Archaeology of a World of Changes. Late Roman and Early Byzantine Architecture, Sculpture and Landscapes: Selected Papers from the 23rd International Congress of Byzantine Studies (Belgrade, 22–27 August 2016) – In memoriam Claudiae Barsanti 9781407354217, 9781407355290

Published in memory of Prof. Claudia Barsanti, Archaeology of a World of Changes provides a selection of papers presente

443 40 87MB

English Pages [413] Year 2020

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Cover
Title Page
Copyright
Of Related Interest
Contents
Preface
Claudia Barsanti (1941–2017)
List of Publications
Introduction
Part 1. Archaeology, Landscape and Topography
1. The ‘Crypt on N. I. Tur’s Land’ in Early Byzantine Chersonesos: Cherson as a Monument of the Formation of the Church in the City
Mikhail V. Fomin
2. Carian Iasos in Byzantine Times
Diego Peirano
3. The Christian Phase of the Rotunda in Thessaloniki
Sophia Akrivopoulou
4. The Archaeology of the Late Roman City of Zaldapa: The Status Questionis in 2016 (with an Appendix on Seasons 2017–2019)*
Dominic Moreau, Georgi Atanasov & Nicolas Beaudry
Introduction
Reinsertion into time and space
The history of excavations and publications until 2014
Christian Basilicas Nos. 3 and 4
Conclusion: archaeological perspectives
Appendix on seasons 2017 to 2019
5. Archaeological Investigations at Golemo Gradište, Konjuh, 2012–2016
Carolyn S. Snively & Goran Sanev
The Episcopal Basilica
The Northern Residence
History of the anonymous city
6. Le monastère des Quarante Martyrs de Sébastée à Saranda (Albanie), un centre important de pèlerinage pendant l’Antiquité tardive
Skënder Muçaj, Kosta Lako, Skënder Bushi & Suela Xhyheri
7. How Early Christian Complexes Developed in Northern Dalmatia and the Kvarner Islands
Ariadna Voronova
8. La nécropole rupestre de Daras (Turquie). De nouvelles perspectives de recherche
I. La datation relative des carrières et des nécropoles
II. La technique de creusement de la ‘Galeri Mezer’, confirmation d’une inspiration perse
III. Les inscriptions inédites
Conclusion
9. Tracing the Cycladic Settled Landscape in Late Antiquity and the Early Byzantine Period (Fourth to Ninth Century AD): The Islands of Paros and Naxos
Konstantinos Z. Roussos
Introduction
The settled landscapes of Paros and Naxos during Late Antiquity
Paros, Naxos and the Cyclades within the interactive Late Antique Mediterranean world
Paros and Naxos in the Early Byzantine period: Two diametrically opposed stories
The involvement of landscape in the development ofthe settlement pattern of Paros and Naxos during LateAntiquity and the Early Byzantine period
Conclusions
10. Life Among the Ruins: Medieval Settlements and Late Antique Cities, Seen Through Recent Excavations in the Region of the Upper Vardar and Bregalnica Rivers
Dejan Gjorgjievski
Part 2. Episcopal Residences
11. The Episcopal Palace in Early Byzantium: Historical Development, Architectural Typologies, Domestic Spaces and the Case of Kos
Isabella Baldini
12. The Episcopal Palaces: Worldly Splendour in the Style of the Ruling Class versus Monastic Virtues
Helen Saradi
13. The Identification of the Bishop’s Palace at Miletus in Caria (Turkey)
Philipp Niewöhner
History of research
Building history
Function
14. Episcopal Basilica of Dion, Greece: The Triconch Building and the Location of the Bishop’s Residence
Kyriakos Fragoulis
15. Les espaces domestiques et économiques du groupe épiscopal protobyzantin de Byllis (Albanie)
Nicolas Beaudry & Pascale Chevalier
Le contexte urbain
Le quartier épiscopal
Le cæur économique de l’évêché
Les espaces domestiques
Conclusions
Part 3. Architectural Sculpture
16. Marble Masons on Stage: Two Case Studies in the Working Procedures of the Byzantine Sculptural Ateliers
Andrea Paribeni
17. New Evidence for Byzantine Sculptures from the Basilica of St Philip at Hierapolis
Silvia Pedone
18. La basilique du Léchaion (Corinthe) : étude des sculptures architecturales
Christina Tsigonaki
Historique de la recherche
Les sculptures architecturales
La datation des sculptures architecturales
Les données de fouille
Conclusions
19. The Sculptural Decoration of Acheiropoietos Basilica (Thessalonike), Re-evaluated in Light of a Recent Architectural Analysis of the Monument
Konstantinos T. Raptis
The ground floor
The gallery level
Conclusion
20. I capitelli imposta del tipo cosiddetto a pannelli nell’ambito della produzione scultorea postgiustinianea
Claudia Di Bello
21. Chapiteaux corinthiens inédits d’Apamée (Syrie)
Catherine Vanderheyde
De la cité hellénistique à la ville romaine puis byzantine : un bref aperçu
Omniprésence des chapiteaux
Le marbre : une denrée rare en Syrie
Des chapiteaux en marbre dans la maison dite « aux colonnes bilobées »
Conclusions
22. Byzantine Capitals of the Architectural Complex of the ‘Basilica of 1935’ in Chersonesos in Crimea
Liudmila G. Khrushkova
The ‘Basilica of 1935’
Composite capitals with fine-toothed acanthus
Impost capitals
Corinthian capitals
23. Transennae from Nikopolis
Eugenia Chalkia
Transennae from Basilica D
Transennae from Basilica B
Conclusions
24. Il rivestimento marmoreo del fonte del battistero di Nocera Superiore: nuove considerazioni
Alessandra Avagliano
25. The Riddle of the Anatolian Cross Stones: Press Weights for Church or Monastic Estates?
Philipp Niewöhner
Columns
Altars
Cross monuments
Mortars
Press weights
Why the decoration, and why with crosses?
Part 4. Decoration and Small Objects
26. The Representations of Dolphins on the Inlaid opus sectile Panels above the Imperial Door in the Church of Hagia Sophia in Constantinople and their Meaning
Anđela Gavrilović
27. Early Byzantine Silver Chalices from Viminacium: Parts of a Tableware Set or Church Vessels?
Ivana Popović
28. Late Antique Bronze Crosses Found in Punta de l’Illa, Cullera, Valencia, Spain: The Archaeological Remains of Emperor Justinian’s Dream?
María J. S. Vicent
Introduction
The bronze crosses found in Punta de l’Illa
Discussion
Conclusions
Part 5. Restoration and Conservation
29. Nouvelle intervention sur la mosaïque du sanctuaire de Mar Gabriel (Turquie orientale)
Patrick Blanc, Véronique Blanc-Bijon & Sébastien de Courtois
30. Recent Restoration Work in the Hagia Sophia Museum
Asnu-Bilban Yalçın
The western façade
The interior: The north-west half of the tympanum wall
Consolidation of the marble floor of the main nave and archaeological investigations
The galleries
Back Cover
Recommend Papers

Archaeology of a World of Changes. Late Roman and Early Byzantine Architecture, Sculpture and Landscapes: Selected Papers from the 23rd International Congress of Byzantine Studies (Belgrade, 22–27 August 2016) – In memoriam Claudiae Barsanti
 9781407354217, 9781407355290

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Archaeology of a World of Changes Late Roman and Early Byzantine Architecture, Sculpture and Landscapes In memoriam Claudiae Barsanti EDITED BY

DOMINIC MOREAU, CAROLYN S. SNIVELY, ALESSANDRA GUIGLIA, ISABELLA BALDINI, LJUBOMIR MILANOVIĆ, IVANA POPOVIĆ, NICOLAS BEAUDRY AND ORSOLYA HEINRICH-TAMÁSKA

B A R I N T E R NAT I O NA L S E R I E S 2 9 7 3

2020

Archaeology of a World of Changes Late Roman and Early Byzantine Architecture, Sculpture and Landscapes Selected Papers from the 23rd International Congress of Byzantine Studies (Belgrade, 22–27 August 2016) in memoriam Claudiae Barsanti EDI TED BY

DOMINIC MOREAU, CAROLYN S. SNIVELY, ALESSANDRA GUIGLIA, ISABELLA BALDINI, LJUBOMIR MILANOVIĆ, IVANA POPOVIĆ, NICOLAS BEAUDRY AND ORSOLYA HEINRICH-TAMÁSKA

B A R I N T E R NAT I O NA L S E R I E S 2 9 7 3

2020

Published in 2020 by BAR Publishing, Oxford BAR International Series 2973 Archaeology of a World of Changes. Late Roman and Early Byzantine Architecture, Sculpture and Landscapes isbn  

978 1 4073 5421 7 paperback isbn   978 1 4073 5529 0 e-format doi  https://doi.org/10.30861/9781407354217

A catalogue record for this book is available from the British Library © the editors and contributors severally 2020 Jonah sarcophagus, Народни музеј Србије / National Museum of Serbia – Ivan Stanić

cov er i m age

The Author’s 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. Links to third party websites are provided by BAR Publishing in good faith and for information only. BAR Publishing disclaims any responsibility for the materials contained in any third party website referenced in this work.

BAR titles are available from: BAR Publishing 122 Banbury Rd, Oxford, em a i l [email protected] phon e +44 (0)1865 310431 fa x +44 (0)1865 316916 www.barpublishing.com

ox 2

7bp,

uk

Of Related Interest Objects in Motion The Circulation of Religion and Sacred Objects in the Late Antique and Byzantine World Edited by Hallie G. Meredith Oxford, BAR Publishing, 2011

BAR International Series 2247

When did Antiquity End? The Proceedings of an International Seminar held at the University of Trento on April 29-30, 2005 on Late Antique Societies, Religion, Pottery and Trade in Germania, Northern Africa, Greece, and Asia Minor Edited by Redha Attoui Oxford, BAR Publishing, 2011

BAR International Series 2268

The Insular System of the Early Byzantine Mediterranean Archaeology and history Edited by Demetrios Michaelides, Philippe Pergola, Enrico Zanini Oxford, BAR Publishing, 2013

BAR International Series 2523

Sub-series: Limina/Limites: Archaeologies, histories, islands and borders in the Mediterranean (365-1556)

Entangled Identities and Otherness in Late Antique and Early Medieval Europe Historical, Archaeological and Bioarchaeological Approaches Edited by Jorge López Quiroga, Michel Kazanski and Vujadin Ivanišević Oxford, BAR Publishing, 2017

BAR International Series 2852

Sub-series: Archaeological Studies on Late Antiquity and Early Medieval Europe (400-1000 A.D.)

For more information, or to purchase these titles, please visit www.barpublishing.com

Published under the patronage of the 23rd International Congress of Byzantine Studies, with the financial support of the Université de Lille, Foundation I-SITE ULNE (DANUBIUS Project) and the HALMA-UMR 8164 research centre.

Contents Preface................................................................................................................................................................................. ix Claudia Barsanti (1941–2017)........................................................................................................................................... xi List of Publications........................................................................................................................................................ xii Introduction......................................................................................................................................................................... 1 Part 1. Archaeology, Landscape and Topography 1. The ‘Crypt on N. I. Tur’s Land’ in Early Byzantine Chersonesos: Cherson as a Monument of the Formation of the Church in the City................................................................................................................................. 5 Mikhail V. Fomin 2. Carian Iasos in Byzantine Times................................................................................................................................. 13 Diego Peirano 3. The Christian Phase of the Rotunda in Thessaloniki................................................................................................ 23 Sophia Akrivopoulou 4. The Archaeology of the Late Roman City of Zaldapa: The Status Questionis in 2016 (with an Appendix on Seasons 2017–2019)...................................................................................................................................................... 35 Dominic Moreau, Georgi Atanasov & Nicolas Beaudry Introduction.................................................................................................................................................................... 36 Reinsertion into time and space..................................................................................................................................... 37 The history of excavations and publications until 2014................................................................................................ 39 Christian Basilicas Nos. 3 and 4.................................................................................................................................... 45 Conclusion: archaeological perspectives....................................................................................................................... 48 Appendix on seasons 2017 to 2019................................................................................................................................ 49 5. Archaeological Investigations at Golemo Gradište, Konjuh, 2012–2016................................................................. 57 Carolyn S. Snively & Goran Sanev The Episcopal Basilica................................................................................................................................................... 58 The Northern Residence................................................................................................................................................. 64 History of the anonymous city....................................................................................................................................... 65 6. Le monastère des Quarante Martyrs de Sébastée à Saranda (Albanie), un centre important de pèlerinage pendant l’Antiquité tardive........................................................................................................................... 69 Skënder Muçaj, Kosta Lako, Skënder Bushi & Suela Xhyheri 7. How Early Christian Complexes Developed in Northern Dalmatia and the Kvarner Islands.............................. 95 Ariadna Voronova 8. La nécropole rupestre de Dara (Turquie). De nouvelles perspectives de recherche............................................. 105 Anaïs Lamesa & Nihat Erdoğan La datation relative des carrières et des nécropoles..................................................................................................... 107 La technique de creusement de la ‘Galeri Mezer’, confirmation d’une inspiration perse............................................111 Les inscriptions inédites............................................................................................................................................... 114 Conclusion................................................................................................................................................................... 115 9. Tracing the Cycladic Settled Landscape in Late Antiquity and the Early Byzantine Period (Fourth to Ninth Century AD): The Islands of Paros and Naxos................................................................................117 Konstantinos Z. Roussos Introduction.................................................................................................................................................................. 117 v

Archaeology of a World of Changes. Late Roman and Early Byzantine Architecture, Sculpture and Landscapes The settled landscapes of Paros and Naxos during Late Antiquity.............................................................................. 118 Paros, Naxos and the Cyclades within the interactive Late Antique Mediterranean world ........................................ 120 Paros and Naxos in the Early Byzantine period: Two diametrically opposed stories.................................................. 123 The involvement of landscape in the development of the settlement pattern of Paros and Naxos during Late Antiquity and the Early Byzantine period............................................................................................................ 129 Conclusions.................................................................................................................................................................. 134 10. Life Among the Ruins: Medieval Settlements and Late Antique Cities, Seen Through Recent Excavations in the Region of the Upper Vardar and Bregalnica Rivers.................................................................... 137 Dejan Gjorgjievski Part 2. Episcopal Residences 11. The Episcopal Palace in Early Byzantium: Historical Development, Architectural Typologies, Domestic Spaces and the Case of Kos........................................................................................................................... 153 Isabella Baldini 12. The Episcopal Palaces: Worldly Splendour in the Style of the Ruling Class versus Monastic Virtues............................................................................................................................................................. 167 Helen Saradi 13. The Identification of the Bishop’s Palace at Miletus in Caria (Turkey) ............................................................. 175 Philipp Niewöhner History of research....................................................................................................................................................... 176 Building history............................................................................................................................................................ 176 Function....................................................................................................................................................................... 180 14. Episcopal Basilica of Dion, Greece: The Triconch Building and the Location of the Bishop’s Residence.......................................................................................................................................................... 189 Kyriakos Fragoulis 15. Les espaces domestiques et économiques du groupe épiscopal protobyzantin de Byllis (Albanie)................... 201 Nicolas Beaudry & Pascale Chevalier Le contexte urbain........................................................................................................................................................ 203 Le quartier épiscopal.................................................................................................................................................... 205 Le cœur économique de l’évêché................................................................................................................................. 206 Les espaces domestiques.............................................................................................................................................. 209 Conclusions.................................................................................................................................................................. 214 Part 3. Architectural Sculpture 16. Marble Masons on Stage: Two Case Studies in the Working Procedures of the Byzantine Sculptural Ateliers.......................................................................................................................................................... 221 Andrea Paribeni 17. New Evidence for Byzantine Sculptures from the Basilica of St Philip at Hierapolis........................................ 231 Silvia Pedone 18. La basilique du Léchaion (Corinthe) : étude des sculptures architecturales...................................................... 243 Christina Tsigonaki Historique de la recherche............................................................................................................................................ 244 Les sculptures architecturales...................................................................................................................................... 244 La datation des sculptures architecturales.................................................................................................................... 245 Les données de fouille.................................................................................................................................................. 249 Conclusions.................................................................................................................................................................. 250 19. The Sculptural Decoration of the Acheiropoietos Basilica (Thessalonike), Re-evaluated in Light of a Recent Architectural Analysis of the Monument...................................................................................................... 253 Konstantinos T. Raptis The ground floor........................................................................................................................................................... 254 vi

Contents The gallery level........................................................................................................................................................... 263 Conclusion................................................................................................................................................................... 265 20. I capitelli imposta del tipo cosiddetto a pannelli nell’ambito della produzione scultorea postgiustinianea............................................................................................................................................................... 267 Claudia Di Bello 21. Chapiteaux corinthiens inédits d’Apamée (Syrie)................................................................................................. 277 Catherine Vanderheyde De la cité hellénistique à la ville romaine puis byzantine : un bref aperçu.................................................................. 278 Omniprésence des chapiteaux...................................................................................................................................... 280 Le marbre : une denrée rare en Syrie........................................................................................................................... 281 Des chapiteaux en marbre dans la maison dite « aux colonnes bilobées ».................................................................. 282 Conclusions.................................................................................................................................................................. 284 22. Byzantine Capitals of the Architectural Complex of the ‘Basilica of 1935’ in Chersonesos in Crimea......................................................................................................................................................................... 291 Liudmila G. Khrushkova The ‘Basilica of 1935’.................................................................................................................................................. 293 Composite capitals with fine-toothed acanthus............................................................................................................ 294 Impost capitals............................................................................................................................................................. 297 Corinthian capitals....................................................................................................................................................... 298 23. Transennae from Nikopolis...................................................................................................................................... 303 Eugenia Chalkia Transennae from Basilica D......................................................................................................................................... 304 Transennae from Basilica B......................................................................................................................................... 310 Conclusions.................................................................................................................................................................. 313 24. Il rivestimento marmoreo del fonte del battistero di Nocera Superiore: nuove considerazioni........................ 317 Alessandra Avagliano 25. The Riddle of the Anatolian Cross Stones: Press Weights for Church or Monastic Estates? ........................... 327 Philipp Niewöhner Columns....................................................................................................................................................................... 328 Altars............................................................................................................................................................................ 329 Cross monuments......................................................................................................................................................... 330 Mortars......................................................................................................................................................................... 331 Press weights................................................................................................................................................................ 331 Why the decoration, and why with crosses?................................................................................................................ 332 Part 4. Decoration and Small Objects 26. The Representations of Dolphins on the Inlaid opus sectile Panels above the Imperial Door in the Church of Hagia Sophia in Constantinople and their Meaning........................................................................... 339 Anđela Gavrilović 27. Early Byzantine Silver Chalices from Viminacium: Parts of a Tableware Set or Church Vessels?.................. 347 Ivana Popović 28. Late Antique Bronze Crosses Found in Punta de l’Illa, Cullera, Valencia, Spain: The Archaeological Remains of Emperor Justinian’s Dream?..................................................................................................................... 357 María J. S. Vicent Introduction.................................................................................................................................................................. 358 The bronze crosses found in Punta de l’Illa................................................................................................................. 360 Discussion.................................................................................................................................................................... 362 Conclusions.................................................................................................................................................................. 363

vii

Archaeology of a World of Changes. Late Roman and Early Byzantine Architecture, Sculpture and Landscapes Part 5. Restoration and Conservation 29. Nouvelle intervention sur la mosaïque du sanctuaire de Mar Gabriel (Turquie orientale)............................... 367 Patrick Blanc, Véronique Blanc-Bijon & Sébastien de Courtois 30. Recent Restoration Work in the Hagia Sophia Museum...................................................................................... 379 Asnu-Bilban Yalçın The western façade....................................................................................................................................................... 380 The interior: The north-west half of the tympanum wall............................................................................................. 382 Consolidation of the marble floor of the main nave and archaeological investigations.............................................. 383 The galleries................................................................................................................................................................. 386

viii

Preface Three years ago (22–27 August 2016), the 23rd International Congress of Byzantine Studies was held in Belgrade. It was the first congress to be organised in accordance with the new guidelines of the Association des Études Byzantines, which replaced separate plenary papers by plenary sessions with papers discussing the same topic but from different angles; special attention was given to roundtables, which in a way became the very backbone of the Congress. This change allowed the roundtables to evolve into veritable little symposia.

the basis for a wider publication, an approach not unheard of in earlier global meetings of Byzantine scholars. In this sense, an excellent example of a realised publication reflecting our hopes is the book Archaeology of a World of Changes: Late Roman and Early Byzantine Architecture, Sculpture and Landscapes. Selected Papers from the 23rd International Congress of Byzantine Studies (Belgrade, 22–27 August 2016) – the book hereby presented to the reader and edited by Dominic Moreau, Carolyn S. Snively, Alessandra Guiglia, Isabella Baldini, Ljubomir Milanović, Ivana Popović, Nicolas Beaudry and Orsolya HeinrichTamáska, who were all involved in organising a session and/or the presentation of a paper at Belgrade.

Slightly unexpectedly for the organisers of the Belgrade Congress, it was the largest congress of its kind to date: six plenary sessions, each with at least three reporters and a moderator; 49 roundtables; 117 sessions containing free communications; several excursions and accompanying exhibitions – a total of 1,649 participants accredited in various capacities, with over 1,400 scholars from different countries, including the host country (49 in total), presenting the results of their research. In view of the projected role of the roundtables, the proportions of this meeting of researchers of varying profiles facilitated the organisation of a range of symposia as part of the Congress. This possibility thus largely came to fruition.

Their project was to bring together in one volume a selection of the papers from all the thematic sessions and roundtables on Late Roman and Early Byzantine archaeology and history of art, for which publication was not already planned. So, as compared to the congress session from which the idea originated, the number of contributors has tripled and the range of topics discussed in their contributions encompasses the vast distance from Spain through Italy and the Balkans to Anatolia, and includes archaeological evidence of remarkably diverse natures and provenances. In fact, just as we had hoped when we formulated the programme of the Congress, this has brought us into the domain of separate publications, which can be published in different environments and on different topics, but whose cohesive element remains the Congress, with its primary academic focus on studying shifts in the Byzantine world throughout the centuries. The result seems to have surpassed our expectations. We now have a book whose texts, in their diversity and convergence, yield a remarkable synthesis in the field of archaeology of the Late Antique and Byzantine world.

Like every congress of this kind, the international meeting of Byzantine scholars in Belgrade had its own title: Byzantium – A World of Changes, which was meant to represent the final farewell to the still living myth of the static nature of Byzantine civilisation as a historical phenomenon that had experienced little change during its millennium-long existence. It was precisely this idea that lay in the focus or subtext of many congress events. A characteristic example was the thematic session held on 25 August under the relatively flexible title Byzantine Archaeology, which nonetheless had remarkable research potential and was chaired by Dominic Moreau (Université de Lille) and Mihailo Milinković (University of Belgrade). Nine participants applied for attendance and submitted communications with varying content and chronological frameworks, with the order of presentation reflecting the aforementioned main intention of the Congress. It was then that the co-creator of this session, our colleague Moreau, suggested the possibility of a separate publication that would include the communications presented at this little meeting but which could potentially have an even wider basis.

We hope that the organisers of other roundtables and thematic sessions at the Belgrade Congress will follow in the footsteps of the one on archaeology and that other ideas about separate thematic publications will come to fruition. Some have already been announced. Their potential publication before the planned congress in Istanbul in 2021 would complete the achievements and messages of the Belgrade congress. Ljubomir Maksimović Византолошки институт Српске академије наука и уметности / Institute for Byzantine Studies of the Serbian Academy of Sciences and Arts Belgrade, 24 April 2019

This idea coincided with the organisers’ wishes for the moderators of thematic sessions to transform their respective sessions into thematic collections of papers, i.e. separate books, wherever the material presented lent itself to such an endeavour. A session could also provide ix

Claudia Barsanti in Hagia Sophia, 2 June 2014 (photo: Alessandra Guiglia).

Claudia Barsanti (1941–2017) and meticulous monograph San Clemente. La scultura del VI secolo, which she produced together with Federico Guidobaldi and Alessandra Guiglia. In Claudia’s scholarly career, this work was the first in an important series of collaborative efforts conducted together with a small équipe of colleagues with whom she shared a frank and productive exchange of ideas.

Claudia Barsanti was born in Rome on 20 October 1941. From her earliest childhood, her family transmitted to her a passion for and an acquaintance with works of art. This was especially the case with ancient sculpture, thanks to the activity of her great-uncle Alfredo and her father Mario Barsanti, both of whom were prominent antiquarians and collectors in Rome in the first half of the twentieth century. From the very beginning, her family background favoured her love of travel and desire to know faraway places, thereby shaping her character and her future life as a scholar.

At the same time, Claudia started work on other subjects, which she developed in a manner both refined and distinctive. Her series of studies dedicated to the topography of Constantinople was in part also the product of her wandering the streets and visiting the monuments of Istanbul – as well as of her study of the ancient sources and travellers’ reports, of course. She approached this complex subject from two different perspectives. On the one hand, attentive to origins, she followed the formation of the city from its beginnings as Roman Byzantium through its status as the capital of Constantine to the monumental additions of the Theodosian age. These studies were in large part published in the Milion series that Fernanda de’ Maffei established in 1988. On the other hand, she also took a retrospective view of the epilogue of the thousandyear history of Byzantium, by studying the extraordinary documentation offered by the panorama of Cristoforo Buondelmonti, which has come down to us in numerous versions realised over the course of the fifteenth century. Once again it was the Rivista dell’Istituto Nazionale di Archeologia e Storia dell’Arte that published her long monographic study ‘Costantinopoli e l’Egeo nei primi decenni del XV secolo: la testimonianza di Cristoforo Buondelmonti’ in 2001. This work constitutes another fundamental reference work for the history of the capital on the Bosporus.

She received her first degree (laurea) from the University of Rome ‘La Sapienza’ in 1977. Under the direction of Fernanda de’ Maffei, she wrote a substantial thesis on the church of St Mary of Ephesus. It was this thesis that marked the beginning of her work on Byzantine art in its early centuries, with a particular focus on architecture and above all on sculpture, while never neglecting the analysis of sources and historical context. Thanks to her brilliant abilities as a young scholar, and in collaboration with Fernanda de’ Maffei, she took part in numerous research expeditions in Asia Minor, Israel and Egypt during the 1970s. From these travels she drew material for various studies presented at international conferences and in scientific journals. In particular, the investigations in Turkey led to the formulation of a research topic that was decidedly innovative at that time, viz. the Middle Byzantine sculpture of Anatolia. This became the subject of her doctoral thesis, which was the first to be defended in the Department of Art History at the University of Rome ‘La Sapienza’. During the 1980s, she undertook work on those studies that most clearly define her scholarly production. These are focused upon the production of architectural sculpture in the fifth and sixth centuries that came out of the workshops of the Proconnesus and was exported from Constantinople to all of the regions adjoining the Mediterranean and Black Sea. Her 1989 study ‘L’esportazione di marmi dal Proconneso nelle regioni pontiche durante il IV–VI secolo’, which was published in the Rivista dell’Istituto Nazionale di Archeologia e Storia dell’Arte, to this day remains the standard reference work; the piece is fundamental for anyone who wishes to work on Early Byzantine sculpture. From the analysis of the phenomenon on a large scale, Claudia Barsanti goes on to examine an emblematic instance of the importation of marble artefacts from the Proconnesus, viz. the sixth-century marbles today re-employed in the schola cantorum of the Roman basilica of San Clemente. From this study came the dense

In 2014, there came to an end yet another long period of study and research, which had its beginnings amongst the marbles of Hagia Sophia in Constantinople in 1999. Of especial note in this context is the publication in 2004 of the volume Santa Sofia di Costantinopoli. L’arredo marmoreo della Grande Chiesa giustinianea, which was produced by the collaborative effort of the research group of Alessandra Guiglia, Mauro della Valle, Roberta Flaminio, Andrea Paribeni, and Asnu-Bilban Yalçın. Subsequently, the small guide to the lapidary collection of the Ayasofya Müzesi appeared in print in 2010. This publication constitutes a preliminary study for the catalogue of sculptures coming from Hagia Sophia itself as well as from numerous other sites in the city. This volume too was the product of a collaborative effort, and the members of the équipe – Alessandra Guiglia, Roberta Flaminio, Andrea Paribeni, xi

Alessandra Guiglia, Andrea Paribeni & Silvia Pedone List of Publications

Silvia Pedone, Eugenio Russo and Alessandro Taddei – hope to be able to complete the catalogue as an homage to Claudia, who was particularly fond of this project.

‘Le architetture “ad limitem” del Menologio di Basilio II (Cod. Vat. Greco 1613) e la miniatura con la commemorazione del Patriarca Ignazio’, in Commentari, 28, 1977, pp. 3–25.

The last monograph that Claudia published, writing in collaboration with Alessandra Guiglia and Roberta Flaminio, is the volume of the Corpus della Scultura Altomedievale that came out in 2015 as part of the project curated by the Fondazione Centro Italiano di Studi sull’Alto Medioevo. It is dedicated to the sculptural evidence belonging to the third ecclesiastical regio of Rome.

‘Problemi efesini’, in Commentari, 29, 1978, pp. 3–21. ‘Ancora sulla lastra con i monogrammi del monaco Giacinto, il presunto fondatore della Koimesis di Nicea’, in Annuario dell’Istituto di Storia dell’Arte. N.S., 1, 1981/2, pp. 5–15. ‘Scultura e architettura, ovvero alcuni aspetti del decoro scolpito negli edifici costantinopolitani del V–VI secolo’, in XVI. Internationaler Byzantinistenkongress, Wien, 4.–9. Oktober 1981. Akten, II/4 = Jahrbuch der österreichischen Byzantinistik, 32/4, 1982, pp. 419–428.

What we have outlined above are just a few of the more important areas of Claudia Barsanti’s research, but one need merely look at the titles of her more than 130 publications – to which others currently in press have yet to be added – to understand just how wide-reaching and profound were the scientific interests that led her to investigate not only Byzantium but also Venice, the Adriatic coasts, and even the Marche in the Middle Ages. Her versatility as a scholar and researcher is also evident from the many encyclopaedic entries that she wrote on a wide variety of topics as a result of her long editorial activity for the Enciclopedia dell’arte medievale.

‘Una nota sulle sculture del tempo di Giacinto nella chiesa della Dormizione (Koimesis) a Iznik-Nicea’, in Storia dell’Arte, 46, 1982, pp. 201–208. ‘Ritrovamenti a Jesi’, in Bollettino d’Arte. S. VI, 18, 1983, pp. 109–116. ‘Testimonianze bizantine in Ancona: le spoglie paleocristiane del San Ciriaco’, in Atti del VI Congresso Nazionale di Archeologia Cristiana, Pesaro-Ancona, 19–23 settembre 1983, Florence, 1986, pp. 387–404.

Her life of scholarship, research and travel was led alongside intense teaching activity at the university level in various places in Italy. This teaching ranged from Macerata to Udine, from the Sapienza in Rome to Aquila, to the University of Rome Tor Vergata, where she was Associate Professor until her retirement in 2011.

‘Una proposta d’identificazione per il committente dell’iconostasi della chiesa nord di Sebaste di Frigia’, in G. Vikan (ed.), Abstracts of Short Papers. The 17th International Byzantine Congress, Dumbarton Oaks/ Georgetown University, Washington, D.C., August 3–8, 1986, Baltimore, 1986, pp. 28–29.

Her immense enthusiasm and unquenchable scientific curiosity were infectious, as is attested by those colleagues and students who were closest to her and with whom she often worked on site and had as collaborators in the publications coming out of this work. Lasting testimony to this are the contributions made to the publication of the proceedings of the Conference held in her honour at Rome on 20 October 2011, which were aptly entitled Con lo sguardo a Levante.

‘Un capitello mediobizantino conservato nel museo di Nicopolis’, in E. Chrysos (ed.), Νικόπολις 1. Πρακτικά του πρώτου Διεθνούς Συμποσίου για τη Νικόπολη, 23–29 Σεπτεμβρίου 1984 – Nicopolis 1. Proceedings of the first International Symposium on Nicopolis, 23–29 September 1984, Preveza 1987, pp. 349–360 and 560–562. (ed. with A. Guiglia Guidobaldi, A. Iacobini), Gruppo nazionale di coordinamento C.N.R. “Storia dell’arte e della cultura artistica bizantina”. Atti della Giornata di Studio, Roma, 4 dicembre 1986, Rome, 1988 (Milion. Studi e ricerche d’arte bizantina, 1).

The Thematic Session of the International Congress of Byzantine Studies held in Belgrade in 2016 that was dedicated to Byzantine sculpture – Studying Byzantine Sculpture in the 21st Century: New Perspectives and Approaches – was in fact organised and prepared by Claudia Barsanti and Alessandra Guiglia. Already during the Congress, and now here in the publication of the proceedings, the topics presented range from Late Antique production to that of the post-Justinianic period and touch upon different categories of materials, problems and geographical regions ranging from Italy to Greece, from the eastern Mediterranean to Anatolia and the Black Sea. The Thematic Session witnessed the participation of many of the foremost scholars working on Byzantine sculpture today, and included many young and promising scholars who give cause to hope for the future increase and consolidation of this study of one of the more important aspects of Byzantine artistic culture.

‘Scultura anatolica di epoca medio bizantina’, in C. Barsanti, A. Guiglia Guidobaldi, A. Iacobini (eds.), Gruppo nazionale di coordinamento C.N.R. “Storia dell’arte e della cultura artistica bizantina”. Atti della Giornata di Studio, Roma, 4 dicembre 1986, Rome, 1988 (Milion. Studi e ricerche d’arte bizantina, 1), pp. 275–306. ‘Siria e Palestina, Egitto e Africa Settentrionale’, in L’arte bizantina nell’età di Giustiniano = Art e Dossier, 23, 1988, pp. 25–47. ‘Appunti per una ricerca sugli stucchi di ambito siciliano e calabrese in epoca normanna’, in Atti del congresso xii

Claudia Barsanti (1941–2017) internazionale su S. Nilo di Rossano, 28 settembre–1 ottobre 1986, Rossano/Grottaferrata 1989, pp. 351–364.

(with A. Guiglia Guidobaldi) ‘Gli elementi della recinzione liturgica ed altri frammenti minori nell’ambito della produzione scultorea proto bizantina’, in F. Guidobaldi, C. Barsanti, A. Guiglia Guidobaldi (eds.), San Clemente. La scultura del VI secolo, Rome, 1992 (San Clemente Miscellany, IV/2), pp. 67–270.

‘L’esportazione di marmi dal Proconneso nelle regioni pontiche durante il IV–VI secolo’, in Rivista dell’Istituto nazionale di archeologia e storia dell’arte. S. III, 12, 1989, pp. 91–220.

‘L’importazione di manufatti marmorei degli opifici del Proconneso nelle città del Mar Nero durante il IV– VI secolo’, in V. T. Gyuzelev (ed.), Bulgaria Pontica [Medii Aevi] III. Nessebre, 27–31 mai 1985 – България Понтика III. Несебър, 27–31 май 1985 г., Sofia, 1992, pp. 197–225.

(ed. with F. de’ Maffei, A. Guiglia Guidobaldi), Costantinopoli e l’arte delle province orientali, Rome, 1990 (Milion. Studi e ricerche d’arte bizantina, 2). ‘Istanbul: un inconsueto capitello ionico ad imposta di età giustinianea’, in F. de’ Maffei, C. Barsanti, A. Guiglia Guidobaldi (eds.), Costantinopoli e l’arte delle province orientali, Rome, 1990 (Milion. Studi e ricerche d’arte bizantina, 2), pp. 423–428.

‘Recensione: C. Jolivet-Levy, Les églises byzantines de Cappadoce. Le programme iconographique de l’abside et de ses abords, Paris, 1991’, in Arte medievale. S. III, VI,2, 1992, pp. 173–175.

‘Madonna con Bambino’, in G. Morello (ed.), Splendori di Bisanzio. Testimonianze e riflessi d’arte e cultura bizantina nelle chiese d’Italia. Catalogo della mostra, Ravenna 1990, Milan, 1990, pp. 286–287 (no 116).

‘Riflessi bizantini nell’architettura monastica delle Marche’, in E. Simi Varanelli (ed.), Le abbazie delle Marche. Storia e arte. Atti del convegno internationale, Macerata, 3–5 April 1990, Rome, 1992, pp. 377–396.

‘Madonna in trono con Bambino benedicente’, in G. Morello (ed.), Splendori di Bisanzio. Testimonianze e riflessi d’arte e cultura bizantina nelle chiese d’Italia. Catalogo della mostra, Ravenna 1990, Milan, 1990, pp. 282–283 (no 114).

(ed. with F. Guidobaldi, A. Guiglia Guidobaldi), San Clemente. La scultura del VI secolo, Rome, 1992 (San Clemente Miscellany, IV/2). ‘Calice, area bizantina’, in Enciclopedia dell’arte medievale, IV, Rome, 1993, pp. 80–82.

‘Note archeologiche su Bisanzio romana’, in F. de’ Maffei, C. Barsanti, A. Guiglia Guidobaldi (eds.), Costantinopoli e l’arte delle province orientali, Rome, 1990 (Milion. Studi e ricerche d’arte bizantina, 2), pp. 11–72.

‘Candelabro, area bizantina’, in Enciclopedia dell’arte medievale, IV, Rome, 1993, pp. 129–133. ‘Capitello, area bizantina e area islamica’, in Enciclopedia dell’arte medievale, IV, Rome, 1993, pp. 200–211.

‘Recension : V. T. Gyuzelev (ed.), Bulgaria Pontica [Medii Aevi] II. Nessebre, 26–30 mai 1982 – България Понтика II. Несебър, 26–30 май 1982 г., Sofia, 1988’, in Arte medievale. S. II, 4,1, 1990, pp. 195–196.

‘Castello, Armenia’, in Enciclopedia dell’arte medievale, IV, Rome, 1993, pp. 434–437. ‘La scultura paleocristiana e bizantina’, in P. Zampetti (ed.), Scultura nelle Marche, Florence, 1993, pp. 59–82.

‘Tunisia: indagine preliminare sulla diffusione dei manufatti di marmo proconnesio in epoca paleo bizantina’, in F. de’ Maffei, C. Barsanti, A. Guiglia Guidobaldi (eds.), Costantinopoli e l’arte delle province orientali, Rome, 1990 (Milion. Studi e ricerche d’arte bizantina, 2), pp. 429–436.

‘Capitello bizantino’, in Enciclopedia dell’arte antica classica e orientale. Secondo supplemento 1971–1994, I, Rome, 1994, pp. 867–873.

‘Anatolia’, in Enciclopedia dell’arte medievale, I, Rome, 1991, pp. 550–566.

‘Costantinopoli’, in Enciclopedia dell’arte antica classica e orientale. Secondo supplemento 1971–1994, II, Rome, 1994, pp. 318–323.

‘Ancona’, in Enciclopedia dell’arte medievale, I, Rome, 1991, pp. 570–578.

‘Costantino VII Porfirogenito’, in Enciclopedia dell’arte medievale, V, Rome, 1994, pp. 380–381.

‘Ankara’, in Enciclopedia dell’arte medievale, II, Rome, 1991, pp. 33–35.

‘Costantinopoli, scultura’, in Enciclopedia dell’arte medievale, V, Rome, 1994, pp. 401–406.

‘Antonio di Novgorod’, in Enciclopedia dell’arte medievale, II, Rome, 1991, pp. 136–137.

‘Costantinopoli, arti suntuarie’, in Enciclopedia dell’arte medievale, V, Rome, 1994, pp. 428–436.

‘Costantinopoli: testimonianze archeologiche di età costantiniana’, in G. Bonamente, F. Fusco (eds.), Costantino il Grande dall’Antichità all’Umanesimo. Colloquio sul Cristianesimo nel mondo antico, Macerata, 18–20 dicembre 1990, I, Macerata, 1992 (Università degli studi di Macerata. Facoltà di lettere e filosofia, 67; Atti, 21), pp. 115–150.

‘Edessa’, in Enciclopedia dell’arte medievale, V, Rome, 1994, pp. 762–763. ‘Efeso’, in Enciclopedia dell’arte medievale, V, Rome, 1994, pp. 770–771. ‘I mosaici di Antiochia: riflessioni sulla documentazione archeologica superstite’, in R. Farioli Campanati (ed.), xiii

Alessandra Guiglia, Andrea Paribeni & Silvia Pedone Atti del 1o Colloquio dell’Associazione italiana per lo studio e la conservazione del mosaico. Ravenna, 29 aprile–3 maggio 1993, Ravenna, 1994, pp. 579–607.

(with A. Guiglia Guidobaldi, J.-P. Sodini) ‘Scultura architettonica in Oriente nei secoli VI–VII’, in P. Pergola (ed.), Seminari di archeologia cristiana (archeologia e cultura della Tarda Antichità e dell’Alto Medioevo). Resoconto delle sedute dell’A.A. 1994–1995 = Rivista di archeologia cristiana, LXX,1–2, 1996, pp. 496–498.

(with A. Guiglia Guidobaldi) ‘Introduction pour un catalogue des plaques et des caissons de fenêtres de Sainte-Sophie à Constantinople’, in C. Mango, G. Dagron (eds.), Constantinople and its Hinterland. Papers from the Twenty-Seventh Spring Symposium of Byzantine Studies, Oxford, April 1993, Aldershot/ Brookfield (VT), 1994 (Society for the Promotion of Byzantine Studies Publications, 3), pp. 67–68.

Guida di Macerata (Comune di Macerata), Macerata, 1997. ‘Il Genio di Salvator Rosa’, in R. Bigliardi Parlapiano (eds.), La Biblioteca Planettiana del Comune di Jesi, Florence, 1997, p. 86. ‘Vedute di Stefano della Bella’, in R. Bigliardi Parlapiano (eds.), La Biblioteca Planettiana del Comune di Jesi, Florence, 1997, pp. 88–89.

‘Mosaici pavimentali paleocristiani in Ancona: alcune riflessioni’, in I Bragnatine, F. Guidobaldi (eds.), Atti del II Colloquio dell’Associazione Italiana per lo Studio e la Conservazione del Mosaico, Roma 5–7 dicembre 1994, Bordighera, 1995, pp. 181–200.

‘Il libello polemico di Giovanbattista Piranesi’, in R. Bigliardi Parlapiano (eds.), La Biblioteca Planettiana del Comune di Jesi, Florence, 1997, p. 106.

‘Panorama storico artistico di Nicea’, in S. Leanza (ed.), Il Concilio Niceno II (787) e il Culto delle immagini. Convegno di studi per il XII centenario del Concilio Niceno II, Messina, settembre 1987, Messina, 1994, pp. 77–107.

‘Il monumento a Pio VI di Andrea Vici a Treia’, in R. Bigliardi Parlapiano (eds.), La Biblioteca Planettiana del Comune di Jesi, Florence, 1997, p. 114. ‘Mistrà’, in Enciclopedia dell’arte medievale, VIII, Rome, 1997, pp. 350–352.

‘Il Foro di Teodosio I a Costantinopoli’, in A. Iacobini, E. Zanini (eds.), Arte profana e arte sacra a Bisanzio, Rome, 1995 (Milion. Studi e ricerche d’arte bizantina, 3), pp. 9–50.

‘Oreficeria bizantina’, in Enciclopedia dell’arte medievale, VIII, Rome, 1997, pp. 860–863. ‘Un mosaico dagli scavi settecenteschi di Falerone (Ascoli Piceno)’, in R. M. Carra Bonacasa, F. Guidobaldi (eds.), Atti del IV Colloquio dell’Associazione italiana per lo studio e la conservazione del mosaico, Palermo, 9–13 dicembre 1996, Ravenna, 1997, pp. 869–882.

‘Alcune riflessioni sulla diffusione dei materiali di marmo proconnesio in Italia e in Tunisia’, in Akten des XII internationalen Kongresses für christliche Archäologie, Bonn, 22–28 September 1991, Vatican City/Münster, 1995 (Studi di Antichità cristiana, 52; Jahrbuch für Antike und Christentum. Ergänzungsband, 20), pp. 515–523.

‘Il mosaico della pantera scoperto nell’anno 1777 durante gli scavi di Venceslao Pezzolli a Falerone’, in K. E. Werner, Die Sammlung antiker Mosaiken in der Vatikanischen Museen, Vatican City, 1998, pp. 83–85.

‘Materiali bizantini poco noti o inediti della Bitinia’, in XLII Corso di cultura sull’arte ravennate e bizantina. Seminario internazionale sul tema Ricerche di archeologia cristiana e bizantina, Ravenna, 14–19 maggio 1995, Ravenna, 1995, pp. 47–69.

(with J.-P. Sodini, A. Guiglia Guidobaldi) ‘La sculpture architecturale en marbre au VIe siècle à Constantinople et dans les régions sous influence constantinopolitaine’, in N. Cambi, E. Marin (eds.), Radovi XIII međunarodnog kongresa za starokršćansku arheoligiju, Split-Poreč, 25.9.–1.10.1994 – Acta XIII congressus internationalis archaelogiae christianae, Split-Poreč, 25.9.–1.10.1994, II, Vatican City/Split, 1998 (Studi di Antichità cristiana, 54/2; Vjesnik za arheologiju i historiju dalmatinsku. Suplement, 88), pp. 301–376.

(ed. with M. della Valle, A. Guiglia Guidobaldi, A. Iacobini, C. Pantanella, A. Paribeni), Bisanzio e l’Occidente: arte, archeologia, storia. Studi in onore di Fernanda de’ Maffei, Rome, 1996. ‘Imperatore bizantino’, in Enciclopedia medievale, VII, Rome, 1996, pp. 346–350.

dell’arte

‘Un inedito pluteo costantinopolitano a Jesi’, in F. Guidobaldi (ed.), Domum tuam dilexi. Miscellanea in onore di Aldo Nestori, Vatican City, 1998 (Studi di Antichità cristiana, 53), pp. 23–48.

‘Il pavimento medievale del Duomo di Osimo’, in F. Guidobaldi, A. Guiglia Guidobaldi (eds.), Atti del III Colloquio dell’Associazione italiana per lo studio e la conservazione del mosaico, Bordighera, 6–10 dicembre 1995, Bordighera, 1996, pp. 445–456.

‘Scultura, Bisanzio’, in Enciclopedia dell’arte medievale, X, Rome, 1999, pp. 594–602.

(with A. Guiglia Guidobaldi) ‘Premessa ad un catalogo della scultura della Santa Sofia di Costantinopoli’, in C. Barsanti, M. della Valle, A. Guiglia Guidobaldi, A. Iacobini, C. Pantanella, A. Paribeni (eds.), Bisanzio e l’Occidente: arte, archeologia, storia. Studi in onore di Fernanda de’ Maffei, Rome, 1996, pp. 79–104.

‘Santa Maria di Chiaravalle della Castagnola: un’abbazia cistercense tra leggenda e realtà’, in A. Cadei, M. Righetti Tosti-Croce, A. Segagni Malacart, A. Tomei (eds.), Arte d’Occidente temi e metodi. Studi in onore di Angiola Maria Romanini, I, Rome, 1999, pp. 391–404. xiv

Claudia Barsanti (1941–2017) ‘Una breve nota sui plutei di Siponto, Monte Sant’Angelo e Benevento’, in M. Mazzei (ed.), Siponto antica, Foggia, 1999, pp. 224–229.

‘Aspetti e problemi della scultura di età giustinianea: le lastre in opera nelle gallerie della Santa Sofia di Costantinopoli’, in T. Creazzo, G. Strano (eds.), Atti del VI Congresso nazionale dell’Associazione italiana di studi bizantini, Catania-Messina, 2–5 ottobre 2000 = Siculorum gymnasium. N.S., 57, 2004, pp. 31–50.

‘Un panorama di Costantinopoli dal ‘Liber insularum archipelagi’ di Cristoforo Buondelmonti’, in A. Iacobini, M. della Valle (eds.), L’arte di Bisanzio e l’Italia al tempo dei Paleologi 1261–1453, Rome, 1999 (Milion. Studi e ricerche d’arte bizantina, 5), pp. 35–54.

(with A. Guiglia Guidobaldi) ‘Introduzione’, in A. Guiglia Guidobaldi, C. Barsanti (eds.), Santa Sofia di Costantinopoli. L’arredo marmoreo della Grande Chiesa giustinianea, Vatican City, 2004 (Studi di Antichità cristiana, 60), pp. 3–52.

‘Sculture e arredi liturgici’, in V. Pace (ed.), Tesori dell’arte cristiana in Bulgaria. Catalogo della mostra, Roma, 22 maggio–15 luglio 2000, Sofia 2000, pp. 67–77.

‘I plutei degli intercolumni delle gallerie. Le cimase, Le transenne’, in A. Guiglia Guidobaldi, C. Barsanti (eds.), Santa Sofia di Costantinopoli. L’arredo marmoreo della Grande Chiesa giustinianea, Vatican City, 2004 (Studi di Antichità cristiana, 60), pp. 315–529.

‘Costantinopoli e l’Egeo nei primi decenni del XV secolo: la testimonianza di Cristoforo Buondelmonti’, in Rivista dell’Istituto nazionale di archeologia e storia dell’arte. S. III, 56, 2001, pp. 83–253. ‘Decorazione musiva in esterno: il caso di San Benedetto di Conversano’, in A. Paribeni (ed.), Atti del VII Colloquio dell’Associazione italiana per lo studio e la conservazione del mosaico : Pompei, 22–25 marzo 2000, Ravenna, 2001, pp. 167–182.

(ed. with A. Guiglia Guidobaldi), Santa Sofia di Costantinopoli. L’arredo marmoreo della Grande Chiesa giustinianea, Vatican City, 2004 (Studi di Antichità cristiana, 60). ‘Recensione: I. Favaretto, E. Vio, S. Minguzzi, M. Da Villa Urbani (eds.), Marmi della Basilica di San Marco. Capitelli, plutei, rivestimenti, arredi, Milano, 2000’, in Journal für Kunstgeschichte, 9/1, 2005, pp. 77–82.

‘Capitelli di manifattura costantinopolitana a Roma’, in F. Guidobaldi, A. Guiglia Guidobaldi (eds.), Ecclesiae Urbis. Atti del Congresso internazionale di studi sulle Chiese di Roma, IV–X secolo, Roma, 4–10 settembre 2000, Vatican City, 2002 (Studi di Antichità cristiana, 59), pp. 1443–1478.

‘Sculture mediobizantine di Monemvasia’, in S. Pasi (ed.), Studi in memoria di Patrizia Angiolini Martinelli, Bologna, 2005 (Studi e Scavi. N.S., 10), pp. 49–73.

‘Il Duomo di Osimo’, in Il Museo Diocesano di Osimo, Ancona, 2002.

(with S. Pedone) ‘Una nota sulla scultura ad incrostazione e il templon della Panaghia Episcopi di Santorini’, in F. Baratte, V. Déroche, C. Jolivet-Lévy, B. Pitarakis (eds.), Mélanges Jean-Pierre Sodini, Paris, 2005 (Travaux et mémoires, 15), pp. 407–425.

‘Una nota sulla diffusione della scultura a incrostazione nelle Marche medievali’, in M. Canti Polichetti (ed.), Il duomo di San Ciriaco, Ancona, 2002, pp. 284–293. ‘Venezia e Costantinopoli: capitelli di reimpiego nelle dimore lagunari del Duecento’, in E. Concina, G. Trovabene, M. Agazzi (eds.), Hadriatica. Attorno a Venezia e al Medioevo tra arti, storia e storiografia. Scritti in onore di Wladimiro Dorigo, Padua, 2002, pp. 59–69.

‘I “Catini d’oro” di Padova: spoglie costantinopolitane di VI secolo’, in G. Trovabene (ed.), Florilegium artium. Scritti in memoria di Renato Polacco, Padua, 2006 (Facoltà di lettere e filosofia dell’Università di Venezia. Miscellanea, 8), pp. 37–48. (with M. della Valle, R. Flaminio, A. Guiglia Guidobaldi, A. Paribeni, A.-B. Yalçın) ‘La collezione delle sculture bizantine nel Museo della Santa Sofia a Istanbul’, in Rivista on line di storia dell’arte, 4, 2006, pp. 1–11.

‘I plutei di Siponto, Monte Sant’Angelo e Benevento’, in E. Russo (ed.), 1983–1993: dieci anni di archeologia cristiana in Italia. Atti del VII congresso nazionale di archeologia cristiana, Cassino, 20–24 settembre 1993, Cassino, 2003 (Pubblicazioni dell’Università degli studi di Cassino. Sezione atti, convegni, miscellanee), pp. 745–760.

(with M. della Valle, R. Flaminio, A. Guiglia Guidobaldi, A. Paribeni, A.-B. Yalçın) ‘Ayasofya Müzesi Bizans Plastik Eserler Koleksiyonu’, in K. Olşen, F. Bayram, A. Özme (eds.), 23. Araştırma Sonuçları Toplantısı, 30 Mayis–3 Haziran 2005, Antalya, I, Ankara, 2006 (T.C. Kültür ve Turizm Bakanlığı, 3053/1; Kültür Varlıkları ve Müzeler Genel Müdürlüğü, 114/1), pp. 319–330.

‘Marmi bizantini nel Palazzo del Gran Maestro dei Cavalieri a Rodi’, in A. Iacobini (ed.), Bisanzio, la Grecia e l’Italia. Atti della giornata di studi sulla civiltà artistica bizantina in onore di Mara Bonfioli, Università di Roma ‘La Sapienza’, 22 novembre 2002, Rome, 2003, pp. 17–31.

(with A. Guiglia Guidobaldi, A. Paribeni, A.-B. Yalçın) ‘Ayasofya Müzesi Projesi, 2005: Beyazıt Bölgesi Mermer Buluntuları’, in F. Bayram, B. Koral (eds.), 24. Araştırma Sonuçları Toplantısı, 29 Mayis–2 Haziran 2006, Çanakkale, II, Ankara, 2007 (T.C. Kültür ve Turizm Bakanlığı, 3080/2; Kültür Varlıkları ve Müzeler Genel Müdürlüğü, 120/2), pp. 311–324.

‘The Iznik-Nicaea’s Archaeological Museum: In search of a catalogue’, in I. Akbaygil, H. Inalcık, O. Aslanapa (eds.), Iznik throughout History, Istanbul, 2003, pp. 265–301. xv

Alessandra Guiglia, Andrea Paribeni & Silvia Pedone (with A. Paribeni) ‘Broken Bits of Byzantium: frammenti di un puzzle archeologico nella Costantinopoli di fine Ottocento’, in A. Calzona, R. Campari, M. Mussini (eds.), Immagine e Ideologia. Studi in onore di Arturo Carlo Quintavalle, Milan, 2007, pp. 550–565.

(with A. Guiglia) ‘St. Sophia Museum Project 2006: The marble sculpture of the Middle-Byzantine Period’, in F. Bayram, A. Özme, B. Koral (eds.), 25. Araştırma Sonuçları Toplantısı, 28 Mayis–1 Haziran 2007, Kocaeoli, II, Ankara, 2008 (T.C. Kültür ve Turizm Bakanlığı, 3112/2; Kültür Varlıkları ve Müzeler Genel Müdürlüğü, 125/2), pp. 117–132.

