable amount of dolia sherds and Terra Sigillata Hispanica, mostly from the first centuries of Roman presence, was found here. As the place was already classified as a site, and the presence of material was obvious, we made a selective collection with a view to chronology rather than sherd densities. (b) Field 5: The material collected here (mostly dolia sherds) suggests the existence of a place of storage for agricultural produce. It also dates from the earlier Roman period. We also found some tegulae sherds, and a preserved wall at one edge of the field. (c) Field 34: There is a great, and quite homogenous, density of material in the whole area surrounding

81

The Urban Development of Leptis Magna in the Late First Century BC and Early First Century AD Ibrahim Kreetallah

Leptis has the same origin as that of Libya, which might derive from the name of the ancient Libyan tribe Lebu (rebu) as it is mentioned in the Egyptian records of the late second millennium BC (Baker 1981: 6).

Introduction This paper will examine the urban development of one of the most important Roman cities in North Africa, Leptis Magna. It will include a discussion of the site, its location and relationship with the countryside. By way of illustration, several early building complexes such as the theatre and market (macellum), dating to the late first century BC and early first century AD will be discussed.

The city acquired the epithet ‘the great’ early on to distinguish it from another city with the same name further west in Tunisia known as Leptis Minor, which was also founded by Phoenician traders (Mattingly 1995: 116).

Name, Foundation and Site Leptis Magna is located on the north coast of Libya and lies about three kilometres east of the modern town of AlKlhmus and about hundred and twenty-three kilometres from the centre of Tripoli, the capital of modern Libya. It was built on the coast at the mouth of the wadi Lebda (Manton 1988: 62) Leptis, with the cities of Sabratha and Oea, in antiquity formed the Tripolis, which means ‘the three cities’. Leptis has had a long and chequered history since its foundation by Phoenician merchants in the first millennium BC. The earliest foundation of the three cities, as shown by the evidence of the pottery is as follows: Leptis dates back to the seventh century BC, Sabratha to sixth century BC and Oea to fifth century BC. There are two variants of the name of the city: Leptis Magna and Lepcis Magna; Lepcis may derive from the Phoenician name of the city: LPOY or LBOY. In addition, it is also known as Lepcis and Lepcis megali (the great Leptis), According to some Greek sources in the early first century AD, it was called Neapolis (the new city) (Baker 1981: 6).

According to Manton, Leptis Magna was the most important city in Tripolitania during the Roman period and is one of the best-preserved cities of antiquity (1988: 62). It came under the authority of Carthage in the third century BC and had to pay daily tribute of one talent of silver to the city (Baker 1981: 10). After the fall of Carthage, Leptis came under the control of Numidia and was then integrated into the region of Africa Nova by Julius Caesar in 46 BC. It became a muncipium under the Flavians and was made a colony under Trajan. The Importance of the Location of Leptis MacKendrick argues that Leptis Magna and other cities in Tripolitania were founded by Phoenician merchants who were making regular voyages to Spain from the eastern Mediterranean (1980: 143). During these voyages, sometimes in starless or stormy nights, it was necessary for the Phoenicians to learn where they could land their ships on the coast to avoid storms and to obtain provisions. For this reason, they gradually established several trading posts in the western Mediterranean, in northern Africa and Spain—for example, Gades in Spain, Carthage in Tunisia and Leptis, Sabratha, and Oea in Tripolitania. Leptis probably began as a trading station and occasional harbour (Baker 1981: 7).

Mattingly states, ‘In many Latin sources the name was transliterated as Leptis Magna, but the local Latin epigraphy uses Lepcis which relates more closely to the Libyo-phoenician original Lpqy’ (1995: 116). As to the meaning or the origin of the name of Leptis or Lepcis, it is not certain whether it is Phoenician or native Libyan in origin. A number of suggestions have been put forward. One of them is that it is derived from the name of the ancient Libyan tribe Luatha and changed into Libate or Libade. Another suggestion is that it might be a Phoenician derivation of Libadah. A third idea is that the name of

Tripolitania, and North Africa in general, provided very important coasts for the Phoenicians for two reasons: firstly Tripolitania supplied them with to the rich resources of the Sudan and tropical Africa, for goods such as gold, precious stones, ivory, ebony, slaves and wild animals (Manton 1988: 62). Secondly, another attraction was the proximity 83

SOMA 2004 of Leptis to other Phoenician centres along the coast of Spain, which was a source of metals for the ancient world. Perhaps the most important reason for the Phoenicians’ choice of Leptis as a trading port was its geographical location, with a natural harbour at the mouth of a wadi.

Wadi Lebda, just southeast of the city, was potentially a good water source, though only periodically. As a result, the city looked further afield to the Wade Caam, the famously fertile Cinyps of Herodotus, which was located near the city to the east (Mattingly 1995: 117). Therefore, Leptis became the port of the rich grain harvests of the Wadi. In addition, the location of Leptis on the sea allowed it to become a gateway for trade into the interior with the best and shortest caravan route and with water available along the way to the Fezzan region to the south (Mattingly 1995: 155-7). Finally, Leptis’ location on the sea allowed access to other regions in the Mediterranean and made trading relationships possible with them.

The Advantage of Leptis’ Location and its Early Development It is clear from the previous discussion that the geographical location of the site is very important. The early Phoenician occupation of the site appears to have depended on the protection afforded to ships by a number of islands at the mouth of Wadi Lebda, which later became the Roman harbour. In addition, the sea conditions around Leptis were generally good during most of the year (Mattingly 1995: 117). Leptis is also situated in a dry farming region, unlike Oea and Sabratha. The Gebal Msellata, mountains around Leptis, curve to the north to meet the coast just to the west of Lebda and offer a good protection to the city from environmental conditions, such as strong winds from the northwest.

I would suggest that Leptis Magna developed very quickly. Although the city was initially laid out on neatly orthogonal lines, the city planner never managed to look far enough ahead, and as a result, the individual quarters of the city, in some cases, were out of alignment with their neighbours to varying degrees. The Relationship of Leptis with the Hinterland The hinterland was very valuable to Leptis, though its economy was also based on the caravan trade, such as the caravans of the Garamantes that connected the interior to the coast (Manton 1988: 62). The other economic

In the ancient world, two particularly influential factors for any city in the Mediterranean were a regular water supply and a profitable relationship with its hinterland. Both of these factors were available in the site of Leptis Magna.

FIGURE 1 THE SEVERAN BUILDINGS OF LEPTIS MAGNA (WARD PERKINS, J.B. 1993: 2) 84

IBRAHIM KREETALLAH resource that Leptis depended on was agriculture. Leptis was surrounded by rich agricultural land; there were olive groves in the hinterland and part of the city’s economy depended on the production of olive trees. MacKendrick points out that by the mid-first century BC, Leptis had to send a heavy duty in olive oil to Rome (1980: 145), a vast amount totalling three million pounds per year in modern terms (Mattingly 1995: 140).

military road that was also the commercial artery of the olive oil trade (MacKendrick 1980: 145). Early Building Complexes From the late first century BC to early first century AD, Leptis developed and flourished. During the first century BC, major development took place westwards and southwards. The lower ground to the east, toward Wadi Lebda lay empty until the construction of the great Hadrianic baths, which were dedicated in AD 126 (Haynes 1956: 75). The remains of the early public buildings include the Old Forum 5 BC to 2 AD, the market or macellum 9-8 BC, the theatre 1 to 2 AD, and the Chalcidicum 11 to 12 AD (Haynes 1956: 86-93). I will take the market and theatre as examples to illustrate the development of Leptis Magna at this time. Both were built with funds provided by the same

This signifies that Leptis had a good relationship with the its neighbours because most of the olive trees were farmed in the countryside areas, like Msellata and Trhonaa. In these areas, much evidence has been found for ancient olive presses (Mattingly 1988: 35-8), and they still have the best-farmed olives in modern Libya. Early in the reign of Tiberius c. 15-16 AD, Leptis became the terminus of the

FIGURE 2 TRIPOLITANIA (MATTINGLY, D.J. 1995: 117, FIG. 6:1) 85

SOMA 2004 local citizen whose name is a combination of Phoenician and Latin: Annobal Tapapius Rufus, a Romanised Libyan. (Reynolds and Ward Perkins 1955: 319, 321-2; Mattingly 1995: 170).

the middle of the second century AD. In the portico behind the stage building there is a temple to the deified emperor, which was dedicated in 42 AD by Iddibal Tapapius, a member of the same family as the builder of the theatre.

The Market or Macellum The market is located in the northwest part of the city near the theatre, bordering along the cardo maximus. The original market was a rectangular open space surrounded by porticoes on all four sides, with two tholoi or kiosks occupying the centre. At the southeast side of the market was an enclosure wall, which took the form of an open arcade with alternate large and small arches, but on the southwest side the main entrance was in the middle of the wall. Other sides, on the northwest, northeast and southeast were filled by tabernae. The market was orientated on a different street grid to adjust to the growing city. Building and decorative materials used were local at that time and did not include marble (Ward-Perkins 1981: 373-6). It is noteworthy that the closest parallels for plan of the macellum are to be found in Italy, notably at Puteoli and Pompeii (Ward-Perkins 1981: 166).

The theatre at Leptis Magna has great significance in a more general context; not only is it a distinctly Romanstyle building constructed at the expense of a prominent member of the local population, but it is also a very early example of type in North Africa. Conclusion It is clear from the surviving public monuments that Leptis Magna developed into one of the most impressive cities in North Africa, and it is particularly significant to note the development that had already taken place by the time of Augustus. As well as economic success, the city and region clearly witnessed a successful meeting of Latin and North African culture; much of this early development, at least in terms of the public buildings, was at the instigation of members of the local elite. This success was to continue for several centuries to come. The rich agriculture of the hinterland favoured the city’s success, and the choice of its location by Phoenician traders provided an ideal position to trade the goods and produce with the Mediterranean world. This is also shown by the city’s theatre and macellum, combining typical Roman elements with the newer idea of two kiosks, demonstrating the influence of the native inhabitants. Along with the theatre, it shows that the inhabitants of Leptis travelled and had contacts in Rome and throughout the Mediterranean. While we can tell much from the public works here, the private domestic side of life has yet to be understood.

Many different activities took place in the market, such as sales and purchases and observation of the products and of the vendors, and there still remain some standards of calibrated measures, such as measures of fish, grain and other products. At the time of the Severans, a new entrance was built at the southeast end, with a wide flight of steps. The colonnade was rebuilt with granite columns, and the columns (or pavilions) were constructed in green cipollino marble with white Corinthian capitals and lotus leaf work (Ward-Perkins 1981: 376; Haynes 1956: 91). The Theatre The theatre is located to the northwest of the Chalcidicum and was built and presented to the Emperor Augustus in 1 to 2 AD. It is similar in plan and in general character to other Roman theatres. The auditorium, or cavea, is semicircular with rows of seats facing the stage, or pulpitum, with a semi-circular orchestra. The orchestra had low steps for the seating of prominent citizens and had two exits where an inscription in neo-Punic and Latin can be seen recording that this theatre was dedicated in the year 1 to 2 AD by the same wealthy man who built the market, Annobal Rufus (Reynolds and Ward Perkins 1955: 319, 321-2; Mattingly 1995: 170). Another inscription, from 35 to 36 AD, at the top of the auditorium recorded the dedication of a small shrine to the goddess Ceres, also funded by one of the same family: Suphunibal, daughter of Annobal (Haynes 1956: 94).

Bibliography Baker, T. (1981), Leptis Magna, 2nd edition, Department of Antiquities, Tripoli, Libya. Haynes, D.E.L. (1956), The Antiquities of Tripolitania, London, Darf, MacKendrick, P. (1980), The North African Stones Speak, London, Croom Helm. Manton, E.L. (1988), Roman North Africa, London, Seaby. Mattingly, D.J. (1988) Oil for export, Journal of Roman Archaeology 1. Mattingly, D.J. (1995) Tripolitania, London, B.T Batsford Limited. Reynolds J.M. and J.B. Ward Perkins (1955) The Inscriptions of Roman Tripolitania, Rome, British School at Rome. Ward-Perkins, J.B. (1981) Roman Imperial Architecture, 2nd edition, London, Harmondsworth, Penguin, Ward-Perkins, J.B. (1993) The Severan Buildings of Lepcis Magna, The Society for Libyan Studies, Gordon Square, London.

Behind the stage was an ornamental wall, or scaenae frons. The scaenae frons was originally decorated with columns and statues of local limestone. These statues were replaced with marble during the reign of Antoninus Pius in

86

Eteocretans in The Bronze Age? The Far East of Crete During The LM II-III Periods (1450-1100 BC) Charlotte Langohr

over of (part of) the island. This is no longer accepted by all working in the field, and more interest is now taken in the study of characteristic archaeological features – the so-called ethnic markers – that can be successfully used to suggest the presence of a non-Minoan population on the island. If a non-Minoan population group indeed came over, the date of their establishment also needs to be defined, and here too opinions differ between LM II (Popham 1984: 263-4; Rehak and Younger 1998: 148-66, and for a list of scholars defending this date, see Niemeier 1985: 2, n. 20) or only in LM IIIA2/B (Hallager 1977; Niemeier 1985: 217-31, and for an exhaustive list of publications and authors supporting this view, see Cline 1997: 1634, n. 7-9). This approach, however, has an inherent danger, since archaeological characteristics associated with specific cultural groups within a society are then equated with an archaeological reconstruction of ethnic identities and geographical origins. Recent reassessments of archaeological excavations and finds both on the island and on the mainland have revealed a mixture of Cretan and Greek traditions that seem to have resulted in an entirely different and original society (for the most recent critical examination, see D’Agata and Moody forthcoming).

‘Mycenaean Crete’? The present study focuses on the Sitia peninsula during Late Minoan II and III, a period dubbed ‘postpalatial’ following the collapse of Minoan palace civilisation around 1450 BC. In the Archaic period, the east part of the island was often called ‘land of the Eteocretans’ and indeed Greek literary sources, including Homer, list the Eteocretans among the first inhabitants of the island, occupying the eastern part and who were, together with the Kydonians (probably the inhabitants of the Khania region) supposed to represent the only native populations of Crete (Odys. XIX, 172-177; see also Duhoux 1982: 9-12). We may ask whether this is a creation of eighth century writers or if it refers to an older, historical reality. Inscriptions identified as ‘Eteocretan’, using Greek characters to transcribe a non-Greek language and dating to the Archaic and Hellenistic periods, have been discovered in this precise region, principally at Praisos (Whitley 1998: 27-30). Language, however, seems only one of the particularities of this region which are revealed by archaeology. One of the purposes of the present study is to assess if this ethnic particularism can already be identified in the material culture of the Sitia peninsula in the later phases of the Bronze Age, and in particular, if a distinct material culture to which possible ethnic connotations can be connected, can be identified at the site of Palaikastro in the very east of the island. The present study is not intended to be exhaustive but it proposes to identify and examine a number of key traits found at Palaikastro, in order to appreciate the degree of originality of the site and to check for the potential presence of external elements resulting from regional, interregional, or external contacts which Palaikastro most likely maintained at this time.

LM II and III Palaikastro: Cultural Traits and Comparisons Old and recent excavations in the Minoan town of Palaikastro have clarified the considerable scale of reoccupation after the destructions at the end of LM IB. This reoccupation seems to have begun as early as the LM II period.1 This contrasts with the presumed gap in eastern Crete in the sequence between the LM IB destructions and the LM IIIA1 reoccupation which led some scholars to posit an ‘abandoned landscape’ (Bennet 1987: 79-81. For recent

The island, during the LM II to LM IIIA-C ceramic periods, roughly between 1450 and 1100 BC, is sometimes termed ‘Mycenaean Crete’, a label concealing much of the historical complexity of the period, including questions about the ancient and multifaceted relationship between the island and the Greek Mainland (Driessen and Farnoux 1997). The Linear B tablets found at Knossos and now also those discovered at Khania have been considered by some as a convincing argument in favour of a Mycenaean take-

1. Archaeological evidence for a very early reoccupation consists of a sherd of a LM II Ephyrean goblet near Building III, a possible LM II sherd from Block Π, a foundation deposit in Building I, including a bridgespouted jar similar to the LM II examples of the Unexplored Mansion at Knossos, and a fragment decorated in Palace Style; moreover, in Building VII, there is a stratigraphical level posterior to the LM IB destruction and situated below the LM IIIA1 levels; finally, a LM II strainer was found in a tomb of the South Cemetery against the slopes of Petsophas.

87

SOMA 2004

FIGURE 1 PALAIKASTRO, BLOCK Δ, LM III NORTHERN FAÇADE, ASHLAR (PHOTOGRAPH BY AUTHOR) evidence for an early reoccupation at Mochlos and with a particular examination of questions of cultural continuity and identity, see Brogan, Smith, et al. 2002, especially 103-5). During the LM II and III periods, Palaikastro is one of the few larger towns still existing on Crete. The first reoccupation during LM II-IIIA1, 1450 to 1370, is still only sporadic but from early LM IIIA2 onwards, the Roussólakkos settlement is largely reconstructed and occupied until an advanced moment in LM IIIB during the thirteenth century when it is suddenly abandoned. A number of LM II to LM IIIB tombs have been excavated in various places around the town, forming small cemeteries. Palaikastro probably played an important role within the complex relationships that existed between the surviving settlements of eastern Crete but the nature of these links remains to be studied in detail (for a first attempt, see Tsipopoulou 1995). An analysis of the archaeological material of the site allows us to highlight some distinctive features.

one. As yet, no centrally located hearths like those on the mainland and at Khania or Malia-Nu have been found; if main rooms comprise two axially placed columns, there is never a trace of a fire-place between them, as is the case in Quartier Nu at Malia, for instance (Driessen and Farnoux 1994: 60) and this becomes the habit in Post-Minoan Crete. The only possible case for the presence of a Mycenaeanising axially organised building is a badly known structure in Block X. Elsewhere on the island, however, more substantial changes can be observed (Hayden 1981; Hayden 1987: 199-234; Hayden 1990: 203-13). Hence, axially placed two- or three-room buildings appear in Kephala Chondrou, Phaistos, Kommos or Amnisos and the Megaron plan bordered by a corridor opening onto auxiliary rooms has been identified at Plati, Gournia and Aghia Triada. These last examples, however, already show a hybrid character in that they combine Minoan and Mycenaean features; they lack a central hearth and, in Minoan fashion, include courts or have open façades. A good example is Quartier Nu at Malia, a LM IIIA2-B complex with axially laid out rooms and an insula organisation that corresponds well with certain Mycenaean structures, but is nevertheless organised around a small court following a Minoan tradition. As mentioned, Quartier Nu includes at least two examples of central hearths between columns (Driessen and Farnoux 1994). The architectural changes between the Neopalatial and Postpalatial phases in the Khania Aghia Aikaterini Square buildings are even more striking: different orientations, different building materials, larger rooms, absence of upper floors and, of course, the presence of permanent circular central hearths of clear Mycenaean type (Tzedakis and Hallager 1983: 14; Tzedakis and Hallager 1985: 22; Hallager and Hallager 2003: 186-94). In Aghia Triada, the LM IIIA2 ‘Edificio N-O’ and ‘Edificio P’ are compared to the Korridorhaus on the mainland but

Architecture Where architecture is concerned, it should be noted first that rebuilding during the reoccupation period often followed the same lines as earlier Minoan construction. There is a surprising continuity, indeed, in location, plan and orientation of buildings (Hayden 1981: 169; Hayden 1987: 204-5 and 216). The rapid reoccupation and the continuity indeed seem to imply a stability and continuity of population. Certain architectural features such as the exceptional ashlar façade of Block Δ (Fig. 1), still preserved for a length of more than 10 metres – if it indeed dates to the post-LM IB phase – and the relaying of the main road with cut stone slabs testify to the flourishing of the town. Architectural features characterised as ‘Mycenaean’ elsewhere on the island are largely absent: rooms are still agglutinating and non-axially laid out, houses are not isolated but adjoining, courts and light wells are totally integrated within houses, entrances from the street very often give a direct access via stairs to the upper floor as in LM I, and indeed, the entire town plan seems to some degree a copy of the extremely structured Neopalatial

2. The LM II and III Palaikastro pottery has been presented elsewhere (C. Langohr, mémoire de DEA, Université Catholique de Louvain, Addendum) and was based on an analysis carried out by J.A. MacGillivray in J.A. MacGillivray 1997a. I thank J.A. MacGillivray for letting me consult his forthcoming study of Palaikastro pottery (J.A. MacGillivray forthcoming).

88

CHARLOTTE LANGOHR the architectural solutions preferred are both Minoan and Mycenaean (La Rosa 1997; Cucuzza 1997).

(a stemmed drinking vessel) which are frequent in western and central Crete but rare in Palaikastro and eastern Crete. Published archaeological reports concerning LM II and III sites in the peninsula of Sitia show that the presence of kylikes or of its Minoan derivative the Champagne Cup is really quite exceptional. At Palaikastro, sherds of two LM II Ephyraean goblets were discovered in houses, an imported LM IIIA1/2 kylix comes from the Block Γ and a few fragments of kylikes were recovered from wells 576 and 605. Finally the LM IIIB to final LM IIIC settlement on Kastri revealed several late examples. Moreover, since many of these typical local Pulled-rim bowls are now safely dated to the LM II to IIIA1 period in Palaikastro, some kind of shift towards regionalism in eastern Crete seems to happen directly after the LM IB destructions of the Neopalatial ‘Palaces’ and not only from LM IIIA2 when the last ‘Palace’ of Knossos is destroyed.

Ceramics Where pottery is concerned, we may start off by noting that the LM II to IIIB pottery found in Palaikastro is impressive, both in terms of its quantity and quality.2 Ceramic analyses have already shown how common, domestic wares entirely lack parallels with other contemporary Cretan or Mycenaean examples. Many local shapes that existed in LM IB continue to develop. Again, this suggests an uninterrupted site occupation through the LM IB destructions. Pottery from Palaikastro is locally made without major references to the important centres of the period. This generated a pottery with an original and distinctive fabric, surface treatment and shape – features already used by Athanasia Kanta to identify the ‘Palaikastro workshop’ (Kanta 1980: 288-90). The very red local clay is uniformly covered by a creamy to pale yellow slip on which red to deep black motifs were added (PKU: 76).

Another distinctive product of the ‘Palaikastro workshop’ is the jug or the circular mouth amphora often decorated in the Blot and Trickle Style and distributed quite widely in eastern Crete in LM IIIA2-B. A decoration with ornamental and figurative motives organised in friezes or panels is also quite particular to its production, painted for example on piriform jars or amphoroid kraters.

Amongst the favourite shapes there is above all the famed Pulled-rim Bowl, an often large ogival bowl with the rim pulled into a spout and two opposing bosses added at right angles to the spout just below the rim. Its distinctive profile clearly shows how it represents an evolution of the local LM IB ogival cup and it illustrates continuity of local pottery traditions after the destructions of LM IB. The pulled-rim bowl is most frequently decorated in styles that are typical for the site, mainly the Dip and Run Style, sometimes the Dip-disk Style (Fig. 2). This bowl represents the most typical drinking vessel of LM II-III Palaikastro but also elsewhere in eastern Crete with local productions at Mochlos and Myrsini (Smith 2002: 170-1). It seems to be the functional equivalent of the LM/LH goblet and kylix

Thanks to such typical shapes and decorative patterns as the Dip-Disk style, the Dip and Run Style and the Blot and Trickle Style, we can follow the diffusion of the ‘Palaikastro workshop’, all throughout eastern Crete and even in central Crete, such as at Karoumes, Gournia (Boyd Hawes 1908: pl. X. 4), Episkopi Ierapetras (Kanta 1980: fig. 57 :7, 59 :1), Katsounakia (MacGillivray 1997b: 278, n. 30), ViannosKera Limaniotissa (Banou and Rethemiotakis 1997: 32), Katsambas (Alexiou 1967: 45, pl. 8b), Malia (J. Driessen, personal communication) and even in Khania in west Crete (MacGillivray 1997a: 206; Hallager and Hallager 2003: 248, 288) and outside of Crete, on the island of Karpathos (Melas 1985: among others, the amphoroid kraters 109114, fig. 93 (987), fig. 94 (989)). Oddly enough, ceramic production in the Viannos region is entirely influenced by Knossian prototypes during LM II-IIIA1, whereas, during LM IIIA2, shapes and motives were strongly influenced by Palaikastro. May we ask whether, after the early LM IIIA2 destruction of the palace of Knossos, Palaikastro grabbed a piece of the cake by imposing itself on this central Cretan trade network?3 Elsewhere on Crete, ceramic production shows many more Mycenaean characteristics: shapes such as kylikes or goblets, pyxides, deep bowls, one handled cups or stirrup jars are well represented especially at Knossos and Khania but are rare or not common at Palaikastro or elsewhere in the Sitia peninsula, except as imports. The Khania workshop is most sensitive to Mycenaean influences but keeps its traditional Minoan character. It has its heyday during LM IIIB1 with pottery from the ‘Kydonian workshop’ being distributed throughout the island and abroad (Hallager and Hallager 2003: 197-265 for the LM IIIB2 period).

FIGURE 2 PALAIKASTRO, BLOCK κ, LUSTRAL BASIN, PULLED-RIM BOWL WITH A DIP-DISK STYLE DECORATION (J.A. MACGILLIVRAY 1997A: FIG. 4B). 3. This particular point is discussed in a forthcoming paper of the author. By taking a census of sites where pottery of the ‘Palaikastro Workshop’ has been recognised, an attempt is made to trace the diffusion of workshop production.

89

SOMA 2004 Mycenaean characteristics in pottery can be explained by acculturation and do not need to be understood in terms of the establishment of groups of persons originating from the mainland. Real Mycenaean imports are rare throughout the Late Bronze Age. Sites such as Malia in LM III adopt some Mycenaean practices, essentially concerning elite drinking vessels. The kylix is imitated and locally produced, but always with some local particularities such as a perforated stem; it is often coupled with the production of the Champagne Cup, the typical Cretan counterpart of the kylix (Farnoux 1997). The situation is similar at Archanes-Tourkogeitonia, for example, in the Knossos hinterland (Andrikou 1997) or at Khania as noted above, even if important ceramic differences exist between these regions. The danger of defining distinctive archaeological elements associated with specific cultural groups within a society and in equating them with an archaeological reconstruction of ethnic identities and geographical origins is hence more than real.

the symbols of powerful foreign groups. At the same time [these expressions] were set in a more familiar context by including symbols of local group identity (typically local drinking shapes) that were common throughout the Mochlos-Myrsini Cemeteries’. By the time of LM IIIA2B, the limited use of shapes with specific connections to a Knossian or Mycenaean identity suggest that the ‘elite of Mochlos-Myrsini were firmly entrenched and no longer needed to promote their authority with references to foreign symbols of power’ (Smith 2002: 276-7). Finally, comparing the LM III pottery of the cemeteries of Mochlos-Myrsini with those of Knossos, Smith concludes that ‘instead of categorizing the material as a part of an implicitly ethnic culture, it is possible to achieve a multidimensional picture of the strategies used by groups and individuals in the negotiation of social, and even political, power’ (Smith 2002: 278). Funerary Practices It may be interesting to look at the funerary practices since, at the beginning of the twentieth century quite a few LM III tombs were excavated at Palaikastro. It should be mentioned that Neopalatial tombs are extremely rare but this observation applies to the entire island: other ways of formal disposal must have been practiced. In LM III, larnakes, terracotta coffins shaped like a bathtub or a chest, sometimes decorated with painted motifs similar to those on pottery, appear in the Palaikastro tombs. The dead are accompanied by some offerings, and the practice of secondary burial, which existed in the Early Minoan and Middle Minoan periods, is reactivated. The larnakes are placed in rock crevices or dug out pits, sometimes stone lined. The entire Sitia peninsula shows a remarkable unity in these funerary practices and some cemeteries even suggest an itinerant workshop where the production of larnakes is concerned (Tsipopoulou and Vagnetti 1995: 131-45). Largely lacking is the ostentation that exists in central and west Crete, apart from the Achladia-Platyskinos tholos and a dubious case at Palaikastro-Angathia, large tombs are absent; the dead are never given figurines or complex vases.

If it is not necessary to explain such changes and developments by a real presence of, or a takeover by, Mycenaeans; there still has to be a mechanism by which such an adoption of new habits is considered profitable. Here some comparative studies may be of help. P. Day and his colleagues have tried to identify evidence for Cycladic ethnicity in their study of the pottery from the Early Bronze Age Cemetery at Aghia Photia in northeastern Crete. As stressed: ‘even if pottery and other material culture is rejected as a simple reflection of ethnic groups in the past, an awareness of its potential use in self-definition and social reproduction is important in considering funerary contexts’ (Day 1998: 140). This should be compared with A.L. D’Agata’s assertion that kylikes found in the tombs at Knossos indicate a distinction between burials of Mycenaean ethnicity and Minoan ethnicity (1999). She argues that, within the Mycenaean aristocracy of Knossos, kylikes marked out a specific group, assuming that social drinking practices ‘probably played [there] a decisive role in indicating the individual’s social connotations’ (D’Agata 1999: 52). R.A.K. Smith also underlines the presence of the kylix in some Knossian tombs but explains this somewhat differently, reflecting the adoption by some of the elite of Knossos of a particular type of drinking practice with which this vessel was associated and which originated in the Mycenaean world, in the scope of ‘aligning themselves to the newly emerging power and influence of the Mycenaean world, instead of the older traditions of the Neopalatial Minoan world’ and then ‘consumption of kylikes … does not necessarily indicate the presence of ethnic Mycenaeans in these tombs. It does, however, signal a connection to a specific Mycenaean drinking practice, and likely identifies the occupant as a member of a distinguished group’ (Smith 2002: 269-70). In his study of the LM III pottery of the tombs of Mochlos and Myrsini, the same author argues that some features connected with LMII-IIIA1 Knossos or Mycenaean practices were ‘elite expressions of individual identity [which] established social status through links to

Elsewhere on the island, the variety of funerary practices is also larger, such as pithos burials in the Mochlos, Pachyammos, and Viannos regions (Banou and Rethemiotakis 1997: 46), larnax burials (rare at Khania but more common in the Rethymnon area), and chamber tombs (especially common in central and west Crete); pit graves in enclosures which are seen as a Mycenaean feature as at Knossos and Archanes, funerary steles such as at Archanes and Rethymnon-Armenoi, shaft graves, and monumental tholos tombs. Most of these are features that show a more common spirit between Cretan and Mainland funerary practices. Moreover, valuable funerary offerings, weapons and metal vases, occur both on the mainland and in central and western Crete, but are almost absent in the Sitia peninsula. Whereas, elsewhere, a mixture of Minoan and Mycenaean features can perhaps be explained as a deliberate strategy by an elite trying to impose itself on 90

CHARLOTTE LANGOHR Archives Finally, we may add that thus far no Linear B document – either a tablet, a stirrup jar or a sealing – has been found in the far east of the island and that none of the toponyms mentioned in the Knossos Linear B tablets seems to refer to the east of the island. This too may suggest that the very east of the island was in some way independent from the Knossos administration at the time of the tablets. Conclusion According to S. Jones, ethnic groups may be defined as ‘culturally ascribed identity groups, which are based on the expression of a real or assumed shared culture and common descent’ (1997: 84). She also insists on the economic and political importance of ethnicity: social groups require organizational bonds necessary to maintain boundaries in order to secure access to resources and, therefore, ethnicity can be use in a strategic way as ‘a means to an end’ (Smith 2002: 275). Indeed, there is ‘rarely a straightforward correlation between cultural similarities and differences and ethnic boundaries’ (Jones 1996: 67). Palaikastro and the far east of the island are notably different from the rest of Crete during the LM III period and differences can be observed in all domains. As noted by Metaxia Tsipopoulou, the Sitia Mountains east of the Isthmus still represent a formidable barrier, isolating the region from the centre and orienting it more to the south-eastern Aegean and eastern Mediterranean (1995: 177). Certain distinctive features already existed in Palaikastro during the Early to Late Minoan I periods but, by LM I, its material culture is closely similar to that of, for instance, Knossos and Khania. Not so after LM IB, however. Material culture shows a remarkable originality, surprisingly free from Knossian and/or Mycenaean influences. We may indeed be facing the birth of the Eteocretan people. But since we have also highlighted the surprising continuity between LM I and LM III at Palaikastro, it seems more likely that east Crete was always unique and perhaps only less affected than the rest of the island during LM III by foreign influences. Distinct material culture and a lesser sensitivity to foreign influences – despite an excellent geographical location do not mean isolation: external elements and imports were present at Palaikastro at this time and its pottery and other ceramic productions show a diffusion in and out of Crete, both showing the maintenance of regional and external contacts. Finally, the remarkable lack of Mycenaean characteristics or goods at Palaikastro during this period does not help to solve the much vexed question of if and when a new group from the Greek mainland arrived. If Mycenaeans ever came over, they never seem to have made it to Palaikastro.

FIGURE 3 FIG. 3: PALAIKASTRO, BLOCK ∆, ROOM 44, GROUP OF LYREPLAYER AND DANCERS (PKU: FIG. 71) a Minoan population, by legitimising its place through a Minoan symbolism, this seems largely absent in the far east. Cult practices All through the island, LM III cult practices take the form of domestic sanctuaries often with benches. At Palaikastro the continuity noted in architecture, pottery and, to some degree, funerary practices, is repeated: buildings that were used for cult practices during the Neopalatial period seem to maintain this function. The most surprising example is the presence of an axial shrine with figurines above the northwest rooms of Building V where the famous LM I chryselephantine kouros was found. But elsewhere too, earlier religious practices are paralleled by LM III ritual activities such as the head of a terracotta figurine and an important collection of shells in Building I, a male figure with hands against the chest in building IV, a bull-head rhyton in building VII and the terracotta group of four dancers from Block Δ room 44 (Fig. 3). Only a single fragment of an imported Mycenaean figurine has been found at Palaikastro whereas they are especially common in west Crete, both imported and locally made (Tzedakis and Hallager 1983: 14; Hallager and Hallager 2003: 191, n. 18-19, 270-3). Female figures with upraised arms are common in central and west Crete but hitherto largely absent from the far east of the island. A fragment of a terracotta hand and the head of a female figure from Building I show that such figurines may have existed at the site but in much smaller numbers than elsewhere. Regional sanctuaries such as Psychro, Kato Symi, Ida and Iuktas remain in use but thus far none have been found in east Crete (Bennet 1986: 56-60). Palaikastro itself was of course home of an important regional sanctuary for Diktaean Zeus from the Geometric period onwards, but apart from assumptions based on the chryselephantine statue, there is no actual proof that this practice went back as far as the Late Bronze Age.

Acknowledgements This paper forms part of the Project ‘A Topography of Power in Minoan Crete’, based at the Université Catholique de Louvain (Belgium). For critical and encouraging comments on the different versions of this paper I thank Prof. J. Driessen and Prof. J.A. MacGillivray. 91

SOMA 2004 Thank you too to my friends and colleagues Quentin Letesson, Tim Cunningham, Peter Tomkins and Klaas Vansteenhuyse for the lively and stimulating discussions. Finally, I would like to thank Killian McAleese of the SOMA 2004 Committee for his thorough final corrections on this paper.

Day P.M. and E. Kiriatzi (1998) Pots, Labels and People: Burying Ethnicity in the Cemetery at Aghia Photia, Siteias. IN Branigan, K. ed. (1998) Cemetery and Society in the Aegean Bronze Age, Sheffield, 133-49. Driessen, J. and A. Farnoux (1994) Mycenaeans at Malia?, Aegean Archaeology, 1, 54-64. Driessen, J. and A. Farnoux eds. (1997) La Crète Mycénienne, Actes de la table ronde internationale organisée par l’Ecole Française d’Athènes, Bulletin de Correspondence Hellénique suppl., 30, Athens. Duhoux, Y. (1982) L’Étéocrétois. Les textes-La langue, Amsterdam. Farnoux, A. (1997) Malia aux Minoen Récent II et IIIA1, IN Driessen, J. and A. Farnoux (1997) 135-47. Hallager, E. (1977) The Mycenanean Palace at Knossos: Evidence for Final Destruction in the LM IIIB Period, Stockholm. Hallager, E. and B.P. Hallager eds. (2003) The GreekSwedish Excavations at the Agia Aikaterini Square. Kastelli, Khania. 1970-1987 and 2001. The Late Minoan IIIB2 Settlement, vol. III. 1, Stockholm. Hayden, B. (1981) The Development of Cretan Architecture from the LM IIIA through the Geometric Period, Ann Arbor. Hayden, B. (1987) Crete in Transition: LMIIIA-IIIB Architecture, a Preliminary Study, Studi Micenei ed Egeo-Anatolici, 26, 199-234. Hayden, B. (1990) Aspects of Village Architecture in the Cretan Post-palatial Period. IN Darcque, P. and R. Treuil eds. (1990) L’habitat égéen préhistorique, Actes de la Table Ronde internationale tenue à l’Ecole Française d’Athènes, Athens, 23th-25th June 1987, Bulletin de Correspondence Hellénique Suppl., 19, Paris, 203-13. Jones, S. (1996) Discourses of Identity in the Interpretation of the Past. IN Graves-Brown, P., Jones S. and C. Gamble eds (1996) Cultural Identity and Archaeology: the Construction of European Communities, New York, 62-80. Jones, S. (1997) The Archaeology of Ethnicity, New York. Kanta, A. (1980) The Late Minoan III Period in Crete. A Survey of Sites, Pottery and their Distribution (SIMA LVIII), Göteborg. La Rosa, V. (1997) Aghia Triada à l’époque mycénienne : l’utopie d’une ville capitale. IN Driessen, J. and A. Farnoux (1997), 249-66. MacGillivray, J.A. (1997a) Late Minoan II and III Pottery and Chronology at Palaikastro: An Introduction. IN Hallager E. and B.P. Hallager eds. (1997) Late Minoan III Pottery. Chronology and Terminology, Acts of a Meeting held at the Danish Institute at Athens, 12th14th august 1994, Athens, 193-207. MacGillivray, J.A. (1997b) The Re-occupation of Eastern Crete in the LMII-IIIA1/2 Periods. IN Driessen, J. and A. Farnoux (1997), 275-9. MacGillivray, J.A. (forthcoming) The Late Minoan Pottery. IN Sackett L.H., MacGillvray, J.A. and J. Driessen eds., Palaikastro: Two Late Minoan Wells and their History.

Bibliography Alexiou, S. (1967) Υστερομινωικοί τάφοι λιμένος Κνοσσού (Κατσαμπά), Athens. Andrikou, E. (1997) Thoughts and Considerations on the Mycenaeanisation of Crete motivated by Pottery from Archanes. IN: Driessen, J. and A. Farnoux (1997), La Crète Mycénienne, Actes de la table ronde internationale organisée par l’Ecole française d’Athènes, Bulletin de Correspondence Hellénique suppl., 30, Athens, 9-22. Banou, E. and G. Rethemiotakis (1997) Centre and Periphery : New Evidences for the Relations between Knossos and the Area of Viannos in the LMII-IIIA Periods. IN Driessen, J. and A. Farnoux (1997), La Crète Mycénienne, Actes de la table ronde internationale organisée par l’Ecole française d’Athènes, Bulletin de Correspondence Hellénique suppl., 30, Athens, 23-57. Bennet, J. (1986) Aspects of the Administrative Organization of LMII-IIIB Crete: A Study Based on Archaeological and Textual Data (unpublished PhD), Cambridge. Bennet, J. (1987) The Wild Country East of Dikte: the Problem of East Crete in the LMIII Period, Minos XXXXII, Salamanque, 77-88. Hawes, H.B., Williams, B.E., Seager, R.B. and E.T. Hall, 1908. (1908) Gournia, Vasiliki and other Prehistoric Sites on the Isthmus of Hierapetra, Crete. Brogan, T.M., Smith, R.A.K. and J.S. Soles (2002) Mycenaeans at Mochlos? Exploring Culture and Identity in the LM IB to LM IIIA1 Transition, Aegean Archaeology, 6, 89-118. Cline, E. (1997) A Wrinkle in Time: Orientalia and the Mycenaean Occupation(s) of Crete. IN Phillips, J. ed. (1997), Ancient Egypt, the Aegean, and the Near East. Studies in Honour of Martha Rhoads Bell, vol. I, San Antonio, 163-7. Cucuzza, N. (1997) The North Sector Buildings of Haghia Triada. IN Driessen, J. and A. Farnoux (1997), La Crète Mycénienne, Actes de la table ronde internationale organisée par l’Ecole française d’Athènes, Bulletin de Correspondence Hellénique suppl., 30, Athens 73-84. D’Agata, A.L. (1999) Hidden Wars: Minoans and Myceaneans at Haghia Triada in the LM III Period. The Evidence from Pottery. IN Laffineur, R. ed. (1999) Polemos. Le contexte guerrier en Egée à l’Âge du Bronze (Aegaeum 19), Liège-Austin, 47-55. D’Agata, A.L. and J. Moody, (forthcoming) Ariadne’s Threads. Connections between Crete and the Greek Mainland in the Post Palatial Period (Late Minoan IIIA2 to Sub-Minoan), Proceedings of the International Workshop held in Athens, April 5-6 2003 92

CHARLOTTE LANGOHR Melas, E.M. (1985) The Islands of Karpathos, Saros and Kasos in the Neolithic and Bronze Age (SIMA LXVIII), Göteborg. Niemeier, W.-D. (1985) Die Palaststilkeramik von Knossos, Stil, Cronologie und historischer Context, Berlin. PKU = Bosanquet, R.C. and R.M. Dawkins (1923) The Unpublished Objects from the Palaikastro Excavations 1902-1906, BSA Supplementary Paper, 1. PK VII = Sackett, L.H. and M.R. Popham (1970) Excavations at Palaikastro. VII, Annual of the British School at Athens, 65, 203-42. PK 1988 = Sackett L.H., J.A. MacGillivray, et al. (1989) Excavations at Palaikastro, 1988, Annual of the British School at Athens 84, 417-45. Popham, M.R. (1984) The Minoan Unexplored Mansion at Knossos, BSA Supplement Paper, 17, London. Rehak, P. and J.G. Younger (1998) Review of Aegean Prehistory VII: Neopalatial, Final Palatial and Pospalatial Crete, American Journal of Archaeology, 102, 91-173. Smith, R.A.K. (2002) The Tombs of Mochlos and Myrsini: Pottery and Cultural Regionalism in Late Minoan

III Crete (UMI Dissertation Services 3077184), Ann Arbor. Tsipopoulou, M. (1995) Late Minoan III Sitia. Patterns of Settlement and Land Use. IN Tsipopoulou, M. and L.Vagnetti (1995), 177-92. Tsipopoulou, M. and L.Vagnetti (1995) Achladia. Scavi e ricerche della Missione Greco-Italiana in Creta Orientale (1991-1993), Roma. Tzedakis Y. and E. Hallager (1983) The Greek-Swedish Excavations at Kastelli, Khania, 1980, Athens Annals of Archaeology XVI, 3-17. Tzedakis Y. and E. Hallager (1985) The Greek-Swedish Excavations at Kastelli, Khania, 1984, Athens Annals of Archaeology XVIII, 9-28. Whitley, J. (1998) From Minoans to Eteocretans : the Praisos Region 1200-500 BC. IN Cavanagh W.G. and M. Curtis eds. (1998), Post-Minoan Crete, Acts of the first Symposium held at the British School of Athens and at the Archaeological Institute of the University College of London, 10th-11th Nov. 1995, London.

93

Reconstructing Minoan Monumentality Quentin Letesson

amount of energy at the moment of construction – and in their subsequent maintenance, as in Minoan Crete – but it is especially their visibility over a long time span that is extraordinary. There is thus a clear and direct relationship between a monument’s design and its communicative potential. The imageability (Lynch 1960) and legibility of a monument is foremost shaped by what we can see. To comprehend monuments one thus has to place oneself into the viewer’s perception of those monuments.

No one will ever be able to tell what it felt like to gaze at the West Façade. Klynne, A. Introduction It is commonplace for architectural studies to focus on functional problems. Whatever method used, the interpretation of architectural spaces often leads to the question of their use (Palyvou 1997: 155-6). Furthermore, even if aesthetic and typological approaches that characterise the work of archaeologists become more semiological (Sanders 1988: 496-7), the intimate relationship between a built space and those who frequented this space is often neglected. It is the aim of this paper to analyse, by means of theoretical concepts of visual perception, the impact certain buildings had on their users. Monumentality will be our primary interest and Cretan Bronze Age structures traditionally described as ‘Palaces’ – perhaps better termed Minoan Court Complexes (Driessen, Schoep, et al. 2002: 1-13) – our principal field of investigation.

A cognitive approach of this kind requires some theory. In his visual analysis of the Japanese landscape, Tadahiko Higuchi defines a series of variables which can be useful in understanding, in terms of visual impact, the way the Ancients perceived their built environment (1983). To some degree, these variables may help us to clarify some of the architectural features of the Minoan ‘palaces’. For a start, it is good to remember that the human maximum field of vision ranges between 50-55 degrees above the horizontal to 70-80 degrees below it. The normal line of sight of a standing person is approximately 10 degrees below the horizontal and 15 degrees when sitting. The optimum viewing zone – the area visible when moving the eyes but not the head – ranges from 25 degrees above the horizontal to 30 degrees below (Higuchi 1983: 40, fig.4.4). The angle of elevation or the difference between the horizontal and the line of vision going from our eyes to the object we look at is one of the most important indexes in the study of monumentality. An upward view is limited in more than one way because the process of looking upwards at an object tends to limit the mobility of the human body and to cut off the line of vision at a point above the horizontal. With the most stable line of vision for the average person being about 10 to 15 degrees below the horizontal, it follows that the very process of looking up involves a certain amount of stress. Looking up to someone or something requires a visual effort (Higuchi 1983: 46-7). The so-called Märtens’s rule, adopted by Higuchi, is a development of this idea (Higuchi 1983: 47-8). In short, Märtens took the objective position that the total aesthetic effect of a given object is related to the special characteristics of the human eye’s range of vision

Monumentality According to Trigger, the principal defining feature of monumentality is that its scale and elaboration exceed the requirements of any practical functions that a building is intended to perform (Trigger 1990: 119-20). But it is quite clear that a monumental building should foremost have a strong communicative potential, that it is able to express political and/or religious messages that help societal integration. Indeed, it is assumed that when a society increases in size, the resolution of social problems becomes a more problematic issue and that people loose the ability to engage in face-to-face contact. Such societies need intermediaries to solve problems that may arise (Moore 1996: 97; Lynch 1960) and monuments may often serve as unambiguous markers of social relationships, not only because of their scale, but also because of their visual prominence (Driessen 1999: 122; Vansteenhuyse 2002: 235-45). The key to monuments is, therefore, their visual impact and legibility, their capacity to convey meaning, and the clarity with which this meaning can be read (Moore 1996: 97). Monuments concentrate a large 95

SOMA 2004 (which forms an elliptical cone) and to a human being’s visual powers. This means that it is not only the absolute height of a monument that determines its visual perception but also the relative position of the viewer. There appear, indeed, to be several thresholds of vision. By approaching an object, the latter will gradually fill the individual’s vision. According to the Märtens’s rule, visual perception changes at 18 degrees when an object first takes on a monumental nature. At 27 degrees, the object fills the viewer’s range of vision and larger details become visible. Eventually, when the viewer reaches a point where the angle is 45 degrees, he is at the best place for observing comparatively small details.

and a below reinforced by gravity; the forward-backward axis – an axis of the walk, of movement in space without slopes; and the transversal axis, or right-left axis, closely linked with the concept of symmetry. These three axes play a decisive role in our perception of space, especially where movement is concerned. They give space its dynamic or static character. In the perception of space through vision, the existence of the axes is in fact less important than the oppositions that affect them. When an axis is blocked, it creates a stop in a particular direction. When it is promoted, it creates a possibility of movement. It is the positive-negative and dynamic-static nature that makes architecture live. Actually, through our vision, it interacts with our perception of space. Cousin argues that a human being sees through an ellipse with golden rectangle proportions (1980: 60-1). This ellipse allows us to perceive space in a comfortable and not maximal way, to understand space without having to move our eyes too much. Although it can vary from one person to another, the ellipse is considered to have an angle of 54 degrees horizontally and of 37 degrees vertically, which in fact implies 23 degrees below the horizontal and 14 degrees above.

Obviously, visibility not only changes with the relative position of the viewer but constructed landscapes and built spaces can also be manipulated to block or enhance visual perception and the communicative potential of monuments. We should consider that it is not the object we are looking at that gives us a visual sense of space but the object’s background. This is exactly the case with the perception of depth. A second theoretical contribution is proposed by Jean Cousin in L’espace vivant (1980). It concerns the influence a built environment may have on the people whom it confronts. In his writings, a human being is considered an intelligent organism strongly affected by the spatial relationships of his environment (Cousin 1980: 15). Following this phenomenological point of view, Cousin underlines the ability of the human being to identify himself with, or project himself onto, volumes and spaces surrounding him. This capacity is inherent in the spatiality of the human body and closely linked to vision, viewed as the main intermediate between space and the body (Cousin 1980: 16). The features of the human body that influence his perception of space are dubbed ‘the bubble’ and ‘the dynamic axis’. The ‘bubble’ is a kind of invisible personal space with a strong physical existence (Cousin 1980: 28; Hall 1971; Moles and Rohmer 1972). Without any confrontation by a specific space, this ‘bubble’ is equivalent to the sphere we occupy when opening our arms and spreading our legs. The ‘bubble’ defines Cousin’s first characteristic of space: its positive or negative nature. Confronted by a particular space – or arrangement of spaces – the ‘bubble’ has the ability to extend its own limits and then to facilitate the identification of the viewer with this space. This kind of space is called positive space, since it is integrated in a limited visual field and is the extension of our ‘bubble’ with a clear boundary. It is closed, all-encompassing, contrary to the space which surrounds it which is negative, outside of our extended bubble. A negative space is essentially undetermined; it is a space without clear boundaries, an open area without a precise visual field. According to Cousin, those properties of space contribute to our intimate perception of architecture (Cousin 1980: 46).

In Practice: the Minoan ‘Palaces’ The ‘palaces’ of Minoan Crete are undoubtedly buildings that performed a central function in their community and they most certainly had a communicative potential. The nature of this message is highly debated, as are the functions of these constructions. Part of the problem is that, in most cases, only the location of the palace itself is well-known and little of the surrounding settlement. With some luck, we have a reasonable idea of their ground plan. (For example, Knossos has very complex architectural phasing, Malia is rather complete, but Phaistos lacks most of its east and south wing). We have hardly any idea, however, about the absolute height of these buildings or about the architectural features of their elevations, such as open corridors, balconies, or windows. It is assumed, however, that the ‘palaces’ were the stage for certain rituals. Following Egyptian parallels, Hägg, for instance, suggests the presence of a so-called ‘window of appearance’ in the West façade, facing the West Court (1987: 129-34), whereas Arthur Evans assumed the presence of a tripartite shrine in the east façade of the West Wing, facing the Central Court (1964: 62, plate XVI). Two miniature frescoes from Knossos, the so-called ‘Sacred Grove and Dance’ fresco and the ‘Grandstand’ fresco, illustrate how courts were actually used by Minoan society (Davis 1987: 157-61) and it may reasonably be assumed that the large open spaces featuring in these frescoes represent the West and Central Courts. We are shown a large crowd of spectators/ viewers and several scenae or stages where several acts are being performed. Add to this the labyrinthine aspect of the ‘palace’ itself, the possible use of a ‘window of appearance’ and astronomical observations involving the use of sun (or moon) light (Goodison 2001: 77-89), and it is clear that the Central and West Courts were important

Secondly, Cousin identified three dynamic axes (Cousin 1980: 36): the vertical axis – an awareness of an above 96

QUENTIN LETESSON stages for Minoan performances. In these performances the viewers’ perception would have been important.

direction to it and to pass around the obstacle. As Cousin observed, this type of space provides a powerful incitement to move, creating a strong dynamic act (Cousin 1980: 97). The raised causeways that are located relatively close to the façade especially at Malia and Knossos most probably have something to do with this movement. Moreover, it is sometimes suggested that they led to the north entrance of the ‘palaces’ by passing round the west façade.

The theoretical approaches just described allow us to study Minoan architectural spaces and especially the open spaces of the ‘palaces’ in two different ways: using Higuchi’s angles of vision, some of their architectural features can be investigated, and, using Cousin’s concepts, we can focus on the influence of ‘palatial’ space on its users. The importance of the courts in the functioning of the Minoan ‘palaces’ has recently been stressed (Driessen 2002: 114; Palyvou 2002: 167-78; Marinatos 1987: 135-43), and these courts constitute an excellent point of view (one can easily look at the ‘palace’ from the courts), contributing, therefore, to the perception of the building itself (Van Effenterre 1987: 85-7; Vansteenhuyse 2002: 235-48). It is perhaps useful to emphasise the main formal features of palatial courts because they strongly influence the perception of their spatiality.

The central court, on the other hand, is a space composed of four parallel planes formed by the wings of the complex and a horizontal plane formed by the court itself. It constitutes a second-degree static and positive space, which means that no single dynamic axis is blocked. Nevertheless, the numerous openings that connect the court with the spatial cells of the wings surrounding it lower its static character. This space is open to the sky, and according to Cousin, such a space without a ceiling, open to the exterior void, is essentially perceived through the architectural structures surrounding it – often walls. Our gaze is instinctively attracted by large free spaces and, as a consequence, the vertical planes become very important. If they were too high, a viewer standing in the court would have been forced to raise his eyes considerably to see the sky, which would have resulted in a strong visual stress (Cousin 1980: 146). In this case, the façades would have been perceived as an enclosing obstacle.

West courts are in general longer (north-south axis) than wider (east-west axis). They are often crossed, especially but not only during the Protopalatial period, by raised causeways that run relatively close to the west façade; court and west façade act together, visually constituting a rectangular element. Central courts have a clear rectangular form, are always longer (north-south axis) than wider (east-west axis), and are surrounded by the different wings of the building, which create a clearly defined space. In most cases there were colonnades or a succession of pillars along one or more sides, supporting balconies.