‘In memoria del reliquiario del Sacro Capo dell’apostolo Andrea’, in M. Sodi, A. Antoniutti (eds.), Enea Silvio Piccolomini. Pius Secundus Poeta Laureatus Pontifex Maximus. Atti del convegno internazionale, 29 settembre–1 ottobre 2005, Roma, e altri studi, Roma 2007, Rome, 2007, pp. 319–340.

‘Una nota sulla diffusione della scultura ad incrostazione nelle regioni adriatiche del meridione d’Italia tra XI e XIII secolo’, in Ch. Pénnas, C. Vanderheyde (eds.), La sculpture byzantine VIIe–XIIe siècles, Actes du colloque international organisé par la 2e Éphorie des antiquités byzantines et l’École française d’Athènes, 6–8 septembre 2000, Athens 2008 (Bulletin de correspondance hellénique. Supplément, 49), pp. 515–557.

‘La scultura architettonica di epoca omayyade: tra Bisanzio e la Persia sasanide: i capitelli di Qasr al-Muwaqqar in Giordania’, in A. C. Quintavalle (ed.), Medioevo mediterraneo: l’Occidente, Bisanzio e l’Islam, Atti del Convegno internazionale di studi, Parma, 21–25 settembre 2004, Milano, 2007 (I Convegni di Parma, 7), pp. 436–446.

(with A. Guiglia, S. Pedone) ‘St. Sophia Museum Project 2007: The marble sculpture of St. John Studius (Imrahor Camii)’, in F. Bayram, A. Özme (eds.), 26. Araştırma Sonuçları Toplantısı, 26–30 Mayis 2008, Ankara, III, Ankara, 2009 (T.C. Kültür ve Turizm Bakanlığı, 3172/3; Kültür Varlıkları ve Müzeler Genel Müdürlüğü, 131/3), pp. 311–328.

‘La scultura mediobizantina fra tradizione e innovazione’, in F. Conca, G. Fiaccadori (eds.), Bisanzio nell’età dei Macedoni: forme della produzione letteraria e artistica. VIII Giornata di studi bizantini (Milano, 15–16 marzo 2005), Milano, 2007 (Quaderni di Acme, 87), pp. 5–49.

(with A. Guiglia Guidobaldi) ‘Le porte e gli arredi architettonici in bronzo della Santa Sofia di Costantinopoli’, in A. Iacobini (ed.), Le porte del paradiso. Arte e tecnologia bizantina tra Italia e Mediterraneo. Convegno internazionale di studi, Istituto Svizzero di Roma, 6–7 dic. 2006, Rome, 2009 (Milion. Studi e ricerche d’arte bizantina, 7), pp. 81–123.

‘Le chiese del Grande Palazzo di Costantinopoli’, in A. C. Quintavalle (ed.), Medioevo: la Chiesa e il Palazzo. Atti del Convegno internazionale di studi, Parma, 20– 24 settembre 2005, Milano, 2007 (I Convegni di Parma, 8), pp. 87–100. ‘Rodi descritta e illustrata nel Liber Insularum Archipelagi di Cristoforo Buondelmonti’, in 15 χρόνια έργων αποκατάστασης στη Μεσαιωνική πόλη της Ρόδου – 15 Years of Restoration in the Medieval Town of Rhodes. Πρακτικά Διεθνούς επιστημονικού συνεδρίου, Ρόδος, 14–18 Νοεμβρίου 2001, Athens 2007, pp. 266–277.

‘I mosaici del Grande Palazzo imperiale di Costantinopoli: alcune riflessioni’, in M. C. Lentini (ed.), Mosaici mediterranei, Caltanissetta, 2009, pp. 55–73. (with M. Pilutti Namer) ‘Da Costantinopoli a Venezia: nuove spoglie della chiesa di S. Polieucto, Nota preliminare’, in Nέα Ρώμη, 6, 2009, pp. 133–156.

‘L’isola di Rodi nel Liber insularum archipelagi di Cristoforo Buondelmonti’, in Nέα Ρώμη, 8, 2008, pp. 357–380.

‘Les Cyclades dans le Liber insularum archipelagi de Cristoforo Buondelmonti’, in Πρακτικά Γ’ Διεθνούς Σιφναϊκού συμποσίου, Σίφνος 29 Ιουνίου–2 Ιουλίου 2006. Εις μνήμην Νικόλαου Βερνίκου-Ευγενίδη – Proceedings of the 3rd International Sifnean Symposium, Sifnos 29 June–2 July 2006. In memoriam Nikolaos Vernicos-Eugenides, Athens, 2009, pp. 67–78.

‘Il Medaglione d’oro di Morro d’Alba’, in C. Barsanti, A. Paribeni, S. Pedone (ed.), Rex Theodericus. Il medaglione d’oro di Morro d’Alba, Rome, 2008 (Espera Archeologia, 2), pp. 3–9. ‘Ravenna: gli arredi architettonici e liturgici negli edifici di età teodericiana’, in C. Barsanti, A. Paribeni, S. Pedone (ed.), Rex Theodericus. Il medaglione d’oro di Morro d’Alba, Rome, 2008 (Espera Archeologia, 2), pp. 185–202.

‘L’Antico Bagno (Eski Kaplica) di Bursa: ieri e oggi’, in A. Armati, M. Cerasoli, C. Luciani (eds.), “Alle gentili arti ammaestra”. Studi in onore di Alkistis Proiou, Rome, 2010 (Testi e Studi bizantino-neoellenici, 18), pp. 125–161.

(with P. Pensabene) ‘Reimpiego e importazioni di marmi nell’Adriatico paleocristiano e bizantino’, in G. Cuscito (ed.), La cristianizzazione dell’Adriatico. Atti della XXXVIII Settimana di Studi aquileiesi, 3–5 maggio 2007, Trieste, 2008 (Antichità altoadriatiche, 66), pp. 455–490.

(with A. Guiglia) ‘Late Roman and Early Byzantine Capitals’, in C. Barsanti, A. Guiglia (eds.), The Sculptures of the Ayasofya Müzesi in Istanbul. A Short Guide, Istanbul, 2010, pp. 79–100. (with A. Guiglia Guidobaldi, A. Paribeni) ‘Le officine dell’imperatore: marmora byzantina’, in A. C. Quintavalle (ed.), Medioevo: le officine. Atti del convegno internazionale di studi, Parma, 22–27

(ed. with A. Paribeni, S. Pedone), Rex Theodericus. Il medaglione d’oro di Morro d’Alba, Rome, 2008 (Espera Archeologia, 2) xvi

Claudia Barsanti (1941–2017) settembre 2009, Milano, 2010 (I convegni di Parma, 12), pp. 118–151.

(with A. Guiglia) ‘Il progetto di ricerca sui marmi della Santa Sofia’, in A. Acconcia Longo, G. Cavallo, A. Guiglia, A. Iacobini (eds.), La sapienza bizantina. Un secolo di ricerche sulla civiltà di Bisanzio all’Università di Roma. Giornata di studi, Sapienza Università di Roma, 10 ottobre 2008, Rome, 2012 (Milion. Studi e ricerche d’arte bizantina, 8), Rome, 2012, pp. 55–78.

(with A. Guiglia, A. Paribeni) ‘Saint Sophia Museum Project 2008: the Byzantine Marble Capitals in the Ayasofya Müzesi, Istanbul’, in H. Dönmez, C. Keskin (eds.), 27. Araştırma Sonuçları Toplantısı, 25–29 Mayis 2009, Denizli, I, Ankara, 2010 (T.C. Kültür ve Turizm Bakanlığı, 3250/1; Kültür Varlıkları ve Müzeler Genel Müdürlüğü, 142/1), pp. 413–432.

‘Restes de la reine des villes/Broken Bits of Byzantium. Introduction à l’édition critique. Première partie’, in Eurasian Studies, 10/1–2, 2012, pp. 127–151 and i–iv.

(with A. Guiglia) ‘The Byzantine Sculptures of the Ayasofya Müzesi in Istanbul: Ten Years of Researches (1999–2009)’, in Ayasofya Müzesi Yıllığı – Annual of Hagia Sophia Museum, 13, 2010, pp. 135–154.

‘The fate of the Antioch mosaic pavements: Some reflections’, in Journal of Mosaic Research, 5, 2012, pp. 25–42. ‘Costantinopoli’, in Costantino I. Enciclopedia costantiniana sulla figura e l’immagine dell’imperatore del cosiddetto Editto di Milano 313–2013, I, Rome, 2013 (Orsa maggiore), pp. 471–491.

(ed. with A. Guiglia), The Sculptures of the Ayasofya Müzesi in Istanbul. A Short Guide, Istanbul, 2010. ‘The sculptures of the Mosaic Museum and the Arasta Bazaar’, in C. Barsanti, A. Guiglia (eds.), The Sculptures of the Ayasofya Müzesi in Istanbul. A Short Guide, Istanbul, 2010, pp. 119–126.

(with M. Beccaloni) ‘La chiesa di Santa Maria Liberatrice: dal Foro Romano al Testaccio’, in D. Gavrilovich, C. Occhipinti, D. Orecchia, P. Parenti (eds.), Miti antichi e moderni, Rome, 2013 (Arti dello spettacolo/Performing Arts; Miscellanea di studi interdisciplinari), pp. 253–269 and 484–486.

(with A. Guiglia, R. Flaminio) ‘Saint Sophia Museum Project 2009: The collection of Byzantine marble slabs in the Ayasofya Müzesi, Istanbul’, in A. N. Toy, C. Keskin (eds.), 28. Araştırma Sonuçları Toplantısı, 24–28 Mayis 2010, Istanbul, II, Ankara, 2011 (T.C. Kültür ve Turizm Bakanlığı, 3266/2; Kültür Varlıkları ve Müzeler Genel Müdürlüğü, 148/2), pp. 361–376.

‘Le cisterne bizantine di Istanbul: nuovi dati sulla scultura dal IV al VI secolo’, in O. Brandt, S. Cresci, J. López Quiroga, C. Pappalardo (eds.), Acta XV Congressus internationalis archaeologiae Christianae, Toleti (8– 12.9.2008). Episcopus, civitas, territorium, Vatican City, 2013 (Studi di Antichità cristiana, 65), pp. 1481–1496.

‘The marble floor of St. John Studius in Constantinople: A neglected masterpiece’, in M. Şahin (ed.), XI. Uluslararasi Antik Mozaik Sempozyumu, 16–20 Ekim 2009 Bursa, Türkiye. Türkiye Mozaikleri ve Antik Dönemden Ortaçağ Dünyasina Diğer Mozaiklerle Paralel Gelişimi. Mozaiklerin Başlangicindan Geç Bizans Çağina Kadar Ikonografi, Stil ve Teknik Üzerine Sorular – 11th International Colloquium on Ancient Mosaics, October 16th–20th, 2009, Bursa, Turkey. Mosaics of Turkey and Parallel Developments in the Rest of the Ancient and Medieval World. Questions of Iconography, Style and Technique from the Beginnings of Mosaic until the Late Byzantine Era, Istanbul, 2011 (Uludag University Mosaic Research Center Series, 1), pp. 87–98.

‘La sculpture byzantine’, in M. Poulain, F. Queyrel, G. Paquot (eds.), Éclats d’antiques. Sculpture et photographies. Gustave Mendel à Constantinople. Exposition, Paris, Institut National d’Histoire de l’Art, 24 avril–20 juillet 2013, Paris, 2013 (Armand Colin. Recherches), pp. 81–93. ‘Una ricerca sulle sculture in opera nelle cisterne bizantine di Istanbul: la Ipek Bodrum Sarnici (la cisterna n. 10)’, in A. Rigo, A. Babuin, M. Trizio (eds.), Vie per Bisanzio. VII Congresso Nazionale dell’Associazione Italiana di Studi Bizantini, Venezia, 25–28 novembre 2009, Bari, 2013 (Due punti, 25), pp. 477–508.

‘Un inedito disegno delle rovine del complesso costantinopolitano del Boukoleon’, in W. Angelelli, F. Pomarici (eds.), Forme e storia. Scritti di arte medievale e moderna per Francesco Gandolfo, Rome, 2011, pp. 45–58.

(with A. Guiglia) ‘Alcune riflessioni sulle travi lignee scolpite della Santa Sofia a Costantinopoli e sui restauri dei fratelli Fossati’, in G. Bordi, I. Carlettini, M. L. Fobelli, M. R. Menna, P. Pogliani (eds.), L’officina dello sguardo. Scritti in onore di Maria Andaloro, I, Rome, 2014, pp. 271–284.

‘Un taccuino di disegni costantinopolitani al Victoria & Albert Museum di Londra’, in O. Brandt, P. Pergola (eds.), Marmoribus vestita. Miscellanea in onore di Federico Guidobaldi, Vatican City, 2011 (Studi di Antichità cristiana, 62), pp. 135–157.

‘Marmi costantinopolitani a Cipro’, in M. Gianandrea, F. Gangemi, C. Costantini (eds.), Il potere dell’arte nel Medioevo. Studi in onore di Mario D’Onofrio, Rome, 2014 (Saggi di storia dell’arte, 40), pp. 195–206.

(with A. Guiglia) ‘Aspetti meno noti della decorazione della Santa Sofia di Costantinopoli’, in Arkeoloji ve Sanat. Journal of Archaeology & Art, 139, 2012, pp. 191–201.

‘Una rara scultura costantinopolitana di epoca paleologa a Murano’, in Venezia Arti, 27/24, 2014, pp. 10–25.

xvii

Alessandra Guiglia, Andrea Paribeni & Silvia Pedone ‘Episodi di reimpiego nel San Ciriaco di Ancona’, in Temporis signa. Archeologia della Tarda Antichità e del Medioevo, 10, 2015, pp. 85–108. (with A. Guiglia) ‘Il ruolo dei marmi bizantini nella produzione scultorea della Sardegna tardoantica e paleocristiana’, in R. Martorelli, A. Piras, P. G. Spanu (eds.), Isole e terraferma nel primo cristianesimo. Identità locale ed interscambi culturali, religiosi e produttivi. Atti del XI congresso nazionale di archeologia cristiana, Cagliari, Dipartimento di Storia, Beni Culturali e Territorio – Sant’Antioco, 23– 27 settembre 2014, Cagliari, 2015 (Studi e ricerche di cultura religiosa. N.S., 8), pp. 349–368. (with R. Flaminio, A. Guiglia) La diocesi di Roma. La III Regione ecclesiastica, Spoleto, 2015 (Corpus della scultura altomedievale, VII/7). (with A. Paribeni) ‘La diffusione del marmo proconnesio nelle Marche in età classica e paleocristiana: il ruolo del porto di Ancona’, in Hortus Artium Medievalium, 22, 2016, pp. 200–214. ‘Scultura dipinta a Bisanzio’, in P. A. Andreuccetti, D. Bindani (eds.), Il colore nel Medioevo. Arte, Simbolo, Tecnica tra materiali costitutivi e colori aggiunti. Mosaici, intarsi e plastica lapidea, Atti delle Giornate di Studi, Lucca, 24–25–26 ottobre 2013, Lucca, 2016 (Collana di studi sul colore, 5), pp. 61–85.

Italian Lens on Byzantine Art in Anatolia,1960–2000, Istanbul, 2018, pp. 67–82. (with A. Guiglia Guidobaldi) ‘Il Corpus della scultura altomedievale della Fondazione CISAM Centro Italiano di Studi sull’Alto Medioevo di Spoleto: nuove prospettive’, in Hortus Artium Medievalium, 24, 2018, pp. 53–58. (with A. Paribeni) ‘La scultura in funzione architettonica a Costantinopoli tra V e VI secolo: aspetti tecnici, tipologici e stilistici’, in Acta ad archaeologiam et artium historiam pertinentia, 30 – N.S., 16, 2018, pp. 22–72. ‘Restes de la reine des villes/Broken Bits of Byzantium. Introduction à l’édition critique. Deuxième partie. Con una nota introduttiva di Andrea Paribeni’, in S. Pedone, A. Paribeni (eds.), “Di Bisanzio dirai ciò che è passato, che passa e che sarà”. Scritti in onore di Alessandra Guiglia, Rome, 2018, pp. 235–249. (with A. Guiglia) ‘Spolia in Constantinople’s Hagia Sophia from the Age of Justinian to the Ottoman Period: The phenomenon of multilayered reuse’, in I. Jevtić, S. Yalman (eds.), Spolia Reincarnated. Afterlives of Objects, Materials and Spaces in Anatolia from Antiquity to the Ottoman Era, Istanbul, 2018, pp. 97–124.

‘Un capitello ionico a imposta di epoca protobizantina reimpiegato nella Ulu Camii di Manisa’, in M. Guidetti, S. Mondini (eds.), “A mari usque ad mare”. Cultura visuale e materiale dall’Adriatico all’India. Scritti in memoria di Gianclaudio Macchiarella, Venice, 2016 (Eurasiatica. Quaderni di studi su Balcani, Anatolia, Iran, Caucaso e Asia Centrale, 4), pp. 49–61.

‘A margine delle colonne-clava dell’arco del Foro di Teodosio I a Costantinopoli’, in S. Cosentino, M. E. Pomero, G. Vespignani (eds.), Dialoghi con Bisanzio. Spazi di discussione, percorsi di ricerca. Atti dell’VIII Congresso dell’Associazione Italiana di Studi Bizantini, Ravenna, 22–25 settembre 2015, Spoleto, 2019 (Quaderni della Rivista di Bizantinistica, 20), pp. 45-61.

‘Capitelli corinzi del tipo cosiddetto “a medaglione” con inserti decorativi’, in  P. Pensabene, M. Milella, F. Caprioli (eds.), Decor. Decorazione e Architettura nel Mondo Romano. Atti del Convegno Internazionale, Roma, 21–24 maggio 2014, Rome, 2017 (Thiasos monografie, 9), pp. 391–401.

Forthcoming: (with A. Paribeni) ‘Sculture bizantine nelle Marche meridionali: provenienza, reimpiego, nuove funzionalità’, in P. Pistilli, F. Gangemi (eds.), Il Piceno prima di Fiastra. Topografia, architettura ed arte, Giornate di studi sul territorio piceno nell’età di mezzo, Poggio San Costanzo, Macerata, 14–15 maggio 2010.

‘Il panorama di Cristoforo Buondelmonti e le chiese latine di Costantinopoli’, in C. Monge, S. Pedone (eds.), Domenicani a Costantinopoli prima e dopo l’impero ottomano. Storie, immagini e documenti d’archivio, Florence, 2017 (Biblioteca di memorie domenicane, 17), pp. 51–67. ‘Un nuovo capitello polilobato costantinopolitano’, in C. L. Schiavi, S. Caldano, F. Gemelli (eds.), La lezione gentile. Scritti di storia dell’arte per Anna Maria Segagni Malacart, Milano, 2017 (Storia dell’architettura e della città, 8), pp. 47–58. (with A. Guiglia) ‘Sapienza İstanbul’da: Ayasofya projesi / Sapienza in Istanbul: The Ayasofya project’, in L. Bevilacqua, G. Gasbarri (eds.), Yitik İmparatorluğu Resmetmek: İtalyan Merceğinden Anadolu’daki Bizans Sanatı, 1960–2000 – Picturing a Lost Empire: An xviii

Alessandra Guiglia Sapienza Università di Roma (Italy) Andrea Paribeni Università degli Studi di Urbino Carlo Bo (Italy) Silvia Pedone Accademia Nazionale dei Lincei (Italy)

Introduction Since its very first conference in 1924, the International Congress of Byzantine Studies (ICBS) has been truly multidisciplinary, and has distinguished itself from other international scientific events of the same type by giving to Archaeology and Art History a place similar to those given to History and Philology.1 It is not always recognised, but the first congresses were, together with the International Congresses of Christian Archaeology (since 1894), among the first world-class scientific meetings to focus on the Late Roman and Early Byzantine world as a whole, geographically and temporally, at a time when most Ancient History textbooks stopped abruptly at Constantine I, Theodosius I or, at best, Romulus Augustulus, without any actual presentation of the civilisation of that time, even from the Western point of view.2

Roman and Early Byzantine period. A call for papers was therefore sent to the participants of six thematic sessions of free communications: • Byzantine Archaeology, chaired by Dominic Moreau and Mihailo Milinković; • Late Antique and Early Byzantine Art, chaired by Alicia Walker and Ljubomir Milanović; • Late Antique and Early Byzantine Architecture, chaired by Elizabeta Dimitrova and Skënder Muçaj; • Material Culture, chaired by Orsolya HeinrichTamaska and Ivana Popović; • Studying Byzantine Sculpture in the 21st Century: New Perspectives and Approaches, chaired by Claudia Barsanti and Alessandra Guiglia; • Varia Archaeologica, chaired by Vesna Bikić and Özgü Çömezoğlu Uzbek;

With the development of Late Roman and Early Byzantine studies throughout the twentieth century, the participation in the ICBS of scholars focusing on the first centuries of the medieval phase of the Roman Empire has increased with each new meeting. The 22nd and 23rd conferences, held in Sofia (2011) and Belgrade (2016), which were the first to be organised in Bulgaria and Serbia in more than eighty years (respectively 1934 and 1927 for the previous ones), have without doubt offered the largest number of papers on the transition to the Middle Ages, especially in the Balkans, and in the field of Archaeology and Art History.3 Moreover, the number of participants has steadily increased over the years, so that a full publication of all the papers is no longer possible.

and one round table: • The Episcopal Palace in Early Byzantium: Historical Development, Architectural Typologies, Domestic Spaces, chaired by Isabella Baldini. Given the number of sessions, the heterogeneous nature of the papers and the international character of the authors, it seemed important to bring together a plural editorial board, respecting as much as possible the chairs of the sessions (obviously, on a volunteer basis) and including representatives of all the languages chosen for the publication, which correspond to the four official Western European languages of the Congress: English, French, German and Italian. Therefore, the reader will find a summary of all the papers in each of these languages, although no author provided a main text in German.

For all these reasons, but also to report on the richness and originality of communications in the field of the Archaeology and Art History of Late Antiquity presented within the ICBS, it seemed worthwhile after the Congress of Belgrade (entitled Byzantium – A World of Changes) to gather several papers from different sessions, in order to form a consistent volume on architecture, sculpture and landscapes, under a general theme linked to that of the meeting: Archaeology of a World of Changes, viz. the Late

This plural and international Editorial Committee is composed of: • Dominic Moreau (Université de Lille / HALMA-UMR 8164 research centre, France); • Carolyn S. Snively (Gettysburg College, USA); • Alessandra Guiglia (Sapienza Università di Roma, Italy); • Isabella Baldini (Università di Bologna, Italy); • Ljubomir Milanović (Византолошки институт Српске академије наука и уметности / Institute for Byzantine Studies of the Serbian Academy of Sciences and Arts, Serbia); • Ivana Popović (Археолошки институт, Београд / Institute of Archaeology, Belgrade, Serbia); • Nicolas Beaudry (Université du Québec à Rimouski, Canada);

1 See M. Marinescu (ed.), 1925. Compte-rendu du Premier Congrès international des études byzantines, Bucarest, 1924, Bucharest. 2 Obviously, the then edition of The Cambridge Ancient History and The Cambridge Medieval History are not really included in this statement, but they are not just simple textbooks (although it should be noted that the first – first published in 1939 – ends in AD 324 and the second – first published in 1911 – begins with Constantine, the Christianisation of central power being considered as the event marking the passage from one period to another). 3 See I. Iliev (ed.), 2011. Proceedings of the 22nd International Congress of Byzantine Studies, Sofia, 22–27 August 2011, I–III, Sofia; and, online, the three volumes of The Proceedings of the 23rd International Congress of Byzantine Studies: www.byzinst-sasa.rs/eng/archive/47/2017/11/30/ the-proceedings-of-the-23rd-international-congress-of-byzantinestudies.html.

1

Dominic Moreau, Carolyn S. Snively & Alessandra Guiglia publication process, beyond her participation in the Scientific Committee, a wish that was a special honour for us. She unfortunately left us too early, before the editing work had really begun. We sincerely hope that our contribution to Late Roman and Early Byzantine Archaeology and Art History is worthy of the tribute we all want to pay her.

• Orsolya Heinrich-Tamáska (Leibniz-Institut für Geschichte und Kultur des östlichen Europa–GWZO, Germany). Every member of the Editorial Committee was also a member of the Scientific Committee, which was completed by:

Dominic Moreau, Carolyn S. Snively and Alessandra Guiglia (on behalf of the Editorial Committee) 19 August 2019

• Claudia Barsanti (Università degli Studi di Roma ‘Tor Vergata’, Italy); • Pascale Chevalier (Université Clermont Auvergne / ArTeHiS–UMR 6298 research centre, France); • Elizabeta  Dimitrova (Универзитет ‘Св. Кирил и Методиј’ / Ss. Cyril and Methodius University in Skopje, North Macedonia); • Skënder Muçaj (Instituti i Arkeologjisë / Institute of Archaeology, Albania). In order to offer a quality publication, each paper was peerreviewed first by two members of the Editorial/Scientific Committee, then by three anonymous reviewers appointed by the publishing house. The result is particularly interesting, as it brings together older research, mainly through status quaestionis, with new projects and discoveries. The arrangement of the material was not easy, given that it was impossible to find real guidelines for most of them – apart from the papers from the round table on episcopal palaces – but five sections have finally emerged: • • • • •

Archaeology, Landscape and Topography; Episcopal Residences; Architectural Sculpture; Decoration and Small Objects; Restoration and Conservation.

The editing work took longer than expected, in particular because a large number of authors were not writing in their mother tongues. The Editorial Committee wishes to thank them all, together with BAR Publishing, for their patience. We would also like to thank Jane Burkowski, Ruth Fisher and Lisa Eaton for their editing work on the manuscript, which helped us to bring everything up to the standard of the BAR International Series, as well as the Organising Committee of the 23rd ICBS, especially Prof. Ljubomir Maksimović, who has believed in this project from the very beginning. Our thanks are also due: to Thomas Nicq, the scientific imaging photographer of the HALMA-UMR 8164 research centre (Lille, France), for his magnificent work on the images; to the same research centre and the I-SITE ULNE Foundation, through the DANUBIUS Project (https://danubius.huma-num.fr), for financing the publication of the book; and to Ivan Stanić (Музеј науке и технике / Museum of Science and Technology, Serbia) for the beautiful picture on the cover, of the famous Jonah sarcophagus from the Народни музеј Србије / National Museum of Serbia. Finally, we dedicate our book to Claudia Barsanti, because she sincerely believed in it and she had repeatedly expressed her intention to become fully involved in the 2

Part 1 Archaeology, Landscape and Topography

1 The ‘Crypt on N. I. Tur’s Land’ in Early Byzantine Chersonesos: Cherson as a Monument of the Formation of the Church in the City Mikhail V. Fomin Харківська державна академія культури / Kharkiv State Academy of Culture (Ukraine) Chersonesos-Cherson is a Byzantine centre on the border between the Byzantine civilisation and the ‘barbarian world’. The ancient city is located outside the modern city, and regular archaeological excavations have been held for more than a hundred years. One of the most interesting monuments of the period of the transformation of the Late Antique town into the medieval city (the fourth to the early fifth century) is the ‘Crypt on N. I. Tur’s Land. The catacomb was planned as a typical burial construction. The study of this monument, its architecture and wall paintings has shown that it was rebuilt into a theophanic martyrium where church services associated with the memory of the early history of Christianity in the city took place. It became a monument of the activities of St Basil, the first bishop of Chersonesos. Chersonèse/Cherson est un centre byzantin à la frontière de la civilisation byzantine et du « monde barbare ». L’ancienne ville, située en dehors de la zone urbaine moderne, fait l’objet de fouilles archéologiques depuis plus de cent ans. L’un des monuments de l’époque de la transformation de la ville tardo-tardive en ville médiévale (IVe siècle – début du Ve siècle) est la « Crypte sur la terre de N. I. Tur ». Cette catacombe a été conçue comme un édifice funéraire typique. L’étude du monument, de son architecture et de ses peintures murales a démontré qu’il a été reconstruit en martyrium théophanique où se tenaient des services religieux liés à la mémoire des premiers temps du christianisme dans la cité. Il est ainsi devenu un monument dédié aux activités de saint Basile, le premier évêque de Chersonèse. Chersonesos/Cherson gilt als ein byzantinisches Zentrum an der Grenze zwischen Byzantium und der Welt der „Barbaren“. Die antike Stadt befindet sich außerhalb der modernen Bebauung, wo seit über hundert Jahren regelmäßig Ausgrabungen durchgeführt werden. Eines der interessantesten Denkmäler aus der Zeit der Umwandlung der antiken Stadt während des 4. und 5. Jhs. ist die „Krypta auf dem Boden von N. I. Tor“. Die Katakombe wurde als eine typische Grabkonstruktion geplant. Die Untersuchung des Denkmals, seiner Architektur und Wandmalereien zeigte, dass es in ein theophanisches Martyrium umgebaut wurde, wo Gottesdienste zelebriert wurden, die an die frühe Geschichte des Christentums in der Stadt erinnerten. Es wurde zu einer Erinnerungsstätte für das Wirken des heiligen Basilius, des ersten Bischofs von Cherson. Chersoneso – Cherson, è un centro bizantino al confine tra la civiltà bizantina e il «mondo dei barbari». L’antica città si trova al di fuori della città moderna e per questo gli scavi archeologici vi sono stati condotti già per oltre cento anni. Uno dei monumenti più interessanti dell’epoca della trasformazione della città tardo antica in quella medievale (IV – inizio del V secolo) è la «cripta sul terreno di N.I. Tur». La catacomba fu progettata come una tipica costruzione funeraria. Lo studio del monumento, della sua architettura e delle pitture murali ha dimostrato che esso è stato ricostruito come martyrium teofanico, dove si svolgevano le celebrazioni liturgiche legate alla memoria della storia precoce del cristianesimo nella città. Esso divenne un monumento alle attività di San Basilio, il primo vescovo di Chersoneso.

5

Mikhail V. Fomin A series of unique works of art of Late Antiquity and the Early Christian period was discovered at the archaeological site and in the nearby territory. They are comparable to the Roman catacombs and the Dura-Europos wall-paintings. Research into them makes it possible to see the process of formation of the Early Christian artistic tradition and the ideological transformation during the period of the evolution of the Late Antique town into the medieval city. Among these monuments, first place is taken by the Chersonesos crypts with wall paintings and the fragments of frescoes and mosaic floor of the early ‘Basilica of 1935’.2 The monument which is mentioned in the literature as the ‘crypt on N. I. Tur’s land’ stands out among the painted crypts of Chersonesos. Discussion about the monuments has already been held for a long time, and one of the most disputed issues is the dating.

One of the significant problems associated with studying the formation of Early Christian art is the insufficient amount of research on Early Christian centres. Many Late Antique centres are not available for research. This problem has been mentioned in the literature many times.1 As a result, the monuments of the Northern Black Sea region are sites of particular interest. A special place among them is occupied by Chersonesos/Cherson, a local centre on the border between the Byzantine civilization and the ‘barbarian world’. Its uniqueness is underlined by the fact that the site is located outside the modern city, unlike many other Late Antique and Byzantine centres such as Thessaloniki, Athens, Ravenna or Constantinople/ Istanbul. Regular archaeological excavations have already been conducted on the territory of the ancient city for more than a hundred years. There is a unique scientific complex at the site that includes laboratories, storage units and a museum – a rare circumstance for Byzantine monuments around the world. As a result, a great quantity of material, which needs to be studied and interpreted, has been accumulated (fig. 1). 1

The famous M. I. Rostovtsev, who was also a specialist in Late Antiquity, dated the monuments to the second half of the fourth and the early fifth century. The basis for this dating

Romanchuk 2007, pp. 133–145.

2

Fomin 2014, pp. 299–390.

Fig. 1. Location of Chersonesos/Cherson and the ‘crypt on N. I. Tur’s land’.

6

The ‘Crypt on N. I. Tur’s Land’ in Early Byzantine Chersonesos was numismatic material.3 Later, some modern scholars, including P. D. Diatroptov,4 L. G. Khrushkova5 and I. A. Zavadskaya6 supported this conclusion. S. B. Sorochan also placed significant emphasis on the group of crypts, with several publications that were dedicated to this particular catacomb and that refer to its creation before the end of the fourth century.7 A. E. Philippov dates the crypts to the third to fourth century.8 V. M. Zubar paid much attention to the monuments in his works and offered a different interpretation. According to his conclusions, the crypts were painted in the fifth to sixth century. 9

The ‘apse’ architecture has some distinctive features. During the reconstruction, a stone ledge was left in the centre of the apse; above it, a small niche was made (fig. 5). It can be assumed that they were used as an altar for liturgy, as similar early liturgical constructions were widespread. They are known in the cave churches of Cappadocia and the Mountainous Crimea, in the regions of the Danube and the Dniester, and are described by Y. Y. Shevchenko.13 The wall-painting system could indirectly confirm this assumption. The space was outlined by red lines along its perimeter. Above the ledge, two peacocks facing each other hold the semicircle of a garland. Above the garland and peacocks, there are ‘flowers’ consisting of four ‘heartshaped’ petals and ‘green tendrils between them’.14 The same flower appears in the centre of the garland (fig. 5). Similar images can be found on other monuments of Early Christian art, for example, in the decorative painting of Ravenna, frescoes of the Roman catacombs, on Coptic textiles and Early Christian sarcophagi; they can also be found in other crypts at Chersonesos (fig. 4, 5). Y. Matveeva has suggested that these flowers are symbols of Eucharistic bread.15 Thus, the entire painting system, which includes Early Christian symbols (peacocks – a symbol of eternal life; laurel garland or wreath – a symbol of victory, i.e. victory over death; flowers – a symbol of Eucharistic bread) indicates the liturgical purpose of the ledge.

The results of archaeological research on two crypts which were discovered in 2006 put an end to this dispute. Analysis of archaeological finds and the wall-painting system allowed them to be dated to the middle of the fourth century.10 The paintings of the ‘crypt on N. I. Tur’s land’ are identical in their colour range, composition, style and technique of execution. The monument has been repeatedly mentioned in scientific works devoted to research on Late Antique art in Chersonesos. It was published for the first time in the fundamental work by M. I. Rostovtsev.11 The author gives a description, plans and sketches of murals, and makes an attempt to interpret them (fig. 2, 3, 4). Unfortunately, the location of the crypt is lost; this publication is the only documentary description. Subsequent researchers have thus used the materials of M. I. Rostovtsev.

As in most of the other crypts of Chersonesos, two diagonal garlands and a wreath with a Chi Ro in its centre were drawn on the ceiling.16 The system of wall-painting on the right wall has become the subject of discussion.17 On the left part of the wall were remnants of the image of a plant with fruits or flowers (fig. 3). Some experts have considered it to depict grapes. At some distance from the bush was depicted a woman in a long tunic. Her right arm is bent, with the forearm directed to the right; the left arm is not preserved. One foot stands frontally, and the position of the second foot is not preserved.18 To the right of the woman, the figure of a man was placed. The figure is dressed ‘almost in the same way’ as the male figure from the ‘Uvarov crypt’ (fig. 6).19 The man’s arms are bent at the elbows; forearms and feet are directed to the right. The figures’ wrists, shoulders, and heads are not preserved, but by analogy with the other monuments, it can be assumed that they were placed frontally. The gesture of the forearms and the position of the feet transmit the direction of movement. The figures ‘are not resting on the surface’, but the depiction of the ground was rendered by wavy brown lines.20

The catacomb was discovered in 1894 but was excavated only in 1912. Initially, it was planned as a typical burial construction (2.48–2.13 х 2.79 х 1.68–1.60 m). In the left and back walls under the ceiling, niches were located. During construction, the bench was made (fig. 7). Later, the inner area was significantly reconstructed. The loculeus, located opposite the entrance, was deepened to the level of the floor. Approximately in the middle of the loculeus, a small protrusion and a small niche behind it were made. In the front wall, the passage was made. As a result of the restructuring, a small area similar to a church apse was built. Its decoration is of special interest (fig. 8). As M. I. Rostovtsev wrote, on the lowest part of the wall the stripes of a deformed meander were depicted. The wall-painting could be an imitation of architectural details; ‘the wall on the left of the niche and at the bottom of the niche, i.e. the altar barrier, were decorated with something similar to the grid ornament typical for antique balustrades and Early Christian altar barriers’.12 Rostovtsev 1914, pp. 451, 457, 462, 469 and 503–507. Diatroptov 2002, рp. 34–36. 5 Khrushkova 2004, рp. 168–194. 6 Zavadskaya 2005, рp. 258–288. 7 Sorochan 2006, pp. 140–154. 8 Philippov 2009, р. 315. 9 Zubar, Khvorostyany 2000, рp. 144–154; Zubar 1988, р. 12; 1999, рp. 300–311. 10 Fomin 2014, рp. 299–390. 11 Rostovtsev 1913; 1914. 12 Rostovtsev 1914, р. 475. 3 4

Shevchenko 2011, pp. 55–147. Rostovtsev 1914, р. 469. 15 Matveeva 2016, рp. 66–70. 16 Rostovtsev 1914, р. 478. 17 Rostovtsev 1914, рp. 473–474; Zubar 1991, р. 22; Sorochan 2005, рp. 1069–1070; Fomin 2007, рp. 164–165. 18 Rostovtsev 1914, р. 455. 19 Rostovtsev 1914, р. 433; Fomin 2014, pp. 325–327, fig. 7–10. 20 Rostovtsev 1914, р. 455. 13 14

7

Mikhail V. Fomin

Fig. 2. Plan of the ‘crypt on N. I. Tur’s land’ (Rostovtsev 1914, tab. CXVI).

To the right of the figures was situated a depiction of the city walls with gates and towers. To the right of it, plants (trees, reeds?) were depicted with black lines. Researchers interpret the remnants of wall-painting under the image of the city variously (fig. 3).

The history of the Christianisation of Late Antique Chersonesos is known from the hagiographic texts ‘The Lives of Holy Bishops of Cherson’.21 The original text that was the basis of the known variations was written probably in the fifth to sixth century by a local man who was familiar with the history of the church of Chersonesos/Cherson. The original text contained an accurate description and recognisable place names in the city. The most interesting text for the interpretation of the crypt is a story about the saint Bishop Basil. In the text of ‘The Lives of Holy Bishops of Cherson’, a story tells that the saint resurrected the son of one of the city rulers. The hagiographic source indicates that the boy had been buried in a crypt not far from the city. His parents found St Basil, who had been

On the left side of the back wall, a male figure was drawn, but only images of his feet have survived (fig. 4). Thus, after reconstruction, the inner space of the burial complex becomes similar to the apse with altar and altar barrier. All those changes make it possible to assume that the crypt was used for church services, including liturgy. A special attitude of the citizens towards the crypt and its transformation into a church leads to the conclusion that the construction played a special role in the religious life of the first Christians of Chersonesos. Probably the crypt became a martyrium.

21

8

Mogarichev et al., 2012.

The ‘Crypt on N. I. Tur’s Land’ in Early Byzantine Chersonesos

Fig. 3. The system of wall-painting of the right wall: ‘man and woman walking in the city’ (Rostovtsev 1914, tab. CXVI).

Fig. 4. The elements of the crypt painting (Rostovtsev 1914, tab. CXVI).

built. Later, this place became the centre of the complex of the Western Basilica. In the sixth century, the cross-shaped church–martyrium was built on the site of his death. The place where the miracle happened was especially revered, and the crypt could be a theophanic martyrium. Analysis of the wall-painting of the crypt–martyrium may indirectly confirm this assumption. It can be assumed that the murals are illustrations of the text ‘The Lives of Holy Bishops

hiding in the city, and asked him to resurrect their son. Basil washed the body ‘as if christening him’ and the boy was resurrected in response to his prayer. As a result of the miracle, the child’s parents and their relatives were baptised, but in spite of this, St Basil was murdered. In the city, there were many monuments associated with the memory of the first bishop. Above his grave, a church was 9

Mikhail V. Fomin

Fig. 5. The painting of the ‘apses’ (Rostovtsev 1914, tab. CXVI).

Fig. 7. The plan of the crypt before reconstruction (Rostovtsev 1914, tab. CXVI).

Fig. 6. The elements of the ‘Uvarov crypt’ painting (Rostovtsev 1914, tab. CVI).

of Cherson’. The man and the woman who are moving towards the city could possibly be the parents of the child, and the man who is coming out of the town, the saint Bishop Basil. The city is probably a symbolic depiction of Chersonesos. Fig. 8. The plan of the crypt after reconstruction (Rostovtsev 1914, tab. CXVI).

Thus, the site which is known as the ‘crypt on N. I. Tur’s land’ could initially have been a tomb owned by members of the local aristocracy. After the event that is described in ‘The Lives of Holy Bishops of Cherson’, the crypt became a particularly venerated site that was associated with the memory of St Basil, the first bishop of Chersonesos. At the end of the fourth century, the crypt was rebuilt into a theophanic martyrium where church services were situated, associated with the memory of the early history of Christianity in the city.

Bibliography Abbreviation: МАИАСК = Материалы по археологии и истории античного и средневекового Крыма – Materials in Archaeology and History of Ancient and Medieval Crimea. 10

The ‘Crypt on N. I. Tur’s Land’ in Early Byzantine Chersonesos Diatroptov P. D., 2002. ‘Еще раз к вопросу о датировке росписей херсонесских христианских склепов’, in Церковная археология южной Руси. Сборник материалов международной конференции “Церковная археология: проблемы, поиски, открытия” (Севастополь, 2001 г.), Sevastopol, pp. 34–37.

Zavadskaya I. A., 2005. ‘Декоративно-символическая система росписи херсонесских христианских склепов’, in МАИАСК, 11, рp. 258–288. Zubar V. M. 1988. ‘По поводу датировки христианской росписи склепов из некрополя Херсонеса’, in Научно-атеистические исследования в музеях, Leningrad, рp. 4–18.

Fomin M. V., 2007. ‘О почитании святых в Херсонесе – Херсоне по материалам некрополя’, in N. N. Bolgov (ed.), Мир Византии: материалы Международного научного семинара (г. Белгород, 27–28 октября 2006 г.), Belgorod, рp. 164–165.

Zubar V. M., 1991. ‘Проникновение и утверждение христианства в Херсонесе Таврическом’, in P. P. Tolochko (ed.), Византийская Таврика. Сборник научных трудов (к XVIII конгрессу византинистов), Kiev, pp. 8–29.

Fomin M. V., 2014. ‘Раннехристианская живопись позднеантичного Херсонеса’, in МАИАСК, 6, pp. 299–390.

Zubar V. M. 1999. ‘К интерпретации одного из сюжетов росписи склепа 1912 г. из Херсонеса’, in Херсонесский сборник, 10, рp. 300–311.

Khrushkova L. G., 2004. ‘Христианские памятники Крыма (состояние изучения)’, in Византийский временник, 63, рp. 168–194.

Zubar V. M., Khvorostyany A. I., 2000. От язычества к христианству. Начальный этап проникновения и утверждения христианства на юге Украины (вторая половина III – первая половина VI вв.), Kiev.

Matveeva Y., 2016. Декоротивные ткани в мозаиках Равены. Семантика и культурно смысловой контекст, Kharkiv. Mogarichev Y. M., Sazanov A. V., Sargsan T. Z., Sorochan S. B., Shaposhnikov A. K., 2012. Жития епископов Херсонских в контексте истории Херсонеса Тавричесского – The Lives of Bishopsof Cherson within the Context of the History of Tauric Chersonesos (Нартекс. Byzantina Ukrainensis, 1–2), Kharkiv. Philippov A. E., 2009. ‘Росписи раннехристианских склепов Херсонеса Тавричесского в контексте художественной традиции: синтез с архитектурой, символика, литургическая основа’, in S. A. Belyev (ed.), Очерки по истории христианского Херсонеса, I, Saint Petersburg, pp. 193–322. Romanchuk A. I., 2007. Исследования ХерсонесаХерсона: Раскопки. Гипотезы. Проблемы, II, Ekaterinburg. Rostovtsev M. I., 1914. Античная декоративная живопись на юге России, Saint Petersburg. Shevchenko Y. Y., 2011. ‘К вопросу о методике датировки первохристианских пещерных храмов восточной Европы’, in МАИАСК, 3, рp. 55–148. Sorochan S. B., 2005. Византийский Херсон: вторая половина VI – первая половина Х вв.очерки истории и культуры, Kharkiv. Sorochan S. B., 2006. ‘Мартирий Воскрешения, или еще раз о херсонесском склепе на земле Н.И. Тура’, in Античная древность и средние века, 37, pp. 140–154. Turovsky E. Y, 2011. ‘Участок херсонесского некрополя у загородного храма Богородицы Влахернской: некрополь святых?’, in Херсонес — город святого Климента. Тезисы и сообщения VI Международной конференции по Церковной археологии (Севастополь, 12–18 сентября 2011 г.), Sevastopol, pp. 40–41. 11

2 Carian Iasos in Byzantine Times Diego Peirano Ministero dell’Istruzione, dell’Università e della Ricerca (Italy) Iasos, during the long period from Late Antiquity to Turkish settlement, underwent a variety of urban changes. A first economic expansion phase, due to the exploitation of the local marble quarries, was accompanied by the building of important basilicas. Later on, a progressive occupation of public spaces occurred, first of all in the agora area, where the initial construction of houses was followed by manufacturing activities. These difficulties seem to reach their apex in the seventh century, when the circulation of money stopped, restarting in the ninth to tenth century. Recovery, following the eleventh century, brought important projects for restoration and widening of the fortifications, having as their main objective control of the harbour rather than defence of the town, which by that time appears to have been depopulated and of little importance. Durant la longue période qui s’étend de la plus haute Antiquité jusqu’à la domination turque, l’urbanisme de Iasos a connu plusieurs transformations. Une première phase d’expansion économique, liée à l’exploitation des carrières de marbre locales, s’est accompagnée de la construction de vastes basiliques. Survint ensuite une occupation progressive des espaces publics, à commencer par l’agora, où s’installèrent d’abord des habitations, puis des activités de production. Ces difficultés semblent atteindre leur sommet au VIIe siècle, marqué par l’arrêt de la circulation monétaire qui ne reprit qu’à partir des IXe-Xe siècles. La reprise qui suivit au XIe siècle amena d’importants travaux de restauration et d’agrandissement des fortifications, avec pour objectif le contrôle des ports, plutôt que la défense de la ville qui apparaissait désormais dépeuplée et sans importance. Während der langen Zeit von der Spätantike bis zur türkischen Besiedelung machte Iasos mehrere urbane Veränderungen durch. Die erste wirtschaftliche Blüte war durch die Ausbeutung der lokalen Marmorsteinbrüche mit der Einrichtung von wichtigen Basiliken verbunden. Später folgte die fortschreitende Aneignung öffentlicher Räume, vor allem die Aneignung der Agora, wo man zuerst Häuser errichtete und anschließend handwerkliche Tätigkeiten ausübte. Die Krise erreichte ihren Höhepunkt im 7. Jh., als der Münzverkehr erlosch. Erst während des 9.10. Jhs. lässt er sich erneut nachweisen. Der Aufschwung setze sich im 11. Jh. fort, es wurden wichtige Bauprojekte wie die Erweiterung der Befestigung durchgeführt. Diese zielte mehr auf die Sicherung des Hafens als auf die der Stadt, deren Einwohnerzahl um diese Zeit zurückging und die insgesamt zu dieser Zeit weniger wichtig zu sein scheint. Iasos, nel lungo periodo che intercorre tra la tarda antichità e lo stanziamento turco, dopo una prima fase di espansione economica legata allo sfruttamento delle locali cave di marmo, cui si accompagnò anche la costruzione di ampie basiliche, vide una progressiva occupazione degli spazi pubblici, prima tra tutti l’agorà, entro cui si ricavarono inizialmente abitazioni e poi attività produttive. Queste difficoltà sembrano raggiungere il loro apice nel VII secolo quando s’interrompe la circolazione della moneta che riprende nel IX-X secolo. Alla ripresa seguente l’XI secolo si accompagnarono importanti opere di recupero e di ampliamento delle fortificazioni, finalizzate però più al controllo dei porti che alla difesa della città, che appare oramai spopolata e priva d’importanza. Located on one of the deep inlets that indented the Carian coast, Iasos was linked in the north-west to the road that

followed the coastline toward Didyma and then Miletus, in the east to the main city of the hinterland, Mylasa, and 13

Diego Peirano in the south-west to the Halikarnassos peninsula.1 From 1960 to 2013, archaeological investigations were carried out under the auspices of the Italian Archaeological Mission.2 Iasos had two ports, and based its ancient economy on olive oil and wine production,3 fishing4 and the manufacture of columns made of local marble.5 From the mid-fourth century BC, the city had a circuit wall, first enclosing the peninsula, then a part of the mainland (fig. 1).6 The transformation of the peninsula circuit wall began in the second half of the third century AD, when the gateway from the western port to the agora was closed and a chamber in a nearby tower was filled with debris and soil to reinforce it.7 The course of the walls was modified by the inclusion of the northern quarters, the end of the aqueduct with its cisterns, and a large bath building (fig. 2.1).8 These changes may be related to the Gothic incursions that reached nearby Ephesus. A hoard, which was buried under the floor in the western stoa,9 may also be associated with these events.

Fig. 1. Plan of Iasos’ fortifications (Judeich 1890).

This seems to have been the first moment of crisis for the ancient city. Later, in the fifth century, some sectors of the agora’s porticoes fell, as attested by two coins, one of Valens and one of Theodosius II, found among porch excavations. Meanwhile, in the western sector of the northern stoa, a number of dwellings were built within the portico, respecting the original alignments (fig. 2.2).10 The building technique includes clay used as mortar, blocks, re-used limestone debris and schist slabs. Near the stylobate, these walls include parts of columns and entablature, evidently spolia from the fallen porticoes. These dwellings preserved the pavement of the stoa. Afterwards, the porticoes were without pavements. These constructions were levelled and rebuilt many times. But not all the porticoes had fallen; scratched crosses on the columns of the east portico suggest that some parts of stoai were still standing. Furthermore, the system of access to the agora was re-arranged: while its south-west entrance From Iasos to Miletus: Hild 2013, p. 26, Ragone 1993, p. 871, Benoit, Pierobon-Benoit, Ragone 1993, pp. 874 ff. From Iasos to Mylasa: Hild 2013, pp. 37 ff., From Iasos to the Halikarnassos peninsula: Hild 2013, p. 28, fig. 55. 2 Annual reports of excavation activities from 1963 to 1972 may be found in Annuario della Scuola archeologica italiana ad Atene; from 1969 to 1974, in Türk Arkeoloji Dergisi; from 1974 to 1983, in Anatolian Studies; from 1983 to 2013, in Kazı Sonuçları Toplantısı. 3 A probable oil press was found in a fourth- to fifth-century building superimposed on the prehistoric necropolis. Another oil press was found in the House of Mosaics. On these, see: Giberti 2012, p. 20; Angiolillo 2001, pp. 339–352. 4 Different sources mention fishing in Iasos. On these, see: Benoit, Pierobon Benoit 1993, pp. 902–919. Strabo relates how, due to the rather poor chora, the city drew its sustenance from the sea: Strabo, 14.2.21. Even in Byzantine times on some pottery there appears a large fish (tuna?): Berti, Desantis, 2003, p. 26. 5 Berti, Peirano 2014, p. 46. The beginnings of the export of Iasian marble columns can be moved forward to the first century AD, as the re-used column of the Bari cathedral is not made of Iasian marble. 6 Compare: Berti 2012, p. 106; Pierobon Benoit 2005, pp. 6–12. 7 Berti 2003–4b, pp. 297–300. 8 Viscogliosi 2009, pp. 6–13. 9 Compare Berti 2003–4a, pp. 25–28; Tondo 2003–4, pp. 38–39. 10 Romagnoli 2012, p. 12. 1

Fig. 2. Iasos: plan of the isthmus with the buildings mentioned in this chapter highlighted.

was blocked, a new entrance into the western stoa was opened.11 11

14

Pagello 1985, p. 137.

Carian Iasos in Byzantine Times

Fig. 3. A block conserving various traces of cutting (photo from D. Levi’s excavations, courtesy of the Scuola Archeologica Italiana di Atene).

The first centre of the Christian polis seems recognisable in an extramural basilica, built in a Roman necropolis that extended between the two harbours. The basilica was roughly square in plan (35.5 x 31 m), with three aisles divided by columns (fig. 2.3). It had a narthex with two side rooms; two other annexes flanked the apse, resulting in a straight eastern wall. This building, given its resemblance to a number of Cilician churches, and to the nearby Knidos church D, may be related to the transition from these Cilician forms to those of western Asia Minor.12 Even if the first attestation of a Christian community dates back to 431, I would be cautious in proposing too early a date: many Cilician churches date to the late fifth century,13 and several inhumations located around the church are later than the sixth century, as the following will clarify. I would, however, point out that such a marginal position in relation to the city core could indicate resistance to Christianity by local elites and/or a still densely occupied urban area.14

(fig. 2.4). Here were found marble blocks showing saw marks, abandoned above thick layers of emery (fig. 3).15 As evidence of Iasian marble slabs begins to appear in the sixth century, the tomb’s transformation probably took place at that time. The water-powered stone saw used water from the nearby aqueduct, so both quarries and workshop may have been imperial property. Furthermore, a sledge road was found between the quarry of Karaoğlan Deresi and the port below; in some parts it splits into two tracks on different levels, in order to facilitate the transit of men, animals and handcarts.16 From here, marble reached the eastern harbour; from there, it could be transported to the nearby workshop and cut or loaded onto ships. Marble probably continued its journey towards Ephesus, the first emporium of Asia Minor, and most likely the first step toward Constantinople. The sea was the main route for the distribution of stone, rarely attested in neighbouring regions, and, at the same time, the sea made extraction economical. The map of finds shows how marble is present almost exclusively along the coasts. In the sixth century, Iasian marble began to be widespread, and the polychromy of the stone, skillfully laid using the ‘open stain’ technique, was admired and celebrated.17 It appeared

At the time in question, this area seems to have been extremely vital: near the basilica, the quadriporticus of a Roman mausoleum was transformed into a sawmill Serin 2004, pp. 169–177, and the related bibliography. Gough 1972, pp. 199 and 210–212. 14 A model for the area’s Christianisation has been proposed by Blid 2012, pp. 259–268. 12

Bruno 2012, p. 707. Pierobon Benoit 2008, p. 8. 17 Paul the Silentiary, Descriptio Sanctae Sophiae 623. 15

13

16

15

Diego Peirano

Fig. 4. Plan of the agora (N. Masturzo, with author’s modifications).

in Bosra in 512–513 in the church of SS Sergius, Bacchus and Leontius,18 as well as in Constantinople in the church of St Polyeuktos, built between 524 and 527.19

and St Sergius in Gaza.23 Considering the degree to which it was appreciated, red marble remained a product that was rarely used locally. It was more advantageous to export it and to purchase other marbles; for this reason, then, it entered into local everyday life via the lower-level production of tables, basins, bowls and mortars, which probably lasted into the seventh century.24

During the reign of Justinian, Iasian marble reached its floruit. At that time, it appeared in places linked to the patronage of the emperor and to his entourage: the churches of the Holy Apostles in Constantinople,20 San Vitale and Sant’Andrea Maggiore in Ravenna,21 St John in Ephesus22

Returning to the workshop: close to its entrance, part of a necropolis connected to the nearby basilica was found. Interred beneath a layer of yellowish sandy soil were ten burials. A block sawn within the workshop and used as a

Masturzo 1995, p. 378 Harrison 1993, p. 42. 20 Theophanes the Confessor, Chronographia, A.M. 6058. 21 Agnellus of Ravenna, Liber pontificalis Ecclesiae Ravennatis (ed. Holder Hegger 1878, p. 329); Deichmann 1972, pp. 61–64. 22 Deichmann 1976, p. 216. 18 19

23 24

16

Mango 1972, p. 60, n. 25. Berti, Peirano 2014, pp. 48–50.

Carian Iasos in Byzantine Times

Fig. 5. The tower on the eastern jetty of the western harbour (E. Garberoglio).

closure slab dates this sector of the necropolis. The sandy soil, the same as that observed within the workshop, relates this area to the atelier landfill.25

The Acropolis basilica was built in the late fifth to early sixth century, on a terrace that still preserves traces of earlier building. It was preceded by an atrium or forecourt and a narthex. The whole complex measured 18.85 x 30.10 m, or 13.30 x 19.73 m excluding the atrium and narthex. The complex had an unusual orientation, with the apse facing SSE, in order to follow the level curves of the hill and to adapt to the existing terrace. The acropolis church included a ciborium, an imported ambo and geometric mosaics in the three aisles.30 In the right aisle, a mosaic inscription mentioning Artemisios – a presbyter – provides us with information about the patronage of these interventions (fig. 2.5).31

Between the late fifth century and Justinian’s era, two basilicas were built with reused materials in areas in transformation: one on the acropolis and one in the former agora.26 The latter was aligned with the stoai; it was threeaisled, measured 14.30 x 22.50 m and had a narthex with side chambers. The apse was also flanked by two rooms that originated from the extension of the aisles (fig. 4.1). The basilica, probably thanks to the exchange of marble products, included imported openwork transennae made of white marble, using Constantinopolitan patterns.27 In front of the basilica and towards the dipylon (fig. 4.3), the monumental fourth-century BC east gate, still in use, a pavement of large slabs was installed, which was absent from the previous agora (fig. 4.2). By the mid-sixth century, the surrounding area of porticoes was being transformed into a manufacturing area with furnaces that produced iron28 and ceramics. At the southeast corner of the agora, the tripartite public building, perhaps a place of worship dedicated to oriental cults, saw the transformation of its central room into a chapel (fig. 4.4).29

At this time, the agora still appears to have been the most vital area: in the eastern part of the tripartite building, a furnace was created; a seventh-century date is provided for this transformation by a Sassanian coin found within an ash layer.32 The agora basilica was in use until the seventh century, as attested by a bishop’s seal of the seventh century found in the eastern part of the south aisle.33 Perhaps due to Arab raids, well attested in the late seventh century on the Carian coast and nearby islands, coin circulation stopped and did not resume until the ninth and tenth centuries.34 Serin 2004, pp. 105–161, Peirano, Garberoglio 2010–2011, pp. 340– 344, Peirano 2012, pp. 25–33. 31 On the mosaics, see: Berti, 1985, pp. 156–158; Serin, 2004, pp. 140– 148; as well as the forthcoming study ‘La basilica cristiana dell’acropoli di Iasos’ by D. Peirano. 32 This furnace replaced an ancient well. Cf. Baldoni 2013, p. 9. 33 Cosentino 2013, p. 22. 34 Delrieux 2012, pp. 388 ff. 30

Baldoni 1996, p. 6. 26 Serin 2004, pp. 38–49. 27 Peirano 2011, pp. 13–14; Berti, Peirano 2016, p. 185. 28 Pagello 1985, p. 137. 29 On furnaces, see: Ghini 2000, pp. 16–17; Spanu 2013, p. 3; Baldoni 2013, p. 9; Pagello 1985, p. 137. Regarding the tripartite building, see: Baldoni 2014, pp. 369–383. 25

17

Diego Peirano

Fig. 6. Plan of the acropolis castle (N. Masturzo – M. Molinari).

to systematic spoliation during the early Byzantine period: the marble columns were smashed with sledgehammers and wedges to obtain new building material or lime. Around the late eighth to early ninth century, this area, close to the western harbour, may have been a place frequented by Arabs, as a wall inscription in a room, which mentions a mosque, may suggest.39 Ruggieri has proposed that this stretch of coast was at the time travelled by Arabs in search of cypress lumber.40

Between the eighth and ninth century, a triconch church was built at the end of the former acropolis basilica. In the tenth century, a necropolis began to spread around it. Similarly, around the late seventh to eighth century, the agora basilica also fell;35 its façade was left where it had collapsed and after some time it was replaced by a church whose unique nave included paintings and an epistyle. From the eighth century, a necropolis also began to spread here.36 Reconstruction of the church has been dated variously between the eighth and the eleventh century; the detail of the façade left as it had fallen suggests a hiatus between the loss of the basilica and the subsequent reconstruction in reduced form. Furthermore, the later date of the epistyle, around the tenth to eleventh century, should be considered.37

During the Byzantine era, the mainland fortification walls seem also to have still been in use. Near them were found traces of buildings that cannot be precisely dated; from there, the inhabitants were able to enter the fortifications in case of danger.41

Near the agora was an anonymous temple, now associated with the Aphrodite cult (fig. 2.6);38 its stoai were subject

In the second half of the eleventh century, the fortifications were renewed with the construction of the isthmus castle between the ports; the castle recovered the northern

Serin 2004, p. 99. Berti 1988, pp. 2–3. 37 Serin 2004, p. 97. 38 Masturzo 2016, pp. 75 ff. 35

On the inscription, see: Ruggieri 2005, p. 249; 2009, p. 211, fig. 7–8. Ruggieri 2005, p. 249. 41 Pierobon Benoit 2007, p. 8.

36

39 40

18

Carian Iasos in Byzantine Times sectors of existing walls, including the north-west agora corner. The new fortification shows traces of refacing and repair. Excavations have shown that its foundations were not particularly deep or robust;42 where possible, existing structures were reused for the walls. The castle used both circular and octagonal towers, and its outer space remained without buildings (fig. 2.7). From the excavations of this sector, two coins from the mid-eleventh century43 were recovered. Therefore, the castle construction seems to be related to the first incursions of the Turks in the late eleventh century, which affected the hinterlands of many cities, including nearby Miletus. Fortified farmhouses44 were built there, and the castle located on top of the theatre harbour was also rebuilt at that time.45 In Iasos, the main entrance to the castle was located to the north, oriented towards the interior roads, bypassing the former agora and the peninsula, which was by then probably depopulated and without importance.

using mostly reused materials. The church was a vaulted three-aisled basilica, divided by pillars and 13.80 x 15.40 m in size. It is also possible that the ruin now visible is a restoration of an older church (fig. 2.9).49 Restoration of the Iasian fortifications continued with the construction of the acropolis castle. This was built at the top of the peninsula and had two entrances: the main entrance to the north, facing towards the city, or rather towards its hinterland, and the smaller one on the other side, allowing quick access to the western port below. The towers had wooden attics fireproofed with clay, spread over planking. There was no access to the towers from inside the castle, but only from the upper walkways, which were linked by 13 stairways to the ground, nine single and four double. Some of the towers had fireplaces. Ten casemates, with from one to four rooms, were located around the perimeter. In the centre, there was a vaulted cistern of approximately 250 cubic metres (fig. 2.10 and fig. 6). The complex may be related to the transition from the Byzantines to the Turks, from the very late thirteenth century to the early fourteenth: in fact, in 1337, a treaty was signed between the Venetians and the Menteshe in which the local port was identified as a stopover for Venetian ships that purchased alum.50 At the time, the city seemed no longer to exist, so the fortification is probably related to control of the ports.

Similar intentions seem to have been associated with the construction of a tower on the eastern jetty of the western harbour (fig. 2.8 and fig. 5). This is fairly well preserved, at least on the north and west sides. They are reinforced with robust buttresses and have two rows of loopholes for the use of war machinery. The tower was built on two levels: while the lower level had a small cistern, the upper had a fireplace. The entrance was oriented towards the peninsula and raised by approximately 2 m; the tower was probably linked to the city walls by means of a raised passage. The date is indicated by the strong buttresses, similar to those of a number of isolated towers from the late eleventh century.46

In Iasos, the Byzantine period seems to have ended with the excavation of mass graves for the inhumation of many individuals who had perhaps died in a plague;51 these were found around the Lascarid church, outside the city and near a port, where these kinds of diseases usually originated.

From the tenth to the eleventh century there are various remains of constructions provided with fireplaces and superimposed on the western sector of the northern agora porch. The scant excavated material suggests that after every collapse the debris was reused. Alternatively, the walls may be the foundations of structures made of perishable materials. In fact, in two layers, numerous post holes containing stones have been identified; they were intended to hold the poles of sheds or fences.47

In conclusion, I would point out that not all the findings of Byzantine structures in Iasos have been mentioned here. Various city sectors show the remains of Byzantine buildings, above all the east gate complex, where the archaeological finds range from Late Antiquity to late Byzantine times. This area was excavated in the early years of the archaeological mission and is difficult to interpret today (fig. 2.11). A number of other structures, such as two nearby churches, are located on private properties, and thus were never excavated. Conversely, in the sectors of the agora, the acropolis, the harbour areas and the isthmus, excavations were more intensive and more reliable longterm data were collected.

A dense network of productive structures was found in the west stoa and in the courtyard, as is attested by pits and holes with a burned perimeter; ducts and furnace waste point to molten metals in the area. These remains are dated by two coins to the mid-eleventh century. Other furnaces, not always easy to date, and used to produce metals, glass and pottery, were found scattered across the entire agora area.48

Bibliography Abbreviations:

Harbours seem to have maintained their importance. The so-called ‘Lascarid church’ was built outside the city’s walls and facing the east port in the thirteenth century,

AAFerrara = Atti dell’Accademia delle Scienze di Ferrara. ASNP = Annali della Scuola Normale Superiore di Pisa. Classe di Lettere e Filosofia. BA = Bollettino d’arte.

Berti 2011, pp. 181–182. Berti 2011, p. 181–182; Romagnoli, 2012, p. 13. 44 Thonemann 2011, pp. 261–262. 45 Müller Wiener 1967, p. 285. 46 Masturzo 1993–1994, p. 157. 47 Romagnoli 2012, p. 12. 48 Baldoni et al. 2004, p. 66. 42 43

Serin 2004, p. 192; Serin 2005, p. 177. On the date of the acropolis castle, see: Berti et al. 2010, p. 395. On treaty: Zachariadou 1983, s.v. Asinkalesi, Lasso, Iassos. 51 Berti 2004, p. 8. 49 50

19

Diego Peirano BAIC = Bollettino dell’Associazione Iasos di Caria.

Berti F., Peirano D., 2016. ‘Iasos di Caria e il rilancio del suo porto in età bizantina: il ruolo del marmo’, in Hortus Artium Medievalium, 22, pp. 178–189.

BN = Bollettino di numismatica.

Blid J., 2012. Felicium Temporum Reparatio. Labraunda in Late Antiquity, PhD dissertation, Stockholms Universitet.

Angiolillo S., 2001. ‘Un impianto per la lavorazione dell’olio a Iasos’, in Architettura, arte e artigianato nel Mediterraneo dalla Preistoria all’Alto Medioevo. Atti della Tavola rotonda internazionale in memoria di Giovanni Tore, Cagliari, 17–19 dicembre 1999, Oristano, pp. 339–352.

Bruno M., 2012. ‘Quarry blocks in marmor iassense from the Balık Pazarı at Iasos (Turkey)’, in A. Gutiérrez Garcia-Moreno, P. Lapuente Mercadal, I. Rodà de Llanza (eds.), Interdisciplinary Studies on Ancient Stone. Proceedings of the IX Association for the Study of Marbles and Other Stones in Antiquity (ASMOSIA). Conference, Tarragona, 2009, Tarragona (Documenta, 23), pp. 706–714.

Baldoni D., 1996. ‘Tombe bizantine all’esterno del Balik Pazari’, in BAIC, 2, pp. 6–8. Baldoni D., 2008. ‘Un edificio tripartito  nell’agora di Iasos: alcune considerazioni preliminari’, in AAFerrara, 83–84, pp. 17–33.

Cosentino S., 2013. ‘Un inedito sigillo bizantino da Iasos di Caria’, in BAIC, 19, pp. 21–23.

Baldoni D., 2013. ‘Lo scavo nel vano centrale dell’edificio tripartito: campagna 2012’, in BAIC, 19, pp. 7–10.

Deichmann F., 1972. ‘Studi sulla Ravenna scomparsa’, in Felix Ravenna, 103–104, pp. 61–112.

Baldoni D., 2014. ‘Culti orientali a Iasos: ipotesi interpretativa di un edificio di età romana’, in L. Karlsson, S. Carlsson, J. Blid Kullberg (eds.), ΛΑBΡΥΣ. Studies presented to Pontus Hellström, Uppsala (Acta Universitatis Upsaliensis. Boreas, 35), pp. 369–383.

Deichmann F., 1976. Ravenna. Hauptstadt des spätantiken Abendlandes, II/2, Wiesbaden. Delrieux F., 2012. ‘La circulation monétaire à Iasos durant la période byzantine, d’Anastase Ier Dikoros à la Quatrième Croisade’, in AAFerrara, 88, pp. 383–426.

Baldoni D., Franco C., Belli P., Berti F., 2004. Carian Iasos, Istanbul (Homer Archaeological Guides, 2).

Ghini G. P., 2000. ‘Iasos: l‘età bizantina’, in BAIC, 6, pp. 15–17.

Benoit J., Pierobon Benoit R., 1993. ‘Il territorio a N di Iasos’, in ASNP. S. III, 23/3–4, pp. 902–919.

Giberti F., 2012. ‘L’edificio tardoromano nell’area della necropoli preistorica. Notizie preliminari: l’impianto di spremitura (vano A)’, in BAIC, 18, pp. 19–23.

Berti F., 1985. ‘I mosaici degli edifici di culto cristiano’, in BA, 31–32 – Supplemento, pp. 155–162.

Gough M., 1972. ‘The Emperor Zeno and some Cilician churches’, in Anatolian Studies, 22, pp. 199–212.

Berti F., 1988. ‘Les travaux de la Mission Archéologique Italienne à Iasos en 1987’, in Kazı Sonuçları Toplantısı, 10, pp. 1–9.

Harrison R. M., 1993. ‘Monumenti e urbanistica nella città’, in A. Guillou (ed.), La civiltà bizantina, oggetti e messaggio. Architettura e ambiente di vita, Rome (Corsi di studi, 6), pp. 27–74.

Berti F., 2003–4a. ’Iasos 1969: il rinvenimento del tesoro’, in BN, 40–43, pp. 25–28. Berti F., 2003–4b. ‘Le vicende di una torre di difesa’, in BN, 40–43, pp. 297–300.

Hild F., 2014. Meilensteine, Straßen und das Verkehrsnetz der Provinz Karia, Wien (Österreichische Akademie der Wissenschaften. Philosophisch-historische Klasse. Denkschriften, 464; Veröffentlichungen zur Byzanzforschung, 33).

Berti F., 2004. ‘Iasos, campagna di scavo 2003’, in BAIC, 10, pp. 4–9. Berti F., 2011. ‘Iasos, 2009 Campaign’, in A. N. Toy, H. Dönmez, Ö. Ötgün (eds.), 32. Kazı Sonuçları Toplantisi, 24–28 Mais 2010, İstanbul, I, Ankara (T.C. Kültür ve Turizm Bakanlığı, 3267/2; Kültür Varlıkları ve Müzeler Genel Müdürlüğü, 147/2), pp. 176–187.

Holder-Egger O. (ed.), 1878. ‘Agnelli qui et Andreas Liber pontificalis Ecclesiae Ravennatis’, in Monumenta Germaniae historica. Scriptores rerum Langobardicarum et Italicarum saec. VI–IX, I, Hannover, pp. 265–391.

Berti F., Desantis P., 2003. ‘Indagini subacquee a Iasos di Caria (Turchia)’, in A. Benini, M. Giacobelli (eds.), Atti del II Convegno nazionale di archeologia subacquea, Castiglioncello, 7–9 settembre 2001, Bari (Bibliotheca archaeologica, 12), pp. 21–34.

Mango C., 1972. The Art of the Byzantine Empire, 312– 1453. Sources and Documents, Englewood Cliffs (Sources and Documents in the History of Art Series). Masturzo N., 1993–1994. ‘La torre del porto occidentale e le fortificazioni post-classiche di Iasos’, in AAFerrara, 71, pp. 155–183.

Berti F., Mengoli D., Molinari M., 2010. ‘Iasos (Turchia). Relazione preliminare sulle ricerche archeologiche compiute nel castello dell’acropoli’, in Archeologia medievale, 37, pp. 385–396.

Masturzo N., 1995. ‘Architettura siriana pre-giustinianea: la chiesa dei SS. Sergio, Bacco e Leonzio. Problemi interpretativi e di conservazione’, in XLI Corso di

Berti F., Peirano D., 2014. ‘Il marmo iasio e alcuni tra i manufatti a esso collegati’, in Marmora, 10, pp. 45–58. 20

Carian Iasos in Byzantine Times cultura sull’arte ravennate e bizantina. Seminario internazionale di studi su: Ravenna, Costantinopoli, Vicino Oriente, Ravenna, 12–16 settembre 1994. In memoria del Prof. F. W. Deichmann, Ravenna, 1994, pp. 369–386.

Thonemann P., 2011. The Maeander Valley: A Historical Geography from Antiquity to Byzantium, Cambridge (Greek Culture in the Roman World). Tondo L., 2003–4. ‘Il tesoro dell’agorà di Iasos. Un archivio d’argento dell’epoca di Plotino’, in BN, 40–43, pp. 38–39.

Masturzo N., 2016. Iasos. L’area a sud dell’agorà, I, Rome (Archaeologica, 176; Missione archeologica italiana di Iasos, 6).

Viscogliosi A., 2009. ‘Il “Castello di terraferma” a Iasos’, in BAIC, 15, pp. 6–13.

Müller Wiener W., 1967. ‘Das Theaterkastell von Milet’, in Istanbuler Mitteilungen, 17, pp. 279–290.

Zachariadou E. A., 1983. Trade and Crusade. Venetian Crete and the Emirates of Menteshe and Aydin (1300– 1415) – Εμπόριο Και Σταυροφορία. Η Βενετοκρατούμενη Κρήτη και τα Εμιράτα του Μεντεσέ και του Αϊδινίου (1300–1415), Venice (Library of the Hellenic Institute of Byzantine and post-Byzantine Studies, 11).

Pagello E., 1985. ‘Il Foro imperiale romano. Considerazioni preliminari’, in BA, 31–32 – Supplemento, pp. 137–150. Peirano D., 2011. ‘Ricerche su assetti, arredi, percezione nella basilica dell’agorà di Iasos’, in BAIC, 17, pp. 10–16. Peirano D., 2012. ‘Il ciborio della basilica dell’acropoli’, in BAIC, 18, pp. 25–33. Peirano D., Garberoglio E., 2012. ‘Amboni iasii: una rilettura con due inediti’, in AAFerrara, 88, pp. 339–358. Pierobon Benoit R., 2005. ‘La campagna di ricognizione del 2004: la cinta di terraferma e l’Akarca Tepe’, in BAIC, 11, pp. 6–12. Pierobon Benoit R., 2007. ‘Ricognizioni archeologiche nel Golfo di Mandalya. Relazione sulla campagna 2006’, in BAIC, 13, pp. 7–13. Pierobon Benoit R., 2008. ‘Un tempio extraurbano nel territorio di Iasos? Nuovi dati dalle ricognizioni archeologiche nel golfo di Mandalya. Campagna 2007’, in BAIC, 14, pp. 7–9. Ragone G., 1993. ‘Da Mileto a Iasos. Toponomastica antica, itinerari antiquari, ricognizioni moderne’, in ASNP. S. III, 23/3–4, pp. 871–902. Romagnoli G., 2012. ‘Scavi nella stoà settentrionale dell’agorà di Iasos. Campagna 2011’, in BAIC, 18, pp. 10–14. Ruggieri V., 2005. La Caria bizantina: topografia, archeologia ed arte. Mylasa, Stratonikeia, Bargylia, Myndus, Halicarnassus, Soveria Mannelli. Ruggieri V., 2009. ‘The Carians in the Byzantine period’, in Rumscheid, F. (ed.), Die Karer und die Anderen. Internationales Kolloquium an der Freien Universität Berlin, 13–15. Oktober 2005, Bonn, pp. 207–218. Serin U., 2004. Early Christian and Byzantine Churches at Iasos in Caria. An Architectural Survey, Vatican City (Monumenti di Antichità cristiana, 17). Serin U., 2005. ‘Some observations on the Middle Byzantine church outside the east gate at Iasos’, in La parola del passato, 60, pp. 156–178. Spanu M., 2013. ‘La campagna 2012 a Iasos’, in BAIC, 19, pp. 2–6.

21

3 The Christian Phase of the Rotunda in Thessaloniki Sophia Akrivopoulou Εφορεία Αρχαιοτήτων Ροδόπης / Ephorate of Antiquities of Rhodope (Greece) This chapter sets out to prove that the Rotunda was converted into a Christian church, with no additions or major modifications in the fifth century, and that the lavish second phase was added later. The first sanctuary was situated in the eastern interior niche (no 6), as the inner arrangement, the brick crosses in the exterior masonry of the dome and ambulatory and the two vaulted tombs in the later ambulatory indicate. The second construction phase, which transformed the Roman Rotunda into a concentric double-shell building, took place at the beginning of the sixth century. The depiction of Cosmas and Damian in the mosaics of the dome, with a medical box between them, reinforces this hypothesis. Cet article cherche à démontrer que la Rotonde a été transformée en église chrétienne au Ve siècle, sans ajout ni modification majeure, et que son caractère somptueux appartient à une phase postérieure. Le premier sanctuaire était situé dans la niche no 6, à gauche de la grande abside actuelle, comme l’indiquent l’aménagement intérieur, les briques en croix de la maçonnerie extérieure de la coupole et du déambulatoire, ainsi que les deux tombes voûtées du déambulatoire tardif. La représentation de Côme et Damien dans la mosaïque de la coupole renforce l’hypothèse selon laquelle la seconde phase de construction daterait du début du VIe siècle. In diesem Artikel wird dargelegt, dass die Rotunde bereits im 5. Jh. in eine christliche Kirche umgewandelt wurde. Dies erfolgte zunächst ohne bedeutende architektonische Eingriffe, erst während der zweiten Bauphase im 6. Jh. kam es zu baulichen Veränderungen. Der älteste Chorraum befand sich in der Nische 6, links neben der heutigen Apsis. Außer einer entsprechenden Umgestaltung des Nischeninnenraums weisen Tonkreuze im Tympanon über der Nische auf seine Stelle hin, ebenso wie zwei frühchristliche Gräber im Süden, im Bereich des während der zweiten Bauphase errichteten Chorumgangs. Während der zweiten Bauphase, zu Beginn des 6. Jhs., wurde die römische Rotunde zu einem konzentrischen doppelschaligen Gebäude umgestaltet. Die Darstellung der Heiligen Kosmas und Damian mit einem Medizinkasten in ihrer Mitte auf den Kuppelmosaiken untermauert diese Annahme. La costruzione della Rotonda di Tessalonica fu portata a termine al momento della trasformazione in chiesa, entro e non oltre la fine del V secolo. Nella prima fase cristiana non vi furono mutamenti architettonici né aggiunte; l’ingresso rimase sempre a Sud e il santuario nella nicchia nord-orientale, vicino e a sinistra dell’abside attuale. La Rotonda divennne un importante centro di pellegrinaggio attorno al quale si collocarono le sepolture «ad sanctos». All’inizio del VI secolo, l’edificio fu completamente ristrutturato con la costruzione di un nuovo santuario, di un deambulatorio, nel quale furono incluse le tombe più antiche, di annessi; infine fu aggiunta la decorazione. La rappresentazione di Cosma e Damiano con un contenitore per strumenti medici posto tra loro, nella fascia musiva dei martiri nella cupola, rafforza la datazione del rinnovamento all’inizio del VI secolo. The Rotunda in Thessaloniki is currently the subject of three major disputes: these centre on whether it was built as a mausoleum or as a Roman temple, whether the cupola was completed in the first phase, and whether its transformation into a Christian church took place at the end of the fourth, in the mid-fifth, or in the early sixth century.

but he insisted that the building had been completed in a single phase.1 According to him, the Rotunda was a Roman temple built by Galerius. He assumed that the original 1 Hébrard 1920, p. 24. Dyggve doubted the double curvature of the Rotunda’s cupola (Dyggve 1941, p. 65). Torp’s arguments against Hébrard are based mostly on the iron cramps found in the masonry of the higher curvature of the dome (Torp 1953, p. 491). Theocharidou 1991–92, p. 66, agreed that the Rotunda remained unfinished, but on completely different grounds.

Ernest Hébrard was the first to observe that the dome of the Rotunda was in two parts, each with a different radius, 23

Sophia Akrivopoulou

Fig. 1. Rotunda, Hébrard’s excavation, 1917–1918, ground plan (Hébrard 1920, pl. III–IV). Digital processing by Hamza Memet Emin, architect.

new finds appeared in the three northern niches (no 4, 5, 6 in Hébrard’s plan). The new finds were not recorded on the ground plan, and the few photos that were taken remained unpublished (fig. 2).

building was still untouched at the turn of the fifth century; the lavish second phase, with its ambulatory, sanctuary and mosaics, was added in the mid-fifth century.2 In this chapter, I intend to show that the Rotunda became a Christian church with – at first – no alterations or additions, and that the second construction phase (Christian renovation) came later. The arguments supporting this view lie both in the archaeological material gathered to date and in an iconographical approach to the mosaic decoration.

Niche no 6 had a unique feature: it was separated from the central hall by a low wall with mural decoration on the external face. The wall stood on a marble sill that followed the chord of the niche from one pier to the other. The inner space was divided into three bays by two parallel walls, seemingly erected on the marble surface of the Early Christian floor (fig. 2). All these constructions remained intact until 1953, when the monument, which at that time served as the city’s Byzantine Museum, underwent major repairs in view of the Ninth International Byzantine Congress held in Thessaloniki that same year (fig. 3).4

Hébrard’s excavation lasted from 24 September 1917 to 30 April 1918, the ground plan having been finished by March 1918 (fig. 1). Nevertheless, the work inside the Rotunda did not end then. The inner space was cleared down to the level of the marble Early Christian floor that Hébrard had recently discovered.3 During this extensive work, which lasted at least three more months at the Ephorate’s expense, 2 3

Of course, the purpose and date of this arrangement cannot be determined with accuracy from a few photographs. The sill, however, kept in situ during the 1953 repairs

Hébrard 1910, p. 26. Vokotopoulou 1986, p. 36; Ηébrard 1920, p. 28.

4

24

Pelekanidis 1960, p. 224; Akrivopoulou, in press, fig. 12.

The Christian Phase of the Rotunda in Thessaloniki

Fig. 2. The northern niches of the Rotunda after the clearing of the interior, 1918 (Ephorate of Antiquities of Thessaloniki City).

Fig. 3. Rotunda, niches no 6 and 7, 1945–1953 (ELIA/MIET).

25

Sophia Akrivopoulou

Fig. 4. Rotunda, eastern elevation (Theocharidou 1992, fig. 1).

and still visible today, seems well accommodated to the marble floor of the second phase. We may presume that niche no 6 had a specific function, at least at the time of the Christian renovation of the building. It is possible that this arrangement reflected an earlier, special use of the niche that was preserved and emphasised in the second construction phase. During the Christian renovation, niche no 6 was marked by brick crosses on the outer face of the masonry of the drum (fig. 4).5 One cross defines the centre of the niche, while another marks its eastern end. A third one must have existed to define the western end, but it was destroyed when this part of the dome collapsed.6 Another brick cross was revealed during the excavation of 1973–74 on the outer face of the ambulatory, next to the northern wall of the sanctuary, marking the lower northern end of niche no 6 (fig. 5). Niche no 6 is oriented almost precisely to the east, in contrast to the later sanctuary (niche no 7), which tends more to the south than the east, and so it was the obvious place for the first sanctuary of the newly sanctified Christian Rotunda (fig. 6). The brick crosses incorporated into the masonry during the Christian renovation reinforce this hypothesis. The older sanctuary could not have been abolished; it would have remained a holy place – a chapel probably, next to the 5 The external masonry of the drum belongs to the Christian renovation. Cf. Ηébrard 1920, 28, plan V. 6 Theocharidou 1992, pp. 63–64, 66; Velenis 2003, pp. 51–56.

Fig. 5. Brick cross on the external face of the northern ambulatory (Ephorate of Antiquities of Thessaloniki City).

26

The Christian Phase of the Rotunda in Thessaloniki exist; however, none of them can be dated before the end of the fifth century.11 If this vague chronological schema can be accepted, the conversion of the Rotunda into a church should be dated to the fifth century, before the construction of the vaulted tombs, and the Christian renovation should be dated after their construction, probably in the first decades of the sixth century. It has been suggested that the transformation of the Rotunda into a church should be associated with Galerius and his persecution of the Christians.12 No archaeological evidence can support such a theory, especially based on the acceptance that the Rotunda was built as a mausoleum for Galerius.13 The transformation of a pagan building into a church with no architectural additions, as part of a rather slow process that reflected social and religious interrelations inside the Late Roman city, seems only logical. Alison Franz outlined three stages in the process of transformation of pagan temples to churches in Athens: the removal of the cult statue, the use of the building as a place of worship with no structural alteration, and the remodelling of the building. This pattern seems to suit the Rotunda very well.14 The Christian renovation represents the second major construction phase of the building. Outside the Rotunda, a circular timber-roofed ambulatory was added; the two earlier vaulted tombs were left untouched beneath its pavement. A sanctuary with an altar and reliquary tomb (niche no 7)15 was added on the east side, while a new entrance (niche no 3) was pierced in the west side, with a rectangular porch, an apsidal annex to the north, and an underground barrel-vaulted compartment in front of it.16 In this complex of annexes one may see the narthex of the new church,17 and perhaps an atrium as well.18 The original southern entrance was moved to the outside of the ambulatory as a monumental rectangular propylon with two concentric buildings annexed to its sides. The furnishings included at least the marble ambo (fig. 1).19 With this renovation the Rotunda was transformed into a concentric double-shell building similar to San Vitale in Ravenna.20

Fig. 6. Ground plan of the Roman Rotunda. In grey, the first Christian sanctuary (after Moutsopoulos 2013).

sanctuary that was in use. The absence of an altar tomb in the floor of the first sanctuary is not at all astonishing; in that very early period the cult of relics could not yet have been fully developed, since the opening of graves and the translation of relics was strictly forbidden.7 Hébrard excavated two vaulted tombs south-east of the Roman building, in the area of the later ambulatory (fig. 1).8 The tombs follow the regular east–west orientation; they are almost parallel with niche no 6, the one presumed here to be the first sanctuary.9 By contrast, all the other tombs in the Rotunda, scattered among the remains of the annexes of the second phase on the south side, follow precisely the orientation of those buildings. When the tomb next to the south entrance was excavated again in 1979, it became clear that it was already there when the coarse mosaic pavement of the ambulatory was laid on top of it (fig. 7).10

The date of the Rotunda’s second construction phase is still under dispute. Hébrard’s mid-fifth-century dating was challenged by Dyggve, who proposed the end of the fourth century instead, associating the conversion of the Rotunda with Theodosius I.21 In 1939, the year of Akrivopoulou 2017, pp. 51–52. Nasrallah 2005, p. 467; Fallone 2013, p. 23. 13 The mausoleum theory is controversial; cf. Johnson 2009, pp. 74–75. Moreover, the dedication of the Rotunda to St George is late and cannot be used as an argument for ‘a direct commentary to Galerius’ (Fallone 2013, pp. 23–24). The original dedication, if any, is under dispute. As Spieser 1999, p. 34, has pointed out, the consecration of a church at the time did not necessarily mean a dedication to a saint, as is the practice today. 14 Franz 1965, p. 201. 15 Pelekanides 1956, fig. 5; Torp 2011, fig. 14–15. 16 Hébrard 1920, p. 28, pl. III–IV; Dyggve 1941, fig. 14; Torp 2011, fig. 16. 17 Velenis 2018, p. 58, fig. 8. Torp suggests the ‘salle de l’ambon’ as narthex (Torp 2018, vol. 1, p. 35). 18 Hébrard 1920, p. 28. 19 Hébrard 1920, pp. 28–30; Sodini 1976. 20 Mentzos 2002, p. 67. 21 Dyggve 1940, p. 154; Dyggve 1957, p. 353. 11

12

Thus, the two vaulted tombs mentioned above should be separated from the Christian renovation and the bulk of the later Christian cemetery. Moreover, they cannot be dated before the end of the fifth century; Early Christian burials within the walls of Thessaloniki are scarce but they do Mango 1990, pp. 51–52; Spieser 1999, p. 30; Akrivopoulou 2017, p. 40. Vokotopoulou 1986, p. 35; Akrivopoulou 2017, p. 49. 9 Velenis 2018, p. 55, has recently suggested the northern niche (no 5) as the first sanctuary of the Rotunda. 10 Moutsopoulos 2013, pp. 181–182, fig. 299–301, 304–305. 7 8

27

Sophia Akrivopoulou

Fig. 7. Vaulted tomb under the coarse mosaic pavement of the ambulatory. Ground plan and section (Moutsopoulos 2013, fig. 299–300).

Dyggve’s excavation in Thessaloniki, Weigand’s paper on the iconography of the ‘Kalenderfries’ was published.22 Weigand’s proposal of a sixth-century dating for the mosaics was accepted the following year by Schoenebeck, Dyggve’s principal associate.23 Torp has consistently supported Dyggve’s Theodosian dating, since 1953.24 Grabar vaguely suggested a fifth-century date for the mosaics.25 The first scholar to re-introduce a sixth-century dating for the mosaics was Spieser.26 A Constantinian dating for both the original building and the mosaics, suggested recently by Bakirtzis and Mastora, has added to the confusion.27 I shall present here two arguments to reinforce my thesis of a sixth-century dating for the Rotunda’s second construction phase: the brick crosses inside and outside the building, and the crosses and the iconography of a pair of saints in the mosaics. Three brick crosses have already been mentioned above, in connection with niche no 6. Two more brick crosses are found on the walls of the new sanctuary of the sixth-century renovation, situated indoors, one at the northern end of the sanctuary28 and one in the middle of it. All of them have flattened ends. The sanctuary’s central cross is set upon an arched base that is hollow now, but initially must have been stepped (fig. 8). A gold cross with flattened ends is depicted in the mosaic of the vault of niche no 1, where the original entrance of the Roman Rotunda was located.29

Fig. 8. Brick cross on a stepped base on the inner face of the sanctuary of the Rotunda (S. Akrivopoulou).

All these crosses in the walls of the neighbouring niches no 6 and 7 should be explained in the same way, as φυλακτήρια of the holiest sections of the building. They could be associated with the edict of Theodosius II of 435, ordering the demolition of the last pagan temples and the purification of their sites by the setting up of a cross.30

Weigand 1939. Schoenebeck 1940, p. 481. 24 Torp 1955, p. 491; Torp 2018, vol. 1, pp. 35–36, 127, 467ff. Since 1991, Torp has suggested a four-stage chronological scheme for the Rotunda: period I, Galerius; period II, Theodosius I (conversion/second construction phase); period III, sixth century (alterations and additions to the second phase); period IV, seventh century (Torp 1991, pp. 16–17; Torp 2017, pp. 37–38; Torp 2018, vol. 1, pp. 35–38, 51–52). 25 Grabar 1967, p. 81. 26 Spieser 1984, pp. 125–164; Spieser 2005. 27 Bakirtzis, Mastora 2010; Bakirtzis 2012, p. 116. 28 Dyggve 1941, fig. 13. 29 Spieser 1984, pp. 135–141; Torp 2018, vol. 2, fig. IV.1. 22 23

The type of cross with flattened ends on a stepped base, which was introduced on coins of Tiberius II (574–582), is believed to depict a real object, most probably the jewelled 30

28

Codex Theodosianus 16.10.25; Franz 1965, p. 187.

The Christian Phase of the Rotunda in Thessaloniki

Fig. 9. Zone of the martyrs, south-west panel, Cosmas and Damian with a medical box (Bakirtzis 2012, pl. 63).

golden cross that Theodosius II erected at Calvary.31 The best depiction of the type is in Hagia Sophia in Constantinople, on an opus sectile panel above the Royal Door.32 In the same monument there were also three brick crosses, at the base of the vault of the sanctuary.33 Most preserved examples of this kind of gilded cross with suspended pendoulia or monograms and inlaid precious gemstones are dated to the sixth century and later.

The most interesting iconographic element that leads to the dating of the renovation phase in the sixth century is the representation of Cosmas and Damian in the southwest panel, over niche no. 2 (fig. 9). This observation was first made by Weigand in 1939 but was later treated with scepticism.36 Cosmas and Damian are the oldest pair of the group of the Anargyroi. The three earliest representations of them are found in the West: at SS Cosma e Damiano in Rome,37 at S. Michele in Africisco in Ravenna (Bode Museum),38 and at the Euphrasian Basilica in Parentium.39 All three works are from the sixth century. This led Weigand to the logical hypothesis that not only do the mosaics of the Rotunda date from the same era but also that the veneration of these saints began in Rome.40

The luxurious version of this type of cross is found in the Rotunda in the mosaics in the dome. Tall jewelled golden crosses rise in the middle of the northern and southern panels of the martyrs’ zone. These crosses, as well as fantastical buildings with jewelled architectural elements in the background,34 remind us of the rich architectural decoration of Hagios Polyeuktos.35

Since 1939, no representations of Cosmas and Damian older than the sixth century have been found.41 The origins of their veneration, however, have become clear. They

Large crosses were erected early at the Mount of Olives (by Helen) and at Calvary. Eudocia replaced the cross at the Mount of Olives (Heid 2000, pp. 140–143, 162–164) and her husband Theodosius II replaced the cross of Calvary (Heid 2000, pp. 229–242, especially 238 and n. 473). 32 Underwood 1960, pp. 206–207, fig. 2–3; Grabar 1965, 7 pp. 2–73, fig. 17–17bis. 33 Teteriatnikov 1995. I thank Natalia Teteriatnikov, who kindly brought this article to my attention. 34 For an interpretation of the imaginary buildings, see Carile 2012, pp. 49–100. 35 Spieser 1984, pp. 135–141; Spieser 2005, p. 438. 31

Weigand 1939, pp. 124–125; Harrold 2007, p. 97; Gounaris 1972. Harrold 2007, pp. 88–93. 38 Effenberger 1975. 39 Harrold 2007, pp. 93–95. 40 Weigand 1939; Harrold 2007, pp. 33–36. 41 Xyggopoulos 1977. 36 37

29

Sophia Akrivopoulou found once more in the north-west panel, over niche no 4, where two other saints, like Cosmas and Damian, are depicted dressed as priests. Only one name is preserved. It is Ananias, whose feast day is in January according to the inscription. Most probably, however, he is the bishop of Damascus whose feast day is in October,52 and who performed miraculous cures during his lifetime.53

were so popular that eventually three equivalent pairs, from Syria, Arabia and Rome, were created.42 Today, the feast day of the Roman pair is in September, as mentioned in the inscription in the Rotunda. This month is not mentioned in the Synaxarium Constantinopolitanum,43 but this is not unusual; the dates of saints’ feasts have changed a great deal through the centuries, and there are many local particularities.44

The saint depicted next to Ananias is most probably Alexander of Pydna or Thessaloniki, to whom, after his death, miraculous healing properties were attributed, like those of St Demetrius and St Artemius.54 If this hypothesis is correct, in the pair of Ananias–Alexander we have the second version of medicine in the Christian world: the impersonation of Christ himself, who healed by laying on of hands and God’s grace. By contrast, Cosmas and Damian represent conventional medicine, the discipline that was taught in the school at Alexandria or on the battlefield at the Roman Limes.55

The veneration of Cosmas and Damian originated in Syria, but it became established with the erection of the Cosmidion in Constantinople, around 480.45 The basic attribute in the early iconography of the saints was a medical box worn over the shoulder on a strap. This element is present in the apse of SS Cosma e Damiano and later in the Chapel of the Physicians of Santa Maria Antiqua.46 In the church of S. Michele in Africisco they hold thecae vulnerariae, cylindrical cases of medical instruments, made of copper. In Parentium, they were depicted with a medical bag.

Conclusions

In Thessaloniki, another representation of Cosmas and Damian was preserved, in the ancient Agora; it has also been dated to the sixth century.47 Until recently, it was thought that in both representations in the city the saints were depicted without symbols of the medical profession.

The Roman Rotunda was converted into a Christian church, with no additions or major modifications, in the fifth century. The first sanctuary was situated in the eastern interior niche (no 6), as the inner arrangement and the brick crosses in the exterior masonry of the dome and ambulatory indicate. The two vaulted tombs that were excavated by Hébrard in the later ambulatory can be dated to this Early Christian phase.

In the Rotunda, between the two saints a rectangular object is depicted, resting on a piece of furniture. This object has generally been interpreted as a book.48 But it looks more like a box with two lids, decorated with οmphalia. It is probably a medical box, like the wooden or copper ones found in Italy and on the Limes.49 Thus, we find in the Rotunda one of the earliest depictions of the Anargyroi with a medical box without a strap; the next example dates from the eighth century, and is found in the portrait of St Abbacyrus in the atrium of Santa Maria Antiqua.50 Moreover, the same misunderstanding occurred with the medical boxes as with the other attributes: due to the lack of an iconographic model, their rendering looks like a book, just as the rendering of the thecae recalls a scroll and the rendering of a medical instrument in the hands of St Abbacyrus resembles a stylus.51

The proposed dating of the mosaics of the Rotunda to the sixth century has already been developed by Weigand, Spieser and Fourlas,56 while the views of Kleinbauer and Vickers57 are not distant. The analysis of the mosaic mortars seems to confirm this dating,58 and the ambo is safely dated to the same period.59 In the same context, the Acheiropoietos has recently been dated to the beginning of the sixth century.60 The type of the brick crosses of the sanctuary, the jewelled crosses on the northern and southern panels of the mosaics, the jewelled architectural elements of the imaginary buildings and the iconography of SS Cosmas and Damian, with a medical box between them, all reinforce the dating of this extended renovation to, most probably, the early sixth century.

The panels of the zone of the martyrs are organised in iconographic pairs and consequently the medical box is Csepregi 2007, pp. 53–54; Harrold 2007, pp. 27–28. Deubner 1907, pp. 38–40. 44 Torp 2011, pp. 23, 28. 45 Mango 1994 and Csepregi 2007, pp. 59–61, support an earlier dating of the Cosmidion. Cf. Harrold 2007, pp. 31–32. 46 Nordhagen 1968, p. 58. 47 Xyggopoulos 1977; Bonnekoh 2013, pp. 237–247. 48 Torp 2011, pp. 20–21. 49 Künzl 1983, p. 68 and fig. 43, pp. 93–95 and fig. 74, 76; Künzl 2002, Β119–Β120 and tables 37–38, C77 and table 54. 50 Knipp 2002, pp. 18–19 and fig. 12–15. 51 For the mixing of attributes in early iconography, two panels of diptychs from Sinai provide characteristic examples. On one, Cosmas holds in one hand an instrument, and in the other a codex and a theca vulneraria, while on the other one, Damian holds a scroll that Weitzmann (1976, B47 and B18) interprets as a theca vulneraria. He dates Cosmas to the eighth century and Damian a little earlier. 42 43

Synax. Const., 95. Synax. Const., 783–784. 54 Dimitrakopoulos 1976; Torp 2011, p. 24. 55 Although known as doctors from their vitae, after their death they were famous for performing miraculous healing through incubation. 56 Weigand 1939; Spieser 1984, p. 164. Fourlas 2012, pp. 193–195. 57 Vickers 1970; Kleinbauer 1972. 58 Korozi et al. 2001, p. 319. The last of the three mosaic substrata, the one that was tangential to the masonry, was dated with 14C to 497 with a variance of ±51 years. 59 Sodini 1976. 60 Raptis 2016; Fourlas 2012. 52 53

30

The Christian Phase of the Rotunda in Thessaloniki Bibliography

Frühjahr 1939’, in Laureae Aaquincenses memoriae Valentini Kuzsinszky dicatae, II, Budapest (Dissertationes Pannonicae ex Instituto Numismatico et Archaeologico Universitatis de Petro Pázmány nominatae Budapestinensis provenientes. Series II, 11), pp. 254–260.

Abbreviations: ACTA = Acta ad archaeologiam et artium historiam pertinentia. BCH = Bulletin de correspondance hellénique.

Dyggve E., 1957. ‘La région palatiale de Thessalonique’, in Acta Congressus Madvigiani, Hafniae 1954, I, Copenhagen, pp. 353–365.

BZ = Byzantinische Zeitschrift. ΔΧΑΕ = Δελτίον της Χριστιανικής Αρχαιολογικής Εταιρείας.

Effenberger A., 1975. Das Mosaik aus der Kirche San Michele in Africisco zu Ravenna. Ein Kunstwerk in der Frühchristlich-byzantinischen Sammlung, Berlin.

DOP = Dumbarton Oaks Papers. Synax. Const. = Synaxarium Ecclesiae Constantinopolitanae (ed. H. Delahaye, Acta sanctorum, November, Propylaeum, Brussels, 1902).

Fallone A., 2013. ‘Christian conversion: The spiritual transformation of Eastern pagan structures in Late Antiquity’, in Art Journal, 2013/1, pp. 17–29.

Akrivopoulou S., 2017. ‘Οι πρώτες ενάστειες ταφές στη Θεσσαλονίκη’, in I. D. Varalis, F. Karagianni (eds.), Κτίτωρ. Αφιέρωμα στον δάσκαλο Γεώργιο Βελένη, Thessaloniki, pp. 39–54.

Fourlas B., 2012. Die Mosaiken der AcheiropoietosBasilika in Thessaloniki. Eine vergleichende Analyse dekorativer Mosaiken des 5. und 6. Jahrhunderts, Berlin/Boston (Millennium-Studien zu Kultur und Geschichte des ersten Jahrtausends n. Chr., 35).

Akrivopoulou S., in press. ‘Η Ροτόντα και ο περιβάλλον χώρος της’, in Το Αρχαιολογικό Έργο στη Μακεδονία και τη Θράκη, 29.

Franz A., 1965. ‘From paganism to Christianity in the temples of Athens’, in DOP, 19, pp. 185–2015.

Bakirtzis Ch., 2012. ‘Ροτόντα’, in Ch. Bakirtzis (ed.), Ψηφιδωτά της Θεσσαλονίκης, 4ος–14ος αιώνας, Athens, pp. 51–117.

Gounaris G., 1972. ‘Αι εορταστικαί επιγραφαί των ψηφιδωτών του τρούλλου του Αγιου Γεωργίου Θεσσαλονίκης’, in Μακεδονικά, 12, pp. 201–227.

Bakirtzis Ch., Mastora, P. 2010. ‘Are the mosaics in the Rotunda into Thessaloniki linked to its conversion to a Christian Church?’, in Ниш и Византија – Niš and Byzantium, 9, pp. 33–45.

Grabar A., 1967. ‘À propos des mosaïques de la coupole de Saint-Georges à Salonique’, in Cahiers archéologiques, 17, pp. 59–81.

Bonnekoh P., 2013. Die figürlichen Malereien in Thessaloniki vom Ende 4. bis zum 7. Jahrhundert. Neue Untersuchungen zur erhaltenen Malereiausstattung zweier Doppelgräber, der Agora und der DemetriosKirche, Düsseldorf (Nea Polis, 7).

Harrold Η. J., 2007. Saintly Doctors: The Early Iconography of SS. Cosmas and Damian in Italy, PhD Thesis, University of Warwick. Hébrard Ε., 1920. ‘Les travaux du Service archéologique de l’armée d’Orient à l’Arc de triomphe de Galère et à l’église Saint-Georges de Salonique’, in BCH, 44, pp. 5–40.

Carile M. C., 2012. The Vision of the Palace of the Byzantine Emperors as a Heavenly Jerusalem, Spoleto (Studi e Ricerche di Archeologia e Storia dell’Arte, 12).

Heid S., 2001. Kreuz, Jerusalem, Kosmos. Aspekte frühchristlicher Staurologie, Berlin (Jahrbuch für Antike und Christentum. Ergänzungsband, 31).

Csepregi I. 2007. The Compositional History of Greek Christian Incubation Miracle Collections: Saint Thecla, Saint Cosmas and Damian, Saint Cyrus and John, Saint Artemios, PhD Dissertation, Central European University.

Johnson M. J., 2009, The Roman Imperial Mausoleum in Late Antiquity, Cambridge. Kleinbauer E. 1972. ‘The iconography and the date of the mosaics of the Rotunda of Hagios Georgios’, in Viator, 3, pp. 27–107.

Ćurčić S., 2000. Some Observations and Questions Regarding Early Christian Architecture in Thessaloniki, Thessaloniki.

Knipp D., 2002. ‘The Chapel of the Physicians in Santa Maria Antiqua’, in DOP, 56, pp. 1–23.

Deubner L., 1907. Kosmas und Damian. Texte und Einleitung, Leipzig/Berlin.

Korozi M., Fakorellis Y, Maniatis Y. 2001. ‘Μελέτη και χρονολόγηση µε άνθρακα-14 ασβεστοκονιαµάτων εντοίχιων ψηφιδωτών’, in Πρακτικά του Γ΄ Συµποσίου της Ελληνικής Αρχαιοµετρικής Εταιρείας (Αθήνα 6–9/11/1996), Athens, pp. 317–326.

Dimitrakopoulos F., 1976. ‘Άγιος Αλέξανδρος Πύδνης ή Θεσσαλονίκης’, in Ελληνικά, 29, pp. 268–277. Dyggve E., 1940. ‘Compte rendu succinct des fouilles de Thessalonique 1939’, in RAC, 17, pp. 149–156. Dyggve E., 1941. ‘Kurzer, vorlaufiger Bericht über die Ausgrabungen im Palastviertel von Thessaloniki, 31

Sophia Akrivopoulou Conference of the International Committee for the Conservation of Mosaics (ICCM). Wall and Floor Mosaics: Conservation, Maintenance, presentation, Thessaloniki, 29 October–3 November 2002. Proceedings, Thessaloniki, pp. 437–446.

Künzl E., 1983. Medizinische Instrumente aus Sepulkralfunden der romischen Kaiserzeit, Cologne/ Bonn (Kunst und Altertum am Rhein, 115). Künzl E., 2002. Medizinische Instrumente der römischen Kaiserzeit in Römisch-Germanischen Zentralmuseum, Mainz/Bonn (Kataloge vor- und frühgeschichtlicher Altertümer, 28).

Teteriatnikov Ν., 1995. ‘The hidden cross-and-tree program in the brickwork of Hagia Sophia’, in Byzantinoslavica, 56, pp. 689–699.

Mango C., 1990. ‘Constantine’s Mausoleum and the translation of relics’, in BZ, 83, pp. 51–62.

Theocharidou K., 1992. ‘Η Ροτόντα της Θεσσαλονίκης. Νέα στοιχεία και αποσαφηνίσεις με αφορμή τις αναστηλωτικές εργασίες’, in ΔΧΑΕ, 16, pp. 57–75.

Mango C., 1994. ‘On the cult of Saints Cosmas and Damian at Constantinople’, in Θυμίαμα στη μνήμη της Λασκαρίνας Μπούρα, Athens, pp. 189–192.

Torp H., 1955. ‘Quelques remarques sur les mosaïques de l’église Saint-Georges à Thessalonique’, in St. Kyriakidis, A. Xyngopoulos and P. Zepos (eds.), Πεπραγμένα του Θ΄ Διεθνούς Βυζαντινολογικού Συνεδρίου, Θεσσαλονίκη 12–19 Απριλίου 1953, I, Athens (Ελληνικά, Παράρτημα. Περιοδικόν σύγγραμμα Εταιρείας Μακεδονικών σπουδών, 9), pp. 489–498.

Mentzos A., 2001. ‘Reflections on the interpretation and dating of the Rotunda of Thessaloniki’, in Egnatia, 5, pp. 57–82. Moutsopoulos Ν., 2013. Η Ροτόντα του Αγίου Γεωργίου στη Θεσσαλονίκη. Αρχαιολογική έρευνα και αναστήλωση του μνημείου – The Rotunda of St. George in Thessaloniki. Archaelogical Research and Restoration of the Monument, Thessaloniki. Nasrallah L. S., 2005. ‘Empire and Apocalypse in Thessaloniki: Interpreting the Early Christian Rotunda’, in Journal of Early Christian Studies, 13/4, pp. 465–508.

Torp H., 1991. ‘The date of the conversion of the Rotunda at Thessaloniki into a church’, in Ø. Andersen and H. Whittaker (eds.), The Norwegian Institute at Athens. The First Five Lectures, Athens (Papers from the Norwegian Institute at Athens, 1), pp. 13–28.

Nordhagen J. 1968. ‘The frescoes of John VII (AD 705–707) in S. Maria Antiqua in Rome’, in ACTA, 3, pp. 115–120.

Torp H., 2011. ‘An interpretation of the Early Byzantine martyr inscriptions in the mosaics of the Rotunda at Thessaloniki’, in ACTA, 24, pp. 11–43.

Pelekanides S., 1956. ‘Η έξω των τειχών παλαιοχριστιανική βασιλική των Φιλίππων’, in Αρχαιολογική Εφημερίς, pp. 114–179.

Torp H., 2017. ‘Considerations on the chronology of the Rotunda mosaics’, in A. Eastmond, M. Hatzaki (eds.), The Mosaics of Thessaloniki Revisited. Papers from the 2014 Symposium at The Courtauld Institute of Art, Athens, pp. 35–47.

Pelekanidis S., 1960. ‘Μεσαιωνικά Μακεδονίας. 1. Θεσσαλονίκη’, in Αρχαιολογικόν Δελτίον, 16 – Κείμενον, pp. 222–225.

Torp H., 2018. La Rotonde palatine à Thessalonique. Architecture et mosaïques, Athens, 2018.

Raptis K., 2016. Αχειροποίητος Θεσσαλονίκης. Αρχιτεκτονική και γλυπτός διάκοσμος, PhD Thesis, Αριστοτέλειο Πανεπιστήμιο Θεσσαλονίκης.

Underwood P., 1960. ‘Notes on the work of the Byzantine Institute in Istanbul’, in DOP, 14, pp. 215–222. Velenis G., 1974. ‘Some observations on the original form of the Rotunda in Thessaloniki’, in Balkan Studies, 15, pp. 298–307.