Higuchi’s thresholds of vision and optimum viewing zone inform us of the visual impact. The results are merely indicative and should be confronted with other archaeological data to actually be meaningful. It is generally assumed that the west façade was the most important one of the ‘palaces’. Van Effenterre (1987), for instance, considered that it had a monumental appearance without developing this idea. By using the visual thresholds and especially the 18 degree angle when an object first takes a monumental nature, we obtain the following height estimates for the west façade, and this from the point of view of a person standing at the west end of the court:

The west court is a horizontal plane: it forms a pre-static and pre-positive area, a space in formation, unstable, of which the character can be modified by the smallest bordering element (Cousin 1980: 89). This kind of space is rarely considered in its own right: boundaries often limit it. Moreover, the texture of the west court generally allows a distinction between this plane and the surrounding spaces (Cousin 1980: 92), mostly because of the paving. If this kind of space has large dimensions, the distinction between the positive and the negative spaces is less clear. Then, if the positive character of this large space had to be expressed, proportionally strong boundaries should have existed. The particular case of the west court and the functions that are generally associated with it plead in favour of it being a positive space (Preziosi 1983: 87; Marinatos 1987: 13543). If this is true, there must have been strong boundaries around it. It is quite possible that buildings existed around the court, but its eastern limit was clearly defined by the most impressive façade of the ‘palace’: the west façade, which was often linked with the court – even functionally. In visual terms, the west façade is an elongated vertical plane (Cousin 1980: 147). The dimensions of this plane were such that it lost its pre-static nature but remained pre-positive. This means that it could have a protective character but allowed a movement to take place and even provoked it. As a matter of fact, close to the façade, the viewer/visitor will automatically have had the tendency to walk in a parallel

‘Palace’

Estimated Width Estimated height of of the the West Facade with West Court Visual angle (18°)

Malia

30 m

10 m +1 m 60

Knossos

44 m

14,6 m + 1 m 60

Phaistos

42 m

14 m + 1m 60

TABLE 1 ESTIMATED HEIGHT OF THE WEST FACADE WITH VISUAL ANGLE (18°), FROM THE WEST COURT Estimated Width = Minimum preserved width of preservation of the west court. Visual angle = these measures represent the supposed height of the facade if it is considered as given a monumental feeling from the end of the west court. 1m60 is added to each result as an average height for the viewers. 97

SOMA 2004 ‘Palace’

Malia A B C Knossos A B C Phaistos A B C

Distance between point of vision and West Façade

Estimated height of the Window of Appearance with Visual angle (18°)

Estimated height of the Window of Appearance with Visual angle (27°)

Estimated height of the Window of Appearance with Visual angle (45°)

10 m 7,5 m 5m

3,3 m + 1 m 60 2,5 m+ 1 m 60 1,6 m+ 1 m 60

5 m+ 1 m 60 3,75 m+ 1 m 60 2,5 m+ 1 m 60

10 m+ 1 m 60 7,5 m+ 1 m 60 5 m+ 1 m 60

2m 5m 2m

0,7 m+ 1 m 60 1,7 m+ 1 m 60 0,7 m+ 1 m 60

1 m+ 1 m 60 2,5 m+ 1 m 60 1 m+ 1 m 60

2 m+ 1 m 60 5 m+ 1 m 60 2 m+ 1 m 60

10 m 5m 16 m

3,3 m+ 1 m 60 1,7 m+ 1 m 60 5,3 m+ 1 m 60

5 m+ 1 m 60 2,5 m+ 1 m 60 8 m+ 1 m 60

10 m+ 1 m 60 5 m+ 1 m 60 16 m+ 1 m 60

TABLE 2 ESTIMATED HEIGHT OF THE WINDOW OF APPEARANCE WITH VISUAL ANGLE (18°-27°-45°), FROM THE WEST COURT. A, B and C are selected viewpoints on the raised causeways These estimates indicate that a height at least equivalent to a third of the width of the west court was necessary for the west façade to have a monumental impact for the viewers standing at the western side of the court. This is, according to present evidence, extremely unlikely.

at Knossos and between 4,1m and 9,6m at Phaistos. It will have created a visually uncomfortable appearance (at 45°) between 6,6m and 11,6m at Malia, between 3,6m and 6,6m at Knossos and between 6,6m and 17,6m at Phaistos. In a forthcoming paper (Letesson and Vansteenhuyse forthcoming), the position of the ‘window of appearance’ is further examined using the visual thresholds in order to improve our understanding of the west façade.

Although the precise function of the raised causeways is not known, it is assumed that they were a kind of processional way, or ‘red carpet’ leading to the entrance of the building (Marinatos 1987: 138). They could, of course, reinforce the incitement to move, created by the elongated vertical plane of the west façade. It is interesting then to test Märtens’s thresholds of vision from different points on the raised causeways in connection with the west façade (See Table 2).

The results of the visual thresholds applied to the west court benefit when considered in relation to those of the central court, so let us first look at the thresholds with a viewing point in the middle of the central court in relation to the west wing (see Table 3). It appears that the data for a visual angle of 45 degrees are in general equal to those given by the lowest threshold (18 degrees) from the west side of the west court. It may be interesting in this regard to take into account the hypothesis that some palaces were once U-shaped and open to the south (MacDonald 2002: 35-55). In Knossos, the principal entrance to the Central Court always seems to have been from the south anyway. Entering the Central Court from the south would, however, immediately focus the attention

The results suggest that, for a viewer standing on the raised walks (on points A, B, C), the west façade would have first (at 18°) given a monumental feeling if it attained a height between 3,2m and 5,9m at Malia, between 2,3m and 3,3m at Knossos and between 3,3m and 6,9m at Phaistos. Using the second threshold (at 27°), the façade would have occupied the whole visual field when attaining a height between 4,1m and 6,6m at Malia, between 2,6m and 4,1m ‘Palace’

Estimated height of Estimated height of Estimated height of the the West Wing with the West Wing with West Wing with Visual Visual angle (18°) Visual angle (27°) angle (45°)

Malia

Distance from the middle of the Central Court to the West Wing 13,5 m

4,25 m + 1 m 60

6,75 m + 1 m 60

13,5 m + 1 m 60

Knossos Phaistos

14 m 11,15 m

4,5 m + 1 m 60 3,90 m + 1 m 60

7 m + 1 m 60 5,57 m + 1 m 60

14 m + 1 m 60 11,15 m + 1 m 60

TABLE 3 ESTIMATED HEIGHT OF THE WEST WING WITH VISUAL ANGLE (18°-27°-45°), FROM THE CENTRAL COURT. 98

QUENTIN LETESSON ‘Palace’

Length of court

Malia Knossos Phaistos

48 m 52 m 46,5 m

Estimated height of North Façade with Visual angle (18°) 16 m + 1 m 60 17,3 m + 1 m 60 15,5 m + 1 m 60

Estimated height of North Façade with Visual angle (27°) 24 m + 1 m 60 26 m + 1 m 60 23,2 m + 1 m 60

Estimated height of North Façade with Visual angle (45°) 48 m + 1 m 60 52 m + 1 m 60 46,5 m + 1 m 60

TABLE 4 ESTIMATED HEIGHT OF NORTH FAÇADE WITH VISUAL ANGLE (18°-27°-45°), FROM THE CENTRAL COURT on the north wing. We may then try to estimate the height of this wing for a viewer standing at the southern side of the Central Court and to compare the results with the heights obtained for the West Court (See Table 4).

Conclusions Obviously, the results of these different visual analyses do not provide irrefutable evidence on the height of the elevation of the different ‘palaces’ nor on the precise visual impact they had on their users. Nevertheless, it appears that the ‘palaces’ probably did not achieve a monumental aspect through their verticality or elevation but rather through their horizontality and the length of the façades (Palyvou 2000) as well as their specific location and accessibility in the city.

It is immediately obvious that the north façade of the Central Court was not monumental because an elevation with a height close to 20 meters is too excessive. Furthermore, as will be shown when considering the optimum viewing zone, such a height would have a negative effect in terms of impact of space through vision on the users of the Central Court.

These results open approaches for future studies. Since visual impact is one of the most important features of a monumental building (Driessen 1999: 123), we can eventually hope to reconstruct the essential nature of the buildings traditionally called ‘palaces’ by adopting, as much as possible, a Minoan point of view, by trying to reconstruct what Bronze Age Cretans may have experienced and, in this way, gradually reconstruct an archaeology of performance.

The central court is unanimously considered as the diagnostic feature of the ‘palaces’. On the basis of iconography and the spaces surrounding it, it is generally assumed to have had a ritual function. If this space was heavily frequented and was the theatre for important ritual activities, it is not likely to have been visually oppressive: too high surrounding walls would contribute to install an uneasy feeling for people standing in the court. Applying Higuchi’s optimum viewing zone and Cousin’s ellipse of vision to the wings for a viewer standing in the middle of the central court, we obtain the following results:

‘Palace’ Malia Knossos Phaistos Zakros

Estimated Height of East and West wings +/- 8,30 m + 1 m 60 +/- 8,30 m + 1 m 60 +/- 7 m + 1 m 60 +/- 4,50 m + 1 m 60

Estimated Height of North and South wings +/- 11,25 m + 1 m 60 +/- 12,5 m + 1 m 60 +/- 11 m + 1 m 60 +/- 8 m + 1 m 60

TABLE 5 OPTIMUM VIEWING ZONE FROM THE CENTRAL COURT ‘Palace’

Estimated Height of East and West wings

Estimated Height of North and South wings

Malia Knossos Phaistos Zakros

+/- 3,4 m + 1 m 60 +/- 3,6 m + 1 m 60 +/- 3 m + 1 m 60 +/- 1,6 m + 1 m 60

+/- 6,4 m + 1 m 60 +/- 7 m + 1 m 60 +/- 6,2 m + 1 m 60 +/- 4 m + 1 m 60

TABLE 6 ELLIPSE OF VISION FROM THE CENTRAL COURT Heights constitute maxima above which the situation would be visually uncomfortable. (All tables are the work of the author) 99

SOMA 2004 Letesson, Q. and K. Vansteenhutse, (forthcoming). An Archaeology of Perception: Experiencing Minoan Court Complexes. Aegean Archaeology. Lynch, K. (1960) The Image of the City, Cambridge, Cambridge University Press. MacDonald, C. (2002) The Neopalatial Palace of Knossos. IN: Driessen, J., Schoep, I. and R. Laffineur, eds. Monuments of Minos. Rethinking the Minoan Palaces Liège-Austin, AEGAEUM. Annales d’archéologie égéenne de l’Université de Liège et UT-PASPXXIII, 35-55. Marinatos, N. (1987) Public Festivals in the West Courts of the Palaces. IN : Hägg, R. and N. Marinatos, eds. The Function of the Minoan Palaces, Stockholm, Skrifter utgivna av Svenska Institutet i Athen, 135-43. Moles, A. and E. Rohmer (1972) Psychologie de l’espace, Paris, Casterman. Moore, J.D. (1996) Architecture and Power in the Ancient Andes. The Archaeology of Public Buildings, Cambridge, Cambridge University Press. Palyvou, C. (1997) Session on the Functional Analysis of Architecture. IN: Hägg, R. ed., The Function of the Minoan Villa, Stockholm, Skrifter utgivna av Svenska Institutet i Athen, 155-6. Palyvou, C. (2000) Concepts of Space in Aegean Bronze Age Art and Architecture, IN: Sherratt, S. ed. The Wall Paintings of Thera. Proceedings of the First International Symposium: Athens: Thera Foundation – Petros M. Nomikos and the Thera Foundation, 41335. Palyvou, C. (2002) Central Courts: The Supremacy of the Void. IN: Driessen, J., Schoep, I. and R. Laffineur, eds. Monuments of Minos. Rethinking the Minoan Palaces AEGAEUM. Liège-Austin, Annales d’archéologie égéenne de l’Université de Liège et UT-PASP XXIII, 167-78. Preziosi, D. (1983) Minoan Architectural Design, New York, Mouton Publishers. Sanders, D.H.(1988) Architecture - the Neglected Artifact. IN: French, E.B. and K.A.Wardle, eds., Problems in Greek Prehistory, Bristol, 489-99. Trigger, B.G. (1990) Monumental Architecture: A Thermodynamic Explanation of Symbolic Behaviour. World Archaeology, 22 (2), 119-32. Van Effenterre, H. (1987) The Function of Monumentality in the Minoan Palaces. IN: Hägg, R. and N. Marinatos, eds., The Function of the Minoan Palaces, Stockholm, Skrifter utgivna av Svenska Institutet i Athen, 85-7. Vansteenhuyse, K. (2002) Minoan Courts and Ritual Competition. IN: Driessen, J., Schoep, I. and R. Laffineur, eds. Monuments of Minos. Rethinking the Minoan Palaces, Liège-Austin, AEGAEUM. Annales d’archéologie égéenne de l’Université de Liège et UTPASP XXIII, 235-45.

Acknowledgements This paper forms part of the project ‘A Topography of Power in Minoan Crete’ based at the Université Catholique de Louvain (UCL). I would like to thank Prof. Jan Driessen and Klaas Vansteenhuyse for their help, useful comments and corrections, as well as Charlotte Langohr and Peter Tomkins for their stimulating discussions. All mistakes obviously remain the responsibility of the author. Bibliography Cousin, J. (1980) L’espace vivant. Introduction à l’espace architectural premier, Paris, Editions du Moniteur. Davis, E.N. (1987) The Knossos Miniature Frescoes and the Function of the Central Courts, IN: R. Hägg and N. Marinatos, eds., The Function of the Minoan Palaces, Stockholm, Skrifter utgivna av Svenska Institutet i Athen. 157-61. Driessen, J. (1999) ‘The Archaeology of a Dream’: The Reconstruction of Minoan Public Architecture. Journal of Mediterranean Archaeology, 12.1, 121-7. Driessen, J. (2002) ‘The King Must Die.’ Some Observations on the Use of Minoan Court Compounds, IN: Driessen, J., Schoep, I. and R. Laffineur, eds., Monuments of Minos. Rethinking the Minoan Palaces Liège-Austin, AEGAEUM . Annales d’archéologie égéenne de l’Université de Liège et UT-PASP XXIII, 1-14. Driessen, J., Schoep, I. and R. Laffineur eds. (2002) Monuments of Minos. Rethinking the Minoan Palaces Liège-Austin, AEGAEUM. Annales d’archéologie égéenne de l’Université de Liège et UT-PASP XXIII. Evans, A. (1964) The Palace of Minos, London, Hafner Publishing Co. Goodison, L. (2001) From Tholos to Throne Room: Perceptions of the Sun in Minoan Ritual. IN: Laffineur, R. and R. Hägg eds., Potnia. Deities and Religion in the Aegean Bronze Age, Liège-Austin, AEGAEUM. Annales d’archéologie égéenne de l’Université de Liège et UT-PASP XXII, 77-89. Hägg, R. (1987) On the reconstruction of the west façade of the palace at Knossos. IN: Hägg, R. and N. Marinatos, eds., The Function of the Minoan Palaces, Stockholm, Skrifter utgivna av Svenska Institutet i Athen., 12934. Hall, E.T. (1971) La dimension cachée, Paris, Editions du Seuil, Higuchi, T. (1983) The Visual and Spatial Structure of Landscape, Cambridge, Cambridge University Press. Klynne, A. (1998) Reconstructions of Knossos: Artists’ Impressions, Archaeological Evidence and Wishful Thinking. Journal of Mediterranean Archaeology, 11.2, 206-29.

100

Terracotta Human Figures and Figurines from the LBA Cyclades (The Cases of Ayia Irini and Phylakopi): Art, Religion, and Politics Dionysia Marinou

desired. A three-dimensional quality would have been more difficult and costly to achieve using marble, while clay would have been repaired easier and faster.

Introduction Traditional methods for determining the nature of the relationship between the Cyclades and their neighbours in the Late Bronze Age period often include two main schools of thought: either complete political control of the islands by Minoan Crete, and later by Mycenaean Greece, involving the actual presence of Minoan/Mycenaean settlers and/or consuls on the islands, or explanation of the cultural traits shared by the sites in question in terms of trading connections, movement of artists, and cultural exchange. In this debate the human figures and figurines from Kea and Melos have repeatedly been used as evidence to argue for the ‘Minoanisation’ of Ayia Irini and ‘Mycenaeanisation’ of Phylakopi, as elements known from Crete and the mainland can be recognised at these sites by their general appearance. I believe, however, that we need to examine the figures from two completely different angles, considering their artistic values and their meaning separately. This is because, as I will try to prove, there is a possibility that even if their style follows the Minoan or Mycenaean ‘fashion’ of the time, they continue, at least to a degree, to express the beliefs of Cycladic people.

Regarding the technique employed for the making of the Late Bronze Age examples, it is essentially different from that of the Early Bronze Age period. The figures are still hand-made, but they are now hollow instead of solid, built over an internal wooden structure with grass strings attached. The fact that hollow examples are also known from Crete even from earlier contexts may be used as an argument for Minoan influence, but its significance is rather dubious. Hollow finds are indeed known from Palatial Crete, for example, the ‘Goddess of Myrtos’ (Warren 1973), but their form is very limited. They comprise vessels and, with the exception of some human limbs, actual hollow human figures have not yet been found. Besides, there is no evidence in the Cretan examples for the use of wooden rods for inner support, as neither traces of burnt organic material, nor air-holes for the escape of gases have yet been identified. As far as the find context is concerned, the ‘Keros Hoard’ (Broodbank 2000: 229-30) may be an indication that Early Cycladic marble figurines were used in cultic as well as funerary settings, and in this respect the Late Bronze Age temple might have replaced an earlier type of shrine, where the figurines were engaged in a similar way. Besides, some concepts behind the Keian figures were perhaps not unknown to earlier Cycladic inhabitants. The importance of music and dance, for example, the Early Cycladic harp player, and the exposed breasts, perhaps emphasising female sexuality and procreative potential, are aspects known from both Early and Late Bronze Age periods. Another, rather controversial similarity is the bent knees, which according to Caskey could be a posture associated with dancing (1986). In my opinion, this may at the same time be an expression of salutation, respect, or obedience, in a context of prayer or adoration, deliberately chosen to depict both concepts. Bowing, or ‘bending the knee’, is still perceived as a token of reverence, respect, or submission.

Case 1: Ayia Irini, on Kea In the case of Ayia Irini fragments of at least fifty-six clay figures were recovered from the rooms of a building which, according to the excavator, was possibly a temple (Caskey 1971: 292-395). Their chronology is not without problems, but it seems likely that they are of a MM III - LM IB date, except one that is of LH III date (Caskey 1986: 32-5). They are essentially alike, representing upright females standing or dancing, dressed in the Minoan costume and ranging in size from 70 to 135 cm. It is very interesting that the white paint used for the flesh of the figures was derived from marble sources. I wonder whether this may be a reflection of the possible significance attached to this material in the Early Bronze Age, and an indication that it partly survived into the Late Bronze Age. Although at this later stage marble has been replaced by clay, we have to question whether this change was not so much determined by imitation of Minoan practice, but rather was dictated by the results 101

SOMA 2004 that these finds have local antecedents even though there is a great chronological gap between them. On the other hand, some of these figures, for example the group with thin, square-like torsos and prominent genitalia, compare closely with late Cretan examples (Barber 1987: 242, 246). An impressive range of Post-Palatial male figures are known from the Perati tombs and the Ayia Triadha Piazzale dei Sacelli, while some Late Minoan examples, for example the small figurine carrying a dove, from the shrine of the Double Axes at Knossos, and the standing ‘warrior’ from the villa at Kannia, near Gortyna, might have been copied from the Phylakopi figures. Yet, only a couple of fragmentary male figurines are known from the mainland, from the Acropolis and Tiryns, and except for the possible male drawings in the Room of Frescoes at Mycenae, there are no clear indigenous male representations with religious connotations in that period from mainland Greece, where even the bronze smiting gods from Mycenae and Tiryns have been imported.

Case 2: Phylakopi, on Melos Turning to the numerous human figures and figurines from the sanctuary of Phylakopi (French 1985: 209-80) we cannot fail to notice that they comprise both imported and local products. Their individuality makes it quite difficult to suggest a date of manufacture for all of them, but they seem to range from LH IIIA to IIIC, as they pre-date the LH IIIC, phase 2b collapse of the West Shrine. The small female figurines found at Phylakopi are of wellknown Mycenaean types, while the large female figures fall into four categories, three of which are also known from elsewhere (French 1985: 210). The first group resembles the Minoan ‘bell-skirt’ type, but no associated paraphernalia such as snakes, horns of consecration or double axes were found at Phylakopi. On the other hand, in Crete, human and animal figurines have not been found together with the Goddess with Upraised Arms. The second group includes figures with tall cylindrical or conical stem topped by a rounded or pinched body and an elaborate head, such as the Lady of Phylakopi. The face seems to have affinities with heads from Mycenae and Asine, the decoration recalls the Curved Stripe motive and the Wavy Border known from LH III pottery, while the zigzag pattern in the horizontal form is common in Crete, for example at Knossos. The third Phylakopi group consists of pots on which arms have been attached and a head placed on the neck of the pot, as known earlier from both Crete and the mainland.

Suggestions Our impression of cult in LBA III at Phylakopi is very different from that in LBA I at Ayia Irini, and therefore it would be very interesting to look for the reasons behind these differences as documented by the finds at these two sites. The first question to emerge is whether there is enough evidence to associate the cult practiced in the temple at Ayia Irini and the shrines at Phylakopi with what is known from Minoan Crete and the Mycenaean mainland respectively? Although the Keian figures are dressed in the Minoan costume and even though there are some indications that large figures might have been produced in Crete, no complete examples have yet been found, let alone in the size and quantity found at Ayia Irini. Moreover, accessories associated with Cretan figures, such as the tiara or crown, plastic snakes or birds, discs or poppies, are not depicted on the figures from Ayia Irini. Most importantly, double axes and horns of consecration are not found together with the Keian figures, even though some other objects, also known from the Cretan sanctuaries, such as clay bovids, animal rhytons, kernoi, clay dolphins, stone libation tables, clay and bronze boats, were recovered from the temple (Caskey 1981: 132). Finally, the benches against the walls of the temple might recall the platforms from Middle Minoan shrines, but such an independent building accommodating life-size human figures seems like a Cycladic invention (Renfrew 1985).

A unique figure from group four is made of local clay, poorly baked, and preserves some traces of paint. It was found beside the Lady of Phylakopi and it is thought that the inverted ‘U’ on the skirt is an explicit sexual attribution. This find is without known parallels. It is a rather crude and poorly baked figure, probably older than the rest of the Phylakopi group, but its simultaneous usage alongside highly stylised figures such as the ‘Lady of Phylakopi’ might indicate its great value and significant meaning for the people of Phylakopi. It might have been placed next to the latter to emphasize its sex, as suggested by French (1985: 215-6), but it is also possible that it was venerated as ancient, as the holy cult image of a deity worshipped by local people before the introduction of the sophisticated, but otherwise imported ‘Lady of Phylakopi’. Thus, it may have represented the views and religious beliefs of an earlier society. It is uncertain whether the two figures depict the same deity, or express different ideas of Cycladic and Mycenaean cult respectively, but in either case it is important that a unique, non-Mycenaean-influenced figure, and thus indigenous Melian ideas, seem to have survived down to LH III times at Phylakopi.

It would, therefore, be unfounded to suggest that the rituals practised in the temple were the same as those practised in Minoan sanctuaries, for basic Minoan cult equipment is missing from Ayia Irini. On the same basis, the absence of any indication of double axes, horns of consecration or snakes at the sanctuary of Phylakopi is noteworthy, even though the bell-skirt group of figures, some decoration patterns on the ‘Lady of Phylakopi’, and perhaps the baetylic stone found just outside the shrines may argue for some Cretan influence. The later finds from Phylakopi seem to have rather Mycenaean affinities, with the general

A further innovation on Melos seems to be the relative frequency with which male figurines appear, almost unparalleled in the Mycenaean world. Certain features, such as the polos cap and the dagger, seen on some Phylakopi examples, are well-known attributes of Early Cycladic origin, as seen for example on the Early Cycladic ‘hunter-warrior’ type. I believe, therefore, that it is possible 102

DIONYSIA MARINOU layout of the shrines being comparable to mainland examples and most of the female figures and figurines following well-known Mycenaean types. At Ayia Irini, on the other hand, Mycenaean influence is not so strong in the later stages of LBA III, even though Mycenaean figurines have been recovered from the site (but not from the temple) and since the large figure constructed in that period follows closely the tradition established in LM I-II. It seems thus that the differences noticed between the two sites are not only of a temporal nature, but they may also indicate a regional, as well as social diversity. Firstly, at Ayia Irini no one figure stands out as the representation of a deity among the unusually large number of figures found, while at Phylakopi at least one deity, if not more, has been proposed. The possibly upraised arms on the ‘Lady of Phylakopi’ have been interpreted as a gesture of sanctity not only here, but also on LM IIIB Cretan figures, which makes this figure a valid candidate for a cult image. Moreover, the largest of the male figures might also be viewed as a cult image, or as a votary, perhaps highlighting the importance of a male god, fitting in with the evidence from at least Crete that male figures were being produced towards the end of the Bronze Age.

weaving (?), equipped with platforms or benches perhaps for the display of certain cult objects, accommodating large human figures, and including similar cult equipment, e.g., eating and drinking vessels, such as kernoi and kylikes, and votive offerings, such as animal figurines and bronze human figurines, all seem to be aspects of a common pattern of ritual expression. Renfrew has argued that this degree of community in expression, which, however, does not prove uniformity, is linked with the process of ‘peer polity interaction’ (1982: 286). Such a hypothesis is, of course, difficult to prove, but it is likely that frequent interactions between people living so close to each other, not only in the Bronze Age but even much earlier, would have led to a sharing of certain artefacts and to a number of common modes of expression. I believe this is precisely the reason for the wide distribution of Minoan-inspired-symbols, e.g., double axes and horns of consecration, in the earlier Late Bronze Age period and Mycenaean-inspired-material, e.g., human and animal figurines, in the later stages of Late Bronze Age. Thus, this might imply that there were ‘Aegean rites and practices’ (Hägg 1981: 39) shared by corresponding cults in Crete, the mainland, and the Cyclades.

It is thus remarkable that on Melos the divinity might have been represented by a female and/or a male figure, while at Kea its representation is possibly aniconic. Of course, the baetyl found at the doorstep of the shrines at Phylakopi might also argue for the worship of a deity not depicted in human form at all. Whether such a deity is identical with the one worshipped at Ayia Irini is of course difficult to prove. Bearing in mind that the baetylic stone is often featured in Minoan settings and that at least some Cretan influence is certainly shown on a number of the Keian features, a common Minoan derivation seems possible, even though it is uncertain whether it has retained the precise meanings associated with it when used in Crete. If this item falls into the category of random imported prestige objects and it does not imply the local acknowledgement of Minoan symbols, then we could be seriously misled. Overall though, the two assemblages are diametrically different and what we witness is most probably a distinction between official and popular cult, as already discussed by Hägg (1981: 35-9) for Mycenaean Greece.

Conclusion In conclusion, I believe that although there seems to have been a sufficient degree of interaction between the Cyclades and the rest of the Aegean world, under whatever political circumstances, to expect a degree of community in religious symbolism and practice, this does not necessarily prove that the same deities were worshipped everywhere, or that similar symbols had identically ascribed meanings in each place. Of course, it may imply that some modes of expression and symbolism were common in Crete, the Cyclades, and the mainland, as a result of either a Minoan/ Mycenaean empire or simply because of commercial contact, but we must remember that religious beliefs have undergone a series of transformations over time, as well as that local elements have obviously affected local art, and apparently cult. Bibliography

It is notable, however, that even if we accept that there is a regional diversity in the use of symbols and thus a pattern of different beliefs from site to site, we cannot fail to recognise that there is some basic similarity in the ways beliefs are expressed. It may be true that the Ayia Irini figures were of Keian origin and they simply adopted certain Minoan features, while the Phylakopi finds were rather of Mycenaean derivation with merely local features, but in both cases external influence is an undisputable factor, no matter where it lies or under what circumstances it was developed. Similarly in both cases independent buildings, set apart from administration centres or open-air places but perhaps being in close proximity to economic/ industrial activity, e.g., evidence for metal-casting and

Barber, R.L.N. (1987) The Cyclades in the Bronze Age. London, Duckworth. Broodbank, C. (2000) An Island Archaeology of the Early Cyclades. Cambridge, Cambridge University Press. Caskey, J.L. (1971) Investigations in Keos: Part I, Hesperia 40, 292-395. Caskey, M.E. (1981) ‘Ayia Irini, Kea: The Terracotta Statues and the Cult in the Temple’, IN: Hägg R. and N. Marinatos eds. Sanctuaries and Cults in the Aegean Bronze Age Stockholm, Paul Åströms Förlag, 127-36. Caskey, M.E. (1986) Keos II: The Temple at Ayia Irini. Part I: The Statues. Princeton, The American School of Classical Studies. 103

SOMA 2004 French, E.B. (1985) ‘The Figures and Figurines’, IN: Renfrew, C. ed. The Archaeology of Cult: The Sanctuary at Phylakopi, London, Thames & Hudson, 209-80. Hägg, R. (1981) Official and Popular Cults in Mycenaean Greece, IN: Hägg, R. and N. Marinatos eds. Sanctuaries and Cults in the Aegean Bronze Age, Stockholm, Paul Åströms Förlag, 35-9. Renfrew, C. (1982) Polity and Power: Interaction, Intensification and Exploitation, IN: Renfrew, C. and

M. Wagstaff eds. An Island Polity: The Archaeology of Exploitation in Melos, Cambridge, Cambridge University Press, 264-90. Renfrew, C. (1985) The Archaeology of Cult: The Sanctuary at Phylakopi. London, Thames & Hudson. Warren, P.M. (1973) The Beginnings of Minoan Religion, IN: Rizza, G. ed. Antichita Cretesi Studi In Onore Di Doro Levi, tania, Università di Catania, Istituto di archeologia, 137-47.

104

The Myth of the Synagogue on Delos Lidia Matassa

Introduction The issue of the identification of a synagogue on Delos has been problematic since it was first suggested. While there is some literary evidence relating to Jews on Delos, not a single piece of it refers to the existence of a synagogue. Indeed, other than a few passing references to the presence of Jews on the island, there is no material relating to any aspect of their lives or how they worshipped. Therefore, when we come to look at how a building on the island was identified as a synagogue, we find a substantial gap between what has been proposed—and widely accepted—and what has actually been found. I am going to look at how the identification of a synagogue was made, and at the literary, archaeological and epigraphical evidence suggesting the presence of the Jewish community on the island, and I will begin by giving a brief history of Delos.

In 69 BCE, Gaius Triarius, the Roman Legate, repaired some of the damage caused by the destructions and built a wall round the town centre of Delos (Phlegon of Tralles, FGrHist 257, fr. 12.13). In 66 BCE, Cicero lectured the Roman Senate on how the friends, allies and subjects of Rome had been at the mercy of pirates until Pompey drove them away (Cicero, Leg. Man. 31-5; 54-7; de Souza 1999: 162-3) By the mid-first century BCE, Delos was in its death throes. Trade routes had altered so as to avoid these dangers leaving Delos increasingly outside the commercial loop until, eventually, Athens stopped bothering to send a governor to the island. In the second century CE, a last-ditch attempt by the philhellenic emperor Hadrian to revive the old Delian festivals was unsuccessful (McLean 1996: 189), and by then, according to Pausanias (8.33.2), the island was very sparsely inhabited. Delos was finally abandoned around the fifth century CE (Brunet 1990: 669-82).

Delos is a small island in the centre of the Cycladic Islands off mainland Greece. The mythological birthplace of Apollo and Artemis, it was a major cult centre by the seventh century BCE and was mentioned in Homer and in a Homeric Hymn (Homer, Odyssey 6.160-9. See Crudden, Hymn 3 to Apollo 2001: 23-42).

The Excavations The Ecole française d’Athènes commenced excavations on Delos in 1873. Between 1904 and 1914 they excavated much of the island. There were further extensive excavations between 1958 and 1975 and the French School continue to run excavations on the island in conjunction with the Cycladic Ephoreia, and maintain a permanent presence there.

According to Thucydides (Peloponnesian Wars, 1.96.2; 6.76.3), during incursions into mainland Greece in the first quarter of the fifth century BCE, the Persian king Xerxes sacked and burned the Athenian sanctuaries, prompting the formation of an alliance between Greek city states in 478 BCE. Delos was chosen to be the treasury of this alliance, which was known as the Delian League. By the end of the third century BCE, a continuous stream of foreigners began to arrive on Delos from all over the Aegean, many of them establishing their ancestral cults and associations on the island (McLean 1996: 189).

I will refer to all built structures hereafter according to their designations in the seminal guide to the excavations on Delos, the Guide de Délos (Bruneau and Ducat 1983). Thus, the building known to many as the Delian synagogue is GD80. Likewise, I will refer to all inscriptions from Delos according to their designations in the Inscriptions de Délos (Durrbach 1926; Durrbach 1929; Durrbach and Roussel 1935; Roussel and Launey 1937; Plassart 1950). Thus, inscription 1 is referred to as ID2329, and so forth.

A thriving and strategically desirable maritime trade centre in the centre of the Aegean, Delos was often caught between warring factions, and throughout the three Mithridatic wars (88-84 BCE, 83-81 BCE, 74-63 BCE) there were a series of devastating attacks on the island from which it never fully recovered (de Souza 1999: 162-3).

During the excavations of 1912 and 1913, André Plassart, of the Ecole française d’Athènes identified building GD80 105

SOMA 2004

FIGURE 1 MAP OF DELOS AND AREA LOCAL TO GD80 (AFTER BRUNEAU AND DUCAT 1983: PLAN III AND P.206) DRAWING: LIDIA MATASSA as a synagogue (1914: 529) Plassart relied on six Greek inscriptions, the principal one having been found some 90m north of GD80 (Fig. 1). This inscription contained the name ΛΥΣΙΜΑΧΟΣ (Lysimachos) and the word ΠΡΟΣΕΥΧΗΙ (proseuche) which, Plassart said, referred to a Jewish ‘house of prayer’ or ‘synagogue’ (1914: 528). The other five inscriptions were found scattered within the two main spaces of GD80 (Fig. 2), some lying on benches, others at the foot of walls, as well as on the

internal configuration and furnishings of the structure he had uncovered. This structure had marble benches running around the walls and a marble throne facing east such that, according to Plassart, it was similar to the configuration of later synagogues (Plassart 1914: 529). Among the inscriptions Plassart found in GD80 was one which contained one of the donor names found in the proseuche inscription. Three of the inscriptions contained the epithet ΘΕΩΙ ΎΨΙΤΩΙ (Theo Hypsisto), and one contained ΎΨΙΤΩΙ (Hypsisto). Plassart’s final inscription retained only two legible words, ΓΕΝΌΜΕΝΟΣ ΕΛΕΥΘΕΡΟΣ (genomenos eleutheros) (‘… became free’) (Plassart 1914: 528). André Plassart argued that the use of the epithets Hypsisto and Theo Hypsisto indicated a tendency towards monotheism and that they therefore—in conjunction with the proseuche inscription—referred to the Jewish deity (1914: 529). He chose to associate the Proseuche and Lysimachos inscriptions with one another, considered the use of Theos Hypsistos and Hypsistos in the other inscriptions to be Jewish, looked at the configuration of the furnishings of the building, and declared it to be a synagogue (1914: 529). I will show that the word proseuche in the context in which Plassart found it refers to the fulfilment of a prayer or votive offering, and not to a building and, indeed, not even necessarily to a Jewish context. I will demonstrate that the

FIGURE 2 PLAN OF GD80 (AFTER BRUNEAU 1970: PLANCHE B) DRAWING: LIDIA MATASSA 106

LIDIA MATASSA reoccurrence of the name Lysimachos is coincidental and that the inscriptions using the epithets Theos Hypsistos and Hypsistos refer either to the Greek deity Zeus Hypsistos (whose sanctuary on Delos is located on Mount Cynthus not far from GD80), or even to the Theos Hypsistos of the Hypsistarian cult, and not to the God Most High of Judaism.

worthy deeds on our behalf and their goodwill towards us.

From this text, we can hypothesise that the Jews on Delos were for some time not permitted the same privileges as Jews in Rome. This prohibition relates to the period when GD80 is said to have functioned as a synagogue, that is, the mid-first century BCE. Of course, it may be that at some point after this, the Jews on Delos (and elsewhere) were allowed to live in accordance with their customs, but we only have the edict as recorded in Josephus. Moreover, the reference to ‘Jews on Delos and some of the neighbouring Jews’ could indicate that the neighbouring Jews may have been on other islands, either in the Cyclades or the Dodecanese, or even Aegina, Crete, Rhodes or Cyprus; not necessarily on Delos itself.

I will also demonstrate that the arguments relating to the form, style, furnishings and artefacts found in GD80 are not relevant to its identification as a synagogue. In short, I hope I will convince you that there are no sound reasons to continue to consider GD80 a synagogue. The Literary Evidence The earliest reference to Jews on Delos is found in 1 Maccabees 15.15-23 and incorporates a letter from Lucius, a Roman consul. In this passage, the Jews, during the reign of the High Priest Simon (143-135 BCE), made an offering to the Romans of a gold shield. The Romans responded, renewing their alliance with the Jews with a letter announcing Rome’s continuing support of the Jews on Delos and elsewhere. Delos is mentioned only in passing, and no specifics are given. There is some debate about the chronology of this passage, but it is not relevant to this discussion.

Plassart cited the passage in Josephus (AJ 14.213-216) as “undertaking to repeal the decree in which the Jews had been forbidden from organising communal meals that would have taken place in the vast premises of the synagogue’ (1913: 201-15). However, as is clear from the text, there is nothing in the text to suggest that GD80 had ever been the location of a synagogue. There is nothing, in fact, to connect GD80 with a reading of this text other than Plassart’s identification of the structure as a synagogue’which was based on his interpretation of the inscriptions he found as being Jewish.

The second piece of text is another account of the same incident as above found in Josephus (AJ, XIV.145-148), in which Delos is not specifically mentioned at all.

The Inscriptions As I have said, Plassart’s justification for identifying GD80 as a synagogue consisted of six inscriptions. The principal one, around which Plassart’s entire argument is based, was not even found in GD80, but in house IIA of GD79 some 90m north of GD80 at the southeastern side of the stadium district in a residential area (Fig 1).

The third and most specific text comes to us via Josephus and is dated to around the middle of the first century BCE (AJ 14.213-16): Julius Gaius, Praetor, Consul of the Romans, to the magistrates, council and people of Parium, greeting. The Jews in Delos and some of the neighbouring Jews, some of your envoys also being present, have appealed to me and declared that you are preventing them by statute from observing their national customs and sacred rites. Now it displeases me that such statutes should be made against our friends and allies and that they should be forbidden to live in accordance with their customs and to contribute money to common meals and sacred rites, for this they are not forbidden to do even in Rome. For example, Gaius Caesar, our consular praetor, by edict forbade religious societies to assemble in the city, but these people alone he did not forbid to do so or to collect contributions or to hold common meals. Similarly do I forbid other religious societies but permit these people alone to assemble and feast in accordance with their native customs and ordinances. And if you have made any statutes against our friends and allies, you will do well to revoke them because of their

Inscription 1 (ID2329) (author’s translation) ΑΓΑΘΟΚΛΗΣ ΚΑΙ ΛΥΣΙΜΑΧΟΣ ΕΠΙ ΠΡΟΣΕΥΧΗΙ ‘Agathokles and Lysimachos for an offering/prayer’. Discovered in House 79 of the residential area abutting the stadium (Fig. 1), this inscription dates to around the first century BCE and is carved on an undecorated plain marble stele with a lead fixing on the top, indicating it held a statue or votive offering, definitely not a Jewish custom. In interpreting the meaning of the inscription, Plassart said the word proseuche referred specifically to a synagogue building (1913: 201-15). Following Plassart most scholars have agreed with this interpretation. Nevertheless, the presence of the lead fixing is convincing support for the argument that this inscription base was a votive or dedication to a Greek or other pagan deity and must certainly be translated as ‘for an offering/prayer’, and not ‘for the synagogue’. On the basis of his supposition that this inscription indicated a synagogue, Plassart identified the two 107

SOMA 2004 names in this inscription as Jewish (1913: 201-15). As a consequence of this, the names Agathokles and Lysimachos on Delos are listed in the Lexicon of Greek Personal Names (LGPN) as Jewish. This has created an entirely circular argument for anyone looking for external corroborating evidence.

the Lexicon as possibly being Jewish, but this is again only on the basis of Plassart’s identification. As before, there are other instances of Laodike from Delos which are not identified as Jewish, such as ID2628, found among a list of donor names on a marble plaque, which was discovered in the Theatre of the Syrian Sanctuary.

Furthermore, there are other contemporary epigraphic instances of the name Agathokles from Delos: one from the Agora of the Italians (ID1735), one from the Agora of the Competalists (ID1760), and one from the list of donors/subscribers to Sarapeion C (ID2618). These other instances of the name Agathokles on Delos are not identified as Jewish in the scholarship on the subject.

Inscription 4 (ID2331) (author’s translation) ΖΩΖΑΣ ΠΑΡΙΟΣ ΘΕΩΙ ΥΨΙΣΤΩΙ ΕΥΧΗΝ ‘Zozas of Paros to the God Most High, an offering’. Found in GD80, this is a small base of white marble, in the shape of a horned altar (Fig. 2). Zozas is identified in the LGPN as possibly belonging to a manumitted slave, but is not specifically identified as being Jewish. The ‘horned altar’ style of this base and that of ID2328 is very similar.

Inscription 2 (ID2328) (author’s translation) ΛΥΣΙΜΑΧΟΣ ΥΠΕΡ ΕΑΥΤΟΥ ΘΕΟΙ ΧΑΡΙΣΤΉΡΙΟΝ ‘Lysimachos for himself [to] God Most High [for a] thankoffering’.

Inscription 5 (ID2332) (author’s translation) ΥΨΙΣΤΩΙ ΕΥΧΗΝ ΜΑΡΚΙΑ ‘[The] Most High from Markia’.

This small, white marble inscription was found at the foot of wall W in Room B of GD80 and dates to the first century BCE (Fig. 2). The name Lysimachos in this inscription caused Plassart to associate IDs 2329 and 2328 with one another. Again, the designation of the name Lysimachos as Jewish in the LGPN was made on the basis of Plassart’s identification and there are other inscriptions from Delos containing the name Lysimachos. The name appears on ID1758, a list relating to the associations of the Hermaists, Apollonists, and Poseidonists, which was found in the Agora of the Italians, and on ID1764, relating to the association of Competalists. Most tellingly, the name appears again in ID2616, a list of subscribers to Sarapeion C, along with the name of our other protagonist, Agathokles.

Found on a bench in area A, this is a small marble base and dates to around the first century BCE (Fig. 2). The name Markia is again listed as Jewish in the LGPN only on the basis of Plassart’s identification. Inscription 6 (ID2333) (author’s translation) […] ΓΕΝΟΜΕΝΟΣ ΕΛΕΥΘΕΡΟΣ ‘… became free’. This badly damaged inscription base is a small rectangular base of white marble, discovered in the southeast corner of room B (Fig. 2). The marble is very badly damaged and only two words can be made out. Given the status of Delos as one of the main centres of the Aegean slave trade, it is hardly surprising to find that there are many inscriptions relating to the freeing of slaves found on the island, and it is evident that other than its proximity to the inscription bases found in GD80 and discussed by Plassart, there is nothing Jewish about this inscription and it is only Plassart’s identification and assumptions that have linked it to the others.

The fact that the names Lysimachos and Agathokles both appear in lists of subscribers to Sarapeion C (and elsewhere) is interesting, and it is important to note here that the internal furnishings of GD80 (the supposed synagogue), GD91 (Sarapeion A) (Fig. 5) and GD100 (Sarapeion C) are very similar (they are all basically rectangular structures with benches lining the internal walls). What these similarities mean is, of course, open to interpretation, but it is at least clear that this is all supportive of my contention that the names Lysimachos and Agathokles are not in themselves an indicator of Jewishness on Delos.

Plassart outlined his belief that the use of the epithets Hypsisto or Theo Hypsisto in the inscriptions indicated ‘a tendency towards Jewish monotheism’ (1914: 528). He further noted, however, noted that an inscription bearing the epithet Thea Hypsista had been found in Lydia, and probably referred to the Great Mother Goddess (Plassart 1913: 529). However, the inscriptions that refer to Hypsistos may also refer to the Greek deity Zeus Hypsistos, whose sanctuary on Delos is located on Mount Cynthus, only some 400m from GD80. There are many inscriptions from Delos containing the epithets Theos Hypsistos or Hypsisto, but Plassart has only identified the ones found in GD80 as Jewish.

Inscription 3 (ID2330) (author’s translation) ΛΑΟΔΙΚΗ ΘΕΩΙ ΥΨΙΣΤΩΙ ΣΩΘΕΙΣΑ ΤΑΙΣ ΥΦ’ ΑΥΤΟΥ ΘΑΡΑΠΗΑΙΣ ΕΥΧΗΝ ‘Laodike to God Most High for healing him/her of his/her infirmities, an offering’. Found in GD80, this inscription is dated to around 108/107 BCE (Fig. 2). This is a healing inscription and, again, is in the style of a Greek votive offering from a Greek sanctuary rather than a Jewish dedication, since Jewish dedications in this form are unknown. The name Laodike is listed in

Hypsistarians Another group that used these terms in the period with which we are concerned are the adherents of the Hypsistarian cult 108

LIDIA MATASSA who used the epithet Hypsistos to refer to their god as being above all others. Lamps, fire, and worship in the open air were an essential part of this cult which was associated with heaven and the sun, and worshippers established links with the deity by the dedication of lamps (Mitchell 1999: 92-9).

ΚΑΤΑΣΚΕΥΑΣΑΝΤΑ ΚΑΙ ΑΝΑΘΕΝΤΑ ΕΚ ΤΩΝ ΙΔΙΩΝ ΕΠΙ ΠΡΟΣΕΥΧΗ ΤΟΥ ΘΕ[ΟΥ] ΤΟΝ [- - - - - - - - - - - - - -] ΟΛΟΝ ΚΑΙ ΤΟ [- - - - - ΚΑΙ ΕΣΤΕΦΑΝΩΣΑΝ] ΧΡΥΣΩΙ ΣΤΕ[ΦΑ-]ΝΩΙ ΚΑΙ [- - - - - - - - - - - - - - - - -] ΚΑ - - Τ--

I would suggest that, as a total of forty-one lamps were found in GD80 during the excavations, it seems at least arguable that if GD80 had been used for cultic worship during its final phase, it could have been used by the Hypsistarians as easily as by any other group.

[The] ‘Israelites who make first-fruit offerings to holy Argarizim honour Menippos, son of Artemidoros of Heraclea, himself as well as his descendants to have established and dedicated its expenses, for an offering/ prayer [to God], [- - - - - - -] and [- - - - -] and crowned it with a golden crown and [- - - ]’ (Bruneau 1982: 469474).

The Samaritan Inscriptions In 1979, two inscriptions were found on Delos by Philippe Fraisse of the Ecole française d’Athènes (White 1987: 141). Both were found in an unexcavated area just beneath current ground level, at the foot of a north-south wall, near the shoreline about 100m north of GD80 (Fig. 1). Both are dedicated by the ‘Israelites who offer to Holy Argarizein’. The term Argarizein is the Greek rendering of the Hebrew term ‫( םזרג רה‬Har Garizim – Mount Garizim), the central locus of the Samaritan religion, and these two inscriptions certainly provide evidence of a hitherto unknown community of Samaritans on the island (White 1987: 142).

The second inscription was dated to around 250-175 BCE and is carved onto a heavily damaged white marble stele (Bruneau 1982: 469-474). This inscription refers only to Israelites as the dedicators, not to Israelites on Delos as indicated in Bruneau’s translation, as he assumes that the on Delos should be there. It is very interesting to note that Bruneau translated the phrase epi proseuche in the second Samaritan inscription as ‘for a prayer/offering’, whereas in relation to ID2329 he accepted Plassart’s interpretation of it as ‘for the synagogue’ (White 1987: 142).

Samaritan Inscription 1 ΟΙ ΕΝ ΔΕΛΩΙ ΙΣΡΑΕΛΕΙΤΑΙ ΟΙ ΑΠΑΡΧΟΜΕΝΟΙ ΕΙΣ ΙΕΡΟΝ ΑΡΓΑΡΙΖΕΙΝ ΣΤΕΦΑΝΟΥΣΙΝ ΧΡΥΣΩΙ ΣΤΕΦΑΝΩΙ ΣΕΡΑΠΙΩΝΑ ΙΣΟΝΟΣ ΚΝΩΣΙΟΝ ΕΥΕΡΓΕΣΙΑΣ ΕΝΕΚΕΝ ΤΗΣ ΕΙΣ ΕΑΥΤΑΟΥΣ

The two inscriptions show that the dedicators (on Delos or elsewhere) were connected to Mount Garizim in Samaria. It could be that offerings were sent to Mount Garizim while the Samaritan temple still stood there or that offerings continued to be made and sent to Samaria after the destruction of the temple. Or, indeed, it could be that the offerings were made on Delos, perhaps in the form of votives and were dedicated by Samaritan visitors to the island, Samaritan residents of the island, or even friends or business partners of Samaritans elsewhere on

‘The Israelites on Delos who make first-fruit offerings to Holy Argarizin crown with a golden crown Sarapion son of Jason of Knossos for his benefactions on their behalf’. (Translation by Bruneau 1982: 469). This inscription has been dated to between 150-50 BCE (Bruneau 1982: 469-74) and honours Sarapion for his benefactions on behalf of the ‘Israelites of Delos’. It is not clear whether the honouree is himself a Samaritan, Jew, pagan, resident of or visitor to Delos. It does however identify the dedicators as ‘the Israelites on Delos’, and there seems little doubt that this refers to a Samaritan community of some sort on the island. Samaritan Inscription 2 ΙΣΡΑΗΛΙΤΑΙ ΟΙ ΑΠΑΡΧΟΜΕΝΟΙ ΕΙΣ ΙΕΡΟΝ ΑΡΓΑΡΙΖΕΙΝ ΕΤΙΜΗΣΑΝ ΥΑΣ. ΜΕΝΙΠΠΟΝ ΑΡΤΕΜΙΔΩΡΟΥ ΗΡΑΚΛΕΙΟΝ ΑΥΤΟΝ ΚΑΙ ΤΟΥΣ ΕΓΓΟΝΟΥΣ ΑΥΤΟΥ 1. Due to difficulties during the editing process, we were unable to use lowercase Greek fonts with accents and other diacritical marks. Therefore, while I have used Philippe Bruneau’s translations for the two Samaritan inscriptions, they have been transferred into uppercase Greek letters.

FIGURE 3 PHOTOGRAPH OF BENCHES AND THRONE IN GD80 (PHOTOGRAPH BY LIDIA MATASSA: OCTOBER 2003) 109

SOMA 2004

FIGURE 4 PHOTOGRAPH OF VIEW FROM STYLOBATE TO CLIFF EDGE, LOOKING EAST (PHOTOGRAPH BY LIDIA MATASSA: OCTOBER 2003)

FIGURE 5 SARAPEION A, LOOKING NORTH (PHOTOGRAPH BY LIDIA MATASSA: OCTOBER 2003)

110

LIDIA MATASSA their behalf—as the two inscriptions are the only evidence of Samaritans on the island, it is impossible to know.

both areas B and D (Fig. 6), as its highest point is just over 30cm off the floor level! The cistern itself is deep (the bottom of the fault lies at 4m in places), and is 6.08m in length. In 1962 the cistern was excavated and cleaned out by Philippe Bruneau of the Ecole française d’Athènes (Bruneau 1970: 481).

GD80 - The Physical Evidence While I will briefly describe the structure and configuration of GD80, it is important to note that there is nothing in or around the building that is Jewish in nature and even the internal arrangement of benches in GD80 (which reminded André Plassart of the configuration of later synagogues) is not unique on Delos (as referred to above in the discussion on the inscriptions).

It has been claimed that the cistern could have functioned as a ritual bath (mikveh) (Bruneau 1970: 490-1 and Binder 1999: 306). However, even with the excavated opening, access from area D is difficult and precarious as the opening is under and extends only a metre from the arch and there is a sharp drop from the floor level to the bottom of the cistern from both areas B and D and, as I have said above, the highest point of the arch is a mere 30cm off the floor level. There are no steps built into the cistern, and there is little space in the opening for access via a ladder (Fig. 6).

GD80 lies on the eastern shore of Delos, in the Bay of Gournia (Fig. 1). The eastern part of the structure (area C) has collapsed into the sea along a line roughly 2m from the edge of a white marble stylobate and down a small cliff face, about a 1.5m drop (Fig. 4). The surviving structure comprises a rectangular building measuring 14.25m by 28.5m, divided into four areas: A, B, D and C (Fig. 2). The two main areas of GD80 (A and B) have some marble benches around the walls, as well as a marble throne in area A facing east (Fig. 3). However, the configuration of the internal furniture of GD80 is not unique on Delos, as areas of Sarapeions A (Fig. 5) and C are laid out identically with benches lining the walls. The marble benches and throne in GD80 probably came from the Ephebium of the gymnasium, a short distance from GD80 (Fig. 1). An inscribed statue base of the Gymnasiarch Poses (ID1928) was reused in GD80 after the destruction or removal of the statue which it carried, and an inscription set up by the ephebes of the gymnasiarch Theodosion, son of Jason (ID 1923b) was also reused in GD80. The phase of GD80 relating to its purported usage as a synagogue dates to some time after the third Mithridatic war, that is, after 63 BCE and by then it appears to have lost all but areas A and B, some small ancillary chambers on the southern end of the building (area D), the cistern, and part of area C. Plassart and other scholars (notably: Mazur 1935; Bruneau 1970; White 1987 and 1990; Goodenough 1988; Kraabel 1992; Binder 1999 and Levine 2000) have interpreted the physical layout of GD80 in several ways, none of which really has much bearing on its identification as a synagogue, perhaps other than the orientation of the surviving structure towards the east and the fact that it has benches arranged around the walls of the main areas. In fact, it is a perfectly standard Hellenistic Delian building, and there are other examples of such structures on the island (Fig. 7).

FIGURE 6 VIEW INTO THE CISTERN IN GD80 ACROSS AREAS B (LEFT) AND D (RIGHT) (PHOTOGRAPH BY LIDIA MATASSA: OCTOBER 2003) Moreover, there is no running water in the cistern, and it would undoubtedly have presented a very unsatisfactory arrangement in which to bathe. Furthermore, on an island devoid of a surface supply of water, bathing would have rendered the cistern useless for collecting water for drinking or cooking, and there is no other water supply related to this building.