Schoenebeck H., 1940. ‘Die Stadtplanung des römischen Thessalonike’, in Bericht über den VI. Internationalen Kongress für Archäologie, Berlin 21.–26. August 1939, Berlin, pp. 478–482.

Velenis G. (ed.), 2003. Μεσοβυζαντινή ναοδομία στη Θεσσαλονίκη. Μνήμη Μανόλη Χατζηδάκη, Ακαδημία Αθηνών, 5 Μαρτίου 2002, Athens.

Sodini J.-M., 1976. ‘L’ambon de la rotonde Saint-Georges: remarques sur la typologie et le décor’, in BCH, 100, pp. 493–510.

Velenis G., 2018. ‘Βυζαντινή Θεσσαλονίκη. Η πόλη και τα μνημεία της’, in F. Karagianni (ed.), Το ημέτερος κάλλος. Βυζαντινές εικόνες από τη Θεσσαλονίκη – Our Sacred Beauty. Byzantine Icons from Thessaloniki, Thessaloniki, pp. 49–75.

Spieser J.-M., 1984. Thessaloniki et ses monuments du IVe au VIe siècle, Athens/Paris. Spieser J.-M., 1999. ‘Les fondations d’Ambroise à Milan et la question de Martyria’, in ΔΧΑΕ, 20, pp. 29–34.

Vickers M., 1970. ‘The date of the mosaics of the Rotunda at Thessaloniki’, in Papers of the British School at Rome, 38, pp. 183–187.

Spieser J.-M., 2005. ‘Remarques sur les mosaïques de la Rotonde de Thessalonique’, in 8ο Συνέδριο Διεθνούς Επιτροπής για τη Συντήρηση των Ψηφιδωτών (ICCM). Εντοίχια και επιδαπέδια ψηφιδωτά: συντήρηση, διατήρηση, παρουσίαση, Θεσσαλονίκη, 29 Οκτωβρίου–3 Νοεμβρίου 2002. Πρακτικά – VIIIth

Vokotopoulou Ι., 1986. ‘Τα πρώτα 50 χρόνια της Εφορείας Κλασσικών Αρχαιοτήτων Θεσσαλονίκης’, in Η Θεσσαλονίκη μετά το 1912. Πρακτικά. Συμπόσιο, 32

The Christian Phase of the Rotunda in Thessaloniki Θεσσαλονίκη, 1–3 Νοεμβρίου 1985 – Thessalonique après 1912. Actes. Colloque, Thessalonique, 1–3 Novembre 1985, Thessaloniki (Αυτοτελείς εκδόσεις, 2), pp. 1–65. Weigand Ε., 1939. ‘Der Kalenderfries von Hagios Georgios in Thessaloniki’, in BZ, 39, pp. 116–145. Weitzmann K., 1976. The Monastery of Saint Catherine at Mount Sinai. The Icons, I, Princeton. Wortley J., 1999. ‘The Byzantine component of the relic-hoard of Constantinople’, in Greek, Roman and Byzantine Studies, 40, pp. 353–378. Xyggopoulos A., 1977. ‘Η παλαιοχριστιανική τοιχογραφία της Ρωμαϊκής Αγοράς της Θεσσαλονίκης’, in Βυζαντινά, 9, pp. 411–417.

33

4 The Archaeology of the Late Roman City of Zaldapa: The Status Questionis in 2016 (with an Appendix on Seasons 2017–2019)* Dominic Moreau Université de Lille, HALMA–UMR 8164 Research Centre (France)

Georgi Atanasov Регионален исторически музей – Силистра / Silistra Regional History Museum (Bulgaria)

Nicolas Beaudry Université du Québec à Rimouski (Canada) (with the collaboration of Ioto Valeriev, Albena Milanova, Brahim M’Barek, Elio Hobdari and Irina Achim)** The site currently identified as Zaldapa is the largest fortified Romano-Byzantine city – 25 ha intra muros – in the hinterland of present-day Dobrudja and North-eastern Bulgaria. Considering the size of Zaldapa, it is curious that it appears in only seven written sources, all from the sixth to eleventh centuries AD. Moreover, the site has been little explored in the last hundred years. Since 2014, however, a Bulgarian team has decided to resume field-work, which has led to the discovery of a new Christian basilica, larger than those previously known. Following that important discovery, the Bulgarian team invited French and Canadian scholars to visit them on the site, in order to evaluate its overall potential and to set up an international mission. During the summer of 2015,  excavations in the sanctuary of Basilica  ‘No 3’ allowed the release of a crypt and other interesting unknown structures. This paper describes the state of the art up to the end of the 2016 field season, as presented at the International Congress of Byzantine Studies by the above-mentioned French and Canadian scholars, together with the Bulgarian teams, as their first joint contribution. An appendix is added to this work, with the goal of briefly reporting the explorations up to 2019, as well as the beginning of the International Archaeological Mission at Zaldapa. Le site actuellement identifié comme Zaldapa est la plus grande ville romano-byzantine fortifiée – 25 ha intra-muros – de l’arrière-pays des actuelles Dobroudja et Bulgarie du Nord-Est. Compte tenu de la taille de Zaldapa, il est curieux qu’elle ne soit mentionnée que dans sept sources écrites, qui sont toutes des VIe-XIe siècles. De plus, le site a été peu exploré dans les cent dernières années. Depuis 2014, une équipe bulgare a, toutefois, décidé de reprendre le travail sur le terrain, ce qui a mené à la découverte d’une nouvelle basilique chrétienne, plus grande que celles connues jusqu’alors. À la suite de cette découverte importante, l’équipe bulgare a invité des chercheurs français et canadiens à venir leur rendre visite sur le site, afin de mettre en place une mission internationale. Pendant l’été 2015, des fouilles dans le chœur de la basilique « No 3 » ont permis le dégagement d’une crypte et d’autres structures intéressantes. Cet article propose l’état de la

This chapter is based both on the paper ‘La forteresse romaine tardive de Zaldapa (Dobroudja du Sud) et la crypte de sa basilique paléochrétienne “No 3”’ (D. Moreau, with the collaboration of G. Atanasov, V. Yotov, I. Valeriev, P. Chevalier and N. Beaudry), and on the poster ‘Zaldapa (South Dobrudja) and the surrounding Late Antique landscape: A preliminary study to fieldwork’ (D. Moreau and B. M’Barek), presented at the 23rd ICBS.

Respectively: Регионален исторически музей – Бургас / Burgas Regional History Museum (Bulgaria); Софийски университет ‘Свети Климент Охридски’/ Sofia University ‘St. Kliment Ohridski’ (Bulgaria); Éveha (France); Instituti i Arkeologjisë, Tiranë / Institute of Archaeology, Tirana (Albania); and Institutul de arheologie ‘Vasile Pârvan’ / Institute of Archaeology ‘Vasile Pârvan’ of the Academia Română / Romanian Academy (Romania).

*

**

35

Dominic Moreau, Georgi Atanasov & Nicolas Beaudry question jusqu’à la fin de la campagne de 2016, présenté au Congrès international des études byzantines, par lesdits chercheurs français et canadiens, avec l’équipe bulgare, comme première contribution conjointe. Un appendice est ajouté à ce travail, avec pour objectif de rendre compte brièvement des explorations jusqu’en 2019, de même que du début de la Mission archéologique internationale à Zaldapa. Der Fundplatz, der als Zaldapa identifiziert wird, ist mit 25 ha intra muros die größte befestigte römisch-byzantinische Stadt auf dem Gebiet der heutigen Dobrudscha und des nordöstlichen Bulgarien. Bei der Größe von Zaldapa ist es verwunderlich, dass die Stadt nur sieben Mal in den Schriftquellen Erwähnung findet, die alle aus dem 6.-11. Jh. stammen. Außerdem wurde der Platz während der letzten hundert Jahre wenig erforscht. Ab 2014 entschied sich ein bulgarisches Team die Feldarbeiten wieder aufzunehmen. Diese Arbeiten führten zu der Entdeckung einer frühchristlichen Basilika, größer als die beiden bis dahin bekannten Kirchen. Infolge dieser wichtigen Entdeckung lud das bulgarische Team französische und kanadische Wissenschaftler ein, um das allgemeine Potenzial des Ortes zu prüfen und eine internationale Zusammenarbeit einzuleiten. Während des Sommers 2015 brachten Ausgrabungen im Chor der Basilika drei Überreste einer Krypta und andere bis dahin unbekannte Strukturen zur Tage. Dieser Beitrag stellt den Stand der Ergebnisse bis zum Ende der 2016er Kampagne vor, die beim Internationalen Kongress für Byzantinische Studien durch französische und kanadische Wissenschaftler zusammen mit den bulgarischen Kollegen vorgestellt worden war, als ihr erster gemeinsamer Beitrag. Der Aufsatz ist durch einen Anhang ergänzt, mit dem Ziel einen kurzen Bericht der Forschungen bis 2019 zu geben und über den Beginn der internationalen archäologischen Kooperation in Zaldapa zu informieren. Il sito attualmente identificato come Zaldapa è la più grande città fortificata romano-bizantina – 25 ettari entro le mura – dell’entroterra dell’attuale Dobrugia e della Bulgaria nord-orientale. Date le dimensioni di Zaldapa, è curioso che il sito sia menzionato solo in sette fonti scritte, tutte risalenti al VI-XI secolo. Inoltre, esso è stato poco esplorato negli ultimi cento anni. Dal 2014, una squadra bulgara ha tuttavia deciso di riprendere il lavoro sul campo, che ha portato alla scoperta di una nuova basilica cristiana, più grande di quelle conosciute fino ad allora. Dopo questa importante scoperta, il team bulgaro ha invitato ricercatori francesi e canadesi a visitare il sito per creare una missione internazionale. Durante l’estate del 2015, gli scavi nel presbiterio della basilica «No. 3» hanno permesso il rinvenimento di una cripta e di altre interessanti strutture. Questo articolo propone lo stato dell’arte fino alla fine della campagna del 2016, presentata al Congresso Internazionale di Studi Bizantini dai citati ricercatori francesi e canadesi, con il team bulgaro, come primo contributo congiunto. A questo lavoro si aggiunge un’appendice, con l’obiettivo di rendere note brevemente le esplorazioni fino al 2019, così come l’inizio della Missione Archeologica Internazionale a Zaldapa. Introduction

stable and lasting Roman institution, and the only one to survive the ‘great invasions/migrations’.

As a border region conquered and Romanised fairly late, the Lower Danubian world is not as well represented in the written sources as other areas of the later Roman Empire. While the texts inform us about major military, political and religious issues involving the provinces on the Lower Danube, these sources tell us little about the fate of the cities or countryside.1 Over the past two decades, the publication of major archaeological works has, however, contributed to a better understanding of the important changes in these provinces during Late Antiquity, of the militarisation of landscapes and cities, but also of the effects of Christianity on urban forms and functions,2 particularly in settlements with an episcopal see. The impact of the Christian religion was very important, in that the episcopacy was the most

It is therefore mainly through archaeology that we can understand life in the Lower Danube provinces during Late Antiquity, and one site is particularly relevant to this understanding: Zaldapa (near modern Abrit, Dobrich district, Bulgaria). Despite being the largest known Romano-Byzantine stronghold of the hinterland of the provinces of Scythia and Moesia Secunda, the site has remained hardly explored until recently. The site is even more interesting given that it is located in the Danubian hinterland, which is less explored than the riverbank. In order to present its remarkable archaeological potential, in view of the design of a new international project on its urban fabric, this chapter presents a comprehensive status quaestionis on Zaldapa, focusing on the history of excavations until 2016, by highlighting the impressive discoveries made by the Bulgarian team who took over the exploration of the site in 2014. Before coming to this

Poulter 2007b. Cf.  especially Petrović 1996; Zahariade 1998; von  Bülow, Milčeva 1999; Slokoska et al. 2002; Poulter 2007a; Vagalinski 2007; Vagalinski et al. 2012. 1 2

36

The Archaeology of the Late Roman City of Zaldapa point, however, it seems important to quickly historicise and locate it, by first looking at the written sources.

Zaldapa may even have been the headquarters of the Lower Danube rebellion against Constantinople in the years 513–518. Given the known details of this story, it would be surprising if the rehabilitation of its wall were reassigned to Anastasius, as could be done for many other cities of the Danubian regions, unless such works are dated before 513.9

Reinsertion into time and space The aforementioned paradox between our limited knowledge of the site and its imposing nature could also be emphasised in terms of its presence in written sources. No inscription mentioning Zaldapa has been found to this day, the site being mentioned only in seven literary sources.3 All of them are in Greek, at least under the known forms of the toponym. It must be understood that although the name ‘Zaldapa’ is now favoured by the scientific literature, there is no consensus among ancient authors. Moreover, there is no occurrence of the name prior to the sixth century, which is rather surprising, because it seems impossible that the settlement is an ex nihilo foundation from this period, given its size and its name.4

At about the same time as John of Antioch, namely in the first half of Heraclius I’s reign, Theophylact Simocatta evokes Ζαλδαπά (oxytone according to Carl de Boor [1848–1923]) on four occasions. Two of them recall the sack of the city, first by the Avars in the autumn of 586 and, later, by the Slavs in the autumn of 593.10 The other two occurrences mention the recovery of the city by the Romans, as a result of these events, by the magistri utriusque militiae Castus and Petrus, in 587 and 593 respectively.11 This demonstrates without a doubt the strategic importance of the place at that moment.

The oldest mention is found in the Synekdemos of Hierocles, generally dated to the first quarter of the sixth century, and certainly before 535. The site is listed under the spelling Ζέλδεπα/Ζελδέπα, as the seventh πόλις of the province of Scythia.5 The following mention, in chronological order, is found in the Buildings of Procopius of Caesarea, probably written in the 550s, where the site is referred to under the name of Ζάλδαπα, without further details.6 It is only a very dry evocation, in one of the many lists of fortifications supposedly restored by Justinian in the Balkans.

Zaldapa is then mentioned as Ζελδίπας in the third Notitia episcopatuum Ecclesiae Constantinopolitanae, according to the numbering of the notitiae established by Jean Darrouzès (1912–1990).12 This occurrence is very problematic, since the hypotheses about the date and the object of this notitia do not really constitute an exact science. Generally dated between 750 and 869 (rather between 750 and 800, and more precisely around 787), the third notitia is clearly inspired by the Synekdemos of Hierocles.13 Thus, it does not represent the actual situation at the end of the eighth century, but rather a patchwork of information, some going back to the sixth century, corresponding most probably to an ideal Byzantine ecclesiastical hierarchy in a context of territorial claims after the violation in 756 of the Bulgarian-Byzantine peace treaty signed fifty years earlier.

At the beginning of the following century, John of Antioch, in a fragment of his universal chronicle that reached us through Constantine Porphyrogenitus’ De insidiis (first half/middle of the tenth century), presents Ζάλδαβα as the birthplace of one of the greatest enemies of Emperor Anastasius I, the famous Flavius Vitalianus.7 His revolt against Constantinople is a most important event as it contributed to the rise to power of the Justinianic dynasty.8

The links with the Synekdemos has led some researchers to consider Zaldapa as an actual episcopal see, claiming that Hierocles’ work would, in fact, be a picture of a reform of the ecclesiastical administration wanted by Justinian on the eve of the reconquest of the West.14 Despite the high probability of this most interesting and stimulating

3 The absence of inscriptions clearly mentioning Zaldapa obviously does not imply the absence of such documents coming from this site or the surroundings. To this day, most of them are spolia from the sanctuary of Heros Hephaistos Dabatopios, near modern Telerig (Dobrich district). For a more or less complete list (not all equivalences are given), cf. Kalinka 1906, nos. 23, 119 (= AE 1895, 55), 274, 306 (= IGBulg, II, 872), 416 and 422; IGBulg, II, 868–873; CMRED, I, 90; CCET, II/1, 186–188 (= IGBulg, II, 868–869 and 867bis) and 189; Torbatov 2000, pp. 33–34, 42–43, 63–64, 94, 96 and 100; AE, 2004, 1275. Thanks to Dan Dana (CNRS, ANHIMA–UMR 8210 research centre, France) who has helped in completing this list. For recent developments, cf. infra, n. 79. On the sanctuary of Telerig, cf. Torbatov 2001; 2005. 4 ‘Zaldapa’ would be a Thracian toponym, meaning ‘yellow water’ or ‘cold water’. Cf. Tomaschek 1894, p. 77. A milestone discovered in the nearby village of Aleksandria (Kapaklii before 1882; cf. Michev 2005, p. 28) – now lost? – could be read: Fines terrae vici. Is this a reference to Zaldapa’s original status? Cf. AE 1895, 55 = Kalinka 1906, no. 119 = Popa-Lisseanu 1921, p. 83, no 14. Cf. also Pârvan 1912, p. 21. 5 Hierocles, Synekdemus 637.7 (ed. Honigmann 1939, p. 13). On the date of composition, see ibid., pp. 1–2. 6 Procopius of Caesarea, De aedificiis 4.11 (ed. and transl. Veh 1977, pp. 244–245). 7 John of Antioch, Fragmenta ex Historia chronica 311.3 (ed. and transl. Roberto 2006, pp. 534–535) / 242.1 (ed.andtransl. Mariev 2008, pp. 452– 453) = Constantine Porphyrogenitus, De insidiis, Excerpta ex Ioanne Antiocheno, 103 (ed. de Boor, p. 143). 8 On Vitalian’s revolt against Anastasius, cf. now Moreau 2017, pp. 191–194.

9 On the attribution to Justinian of the reconstructions and repairs of Anastasius in the Lower Danube, cf. in particular the discussion and the bibliography in Cameron 1985, pp. 219–220. 10 Theophylact Simocatta, Historia 1.8.10 and 7.2.2 (ed. de Boor 1972, pp. 55 and 247). 11 Theophylact Simocatta, Historia 2.10.10 and 7.2.16 (ed. de Boor 1972, pp. 90 and 249). On the generals Castus and Petrus, cf. PLRE, III, pp. 274– 275 (Castus) and 1009–1010 (Petrus 55). Although Theophylact speaks of the recovery of Zaldapa by Roman troops, it is generally considered in the historiography that the site was destroyed by the successive invasions of Avars and Slavs, just before being finally abandoned. This, however, remains to be fully proven, and some elements could demonstrate a possible medieval occupation of these places (not necessarily in the same form as in Antiquity). See, for example, Valeriev 2015. 12 Notitiae episcopatuum Ecclesiae Constantinopolitanae, 3.652 (ed. Darrouzès 1981, p. 242). 13 See Darrouzès 1981, p. 20-33. 14 It was Emilian Popescu who offered the most developed version of this hypothesis, originally issued by Vasile Pârvan (1882–1927). Cf. Pârvan 1924; Popescu 1988 (which is the definitive version of this study, previously published in French and Romanian).

37

Dominic Moreau, Georgi Atanasov & Nicolas Beaudry the eleventh century. This list is actually a copy-and-paste of the Synekdemos of Hierocles.

proposal, it still remains a hypothesis among others, as the assumed Justinianic reform is not clearly documented by the sources and none of the arguments put forward are decisive.15 Vasile Pârvan was perhaps right, when he asserted that the fourteen new episcopal sees emanating from the so-called reform should eventually be considered “come possibilità autorizata e non come realtà attiva” (Pârvan 1924, p. 135).

The number of occurrences of Zaldapa in the literature is thus not negligible. Nevertheless, the details given about the site itself are very thin. In particular, there is nothing to locate it on a map, except that it was located in the province of Scythia (its placement in Moesia Secunda being a hapax). Actually, the association between Zaldapa and the site which is presented here is recent. Karel Škorpil (1859–1944) was the first to link that site to an ancient city name: Abrit(t)us.18 We do not know all the reasons for this choice, but the monumental character of the ruins of the Late Roman settlement currently known as Zaldapa, together with their location on strategic roads (fig. 1), and that its immediate environment was swampy, certainly contributed to its identification as the place of death of Emperor Decius.

In any case, some decades after the very difficult-tointerpret mention in the third notitia, Theophanes the Confessor evokes Ζάρδαπα (with a rho) twice in his Chronographia, composed between 810 and 815.16 These two occurrences are, however, only paraphrases of Theophylact Simocatta’s passages on the events of 586–587. The last known mention of Zaldapa is from the continuation of Constantine Porphyrogenitus’ De thematibus, which was composed shortly after 998, perhaps by Joseph Genesius.17 Ζέλπα is mentioned in the seventh place among the πόλεις of the province of Moesia Secunda (here, not Scythia), in a list of cities which is more an idealised situation than the reality of the ground, written on the eve of the Byzantine recovery of the northern Balkans by Basil II’s troops in the first quarter of

In 1954, a first inscription allowing the relocation of Abrit(t) us to Razgrad was found by Teofil Ivanov (1918–1999).19 News of the find was immediately disseminated in Bulgaria and Romania. Radu Vulpe (1899–1982) presented, as soon as 1955, a first paper at the Academia Republicii Populare Române (Romanian People’s Republic Academy) reporting the discovery and trying to find a solution for the site studied here, by proposing an association with Zaldapa.20 One had to wait until 1962, however, for the publication of an actual demonstration supporting this hypothesis, in a study by Veselin Beshevliev (1900– 1992).21 Their hypothesis, which is today the most widely accepted – even if no ancient text proving it has been discovered to this day – took some years to be recognised as conclusive. For example, in 1967, the Pauly–Wissowa was still presenting all the proposals about its location as potential solutions (except for the identification with Abrit(t)us).22

The main arguments raised by the defenders of this hypothesis are as follows (cf. Popescu 1988, pp. 84–91 [repr. pp. 131–136]): (1) there would be a change in the titulature of the bishop of Tomis between the middle of the fifth century – episcopus Tomitanus/episcopus Tomitanae civitatis (provinciae Scythiae) – and the first quarter of the sixth century – episcopus provinciae Scythiae metropolitanus (Collectio Avellana, 234.13 [ed. Günther 1898, p. 714])–, which would testify to a change in the civic organisation of the province of Scythia; (2) in 519, a letter from papal ambassadors mentions accusations against the episcopi of a specific provincia, among which Paternus of Tomis (Collectio Avellana, 217.6 [ed. Günther 1898, p. 678]); (3) the phrase sed et universos ad tuam pertinentes ordinationem commoneas, which can be read in a letter of 550 from Pope Vigilius to Valentinianus of Tomis (Concilium oecumenicum Constantinopolitanum (II) anni 553, actiones, 7.9.5 [ed. ACO, IV/1, p. 196]), would testify to an episcopal hierarchy under the latter’s orders; (4) an inscription mentioning bishops, among which one is named Stefanus/Stefanos, on a sixth-century cross found in CallatisMangalia (Popescu 1976, no 91) would, at least, attest the existence of an episcopal see in this city; (5) the ruins of the late fifth- or sixthcentury basilicas in each of the πόλεις of Scythia mentioned in Hierocles’ Synekdemos and the third notitia would be an additional argument for the existence of an episcopal see in each of them at that time. Counter arguments were presented by Dominic Moreau in a paper entitled “To Baptise: An Episcopal Prerogative in Late Antiquity?”, at the symposium Cleric in Church and Society up to 700, held in Warsaw on the 26th and 27th of April 2019 (http://clericsconference.ihuw.pl). The publication of this paper is currently in preparation, but we can already say that the last known bishop of Tomis in the sixth century, Valentinianus, is mentioned at the Council of Constantinople II in 553 as the episcopus Scythiae, together with the episcopus Arelati, Aurelianus. We can thus see that the Fathers of Constantinople did not even consider it necessary to specify the city of Valentinianus, and mention only his province, where the city of another bishop, who is also a metropolitanus, is clearly specified. Cf. Concilium oecumenicum Constantinopolitanum (II) anni 553, actiones, 1.7.12, 7.4.4, and 7.9.1 (ed. ACO, IV/1, pp. 12, 186 and 195). Moreover, if the so-called Justinianic reform of the ecclesiastical administration seems to work for Scythia, this is not the case at all for other provinces, among them Dacia Ripensis, for which neither Hierocles nor the third notitia are describing the episcopal organisation of the time of Justinian. See Gargano, Moreau, in press; meanwhile, see the first doubts about this hypothesis expressed in Moreau 2018, pp. 959–960. 16 Theophanes the Confessor, Chronographia, A.M. 6079 (ed. de Boor 1883, p. 257). 17 Constantine Porphyrogenitus, De thematibus 2.1.59 (ed. Pertusi 1952, p. 86). On the date of the continuation of the De thematibus, traditionally considered as its second book, see ibid., pp. 39–49. 15

The site that is now associated with Zaldapa, and previously with Abrit(t)us – but which was traditionally called ‘Abtaatsko kale’, ‘Kaz kale’, ‘Dorbi(n) kale’, ‘Abtakalesi’ or, more commonly in the scientific literature, ‘Hisar kale’23 – is located slightly south-east of the village of Abrit (Aptaat [from Abdul Ehat] before 1942),24 currently part of the Municipality of Krushari (Armutlii before 1942),25 in the district of Dobrich (Bulgaria). During Antiquity, if the current location is correct, Zaldapa was therefore at the crossing of roads Cf. infra, n. 30. Details and full bibliography are given in Moreau, Carrié 2016, pp. 239–241 and 247. Cf. also Carrié, Moreau 2015, p. 602. 20 Vulpe 1955. Before that, Zaldapa was located in various places by scholars, among others near the current Dobrich in Bulgaria. Cf. PopaLisseanu 1921, pp. 86–93. 21 Beshevliev 1962, pp. 2–4 (German translation: Beševliev 1962, pp. 59–60). As for R. Vulpe, it was only in 1970 that he published an extensive and complete version of his own demonstration. Cf. Vulpe 1970 (Romanian updated translation: 1972). 22 Danoff 1967. 23 Torbatov 2000, pp. 5 and 89; 2003, p. 87. 24 Michev 2005, p. 25. 25 Michev 2005, p. 209. 18 19

38

The Archaeology of the Late Roman City of Zaldapa

Fig. 1. The region of Zaldapa in Roman times according to archaeological maps (B. M’Barek, with D. Moreau).

leading from Durostorum to Marcianopolis and Odessos, in the north–south direction, and from Abrit(t)us to Tropaeum Traiani, Tomis and Callatis, in the east–west direction (fig. 1).

the help of his brother Hermann (1858–1923), between the late 1880s and the very beginning of the 1910s, but with greater regularity between 1897 and 1906.30 They collected new information with increased precision again in 1918, which, however, remained unpublished until recently, so we do not know the exact conditions of their access to the field.31 Their research was then part of wider research on the Late Roman and Early Byzantine fortifications in northeastern Bulgaria. On the site then identified with Abrit(t)us, their work focused initially on the entire fortification, and

The site itself is a fortress situated on a promontory, which is about 1200 m long and 500 m wide.26 Some 2050 m of walls surround an area of about ​​ 25 hectares and exclude the whole north-eastern part of the plateau, which forms a kind of extra muros peninsula.27 The fortification numbers 32 towers and has three main gates, as well as up to five posterns according to Karel Škorpil’s observations, although only one is detectable today.28 Zaldapa is nothing less than the largest Roman-Byzantine fortified town in the hinterland of the provinces of Scythia and Moesia Secunda. As for the intra muros part, it is so densely built according to the satellite photographs (fig. 2 a–c) that it was dubbed the ‘Bulgarian Pompeii’ by the Bulgarian media.29

The results of the Škorpil brothers’ investigations in ‘Abrit(t)us’/ Zaldapa were only partly published in studies about other sites. By far their most detailed published accounts can be read in: Škorpil 1905, pp. 493–499 (together with Uspenskіj 1905, pl. CIX,b and CXI,c). Cf. also Kalinka 1906, cols. 349–358. The Научен архив на Българска академия на науките / Scientific Archives of the Bulgarian Academy of Sciences hosts most of K. Škorpil’s archives, among which is an important file entitled Abrittus (ф. 165 k, оп. 1, а. е. 493), which contains most of the notes and plates he had composed for the publication of a monograph on our site. Part – but not all – of this material was used for the composition of: Torbatov 2000; 2003. Other elements can be found in the Варненски археологически музей / Varna Archaeological Museum (unfinished manuscript on the Christian monuments in Northern Bulgaria, most probably the same that is mentioned in Netzhammer 2005a, p. 107, n. 137) and in the Музей в Националния историко-археологически резерват „Плиска“ / Museum of Pliska National Historical-Archaeological Reserve (excavation notebooks – thanks to Slavi Kirov [Postdoctoral Fellow, ERC project Patrimonium, France / Research Fellow, Център по тракология „Проф. Александър Фол“ / Centre of Thracology ‘Prof. Aleksandar Fol’, Bulgaria] for this information), but they cannot by consulted without special permission. For recently published archives and history of the excavations in Zaldapa, cf. Valchev 2017; Valeriev 2017, pp. 161-166. 31 Most likely in the context of the temporary recovery of Southern Dobrudja by Bulgaria (1916–1918) during World War I. See Torbatov 2000, pp. 22–24 and 91; 2003, p. 93. 30

The history of excavations and publications until 2014 Today, the ruins of the site are less visible than at the time of the first scientific explorations by the father of Bulgarian archaeology, Karel Škorpil. He investigated Zaldapa, with Torbatov 2000, pp. 5 and 89; 2003, p. 87. Torbatov 2000, pp. 5, 10, 89 and 91; 2003, pp. 87 and 89. 28 Torbatov 2000, pp. 15–16 and 92; 2003, pp. 91–92. On the defensive system of Zaldapa, see Torbatov 2000, pp. 8–34 and 90–94; 2003, pp. 89–95. 29 See, for example, https://www.novinite.com/articles/163300/Ancient+ Episcopal+Basilica+Found+in+Bulgaria%27s+Zaldapa+Excavations. 26 27

39

Dominic Moreau, Georgi Atanasov & Nicolas Beaudry

Fig. 2 a. Satellite image of Zaldapa in 2013 (Google Earth). Fig. 2 b. Satellite image of the south part of Zaldapa in 2013 (Google Earth). Fig. 2 c. Site plan of Zaldapa in 2019, drawn from earlier plans, recent satellite images and observations on the field (B. M’Barek, with contributions from D. Moreau, M. Valchev, E. Hobdari and N. Beaudry).

40

The Archaeology of the Late Roman City of Zaldapa later on the surrounding landscape. They also studied the urban fabric, as well as the road network and the buildings.

until 1906. Relying only on the plan of the building, some scholars argue that it is most probably a granary from the tetrarchic/Constantinian period, and more precisely a double horreum or two joint horrea, rather than a basilica civilis/forensis.34

They identified up to six structures,32 but studied – partly at least – only three of them. Of these three buildings, the two which were first discovered are identified, according to the traditional nomenclature, as the Civic Basilica and Christian Basilica No. 2. The first of these, which is located in the centre of the intra muros area, measures approximately 101  ×  18  m (fig. 3).33 It was identified sometime before 1905, but was not actually excavated

As for Christian Basilica No. 2, it is located in the northern part of the fortress (fig. 4).35 This building was also discovered before 1905, and it is currently even less well known than the previous one. The only details that have been published to date are that: (1) it would have been, in

Fig. 3. Plan of the “Civic Basilica”/horreum (B. M’Barek, adapted from K. Skorpil and S. Torbatov).

Fig. 4. Sketch plan of Christian Basilica No. 2 before 2016 (B. M’Barek, adapted from K. Skorpil and S. Torbatov).

Letters G, H, L (= Christian Basilica No. 3?), M (= Christian Basilica No. 2), R (= the so-called Civic Basilica) in Uspenskіj 1905, pl. CXI, c (reprinted in: Tobartov 2000, p. 9, fig, 1; 2003, p. 89, fig. 1), to which we must add Christian Basilica No. 1, discovered near the south-west gate in 1906. Cf. infra. 33 For a full description, see Torbatov 2000, pp. 37–47 and 95–97; 2003, pp. 96–98.

34 Dinchev 2005, pp. 282–284; Rizos 2013, pp. 673–674. A solution to this question will require the resumption of fieldwork and would benefit from comparison with similar sites, such as Tropaeum Traiani (which is a kind of smaller version of Zaldapa), where a building identified for a long time as a basilica civilis/forensis could also be a horreum. Cf. Dinchev 2005, pp. 283; Rizos 2013, pp. 672–673. 35 Torbatov 2000, pp. 58–59 and 99; 2003, p. 102.

32

41

Dominic Moreau, Georgi Atanasov & Nicolas Beaudry

Fig. 5. Plan of Christian Basilica No. 1 before 2016 (B. M’Barek, adapted from K. Skorpil and S. Torbatov).

when the Bulgarian mission at Zaldapa uncovered part of the church to assess its state of conservation (fig. 6).41 This new study of the building led to the discovery of a crypt or altar pit, completely looted at the time of discovery.

some way, connected directly to the north-eastern wall;36 (2) the church itself would be 25 paces (крачки) away from this wall, and would include a 10-pace-long court or room at its entrance – an atrium, or a narthex? – while the northern and southern walls of its nave would be 37 paces long;37 (3) its ground plan would be basilical. Karel Škorpil’s sketch plan of the structure even implies a circus-like basilica, which would be highly unlikely considering the region, but it should be noted that this is only a sketch plan, and that he has not conducted excavations in the apse.38

Following the annexation of Southern Dobrudja by Romania in 1913, field investigations were conducted by Romanian archaeologists, but they remain unpublished and their history is difficult to reconstruct.42 Relying on a vague assertion by Radu Vulpe, Serguey Torbatov has suggested that a team led by George G. Mateescu (1892– 1929) explored the fortification between 1913 and 1915.43 However, the Anuarul Comisiunii monumentelor istorice

The third structure explored by the Škorpils, which is located in the south-western part of Zaldapa, is also a church, commonly known as Christian Basilica No. 1 (fig. 5).39 This early sixth-century three-naved church, with a projecting apse and an inner narthex, was uncovered and excavated in 1906. It measures 27.75 × 16.50 m. The study of that building has yielded a number of elements of architectural sculpture, the base of the ambo, as well as a mosaic that has since disappeared. The looting of all visible elements of the church was already observed in 1922 by Raymund Netzhammer (1862-1945) – who at the time blamed the villagers40 – and was documented again in 2016,

Atanasov et al. 2017b. Irina Achim from the Institutul de arheologie ‘Vasile Pârvan’ of the Academia Română in Bucharest is currently working, with the collaboration of Florian Matei-Popescu (from the same institution), on Romanian excavation archives, in order to clarify the situation. Some results are given here, but the details of this investigation will be published extensively in another study. We must note that the fate of Dobrudja’s ancient heritage during the various conflicts for its possession between Bulgaria and Romania is an extremely complex and delicate topic, which still affects national sensitivities. Even today, the ‘occupation’ of the territory by one party or the other is perceived differently on both sides of the border, and there is very little study of this topic. For an example of recent work, see Boroneanț 2007. 43 Torbatov 2000, pp. 8 and 90; 2003, p. 89; which rely on Vulpe 1938, p. 336: ‘À Abrittus, G. G. Mateescu commença des fouilles systématiques pour la recherche des murs de la cité; on ne les a pas terminées à cause de la guerre.’ The archives of the Muzeul Național de Antichități (MNA) in Bucharest, the precursor of the Institutul de arheologie ‘Vasile Pârvan’, tell us that V. Pârvan had officially requested 10,000 lei from the Ministry of Internal Affairs, for excavations at ‘Kaliakra/Abrit(t)us’, as early as 1914, but there is no trace of any excavation for this same year. Cf. IAVP Archives, MNA Fonds, Volume D20/1914, Folder 2. 41 42

Škorpil 1905, p. 497. Torbatov 2000, pp. 58–59 and 99; 2003, p. 102. 38 Torbatov 2000, p. 59, fig. 34; 2003, p. 102, fig. 14. 39 For a full description, cf. Torbatov 2000, pp. 47–58 and 97–98; 2003, p. 99–101. 40 Netzhammer 2005a, p. 107, n. 137. For the complete account of his visit on the site on the 15th of May 1922, see Netzhammer 2005b, pp. 1116-1117; for his previous visit on the 6th of May 1914, while the walls of Basilica No. 1 were still standing, see pp. 500-501. 36 37

42

The Archaeology of the Late Roman City of Zaldapa

Fig. 6. Apse and crypt of Christian Basilica No. 1 in 2016, looking east (N. Beaudry).

and Buletinul Comisiei monumentelor istorice do not report any activities by G. G. Matesscu in ‘Abrit(t)us’/ Zaldapa during these years (but these publications are not necessarily exhaustive), while V. Pârvan mentions a presence in the field in 1916 instead.44 According to some official papers of the Romanian archaeological authorities of the 1920s, the Germans and their allies looted artefacts in ‘Abrit(t)us’/Zaldapa between 1916 and 1918.45 We also know from published material that some equipment for excavation was stored by the Romanian archaeologists in 1918 – the year the Škorpils returned for the last time to the field – in the nearby village of Dobrin (called ‘Devedji koy’ or Călimaru before 194246), but we do not know whether fieldwork actually took place.47 From this point onwards, everything is even less clear. According to

Dinu Adameșteanu (1913–2004), fieldwork was conducted in 1938 on Christian Basilica No. 1 (already dismantled)48 and on the Western Gate (porta decumana), but it seems rather that the missions of this period were organised in 1939-1940.49 Unfortunately, not much is known of the results of this last phase of the Romanian exploration of Zaldapa. The year 1940 was marked by the recovery of Southern Dobrudja by Bulgaria, but World War II stopped the exploration on the site. Actually, the last excavations prior to 2014 that we know of took place in 1949, prior to the construction of an artificial lake at the western edge of the site. They were led by Milko Mirchev as a rescue mission of the structures of the north-western flank of the hill.50 They revealed a cistern (fig. 7), which had most probably

Pârvan 1924, p. 129, n. 90. Cf. also Vulpe 1928, p. 136; 1935, p. 191; Micu 1938, p. 78l; 1939, p. 65; Valeriev 2017, p. 166-167. The archive of the MNA confirms V. Pâvan’s statements. These documents even present him as the main organiser, especially for intercession with local public authorities, while the then Deputy Director of the Museum, Dimitrie M. Teodorescu (1881–1947), and his assistant, George G. Mateescu, were put in charge of the field. However, G. G. Mateescu was the true supervisor of these excavations, D. M. Teoderescu being occupied with other tasks in July 1916. In his team, we find in particular Paul Nicorescu (1890–1946). V. Pârvan had also planned the participation of Scarlat Lambrino (1891–1964), but there seems to be no trace of his presence on the field. The entry of Romania into the First World War was certainly the cause of the interruption of the mission in August 1916. Cf. IAVP Archives, MNA Fonds, Volume D22/1916, Folder 1916, and Folder Acțiunea intentată de Luigi Sinigallia MNA-ului. 1916-1927. 45 Boroneanț 2007, p. 263. 46   Michev 2005, p. 126. 47 Boroneanț 2007, p. 247. The above-mentioned archives of the MNA also report that some equipment had been stored in Koriten (called Hardalii before 1942; cf. Michev 2005, p. 201) 1916, but it must have disappeared during the War. 44

Cf. supra, n. 40. Adameșteanu 1958; to be read in the light of Micu 1939, and Valeriev 2017, pp. 167-168. The archives of the MNA contain a lot of details about expenses incurred for the resumption of fieldwork in Abrit(t)us/Zaldapa, at S. Lambrino’s initiative and under the patronage of the Museum, during the summer of 1939. His team consisted of Bucur Mitrea (1909–1995) and Dorin Popescu (1904–1987), and, again, the village of ‘Devedji koy’/ Dobrin, known then as Călimaru, was used for the archaeological base. Cf. IAVP Archives, MNA Fonds, Volume D37/1939A, Folder No II – Fonduri, subvenții, acte justificative; and Volume D37/1939B, with Folder Acest borderou cuprinde 88 (optzeci și opt) de file. S. Lambrino. The event was announced with great fanfare in the national newspapers (see for example several articles published between July and August 1939 in the journal Universul). We also learn from these archives that S. Lambrino did not get all the funding requested in 1940 and that he tried, nevertheless, to organize the mission with the local authorities. Cf. IAVP Archives, MNA Fonds, Volume D38/1940, Folder MNA 1940 – V. Săpături, cercetări, descoperiri. Colecțiuni particulare, Com. Mon. Ist., Muzee regionale. 50 Mirchev 1951. 48 49

43

Dominic Moreau, Georgi Atanasov & Nicolas Beaudry was unearthed again by the Регионален исторически музей – Добрич (Regional Historical Museum of Dobrich) in the spring of 2016 to make it visible to visitors, but unfortunately no conservation plan has yet been established (fig. 8).

been seen, at least in part, by K. Škorpil, who locates a spring or a well on his 1905 plan at the very place where the building was found.51 The cistern was composed of two basins, one of which measures approximately 10.80 × 5.30 m (according to recent measurements) and is connected by a tunnel to the fortification. The same basin

After R. Vulpe and V. Beshevliev’s proposals to identify the site with Zaldapa, it was not really mentioned in publications until the very end of the twentieth century, as if the loss of its association with Abrit(t)us had deprived it of all archaeological interest. The situation changed in 2000, when Sergey Torbatov published the first complete monograph on Zaldapa, a work that is currently the most complete reference on the site.52 The idea of this work, based on the earlier publications, on the records of Karel Škorpil, on surface surveys and on aerial photography, had come to him as he was preparing his work on the late Roman defensive system of the province of Scythia, published two years later.53 This important book on Zaldapa placed the site back in the corpus of Bulgarian sites cited as examples in the literature on Roman cities.54 It took fourteen more years, however, for an archaeological team to return to the site. Excavations resumed in Zaldapa in 2014 under the direction of Georgi Atanasov (Регионален исторически музей – Силистра/Silistra) and Valeri Yotov (Археологически музей – Варна/

Fig. 7. Plan of the cistern before 2016 (B. M’Barek, adapted from M. Mirchev and S. Torbatov).

Fig. 8. The cistern in 2016, looking north (N. Beaudry). Torbatov 2000. Torbatov 2002. 54 Torbatov 2003.

51 Uspenskіj 1905, pl. CXI,c (reprinted in: Torbatov 2000, p. 9, fig, 1; 2003, p. 89, fig. 1). More dotted circles on the plan may indicate other springs or wells.

52 53

44

The Archaeology of the Late Roman City of Zaldapa Varna), in collaboration with Ioto Valeriev (then Регионален исторически музей – Добрич/Dobrich, but now Регионален исторически музей – Бургас/Burgas). Their team was completed recently by Albena Milanova (Софийски университет ‘Свети Климент Охридски’ / Sofia University ‘St. Kliment Ohridski’). The results of the 2014 campaign were important, including the discovery of a new Christian basilica.55

identified the newly found basilica as its cathedral and dated it from the end of fifth to the middle of the sixth century.59 After three campaigns, it is now known that it was a richly decorated60 church with a basilical plan, composed of three naves, a semicircular apse in continuity with the central nave, a tripartite narthex and an atrium (fig. 10). Without considering the atrium – as it had not yet been well delimited in 2016 – the church measures 30/34 × 21/22 m.61 The basilica was part of a complex that included a number of structures annexed to or articulated with the accesses to the church; the one to the south-east was identified as the episcopal residence or part of it.62

Christian Basilicas Nos. 3 and 4 By looking at their plan of the site published in 1905, we can understand that the Škorpil brothers had already detected a structure where the new basilica was discovered, without, however, documenting it.56 Also, a sketch plan prepared from satellite imagery by Brahim M’Barek (Éveha and HALMA-UMR 8164 research centre, France) and Dominic Moreau (Université de Lille, HALMA-UMR 8164 research centre, France) for the Bulgarian mission of 2014 showed that there was indeed an important building in that area (fig. 2 c). For all these reasons, and because it was being targeted by looters, the decision was made to conduct excavation in this area, even though the mission was originally intended for the study of Christian Basilica No. 2 and a Late Antique domestic structure located close by (fig. 9).57

The brick pavement of the central nave was highly damaged, but still shows evidence of decorative patterns (fig. 11). Excavation also yielded some column bases, the base of an ambo (fig. 11-12) and the remains of a chancel (fig. 13), most of them preserved in situ. The chancel screen was carried by a strong foundation, 0.65 m to 0.70 m wide by 0.75 m to 0.80 m high; the marble screen itself was an openwork lattice of foliage featuring acanthus leaves and birds (fig. 14), very similar to the chancel screen of the episcopal church of Histria, in present-day Romania.63 The excavation of the sanctuary in 2015 brought to light a rectangular, transversal arrangement of small stones joined by yellow clay (4.00 × 3.10 m), under which a crypt was discovered.64 Its walls, made of small and medium stones held together by a light pink mortar mixed with crushed brick, are approximately 0.40 m to 0.47 m thick and rise up to 1.40 m. This rectangular space is surmounted by a vault made of square bricks, held together by a red mortar. The external dimensions of the structure are 3.80 × 2.50 × 2.40 m (L × W × H); its interior dimensions are 2.80 × 1.95 × 2.04 m (fig. 15). The vault is pierced in its center by a pit about 1.40 m in diameter, possibly where the altar stood.

Standing by V. Pârvan’s and R. Vulpe’s interpretation of Zaldapa as an episcopal see58, the Bulgarian team

The crypt is accessible from the south, from a small staircase of seven steps. The staircase was discovered sealed by an imposing rectangular stone (0.95 × 1.39 × 0.12 m). The interior of the crypt is covered with a pale yellow mortar. A cross is inscribed in the centre of the northern wall and another, similar, but in a poorer condition, is in the centre of the eastern wall. The floor is paved with bricks. The interior space was discovered filled with soil and numerous pieces of marble, including four column bases, complete or broken Corinthian and Ionic capitals, several fragments of marble or limestone columns, as well as Fig. 9. Plan of the domestic structure between the northeastern gate and Christian Basilica No. 2, 2016 (B. M’Barek).

Atanasov et al. 2015, pp. 422–423; 2016, p. 460; 2017a, p. 124. In particular, by the use of architectural elements taken from earlier monuments. See, for example, Dimitrov 2017. 61 Earlier reports give 34 × 22 m. Cf. Atanasov et al. 2015, p. 422. More recent ones give 51 × 21 m, taking into account the atrium – delimited in 2017 – which measures 21 × 22.5/27 m. Cf. Atanasov et al. 2017a, p. 123; 2017b, p. 296; 2018, pp. 246 and 248. 62 Atanasov et al. 2016, pp. 460 and 463; 2017a, p. 124. 63 Bounegru, Iaţcu 2007, pp. 57–66 (together with Suceveanu 2007, pl. XXXI–XLII); Atanasov et al. 2017a, pp. 124–125. 64 Atanasov et al. 2016; 2017a, pp. 125–126. 59 60

Atanasov et al. 2015. Uspenskіj 1905, pl. CXI,c (reprinted in: Torbatov 2000, p. 9, fig, 1; 2003, p. 89, fig. 1). Some recently published plans and pictures from K. Škorpils’ archives show that he knew it was a Christian basilica: see Valchev 2017; Valeriev 2017. 57 Atanasov et al. 2015, pp. 422–423. 58 Cf. supra, n. 14. 55 56

45

Dominic Moreau, Georgi Atanasov & Nicolas Beaudry

Fig. 10. Plan of Christian Basilica No. 3, 2016 (B. M’Barek, adapted from G. Atanasov, V. Yotov and I. Valeriev).

Fig. 11. Plan of the ambo and brick pavement of Christian Basilica No. 3, 2016 (E. Hobdari).

pieces of the chancel, altar table, and ambo. This space is thought to have been filled by devotees to avoid the desecration of the crypt and sacred furniture after the destruction of the church, probably in the early seventh century.65 The relics would have been moved to another 65

location, but three carpal bones of a human hand were found in the debris. Comparing this crypt to the known crypts and monumental tombs of the Lower Danube and beyond, the archaeologists concluded that its architecture is closer to that of a number of fourth- and fifth-century vaulted tombs, like those in

Atanasov et al. 2017a, pp. 126–127.

46

The Archaeology of the Late Roman City of Zaldapa

Fig. 12. Base of the ambo of Christian Basilica No. 3, 2015 (N. Beaudry).

Fig. 13. Base of the chancel of Christian Basilica No. 3, 2016, detail (N. Beaudry).

47

Dominic Moreau, Georgi Atanasov & Nicolas Beaudry potential. It can certainly shed light on the processes of militarisation and of Christianisation of the cities of the Lower Danube. The issue of Christianisation is particularly complex in the province of Scythia, where the Christian communities may have been organised along different lines than those of the traditional model in which a city necessarily corresponded to an episcopal see. This potential of Zaldapa is clearly perceptible in the results obtained by Georgi Atanasov, Valeri Yotov and Ioto Valeriev since 2014. In 2015 and in 2016, they invited an foreign team directed by Dominic Moreau and Nicolas Beaudry (UQAR, Canada) with the objective of putting together an international project on Zaldapa.70 Thus far, the contributions of the foreign team included an updated plan of the city and of its defensive system based on satellite imagery and field walking, as well as drawn and photographic records of Christian Basilica No. 1, Christian Basilica No. 3 and the cistern. Above all, its visits to Zaldapa allowed an assessment of the archaeological interest and potential of the site, and the design of a joint project to investigate the city’s urban fabric. An international research project at Zaldapa opens new and exciting perspectives on the changing urban landscape of a Late Roman fortress that otherwise remains barely explored. The imposing character of its defensive structures suggests a strong military presence which has yet to be studied; it is even possible that Flavius Vitalianus himself took refuge in his home town, which was one of the main strongholds of the hinterland of the Lower Danube, when he withdrew to northern Thrace after a failed attempt to lay siege to Constantinople in 515 (cf. supra). The religious importance of Zaldapa is suggested by the density, size and decor of its Christian monuments; it is also suggested by sources that link Vitalianus’ revolt to a Christological formula that was central to sixth-century theological debates.71 The historical significance of the army and of the Church in Zaldapa give a measure of the archaeological potential of their material footprints, and the site offers excellent conditions for investigating their interrelated effects on the urban fabric.

Fig. 14. Fragments of the openwork chancel screen of Christian Basilica No. 3, 2015 (D. Moreau).

Axiopolis and Durostorum (fig. 16 a–b), than to that of crypts66. In 2016, the study of the structure and foundations of the crypt led to the discovery of the apse of an earlier Christian basilica, located under Basilica No. 3 (fig. 17)67 and tentatively dated to the fourth century.68 The work then focused on this new building, named Christian Basilica No. 4, and led to the discovery of its own, even larger crypt, composed of two rooms (south room 2.80 × 1.35 m; north room 2.80 × 2.40 m; external dimensions 6.60 × 4.60 m). The 2016 season also allowed the discovery of a necropolis south-west of the fortress. Informed by the villagers of massive looting at its location, the Bulgarian team informed the regional archaeological authorities, who organised rescue excavations. The necropolis was dated from the mid-sixth to the early seventh century and may point to an Alan presence.69

The joint archaeological project will thus seek to deliver a dynamic portrait of Zaldapa’s urban landscape, of its economy, and of its religious and military environments during Late Antiquity. The exceptional preservation of the site, which is thought to have been completely deserted after Antiquity, is expected to yield a rich archaeological

Conclusion: archaeological perspectives

In addition to their directors, the visiting foreign team included: Pascale Chevalier (lecturer, Université Clermont Auvergne / ArTeHiS– UMR 6298 research centre, France – 2015-2016); Elio Hobdari (Instituti i Arkeologjisë, Albania –2016); Adrien Leblond (doctoral student, Université de Lille, France – 2015-2016); Brahim M’Barek (cf. supra – 2016); Julia Reveret (doctoral student, Université Clermont Auvergne and Université de Fribourg, France and Switzerland – 2015-2016); Nadia Saint-Luc (lecturer, Lycée Claude Fauriel and École nationale d’architecture de Clermont-Ferrand, France – 2015-2016). 71 Moreau 2017. 70

In the light of the short assessment drawn up here, one can see that Zaldapa offers an exceptional archaeological Atanasov et al. 2017a, pp. 127–131. Atanasov et al. 2017b. 68 Atanasov, Valeriev 2018. 69 Ivanov et al. 2017. 66 67

48

The Archaeology of the Late Roman City of Zaldapa

Fig. 15. The crypt of Christian Basilica No. 3, 2015, looking north (G. Atanasov and I. Yotov).

archive in an area where well-documented sites remain rare. The project will allow an assessment of the combined effects of militarisation and Christianisation on the urban forms and functions of a city of the Danubian hinterland, which has been less explored than the front line; it will also allow a critical assessment of models in which the army would have been the main vehicle of Christianity along the limes. The study of this fortified, possible episcopal see of the hinterland will thus offer an original parallel to that of the great forts of the Danube and usefully complement the documented sites of the hinterland.

on the north wall of the crypt of Basilica 4. In 2018, the exploration focused on the walls and floor of Basilica No. 4, as well as on the decor and passageways of Basilica No. 3, specifically around its exterior walls.73 In parallel with the work of the Bulgarian mission, a new international project was launched in the wake of the exploratory missions of 2015 and 2016. Cooperation agreements were signed to this end in 2017 and 2018 between the Université de Lille (France), the Université du Québec à Rimouski (Canada), Софийски университет ‘Свети Климент Охридски’ (Sofia University ‘St Kliment Ohridski’, Bulgaria) and the Регионален исторически музей – Добрич (Regional Historical Museum of Dobrich, Bulgaria). The Zaldapa International Archaeological Project is codirected by G. Atanasov and N. Beaudry, with the collaboration of D. Moreau, A. Milanova and I. Valeriev.74 It is funded mainly by the Social Sciences and Humanities Research Council of Canada and the Danubius Project of the

Appendix on seasons 2017 to 2019 Since this paper was read at the 23rd ICBS in Belgrade in September 2016, research in Zaldapa has developed to such an extent that it was thought useful to briefly report on the work of the Bulgarian team and of what has become the Zaldapa International Archaeological Project. The Bulgarian mission has carried on its study of Christian Basilicas Nos. 3 and 4 in 2017 and 2018, and a new campaign was scheduled for September 2019. The 2017 campaign was marked by the end of the excavation and documentation of the atrium of Basilica No. 3, and of the crypt of Basilica No. 4.72 A burial was found under the atrium of Basilica 3 and traces of wall painting were found 72

Atanasov, Valeriev 2019. Other participants (excluding external service providers) were: Anthony Carneaux (Master’s student, Université de Lille, France – 2019); Jérémy Gribaut (Master’s student, Université de Lille, France / Master’s student, UQAR, Canada – 2018-2019); Elio Hobdari (cf. supra, n. 70 – 2018); Slavi Kirov (cf. supra, n. 30 – 2018-2019); Lyubomir Malinov (Master’s student, Université de Montpellier, France – 2019); Brahim M’Barek (cf. supra – 2018-2019); Aleksandar Ivanov (PhD student, Софийски университет ‘Свети Климент Охридски’ / Sofia University ‘St. Kliment Ohridski’, Bulgaria – 2018-2019); David L. Tremblay (undergraduate student, UQAR, Canada – 2018). 73 74

Atanasov, Valeriev 2018.

49

Dominic Moreau, Georgi Atanasov & Nicolas Beaudry

Fig. 16a. Vaulted tomb in Axiopolis (archives of the Muzeul de istorie naţională şi arheologie Constanța, printed with permission).