The Cistern The first phase of GD80 appears to have been constructed around a natural rock-fault which was extended underneath areas B and D into a cistern and vaulted to support the weight of the building above it. The arch itself provides very limited access to the cistern from 111

GD57 HOUSE OF THE POSEIDONISTS

GD89 HOUSE OF THE HERMES

GD111 HOUSE OF THE DAUPHINS

FIGURE 7 OTHER HELLENISTIC BUILDINGS ON DELOS (AFTER BRUNEAU AND DUCAT 1983: 175, 216, 240). DRAWINGS: LIDIA MATASSA

112

LIDIA MATASSA While it is possible that there was a synagogue (Samaritan or Jewish, or both) on Delos, there is no evidence that it has been found or indeed that such discrete structures existed at this time. Because of the restrictions on the traditional practices of some cults and associations in the first century BCE and CE, it is possible that when Jews assembled for religious purposes, they did so in private or even in the open air, perhaps by the sea (as in the Letter of Aristeas). Moreover, the edict preserved in Josephus (AJ 14.213-216) relating to the Jews being forbidden to follow their religious traditions and customs is dated to precisely the time that Plassart considered GD80 to function as a synagogue, that is, to the middle of the first century BCE.

The Lime Kiln After the building was no longer used for any other purpose, a fairly substantial lime kiln was built in area A (Fig. 2). When Plassart found his inscription bases in areas A and B of GD80, he said that they were not associated with the lime kiln (Plassart 1914: 526), but his argument is based only on his belief that the inscriptions found in GD80 were Jewish and that GD80 was a synagogue. The burning down of marble for lime generally only occurred when the Mediterranean marble trade had tapered off, perhaps from the late third or fourth centuries CE, after GD80 was ruined and abandoned. Furthermore, all of the inscriptions found in GD80 were marble, including some small horned-altar type inscription bases (whose faces were so eroded that no inscriptions were visible on them) that Plassart did not mention and other scholars did not discuss, and there are many larger pieces of marble in and around GD80 (marble column drums that clearly do not belong to the building, and miscellaneous pieces of marble masonry) that were most likely brought there to be sawn into smaller pieces and melted down for use by those still cultivating land on the island and for water-proofing the cisterns.

Since we know of the Jewish population on Delos only from Maccabees and Josephus, and of the Samaritans only from the two inscriptions, it is difficult to see how these questions can be resolved unless and until there are substantial and extensive excavations of the area east of the stadium on Delos. Acknowledgements The Ecole française d’Athènes maintain a number of houses on the island for the purpose of accommodating their archaeologists during the digging season and I am extremely grateful to their Director of Studies, Mde. Michèle Brunet for arranging to open one of them to me, and to Κος. Π. Χατζηδακης of the Cycladic Ephoreia, for giving me permission to stay overnight on the island in 2003. This added a great deal to my understanding of the island as a sacred place, as well as allowing me time to view the structures cited in this article in the most peaceful environment imaginable, and one I cannot envisage being duplicated anywhere else.

Conclusions The early identification of GD80 as a synagogue has led to a distortion of the history of synagogues, with some scholars dating ‘the synagogue’ to the Hellenistic origins of the building in the third century BCE, all on the basis of the first inscription that Plassart discovered 90m north of GD80. While it is safe to say that there is nothing that would in itself prohibit GD80 from being a synagogue, not a single piece of evidence that has been linked with the structure would suggest that it actually was a synagogue. All hypotheses arguing that GD80 was a synagogue are based on the association of the first two inscriptions by Plassart back in the excavations of 1912 and 1913. Even the discovery in 1979 of the two Samaritan inscriptions does not go to support the identification of GD80 as a synagogue, as they were not found within or even close by that structure. What can be said is that there were Jews or Samaritans (or both) on Delos, possibly from as early as the third century BCE. It is simply not known where, how, or even if they worshipped on Delos.

Bibliography Binder, D. (1999) Into the Temple Courts. The Place of the Synagogues in the Second Temple Period, Society of Biblical Literature, Dissertation Series 169, Atlanta, Georgia. Bruneau, P. (1970) Recherches Sur Les Cultes de Délos a L’Époque Hellénistique et a L’Époque Impériale, Éditions E. De Boccard, Paris. Bruneau, P. and J. Ducat (1983) Guide de Délos, (third edition). Bruneau, P. (1982) Les Israélites de Délos et la juiverie délienne, in Bulletin de Correspondence Hellénique, 106: 465-504. Brunet, M. (1990) Delos. Bulletin de Correspondence Hellénique 114, 669-682. Crudden, M. (2001) The Homeric Hymns, Oxford, Oxford University Press. de Souza, P. (1999) Piracy in the Graeco-Roman World, Cambridge, Cambridge University Press, UK. Durrbach, F. and P. Roussel (1935), Inscriptions de Délos. Actes des Fonctionairres Athéniens Préposés

The physical evidence is complicated because in this period it is not certain that we should be looking for synagogues, since religious structures are bound to be of an ambiguous nature if worship was forbidden by local or Athenian law on Delos. For instance, during the period of Roman persecution there were no Christian churches; instead, private houses, bath houses, crypts, and even catacombs were used as meeting places. Overt architectural statements of identity only emerged when the political climate of religious tolerance made it safe for them to develop. 113

SOMA 2004 M Frede, eds., Pagan Monotheism in Late Antiquity, Oxford, Oxford University Press. Plassart, A. (1913) ‘La Synagogue juive de Délos’ in Mélanges Holleaux, recueil de memoirs concernant l’antiquité grecque, 201-215, Paris, Picard. Plassart, A. (1914) La synagogue juive de Délos. Revue Biblique, 23. Plassart, A. (1950) Inscriptions de Délos. Périodes de l’Amphictyonie Ionienne et de l’Amphictyonie AtticoDélienne, Dédicaces et Textes Divers. Ecrits dans les Alphabets Cycladiques (Nos. 1-35). Dédicaces, Bornes, Règlements, d’Alphabet Ionien Classique (Nos. 3670). Décrets Déliens. Ordonnance Lacédemonienne. Décrets Athéniens (Nos. 71-88), Paris, Librairie Ancienne Honoré Champion. Roussel, P. and M. Launey (1937) Inscriptions de Délos. Decrets Postérieurs A 166 AV. J.-C. (Nos. 1497-1524). Dédicaces Postérieures A 166 AV. J.-C. (Nos. 1525-2219), Dédicaces Postérieures A 166 AV. J.-C. (Nos. 22202528). Textes Divers, Listes et Catalogues, Fragments Divers Postérieurs A 166 AV. J.-C. (Nos. 2529-2879), Paris, Librairie Ancienne Honoré Champion. White, L.M. (1987) The Delos Synagogue Revisited: Recent Fieldwork in the Graeco-Roman Diaspora. Harvard Theological Review, 135-54. White, L.M. (1990) Building God’s House in the Roman World: Architectural Adaptation Among Pagans Jews, and Christians, Baltimore and London, Johns Hopkins University Press.

A L’Administration des Sanctuaires Après 166 AV. J.-C. (Nos. 1400-1479). Fragments de Actes Divers (Nos. 1480-1496), Paris, Librairie Ancienne Honoré Champion. Durrbach, F. (1926) Inscriptions de Délos. Comptes des Hiéropes (Nos. 290-371), Paris, Librairie Ancienne Honoré Champion. Durrbach, F. (1929) Inscriptions de Délos. Comptes des Hiéropes (Nos. 372-498). Lois ou Réglements, Contrats d’Entreprises et Devis (Nos. 499-509), Paris, Librairie Ancienne Honoré Champion. Goodenough, E. (1988) Jewish Symbols in the GrecoRoman Period, Princeton, NJ and Oxford, Princeton University Press. Kraabel, A.T. (1992) Diaspora Jews and Judaism. Essays in Honor of, and in Dialogue with, A. Thomas Kraabel, South Florida Studies in the History of Judaism, Number 41, Atlanta, Scholars Press. Levine, L.I. (2000) The Ancient Synagogue. The First Thousand Years, New Haven and London, Yale University Press. McLean, B.H. (1996) The Place of Cult in Voluntary Associations and Christian Churches on Delos. IN: Kloppenborg, J.S. and S.G. Wilson, eds., Voluntary Association in the Graeco-Roman World, London, Routledge. Mazur, B.D. (1935) Studies on Jewry in Greece, Athens, Printing Office Hestia. Mitchell, S. (1999) The Cult of Theos Hypsistos between Pagans, Jews, and Christians, in Athanassiadi, P. and

114

A ‘Hall of the Muses’ at Cyrene (Libya)? Oscar Mei

Immediately east of the Agora at Cyrene, next to the Temple of the Octagonal Basis, is the so-called Hall of the Muses (Fig. 1). It faces north-south and fronts onto the Skyrota, the main Greek plateia of the town and still one of the most important decumanis in Roman times (Stucchi 1967: 24-5; Luni 1988: 271-7; 1990: 87-120). The eighteenth, nineteenth and early twentieth century travellers at Cyrene neither mention the building in their reports nor indicate it on their maps. The monument was dug out in a hurry in 1916 by Italian soldiers clearing the hill of the Agora quarter from detritus, in order to get building material for a defensive wall on the summit of Cyrene hill. Since then, the building has been mentioned persistently yet concisely in both scientific and popular publications, where it has been indirectly hinted at in connection with other buildings or with the discovery of significant finds in its vicinity.

monument to the east of the Agora is ‘a building basement of Roman age with a decoration of scenic masks in the left corner; perhaps it served as a musical auditorium’ (1929: 501). Gaspare Oliverio, in 1931, refers to the building as the ‘House of the Muses’: ‘… the so-called House of the Muses, whose medallions are in the middle of the mosaic floor; it is finely adorned with tragic and comic masks, carved on the capitals of the external pillars’ (1931: 20). Richard Goodchild points out that

In the first guide to Cyrene, edited in 1923, Amilcare Fantoli mentions ‘a building of roman age, consisting of a square room with a ruined pronaos. There was a polychrome mosaic on the floor, with medallions, each portraying a head’. Fantoli also maintains that ‘in front of the building two statues were found (…); two torsos are still on the ground near the back corners’ (1923: 149). In his guide to Lybia, edited in 1929, Rodolfo Micacchi claims that the first

in the south-west corner of the insula is the “Hall of the Muses”, which is so renowned on account of its mosaic floor. It has badly destroyed round medallions, portraying the heads of the Muses. Two statues wearing a toga were found next to the front entrance: one of them surely represents a Roman emperor. The statue of a Roman empress, possibly Julia Mamea, was found inside the hall, lying on the floor. There were four Corinthian columns in front of the hall. Tragic masks had been carved on the lower part of the corner pillars. The building was damaged in the closing stages of its existence. The entrance steps were demolished and shacks were built in their place. The original function of the building is still unknown (1971: 91).

FIGURE 1 THE CYRENE AGORA AND THE TEMPLE OF THE MUSES (CENTRE FOREGROUND) IN 1931 (CYRENE NEG. 33388)

FIGURE 2 THE TEMPLE OF THE MUSES SEEN FROM THE SOUTH-WEST (O. MEI) 115

SOMA 2004 Members of the Urbino University Archaeological Mission have systematically studied the building through surveys and soundings since 2000 in order to gain a deeper understanding of it. The building is made of local soft yellow limestone: the blocks are partially destroyed. Four courses belonging to the podium walls are still standing, as are two courses on both the naos and pronaos walls. Only a few portions of the dilapidated mosaic pavement can be seen today.

Four rows of blocks make up the podium. It is about 1.5 m high and was built in two different stages. Even though the column-bases are completely ruined, all four of them are still in situ, rising directly from the continuous stylobate. The shafts are made of smooth drums of various sizes. The best-preserved remains are those belonging to the first column in the west. They show that the columns had a slight enthasis at a height of about 120 cm and then tapered regularly towards the upper end. We were able to calculate the column’s original height on the basis of the recovered drums and of the gradual tapering towards the upper end. It was 8.70 m high, base and capital included. The relation of diameter to height is 1:9.

This temple is of the Corinthian order and stands on a low podium and has a tetrastyle pronaos and a large naos (Figs. 2-3). It is 17 m long and 12 m wide. The front steps of the entrance were demolished to build shacks against the building, probably after the terrible earthquake of AD 262 (Stucchi 1965: 292-4, 298-301; Luni 1988: 273-4). The walls were skilfully built of ashlar masonry in a drystone technique, with courses of alternating blocks head by side (Lugli 1957: 177; Adam 1996: 118).

Accurate research on the ruined area of the temple has allowed us to identify some Corinthian capital fragments pertaining to the building, as well as one well-preserved capital. In spite of its good condition, the capital is heavily worn near the volutes; it is 110 cm high and 105 cm wide; its lower diameter is 78 cm. A fragment of a capital top was also found inside a trench: it had been set into the foundations of a late wall leaning to the temple podium. The Corinthian capital of the Temple of the Muses is more likely to be related to eastern prototypes than to western ones, especially on account of the leaf sharpness, the marked light and shade effect, and of its elegant carving. It can be compared with some capitals from the Corinthian Basilica of Hermoupolis Magna (Pensabene 1993: 396-7), with some from Sabratha (Tomasello 1983: 87-103; Joly and Tomasello 1984: 137-8, 162) and with

The pronaos is about 3 metres deep. The stylobate is formed of short side-facing blocks, firmly arranged in a homogeneous structure 130 cm wide. The side interaxial of the front columns measures 2.45 m, while the central one is more than twice as wide at 5.45 m. This peculiarity suggests that the front architrave was of Syrian type (Stucchi 1975: 249-50). The naos, 7.42 m by 10.09 m, is of rectangular plan. A rectangular statue base, 3.80m by 1.25 m, originally leant against the back wall. It was demolished in the last building stage and replaced with a niche.

FIGURE 3 PLAN OF THE TEMPLE OF THE MUSES (RECONSTRUCTION M. LUNI AND O. MEI, DRAWING O. MEI)

FIGURE 4 PROPOSED RECONSTRUCTION OF THE FACADE OF THE TEMPLE OF THE MUSES (RECONSTRUCTION M. LUNI AND O. MEI, DRAWING O. MEI) 116

OSCAR MEI the Agora of Side, in Pamphylia (Tomasello 1983: 90). The aforementioned examples are all dated to the Late Antonine period.

(reconstructed in AD 195; Pensabene 1990: 251-93), the Temple of Ceres in Thuburbo Maius (reconstructed during the time of Commodus; Pensabene 1990: 253-4) and the Temple of Aesculapius in Djebel Oust (Pensabene 1990: 263-5). Therefore, the Temple of the Muses can be dated to the end of the second century AD on the basis of these parallels.

Further to the west, in the corner of the street crossing, rows of architectural fragments relating to the collapse of the building’s entablature have been placed on the ground; they include a whole block of ionic architrave, 3.86 m long. Its length excludes the possibility that it was located on the façade, thus allowing us to definitely associate it with the temple’s lateral epistylium. As a matter of fact, the collapse in situ itself supports such a view. The most distinctive architectonic feature visible on the façade of the Temple of the Muses is its broken entablature connected with the central intercolumniation and crowned with an arch whose span is 4.80 m. As stated above, the central interaxial of the façade columns is indeed much wider than that of the lateral ones. Besides, among the collapse material east of the temple we have found two blocks pertaining to a similar arch whose dimensions match those of the monument. One of them seems to be a keystone, the other one is a springer stone.

In the final stage of its existence, the temple received a mosaic polychrome pavement. It was heavily damaged at the time of its discovery and almost nothing of it survives today. According to the discoverers’ reports, it was typified by round medallions portraying the busts of the nine Muses: data and reports definitely agree on this fact, so much so that the building drew its name from these very female figures. It was at first referred to as the ‘House of the Muses’, then as the ‘Hall of the Muses’ and eventually as the ‘Temple of the Muses’. What we currently have of the mosaic is limited to a few fragments recently found inside the naos, as well as some remains in the internal corners. Such small pieces are nevertheless important in order to unravel the general pattern of the mosaic. Its plan is rectangular (9.54 by 7.42 m); the outermost area comprised large circles, consisting of black tesserae on a white background. They were followed by a pattern of large red-bordered diamonds with a small polychrome square inscribed. A square frame made of blue, red, yellow and grey tesserae surrounded a simple polychrome guilloche. The centrally placed medallions portraying the nine Muses stood out clearly. Accurate technique and regular tessera dimensions surely underline the highly skilled style of the mosaic.

The pattern of the true Syrian pediment, with its curved and continuous entablature, reached its apogee in Syria in the second century AD (Crema 1959: 142-3; Frova 1961: 745; Mansuelli 1981: 107). Another type of Syrian arch also exists: the broken architrave of the façade is crowned with a distinct arch, as is the case with our Temple. The Corinthian Temple of Termessos Agora is a particularly good example of this kind: its broken entablature is connected with the central intercolumniation and crowned with an arch penetrating the tympanum (Lanckoronski 1892: 84-90; Crema 1959 389; Giuliano 1981: 77-8; Gros 1996: 190). The building is dated to the second century AD, like the contemporary Tomb of Mamastis in Termessos (Lanckoronski, Niemann, et al. 1890-1893: 40, 88, 152; Crema 1959: 389, 501; Gros 1996: 190) whose façade has the same architectonic pattern. Such peculiarity of the central arch can be also found on tombs, gates (Crema 1959: 446, 528) and on nymphaei (Crema 1959: 533, 748). The arch typology in the mausoleum of Claudia Antonia Sabina in Sardis, dated to AD 191 (Butler 1922: 170; Crema 1959: 564; Giuliano 1981: 100), is definitely comparable to that of the Temple of the Muses. No blocks of the frieze have been found, but we have ten preserved blocks from the cornice. This has allowed us to definitely restore the façade of our building (Fig. 4).

The lateral pattern is common to many other mosaics at Cyrene, especially to those of the House of Jason Magnus and of the Hall of Orthostats (Mingazzini 1965: 19-24) and of the Capitolium (Ghislanzoni 1916: 210; Mingazzini 1961: 101-3), dated to the late second and early third century AD. In contrast, the central panel can only be reconstructed based on an archive photograph of the floor showing indistinct, heavily damaged medallions arranged in a circle, each inscribed in an octagon. Taking for granted the symmetry of the design, it seems reasonable to suggest eight medallions arranged in a circle around a single central one (Fig. 3). For a full analysis, we should also take into consideration other Roman mosaics portraying the Muses and their general significance. As a matter of fact, the Muses are represented in this manner on many contemporary mosaic pavements. Nevertheless, the attributes of such depictions vary considerably. On the whole, these charismatic females gradually lost their particular religious connotations and eventually became used as purely decorative elements. This kind of mosaic was meant to celebrate the erudition of either the house owner or the patronus. Patroni are indeed assumed to have chosen the Muses to communicate their commitment to the Classics as well as their interest in the fine arts.

The longitudinal walls of the naos are excessively wide (1.90 m). This measurement includes the thickness of the wall proper as well as a series of niches facing inwards (at least four per side) and intended for statues. This is an uncommon feature for the temples of Cyrene (Stucchi 1975: 101). In contrast, it certainly applies to many temples elsewhere in the Roman world, such as the Serapaeum in Ephesus (second half of the second century AD; Giuliano 1981: 25-6), the Temple of Isis in Sabratha (BrouquierReddé 1992: 58-63) the Temple of Saturn in Dougga 117

SOMA 2004 An extremely interesting mosaic from Kos dating to the second to third century AD (Theophilidou 1984: 2447; Faedo 1997: 1011) has nine circular medallions, each portraying a bust and its attributes, arranged in three straight rows and joined together by a curvilinear guilloche. The lateral pattern of large diamonds closely resembles that of the Cyrene mosaic. Actually, the Muses were very popular in many African cities, particularly in Sousse, El Djem and their environs.

contrast, the female (Goodchild 1971: 91), headless, was found in 1916 inside the naos and is still in the storeroom of the Department of Antiquities of Cyrene: Rosenbaum dated it to the Severan period (1960: 96). The head was found in the naos as well and is also kept in the storeroom of Cyrene Museum. It is a high-quality sculpture, whose hairstyle belongs to a type Julia Domna wore in her early portraits (Rosenbaum 1960: 70). The other two statues were rediscovered by the archaeological mission of the University of Urbino during the campaign at Cyrene in September 2000. They are headless male statues whose size is slightly bigger than life-size (Fig. 5). Their features make it plausible to date them to the years between the end of the second century and the early third century AD, like the headless female statue and the portrait identified as Julia Domna (Rosenbaum 1960: 69-70, 80-1). Thus, the statues all seem to belong to one and the same imperial group from the Severan period. Therefore, we can conclude that the temple was remodelled for the last time during the Severan Period: the floor in the naos was covered with a mosaic and then a niche was built in the back wall. I think that, as far as this period is concerned, the so-called Temple of the Muses can plausibly be considered as a sacred space dedicated to the Severan dynasty.

Besides cultural reasons, the patronus’ choice could have disguised some practical need; as was the case with the Four Seasons and the Dionisiac retinue, the Muses represented a ‘convenient’ group of symmetrical figures to cover a large and demanding floor area. The Muses could sometimes emphasise the function of a room: in the House of the Muses of Althiburus, second half of the third century, fulllength Muses are depicted in a room with a semicircular apse facing the peristyle to the right of a central oecus (Dunbabin 1978: 268; Theophilidou 1984: 325-6). Such rooms are common in many houses in Roman Africa and possibly acted as halls where guests could attend musical performances and recitals. Two relevant mosaics have been also found in the Maison des Mois (Thysdrus). One of them (Room 3; AD 210-235) depicts laurelled medallions portraying the busts of the Muses (Dunbabin 1978: 258; Theophilidou 1984: 320-2; Lancha and Faedo 1997: 1017); in the other (Room 4, i.e. a cubiculum; late third century AD) the medallions, each inscribed in an octagon, are surrounded in their turn by a geometric pattern (Dunbabin 1978: 260; Theophilidou 1984: 323; Lancha 1997: 56-7; Lancha and Faedo 1997: 1017).

Now, one question remains unanswered: why did a temple dedicated to the imperial dynasty hold a mosaic pavement portraying the nine Muses? They were certainly not there by chance; there must have been some connection between them and the functions the temple fulfilled. We know that different kinds of events, including recitals, took place in temples (Stambaugh 1978: 587-8). For instance, Rome unquestionably had a temple dedicated to Hercules Musarum: M. Fulvius Nobilior built it after Ambracia was captured in 189 BC and had it adorned with statues of the nine Muses which had been plundered from

As a matter of fact, it is rather difficult to date our lost mosaic, since neither a drawing nor a photo of it exists. We rely entirely on second-hand excavation reports. Nevertheless, I believe that connections with, and similarities to, other mosaics can be suggested. We know for a fact that some of the known mosaics portraying the Muses date from the second half of the second century AD, and some others even to the third and fourth centuries AD. The typical style of medallions portraying the Muses dates from the late second to the first half of the third century AD, as indicated by the Maison des Mois (Thysdrus, Room 4) and the Kos mosaic. What’s more, such a span would fit the pattern of large diamonds, that is, the only extant piece of the Cyrene mosaic. At least four statues and one portrait were discovered in the 1916 excavation (Fantoli 1923: 73; Goodchild 1971: 91). Two of them, specifically the example portraying a crowned emperor and that of a female, in addition to the portrait were sent to Bengazi Museum (Fantoli 1923: 149). The others are still in situ at the back of the niche where they fell when the earthquake occurred. Unfortunately, the crowned statue of an emperor is no longer identifiable due to the confusion caused by World War II in Cyrenaica. In

FIGURE 5 THE MALE STATUES REDISCOVERED BY THE ARCHAEOLOGICAL MISSION OF THE UNIVERSITY OF URBINO IN SEPTEMBER 2000 (O. MEI) 118

OSCAR MEI Ambracia. This temple, connected to the amphitheatre to the north-east of the Porticus Octaviae was an important building where poets and musicians gathered to teach and read their works, surrounded by statues of the inspiring Muses (Tamm 1962; Richardson 1977). Our Temple might similarly have acted as a meeting place for poets, musicians and intellectuals deriving their inspiration from the Muses and Apollo. This would mean that the Severans were also seen and celebrated as patrons of the arts, and therefore the temple could be an emblem of the peace and stability which stemmed from their reign.

Mingazzini 1965: 73-5): they had previously been used to adorn another major building. This group, in pentelic marble, was used to decorate the south side of the courtyard: it faces a room paved with coloured marble inlays and a mosaic surround (a triclinium). Five Muses currently stand between the columns on remodelled octagonal bases (Fig. 6), a female head and the fragment of a torso are lying nearby. One of them is far bigger than the others: Paribeni identified this piece as Ariadne, taking his cue from the remains of a bunch of grapes on her abdomen (1959: 125). However, the connection with Ariadne, though plausible, is not the only possibility: such iconography might have been used to represent different types of female figures, such as maenads and muses. The statue could also portray either Calliope (the most important Muse as well as their leader, according to the ancients; Wegner 1960: 286) or Mnemosyne, the mother of the Muses. The latter is actually portrayed as bigger than the others on the votive relief from Bovillae, which is kept at the British Museum (Faedo 1997: 1004).

In the light of this argument, we can propose that the Hall of the Muses was constructed with the aim of hosting meetings of intellectuals and poets in the framework of the redevelopment which was carried out after the Jewish Revolt. It was probably completed at the end of the second century AD. It was subsequently remodelled during the first Severan period when it was, in all probability, dedicated to the new imperial family. The statues of the Muses were removed from the temple and displayed elsewhere, possibly in another prestige building.

Given that the bases have been sawn in half in order to fit the courtyard intercolumnia, we can reasonably assume that the statues were placed in their current position only in later reuse and that they were not originally meant for the peristyle. This view is supported by the fact that the fronts of the statues are carefully carved, whereas their backs are flat and unfinished. This indicates that originally they were intended to be placed either in niches or against a wall. Their size matches that of the eight niches which are thought to have existed inside the naos of our temple. The bigger one, Calliope or Mnemosyne, might belong

In this regard it is worth adding that at least three statuary groups of Muses have been found on the site. A single Muse, which can be dated to the third century AD, has been discovered in Theatre 4; it probably belongs to an unknown group (Paribeni 1959: 141). Another group was discovered during the 1962 excavations in the theatre west of the Caesareum (Goodchild 1971: 87-8; Menozzi 1998). At least six Muses have been found in the peristyle of the House of Jason Magnus (Paribeni 1959: 125, 138-41;

FIGURE 6 THE COURTYARD OF THE HOUSE OF JASON MAGNUS, WITH THE STATUES OF FIVE MUSES (O. MEI) 119

SOMA 2004 to the base adjoining the back wall. It might have been placed next to a statue portraying Apollo Musagetes. As a matter of fact, ancient pieces singling out one of the Muses are quite common: eight busts of Muses surround an emblema depicting Apollo and Erato on the mosaic floors from Sousse (Lancha and Faedo 1997: 1020-1) and Mérida (Lancha and Faedo 1997: 1025), while another mosaic pavement (from Sfax) displays a central image of Calliope with the poet Ennius (Lancha and Faedo 1997: 1024). Furthermore, Roman sarcophagi testify to the common habit of depicting eight Muses plus the deceased acting as the ninth. For example, this composition is illustrated on a Louvre sarcophagus, dated to 150 to 160 AD, on which the deceased embodies Calliope and leads the procession of the Muses (Lancha and Faedo 1997: 1045). On a contemporary piece from Civita Castellana the deceased seems to embody Erato (Lancha and Faedo 1997: 1045).

Ghislanzoni, E. (1916) Statua di Giove ed iscrizione onoraria agli imperatori Adriano ed. Antonino Pio, Notiziario Archeologico del Ministero delle Colonie, 2, Roma, Alfieri, 193-216. Giuliano, A. (1981) Le città dell’Apocalisse, La Spezia, Fratelli Melita. Goodchild, R. (1971) Kyrene und Apollonia, Zurich, Raggi Verlag. Gros, P. (1996) L’Architecture Romaine, I. Les monument publics, Paris, Picard. Joly, E. and F. Tomasello (1984) Il Tempio a divinità ignota di Sabratha, Roma, L’Erma di Bretschneider. Lancha, J. (1997) Mosaïque et culture dans l’Occident romain (I-IV sec.), Roma, L’Erma di Bretschneider. Lancha, J. and L. Faedo (1997) s.v. Musae, Lexicon Iconographicum Mythologiae Classicae, 8, ZürichDüsseldorf, Artemis, 1013-59. Lanckoronski, K. (1892) Städte Pamphyliens und Pisidiens, 2, Wien, F. Tempsky, Leipzig. Lanckoronski, K., Niemann, G. and E. Petersen (18901893) Les villes de la Pamphylie et de la Pisidie, II, Paris, Firmin-Didot. Lugli, G. (1957) La tecnica edilizia romana con particolare riguardo a Roma e Lazio, I, Roma, G. Bardi. Luni, M. (1988) Il Foro di Cirene tra secondo e terzo secolo. IN: Mastino, A. ed., L’Africa romana – Atti V Convegno di studio: Sassari, 11-13 dic. 1987, Sassari, Dipartimento di Storia, Università degli Studi di Sassari, 271-7. Luni, M. (1990) Il Ginnasio–“Caesareum” di Cirene nel contesto del rinnovamento urbanistico della media età ellenistica e della prima età imperiale. IN: Stucchi, S. ed., Giornata Lincea sulla Archeologia Cirenaica, Roma, 3 nov. 1987, Roma, L’Erma di Bretschneider, 87-120. Mansuelli, G.A. (1981) Roma e il mondo romano, 2, Torino, UTET. Menozzi, O. (1998) Un gruppo inedito di Muse da Cirene. IN: Catani, E. and S.M. Marengo, eds., La Cirenaica in età antica. Atti del Convegno internazionale di studi, Macerata 18-20 maggio 1995, Pisa-Roma, Istituti Editoriali e Poligrafici Internazionali, 405-27. Micacchi, R. (1929) Cirene, in Bertarelli, L.V. ed. (1929), Guida d’Italia del TCI, Possedimenti e Colonie, Milano, Touring Club Italiano, 501. Mingazzini, P. (1961) Sulla vera denominazione del cosiddetto Capitolium di Cirene, Quaderni di Archeologia della Libia, 4, Roma, L’Erma di Bretschneider, 101-3. Mingazzini, P. (1965) L’insula di Giasone Magno a Cirene, Roma, L’Erma di Bretschneider. Oliverio, G. (1931) Scavi di Cirene, Bergamo, Istituto italiano d’arti grafiche. Paribeni, E. (1959) Catalogo delle sculture di Cirene, Roma, L’Erma di Bretschneider. Pensabene, P. (1990) Il Tempio di Saturno a Dougga. IN: Mastino, A. ed., Africa Romana – Atti del VII convegno di studio, Sassari 15-17 dic. 1989, 7, I, Sassari, Dipartimento di Storia, Università degli Studi di Sassari, 251-93.

In conclusion, on account of the fact that the temple was restored and the Severan statues placed in the naos, the group of the Muses might consequently have been transported to the courtyard of the House of Jason Magnus. This would probably justify the remodelling of perystile capitals into scenic masks during the reign of Septimius Severus (Mingazzini 1965: 87-8). The building existed until AD 262 when it was destroyed by a terrible earthquake, which caused many other buildings to collapse. Trenches TM00A /TM00C have shown that the rubble was levelled and shacks were built with any material on hand, such as the front steps of the temple and the abacus of one of the Corinthian capitals of the pronaos. The walls of the shacks, leaning against the building, show how scant the available resources were and how hastily the city was reconstructed. The temple and the adjoining shacks were utterly destroyed after the 365 AD earthquake, as attested by some coins which have been recently found on shack pavements. Bibliography Adam, J.P. (1996) L’arte di costruire presso i Romani, Milano, Longanesi. Brouquier-Reddé, V. (1992) Temples et cultes de Tripolitanie, Paris, Editions du Centre National de la Recherche Scientifique. Butler, H.C. (1922) Sardis, I, 1, Leiden, E. J. Brill. Crema, L. (1959) L’Architettura romana, Enciclopedia Classica, Sez. III, 12, Tomo I, Torino, SEI. Dunbabin, K.M.D. (1978) The Mosaics of Roman North Africa, Oxford, Clarendon Press. Faedo, L. (1997) s.v. Mousa, Mousai, Lexicon Iconographicum Mythologiae Classicae, 8, ZürichDüsseldorf, Artemis, 991-1013. Fantoli, A. (1923) Guida della Libia, II. Cirenaica, Milano, Touring Club Italiano. Frova, A. (1961) L’arte di Roma e del mondo romano, Torino, UTET. 120

OSCAR MEI Pensabene, P. (1993) Elementi architettonici di Alessandria e di altri siti egiziani, Repertorio d’Arte dell’Egitto Greco-romano, Serie C, 3, Roma, L’Erma di Bretschneider. Richardson, L. (1977) Hercules Musarum and the Porticus Philippi in Rome. American Journal of Archaeology, 81, Boston, Archaeological Institute of America, 35561. Rosenbaum, E. (1960) A catalogue of Cyrenaican portrait sculpture, London, British Academy. Stambaugh, J.E. (1978) The Functions of Roman Temples, Aufstieg und Niedergang der Römischen Welt, II, 16.1, Berlin-New York, W. de Gruyter, 554-608. Stucchi, S. ed. (1965) L’Agorà di Cirene, I. I lati Nord ed Est della Platea Inferiore, Roma, L’Erma di Bretschneider.

Stucchi, S. ed. (1967), Cirene 1957-1966. Un decennio di attività della Missione Archeologica Italiana a Cirene, Tripoli, Istituto Italiano di cultura. Stucchi, S. (1975) Architettura Cirenaica, Roma, L’Erma di Bretschneider. Tamm, B. (1962) Le Temple des Muses à Rome, Opuscola Romana, 3, Lund, Glerup, 157-67. Theophilidou, E. (1984) Die Musenmosaiken der Römischen Kaiserzeit, Trierer Zeitschrift, 47, Trier, Rheinischen Landesmuseum. Tomasello, F. (1983) Un prototipo di capitello corinzio, Quaderni di Archeologia della Libia, 13, Roma, L’Erma di Bretschneider, 87-103. Wegner, M. (1960) s.v. Muse, Enciclopedia dell’Arte Antica, 5, Roma, Istituto dell’Enciclopedia italiana, 286-97.

121

Disentangling Material Cultures: Late Roman and Sasanian Facet Cut Glassware in Late Antiquity Hallie Meredith-Goymour

Outlining the Question The history of late Roman and Sasanian relations was a divisive one characterised by periodic invasions interspersed with periods of general animosity. The Sasanian dynasty lasted from the first quarter of the third century AD to the Islamic Conquest c. 651 AD. Constantinople, the capital of the later Roman Empire and Ctesiphon, the capital of the Sasanian Empire, were in relative proximity to one another. The frontier zone between the eastern late Roman Empire and the western Sasanian Empire was fought over throughout late antiquity. The Sasanian Empire had its eastern limits around Bactria in modern Afghanistan, and it was within northern Sasanian territory that the socalled Silk Road from China split, creating two principal routes to the Sasanian and late Roman capital cities, respectively. This route is arguably the way in which high value goods were traded long distances overland, to and from the Far East. Since overland trade via the Silk Road crossed Sasanian territory, sometimes with goods en route to the Roman east, excavations of the frontier zones have the potential to uncover a wealth of evidence concerning traded luxury items such as facet cut glassware.

FIGURE 1 SASANIAN FACET CUT GLASS BOWL. CORNING MUSEUM OF GLASS (60.1.3), AMLASH, IRAN (?), FOURTH TO SIXTH CENTURY (?), HEIGHT: 89 MM, DIAMETER: 130 MM (WHITEHOUSE 1993: 264, FIG. 113) known as open-work vessels, cage cups or diatreta, which are found almost exclusively within the Roman Empire. This form of fourth century glassware has never been found in Sasanian Persia. This distribution pattern may be the result of retrieval bias in Sasanian Persia, or it may represent differences between Roman and Sasanian tastes and preferences. Alternatively, it may suggest ongoing trade restrictions placed upon luxury glass imports or exports between the Roman and Sasanian Empires.

Sasanian merchants provided the long-distance link for trade in luxury objects between the later Roman Empire and the Far East. As a case in point, whilst glass making has been known since c. 1500 BC, glass blowing was developed by the Romans between 50 BC–50 AD, out of a western Asiatic glass making tradition (Israeli 1991: 46-55). According to Chinese sources, knowledge of the craft of glass making was not made available to China until Soghdian merchants on the north-eastern periphery of Sasanian territory ‘sold Roman techniques of glassmaking to the emperor of China’ in the fifth century AD (Brown 1971: 162). Therefore, any blown faceted glass found in China before they could produce it locally was necessarily imported from the late Roman and Sasanian west, the result of long-distance trade (eg. Fig. 1).

The production of facet cut glassware is of special interest as it overlaps cultural boundaries. Faceted glass can potentially contribute to our understanding of possible trade restrictions between later Romans and Sasanians, or, alternatively, concurrent well-established trade relations with the Far East. Based on studies involving the circulation of other types of luxury vessels, especially Sasanian silver royal hunting plate, the limited evidence suggests a history of general foreign influence but with very restricted trade between these two particular neighbouring empires (Harper 1988: 154).

Faceted glass was the most laborious form of glassware produced by Sasanians. Roman craftsmen were producing a more intricate and highly crafted type of engraving, 123

SOMA 2004 Situated in modern Syria, Palmyra was a town on a caravan route that lay within the contested frontier zone, several days journey from the Sasanian capital near modern Baghdad. A tomb discovered in Palmyra included a carved stone statue holding a faceted bowl (Fukai 1968: fig. 8). Although there is no way of knowing for certain whether the material represented in stone is glass, similar glass vessels have been found in excavations in Syria, for example, at Dura Europos (Clairmont 1963: 56-80). This type of funerary sculpture demonstrates the high status attributed in antiquity to facet cut vessels, most likely of glass. Facet cut glass vessels were manufactured contemporaneously in the Sasanian and Roman Empires, and the faceted vessel depicted may have been either Roman or Sasanian. The central question addressed is to what extent can Sasanian and Roman luxury facet cut glassware be differentiated?

FIGURE 3 FACET CUT GLASS BOWL. TREASURY OF SAN MARCO (67), UNKNOWN PROVENANCE, COLOURLESS, TRANSPARENT GLASS OF UNCERTAIN DATE WITH ELEVENTH CENTURY SILVER-GILT MOUNT, HEIGHT: 80 MM, DIAMETER: 270 MM (GRABAR 1971: 75-6, PL. LXIII)

The artefacts considered here provide a brief survey of Roman and Sasanian faceted glass vessels in order to establish some of the salient cultural differences in production. The discussion omits other forms of glassware unique to Roman manufacture. The approach discussed here has been developed as an attempt to distinguish between glass vessels produced using a similar technique to achieve somewhat different effects: faceted glass vessels produced by late Roman and Sasanian craftsmen. Subsequently, two dissimilar pairs are discussed in order to demonstrate the utility of this approach. The aim is to determine to what extent the criteria presented could serve as a tool with which to differentiate between Roman and Sasanian facet cut glass vessels, independent of archaeological context.

1992: 386). Facets are found adorning glass vessel shapes such as tubes, conical vessels, amphora-like forms, small unguentaria, as well as other forms. Within the late Roman and Sasanian Empires, facets frequently decorated widenecked glass vessels that were typically palm-sized, making the glass easy to hold, even when wet. The criteria which, when taken together, could serve to differentiate late Roman from Sasanian facet cut glass derive from three overlapping aspects of production: glass making, glass working and decoration. It is important to note that glass making and glass working, are two separate stages in glass production. Whereas glass making refers to the aspects of production which affected the composition of glass as a raw material, glass working is the physical manipulation or shaping of glass using hot or coldworking techniques. Glass working refers to consistent culturally-based practices, irrespective of decoration. Decoration refers to the pattern which adorns the surface of a glass vessel. Each of the characteristics represents a single aspect of production. It is only when considered in conjunction with one another that cultural differences in production can be discerned.

Criteria for Differentiating Facet Cut Glass Vessels Facets are essentially one or more smooth, planar faces which comprise a depression on the exterior surface of a glass vessel (Figs. 1-2). Facets in relief consist of a depression on the outer surface of a shallow circular projection, and are less commonly found (Fig. 3). A single facet is one of the small cut and polished faces on a surface that produces a glittering effect in light (Simpson

Glass Making Glass produced by Sasanian and late Roman craftsmen was made by melting a mixture of soda, lime and silica in varying proportions (Brill 1988: 264-9; Brill 2001: 259). Lead-rich glasses were relatively uncommon outside of the Far East. Ingots and other transportable forms of raw glass have been found or suggested from throughout the late Roman Empire and possibly in Sasanian sites in central and southern Iraq (Negro Ponzi Mancini 1984: 356; Simpson pers. comm. February 2004). Thus, scientific analyses of the composition of glass produced in different periods and cultures do not necessarily identify the place of production. However, scientific analyses which establish the quantities and proportions of chemicals found in glassware are furthering our knowledge of chronological and cultural differences in glass making practices.

FIGURE 2 ROMAN FACET CUT GLASS BOWL. RÖMISCH-GERMANISCHES MUSEUM (474), COLOGNE, THIRD CENTURY, HEIGHT: 65 MM (CLAIRMONT 1963: PL. XXXVII) 124

HALLIE MEREDITH-GOYMOUR Roman glass was naturally a pale bluish-green colour. According to Pliny the Elder, writing in the mid-first century AD, Romans preferred colourless, transparent glass because of its similarity to rock crystal (Pliny HN 36.198-9). In order to obtain clear glass, blue-green glass was intentionally decolourised. This was achieved by the addition of specific quantities of chemical agents during the process of glass making.

integrity of the pattern from the mould is retained (Negro Ponzi Mancini 1984: 35). After mould-blown glass had cooled, the outer surface of the glass vessel would then have been engraved using cold-working techniques to further define the facets. This method of production involves both cold and hot-working glass techniques, which has widespread implications for trade. This sequence of production was a process that involved both craftsmen, who were skilled glass blowers, and glass engravers. This may have required trade at various levels of production or a division of specialised labour in a producing centre. It is primarily through this kind of indirect evidence found in the finished products that archaeologists can establish a clearer picture of late antique workshop production and trade.

Sasanians did not exhibit any such predilection for clear glass. No known Sasanian glassware is colourless or intentionally decolourised. Although such an established trend could certainly indicate cultural preference, it raises a related question concerning trade: could this indicate an inability of the Sasanians to acquire the decolourant? Most faceted Sasanian glassware appears pale in colour, commonly green, yellow, amber, or brown (Negro Ponzi Mancini 1984: 36; Simpson 1992: 372). Since not all Roman glass was decolourised, coloured glass could be either Roman or Sasanian; however when faceted glass is found that is colourless, it is one characteristic that may be taken to indicate Roman rather than Sasanian manufacture. Ideally, however, this should be tested through compositional analysis and typological comparison.

Relative to their late Roman counterparts, the wall thickness of Sasanian glassware was typically thicker in proportion to the vessel as a whole. Measurements taken of a representative sample reveal that, whereas Roman facetted vessels have approximate wall thicknesses of between 1.5 and 3 mm, Sasanian facetted vessels have approximate wall thicknesses of between 4 to 6 mm. This consistent difference in relative wall thickness strengthens the argument for identification by typical characteristics exhibited in production within cultural traditions.

It is not always possible to determine the original treatment of the surface beneath layers of weathered and decaying glass. As summarised by Newton and Davison, Weathering refers to the ‘[c]hanges on the surface of glass (caused by exposure to adverse conditions), which appear as dulling, frosting, iridescence or decomposition’ (1997: 283), while Iridescence is ‘the rainbow-like play of different colours, changing according to the angle of view’ (1997: 280), caused by the diffraction of light from several layers that have air trapped between them. Iridescent layers flake off, because the glass is decomposing. Therefore, iridescence is a visual indication that the surface layers are coming apart.

Decoration No typology has been established for late Roman faceted glass vessels. However, continuity in decoration is a useful way of characterising some distinctly Roman aspects of facet production in glassware (Ekholm 1963: 29-37). Both late Roman and Sasanian types have faceted decoration, which exclusively adorns the exterior surface of the vessel. This patterning continues on the lower walls and underside of the base. When facet patterns are not adjacent, producing a honeycomb effect, facets are generally arranged in horizontal bands.

Weathering can have a serious effect upon the interpretation of an assemblage. The proportions used in the composition of glass are crucial in determining its rate of deterioration. Factors contributing to a weathered layer on a glass surface include composition, time since burial, and environmental conditions (Freestone 2001: 615-25). In a pilot study carried out by the British Museum, it was provisionally determined that Sasanian glassware decayed, or weathered, around ten times faster than Roman glassware, establishing a useful diagnostic tool for differentiation (Freestone 2001: 619).

Unlike Sasanian artefacts, late Roman facet cut vessels usually contain one to two bordered, but otherwise plain, upper registers (Figs. 2 and 5). Such plain upper bands may have continued to be included due to ongoing use of traditional production techniques. As part of a related overall division of space, the late Roman decorative scheme divided the exterior surface into segments, each with similar patterning, but usually differentiated from one another. The Sasanian facet patterns generally extended across the entire surface of the vessel (Figs. 1 and 6). Concave depressions are most typically adjacent, as in the honeycomb pattern, or arranged horizontally in discontinuous rows (Harada 1962: pl. III, fig. 6; Jiayao 1984: 3 and 7, figs. 3 and 11).

Glass Working When the surface of the glass is not obscured by weathering, the uniformity of layout and relative facet diameter in some Sasanian and late Roman faceted glass vessels is evidence of mould-blowing (Fig. 1). Mould-blown glass is glass that is freeblown directly into a mould when in a malleable, molten state. An inflated glass bubble takes the shape of the negative-space pattern, typically on the walls and base of a mould. With limited further inflation and shaping, the

In general, late Roman patterns appear more varied in terms of their composition and placement. Whereas facets decorating late Roman glassware are commonly mixed with other kinds of repeated engraving (Fig. 2), this is not 125

SOMA 2004

FIGURE 4.1 FACET CUT GLASS BOWL. TREASURY OF SAN MARCO (93), UNKNOWN PROVENANCE, GREENISH GLASS OF UNCERTAIN DATE WITH TENTH TO ELEVENTH CENTURY SILVER-GILT, STONE AND PEARL MOUNT, HEIGHT: 60 MM, DIAMETER: 170 MM (GRABAR 1971: 73, PL. LX). often the case in Sasanian faceted vessels. (A noteworthy exception are the vessels from the early Sasanian site of Choche; see Negro Ponzi Mancini 1984: 40, fig. 4, nos. 1, 2, 4, 5, 10-13). If different shapes were utilised as Sasanian facets, the overall patterning was generally a mixture of ordered faceted shapes.

The first bowl is described as colourless and transparent (Buckton 1984: 191). Therefore, it would have been decolourised, indicating non-Sasanian glass making. It is decorated with facets in relief, consisting of projecting circles combined with small conical protrusions, and has a ring base (Fig. 3). The projecting facets on the bowl appear similar in form to a type of Sasanian facet discussed by Fukai, chronologically later than the honeycomb pattern. Just visible below the inscribed metal addition is the lower limit of a horizontal line. The upper limit of the interior of the glass rim is visible. The metal band roughly follows the contour of the original rim, hindering measurement of the thickness of glass at the lip. The exterior metal band covers a wide, plain horizontal border mirroring one aspect of late Roman decoration, although the vessel as a whole does not adhere to any decorative scheme typical of late Roman or Sasanian faceted glass vessels.

At Sasanian Choche, several varied combinations of facets and lines were found dating to the mid to late fourth century AD. According to Negro Ponzi Mancini , such locally produced faceted glass decoration included ‘rows of isolated oval facets, rows of slice facets between bands of engraved lines, alternating rows of oval and slice facets with or without intermediate strips of engraved lines and occasionally rows of facets alternating with various elements, such as oblique lines’ (1984: 35-6). In general, the greatest degree of variation in decorative patterning appears to have occurred in the use of facets in the form of concave depressions. Other later types of circular Sasanian facets are known, some of which are discussed by Fukai (for facets in relief, see Fukai 1977: 43-6; von Saldern 1963: pl. 4, fig. 7; for double circular facets, see Fukai 1977: 46-8; see von Saldern 1963: pl. 5, figs. 11-12). Applying this Approach Two groups of vessels will be considered here. The first group consists of two glass bowls, part of a problematic set of material from the Treasury of San Marco, Venice (Figs. 3, 4.1–4.2). The glass itself is most likely from late antiquity; however, the Byzantine metal additions (mounts) are from around the tenth or eleventh centuries. This kind of Byzantine re-use has led to ongoing debates regarding the cultures that produced the original vessels. Since there is no evidence of decay resulting from a chemical reaction to soil conditions, neither object appears to have been buried. The second group comprises two similar, narrownecked glass vessels (Figs. 5 and 6). The criteria outlined earlier will be applied to the question: are these vessels late Roman or Sasanian?

FIGURE 4.2 DETAIL, BASE OF FIG. 4.1 (BUCKTON 1984: 197, FIG. 26B) 126

HALLIE MEREDITH-GOYMOUR The second bowl has been described as greenish (Buckton 1984: 195) and is decorated with honeycomb facets (Figs. 4.1–4.2). It has a metal addition serving as a ring base, a wide, bejewelled metal rim, and four metal straps connecting them. The five-sided facets extend to the upper limit of the exterior of the glass rim, which is visible only from below. As in Sasanian honeycomb faceted glassware (c.f. Figs. 1 and 6), there is no upper border, neither does it appear that the glass vessel originally had a ring base. This is similar to early Sasanian artefacts from fourth century Choche (Negro Ponzi Mancini 1984: 40, fig. 4, no. 7). However, as seen most clearly on the detail of the base, squares are interspersed between the rows of facets (Fig. 4.2). The presence of the squares in the decorative programme found on the base is not characteristic of Sasanian faceted glass bases (eg. Fukai 1968: figs. 2, 6, 10, 12, 14 and 18; Fukai 1977: figs. 21, 32, 33, 36, 37, 42-4). The facets on the underside of the base are adjacent, but the triangular shapes typically found between circular facets surrounding the central base facet are entirely absent. Omitting discussion concerning the nature or function of the Byzantine metal additions, the movement or the acquisition of these vessels, solely on the basis of the criteria already outlined, the diagnostic features of the first bowl suggest late Roman origin. Based on stylistic considerations, the second bowl may be earlier than the first. The extension of the facets to the rim of the second bowl suggests Sasanian manufacture. Although described as ‘greenish’, the glass is extremely pale in colour and is arguably decolourised. Scientific analysis could reveal decolourisation by the addition of manganese, strongly suggesting Roman manufacture. Since no such tests have been performed, and wall thickness measurements remain unavailable, the attribution of the second bowl is still in question. This analysis, therefore, suggests that: (1) if the glass was decolourised, then this bowl is most likely late Roman; (2) however, if the glass was not decolourised, then based primarily on aspects of decoration, this bowl is more likely of Sasanian origin.

FIGURE 5 FACET CUT GLASS JUG. BRITISH MUSEUM (GR1856.1226.1203), UNKNOWN PROVENANCE, DECOLOURISED GLASS C. 50-100, HEIGHT: 120 MM, UPPER DIAMETER: 77 MM (HARDEN, ET AL. 1987: 193, FIG. 103)

The second group consists of two narrow-necked, conical vessels (Figs. 5 and 6) which in several respects appear similar. Independently from the approach discussed, one vessel has been assumed to be Roman, the other Sasanian (Harden, et al. 1987: xi; Whitehouse 1993: 257). These identifications are in keeping with the evaluation resulting from the approach to disentangling material cultures put forward. Despite the fact that both glass vessels employ the continuous honeycomb pattern across the outer surface, and are similarly shaped, colour and division of space aid in the attribution of these pieces. Whereas the Sasanian vessel has no delineated band below the lip (Fig. 6), the Roman vessel has a wide, plain border just below the rim (Fig. 5). Both vessels have a central, faceted disc at the centre of the base. Most likely serving different functions, the Roman vessel has a sparsely decorated applied handle. The Sasanian vessel has a hole pierced in the base, created after production, the function of which is unknown.

FIGURE 6 FACET CUT GLASS VESSEL. CORNING MUSEUM OF GLASS (62.1.4), UNKNOWN PROVENANCE, TRANSPARENT PALE YELLOWISH BROWN GLASS, FOURTH TO SIXTH CENTURY (?), HEIGHT: 202 MM (WHITEHOUSE 1993: 257, FIG. 105). 127

SOMA 2004 Although difficult to tell from the effects of decay and the colours visible today, the Roman vessel was intentionally decolourised (Harden, et al. 1987: 193). The Sasanian vessel was not decolourised. Thus, the identifications of Roman for the manufacture of the former, and Sasanian for the latter are in agreement with the application of the proposed criteria.

Acknowledgements I would like to thank Lincoln College, Oxford’s Graduates’ Research Fund for their kind support, Dr. Gasparotto of the Treasury of San Marco, Marlia Mango, St. John Simpson, the anonymous reader and Dan Meredith-Goymour for their comments. Any errors or omissions are those of the author entirely.

It can be difficult to determine whether glassware of a similar technique is late Roman or Sasanian. The difficulty can be further exacerbated by the absence of scientific analyses concerning glass composition. However, late antique faceted glassware can potentially be distinguished in a number of ways. Although similar in chemical composition, late Roman and Sasanian glassware differ in decolourisation of glass and rate of decay. Relative wall thicknesses typically differ. Lastly, unlike Sasanian faceted glass, the late Roman equivalent more commonly mixes circular facets with other patterns of engraving, and almost always has a plain border below the lip, or a similar framing device.

Bibliography Brill, R.H. (1988) Scientific Investigations of the Jalame Glass and Related Finds. IN: Weinberg, G.D. ed., Excavations at Jalame: Site of a Glass Factory in Late Roman Palestine. Columbia, MO., University of Missouri Press, 257-83. Brill, R.H. (2001) Some Thoughts on the Chemistry and Technology of Islamic Glass. IN: Carboni, S. and D. Whitehouse eds., Glass of the Sultans. New York, Metropolitan Museum of Art, 25-45. Brown, P. (1971) The World of Late Antiquity, AD 150750. London, Thames & Hudson Ltd. Buckton, D. (1984) The Treasury of San Marco, Venice. Milan, Olivetti. Clairmont, C.W. (1963) The Glass Vessels. The Excavations at Dura-Europos: Final Report IV, Part 5. New Haven, CT., Dura-Europos Publications. Ekholm, G. (1963) Scandinavian Glass Vessels of Oriental Origin from the First to the Sixth Century. Journal of Glass Studies, 5, 29-37. Freestone, I. (2001) Post-depositional Changes in Archaeological Ceramics and Glasses IN: Brothwell, D. and A. Pollard eds., Handbook of Archaeological Sciences. Chichester, John Wiley & Sons, Ltd., 61525. Fukai, S. (1968) Study in Iranian Art and Archaeology: Glass and Metalwork. Tokyo, Yoshikawa Kobunkan. Fukai, S. (1977) Persian Glass. Crawford, E.B. trans. New York, Weatherhill & Tankosha. Grabar, A. (1971) Calici bizantini e patene bizantine medievali. IN: Hahnloser, H.R. ed., Il Tesoro di San Marco. Sansoni: Firenze, 55-90. Harada, Y. (1962) Ancient Glass in the History of Cultural Exchange between East and West. Acta Asiatica: Bulletin of the Institute of Eastern Culture, 3. Tokyo, the Tōhō Gakkai, 57-69. Harden, D.B., Hellenkemper, H., Painter, K. and D. Whitehouse (1987) Glass of the Caesars. Milan, Olivetti. Harper, P.O. (1988) Sasanian Silver: Internal Developments and Foreign Influences in Baratte, F. ed., Argenterie Romaine et Byzantine. Actes de la Table Ronde, Paris 11-13 Octobre 1983. Paris, de Boccard, 153-62. Israeli, Y. (1991) The Invention of Blowing. IN: Newby, N., and K. Painter, eds., Roman Glass: Two Centuries of Art and Invention, 13. London, Society of Antiquaries of London, 46-55.

Conclusions Faceted glass vessels were made and widely circulated throughout late Roman territory, Sasanian Persia, the Caucasus and the Far East. The archaeological evidence demonstrates that both Sasanian and late Roman artisans were producing a similar type of decorated vessel, often both hot and cold-worked according to their culturallybased practices and tastes. Early Sasanian glass artisans could look back to over a century of faceted glass production by their glass producing neighbours in the eastern Roman Empire. However, the aesthetic of the facet inherited from the west developed into a distinctly Sasanian tradition. Importantly, the culturespecific late antique forms created contemporaneously by late Roman and Sasanian glass craftsmen display a similar development from early Roman faceted glassware by expanding the decoration to the underside of the vessel. This may further illustrate the idea that Sasanians appropriated a late Roman facet cutting practice. The extent to which both late Roman and Sasanian faceted glassware developed in divergent ways is more difficult to determine when one considers the possibility of imitation. However, as proposed here, there are key diagnostic features which considered together have the potential to differentiate late Roman from Sasanian faceted glassware. The value of such a culturally distinguishable method of identification is two-fold. Firstly, such studies may aid in the identification of glass exported from either the Roman or the Sasanian Empire, independent of final depositional context. Secondly, when systematically applied, the identification of glassware as either Sasanian or late Roman can function as another tool with which to disentangle intercultural activity at complex frontier sites. 128

HALLIE MEREDITH-GOYMOUR Jiayao, A. (1984) Early Chinese Glassware. Henderson, M. trans. Translations Number Twelve, the Oriental Ceramic Society. Hong Kong, Millenia Publishers. Negro Ponzi Mancini, M.M. (1984) Glassware from Choche (Central Mesopotamia) in Boucharlat, R., and J.-F. Salles, eds., Arabie Orientale, Mésopotamie et Iran Méridional de L’age du Fer au Début de la Période Islamique, Mémoire 37. Paris, Histoire du Golfe. Editions Recherche sur les Civilisations, 3340. Newton, R., and S. Davison. (1997) Conservation of Glass. Oxford, Butterworth-Heinemann.

Pliny. (1971) Natural History. Eichholz, D.E. trans. Warmington, E.H. ed., Loeb Classical Library, X, Cambridge, MA., Harvard University Press. Simpson, S.J. (1992) Aspects of the archaeology of the Sasanian period in Mesopotamia. D.Phil. Thesis, University of Oxford. von Saldern, A. (1963) Achaemenid and Sasanian Cut Glass. Ars Orientalis, V, 7-16. Whitehouse, D. (1993) Le Verrerie IN: Overlaet, B. ed., Splendeur des Sassanides: l’empire perse entre Rome et la Chine (224-642). Brussels, Musées royaux d’Art et d’Histoire: 109-111, 257 and 264.

129

Cities Under Siege? A Look at Bronze Age Iconography1 Angelos Papadopoulos

The aim of this paper is to discuss the evidence for siege warfare in the Aegean region during the Late Bronze Age. As the surviving fortified sites and the Linear B documents do not provide satisfactory information, various images will form the basis of the discussion. Depictions of martial scenes will be presented and examined in order to suggest that, although some of these images are considered to depict settlements under siege, in fact this could not be the case in the Aegean region. A brief comparison with Egyptian and Near Eastern iconography will support this hypothesis.

method of carrying water with the help of animals never stopped (Iakovidis 1999: 202-3). On the Aegean islands, settlements were already fortified from the Early Cycladic period. Sites such as Panormos on Naxos and Kastri on Syros were relatively small in size and had small entrances and semicircular protective towers, a model for which was possibly the defensive system of EH II Lerna on the mainland (Barber 1987: 66-7). It appears that at that time the defenders were not facing very sophisticated attacks and possibly the most common way to attack a fortified site was with the use of arrows, hand to hand combat and slings (Barber 1987: 67). A large amount of stones that could have been used as missiles were found at sites such as Panormos on Naxos (Doumas 1964: 411-12) and Kastri on Syros (Tsountas 1899: 116), and, most likely, this is evidence for an attack. There is no imagery that can provide more information.

The Mycenaean civilisation of the Late Bronze Age flourished in the Aegean region during the Late Helladic period, from 1600 to 1100 BC. Probably the most characteristic feature is the palace built on a citadel which controlled the surrounding areas. It seems that the citadels were not fortified until the mid fourteenth century when the so-called ‘Cyclopean’ walls were introduced (Iakovidis 1999: 199), giving the citadels the look that survives until today, for example at Mycenae and Tiryns. The fortification walls were most likely not just for defensive purposes, but as much for status and to impress those outside the polity. The walls were almost 900m and 700m respectively in length; they reached a height of 7.5 m and a thickness of 8 m, which could reach 17 m where galleries were built into them (Iakovidis 1999: 200; Treuil, Darcque, et al. 1996: 501), more than enough to withstand any possible attack (Fig. 1). But it was not until the very end of this period that extensions were built in order to secure water supplies for the inhabitants of the citadels, as can be seen at Tiryns, Mycenae and Athens. The supply of water was achieved by digging infiltration galleries at Tiryns, constructing a pipeline to connect a spring with a subterranean fountain at Mycenae and digging a fountain in Athens; at Midea and Gla it seems that the traditional

During the Middle Cycladic period major towns like Phylakopi on Melos and Ayia Irini on Keos were fortified, but it was during the Late Cycladic period that the closest parallel to the Mycenaean citadels, the defences of the site of Koukounaries on Paros, were built. The town had two protective walls – an inner wall of almost 3.5 m thickness and an outer wall of 1.2 m thickness – constructed in a ‘Cyclopean’ way as Barber comments (1987: 68-9). Generally the defensive walls of the Cycladic towns were never as impressive and massive as the mainland ones. Apart from the fact that the final phase of the fortifications was built in order to secure water supplies for the palaces, there is no evidence from the archaeological record that would suggest the practice of siege warfare in the Aegean. In addition, contemporary documents, that is, the Linear B tablets, are not of great use as they are mainly lists and catalogues for bureaucratic purposes (Dickinson 1994: 194). Although there is mention of weapons (Chadwick, 1973: 160-81), no siege devices are described, and the only clue for an expected attack comes from the Pylian An series of tablets that lists a total of 770 watchers at 10 coastal sites (o-ka) and led by an officer (e-qe-ta)

1. I would like to express my gratitude to my supervisor Prof. Christopher B. Mee for his comments on earlier drafts and for his patience, advice and help. I am also grateful to the University of Nottingham for the Thomas Wiedemann Memorial Fund that helped me to participate in this year’s SOMA. Finally, I would like to thank Dr. Christina Televantou for her kind permission to use her drawing of the Landing Party from her work on the Miniature Frieze from Akrotiri.

131

SOMA 2004

FIGURE 1 THE FORTIFICATIONS OF TIRYNS (FRICKENHAUST 1912: 2, FIG. 1) (Chadwick 1973: 184-5; Treuil, Darcque, et al. 1996: 517). It must be mentioned, though, that a detailed analysis of these texts suggests that the term e-qe-ta has no military significance and probably the An tablets have to do with ploughing (Uchitel 1984: 162-3).

defend the settlements from the fortified walls and dead warriors can be seen. The Assyrian depictions date to a later period than the Aegean scenes, but their similarity to the Egyptian wall-paintings show that there were specific ways of practicing and depicting siege warfare in the Eastern Mediterranean that lasted many millennia.

Iconography, if studied properly, can give a valuable insight into the societies of that period, and the aim of this paper is to discuss iconographic elements that may provide information on the practice of siege warfare.

As can be seen from various examples that even date to much earlier periods, Eastern Mediterranean peoples developed siege techniques and elaborate machinery was invented and improved (Kern 1999: 16). Although there is no pictorial evidence from this period, literary evidence suggests that battering rams were in use in Mesopotamia as early as 2500 BC (Kern 1999: 16). By that time siege warfare was so advanced that large protection shields for the attacking archers were in use (Kern 1999: 51), but even then none of these devices existed in the Greek region. The same absence seems to apply to siege ladders, such as those shown in Egyptian depictions (Shaw 1991: 37-8, fig. 28; 1996: 243, fig. 2).2

Although hunting and fighting scenes seem to be a favourite subject of Mycenaean artists (Immerwahr 1990: 122-33) and they are depicted in every art form, there is not a single siege scene, at least not in the way that this is shown in contemporary Egyptian (Smith 1981; Shaw 1991, especially chapter 3) and later Near Eastern art (Frankfort 1996, especially chapter 7). The theme of siege warfare is depicted in a very similar way in the art of both regions: various wall-paintings in Egypt from the Middle and New Kingdoms depict fortified settlements under attack. The defenders are behind battlements, and they use all the means they have, such as spears and arrows, in order to resist the enemy who has surrounded the town and uses ladders to climb the fortifications and sack the city. Primitive kinds of battering rams can be seen (Shaw 1991: 17, fig. 9; Kern 1999: 16, fig. 4). The late Assyrian period stone reliefs (Frankfort 1996: 159, fig. 84; 181, fig. 206) depict armies with battering rams attacking and sacking cities and, as in the Egyptian scenes, archers

In the Aegean region no such depictions survive, and it seems unlikely that any will be found. Warfare was most likely practiced under different circumstances, and besieging an important settlement had a totally different meaning. Two compositions provide what are probably the most informative representations of an attack on a city, 2. A group of soldiers is depicted using a ladder during the siege of an Asiatic fortress, Fifth or Sixth Dynasty.

132

ANGELOS PAPADOPOULOS

FIGURE 2 DRAWING OF THE SCENE OF THE SHIPWRECK AND THE MARCHING PARTY (TELEVANTOU 1994: SUPPLEMENTARY PLANS) (All figures are not to scale) but even these are not complete enough and have been the subject of many different interpretations. They date to a much earlier period than the final phase of the Cyclopean walls, but they are the only pictorial evidence so far.

FIGURE 3 A DETAIL OF THE SILVER SIEGE RHYTON FROM MYCENAE (EVANS 1930: 93, FIG. 52) There are various other fragments of wall-paintings, pottery and stone vessels from the Mycenaean period that provide information for siege warfare, but the aforementioned compositions are the only ones that actually tell a story, and, together with a number of the fragments, they seem to be a part of a narrative cycle. Already in 1930 Evans had suggested that the theme of the ‘Beleaguered City’ was introduced to the Aegean world from Egypt, where the first iconographic examples occur at the end of the Old Kingdom (Evans 1930: 82). According to his view the MM IIIA (1700-1650 BC) Town Mosaic is the earliest representation of this theme in the Aegean, although the exact date of the mosaic is not certain (Immerwahr 1990: 68). It certainly depicts animals, plants, buildings and people with weapons (a bow and possibly spears), but its military character is strongly debated by Foster (1979: 115). But apart from the highly debatable case of the Town Mosaic, Evans puts into the same thematic cycle various fragmentary works of art, such as the Mycenaean Siege Rhyton and the stone vessel from Epidauros. The latter consists of fragments dating to LH III (1400 BC), and it shows the familiar scene of a ship near the coast and a group of individuals of military character (if they indeed carry shields) marching there (Sakellariou 1971: 3-14). Some wall-painting fragments from Knossos depicting groups of male individuals with necklaces, holding what Evans interpreted as javelins, were also considered by him to be warriors attacking enemy fortifications and throwing their spears at the defenders (Evans 1930: 82-3). It has also been argued that the wallpaintings that decorated the Throne Rooms at Mycenae and Pylos and the few fragments that survive from the possible Mycenaean palace at Boeotian Orchomenos also depict fighting before a walled city (Immerwahr 1990: 123-8). Generally, scholars believe that the theme of the

Firstly, the Mycenaean Siege Rhyton (Fig. 3), a conical libation vase was found in a very fragmentary condition in Shaft Grave IV of Grave Circle A at Mycenae and dates to LH I. Its reconstruction was a major issue as most of the fragments did not join, but it is clear that it shows helmeted warriors attacking a fortified coastal city in front of which naked individuals, possibly the defending inhabitants, are fighting back with slings, two types of bows and what seem to be shields. At a different level, bodies of dead individuals or swimmers can be seen in the water. Generally, the very compact and tightly arranged depictions of the figures and structures are reminiscent of a wall-painting arrangement, so that it has been suggested that it was created by a wall-painter (Evans 1930: 81-106; Sakellariou 1975: 195-208; Hood 1993: 196-9).3 Secondly, the miniature frieze from the West House at Akrotiri on Thera (Fig. 2) ran around all four walls of room 5, on the first floor of what seems to have been a private house. It depicts coastal towns, river landscapes, a large fleet of ships and other boats, animals and shepherds and, interestingly enough, a shipwreck scene and a group of warriors with boar’s tusk helmets, tower shields, long spears and swords who seem to form a landing party (Morgan 1988; Televantou 1994).4 The entire composition has been dated to the LC I period, the same time that the Siege Rhyton was made. 3. The bibliography on the Siege Rhyton is much larger than the one presented here, which is for basic reference and provides a good description of the artifact. 4. The bibliography on the Miniature Frieze is endless and these two publications are the most complete concerning presentation of the material, discussion of the key elements and attempts at interpretation.

133

SOMA 2004 besieged coastal city was the subject of various works of art, perhaps recollecting a historical event that stayed for many years in the memories of people as a glorious victory (Televantou 1994: 333).

eastern Mediterranean representations of siege warfare, and if their aim was to depict cities under attack, then the Aegean artists followed a different artistic vocabulary. The defenders on the Rhyton do not fire slings or arrows from the top of the walls, like the Egyptians on the wallpaintings from the Middle Kingdom tomb of Khety at Beni Hasan, circa 1900 BC on which a fortified town under siege is depicted (Shaw 1996: 244). This would cause major loses to the attacking army. This can be clearly seen by the remains of sixty Eleventh Dynasty soldiers that were buried in a mass grave in western Thebes in Egypt; many of these have suffered severe head wounds (one of them still had an arrowhead embedded in his left eye socket) most likely a result of attacking enemy walls (Shaw 1996: 240), thus siege warfare. On the mainland, the fortification walls have not survived to their original height, but it has been suggested that there were no battlements upon them (Iakovidis 1999: 200). However, a fresco fragment from Thebes shows what could be a helmeted female looking out from a square construction that could be a battlement (Immerwahr 1990: 128), but once again there are strong debates on this interpretation as there is the possibility that it may have been a kind of religious procession and not a battle scene (Boulotis 2000: 1121-6).7 According to Televantou, the black triangular projections of the building K1 of Town II are battlements (1994: 273-4); Town II is painted in the middle of the north wall of the West House miniature frieze. These triangles look very similar to the projections depicted on Egyptian, Syrian and Mesopotamian iconography already from the First Dynasty and the end of the second millennium BC respectively, in which defensive edifices, fortified settlements and fortresses are represented. In addition, the location of building K1 at the harbour of the town has led Televantou to the suggestion that it could function as a kind of fortress (1994: 274, 331).8 What may have been a representation of Aegean warriors climbing a siege ladder is depicted on a pottery fragment from Kos depicting three helmeted individuals on different levels (Fig. 4), but as the rest of the vessel is missing this is only a hypothesis (Vermeule and Karageorghis 1982: 160).

None of the above works of art with the exception of the Akrotiri paintings and the Siege Rhyton provide enough information to strongly support their interpretation as siege scenes due to their state of preservation. Therefore, it is difficult to use them as a source of information on siege tactics. As for the two aforementioned compositions, whether they commemorate a historical event or not, these scenes are the only ones in Aegean iconography that combine the elements of a fortified city and fighting parties and therefore are the closest to siege warfare. The interpretation of what was meant to be depicted on these items, especially on the Miniature Frieze, has been the subject of much debate.5 The origin of the people, the location of the scenes and the message that was supposed to be conveyed are just a few of the key debated points. Nevertheless, what is of interest for the aims of this paper is the fact that Aegeans are shown (a term that is preferable to ‘Mycenaeans’ or ‘Cycladeans’), and, therefore, their military techniques can be analysed. The fact is that these compositions do not give us much direct evidence about siege warfare. Concerning what is visible, one can see fortified coastal towns with closed gates, a rocky landscape, flora and fauna, sea, ships, bodies in the water (maybe a result of a sea-battle although no evidence supports the hypothesis that any fighting was conducted at sea on board ships at this time),6 a land battle (in the case of the Rhyton) and maybe battlements in the case of the frieze. However, no siege devices are visible, and it seems that the town on the frieze was not fortified. These compositions do not share elements with the 5. Televantou (1994: 309-38) presents the many different interpretations based on various parallels. 6. The earliest pictorial evidence showing a definite naval battle in the Aegean region dates to the LH IIIC period and comes from the settlement of Pyrgos Livanaton in central Greece. Fragments of large kraters depict warriors on the deck of an oared ship preparing to attack another ship and a scene of fighting that takes place on board a vessel. Various fragments of schematic warriors survive. The ships are probably near the coast, but almost definitely they are in the open sea as is suggested by the presence of fishes. No rams are shown and therefore it differs from a naval battle in the Classical period (Dakoronia 1987: 117-22; 1999: 119-28; 2002: 28390). It has been suggested that long spears, the Homeric ναύμαχα ξυστά were used as weapons for sea battles (Dakoronia 1999: 221). Marinatos identified as such the long spears that are depicted on the Miniature Frieze fleet (Televantou 1994: 221). Fighting between the longboats where spears would reach the enemy ships or even skirmishes could have also been a reality during the Early Bronze Age (Broodbank 2000: 253). Aggressive scenes on board ships are known from the Early periods as well. A stone plaque from Naxos, thought to date to EC II, show two individuals facing each other in a ship in a very violent way (Doumas 1966: 53, fig. 7) and the EC III-MM I seal from central Crete also depicts two individuals on board a ship facing each other, but this time they most likely hold weapons (Hatzi-Vallianou 1979: 384). In the case of the miniature frieze only the wrecks of ships are shown, the moments immediately after the action. As this is not enough evidence that the result of an open sea battle is represented, the possibility that these ships were destroyed on the beach by land troops seems plausible.

So what does all this evidence tell us about siege practices? It has been suggested that the narrative cycle, to which the two compositions belong, describes raids on rich coastal 7. For a photograph of the fragment, see Morgan (1988: fig. 156). Boulotis comments that there are two figures appearing from two rectangular architectural features, but their gender is uncertain (2000: 1121-6). It has been suggested that these paintings are copies of the well-known ivory pieces that depict the heads of helmeted warriors or that they may show a religious procession and because of the white skin the figures cannot be anything else but female. The problem centres on a small black line that starts from the eyebrows and reaches the chin that could either be a beard or leather part of the helmet. After considering and discussing various suggestions, Boulotis concludes that the scene possibly represents a female warrior deity. 8. Televantou comments that the triangles depicted on top of some of the buildings of the Siege Rhyton could be interpreted as battlements as well and that these features can be paralleled to the miniature frieze Town II projections (Televantou 1994: 274; 331).

134

ANGELOS PAPADOPOULOS A possible explanation using this evidence is that, at least on the islands, enemy forces would reach and land on the coast. The inhabitants would rush behind the city walls, and the enemy would attack them, as in the case of the bearded archer who shoots at a high target shown on a fragment of a serpentine rhyton from Knossos (Evans 1930: 100, fig. 59; Warren 1969: 85, P 473; 1979: 127). Recognising the danger of loosing their property, the defenders would come out to counter attack. But would the richly buried individual in Grave IV at Mycenae have ordered a precious item to be made for him showing nothing more than a piratical raid? On the other hand, as has been mentioned above raids seem to have been a common practice during later periods in the Greek world. War was a profit-making enterprise (Sage 1996: 17). Therefore, attacking a wealthy settlement to acquire goods and supplies may have been an act worth commemorating. Another explanation is possible: namely that the actual target of the attacks were the harbours. The shore side buildings that are depicted on the wall-paintings not only from Thera but also from Keos could have been used as storage facilities for nautical gear, trade goods and even certain types of ships, in order as Shaw points out ‘to avoid carrying it back from town to shore’ (1990: 433). The archaeological record supports this hypothesis as in the case of the Kommos harbour constructions (Shaw 1990: 425-6). A successful raid would involve looting the goods that were stored there and also causing damage to important buildings. If indeed for the islanders trade was the lifeblood of their communities, then it is logical to assume that destroying or conquering important ports would give the attacker a clear advantage. Maybe the destroyed ships shown at Akrotiri were the result of such an attack, although the most common interpretation is that this was the enemy’s fleet destroyed after the allied Aegeans went to help a friendly city somewhere in the eastern Mediterranean (Televantou 1994: 338).

FIGURE 4 A KRATER FRAGMENT DEPICTING THREE HELMETED INDIVIDUALS (RE-DRAWN BY AUTHOR AFTER VERMEULE AND KARAGEORGHIS 1982: FIG. XII 30). towns for booty, animals and women (Warren, 1979: 129). Piracy is known from earlier times in the Aegean (Renfrew 1972: 262-4, 398-9) and was common later, for example in the Classical period when traditional Greek warfare often aimed not to conquer enemy towns, but to steal livestock and destroy crops (Kern 1999: 91). A stone fragment depicting a man dragging a goat, where a boar’s tusk helmet is also visible could support this hypothesis (Fig. 5) (Warren 1969: 89; 1979: 126-7). Linear B tablets from Pylos mention the place of origin of groups of women from outside the Mycenaean world, using the word ‘slaves’ for some and in one case even the word ‘captives’ for a specific group, but not for the rest. ‘Slaves’ may have been acquired from trading posts in Anatolia, but also from raids (Chadwick 1973: 166-8; 1976: 80).

As there is no pictorial evidence showing an actual siege in the Aegean world, one is tempted to assume that the local peoples were unaware of such practices. Although the absence of evidence is not necessarily evidence of absence, this may be the case. If their aim was not to capture a town, armies would simply strike at key structures and destroy ships and supplies. In the Aegean world, the technological ability and previous experience were available in order to do that, as in the Mycenaean period a ship type was developed that was ‘well adapted for lightning visits to recalcitrant subjects, as well as opportunist raids on non-aligned settlements’ (Wedde 1991: 92). On the mainland, one would expect the fortified citadels to be the last resort of a limited number of people, possibly the royal family and the élite, and the fighting would take place outside the walls, where most of the people would have had their houses and farms. Unlike the island fortifications, defensive works at citadels were meant to protect the megaron and the king, the storage areas and some workshops (Treuil, Darcque, et al. 1996: 501), but not the majority of the population, although

FIGURE 5 A STONE FRAGMENT SHOWING A MAN DRAGGING A GOAT; A BOAR’S TUSK HELMET IS CLEARLY VISIBLE AT THE LOWER PART (EVANS 1930: 185, FIG. 128) 135

SOMA 2004 the length of these walls would suggest that at a time of crisis many defenders must have been available inside the citadel in order to defend it properly and not leave large exposed gaps.

Frankfort, H. (1996) The Art and Architecture of the Ancient Orient, 5th ed. Yale University Press/ Pelican History of Art. Frickenhaust, A. (1912) Tiryns II: Die ergebnisse der ausgrabungen des instituts, Kaiserlich Deutsches Archaeologisches Institut an Aten. Hatzi-Vallianou, D. (1979) Κεντρική Κρήτη, Νομός Ηρακλείου, Αρχαιολογικόν Δελτίον 134 (β2), 382-5. Hood, S. (1993) Η Τέχνη στην Προïστορική Ελλάδα, Αθήνα, Εκδόσεις Καρδαμίτσα. Iakovidis, S. (1999) Late Helladic Fortifications. IN: Laffineur, R. ed. Polemos. Le context guerrier en Égée à l’ âge du Bronze, Actes de la septième Rencontre égéenne internationale, Université de Liège, (1418 Avril 1998), Aegaeum 19, Université de Liège, Histoire de l’art et archéologie de la Grèce antique and University of Texas at Austin, 199-204. Immerwahr, S.A. (1990) Aegean Painting in the Bronze Age, Pennsylvania, The Pennsylvania State University Press. Kern, P.B. (1999) Ancient Siege Warfare, Bloomington and Indianapolis, Indiana University Press. Morgan, L. (1988) The Miniature Wallpaintings of Thera: A Study in Aegean Culture and Iconography, Cambridge, Cambridge University Press. Renfrew, C. (1972) The Emergence of Civilisation: The Cyclades and the Aegean in the Third Millennium B.C., London, Methuen and Co Ltd. Sage, M.M. (1996) Warfare in Ancient Greece: A Sourcebook, London and New York, Routledge. Sakellariou, A. (1971) Scène de bataille sur un vase Mycénien en pierre? Revue Archéologique, Fascicule 1. Sakellariou, A. (1975) La scène du “siège” sur le rhyton d’argent de Mycènes d’après une nouvelle reconstitution. Revue Archéologique, Fascicule 2. Shaw, J.W. (1990) Bronze Age Aegean Harboursides. IN: Hardy, D.A., Doumas, C.G. Sakellarakis, J.A. and P.M. Warren, eds., Thera and the Aegean World III. Vol. 1: Archaeology. Proceedings of the Third International Congress, Santorini, Greece, 3-9 September 1989, London, The Thera Foundation, 420-36. Shaw, I. (1991) Egyptian Warfare and Weapons, Princes Risborough, Shire Publications Ltd. Shaw, I. (1996) Battle in Ancient Egypt: The Triumph of Horus or the Cutting Edge of the Temple Economy? IN: Lloyd, A.B. ed., Battle in Antiquity, London, Duckworth, 239-69. Smith S.W. (1981) The Art and Architecture of Ancient Egypt, 2nd ed. revised by W.K. Simpson, New Haven/ London, Yale University Press/ Pelican History of Art. Televantou, C. (1994) Ακρωτήρι Θήρας. Οι Τοιχογραφίες της Δυτικής Οικίας, Η εν Αθήναις Αρχαιολογική Εταιρεία. Treuil, R., Darcque, P., Poursat, J-Cl. and Touchais, G. (1996) Οι Πολιτισμοί του Αιγαίου κατά τη Νεολιθική και την εποχή του Χαλκού, Αθήνα, Εκδόσεις Καρδαμίτσα.

To conclude, the lack of siege warfare iconography, at least in comparison with the rich pictorial evidence from Egypt and the Near East, may well mean that the Aegeans did not use it on a regular basis, but preferred sudden strikes. Even if they were besieged, it is unlikely that they would depict their own settlements being attacked (despite the fact that they probably survived to tell the story). But towards the end of the Bronze Age, a time of danger and tension in the whole of the eastern Mediterranean, they may have had to face this kind of warfare, and it was only then that they tried, unsuccessfully it seems, to protect themselves. Bibliography Barber, R.L.N. (1987) The Cyclades in the Bronze Age, Iowa City, University of Iowa Press. Boulotιs, Ch. (2000) Η τέχνη των τοιχογραφιών στην Μυκηναïκή Βοιωτία, Αραβαντινός, Β. επιμ. Επετηρίς της Εταιρείας Βοιωτικών μελετών, τόμος Γ’, τεύχος α’, Αρχαιολογία, Γ’ Διεθνές Συνέδριο Βοιωτικών Μελετών, Θήβα 4-8 Σεπτ. 1996, 1095-154. Broodbank, C. (2000) An Island Archaeology of the Early Cyclades, Cambridge, Cambridge University Press. Chadwick, J. (1973) Documents in Mycenaean Greek, 2nd ed., Cambridge, Cambridge University Press. Chadwick, J. (1976) The Mycenaean World, Cambridge, Cambridge University Press. Dakoronia, F. (1987) Warships on sherds of LH III C kraters from Kynos. IN: Tzalas, H. ed., TROPIS II: 2nd International Symposium on Ship Construction in Antiquity, Delphi, Proceedings, Athens, Hellenic Institute for the Preservation of Nautical Tradition, 117-22. Dakoronia, F. (1999) Representations of sea-battles on Mycenaean sherds from Kynos. IN: Tzalas, H. ed. TROPIS V: 5th International Symposium on Ship Construction in Antiquity, Nauplia 1993, Proceedings, Athens, Hellenic Institute for the Preservation of Nautical Tradition, 119-28. Dakoronia, F. (2002) Further finds from Kynos. IN: Tzalas, H. ed. TROPIS VII: 7th International Symposium on Ship Construction in Antiquity, Pylos 1999, Proceedings, Athens, Hellenic Institute for the Preservation of Nautical Tradition, 283-90. Dickinson, O. (1994) The Aegean Bronze Age, Cambridge, Cambridge University Press. Doumas, Ch. (1964) Κυκλάδες, Νάξος, Αρχαιολογικόν Δελτίον 19, 411-12. Doumas, Ch. (1966) Κορφή τ’ Αρωνιού, Αρχαιολογικόν Δελτίον 20 (α), 41-64. Evans, A.J. (1930) The Palace of Minos at Knossos, vol. IIΙ, London, MacMillan and Co Ltd. Foster, K.P. (1979) Aegean Faience in the Bronze Age, New Haven and London, Yale University Press. 136

ANGELOS PAPADOPOULOS Tsountas, Ch. (1899), Κυκλαδικά ΙΙ. Αρχαιολογική Εφημερίς, 74-134. Uchitel, A. (1984) On the ‘Military’ Character of the o-ka Tablets. Kadmos 23.2, 136-63. Vermeule, E. and V. Karageorghis eds. (1982) Mycenaean Pictorial Vase Painting, Cambridge, Mass., Harvard University Press. Warren, P. (1969) Minoan Stone Vases, Cambridge, Cambridge University Press. Warren, P. (1979) The Miniature Fresco from the West House at Akrotiri, Thera and its Aegean Setting. Journal of Hellenic Studies, 99, 115-29.

Wedde, M. (1991) Aegean Bronze Age Ship Imagery: Regionalisms, a Minoan Bias, and a ‘Thalassocracy’. IN: Laffineur R. and L. Basch eds. Thalassa: L’ Ègée Prèhistorique et la Mér, Actes de la troisième Rencontre égéenne internationale de l’ Universitè de Liége, Station de recherches sous-marines et océanographiques (StaReSo), Calvi, corse (2325 avril 1990), Aegaeum 7, Université de Liège, Histoire de l’art et archéologie de la Grèce antique, 73-94.

137

Symbols of Superiority: Western Greek Votive Offerings at Delphi Elvira Rocca

It has been assumed by most modern scholars that an ongoing interaction existed between the Magna Grecian colonies and the panhellenic sanctuary of Delphi involved in their foundation. This paper will examine, through literary, epigraphic and archaeological evidence, how and why this relationship was maintained and articulated between the late Archaic and the early Classical period. The display and the location of the western Greek offerings in the Apollonion may provide the answer. Religion is the obvious medium through which veiled political messages were expressed. Competition and rivalry between mainland cities as well as among western Greek colonies were reflected materially by erecting monuments to the wealth and success of various states designed to impress other dedicants. In order to explain the symbolic value of votives, we shall look firstly at the overall layout of the sanctuary in order to investigate the relationship between the topographical features of the offerings and the historical events commemorated. In the second part, we shall attempt to apply this general assumption to a detailed study case: the tripod of Croton.

sense. On the one hand, events such as the war of Croton, Metapontum and Sybaris against Siris (in the middle of the sixth century BC), the Battle of the Sagras River fought between Locri Epizephyrii and Croton (during the first half of the sixth century BC), the destruction of Sybaris inflicted by the army of Croton near the Traeis River (510 BC), and the Rhegine victory over Hyele and the foundation of Pyxous (about 471/0 BC) as requested by the regent of Anaxilas, Mikytos ‘of Rhegion and Messene’, all attest to the long-term oracular activity dealing with western Greek affairs. On the other hand, it is necessary to take into account the nature of western Greek dedications, linked by ancient sources to specific events such as victorious battles, or to give thanks to Apollo for his goodwill (Strabo VI 275 C; Diod. Sic. V, 9; Paus. V, 27,1; X, 11, 5; X, 16,7). Their location might help us to understand their historical origins in the light of the panhellenic context. In the late Archaic temenos, prestigious archaeological finds have marked three likely places on the main route towards the great temple (Amandry 1977: 273-93; Bousquet 1985: 725-6) as the main hubs of an ‘architectural project of the Apollonion’ (Chaney 1991; Jacquemin 1999: 245): the Aire, where tradition sets the well-known omphalos, symbol of the goddess Ge and centre of the world (Strabo IX 3,12 C 422), the first stretch of the Sacred Way and the terrace of the temple (Hansen 1960: 423-33; La Coste Messelière 1969: 730-58; Laroche 1988: 291-305; Hansen 1992: 1257; Amandry 1993: 273-6; Partida 2000: 29-30). According to mainland donors, Magna Grecian dedicants respected these topographic prescriptions. Western state votives were set up near the Athenian public offerings. Near the main entrance to the sanctuary, close to the dedication of Marathon, the Cnidian people of the Aeolian Islands, the Tarantines and the Rheginoi built their votive offerings (Strabo VI 275 C; Paus. X 10, 6; X 13, 10). Next to the Sitalkas Apollo (Paus. X 15, 2), in front of the temple, the fame of the Greek ethnos is recalled by the upper Tarantine votive and by the tripods of the Deinomenes (Plut. De Pyth. Or. 8 397 E; Diod. Sic. XI 26, 7), while the local Achaean struggles of southern Italy are emphasised by the tripod of Croton (see Fig. 1). Their location illustrates

The Religious and Historical Route of the Pythian Temenos From the early colonial period, the ancient relationship established between many western Greek colonies and the Apollonion of Delphi was founded on two different values, religious tradition and political needs (Hdt. I 165; Plato. Resp. IV, 427 b-c; Leg. 738 b-d). During this period, the Pythian oracle was the spiritual centre of many groups of mainland colonists who left for southern Italy. As a mothercity oikist, Apollo assumed an ethical and normative role on behalf of the new Greek communities founded under his auspices (Parke and Wormell 1956: 49-79; Forrest 1957: 162-75; Amandry 1959: 400-47; Piccirilli 1972: 44-8; Valenza Mele 1977: 493-524; Malkin 1987: 47-52; Rougemont 1991: 157-92; Lévêque 1996: 1112-15). In the late Archaic period the Delphic oracle seems to renew its role. A sequence of literary evidence backs up the connection between the Pythian oracle and western Greek political events (Strabo VI 1, 10, C 241; Just. XX 2, 35). Such an active bond should be interpreted in a double 139

SOMA 2004

FIGURE 1 TEMENOS OF APOLLO, DELPHI (ATLAS, RELEVÉS EXÉCUTÉS PAR UN GROUP D’ARCHITECTES DANOIS, PLATE III) the desire of Magna Grecian cities to include their own history in the mainland history in order to strengthen the unity of Greece and celebrate it. Moreover the language of the votives sets up a symbolic space where many rivalries between Greek cities are brought to life again. The votive monument of the Nauarches, dedicated by Sparta after its triumph over the Athenian fleet at Egospotami (405 BC), was built in front of the so-called ‘basement of Marathon’ (Atlas 110) taking away its privileged space near the main entrance of the sanctuary (Bommelaer 1992a: 281-6; Jacquemin 1999: 251-2). The placing might be compared with the setting up of the tripod of Croton and the upper monument of Taras next to the votives of the Deinomenes. This offering location would better express the belligerent situation between these two poleis and Syracuse, which begins to interfere in Magna Grecian affairs during the first

half of the fifth century BC as related by many ancient authors (Schol. Pind. Pyth. II 36c, 38, Strabo VI 1, 5 c.255; Paus. VI, 13, 1; Just. XXI 3, 2). Through the display of votives, the temenos of Delphi becomes a cosmopolitan recipient of political messages addressed to those present at the moment of the dedication and even more so to subsequent visitors (Morgan 1993: 18-20; Shepherd 2000: 68-70). It means that these offerings, as mnèmata, claim a connection between the permanence of the gift to Apollo and the memory of actual events (de Polignac 1996b: 63; Osborne 2004: 4-5). The custom of ritualised competition at Delphi from about 800 BC (de Polignac 1996a: 11-13; 1996b: 636) is very much resumed by the tripod of Croton, an exemplary model of western Greek votives. This gives us 140

ELVIRA ROCCA

FIGURE 2 NORTH-EAST REGION, APOLLONION OF DELPHI (ATLAS, RELEVÉS EXÉCUTÉS PAR UN GROUP D’ARCHITECTES DANOIS FIG. 11) the evidence required to explain the ‘Delphic pattern’ by which competitions and rivalries were transferred to this interstate context through commemorative objects. The Tripod of Croton During archaeological excavations in the north-eastern region (Fig. 2), Homolle, persuaded by many pilgrims’ accounts, tells us that ‘in this region the base of the Tripod of Plataea was recognised’ (1898: 572-9) as described by Herodotus (IX 81) and Pausanias (X 13 9). The nineteenth century topographical surveys certainly assigned the Atlas, Relevés exécutés par un group d’architectes danois 408 monumental base to the bronze Snake column (Sismondo Ridgway 1977: 374-9; Bommelaer 1992b: 178-9; Laroche 1993: 631-3 fig. 19). Although modern analyses have been carried out which favour this solution, two weak points remain: the place where the fragments were discovered near the Tarantine offering site described by Pausanias XIII, 9-10 and the shape of the stone and the gamma sealings (Amandry 1987: 79-131). Moreover, the cavity for the location of the column, the base of which was missing, was identified in a fragment of the ‘G’ base discovered under the ‘C’ entrance. This observation led Laroche (1989: 185-6, 191-6) to argue that on the basis of their structural relationship, this fragment and another one of the same base (G) might have both belonged to the offering of Plataea; both elements set up a bell-shaped support linking the quadrangular base to the vertical cylindrical support. The doubts of Amandry (1987: 103) about the authenticity of the lower side, rebuilt hastily with various fragments on the Atlas, Relevés exécutés par un group d’architectes danois 408 base in 1903, have been refuted by the final proposal of Jacquemin and Laroche (1990: 304-7): i.e. that this base and the fragments of the dedication of Croton to Apollo, discovered in 1894 to the east of the Upper Tarantine offering, are connected structurally (Bourget 1929: 1-3).

FIGURE 3 REBUILDING OF THE ATLAS 408 SOCLE (JACQUEMIN AND LAROCHE 1990, FIG. 9) base, have shed light on the real structural role of the marble base wrongly rebuilt on the Atlas, Relevés exécutés par un group d’architectes danois 408 foundation (Fig. 3). According to Jacquemin and Laroche, this base must be recognised as a support element of the monumental tripod dedicated by Croton to Apollo during the first half of the fifth century BC. This alternative architectural restoration shows a sequence of three squared bases, supported by an euthynteria base, over which there are two circular calcareous bases decorated by a third base (Laroche 1993: 631-41) (Fig. 4). Moreover, the correspondence of their diameters and the place where these inscribed fragments were discovered seems to

New examinations of these fragments inscribed on a circular marble block, and the resetting of the original squared bases found among reused elements of the Atlas, Relevés exécutés par un group d’architectes danois 408

FIGURE 4 RECONSTRUCTION OF THE ORIGINAL BASE (LAROCHE 1993, FIG. 20) 141

SOMA 2004 Material evidence drawn from the architectural reconstruction of the tripod has revealed the most appropriate restoration of this inscription, which appears to be closely linked to the historical background of the dedication (Jacquemin and Laroche 1990: 304-7, 312-18). Two crucial points allow us to recognise it: a) according to the length of the rebuilt circular base, ca. 26/27 letters are missing (Jacquemin and Laroche 1990: 304-5); b) the name of the defeated enemy appears in the genitive case, with or without the  preposition on many inscribed votives dedicated to Pythian Apollo in the first half of the fifth century BC (Lazzarini 1976: 163-6; Jacquemin 1995: 144 n.4; 1999: 92-3). This argument invites us to accept the modern restoration: ΚΡΟΤΟ[Ν]ΙΑΤΑΙ [ΑΠΟ ΣΥΒΑΡΙΤΟΝ ΑΠΟ]ΛΛΟΝ[Ι ΑΝΕΘΕΣ]ΑΝ (Jacquemin and Laroche 1990: 322) rather than the traditional ΚΡΟΤΟ[Ν]ΙΑΤΑΙ Α[ΠΟ]ΛΛΟΝ[Ι ΑΝΕΘΕΚ-or ΔΕΚΑΤ]ΑΝJeffery 1961: 258-9; 1990: 258; Lazzarini 1976: 306) because the former, taking its rightful place on the rebuilt base, correctly assigns a space of nine or ten letters to the mention of the origin of the war booty, as was the case with other Delphic dedications. Then on the basis of the letter-forms, dated earlier in the first quarter (Bourget 1929: 1-2) or in the second quarter of the fifth century (Jeffery 1961: 258-9; Lazzarini 1976: 306; Jeffery 1990: 249, 257-8), we should assume that this tripod was undoubtedly consecrated by Croton within the first half of the fifth century (Maddoli 1984: 331-2) to display to the Delphic community a military victory against its Achaean rival. But this raises the broader question of whether the literary and material evidence offer us any indication of the likely success of the war related to this dedication?

FIGURE 5 MODERN ANASTYLOSIS. EAST SIDE (LAROCHE 1993, FIG. 21) support the setting of the base of Croton over the Atlas, Relevés exécutés par un group d’architectes danois 408 base. This hypothesis was confirmed by the recognition of extraneous elements in the anastylosis of 1902 and by the resetting of the original blocks on the offering dedicated by Croton (Fig. 5). The Dedicatory Inscription and its Historical Context After the discovery at the end of the nineteenth century (1893-1895), of four blocks of a circular base inscribed with a dedication by Croton to ApolloFig. 6), most early scholars turned their attention to investigating the original function of this monumental base with particular regard to the restoration of the small, neat inscription (SIG.3 n.30; Supplementum Epigraphicum Graecum XLV n. 494; Karo 1909: 233-7). Firstly Hauvette and Pomtow, followed by Bourget (1911: 149-62; 1929:1), classed it among the public Delphic inscriptions: ΚΡΟΤΩ[Ν]ΙΑΤΑΙ (Α)-[ΠΟ]ΛΛΩΝ[Ι ΑΝΕΘΗΚ]ΑΝ [ΔΕΚΑΤΑΝ].They remarked that the dedication may have commemorated a votive gift made by Croton to Apollo as a tithe from the spoils taken from defeated enemies. This early allusion to belligerence as the original context of this inscription and of the dedicated object, acquires validity in light of the modern rebuilding of the tripod.

In recent years, historical interpretations have been renewed by Jacquemin’s assumption (1995: 144; 2001: 106). In support of a late dating, she connects this votive dedication to the victory of Croton over Sybaris in 477/6 BC and identifies the defeated enemies as the Syracusans. This assertion seems to be justified by a passage of Diodorus (XI 48,4-5), in which he stresses the military alliance that Sybaris, anxious to recreate a new independent community after the demise of 510 BC, forms in these years with Syracuse against Croton (De Sensi Sestito

FIGURE 6 DEDICATORY INSCRIPTION (BOURGET 1929, FIG. 1) 142

ELVIRA ROCCA and Olympia, the possibility of a private dedication made by the fifth-century tyrant Kleinias also disappears. The tyrant dedicatory inscriptions generally respect a typical formulary: the name of the dedicant-tyrant who offers the votive to the divinity at his own expense is expressed in the nominative case; there is no mention of the political community to which he belongs (Lazzarini 1976: 218 n.298, 219 n. 307, 300 n.858). The name of the people seldom follows that of the worshipper-tyrant (Lazzarini 1976: 164, 317 n.964). In our case, this public inscription indicates that this tripod was built from a war booty won by free citizens of Croton. Consequently, this offering acquires a specific state value because it had been erected to display to the international Greek community of the Apollonion, a war success in which the whole polis took part.

1987: 258-62; Giangiulio 1989: 260-77; De Sensi Sestito and Intrieri 1992: 46-7; Lombardo 1993: 284-9). Although this hypothesis is well argued, there remain several unanswered problems with it: a) the greatest military success of Croton was, of course, the victory over Sybaris in 510 BC on the banks of the Traeis River (Hdt. V 44-45, 47; VI 21; Diod. Sic. XII 9, 2-10; Strabo VI 1, 13 C 263); b) the rapid decline of Croton, from the end of the second decade of the fifth century, seems to preclude the 477 BC victory over Sybaris as the origin of the dedication; c) in the historical sources there is no mention of the triumph of Croton in 477/6 BC. Firstly, a closer look at the ‘post-510 BC’ events should be instructive. Although the expansionist policy of Croton made political and material capital out of the rich region of Sybaris, Iamblichus (VP 248-264) informs us about a series of inner contrasts and political tensions between the Pythagoreans and the rest of the population due to the territorial management of the recently conquered area (Mele 1982: 55-80; 1984: 29-33; De Sensi Sestito 1983: 38-55; Musti 1990: 35-65; Osanna 1992: 180-2; Lombardo 1993: 265-7). Many historical sources also set the revolt against Pythagoras, headed by the aristocrat Cylon (Iambl., VP 248) and the tyranny of Kleinias (Bicknell 1976: 5-25; Mele 1982: 51; Lombardo 1993: 267) in this period. Moreover, at the same time, the alliance between the Deinomenes of Syracuse and Locri causes Croton to lose many territories previously inherited from Sybaris (De Sensi Sestito and Intrieri 1992: 45-8). These historical clues might well clarify why Croton was unable to organise a defensive attack against Sybaris between the first and the second quarter of the fifth century. In order to verify the last objection one other possibility remains. It would therefore be appropriate to point out a passage of Polyaenus (Stratag., I 29, 2) in which he refers to war-captives captured by Ieron. If we connect this passage to the expedition of Polyzalos on behalf of Sybaris (Schol. Pind., Ol. II, 29b; Diod. Sic., XI, 48,4-5) and hypothesise that these captives came from Croton, positive evidence to the contrary can be inferred from the argument ex absentia: a Syracusan victory over Croton in 477 BC (De Sensi Sestito 1981: 634; 1987: 260). This might also explain the lack of any reference to this Achaean success in the literary sources. Therefore, on the one hand, in light of the history of Croton during the first half of the fifth century, no successful events allow us to date the dedication of the tripod to Apollo to the 470s, and on the other hand, a date as early as ‘post-510 BC’ might be suggested not only by political events, but also by epigraphic and stylistic evidence (Krumeich 1991: 50-2; Lippolis, Garraffo, et al. 1995: 306-9). Indeed this earlier context does not contradict the date suggested by the letter-forms (Jeffery 1990: 258). Although it would seem that this inscription was engraved by a Phokian mason, its neat and squared lettering might attest to the use of a formal Phokian Delphic script, as shown by other votive Delphic inscriptions (Jeffery 1990: 102-3). More to the point, if we compare the legend ΚΡΟΤΟΝΙΑΤΑΙ to the inscribed votives dedicated by Greek tyrants at Delphi

Other material clues drawn from the monumental base support an earlier dating for this dedication. Le Roy’s (1967: 80-4) and Mertens’ (1984: 214-17) stylistic analyses have highlighted some connections between a roof of a votive offering found in the temenos of Zeus at Olympia and another one belonging to a ‘Delphic treasure’ with decorative features similar to the typical late archaic roofs of Croton (architectural terracottas) from the second half of the sixth century BC (Shoe 1952: 22-32; Mertens Horn and Viola 1990: 244-8; Aversa 1992: 212-14; Mertens 1993: 113-15). As well as stylistic links permitting these scholars to connect the roofs to sacred buildings dedicated by Croton in both sanctuaries after the famous 510 BC victory, material evidence allows us to put the tripod of Croton in the same chronological framework. The stylistic pattern of its base recalls the ionic terracotta mouldings of the late Archaic western Greek roofs and the mixed style of the ‘Achaean eclecticism’ in which Croton played an innovative role between the end of sixth and the beginning of the fifth century BC (Mertens 1984: 214-24). At this point, textual and material evidence drawn from this analysis allow us to go beyond traditional assumptions regarding the lack of a provable historical context for the tripod of Croton (Jacquemin and Laroche 1990: 3223). It is likely that it celebrated the warranted success of Croton over Sybaris in 510 BC. Through this dedication, the Crotoniates wished to display their victory within the interstate sanctuary of Delphi as its topographic location seems to confirm. The Placement of the Votives Epigraphic material from the third century BC testifies that the authorities in charge of the Apollonion seemed to respect an urban planning of the temenos (Sylloge Inscriptionum Graecarum n.523; Corpus des inscriptions de Delphes. I. Lois sacrées et règlements religieux IV 44; Corpus des inscriptions de Delphes. I. Lois sacrées et règlements religieux IV 33, 28-31). They entrusted the maintenance of order within the sanctuary precinct to the architects of the temple who had also to supervise the placement of the votives. Worshippers who wanted to dedicate an offering 143

SOMA 2004 were subject to the permission of the persons in charge of that sacred space. Nonetheless, as the valuable study of van Straten has pointed out, sometimes for practical reasons the religious authorities laid down certain rules in order to restrict the placement of ex votos (1981: 65-151; 2000: 190223). This evidence is well documented by an inscription from the inventories of the Athenian Asklepieion (Lois sacrées des cités grecques. Supplément 107; Linders 198990: 282-5; 1992: 9-12; Aleshire 1989: 72-88; 1992: 85-98) and by another inscription from Athens (Lois sacrées des cités grecques. Supplément 123; van Straten 2000: 21314) that testify to the authority of the temple personnel to forbid the placement of votives in certain areas of the sacred precincts. But as with the Delphic decrees, we can date these inscriptions no earlier than the third or the second century BC. This means that there is no material evidence to support the assumption that this procedure was to be followed in the Archaic period and up to the fourth century BC when the custom of overcrowded Greek sanctuaries still caused the irritation of Plato (Leg. 909e-910a). We should not be surprised therefore if in the late Archaic period the amphiktyones might have granted the dedicants’ wishes for the placement of their votives (Herondas 4. 19-20; Jacquemin 1999: 103-5) and allowed a disorderly placement of ex votos in peoples’ favourite places. In these areas tripods and pillars were erected and placed high off the ground, taking up lesser sacred space. For instance, in the north-east region of the Apollonion, topographic surveys have unearthed a few humble offerings dated after the fire of 548 BC when this temenos was rearranged; it is only after 480-479 BC that a large quantity of these votives accumulated in this area (La Coste Messelière 1936: 7 n.1, 472 n.3, 481-3; Jacquemin 1991: 250-8). Although many scholars have regarded this hoarding of votives as starting after the dedication of the Plataean tripod (see Meiggs and Lewis 1971: 60-2; Rougemont 1991: 159-67; Jacquemin 1992: 196), its date must be advanced to the early years of the fifth century BC when Croton dedicates its tripod. By dating this votive offering after 510 BC, we have confirmation of its ideological and historical value. As a consequence of the victory over its Achaean rival, Croton, wishing to consolidate its political/religious interaction with this interstate sanctuary, wanted to show itself as the leader of the Achaean koine of southern Italy. Thus, while the Plataean tripod inspires the tripods of the Deinomenes, the Upper Tarantine votive offering and that of the Cnidian people in the name of that ‘national’ feeling that strengthened during the Persian War, the tripod of Croton still seems to be part of a local reality in which Magna Grecian cities fight each other. Only during the Persian War was Croton able to participate fully in the ‘Hellenic spirit’ and overcome its confined context (Paus. X 9,2). Indeed this tripod, symbolising current political interests, provides us with revealing clues as to how the Delphic dedications should be interpreted.