Fig. 16b. Vaulted tomb in Durostorum (R. Dimitrov ; Atanasov 2014).

50

The Archaeology of the Late Roman City of Zaldapa

Fig. 17. The crypt of Christian Basilica No. 4, 2017, looking east (N. Beaudry).

Université de Lille,75 with contributions from the HALMA–UMR 8164 research centre. The International Project focuses, in its first phase, on the north end of the site, where the city wall forms a salient. A test trench opened in 2014 by the Bulgarian team suggests a dense habitat (cf. supra) but since looters’ trenches do not show on satellite imagery, this area could not be planned unlike most of the intra muros city. In July 2018 and 2019, excavations were simultaneously conducted on Christian Basilica No. 2 and on the north-eastern gate. The 2018 season allowed the completion of the sketch plans drawn by K. Škorpil of both structures: Basilica No. 2 proved to be a three-naved basilica with a polygonal apse, while the west, U-shaped of the gate (Tower No. 2) is attached to the rampart at a different angle than was previously thought (fig. 18).76 More of the plan of the basilica was exposed in 2019 (fig. 19): its narthex and a possible atrium were identified and the exploration of its sanctuary continued, while different phases of the north-eastern gate and its passageway were identified. Fig. 18. Plan of Tower 2, 2018 (B. M’Barek).

In addition, the International Project has contributed over the last three years to the publication outside Bulgaria of the site and of ongoing fieldwork at Zaldapa. The

collaboration of the Bulgarian and foreign teams has also led to a joint publication of inscriptions found during the excavations of Christian Basilicas Nos. 3 and 4.77

75 The DANUBIUS Project intends to study the Christianisation of the Lower Danube, between the third and the eight centuries AD. It is mainly co-funded for 2018–2021 by the Agence nationale de la Recherche (ANR) and the I-SITE ULNE Foundation. For more information see https://danubius.huma-num.fr. 76 Atanasov et al. 2018.

Dana et al. 2017 (erratum: p. 158 – …Abritus (relocalisée à Razgrad depuis 1954)…); 2019 (erratum: p. 74 – …in 2015, among the spolia,…).

77

51

Dominic Moreau, Georgi Atanasov & Nicolas Beaudry

Fig. 19. Plan of Christian Basilica No. 2, 2019 (N. Beaudry).

Bibliography

Atanasov G., Valeriev I., 2018. ‘Археологическо проучване на базилики No 3 и 4 в Залдапа, община Крушари, област Добрич’, in Археологически открития и разкопки през 2017 г., pp. 246–248.

Abbreviations ACO = Schwartz E., Straub J., Schieffer R., Riedinger R., 1914–84, Acta conciliorum oecumenicorum [1st series], Strasbourg/Berlin/Leipzig.

Atanasov G., Valeriev I., 2019. ‘Раннохристиянски базилики No 3 и 4 в Залдапа’, in Археологически открития и разкопки през 2018 г. – Archaeological Discoveries and Excavations in 2018, pp. 262–264.

AD = Analele Dobrogei. AE = Année épigraphique.

Atanasov G., Yotov V., Valeriev I., 2015. ‘Античен град Залдапа, община Крушари, област Добрич. Раннохристиянска базилика’, in Археологически открития и разкопки през 2014 г., pp. 421–423.

CCET = Gočeva Z., Oppermann M., Hampartumian N., Cermanović-Kuzmanović A. (ed.), 1979–84. Corpus cultus equitis Thracii (CCET), Leiden (Études préliminaires aux religions orientales dans l’Empire romain, 74).

Atanasov G., Yotov V., Valeriev I., 2016. ‘Античен град Залдапа, община Крушари, обвласт Добрич. Раннохристиянска базилика’, in Археологически открития и разкопки през 2015 г., pp. 460–463.

CMRED = Tudor D. (ed.), 1969–76. Corpus monumentorum religionis equitum Danuvinorum (CMRED), Leiden (Études préliminaires aux religions orientales dans l’Empire romain, 13).

Atanasov G., Valeriev I., Yotov V., 2017a. ‘The crypt in the sanctuary of the Basilica No 3 at the ancient city of Zaldapa (province of Scythia)’, in Ниш и Византијa – Niš and Byzantium, 15, pp. 123–132.

IGBulg = Mihaylov G. (ed.), 1958–97. Inscriptiones Græcae in Bulgaria repertae – Гръцките надписи намерени в България, Sofia.

Atanasov G., Yotov V., Valeriev I., 2017b. ‘Античен град Залдапа, община Крушари, област Добрич. Раннохристиянска базилика’, in Археологически открития и разкопки през 2016 г., pp. 296–297.

PLRE = Jones A.H.M., Martindale J., Morris, 1971–1992. The Prosopography of the Later Roman Empire, Cambridge. SCIVA = Studii şi cercetări de istorie veche [şi arheologie] – Éudes et recherches d’histoire ancienne [et d’archéologie].

Atanasov G., Beaudry N., Milanova A., Moreau D., Valeriev I., 2019. ‘Базилика Nо 2 и централна северна порта на Залдапа’, in Археологически открития и разкопки през 2018 г. – Archaeological Discoveries and Excavations in 2018, pp. 260–262.

Adameșteanu D., 1958. ‘Abrittus’, in R. B. Bandinelli, G. Becatti (eds.), Enciclopedia dell’arte antica classica e orientale, I, Rome, p. 6.

Beševliev V., 1962. ‘Zur Geographie Nordost-Bulgariens in der Spätantike und im Mittelalter’, in Linguistique balkanique, 4, pp. 57–80 (repr. in V. Beševliev, 1978. Bulgarisch-Byzantinische Aufsaetze, Burlington/

Atansov G., 2014. The Roman Tomb in DurostorumSilistra, Silistra. 52

The Archaeology of the Late Roman City of Zaldapa Singapore/Sydney [Collected Studies Series, 80)], no XXIII).

Professoris Ludmili Getov, Sofia (Studia Archaeologica Universitatis Serdicensis. Suppl., 4), pp. 277–295.

Beshevliev V., 1962. ‘Из късноантичната и средновековната география на Североизточна България’, in Известия на Археологическия институт – Bulletin de l’Institut archéologique, 25, pp. 1–18.

Gargano I, Moreau D., in press. ‘Some remarks on the Christian topography of Dacia Ripensis (second half of the 5th century–6th century)’, in I. Topalilov, Z. Gerdzhikova (eds.), Създаването на късноантичния свят на Балканите. Дискусионен семинар, 8–10 ноември 2018 г., София – Creation of the Late Antique World on the Balkans. Discussion Seminar, 8–10 November 2018, Sofia, Sofia.

Boroneanț A., 2007. ‘Vasile Pârvan şi patrimoniul arheologic dobrogean în timpul Primului Război Mondial. Documente în Arhiva Muzeului Naţional de Antichităţi’, in SCIVA, 58/3–4, pp. 229–264.

Günther O. (ed.), 1898. Epistulae imperatorum pontificum aliorum inde ab a. CCCLXVII usque ad a. DLIII datae Avellana quae dicitur collection, II, Vienna/ Prague/Leipzig (Corpus scriptorium ecclesiasticorum Latinorum, 35/2).

Bounegru O., Iaţcu I., 2007. ‘Éléments de décoration intérieure’, in Suceveanu 2007, pp. 57–72, Cameron A., 1985. Procopius and the Sixth Century, London/Berkeley/Los Angeles (Classical Life and Letters / The Transformation of the Classical Heritage, 10).

Honigmann E. (ed.), 1939. Le Synekdèmos d’Hiéroklès et l’opuscule géographique de Georges de Chypre, Bruxelles (Corpus Bruxellense historiae Byzantinae. Forma Imperii Byzantini, 1).

Carrié J.-P., Moreau D., 2015. ‘The archaeology of the Roman town of Abritus: the status quaestionis in 2012’, in L. Vagalinski, N. Sharankov (eds.), Limes XXII. Proceedings of the 22nd International Congress of Roman Frontier Studies, Ruse, Bulgaria, September 2012, Sofia (Bulletin of the National Archaeological Institute, 42), pp. 601–610.

Ivanov B., Rusev N., Vardanov V., Dobrev D., 2017. ‘Късноантичен некропол до крепостта Залдапа’, in Археологически открития и разкопки през 2016 г., pp. 470–474. Kalinka E. (ed.), 1906. Antike Denkmäler in Bulgarien, Vienna (Schriften der Balkankommission. Antiquarische Abteilung, 4).

Dana D., Valeriev I., Moreau D., 2017. ‘Un théonyme et des noms thraces nouveaux dans une dédicace grecque découverte à Zaldapa (Mésie inférieure)’, in Zeitschrift für Papyrologie und Epigraphik, 202, pp. 158–162.

Mariev S. (ed. and transl.), 2008. Ioannis Antiocheni fragmenta quae supersunt omnia, Berlin/New York (Corpus Fontium Historiae Byzantinae, Series Berolinensis, 47).

Dana D., Moreau D., Kirov S., Valeriev I., 2019. ‘A new Greek dedication from the sanctuary of Telerig among the spolia at Zaldapa’, in Archaeologia Bulgarica, 23/2, pp. 71–78.

Michev N., 2005. Речник на имената и статута на населените места в България 1878–2004, Sofia. Micu I., 1938. ‘Cetăți, muzee și colecții de antichități din sudul Dobrogei’, in AD, 16, pp. 67–84.

Danoff C. M., 1967. ‘Zaldapa’, in K. Ziegler (ed.), Paulys Realencyclopädie der classischen Altertumswissenschaft. Neue Bearbeitung. Zweiter Reihe, IX/2, Stuttgart, cols. 2296–2298.

Micu I., 1939. ‘Abrittus’, in Pontica, 1/3, pp. 65–66. Mirchev, M. 1951. ‘Водният резервоар и каптаж при крепостта Абрит’, in Известия на Варненското археологическото дружество – Bulletin de la Société archéologique à Varna, 8, pp. 99–103.

Darrouzès J. (ed.), 1981. Notitiae episcopatuum Ecclesiae Constantinopolitanae, Paris (Géographie ecclésiastique de l’Empire byzantin, 1). de Boor C. (ed.), 1883. Theophanis Chronographia, I, Leipzig.

Moreau D., 2017. ‘Les moines scythes néo-chalcédoniens (de Zaldapa?). Étude préliminaire à une prosopographie chrétienne du Diocèse des Thraces’, in Yotov, Gancheva 2017, pp. 287–202.

de Boor C. (ed.), 1905. Excerpta historica iussu Imp. Constantini Porphyrogeniti, III, Berlin. de Boor C. (ed.), 1972. Theophylacti Simocattae Historiae, rev. ed. by P. Wirth, Stuttgart (Bibliotheca scriptorum Graecorum et Romanorum Teubneriana).

Moreau D., 2018. ‘The episcopal sees on the Roman ‘limes’ up to AD 610: a prosopographical investigation’, in C. S. Sommer, S. Matešić (eds.), Limes XXIII. Proceedings of the 23rd International Congress of Roman Frontier Ingolstadt 2015 – Akten des 23. Internationalen Limeskongresses in Ingolstadt 2015, Mainz (Beiträge zum Welterbe Limes. Sonderbande, 4), pp. 957–967.

Dimitrov Z., 2017. ‘Коринтски капител от Залдапа: уникален модел на ордената архитектура в Долна Мизия през принципата’, in Yotov, Gancheva 2017, pp. 177–186. Dinchev V., 2005. ‘Късноантични обществени складове от Thracia и Dacia’, in T. Stoyanov, S. Angelova, I. Lozanov (eds.), Stephanos Archaeologicos in honorem

Moreau D., Carrié J.-P., 2016. ‘L’agglomération romaine d’Abritus (Mésie inférieure/Mésie seconde) : sources 53

Dominic Moreau, Georgi Atanasov & Nicolas Beaudry Škorpil K. V., 1905. ‘Нѣкоторыя изъ дорогъ восточной Болгаріи’, in Uspenskіj 1905, pp. 443–502.

textuelles et bilan archéologique’, in C. Freu, S. Janniard, A. Ripoll (eds.), Libera curiositas. Mélanges d’histoire et d’Antiquité tardive offerts à Jean-Michel Carrié, Turnhout (Bibliothèque de l’Antiquité tardive, 31), pp. 229–256.

Slokoska L., Ivanov R., Dinchev V. (eds.), 2002. The Roman and Late Roman City: The International Conference (Veliko Turnovo, 26–30 July 2000) – Римският и късноантичният град. Международна научна конференция (Велико Търново, 26–30 юли 2000), Sofia.

Netzhammer R., 2005a. Antichităţile creştine din Dobrogea, ed. A. Barnea, transl. G. Guţu, Bucharest. Netzhammer R., 2005b. Episcop în România. Într-o epocă a conflictelor naţionale şi religioase, 2nd rev. ed. N. Netzhammer, transl. G. Guţu, Bucharest.

Suceveanu A. (dir.), 2007. Histria, XIII, Bucharest. Tomaschek W., 1894. ‘Die alten Thraker. Eine ethnologische Untersuchung. II. Die Sprachreste. 2. Hälfte. Personen und Ortsnamen’, in Sitzungsberichte der PhilosophischHistorischen Classe der Kaiserlichen Akademie der Wissenschaften, 131, no I (repr. in W. Tomaschek, 1975. Die alten Thraker. Eine ethnologische Untersuchung, Osnabrück, pp. 201–303; W. Tomaschek, 1980. Die alten Thraker. Eine ethnologische Untersuchung, Vienna, no 3).

Pârvan V., 1912. Cetatea Tropaeum. Consideraţii istorice, Bucharest. Pârvan V., 1924. ‘Nuove considerazioni sul vescovato della Scizia minore’, in Atti della Pontificia Accademia romana di archeologia (Serie III). Rendiconti, 2, pp. 117–135. Pertusi A. (ed.), 1952. Costantino Porfirogenito, De thematibus, Vatican City (Studi e testi, 160).

Torbatov S., 2000. Късноантичният град Залдапа – The Late Antique City of Zaldapa, Sofia.

Petrović P. (ed.), 1996. Римски лимес на средњем и доњем дунаву – Roman Limes on the Middle and Lower Danube, Belgrade (Ђердапске свеске. Посебна издања – Cahiers des Portes de Fer. Monographies, 2).

Torbatov S., 2001. ‘Антично култово средище край село Телериг, Добричко (предварително съобщение)’, in Научни съобщения на Съюза на учените в България. Клон Добрич, 3, pp. 141–147.

Popa-Lisseanu G., 1921. Cetăţi şi oraşe greco-romane în noul teritoriu al Dobrogei, 2nd ed., Bucharest (Casa Școalelor. Biblioteca Secundară, 4).

Torbatov S., 2002. Укрепителната система на провинция Скития (края на III–VII в.) – The Defence System of the Late Roman Province of Scythia (the End of the 3rd–the 7th Century A.D.), Veliko Tarnovo.

Popescu E. (ed.), 1976. Inscriptiones intra fines Dacoromaniae repertae Graecae et Latinae anno CCLXXXIV recentiores – Inscripţiile greceşti şi latine din secolele IV–XIII descoperite în România, Bucharest (Inscriptiones Daciae et Scythiae Minoris antiquae – Inscripţiile antice din Dacia şi Scythia Minor).

Torbatov S., 2003. ‘Залдапа (Zaldapa)’, in R. Ivanov (ed.), Римски и ранновизантийски селища в България – Roman and Early Byzantine Settelements in Bulgaria, II, Sofia, pp. 87–109.

Popescu E., 1988. “Die kirchliche Organisation der Provinz Scythia Minor vom vierten bis ins sechste Jahrhundert”, in Jahrbuch der Österreichischen Byzantinistik, 38, pp. 75–94 (repr. E. Popescu, 1994. Christianitas Daco-Romana. Florigelium studiorum, Bucharest, pp. 124-138).

Torbatov S., 2005. ‘ Ἥρος Ἥφαιστος (предварителни данни за античния култов център край Телериг, Южна Добруджа)’, in T. Stoyanov, M. Tonkova, Ch. Preshlenov, Ch. Popov (eds.), Heros Hephaistos. Studia in honorem Liubae Ognenova-Marinova, Veliko Tarnovo, pp. 80–91.

Poulter A. G. (ed.), 2007a. The Transition to Late Antiquity on the Danube and Beyond, Oxford (Proceedings of the British Academy, 141).

Uspenskіj F. I. (dir.), 1905. Матеріалы для болгарскихъ древностей. Абоба-Плиска = Извѣстія Русскаго Археологическаго Института въ Константинополѣ – Bulletin de l’Institut archéologique russe de Constantinople, 10.

Poulter A. G., 2007b. ‘The transition to Late Antiquity’, in Poulter 2007a, pp. 1–50.

Vagalinski L. (ed.), 2007. The Lower Danube in Antiquity (VI c. BC–VI c. AD): International Archaeological Conference Bulgaria-Tutrakan, 6–7.10.2005, Sofia.

Rizos E., 2013. ‘Centres of the late Roman military supply network in the Balkans: a survey of horrea’, in Jahrbuch des Römische-Germanischen Zentralmuseums Mainz, 60, pp. 659–693.

Vagalinski L., Sharankov N., Torbatov S. (eds.), 2012. The Lower Danube Roman Limes (1st–6th c. AD), Sofia.

Roberto U. (ed. and transl.), 2006. Ioannis Antiocheni Fragmenta ex Historia chronica, Berlin/New York (Texte und Untersuchungen zur Geschichte der altchristlichen Literatur. Archiv fur die Ausgabe der Griechischen Christilichen Schriftsteller der ersten Jahrhunderte, 154).

Valchev M., 2017. ‘Първите планове на Залдапа в архива на Карел Шкорпил’, in Yotov, Gancheva 2017, pp. 187–202.

54

The Archaeology of the Late Roman City of Zaldapa Valeriev I., 2015. ‘Оловен печат на Георги Теодорокан, протокуропалат, от района на късноантичния град Залдапа’, in Добруджа. Сборник, 30, pp. 467–468. Valeriev I., 2017. ‘От архивите – за античния град Залдапа (тогава все още “Абтаат кале/Абритус”)’, in Yotov, Gancheva 2017, pp. 187–202. Veh O. (ed. and transl.), 1977. Prokop, Bauten, Munich (Tusculum-Bücherei). von Bülow G., Milčeva A. (eds.), 1999. Der limes an der unteren Donau von Diokletian bis Heraklios. Vorträge der internationalen Konferenz, Svištov, Bulgarien (1.–5. September 1998), Sofia. Vulpe R., 1928. ‘Activitatea archeologică în Dobrogea în cei 50 de ani de stăpânire românească’, in AD, 9/1, pp. 117–144. Vulpe R., 1935. ‘Noutăţi archeologice dobrogene’, in AD, 16, pp. 185–192. Vulpe R., 1938. ‘Histoire ancienne de la Dobroudja’, in La Dobroudja, Bucharest (Connaissance de la terre et de la pensée roumaines, 4), pp. 35–454. Vulpe R., 1955. ‘Comunicări privitoare la arheologie şi istorie veche: 11 noiembrie 1955. R. Vulpe: “Graniţa de sud a provinciei Scythia şi localizarea oraşelor Abrittus şi Zaldapa”’, in SCIVA, 6/3–4, p. 939. Vulpe R., 1970. ‘La limite méridionale de la province romaine de Scythie’, in V. Beševliev, V. TăpkovaZaimova (eds.), Recherches de géographie historique, Sofia (Studia Balcanica, 1), pp. 33–47. Vulpe R., 1972. ‘Limita meridională a provinciei romane Scythia’, in Pontica, 5, pp. 205–221. Yotov V., Gancheva S. (eds.), Сборник в чест на 60 години проф. дин Георги Атанасов = Добруджа. Сборник, 32. Zahariade M. (ed.), 1998. The Roman Frontier at the Lower Danube, 4th–6th Centuries: The Second International Symposium (Murighiol/Halmyris, 18–24 August 1996), Bucharest (Studia Danubiana. Pars Romaniae. Series Symposia, 1).

55

5 Archaeological Investigations at Golemo Gradište, Konjuh, 2012–2016 Carolyn S. Snively Gettysburg College (USA)

Goran Sanev Археолошки Музеј на Република Северна Македонија / Archaeological Museum of the Republic of North Macedonia Excavation of the anonymous Late Antique city located at Golemo Gradište, Konjuh, North Macedonia, by a Macedonian–American team, continued during the 2012–2016 seasons. The focus of attention has been on the lower city on the northern terrace of the site, and more specifically on the Episcopal Basilica and the Northern Residence, which were centrally located on the terrace and which were shown in 2016 to have formed a single complex in a late phase of their sixth-century existence. Annexes on the south, west and north sides of the basilica included a baptistery, several rooms with benches and two possible rooms for deposition of offerings. Excavation of the entire Northern Residence has shown that it was a peristyle house of significant size with at least two major phases of construction. The west wall of the residence both connected it with the basilica and cut off the original western part of the building. A catastrophic event in the late sixth century damaged or destroyed both basilica and residence and caused the inhabitants of the terrace to flee to the acropolis of the site for refuge. La fouille par une équipe américano-macédonienne de la ville tardo-antique anonyme située à Golemo Gradište, en Macédoine, s’est poursuivie au cours des saisons 2012-2016. Le centre d’intérêt était la ville basse située sur la terrasse septentrionale du site et, plus précisément, la basilique épiscopale, de même que la résidence nord, situées au centre de cette terrasse, dont l’on a constaté, en 2016 qu’elles constituaient un unique complexe vers la fin de leur existence au VIe siècle. Des annexes sur les côtés sud, ouest et nord de la basilique comprenaient un baptistère, plusieurs salles munies de bancs et deux salles destinées peut-être au dépôt des offrandes. La fouille complète de la résidence nord a montré qu’elle était une maison à péristyle de grandes dimensions, qui a connu au moins deux phases de construction. Le mur ouest de la résidence la liait à la basilique tout en la coupant de la partie occidentale du premier état. Un événement catastrophique survenu vers la fin du VIe siècle a endommagé ou détruit la basilique et la résidence, poussant les habitants de la terrasse à chercher refuge sur l’acropole. Die Ausgrabungen der anonymen, spätantiken Stadt von Golemo Gradište, Konjuh in Mazedonien wurden durch ein mazedonisch-amerikanisches Team von 2012 bis 2016 fortgesetzt. Sie konzentrierten sich auf die untere Stadt an der Nordterrasse des Fundplatzes, insbesondere auf die Bischofsbasilika und auf die nördliche Residenz. Letztere befand sich zentral auf der Terrasse und wie die Ergebnisse des Jahres 2016 zeigten, bildete sie einen eigenen Komplex während des späten 6. Jhs. Anbauten an der Süd-, West- und Nordseite der Basilika dienten als Baptisterium, einige besaßen Priesterbänke und zwei Räume nutzte man wohl für die Deponierung von Opfergaben. Ausgrabungen in der gesamten nördlichen Residenz zeigten, dass es sich um einen großen Peristylbau mit zwei Hauptbauphasen handelt. Die Westmauer der Residenz wurde mit der Basilika verbunden und sie trennte die ursprüngliche Westseite des Gebäudes ab. Eine Katastrophe während des späten 6. Jhs. zerstörte die Kirche und die Residenz und die Einwohner der Terrasse flohen und suchten an der Akropolis des Ortes Schutz.

57

Carolyn S. Snively & Goran Sanev Gli scavi dell’anonima città antica, situata a Golemo Gradište, Konjuh, in Macedonia, sono stati intrapresi da un gruppo archeologico macedone-americano e si sono svolti con continuità durante il periodo 2012 – 2016. L’attenzione si è concentrata sulla città inferiore, sulla terrazza settentrionale del sito, e in particolar modo sulla Basilica Episcopale e sulla Residenza Settentrionale. Entrambe erano posizionate al centro della terrazza e nel 2016 è stato dimostrato che queste formavano un complesso unico nella fase avanzata della loro esistenza nel VI secolo. Sui lati sud, ovest e nord della basilica si trovavano edifici annessi come un battistero, numerose aule con dei bancali e due possibili ambienti per il deposito delle offerte. Gli scavi dell’intera Residenza Settentrionale dimostrano che questa era una casa con peristilio di grandi dimensioni, con almeno due fasi principali di costruzione. Il muro occidentale della residenza, tuttora esistente, la connetteva alla basilica e tagliava la parte originale occidentale della residenza stessa. Negli ultimi anni del VI secolo un evento catastrofico danneggiò o distrusse completamente sia la basilica che la residenza, e obbligò gli abitanti della residenza a fuggire sull’acropoli del sito per trovare rifugio. Since a report on the progress of our investigations was given at the 22nd International Congress of Byzantine Studies in Sofia in 2011, a Macedonian–American project has continued to excavate at the archaeological site of Golemo Gradište at Konjuh, county of Kratovo, North Macedonia. The anonymous Late Antique city that once stood there consisted of three parts within the fortification walls: (1) an east–west ridge, separately fortified as the acropolis of the city; (2) an area south of the acropolis, lying between the acropolis ridge and a lower bedrock ridge known as Malo Gradište; and (3) the northern terrace where the lower town once stood, located between the foot of the acropolis and the Kriva River (fig. 1). Investigation of the site has been the main focus of a long-term international project sponsored by Gettysburg College, in Gettysburg, Pennsylvania, USA, and the Archaeological Department of the National Archaeological Museum of North Macedonia in Skopje. After a pilot project in 1998, systematic excavations began in 2000 and are continuing.1

buildings became obvious during the 2016 season; the west wall of the residence, in its final major phase, was also the east wall of Annex Room 11 of the basilica complex (fig. 3). The Episcopal Basilica By the end of the 2011 season, the apse and presbyterium, nave, aisles and narthex of the Episcopal Basilica had been almost entirely revealed.2 At first glance, the building appears to be a standard three-aisle basilica, c. 33 m in length from apse through narthex plus rooms west of the narthex. The total width of nave and aisles is almost 15 m. Several reasons why it is not the usual sixth-century basilica have been described in earlier articles.3 They include the annular corridor that communicated with a tomb or crypt under the apse; the long presbyterium taking up nearly a third of the length of the nave; two ambos, a small one in the presbyterium and a larger one at the south side of the nave; the horseshoe-shaped enclosure on the south side of the nave, just west of the ambo; and the colonnades, which included a pier beside the presbyterium and a c. 1.50 m high barrier between lateral aisles and nave.

Since 2011, attention has been focused on the excavation and elucidation of the Episcopal Basilica and the Northern Residence on the northern terrace, with two brief forays into cemetery areas outside the city. Here we describe recent work on the northern terrace, with primary focuses on the basilica and the residence and the questions they raise. Finally, we will outline our present understanding of the history of the anonymous city at Golemo Gradište – insofar as one may speak of history without written sources.

Various compartments were located in the space between the retaining wall at the south side of the church complex and the south wall of the basilica. In 2014, parts of several rooms were revealed at the east end of that space; their functions remain unclear. Apparently, a corridor ran along outside the south aisle of the basilica for most of its length.

The Episcopal Basilica and the Northern Residence are centrally located in the lower city on the northern terrace (fig. 2). A massive east–west retaining wall marked the south edge of a relatively level space for the church complex, although floors of the southern annexes were 1.0–1.5 m higher than those to the north. The Northern Residence stood north-east of the church, on another relatively level space c. 1.5 m or more below the northern annexes of the church. The connection between the two

From the south end of the narthex, a doorway led into Room 1. Steps and a second doorway led south into Room 2. In the east wall of Room 1, a doorway without a threshold block gave access to a space that, c. 3 m from its west wall, divided into two corridors. The northern one, mentioned above, ran beside the south wall of the basilica; the shorter southern one on a higher level apparently was included within the baptismal complex. 2 Carolyn S. Snively, with the assistance of other archaeologists, has supervised excavation in the basilica since its unexpected discovery in 2008. 3 Snively 2011; 2013.

The co-directors of the project are Carolyn S. Snively and Goran Sanev. Project publications include Sanev et al. 2012; Snively 2002, 2006, 2008, 2010, 2014 and 2017; Snively, Sanev 2013 and 2018. 1

58

Archaeological Investigations at Golemo Gradište, Konjuh, 2012–2016

Fig. 1. Plan of Golemo Gradište, Konjuh, 2015 (G. C. McArdle, after I. Mikulčić, M. Milojević and the National Survey Institute, with additions by C. S. Snively).

Room 2, in addition to being accessible from the narthex via Room 1, had an eastern doorway into the baptistery and a western one into Room 3. Possibly Room 2 served as a catechumeneion. Its only preserved furnishings consisted of two columns flanking the doorway and a smaller column (or table support?) in the south-east corner. A cist burial occupied the north-west corner of Room 2. The deceased, a middle-aged male, wore a small bronze cross. Although the chronological relationship of the burial to the church remains uncertain, the grave was most likely constructed shortly after the basilica was damaged and ceased to function, but before the roof collapsed.

The corridor of the baptismal complex ended 1.5 m east of the doorway into the baptistery. Its north side for a length of 6 m consisted of three built piers that supported two large stone slabs with mouldings. They formed a balustrade from which one could look down into the lower northern corridor.4 At the east end of the baptistery corridor, a 4 A number of well-cut stone blocks, of which several clearly belong to an arch or arches, were found within the debris between the baptistery and the south wall of the basilica. Many of them appeared in the lower corridor below the balustrade. It now seems likely that the piers supported two arches above the two screen slabs. But why such an elaborate and massive construction would have been placed in this position remains unclear.

59

Carolyn S. Snively & Goran Sanev

Fig. 2. The Northern Residence and the Episcopal Basilica, on the northern terrace, 2016. Photo from the acropolis, from south and above (Konjuh Excavation Project photo Kon 18 Jl 16 32).

purpose or the piscina was later added to an existing space (fig. 4).

doorway led into a courtyard that extended from the wall of the lower, northern corridor to the southern terrace wall. The apse of the baptistery protrudes into the west side of the courtyard. The extent of the courtyard to the east and its connections with the rooms at the south-east corner of the basilica await excavation.

Outside the line of the west wall of the narthex, a row of at least five rooms begins at the southern terrace wall and runs to the north. At the south-west corner of the basilica complex, the 0.90 m wide west wall of Room 3 also served as a retaining wall; together with the southern terrace wall, it protected the church from rising terrain to the southwest. Benches lined the east, north and west walls of Room 3; a step gave access to a 1 m wide platform against the south wall (fig. 5).

The baptistery itself consists of a rectangular space with a raised eastern apse. While multiple conches were common, baptisteries with apses were relatively rare.5 Six columns once surrounded the round piscina and supported a baldacchino; steps on the east and west sides led into the oval basin. The piscina was not centered but stood in the south side of the room, slightly closer to the apse than to the west wall. The doorway from the corridor into the north-east corner and that from Room 2 into the northwest corner would have brought clergy and catechumens into the open, northern side of the room. A perforated screen slab6 standing near the west doorway may have guided those entering from Room 2.

A doorway at the west end of the north wall of Room 3 led into Room 4. North of the threshold, two semicircular steps bridged the drop in level. Benches lined the walls of Room 4 as well. A modest brick arch, whose shape was visible at the time of excavation, once formed a small opening between Rooms 1 and 4. Benches were the characteristic feature of Rooms 3 and 4. Perhaps the platform or dais at the south end of Room 3 supported the seat of a church official presiding over a meeting of those sitting on the benches. On the other hand, the supports and the fragmentary top of a stone table in Room 4 raise the possibility that this compartment, easily accessible from the adjacent room with a doorway to the

The room with the piscina, Room 2 and the southern corridor form a unit. Nevertheless, it remains unclear whether the baptistery was constructed originally for that 5 6

Ristow 1998, pp. 17–20, especially p. 20 for baptisteries with apses. It will be published by Sabrina Higgins.

60

Archaeological Investigations at Golemo Gradište, Konjuh, 2012–2016

Fig. 3. Combined plan of the Northern Residence, in 2015, and the Episcopal Basilica, in 2016 (K. Haas, C. Maxwell-Jones, M. Nichol, G. Sanev and C. S. Snively).

outside, served as a place for those entering the basilica to deposit their offerings.7

forms Room 6. No evidence for function was observed, although a cache of mostly unidentifiable iron objects was found in the north-west corner. Benches lined parts of the east and west walls; the eastern bench also served as a step for entrance from the narthex.

In the doorway at the west side of Room 5, a 1 m wide threshold displayed cuttings for a door. This modest doorway was the only western entrance to the basilica in its final phase. From Room 5 one could proceed east into the narthex, go south into Room 4 or, in an earlier phase before that doorway was blocked, move north into Room 6.

The north wall of Room 6 is in line with the north wall of the narthex. The east and west walls of Room 7, however, lie c. 0.5 m east of the walls of Rooms 5 and 6. Thus, although the doorway between Rooms 6 and 7 is located at the east end of the north wall of Room 6, there was space for a small, stone-paved compartment to be tucked into the south-east corner of Room 7. A narrow wall and two perforated screen slabs separated that space (Room 7b)

The long, narrow compartment directly west of the narthex and accessible through one doorway from the narthex 7

For a more detailed discussion of this possibility, cf. Snively 2018.

61

Carolyn S. Snively & Goran Sanev

Fig. 4. The Baptistery, from the south-west, in 2015 (Konjuh Excavation Project photo Kon 10 Jl 15 06).

Fig. 5. The Episcopal Basilica, from the south-west. Room 3 in foreground, baptismal complex to right, Room 4 to left, in 2015 (Konjuh Excavation Project photo Kon 11 Jl 15 39).

62

Archaeological Investigations at Golemo Gradište, Konjuh, 2012–2016

Fig. 6. Room 7 of the Episcopal Basilica complex, in 2016, from the west (Konjuh Excavation Project photo Kon 05 Jl 16 13).

from the rest of the room. Several pieces of a white marble table top along with fragments of a small pithos came to light in Room 7b, which may have served as a place for deposition of offerings.8 A larger pithos and a fire pit were found at the north side of Room 7a. At the south end of the west wall, a 2 m wide doorway had been blocked up in the final phase (fig. 6).

made from a Corinthian capital came to light. These finds in the 2011–2012 seasons, before the discovery of the baptistery south of the basilica in 2013, were interpreted as a very thoroughly destroyed baptistery, in which the pit represented the location of the piscina. Perhaps completion of the excavation of Room 9 will provide evidence to identify the space either as an earlier baptistery or as some other element in the ecclesiastical complex.9

At the north end of the narthex a wide doorway opens into a square room (8) whose north side has not been completely excavated. Nevertheless, it appears that other annexes are located to the north of Rooms 7 and 8 as well as to the east. From the east side of Room 8, a wide doorway and steps lead down into a large compartment (Room 9) beside the north wall of the basilica; its extent to the north remains unexcavated. A doorway near the south end of the east wall gives access to Rooms 11 and 12 further east. Several columns and bases were found in a heap in the west part of Room 9.

In the final phase, the doorway in the south-east corner of Room 9 led east through a small room (12a/south end of Room 12) and through a second doorway into a blind corridor (11a/south end of Room 11), which stopped at the south end of the west wall of the Northern Residence. Before the construction of a wall 3 m north of the north wall of the basilica, Rooms 11 and 12 were substantial compartments extending north from the basilica. Benches lined the east, south and west walls of Room 11; a solid construction of brick, spolia and mortar rested against the north wall. A small rectangular ‘window’, closed in the final phase, once opened between Rooms 11 and 12. A

Within Room 9, a low barrier of large blocks marked off a rectangular space in the north-east part of the excavated area, i.e., Room 9b. In the middle of the north side of that space, along the north edge of the trench, a large, roughly rectangular pit was excavated. The debris in this area included several large worked but undecorated stone slabs. At the east side of the space, a stone vessel 8

Excavation in 2018 tentatively confirmed the hypothesis of Yannis Varalis, who had suggested that a monumental entrance to the basilica complex might have been located here, at the north side of the church. Room 9b now appears to be a long courtyard, once surrounded by corridors and rooms on east, south and west sides, i.e., possibly an atrium. The usual location for an atrium is west of the narthex. Cf. de Blaauw 2011; Ončevska-Todorovska 2015. The basilica at Plaošnik, Ohrid, has an atrium on the north side of the church. Cf. Bitrakova Grozdanova 2009, p. 29, fig. 3–4.

9

Snively 2018.

63

Carolyn S. Snively & Goran Sanev buried pithos and, near the west and north walls of Room 3, a cache of smashed cooking pots and four small pithoi, as well as a large decorated bone artefact, fragments of glass vessels and a small hoard of coins suggested the possibility of a residence; the courtyard and an apparent kitchen confirmed that hypothesis.

patch of brick and mortar floor was found in Room 12, which displayed in its west wall two doorways leading into the northern part of Room 9. The walls that cut off the north part of Rooms 11 and 12 (Rooms 11b and 12b) originally included doorways that allowed access to the northern parts, but after those doorways were also blocked, Room 11b was completely abandoned. It is unclear whether Room 12b continued to be accessible through Room 9.

By 2013, the whole eastern side of the house had been excavated, with Rooms 4 and 5 revealed to the north, together with the north wall and most of the courtyard between the south and east corridors. The house measures c. 27 m along the east side, c. 12.7 m along the south wall of the south-east wing, and c. 24 m along the north side, as indicated by recent campaigns. In 2014, a west wall came to light. But, although the west wall abutted the north wall of the basilica at its south end, it did not extend all the way to the north wall of the residence, but made a corner and turned west. The 2015 and 2016 seasons showed that the north-west corner of the building extended to the west beyond the line of the west wall of the southern part of the building. Between the north-west room (7) and the north end of the west wall stood an elaborate entryway that possibly included a staircase to a second storey (fig. 7). Among the evidence for an earlier phase of the entry was a well-built drain that was truncated by the construction of the west wall of the residence, as well as a column that may once have formed part of a double arched entrance to the building.

Rooms 11 and 12 and an unknown number of spaces north of them as far as the entryway of the Northern Residence formed a rectangular unit that at one time connected the basilica and the Northern Residence, i.e., at some time the two buildings formed a single complex. The unexplored spaces in the northern part of the unit were not accessible from Room 11 or 12; their connection with either basilica or residence remains unclear. The assumption in 2011–2012 was that Room 9 had been destroyed – with great thoroughness – at the same time as the rest of the basilica complex. More recent discoveries raise the question of whether it may have been dismantled at an earlier time. Whether or not the corridor along the north side of the basilica, through Rooms 9, 12a and 11a, continued to function, Room 11b certainly and Room 12b probably had ceased to be an active part of the church complex in its final phase before it was damaged and went out of use in the later sixth century.

The Northern Residence belongs to the category of the peristyle house, a relatively rare construction in the sixth century. Its defining feature was the courtyard, surrounded by colonnaded porticos, behind which appeared various rooms. Column bases are still in situ on the north, east and south sides of the courtyard. Only in the western part of the courtyard next to the west wall was rough stone paving found; this area was used as a kitchen after construction of the west wall.

Investigation of the basilica has raised a chronological issue. The church was constructed above a secular building, probably a large residence, that belonged to the general rebuilding of the city in the second quarter of the sixth century. After that earlier building was dismantled, the basilica was erected near the middle of the sixth century; it existed for no more than a few decades and ceased to function in the later sixth century.10 Yet the church displays numerous renovations and rebuildings that are difficult to fit into those few decades. In particular, the construction of Rooms 11 and 12 and other spaces to the north as part of a unit bounded by the west wall of the Northern Residence, followed by their gradual abandonment, presents a problem.

Excavation in 2014 and 2016 in the irregular space between the west wall and the south-east wing revealed three large pits, and three burials in or above the pits (fig. 8).12 While the burials must belong to the time after abandonment of the terrace in the late sixth century, it is unclear whether the pits functioned together with some phase of the residence or post-dated its destruction. The means of access to this space from the residence or from outside is unclear. Several walls were found that may belong to a building from the fifth-century settlement on the terrace; construction of the south-east wing had truncated and damaged those walls.

The Northern Residence Excavation of the Northern Residence began in 2005, but the nature and extent of the building remained uncertain for several seasons.11 By 2011 the south-east wing had been cleared, revealing three rooms, i.e., a long corridor and two rooms to its east, while to the north the eastern colonnade of the courtyard had come to light. Investigation continued to the north and west, in order to determine the size of the building. The discovery in Room 2 of a

The plan of the Northern Residence illustrates a number of issues. Two or three different orientations are visible in the building; the west wall is not parallel with any other wall. Clearly, construction of the west wall substantially changed the overall plan of the residence. It and the south

The date given for abandonment and/or destruction of many cities in the region is AD 586; the date is based on coin hoards. Cf. Hadzi-Maneva 2009. 11 Since 2007, Goran Sanev, with the assistance of several other archaeologists, has supervised excavations in the residence. 10

12 For these and other burials on the northern terrace, cf. Snively, Sanev 2018.

64

Archaeological Investigations at Golemo Gradište, Konjuh, 2012–2016

Fig. 7. The entryway of the Northern Residence, in 2016, from the west (Konjuh Excavation Project photo IMG_1972).

wall of the entryway represent the final major phase of the residence, when the west wall truncated the south wall of the courtyard as well as the dismantled walls of an earlier building. The entryway in its present form belongs to that final phase. As already described above, the south wall of the entryway, the west wall of the residence and the north wall of the basilica form three sides of a rectangle that once connected the two buildings.

is documented in the area from at least the second century. Present evidence indicates that the city on the northern terrace was not established until the fifth century. Relatively little is known about the fifth-century settlement. It was rebuilt no later than the second quarter of the sixth century, probably in the 520s or 530s, but possibly even earlier. At that time, the fortress was established on the acropolis, and the fortification walls were constructed around the entire site.

Evidence for earlier phases or perhaps a previous building is visible or was found in tests in the north-east rooms as well.13 Thus, a major issue in the Northern Residence will be the elucidation of the various phases of construction on its site and the relationship between those buildings and phases and those of the Episcopal Basilica.

Walls found in tests below the Southern Residential Complex belong to the fifth-century city; walls found below Room 4 and the space west of the south-east wing of the Northern Residence may be of similar date. Based on evidence now available, the Northern Residence and the secular building under the Episcopal Basilica were part of the early sixth-century reconstruction. As noted above, the Episcopal Basilica replaced its predecessor near the middle of the sixth century.

History of the anonymous city Cemeteries of the Roman period located in the vicinity of Golemo Gradište point to the existence of a nearby settlement, which has not been located.14 Two grave inscriptions from the site have recently been dated to no later than the second century AD;15 thus, a Roman presence

The latest coins in a small hoard found in Room 3 of the Northern Residence were of Justinian I, thus giving a terminus post quem of AD 565 for the destruction of the building. For the end of the basilica, the dating evidence is not so specific, but more is known about the sequence of events. The eastern part of the church was damaged by fire and it went out of use, but the building continued to stand for some time until it gradually collapsed. Although the assumption is that both basilica and residence ceased to function at the same time, as the

13 Neolithic material appeared in a deep test trench in Room 4 in 2006. Our investigations so far suggest that there was only sporadic use of the northern terrace between the Neolithic and Late Antique periods. 14 For a discussion of the evidence for a Roman settlement, cf. Snively 2017; Higgins 2013. 15 Personal communication from Slavica Babamova, epigrapher at the Archaeological Museum in Skopje, who will publish the inscriptions from Golemo Gradište.

65

Carolyn S. Snively & Goran Sanev

Fig. 8. Excavation in the space between the south-east wing and the west wall of the Northern Residence, with pits and early walls, in 2014, from the north-west (Konjuh Excavation Project photo Kon 3 JL 14 55).

Bibliography

result of the same catastrophic event, it cannot be proved. What is clear, however, is that the residents of the terrace fled to the acropolis for refuge in the late sixth century. Some spaces there were reconstructed for storage of pithoi, and the entire area of the acropolis was filled with small dwellings, some of which were quarried out of the bedrock. The number of storage vessels makes it clear that the population expected and feared a siege. The evidence for an early seventh-century destruction shows that their fears and expectations were realistic. Although traces of later habitation have been observed on the acropolis and later but unstratified material has been found here and there on the northern terrace, the early seventh-century destruction on the acropolis seems to mark the end of urban life on the site.16

Abbreviations: MAA = Macedoniae Acta Archaeologica. NišByz = Ниш и Византија – Niš and Byzantium. Bitrakova Grozdanova V., 2009. ‘Lychnidos à l’époque paleochrètienne et son noyau urbain’, in NišByz, 7, pp. 23–36. de Blaauw S. L. 2011. ‘The church atrium in Rome as a ritual space. The Cathedral of Tyre and St. Peter’s in Rome’, in F. Andrews (ed.), Ritual and Space in the Middle Ages. Proceedings of the 2009 Harlaxton Symposium, Donington (Harlaxton Medieval Studies, 21) pp. 30–43.

For evidence of medieval activity at the site, cf. the article by D. Gjorgjievski in this volume. A recent article, Tomović 2013, gives a fascinating picture of the third quarter of the fourteenth century at the castel of Konjuh on the basis of a grave inscription and written sources; unfortunately, so far little evidence has been found at the site to confirm or add archaeological detail to that picture.

16

Hadzi-Maneva M., 2009. ‘Coin hoards from late 6th and 7th century discovered in the Republic of Macedonia’, in M. Wołoszyn (ed.), Byzantine Coins in Central Europe 66

Archaeological Investigations at Golemo Gradište, Konjuh, 2012–2016 Between the 5th and 10th Century. Proceedings from the Conference Organised by Polish Academy of Arts and Sciences and Institute of Archaeology, University of Rzeszów, under the Patronage of Union académique international (Programme No. 57 Moravia Magna), Kraków, 23–26 IV 2007, Kraków (Moravia magna. Seria Polona, 3), pp. 47–56

Giving Gifts to God. Evidences of Votive Offerings in the Sancturies, Temples and Churches. Proceedings of the 1st (Kokino: Archaeological and Astronomical Aspects – Parallels and Experience) & 2nd (Kokino: Giving Gifts to God) International Archaeological Conference held in Skopje & Kumanovo, 2016–2017, Kumanovo, 2018, pp. 175–182.

Higgins S., 2013. ‘Riding into Late Antiquity: A Thracian rider relief from Golemo Gradište, Konjuh’, in NišByz, 11, pp. 173–185.

Snively C. S., Sanev. G., 2013, ‘Late Antique residences at Golemo Gradište, Konjuh, R. Macedonia’, in NišByz, 11, pp. 157–172.

Ončevska Todorovska M., 2015. ‘The atrium in Early Christian church architecture in East Illyricum and the large Early Christian basilica with an atrium in Scupi’, in E. Dimitrova (ed.), Folia Archaeologica Balkanica, III, Skopje, pp. 267–288.

Snively C. S., Sanev G., 2018. ‘Life – and death – in the Late Antique city at Konjuh’, in Acta Musei Tiberiopolitani, 2, Strumica, pp. 157–163. Tomović G., 2013. ‘Надгробни натпис Воихне из Коњуха’, in B. Miljković-Katić (ed.), Споменица др Данице Милић, Belgrade (Зборник радова, 27), pp. 147–166.

Ristow S., 1998. Frühchristliche Baptisterien, Münster (Jahrbuch für Antike und Christentum. Ergänzungsband, 27). Sanev G., Snively C. S., Stojanoski M., 2012. ‘Excavations on the Northern Terrace at Golemo Gradište, Konjuh, 2007–2010’, in MAA, 20, pp. 347–364. Snively C. S., 2002. ‘Golemo Gradište at Konjuh: Report on the excavations in 2000’, in Dumbarton Oaks Papers, 56, pp. 293–302. Snively C. S., 2006. ‘Golemo Gradište at Konjuh, Republic of Macedonia: Prolegomena to the study of a Late Antique fortification’, in NišByz, 4, pp. 229–244. Snively C. S., 2008. ‘Archaeological excavations on the Acropolis of Golemo Gradište, Konjuh: 2000–2004’, in MAA, 18, pp. 335–351. Snively C. S., 2010. ‘Golemo Gradište at Konjuh, 2005 and 2006 seasons: The Northern Terrace’, in MAA, 19, pp. 375–389. Snively C. S., 2011. ‘The New Basilica at Golemo Gradište, Konjuh: A sixth century church in the province of Dardania’, in NišByz, 9, pp. 187–201. Snively C. S., 2013. ‘Golemo Gradište at Konjuh: An unidentified Late Antique city and its churches’, in O. Brandt, S. Cresci, J. López Quiroga, C. Pappalardo (eds.), Acta XV Congressus internationalis archaeologiae Christianae, Toleti, 8–12.9.2008. Episcopus, civitas, territorium, Vatican City (Studi di Antichità cristiana, 65), pp. 403–414. Snively C. S., 2014. ‘The rock-cut room on the acropolis at Golemo Gradište, Konjuh: Date and purpose’, in NišByz, 12, pp. 193–204. Snively C. S., 2017. ‘Golemo Gradište at Konjuh: A new city or a relocated one?’, in E. Rizos (ed.), New Cities in Late Antiquity. Documents and Archaeology, Paris (Bibliothèque de l’Antiquité tardive, 35), pp. 205–220. Snively C. S., 2018. ‘Spaces for deposition of offerings in Early Byzantine churches: Possible sacristies at Golemo Gradište, Konjuh’, in D. Gjorgjievski (ed.), 67

6 Le monastère des Quarante Martyrs de Sébastée à Saranda (Albanie), un centre important de pèlerinage pendant l’Antiquité tardive Skënder Muçaj Instituti i Arkeologjisë, Tiranë / Institut d’archéologie, Tirana (Albanie)

Kosta Lako Instituti i Arkeologjisë, Tiranë / Institut d’archéologie, Tirana (Albanie)

Skënder Bushi Muzeu Historik Kombëtar, Tiranë / Musée historique national, Tirana (Albanie)

Suela Xhyheri Universiteti Eqrem Çabej / Université Eqrem Çabej (Albanie) Les ruines du monastère des Quarante Martyrs, situé à proximité du port d’Onhesmos, a servi de point de repère pour les bateaux qui naviguaient sur la route maritime qui liait l’Occident et l’Orient. Le monument des Quarante Martyrs constitue un cas unique dans toute la Méditerranée pour son architecture, son plan, ses espaces en surface et en sous-sol, sa technique de construction, sans mentionner ses fonctions. En ce qui concerne son architecture, le modèle le plus proche est l’église Saint-Polyeucte à Constantinople. Les données archéologiques provenant du monument sont nombreuses et diverses ; elles concernent l’architecture, la sculpture, les peintures murales, les inscriptions, les monnaies, la céramique (des milliers d’amphores et des centaines de lampes), des objets métalliques, des objets d’os et de verre, etc. La plupart de ces objets appartiennent à la deuxième moitié du Ve et à la première moitié du VIe siècle. The ruins of the Forty Martyrs Monastery near the port of Onhesmos served as a reference point for the boats that travelled on the sea route linking the west and the east. The monument of Forty Martyrs, in terms of its architecture, its plan, its spaces under and above ground, its technique of construction, to say nothing of its functions, constitutes a unique case for the whole of the Mediterranean, for its preservation both at the underground level and on the surface. As far as architecture is concerned, the closest model is St Polyeuctus’ Church in Constantinople. The archaeological data from the monument are numerous and various, concerning the architecture, the furniture, the mural inscriptions, coins, thousands of amphorae and hundreds of lamps, metallic objects, glass etc. Most of these objects belong to the second half of the fifth century and the first half of the sixth century. Die Ruinen des Klosters der Vierzig Märtyrer in der Nähe des Hafens von Onhesmos war ein Anlaufpunkt für Schiffe auf der Seeroute zwischen West und Ost. Das Kloster stellt, ohne im Einzelnen hier auf die Funktion einzugehen, hinsichtlich seiner Architektur, seines Grundrisses und seines Aufbaus einen einzigartigen Fall im gesamten Mittelmeerraum dar, sowohl bei den unter- als auch bei den oberirdischen Befunden. Architektonisch steht die Kirche von Saint Polyeuctus in Konstantinopel dem am Nächsten. Archäologische Funde aus dem Areal sind zahlreich erhalten, ob Bauplastik, Möbel, Wandinschriften, Münzen oder Tausende von 69

Skënder Muçaj, Kosta Lako, Skënder Bushi & Suela Xhyheri Amphoren, Hunderte von Lampen, Metall- und Glasobjekte. Die meisten dieser Gegenstände gehören in die zweite Hälfte des 5. und in die erste Hälfte des 6. Jhs. Le rovine del Monastero dei Quaranta Martiri situate vicino al porto di Onhesmos fungevano da punto di riferimento per le barche che navigavano sulla rotta marittima che collegava l’ovest e l’est. Il complesso dei Quaranta Martiri, con la sua architettura, la sua pianta, gli spazi sotto e sopra terra, la tecnica costruttiva, senza menzionare le funzioni, è unico nel Mediterraneo. Per quanto riguarda l’architettura, il modello più vicino è la chiesa di San Polieucto a Costantinopoli. I dati archeologici emersi dal monumento sono diversi e riguardano architettura, scultura, dipinti murali, iscrizioni, monete, ceramiche (migliaia di anfore e centinaia di lampade), oggetti in metallo, in osso e vetro, ecc. La maggior parte di questi oggetti appartiene alla seconda metà del V secolo e alla prima metà del VI secolo. de la flotte britannique (fig. 2-3). De 2002 à 2013, une équipe de l’Institut d’archéologie de Tirana a réalisé une fouille presque complète du monument des Quarante Martyrs (fig. 4).3 Les photographies datant d’entre la Première Guerre mondiale (fig. 5) et l’année 1937 ainsi que l’exploration archéologique montrent que l’ensemble monumental constituait un véritable centre de pèlerinage pendant l’Antiquité tardive.

Les ruines du monastère des Quarante Martyrs se trouvent au sommet de la colline sur laquelle est construite l’actuelle ville de Saranda (anciennement Onhezm)1, en un point où l’on a une vision à la fois sur la mer et sur l’intérieur des terres. Le monastère, situé non loin du port de l’antique Onhesmos2 (fig. 1), a servi de point de repère pour les bateaux naviguant sur la route maritime qui liait l’Occident et l’Orient par la baie de Saranda et le détroit de Corfou. Les murs du complexe principal ont été préservés à une hauteur considérable jusque vers la fin de la Seconde Guerre mondiale, plus précisément jusqu’en octobre 1944, quand le monument a été détruit par les bombardements

D’après la légende, les Quarante auraient été des soldats de la XIIe légion Fulminata en garnison à Mélitène, en Petite-Arménie, où aurait été entreprises des persécutions

Fig. 1. Photo aérienne du port d’Onhezm/Saranda et du monastère des Quarante Martyrs en 1937. Lako 1984; Lako, Muçaj, Bushi, Xhyheri 2014. Maraval 2004, p. 102. Dans cet ouvrage, P. Maraval mentionne un grand nombre de centres de pèlerinage situés à proximité des ports. Sur le culte des reliques dans les églises extra-muros, voir aussi Brown 1981, pp. 7-8 et 42-43.

1 2

3 Muçaj 2012a; et 2012b, pp. 14-15; Muçaj, Lako, Bushi, Xhyheri 2013 et 2014.

70

Le monastère des Quarante Martyrs de Sébastée à Saranda (Albanie), un centre important de pèlerinage....

Fig. 2. Le monastère des Quarante Martyrs en 1937.

Fig. 3. Le monastère des Quarante Martyrs (L. M. Ugolini).

71

Skënder Muçaj, Kosta Lako, Skënder Bushi & Suela Xhyheri

Fig. 4. Le bâtiment principal du monastère, fouillé par l’équipe de l’Institut d’archéologie de Tirana.

alors été condamnés à passer toute une nuit dehors, nus, sur un lac gelé, et tous en seraient morts. Rapidement, les premiers sanctuaires dédiés aux Quarante Martyrs auraient été construits à Sébastée, avant que leur culte ne se répande peu de temps après en Méditerranée orientale

de chrétiens pendant le règne de Licinius (308-324). Vers l’an 320, les Quarante auraient été appelés à Sébastée devant le gouverneur de la province, Agricola, où ils auraient refusé d’abjurer leur religion malgré les pressions de leur juge et du général Lysias. Ils auraient

Fig. 5. Carte postale datant de la Première Guerre mondiale.

72

Le monastère des Quarante Martyrs de Sébastée à Saranda (Albanie), un centre important de pèlerinage.... et, dès la fin du IVe siècle, en Méditerranée occidentale.4 Hormis les nombreuses églises dédiées aux Quarante Martyrs, des martyria de plusieurs formes leur été consacrés, parmi lesquels le péristyle à 40 colonnes de Nysse (Cappadoce).5 De très nombreux monastères ont aussi hébergé leurs reliques.

la fin du XIIe siècle, en référence à la cité en contrebas de la colline sur laquelle se situait le monument. Selon Strabon, Denys d’Halicarnasse et Claude Ptolémée, Onhesmos était, jusqu’au IIe siècle av. J.-C., un port de la Chaonie, tandis qu’elle est mentionnée comme cité et siège épiscopal pendant les Ve et VIe siècles sous le nom d’Anchiasmos.6 Au cours de cette même période, elle devint un centre important pour l’accueil et l’hébergement des pèlerins chrétiens. Les incursions de Totila en 551 auraient causé beaucoup de destructions dans la cité ; elle fut fortifiée par la suite, mais seul un tiers de son ancien territoire a été défendu.7 Le matériel archéologique montre qu’à partir du dernier quart du VIe siècle la cité commença à être abandonnée et devint un petit port, ce qu’elle demeura jusqu’à au milieu du VIIe siècle (fig. 6-13 et 3438).8 Le monastère des Quarante Martyrs déclina lui aussi.

Les circonstances de la diffusion du culte des Quarante Martyrs dans la province d’Epirus Vetus pourraient être mieux comprises par l’étude du monument qui fait l’objet du présent article. Les seules autres sources tangibles à notre disposition, y compris pour les périodes les plus tardives, byzantine et post-byzantine, sont des peintures murales des régions de Saranda, Ohrid et Korça. Il n’y a effectivement pas de source écrite de l’Antiquité tardive ou du Moyen Âge sur le monastère ici étudié. Le nom Sancta Karentet est toutefois mentionné pour la première fois vers

Fig. 6. Fragments de lampes de verre et de vitre (VIe siècle). 6 Perry 1881, p. 16; Meksi 2004, p. 18. Les sources indiquent que l’évêque Christodorus d’Anchiasmos signa une lettre envoyée à l’empereur Léon Ier (457-474) en 458, puis une autre envoyée au pape Hormisdas en 516. 7 Lako, Muçaj, Bushi, Xhyheri 2014. 8 Muçaj, Lako, Bushi, Xhyheri 2011.

Wilpert 1916, pp. 709-710 et 722-725; Hülsen 1927, pp. 425-27; Janin 1953, pp. 498-502; Brown 1981, p. 95; Maraval 1999; Gulowsen 2001, p. 241; Maraval 2004, pp. 338, 372, 374-375, 375-378, 392 et 407. 5 Maraval 1999, pp. 199-200. 4

73

Skënder Muçaj, Kosta Lako, Skënder Bushi & Suela Xhyheri

Fig. 7. Fragments de lampes de terre cuite (VIe siècle).

74

Le monastère des Quarante Martyrs de Sébastée à Saranda (Albanie), un centre important de pèlerinage....

Fig. 8. Lampes de terre cuite (VIe siècle).

Les premières données précises sur le monastère nous proviennent de voyageurs, de théologiens et d’archéologues et remontent à la période comprise entre le XIXe  siècle et les années 1920 (fig. 33). Parmi ces sources « primaires », mentionnons par exemple Athanasios Psalidas, Kosmas Thesprotos, William Martin Leake, Panagiotis Aravantinos, Evangelos Demetriou, ou encore Demetrios Evangelidis.9 Après

cette phase initiale, d’autres chercheurs se sont intéressés au monastère avant que la mission des années 2002-2013 ne prenne la relève.10 Ainsi, les photographies faites par Luigi Ugolini et Ejnar Dyggve (fig. 3), de même que les photos aériennes de l’année 1937 réalisées par la flotte aérienne italienne pour le compte du Royaume albanais (fig. 1-2), sur lesquelles on distingue le plan 10 Ugolini 1927, pp. 147-148; Dyggve 1940, pp. 405-406; Meksi 1985, pp. 26-27; Ceka 1999, pp. 68-70; Bowden, Mitchell 2001; Muka 2002; Mitchell 2004.

Leake 1835, p. 35; Aravantinos 1856, pp. 3-4; Demetriou 1913; Evangelides 1919, pp. 39-40; Thesprotos, Psallidas 1964, p. 83.

9

75

Skënder Muçaj, Kosta Lako, Skënder Bushi & Suela Xhyheri

Fig. 9. Support en fer de lampes de verre.

Fig. 12. Fragment de table d’autel en marbre.

Fig. 10. Élément de décor en fer d’un accessoire d’éclairage.

Fig. 11. Plaque de parapet en calcaire.

Fig. 13. Fragment d’une inscription grecque sur marbre.

76

Le monastère des Quarante Martyrs de Sébastée à Saranda (Albanie), un centre important de pèlerinage....

Fig. 14. L’église du XIXe siècle et la base du chancel.

l’ouest, jusqu’à devenir le monument que nous connaissons aujourd’hui (fig. 32). À l’église s’est ajouté un portique adossé aux murs extérieurs des nefs latérales et du narthex (fig. 21). Le portique est érigé sur les arcades doubles d’un nouveau couloir qui entoure la crypte et dont les murs sont couverts d’inscriptions portant le nom de différents donateurs (fig. 22). Au nord-est ont été ajoutés des espaces pour l’accueil des pèlerins, avec un bain et deux citernes. Le grand hall de la crypte a été transformé lors de cette phase en une citerne qui fut mise en communication avec trois autres nouveaux bassins d’accumulation (fig. 20). C’est à la deuxième phase qu’appartiennent les nombreuses constructions liées au baptême des pèlerins : des annexes et des fonts baptismaux, à la fois dans le baptistère, dans les nefs de l’église, dans le portique et dans le nouveau couloir de la crypte. La piscine au sud perdit sa fonction pendant cette phase et une nouvelle piscine fut construite à l’est de l’église.

d’une église, aux caractéristiques particulièrement peu communes parmi les structures paléochrétiennes connues, constituent autant de sources importantes pour l’histoire de l’architecture ecclésiastique. Le complexe étudié est composé d’une église d’un type particulier avec un portique de 40 arcades, d’un baptistère (fig. 18) et de fonts baptismaux extérieurs (fig. 16-17), d’espaces pour l’accueil, de citernes (fig. 19-20), ainsi que de grandes piscines qui symbolisaient le lac de Sébastée où les soldats auraient été martyrisés (fig. 23 et 25). La partie souterraine est particulièrement bien conservée : principalement occupée par la crypte, elle est également composée de chapelles (fig. 31), d’espaces de méditation, de halls, de couloirs, de niches dans les murs pour les reliques des martyrs (fig. 26), de citernes et de fonts baptismaux. Le monument comporte deux phases principales de construction, même si sa structure a connu des interventions postérieures. Pendant la première phase ont été construits l’église à trois nefs et son narthex (fig. 4) ; la partie principale de la crypte, avec un couloir en ovale, précédé d’un hall couvert par une arcade ; une piscine sur le côté sud de l’église, de même que l’ensemble principal du baptistère (fig. 30). Dans une deuxième phase, l’édifice a été complété par des constructions au nord, au sud et à

Cette dernière se présente comme une grande construction rectangulaire de 40 m de longueur par 23 m de largeur, avec une abside à l’est et trois autres sur chacune des parties extérieures des deux nefs latérales. Les murs de ces six grandes conques latérales et de l’abside orientale sont interrompus par plusieurs portes, fenêtres et niches (fig. 24 et 29). 77

Skënder Muçaj, Kosta Lako, Skënder Bushi & Suela Xhyheri

Fig. 15. Ambon dans la nef centrale.

Fig. 16. Fonts baptismaux (A 18-3 B).

78

Le monastère des Quarante Martyrs de Sébastée à Saranda (Albanie), un centre important de pèlerinage....

Fig. 17. Fonts baptismaux (A 7-1).

Fig. 18. Baptistère.

79

Skënder Muçaj, Kosta Lako, Skënder Bushi & Suela Xhyheri

Fig. 19. Système de citernes d’eau potable.

À l’est se trouve le sanctuaire, séparé de la nef centrale par la clôture du chancel et équipé d’un synthronon dont seuls les bancs latéraux ont été conservés (fig. 14). Un ambon circulaire se trouve dans la partie sud de la nef centrale (fig. 15). Dans les nefs latérales, des fonts baptismaux de différentes profondeurs ont été aménagés dans de petites conques (fig. 16-17). Le bâtiment est fait de pierres calcaires plates, associées à des briques dans les arcades et les voûtes. Le toit des nefs latérales était couvert de voûtes semi-sphériques, alors que celui de la nef centrale l’était de voûtes cruciformes.

sol porté par des pilettes mais qu’il n’a jamais été réalisé. À l’inverse des sols qui sont assez simples dans tout le monument, les murs sont ornés de peintures murales, dont une grande partie est conservée en sous-sol. Ces peintures appartiennent à deux phases distinctes : la première présente des motifs d’arcs surplombant des croix (fig. 27), alors que la deuxième représente des scènes de l’Évangile (fig. 28).11 11 Dans l’une des chambres où les fresques sont bien conservées, on peut distinguer l’appel des quatre premiers disciples (Luc, 5.1-11) ; la multiplication des pains (Matthieu, 14.13-21 et 15.32-39) ; une femme penchée sur un sarcophage (référence biblique non identifiée) ; une scène inhabituelle, dans laquelle deux apôtres se tirent respectivement la barbe (Jean, 13.29 ?), où Judas Iscariote serait théoriquement présenté comme le « gardien de la bourse ». Cette dernière scène pourrait aussi être expliquée comme un « baiser sacré », une expression d’amitié courante dans le monde juif (p. ex. 2 Samuel, 20.9, où Joab prend Amasa par la barbe pour l’embrasser, avant de le tuer).

Les sols sont simples et faits de dalles de pierre ou de tuiles, placées sur une couche d’argile. Dans leur construction initiale, les voûtes des niches de la crypte comportaient des tubes d’aération, qui ont été ultérieurement fermés par une plaque. On en déduit que le projet de départ prévoyait un 80

Le monastère des Quarante Martyrs de Sébastée à Saranda (Albanie), un centre important de pèlerinage....

Fig. 20. Grande citerne d’eau potable, modifiée dans une phase tardive (B 11 [1495]).

Fig. 21. Le portique sud.

81

Skënder Muçaj, Kosta Lako, Skënder Bushi & Suela Xhyheri

Fig. 22. Le monument, vu de l’ouest.

Fig. 23. Piscine représentant le lac de Sébastée.

82

Le monastère des Quarante Martyrs de Sébastée à Saranda (Albanie), un centre important de pèlerinage....

Fig. 24. Niche A 19,3 A.

Fig. 25. Aire au sud du porche sud.

83

Skënder Muçaj, Kosta Lako, Skënder Bushi & Suela Xhyheri

Fig. 26. Tombes au sous-sol.

Fig. 27. Fresque appartenant à la première phase.

84

Le monastère des Quarante Martyrs de Sébastée à Saranda (Albanie), un centre important de pèlerinage....

Fig. 28. Fresque appartenant à la deuxième phase.

Les inscriptions que l’on trouve sur les murs de l’église, composées de céramique, montrent que des membres de la communauté juive d’Onhesmos ont participé à la construction12 ; cette communauté juive est également connue par une synagogue dans la cité13.

–, les objets de métal14 (fig. 9-10), d’os et de verre (fig. 6), etc. La plupart de ces objets appartiennent à la deuxième moitié du Ve et à la première moitié du VIe siècle. Les céramiques, qui constituent la plus grande part du mobilier archéologique, peuvent être associées à trois périodes principales. Le premier groupe vient des pièces du baptistère et des résidences ; on le date la deuxième moitié du Ve siècle et il est représenté par la sigillée africaine de type ARS 87A-B et des amphores de type LRA 2A, LRA 4B, LRA 4B1, bag-shaped 1, spatheion 1, spatheion 2A, Keay 55, Keay 56, Keay 57 et Keay 62Q (fig. 34).

Les données archéologiques provenant du monument sont nombreuses et variées. Elles concernent non seulement l’architecture (fig. 13), la sculpture (fig. 11-12), la peinture murale et les inscriptions, mais également les monnaies, la céramique – comme en témoignent des milliers d’amphores et des centaines de lampes (fig. 7-8 et 34-38)

Le deuxième groupe est principalement associé au corps central du complexe et à sa phase de destruction (attribuée à l’attaque de Totila en 551). On y retrouve les types suivants : ARS 97, ARS 99B, ARS 104B, LRC 3F et LRC 3G ; amphores LRA 1B2, LRA 2AB, LRA 3A, LRA 3A2, LRA 4B2, LRA 4B2-3, bag-shaped 3, LRA 10, crétoises, Ayala Axum, spatheion 2B, Keay 8B, Keay 62

L’une des inscriptions se trouve sur la façade nord des espaces souterrains, recouverte par les galeries latérales dans une deuxième phase ; comme les autres, elle est en grec et composée de fragments de céramique. Elle mentionne d’un donateur nommé Paregorios (Παρηγόριος), un nom très fréquent parmi les Juifs de l’Antiquité tardive (cf. Feissel 1983, no 193, p. 167 et no 294, p. 243) mais qui fait généralement défaut dans les inscriptions chrétiennes, bien que ce soit celui de l’évêque de Skopje en 343. Cette inscription est la seule du site qui ne soit pas accompagnée d’une croix, ce qui laisse penser à un membre de la communauté juive d’Anchiasmos. 13 On peut aussi mentionner une inscription de Venosa, une épitaphe en latin et en hébreu de la première moitié du VIe siècle : « Ici repose Augusta, l’épouse du vir laudabilis Bonus, fille d’Isa le Père, d’Anchiasmon, (...) ». Voir Noy 1993, no 107, pp. 137-140. 12

14 Antonaras 2009, pp. 118-119. Parmi les objets métalliques retrouvés, on peut mentionner un candélabre de fer, en forme de croix, semblable à celui de l’église Sainte-Sophie de Thessalonique.

85

Skënder Muçaj, Kosta Lako, Skënder Bushi & Suela Xhyheri

Fig. 29. Plan d’ensemble.

Fig. 30. Plan de l’église dans sa première phase.

86

Le monastère des Quarante Martyrs de Sébastée à Saranda (Albanie), un centre important de pèlerinage....

Fig. 31. Plan du complexe souterrain.

Fig. 32. Plan des premier et deuxième étages.

87

Skënder Muçaj, Kosta Lako, Skënder Bushi & Suela Xhyheri

Fig. 33. Plan de l’église dans sa période tardive (XIXe siècle).

(variantes D, F, K, J, I, V), et Benghazi LRA7 (variantes A, B et Mandruzzato C54) (fig. 35-37).

des pèlerins venait de territoires lointains où les croyants n’étaient pas encore baptisés.

Le troisième groupe de céramiques provient des annexes réutilisées du bâtiment principal, des installations d’hébergement et des citernes. Il est composé d’un nombre limité d’ARS 109, d’ARS 104B et C, d’amphores de type LRC 10 phocéenne, spatheion 3C et 3D, Keay 61B, Keay 50, “Orlo a fascia” et Castrum Perti, ainsi que de céramiques culinaires modelées à la main dites « slaves ». Tous ces éléments datent de la fin du VIe et du début du VIIe  siècle (fig.  38). Ces trois groupes comportent également un nombre très important de lampes de céramique et de verre.

Le sous-sol de l’église des Quarante Martyrs constitue un cas particulier, parmi les exemples connus, en matière d’espaces voués au pèlerinage. En plus d’être un lieu de préservation des reliques – il compte 4 tombes construites pour être observées par les visiteurs (fig. 26) –, il comporte une chapelle et d’autres espaces de méditation. La partie supérieure est pourvue d’un portique qui entoure la basilique sur trois côtés et dont les 40 arcades représentent les Quarante Martyrs. Il rappelle un autre martyrium, celui de Nysse.15 Les portiques servaient non seulement à permettre aux visiteurs de faire la queue pour voir les reliques ou pour être baptisés, mais également comme un lieu d’accueil lors des cérémonies.

Ce mobilier archéologique montre que le monastère a été bâti dans la deuxième moitié du Ve siècle et qu’il a été un centre important pour les nombreux pèlerins de passage au VIe siècle. Certaines parties ont connu plusieurs reconstructions et modifications pendant une période d’environs trois quarts de siècle, visiblement en fonction de l’augmentation du nombre des pèlerins. Le nombre considérable des fonts baptismaux (neuf) au sein même du baptistère, dans les portiques et dans les nefs montre clairement qu’une partie importante

L’augmentation du nombre des pèlerins semble avoir continué jusqu’au moment de la destruction de la partie supérieure du monument, lors des attaques des Goths

15

88

Maraval 1999, pp. 199-200.

Le monastère des Quarante Martyrs de Sébastée à Saranda (Albanie), un centre important de pèlerinage....

Fig. 34. Amphores (Ve siècle).

d’Italie dirigés par Totila en 551.16 Après cette destruction, certains espaces souterrains semblent avoir été utilisés par les moines, avant d’être finalement abandonnés vers le milieu du VIIe siècle. Le monument a néanmoins toujours

été considéré comme un lieu sacré et il a été régulièrement visité par des membres d’autres religions après son abandon. Le monument des Quarante Martyrs de Saranda, qui est aujourd’hui entièrement fouillé, constitue un exemple particulièrement manifeste de centre de pèlerinage dans

Muçaj 2008, pp. 382-383. L’essentiel des monnaies trouvées sur le site sont antérieures à Baduila/Totila, comme dans d’autres lieux de l’Albanie, ce qui témoignerait de l’invasion.

16

89

Skënder Muçaj, Kosta Lako, Skënder Bushi & Suela Xhyheri

Fig. 35. Amphores (VIe siècle).

l’Antiquité tardive. La multiplication des citernes pour l’accumulation des eaux ainsi que le nombre d’amphores trouvées sur le site témoignent que le nombre des pèlerins qui visitaient ce centre religieux était très important.

Le monastère des Quarante Martyrs constitue, par son architecture, son plan, ses espaces, y compris souterrains, ses techniques de construction, sans mentionner ici ses fonctions, un cas unique dans toute la Méditerranée en 90

Le monastère des Quarante Martyrs de Sébastée à Saranda (Albanie), un centre important de pèlerinage....

Fig. 36. Amphores (VIe siècle).

raison de son état de préservation, à tous les niveaux. En matière d’architecture, le modèle le plus proche est l’église Saint-Polyeucte à Constantinople.17

Si son état de préservation est exceptionnel, les données parvenues jusqu’à nous montrent que le monastère des identifie Anicia Juliana, arrière-petite-fille de l’emnpereur Théodose Ier, comme la fondatrice de l’église Saint-Polyeucte. Les fouilles ont permis de dater cette église de 508-522 et, avec les données présentées dans ces poèmes, de proposer une restitution de son plan et de ses élévations.

Bardill 2006; 2011. Un poème de l’Anthologie palatine (1.10) – ouvrage dont l’unique manuscrit date de 940 et qui offre le principal témoignage écrit concernant plusieurs églises constantinopolitaines tardo-antiques – 17

91

Skënder Muçaj, Kosta Lako, Skënder Bushi & Suela Xhyheri

Fig. 37. Amphores (VIe siècle).

Quarante Martyrs est semblable aux autres centres de pèlerinage du monde chrétien de l’Antiquité tardive. Sur le territoire albanais, il faut le rapprocher de celui que l’on trouve dans la baie de Lin, en face de Lychnida/ Ohrid.18 18

Ces deux centres de pèlerinage ont été construits à côté d’axes importants qui liaient l’Occident et l’Orient, non seulement par voie maritime, mais aussi par la voie terrestre de la via Egnatia. Les données des fouilles sur le monastère des Quarante Martyrs à Saranda et dans la baie de Lin permettent de compenser l’absence des sources écrite sur le développement du culte de ces martyrs.

Muçaj 2012b, pp. 14-15.

92

Le monastère des Quarante Martyrs de Sébastée à Saranda (Albanie), un centre important de pèlerinage....

Fig. 38. Amphores (VIIe siècle).

Bibliographie

Bardill J., 2006. ‘A new temple for Byzantium: Anicia Juliana, King Solomon, and the gilded ceiling of the church of St. Polyeuktos in Constantinople”, in W. Bowden, A. Gutteridge, C. Machado (éd.), Social and Political Life in Late Antiquity, Leyde/Boston (Late Antique Archaeology, 3/1), pp. 339-372.

Aravantinos P., 1856. Χρονογραφία της Ηπείρου, II, Athènes. Antonaras A., 2009. ‘Ορειχάλκινα ευρήματα από το ιερό της βασιλικής κάτω από την Αγία Σοφία Θεσσαλονίκης’, in M. Giannopoulou, Ch. Kallini (éd.), ᾽Ηχάδιν. Τιμητικός τόμοςγια τη Στέλλα Δρούγου, I, Athènes, pp. 116-132.

Bardill J., 2011. ‘Église Saint-Polyeucte à Constantinople : nouvelle solution pour l’énigme de la reconstitution’, in 93

Skënder Muçaj, Kosta Lako, Skënder Bushi & Suela Xhyheri J.-M. Speiser (éd.), Architecture paléochrétienne, Gollion (Grèce-Rome-Byzance. Études fribourgeoises d’histoire, d’archéologie et d’histoire de l’art), pp.  77-103.

Meksi A., 1985. ‘Arkitektura paleokristiane në Shqipëri’, in Monumentet, 30, pp. 13-44. Meksi A., 2004. Arkitektura e kishave të Shqipërisë (Shekujt VII-XV), Tirana.

Bowden W., Mitchell J., 2001. ‘The church of Forty Martyrs’, in Minerva, 31, pp. 31-33. Brown P., 1981. The Cult of the Saints: Its Rise and Function in Latin Christianity, Chicago/Londres (The Haskell Lectures on History of Religions. New Series, 2).

Mitchell J., 2004. ‘The archaeology of pilgrimage in Late Antique Albania: the basilica of the Forty Martyrs’, in W. Bowden, L. Lavan, C. Machado (éd.), Recent Research on the Late Antique Countryside, Leyde (Late Antique Achaeology, 2), pp. 145-186.

Ceka N., 1999. Butrint. A Guide to the City and its Monuments, trad. S. Martin, Londres.

Muka G., 2002. ‘Bazilika që i dha emrin Sarandës’, in Monumentet, 44, pp. 7-41.

Demetriou E. M., 1913. ‘Ὁ ναός καὶ αἱ κατακόμβαι τῶν Αγίων Τεσσαράκοντα Μαρτύρων’, in Ἱερός σύνδεσμος. Ἐκκλησιαστικόν περιοδικόν του ομώνυμου συλλόγου. Έτος Θ‘. (ΙΖ‘). Περίοδος Β‘, 198, pp. 8-10.

Muçaj S., 2008. ‘Aperçu des données archéologiques dans le territoire albanais (IVe-VIIe siècles)’, in Mélange de l’École française de Rome, Moyen Âge, 120/2, pp. 377-384.

Dyggve E., 1940. ‘Die Altchristlichen Kultbauten an der Westküste der Balkanhalbinsel’, in Atti del IV Congresso internazionale di archeologia cristiana, Città del Vaticano, 16-22 Ottobre 1938, I, Rome (Studi di Antichità cristiana, 16), pp. 391-414.

Muçaj S., 2012a. ‘Manastiri i 40 Shenjtorëve, Sarandë’, in Iliria, 36, pp. 447-449. Muçaj S., 2012b. ‘Vendet e shenjta të pelegrinazhit në Shqipëri’, in Candavia, 4, pp. 5-18. Muçaj S., Lako K., Bushi S., Xhyheri S., 2011. ‘Sigilata nga qyteti i Onhezëm-Ankiazmit (Sarandë)’, in Candavia, 3, pp. 41-103.

Evangelides D., 1919. Ἠ Βόρειος Ἤπειρος, Athènes. Feissel D., 1983. Recueil des inscriptions chrétiennes de Macédoine du IIIe au VIe siècle, Athènes (Bulletin de correspondance hellénique. Supplément, 8).

Muçaj S., Lako K., Bushi S., Xhyheri S., 2013. ‘Monumenti i 40 Shenjtorëve – 2012’, in Iliria, 37, pp. 379-386.

Gulowsen K., 2001. ‘The cult of the Forty Martyrs on the Forum Romanum’, in J. Fleischer, J. Lund, M. Nielsen (éd.), Late Antiquity. Art in Context, Copenhague (Acta Hyperborea, 8), pp. 235-248.

Muçaj S., Lako K., Bushi S., Xhyheri S., 2014. ‘Monumenti i 40 Shenjtorëve, 2013’, in Iliria, 38, pp. 321-326. Noy D. (éd.),1993. Jewish Inscriptions of Western Europe, I, Cambridge.

Hülsen C., 1927. Le chiese di Roma nel Medio Evo. Cataloghi ed appunti, Florence.

Perry S. G. F. (trad.), 1881. The Second Synod of Ephesus, Together with Certain Extracts Relating to it, Dartford.

Janin R., 1953. Le Siège de Constantinople et le Patriarcat œcuménique, III, Paris (La géographie ecclésiastique de l’Empire byzantin, 1).

Price R., Gaddis M. (trad.), 2005. The Acts of the Council of Chalcedon, Liverpool (Translated Texts for Historians, 45).

Lako K., 1984. ‘Kështjella e Onhezmit’, in Iliria, 14/2, pp. 153-205.

Thesprotos K., Psallidas A., 1964. Γεωγραφία Αλβανίας και Ηπείρου. Εξ ανέκδοτου χειρογράφου του Κωσμά Θεσπρωτού με τοπογραφικά σχεδιογραφήματα και γεωγραφικούς χάρτας του ίδιου, Ioannina.

Lako K., Muçaj S., Bushi S., Xhyheri S., 2014. ‘Anchiasmos (Onhesmos) in the 5th-7th century: city, pilgrimage centre and port”, in L. Përzhita, I. Gjipali, G. Hoxha, B. Muka (éd.), Proceedings of the International Congress of Albanian Archaeological Studies: 65th Anniversary of Albanian Archaeology (21-22 November, Tirana 2013), Tirana, pp. 613-624.

Ugolini L. M., 1927. Albania Antica, I, Rome. Wilpert J. 1916. Die römischen Mosaiken und Malereien der kirchlichen Bauten vom IV. bis XIII. Jahrhundert, II, Fribourg-en-Brisgau.

Leake W. M., 1835. Travels in Northern Greece, I, Londres. Maraval P., 1999. ‘Les premiers développements du culte des Quarante Martyrs de Sébastée dans l’Orient byzantin et en Occident’, in Vetera Christianorum, 36, pp. 193-209. Maraval P., 2004. Lieux saints et pèlerinages d’Orient. Histoire et géographie des origines à la conquête arabe, 2e éd., Paris (Cerf. Histoire).

94

7 How Early Christian Complexes Developed in Northern Dalmatia and the Kvarner Islands Ariadna Voronova Православный Свято-Тихоновский гуманитарный университет / St. Tikhon’s Orthodox University (Russia) Early Christian complexes in Northern Dalmatia and the Kvarner Islands were formed on the basis of Antiquity in three ways: first, by building upon the foundations of pagan religious structures, next, by using antique profane buildings or their parts, and third, by transforming the mausolea in private villas. Through their typological and constructional features and spatial treatment, these monuments are closely related to examples in metropoleis and the regional centres of Ravenna and Aquileia area, as well as to features of Early Byzantine architecture, which was typical of the architecture of the Adriatic cultural circle. Les édifices paléochrétiens de la Dalmatie du Nord et des îles de Kvarner trouvent leurs origines dans l’architecture antique, de trois manières différentes : la construction sur les fondations de lieux de culte païens, le remploi de bâtiments civils antiques ou de certaines de leurs parties, ou encore la conversion de mausolées en villas privées. Par leurs traits typologiques, leur construction et leur traitement de l’espace, ces monuments sont étroitement liés aux modèles des métropoles et des centres régionaux d’Aquilée et de Ravenne ainsi qu’aux caractéristiques du premier art byzantin, ce qui est typique de l’architecture du cercle culturel de l’Adriatique. Frühchristliche Sakralkomplexe in Norddalmatien und auf den Kvarner Inseln entstanden auf den Grundlagen antiker Stätten. Es können drei Varianten unterschieden werden: erstens auf den Fundamenten heidnischer Kultstätten, zweitens über profanen antiken Gebäuden bzw. Gebäudeteilen und drittens durch die Transformation von Mausolea privater Villen. Diese Denkmäler sind durch ihre typologischen und strukturellen Charakteristika bzw. ihre Raumeinteilung eng mit Denkmälern in den Metropolen und regionalen Zentren von Ravenna und Aquileia sowie mit der frühbyzantinischen Architektur des adriatischen Kulturkreises verbunden. I complessi sacri protocristiani della Dalmazia settentrionale e delle isole Quarnerine si sono formati su preesistenze antiche, in tre modi: sulle fondamenta di edifici di culto pagani, con il riutilizzo di antichi edifici profani o di loro parti, con la trasformazione di mausolei appartenenti a ville private. Questi monumenti sono strettamente legati nei loro tratti tipologici e costruttivi, così come nell’organizzazione dello spazio, ai modelli presenti sia nelle metropoli, sia nei centri regionali dell’area di Ravenna e Aquilea, ma anche ad alcuni aspetti di matrice bizantina, il che è caratteristico dell’architettura dell’ambito culturale adriatico. the same time, Early Christian complexes in Northern Dalmatia are expressive of the architecture of the local cultural circle, which gives us vivid evidence of its own morphological features. Early Christian sacral complexes in Northern Dalmatia and the Kvarner islands1 developed from Antiquity in three ways: 1) by building upon the foundations of pagan religious structures, 2) by using

The principles of formation of the Early Christian architectural heritage in Dalmatia were rooted in the Late Antique architecture of the Adriatic cultural circle, which was the meeting point of the artistic styles of the Roman East and West that were brought to the Dalmatian coast via sea routes through the harbours of the Ravenna– Aquileia coast and the Aegean Sea. In the period from the fourth to the seventh century, interrelated forms of sacral architecture arose in this cultural circle. These monuments are closely linked in their typological and constructive features, construction methods and spatial solutions; at

Chevalier 1995, pp. 68, 70, 80–84 and 97–99. The beginning of the research: Cambi 1978; 1984–85, pp. 42–44. The latest discoveries: Migotti 2012, pp. 192–197.

1

95

Ariadna Voronova parts. Examples are found on the Kvarner islands; for instance, on the island of Krk, the Christian complex is represented by the Cathedral Church of the Holy Virgin Mary of Krk (fig. 3). It was placed in an insula of ancient Curicum, along the main cardo. Roman thermae were revealed on the site to the north; a part of the thermae was adapted for the first oratory and Early Christian baptistery. South of the thermae above the palaestra, a single-aisle basilica was constructed, whose semicircular apse was located partly within the space of the church. There is an assumption that the Early Christian structure initially had a tripartite presbytery.6 The wall masonry is of cut stone with inset bricks typical of the architecture of Byzantine provinces.7

antique secular buildings or their parts, and 3) through transformation of the mausolea of private villas. The first method is found only once in the region, represented by the Christian centre of Iadera (Zadar).2 Unlike other towns of the Adriatic region, where Christian centres rose far from pagan cult centres, the Christian centre in Zadar appeared in the forum, not long after AD 313. The core of the Episcopal complex was the singleaisle basilica of St Anastasia (fig. 1),3 with a synthronon in the semicircular apse, a triumphal arch in the form of a tribelon, a baptistery,4 and a catechumenon with a synthronon (fig. 2).5 The proportion of width to length of the basilica is 2:3 (without apse), the building measures 20 x 30 m (without apse), and the inner span of the apse is wide compared to the width of the central nave; all of these traits were derived from the architecture of secular and of subsequent Christian basilicas of Rome and, later, of the whole Empire.

On the island of Cres, the Cathedral complex of St Mary in Osor took its shape gradually as well. Initially, a Christian single-nave basilica without an apse was established on the site of a pre-Christian structure, probably through adaptation of part of a former villa rustica; a baptistery was constructed in the fourth or fifth century. The complex was thoroughly renovated in the fifth and sixth centuries; an apse was added to the basilica, and the interior acquired

The second method of formation for Early Christian complexes from Antiquity in Northern Dalmatia consists in the use of the foundations of secular structures or their

Fig. 1. Zadar. St Anastasia Basilica, fourth to fifth century, twelfth century, alterations of the fourteenth to fifteenth century, and eighteenth-century interior (A. Voronova). Vežić 2005, pp. 11–13. Vežić 2002, pp. 22 and 29–30. 4 Vežić 2002, pp. 41–47 and 152. 5 Vežić 2013, pp. 32–34. 2

The transformation of thermæ into churches took place very often at that time. Cf. Doig 2008, p. 6. About the tripartite presbytery, cf. ibid., p. 77. 7 Marasović 2009, pp. 42–47. Suić 2003, p. 37.

3

6

96

How Early Christian Complexes Developed in Northern Dalmatia and the Kvarner Islands

Fig. 2. Zadar. St Anastasia Basilica, apse of сatechumenon, fourth to fifth century (A. Voronova).

the appearance of a single-aisle basilica. Another basilica with a semicircular apse was constructed to the south, which resulted in a double basilica.8

church was established by the addition of a semicircular apse, naos and narthex to one wall of a Roman house. In the fifth to sixth century a cross-shaped font was constructed. Parallel to the longitudinal oratory, yet another church was built to the north.10

Other examples of double basilicas developing from single-nave churches are found in Northern Dalmatia. Thus, in Srima near Šibenik (fig. 4), the north church and a baptistery were the first to appear; the south church was added later, with a cistern between the churches. Numerous graves were discovered inside the church and around it; they included burial chambers with entrances, benches and ossuaries. The cross-shaped baptismal font of the baptistery was later replaced by a rectangular one. The cistern was formed from the south longitudinal pastophorium that presumably functioned as a catechumenon.9 The formation of the complex of double basilicas at the Holy Trinity of Nin began when the first 8 9

The single-nave triconch is a distinct architectural type in Northern Dalmatia, derived from burial memoriae in Roman private villas. The complex in Muline on the island of Ugljian has been most studied. It consists of three objects situated 100 m distant from one another: a mausoleumaedicula; a memoria with two semicircular apses, atrium, and annexes built in stages; and a single-aisle basilica with a synthronon in the apse and pastophoria in the narthex, that later became a parish church with a baptistery.11 Given the aediculæ annexed to the initial cella, the memoria in Muline is similar to the cella in the Christian necropolis

Marasović 2009, pp. 86–93. Chevalier 1995, pp. 161, 167 and 171. Gunjača 1991, pp. 269–287. Chevalier 1995, pp. 105, 109, 169 and 171.

10 11

97

Vežić 2005, pp. 113–118. Vežić 2011, p. 29.

Ariadna Voronova

Fig. 3. Island of Krk, city of Krk, St Mary’s Cathedral, fifth-century interior (A. Voronova).

Fig. 4. Double basilicas in Srima, fourth to sixth century, apses (A. Voronova).

98

How Early Christian Complexes Developed in Northern Dalmatia and the Kvarner Islands

Fig. 5. St Martin’s Church in Pridraga, fourth to sixth century (A. Voronova).

in Salona, Manastirine, dating from the fourth through the sixth century.12

a parish church; annexes around the naos were arranged similarly, i.e., a narthex and side structures of which the southern one served as a baptistery.14 An analogous example is the triconchal memoria in Gata near Omiš (fig. 7) on the site of ancient Gedate, where the remains of a villa rustica were uncovered. The central conch, whose smooth exterior walls were divided by semicircular niches with lesenes at the corners, was initially a free-standing crossshaped memoria, with a confessio in the main apse. In its post-Justinianic period, buttressed walls were built around the triconch, so that an ambulatory was formed around the nucleus; a narthex was added as well. Lateral sections of the narthex indicate adjoining secular structures on both sides of the complex; traces of a staircase were found in the south section of the narthex.15 The Mausoleum of Galla Placidia in Ravenna, the cruciform baptistery in Pula, and burial chapels in the Basilica of St Mary Formosa in Pula may serve as counterparts,16 although these monuments are not triconches.

Unlike the memoria discussed earlier, the Church of St Martin in Pridraga (fig. 5) was not composed gradually; it was originally planned as a triconch. Nevertheless, the plan of the whole villa complex, with the hexaconchal oratory of St Michael and the Church of St Martin 200 m away, is similar to the previous example. Initially, the triconch probably served as a memoria in private ownership late in the fifth century or at the beginning of the sixth century; it was adapted from a Roman subsidiary structure. A narthex, an octagonal baptistery (fig. 6) with a protiron and hexagonal font (whose apse was added later) were built subsequently, as were some other structures; thus, an ambulatory or inner corridor was formed, which enveloped the church on three sides. As a result, the memoria was turned into a basilica complex that functioned as a parish church; it was established along with the village around an antique villa rustica and its former memoria.13

At the site of Crkvine, in Galovac, there are remains of a sacral complex dating from the sixth century; it consists of a church and a rectangular memoria, with an apse and two side structures running along the building. In the south structure, there is a tomb with a semicircular vault;

In Bilice, near Skradin, there are modest remains of a villa rustica and a Christian memoria in the shape of a triconch 100 m apart, analogous to the previous example. The memoria subsequently grew into a complex structure and Vežić 2005, p. 82. Marasović 2009, pp. 397–398. Dyggve 1996, p. 98. Dyggve 1996, pp. 85–94. Marasović 2009, pp. 238–242. Cf. also Chevalier 1995, p. 163 (project of the reconstruction with a prismatic dome at p. 83).

12

Vežić 2011, pp. 37 and 42. Vežić 2011, pp. 32–33, 44. Marasović 2011, pp. 478–480. 16 Ujčić 2005, pp. 5–42. 14

13

15

99

Ariadna Voronova

Fig. 6. St. Martin’s Church in Pridraga, fourth to sixth century, baptistery (A. Voronova).

Fig. 7. Triconch in Gata, fifth to sixth century, northern wall (A. Voronova).

100

How Early Christian Complexes Developed in Northern Dalmatia and the Kvarner Islands of private burial chapels into parish churches, e.g. the church in Gata, the triconch at Cim near Mostar from the fifth century, with extensions dating from the sixth.22 Nevertheless, the memorial tradition of the structure could have been preserved along with the liturgical functions of the parish church. This feature reflects the symbiosis of Christian religion and Roman agrarian culture.

graves were found outside the church as well. The second memoria was constructed nearby, almost similar to that in Muline, although it has a single nave, with a narthex and an apse, semicircular inside and polygonal on the outside, with a synthronon. On the south side, an annex was built simultaneously with the church; under its floor, a sarcophagus was uncovered, carved from the antique architrave of a monumental Roman structure. Another structure was added later; it served as a baptistery and included a narthex with an entrance on the south side; on the north side, a structure was added to the presbytery. A Roman villa was situated not far from the sacral complex, in the west part of the site.17

The particularity of the northern part of Dalmatia consists, first of all, in the fact that historically an Illyrian tribe, the Liburnians, inhabited this region; they were not the Delmatians, from whom the name Dalmatia stems. Hence, the region including the Kvarner archipelago was called Liburnia.23 It also included the eastern coast of the Istria peninsula, where autochthonous settlements were built according to a radial-concentric plan around a Histrian tribe ritual site located on top of a hill.24 The principle of adding buildings around a sacred centre may have influenced the formation of church complexes in Northern Dalmatia.

At the cemetery in Lepuri, on the site of the church of St Martin, a basilica complex once stood; its nucleus, initially, was a memoria or a single-nave Early Christian private chapel, with an apse, which was semicircular inside and polygonal on the outside. During the second phase, in the fifth through the seventh century, a narthex with long corridors on the sides was added, without traces of transverse walls, unlike other basilica complexes.18 Moreover, archaeological remains of several other basilica complexes located at villae rusticae survive in the region, namely, Begovačа at Donji Biljane, Vagići at Mokro Polje near Knin, Bičinа at Polačе, the church of St Nediljica at Korlati, and Glavičina at the village of Podvršje (double basilicas); many other complexes could be mentioned here as well.19

The particularity of Northern Dalmatia in matters of Christianity is accounted for by its proximity (unlike Central Dalmatia or, even more, Southern Dalmatia) to the western parts of the Roman Empire. During the spread of Christianity, Rome and Constantinople competed in their proselytising activities in those provinces that were equidistant from both capitals. The jurisdiction of each capital included a number of Christian centres that were quickly developing both spiritually and culturally; they included established ones as well as new episcopal sees. In the Early Christian centres of the Adriatic cultural circle, i.e., Aquileia, Parentium and Salona, there existed small, well-organised Christian communities as far back as the turn of the third to the fourth century. Christianity spread from those centres to smaller ones on the islands and on the continent.

The transformation of the discussed triconches in Northern Dalmatia relates to the period not prior to the middle third of the sixth century when mass baptisms occurred. During that period, a large number of earlier churches became the centres of new complexes.20 One can hardly call such buildings basilicas, bearing in mind their singlenave interior space, without side aisles; nevertheless, their exterior appearance resembles that of a basilica. Lateral structures are covered by shed roofs, while the middle one is covered by a gable roof; the middle structure rises above the lateral ones; it has window openings along its side walls, forming a clerestory.21

Salona, the ancient metropolis of Dalmatia, and its zone of influence lay immediately south of Northern Dalmatia. Salona may have played a vital role in the spread of Christianity in Northern Dalmatia. In AD 481, Salona and the whole of Dalmatia came under the influence of Arian Goths; nevertheless, Salona remained the flourishing cultural centre on the Adriatic.25 The bishop of Salona was influenced greatly by Byzantium, where he was titled as Metropolitan (that is, the head of a metropolis), as well as by Rome, which granted the bishop the Pallium more than once. But Christian Salona remained powerful and politically independent at the crossroads of the interests of East and West.26

Thus, two phases of Christian buildings survive in the rural centres of the Adriatic region. The first phase – not later than the fifth century – comprises private oratories and memoriae at the villae rusticae of wealthy citizens. The second phase, which is not earlier than the sixth century (and there are seventh-century examples), includes the adjustment of earlier churches or their annexes to serve as parish churches. In the process of that architectural development, a circular nave or aisle was formed, carried around the triconch, either in part or fully, creating some sort of ambulatory and testifying to the transformation

Salona was a growing city, where large-scale secular and religious complexes were extensively built up until the seventh century AD, when Salona was destroyed in the

Marasović 2009, pp. 411–416. Marasović 2009, pp. 265–268. 19 Marasović 2009, pp. 250–253. Vežić 2005, pp. 96–111. On the church at Bičinа in Polačе, see Chevalier 1995, pp. 97, 125, 165 and 169. 20 Migotti 2012, p. 198. 17

Vežić 2011, pp. 38–39 and 43–44. Pliny the Elder, Naturalis Historia 3.23 (19). 24 Suić 2003, p. 110–111, fig. 5–6, and p. 126. 25 Cambi 2002, p. 117. 26 Dyggve 1996, pp. 24 and 33. 22

18

21

23

Vežić 2011, p. 41.

101

Ariadna Voronova were observed at the graves in small chapels, memoriae and martyria, in whose place large basilicas were later constructed. Basilicas with an inscribed apse, like the cemetery basilicas in Zadar and Novaljiа, as well as the one on the island of Krk at Mira near Baška, which was formed by adapting part of the former villae rusticae29 – all of these examples testify to the elaborate burial cult that was important for the culture and Early Christian tradition and that was found in all significant Christian centres in the Adriatic region in the fourth and fifth century.

Slavo-Avar invasion. The architecture of those structures had a great impact on the formation of Early Christian churches in the northern part of Dalmatia. The Early Christian city centre of Salona developed through the adaptation of buildings from Antiquity; the first oratory appeared on the site of the thermae of a private house in the second half of the third century. Close to the oratory, a small basilica with a baptistery was constructed in the fourth century, while in the fifth century a new and much larger basilica was built, with a new baptistery, consignatorium and catechumenon.27 In the sixth century, the south basilica was replaced by a cruciform church, while a common narthex connected the two churches (fig. 8). Double basilicas allowed for important ceremonial interior spaces for the Church Councils of 530 and 533 that were held in Salona. The Episcopal Palace was built in the north part of the complex, which led to the development of a new city district.28

The architecture of cemeteries in the surroundings of Salona had a great impact on the development of Early Christian church complexes in Northern Dalmatia. In the necropolis of Marusinac near Salona, there are private chapels and the grave of a martyr dating from 300–305, as well as a large burial complex with the mausoleum of St Anastasios, c. AD 300 (fig. 9), a basilica c. AD 426, and private martyria.30 Salona was devastated by many barbarian raids, beginning c. AD 400; as a result, a lot of basilicas were partly destroyed and were later reduced to either the transverse nave or the central nave. The oratory of St Anastasios at the necropolis of Marusinac was the only religious structure to have survived the devastation

A distinct phenomenon of Early Christian architecture can also be found in Northern Dalmatia, namely, cemetery basilicas built at city necropoleis. They had been built over the graves of Christian martyrs and ordinary Christians since the times of persecution, when the first liturgies

Fig. 8. The Episcopal complex of Salona, fourth to sixth century (A. Voronova). 27 28

Rendić-Miočević 2011, pp. 45–55. Suić 2003, p. 365; Dyggve 1996, pp. 35–36.

29 30

102

Vežić 2005, pp. 94–96. Rendić-Miočević 2011; Marin 1994, p. 58.

How Early Christian Complexes Developed in Northern Dalmatia and the Kvarner Islands

Fig. 9. St Anastasios’ oratorium at the necropolis Marusinac near Salona, fourth century (A. Voronova).

Bibliography

in Early Christian Salona and, together with the partly dismantled churches in other necropoleis, it kept the architectural and memorial continuity of the city.31

Cambi N., 1978. ‘Starokršćanska crkvena arhitektura na području salonitanske metropolije’, in Arheološki Vestnik, 29, pp. 606–625.

As far back as the period of Antiquity, close ecclesiastical links were established between Dalmatia and Aquileia and were maintained in the fifth century, especially in the field of ecclesiastical arts. Beyond that, for Byzantium the region was always the key to the Adriatic Sea and safe sea routes there. These circumstances also account for complex cultural trends and forms that emerged in Dalmatian architecture. In their stylistic features, these structures followed the examples from metropoleis, but during the period from Theodoric to Justinian, in the provincial centres of Istria and Dalmatia one can observe the regional influences of the Ravenna and Aquileia areas; after the re-conquest of Justinian they bear traits of Early Byzantine examples, thus proving deep links with regional architecture in Macedonian lands. Culturally and stylistically, the phenomenon showed the interrelation of the traditions of the western and eastern coasts of the Adriatic Sea as parts of the Mediterranean culture, where, due to constant connections via sea, the processes of mixing of different traditions were extremely intensive. 31

Cambi N., 1984–85. ‘Arhitektura Narone i njezina teritorija u kasnoj antici’, in Radovi. Razdio povijesnih znanosti, 24 (11), pp. 33–58. Cambi N., 2002. ‘Kiparstvo’, in E. Marin (ed.) Longae Salonae, I, Split (Arheološki Muzej-Split Niz Salona, 11), pp. 117–174. Čaušević-Bully M., 2012. ‘L’architecture paléochrétienne dans le Kvarner: état des connaissances et nouvelles découvertes’, in Hortus Artium Medievalium, 18/1, pp. 133–142. Čaušević-Bully M., Bully S., 2013. ‘Organisation et architecture des sites ecclésiaux paléochrétiens de l’archipel du Kvarner (Croatie) – nouvelles pistes’, in Y. Marion, F. Tassaux (eds.), AdriAtlas et l’histoire de l’espace adriatique du VIe s. a.C. au VIIIe s. p.C.. Actes du colloque international de Rome, 4–6 novembre 2013, Bordeaux (Scripta antiqua, 79), pp. 247–272. Chevalier P., 1995. Salona. Recherches archéologiques franco-croates à Salone – Francusko-hratska

Dyggve 1996, pp. 63–70.

103

Ariadna Voronova arheološka istraživanja u Saloni, II, Rome (Collection de l’École française de Rome, 194/2). Doig A., 2008. Liturgy and Architecture. From the Early Church to the Middle Ages, Farnham/Burlington. Dyggve E., 1996. Povijest salonitanskog kršćanstva, Split (Biblioteka znanstvenih djela, 84). Gunjača Z., 1991. ‘Cisterna starokršćanske dvojne baziliki na Srimi’, in Diadora. Glasilo Arheološkoga muzeja u Zadru – Journal of the Archaeological Museum in Zadar, 13, pp. 269–290. Marasović T., 2009. Dalmatia praeromanica. Ranosrednjovjekovno graditeljstvo u Dalmaciji, II, Split (Biblioteka Knijga Mediterana, 57). Marasović T., 2011. Dalmatia praeromanica. Ranosrednjovjekovno graditeljstvo u Dalmaciji, III, Split (Biblioteka Knijga Mediterana, 65). Marin E., 1994. ‘Civitas splendida Salona. Geneza, profil i transformacija starokršćanske Salone’, in E. Marin (ed.) Salona Christiana, Split, pp. 15–28. Migotti B., 2012. ‘Early Christian archaeology of Dalmatia: The state of research and selected problems’, in O. Heinrich-Tamáska, N. Krohn, S. Ristow (eds.), Christianisierung Europas. Entstehung, Entwicklung und Konsolidierung im archäologischen Befund – Christianisation of Europe. Archaeological Evidence for its Creation, Development and Consolidation. Internationale Tagung im Dezember 2010 in BergischGladbach. Regensburg, pp. 187–212. Rendić-Miočević D., 2011. Dalmatia Christiana. Opera omnia, Zagreb/Split (Collectanea archaelogica Musei Archaeologici Zagrabiensis / Sabrani radovi – Arheološki muzej u Zagrebu, 1; Biblioteka Knjiga Mediterana, 64), pp. 205–219. Suić M., 2003. Antički grad na istočnom Jadranu, 2nd ed., Zagreb (Manualia universitatis studiorum Zagrabiensis). Ujčić Ž., 2005. Ranokršćanska Bazilika sv. Marije Formoze u Puli. Izložba – La basilica paleocristiana di S. Maria Formosa a Pola. Mostra, Pula (Arheološki muzej Istre. Katalog, 68). Vežić P., 2002. Sveti Donat. Rotonda Sv. Trojstva u Zadru, Split (Muzej hrvatskih arheoloških spomenika. Katalozi i monografije, 13). Vežić P., 2005. Zadar na pragu kršćanstva. Arhitektura ranoga kršćanstva u Zadru i na zadarskome području. Zadar. Vežić P., 2011. ‘Dalmatinski trikonhosi’, in Ars Adriatica. Časopis Odjela za povijest umjetnosti Sveučilišta u Zadru, 1, pp. 27–66. Vežić P., 2013. Episkopalni kompleks u Zadru – Episcopal Complex in Zadar, Zadar.

104

8 La nécropole rupestre de Dara (Turquie). De nouvelles perspectives de recherche Anaïs Lamesa UMR 8167 – Orient & Méditerranée (France) et Institut français d’études anatoliennes (France/Turquie)

Nihat Erdoğan* Mardin Müzesi / Musée de Mardin (Turquie) Dara est située à proximité de la ville de Mardin, en Anatolie orientale. L’histoire du village est bien connue grâce aux sources antiques et médiévales : il fut fortifié en 505-507 ap. J. C. sur ordre de l’empereur Anastase Ier. Vingt ans plus tard, la fortification fut restaurée par l’empereur Justinien. La préservation des fortifications et des citernes est remarquable. Quelques monuments et une partie de la ville ont été fouillés dans les années 1990 par Metin Ahunbay (Université d’Istanbul) et durant les années 2000 par Nihat Erdoğan, actuel directeur du musée archéologique de Mardin. En outre, le village est entouré de carrières dont le plus impressionnantes s’étendent au sud-ouest des fortifications. Certaines sont devenues des nécropoles. Plusieurs études ont été publiées sur ces fortifications ou sur le système d’adduction d’eau, quelques-unes sur les nécropoles, mais aucune sur les carrières. En nous appuyant sur l’archéologie des techniques, nous souhaitons formuler quelques hypothèses au sujet de l’occupation des carrières. Seront également discutées certaines propositions autour de la Galeri Mezer. Enfin, nous présenterons quelques inscriptions inédites. Dara is located close to the city of Mardin in Eastern Anatolia. The history of the village is well known from ancient and Byzantine written sources; it was fortified in AD 505–507 by order of the emperor Anastasius I. These fortifications were rebuilt twenty years later under the emperor Justinian I. The fortifications as well as the cisterns are exceptionally well preserved. Some monuments and parts of the village were excavated during the 1990s by Metin Ahunbay (Istanbul University) and in the 2000s by Nihat Erdoğan, current director of the Museum of Mardin. The village is furthermore surrounded by quarries, of which the most impressive extend to the south-west of the fortifications. Some of these quarries were later turned into necropoleis. While several studies have been published on the fortification and the systems of water supply as well as the necropoleis, no systematic research has been conducted on the quarries. Based on the study techniques, this paper advances some hypotheses about the use of the quarries. Some proposals are also discussed concerning the Galeri Mezer and some unpublished inscriptions are presented. Dara liegt in der Nähe der Stadt Mardin, im östlichen Anatolien. Die Geschichte der Siedlung ist aus antiken und byzantinischen Quellen gut bekannt. Sie bekam 505 n. Chr. auf Befehl des Kaisers Anastasius I. eine Befestigung. Zwanzig Jahre später wurde sie unter Kaiser Justinian I. erneuert. Die Wehranlagen und Zisternen sind außergewöhnlich gut erhalten. Teile davon sowie Teile der Siedlung wurden in den 1990er Jahren durch Metin Ahunbay (Universität Istanbul) und in den 2000er Jahren durch Nihat Erdoğan, durch den Direktor des Museums von Mardin, ausgegraben. Außerdem finden sich um die Siedlung Steinbrüche, die eindrucksvollsten liegen südwestlich der * Les auteurs souhaitent remercier le Ministère du Tourisme et de la Culture turc de les avoir autorisés à travailler à Dara ainsi que le Musée archéologique de Mardin pour son soutien logistique. Ils souhaitent également remercier chaleureusement Andrew Palmer qui les a particulièrement aidés à comprendre les inscriptions découvertes ainsi

que Jean-Pierre Grélois pour son aide précieuse et sa relecture attentive. Enfin, leur gratitude va à Elif Keser Kayaalp qui a eu la gentillesse de leur transmettre l’un de ses articles sous presse. Enfin, les auteurs souhaitent remercier les éditeurs du présent volume, pour leurs relectures attentives. N.B. Sauf mention contraire, les photographies sont d’Anaïs Lamesa.