Apollonion. The Pythian winners as well as the ancient worshippers who looked at these votive offerings were attracted by antagonism rather than by religious piety. Sacred objects shone with political implications; ancient myths and religious rituals were mixed with actual events. Pragmatism and self-interest, lying behind ritual practices, assured the maintaining of western colonial identities. Here the Pythian oracle that legitimised the foundation of the Magna Grecian colonies is recalled as a witness to their historical development. In this sacred scenery where symbols and practices of religious traditions were used as symbols of superiority, political messages acquire an imperishable memory. Each dedicant promotes his own interests through precious ex-votos by recording his general prosperity and wealth, and in so doing he guarantees his memory for posterity. Like many mainland dedications, the Magna Grecian votives, in adapting this Delphic tradition, illustrate the history of each polis, which has been handed down over the centuries from ancient pilgrims to modern tourists. Bibliography Aleshire, S. (1989) The Athenian Asklepieion, Amsterdam, Gieben. Aleshire, S. (1992) The Economics of Dedication at the Athenian Asklepieion. Boreas, 21, 85-98. Amandry, P. (1959) Oracle, littérature et politique. Revue des études anciennes, 61, 400-47. Amandry, P. (1977) Statue de taureau en argente. Bulletin de correspondance hellénique Suppl., IV, 273-93. Amandry, P. (1987) Trépieds de Delphes et du Péloponnèse. Bulletin de correspondance hellénique, 111, 79-131. Amandry, P. (1993) Notes des topographie et d’architecture delphique. IX. L’opisthodome du temple d’Apollon. Bulletin de correspondance hellénique, 117, 263-83. Aversa, G. (1992) Intervento. Atti del Convegno di Studi sulla Magna Grecia, 31, 212-14. Bicknell, P.J. (1976) The Tyranny of Kleinias at Croton. Klearchos, 18, 5-25. Bommelaer, J.F.(1992a) Monuments argiens de Delphes e d’Argos. Bulletin de correspondance hellénique, Suppl., XXII, 265-93. Bommelaer, J.F. ed.(1992b) Delphes. Centenaire de la “Grande Fouille” réalisée par l’École Française d’Athènes (1892-1903). Leiden - New York, Brill. Bourget, È.(1911) Inscriptions de Delphes. Bulletin de correspondance hellénique, 35, 149-76. Bourget, È (1929) Inscriptions de l’entrée du Sanctuaire au Trésor des Athéniens, Fouilles de Delphes, III. 1. Athènes-Paris, École Française d’Athènes. Bousquet, J. (1985) L’hoplothèque de Delphes. Bulletin de correspondance hellénique, 109, 717-26. Chaney, F. (1991) Delphi and Kos: The Symbolic order of two Greek Sanctuaries. M. Phil. Degree, Cambridge University. De Polignac, F. (1996a) Mediation, Competition and Sovereignty: The Evolution of Rural Sanctuaries in Geometric Greece. IN: Alcock, S. and R. Osborne eds.,

Religion is the common filter through which the assertion of power and success translated into monumental forms was spread to the international Greek community of the 144

ELVIRA ROCCA La Coste Messelière, P. ed. (1936) Au Musée de Delphes. Paris, De Boccard. La Coste Messelière, P. (1969) Topographie delphique. Bulletin de correspondance hellénique, 93, 730-58. Laroche, D.(1988) L’emplacement du trésor de Cyrene à Delphes. Bulletin de correspondance hellénique, 112, 291-305. Laroche, D.(1989) Nouvelles observations sur l’offrandes de Platées. Bulletin de correspondance hellénique, 113, 183-98. Laroche, D.(1993) Travaux de l’École Française en Grèce en 1992. Bulletin de correspondance hellénique, 117, 629-44. Lazzarini, M.L. (1976) Le formule delle dediche votive nella Grecia arcaica. Atti della Accademia nazionale dei Lincei, Classe di scienze morali, storiche e filosofiche, Ser.8, 19/2, 47-354. Le Roy, C.(1967) Les terres cuites architecturales. Fouilles de Delphes II, Athènes-Paris, École Française d’Athènes. Lévêque, P.(1996) Anfizionie, comunità e concorsi panellenici. IN: Settis, S. ed., I Greci, storia, cultura, arte e società II, Torino, Einaudi, 1111-139. Linders, T. (1989-90) The melting down of discarded metal offerings in Greek Sanctuaries, Scienze dell’ Antichità, 3-4, 281-5. Linders, T. (1992) Sacred finances: Some observations. Boreas, 21, 9-12. Lippolis, E., Garraffo, S., Nafissi M. (1995) Taranto. Culti greci in Occidente: fonti scritte e documentazione archeologica, I. Taranto, Istituto per la storia e per l’archeologia della Magna Grecia. Lombardo, M. (1993) Da Sibari a Thuri. Atti del Convegno di studi sulla Magna Grecia, 32, 255-328. Maddoli, G. (1984) I culti di Crotone. Atti del Convegno di studi sulla Magna Grecia, 23, 310-35. Malkin, I. ed. (1987) Religion and colonization in Ancient Greece. Leiden, Brill. Meiggs R. and D. Lewis, eds. (1971) A Selection of Greek Historical Inscriptions to the End of the Fifth Century B.C. Oxford, Clarendon Press. Mele, A. (1982) La Megale Hellàs pitagorica: aspetti politici, economici e sociali. Atti del Convegno di studi sulla Magna Grecia, 21, 33-80. Mele, A. (1984) Crotone e la sua storia. Atti del Convegno di studi sulla Magna Grecia, 23, 9-87. Mertens, D.(1984) I santuari di Capo Colonna e Crimisa: aspetti dell’architettura crotoniate. Atti del Convegno di studi sulla Magna Grecia, 23, 189-228. Mertens, D. ed. (1993) Der Alte Heratempel in Paestum und die Archaische Baukunst in Unteritalien. Mainz, von Zabern. Mertens Horn, M. and L. Viola (1990) Archaische Tondächer Westgriechisher Typologie. Hesperìa, 59, I, 240-8. Morgan, C. (1993) The origins of Pan-Hellenism. IN: Marinatos, N. and R. Hägg eds., Greek Sanctuaries. New Approaches, London-New York, Routledge, 1837.

Placing the Gods. Sanctuaries and Sacred Space in Ancient Greece, Oxford, Clarendon Press, 3-18. De Polignac, F. (1996b) Offrandes, mémoire et compétition ritualisée dans les sanctuaires grecs à l’époque géométrique. Boreas, 24, 59-66. De Sensi Sestito, G. (1981) I Dinomenidi nel basso e medio Tirreno tra Imera e Cuma. Mélanges de l’École française de Rome. Antiquité, 93, 617-42. De Sensi Sestito, G. (1983) Gli oligarchici sibariti, Telys e la vittoria crotoniate sul Traente. Miscellanea di Studi Storici, III, 37-56. De Sensi Sestito, G. (1987) La Calabria in età arcaica e classica: storia, economia, società. IN: Settis, S. ed., Storia della Calabria antica, I. Roma-Reggio Calabria, Gangemi, 229-89. De Sensi Sestito, G. and M. Intrieri (1992) Crotone in età greca e romana. IN: Mazza, F. ed., Crotone. Storia, cultura, economia, Soveria Mannelli (CZ), Rubbettino, 23-78. Forrest, W.G. (1957) Colonization and the rise of Delphi. Historia, 6, 160-75. Giangiulio, M. ed. (1989) Ricerche su Crotone arcaica. Pisa, Scuola Normale Superiore. Hansen, É. (1960) Les abords du trésor de Siphnos à Delphes. Bulletin de correspondance hellénique, 94, 387-433. Hansen, É. (1992) Autour du temple d’Apollon (1). IN: Bommelaer, J.Fr. ed., Delphes. Centenaire de la “Grande Fouille” réalisée par l’ École Française d’Athènes (1892-1903). Leiden-New York, Brill, 120-8. Homolle, T. (1898) Topographie de Delphes. Bulletin de correspondance hellénique, 22, 572-9. Jacquemin, A. and D. Laroche (1990) Une offrande monumentale à Delphes: le Trépied des Crotoniates. Bulletin de correspondance hellénique, 116, 299-323. Jacquemin, A. (1991) Le chantiers de Pankratès, d’Agathôn et d’Euainètos au péribole du sanctuaire d’Apollon à Delphes. Bulletin de correspondance hellénique, 115, 244-58. Jacquemin, A. (1992) Offrandes monumentales italiotes e siceliotes à Delphes. Atti del Convegno di studi sulla Magna Grecia, 31, 193-204. Jacquemin, A. (1995) Ordres des termes des dédicaces delphique. Annali dell’Istituto Orientale di Napoli. Sezione di Archeologia e Storia antica, 2, 141-57. Jacquemin, A. ed. (1999) Offrandes monumentales à Delphes. Paris, École Française d’Athènes. Jacquemin, A. (2001) Delphes au Ve siècle ou un panhellénisme difficile à concrétiser. PALLAS, 57, 93110. Jeffery, L.H. ed. (1961) The local scripts of archaic Greece, Oxford, Clarendon Press. Jeffery, L.H. ed. (1990) The local scripts of archaic Greece, 2nd edn. Oxford, Clarendon Press. Karo, G. (1909) En marge de quelques textes delphiques, Bulletin de correspondance hellénique, 33, 201-37. Krumeich, R. (1991) Zu den goldenen Dreifüssen der Deinomeniden in Delphi. Jahrbuch des Deutschen Archäologischen Instituts, 106, 37-62. 145

SOMA 2004 Musti, D. (1990) Le rivolte antipitagoriche e la concezione pitagorica del tempo. Quaderni Urbinati di Cultura Classica, 36, 35-65. Osanna, M. ed. (1992) Chorai coloniali da Taranto a Locri. Documentazione archeologica e ricostruzione storica. Roma, Istituto Poligrafico e Zecca dello Stato. Osborne, R. (2004) Hoards, votives, offerings: the archaeology of the dedicated object. World Archaeology, 36/1, 1-10. Parke, H.W. and D.E. Wormell eds.(1956) The Delphic oracle. I. The History. Oxford, Blackwell. Partida, E. (2000) The treasuries at Delphi. An architectural study. Jonsered, Paul Åströms Förlag. Piccirilli, L . (1972) Aspetti giuridici dell’anfizionia delfica e suoi rapporti con la colonizzazione greca. Annali della Scuola Normale Superiore di Pisa, III 2 (1), 35-61. Rougemont, G. (1991) Delphes et les cités grecques d’Italie du Sud e de Sicilie. Atti del Convegno di studi sulla Magna Grecia, 21, 157-92. Shepherd, G. (2000) Greeks bearing gifts: religious relationships between Sicily and Greece in the archaic

period. IN: Smith, C. and J. Serrati eds., Sicily from Aeneas to Augustus. New approaches in Archaeology and History, Edinburgh, Edinburgh University Press, 55-70. Shoe, L.T. ed. (1952) Profiles of Western Greek mouldings. Roma, American Academy in Rome. Sismondo Ridgway, B. (1977) The Platean tripod and the serpentine column. American Journal of Archaeology , 81, 374-9. Valenza Mele, N. (1977) Hera ed Apollo nella colonizzazione euboica d’Occidente. Mélanges de l’ École française de Rome. Antiquité, 89 (2), 493-524. Van Straten, F. (1981) Gifts for the Gods. IN: Versnel, H.S. ed., Faith, Hope and Worship: Aspects of Religious Mentality in the Ancient World, Leiden, Brill, 65-151. Van Straten, F. (2000) Votives and Votaries in Greek Sanctuaries. IN: Buxton, R. ed., Oxford Readings in Greek Religion, Oxford, Oxford University Press, 191223.

146

Sacred or Profane: Swallow-Painted Nippled Ewers from Akrotiri Heather Mae Russell

In August of 1970, Spyridon Marinatos excavated a room known as Delta 2 at Akrotiri, a large Late Bronze Age settlement on the southern coast of the Cycladic island of Thera. In that room, Marinatos found a vessel suspended in volcanic ash emitted from the volcano that buried the site c. 1628 BCE. Evidently fallen from the shelf above, this ewer was gently caught in ‘mid air’ as ash filled the room (Marinatos 1970: 20). Almost thirty-six hundred years later Marinatos removed the object whole and in pristine condition (Fig. 1). Spyridon Marinatos discovered more of these so-called ‘nippled ewers’ in rooms across the length of the site at Akrotiri.

and to offer an interpretation of their symbolic meaning and possible function within the larger scope of Bronze Age Theran religion. It approaches nippled ewers as objects that unify a variety of ideas, and in doing so they reveal themselves as ritual objects essential to a rite of fertility held in honour of a great nature goddess. Manufacture Potters first made nippled ewers in the Middle Cycladic period (henceforth, ‘MC’), and they continued making them in the Late Cycladic period (henceforth, ‘LC’) using the same techniques. During the transition from the MC to the LC period, potters began to make pottery using the wheel. Despite the corresponding increase in the variety of pottery shapes and the industrial scale of production, potters continued to make a few vessel types by hand (Papagiannopoulou 1990: 57). According to Papagiannopoulou, ‘…by the Late Cycladic period the wheel was employed to the virtual exclusion of other techniques. The few pots which continued to be made by hand were always traditional Cycladic shapes, such as the breasted jug or jug with backwards tilted neck’ (Papagiannopoulou 1990: 62). Nippled ewers apparently numbered among these because they lack evidence of a cutting string on the bottom or rilling on the interior.

FIGURE 1 FOUR VIEWS OF THE DELTA 2 NIPPLED EWER (DRAWING: STEFAN KLEINSCHUSTER) Marinatos first described the vase as a mastoprochos, a ‘nippled ewer’, and characterised it as a ritual vessel in his reports from the third season of excavation (1969). Such ewers are ‘jugs whose characteristics are a narrow neck, and a long beak-like spout … usually they have a pair of breasts and thus fertility is emphasised. Almost always other features taken from the animal world show their influence, namely painted or plastic eyes and a mouth in the form of a beak’ (Marinatos 1970: 41). Further evidence supports these initial thoughts on the symbolism of nippled ewers and their ritual character.

Clearly the MC nippled ewers on display in the New Museum of Prehistoric Thera are hand-built (Fig. 2). The shapes of MC nippled ewers can range from a highly angular form to a softer shape with smooth transitions from foot to belly and through the shoulder and neck. The crisp profiles of vessels like the Delta 2 nippled ewer challenge the notion that in the LC period potters continued to build them by hand. There is a distinct possibility that the upper portion, shoulder, neck and lip continued to be handbuilt, while the lower foot and belly were wheel-thrown. Evidence of rilling on the lower interior may not be visible from the view afforded by the open mouth.

This paper examines the manufacture, surface decoration, and symbolism of those nippled ewers painted with swallows. The objective of this paper is to examine the nippled ewers discovered at Akrotiri (especially those painted with swallows), to assess their form and context,

Papagiannopoulou associates the technical continuity of basic form from MC to LC with a putative symbolic meaning of the type: ‘… the [nippled] jug changed very 147

SOMA 2004 What does this use of white achieve? Marthari comments, ‘… [creating contrast between surface and decoration] cannot be the reason for coating the surface with white, a view further supported by the fact that there are Theran vases with white coats but without painted decoration’ (Marthari 1990: 454). Marthari concludes her argument by stating that it was used either to ‘mask technical weaknesses or to enhance the expression of the Theran aesthetic’ (Marthari 1990: 456). She also comments that Aloupi and Maniatis have suggested talc could give a smooth effect ‘not achievable by simple burnishing of the semi-coarse Theran clay. Consequently, the talc coat produced a finer quality light surface’ (Marthari 1990: 454). Papagiannopoulou, on the other hand, advances an even more abstract explanation: ‘Regardless of differences in opinion as to the mechanism involved in the decoration of a pot, it is generally agreed that the style of a family of similarly decorated ceramic containers used the same system of designs because this denoted the same symbolic meaning for the users of the pots’ (Papagiannopoulou 1990: 63). As Papagiannopoulou indicates, it is thus likely that this particular use of white slip denoted a specific symbolic meaning.

FIGURE 2 MIDDLE CYCLADIC NIPPLED EWERS ON DISPLAY IN THE NEW MUSEUM OF PREHISTORIC THERA (PHOTOGRAPH: AUTHOR) little, if at all, in the Late Cycladic period. The survival of unaltered local shapes for ritual use suggests strong continuity of tradition … [Those vessels] which were still made without a wheel [or those whose wheel-throne form imitates MC hand-built types?], must come from local workshops reluctant to abandon the traditional technique [and form], as long as some old wares were still in demand …’ (Papagiannopoulou 1990: 62-3). Here Papagiannopoulou suggests that traditional form and manner of production lend nippled ewers a specific quality and that these vessels constituted part of a class of pottery used specifically for ritual.

The extensive efforts made to use white slip in this way suggest that there was more to its use than simply providing a smooth white surface on which to paint. As in the art of other Bronze Age Aegean cultures borrowing from Egyptian wall-painting conventions, the flesh of women in Cycladic wall painting is white. In the case of Theran wall painting, the white of fine lime plaster denotes female flesh, while black outlines the contour of bodies. The white slip, I propose, may therefore make explicit an association between the white ‘flesh’ of the anthropomorphic vessel and the conventional white flesh of women in Theran frescoes such as those on the east wall of Room 3, Xeste 3. Depictions of women in Theran wall paintings illustrate them in traditional Bronze Age Aegean bare-bodiced costume. This combination of bare-breasts and white flesh corresponds to the surface and relief decoration of nippled ewers, and in doing so it creates a coordination of meanings. I suggest here that the user of these pots was meant to read nippled ewers as evoking the female body.

This conservatism in form and technique is also evident in surface decoration. The fabric of nippled ewers is local calcareous clay rich in iron and potash (potassium hydroxide). The clay is naturally buff coloured and, because it has variable calcite content, will turn a variety of green or orange when fired at certain temperatures (Marthari 1990: 452). The dark slip used for decoration is made with manganese rich clay. The potter coated the surface of these hand-built vessels in white slip, and then burnished them to create a smooth satin finish. On many nippled ewers, however the white coat is no longer visible. Marthari points out that this use of white to coat an already buff coloured surface is rare at Phylakopi and Ayia Irini but common at Akrotiri. It is especially distinct when compared with the sparse use of white on Kamares ware from Phaistos on Crete. Cretan potters restricted the use of white to sparse highlights in the decoration, whereas on nippled ewers the mineral talc-based slip coats the entire surface. Marthari attributes this to the low temperature at which the wares were fired, since talc sinters at 1050 degrees Celsius. If the ewers were fired at 700-800 degrees, the white coat would have quickly vanished (Marthari 1990: 452). In addition, the mineral talc base does not easily adhere to the clay. Thus, the talc must have been suspended in some organic material and then brushed on the surface. (Marthari 1990: 454-6)

Surface Decoration The application of relief or barbotine breasts on nippled ewers occurs in both the MC and LC periods. MC nippled ewers such as those on display in the New Museum of Prehistoric Thera have nipples that are little more than small balls of clay pressed onto the surface of the vessel (Fig. 2). Their placement ranges from high on the neck to high on the shoulder. The MC vessel in Fig. 3 has sharply moulded nipples, each emphasised with dark slip and an encircling aureole of dots. The use of dots and dark slip accents continues, and in the Mature LC I phase, ceramicists render the nipples with increasing naturalism. 148

HEATHER MAE RUSSELL Swallow Symbolism In addition to breasts, beading rings, bands, and abstract motifs, painters occasionally decorated nippled ewers with polychrome pictorial representations drawn from the natural world. Among these ‘representations from real life, with real scenes of action’ exist motifs such as the swallow (Fig. 4), myrtle plant, and barley (Marthari 1987: 377). Sara Immerwahr concludes, ‘The swallow is an indigenous Theran motif that begins in the MC period and in turn influenced the wall-paintings of LCI, particularly the “Spring Fresco” from Delta 2’ (Immerwahr 1990b: 237). Furthermore, such frescoes in turn influenced decorative motifs on pottery from the Mature LCI phase.

FIGURE 3 MIDDLE CYCLADIC NIPPLED EWER (DRAWING: STEFAN KLEINSCHUSTER) The round belly of the Delta 2 ewer and the bulge of the belly on other nippled ewers also evoke female anatomy, which is accentuated by the white surface and the plastic painted nipples. This bare-breasted aesthetic is also observable in Bronze Age female costume. The frescoes from Room 3 Xeste 3 illustrate female worshippers participating in the ritual gathering of saffron. All of the women wear the same type of Aegean festal dress, but the individual details of their costume differ. The fitted bodice with accentuated seams exposed the breasts, thus conspicuously drawing attention to them; the sleeves are short with tassels; the flounced skirt hangs over an undergarment. Immerwahr observes, ‘Nowhere does the close union between the goddess and her attendants come out better than in the frescoes of Xeste 3’ (Immerwahr 1990a: 59). Both the goddess and her attendants wear this type of dress. Clearly these costumes suggest by their design that the breasts symbolised something for both the participant wearing the costume and the observer of a ritual.

FIGURE 4 SWALLOW-PAINTED NIPPLED EWER FROM ROOM BETA 2, AKROTIRI (PHOTOGRAPH: AUTHOR) Essentially, Immerwahr suggests that the swallow began as a pictorial motif on MC vessels that influenced wall painters of the subsequent period (Immerwahr 1990b: 237). In turn, wall paintings such as the ‘Spring Fresco’ influenced contemporary vase painters. Regardless of who inspired whom, the fluid relationship between vase painting and wall painting is clear when swallows from the ‘Spring Fresco’ are compared to the swallows painted on the nippled ewer in Fig. 4. The result manifested itself in vessels like the one in Fig. 4, which exhibits a simple harmony between the image of the swallow and the shape of the vessel.

In the introduction to Thinking Through the Body: Archaeologies of Corporeality, the authors suggest that the body carries with it its own history of making its own genealogy within a given context (Hamilakis, Pluciennik, et al.: 2002, 1-21). In other words, the ways in which a culture perceives the human body have a history of promoting specific aspects of it. We have seen here that public communicators such as costume convey perceptions of the body in the Bronze Age Aegean. Emphasis on the breasts in female Aegean festal costume worn by both the participants and the goddess conveys their importance most likely as symbols of nourishment or fertility. The jewellery and adornment of the goddess’ costume, with motifs drawn from nature such as dragonflies, waterfowl, and crocus stamens, feature her as a goddess of nature and fertility.

Together the continuity of form and surface decoration imbues nippled ewers with significance: the surface decoration evokes divine or mortal female flesh and emphasises the breasts; their anthropomorphic form is combined with avian anatomy which is suggested by the shape of the neck and mouth. Marinatos’ excavation turned up many nippled ewers in a variety of locations across the site. When considered next in light of context and the frescoes at Akrotiri, nippled ewers seem the sum of coordinated ideas, on the one hand, those of a breasted 149

SOMA 2004 functional pouring vessels—that they were themselves ‘iconic’ in character. Thus, nippled ewers had a third more abstract, ‘transactional’ function. By ‘transactional’ I mean this type of vase was essential to a ritual transaction between mortal and divinity. Using these vessels was the same as conducting a conversation with the divine. A ritual participant would offer the contents of a nippled ewer, ask for fertility in all things, and hope that in conducting this service he/she would be blessed. It was the act, or the transaction, of using these particular vessels, with their particular qualities of form, manufacture and decoration that permitted interpersonal communication with the divine. The patron used these objects in performing a ritual that established a contract of service and blessing with the divine. As Suzanne Preston Blier states, ‘Rituals, as markers of life, offer through their formality and relative fixidity a means of measuring, mastering, and making sense of the world at large’ (Blier 1996: 189). There is more to ritual, however, than the rationalising or ordering of experience and belief. The performance of ritual—ritual action—gives individuals an opportunity to participate in sanctioned communal or private events. The artefacts of ritual, including setting, costume and ritual objects, convey the essence of a ritual and the preoccupations of cult. The white-coated surface and relief nipples, the avian shape of neck and mouth and the painted motif of flying swallows are specifically chosen to convey meaning.

FIGURE 5 BARLEY PAINTED NIPPLED EWER FROM WESTERN ROOM 2, SECTOR ALPHA, AKROTIRI WITH AN EYED-JUG PAINTED WITH A NIPPLED EWER ON DISPLAY IN THE NEW MUSEUM OF PREHISTORIC THERA (PHOTOGRAPH: AUTHOR) goddess and her devotees, and on the other, avian images at Akrotiri.

It is argued here that the ritual in which participants used swallow-painted nippled ewers, with their sum of anthropomorphic and avian symbols, was held in honour of a nature goddess and most likely emphasised fertility. It would have occurred in September around the autumnal equinox following the first rain to end summer’s drought and begin the New Year—a time when swallows flit about while crocus flowers and lilies bloom. The potter and painter convey these details through their choice of shape and decoration especially in using the swallow motif.

Nippled ewers were found in rooms dotted evenly across the site, and there is a range in the quality of craftsmanship that suggests a wide range of the population used them. The inhabitants of the Bronze Age community at Akrotiri most clearly conveyed the importance of nippled ewers as ritual objects or as conveyers of meaning by representing them on other pots such as pithoid amphorae and the eyed-jug on display in the New Museum of Prehistoric Thera (Fig. 5). The profile includes a clearly drawn foot and belly. The neck, ‘horns,’ and handle are not as conspicuous, but they are plainly rendered. Most importantly, the painter paid specific attention to delineating both nipples: the viewer sees only one on a ewer presented in pure profile.

Just as the image of a nippled ewer painted on other pots conveyed its significance, the image of a swallow painted on walls and pots also held significance. The swallow in particular is a characteristically social bird. It has ‘relatively unpredictable, wheeling flight and a habit of flying unexpectedly through open doors and windows; it is willing to nest in and around human habitation; and it has a long migratory journey back and forth from parts unknown’ (Foster 1995: 417). Despite the fact that the swallow is only represented in Theran art, it is the third most commonly represented bird in the Bronze Age Aegean. As noted by Immerwahr, the swallow appears to be a ‘local Theran specialty’ (Immerwahr 1990b: 47). Of the thirty-five birds in published paintings from Thera, fourteen are swallows (Harte 2000: 681). All fourteen are found in the frescoes from Room 4, Xeste 3, room Beta 6, or Delta 2.

These nippled ewers painted on pots intended to contain liquids add a new dimension to functional, ceremonial and decorative interpretations of the nippled ewer type. These images suggest that rather than purely functional vessels, nippled ewers were significant/ritual vessels appreciated for both their functional capacity and their symbolic value. The idea of converting the nippled ewer into an image on other pots suggests awareness on the part of the potter and patron/user that nippled ewers were more than just 150

HEATHER MAE RUSSELL pier-and-door partition they were entering the domain of the goddess. The juxtaposition of the blue monkey and swallow together in a composition occurs once more in the frescoes of Beta 6. The Beta 6 swallow is alone, and it is depicted (in the reconstruction by Nanno Marinatos) as if it were in flight above a rocky landscape with two goats, a crocus clump and myrtle (Marinatos 1984: Fig. 83; Fig. 7). Eight blue monkeys occupy the adjoining wall. The frescoes, like that of Delta 2, occupy the height of the wall and wrap around one corner of the room. The monkeys are not searching for food or picking saffron; here they are performing acrobatics in rocky terrain with a river running below. There are no human figures and the monkeys do not behave as humans; rather, the fresco emphasises the monkeys and swallow as animals inhabiting the natural world. The relationship of monkeys, swallows, and crocus flower in Beta 6 to the monkeys, swallows, crocus, and enthroned goddess at Xeste 3, as well as the large encompassing scale of the wraparound fresco reminiscent of the ‘Spring Fresco,’ together suggest that at Beta 6 the monkeys and swallow are sacred animals of the goddess inhabiting the natural world. The twelve nippled ewers found in room Beta 6 further solidify the connection between nippled ewers and the natural world inhabited by the goddess’ sacred animals and plants. The blossoming crocus flowers and lilies are symbols of rebirth under the tutelage of the goddess and her generative powers throughout the year.

FIGURE 6 SWALLOW FROM THE FRESCOED FRIEZE OF ROOM 4, XESTE 3, AKROTIRI (DRAWING: STEFAN KLEINSCHUSTER) Above the pier-and-door partition between Rooms 4 and 3 in Xeste 3 at Akrotiri is a frescoed frieze filled with painted swallows in flight feeding their young (Fig. 6). These are barn swallows, hirundo r. rustica, of west Eurasia and central-north Africa. The polythyron of Room 4, Xeste 3 most likely functioned as a room in which women or men would have gathered prior to entering Room 3. The composition includes swallows and blue monkeys engaged in human activities. One monkey plays a golden harp and the second carries a golden sword and scabbard (Doumas 1992: Fig. 95-6). Bestowing human qualities upon these monkeys accentuates their position as intercessor between the animal world and the human, and it bestows capabilities that transcend normal boundaries. Just inside Room 3, Xeste 3 a blue monkey approaches an enthroned goddess (Doumas 1992: Fig. 122). A young woman, who pours an offering of saffron into a large basket, accompanies the monkey. On adjoining walls other women hike through rocky crags gathering saffron rich stigmas from crocus blooms. The monkey before the goddess literally stands astride two worlds with one foot on the first step of the shrine, and the other foot level with the saffron basket. Similarly, the six swallows in the Room 4 frieze act as intercessors. Endowed with the ability of flight, they are able to communicate between material and immaterial realms. I suggest the combination of blue monkeys and swallows in the fresco of Room 4, Xeste 3 may have signified to visitors that in stepping through the

FIGURE 8 SWALLOWS FROM THE ‘SPRING FRESCO,’ ROOM DELTA 2, AKROTIRI (DRAWING: STEFAN KLEINSCHUSTER) The remaining seven swallows in flight appear in the ‘Spring Fresco’ of room Delta 2 (Fig. 8). The birds engage one another or fly alone in an aerial display. The artist’s stylisation of the swallow efficaciously conveys the most salient characteristics of the species. The creativity of positions and the ease of the artist’s calligraphic line capture the elegant speed with which the swallows swoop through

FIGURE 7 SWALLOW FRESCO FRAGMENT FROM ROOM BETA 6 (DRAWING: STEFAN KLEINSCHUSTER) 151

SOMA 2004 narrow confines. Immerwahr considers the swallows to be a ‘most remarkable feature because no two are alike’ (Immerwahr 1990b: 47). Indeed, the swallows appear in a variety of poses, tumbling in air. Regardless of the specific nature of their activity, the birds appear to move swiftly and in tumultuous patterns that closely resemble their natural flight. Swallows are social birds that enjoy aerial acrobatics performed at high speeds. When these birds are in flight their performance appears effortless, even divinely inspired, as they soar into the atmosphere and drop low close to the earth, rapidly reducing the distance between. The mural is essentially life-size, with the majority of action taking place above waist level. The thick doublewalls on the north, west and south provided ample room for a fresco that dissolves the walls into ‘landscape.’ In its panoramic effect the ‘Spring Fresco’ brings the exterior world inside at the first sign of blossoming and new growth. Thus, the moment is captured and enabled to continuously occur. The fresco perpetually conveyed the significance of the swallow as an intercessor between the material and divine worlds.

final analysis what makes religion, what makes ritual, is the doing of it. It is what the worshipper does that is of primary importance,’ and it is argued here that by extension the object used in the ritual is also of high importance (Peatfield 2001: 55). The decoration and shape of nippled ewers symbolise the divine while the ewer’s transactional function speaks of communication with the otherworldly. Those pots decorated with iconic images of nippled ewers clearly demonstrate their significance to the inhabitants of Bronze Age Akrotiri. The shape of the vessel, both feminine and avian, fantastically unifies the natural world, inhabited by sacred plants and animals, with the divine realm of the goddess and her devotees. Emphasis on the breasts and female flesh echoes contemporary wall paintings, depicting Aegean festal costume, and suggests that the intent of the ritual action was to secure the blessing of the goddess. Swallows painted on nippled ewers locate the ritual temporally at a time of rebirth when flowers such as the crocus or lilies blossom. The illusion of a swallow in flight speaks of an otherworldly realm. These characteristics are combined in a ritual object, a pouring vessel that is functional, corporeal. In short, swallow-painted nippled ewers are essentially an attempt to unify the divine and material realms. As a collective symbol, swallow-painted nippled ewers signify the presence of the divine and its manifestation in the material world.

Images of birds in flight, like nippled ewers, were well distributed across the present dimensions of the excavation site. The capability of flight enables birds to achieve that which humans cannot: communication with another otherworldly, atmospheric realm. I suggest that in the context of Delta 2 or Beta 6 swallows read as iconic signs because they resemble what they stand for—fleeting moments of communication with a nature deity and the desire to make this seasonal communication permanent. Thus, images of birds in flight from across the site of Akrotiri suggest that communication with the otherworldly had the potential to occur simultaneously and perpetually throughout the town. Such a hypothesis may seem bold when one considers the repercussions, namely that every room in which a flying bird is depicted would have a religious function of sorts. But this idea is comparable to the way an image of Jesus or a crucifix hangs in a room symbolising for some the presence of Jesus in their everyday lives. In the words of Suzanne Preston Blier, ‘Quotidian life and special events are both defined by ritual’ (Blier 1996: 189). This understanding of ritual allows for the existence of images that subtly visualise the presence of the divine in everyday life.

Bibliography Blier, S. P. (1996) Ritual. IN: Nelson, R.S. and R. Shiff, eds., Critical Terms for Art History, Chicago, University of Chicago Press, 187-96. Doumas, Ch. (1992) The Wall-Paintings of Thera. Athens, The Thera Foundation. Foster, K.P. (1995) A Flight of Swallows. American Journal of Archaeology 99, 409-25. Hamilakis, Y., M. Pluciennik and S. Tarlow. (2002) Introduction to Thinking Through the Body: Archaeologies of Corporeality. New York, Kluwer Academic/Plenum Publishers. Harte, K. (2000) Birds of the Thera Wall-Paintings. IN: Sherratt, S., ed., First International Symposium: the Wall Paintings of Thera, vol. II; Thera 30 Aug–4 Sept. 1997, Athens, The Thera Foundation, 681-98. Immerwahr, S. (1990a) Aegean Painting in the Bronze Age Aegean. University Park, Pennsylvania State University Press. Immerwahr, S. (1990b) Swallows and Dolphins at Akrotiri: Some Thoughts on the Relationship of Vase-Painting to Wall-Painting. IN: Hardy, D.A. et al., eds., Thera and the Aegean World III vol. I: Archaeology, Proceedings of the Third International Congress, Santorini, Greece, 3-9 September 1989, London, The Thera Foundation, 237-45.

It is not suggested here, however, that every image of a flying bird served the purpose of communicating with the divinity in the same way that nippled ewers did. Here the functional role of nippled ewers as vessels necessitated participation. The vessels had to be poured. Using the swallow-painted nippled ewers must have held significance beyond using one of the basic nippled ewer type that lacked pictorial decoration. Yet even those nippled ewers such as the ewer from Delta 2 subtly evoke avian anatomy in the shape of the neck and mouth. Nippled ewers are utilitarian objects imbued with a greater meaning by their decoration. Alan Peatfield states, ‘In the 152

HEATHER MAE RUSSELL Marinatos, N. (1984) Art and religion in Thera: reconstructing a Bronze Age Society. Athens, D.&I. Mathioulakis. Marinatos, N. (1969) Excavations at Thera II: 1968 Season. Athens, University of Athens. Marinatos, N. (1970) Excavations at Thera III: 1969 Season. Athens, University of Athens. Marthari, M. (1990) Investigation of the Technology of Manufacture of the Local LBA Theran Pottery: Archaeological Consideration. IN: Hardy, D.A. et al., eds., Thera and the Aegean World III vol. I: Archaeology, Proceedings of the Third International Congress, Santorini, Greece, 3-9 September 1989, London, The Thera Foundation, 449-58. Marthari, M. (1987) The Local Pottery Wares with Painted Decoration from the Volcanic Destruction Level of Akrotiri, Thera. Archäeologischer Anzeiger, 359-79.

Papagiannopoulou, A. (1990) Some Changes in the BA Pottery Production at Akrotiri and Their Possible Implications. IN: Hardy, D.A. et al., eds., Thera and the Aegean World III vol. I: Archaeology, Proceedings of the Third International Congress, Santorini, Greece, 3-9 September 1989, London, The Thera Foundation, 57-66. Peatfield, A. (2001) Divinity and Performance on Minoan Peak Sanctuaries. IN: Laffineur, R. and R. Hägg, eds., Potnia: Deities and Religion in the Aegean Bronze Age. Proceedings of the 8th International Conference, Göteborg University 12-15 April 2000, Université de Liège, Histoire de’ l’art et archéologie de la Grèce antique and University of Texas at Austin, Program in Aegean Scripts and Prehistory, 51-5.

153

Is There ONE Crusader Archaeology? James G. Schryver

Superficially, the idea of one all-encompassing Crusader archaeology that includes all of the areas, artifacts and activities that scholars associate with the idea of the Crusades is quite attractive. The connections that existed among various groups of Westerners (Franks) yielded similarities between the lived culture and the material culture of the various Crusader states.1 The Franks who settled on Cyprus, for example, shared aspects of both culture and material culture with those Franks living along the Levantine coast. These connections were particularly strong among the nobility living in these areas. However, despite the similarities shared by the various communities living in the Crusader states, the concept of one, all-encompassing Crusader archaeology is incapable of providing us with a complete understanding of these communities and their individual characteristics. My purpose, therefore, is to caution that we must be careful about how fully (or even to what degree) we embrace this concept. While the generalities concerning the Crusader states that such an approach suggests may be attractive, they obscure the subtle nuances that existed between the various societies that arose in the different Crusader states. These nuances are becoming clearer as we learn more and more about the societies of the Latin East, Greece, Cyprus, and the area of the Transjordan. As the available evidence for each of these areas increases, the circumstances particular to each become clearer.

of Crusader archaeology, they are really talking about the material culture of the Kingdom of Jerusalem, mainly because this area (modern day Israel/Palestine) is the one best known to scholars. Over the last few decades, historians such as Joshua Prawer, Benjamin Kedar, and Meron Benvenisti have used the documentary evidence for the Kingdom of Jerusalem to illuminate life in the Latin East in general (Prawer 1972, 1980; Kedar et al. 1982, 1993; Benvenisti 1970). Archaeologists have followed suit, building upon the strong foundation of evidence that the aforementioned scholars excavated from the texts. Scholars focusing on the Kingdom of Jerusalem have also shaped the way we view society in the Latin East. The most recent example of this trend is Ronnie Ellenblum’s important and convincing argument for a societal model whereby the Franks settled in both urban and rural contexts, but according to the settlement patterns of the local Arab Christian population (1998). The implications of Ellenblum’s model are that contact and assimilation did occur between Franks and locals but was limited to the Latin and Arab Christian populations. In a way, his model is a reaction to the previous model of Frankish society, which argued for a situation in which the Franks and the local populations remained segregated from one another, and which he sees as heavily influenced by the conditions surrounding the growing pains of the newly-born state of Israel. Scholars such as Otto Smail developed this model of a segregated society in the years following World War II (1956). The main proponents of the segregated model were in turn reacting to a model developed in the nineteenth century, according to which the settlement of the Franks in the Latin East resulted in an integrated society, and which these later scholars saw as unduly influenced by French colonialist propaganda. Generally, archaeologists and art historians have examined the material evidence of this period in light of ideas provided by the model of society current at the time of their work. It is important for us to recognise the influence of these various models on the interpretation of the archaeological evidence. One result of this influence is the focus on the archaeology of the Kingdom of Jerusalem.

I emphasise the issue of availability of data for a reason. Typically, when scholars today discuss different aspects 1. One of the difficult distinctions that must be made in the field of Crusader archaeology, or in any other field of enquiry occupied with some aspect of the Crusades or the Crusader states, is the distinction between “Crusader” and “Frank.” In this article, I limit the application of “Crusader(s)” to those Westerners who actually took part in one of the Crusades. Frank (al-Franj or Ifranj in Arabic, o Frankos in Greek) will be used to refer to all other westerners who settled and/or were born in the different areas of the Crusader states in what are today part of Cyprus, Egypt, Greece, Israel, Jordan, Lebanon, Palestine and Syria. The area including Israel, Jordan, Lebanon, Palestine and Syria is generally referred to as the Latin East.

155

SOMA 2004 Archaeologists’ focus on this area is not the result of a lack of excavation and research in other areas where the Crusaders built castles and the Franks settled. Scholars have conducted work in Cyprus, Greece, and areas and countries to the north, south and east of the Kingdom of Jerusalem. However, with laudable exceptions, much of this work remains unpublished or incompletely published and therefore unavailable to the scholarly community at large.

As mentioned, the notion of one all-encompassing Crusader archaeology is supported by certain aspects of the material culture; the connections maintained by the members of the Frankish nobility in the Latin East with the West and with each other resulted in, and reinforced, certain shared aspects of culture and material culture. Those living in the Latin East were kept abreast of developments in France, for example, in numerous ways. These included the stream of arrivals of clergy and nobility, the members of their entourages and their own travels back and forth. The results of these personal connections with the West can be seen in the realms of fashion, literature and architecture (Rice 1937: 100-39; Prawer 1972: 396; Jacoby 1982, 1986; Enlart 1987). Social events, such as coronations, knighting ceremonies, and weddings, also served to update the Frankish nobility of the Latin East with regard to the news and trends of the courts of the West (mainly France). At the same time, their participation in these events and adoption of these contemporary trends reinforced their connections with one another (Keen 1984: 93; Philip of Novara 1887: 220, §439; 1913: 3, §IX (89) and 7, §VI (112)).

Unfortunately, the current political situation in this part of the world plays a large role in preventing archaeologists from certain countries from communicating directly with one another and from sharing their findings in regional conferences. The predominance of the evidence regarding the Kingdom of Jerusalem has not passed entirely unnoticed by scholars, however. Adrian Boas, in the beginning of his recent book entitled Crusader Archaeology, notes that Much of the archaeological activity concerned with the Crusader period has concentrated in the modern State of Israel, which covers most of the territory of the kingdom of Jerusalem … apart from some important architectural studies of the monuments of Syria and Lebanon, the remainder of the Latin East is poorly represented in publications … (1999: xv).

Other evidence for shared culture and material culture comes to us from the economic sphere. Denys Pringle, for instance, has pointed out that the same general pottery assemblage is found at many of the sites along the coast of the Levant (1982: 111; 1986: 467). This type of assemblage includes pieces of Zeuxippus Ware, Syrian Imports, Port St. Symeon Ware, Slip-Painted Ware and Proto-Maiolica. My own research on Frankish Cyprus shows that this same assemblage type is also to be found on the sites associated with the Franks who settled on the island after it was conquered by Richard the Lionheart in 1191 and sold to Guy de Lusignan (who had recently lost Jerusalem to Saladin) in 1192. In addition, once the very distinct Cypriot sgraffito ware begins to be produced in the thirteenth century, it appears at sites across Cyprus and up and down the Levantine coast (see also Stern 1995: 32535). I believe that the explanation for these similarities lies in the contacts between these two areas in the spheres of trade, religion and administration, as well as through individuals travelling from one area to the other for social purposes. The connections between the Kingdom of Jerusalem and Cyprus, especially, were reinforced throughout the thirteenth century by the waves of Frankish refugees fleeing and moving to the island ahead of the advances of the Mamluk armies (Richard 1979; 1987: 404; 1991: 238). Other evidence indicates that in addition to the Lusignans’ claim to the throne of Jerusalem, certain private individuals and institutions owned land in both places simultaneously (Edbury 1994: 76-9, 101). Men like Jean d’Ibelin invested in the sugar industry in both areas and would have been responsible for the spread of technology, developed on the mainland, to the island. An examination of the various sugar mills recently excavated in the eastern Mediterranean suggests that the similarities among installations on Cyprus, already noted by scholars, may extend to those installations constructed on the mainland. I refer here to the mills excavated at Jericho, Tawahin es-

In addition, the work that has been published is in languages that many scholars cannot read and in journals that others cannot find. These challenges are further complicated by the negative light in which the Crusades are currently viewed in some areas where the Franks settled. As a result, while scholarship is advancing, important parallels remain unexamined and interesting paths of research unexplored. Thus, regardless of how one feels about countries such as Greece and Jordan publishing the excavations carried out by their respective antiquities departments in their own language, one must recognise that the recent publication of three books focusing on either the city or the Kingdom of Jerusalem in English, both affects and effects interest in this area (Boas 1999, 2001; Ellenblum 1998). For those scholars actively engaged in research on the Crusades, the caution raised by Boas is hopefully very apparent, and the temptation to draw conclusions for the entire Latin East based on the Kingdom of Jerusalem is minimal. However, for those who do not focus on this field, the archaeology of Crusader Jerusalem as published may in fact come to represent the material culture of all of the Crusader states. In turn, such a representation would also apply to the culture and society that it illuminates, including all of the areas, artifacts and activities that one might associate with Frankish settlement in the eastern Mediterranean. For many it will not be evident that there is more to Crusader archaeology than that which is found between the covers of these, nonetheless, important books. In addition, information that is either unpublished or unavailable will not have any role in convincing them otherwise. 156

JAMES G. SCHRYVER in a line running north-south parallel to the Jordan River, along the Hijaz route. We also know, for example, that the castles of Kerak and Shawbak were built specifically for controlling this very trade route (Boas 1999: 117; Pringle 2001: 677-84). The presence of the remains of parrot fish from the Red Sea at sites linked to the Crusaders in Petra (Schmid 2002: 266) also seems to support the conclusion that the Franks living in this area participated in the trade carried out along this route as opposed to being involved in the trade occurring along the Mediterranean coast. The possibility that isolation from the coastal trade and reliance on the local trade routes hastened any assimilation of the Franks with the local communities in these areas is an issue that deserves more attention in the future.

Sukkar south of the Dead Sea and other sites in the Jordan Valley, all dating generally to the Ayyubid/Mamluk period (von Wartburg 2001; Abu Dalu 1995; Jones, Tompsett, et al. 2000; Photos-Jones, Politis, et al. 2002; Taha 2001).2 We will learn more as the excavation and studies of these sites continues. Although cultural parallels and similarities in material culture do exist between the Frankish populations in the various Crusader states, they should not blind us to the existence of important nuances in both culture and material culture in these areas. Many of these nuances are tied to local cultural influences already present in each of the areas where Franks settled. As a result, these local influences played an important role in the way that a particular group of Franks reacted to their new host environment and in the various types of societies that formed there. Differences in the chronologies of Frankish settlement in different parts of the Crusader states also contributed to this variety. The Frankish presence in the Transjordan, for example, came to an end two-to-three years before Guy de Lusignan began to settle the island of Cyprus, thus removing any possibilities for either kingdom to interact with and influence the other.

To propose one material culture or archaeology (and thus one notion of society or culture) for all of the different Crusader principalities is dangerous as it ignores the differences previously outlined and the role of local influences in the creation of these differences. An understanding of the nature of the local culture and material culture and their effects on those of the incoming group is especially important for our understanding of the development of varied archaeological assemblages in the region. It is also important for our analysis of what these assemblages can tell us about the societies and cultures that developed in the different Crusader states. The added information that an understanding of these elements can provide is of paramount importance to the study of expressions of identity in the Latin East, be they cultural, ethnic, social or religious. Such investigations can only take place as part of a contextual approach that seeks to understand as much as possible of the historical, political, and social contexts (the nuances) that affect the production of this material culture.

Some of the resulting differences in the archaeology produced in these areas are more obvious than others. For example, a survey of the ceramic record of various Crusader sites shows that the similarities in pottery assemblages already discussed do not extend to all of the areas where the Crusaders settled. Once one moves away from the coast and the trade routes that served it, the pottery assemblages begin to change. The further away one moves, the more likely it becomes that sites will have been part of different trade routes or networks, and, consequently, the more likely it will be that the material culture of the site will be different. Once across the Jordan River Valley, the difference in the ceramic assemblage is very great. Although in this case we must consider the issue of chronology mentioned above, it is still telling that almost none of the contemporary wares so numerous along the coast have been found during excavations in these areas. The difference between coastal, inland and Transjordanian sites is especially evident in terms of Proto-Maiolica ware whose presence decreases dramatically as one moves inland (Pringle 1982: 108-9). Once again, I would suggest that this difference is a result of these sites being tied into different trade routes. Pringle’s observation that the commerce in certain ceramic types would not necessarily have been dependent upon Frankish markets like Acre is extremely important to bear in mind in this regard (1986: 461-3). The sites in the Transjordan, for example, are all located

Recently, other scholars focusing on the historical record of various areas of the Crusader states have also remarked upon the importance of local influences. Concerning the notion that the situation was the same in each of the areas where the Latins settled, Angel Nicolaou-Konnari has remarked, The typology of the encounter of the Greek Orthodox and the Frankish Latin cultures in thirteenth-century Cyprus, however, is characterised by particularities that pertain to the social and historical realities of the two ethnic groups involved (1999: 286).

Christopher MacEvitt’s study of the historical sources pertaining to the interaction between Franks and different local Christian groups in northern Syria has reached a similar conclusion and presents an interesting example of the importance of the local context for an understanding of how these groups interacted with one another. His research indicates that the Franks and local Christians quickly placed each other into categories intelligible within the framework/context of their own worldviews.

2. Birgitte-Porëe (1995: 380) has made similar assertions. However, it is important to exercise caution when using this source. M.-L. von Wartburg has voiced the following criticism: ‘The list of Cypriot sites in Porëe 1995 … represents a second-hand compilation without reference to the original documentary sources and without first-hand knowledge of the archaeological material. To try to correct all errors and dubious conclusions would be tedious.’ (2000: 397, note 70).

157

SOMA 2004

FIGURE 1 COMPARISON OF THE PLAN OF SARANDA KOLONES WITH THAT OF BELVOIR. (AFTER ROSSER 1987, FIGS. 1-2) As a result, in the eyes of the Armenian rulers living in the area of Odessa, the Franks were just another group of conquerors in a long history of conquerors that had come and gone. The Armenians therefore treated them as just the latest foreign, political player in a game that had been played for centuries and thus saw little need to change their own affairs completely or bring their worldview into line with that of the Franks. The image of Frankish society that emerges from MacEvitt’s study is also one of a society ‘that was rooted in its local human and physical surroundings more than previously suggested’ (2002: 3). This new reading of the historical sources may also apply to situations where the contact was between the Franks and various communities seen as belonging to a larger ethnic group, such as the Greeks. In other words, perhaps we should not expect every encounter between Franks and Greeks to produce the same results.

they also, to their benefit, adapted many elements of the local Byzantine governmental structures (Edbury 1994: 184; Grivaud 1993: 151-2). At this point I would like to return to the material evidence. It is important to note that not all of the differences in the archaeology of the various Crusader states are as obvious as those indicated above. However, I would add that it is these more subtle differences that should serve as reminders of the particularities of each situation. I will illustrate my argument with an example from Frankish Cyprus, where the material culture of the Frankish community during the period from 1191-1291 is generally the same as that of the contemporary Kingdom of Jerusalem, but nonetheless contains variations. As the scholars who have studied and excavated the two castles have pointed out, Saranda Kolones on Cyprus (built in the 1190s) and Belvoir in Israel (modified after 1168) are almost identical in plan (Megaw 1994: 48-50; Rosser 1987: 187-8; Ben-Dov 1993: 182-6). Both castles are of the double-castra or concentric type whereby there exists a castle within a castle (Fig. 1). They also share a number of other characteristics: both have a special oblique entrance; both castles have towers at the corners and midpoints of the outer walls; both castles are surrounded by a moat and both were built on bedrock with stones quarried from the area of this moat.3 These similarities have led Peter Megaw, the lead excavator of Saranda Kolones, to suggest that some of the same people may have been involved in the construction of both (1994: 50).

A comparison of the different areas where these two groups were in contact gives credence to this caution. Some historians working on Frankish Cyprus have pointed out that in contrast to what occurred in the principalities of Frankish Greece and on Crete, on Cyprus the local class of Greek lords was not incorporated into the ruling hierarchy. Thus, they could not serve as a secular leadership around which political, cultural or even armed resistance could form, as was the case on Crete in particular (Coureas 2000: 32; Edbury 1994: 20. cf. Grivaud 1993). Nor were there the same incentives or causes for resistance in all these areas. Other differences among the Franks themselves and the various settlement approaches they enacted are also visible. For example, although the Kings of Cyprus brought many elements of their government with them from the Crusader mainland

3. Many of these characteristics are also found at ‘Ajlun Castle (1185-86) built by ‘Izz al-Din Usamah and located across the Jordan Valley from Belvoir (Minnis and Bader 1988: 255-63).

158

JAMES G. SCHRYVER There are also, however, some subtle differences in the plans of the two castles, such as the shape of the towers. Although the excavators’ opinions concerning the origin of Saranda Kolones have changed over time, they now acknowledge that the castle was built de novo by the Crusaders. Meanwhile, they have maintained their claim that it may contain local stylistic elements in addition to the physical elements of local Byzantine structures that the builders incorporated into its construction (Megaw 1967: 28; 1972: 179-81; 1988: 285; 1994: 50; Rosser 1985; 97). Megaw has suggested, for example, that the adaptations and differences in plan and composition of Saranda Kolones from those of Belvoir, such as the multiple shapes of the towers, may be suggestive of the participation of local Cypriot builders in its construction (1994: 50).

Excavations in the Holy Land. New York, Simon & Schuster, 182-6. Benvenisti, M. (1970) The Crusaders in the Holy Land. Trans. P. Fitton. Jerusalem, Israel Universities Press. Birgitte-Porëe, P. (1995) Les Moulins et Fabriques a Sucre de Palestine et de Chypre: Histoire, Geographie et Technologie d’une Production Croisee et Medievale. IN: Coureas N. and J. Riley Smith eds., Cyprus and the Crusades: Papers Given at the International Conference ‘Cyprus and the Crusades’, Nicosia, 6-9 September, 1994. Nicosia: Society for the Study of the Crusades and the Latin East and the Cyprus Research Centre, 377-510. Boas, A. (1999) Crusader Archaeology. London, Routledge. Boas, A. (2001) Jerusalem in the Time of the Crusades. London, Routledge. Coureas, N. (2000) The Latin Elite on Cyprus: Trying to Keep Apart. Journal of Mediterranean Studies, 10, nos. 1-2, 31-45. Edbury, P.W. (1994) The Kingdom of Cyprus and the Crusades 1194-1374. Cambridge, Cambridge University Press. Ellenblum, R. (1998) Frankish Rural Settlement in the Latin Kingdom of Jerusalem. Cambridge, Cambridge University Press. Enlart, C. (1987) Gothic Art and the Renaissance in Cyprus. Trans. David Hunt. London, Trigraph. Grivaud, G. (1993) Les Lusignans et leurs Archontes Chypriotes (1192-1359). IN: Actes du Colloque, Les Lusignans et L’Outre-Mer, Poitiers-Lusignan 20-24 octobre 1993. Poitiers, Programme Com’science, Conseil régional Poitou-Charentes, 150-8. Jacoby, D. (1982) La littérature française dans les États latins de la Méditerranée Orientale à l’époque des croisades. IN: Limentani A., M.L. Meneghetti, R. Brusegan, L. Milone, G. Peron and F. Zambon eds., Essor et fortune de la chanson de geste dans l’Europe et l’Orient latin: Actes du IXe Congrés international de la Société Rencesvals pour l’étude des épopées romanes, Padoue-Venise, 25 août–4 septembre 1982. Modena, Mucchi Editore, 617-46. Jacoby, D. (1986) Knightly Values and Class Consciousness in the Crusader States of the Eastern Mediterranean. Mediterranean Historical Review, I, 158-86. Jones, R.E., G. Tompsett, K.D. Politis and E. PhotosJones (2000) The Tawahin as-Sukkar and Khirbat ash-Shaykh ‘Isa Project Phase I: The Surveys. Annual of the Department of Antiquities of Jordan, 44, 523-34. Kedar, B.Z., H. E. Mayer and R.C. Smail eds. (1982) Outremer: studies in the history of the crusading kingdom of Jerusalem presented to Joshua Prawer. Jerusalem: Yad Izhak Ben-Zvi Institute. Kedar, B.Z. (1993) The Franks in the Levant, 11th to 14th centuries. Aldershot, Hampshire; Brookfield, Vt., Variorum. Keen, M.H. (1984) Chivalry. New Haven, Yale University Press.