105

Anaïs Lamesa & Nihat Erdoğan Befestigung. Einige dieser Steinbrüche wurden als Nekropolen genutzt. Sowohl die Befestigung als auch die Wasserversorgung und die Nekropolen sind gut erforscht, die Steinbrüche an sich blieben aber unberücksichtigt. Mittels neuer archäologischen Studien sollen einige Hypothesen über die Nutzung der Steinbrüche aufgestellt werden. Zudem werden die Galeri Mezer und einige unveröffentlichte Inschriften diskutiert. Dara è situata nei pressi della città di Mardin, nell’Anatolia orientale. La storia del villaggio è conosciuta grazie alle fonti scritte antiche e medievali: la fortificazione fu costruita nel 505-507 d.C. per ordine dell’imperatore Anastasio I. Vent’anni più tardi, essa fu ristrutturata sotto l’imperatore Giustiniano. Lo stato di conservazione delle fortificazioni e delle cisterne è veramente notevole. Qualche monumento e una parte della città sono stati indagati durante gli anni ‘90 da Metin Ahunbay (Università di Istanbul) e durante gli anni 2000 da Nihat Erdoğan, l’attuale direttore del museo archeologico di Mardin. Inoltre, il villaggio è circondato da cave tra cui le più impressionanti si estendono a sud-ovest delle fortificazioni. Alcune sono diventate delle necropoli. Diversi studi sono stati pubblicati sulle fortificazioni, sul sistema di approvvigionamento delle acque o sulle necropoli ma nessuna ricerca è stata condotta sulle cave. Prendendo spunto dall’archeologia delle tecniche, ci proponiamo di formulare qualche ipotesi riguardo all’uso delle cave. Saranno inoltre discusse alcune proposte interpretative sulla Galeri Mezer. Infine saranno presentate alcune iscrizioni inedite. Situé à environ 30 km à l’est de Mardin, l’actuel village d’Oğuz est bâti sur l’ancien village tardo-antique de Dara (fig. 1). Cette cité dépendait de la province romaine de Mésopotamie et de l’évêché d’Amida (actuelle Diyarbakır). Elle obtint, un temps, le statut de métropole (au VIe siècle), à la tête de trois cités suffragantes et abrita entre 527 et 532 le dux de la Mésopotamie.1

La préservation de Dara est remarquable, la plupart des vestiges présentant encore des élévations importantes (fig.  2). Les murs et les tours de la fortification sont bien visibles, notamment dans la partie méridionale du site (fig.  3)  ; les trois citernes sont en parfait état de conservation, l’acropole ainsi qu’une partie de la nécropole ont fait l’objet de plusieurs campagnes de fouille menées par Metin Ahunbay, puis par Nihat Erdoğan7. Seule l’église est en grande partie détruite et utilisée à présent comme habitat. Elle est construite au-dessus d’une des trois citernes8.

Dara fut le théâtre d’affrontements entre les Sassanides et les Byzantins tout au long du VIe siècle. La forteresse enserrant le village fut édifiée par l’empereur Anastase Ier entre 505 et 507 et reconstruite par l’empereur Justinien (527-565) vers 530, peu avant la paix éternelle signée avec Chosroès Ier.2 La ville fut par deux fois occupée par les Perses entre 573-590 et entre 605-628. Elle a été définitivement perdue par les Byzantins en 639.3

À l’extérieur des murs de fortification a été mise au jour une structure décorée d’une mosaïque de pavement dont l’emblèma portait une inscription désignant Anastase comme le fondateur de la ville (fig.  4)9. Cette structure désignée sous le vocable de « Mozaikli Yapı » est située en face d’une grande zone de carrières qui suit l’affleurement rocheux. Il s’agit de calcaires tendres formés entre l’Éocène et l’Oligocène. D’autres carrières s’observent au nord et à l’est du site10. Une partie des carrières occidentales a été transformée en nécropole. Le secteur situé à proximité des fortifications a été fouillé par le musée et ouvert au public.

Avant d’acquérir une telle importance militaire, Dara était une simple agglomération.4 Le site semble avoir été choisi pour sa richesse en eau (le Cordes traversait le village) et en pierre ainsi que pour sa position naturelle sur une colline, aisément défendable.5 Sa situation, non loin de la bourgade perse de Nisibe, paraît également avoir été un critère stratégique décisif, mentionné dans le traité passé entre Justinien et Chosroès retranscrit par Ménandre.6

7 Les résultats de fouilles sont partiellement publiés : Ahunbay 1990; Ahunbay 1991; Can, Erdoğan 2014. Pour une vue générale des travaux réalisés : Keser-Kayaalp, Erdoğan 2017. Il est à souligner que la citerne nord-ouest a été, un temps, identifiée à un grenier à blé suivant la description de Jean Malalas, Chronographia, 16.10 (éd. Thurn 2000, pp. 326-327) : Bell 1982, pp. 103-104; Whitby 1986, p. 750; Furlan 1995, p. 51; Gregory 1996, pp. 85-86, contra Preusser 1911, pp. 4546; Garbrecht 2004, pp.  112-113; Can, Erdoğan 2014, p.  357; KeserKayaalp, Erdoğan 2017, p. 158. 8 Preusser 1911, pp. 47-48; Furlan 1988; Brands 2004 considère l’ensemble église/citerne comme un complexe. Whitby estimait, en s’appuyant sur la description de Procope, que l’église était postérieure à la citerne : Whitby 1986, p. 762. Cette hypothèse semble confirmée par les récentes découvertes archéologiques : Keser-Kayaalp, Erdoğan 2017, pp. 154-156. 9 L’inscription est en cours de publication. Cf. Can, Erdoğan 2014, p. 352; Keser-Kayaalp, Erdoğan 2017. 10 Nicholson 1985, pp. 663-667, mentionne une carrière plus éloignée à proximité du village d’Ambar.

1 Gelzer 1893, p. 261; Vailhé 1907, p. 96; Honigmann 1947, p. 151; Janin 1960, col. 83-84. Voir également Mundell-Mango 1991, p. 588 (Dara). 2 Procope de Césarée, De aedificiis, 2.1.4-27 (éd. Haury 1964, pp. 46-50). 3 Les étapes de ce conflit sont relatées en Greatrex, Lieu 2002; et Garbrecht 2004, p. 108. 4 Pseudo-Josué le Stylite, 90 (éd. Chabot 1927, p. 309); Pseudo-Zacharie le Rhéteur, 7.6 (éd. Brooks, Chabot 1921, pp. 35-36); Procope de Césarée, De aedificiis, 2.1.4-27 (éd. Haury 1964, pp. 46-50); Jean le Lydien, De magistratibus reipublicae Romanae, 3.47.4 (éd. et trad. Dubuisson, Schamp 2006, p. 102); Théophane le Confessseur, Chronographia, A.M. 6000 (éd. de Boor 1883, p. 150), contra Évagre le Scholastique, 3.37 (éd. Bidez, Parmentier 1898, pp. 135-136) qui présente le site comme un champ. Sur les sources mentionnant Dara : Rist 2004. 5 Croke, Crow 1983, p. 150; Lillington-Martin 2007, p. 299. 6 Ménandre le Protecteur, 6.1, et 8 (éd. Blockley 1985, pp. 73-77 et 97). Sur les détails du conflit sur Nisibe entre les Perses et les Byzantins  : Croke, Crow 1983, p. 150; Blockley 1985, pp. 257-258, n. 59.

106

La nécropole rupestre de Dara (Turquie). De nouvelles perspectives de recherche

Fig. 1. Carte de localisation de la ville d’Oğuz.

est de proposer quelques hypothèses formulées grâce à l’observation et à l’étude des carrières et des nécropoles. Il s’agira de proposer d’une part un phasage de l’implantation des nécropoles liées intrinsèquement à l’exploitation des carrières. Une analyse fine des techniques de creusement du tombeau ‘Galeri Mezer’ permettra, d’autre part, d’étayer l’hypothèse d’Oliver Nicholson quant à l’origine du monument. Enfin, cet article sera l’occasion de présenter trois inscriptions inédites découvertes dans deux tombeaux. Il s’agit d’un prélude à une étude monographique des carrières occidentales et des nécropoles du site. La datation relative des carrières et des nécropoles Les prospections menées lors de deux missions ont permis de déterminer cinq zones de nécropoles, divisées en plusieurs secteurs. Les zones de nécropole 2 à 4 n’ont pas été fouillées. Elles sont principalement constituées de tombeaux à chambres et d’arcosolia. La nécropole 3A comporte une trentaine de sarcophages creusés directement dans le sol rocheux et deux tombeaux à dromos (fig. 5)12. Des sarcophages à ciste ont également été mis au jour dans la nécropole 5. La prospection n’est pas achevée et il nous reste à étudier la nécropole 5C. À ce jour, il a été possible de recenser 202 tombeaux à chambre et 119 arcosolia.

Fig. 2. Plan du site, repris de Metin Ahunbay avec la répartition des nécropoles (Musée archéologique de Mardin).

Le site de Dara fait l’objet de recherches depuis le début du XXe siècle, mais les nécropoles et, surtout, les carrières demeurent peu étudiées11. Le but de cet article 11 La tombe désignée comme ‘Galeri Mezer’ est l’objet principal des descriptions et des études portant sur les nécropoles. La plupart des auteurs évoquent également la morphologie des tombeaux rupestres : Preusser 1911, pp. 47-48; Mundell 1975; Nicholson 1985; Can, Erdoğan 2014, pp. 352-354; Keser-Kayaalp, Erdoğan 2017, pp. 165-171.

Désignés sous le nom de chamasoria par Keser-Kayaalp, Erdoğan 2017, p. 165.

12

107

Anaïs Lamesa & Nihat Erdoğan

Fig. 3. Fortifications du site, vues du mur oriental.

Les tombeaux à chambre peuvent être dotés d’un à huit arcosolia. Certains sont décorés en façade ou dans leur chambre.

et l’amélioration du système défensif de Dara par Justinien autour de 53015. Les carrières ont donc connu deux phases d’exploitation extensive au VIe siècle.

Selon les informations publiées jusqu’à présent, certains tombeaux peuvent être datés du VIe siècle, grâce à l’iconographie de certaines façades (présence de croix, bestiaires représentés) (fig. 6)13. L’occupation de la zone 5 semble s’étaler entre le VIe et le VIIIe siècle, comme l’attestent les objets trouvés en fouille14. Mais peut-on être plus précis ? Nous pensons, en effet, que l’étude de l’organisation des carrières peut être un indice important pour préciser la datation de l’implantation des nécropoles.

Une première hypothèse peut donc être formulée : les nécropoles s’étendant sur environ 1/3 de la superficie globale des carrières, les tombeaux auraient pu être réalisés après la seconde phase d’exploitation des carrières, donc après les années 530. Néanmoins, la fouille de la nécropole 5 a mis au jour deux étages de tombeaux à chambre (fig. 7). Le premier étage suit la ligne de crête ou, dans certains cas, se situe à plus de 3 m du sol ; le second est au niveau du sol actuel. Ces deux étages de tombeaux semblent correspondre à deux phases d’exploitation des carrières. En effet, l’implantation suit très clairement une reprise d’extraction, visible le long de la paroi. Cette reprise sépare très nettement les deux étages de tombeaux. Elle met en lumière la présence de deux sols de carrières, le premier ayant disparu avec la seconde phase d’exploitation.

Les carrières occidentales du site sont exploitées de manière organisée, ce qui prouverait une exploitation extensive. Elles s’étendent sur plus de 1 km. Il semble que plusieurs ateliers aient œuvré en même temps : en effet, des secteurs se dessinent très clairement sur la photographie satellite. Au regard de l’histoire du site, deux événements expliquent cette organisation rationnelle des carrières. Le premier peut être identifié à la construction de la forteresse par Anastase entre 505 et 507. La réalisation de murs de plus de 10 m, jalonnés de tours, a créé une forte demande de pierres de taille. La proximité des carrières ouest avec la fortification conforte l’hypothèse de leur exploitation durant cette première phase. Le second événement est la restauration

Cette donnée permet d’affiner encore l’hypothèse de départ : une première phase d’implantation des nécropoles pourrait avoir eu lieu entre 507 et les années 520, et elle se situerait logiquement après la première phase d’exploitation des carrières. Une seconde phase 15 Voir le tableau récapitulatif dans Garbrecht 2004, p. 108, contra Gregory 1996, p. 80. Procope ne mentionne pas le siège de Dara qui précéderait la bataille de 530 entre les Byzantins et les Perses.

Mundell 1975, pp. 210 et 226-227. 14 Keser-Kayaalp, Erdoğan 2017, p. 171. 13

108

La nécropole rupestre de Dara (Turquie). De nouvelles perspectives de recherche

Fig. 4. Inscription de fondation en mosaïque (Musée archéologique de Mardin).

d’occupation aurait eu lieu après la restauration de Justinien, donc à partir de l’an 532.

funéraires. Cette année, en effet, le musée a déblayé un deuxième secteur, la nécropole 1, qui ne possède pas de second niveau (fig.  8). Pour autant, cette observation n’infirme pas la proposition. Elle permet simplement de signaler que la nécropole 1 a été creusée en une seule fois.

Si cette hypothèse de datation s’applique à la nécropole 5, elle demande à être confirmée pour les quatre autres zones 109

Anaïs Lamesa & Nihat Erdoğan

Fig. 5. Vue aérienne de la zone 3A (Musée archéologique de Mardin).

Sa localisation excentrée, à l’extrémité ouest de la zone des nécropoles, pourrait expliquer sa datation plus tardive ce qui la placerait dans la seconde phase d’implantation des nécropoles, c’est-à-dire après 532 : la carrière dans laquelle elle est réalisée pourrait en effet avoir été ouverte durant la seconde phase de construction du site. Toutefois,

aucune inscription ni décoration en façade ne permet de placer ces tombeaux dans la lignée des autres tombeaux rupestres, si ce n’est la morphologie de ses tombeaux dotés d’une chambre funéraire à arcosolia et la présence d’arcosolia autonomes creusés à même la falaise que l’on retrouve dans les autres zones funéraires du site.

Fig. 6. Exemple d’un relief tardo-antique ornant la tombe N3C T05.

110

La nécropole rupestre de Dara (Turquie). De nouvelles perspectives de recherche

Fig. 7. Nécropole 5A avec les deux étages de tombeaux.

Fig. 8. Nécropole 1 constituée d’un seul étage de carrière.

La technique de creusement de la ‘Galeri Mezer’, confirmation d’une inspiration perse

du VIe siècle par M. Mundell-Mango, par O. Nicholson et plus récemment par E. Keser-Kayaalp et N. Erdoğan16. Il pourrait avoir été réalisé par les exilés de Dara revenus

Cette hypothèse de datation ne prend pas en considération la ‘Galeri Mezer’, tombeau monumental rupestre localisé dans la zone de nécropole 5. Celui-ci est daté de la fin

Mundell 1975, p.226-227; Nicholson 1985, pp. 670-671; KeserKayaalp, Erdoğan 2017, p. 171 (reste plus prudents quant à l’utilisation du tombeau comme ossuaire dès la période tardo-antique).

16

111

Anaïs Lamesa & Nihat Erdoğan Deux éléments peuvent être mis en avant grâce à cette première observation : d’une part, la ‘Galeri Mezer’ n’appartient pas à la première phase d’occupation de la nécropole 5. En effet, les tombeaux en hauteur se situent bien plus à l’ouest ou dans la nécropole 5B, qui se situe au nord de l’enclave. D’autre part, la correspondance entre le nu de la paroi du tombeau N5A T24 et le pan nord de la falaise où se trouve le tombeau N5A T30 prouve que le puits a été creusé a posteriori, une fois la deuxième phase d’occupation de la nécropole 5 réalisée. Enfin, la disposition en cour de la ‘Galeri Mezer’ et des autres salles rupestres qui lui sont associées ne correspond pas à l’organisation interne de la nécropole 5A. Les tombeaux sont tous alignés les uns par rapport aux autres, N5A T30 prouvant qu’il existait très probablement des tombeaux sur le pan maintenant ouvert par l’enclave. Il est donc possible d’en déduire que le puits a détruit une partie de la nécropole. L’hypothèse d’une réalisation plus tardive est confortée par la réminiscence d’une plateforme de creusement qui ne suit pas le premier sol de carrière

de Perse autour 590. Au regard de sa situation dans la nécropole 5A et des techniques de creusement employées pour sa réalisation, il est possible d’abonder dans le sens de cette hypothèse. L’exécution de ce monument est, à bien des égards, beaucoup plus technique que celle des autres tombeaux rupestres. Son emplacement au fond de la zone 5A résulte d’un choix précis. En effet, la ‘Galeri Mezer’ se situe dans une enclave dessinée dans la paroi de la carrière, une sorte de puits creusé depuis le haut de la falaise (fig.  5). L’enclave forme une cour où s’ouvre la porte originale, surmontée de reliefs. En avant de cette enclave, on observe à l’ouest un tombeau creusé au niveau du sol actuel (N5A T24). Le nu de la paroi, où est excavé ce tombeau, est au même niveau que le retour formé par le pan nord de la falaise qui délimite l’enclave. Ce pan nordest également creusé d’un tombeau (N5A T30). Enfin, à une cinquantaine de mètres à l’ouest débute le premier étage de tombeaux (fig. 9).

Fig. 9. Vue de l’enclave de la Galeri Mezer et des tombeaux N5A T24 et N5A T30.

112

La nécropole rupestre de Dara (Turquie). De nouvelles perspectives de recherche

Fig. 10. Plateforme de guidage en-dessous du sol de carrière de la première phase et encastrement d’un échafaudage creusé dans la paroi.

(fig. 10). Cette plateforme est visible juste au-dessus de la façade sculptée et était accessible par le haut grâce à des encoches. Il était donc nécessaire de descendre plus bas que le premier sol de carrière pour atteindre la zone de façonnage de la façade de la ‘Galeri Mezer’.

Enfin, la proportion du monument s’éloigne très nettement des autres tombeaux du site. En effet, les monuments sont généralement de petite taille et leurs façades ne dépassent pas 2 m de hauteur. La monumentalité des tombeaux rupestres est un fait rare à partir de l’époque romaine. Sur le territoire de l’actuelle Turquie, cette pratique est d’ailleurs propre aux tombeaux d’époque grecque classique ou de la période hellénistique, par exemple en Carie18. Pour les périodes plus récentes, on observe une disparition progressive des façades monumentales des tombeaux par exemple en Cappadoce, en Paphlagonie ou dans le sud-ouest et l’augmentation de leur concentration dans les nécropoles19. Ainsi, par sa forme, l’iconographie et le style de sa façade ainsi que les techniques employées pour la réaliser, la ‘Galeri Mezer’ s’écarte des traditions de creusement rupestre de Dara, mais également des

La présence de cette plateforme au niveau de la façade monumentale rapproche ce tombeau des monuments de Perse et d’Arabie. Outre le style de la façade proche du relief de Chosroès II à Taq-e Bostan, mentionné par M. Mundell-Mango, le creusement d’une niche pour implanter le tombeau renvoie aux pratiques observées à Qadamgah (Perse) ou à Pétra et Hégra (Nabatène)17. Ce choix rompt avec les techniques utilisées à Dara, hors de la ‘Galeri Mezer’, où les tombeaux ne sont pas implantés dans une conque ou une cavité, mais directement sculptés suivant le nu de la paroi.

18 17

Mundell 1975, p. 227; Bessac 2007, pp. 194-197; 2015, p. 171.

19

113

Henry, 2009. Von Gall 1966; Lamesa 2016.

Anaïs Lamesa & Nihat Erdoğan Inscription n° 2, paroi sud :

autres nécropoles rupestres d’époque romaine ou tardoantique observées en Turquie. Son rapprochement avec des monuments nabatéens ou perses semble confirmer, par ailleurs, que son modèle a pu être perse. L’idée d’Oliver Nicholson, qu’elle ait été façonnée par des tailleurs de pierre de ces régions, retournés d’exil à la fin du VIe siècle, paraît tout à fait plausible.

† ὅπλον ἀκαταμάχητον † Traduction : « l’arme invincible » La hauteur des lettres est de 6 cm environ. Les croix initiale et finale sont visibles, ce qui permet d’affirmer que l’inscription est complète.

Les inscriptions inédites Trois inscriptions ont été découvertes à l’occasion de ces deux campagnes de prospection20. Nous livrons ici uniquement une proposition de lecture de ces inscriptions et quelques observations qui leur sont associées.

L’expression « arme invincible » se trouve dans une prière de l’office byzantin célébré pour la fête de l’Exaltation de la Croix (14 septembre)  : «  Σταυρὸς [...] ὅπλον ἀκαταμάχητον ». Elle est attestée dans d’autres nécropoles, notamment une tombe à ciste dans le parc archéologique de Capo Boeo21. L’inscription italienne est peinte en rouge, débutée et achevée par une croix ; elle est peinte au-dessus d’une croix inscrite dans un cercle. Le corps d’un adulte a été retrouvé dans cette tombe datée de la fin du VIe siècle ou du début du VIIe siècle.

Les deux premières inscriptions sont situées dans le tombeau N2A T01. Il s’agit d’un tombeau localisé dans la nécropole 2A. Ce tombeau est profondément enfoui. Il est à chambre unique et dépourvu d’arcosolium. Il a toutefois la particularité d’être décoré de deux croix en relief situées sur ses parois orientale et méridionale.

Le rapprochement de la Croix avec l’inscription pourrait confirmer notre lecture. Les deux inscriptions, qui paraissent contemporaines tant dans la forme des lettres que dans leur position dans l’espace, pourraient ainsi attester le caractère funéraire de la chambre. Elle aurait abrité le corps d’une certaine Anna, que l’on a voulu protéger pendant son voyage vers l’au-delà, par l’invocation de la sainte Croix en tant qu’« arme invincible ».

Les deux inscriptions sont situées juste au-dessous des deux croix en relief. Elles sont peintes toutes deux en rouge, sur une ligne. Inscription n° 1, paroi orientale : † νεκρὸν Ανα[ς...]

La troisième inscription se situe dans le tombeau N4C T04 de la nécropole 4C (fig.  11). Il s’agit d’un tombeau profondément enfoui, dont la façade ne présente pas l’habituel décor en arche. La chambre funéraire est dotée de trois arcosolia ornés de croix peintes en rouge.

Traduction : « le corps d’Ana » La hauteur des lettres varie entre 4 et 6 cm. Une croix initiale est peinte. La croix finale n’est pas visible ce qui pourrait indiquer que la lecture de l’inscription est incomplète.

Fig. 11. Inscription du tombeau N4C T04. 20 L’ensemble a été remis à Andrew Palmer qui réalise actuellement un recueil des inscriptions syriaques et grecques de Dara.

21

114

SEG, 57, 880 = AE, 2007, 679c.

La nécropole rupestre de Dara (Turquie). De nouvelles perspectives de recherche L’inscription est peinte en rouge au-dessus de l’arcosolium oriental et elle se développe sur deux lignes. La première ligne est en onciale, alors que la seconde est en cursive, ce qui signale deux phases dans sa réalisation. La lecture des deux lignes est difficile.

à l’ouest du site, par exemple, où aucune nécropole n’a pu être observée, permettrait de comprendre pourquoi les tombeaux ont été creusés dans une zone restreinte. De même, la fouille des nécropoles pourrait affiner encore la datation de leurs implantations.

Inscription n° 3 :

Bibliographie



Σαβᾶ ἀδελφοί† ὑ(ι)οῦ Μὰρ Ηβαδα

Abréviations :



κ[αι]...υ

AE = Année épigraphique. KST = Kazı Sonuçları Toplantısı.

Traduction de la première ligne : « les frères de Sabas, fils de Mar Èbada ».

SEG = Supplementum epigraphicum Graecum. Ahunbay M., 1990. ‘Dara-Anastasiopolis’, in KST, 12/1, pp. 391-397.

La hauteur des lettres varie entre 5 et 9 cm.

Ahunbay, M., 1991. ‘Dara-Anastasiopolis 1990 Yılı Çalışmaları’, in KST, 13/1, pp. 197-203.

Le nom « Marebada » pourrait être une transcription grecque d’un nom syriaque théophore hypothétique de la forme Mary cAvdeh et aurait pour sens « Mon Seigneur l’a créé (à savoir : l’homme qui porte ce nom) ». Plus plausible, pourtant, serait le nom Marcavda, laïc baptisé sous l’invocation d’un saint homme du nom cAvda (autre forme cAbdo, dont le sens est esclave). Il s’agit d’un nom assez fréquent que l’on comprend au sens d’esclave volontaire de Dieu, « Mar » pouvant être l’équivalent du titre d’un saint occidental. Cette sorte de nom est bien connue chez les Syriaques. Plusieurs saints syriaques portent d’ailleurs le nom d’Avda22.

Bell G. L., 1982. Churches and Monasteries of the Tur ‘Abdin, réimpr. avec préface et notes de M. C. MundellMango, Londres. Bessac J.-C., 2007. ‘Étude technique et interprétations du monument rupestre de Qadamgah (FARS)’, in Iranica antiqua, 42, pp. 185-206. Bessac J.-C. 2015. ‘Artisans, techniques et économie des chantiers rupestres de Hégra’, in L. Nehmé (dir.), Les tombeaux nabatéens de Hégra, I, Paris (Épigraphie et archéologie, 2/1), pp. 163-201.

Il n’est pas possible de dater cette inscription : la présence d’un alpha à traverse brisée et apices ne suffisent pas, ces formes apparaissant dès la période hellénistique. De plus, les noms hagiophores relevés dans les inscriptions syriaques de Dara (Marabraham et Margabril/Margabriel) et théophores (Maratqen signifiant « le seigneur a établi ») dateraient vraisemblablement du VIIe ou du VIIIe siècle. Nous proposons donc de placer l’ensemble des inscriptions grecques de Dara au début du VIIe siècle23, période où elle appartenait encore à Byzance.

Bidez J., Parmentier L. (éd.), 1898. The Ecclesiastical History of Evagrius with the Scholia, Londres. Blockley R. C. (éd. et trad.), 1985. The History of Menander the Guardsman, Liverpool (ARCA. Classical and Medieval Texts, Papers, and Monographs, 17). Brands G., 2004. ‘Ein Baukomplex in Dara-Anastasiopolis’, in Jahrbuch für Antike und Christentum, 47, pp. 144155. Brooks E. W., Chabot J.-B. (éd.) 1921. Historia Ecclesiastica Zachariae Rhetori vulgo adscripta, II, Paris/Louvain (Corpus scriptorum christianorum orientalium, 84; Scriptores syri. Series tertia, 39).

Conclusion L’étude des carrières de Dara offre des informations intéressantes pour comprendre tant les phases d’occupation des nécropoles que l’histoire du site. Il semble qu’au moins une zone de carrière ait connu deux phases d’extraction, ce qui explique la présence d’un étage dans la nécropole. Le processus de réalisation de la ‘Galeri Mezer’ permet d’abonder dans le sens d’O. Nicholson et de considérer cet ouvrage rupestre comme issu de la tradition perse. Enfin, les inscriptions semblent confirmer la fonction funéraire des monuments rupestres. Elles demandent toutefois à être davantage étudiées.

Can B., Erdoğan N., 2014. ‘Dara. Bisans-Sasani sınırında bir garnizon kenti ve kazıları’, in H. Kasapoğlu, M.  A.  Yılmaz (éd.), Anadolu’nun zirvesinde türk arkeolojisinin 40 Yılı, Ankara, pp. 347-371. Chabot J.-B. (éd.), 1927. Incerti Auctoris Chronicon Pseudo-Dionysianum vulgo dictum, I, Paris (Corpus scriptorum christianorum orientalium, 91; Scriptores syri. Series tertia, 1). Croke B., Crow J., 1983. ‘Procopius and Dara’, in Journal of Roman Studies, 73, pp. 143-159.

Cette recherche sur le site de Dara n’en est qu’à ses débuts et demande à être poursuivie. La fouille des carrières plus 22 23

de Boor C. (éd.), 1883. Theophanis Chronographia, I, Leipzig.

Voir, par exemple, Fiey 2004, pp. 17-19, nos 6-8. Keser-Kayaalp, Erdoğan 2017, p. 171.

115

Anaïs Lamesa & Nihat Erdoğan Dubuisson M., Schamp J. (éd. et trad.), 2006. Jean le Lydien, Des magistratures de l’État romain, II, Paris (Collection des universités de France. Série grecque, 452).

a Colloquium held at Potenza, Acerenza and Matera, Italy, (May 2005), Oxford (BAR. International Series, 1717), pp. 299-312.

Fiey J. M., 2004. Saints syriaques, éd. L. I. Conrad, Princeton (Studies in Late Antiquity and Early Islam, 6).

Mundell M. C., 1975. ‘A sixth-century funerary relief at Dara in Mesopotomia’, in Jahrbuch der Osterreichischen Byzantinistik, 24, pp. 209-227.

Furlan I., 1988. ‘Oikema Katagheion. Una problematica struttura a Dara’, in C. Barsanti, A. Guiglia Guidobaldi, A. Iacobini (éd.), Gruppo nazionale di coordinamento C.N.R. ‘Storia dell’arte e della cultura artistica bizantina’. Atti della Giornata di Studio, Roma, 4 dicembre 1986, Rome (Milion. Studi e ricerche d’arte bizantina, 1), pp. 105-127.

Mundell-Mango M. C., 1991. ‘Dara’, in A. P. Kazhdan (dir.), The Oxford Dictionary of Byzantium, Oxford/ New York, p. 588. Nicholson O., 1985. ‘Two notes on Dara’, in American Journal of Archaeology, 89/4, pp. 663-671. Preusser C., 1911. Nordmesopotamische Baudenkmäler altchristlicher und islamischer Zeit, Leipzig (Wissenschaftliche Veröffentlichung der Deutschen Orient-Gesellschaft, 17/1-2).

Furlan I., 1995. ‘Cisterne a Dara’, in A. Iacobini, E. Zanini (éd.), Arte profana e arte sacra a Bisanzio, Rome (Milion. Studi e ricerche d’arte bizantina, 3), pp. 51-63. Garbrecht G., 2004. ‘Procopius und die Wasserbauten von Dara’, in C. Ohlig (éd.), Wasserbauten im Königreich Urartu und weitere Beiträge zur Hydrotechnik in der Antike, Norderstedt (Schriften der Deutschen Wasserhistorischen Gesellschaft e.V., 5), pp. 105–132.

Rist J., 2004. ‘Der Bau der ostsyrischen Stadt Dara (Anastasiupolis): Überlegungen zum Eigengut in der Kirchengeschichte des Ps.-Zacharias Rhetor’, in M. Tamcke (éd.), Syriaca II: Beiträge zum 3. Deutschen Syrologen-Symposium in Vierzehnheiligen 2002, Münster (Studien zur Orientalischen Kirchengeschichte, 33), pp. 243-266.

Gelzer H., 1893. ‘Ungedruckte und wenig bekannte Bistümerverzeichnisse der orientalischen Kirche’, in Byzantinische Zeitschrift, 1, pp. 245-282. Greatrex G., Lieu S. N. C. (éd.), 2002. The Roman Eastern Frontier and the Persian Wars, II, Londres/New York.

Thurn I. (éd.), 2000. Ioannis Malalae Chronographia, Berlin (Corpus fontium historiae byzantinae. Series Berolinensis, 35).

Gregory S. (dir.), 1996. Roman Military Architecture on the Eastern Frontier, II, Amsterdam.

Vailhé S., 1907. ‘Une notitia episcopatuum d’Antioche du Xe siècle’, in Échos d’Orient, 63, pp. 90-101.

Haury J. (éd.), 1964. Procopii Caesariensis opera omnia, IV, réimpr. anast. avec add. et corr. de G. Wirth, Leipzig (Bibliotheca scriptorum Graecorum et Romanorum Teubneriana).

Von Gall H., 1966. Die paphlagonischen Felsgräber. Eine Studie zur kleinasiatischen Kunstgeschichte, Tübingen (Istanbuler Mitteilungen. Beiheft, 1). Whitby M., 1986. ‘Procopius’ description of Dara (“Buildings” II 1-3)’, in P. Freeman, D. L. Kennedy (éd.), The Defence of the Roman and Byzantine East. Proceedings of a colloquium held at the University of Sheffield in April 1986, Oxford (BAR. International Series, 297; British Institute of Archaeology at Ankara. Monograph, 8), pp. 737–783.

Henry O., 2009. Tombes de Carie. Architecture funéraire et culture carienne VIe-IIe siècle avant J.-C., Rennes (Archéologie et culture). Honigmann E., 1947. ‘The Patriarchate of Antioch: A revision of Le Quien and the Notitia Antiochena’, in Traditio, 5, pp. 135-161. Janin R., 1960. ‘Dara’, in R. Aubert, É. Van Cauwenbergh (dir.), Dictionnaire d’histoire et de géographie ecclésiastiques, XIV, Paris, col. 83-84. Keser-Kayaalp E., Erdoğan N., 2017. ‘Recent research on Dara/Anastasiopolis’, in E. Rizos (éd.), New Cities in Late Antiquity. Documents and Archaeology, Turnhout (Bibliothèque de l’Antiquité tardive, 35), pp. 151-172. Lamesa A., 2016. D’une Cappadoce à l’autre (Ve av.Xe ap.) : problèmes historiques, géographiques et archéologiques, thèse de doctorat, Université ParisSorbonne. Lillington-Martin C., 2007. ‘The archaeological and literary evidence for a battle near Dara, AD 530: topography, text and trenches’, in A. S. Lewin, P. Pellegrini (éd.), The Late Roman Army in the Near East from Diocletian to the Arab Conquest. Proceedings of 116

9 Tracing the Cycladic Settled Landscape in Late Antiquity and the Early Byzantine Period (Fourth to Ninth Century AD): The Islands of Paros and Naxos Konstantinos Z. Roussos Ινστιτούτο Μεσογειακών Σπουδών / Institute for Mediterranean Studies (Greece) This paper offers a fresh approach to the history and archaeology of the Cyclades in Late Antiquity and the Byzantine Early Middle Ages in the light of current archaeological investigations. The neighbouring islands of Paros and Naxos are used as case studies. A new question needs to be answered: to what extent the local diversities of the Parian and Naxian landscapes determined the historical evolution of the settlement pattern and the ways in which these communities interacted with broader worlds under specific conditions and over particular periods of time. Ce travail a pour objectif principal de présenter la problématique et la méthodologie employée pour l’étude du réseau urbain des Cyclades au cours de l’Antiquité tardive et de l’époque protobyzantine, à la lumière des approches actuelles de l’archéologie des paysages et de l’« archéologie insulaire ». Les îles de Paros et de Naxos servent d’étude de cas. Les questions centrales de cette recherche sont de savoir si et à quel degré les différences du paysage ont influencé l’évolution interne du réseau urbain ainsi que de comprendre la façon dont les îles ont interagi avec d’autres régions, insulaires ou continentales. In diesem Aufsatz wird ein neuer historisch-archäologischer Zugang zu der Erforschung der Kykladen während der spätantik-frühbyzantinischen Zeit im Lichte aktueller landschaftsarchäologischer Analysen vorgestellt. Die benachbarten Inseln von Paros und Naxos dienen dazu als Fallbeispiele. Das zentrale Interesse dieser Forschungen zielte auf die gegenseitige Bedingtheit der landschaftlichen Charakteristika und der Besiedlung sowie die Interaktionen zu untersuchen, die es zwischen dieser Insel und den angrenzenden anderen insularen und kontinentalen Regionen gegeben hat. L’obiettivo principale del presente studio è quello di creare un quadro per la riflessione ed una metodologia a riguardo dell’indagine sulla rete di insediamenti Cicladici durante la tarda antichità ed il periodo proto-bizantino alla luce degli approcci moderni dell’“archeologia del paesaggio” e dell’ “archeologia insulare”. Le isole di Paros e di Naxos vengono utilizzate come casi di studio (case-studies). È interessante indagare se, ed in quale grado, le differenze del paesaggio hanno influenzato lo sviluppo interno della rete urbanistica così come i modi in cui le isole interagiscono con mondi più ampi.

Introduction

the conceptual and methodological models of island and landscape archaeology.1

The group of islands known today as the Cyclades is located in the south-central Aegean Sea (fig. 1). Despite the fact that research has made important steps in studying the insular societies of the Aegean, until recently the increasing volume of archaeological data regarding Late Antiquity and the Early Byzantine period from the Cyclades remained without interpretation on the basis of

The aim of this paper is to offer a fresh approach to the history and archaeology of the Cyclades in Late 1 The chronological framework followed by this study is as follows: i) Late Antiquity (from the early fourth to the mid-seventh century AD, and ii) the Early Byzantine period or the so-called ‘Byzantine Dark centuries’ (from the late seventh to the ninth century AD).

117

Konstantinos Z. Roussos

Fig. 1. Map of the Aegean and the Cyclades.

Antiquity and the Early Byzantine period in the light of current archaeological investigations, establishing a new framework for the discussion and examination of the material culture dated between the fourth and the ninth century AD. The methodology proposed is an interdisciplinary approach, which combines archaeological evidence, literary sources, and observations of the sites and micro-landscapes as a whole, with the advantages offered by the application of innovative technologies in archaeological research (Geographic Information Systems). The diversity of the Cycladic islands makes clear that these geographical entities can be approached as individual case studies, and can subsequently be integrated in wider contexts. In order to illustrate this diversity, Paros and Naxos, two of the most highly interactive microcosms in the Aegean, will be used as case studies (fig. 2).

remains (religious buildings and a limited number of public or private structures), marble sculpture, fresco decorations, ceramic material, cemeteries and workshops.2 Putting all these dots on the map for the first time makes it evident that the settled landscapes of Paros and Naxos over the course of the Late Antique centuries consisted of urban centres and a busy countryside (fig. 3–4). The ancient city-ports of Paros and Naxos, each placed almost in the middle of its respective island’s west coastline (fig. 5–6), were still the main socio-economic, administrative and religious reference points, acting as the largest centres within the internal settlement networks. At the same time, they progressively emerged as regional naval nodes within the broader maritime commercial networks of the eastern Mediterranean world. A great

The settled landscapes of Paros and Naxos during Late Antiquity

2 This study is drawn from my PhD dissertation, submitted at Leiden University (2011–2017). This chapter does not intend to discuss the archaeological material in detail, but focuses on its synthetic analysis and interpretation. The thesis was published in 2017 as a monograph in the series Archaeological Studies Leiden University (ASLU). For more details, see Roussos 2017.

The available archaeological material from the islands of Paros and Naxos is composed mostly of architectural 118

Tracing the Cycladic Settled Landscape in Late Antiquity and the Early Byzantine Period

Fig. 2. Map of Paros and Naxos.

deal of imported ceramic evidence3 and numerous marble sculptures4 testify to commercial connections between the city of Paros and the famous production centres of this period, such as Constantinople, mainland Greece, North Africa, Asia Minor, Cyprus, Syria and Palestine.

well-connected world. On the other hand, it reflects the local ecclesiastical elite’s growing civic power and its new role in influencing regional building projects and city management (fig. 7).6 In the rural context, evidence from both islands suggests a busy countryside with intensively exploited landscapes occupied by a wide range of agricultural sites (fig. 3–4). A closer examination of the natural features of the majority of the sites makes clear a consistent correlation between settlement density and land use zoning, showing greater preference for well-watered and fertile arable or grazing7 land. The application of multiple viewshed analysis to most Late Antique rural sites on western Naxos has shown that each one overlooks a specific fertile productive zone, reflecting a form of visual control between the settlements and the surrounding fertile land (fig. 8). In both case studies, this network co-existed with a number of sea-oriented sites, such as ceramic workshops, coastal settlements and small anchorages, which served as collecting, processing and exporting points for the agricultural surplus produced. Evidence from the Bay of Naoussa in North Paros has shown that from the middle sixth century AD along the

It is worth mentioning that the city of Paros reflects the progressive transformation of a pagan urban centre to a Christian coastal city of commercial importance (fig. 5). The erection of the giant religious complex of Ekatontapiliani in the mid-sixth century AD illustrates the new focus on Christian churches and the articulation of the urban space around them, which can be seen in many of the Empire’s other cities during the course of Late Antiquity.5 The dominant and advantageous position of this magnificent complex, in close proximity to the port, indicates that its construction can be viewed as an act of propaganda by the central imperial authority in the context of the Mediterranean’s highly interactive and 3 For a preliminary study of the ceramic material from the ancient port of Paros (modern Parikia), see Roussos 2017, pp. 65–69. My deepest acknowledgements should be addressed to Ioanna Kraounaki, archaeologist of the Ephorate of Underwater Antiquities of Greece, for kindly granting me permission to study the ceramic material from the old rescue excavation of the Ephorate in the port of Parikia (1999–2000). 4 Mitsani 1996–97. 5 For a general overview of this phenomenon, cf. Saradi 2006, pp. 385– 406.

Roussos 2017, pp. 280–283. Mostly for the mountainous island of Naxos. For more, see Roussos 2017, pp. 255–256. 6 7

119

Konstantinos Z. Roussos

Fig. 3. Distribution of Late Antique sites on the island of Paros.

Paros, Naxos and the Cyclades within the interactive Late Antique Mediterranean world

coastline there flourished a well-organised centre for mass production of amphorae (fig. 9).8 These ceramic workshops were associated with a number of rural sites located across the surrounding farming zone and several port installations. This spatial relationship reflects a strong connection between production of goods, manufacture of amphorae, and inter-regional commercial networks, suggesting highly intensive commercial farming and a distinct market orientation.

Archaeological evidence from Paros and Naxos dated to Late Antiquity, especially during the sixth and seventh centuries AD, presents a picture of economically active coastal cities in parallel with a dense network of various sites in the countryside, which fits well with comparative material from other Aegean islands.9 These complex Melos: Renfrew, Wagstaff 1982, p. 145; Keos: Cherry et al. 1991; Thera: Gerousi 2001; Antikythera: Bevan, Conolly 2013; the Dodecanese: Deligiannakis 2008, 2016.

9

For a detailed interpretation of the archaeological material from the Bay of Naoussa, see Roussos 2017, pp. 91–113. 8

120

Tracing the Cycladic Settled Landscape in Late Antiquity and the Early Byzantine Period

Fig. 4. Distribution of Late Antique sites on the island of Naxos.

121

Konstantinos Z. Roussos

Fig. 5. Roman and Late Antique city of Paros.

Fig. 6. Roman and Late Antique city of Naxos.

122

Tracing the Cycladic Settled Landscape in Late Antiquity and the Early Byzantine Period

Fig. 7. The religious complex of Ekatontapiliani and its relationship with the port area.

settlement trends can be understood and interpreted in the framework of the political and economic circumstances pertaining to the eastern Mediterranean during Late Antiquity.

diversity and distinctiveness of each island resulted in the creation of highly interactive microcosms that were integral links in the chain of the broader Late Antique socioeconomic system. The large involvement of the Cyclades’ insular communities in trade patterns and sea routes as well as the effect of the economic activities across the Eastern Mediterranean’s unified economic space during Late Antiquity had, therefore, a strong impact on the daily life of the islands between the fifth and the seventh century AD. This is clearly reflected in the settlement pattern of Paros and Naxos with the development of a dense network of various types of sites, in order to support the islands’ role as smaller links in the chain of Late Antiquity’s unified economic system.

After the establishment of Constantinople as the new growing metropolis, the Aegean Archipelago assumed a crucial role as the busiest maritime zone within the eastern Mediterranean basin.10 The Late Antique economic system was developed as a unified economic space, divided into compartments in order to serve the increasing needs of the large eastern urban centres and the numerous frontier armies.11 A plethora of economic, social and cultural subregions were created and flourished through this complex network. Many regions specialised in a single product, mainly wine, oil and cereals, produced in vast quantities for export purposes.

Paros and Naxos in the Early Byzantine period: Two diametrically opposed stories

In this general context, the growth of maritime commercial economy during Late Antiquity proved very beneficial for the Aegean’s insular communities. The Cyclades formed small but dynamic sub-zones of economic, cultural and social interaction within the Eastern Mediterranean. The 10 11

A great challenge that the reconstruction of insular societies of the Aegean faces is to shed light on the settled landscape in the transition between Late Antiquity and the Early Byzantine period. At the beginning of the Early Byzantine centuries, most of the Aegean world appeared to enter a period of fundamental changes in settlement patterns, material culture and everyday life. The Byzantine–Arabic struggle for control over the Aegean, written sources and

Abadie-Reynal 1989. Carrié 2012.

123

Konstantinos Z. Roussos

Fig. 8. Multiple viewshed analysis of Late Antique rural sites on western Naxos.

the low visibility of archaeological material have resulted in a negative perception of these centuries. The new evidence from Paros and Naxos offers a fresh approach for

the period before the Arab conquest of Crete (824–828). What is more interesting is the different reactions of the two islands during the eighth and ninth centuries AD. 124

Tracing the Cycladic Settled Landscape in Late Antiquity and the Early Byzantine Period

Fig. 9. Distribution of Late Antique sites in the region of Naoussa.

Starting with Naxos, the emerging picture, through the synthetic analysis of the archaeological data, shows the island’s gradual and creative adaptation to the new circumstances pertaining to the Aegean after the late seventh century AD. The major characteristics of the Naxian settlement pattern during this period (fig. 10) can be summarised under six headings: (a) changes in urban context, with the capital’s relocation and the emergence of the Medieval urban centre of Apalirou, (b) the presence of the imperial administration on the island of Naxos, which probably emerged as an administrative and military node of great importance for the empire’s maritime defensive system, (c) in the Naxian countryside, the pattern of settlement distribution in relation to soil productivity, which did not change in the eighth and ninth centuries AD, (d) settlements not only being withdrawn to the mountainous hinterland but also located in lowland localities very close to the shoreline (fig. 11),12 (e) the mountainous interior, which presents a picture of more intensive habitation compared to the previous period (fig. 12–13), and (f) changes in the character of the industrial centres and the secondary export points.

eighth century AD. However, archaeological evidence suggests that life and small-scale economic activities continued in this coastal landscape during the Early Byzantine period.13 From the middle seventh century AD the fortified medieval urban centre of Apalirou made its appearance in the interior of west Naxos.14 Apart from its defensive character (fig. 14), the fortress’s western orientation shows an effort to control and manage internal production across the fertile lowland of Naxos and at the same time indicates a special care to ensure safe maritime circulation through the naval channel between Paros and Naxos (fig. 15–16). Kastro Apalirou was associated with two contemporary fortresses οn the islet of Viokastro15 and Palaiokastro on Ios.16 The intervisibility between them indicates the existence of a regional maritime communication network, probably for administrative and protective purposes. The study of archaeological evidence and the application of viewshed analysis has proved that Kastro Apalirou had a key role in this network (fig. 16). Thus, in terms of internal function, it emerged as the new local administrative centre of Naxos. Ιn a wider Cycladic context, it possibly served as the largest node within a regional maritime defensive network created by the central imperial administration. From this point of view,

The ancient city-port of Naxos lost its urban status and shrank in size into a small coastal settlement during the

13 Lambrinoudakis 1978, pp. 211–217; 1980; 1981, pp. 293–294; 1982, pp. 253–255. 14 Hill et al. 2017. 15 Viokastro is an offshore islet, located less than half of a mile off the north coast of Paros. For more, see Vionis 2012, pp. 125–131. 16 Eberhard 1974–78, pp. 552–554.

Despite considerable progress in the current archaeological investigations, this historiographical topos was until recently very powerful, and it has resulted in the negative perception of these centuries: Malamut 1988, pp. 67–68. For some new approaches, cf. Vionis 2013, pp. 30–31; Roussos 2017, pp. 189–234 and 292–296. 12

125

Konstantinos Z. Roussos

Fig. 10. Distribution of sites on Naxos during the Early Byzantine period.

this development reflects Constantinople’s organised effort to redefine the administration and management of the Cyclades and to secure the maritime routes, indicating a strong imperial presence on the island of Naxos at least until the Arab conquest of Crete.17 17

By contrast, material culture from Paros during the Early Byzantine period tells us a different story, particularly with regard to the countryside. For many years, an inability to recognise Early Byzantine pottery in conjunction with the information derived from the Life of Saint Theoktiste18

For more on this interpretation, see Roussos 2017, pp. 213–227.

18

126

Hero 1996.

Tracing the Cycladic Settled Landscape in Late Antiquity and the Early Byzantine Period

Fig. 11. Distribution of sites in western lowland Naxos during the Early Byzantine period.

Archaeological data, such as the dedicatory wall-painting in the chapel of Hagios Nikolaos (Ekatontapiliani complex),19 a group of ceramic sherds from the old

supported the view that the island was progressively abandoned as a result of Arab raids. It is difficult, however, to accept the pattern of total abandonment of so large Cycladic island as Paros.

19

127

Drossoyianni 1998.

Konstantinos Z. Roussos

Fig. 12. Distribution of sites in the areas of Tragea, Apiranthos and Danakos during Late Antiquity.

Fig. 13. Distribution of sites in the areas of Tragea, Apiranthos and Danakos during the Early Byzantine period.

128

Tracing the Cycladic Settled Landscape in Late Antiquity and the Early Byzantine Period

Fig. 14. The north slopes of the Byzantine Kastro Apalirou. The fortification’s north circular tower is discernible on the top of the mountain.

rescue excavation in the port of Parikia20 and the fact that the religious complex of Ekatontapiliani has remained standing from the sixth century AD to the present day, offer tangible evidence for the continuity of life during the Early Byzantine period. Analysis of the ceramic material from many sites, however, shows a gradual decline in imported pottery from the late seventh century AD. 21 This is an indication that the city gradually ceased functioning as a regional maritime commercial node. It lost its urban status and survived as a small coastal settlement, like the city of Naxos.

naval fortress on the islet of Viokastro from the seventh century AD.25 indicates that the maritime region was of military and commercial importance for the interest of the Byzantine Empire. This military installation reflects the integration of the island of Paros, and the Bay of Naoussa in particular, into a regional maritime defensive system for the protection of the Cyclades, created, probably, by the central administration. Thus, despite limitations imposed by the material evidence, this study suggests continuity of life along Paros’ coastal zone during the Early Byzantine centuries, but on a lesser scale compared to the previous period.

As for the Parian countryside, during our personal field observations a small amount of ceramic material evidence dated to the Early Byzantine period was identified in the Bay of Naoussa and the areas of Kantouna and Glifades (fig. 17).22 This extremely limited but valuable archaeological data, in conjunction with the evidence from Piso Livadi,23 provides a different picture, suggesting low levels of human activity along Paros’ north and east coastline after the early eighth century AD.24 The establishment of a

The involvement of landscape in the development of the settlement pattern of Paros and Naxos during Late Antiquity and the Early Byzantine period A new tantalising question needs to be answered: whether the local diversities of the Parian and Naxian landscapes determined the historical evolution of their settlement pattern, as well as the ways in which these two adjacent insular spaces integrated into and interacted with broader worlds under specific conditions and during specific periods of time.

For more details, see Roussos 2017, pp. 83–84, fig. 4.34. Roussos 2017, p. 84. 22 Roussos 2017, pp. 296–298. 23 Pennas 2001–4, pp. 209–210. 24 For more details, see Roussos 2017. 20 21

25

129

Vionis 2012, pp. 125–131.

Konstantinos Z. Roussos

Fig. 15. Viewshed analysis showing what is visible from the Kastro Apalirou in the island of Naxos. Visible areas marked in turquoise.

A comparative view of the settlement patterns in both case studies during Late Antiquity shows that Paros presents a clear trend towards a higher intensity of sites along the coastal zone, while Naxos presents a more balanced distribution of sites in all types of landscapes (fig. 3–4). This is a reflection of how the localised environmental advantages of different landscapes influence human behaviour and the evolution of settlement patterns over the centuries, even in such neighbouring insular communities.

in four important respects: (a) the landscape’s rich relief, (b) the distribution of fertile productive zones throughout almost the entire island (agricultural and grazing land), which diachronically supported the development of a mixed farming system (fig. 18), (c) the limited but strategically located small natural ports along its coastline, and (d) the abundance of water sources. Hence, habitation was distributed throughout the whole island, in regions that, on the one hand, could offer self-sufficiency and a more stable internal production system but, on the other hand, could provide potential and opportunities for exporting the surplus produced, through maritime trade.

In terms of natural environment, Naxos, which is the largest island among the Cyclades (430 km²), differs from Paros 130

Tracing the Cycladic Settled Landscape in Late Antiquity and the Early Byzantine Period

Fig. 16. Viewshed analysis showing the visibility from Kastro Apalirou. Visible areas marked in turquoise.

By contrast, the landscape of Paros, which is the third largest Cycladic island (196.308 km²), is characterised by: (a) the concentration of a greater percentage of arable land along the flat coastal zone, (b) the high distribution of natural ports along the coastline, which provided numerous safe shelters for ships, (c) the lowland and gentle coastal landscape that facilitates the internal mobility of people and goods, and (d) the barren and dry mountainous hinterland (fig. 19–20). Where land use and habitation patterns are concerned, the Parian landscape has created a diachronically strong contradiction between the low sustainability of the barren island’s interior and the

fertile coastal zone’s great economic potential. This sharp contrast developed an imbalanced ‘coastal’ settlement pattern consisting of sites oriented towards the sea and towards the interior of the flat fertile coastal zone (to a radius of 5 km from the shoreline). This local system was open to contacts and external influences, offering more potential and opportunities for prosperity through trade by receiving great support from the sea. A crucial question that remains open concerns the interpretation of the contrasting picture between Paros and Naxos during the Early Byzantine period. A convincing 131

Konstantinos Z. Roussos

Fig. 17. Sites with signs of human activity on Paros during the late seventh to the ninth century.

Aegean, the fading of Late Antique cities, the reduction of the Empire’s economic space and the subsequent decrease in the demand for commercial goods, the considerable transformation of the economic system from market to subsistence economy, and the essential changes in the state’s character, created a new reality during the Early Byzantine period, having a huge impact on islands like Paros. The high specialization and the market orientation of the production of Paros, as well as the lack of sufficient land in the island’s hinterland to support a more stable mixed farming system, proved key factors in the local population’s inability to adjust to the new conditions and achieve self-sufficiency. By contrast, the development of

explanation cannot be given based only on Arab hostility, since complex socio-economic phenomena are related to multiple parameters. It seems that the specific natural features of local landscapes once more played a role in this process. In the case of Paros, which formed a more fragile microcosm, demographically, culturally and environmentally, any change in the trade patterns and sea routes of Late Antiquity might have had a strong impact on the island’s role in this network, as well as on its socioeconomic situation. A series of external parameters, such as the Byzantine–Arabic struggle for control over the 132

Tracing the Cycladic Settled Landscape in Late Antiquity and the Early Byzantine Period

Fig. 18. Aerial photo of Naxos.