This may seem a trivial point in itself as one might expect different masons to produce different work, but its importance lies in what it symbolises: the influence of local cultures and local material cultures on the emergent crusader material culture. Thus, the differences in material culture may be small in some instances and bigger in others. I would propose that the differences in the cultures are likewise small in reference to some components (the Western or Frankish elements) but larger in reference to others (the local elements). Not all of the Franks settled in the Kingdom of Jerusalem and none of them settled in a socio-cultural vacuum; instead local elements played a significant role in the formation of the various societies in the Crusader states. As a result, these local elements should play just as large a role in our analysis and study of these Crusader societies. Conclusion While the idea of one all-encompassing Crusader Archaeology may appear attractive at first, its usefulness for attempting to understand the complexities of the actual artifactual assemblages that are excavated is limited. The more we learn about the culture and material culture of the various areas of the Crusader states, the more we understand the importance of the differences inherent in the solutions arrived at by those who settled in these areas. Thus, we should look at these Frankish societies as individual and separate entities sharing a number of common threads. These are differences which historians are beginning to focus on, and they are differences that we will continue to see, and should be aware of, in the material record that these Crusader societies produced. Bibliography Abu Dalu, R. (1995) The Technology of Sugar Mills in the Jordan Valley During the Islamic Periods. [In Arabic] Studies in the History and Archaeology of Jordan, 5, Arabic Section, 37-48. Ben-Dov, M. (1993) Belvoir (Kokhav Ha-Yarden). IN: Stern, E. ed., The New Encyclopedia of Archaeological 159

SOMA 2004 MacEvitt, C.H. (2002) Creating Christian Identities: Crusaders and Local Communities in the Levant, 10971187. Ph.D. dissertation, Princeton University. Megaw, A.H.S. (1967) Excavations at Paphos in Cyprus, 1966. Archaeological Reports for 1966-67, 13, 25-8. Megaw, A.H.S. (1972) Saranda Kolones: A Mediaeval Castle Excavated at Paphos. IN: International Congress of Cypriot Studies ed., Praktika tou Protou Diethnous Kyprologikou Synedriou (Proceedings of the First International Congress of Cypriot Studies) 1969. Nicosia, Hetaireia Kypriakon Spoudon [Cyprus Research Center], Tomb B, 173-82. Megaw, A.H.S. (1988) The British School at Athens and Cyprus. Report of the Department of Antiquities Cyprus, 2, 281-6. Megaw, A.H.S. (1994) A Castle in Cyprus attributable to the Hospital? IN: Barber, M. ed., The Military Orders: Fighting for the Faith and Caring for the Sick, Aldershot, Hampshire, Variorum, 42-51. Minnis D. and Y. Bader. (1988) A Comparative Analysis of Belvoir (Kawkab al-Hawa) and Qal’at al-Rabad (‘Ajlun Castle). Annual of the Department of Antiquities of Jordan, 255-63. Nicolaou-Konnari, A. (1999) The Encounter of Greeks and Franks in Cyprus in the Late Twelfth and Thirteenth Centuries Phenomena of Acculturation and Ethnic Awareness. Ph.D. dissertation, University of Wales, Cardiff. Philip of Novara. (1913) Mémoires, 1218-1243. Ed. C. Kohler. Paris, Librairie Ancienne Honoré Champion, Éditeur. Philip of Novara and Gerard of Monréal. (1887) Les Gestes de Chiprois: Recueil de chroniques françaises écrites en orient aux XIIIe et XIVe siécles. Ed. G. Raynaud. Genève, Imprimerie Jules-Guillaume Fick. Photos-Jones, E., K.D. Politis, H.F. James, A.J. Hall, R.E. Jones, and J. Hamer. (2002). The Sugar Industry in the Southern Jordan Valley: An Interim Report on the Pilot Season of Excavations, Geophysical and Geological Surveys at Tawahin as-Sukkar and Khirbat Ash-Shaykh ‘Isa, in Ghawr as Safi. Annual of the Department of Antiquities of Jordan, 46, 591-614. Prawer, J. (1972) The Latin Kingdom of Jerusalem: European Colonialism in the Middle Ages. New York, Praeger. Prawer, J. (1980) Crusader Institutions. Oxford, Clarendon. Pringle, D. (1982) Some More Proto-Maiolica From ‘Athlit (Pilgrims’ Castle) and a Discussion of Its Distribution in the Levant. Levant, 14, 104-17. Pringle, D. (1986) Pottery as Evidence for Trade in the Crusader States. IN: Airaldi, G. and B.Z. Kedar eds., I

comuni italiani nel regno crociato di Gerusalemme: Atti del Colloquio ‘The Italian Communes in the Crusading Kingdom of Jerusalem,’ Jerusalem, 24-28, May 1984, Collana Storica di Fonti e Studi, 48, Genova, Università di Genova, Istituto di Medievistica, 449-75. Pringle, D. (2001) Crusader Castles in Jordan. IN: MacDonald, B., R. Adams and P. Bienkowski eds., The Archaeology of Jordan. Sheffield, Sheffield Academic Press, 677-84. Rice, D. T. (1937). Icons of Cyprus. London, George Allen & Unwin Ltd. Richard, J. (1979) Le peuplement latin et syrien en Chypre au XIIIe siècle. Byzantinische Forschungen, VII, 15773. Richard, J. (1987) Culture franque et culture grecque: le royaume de Chypre au XVème siècle. Byzantinische Forschungen, XI, 399-415. Richard, J. (1991) Culture franque, culture grecque, culture arabe, dans le royaume de Chypre au XIIIème et au début du XIVème siècle. Hawliyat Far’ al-Adab al-‘Arabiyah (Annales du Départment del Lettres Arabes), 6, no. B, 235-45. Rosser, J. (1985) Excavations at Saranda Kolones, Paphos, Cyprus, 1981-83. Dumbarton Oaks Papers, 39, 81-98. Rosser, J. (1987) The Lusignan Castle at Paphos Called Saranda Kolones. IN: Rupp, D.W. ed., Western Cyprus: Connections. Studies in Mediterranean Archaeology, LXXVII, Göteborg, Paul Åströms Förlag, 185-98. Schmid, S.G. (2002) The International Wadi al-Farasha Project (IWFP) Preliminary Report on the 2001 Season. Annual of the Department of Antiquities of Jordan, 46, 257-77. Smail, R.C. (1956) Crusading Warfare 1097-1193. Cambridge, Cambridge University Press. Stern, E. (1995) Export to the Latin East of Cypriot Manufactured Glazed Pottery in the 12th-13th Century. IN: Coureas, N. and J. Riley-Smith eds., Cyprus and the Crusades: Papers Given at the International Conference ‘Cyprus and the Crusades’, Nicosia, 6-9 September, 1994, Nicosia, Society for the Study of the Crusades and the Latin East and the Cyprus Research Center, 325-35. Taha, H. (2001) The Excavation of Tawaheen es-Sukkar in the Jordan Valley. Orient-Express, 3, 68-71. Von Wartburg, M.-L. (2000) Cane Sugar Production Sites in Cyprus. Real and Imagined. Report of the Department of Antiquities Cyprus, 381-401. Von Wartburg, M.-L. (2001) The Archaeology of Cane Sugar Production: A Survey of Twenty Years of Research in Cyprus. The Antiquaries Journal, 81, 30535.

160

Exploring the Divine Cow in Ancient Egypt Nina Shiel

This paper has arisen from my research into ancient and modern concepts of gender in ancient eastern Mediterranean religious thought and religious figures. In particular my interests lie with the so-called ‘Venus’ deities and their symbolism. In this paper, one aspect of that symbolism, as evident in Egypt from the beginning of the Dynastic period to the end of the New Kingdom (c. 3000-1075 BCE), is examined. The Egyptian bovine imagery is usually associated with the solar/celestial deity Hathor, who is more typically described as a goddess of love, joy and music. Another animal closely linked with her is the lioness, which is often interpreted as embodying her destructive and aggressive characteristics. The cow is seen as an antithesis to this and is, almost without exception, interpreted as a maternal, nurturing symbol, regardless of the context in which it occurs.

(Pyramid Texts, Coffin Texts and the Book of the Dead)1, which provide the most coherent set of literary information on Egyptian deities. The bovine elements in Egyptian divine iconography have four fairly distinct manifestations: (1) pure animal form (2) woman with a full bovine head (3) frontal feminine face with bovine ears, and (4) woman with a pair of bovine horns and a sun disc on top of her head (Hornung 1982: 110). While identification of a mix of human and bovine elements as divine, or at least non-mundane, is relatively easy, in theory distinguishing divine cows from mundane ones might present some difficulty. Thankfully, a divine cow is set apart by a number of attributes: (1) captions (2) attributes of divinities worn by the cow: sun disk and/or uraeus, wadjet eye, double plume, menat necklace, lotus necklace, ankh sign (3) location: amidst papyrus or reeds, by a mountain or mound, on a raised platform and/or being adored by other figures, and (4) special physical features: size, markings in the shape of stars, long horns. Short horns appear to relate to mundane, domestic cattle.

The Egyptian cow imagery also serves as a modern symbol for the reductionist attitude prevalent amongst scholars until very recently, concerning ancient goddesses. A set of complicated characteristics and aspects of a female deity are reduced to terms such as ‘motherhood’, ‘fertility’ and ‘sexuality,’ which are indiscriminately applied to goddesses regardless of cultural sphere or their position within the cult and among other deities. The cow imagery may be compared to bull imagery, familiar from all over the Mediterranean and commonly interpreted as symbolising power, rulership, strength and male virility. The apparent disparity between the perceived symbolisms of the two sexes of the same animal can perhaps be explained by the attribution to these animals of the traditional human male-active/female-passive dichotomy, also clearly visible in the respective treatments of male and female deities in academic literature.

The attributes of the divine cow mentioned above are otherwise usually borne by the goddess Hathor. Captions in stelae and paintings often identify a divinity depicted wholly or partially bovine as Hathor, although, interestingly, in literature this connection is not explicitly made. When the literary texts speak of the divine cow, usually no name is given to her. Allusions to apparently bovine elements are made in connection with Hathor and Nut (PT 405, 548), but, among others, BD 181 refers to the horns of Osiris, who is never represented in bovine form. Hathor can be represented in all the forms of the divine cow set out above. Nut, on the other hand, is usually depicted as a nude woman. Another, if somewhat obscure, goddess

In this paper, the symbolism of the divine cow is treated as multi-faceted and far broader than the traditional readings allow. My approach is influenced by feminist archaeology and iconographical reading. The evidence discussed largely comes from funerary context, as this is by definition the nature of a great deal of extant Egyptian material. Bovine iconography is used together with texts written in tombs

1. Utterence/Spell numbers cited in this article are from Faulkner (1972). 2. See Hornung 1982: Pl. III, for a Nineteenth Dynasty depiction with a bovine head only.

161

SOMA 2004 depicted in a full cow form almost without exception is Mehetweret2 the deity of primordial waters of creation. Pinch remarks that Mehetweret in cow form is usually shown recumbent, unlike Hathor (1993: 174). Egyptian religion was by essence fluid and flexible, and the deities could and did assume each other’s forms, attributes, and names without posing any apparent difficulty to Egyptian thought. Yet, the divine cow ‘template’ was not applied indiscriminately. Although the Egyptian deities are by origin celestial almost without exception, it was the goddesses who were perceived as having a particular connection to the sky, and to the sun god, who came to be perceived in bovine terms. A predynastic palette from Girza already shows a bovine head surrounded by stars (Fig. 1).

FIGURE 2A THE NARMER PALETTE, VERSO, IN LINE DRAWING (AFTER SHAW, I. 2000) FIGURE 1 DETAIL FROM THE ‘GERZEAN’ PALETTE SHOWN IN LINE DRAWING (AFTER RAFFAELE, F. 2004)

FIGURE 2B DETAIL FROM THE BELT OF NARMER PALETTE, IN LINE DRAWING (DRAWING: NINA SHIEL) THE KING IN THE

Bovine imagery was used in royal context as early as the first dynasty. The Narmer palette depicts King Narmer smiting his enemies under the gaze of two frontal figures in half-bovine form (Fig. 2a). The same figures are repeated as ornaments hanging from the king’s belt (Fig. 2b). An argument has been made that these figures are in fact bulls referring to the vigour and power of the king (Fairservis 1991), but the iconographic similarities between these figures and those of the contemporary little-known cow goddesses Bat, as well as the later frontal semi-bovine depictions of Hathor render this argument improbable. Baines suggests that these figures allude to the supports of the sky, and function as markers of the cosmos and the ordered world (1995: 120).

in royal context remains high. It has been suggested that King Khufu identified himself with Re, and his queen with Hathor (Hawass 1995, with refs.), presumably in her role as the Eye and companion of Re. His immediate successors were the first kings to assume the title ‘son of Re.’ A major body of literary evidence from the Old Kingdom is the Pyramid Texts, written on the walls of the pyramids of rulers of the fifth-sixth dynasties. These funerary texts, unlike those of later periods, were not illustrated. They are concerned solely with the afterlife of the king: his ascent into the sky and his reception there. Hathor is mentioned only thrice by name (PT 303, 335, 405). Nut is addressed in a great number of Utterances (eg. 427435), often as the mother of the king, to whom he returns in death by becoming a star in the sky. The king’s divine mother is also referred to simply as the Great Wild Cow (PT 412, 554, 582, 675). Accordingly, the king himself is called the Bull of the Sky (PT 254), and Re is addressed as ‘a calf of gold born of the sky’ (PT 485A).

An ivory engraving from this period depicts a recumbent cow, with the caption, ‘Hathor in the marshes of King Djer’s city of Dep’ (Wilkinson 2003: 140). This indicates that Hathor was known already at this point. Barbara Lesko has argued that she was effectively ‘created’ by the elite priests in the fourth dynasty as a consort of Re (1999). However, among others, Pinch refers to the existence of a temple to Hathor at Gebelein in the third dynasty (1993: 165). From the 4th Dynasty onwards, the number of inscriptions concerning and statues depicting Hathor

By claiming descent from Nut, the mother of the sun god, and of the Ennead, the king justified his place among the gods in the sky after his death. Although in 162

NINA SHIEL The first temple to Hathor, under her title Lady of Dendara, at Deir el-Bahri, the great necropolis on the western bank of the Nile, was built by the Eleventh Dynasty King Nebhepetre Mentuhotep. This king also built shrines to her at Gebelein and used the title ‘Son of Hathor’ (Pinch 1993: 4). One of the first surviving representations of a king being suckled by a cow comes from this period at Deir el-Bahri (Woldering 1955: pl. 23). Another early suckling scene is depicted on the pectoral of Amenemhet III of the Twelfth Dynasty from Byblos (Jidejian 1968 pl. 61). The association of Hathor with the Theban necropolis started at least during this period, although Pinch refers to a suggestion that there was a ‘popular’ cult here in a ‘cave’ shrine before this period (1993: 4). The realm of the dead was understood to lie in the west, where the sun set every night. Both the Pyramid and Coffin Texts personify the land of the dead as the ‘Beautiful West,’ which may well be a reference to Hathor. By the early New Kingdom, she had taken on the title Lady of the West, and her iconography shows her in human form wearing the hieroglyph for ‘west’ as her headdress (Fig. 3).

general the king was identified with the god Horus, at this early stage in particular the division between Horus and Re, both solar deities, is anything but clear. One of the major motifs in the Pyramid Texts is the Eye of Horus, mentioned in almost all Utterances from 1 to 200, but in the Coffin Texts of the Middle Kingdom, which developed from the Pyramid Texts, the Eye is that of Re. The apparent rising into prominence of Re and Hathor in the fourth dynasty has been mentioned above; it seems likely that to a degree, Re replaced Horus as the main royal deity. Such confusion is reflected by the fact that a tradition of Hathor being the mother of Horus (PT 303), as opposed to the usual Isis, survived. Isis had no particular connections to the sun, nor was she primarily viewed as a sky goddess, and is not represented with bovine attributes until late in the New Kingdom. Hathor’s name, translated as ‘House of Horus’, is usually understood as indicating her womb and hence a maternal relationship with Horus. Considering that ‘Horus’ refers not only to the god but to the institution of kingship, as the word was used as a synonym to ‘king’, Hathor’s name would seem to indicate her as being the source of kingship. Another evident peak in the evidence for Hathor starts during the First Intermediate Period, the 11th Dynasty, whose rulers were of Theban origin, and continues throughout the Middle Kingdom. It is during this time that the Pyramid Texts developed into longer and more complex Coffin Texts, no longer reserved solely for the use of the King. The elite now also wished to attain eternal life and to ascend into the sky, which so far had been the King’s prerogative. Pictorial representation in the tombs gradually gained prominence. Combined texts and images would in the New Kingdom develop into the Books of the Underworld. The established bovine iconography of Hathor is very evident, however, despite her strong prominence in the Coffin Texts and other literature, which also started to develop during this period, she is not addressed, or explicitly referred to, as a cow. Instead, she is addressed in her solar aspect and in her capacity as the companion of Re. The frequent mentions of the divine cow in the Pyramid Texts significantly decrease in the Coffin Texts. Interestingly, in CT Spell 542 the devotee asserts (Faulker Trans. 1977):

FIGURE 3 HATHOR AS LADY OF THE WEST, IN LINE DRAWING. DETAIL FROM A PAINTED RELIEF IN THE EARLY NINETEENTH DYNASTY

I am one of your sacred cattle, O my mistress Hathor; one who eats the hidden things (…) who spends the night doing their priestly service.

TOMB OF

SETI I (AFTER SHAW, I. 2000)

Hathor’s funerary aspects are traditionally neglected in favour of her more evocative (and much later) aspects of joy and love. In the Coffin Texts she is presented as a gentle recipient of the dead. By this stage, she seems to have mostly replaced Nut in this function. The devotees display great anxiety to make their way to her presence in the afterlife. CT Spell 184 describes what can be expected there: ‘abundance, power over the pleasure of copulation, joy in pleasure when desire comes, and contentment at the pleasure of eating bread...’ Abundance, sexuality, joy, and fertility are all traditional aspects of Hathor.

The title of Herdsman of the Sacred Cows is attested from Middle Kingdom texts at Dendara and Meir (Pinch 1993: 173). It has been suggested that drinking this divine milk may have transformed the king from human to divine, or renewed his vitality and power on a regular basis (Pinch 1993: 176). This suggestions is based on an Eleventh Dynasty relief depicting the king being offered the milk of a cow, still suckling her male calf, adjacent to the depiction of the king before Hathor in human form, holding an ankh symbol to his nose (Habachi 1963: fig. 7). 163

SOMA 2004 ‘The Book of the Cow of Heaven’ is one of the New Kingdom texts known as Books of the Underworld, and it survives inscribed in five royal tombs. It takes its name from an accompanying illustration of the sky depicted as a cow (Pinch 2002: fig. 26). The language of the text is, however, Middle Egyptian, and it seems likely that it was originally devised in the Middle Kingdom (Lichtheim 1976: 198). The text narrates the decision by Re to send his Eye as Sakhmet to slay mankind, only to repent and by trickery distract the fierce Eye. At the end, Re and the other gods leave the earth to ascend into heaven. Re is carried there on the back of the celestial cow. Allusions to this myth exist in the Middle Kingdom text ‘Instructions of Merikare’ (Lichtheim 1973), and, potentially, in the Pyramid Texts Utterance 519 (Faulkner Trans. 1969):

The concept of the divine cow in the marshes became popular enough in the early New Kingdom to be established as a new regular motif in iconography. Pinch divides this motif into two, ‘cow and lotus’ and ‘cow and papyrus’, and suggests that the first motif refers to Mehetweret as the mother of Re-Atum, who was said to be born from the divine cow or the primeval lotus (1993: 175), and so the second motif would refer to the divine cow as the protector of Horus, as mentioned earlier. Regardless of the identity of the figures, in both cases a divine cow functions as the protector of the sun god. A painting from a Theban mortuary shrine depicts a recumbent cow, in its caption referred to as Mehetweret, as per the iconographic custom, but her menat necklace bears the inscription ‘Hathor, Foremost in Thebes’ (Pinch 1993: 175; Desroche-Noblecourt 1976: pl. 45). At this point Hathor’s persona has almost completely absorbed the other manifestations of the divine cow. Interestingly, the new way of depicting her in the marshes seems to recall the first dynasty engraving featuring a recumbent cow, characteristic of Mehetweret, yet labelled Hathor, in the marshes, referred to above.

... give me these your two fingers which you gave to the Beautiful, the daughter of the great god, when the sky was separated from the earth, when the gods ascended to the sky ...

No Egyptian myth describes the creation of mankind beyond the fact that it came from the Eye of Re, potentially in form of tears (Lichtheim 1976: 199, n.5). As the source of the mankind, the Eye was the best candidate to go and punish them for their disobedience. The myth narrates the Eye’s departure and slaughter as Sakhmet, but on its return, it is welcomed back by Re ‘in peace’ as Hathor. Thus, this myth would seem to describe the creation of the two aspects of the Eye of Re: fierce, unforgiving Sakhmet, and peaceful, calm Hathor. In art, Sakhmet is depicted in leonine terms. Pinch suggests, due to the great number of bovine votive items offered to Hathor in the New Kingdom, that Hathor’s cow form represents her benevolent and approachable side (1993: 183).

The strong position of Hathor at this time is not surprising, considering the contemporary gradual evolution of ReAtum, a combined version of the Theban Atum and the traditionally royal Re, towards omnipotence. Hathor as Re’s right hand would have assumed part of his glory. The development of the sun god as the quintessential deity of Egypt reached its peak during Akhenaten’s renouncing of all other god forms in favour of the sun disc Aten. ReAtum as Aten’s main rival suffered the greatest amount of persecution, and it would not be unexpected to find evidence for similar sentiments concerning Hathor. Although Tutankhamun, at least in theory, also followed Aten, a funerary couch shaped as a pair of divine cows was discovered in his tomb (Wilkinson 2003: 174). The idea of the divine cow as the protector of the ruler, and as the transport to the otherworld, persisted even through the brief but intense period of official monotheism. When the Nineteenth Dynasty restored the gods, Hathor too was brought back to power, but transformed. The official philosophy of the Amarna period had been that only the king could be in contact with the god. Now, it seems, the elite, who had been denied the access to the divine realm (it is unlikely that ordinary people were greatly affected by Akhenaten’s ideas), reacted to this denial and took its own back by cultivating their own ideas and practices relating to the gods. From hereon, Hathor’s status gradually diminished in favour of Isis and Mut, both of whom gained bovine and royal attributes. Mut was perceived as the consort of Amun. Texts from later periods depict Hathor as the Golden One, a stereotypical goddess of love and mirth, with no evident connection to the royalty or death.

Most of the evidence usually employed to interpret the concept of the divine cow comes from the New Kingdom. After the turmoil of the Second Intermediate Period, the country was unified by a Theban king, Ahmose, the founder of the Eighteenth Dynasty. The pre-Akhenaten rulers of this dynasty depicted Hathor the divine cow as the source and protector of their power. Amenhotep I, Hatshepsut, and Thutmose III constructed shrines to her at Deir el-Bahri in connection with mortuary and solar temples. Bovine imagery survives from shrines in the form of votive offerings (figurines, jewellery, plaques, vessels, stelae, textiles), statuary, paintings and reliefs. The motif of the ruler being suckled by the divine cow was regularly employed. Another popular motif depicts the divine cow licking the hand of the ruler. Texts in the temple of Hatshepsut equate these gestures with the suckling and licking of the infant Horus, who according to myth was hidden from Seth in the marshes of Chemnis (Pinch 1993: 176). Another text calls Hathor ‘the divine cow, the divine mother, the lady of the sky, the queen of the gods’ (Lesko 1999: 106). All these epithets are distinctly reminiscent of Nut of the Old Kingdom.

In the mid-to late Eighteenth Dynasty, yet another new motif concerning the divine cow had appeared, this time in connection with a mountain. First appearing on stelae and textiles from Deir el-Bahri, the ‘cow and mountain’ motif became common on Theban votive and funerary 164

NINA SHIEL stelae during the Nineteenth – Twenty-first Dynasties. It was not employed in the tomb decoration of the kings, but it does appear in 32 Nineteenth Dynasty tombs in Thebes, and in a few other sites (Pinch 1993: 179). The mountain in question is doubtlessly the ‘Peak of the West’ by the Theban necropolis. The divine cow herself often bears the symbols or epithets relating to the west.

The idea of the divine cow in ancient Egypt retained its significance throughout the millennia in an extraordinary way, despite frequent transformations and new personae adopting the concept. From the beginning, she had clear connections with Hathor and Nut, although the Pyramid Texts stop short at giving her a name. In the Old Kingdom, Nut as the starry sky was unrivalled in her position as the origin and destination of the king, who was the divine Horus on earth. The Fourth Dynasty kings embraced the worship of Re and Hathor, and she was particularly popular among the Eleventh and Eighteenth Dynasties of Theban origin. During the Middle Kingdom, she can be called the main goddess of Egypt, and in art she was regularly shown as a cow. When her status in the later part of the New Kingdom diminished, some of her bovine and royal characteristics were transferred to Isis and Mut, but she retained importance as the funerary manifestation of the divine cow. Throughout the dynastic period she was a solar and astral deity, and she was identified with Mehetweret, the personification of the primordial waters in the sky. All these deities were associated with the sky and the sun god, and all of them came to assume the mantle of the divine cow.

The Spell 186 of the Book of the Dead, a short hymn to Hathor, is illustrated by a vignette showing the divine cow emerging from reeds by a mountainside, accompanied by the hippopotamus goddess Taweret, who stands before a table of offerings (Fig. 4). Taweret is associated with childbirth and is sometimes argued to be an aspect of Hathor. Here the two deities form a balanced equation of death and birth. The divine cow is shown adjacent to the reeds, which recall the marshes of Chemnis, and to the mountain, which evokes the necropolis, or a gateway to the otherworld. BD Spells 141 and 148 mention seven cows and a bull, who reside in the underworld. It can be suggested that the cows’ names match the various aspects of the divine cow, and in particular Hathor, who is also said to have a sevenfold form. ‘The Seven Hathors’ are mentioned in texts foretelling fates. Even though Hathor as a royal goddess lost some of her importance, her association with the divine cow meant that she persisted in funerary context.

To summarise: As the king equated himself with the solar deities, he made the divine cow his mother. Yet the creative properties of the divine cow extended far beyond human

FIGURE 4 HATHOR, EMERGING FROM A MOUNTAIN, AND TAWERET RECEIVING OFFERINGS, IN LINE DRAWING. VIGNETTE FOR THE BOOK OF THE DEAD SPELL 186 (AFTER WASSERMAN, J. 1998) 165

SOMA 2004 fertility and procreation, and physical nourishment. She made the king divine in life with her milk and led him to the sky at his death. Although she appears mainly to have been a beneficent figure, she manifested in the Middle and early New Kingdom as Hathor, whose terrifying aspect is well known. Besides Hathor, the divine cow ‘template,’ originally perhaps an independent godhead, was in the Old Kingdom applied to Nut, and in the New Kingdom to Mehetweret and, to an extent, Isis and Nut. She was a cosmic, liminal figure, present at the creation from the primordial waters, during the retreat of the gods into the sky, and marking the boundaries between this and the other world.

Faulkner, R.O. (1973/1977-1978) The Ancient Egyptian Coffin Texts, Vol. I-III, Warminster, Aris & Phillips Ltd. Habachi, L. (1963) King Nebhepetre Mentuhotep: His Monuments, Place in History, Deification, and Unusual Representations in the Form of Gods. Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 19, 16-52. Hawass, Z. (1995) The Programs of the Royal Funerary Complexes of the Fourth Dynasty. IN: O’Connor, D. and D.P. Silverman eds., Ancient Egyptian Kingship, Leiden/New York/Köln, E.J. Brill, 221-62. Hornung, E. (1982) Conceptions of God in Ancient Egypt. The One and the Many, New York, Cornell University Press. Jidejian, N. (1968) Byblos through the Ages. Beirut, Dar el-Machreq Publishers. Lesko, B. (1999) The Great Goddesses of Egypt, Norman, University of Oklahoma Press. Lichtheim, M. (1973) Ancient Egyptian Literature, Vol. I: The Old and Middle Kingdom, Berkeley and Los Angeles/London, University of California Press. Lichtheim, M. (1976) Ancient Egyptian Literature, Vol. II: The New Kingdom, Berkeley and Los Angeles/London, University of California Press. Pinch, G. (1993) Votive Offerings to Hathor, Oxford, Griffith Institute, Ashmolean Museum, and Geraldine Pinch. Pinch, G. (2002) Egyptian Mythology. A Guide to the Gods, Goddesses, and Traditions of Ancient Egypt. Oxford, Oxford University Press. Raffaele, F. (2004) Corpus of Egyptian Late Predynastic Palettes. [Internet] Available from: [Accessed January 13, 2005] Shaw, I. (2000) The Oxford History of Ancient Egypt. Oxford, Oxford University Press. Wasserman, J. (1998) The Egyptian Book of the Dead. The Book of Going Forth by Day, 2nd ed. San Francisco, Chronicle Books. Wilkinson, R.H. (2003) The Complete Gods and Goddesses of Ancient Egypt, London, Thames & Hudson Ltd. Woldering, I. (1955), Ausgewählte Werke der aegyptischen Sammlung. Bildkataloge des Kestner-Museums, Hannover, Kestner Museum.

Bibliography Ancient Sources BD Faulkner, R.O. (1972) Translation of the Book of the Dead. IN: Wasserman, James. (1998) The Egyptian Book of the Dead. The Book of Going Forth by Day, 2nd ed. San Francisco, Chronicle Books. CT Faulkner, R.O. (1973/1977-1978) The Ancient Egyptian Coffin Texts, Vol. I-III, Warminster, Aris & Phillips Ltd. PT Faulkner, R.O. (1969) The Ancient Egyptian Pyramid Texts, London, University Press. Modern Sources Baines, John. (1995) Origins of Egyptian Kingship. IN: O’Connor, D. and D.P. Silverman eds., Ancient Egyptian Kingship, Leiden/New York/Köln, E.J. Brill, 96-156. Desroches-Noblecourt, C. ed. (1976) Ramses le Grand: Galeries Nationales du Grand Palais. Paris, Galeries Nationales du Grand Palais. Fairservis Jr, W.A. (1991) A Revised View of the Narmer Palette. Journal of the American Research Center in Egypt, 28, 1-20. Faulkner, R.O. (1972) Translation of the Book of the Dead. IN: Wasserman, James. (1998) The Egyptian Book of the Dead. The Book of Going Forth by Day, 2nd ed. San Francisco, Chronicle Books. Faulkner, R.O. (1969) The Ancient Egyptian Pyramid Texts, London, University Press.

166

The Uses of Bow Imagery in a Greek Context: The Enemy Within Joe Skinner

I am not troubled because I must die, but because my death comes at the hands of a womanish archer. Plut., Sayings of the Spartans: Mor., 234 46 (Cole Babbitt, Trans. 1931)

contexts that operated in the Greek world is very much dependent upon how deeply ‘ancient realities’ are probed for data (Sourvinou-Inwood 1991). These realities bear little relation to the conventional argument that archery in Greece was a relatively marginal activity, even within the confines of the hunt. Empirical observation, combined with a patriotic aversion towards bow imagery promulgated by the Great King, was not the sole basis for traditions surrounding the effete, cowardly, ‘Oriental archer’. It must be emphasised, rather, that the polarity of opposites around which Greek civic identity was constructed included, almost from the outset, the pairing of the archer and the hoplite. This juxtaposition was by no means immutable; earlier traditions in which archery enjoyed a status similar to that which it held in the ancient Near East may well have been suppressed by a dominant hoplite ethos, formulated during the early days of the archaic polis (Lissarrague 1990a: 97). The values and identities that these warrior types embodied changed over time in reaction to social, cultural and political tensions wholly internal to the polis. Although these tensions contributed significantly towards the pejorative manner in which Orient and ‘Orientals’ were represented in the artistic and literary traditions of the fifth and fourth centuries BC, the interpretation of such imagery was necessarily a complex and inherently problematic affair –then as now.

This paper challenges current attempts to impose conceptual order upon the imagined world of the Greek viewer and, by extension, polis society as a whole. Whilst literary and iconographic records project a seemingly clear-cut polarity between archer and hoplite, east and west, such formal constructs mask a reality of far greater complexity; divine and heroic archers also existed, hunters and warriors alongside cowards and villains. Although totalising statements referring to notions of Greek ‘culture’ and ‘society’ are both commonplace and a practical necessity within modern scholarship, the use of such broad and generalising terms, when combined with the uncritical application of post-structural approaches such as altérité, leads to increasingly schematised analyses whose simplicity is entirely deceptive. Current misconceptions relating to the function and status of bow imagery in a Greek context reflect the unquestioning acceptance of highly partisan source material, thus serving to perpetuate ancient ideologies and prejudices. This often results in a manifest contradiction between the ways in which bow imagery was evidently employed and understood by contemporaries and subsequent interpretations whereby both archers and archery in general are portrayed as being socially and ideologically isolated: barbarian, low status, and cowardly, in opposition to the civic, hoplite ideal. The fact that bow imagery could be subjected to various ‘competing readings’ ensured that archers remained both elusive and ambiguous figures problematising Greek notions of self and the symbolic systems by which these were articulated.

It is in Homeric epic, a body of tradition fixed c. the end of the sixth century BC that archery first comes unambiguously to the fore—the position of the archery contest at the culmination of festivities at Patroclus’ funeral games being an obvious example. Whether this reflects the persisting influences of Near Eastern thought systems prevalent within a wider Indo-European tradition (Morris 1997; Hall 1989a: 42, no. 132) or the archaic remnants of some indigenous practice is a moot point. Occasions on which archers are portrayed in a negative fashion could conceivably be described as anachronisms whose authenticity is belied by the fact that heroic archers can be found amongst Greeks and Trojans alike (Hom., Il., IV 50-126; V. 171; 246; VIII. 710; XIII. 581; 650). The texts as they now stand were significantly modified,

An investigation of Greek ideologies surrounding the practice of archery in war would go some way towards explaining the way in which bow imagery was ‘understood’ following the outbreak of hostilities with Persia. A coherent understanding of the iconographic and semantic 167

SOMA 2004 therefore, in order to accommodate the new modes of thought and egalitarian values of the citizen soldier (Hall 1989a: 42; Hunt, 1998: 9). The name Telemachus (‘fighter from afar’) in all likelihood reflects an earlier, oral tradition (Hall 1989a: 42). Other aspects of the Odyssey, such as Odysseus’s skill with the bow, were arguably so fundamental to the epic narrative that they could not be erased without causing irreparable damage: they therefore escaped unharmed (Hom., Od., VIII 215; XXI 38; VIII 223-25; XI 605-8). It has been claimed, nevertheless, that the final battle with the suitors was adapted in order to enable a mode of combat more ideologically acceptable than the bow to be brought into play. The survival of such material casts considerable doubt upon whether the rejection of archers and archery was as wholehearted as might previously have been thought. This would to some extent explain the otherwise glaring contradiction that exists between the themes and values implicit in the texts and those to which the various communities that made up contemporary Greek society supposedly subscribed, namely the hoplite ideal.

FIGURE 2 ELECTRUM STATER FROM CYZICUS SHOWING A CROUCHING ARCHER, FIFTH CENTURY. TAKEN FROM CARRACICE, I. (1995)

Evidence from the material and iconographic record indicates that the figure of the archer was firmly entrenched within the imagination of the Greek audience well before the onset of hostilities with Persia. Archer gods and goddesses existed in the realm of the divine; albeit possessing specific attributes and representing specific categories that might set them apart from the hoplite/citizen ideal: ‘Phoebus the far-darter’ being a notable example (Soph., OT., 164). An Attic red-figure krater depicting the slaughter of Niobe’s children has Apollo equipped with a typically ‘Scythian’ gorytus (Fig. 1). Both he and Artemis had long been referred

to by their common epithet toxophoros. Representations of often-unnamed archers on intaglios or coin issues also abound. The obverse of an Electrum stater from Cyzicus shows a crouched archer examining an arrow to see if it is true (fifth century BC; BMC Cyzicus 79; Carradice 1995: pl. III; Fig. 2). The latter is in all likelihood Apollo, patron deity of the polis in question and thus a focus of civic pride and identity as opposed to an object of shame and derision. The Cretans, admittedly an exceptional example, had been employing archer imagery on their coinage ever since it was first adopted as a means of tender. By around the eighth century BC, archery was relatively widespread in mainland Greece and archers themselves a commonplace on the battlefield (Tyrtaeus, 19). Poverty, topography and lack of expertise meant that light-armed troops such as archers were often far more practical than their heavier counterparts. In spite of its never achieving the ‘historical pre-eminence’ of hoplite warfare (Snodgrass 1967: 77), finds from sanctuaries, battlefields and sieges, along with more mundane settings such as isolated rural farmsteads, situated in areas as far flung as Southern Calabria, show that archery was widely practiced in a Greek context (Bova Marina Archaeological Project 2001: Lin Foxhall pers. comm.). The prevalence of such imagery, along with the material evidence, makes it difficult to believe that a ‘national aversion’ towards archery meant that it was almost unknown in the Archaic period, arriving only with the migrant Scyths and later Persians (Snodgrass 1967: 81). Given the fact that such a picture fails quite simply to tally with the evidence on offer, how then can this entirely deceptive dichotomy have come about?

FIGURE 1 APOLLO AND ARTEMIS: THE SLAUGHTER OF NIOBE’S CHILDREN. ATTIC RED-FIGURE KRATER, FIFTH CENTURY. CARTLEDGE, P. (1998). ALTHOUGH THE BOWS DEPICTED ARE HEAVILY STYLISED IN NATURE, APOLLO BEARS A GORYTUS TYPICAL OF A ‘SCYTHIAN’ ARCHER

It is commonly acknowledged that monopoly of the military establishment guaranteed the dominance of 168

JOE SKINNER independent farmers in the political process, shutting out both the higher and lower orders. Both of these may have had cause to identify with the archer figure, on the one hand because of its heroic and aristocratic associations and on the other because they could afford little else by way of panoply. The problem would seem to be that the overwhelming majority of our sources subscribed to this same ideal, raising questions as to the extent to which such a bias has coloured our understanding of both ancient warfare and society in general (Van Wees 1995: 165). Osborne’s discussion of the way in which psiloi were represented in text and image highlights the extent to which some of the oppositions, integral to the construction of an Athenian identity, originated from within the citizen body itself. The argument that light armed troops were marginalised and in some cases, systematically denigrated is arguably applicable to archers who would make up a significant proportion of this category (Osborne 2000). The fact that psiloi incorporate a number of categories is perhaps significant; the image of the archer is often placed in opposition to that of the hoplite, a practise symptomatic of ‘the ideological habit of polarization’ identified by Cartledge as being characteristically ‘Greek’ (1993: 11). This systematic opposition, older certainly than that between warrior and ephebos, is inherently complex.

113-14; 2002: 115). Any attempt to use such images as an indication of the tactics employed by archers on the field of battle would therefore be wholly misguided. Accounts of the marginal role played by archers both, within the military hierarchy and upon the field of battle, certainly seem to owe more to political philosophy than reality (van Wees 1995: 162). The public epitaphs commemorating those who fell in the service of Athens are similarly suspect, as although they present something approaching a cross-section of the population, the extent to which such monuments represented an egalitarian ideal, enforced during a period in which the commemoration of private individuals was in fact prohibited, is debatable. One might justifiably question the extent to which archers as a group were written out of the historical tradition and whether this was in fact a result of authors reacting against social tensions generated by what is referred to as an era of ‘total war’ (Hunt 1998: 1). It being possible, of course, that the idea of ‘total war’ is itself an illusion created by earlier silences in the documentary evidence. Shipley has highlighted the role of warfare as a banal factor of everyday life -presumably encompassing all levels of society (Shipley 1993: 5). Van Wees meanwhile draws attention to the presence of the anonymous masses in the supposedly heroic context of the Iliad: suggestive of a considerable numbers of light troops, including archers (van Wees 1997: 692, 674). The clouds of arrows, referred to during the course of Homeric battles, are unlikely to have come entirely from the bows of heroes. Derogatory reference to the unwarlike and cowardly disposition of common men may itself be interpreted as evidence of social tension, however slight, within the ranks of the Greek host (Hom., Il., II 198-202).

From the second half of the sixth century BC onwards, archers appear in great numbers on Attic vases sporting costume and equipment that is decidedly non-Greek in appearance. Their presence is now interpreted in most quarters as being part of a dialogue intended to define ‘self’ via the portrayal of ‘other’, rather than the veristic portrayal of any military formation. Such images represented an attempt to reinforce Attic norms by encapsulating all that was not in the image of an archer of Thracian, Scythian or Amazonian origin. They might also be interpreted more specifically, however, as an attempt to bolster hoplite and citizen identities in the face of pressure from below.

The ‘traditional character of warfare’ as we see it may, in other words, be a charade, accepted and perpetuated by academics who, in neglecting to probe the issue further, inadvertently consolidate ancient ideologies and prejudices. These are best exemplified in Euripides’ Herakles during an exchange between Lycus and Amphitryon where the former denounces the hero, articulating a view coloured by both cultural and social prejudice.

Modern ‘readings’ of representations of both archers and light troops in general still place chief emphasis upon the identity of the hoplite as hero/citizen/warrior, as opposed to exploring the link between such images and an ongoing power discourse within the polis itself (Lissarrague 1990a: 239-40, fig. 130; 2001: 17). Lissarrague argues that the identity of the hoplite and the archer were considered complementary, the latter being rarely depicted without the former to whom he forms a backdrop or counterpoint. The use of the bow, as opposed to weapons intended for hand to hand combat, along with foreign appearance, exotic dress and origins, meant that the ‘Scythian’ archer formed a convenient foil for the heroic hoplite ideal. Whilst the hoplite is shown at the centre of a whole series of social rituals related to both the oikos and warfare, the archer, when depicted, appears predominantly in the passive role of a spectator (Lissarrague 1990a: 121-22). This would explain why the archer is only depicted singly, or on the margin of battle scenes: as a social and moral inferior he is excluded from the heroic agon (Lissarrague 1990a: 97,

He has never strapped a shield on his left arm, never faced the spear point. He has a bow, basest of weapons, in his hand and was ready to run away! A bow does not show a man’s courage: that is done by standing your ground, looking straight at the swift swathe cut by enemy spears, and holding ranks. (Eur., Heracl., 158-64, Kovacs, Trans. 1998)

We have no way of knowing whether the artifice and sophistry of Amphitryon’s counter argument was intended to be taken entirely seriously (Eur., Heracl., 188-203). The bi-partite opposition between hoplite and archer may in this case be influenced by the ‘agonistic’ nature of the 169

SOMA 2004 sophistic dialogue: whether this was a rhetorical inversion of societies norms and ideals, reflecting a coda of universal values/sense of identity is impossible to tell.

known colloquially as toxotai, to carry out what might essentially be termed police duties (Plassart 1913; Hall 1989b: 44-5; Hunter 1994: 145-6). Aristophanes’ mockery of the accentuated speech and garbled language of these figures leads one to suspect that these figures were, indeed, foreign (Ar., Lys., 18, 184, 241; Lissarrague 1990a: 126). It remains, however, that this measure postdated the imagery adorning Attic vases by at least a century and had died out by the end of the fourth century BC (Hunter 1994: 147).

The primacy of weaponry in defining the power and identity of the warrior is borne out not only in the formulaic descriptions of heroes arming themselves as relayed by the epics but by the emblematic representations of hoplite arms and armour depicted in Attic vasepainting. Attention has also been drawn to the extent to which it was, in these contexts, the use of the bow that determined a figure’s status as ‘other’, in terms of dress and appearance, as opposed to ethnic origin per se (e.g. the ‘Scythianised’ sons of Eurotas) (Lissarrague 1990a: 103-6, figs 58a-b). Erskine’s analysis of the process whereby Trojans came to be represented as Oriental explores issues pertinent to this topic; the internecine rivalries that existed throughout the Greek world often saw states or individuals identifying their opponents with the Oriental other, albeit largely in the aftermath of the Persian Wars and not before (Erskine 2001). Such rivalries were often played out in monumental art and architecture, but a more graphic example can perhaps be found in the ostrakon inscribed ‘Kallias Kratiou, the Persian’, accompanied by an image of an archer, found in the Kerameikos (Lissarrague 2002: 117-18, fig. 10). Hall’s suggestion that Euripides cast Paris in a Persian, as opposed to Homeric, guise by styling him ho toxotas, with the attendant moral, political and ethical connotations that defined him as other, is also highly significant in this context (Hall 1989a: 139; Eur., Or., 1409).

The onset of the Persian Wars had a significant impact upon the way in which the archer was represented in Attic vase-painting; social, cultural, and ethnic prejudice coalesced to produce the image of the Oriental enemy, typically represented as an archer. Contemporary literature provides, in this instance at least, a sound basis from which to interpret such imagery. In Aeschylus’ Persae the quiver functions as a symbol of the bankruptcy of Persian arms in what was seen as the contest of the bow against the spear (Gow 1928: 156; Aesch., Pers., 147-9; 239-40; Hall 1993: 119; 128). The costume of the Scythian archer became indelibly associated therefore with the external, Persian enemy making it impossible to distinguish between the two; new archer types had to be formulated as a result (Goldhill 2002: 54; Lissarrague 1990a: 123; 2002: 117). The semantic value of bow imagery remained constant in some respects, as demonstrated by literary traditions concerning Persian society and education along with the ideological imperatives by which they were underpinned (Briant 2002: 201, 209). Xenophon himself describes slingers, peltasts and archers as using servile weapons, so maybe it is no coincidence that light troops are more usually described in the context of an Oriental army, rather than that of the Greeks (Xen., Cyr., 2. 1. 18).

The function and significance of archers appearing in groups or as individuals without the citizen/hoplite double in attendance has also received considerable attention via work analysing the Athenian image of the foreigner in the iconography of the symposium (notably Thracians, Amazons and Scythians—all potentially archer figures). To be ‘Scythian in Athens’ was to explore the boundaries of acceptable behaviour the better to affirm the values upon which Athenian identity was founded (Lissarrague 1990a; 1990b). The question as to the exact nationality of the archers depicted upon Archaic vases is, however, almost impossible to resolve, since there is no reliable means of establishing whether the apposite to the hoplite paradigm was in fact a Scythian or merely a Greek in Scythian dress. It seems fair to say, nevertheless, that the Scythian archers from Attic vases bear scant resemblance to the horse-borne nomads descried in Herodotus’ mirror (Lissarrague 1990a: 127; Hartog 1988: 329). The question of nationality is thus one of marginal importance if these images perform the same function as Herodotus’ portrait of the Scythians in offering an accurate picture of Greek themes and values inverted, rather than any social or ethnographic observation.

Despite Davidson’s effective deconstruction of the phallo-centric penetration-power schema, modern scholarship habitually ascribes to the Eurymedon Vase, neatly encapsulated in Dover’s memorable phrase ‘We’ve buggered the Persians’, the fact remains that the imagery portrayed offers an unequivocally pejorative view of the Oriental archer and would have been interpreted as such by contemporaries (Davidson 1997: 170-182; Dover 1978: 105, quoted by Davidson, 171) (Fig. 3). Davidson’s colourful exposé of the assumptions underlying the analyses of Dover and, latterly, Foucault, does not detract from the fact that Eurymedon’s display of ‘the decadence of the Asiatic’ provides an effective rationale for his defeat (Foucault 1985: 85-6, 47, 220. See also Hall 1993: 111). The promiscuous nature of the pose, indicative of an insatiable sexual appetite, betrays the cowardly and effeminate nature of the archer/ Persian: ‘he long ago surrendered not to the Greeks but to his desires’ (Davidson 1997: 180-1). Although this same image has been employed to argue for the cultural receptivity Athenian society demonstrated towards Achaemenid Persia (Miller 1997: 13, figs. 1-2), it seems

The regulatory function of the foreign archer was fulfilled in real-life by a contingent of Scythian slaves purchased in the aftermath of the Persia Wars (476 BC onwards), 170

JOE SKINNER Bibliography Briant, P. (2002) History and ideology: the Greeks and Persian decadence, trans. A. Nevill. IN: Harrison, T. E. ed., Greeks and Barbarians, Edinburgh, Edinburgh University Press, 193-211. Carradice, I. (1995) Greek Coins, London, British Museum Press. Cartledge, P. (1993) The Greeks: a Portrait of Self and Others, Oxford, Oxford University Press. Cartledge, P. ed., (1998) The Cambridge Illustrated History of Ancient Greece, Cambridge, Cambridge University Press. Cole Babbitt, F., Trans. (1931) Plutarch, Moralia III, London, William Heinemann Ltd. Davidson, J. (1997) Courtesans and Fishcakes: The Consuming Passions of Classical Athens, London, HarperCollins. Dover, K. J. (1978) Greek Homosexuality, London, Duckworth. Erskine, R. (2001) Troy Between Greece and Rome, Oxford, Oxford University Press. Foucault, M. (1985) The Use of Pleasure, New York, Random House. Goldhill, S. (2002) Battle narrative and politics in Aeschylus’ Persae. IN: Harrison, T. ed., Greeks and Barbarians, Edinburgh, Edinburgh University Press, 50-62. Gow, A. S. F. (1928) Notes on the Persae of Aeschylus. Journal of Hellenic Studies. XLVIII, 134-58. Hall, E. (1989a) Inventing the Barbarian: Greek SelfDefinition through Tragedy, Oxford, Oxford University Press. Hall, E. (1989b) The archer scene in Aristophanes’ Thesmoporiazousae. Philologus 133, 38-54. Hall, E. (1993) Asia unmanned: images of victory in Classical Athens. IN: Rich, J. and G. Shipley eds., War and Society in the Greek World, London, Routledge, 107-33. Harrison, T. (2000) The Emptiness of Asia: Aeschylus’ Persians and the History of the Fifth Century, London, Duckworth. Hartog, F. (1988) The Mirror of Herodotus: The Representation of the Other in the Writing of History, trans. J. Lloyd, Berkeley, University of California Press, Berkeley. Hunt, P. (1998) Slaves, Warfare and Ideology in the Greek Historians, Cambridge, Cambridge University Press. Hunter, V. J. (1994) Policing Athens: Social Control in the Attic Lawsuits, 420-320 B.C., Princeton, Princeton University Press. Kovacs, D., Trans. (1998) Euripides’ Herakles. Cambridge, MA and London, Harvard University Press. Lissarrague, F. (1990a) L’Autre Guerrier: archers, peltasts, cavaliers dans l’imagerie attique, Rome, École Francaise de Rome. Lissarrague, F. (1990b) The Aesthetics of the Greek Banquet: Images of Wine and Ritual, trans. A. SzegedyMaszak, Princeton, Princeton University Press.

FIGURE 3 ‘EURYMEDON’ RED-FIGURED OINOCHOE, MID-FIFTH-CENTURY. CARTLEDGE, P. (1998) churlish to discount a patriotic interpretation purely on the basis that it has no parallels when such sentiments are so readily apparent in dramatic works such as the Persae (Harrison 2000: 103-15). The clash of ideologies that took place at the time of the Persian wars did little to change the way in which the archer was viewed within the polis. It might be argued rather, that an ingrained aversion to bow imagery, rooted largely in social prejudice, helped shape the way in which Oriental iconography was both interpreted and ‘understood’. Formerly a metaphor for social exclusion in the form of an internal, disenfranchised ‘other’, archer imagery became a means of conceptualising an external altérité altogether excluded from both Greek and civic ideal. Occasional glimpses of rival traditions are preserved nevertheless in the cults surrounding divine or mythical archers such as the hero Herakles: just one in a number of cases in which the bow retained specifically heroic connotations. This paper has highlighted both the need for contextual approaches to the iconographic evidence and the fact that simple, bipolar oppositions are unlikely to have much mileage when applied to a cultural consciousness as dynamic and inherently diverse as that of ‘the Greeks’. The archer was ‘other’ but the limitations inherent to this theoretical approach need themselves to be acknowledged. That divine/heroic archers existed alongside callow and effeminate types, raises questions as to the extent to which rigid cultural schemas are in fact applicable to the ephemeral world of the Athenian–Greek imagination: ridden with conflict and contradiction. Bow imagery can be likened to the altérité of which it is supposed to be indicative: difficult to define and subject to competing readings in the minds of both audience and its (now late) author.

171

SOMA 2004 Lissarrague, F. (2001) Greek Vases: The Athenians and their Images, trans. K. Allen, New York, Riverside Inc. Lissarrague, F. (2002) The Athenian image of the foreigner, trans. A. Nevill. IN: Harrison, T. E. ed., Greeks and Barbarians, Edinburgh, Edinburgh University Press, 101-27. Miller, M. C. (1997) Athens and Persia in the Fifth Century B.C.: A Study of Cultural Receptivity, Cambridge, Cambridge University Press. Morris, S. (1997) Homer and the Near East. IN: Morris, I., and B. Powell, eds., A New Companion to Homer, Leiden, Brill, 599-624. Osborne, R. (2000) An other view: an essay on political history. IN: Cohen, B. ed., Not the Classical Ideal: Athens and the Construction of the Other in Greek Art, Leiden, Brill 21-43.

Plassart, A. (1913) Les Archers d’Athénes. Revue des Études Grecques 26, 151-213. Shipley, G. (1993) The Limits of War. IN: Rich, J. and G. Shipley eds., War and Society in the Greek World, London, Routledge, 1-25. Snodgrass, A. (1967) Arms and Armour of the Greeks, Thames & Hudson. Sourvinou-Inwood, C. (1991) ‘Reading’ Greek Culture: Texts and Images, Rituals and Myths, Oxford, Oxford University Press. van Wees, H. (1995) Politics and the battlefield: ideology in Greek warfare. IN: Powell, A. ed., The Greek World, Routledge. van Wees, H. (1997) Homeric warfare. IN: Morris, I., and B. Powell eds., A New Companion to Homer, Brill, Leiden, 668-94.

172

Altars and Cult Installations of Punic Tradition in North Africa Federica Spagnoli

The sacred areas and sanctuaries of North Africa under Punic influence often present identification problems: it is usually difficult to identify the religious functions of the buildings based on reliable evidence such as inscriptions, stelae, or other remains of cult installations or activities. Moreover, sacred buildings of Punic foundation were often restored or rebuilt in the Hellenistic and Roman periods, sometimes preserving previous Punic altars and cult installations, as in some of the temples taken into consideration in this work, hence it is somewhat difficult to distinguish the originals from later superimpositions. Punic structures can thus be identified based on their shape and on the analysis of the archaeological remains surrounding them. Indeed stelae, Punic pottery and Punic terracottas, usually buried close to those installations, may testify to their belonging to a Punic tradition.

FIGURE 1 KERKOUANE (TUNISIA): SANCTUARY, ENTRANCE benches on the southern side (Fig. 1), as is the case in other Phoenician-Punic temples in Sicily, for example Solunto and Motya (Famà 1980: 7-15; Nigro 2003: 83), and a workshop with ovens and water installations on the east side (Fantar 1986: 148-53). The northern side of the court shows two square podia (A on the right, B on the left) oriented north-south, probably supporting two Punic shrines (2.85 x 2.77 x 0.60 m; 2.80/2.90 x 2.82/2.80 x 0.57 m), as suggested by the path of mud brick foundation walls

In the present study, after an analysis of the archaeological context of altars and sacred equipment, three types of cult installations will tentatively be identified: square altars made of stone blocks or bricks, wide podia, and small arulae made of stone or terracotta. The different types were found in four sites in Tunisia and Morocco; these sites were chosen because of the presence of sacred areas of the Punic tradition reused and modified during the Roman period, except for the Sanctuary of Kerkouane in Tunisia that presents only a sporadic reoccupation after the Second Punic War at the end of third century BC. Kerkouane, Tunisia The material remains of Kerkouane, a Punic foundation on the northeastern coast of Cap Bon in Tunisia, are no earlier than the sixth century BC and no more recent than the midthird century BC, since the city was destroyed during the First Punic War (Fantar 1984: 11-12). The main Kerkouane sanctuary looks like a Phoenician building: it has a rectangular plan with a main court (Court I) surrounded by shrines and ancillary rooms, a rectangular hall with two pillars flanking the entrance, a room with

FIGURE 2 KERKOUANE: COURT I, ALTAR AND, IN THE BACKGROUND, PODIUM A ON THE RIGHT, PODIUM B ON THE LEFT 173

SOMA 2004 on the upper surface. In front of podium A, is a square altar (0.75 x 0.74 x 0.45 m) lacking the upper surface made of limestone blocks plastered on the sides (Fig. 2). Some Punic coins, arrowheads, a lead plate and animal bones (ovine and bovine) were found near the altar (Fantar 1986: 153).

artefacts, maybe offerings to the deity or small altars for libations or solid offerings. In addition, at Kerkouane two square altars were found in domestic contexts. The first was located near the rear side of a long room in House 21 of Rue de l’Apotropaion (Fig. 5): the altar is made of limestone and has a smooth surface; its upper table is missing and it is oriented northsouth (Fantar 1985: 104; Fantar 1986: pl. CLXX).

FIGURE 3 KERKOUANE: COURT II, WELL WITH A RECTANGULAR MOUTH, THE CACHETTE AND THE ‘L-SHAPED’ WALL OF MUD BRICKS FIGURE 5 KERKOUANE: ALTAR OF HOUSE 21, RUE DE L’APOTROPAION

On the rear side of Court I is the so called ‘Sacrificial Area’ or Court II (Fig. 3) with three cult installations: an L-shaped mud brick wall, a well with a stone mouth and a votive deposit containing Roman lamps and pottery. On the floor, urns containing ashes and animal bones, black painted and undecorated potsherds, coins and bell-shaped figurines were found (Fantar 1986: 160-1). In the court four small altars or arulae made of limestone or terracotta were also found; one of these has a tapering shaft with a frame and an Egyptian cyma on the top (Fantar 1986: 184; Fig. 4). The dimensions of these arulae are on average 0.23 x 0.18 x 0.20 m; they are thus small and easily removable

The second installation was placed on the eastern part of the court of House 3, Rue des Deux Places, near two benches (0.36-0.40 m high), leaning on the corner (Fig. 6). The altar has a stepped base and pilasters buttress its corners. It is lacking the upper surface, shows traces of white plaster on the sides and is oriented north-south. Near the altar, in the western part of the court are a well and a drain (Fantar 1985: 655-6; Fantar 1986: pl. CLXIX). Haïdra (Hammaedara), Tunisia The next Tunisian site is Haïdra (Hammaedara), situated near the Algerian border. The Temple of Saturn (Figs. 7,

FIGURE 6 KERKOUANE: ALTAR AND BENCHES, COURT OF HOUSE 3, RUE DES DEUX PLACES

FIGURE 4 KERKOUANE: ARULA FROM COURT II 174

FEDERICA SPAGNOLI m) in front of a stepped tripartite cella (Baratte and Ben Abdallah 2000: 52). Terracotta figurines representing a male personage, maybe Saturn, the Roman African deity with characteristics of Punic Baal Hammon, were found in the temple. Lixus, Morocco The archaeological remains at Lixus, a Phoenician city founded, according to Roman tradition, in the twelfth century BC (Plinius, Naturalis Historia, XIX, 63) on the east coast of the River Lukkos, date to no earlier than the seventh century BC (Ponsich 1992a: 264-6). In this period, an apsidal temple (Temple H) was erected on the acropolis and was probably dedicated to Melkart; the cult of the Phoenician deity continued after the Roman occupation of the site. Roman temples in the sacred area on the Acropolis of Lixus were indeed erected on earlier Punic structures of the fourth to fifth centuries BC, the period of greater Carthaginian influence. Temple F was built partially concealing Punic Temple H, during the reigns of Juba II (25 BC to 23 AD) and Ptolemy (23 to 40 AD). Temple F Temple F is located in the central part of a great sacred complex and is oriented north-south. It consists of six sections including the Temple (Fig. 9, no. 1) with a tripartite cella (Fig. 9, no. 5), a cistern (Fig. 9, no. 6) ancillary rooms on the western side (Fig. 9, no. 2), a corridor running along the north and the east sides (Corridor A, Fig. 9, no. 3) and a great open court facing the entrance of the temple (Fig. 9, no. 4). A podium with an altar above it was probably located in the court in front of the triple apse cella, as in the Temple of Saturn at Timgad, Algeria (Leglay 1961: 2747; Pensabene 1990: 265-6), another Punic temple restored and reused in the Roman period. Punic pottery of the sixth century BC was found in the cistern buried under the floor of the western part of the cella (Ponsich 1981: 55-7).

FIGURE 7 HAIDRA (TUNISIA): PLAN OF THE TEMPLE OF SATURN 8), located in the south-eastern district of the city, and its stelae provide the only evidences of a Carthaginian presence (Baratte and Duval 1974: 7). The sanctuary was erected according to the earlier Punic tradition (Leglay 1961: 323-31; Pensabene 1990: 271-81), but it was restored during the Roman occupation of the site from the first century BC. The temple has a rectangular plan divided into a front platform, where a number of Punic and NeoPunic stelae were found and a central court (4.50 x 7.90 m) flanked by porticos with a rectangular altar (2.50 x 0.73

Temple H Temple H, located north of later Temple F (Fig. 9 on the left, 10), is lacking its southern section; it has the same orientation as Temple F and consists of a semicircular apse (diameter 19.40 m) built of limestone blocks, at the centre of which is a niche (3.50 x 1.60 m) where the statue of the deity was probably located. In front of it is a square altar (side 3.0 m, elevation 0.40 m) made of limestone blocks on the outer curtain and small round stones inside. It leans on a masonry base 0.15 m high and shows traces of red plaster. Not far from the altar a few Punic lamps with two spouts, Phoenician dishes, amphorae of the seventh and sixth centuries BC, mushroom rim jugs and bifid handles were found (Ponsich 1981: 100-5). Volubilis, Morocco Volubilis, the Roman city near Meknes, in Morocco (Fig. 11), was erected on a indigenous site of the fourth to third centuries BC, but building techniques, ordinary objects and monumental Punic inscriptions indicate a Punic influence from the third century BC. Neo-Punic inscriptions testify

FIGURE 8 HAIDRA: TEMPLE OF SATURN, FROM NORTH – EAST 175

FIGURE 9 LIXUS (MOROCCO): PLAN OF THE COMPLEX OF TEMPLES F (RIGHT) AND H (LEFT) ON THE ACROPOLIS

FIGURE 10 LIXUS: TEMPLE H, PLAN

FIGURE 11 VOLUBILIS (MOROCCO): PLAN OF THE CITY 176

FEDERICA SPAGNOLI to the persistence of this element even in later periods (Ponsich 1992b: 493-4). Temple A The remains of a Punic sacred building erected on the acropolis of the site were concealed under the Temple of Capitolium: the structure, called Temple A (Fig. 12), was uncovered in court B of the Temple of Capitolium. It consists of a podium (3.77 x 3.70 x 1.46 m) with a flight of stairs made of limestone blocks (saxa quadrata) filled inside with small irregular river stones. On the upper surface of these blocks the existence of grooves demonstrates that it served as a base for an upper structure, perhaps a shrine. Coins of Juba II and Cleopatra Selene date the context to the first century BC (Jodin 1987: 164-6; Behel 1993-95: 36).

FIGURE 13 VOLUBILIS: TEMPLE B, PLAN attached to the back wall of the court facing the entrance (Fig. 14); the podium looks like the stepped podia of Temple B, and, probably, as in the case of the structures mentioned previously, it was intended to contain a shrine. The base of the podium measures 4.50 x 5.50 m. Near the sanctuary, a small altar was found out of context: it has a rectangular shaft in the middle of the upper surface, probably to host a baetyl (Fig. 15). It was reused as a building stone in the sanctuary during Caracalla’s reign and erected over the Punic structures (Jodin 1987: 168-9).

FIGURE 12 VOLUBILIS: TEMPLE A, PLAN Temple B Temple B, placed on the east side of Wadi Fertassa, into the walls that Marcus Aurelius had built to enclose the city, is composed of a square court (about 55 x 55 m, Fig. 13) surrounded on three sides by porticos. In the centre of the court were three podia made of stone blocks (2.50 x 2.45 x 1.30 m; 5.0 x 2.35 m; 2.50 x 2.20 m) that face east; two of them have a flight of stairs on the front side (Jodin 1987: 167-8). The archaeological remains attest to the fact that the temple was dedicated to Saturn, and the presence of a number (more than a hundred) of Neo-Punic stelae dating from the first and second centuries AD, representing a Greek or Punic shrine, and a votive for making offerings or a sacrifice, suggests the persistence of Punic cults during the Roman period (Morestin 1980: 54-9, 63; Jodin 1987: 236-7). Temple C Temple C, located west of Temple B (Fig. 11), has a plan which includes a rectangular court with columns or pillars on three sides and a podium leaning on a raised basis

FIGURE 14 VOLUBILIS: TEMPLE C, PLAN 177

SOMA 2004 Typologies The comparative study of cult installations in the Punic sanctuaries and the residential quarters of the cities of Tunisia and Morocco allows us to put forward a basic typology of built altars connected to different cult activities and rituals. Square Altars This type is typical of sacred and domestic contexts; altars are made of limestone blocks filled with small rounded stones, or of masonry. The shape of the structures can be plain (e.g. Kerkouane: Sanctuary, House 21) or have bases decorated with a raised frame, pilaster strips, and coated with painted plaster. Altars are usually placed at the centre of the temple court or in front of the entrance of the cella (e.g. Haïdra: Temple of Saturn; Lixus: Temple F). Square altars are often connected with other cult installations such as podia, benches and niches. For instance, the altar located in Court I of the Sanctuary of Kerkouane is aligned with podium A (podium B represents a later phase of the sanctuary). At the same site, House 3 presents a domestic shrine located in the corner of the central court consisting of an altar and two plastered benches. Finally, an altar inside the cella of Temple H of Lixus stands in front of the niche, where it is likely that the sacred image of the deity was placed.

FIGURE 15 VOLUBILIS, TEMPLE C: ARULA

Small altars were also found in domestic units on the site: one of these was found inside a house of insula 8. The altar is rectangular and decorated with a raised frame at the top and the bottom; it is pierced by a circle (diameter 11 cm) with a crescent below (Fig. 16). Thymiateria, votive statuettes and Punic lamps came from the same context. These cult objects also date to the beginning of the reign of Juba II (Jodin 1987: 232-3).

None of the square altars show traces of burning on their upper surfaces. Often they are associated with other cultic devices, as well as wells, cavities and drains for water and benches (Kerkouane, House 3). These structures were probably used to receive liquid libations. The characteristic that allows a distinction between square altars and statue bases to be drawn is the different placement inside the temple, and the fact that the bases are not associated with other cult installations. Arulae Small altars and arulae have a large distribution in Punic regions of the Mediterranean, especially in domestic contexts: dimensions, shape, contexts and functions are the same everywhere (Famà 1989: 81-2). Small altars were found both in cultic and domestic contexts. Some specimens come from Court II of the Sanctuary of Kerkouane: they are made of limestone, with square bases and plain sides. The example with a tapered shaft, a frame and an Egyptian cyma on the top, looks like the portable altars found at Khirbet el-Muqanna’ a Philistine site in Palestine (Gitin 1992: 43-9), and altars of the Tofet of Antas and Tharros, in Sardinia (Moscati and Uberti 1985:131). They were votive gifts to the deity, or, more probably, they were used as altars for offerings of grains or for libations. The small pierced installation decorated with a crescent found in a residence, was used in domestic rituals probably involving the offering of grain and libations and incense burning in thymiateria.

FIGURE 16 VOLUBILIS, INSULA 8TH, PIERCED ALTAR WITH CRESCENT BELOW. 178

FEDERICA SPAGNOLI The other structure that the excavator of Volubilis called an ‘altar’ (Jodin 1987: 232), is actually the base of a baetyl, an aniconic image of the god; the structure, lacking the central element, looks like an empty throne.

Bibliography Baratte, F. and Duval, N. (1974) Les ruines d’Ammaedara – Haïdra. Tunis, Société Tunisienne de Diffusion. Baratte, F. and Ben Abdallah, Z.B. (2000) Le sanctuaire de Saturne à Ammædara (Haïdra, Tunisie): documents inédits. Revue Africaine I, 51-79. Behel, M. (1993-95) Un temple punique à Volubilis. Bulletin Archéologique Du Comité des Travaux Historiques et Scientifiques de l’Afrique du Nord 24 (Antiquités Nationales Afrique du Nord), 25-51. Dunand, M. (1971) La piscine du trône d’Astarté dans le temple d’Echmoun à Sidon. Bulletin du Musée de Beyrouth 24, 19-25. Dunand, M. and Saliby, N. (1985) Le Temple d’Amrit dans la Pérée d’Aradus. Institut français d’archéologie du Proche-Orient. Bibliothéque archéologique et historique, 121. Famà, M.L. (1980), L’area sacra con altare a tre betili di Soluto. Sicilia Archeologica 42, 7-15. Famà, M.L. (1989) Il Museo G. Whitaker. Arule e louteria. IN: Ciasca, A., Cutroni Tusa, A., Famà, M.L., Spanò Giammellaro, A., Tusa, V., Mozia. Comitato nazionale per gli studi e le ricerche sulla civiltà fenicia e punica. Itinerari IV, Roma, 81-2. Gitin, S. (1992) New Incense Altars from Ekron: context, typology and function. Eretz Israel 23, 43-9. Jodin, A (1987) Volubilis Regia Iubae. Contribution à l’étude des civilisations du Maroc antique préclaudien, Paris. Fantar, Mh. (1984) Kerkouane I, Cité Punique du Cap Bon (Tunisie). Tome I, Tunis. Fantar, Mh. (1985) Kerkouane II, Cité Punique du Cap Bon (Tunisie). Tome II, architecture domestique, Tunis. Fantar, Mh. (1986) Kerkouane III, Cité Punique du Cap Bon (Tunisie). Tome III, sanctuaires et cultes. Société – économie, Tunis . Leglay, M. (1961) Saturne Africaine. Monuments I, Paris. Morestin, H. (1980) Le Temple B de Volubilis. Études d’Antiquités Africaines, Paris. Moscati, S. and Uberti, M.L. (1985) Scavi al tofet di Tharros. I monumenti lapidei. Collezione di Studi Fenici 21, Roma. Nigro, L. (2003) Mozia X. Rapporto Preliminare. QAFP I, Roma. Pensabene, P. (1990) Il Tempio di Saturno a Dougga. IN: Mastino, A. ed., L’Africa Romana. Atti del VII convegno di studio, Sassari, 15-17 dicembre 1989, Sassari, 251-93. Ponsich, M. (1981) Lixus. Le quartier des temples. Étude préliminaire. Études et Travaux d’Archéologie Marocaine IX, Rabat. Ponsich, M. (1992a) Lixus. IN: Lipińsky, E. ed., Dictionnaire de la Civilisation Phénicienne et Punique, Brépols, 264-6. Ponsich, M. (1992b) Volubilis. IN: Lipińsky, E. ed., Dictionnaire de la Civilisation Phénicienne et Punique, Brépols, 493-4.

Podia Structures with rectangular or square bases and with steps leading up to them, sometimes known only in very schematic plans, such as those of the Temples of Volubilis, can be included among other cult installations. They functioned as pedestals for statues of the gods or were perhaps foundations for shrines containing these images. In Punic temples, podia are placed on the side of the main court facing the entrance, as is shown by those of Temples A and C in Volubilis; otherwise, they stay in the middle of the court. Unfortunately, excavation reports do not give any accounts on the presence of cult remains near these podia, which would have been useful for their interpretation. The podia from Punic sites in North Africa are very similar to those known from the Temples of Amrit and Bostan esSheikh in Phoenicia (Dunand and Saliby 1985; Dunand 1971; Stucky 1991); they show a square base supporting a small shrine or a throne, probably to house the simulacrum of the deity worshipped in the temple, which was located at the centre of an open space against a monumental façade. Possibly these devices were used when the cult statue was moved from the cella of the main temple to open areas for ritual purposes. Conclusions During the period of Roman domination, the structure of the temple changed, Roman cults were introduced but Punic rituals were left: this is shown by the various Punic and Neo-Punic stelae found close to the sacred areas, hence the placement inside the temple and the morphology of cult installations (core of the ritual itself) are conserved. The same conclusions cannot be extended to domestic contexts, since usually the arulae were not found in situ. However, they often depict religious symbolism typical of Punic cults (such as the crescent on the arula found in Volubilis, first century BC), demonstrating the persistency of a Punic substratum. The podia type is strongly related to a kind of Phoenician architecture originating in the East and conserved also in western areas, for instance in Volubilis as well as in Amrit or Sidon, where the podium is built in the middle of a big open space. Through these data it is then possible to demonstrate the persistence of Punic culture and rituals in North African Punic sites, occupied by Romans from the second and third centuries BC.

179

SOMA 2004 Stucky, R.A. (1991) Il Santuario di Eshmun a Sidone e gli inizi dell’ellenizzazione in Fenicia. Scienze dell’Antichità. Storia, Archeologia, Antropologia 5, 461-82.

180

Sun, Sea and Archaeological Sites: Archaeology and the State – A View from Cyprus Deirdre Stritch

In my research, I am concerned with the nature of the relationship between archaeology, nationalism and the nation-state. I seek to examine how these phenomena interact and constrain each other and how this interaction is absorbed at a popular level through the so-called ‘heritage industry’. This relationship is characterised by a number of features, but in this paper I will focus on what I characterise as the manipulation of the past and archaeological remains for financial benefit, as well as for the promotion of ideological agendas. Whilst superficially independent and sometimes incongruous, these characteristics of the relationship between archaeology and the state are in fact interlinked and mutually reinforcing. I argue that in the current situation, whereby the discipline of archaeology is situated within the framework of the nation-state, important decisions about archaeological practice, preservation and public display of archaeological sites are taken out of the archaeologists’ hands and placed in the hands of the state whose primary concern is often one of economic gain or the bolstering of certain principles and ideals. I contend that this fact poses major problems for archaeology and can be examined in the areas where archaeology and the past are presented to the public and the manner in which this is done. It is this that I have explored in my fieldwork in Cyprus, which carried out in summer 2003; some of the results of which are examined here. Specifically, I looked at museum and site displays from all periods, both those open to the public and those where public access is restricted, as well as any other significant interface between archaeology and the public. I also examined whether or not tensions existed between the economic exploitation of archaeological remains and academic or intellectual interest in them and whether this was reflected in the sites’ preservation and study. In this regard I look at time and space as social constructs and the role of both the archaeologist and the government appointed ‘heritage manager’ as the producers of expert knowledge on how to manage, both intellectually and physically, this time and space.

my contention that this fact has a profound effect on the way that an archaeology embedded within state structures operates. Archaeology and the past in general come to be used to bolster notions of homogeneity in culture and identity (Herzfeld 1997: 58-9; Gellner 1987: 18 and Mouliou 1996: 175), and to validate claims to territory and modern political legitimacy (Gellner 1983; Herzfeld 1997: 61-2). An undue emphasis is placed on the notion of the importance of continuity and the ability to trace ethnic groups temporally (Rowlands 1994: 136; Mouliou 1996: 175). In the modern state, these ideological functions of archaeology are often superseded or at least matched on another level by financial imperatives with an equally potent impact on local archaeology. Here archaeology, operating within state controlled agencies, plays a vital role in the ‘heritage industry’, which, as a key element in modern tourism (Urry 1990), is regularly a lucrative and crucial source of revenue for the state. Economic concerns thus receive priority with frequently disastrous consequences for archaeological remains. Indeed, many countries make use of both the ideological and economic benefits of archaeology right from the state’s inception, as their need for economic solvency is just as great as their need for internal cohesion or their need to legitimize their territorial or political claims. In any case, ideologically generated, authoritative narratives, as manifested in the choice of archaeological remains for conservation and their subsequent presentation, can be promoted through the heritage industry, and thus tourism, through its access to a large domestic and foreign audience. As we shall see, ideology and economics are simultaneously at play in Cyprus in its management of archaeology, owing to its continued political problems with the Turkish occupied north of the country, and as a physically and demographically small, resource poor island. I shall begin by examining the importance of archaeology to tourism and of tourism to the state. World wide, the tourist industry is big business generating billions of dollars annually and, as a case in point, provides 13% of total employment in Cyprus alone (Cyprus.gov.cy). According to the data released in June 2002 by the Secretariat of the

Nationalism in its ideological development equated modern state political legitimacy with group cultural antiquity. It is 181

SOMA 2004 World Tourism Organisation, international tourist arrivals amounted to 693 million in 2001; 0.6% down compared to 2000 due to the weakening economies of major tourismgenerating markets and the impact of the terrorist attacks of September 11th (World-tourism.org 2002). Such numbers have an enormous effect on the economies of tourist destinations. Besides the hotel, restaurants and catering sector, there are the travel agencies, rent-a-car agencies and innumerable other small businesses that depend on the tourist trade. Much depends then, on the industry’s development and sustained success.

difficulty in reconciling site preservation and the demands of large visitor numbers. Similar plans were also envisaged for the sites of Kouklia and Choirokoitia. Millions of pounds were allocated for the development of these sites with, in the case of Kouklia and Kourion (at which work is ongoing), 50% of funding being allocated by the Cyprus Tourist Organisation (Hadjisavvas 1995: 6). The participation of parties with such vested interests clearly has consequences for the selection and subsequent presentation of these sites. This money has to be recouped with interest, and while such high visitor numbers is great news financially, it is one of the main factors leading to site destruction through wear and tear and deliberate damage. A pertinent example of the problems arising from such a situation is the site of Kourion in western Cyprus. This large Classical city incorporates several phases and impressive visible remains, including an amphitheatre and a number of mosaics. The aforementioned government ‘Master Plan’ for Kourion was submitted in 1991 and in 1994 an agreement was reached between the Department of Antiquities and a Cypriot firm of consultants for the preparation of plans and tender documents for its implementation (Hadjisavvas 1995: 6). The Cy£3 million project was scheduled to begin in October 2001, having already been delayed and was expected to take 18 months. Work included the construction of a large on-site car park, souvenir shop and cafeteria, the installation of raised walkways, mosaic covers and information panels, all part of what the Director of Antiquities, Sophocles Hadjisavvas, referred to as the ‘beautification’ of the site (Cyprus-mail.com 2001). By the following spring, tourists and tour-guides alike, as well as politicians from the Green Party and other various concerned environmentalists and archaeologists were protesting due to the damage being inflicted on the archaeological remains. Of particular concern was the amphitheatre, from which plaques set in concrete on seats in the 1950s were being removed and other stage structures added. It was further claimed that the mosaics were being damaged by the use of heavy onsite equipment and the laying of the walkways (Cyprusmail.com 2002). By January 2003, the government had succumbed to opposition pressure and agreed to freeze the year’s budget for work on the theatre at Kourion, though in reality, and much to the concern of many, work continued on the site. Roxani Coudounari, a member of the Green Party and an architect specialising in monument restoration and conservation, objected to work on the site on the basis that it contravened international conventions, in particular the Charter of Venice to which Cyprus was a signatory. The principles enshrined in that charter of minimal intervention, the ability to easily distinguish new additions from the original and the ability to reverse all new additions were all being flouted at Kourion (Cyprusmail.com 2003a). However, anger was directed at the government for a completely different and conflicting reason by tour-guides, who claimed that delays to work meant that the theatre was out of bounds to tourists (Cyprus-mail.com 2003b). In November of 2003 Kritonas

Naturally, governments and regional authorities are understandably keen to promote those resources at their disposal in order to attract as large a slice as possible of this highly lucrative market. These resources refer to anything from the natural environment, to local culture, or at least a stylised stereotyped version of it, to archaeological remains. Doumas, in a study on the interpretation of the past in modern Greece, noted that in recent years it was perceived that economic growth was dependent on tourism, in turn dependent on archaeology and museums and thus every village sought to have its own archaeological attraction and related museum (2001: 79-80). It is understood that the development of the historical pedigree of an area will enhance tourism and employment and also play a part in boosting civic pride. A similar process can be noted in Cyprus, as my fieldwork revealed, where everything from old churches to traditional oil presses was promoted for tourist consumption. The tangible past is thus not only a country’s culture but essentially also its economic mainstay; in many countries tourism has been the driving force behind the development of archaeological and heritage sites (Sivan 1997: 51). As cultural tourism grows annually ‘artificial’ archaeological sites such as replica ancient farms and Iron Age villages are being constructed in order to cater to this ever-expanding market (Fowler 1981: 67). Nostalgia has been credited with being the most powerful motivator for making use of the past in the present (Fowler 1992, 119). In the words of Lowenthal, ‘If the past is a foreign country, nostalgia has made it “the foreign country with the healthiest tourist trade of all’ (1985: 4). Consequently, much of what archaeology is about is the presentation of knowledge within the tourist industry. Yet, at the same time, professionals work in a world in which a museum or an archaeological site is not always rated on the quality or quantity of the information it provides, but on how economically viable it is. It is, therefore, no coincidence that two of the most highly visited archaeological sites in Cyprus; The Paphos Mosaics and Kourion, with a combined visitor total of over half a million visitors for 1993 alone, were and are the focus of a developmental ‘Master Plan’ conceived of in 1989 and since put into action (Hadjisavvas 1995: 5-6). The main aim of the master plan was the protection of the sites while at the same time promoting and facilitating visitors. While the aims themselves were commendable, the reality has unfortunately proved quite different, demonstrating the 182

DEIRDRE STRITCH region is one of the key elements underpinning the unique individuality and personality of that region, which, in a market driven by the quest for an experience of the novel and exotic yet authentic, is a key selling point. Studies in nationalism and archaeology have shown how the past and its physical manifestations can become essential tools in the self-imaginings of groups of people bound together by the political and geographic binds of the nation-state.1 Such studies reveal that nationalism, in an effort to inculcate a sense of unity and solidarity, promotes the nation-state as a culturally homogeneous community with a shared history and as distinct from other nations on the basis of that culture and history. An emphasis is placed on the origins of the group and myths of origin are duly provided. These are commonly succeeded by a ‘Golden Age’, followed by a period of decline, to be ended by the current rejuvenation of the group (Hutchinson 1994: 123; Silberman 1995: 253). In Cyprus this can be seen in the focus on Mycenaean colonisation, the Classical and the Byzantine periods and of course the modern state. Individual and group esteem is thus boosted and the pedigree of the state is delineated. This coming full circle implies that the past has no meaning without the present and the present no meaning without the past (Silberman 1995: 253). A thread of ethnic and cultural continuity is thus woven from the dawn of prehistory to the present glorious renaissance. The collective memory of the group is altered through the selective highlighting and suppressing of perceived important events and their physical manifestation (Zerubavel 1994: 73).2

Makriyiannis, President of the Cyprus Mosaics Research Centre, claimed that mosaics and other artefacts were disappearing from the site. The Director of Antiquities, Sophocles Hadjisavvas admitted that indeed one mosaic was missing (Cyprus-mail.com 2003c). Work at the site continues. It is not only archaeological remains that are affected in such a scenario, but the narratives created to elucidate their history and importance. As Silberman points out; Needless to say, when market share and market orientation influence government decisions to “create” archaeological attractions, a nation’s archaeological narratives are deeply affected … In addition, ephemeral prehistoric sites and mudbrick constructions that are difficult to preserve for public viewing will often be excluded, leaving the emphasis on large-scale, impressive, masonry constructions (1995: 259).

In Cyprus, this means an emphasis on large Classical sites like Paphos, Kourion and Ammathus. These sites also have the added benefit of being linked to a modern national self-consciousness, which is dependent on the ability to prove direct kinship with the ancient Hellenes (Argyrou 1995: 198). However, as the case of Cyprus proves, decisions to promote certain archaeological sites for public consumption are not always based on economic, but on ideological principles, which may well seem to contradict the criteria outlined for economic success. The twelfth century BCE site of Maa Palaiokastro is a case in point. This quite small and physically unimpressive site was the subject of enormous investment in presentation and promotion on the basis of the ideological propaganda value it contained. The meagre physical remains at the site were left more or less untouched, but a museum known as the Museum of the Mycenaean Colonisation on Cyprus was built nearby. Professor Vassos Karageorghis is cited on the official website of the Government of Cyprus, Press and Information Office as saying that

The power of the simplicity of the nationalist argument is not to be underestimated, ensuring ease of transmission and intelligibility to a mass audience. Through its reliance on visual metaphor it dispenses with the need for mass literacy and a common tongue, and, in the process, monumentalises the landscape and its archaeological content as physical illustrations to the nationalist text. Physical objects can thereby stand as symbols representing either the group as a whole or some trait or historical episode pertaining to the group or its collective identity. This fact is central to the functioning of modern tourism. As Urry notes, the tourist gaze is ‘… constructed though signs, and tourism involves the collection of signs. When tourists see two people kissing in Paris what they capture in the gaze is “timeless romantic Paris”’ (1990: 3). In

… this is a museum of nothing, a place for reflection and memory. But for Cyprus the significance is much greater. This is where the Hellenisation of the island started, a process which lasted more than 100 years. There we can put our finger on the protagonists of this historical moment and prove with indisputable evidence when and how the island became Greek (Kypros.org 1997).

1. There have been a number of recent studies which examine the relationship between nationalism and archaeology, and nationalism’s use of the past. Chief amongst them are Kohl and Fawcett (1995), DiazAndreu and Champion (1996) and Atkinson, Banks, et al. (1996). 2. Many authors have commented on the construction at a popular level of anti-hegemonic narratives or the subversion of hegemonic narratives (cf. Herzfeld 1991, 1997; Silverman 2002). In the case of Cyprus, it can be argued that while the government might favour imagery of the Mycenaean past and Greek mythology for a host of economic and ideological reasons, for most Cypriots, it is the Christian past that has a much more immediate resonance on a personal ‘everyday’ level. This does not deny the potency for ordinary Cypriots of official narratives in other contexts where for example the desire is felt to defend or promote the ‘Greekness’ of the island particularly in relation to Turkish claims in the north, or make use of the economic potential inherent in such narratives.

By misleadingly conflating ancient Mycenaean with modern Greek identity the archaeological data provided by this site is manipulated to produce a narrative of ethnic continuity spanning over three millennia on the island. The evolving, external nature of social identity, contingent not only on historical but on social circumstances (Rowlands 1994: 132) precludes such an interpretation. A further concern is the nature of information presented at sites and museums. The archaeological heritage of a 183

SOMA 2004 this way, these symbols can then be promoted for sale to both an international as well as domestic audience. At the same time, national self pride can be boosted through the promotion of a glorious and attractive past (Hobsbawm 1998, 28-34). A recent Newsweek headline encapsulates this idea effectively, ‘It is often thought that civic pride can be bought and cultural tourism can pay for it’ (Plagens 2002). According to Sophocles Hadjisavvas, the inclusion of Paphos and Palaeopaphos on the World Heritage List in 1980 was an event of huge significance for the Department of Antiquities and for its Heritage Management policy and that besides the honour and the upgrading of the site, it contributed enormously to local appreciation of the archaeological record. He says:

the Minister of Education, Mrs. Claire Angelidou, at the opening of the international academic symposium ‘Cyprus in the 11th Century BC’ held in Cyprus in 1993: With the difficulty which society as a whole has to adjust to the con-comitant, ever-changing social reality, society instead strives to preserve its memory. This is why the present symposium reminds us of the Greek origins of the island, which we should struggle to preserve intact. Moreover, our foreign guests have a unique opportunity to get to know and love our island and its people, to live the drama of our uprooted people and of our country divided by the Turkish invaders, and to become ambassadors for our just cause in their own countries. We do not ask for anything other than justice. We want to return to our homes: there, where we first saw the light of the sun; there, where we buried our ancestors; there, where we christened our children. There, where the archaeological remains bear witness to the Greek character of the island; in Salamis, Soloi and Lamboussa (Angelidou 1993: xvxvi).

People who would never have had an understanding of their immediate man-made environment and archaeological sites, at last realised through international recognition what they actually possess (1995: 3).

As anyone who has ever visited the site and town of Paphos can attest, it may not be unfair to suggest that the increased local appreciation lay more with the economic possibilities arising from this honour than with the archaeological knowledge it promoted.

This evokes a further issue that I can only point out here and that is the practice of using indigenous artefacts in foreign museums to promote the country of origin. Archaeology and archaeologists thus become cultural and political ambassadors abroad. The country of origin and any ideological issues thereby promoted are thus maintained high in international public awareness and further tourist interest and revenue are generated.

For the purposes of tourism such neatly packaged, concise symbols are easier to market and sell than the more complicated and nuanced reality of the continuously evolving nature of our knowledge of the past and of the historical process. They are more readily digestible in a climate which promotes and succeeds by the sound-bite. The danger inherent in this environment is that such reliance on simplified symbols and metaphors can mean that a stereotyped and one-dimensional image of the nation is presented and promoted. This is largely due to the nature of successful modern advertising and thus Cyprus becomes the island of Aphrodite, home of mythology and inherently Greek in nature. Furthermore, authoritative narratives which seek to promote certain groups/periods to the exclusion of others or to stake territorial claims are bolstered and archaeological remains become a commodity to be bought and sold to the highest bidder—in this case, the tourist industry. Much of the public face of archaeology is concerned with the representation of knowledge within the tourist industry aimed at both a domestic and international audience. This industry is further fostered and financed by government, and thus archaeology becomes a far reaching and versatile vehicle for the transmission of authoritative narratives.

Research into site and museum presentation in Cyprus that I carried out in summer 2003 confirms that the views expressed by Mrs. Angelidou are represented in the visitor information provided at sites and museums in Cyprus. Across the board emphasis was placed on the arrival of the Mycenaeans, associated in certain academic and hence public arenas with the origins of Greek culture and identity on the island. An information plaque in the Cyprus Museum tells the visitor: After the fall of Knossos, just before 1400 BC, the Mycenaeans expanded to the Eastern Mediterranean and established themselves in the large commercial centres mainly Cyprus, from where they traded with the neighbouring countries.

It goes on to describe various pottery types and on getting to Mycenaean IIIC I pottery says that it ‘coincides with the arrival of Mycenaean colonists’. The arrival of the Mycenaeans or Achaeans is a common thread running through most of the information displays at sites and museums. Sometimes, as at Paelopaphos/Kouklia (a World Heritage Site) ancient sources are quoted recounting the legends of the arrival of these groups, and it is stated that archaeological evidence proves the validity of the myths. What is also interesting about this particular site is that the museum is housed in the onsite medieval and later

The selection of particular archaeological sites for promotion to the public is one aspect of this as we have already seen at Kourion and Maa Palaiokastro, but the wording of information where provided at sites and museums is another important factor in the transmission and maintenance of such narratives. Feelings at an official level on what information should be provided and how are well expressed in the following excerpt from an address by 184

DEIRDRE STRITCH Ottoman fort which has been reconstructed. Despite the fact that the fort, too, is an archaeological site, nothing on the site or in the building tells of its history or its role over time in the local community. The same is true of Kolossi castle. This medieval castle was built in the crusader period and remained in use right up until the British colonial period. While it is advertised well both in pamphlets and brochures and on the roadways, the building is left to speak for itself, with absolutely no onsite information provided on the building, its grounds or history. A booklet can be bought for an additional Cy£2.50. It provides good information especially on the debate as to the original building date of the castle and about the nearby mediaeval sugar factory, but it provides little or no information for the subsequent Venetian to modern periods. This is a chronological gap found at many sites with phases from later periods. Limassol Medieval Museum for example, is housed in a restored medieval castle with prison cells added in the Ottoman period, the story of which is ignored in the narrative of the building and enclosed museum. In this way the physical historical environment is manipulated and transformed to provide concrete evidence and reminders of a particular authoritative text.

Acknowledgments The fieldwork for this paper was carried out with the patient assistance (car-driving!) of Dr. Wendy Logue and Miguel Mendez. I thank them both warmly for all their good humour and help; the work could not have been carried out without them. I would also like to thank Professor Demetrios Michaelides for his comments and suggestions following the presentation of an earlier draft of this paper at the POCA Dublin Workshop in 2003. Further thanks are owed to Dr. Christine Morris and Dr. Amanda Kelly for their comments and encouragement throughout the several revisions of this paper. Finally, I am grateful for the useful feedback and comments of an anonymous referee. All mistakes, however, remain with the author. Bibliography Angelidou, C. (1994) Address. IN: Karageorghis, V. ed. Proceedings of the International Symposium, Cyprus in the 11th Century BC. A.G. Nicosia, Leventis Foundation & University of Cyprus, xv-xvi. Argyrou, V. (1995) Greek Cypriot Nationalism and the Poverty of Imagination. IN: Wallace, P.W. ed. Visitors, Immigrants and Invaders in Cyprus: Proceedings of the Cyprus Conference. Albany, Institute of Cyprus Studies, 196-201. Atkinson, J. Banks, I. & J. O’Sullivan, eds. (1996) Nationalism and Archaeology/Scottish Archaeological Forum. Glasgow, Cruithne Press. Cyprus.gov.cy, The Early Years of the Tourist Industry–The Turkish Invasion and the Subsequent Reconstructions, Internet. Available from: http://www.cyprus.gov/cy/ cyphone/govhome.nsf/0/9DBD25AB8A81C9FAC225 6B6B002F977A?OpenDocument&languageNo=1> [Accessed 2 March 2004]. Cyprus-mail.com (2001) £3 million for Curium facelift. Internet. Available from: [Accessed 4 February 2004]. Cyprus-mail.com (2002) Archaeologists Defend Curium Repairs. Internet. Available from: [Accessed 4 February 2004]. Cyprus-mail.com (2003a) Kadar, T. Green Party Slam Curium Work. Internet. Available from: [Accessed 4 February 2004]. Cyprus-mail.com (2003b) Mita,A. Green Fury Over Curium U-turn. Internet. Available from: [Accessed 4 February 2004]. Cyprus-mail.com (2003c) Greek Press – Philelefthertos. Internet. Available from: [Accessed 4 February 2004].

It was made clear to me through my fieldwork that the tensions between the interests and concerns of a government department and the preservation and presentation of archaeological material and knowledge can be overwhelming. Archaeologists produce the raw material for an industry which at the same time supports the archaeologists’ endeavours. It seems to me now that this is a cycle from which it would be very difficult to extricate archaeology and so we must find ways to retain the beneficial elements for both parties whilst protecting each component from the negative impact they might have on the other. In the current model a dominant narrative, albeit born of ideological or financial concerns, silences the nuances and complexities of actual historical processes in time and space by focusing on territorial and ethnic continuities and homogeneity (Graham, Ashworth, et al. 2000: 58) while denying the voice of competing identities and histories in that same time and space. Economic agendas place value for money above preservation or other knowledge-based concerns, whilst at the same time potentially bolstering the aforementioned dominant narratives. Tourism can indeed be of benefit not just to society but to archaeology; however, forceful marketing, which becomes economical with the truth, and excessive visiting numbers that damage sites are pitfalls that need to be avoided (Fowler 1992: 128). It is my contention that we in archaeology have an ethical obligation to represent all who have used the landscape and their respective narratives. Furthermore, we must seek to preserve more of our archaeological heritage than that part which is immediately marketable. We can do this by having a more reflective understanding of the environment in which we work and the inherent constraints on our discipline stemming from the institutionalisation of our profession within the framework of the nation-state. 185

SOMA 2004 Diaz-Andreu, M. & T. Champion, eds. (1996) Nationalism and Archaeology in Europe. London, University College London Press. Doumas, C.G. (2001) Interpreting the Past in Modern Greece. IN: Silberman, N.A. and E.S. Frerichs, eds. Archaeology and Society in the 21st Century: The Dead Sea Scrolls and other Case Studies. Jerusalem, Israel Exploration Society, The Dorot Foundation, 75-92. Fowler, P. (1981) Archaeology, the Public and the Sense of the Past. IN: Lowenthal, D.& M. Binney eds. Our Past Before Us Why do we Save it? London, Temple Smith, 56-69. Fowler, P. (1992) The Past in Contemporary Society, Then, Now. London, Routledge. Gellner, E. (1987) Culture, Identity and Politics. Cambridge, Cambridge University Press. Gellner, E. (1983) Nations and Nationalism. Basil, Blackwell. Graham, B. Ashworth, G.J. & J.E. Tunbridge eds. (2000) A geography of heritage: power, culture, and economy. London, Arnold. Hadjisavvas, S. (1995) Cyprus Department of Antiquities, 60 Years After. The Monuments Branch Report of the Director of Antiquities Cyprus. Nicosia. Herzfeld, M. (1991) A Place in History: Monumental and Social Time in a Cretan Town. Princeton, Princeton University Press. Herzfeld, M. (1997) Cultural Intimacy, Social Poetics in the Nation-State. London, Routledge. Hobsbawm, E. (1998) On History. London, Abacus. Hutchinson, J. (1994) Cultural Nationalism and Moral Regeneration. IN: Hutchinson, D. and A.D. Smith, eds. Nationalism. Oxford, Oxford University Press, 12231. Kohl, P.L. & C. Fawcett, eds. (1995) Nationalism, Politics and the Practice of Archaeology. Cambridge, Cambridge University Press. Kypros.org (1997) Maa Palaeokastro Museum. Internet. Available from: [Accessed 9 November 2003].

Lowenthal, D. (1985) The Past is a Foreign Country. Cambridge, Cambridge University Press. Mouliou, M. (1996) Ancient Greece, its classical heritage and the modern Greeks, aspects of nationalism in museum exhibitions. IN: Atkinson, J. Banks, I. & J. O’Sullivan, eds. Nationalism and Archaeology/ Scottish Archaeological Forum. Glasgow, Cruithne Press, 174-99. Plagens, P. (August 12th 2002) Flights of Folly. Newsweek. Rowlands, M. (1994) The Politics of Identity in Archaeology. IN: Bond, G. & A. Gilliam, eds. Social Construction of the Past: Representation as Power. London, Routledge, 129-43. Silberman, N.A. (1995) Promised Lands and Chosen Peoples: the Politics and Poetics of Archaeological Narratives. IN: Kohl, P.L. & C. Fawcett, eds. Nationalism, Politics and the Practice of Archaeology. Cambridge, Cambridge University Press, 249-62. Silverman, H. (2002) Touring Ancient Times: The Present and Presented Past in Contemporary Peru, American Anthropologist 104 (3), 881-902. Sivan, R. (1997) The Presentation of Archaeological Sites. IN: de la Torre, M. ed. The conservation of archaeological sites in the Mediterranean Region, An international conference organised by the Getty Conservation Institute and the J. Paul Getty Museum, 6-12 May 1995, Los Angeles, The Getty Conservation Institute, 51-9. Urry, J. (1990) The Tourist Gaze, Leisure and Travel in Contemporary Societies.London, Sage Publications. Zerubavel, Y. (1994) The Death of Memory and the Memory of Death: Masada and the Holocaust as Historical Metaphors. Representations 45, 72-100. World-tourism.org (2002) Tourism Proves as a resilient and stable economic sector. Internet. Available from: [Accessed 4 November 2003].

186

Body, Senses and Space in Late Bronze Age Central Macedonia, Greece Vasileios Tsamis

open hearths where communal gatherings could occur. The stand-alone houses mainly served daily practices such as accommodation, cooking, storage and the production of pottery and tool-making. Although there is no evidence that each house was occupied by a single family, these structures could accommodate at least four people. During the Early Iron Age, the existence of large open yards with hearths and stand alone-houses is replaced by small agglomerated spaces with multiple uses. Distinguishing single houses from the agglomeration is difficult and may suggest a communal way of sharing food, space and materials. Mud brick and wattle and daub are the two preferred building materials, and, in some habitation periods, can be found applied within the same building phase.

Introduction Archaeology has been slow in incorporating the body and the senses as central to its field of investigation. It is only recently that studies on embodied practices and experiences have been published (e.g. Meskell and Joyce 2003). However, the majority of such works are still influenced by the dominant role of vision or are focused on representation without engaging on the potential nature of interpreting the corporeal. Recently published material in archaeology has introduced the role of taste, smell and mnemonic consumption (Hamilakis 2002), the importance of sound and memory in the creation of rock art (Goldhahn 2002) and the role of the human body in everyday life (Knapp & Meskell 1997). In addition, these works have rejected and downplayed cognitive perception and the importance of the mind in understanding and sharing experiences. As a result, an emphasis has been placed on studies that illustrate sensory memory in interpreting the world (e.g. Seremetakis 1994, Stoller 1998). Instead of placing an emphasis on the way people in the past perceived, understood and interpreted their surroundings, these publications evoke new ways of experiencing and ‘consuming’ the past. Studies on bodily senses came to the fore in an effort to criticise and reject culture/nature, mind/body dichotomies (e.g. Hamilakis, Pluiennik, et al. 2002, Thomas 1996). The role of the superior mind to an inferior body has been reversed, promoting the value of corporeality. However, what is needed is the appropriate balance in relation to the material under consideration

The prominent role of space in prehistoric Macedonia is used to interpret the past in functional terms, whereas I believe that an embodied interpretation of space can provide a new approach in understanding this past. In general, my research will engage in the interpretation of bodily actions and space, putting people back into the places they created and used, identifying space as a sensual experience. The primary task is to identify ways of experiencing and remembering in Bronze Age societies of central Macedonia through the embodiment of spatial organisation and material culture. The choice to use the notions of ‘sensuous sense’ and ‘modes of subjectification’ was made for the following two reasons. First, I wanted to escape from the mind/ body dichotomy and focus on experiencing as a process. Second, instead of putting an emphasis on groups of people, I emphasise the role of the individual in creating collective experiences. This article introduces the above notions and explains how the terms ‘space’ and ‘memory’ are being used. The discussion continues with a brief account of my selected case studies, providing a set of examples on how space was potentially embodied and memory experienced. The notion of embodied space is suggested as one of central importance in central Macedonia, and one which can provide alternative reconstructions of its past.

Moreover, the aforementioned studies were focused on a specific set of material evidence (human figurines, rock art) without being interested in built space and spatial organisation. In contrast, my research re-introduces space as embodied and experienced. The creation of mound settlements, along with the lack of any burials, illustrates that space had a prominent role in Late Bronze Age – Early Iron Age Macedonian society. During the Late Bronze Age these sites had stand-alone buildings with one or two rooms, while during the Early Iron Age there is a shift towards the agglomeration of small spaces. The early layout of these settlements was dictated by large yards with 187

SOMA 2004 we cannot understand it. However, ‘sensuous sensing’ is not taking place on top of a ‘true’ or normal event; there is no one true experience and interpretation, but potential ways of experiencing.

Embodied Experiences It seems that in the effort to interpret embodied practices we are confused as to whether bodily experiences include or exclude perception and understanding. However, we can suggest that there is a distinction between ways of sensory experience: that of experience, which can have multiple meanings and connotations, and that of experiencing, which is unique and directed towards the individual and the collective. However, understanding collective past bodily experiences is a difficult task, since it involves modern assumptions on the way people in the past experienced themselves and their surroundings. It is even harder to try to identify individual experiences and interpretations. Hence, since individual bodily experiences are not easily detectable in the archaeological record, I decided to implement Deleuze’s notion of collective individuations, where collective experiencing is shaped by individual actions (1995: 98-9).

Embodied Space Spatial archaeology was based on the assumption that the world is composed of geometric entities and has to be understood as a text or language to be deciphered (e.g. Clark 1977; Hodder & Orton 1976). Human experience of space and the importance of remembering were totally ignored in the shaping of identities and relations among people. Space should not be viewed as a language or text with hidden meanings and formations (e.g. Grahame 2000) but as a bodily experience with personal, public and social levels. An attempt to apply phenomenology to archaeology and to understanding the role of space and place in human experience of the world can be found in Christopher Tilley’s A phenomenology of landscape. Influenced by Heidegger’s notion of being-in-the-world, Tilley embarks in an attempt to illustrate how people in the past experienced and understood their world (Tilley 1994: 11). Tilley’s work provides an understanding of human experience of monumental architecture and landscape, illustrating the visual impact and the importance of memory and identity. However, the book never escapes the nature/culture dualism, and it emphasises heavily the visual perception of the world.

An example will illustrate this concept. Open spaces with hearths within settlements that have evidence for cooking and consumption can provide information on the embodied experience of feasting. This bodily experience can be viewed as a group practice that may have links with the past and serves in negotiating power, identity and access to resources. Participation in this event demonstrates its popularity and validity within the community. However, this embodied practice is the result of individual embodied experiences. Hence, the experience of drinking, eating, dancing and sharing are modes of subjectification that express potential individual experiences. In contrast, feasting is a collective individuation, the sum of these experiential events and their potential collective character. Hence, instead of looking for individuals we should seek modes of subjectification which are specific collective individuations relating to a specific event or experience (Deleuze 1995: 138-40). Modes of subjectification refer to events expressing potential individual experiences, while collective individuations are the sum of these experiential events and their potential collective character (Deleuze 1995: 98-99). Hence, past bodily experience is to be found in the potential ways of subjective and collective bodily expressions. This simultaneous uniqueness and multiplicity of human embodiment is acknowledged in my research. Therefore, my approach will include perception as part of bodily experience, and I will acknowledge that sensory experience is both individual and collective.

Therefore, in my research, space is understood as a constantly changing process which is influenced and influences the human body. Space triggers memory and the senses, creating experiences of place within time. I also perceive space as a political, ideological and experiential process which is changing shape and meaning though time. While corporeality of space influences its meaning and shape, space is also interpreted, remembered and forgotten, influencing the formation processes of collective individuations. Since the human body interacts with the world through its senses, it is crucial to investigate in every detail the space in which past societies lived and experienced their surroundings. In my research, space will be seen as embodied, provoking feelings, experiences and identities, constantly changing and being shaped by society, influencing and being influenced by bodily senses. In addition, place will be interpreted as the ‘focus’ of bodily senses, intense embodiments that incorporate stronger experiences for certain people or groups. This does not indicate that there is a distinction between space and place. On the contrary, in my research space is place and place is space (cf. Casey 1997). The difference is in the embodied experiences that have different meaning for different people. Hence, when space triggers embodied practices that have importance for certain people then space is place. The fact that these embodied experiences affect people differently illustrates how someone’s place can be someone else’s space.

Using this guideline, I argue that it is only through the interrelationship between corporeal experiencing and cognition that human embodiment can be fully understood. The human body experiences the world through physical presence and perceptual meaning (Merleau-Ponty 1945: 138). Accordingly, the term ‘sensuous sense’ is a process, whereby experiencing incorporates the corporeal and perception. The bodily senses and the ways of interpreting them are understood as intertwined, interdependent and not subject to categorisation. Without bodily senses, we cannot experience the world, while without interpretation 188

VASILEIOS TSAMIS The embodiment of space indicates the relationship between ‘sensuous sensing’ and ‘modes of subjectification’. Furthermore, the process of embodiment is a choice, since the body-subject experiences space by means of its belongingness and involvement with the social world. The embodiment of space is a set of bodily experiences and preferences and a means by which situations are made meaningful. This does not imply that one cannot choose one’s way of being, but it emphasises that one can never experience or stand in a cultural vacuum (Merleau-Ponty 1945: 434-8). I will apply this theoretical model to Late Bronze Age – Early Iron Age Macedonia in an effort to illustrate how the inhabitants of this region experienced space through the places they created.

Late Bronze Age Macedonia: Assiros, Kastanas and Toumba Thessalonikis My research looks at the province of Central Macedonia which is situated in northern Greece (Fig.1). It has a Mediterranean climate near the coast and a Central European climate further inland. This region is dominated in the north by high mountains, while two major rivers provide natural barriers in its division into east, west and central Macedonia. The period I investigate is the Late Bronze – Early Iron Age, beginning around 1300 BC and lasting until 850 BC (Table 1). During this ‘transitional’ period Macedonia witnesses a widespread appearance of mound settlements. The three sites I am examining are Assiros, Kastanas and Toumba Thessalonikis. All three formed part of a densely habituated region, where mound settlement was the preferred mode of dwelling.

Sensuous Memory Memory is closely integrated with the collective, place, and the senses in the works of major theorists such as Halbwachs (1950), Bachelard (1964), Casey (1987), Connerton (1989) and Nora (1989). Halbwachs stresses the role of society in the creation of memory and that the past is not kept in individual memory (1950: 45). Bachelard analyses domestic space as triggering memories through its smells, spatial orientation and interaction with people (1964: 4-6). Space is viewed as containing compressed time, since it is understood as ‘accumulating’ different experiences from different people. Casey illustrates how the body plays a major role in remembering and forgetting the past (1987: 21316), while Connerton states that memory is embedded in bodily practices having personal and social significance (1989: 71-2). Finally, Nora sees memory as an actual phenomenon in past societies, while in modern society we need ‘places of memory’, crucial for creating a locus of remembrance or forgetting (1989: 7).

Chronological Table of Macedonia Period

From

To

Early Neolithic

6700/6500 BC

5800/5600 BC

Middle Neolithic

5800/5600 BC

5400/5300 BC

Neolithic Age

Late Neolithic

5400/5300 BC

4700/4500 BC

Final Neolithic

470/4500 BC

3300/3100 BC

Early Bronze Age

3300/3100 BC

2300/2200 BC

Middle Bronze Age

2300/2200 BC

1700/1500 BC

Late Bronze Age

1700/1500 BC

1100 BC

1100 BC

700 BC

Bronze Age

Iron Age Early Iron Age

TABLE 1 CHRONOLOGICAL TABLE OF CENTRAL MACEDONIA (BASED ON COMMENTS BY ANDREOU, S., FOTIADIS, M. AND K. KOTSAKIS, 2001: 298-306)

Hence, memory and embodied practices are heavily interdependent and have great significance in interpreting the past. Therefore, in my research memory is not confined to a pure mentalist or subjective sphere. It is also not diminished to behaviourism or habitual sensory experience. Instead, memory is understood as culturally meditated material practice that is activated by embodied experiences and practiced through material culture. Memory is the horizon of sensory experiences, storing and restoring the experience of each sensory dimension as well as finding sensory records outside the body in a surround of objects and places (Seremetakis 1994: 10-11). This approach interprets memory as a bodily experienced process that is triggered by sensory practices and manifested in everyday life. Hence, ‘sensuous memory’ is the mnemonic process intertwined with sensory experience in such a manner as to render each process of remembering and forgetting an embodied practice. Sensuous memory is the interplay, witnessing, experiencing and understanding of sensory experiencing in space and across time.

Located in the Axios river valley, the tell of Kastanas is 110 metres in length, 50 metres wide and 14 metres in height. During the course of its habitation the site was situated within a marshland and was an island. It was inhabited almost continuously from 2400 BC to 200 BC. Characteristic elements include the frequent remodelling of its settlement plan and the consumption of wild animals whose remains are represented in unusually high proportions during the Early Iron Age (Hänsel 1989). The tell of Assiros is located in the Langadas Basin 80km north from the major city of Thessaloniki. It is approximately 110 metres in length, 70 metres wide and 14 metres high. It was situated next to two small streams and was inhabited from 2200BC to the seventh century BC. Due to the presence of large storage rooms during the thirteenth century BC, and the continuity in building plan, it was interpreted by its excavators as playing a redistributive role in the region, resembling the Mycenaean palaces in their early period (Wardle 1983; 1997). 189

SOMA 2004

FIGURE 1 THE SITES OF ASSIROS, KASTANAS AND TOUMBA THESSALONIKIS IN CENTRAL MACEDONIA, GREECE (DRAWING BY VASILEIOS TSAMIS) Situated in the northeast plain of the Thermaic Gulf, Toumba Thessalonikis is located 12km from the sea. This mound rises 23 metres above the ground and is 120 metres wide at its base and 230 metres in length. It is the largest of the three tells and the only one close to the coast. It was inhabited from 2200 BC and abandoned during the fourth century BC. Its proximity to the coast, the evidence for grape pips (indicating wine production) and its storage rooms led to its interpretation as an important centre and one which had trade relations with Mycenaean colonies in the region (Andreou and Kotsakis 1994; Andreou 2001).

memory and links with the past. This distinctive process of habitation is influencing and influenced by distinctive embodied practices and experiences. Since the formation process of these mounds was continuous, these sites are seen as constantly changing and negotiating space. Therefore, events of abandonment and reconstruction are not interpreted as the result of invasion and colonisation, but as periods where new ways of experiencing are introduced. Assiros, Kastanas and Toumba Thessalonikis are sites where the settlement plan of earlier phases was being rebuilt and preserved. This process of rebuilding and reusing of building material is experiential, promoting memory and creating potential ways of interpretation. It also illustrates the aspirations and dreams of the people who lived in these sites and their effort to maintain a way of life that resembled that of the past. This is a continuous effort to maintain existing building plans and specific bodily experiences. However, memory is not always about remembering, but also about forgetting. Constant destructions by fire and earthquake occur in Assiros and

Layers of Memory All three sites were situated within diverse environments. The most striking feature is the creation of distinctive habitation space which resulted in the formation of mound settlements. Instead of focusing on the origins of these mounds and on their regions of influence, I engage in the analysis of potential bodily experiences. Their distinctive shape and prominent location are understood as the result of continuous habitation. This demonstrates the effort to maintain specific corporeal experiences, triggered by 190

VASILEIOS TSAMIS Kastanas. Those which were the result of human activity were interpreted by the excavators as ‘proof’ of invasion and colonisation (e.g. Hänsel 1989). However, the continuity in building plans after events of destruction by fire diminished the impact of such events. Therefore the excavators of the three sites concluded that these events of invasion and colonisation were either small in scale or did not leave any material traces. Instead, I suggest that these events should be understood as processes of forgetting and attempting to provide different bodily experiences. These settlements were not static but interacting, constantly changing and vibrant communities. Hence, they are not investigated as complete bounded entities with specific spatial organisation, but as continuously developing in size and shape.

and shaping identities. Therefore, the centralisation of storage can be viewed as control of ‘sensuous sensing’. The absence of such facilities in later stages in Assiros and Toumba illustrates that bodily experiences are continuous processes with potential ways of interpretation. For me, continuity and absence demonstrate that both strategies exist simultaneously. What is found in the archaeological record is the result of choice. Conclusion In order to create an archaeology of the senses, I have concluded that we need to clarify that the body engulfs the mind and thus that perception and cognition are a simultaneous process of embodiment. Furthermore, in my research I have rejected the distinction of identifying collective and subjective embodied practices. Instead, Deleuze’s notion of collective individuations (1995: 13840) can act as a tool in analysing collective embodied acts in the past, through which personal experiences can emerge. ‘Sensuous sense’ and ‘sensuous memory’ illustrate my focus on the process of corporeal experience that includes perception and the material connotations of remembering through the human senses. Lastly, space and place are understood as being determined by their importance in triggering sensuous sensing and sensuous memory.

Embodied practices The surrounding walls of Assiros and Toumba Thessalonikis along with the earthworks at the base of Kastanas were interpreted as defensive and symbolic structures (Hänsel 1989; Wardle 1989), necessary for the creation of additional living space (terraces). However, rather than focusing on the end product and its implications for the way of life at these sites, I promote the experiential value of building and rebuilding. Building a wall can be viewed as an ongoing process frequently recurring through time, with potential bodily experiences which promote memory continuity, and through this process lead to collective individuations and potential ways of interpretation. The process of rebuilding and reusing material for the construction of a surrounding wall is experiential. People worked together, sharing space and materials. Sounds of building and speaking, smells of sweat, mud and possibly drink may have accompanied the workers. Therefore, the process of construction would include various embodied practices that can be used to promote memory. Collective embodied experiences practiced during the construction of a surrounding wall can serve to highlight links with the past, a line of descent and possibly links with the ancestors.

Archaeology of the body and the senses is a new field in archaeology, but it is more than that. It is a new way of looking and doing archaeology and experiencing archaeology personally. It is a new way of experiencing the embodied selves of the past and their relations with others in their society. Hence, it is a very sensitive theoretical development in terms of experiencing space, material culture and emotions. We have to understand that the parameters engaged in the understanding of the body are many and continuously changing. Archaeologists have to be selective rather than encyclopaedic, since considering everything about the body will result in a parade of overwhelming information. This will create a sense of relativism and nihilism, producing a negative approach towards the understanding of past-embodied experiences.

In Assiros, especially between 1350 BC and 1200 BC, there is evidence for the existence of centralised storage facilities. A similar construction is found in Toumba Thessalonikis during the twelfth and eleventh century BC. In contrast, Kastanas does not have any traces suggesting similar practices. These storage facilities were interpreted as proof for relations with southern Greece. However, the corporeal value of such evidence was not addressed. Storage of specific foodstuff can be viewed as an attempt to control bodily experiences. It can also serve to illustrate social relations and the power to mobilise people and energy in the organisation of communal storage facilities. The memory of such an event can be triggered through the visual impact of a storage building and the smell of stored grains. Consumption of the stored foodstuff can be understood as control over what is consumed, and therefore, through taste, identities are being shaped and negotiated. This illustrates that consumption and storage of grains can also have experiential values incorporating memory

Therefore, my research on the embodiment of space focuses on the relations between the body and its senses and the organisation of domestic and public spaces. Although my approach engages in understanding the embodied experience of past people in central Macedonia, it will provide insights into a new way of doing archaeology by putting people back in the places they lived. Hence, this is an effort to illustrate how people experienced their lives through the spaces they created and inhabited across time. Finally I hope that my approach has demonstrated that control over memory and sensory experience was crucial in the creation of social organisation in Late Bronze Age – Early Iron Age central Macedonia. Bodily experiences can provide alternative interpretations on a region dominated by invasion and colonisation models. 191

SOMA 2004 Therefore, the creation of a tell settlement is viewed as a process of remembering and forgetting with potential ways of experiencing.

Halbwachs, M. (1992) [1950] On collective memory. Edited, translated and with an Introduction by Lewis A. Coser. The heritage of sociology. Chicago, Chicago University Press. Hamilakis, Y. Pluciennik, M. and S. Tarlow eds. (2002) Thinking through the body. New York, Kluwer Academic/Plenum Publishers. Hamilakis, Y. (2002) The past as oral history: towards an archaeology of the senses. IN: Hamilakis, Y. Pluciennik, M. and S. Tarlow, eds., Thinking through the body. New York, Kluwer Academic/Plenum Publishers, 121-36. Hänsel, B. (1989) Kastanas in einem Siedlungshugel der Bronze- und Eisenzeit Makedoniens 1975-1979: Die Grabung und der Baubefund. Prähistorische Archäologie in Sudosteuropa 7(2). Berlin, Volker Spiess. Hodder, I. and C. Orton (1976) Spatial analysis in archaeology. Cambridge, University Press. Knapp, B. A. and L. Meskell (1997) Bodies of Evidence on Prehistoric Cyprus. Cambridge Archaeological Journal 7(2): 183-204. Merleau-Ponty, M. (1945) Phénoménologie de la perception. Gallimard, Paris. Meskell, L. and R. Joyce (2003) Embodied lives: figuring ancient Maya and Egyptian experience. London, Routledge. Nora, P. (1989) Between history and memory: Les lieux de mémoires. Représentations, 26, 11-19. Seremetakis, N. (1994) The senses still: perception and memory as material culture in modernity. Chicago, Chicago University Press. Stoller, P. (1989) Taste of ethnographic things. Philadelphia University Press, Philadelphia. Thomas, J. (1996) Time, Culture and Identity: an interpretative archaeology. London, Routledge. Tilley, C. (1994) A phenomenology of landscape. Oxford, Berg. Wardle, K. (1983) Assiros. A Macedonian settlement of the Late Bronze and Early Iron Age. Αρχαία Μακεδονία, 3, 291-305. Wardle, K. (1989) Excavations at Assiros Toumba 1988. A preliminary report. Annual of theBritish School at Athens, 84, 447-63. Wardle, K. (1997) Change or continuity: Assiros Toumba at the transition from Bronze Age to Iron Age. ΑΕΜΘ, 10, 443-57.

Bibliography Andreou, S. (2001) Exploring the patterns of power in the Bronze Age settlements of Northern Greece. IN: Branigan, K. ed., Urbanism in the Aegean Bronze Age. Sheffield studies in Aegean Archaeology. Sheffield, Sheffield Academic Press, 56-66. Andreou, S. and K. Kotsakis, (1994) «Μυκηναϊκή Παρουσία;», «Μυκηναϊκή Περιφέρεια;»: Ητούμπα Θεσσαλονίκης, μια θέση της εποχής Χαλκού στη Μακεδονία. Η περιφέρεια του Μυκηναϊκού κόσμου. Πρακτικά Ά Διεπιστημονικού Συμποσίου, Λαμία, 10716. Andreou, S., Fotiadis, M. and Kotsakis, K. (2001) Review of Aegean Prehistory V: Neolithic and Bronze Age of Northern Greece. IN: T. Cullen (ed.) Aegean Prehistory: A review. American Journal of Archaeology Supplement I. Boston: Archaeological Institute of America, 259-327. Bachelard, G. (1964) The Poetics of Space. Boston, Beacon Press. Casey, E. S. (1987) Remembering: A phenomenological study (Second edition). Bloomington and Indianapolis, Indiana University Press. Casey, E. S. (1997) The fate of place: a philosophical history. Berkeley, University California Press. Clark, M. (1977) Spatial Archaeology. London, Academic Press. Connerton, P. (1989) How societies remember. Cambridge, Cambridge University Press. Deleuze, G. (1995) Negotiations 1972-1990. New York, Columbia University Press. Goldhahn, J. (2002) Roaring Rocks: An Audio-Visual Perspective on Hunter-Gatherer Engravings in Northern Sweden and Scandinavia. Norwegian Archaeological Review 35(1), 29-61. Grahame, M. (2000) Reading Space: Social interaction and identity in the Houses of Roman Pompeii: A syntactical approach to the analysis and interpretation of built space. British Archaeological Reports, International Series 886, Oxford, BAR Publishing.

192

A New Look at Minoan Metalworking Techniques Thomas Tselios

Introduction The Department of Archaeology of the University of Crete, the National Centre for Scientific Research Demokritos, and the Herakleion and Chania Museums have established a project dealing with Minoan metalworking in Crete. Among the main objects of the project is the study of copper-based artefacts in combination with the evolution of their relevant manufacturing techniques. To accomplish this, the project has adopted an inter-disciplinary approach that employs combined archaeological, analytical and metallographic methods of study on selected copperbased artefacts. The main aim of this paper is to examine the technological information on the metalworking techniques of the Prepalatial (2800-2000 BC) and First Palatial periods (2000-1700 BC), in order to reconstruct the manufacturing history of the artefacts and to detect any possible relations between technological and archaeological information.

The Selection of Samples The 69 artefacts examined consist of 32 long daggers (Fig. 2A), 9 triangular daggers (Fig. 2B), 10 chisels (Fig. 2C), 8 double axes (Fig. 2D), and 3 single axes. This selection of artefacts was based on chronological and spatial criteria. More specifically, the samples chosen were those that were chronologically the earliest, based mainly on the studies of Xanthoudides (1924) and Branigan (1968), and those objects that come primarily from the Mesara plain, in central Crete, as well as from the isthmus of Hierapetra, in eastern Crete (Fig. 1). The artifacts have not been subjected to major conservation, so the structural data which is under investigation here should provide a reliable picture of the ancient technology. Methodology - Metallography in General Metallography is the study of polished sections of metallic materials using an optical microscope that reflects light passing through the objective lens onto the specimens’

FIGURE 1 MAP OF CRETE WITH THE EXAMINED SITES 193

SOMA 2004

FIGURE 2 PHOTOGRAPHS SHOWING A LONG DAGGER, A TRIANGULAR DAGGER, A CHISEL AND A DOUBLE AXE (HERAKLEION ARCHAEOLOGICAL MUSEUM (PHOTOGRAPH: THOMAS TSELIOS) surface (Scott 1991: 57). The reflected light passes back through the objective lens to the eyepiece, which enables the surface structure of the section to be studied. Thus, metallography, with the additional diagnostic power of the polarising filters ‘nicols’, is an important tool that can provide information on: 1.

2. 3. 4.

grinding and polishing on horizontal discs with a diamondimpregnated cloth (grain size 6 and 1 μm) (Buchwald & Leisner 1990: 64-65). Recording the Resulting Observations The next step was the examination of the polished section under a reflected light microscope, when preliminary description took place. Photomicrographs were taken under the microscope. Subsequently, the samples were etched for a few seconds with alcoholic ferric chloride (Scott 1991: 72) and examined again under the microscope (Figs. 3 and 4). This procedure produces grain contrast and thus allows examination of grain boundaries, phases and other features of the microstructure. Photomicrographs were then taken again.

The manufacturing process used to produce the object (the ‘chaine opératoire’ of the metalworking process). For instance, whether it was cast into a mould or worked to shape by hammering and annealing.1 The thermal history of the object. The nature of corrosion products. Much useful and varied information can be obtained from corroded fragments (Piccardo 2000: 13-15). The nature of the metal or alloy employed to make the object. Very often, however, it is extremely difficult, if not impossible, to identify the components just by looking at the microstructure, and there is a need for an additional analytical method to obtain such information.

The following features must be noted in such examinations (Scott 1991: 67): 1. 2.

The range and type of grains. The presence of different phases defined as follows by Scott; ‘Phase is a homogeneous

Sampling, Mounting and Preparing Specimens Wedge-shaped pieces 1-2 mm2 were removed from the cutting edge of each artefact using a very fine saw, so the technology of these different parts could be compared. These pieces were then mounted in synthetic resin in small moulds. This synthetic resin was left for approximately 24 hours to harden and then followed by preliminary FIGURE 3 PHOTOMICROGRAPHS TAKEN UNDER A REFLECTED LIGHT

1. Annealing is the heating of the metal to some 30%, if pure, or 50%, if an alloy, of the appropriate melting point (Northover and Evely 1995: 93).

MICROSCOPE SHOWING THE MICROSTRUCTURE OF TWO OBJECTS

194

THOMAS TSELIOS

FIGURE 4 DRAWINGS SHOWING SOME METALLOGRAPHIC STRUCTURES (AFTER SCOTT 1991, FIG. 11A-F). THE FIRST FIGURE SHOWS THE STRUCTURE OF AN ARTEFACT CAST INTO A MOULD. THE SECOND FIGURE IS THE SAME CAST OBJECT AFTER ANNEALING. IF THIS ARTEFACT IS SUBMITTED TO HAMMERING, WE GAIN THE STRUCTURE C; AND IF THIS ARTEFACT IS SUBMITTED TO ANNEALING AGAIN THEN THE STRUCTURE IN FIGURE D IS PRESENT Code 1 2 3 4 5 6 7 8

Table 1: Types of Manufacturing Techniques (Chaine Opératoire) Cast Cast => Hammering Cast => Annealing Cast => Hammering => Annealing Cast => Annealing => Hammering (Cast)* => Hammering => Annealing => Hammering (Cast)2 => Annealing => Hammering => Annealing Annealing => Hammering => Annealing => (Cast)2 =>

* The casting process cannot be identified after such a long procedure and so many stages.

3. 4. 5. 6. 7.

chemical composition and uniform material, describing one component in a metallic system’ (Scott 1991: 142). Gross heterogeneity between different areas of the sample. Grain size, surface deformation features, or heat-treated zones at cutting edges and worked surfaces. The distribution of inclusions, slag particles and porosity. The distribution of any corrosion products. The presence of twin lines within the grains and whether the twin lines are straight or curved. The presence of strain lines within the grains. The presence of dendrites in cast alloys. The presence of cracks in the specimen.

Hammering

It is clear that the complexity of the metalworking process increases from technique 1 to 8. The term complexity is used here only to denote the number of metalworking steps involved in each manufacturing sequence. More steps may result in tougher and harder objects, but this also depends on the kind of alloy used. The identification of ‘simple’ or more ‘complex’ techniques does not, therefore, necessarily reflect the technological level of metalworking at a specific site or period.

Statistical Analysis The features identified via metallography reflect specific metalworking sequences. After the metallographic examination of the samples, a repetition of the manufacturing stages was noted and is summarised in the classification shown in Table 1.

The triangular daggers (Graph 2), which form one of the earliest dated groups of objects under study (mainly EMII-III),3 were manufactured using techniques with few metalworking steps: casting followed by hammering or annealing. Long daggers, which include both earlier (EMII-MMIA) and later (MMIB-MMII) pieces, were manufactured by both simple and complex techniques (Graph 3). Simpler techniques were used for the earlier pieces and complex techniques for the later examples. Chisels (EMIII-MMII) (Graph 4) and axes (MMI-II) (Graph 5) were both manufactured using complex techniques, as they belong to later periods, and also because of the need for them to be strong enough to be used as tools.

The metallographic examination of these artifacts indicated that the most popular techniques are Nos. 2, 5 and 7, as shown on Graph 1.2

Conclusions After this brief analysis several conclusions can be drawn. The chronologically earlier groups of objects (i.e.

2. The number given on the X-axis of each graph refers to the code in Table 1, which corresponds to the specific metalworking sequence.

3. Early Minoan (EM) and Middle Minoan (MM) period. All dates BC EMII: 2500-2200, EMIII: 2200-2100, MMIA: 2100-2000, MMIB: 20001900, MMII: 1900-1700.

8. 9. 10.

195

SOMA 2004

40

2,2

2,0

Number of Triangular Daggers

Number of Artifacts

30

20

10

0 1

2

3

4

5

6

7

1,8

1,6

1,4

1,2

1,0 ,8 1

8

3

5

7

8

GRAPH 2 METALLOGRAPHY OF TRIANGULAR DAGGERS (GRAPH BY THOMAS TSELIOS)

GRAPH 1 DISTRIBUTION OF METALWORKING TECHNIQUES (GRAPH BY THOMAS TSELIOS)

3,5

12

3,0

8

2,5

6

2,0

Number of Chisels

10

Number of Long Daggers

2

4

2

0 2

4

5

6

7

8

1,5

1,0

,5 2

GRAPH 3 METALLOGRAPHY OF LONG DAGGERS (GRAPH BY THOMAS TSELIOS)

3

5

7

GRAPH 4 METALLOGRAPHY OF CHISELS (GRAPH BY THOMAS TSELIOS)

4,5

20

4,0 3,5 3,0 2,5

Number of artifacts

10

Number of Axes

2,0 1,5 1,0 ,5 1

2

3

5

7

8

type of site Cave Settlement Tomb

0 1

GRAPH 5 METALLOGRAPHY OF AXES (GRAPH BY THOMAS TSELIOS)

2

3

4

5

6

7

8

GRAPH 6 METALLOGRAPHY PER TYPE OF SITE (GRAPH BY THOMAS TSELIOS)

196

THOMAS TSELIOS triangular daggers and the earlier group of long daggers) were manufactured using simpler metalworking techniques, while the later groups of objects (i.e. axes and chisels) used more complicated techniques, involving more metalworking steps. Therefore, the use of different techniques in metalworking seems to have a chronological link, as the acquisition of technological knowledge increased over time.

such a differentiation is insignificant. This observation, in combination with the traces of usage on most objects, indicates that the findings from burials are personal tools of the deceased and were not manufactured specifically for burial purposes. The study is still at a preliminary stage. Future work will include chemical analyses of all the samples using a scanning electron microscope, as well as the metallographic examination of additional objects. This combination of metallographic and analytical data, together with the archaeological information, will be used to define specific technological groups and assess whether particular patterns can be discerned between different chronological periods or geographical areas.

However, different artefacts or even different types of the same artefact are made by different types of metalworking techniques. For instance, a carpenter’s double axe, after casting in a mould, needs many metalworking steps in order to make the cutting edges sharp and tough. On the other hand, a double axe for status or votive purposes does not have to be tough, and so it does not need numerous metalworking steps. This indicates that another parameter should be considered apart from the acquisition of technological knowledge: the intended use of each artefact which determines the metalworking process that will be followed.

Acknowledgements I would like to thank my supervisors, Dr. Yannis Bassiakos (N.C.S.R. Demokritos), Dr. Iris Tzahili (University of Crete) and Dr. Carole Gillis (University of Lund), for their invaluable help and support throughout my work. Funding for this project was kindly provided by the General Secretary for Technology and Research, Greek Ministry of Development.

This question of usage can be partially answered by metallographic examination. As is obvious from Graph 6, most of the examined artefacts come from burial contexts. One would expect that artefacts purely for burial use would be made using the fewest possible metalworking processes and the least amount of labour. However, that the opposite happened cannot be excluded either. So the question that arises is whether these artifacts were made intentionally for burial purposes, or whether they were artefacts from everyday life, deposited following their owner’s death.

Bibliography Branigan, K. (1968) Copper and Bronze Working in Early Bronze Age Crete, SIMA XIX, Lund, Bloms Boktryckeri. Buchwald, V.F. and P. Leisner (1990) A Metallurgical Study of 12 Prehistoric Bronze Objects from Denmark. Journal of Danish Archaeology 9, 64-102. Northover, P. and D. Evely (1995) Towards an Appreciation of Minoan Metallurgical Techniques: Information provided by Copper Alloy Tools from the Ashmolean Museum, Oxford. Annual of the British School at Athens 90, 83-105. Piccardo, P. (2000) Setting up an Investigation Method for the Study of Corrosion of Ancient Burial Bronze Objects. European Microscopy and Analysis 63, 1315. Scott, D.A. (1991) Metallography and Microstructure of Ancient and Historic Metals. Singapore, Tien Wah Press. Xanthoudides, S. (1924) The Vaulted Tombs of Mesara. London, University Press of Liverpool.

As can be seen on Graph 6, the artefacts found in burial contexts are made using the same techniques as those from settlements. As far as triangular daggers are concerned, no significant metallographic difference is noted between settlement and burial context. However, their manufacturing process is generally very short (mainly techniques 2 and 3, Table 1), which makes differentiation between the triangular daggers made for use and those made for possible votive reasons, difficult. Additionally, there are traces of wear that testify to the previous use of most of the triangular daggers. With respect to the other artefacts that were manufactured using more metalworking steps, like the long daggers and axes, it is easier to notice any possible differentiation between artefacts from burial and settlement contexts. Still, as is shown in Graph 6,

197

Debates and Trends on Heritage Management and Architectural Conservation in the Mediterranean Region: The Issue of the Driving Forces Behind Conservation Planning Kalliopi Vacharopoulou

Introduction This paper arises from research on the concept and practice of anastylosis on archaeological monuments of the Mediterranean region, especially in Greece, Italy and Turkey. Anastylosis was introduced as a concept and practice at the beginning of the nineteenth century, and its implementation is related to the various problems associated with archaeological monument preservation in the Mediterranean.

At the dawn of the twenty-first century, conservation and management of archaeological and cultural heritage have expanded and are the subject of various debates related to the degree of intervention that should occur, concepts of authenticity and integrity, issues of heritage ownership and values, the role and involvement of stakeholders in planning processes, and the role of driving forces such as interpretation, education, tourism, and identities, in the making of conservation decisions. These debates should be explored as they have an impact on, and determine our responses to, the protection of archaeological monuments.

Modern concepts about the protection and conservation of archaeological heritage are founded on the new historical consciousness that developed in Europe from the eighteenth century onwards. Their main principles found their first expression at that time (Jokilehto 1996: 56) and were intensified and systematised during the nineteenth century, when, as noted by Lowenthal, ‘the European nations closely identified themselves with their material heritage’ (1985: 385). With the passage of time, a policy of minimum intervention was advocated and induced by conservation principles such as those of minimum alteration of historic fabric, retention of a maximum of the original structure, and minimal risk of loss or damage as a result of interventions (Warren 1996: 39).

Among the most important problems and questions remaining in the area of the preservation and presentation of archaeological monuments is the issue of the aforementioned driving forces that dictate conservation implementation. Various degrees of intervention are employed, from preservation and consolidation to restoration, reconstruction and anastylosis. These interventions do not only aim at structural protection and conservation, but are also interwoven with issues of interpretation, education, identity, and tourism.

Throughout the twentieth century various organisations, such as the United Nations Educational, Scientific and Cultural Organisation (UNESCO), the International Centre for the Study of Preservation and Restoration of Cultural Property (ICCROM), the International Council of Monuments and Sites (ICOMOS), to name the most prominent, were established to develop awareness of heritage and promote international collaboration for its protection and safeguarding. These organisations proceeded to form international conventions, which are legally binding for the signatory countries, and international recommendations or charters that remain moral guidelines and contribute to defining standards for conservation treatments (Jokilehto 1996: 57-8 and 64-6). Moreover, various theories and philosophies of conservation and restoration have developed, accompanied by technological experimentation.

Various Degrees of Intervention: Terminology, Meanings, and Debates According to Pye, conservation activities are thought ‘to prevent and/or rectify the effects of existing or past deterioration, balancing the needs for access and use of heritage with the requirement of long-term preservation’ (2001: 24). Restoration is the form of treatment applied to monuments to return them to a known or presumed earlier state (New Zealand Charter 1992, Article 22) and enhance their aesthetic and historic values (Venice Charter 1964, Article 9). However, in Italy the concept of restoration has remained the main term for describing all treatments (Jokilehto 1996: 67), whilst in other European languages and contexts, restoration is used to cover both the activities of restoration and conservation (Pye 2001: 26). 199

SOMA 2004 A characteristic example of restoration is the intervention in the Arch of Titus in Rome, Italy, around 1800. The arch appears to have survived almost intact, while a closer examination shows the differences between the original damaged centre and the new parts in the outer areas (Schmidt 1997: 41).

Driving Forces Behind Architectural Conservation: Interpretation, Education, Identities, Tourism, and Public Access Various concerns—interpretation and education, scientific experiment and research, tourism development, local and cultural identities—lie behind the increasing number of extensive interventions carried out on archaeological monuments.

Anastylosis is described as the re-assembly of existing but dismembered parts of a monument (Venice Charter 1964, Article 15). It was introduced at the Athens Conference of 1931 and was endorsed as the only acceptable form of reconstruction by the Venice Charter of 1964. However, at national level, especially in Greece and Italy, it implies restoration. An example of anastylosis is the re-erection of the Acropolis monuments by Balanos during 1895-1940 and the re-erection of the façade of the Library of Celsus in Ephesus, Turkey by Hueber and Strocka between 1970 and 1978 (Schmidt 1997: 43 and 46). Other examples of the use of anastylosis are to be found in the Treasury of the Athenians at Delphi, Greece, and the Temple of Hadrian in Ephesus, Turkey, in which anastylosis was carried out mainly with original building elements (Mertens 1995: 125)

Interpretation is particularly important as it can make visitors’ experiences richer and more meaningful. It should depend on the surviving physical evidence, taking into consideration the site in its entirety, size, and aesthetic values (Sivan 1997: 51-2). Associated with interpretation is education, which has become a significant aspect of conservation efforts and forms the basis of heritage presentation because of the growing emphasis on making the past interesting, attractive, accessible and educational to the public (Bower 1995: 34 and 37-8). At the same time, information derived from a restored monument is appreciated by many groups (de la Torre and MacLean 1997: 9). Interpretation and education represent longterm investments and serve different functions for different stakeholders. They can improve the quality of visitors’ experiences by giving context and meaning to a site or a monument, thus making them more enjoyable. For the community, they enable a greater sense of place and feelings of ownership of local heritage—as they provide tangible testimony to the community’s past, and, hence, their continuity and ownership of a place— whilst for heritage managers they raise awareness and understanding of the values and uses of heritage (Hall and McArthur 1998: 168). The above-mentioned notions are best reflected in Tilden’s dictum: ‘through interpretation understanding, through understanding appreciation, through appreciation protection’ (as quoted in Uzzell 1998: 12).

Reconstruction is determined as partial or full-scale, accurate, duplication of a monument that should be documented through archaeology, archival research, or physical evidence (Jameson and Hunt 1999: 35). Indicative of this technique is the reconstruction of the Stoa of Attalos in the Athenian Agora undertaken by the American School of Classical Studies in Athens in the mid-twentieth century. Only small fragments of the structure had survived, but the building was erected as a completely new monument (Schmidt 1997: 42). Another example of reconstruction is found in the Roman baths of Sardis, Turkey, in which full-scale reconstruction was undertaken despite only small amounts of original material surviving (Mertens 1995: 126).

Restoration and reconstruction are favoured as efficient interpretative and educational tools when they are well planned and do not damage the heritage resource (Jameson and Hunt 1999: 60; Colomer 2002: 85). Their main value is found in their power of visual communication (Colomer 2002: 86). However, it is argued by Dimacopoulos that educational value should be the result, not the primary objective of reconstruction (in AA 1996: 43). His main argument, also outlined by Schmidt, is that reconstruction ‘can translate any inconspicuous archaeological ruin into a comprehensible structure’ (1999: 65). This is why interpretation and education are much related and are based on the improvement of legibility, which can make a building’s spatial volumes more easily visualised (Schmidt 1997: 42; Sivan 1997: 52). Thus, acquisition of a third dimension through anastylosis and restoration is considered to facilitate understanding of the character of ruined monuments, as is noted at Epidauros (Maurommatidis 1987: 21; Kyriaki 1988: 11-12), the Athenian Acropolis (Papanikolaou 1994: 147; Papanikolaou 1989: 4-5) in Greece, in the Temple of Trajan in Pergamon (Nohlen

Consequently, it is understood that ambiguity appears in usage and application of these terms for intervention. No strict boundaries are found between terms whilst overlapping areas are apparent, especially due to differences in national terminologies and contexts. The primary objective of each treatment is, ideally, the contribution to the understanding and safeguarding of archaeological heritage. Hence, the common aim of all the above intervention methods is the reinstatement of the form, or part thereof, of the monument, as closely as possible through restoration and anastylosis and as exactly as can be achieved with reconstruction (Feilden and Jokilehto 1998: 65). However, within the scope of this objective, the debate on what constitutes the appropriate scale of intervention is endless. The extent and location of any new work often becomes a controversial issue and requires sensitive and rational contemplation of conservation objectives (Feilden and Jokilehto 1998: 65). 200

KALLIOPI VACHAROPOULOU 1997: 188 and 191) and the Library of Celsus in Ephesus (Hueber and Strocka 1975: 10) in Turkey.

in taking its first steps (Mallouchou-Tufano 1998: 55). Similarly, the reconstruction of the Minoan palace in the site of Knossos in Crete, Greece, is a source of national (Greek) and local (Cretan) pride (Papadopoulos 1997: 99). Furthermore, the archaeological site of Ephesus, as the principal representative of Hellenistic and Roman cities that once existed on the Asia Minor coast, is identified as a symbolic link between Turkey and Europe (Demas 1997: 131). Hence, excavation and restoration works have been undertaken there for the last 100 years.

Furthermore, in the approach known as experimental archaeology, reconstruction is employed to test archaeological theories (Schmidt 1997: 47) and introduce ancient techniques to the general public (Colomer 2002: 87). Use of experimental archaeology to research and understand the past and to present it to, and use it to educate, the public, can be an effective approach to the issue of interpretation and an important aspect of educational visits to sites (Killebrew and Lehmann 1999: 5-6; Stone and Planel 1999: 7). Models of monuments and other structures are constructed based on surviving evidence, and the more ways in which they are tested means that greater interpretational reliability can be placed upon them. Experimental archaeology is valued because of its educational potential and its public appeal, which in turn brings appreciation and support for further work (Coles 1979: 33 and 36). Furthermore, people understand the way in which archaeology builds up its interpretations, making them more in control of their own learning experience next time they visit a site (Colomer 2002: 92).

The idea of nation has been, and still is, articulated through the promotion of ideas of continuity and tradition (Walsh 1992: 90 and 114). Hence, in this regard, national Greek identity is celebrated through a focus on its classical heritage, as became obvious with the first restoration of the Parthenon on the Athenian Acropolis (Mallouchou-Tufano 1998: 55; Jokilehto 1999: 89). However, questions of identity and the nature of the meaning and values inherent in archaeological monuments indicate the possibility of conflicting notions of ownership attached to heritage held by the various stakeholders in it and, therefore, highlight the existence of contradictory sets of values and interests (Hall and McArthur 1998: 4). In Ephesus, the use of ancient monuments for religious purposes illustrates the difficulties of reconciling different values. The religious values of the monuments occasionally conflict with the historical record and, therefore, with the interpretation and use of the monuments in the present (Demas 1997: 145).

Such an example is found in the approach adopted at the Lemba Experimental Village in Cyprus, where a research project on experimental buildings became a means of communication between specialist archaeologists and the public. The aim is to maintain a dialogue between the public and the archaeologists without turning to a more didactic approach, which can be patronising or misleading. The program is carried out as part of research into the Bronze Age period in western Cyprus, which, among other aspects, incorporates a full-scale construction of a Chalcolithic village next to actual archaeological remains of such a village. It has the final goal of testing certain aspects of prehistoric construction, of observing and defining processes of site formation, and of creating an archaeological visitor centre. Visitors can understand the site on many levels depending upon their level of interest and the purpose of their visit (Gordon 1999: 110-20).

In addition, in multi-cultural societies, archaeological heritage can help to develop the population’s awareness of a shared historical identity, either in geographical or cultural terms (Byrne 1991: 275). In this sense, as Herbert indicates (1995: 13-14), many nations have used concepts of common roots and shared past achievements reflected in buildings and monuments in order to create a sense of unity. In Megiddo, a biblical archaeological site in Israel, excavations have revealed archaeological remains from the sixth millennium BC to the fourth century BC. The remains from the Canaanite and Israelite cities are poorly preserved, but in order to attract the local (mainly Jewish) population to the site, plans have been made to illustrate these aspects of Tel Megiddo’s history, including the Assyrian period (Killebrew 1999: 25-6).

The promotion of national and cultural identities is often a major driving force behind the restoration or reconstruction of monuments, because, as Killebrew and Lehmann point out, ‘politics and ideologies are closely interwoven with site interpretations’ (1999: 4). In the case of the Hellenistic Stoa of Lindos, in Rhodes, Greece, initially restored in the 1920s by Italians, the main aim of the reconstruction was to ideologically support the Italian military presence on the island, by promoting the GreekRoman idea. Hence, they proceeded with restoration of classical and roman antiquities (Papadimitriou 1988: 170). On the other hand, restoration works on the Parthenon on the Athenian Acropolis were initiated when Greece had just emerged as an independent state after the long period of Ottoman occupation. Reinforcement of the Greek national identity was essential in assisting the nation, which had shared a common identity in the past,

Yet, in most cases where multiple ethnic groups exist within a country or region, the dominant group in society uses its power to advance its own heritage as it shapes a national identity according to its own image (Byrne 1991: 275). For instance, Qasrin, a small Byzantine-period Jewish village in the Golan Heights was selected for excavation in the 1970s and 1980s to demonstrate ancient Jewish presence in the Golan. Initially, the focus was on the monumental synagogue, but following the reconstruction of a fifth to sixth century AD house, the focus shifted towards the village’s domestic structures. However, though three main periods of occupation exist at the site (the Jewish Byzantine site, the Muslim Mamluk settlement and a recent Syrian 201

SOMA 2004 village), only the Jewish Byzantine-period village is presented to the public, due to political considerations and the desire to emphasise the Byzantine (i.e. Jewish) remains for propaganda purposes. Thus, as argued by Killebrew, presentation of the past at Qasrin gives ‘a somewhat distorted picture of the site’s occupation through history’ (1999: 28). The village has become a symbol of modern Jewish settlement and the setting for political rallies that support Israeli territorial claims in the Golan. At the same time, significant government funds were allocated to the site for further development (Killebrew 1999: 26-8).

Consequently, researchers are deprived of their study material (Herrmann 1989: 31; Blockley 1999: 18) and satisfaction in the experience of heritage deteriorates (Hall and McArthur 1998: 87). This is noted in various cases around the Mediterranean, such as the archaeological site of Ephesus on the western coast of Turkey (de la Torre and Mac Lean 1997: 11; Demas 1997: 146), where millions of tourists circulate unrestricted on the site. In Knossos, Crete, the arrival of large groups of visitors has placed pressure on the monument’s original fabric, as the original paving has been much eroded, both by natural weathering and by foot traffic (Papadopoulos 1997: 113). Therefore, heritage managers should safeguard archaeological sites and monuments, while as Herrmann notes ‘extensive reconstructions and sensational presentations’ should be avoided (1989: 31). The balance must be found between the conflicting objectives of preservation and use, maintenance of historical accuracy, respect of the heritage’s integrity and aesthetic appeal, encouragement of visitors, and the making of profits (Carter and Grimwade 1996: 45-6; Johnson 1985: 13; Shackley 1999, 70; Sivan 1997, 52).

As Jokilehto has observed (1999: 307-8), the concept of ‘national monuments’ is frequently loaded with political values, which in turn can incite reconstruction of desired features of monuments and the elimination or destruction of others that may be contrary to political goals. For instance, in reconstructing ancient Babylon in Iraq, only the ruler power aspect of that culture has been deliberately displayed, rather than the qualities of learning and wisdom of the ancient Mesopotamian civilisation (Blockely 1999: 18).

The solutions usually adopted to address this problem are: restriction and regulation of access, but with sociopolitical and financial restraints; denial of access to endangered monuments and provision of replicas nearby, justified by the already lost sense of originality and the assurance of cultural and educational experiences; and the removal of endangered monuments and reconstruction at new sites (Herrmann 1989: 31-2). However, restricting access is a short-term solution, which becomes less acceptable because it restrains the community from using and appreciating heritage (Carter and Grimwade 1996: 52). The mass tourism in Ephesus has lead to the periodic closing off of large areas and important monuments to visitors and use (Demas 1997: 146). Visitor circulation and its control should be undertaken at all sites (Stubbs 1995: 75). An exemplary solution has been adopted for the site of Akrotiri in the island of Santorini, Greece. Explanatory texts and graphics have been placed along the walkways to accommodate circulation and present the site to visitors, while in order to prevent congestion, only three guided groups are allowed on site at any given time (Doumas 1997: 35). Another example is formed by the creation of a system of raised walkways, resting upon the consolidated wall tops of the Roman Villa at Piazza Armerina in Sicily, Italy. This system makes possible the viewing of all the villa’s mosaics without walking upon them (Stanley Prince 1997: 79; Stubbs 1995: 81).

Furthermore, archaeological and cultural sites have become important economic resources in many parts of the world, and their economic potential can be realised through tourism (de la Torre and Mac Lean 1997: 10-11). For example, the archaeological site of Knossos in Crete is the second most visited site in Greece, representing an important economic resource, both on a national and a local level, as tourism enhances all aspects of local economy through tourist agencies, hotels, restaurants, stores, etc. (Papadopoulos 1997: 99). However, many imbalances, such as the prioritising of heritage’s economic values over other values, become apparent and are probably caused by lack of communication between the tourist industry and the cultural heritage sector. Both sectors may simply pursue their objectives independently, without co-ordination or collaboration (de la Torre and Mac Lean 1997: 10-11). While heritage programs might initiate with education and preservation in mind, they may be forced to compromise on these goals in the face of demands from economy and tourism in order to ensure their long-term survival (Colomer 2002: 89). Archaeological reconstructions can play an important role in advertising and promoting heritage (Colomer 2002: 88). Cultural tourism and economic development provide jobs, lead to modernisation, and opportunities to learn about history and archaeology (McNulty 1985: 34). Investment in the maintenance and development of a site depend on the site’s ability to attract visitors and, thus, generate funds (Killebrew and Lehmann 1999: 4). Archaeologists and heritage professionals have become aware of the destructive potential of tourism. While it can generate funds for conservation and management, it can also create demands for leisure entertainment. The latter becomes a problem when unrestricted and excessive tourist activity can lead to physical destruction of monuments and their surroundings.

Conclusion To address these issues and many others referred to in the introduction, a holistic management approach can be suggested as the best solution. The main aims of a holistic management approach are the accommodating of local, national and cultural identities, the achievement of the educational function of heritage, the provision for controlled tourism, and the safeguarding of the heritage database (Cleere 1989: 10). 202

KALLIOPI VACHAROPOULOU Key steps for its achievement are a management plan, identification and involvement of the key interest groups, assessment of the significance of the site and the values it is perceived to hold, and the definition and subsequent assessment of management policies and strategies (Sullivan 1997: 15-25). Such strategies include visitor management, physical protection of remains, condition monitoring, maintenance and ongoing evaluation (de la Torre and Mac Lean 1997: 13). The importance of physical conservation and heritage restoration still remains, but the conflicting demands of various stakeholders should be prioritised (Hall and McArthur 1998: 8). On the other hand, visitor management is equally important and has become just as significant as the heritage itself (Hall and McArthur 1998: 107; Shackley 1999: 80). Various techniques can be employed in order that the total number of visitors does not detract from true enjoyment of the site, nor prevent proper appreciation of it or cause physical harm to it. Besides, management costs can be reduced, while political support for conservation, foreign currency, jobs and income can be generated (Feilden and Jokilehto 1998: 7 and 100).

Acknowledgements I would like to express my gratitude to both my supervisors (Professor C.A. Price and Tim Williams) for their guidance in developing these ideas and issues in a concrete form. I would also like to thank the Organising Committee of SOMA 2004 for giving me the opportunity to present this paper. Special thanks should also go to Prof. Price for kindly looking at a draft and the revised edition of this paper. All mistakes, though, lay with the author. Bibliography Charters New Zealand Charter: Charter for the Conservation of Places of Cultural Heritage Value, adopted by the ICOMOS New Zealand National Committee, 1992. The Venice Charter: International Charter for the Conservation and Restoration of Monuments and Sites, adopted by the Second International Congress of Architects and Technicians of Historic Monuments, Venice, 1964. Athens Charter: Athens Charter for the Restoration of Historic Monuments, adopted by the First International Congress of Architects and Technicians of Historic Monuments, Athens, 1931.

Unfortunately, though, only few long-term conservation plans are found in the region. As most Mediterranean countries depend on income generated by tourism, archaeological authorities are encouraged to make their sites more attractive to visitors. Presentation and use of sites by the public and development of a tourist infrastructure can enhance the site’s values, but they can also destroy them if these actions are unplanned and uncoordinated. In many cases, these problems could be moderated by practical collaborations among those who have an interest in the survival of these resources, including cultural officials, international and private organisations, and commercial tourism organisations (de la Torre and Mac Lean 1997: 7), as well as the local people. For instance, at Akrotiri in Santorini, Greece, conflict is noted between the aims of site protection and the economic interests of the local population, which could be increased from developing tourist enterprises near the site or from restricted land use by locals. The way this situation is handled by Akrotiri’s heritage managers is realised through talks with municipal representatives and the people most affected (Doumas 1997: 36).

Texts Archaeological Association (1996) Transactions of the Second Convention of the Fellows. Title of the Convention: The Greek Anastyloseis. Mentor. Bulletin of the Fellows of the Athenian Archaeological Association, Volume 37, Athens, Greece. Blockley, M. (1999) Archaeological reconstructions and the community in the UK. IN: Stone P.G. and P.G. Planel eds. The Constructed Past. Experimental Archaeology, Education and the Public. London, Routledge, 15-34. Bower, M. (1995) Marketing Nostalgia: An exploration of heritage management and its relation to human consciousness. IN: Cooper M.A., Firth, A. Carman, J. and D. Wheatley eds. Managing Archaeology. London, Routledge, 33-9. Byrne, D. (1991) Western Hegemony in Archaeological Heritage Management. History and Anthropology, 5, 269-76. Carter, B. and G. Grimwade (1996) Balancing Use and Preservation in Cultural Heritage. International Journal of Heritage Studies 3 (1), 45-53. Cleere, H. (1989) Introduction: The rationale of archaeological heritage management. IN: Cleere, H. ed. Archaeological Heritage Management in the Modern World. London, One World Archaeology, Unwin-Hyman, 1-19. Coles, J. (1979) Experimental Archaeology. Academic Press, London. Colomer, L. (2002) Educational facilities in archaeological reconstructions. Is an image worth more than a thousand words? Public Archaeology (2), 85-94.

Generally, professionals acknowledge the critical approach required by modern conservation. Fundamental changes such as the broader definition of cultural heritage and questions of heritage management have challenged the current methods used for protection and conservation of ancient monuments. These notions are fundamental to successful heritage management and for fulfilling the potential of conservation (Jokilehto 1999: 317-18). It is now the responsibility of current and future generations of heritage specialists, in the Mediterranean and world-wide, to advance this process, by taking discussions forward, critically examining relevant situations, understanding even better and more passionately the world’s heritage. 203

SOMA 2004 de la Torre, M. and M. Mac Lean (1997) The Archaeological Heritage in the Mediterranean Region. IN: de la Torre M. ed. The Conservation of Archaeological Sites in the Mediterranean Region. An International Conference Organised by the Getty Conservation Institute and the J. P. Getty Museum, 612 May 1995. Los Angeles, The Getty Conservation Institute, 5-14. Demas, M. (1997) Ephesus. IN: de la Torre M. ed. The Conservation of Archaeological Sites in the Mediterranean Region. An International Conference organised by the Getty Conservation Institute and the J. Paul Getty Museum, 6-12 May 1995. Los Angeles, The Getty Conservation Institute, 127-49. Doumas, C. (1997) Management considerations at a Mediterranean Site: Akrotiri, Thera. IN: de la Torre, M. ed. The Conservation of Archaeological Sites in the Mediterranean Region. An International Conference organised by the Getty Conservation Institute and the J. Paul Getty Museum, 6-12 May 1995. Los Angeles, The Getty Conservation Institute, 27-40. Feilden, B.M. and J. Jokilehto (1998) Management Guidelines for World Cultural Heritage Sites. Rome, ICCROM. Gordon, Th. (1999) The role of research and education in site management at the Lemba Experimental Village, Cyprus. Conservation and Management of Archaeological Sites (3), 109-23. Hall, C.M. and S. McArthur (1998) Integrated Heritage Management. Principles and Practice. London, The Stationery Office. Herbert, D.T. (1995) Heritage Places, Leisure and Tourism. IN: Herbert, D.T. ed. Heritage, Tourism and Society. London, Mansell, 1-20. Herrmann, J. (1989) World Archaeology - The World’s Cultural Heritage. IN: Cleere, H. ed. Archaeological Heritage Management in the Modern World. London, One World Archaeology, Unwin-Hyman, 30-7. Hueber, F. and V.M. Strocka (1975) Die Bibliothek des Celsus. Eine Prachtfassade in Ephesos und das Problem ihrer Wiederaufrichtung. Antike Welt. Zeitschrift für Archäologie und Urgeschichte (4). Jameson, Jr. J.H. and W.J. Hunt, Jr. (1999) Reconstruction versus preservation-in-place in the US National Park Service. IN: Stone P.G. and P.G. Planel eds. The Constructed Past. Experimental Archaeology, Education and the Public. London, Routledge, 35-62. Johnson, R.N.P. (1985) Report by the chairman of the congress. IN: Second International Congress of Architectural Conservation and Town Planning. Conservation and Tourism. Basle 1-4 April 1985. Heritage Trust, London, 10-13. Jokilehto, J. (1996) International standards, principles and charters of conservation. IN: Marks, S. ed. Concerning Buildings. Studies in Honour of Sir Bernard Feilden. Architectural Press. An imprint of Butterworth Heinemann, 55-81. Jokilehto, J. (1999) A History of Architectural Conservation. Butterworth-Heinemann.

Killebrew, A.E. (1999) From Canaanites to Crusaders. The presentation of archaeological sites in Israel. Conservation and Management of Archaeological Sites (3), 17-32. Killebrew, A.E. and G. Lehmann, (1999) Interpreting the Past: Presenting archaeological sites to the public. Conservation and Management of Archaeological Sites, volume 3, 3-7. Kyriaki, V.E.E.S. (1988) Supplementary Report for the Restoration of the Propylon of the Gymnasium’. Athens, Ministry of Culture, Committee for the Conservation of the Epidauros monuments (in Greek). Lowenthal, D. (1985) The Past is a Foreign Country. Cambridge, Cambridge University Press. Mallouchou-Tufano, F. (1998) Anastylosis of Ancient Monuments in Modern Greece (1834-1939). The Work of the Archaeological Association of Athens and of the Archaeological Service, No 176, Athens, Library of the Archaeological Association of Athens. Maurommatidis I. (1987) Analysis of the Principles Guiding the Anastylosis Proposal for the Avaton. IN: Committee for the Conservation of the Epidauros Monuments. The Stoa of Avaton in the Asklipieion of Epidauros. Proposal for its Conservation and Partial Restoration. Committee for the Conservation of the Epidauros Monuments, Ministry of Culture, Athens, 21-5 (in Greek). McNulty, R.H. (1985) Cultural tourism: new opportunities for wedding conservation to economic development. IN: Second International Congress of Architectural Conservation and Town Planning. Conservation and Tourism. Basle 1-4 April 1985. Heritage Trust, London, 34-41. Mertens, D. (1995) Planning and executing anastylosis of stone buildings. IN: Stanley-Price N. (ed.) Conservation on Archaeological Excavations. With particular reference to the Mediterranean area. Rome, ICCROM, 113-34. Nohlen, K. 1997 Aesthetik der Ruine. Antike Welt (1997), 185-99 (in German). Papadimitriou, V. (1988) The Anastylosis of the Ancient Monuments on theAcropolis of Lindos. Past and Present Attempts. IN: Dietz, S. and I. Papachristodoulou, eds. Archaeology in the Dodecanese. The National Museum of Denmark. Department of Near Eastern and Classical Antiquities. Copenhagen, 169-71. Papadopoulos, J.K. (1997) Knossos. IN: de la Torre, M. ed. The Conservation of Archaeological Sites in the Mediterranean Region. An International Conference organised by the Getty Conservation Institute and the J. Paul Getty Museum, 6-12 May 1995. Los Angeles, The Getty Conservation Institute, 93-126. Papanikolaou, A. (1989) The Restoration Works of the Erectheion. Ministry of Culture, Committee for the Preservation of the Acropolis Monuments. Athens (in Greek). Papanikolaou, A. (1994) The Restoration of the Erechtheion. IN: Economakis, R (ed.) Acropolis Restoration. The CCAM Interventions. Academy Editions, London, 136-49. 204

KALLIOPI VACHAROPOULOU Pye, E. (2001) Caring for the past. Issues in Conservation for Archaeology and Museums. London. James and James. Schmidt, H. (1997) Reconstruction of Ancient Buildings. IN: de la Torre, M. ed. The Conservation of Archaeological Sites in the Mediterranean Region. An International Conference organised by the Getty Conservation Institute and the J. Paul Getty Museum, 6-12 May 1995. Los Angeles, The Getty Conservation Institute, 41-51. Shackley, M. (1999) Visitor Management. IN: Leask A. and I. Yeoman eds. Heritage Visitor Attractions: An Operations Management Perspective. London, Continuum, 69-82. Sivan, R. (1997) The Presentation of Archaeological Sites. IN: de la Torre M. ed. The Conservation of Archaeological Sites in the Mediterranean Region. An International Conference organised by the Getty Conservation Institute and the J. Paul Getty Museum, 6-12 May 1995. Los Angeles, The Getty Conservation Institute, 51-9. Stanley-Price, N. (1997). The Roman Villa at Piazza Armerina, Sicily. IN: de la Torre M. ed. The Conservation of Archaeological Sites in the Mediterranean Region. An International Conference organised by the Getty Conservation Institute and the J. Paul Getty Museum, 6-12 May 1995. Los Angeles, The Getty Conservation Institute, 65-92.

Stone, P.G. and P.G. Planel (1999) Introduction. IN: Stone P.G. and P.G. Planel eds. The Constructed Past. Experimental Archaeology, Education and the Public. London, Routledge, 1-14. Stubbs, J.H. (1995) Protection and presentation of excavated structures. IN: Stanley-Price N. (ed.) Conservation on Archaeological Excavations. With particular reference to the Mediterranean area. Rome, ICCROM, 73-90. Sullivan, S. (1997) A Planning Model for the Management of Archaeological Sites. IN: de la Torre M. ed. The Conservation of Archaeological Sites in the Mediterranean Region. An International Conference Organised by the Getty Conservation Institute and the J. P. Getty Museum, 6-12 May 1995. Los Angeles, The Getty Conservation Institute, 15-26. Uzzell, D. (1998) Interpreting our heritage: a theoretical interpretation. IN: Uzzell D. and R. Ballantyne, Contemporary Issues in Heritage and Environmental Interpretation. The Stationery Office, 11-25. Walsh, K. (1992) The representation of the past: museums and heritage in the post-modern world. London, Routledge. Warren, J. (1996) Principles and Problems: Ethics and Aesthetics. IN: Marks, S. ed. Concerning Buildings. Studies in Honour of Sir Bernard Feilden. Architectural Press. An imprint of Butterworth Heinemann, 34-54.

205