Fig. 19. Aerial photo of Paros.

133

Konstantinos Z. Roussos

Fig. 20. a) Panoramic view of the eastern coastal zone of Paros. b) Panoramic view of the south-western coastal plain of Paros.

a more balanced and stable settlement pattern on Naxos since antiquity, within a completely different landscape compared to Paros, ensured that the island could retain high levels of human activity under all circumstances pertaining to the Aegean. The geographically and militarily advantageous location of the island in the Cyclades, in conjunction with the potential and opportunities offered by its distinctive landscape character, led the central government to select Naxos as one of the administrative centres in the southern Aegean.

Byzantine ‘Dark Ages’ may not have been so ‘dark’ as we think, especially for the insular Aegean world. Cycladic islands may be approached as places of interaction with broader worlds, rather than as places of isolation, since Cycladic landscapes, coastal and mountainous, continued to be settled, cultivated and in other ways interacted with humans between the fourth and mid-ninth century AD. Bibliography Abbreviation:

Conclusions

ΠΑΕ = Πρακτικά της εν Aθήναις Aρχαιολογικής Eταιρείας.

These two case studies are enough to demonstrate that each island can tell a unique story during different periods of its history. The essential distinction between the two islands lies in the diametrically opposed, localised environmental advantages of their landscape character. During the period between the fourth and late seventh century AD, Paros and Naxos, like many other regions in the Aegean world, show evidence of commercial success in functioning as producing, processing and exporting centres. During the following centuries, the island of Naxos shows tangible signs of a stable and vibrant insular microcosm, which held an important position in the defensive network of the Early Byzantine period. Even in the case of Paros, with archaeologically low-visibility sites, some parts of the coastal landscape were occupied in almost every period, and the island continued to interact with the external word in a different way and on a lesser scale. Thus, the so-called

Abadie-Reynal C., 1989. ‘Céramique et commerce dans le bassin Égéen du IVe au VIIe siècle’, in C. Morrisson, J. Lefort (eds.), Hommes et richesses dans l’Empire byzantin, I, Paris (Réalités byzantines), pp. 143–159. Bevan A., Conolly J., 2013. Mediterranean Islands, Fragile Communities and Persistent Landscapes. Antikythera in Long-Term Perspective, Cambridge. Carrié J.-M., 2012. ‘Were Late Roman and Byzantine economies market economies? A comparative look at historiography’, in C. Morrisson (ed.), Trade and Markets in Byzantium, Washington (Dumbarton Oaks Byzantine Symposia and colloquia), pp. 13–26. Cherry J. F., Davis J. L., Mantzourani E. (eds.), 1991. Landscape Archaeology as Long-Term History. 134

Tracing the Cycladic Settled Landscape in Late Antiquity and the Early Byzantine Period Northern Keos in the Cycladic Islands from Earliest Settlement until Modern Times, Los Angeles (Monumenta archaeologica, 16).

the Late Antique and Early Byzantine Centuries, Leiden (Archaeological Studies – Leiden University, 40). Saradi E., 2006. The Byzantine City in the Sixth Century. Literary Images and Historical Reality, Athens.

Deligiannakis G., 2008. ‘The economy of the Dodecanese in Late Antiquity’, in Ch. Papageorgiadou-Banis, A. Giannikouri (eds.), Sailing in the Aegean. Readings on the Economy and Trade Routes, Athens (Μελετήματα, 53), pp. 219–223.

Vionis A. K., 2012. A Crusader, Ottoman and Early Modern Aegean Archaeology. Built Environment and Domestic Material Culture in the Medieval and PostMedieval Cyclades, Greece (13th–20th Century AD), Leiden (Archaeological Studies – Leiden University, 22).

Deligiannakis G., 2016. The Dodecanese and the Eastern Aegean Islands in Late Antiquity, AD 300–700, Oxford (Oxford Monographs on Classical Archaeology).

Vionis A. K., 2013. ‘Considering a rural and household archaeology of the Byzantine Aegean: The ceramic spectrum’, in J. Bintliff, M. Caroscio (eds.), Pottery and Social Dynamics in the Mediterranean and Beyond in Medieval and Post-Medieval Times, Oxford (BAR International Series, 2557), pp. 25–40.

Drossoyianni F., 1998. ‘Παλαιοχριστιανικές τοιχογραφίες στην Εκατονταπυλιανή Πάρου’, in Η Εκατονταπυλιανή και η χριστιανική Πάρος. Πρακτικά Επιστημονικού Συμποσίου (Πάρος 15–19 Σεπτεμβρίου 1996), Paros, pp. 55–81. Eberhard H., 1974–78. ‘Mittelalterliche Burgen auf den Kykladen’, in Επετηρίς Εταιρείας Κυκλαδικών Μελετών, 10, pp. 501–585.

Vionis A. K., 2016. ‘Naxos between Late Antiquity and the Early Middle Ages: Decline or stability?’, in B. Krsmanović, L. Milanović (eds.), Proceedings of the 23rd International Congress of Byzantine Studies, Belgrade, 22–27 August 2016. Round Tables, Belgrade, pp. 329–332 (http://www.byzinst-sasa.rs/srp/uploaded/ PDF%20izdanja/round%20tables.pdf).

Gerousi E., 2001. ‘Thera in Early Christian Times’, in I. Danezis (ed.), Σαντορίνη. Θήρα, Θηρασιά, Ασπρονήσι, Ηφαίστεια, Athens, pp. 254–268. Hero A. C., 1996. ‘Life of St. Theoktiste of Lesbos’, in A.M. Talbot (ed.), Holy Women of Byzantium. Ten Saints’ Lives in English Translation, Washington (Byzantine Saints’ Lives in Translation, 1), pp. 95–116. Hill D., Roland H., Ødegård K., 2017. ‘Kastro Apalirou, Naxos, a seventh-century urban foundation’, in E. Rizos (ed.), New Cities in Late Antiquity. Documents and Archaeology, Paris (Bibliothèque de l’Antiquité tardive, 35), pp. 281–292. Lambrinoudakis V., 1978. ‘Ανασκαφή Νάξου’, in ΠΑΕ, 133, pp. 211–218. Lambrinoudakis V., 1980. ‘Ανασκαφή Νάξου’, in ΠΑΕ, 135, pp. 259–262. Lambrinoudakis V., 1981. ‘Ανασκαφή Νάξου’, in ΠΑΕ, 136, pp. 293–297. Lambrinoudakis V., 1982. ‘Ανασκαφή Νάξου’, in ΠΑΕ, 137, pp. 253–259. Malamut E., 1988. Les îles de l’Empire byzantin, VIIIe– XIIe siècles, Paris (Byzantina Sorbonensia, 8). Mitsani A., 1996–97. ‘Το παλαιοχριστιανικό κιβώριο της Καταπολιανής Πάρου’, in Δελτίον της Χριστιανικής Αρχαιολογικής Εταιρείας, 19, pp. 319–333. Pennas, Ch. 2001–4. ‘Πάρος. Πίσω Λιβαδι. Ναός Αγίου Γεωργίου Θαλασσίτη’, in Αρχαιολογικόν Δελτίον, 56– 59 – Χρονικά Β΄6, pp. 209–210. Renfrew C., Wagstaff M., 1982. An Island Polity. The Archaeology of Exploitation in Melos, Cambridge. Roussos K. Z., 2017. Reconstructing the Settled Landscape of the Cyclades. The Islands of Paros and Naxos During

135

10 Life Among the Ruins: Medieval Settlements and Late Antique Cities, Seen Through Recent Excavations in the Region of the Upper Vardar and Bregalnica Rivers Dejan Gjorgjievski Национална установа Музеј Куманово / National Institution Museum Kumanovo (North Macedonia) The continuity of habitation in Late Roman cities in the medieval period in the south-western Balkan Peninsula has long been neglected, for several reasons. Surprisingly, although there were several indicators of medieval settlements built on Late Antique remains, their modest appearance, compared with the Antique luxury below them, often resulted in ‘archaeological destruction during the process of excavation during the first half of the twentieth century. In this chapter, we examine the most important settlements in the area bordered by two important rivers, the Vardar and the Bregalnica. The latest research clearly indicates that if the towns were not completely destroyed in the Roman period, most of them were re-used in the medieval period. Although only scattered traces survive, they are witnesses to the continuity of cities whose fate mostly depended on internal factors rather than on the migrations. La continuité durant la période médiévale de l’occupation des villes romaines tardives dans le sud-ouest de la péninsule balkanique a longtemps été négligée pour plusieurs raisons. Étonnamment, bien qu’il y ait plusieurs indicateurs d’agglomérations médiévales construites sur des restes de l’Antiquité tardive, leur apparence modeste, comparé au luxe antique sous-jacent, a souvent entraîné leur « destruction archéologique » lors des fouilles de la première moitié du XXe siècle. Dans cet article, nous examinerons les principales agglomérations d’un secteur délimité par deux cours d’eau importants, le Vardar et la Bregalnica. Les dernières recherches indiquent clairement que si les villes n’étaient pas complètement détruites à l’époque romaine, la plupart étaient en usage au Moyen Âge. Bien que seules des traces éparses en aient survécu, elles témoignent de la continuité de villes dont le destin dépendait plus de facteurs internes que des migrations. Die Besiedlungskontinuität spätrömischer Städte auf der südwestlichen Balkanhalbinsel im Mittelalter wurde aus verschiedenen Gründen lange wenig beachtet. Überraschenderweise gab es jedoch mehrere Hinweise darauf, dass mittelalterliche Siedlungen auf spätantiken Überresten enstanden. Ihr verglichen mit dem antiken Reichtum wenig spektaluäres Aussehen führte oft zu ihrer „archäologischen Zerstörung“ während der Ausgrabungen der ersten Hälfte des 20. Jhs. In diesem Artikel werden die wichtigsten Siedlungen in dem durch zwei wichtige Flüsse, dem Vardar und der Bregalnica, begrenzten Gebiet untersucht. Die neuesten Forschungen verdeutlichen, dass die römischen Städte nicht vollständig zerstört und im Mittelalter weiter genutzt worden sind. Obwohl nur geringe Spuren erhalten geblieben sind, sind diese als Zeugen der Kontinuität von Städten zu werten, deren Schicksal hauptsächlich durch interne Veränderungen und nicht durch Migration beeinflusst war. La continuità abitativa delle città tardo-romane nel periodo medievale nella penisola balcanica sud-occidentale è stata a lungo trascurata per diverse ragioni. Sorprendentemente, anche se ci sono stati diversi indicatori di insediamenti medievali costruiti su resti ardoantichi, il loro aspetto modesto, rispetto al lusso delle antiche strutture sottostanti ha spesso provocato una «distruzione archeologica» durante il processo di scavo nella prima metà del XX secolo. In questo articolo, esamineremo gli insediamenti più significativi nell’area delimitata da due importanti fiumi: il Vardar e il Bregalnica. Le ultime ricerche indicano chiaramente che se le città non fossero state completamente distrutte nel periodo romano, la maggior parte di queste sarebbe stata in uso in

137

Dejan Gjorgjievski epoca medievale. Sebbene siano sopravvissute solo tracce sparse, queste sono testimoni della continuità delle città, il cui destino dipendeva principalmente da fattori interni piuttosto che dalle migrazioni. In contrast to the splendour and luxury of Late Antique cities in the south-west Balkans, stand the modesty and mystique of medieval settlements. The ‘official theory’ concerning the fate of Balkan cities after antiquity, drawn from both historical and archaeological data, explains that most of the urban centres were deeply stricken by the barbarian incursions.1 As a result of devastations caused by the attackers, the cities did not survive those turbulent periods, and most of them were not occupied during the medieval period.2 Although there are two different opinions about how that happened – and they differ mostly on the question of whether it happened as a Blitzkrieg or as the result of continuing and increasing attacks over two centuries,3 neither group of researchers treats the results of those events. Did the centres simply ‘die’, to be replaced by new ones, or was their fate slightly different than what has been thought to be the most logical explanation?

after the earthquake, and the city slowly faded away from the historical scene of the sixth century.7 On the other hand, there are reasons to argue for important constructions at Scupi in the sixth century, after the 518 earthquake. The basilica with three eastern apses, discovered in 2008, is simply too developed to be placed in the early sixth century, as the excavator tends to believe.8 The basilica below the theatre was also dated on the assumption that it could not have been built later than 518. After Late Antiquity, there are documented traces of a population that dwelt here in the medieval period. Small fireplaces or hearths were found in the north aisle of the church below the theatre.9 Without any precise material for reliable dating, they can be correlated with the traces of settlement found above the public baths, where at least four horizons of habitation were found, roughly dated to the beginning of the seventh to the eighth century.10 It is interesting to point out that the excavators found no traces of violent events that destroyed any of the houses from that period!11

The fate of the provincial urban centres that were affected by these changes can be observed through the results of archaeological excavations conducted in recent decades in the region that appears as a contact zone of the ‘frontiers’ of the Roman provinces of Dardania, Macedonia Secunda and Dacia Mediterranea.4 The region, which is framed by two large rivers – Vardar (the ancient Axios) and Bregalnica (the ancient Astibo) – had at least three major cities in the Late Antique period. In the north was the Roman (Flavian) colony of Scupi, in the south was the Municipium of Stobi, in the east was the episcopal seat at the city of Bargala. There are two more possible cities – candidates for episcopal seats in the period: the anonymous city at Konjuh and the fortress at Vinica.

Furthermore, among the published finds that testify to crafts and the existence of the applied arts in Scupi in the Late Roman period, two published rings can be attributed without any doubt to the medieval period.12 One is a ring decorated with a heraldic eagle that can be dated to the tenth to twelfth century; another type of ring, decorated with a dove holding a branch, is dated to the same period.13 During the last few years, the antique theatre has been almost completely excavated (fig. 1). The partially published material indicates that above the theatre there stood an entire medieval settlement (and necropolis) (fig. 2, 3). Most of the material can be closely connected with the Middle Byzantine period, but there are artifacts that can be dated earlier, to the eighth and ninth centuries, as confirmed by both ceramic and numismatic finds.14

Scupi was erected as a Roman castra but, due to its position, it soon became one of the most important cities in the south-central Balkans.5 From the time the province of Dardania was created in the late third century, Scupi served as its capital. With all the elements of Roman urban planning, its theatre, public baths, Early Christian churches and wide-open streets have attracted the attention of scholars for more than a century.6 According to the excavators of the site, past and present, the city was devastated in the earthquake of 518 (Marcellinus Comes). Although rebuilt, in the succeeding decades it became a target of barbarian incursions from the north. The city recovered to some extent, but no major building took place

Therefore, if we have documented traces of habitation from the seventh to the twelfth century, it seems that the turbulent Ončevska Todorovska 2009, pp.108–109. Ončevska Todorovska 2009, pp. 111–112; 2011, p. 373; 2015. Compare with Varalis 2006; Snively 2011, pp. 199–200. 9 Garašanin, Koračević 1979, p. 36. 10 Koračević 2002, pp. 165–170; Risteski 2002, pp. 79–81. 11 In this context, see Curta 2012. 12 Ončevska Todorovska 2006, pp. 284–286. 13 For the numerous analogies, cf. Grigorov 2007, pp. 55–56; Bikić 2010, pp. 93–95; Bosselmann-Ruickbie 2011, pp. 295–296. 14 Jakimovski, Tolevski 2016, pp. 179–197; Jakimovski 2017, pp. 321– 366 . I wish to express my gratitude to Prof. A. Jakimovski, director of the excavation of the antique theatre of Scupi, and L. Jovanova, director of the Scupi excavation, for their permission to study the material. I am indebted also to I. Tolevski for his help in the interpretation of the finds and his kindness allowing me to use his papers before they were published. 7 8

On the genesis of the theory and its critics, see Curta 2001. Some general considerations can be found in Haldon 2001, pp. 455–462. 3 Popović 1980; 1982; Dunn 1994; 1999; Morrisson 2007; Rashev 2007, pp. 97–98. 4 On the complex history of the region in this period, see Snively 2010. 5 From the vast literature on the city, the most recent publications are: Jovanova 2015a; Jakimovski 2017; Jovanova, Ončevska Todorovska 2018. 6 Jovanova 2015b, pp. 219–220. 1 2

138

Life Among the Ruins

Fig. 1. Theatre at Scupi, at the end of the excavations (©Museum of the City of Skopje).

Fig. 2. Medieval remains above the theatre (after T. Jakimovski, I. Tolevski).

139

Dejan Gjorgjievski

Fig. 3. General plan of the theatre (after T. Jakimovski, I. Tolevski).

Stobi continued with its church activities in the seventh and eighth centuries.

‘Dark Ages’ did not cause the population to disappear from the city. The well-protected medieval fortress at Skopsko Kale was a far better place to be in time of war, but it seems that the life of this Late Roman town was not just interrupted by the barbarians. It seems obvious that internal socio-economic factors, in combination with external events, caused the decline of the popularity of the city. It did not all happen in the sixth century, however, but was a slow process over the course of the following centuries.

At the Palikura Basilica, approximately 2 km west of Stobi, the architectural sculpture included some unusual pieces, atypical for the fifth or sixth century.18 For example, the top of the columns (and not the capitals!) display carved leaves, crosses, or flutes. As pointed out by Nikolajević, the masons had forgotten the principles of decoration for separate architectural elements and transferred the decoration of the capital to the upper part of the column.19 According to Snively, the date of the material from the Palikura Basilica must be pushed into the seventh century, but how far into that century is difficult to say.

Was this the case with another city centre – the Municipium of Stobi? Slightly less important in the Roman period, but far better known through excavation, the city of Stobi has been assumed to be a typical representative of the events that occurred in the sixth century. Again, because of earthquakes and sudden attacks, urban life ended in the 580s and the city was completely abandoned within the next few decades.15 But, although the standard of living fell below that of Roman urban life, the population of the city did not completely disappear. Historical sources mention two bishops of Stobi, John in 680/81 and Margarites in 692, in the list of those attending church councils. Popović suggests that the briefly restored bishopric represented the Sermesianoi, who had just arrived and settled in the Keramesian plain.16 It is more likely, however, that the Keramesian plain was located near Thessaloniki, as has recently been argued,17 and that

The salvage excavations in 1992 revealed a new extra muros basilica outside the south city wall at Stobi. The unpublished excavations of 2003–2004 investigated a small chapel that had been erected above the presbyterium of that basilica; the excavators dated it tentatively to the seventh century and raised the possibility that its use had been connected with the Sermesianoi and the briefly revived bishopric at Stobi. The alternative hypothesis was an eleventh-century date.20 As shown in Carolyn S. Snively’s paper ‘The cities of Northern Macedonia and Dardania after the sixth century’, presented at the NGO ‘Aimos’ seminar Transfigurations of Byzantine Urban Landscapes (Sixth–Ninth Centuries), Thessaloniki, 19 November 2011 (not yet published). 19 Nikolajević-Stojković 1957, pp. 45–47. 20 Cf. supra, n. 18. 18

Georgiev 2012. Popović 1980, pp. 250–252. 17 Bakirtzis 2007, p. 101, n. 66. 15 16

140

Life Among the Ruins

Fig. 4. The theatre at Stobi.

Here too, as at Scupi, some small houses were built above and inside the theatre (fig. 4).21 Most of the theatre was excavated in the 1960s and, not surprisingly, there are no records of the later layers. In recent years, however, excavations revealed remains of small houses; the latest ones now visible above the theatre can be dated to the second half of the sixth century. Above them are floor levels and fireplaces that cannot be dated with accuracy. In the circular corridor there are floor levels that are roughly dated between the eleventh and fourteenth century.22 Further to the south-east, archaeological material of a similar date was found near the newly discovered Temple of Isis, in 2008.

the church in the ninth century, and a small necropolis dated between the ninth and twelfth century is closely connected with its function.26 Apparently, the settlement is mentioned in the First Chrysobull of Basil the Great on the rights of the Ohrid archbishopric; in reference to the bishopric at Morodvis, the church (actually the presbyters) at Kozjak is mentioned.27 Admittedly, Kozjak is connected with the mountain of that name in the region of Kumanovo or, more precisely, with the small fortress near Kanarevo.28 But it is more probable that the chrysobull refers to the existing village of Goren Kozjak, which is close to Morodvis, and where Bargala is located.29 So far, the only link between the sixth- and seventh/eighth-century settlements consists of a few finds from pits near the church, but their exact dating remains a matter of dispute.30

In 1955, 23 burials were excavated in and around the North Basilica, apparently the surviving remnant of a larger cemetery spread over the area between the North and Central Basilicas. They were dated from the ninth to the twelfth century. Some additional burials were found in 2005 during conservation of the ‘Civil’ Basilica, now identified as a church; it lies between the Central and North Basilicas. Near the Palikura Basilica, above the remains of a Late Antique xenodocheion, 16 medieval burials were excavated and dated to the eleventh to thirteenth century.23 The city of Bargala was briefly excavated by a joint Yugoslav–American project led by Blaga Aleksova and Cyril Mango. They investigated the most outstanding object known so far – a large Episcopal Basilica from the fifth to sixth century.24 In the following decades, excavations were concentrated in the eastern part of the city and around the medieval Church of St George (fig. 5).25 Art historians are now placing the erection of

Something similar may be observed at another Late Antique city situated on the Bregalnica. The fortress at Vinica, known mostly for its unique terracotta reliefs with Christian themes,31 has long been considered another town that was erased by the turbulent events at the end of Late Antiquity. Medieval finds were connected mostly with the small necropolis situated on the hilltop, but the medieval settlement was not located.32 Recently excavated material, especially from the north-east end of the fortress, however, tells a different story (fig. 6, 7). Most of the finds can be dated to the period of the tenth to the twelfth century, with analogies that may go back as early as the eighth century.33 But here, as at Bargala, material from the seventh century has not been recorded. We do not know for sure whether or not the hilltop was in use in the seventh century. But before such a possibility is excluded, it is necessary to remember that the excavations did not provide sufficient

Pavlovski 2018, pp. 200–205. Once again, I express my gratitude to the director of the NI Stobi, S. Blazevska, and to G. Pavlovski, director of the excavations of the theatre, for their willingness to share with me the unpublished results of the latest excavations. 22 Archaeological material which can roughly be dated between the ninth and twelfth centuries, was found near the newly discovered Temple of Isis in 2008. 23 Petrović 1942, p. 492; Aleksova 1981; Ivanovski 1986. 24 Aleksova, Mango 1971; Nacev 2008. 25 Aleksova 1978; Rasolkoska-Nikolovska 1975; Nacev, Stojanovski 2017, p. 277.

Rasolkoska-Nikolovska 2011, pp. 19–20. Snagarov 1924, p. 175. 28 Mikulčić 1996, pp. 221–223. 29 Gjorgjievski 2012, p. 61. 30 Beldedovski 1990, p. 50. 31 The latest examination of the subject is that of Prof. Dimitrova (2012). On the problem of the dating of the reliefs, cf. Gjorgjievski 2015. 32 The necropolis is not completely published; see Guštin, Krstevski 2012, p. 383. 33 Guštin, Krstevski 2012, pp. 384–389.

21

26 27

141

Dejan Gjorgjievski

Fig. 5. The Church of St George at Bargala, just next to the Late Antique fortification wall.

this site. So far, only an eighth-century Byzantine coin has been published,36 although traces of medieval habitation were noticed.37

material for reconstruction of the houses from the early medieval settlement. As Gustin and Krstevski observe, it was impossible to trace the medieval structures layered above the Late Roman ones, partially because of their fragility, but also because they were re-using a lot of destroyed Late Roman structures.

On the northern terrace, some later material has been found, but mostly scattered through the various layers. In 2013, a well-preserved sixth-century baptistery was discovered (fig. 9). A wall divides the apsidal room with piscina from the large room to the west. It is interesting that the wall had been rebuilt from a level a metre or so above the level of the baptistery (fig. 10, 11). Obviously the doorway was closed, beginning from that level. Clearly the building was in use by someone, at a time when more than a metre of destruction debris covered the original floor. On the acropolis also, in a dark ashy layer near the south gate and beside the collapsed fortification wall, artefacts were found that probably date to the Late Byzantine or even Early Ottoman period (fig. 12).

At the village of Konjuh, an anonymous city once stood on one of the roads that connected Thessaloniki with Naissus (fig. 8). The site has been intensively investigated for more than a decade, and the excavation revealed a probable Episcopal Basilica, an Episcopal residence, a rotunda church and two small extramural churches north and south of the city.34 The lower town developed on the northern terrace, but its acropolis, where many rock-cut rooms are found, was in use as well. The city was mentioned as a castel at the end of the fourteenth century; in 1373, a merchant from Ragusa was travelling from Thessaloniki to Novo Brdo, and was robbed by two noblemen from the court of Dragas at the castel of Konjuh.35 But archaeology provided little information about the medieval layers on

After this review of the evidence, it is clear that if Late Roman towns were not completely destroyed in the Roman period, most of them were in use (at least as

A complete bibliography on the site can be found at: http://www. konjuh.mk/en/ (last accessed 23 March 2017). 35 See the most recent work on medieval Konjuh by Gordana Tomović: Tomović 2016. Still, the image of the city as it is presented in the article can hardly be confirmed on the site itself. 34

Mikulčić 1996, p. 226. I would like to thank to Prof. C. S. Snively, co-director of the Konjuh excavations, for information and advice on this subject.

36 37

142

Life Among the Ruins

Fig. 6. Medieval findings from Vinica (after M. Gustin, C. Krstevski).

143

Dejan Gjorgjievski

Fig. 7. Medieval hoard from Vinica (after M. Gustin, C. Krstevski).

Fig. 8. Konjuh, the Episcopal Church (C. S. Snively).

144

Life Among the Ruins

Fig. 9. Konjuh, plan of the Episcopal Church and apsidal baptistery in the south (C. S. Snively).

Fig. 10. Konjuh, a wall dividing the baptistery (C. S. Snively).

145

Dejan Gjorgjievski

Fig. 11. Konjuh, closed doorway in the baptistery (C. S. Snively).

Fig. 12. Medieval layer on the acropolis (C. S. Snively).

146

Life Among the Ruins Aleksova B., 1981. ‘Medieval graves in the North Basilica’, in J. Wiseman, B. Aleksova (eds.), Studies in the Antiquities of Stobi, III, Princeton, pp. 249–261. Aleksova B., Mango C., 1971. ‘Bargala: A preliminary report’, in Dumbarton Oaks Papers, 25, pp. 265–281. Avraméa A, 2005. ‘Les villages de Thessalie, de Grèce centrale et du Péloponnèse (Ve–XIVe siècle)’, in Lefort et al. 2005, pp. 213–224. Bakirtzis Ch., 2007. ‘Imports, exports and autarky in Byzantine Thessalonike from the seventh to the tenth century’, in J. Henning (ed.), Post-Roman Towns, Trade and Settlement in Europe and Byzantium, II, Berlin/New York (Millennium-Studien zu Kultur und Geschichte des ersten Jahrtausends n. Chr., 5/2), pp. 89–118. Beldedovski Z., 1990. Брегалничкиот басен во римскиот и раниот средновековен период, Štip (Зборник Щип, 6). Bikić V., 2010. Vizantiski nakit u Srbiji. Modeli i nasleđe – Byzantine Jewellery in Serbia. Models and Heritage, Belgrade.

Fig. 13. Stobi, photo from the 1960s excavations (courtesy of NI Stobi).

Bosselmann-Ruickbie A., 2011. Byzantinischer Schmuck des 9. bis frühen 13. Jahrhunderts. Untersuchungen zum metallenen dekorativen Körperschmuck der mittelbyzantinischen Zeit anhand datierter Funde, Weisbaden (Spätantike, frühes Christentum, Byzanz. Reihe B, Studien und Perspektiven, 28).

smaller settlements) during the medieval period.38 Although only scattered traces survive, they are witnesses to the continuity of the cities,39 whose fate mostly depended on internal factors rather than on the migrations.

Curta F., 2001. The Making of the Slavs. History and Archaeology of the Lower Danube Region c. 500–700, Cambridge (Cambridge Studies in Medieval Life and Thought, 52).

The reader may find this claim to be too tenuous. It is necessary to remember that the goals of excavation at the time when most of the buildings from the Roman period were excavated were not very favourable toward the latest deposits on the sites. For instance, the excavators of the theatres at both Stobi and Scupi paid little attention to later deposits, blinded by the beauty of the luxurious Roman buildings below (fig. 13). Only recent excavation of what was left intact provides us with information about those later archaeological contexts. But it is obvious that the fate of Late Antique city centres in the Balkan Peninsula was far more complicated than we imagined.

Curta F., 2012. ‘Were there any Slavs in seventh-century Macedonia?’, in Istorija, 47, pp. 61–76. Dimitrova E., 2012. Виничката мистериjа. Керамичката ризница од доцноантичкото кале – The Vinica Mystery. The Ceramic Treasuries of a Late Antique Fortress, Vinica. Dimitrova E. (ed.), 2015. Folia Archaeologica Balkanica, III, Skopje.

Bibliography

Dunn A. W., 1994. ‘The Transformation from polis to kastron in the Balkans (III–VII cc.): General and regional perspectives’, in Byzantine and Modern Greek Studies, 18, pp. 60–80.

Abbreviations: CRAI = Comptes rendus des séances de l’Académie des Inscriptions et Belles-Lettres.

Dunn A. W., 1999. ‘From polis to kastron in Southern Macedonia: Amphipolis, Khrysoupolis, and the Strymon Delta’, in A. Bazzana (ed.), Castrum, V, Madrid/Rome (Collection de la Casa de Velázquez, 55; Collection de l’École française de Rome, 105), pp. 399–413.

MAA = Macedoniae Acta Archaeologica. MH = Macedonian Heritage. Aleksova B., 1978. ‘Баргала: Археолошки истражувања 1966–1976’, in Зборник, Археолошки музеј на Македонија, VIII–IX, pp. 75–84.

Garašanin M., Koračević D., 1979. ‘Базиликата II во Скупи’, in MAA, 5, pp. 31–46.

38 A similar process can be observed in smaller settlements in other parts of the Balkans: Avraméa 2005, pp. 213–217; Rosser 2005; Veikou 2013, pp. 130–133. 39 Völling 2001, pp. 309–310.

Georgiev, Z., 2012. ‘Тези и аргументи за некои моменти од историјата на Стоби’, in Cultural Heritage, 38, pp. 35–51. 147

Dejan Gjorgjievski Nacev T., 2008. Доцноантичкиот град Баргала, Štip.

Gjorgjievski D., 2015. ‘Dating of the Vinica Teracottas’, in Z. Dimoski, B. Paweł, E. Lučeska (eds.), Słowiańskie teksty kultury samokowskie kontynuacje, II, Poznan, pp. 109–117.

Nacev T., Stojanovski D., 2017. ‘The city of Bargala since VII to XII Century’, in Годишник на асоциация за антропология етнология и фолклористика ‘Онгъл’, 15, pp. 272–279.

Gjorgjievski D., 2012. ‘Кумановскиот регион во средновековниот период. Согледувања според археолошките наоди’, in Музејски Гласник, 11, pp. 60–74.

Nikolajević-Stojković I., 1957. Рановизантиска архитектонска декоративна пластика у Македонији, Србији и Црној Гори – La décoration architecturale sculptée de l’époque bas-romaine en Macédoine, en Serbie et au Monténegro, Belgrade.

Grigorov V., 2007. Метални накити от средновековна България (VII–XI vek), Sofia (Национален археологически институт и музей. Дисертации, 1).

Ončevska Todorovska M., 2006. ‘Занаетчиството и применетата уметност во доцната антика преку металните предмети од Скупи’, in MAA, 19, pp. 281– 292.

Guštin M., Krstevski C., 2012. ‘Ranosrednjovjekovni nalazi s nalazišta Viničko kale u Makedoniji’, in T. Šeparović (ed.), Dani Stjepana Gunjače, II, Split, pp. 379–393.

Ončevska Todorovska M., 2009. ‘The great Early Christian basilica in Scupi and the Early Christianity’, in MH, 34–35, pp. 97–112.

Haldon J., 2001. ‘Byzantium in the Dark Centuries. Some concluding remarks’, in Kountoura-Galaki 2001, pp. 455–462.

Ončevska Todorovska M., 2011. ‘Новата голема ранохристијанска базилика во Скупи’, in MAA, 20, pp. 365–379.

Ivanovski M., 1986. ‘Средновековна некропола во реонот на Паликурската базилика во Стоби’, in M. Apostolski (ed.), Зборник посветен на Бошко Бабиќ/Boško Babić – Mélanges Boško Babić, Prilep, pp. 97–104. Jakimovski A. (ed.), 2017. Римски театар Скупи – Roman Theatre Scupi, Skopje.

Ončevska Todorovska M., 2015. ‘The atrium in Early Christian church architecture in East Illyricum and the large Early Christian basilica with an atrium in Scupi’, in E. Dimitrova 2015, pp. 267–288.

Jakimovski A., Tolevski I., 2016. ‘Glagolitic inscriptions on ceramic vessels uncovered during archaeological excavations in Scupi (Skopje)’, in Slověne, 5/2, pp. 179–197.

Pavlovski G., 2018. ‘The cavea of the theater at Stobi, results from the excavations in 2009, 2010, 2012’, in S. Blaz̆ evska (ed.), Studies in the Antiquities of Stobi, IV, Stobi, pp. 156–205.

Jovanova L., 2015a. Scupi. Colonia Flavia Scupinorum, Skopje.

Petrović J., 1942. ‘U Stobima danas’, in Glasnik Hrvatskih zemaljskih muzeja u Sarajevu, 54, pp. 463–525.

Jovanova L., 2015b. ‘Scupi. Transition from Imperial colony to Late Antique metropolis’, in Dimitrova 2015, pp. 219–252.

Popović V., 1980. ‘Aux origines de la slavisation des Balkans: la constitution des premières Sklavinies macédoniennes vers la fin du VIe siècle’, in CRAI, 124/1, Paris, pp. 230–257.

Jovanova L., Ončevska Todorovska M., 2018. Скупи. сектор Југоисточен бедем и комплекс Терма – базилика со атриум – Scupi. Sector Southeastern Defensive Wall and the Thermae – Atrium Basilica Complex, Skopje.

Popović V., 1982. ‘Desintegration und Ruralisation der Stadt im Ost-Illyricum vom 5. bis 7. Jahrhundert n. Chr.’, in D. Pappenfuss, V. M. Strockaeds (eds.), Palast und Hütte. Beiträge zum Bauen und Wohnen im Altertum von Archäologen, Vor- und Frühgeschichtlern. Tagungsbeiträge eines Symposiums der Alexander von Humboldt-Stiftung, Bonn-Bad Godesberg, veranstaltet vom 25.–30. November 1979 in Berlin, Main (Internationale Fachgespräche), pp. 545–566.

Koračević D., 2002. Скупи. Градска територија, Skopje. Kountoura-Galaki E. (ed.), Οι σκοτεινοί αιώνες του Βυζαντίου (7οσ–9ος αι.) – The Dark Centuries of Byzantium (7th–9th c.), Athens (Διεθνή συμπόσια, 9). Lefort J., Morrisson C., Sodini J.-P. (eds.), Les villages dans l’Empire byzantin (IVe–XVe siècle), Paris (Réalités byzantine, 11).

Rashev R., 2007. ‘Към проблема за материалната култура на територията на Република Македония от края на VII до средата на IX в.’, in Археология, 48, pp. 90–101.

Mikulčić I., 1996. Средновековни градови и тврдини во Македонија – Mittelalterliche Städte and Burgen in der Republik Makedonien – Medieval Towns and Castles in the Republic of Macedonia, Skopje.

Rasolkoska-Nikolovska Z., 1975. ‘Археолошки истражувања на црквата Св. Ѓорѓи во Горни Козјак’, in Зборник на Штипскиот народен музеј, 4–5, pp. 125–139.

Morrisson C., 2007. ‘La fin de l’Antiquité dans les Balkans à la lumière des trésors monétaires des VIe et VIIe siècles’, in CRAI, 151/2, pp. 661–684.

Rasolkoska-Nikolovska Z., 2011. ‘Најстарата фреска во црквата Св. Георги во Горни Козјак, во близина на Штип’, in D. Medaković, C. Grozdanov. (eds.), На 148

Life Among the Ruins траговима Војислава Ј. Ђурића, Belgrade/Skopje (Српска академија наука и уметности Одељење историјских наука, 135; Научни скупови, 33), pp. 18–21. Risteski B., 2002. ‘Early medieval findings from Scupi’, in MH, 19, pp. 71–81. Rosser J., 2005. ‘Dark Age settlements in Grevena, Greece (southwestern Macedonia)’, in Lefort et al. 2005, pp. 279–288. Snagarov I., 1924. История на Охридската архиепископия. Отъ основаването й до завладѣването на Балканския полуостровъ отъ Турцитѣ, I, Sofia. Snively C. S., 2010. ‘Macedonia in Late Antiquity’, in J. Roisman, I. Worthington (eds.), A Companion to Ancient Macedonia, Malden/Oxford (Blackwell Companions to the Ancient World), pp. 545–570. Snively C. S., 2011. ‘The New Basilica at Golemo Gradište, Konjuh: A sixth century church in the province of Dardania’, in Ниш и Византија – Niš and Byzantium, 9, pp. 187–201. Tomović G., 2016. ‘Надгробни натпис Воихне из Коњуха’, in B. Miljkobić-Katić (ed.), Споменица д-р. Данице Милић, Belgrade (Зборник радова, 27), pp. 147–166. Varalis Y. D., 2006. ‘Prothesis and Diakonikon: Searching the original concept of the subsidiary spaces of the Byzantine sanctuary’, in A. M. Lidov (ed.), Иеротопия. создание сакральных пространств в Византии и Древней Руси – Hierotopy. The Creation of Sacred Spaces in Byzantium and Medieval Russia, Moscow, pp. 282–292. Veikou M., 2013. ‘Settlements in the Greek countryside from 4th to 9th century: Forms and patterns’, in Antiquité tardive, 21, pp. 125–133. Völling Th., 2001. ‘The last Christian Greeks and the first pagan Slavs in Olympia’, in Kountura-Galake 2001, pp. 303–323.

149

Part 2 Episcopal Residences

11 The Episcopal Palace in Early Byzantium: Historical Development, Architectural Typologies, Domestic Spaces and the Case of Kos Isabella Baldini Università di Bologna (Italy) Throughout Late Antiquity, the most important reference model for the residences of the upper classes was provided by imperial palaces, for their typological variety, for the complexity of their functions and especially for the prestige of their role. It is important to examine episcopia keeping in mind this broader context and the concreteness of the archaeological evidence, which shows an increase in the representative spaces. In the fifth and in the sixth century, many episcopia have boundary walls and an ornate façade, internal pathways, reception halls, and residential and service areas. All these features correspond to the bishop’s role and numerous activities. The case of the episcopium of Kos is significant for the urban and monumental transformations of a large ecclesiastical complex between Late Antiquity and the Middle Ages. Tout au long de l’Antiquité tardive, le modèle de référence le plus important pour les résidences des classes supérieures est représenté par les palais impériaux, pour leur variété typologique, pour la complexité des fonctions et, surtout, pour le prestige de leur rôle. Il est important d’examiner les episcopia en se référant à ce model impérial et au caractère concret des témoignages archéologiques, qui montrent une augmentation progressive des espaces représentatifs. Aux Ve et VIe siècles, de nombreux palais épiscopaux possèdent des murs d’enceinte et une façade décorée, des parcours intérieurs, des salles de réception, des zones résidentielles et de service. Tous ces éléments correspondent au rôle de l’évêque et aux nombreuses activités qu’il doit gérer. Le cas de le palais épiscopal de Kos est significatif des transformations urbaines et monumentales d’un grand complexe ecclésiastique entre l’Antiquité tardive et le Moyen Âge. Während der Spätantike stehen imperiale Paläste mit ihrer typologischen Variabilität, ihrer funktionalen Komplexität und insbesondere mit ihrer Prestigerolle für den Sitz der obersten gesellschaftlichen Klassen. Es ist wichtig, „bischöflich“ in einen breiteren Kontext zu stellen und die archäologische Belastbarkeit der Befunde zu überprüfen, die einen Anstieg im Bereich der Repräsentation zeigen. Während des 6. und 7. Jhs. haben viele episcopia Begrenzungsmauern und eine kunstvolle Fassade, interne Wegenetze, Empfangshallen und Residenz- bzw. Dienstbereiche. All diese Funktionen korrespondieren mit der Rolle des Bischofs und seinen Aktivitäten. Das episcopium von Kos ist bezeichnend für den urbanen und monumentalen Wandel großer bischöflicher Komplexe zwischen der Spätantike und dem frühen Mittelalter. Per tutta la tarda antichità il modello di riferimento più importante per le residenze delle classi elevate è rappresentato dai palazzi imperiali, per la loro varietà tipologica, per la complessità delle funzioni e soprattutto per il prestigio del loro ruolo. È importante esaminare gli episcopia tenendo presente questo quadro di riferimento e la concretezza delle evidenze archeologiche, che mostrano un aumento progressivo degli spazi rappresentativi. Nel V e nel VI secolo molti episcopi hanno mura di cinta e una facciata decorata, percorsi interni, sale di ricevimento, residenziali e di servizio. Tutte questi elementi corrispondono al ruolo del vescovo e alle numerose attività. Il caso dell’episcopio di Kos è significativo delle trasformazioni urbane e monumentali di un grande complesso ecclesiastico tra la tarda antichità e il Medioevo.

153

Isabella Baldini Over the last decade, investigations focusing upon episcopal palaces have greatly expanded the archaeological documentation available on them, enabling scholars to compare new data with earlier methodological reflections.1 These analyses have mostly served to draw the attention of researchers ‘outside’ of the churches or the baptisteries’ in the strictest sense, emphasising some peculiar aspects of this type of architecture and pointing out its importance in the context of public monumentalisation. As a matter of fact, the episcopia represent a particularly significant research field within studies on Late Antiquity and Byzantium for many reasons: for the political, religious and administrative importance that they assume;2 for the function of architectural models within cities; for the development of the conception of the prestigious residence; and for their typological experimentation in connection with the progressive definition of the competences and the role of the bishop and his court, which could at times be very numerous: at least 80 monks, for example, served in the complex of Salamis in Cyprus at the time of Bishop Epiphanius.3

Fig. 1. Ravenna, reconstruction of the Episcopal Palace (Cirelli 2008).

Knowledge of the episcopate as an institution has also expanded in recent decades, thanks to a significant amount of research,4 permitting a better description of the periods of formation and development of the architectural structures that can be defined as ‘episcopal palaces’ (fig. 1).5 Despite this progress in studies, however, the knowledge of episcopia is still partial: to a certain extent, this is due to the heterogeneous nature of the documentation and is the direct consequence of the archaeological practices applied to church architecture in the past. Another difficulty stems from the frequent absence of the movable finds used to identify the functions of rooms when specific typological features are not immediately visible, as in the case of reception halls, triclinia and bath buildings. For this reason, one of the prevalent methods in publications on episcopal palaces is the identification of close analogies within contemporary residential architecture (palaces, praetoria, domus, villae).6 On the other hand, these analogies sometimes make the identification of some residential structures as episcopal

Fig. 2. San Giusto (Apulia), Episcopal complex (Volpe 2009).

This paper was conceived as an introduction to the Round Table ‘The Episcopal palace in Early Byzantium: Historical development, architectural typologies, domestic spaces’, at the 23rd International Congress of Byzantine Studies. To the Round Table were invited P. Chevalier, P. Niewöhner, N. Poulou, E. Rizos, H. Saradi and A. Taddei. On the general development of eastern episcopia, see: Pallas 1971; Malaspina 1975; Müller Wiener 1984; Müller Wiener 1989; Sodini 1989 (Asia Minor); Real 2003; Baldini Lippolis 2005, pp. 102–136; 2007b; Ceylan 2007 (Asia Minor); Seeliger, Krumeich 2007 (Egypt); Marano 2007 (Northern Italy); Uytterhoeven 2007, pp. 39–40 (bibliographical review); Caillet 2010; Baldini 2014; Ceylan 2014 (Asia Minor); Baldini 2015b. 2 Hohlweg 1971; Durliat 1996; Rebillard, Sotinel 1998; Rapp 2000; Brown 2002, pp. 45–73; Rapp 2004; Rapp 2005; AsimakopoulouAtzaka, Parcharidou-Anagnostou 2009 (bishops’ inscriptions in Greece). Cf. also the contribution of H. Saradi in this volume. 3 John (disciple of Epiphanius), Vita sancti Epiphani Constantiniensis, 19. 4 Uytterhoeven 2007, p. 40. 5 On the increase in the dimensions of these complexes, see: Baldini Lippolis 2005, pp. 102–136. 6 Baldini 2015a. 1

palaces difficult. An example is the archaeological recognition of rural episcopia (the seats of the country bishops known from sources); the complex of S. Giusto in Apulia, for example, is a candidate for this function (fig. 2).7 In the mid-fifth century a basilica with baptistery and annexes was added to a Late Roman villa. Between the end of the fifth and the beginning of the sixth century, a cemetery church was built, while the villa was enlarged, and thermal baths were also constructed. The complex could be identified with the seat of Probus episcopus Carmeianensis, present at the Roman councils of the early sixth century. Volpe 2007; Volpe, Annese, Favia 2007; Volpe 2008, 2009, 2011, pp. 355–365; Volpe, Biffino, Giuliani 2001.

7

154

The Episcopal Palace in Early Byzantium

Fig. 3. Aphrodisias, so-called Episcopal palace (I. Baldini).

In many cases, the morphological similarities make the identification of a bishop’s palace difficult, particularly when it has been inserted into a pre-existing architectural context; the famous Palace of Aphrodisias (fig. 3), located at the bouleuterion and at the northern limit of the agora, certainly belonged to a member of the local elite during the fourth century, possibly with public duties (as demonstrated by the paintings found in the building).8 Maybe it was used as a bishop’s seat after the transformation of the temple of Aphrodite into a church during the reign of Theodosius II. It is unclear, however, whether this endowment of the building to the local Church can really be confirmed; in fact, no significant element of a Christian transformation of the building is recognisable, nor are there traces of a structural link between the cathedral and the residential complex.

From this point onwards, over the course of two centuries, a new conception of the episcopal residence matured, in the sense that it became a real palace with legal appurtenances (such as the right of asylum, as described in Codex Theodosianus);9 the acquisition of cells, houses, gardens, baths, open areas and porches), and is defined as an episkopeion for the first time in early fifth-century sources.10 It became a palace not only with regard to its residential characteristics, but also its administrative and economic functions, which were mirrored in all cities in which the local Church played a very central role in the management of public activities. With the development of the residential and representative aspects of the episcopia, an architectural specialisation became more defined over time, built upon precise choices. On the one hand, some residential buildings or rooms remained from the previous residential tradition; on the other hand, new forms of organisation of the space were experimented with, resulting in a wide variety of solutions that depended upon the pre-existing structures and on the mode of aggregation of the building units.

One of the achievements of recent research is the reconstruction of the general lines of development of the Late Antique episcopia, and their gradual conception as palaces. In reality, the relationship between the architectural features and those of the coeval residential architecture changed over the decades; this process began with the lack of any specific importance within the housing sector, which coincided with the simplest solutions in the domus ecclesiae of the third and the fourth centuries, in which the residential rooms were adapted to various ecclesiastical functions: liturgical, productive and charitable, within the centre of the Christian community.

Within this range, some constants can be identified in the fifth and the sixth century. For example, a distinguishing feature of these complexes was the connection between the residential area and episcopal structures (the cathedral, Codex Theodosianus 9.45.3–5. Palladius, Historia Lausiaca 63.2, in reference to the Constantinople complex. 9

10

8

Cormack 1990; Herbert 2000; Baldini Lippolis 2001, pp. 119–120.

155

Isabella Baldini the baptistery, other churches and chapels, as well as rooms for charitable or catechismal functions). Sometimes the connection is via passages located on an upper floor, as is seen in imperial palaces.11 An example has recently been investigated in the episcopal palace of Kos, where two symmetrical stairways lead to the first floor, overlooking the main church (fig. 5).12 Other common features can be identified by the presence of boundary walls13 and by an ornate façade, or in the development of internal pathways, which followed liturgical practices. The dimensions are generally wide, with progressive enlargements (e.g. at Aquileia,14 Poreč,15 Ravenna16 and Salona).17 Within the episcopia, there were also many reception halls and service areas, e.g. housing, kitchens, baths,18 workshops, warehouses, archives and katagoghia, the last known mainly from literary sources.19 Also, in rare cases, offices for the administration of the episcopal finances are recorded,20 reflecting the functions of the episcopate as a management centre.

Fig. 4. Canosa (Apulia), brick with Sabinus’ monogram (Volpe 2007).

These aspects are also linked with some specificity to the production and commercialisation of artefacts. On occasion, bricks and other manufactured articles marked with the name of the bishop show this productive capacity; under Bishop Sabinus, for example, bricks with his monogram were produced in Canosa (fig. 4).21

In the context of the individual choices and the transmission of models, it is interesting to observe that within episcopal architecture some recessive typological elements were sometimes present, as may be seen for example in the triclinia with more than one apse, a style derived from palatial architecture.26 In cases such as the episcopal triclinium in Ravenna27 or the Lateran Palace in Rome,28 the presence of this particular type of hall depended upon factors related to prestige and capacity issues; bishops, in fact, could host a great number of dinner guests (even 200 in Ephesus29) and had the social need to exhibit a luxurious triclinium, as occasionally described by written sources and shown in archaeological remains.30

It should also be remembered that sources mention in a positive manner the role of the bishops in the procurement of quality materials for churches, as in the case of the Constantinople marbles supplied by Archbishop Maximian for the churches of Ravenna,22 or by Laurence of Sipontum in the sixth century.23 By contrast, Antoninus, bishop of Ephesus, was sanctioned for having transferred architectural material from a baptistery and from a church to his residence (for use in the baths and in the triclinium).24 It is a reminder of the public law governing such transfers of architectural elements, which entrusted judges with the authorisation to grant permits to eminent applicants.25

The role of the patriarchal sees in the transmission of typological models is a chapter yet to be written; indeed, little is known of these architectural complexes during the early Byzantine phases. One instance is that of the Patriarchate of Constantinople;31 despite the paucity of archaeological evidence, we know that it included the bishop’s residence, a tribunal, baths, a triclinium with three apses, an atrium with arcades in three orders, a library, a katagoghion and other buildings.

11 E.g. in Constantinople, the so-called Palace of Ormisdas: Janin 1964, p. 137; Baldini Lippolis 2001, pp. 181–182. 12 Giletti 2015, p. 208. 13 Baldini 2015b, pp. 158–159 (Nea Anchialos, Side, Constantinople), with bibliography. 14 Bertacchi 1985; Baldini Lippolis 2005, pp. 114–118. 15 Bertacchi 1985; Matejčić 1995; Baldini Lippolis 2001, pp. 245 and 129–131. 16 Rizzardi 1989; Miller 1991–92; Baldini Lippolis 2001, p. 258;2005, pp. 119–122; Cirelli 2008, pp. 75–76 and 247–248; Mauskopf Deliyannis 2011, pp. 188–196. 17 Bertacchi 1985; Baldini Lippolis 2005, pp. 123–124. 18 Baldini 2015b, p. 159 (e.g. Ravenna, Ephesus, Philippi, Salamis of Cyprus, Jerusalem, Constantinople), with bibliography. 19 Malaspina 1975, pp. 44–45; Baldini 2015b, p. 158. 20 Baldini et al. 2019 21 Volpe 2011, p. 407. 22 Volpe 2007, 2011. 23 Fabbri 1989, p. 181 24 Cf. the contribution of H. Saradi in this volume. 25 Baldini Lippolis 2007a.

The problem of the representative characteristics of episcopia is related to the perception of the same Baldini 2014, pp. 165–166, with references. Cf. n. 16, and Mauskopf Deliyannis 2011, pp. 100–101. 28 Luchterhandt 1999; Baldini Lippolis 2001, pp. 264; Santangeli Valenzani 2011, pp. 27–28. 29 Palladius, Dialogus de vita sancti Joannis Chrysostomi 13. 30 Ellis 1997, pp. 41–51; Baldini Lippolis 2001, pp. 79–83; Dumbabin 2003, pp. 193–195; Vroom 2007a, pp. 318–325; Vroom 2007b, p. 194; Baldini 2015b, p. 167, with bibliography. 31 Guilland 1956; Mango 1959; Janin 1962; Cormack, Hawkins 1977; Müller Wiener 1989, pp. 668–670; Dark, Kostenec 2006; Dark, Kostenec 2014. A bibliography and some preliminary reflections are offered by A. Taddei in Krsmanović, Milanović 2016. 26 27

156

The Episcopal Palace in Early Byzantium buildings as a dominant reality of urban life; the particular connotations of the religious beneficiary of the structure – the bishop – prompt the reflection of an interesting dialectic between huge architecture, luxurious decoration and the conveniences appropriate to sober habits, in keeping with the religious life. These elements especially emerge, even in this case, from literary sources: sensational episodes are recorded, as seen in the example of John Chrysostom and his abandonment of sumptuous episcopal banquets.32 The reception of models par excellence often intersects with local architectural traditions, and with the existence of homogeneous cultural areas. As an example of recently completed research, the case of the episcopal palace at Kos (fig. 5) can be considered significant for the area of Greece and Asia Minor. As one of the most important Mediterranean urban centers of the Hellenistic and Roman periods, the city of Kos was characterised in Early Christian times by its sustained use of roads, as well as its highly important monumental areas, harbour storehouses and private buildings.33 The effects of the destructive earthquakes of AD 469 and 554 (the latter described by the Byzantine writer Agathias, who was an eyewitness to the disaster34), and a plague in 542,35 did not in fact cause any drastic breaks in the settlement’s life; rather, they provided the opportunity for renewal, in accordance with updated models and lifestyles. In the process of Christianisation, the ecclesiastical architecture increasingly engaged most of the available resources. The episcopate of Kos was very ancient and prestigious, and a list of bishops is known from the first half of the fourth century,36 despite the fact that no Christian building has been discovered that can be dated to before the middle of the fifth century. A complex excavated in the 1930s in the area of the Western Baths was recently analysed anew by the University of Bologna.37 The surveys implemented show the process – very common in Late Antiquity – of the transfer of public properties to the ownership of the Church and their reuse for new functions. From the midfirst century AD onward, the block of the Roman Western Baths occupied a former residential area that extended from the eastern stoa of the Hellenistic Gymnasium to the cardo, which connected the south-western districts of the town with the harbour.38 A significant restructuring of the bath complex took place towards the end of the third or the beginning of the fourth century AD.39 Given the large amount of reused material Cf. the contribution of H. Saradi in this volume. For a synthesis, with bibliography, see Baldini 2015a; Bersellini 2015; Casadei 2015; Cosentino 2015a; Livadiotti 2015a; Orlandi 2015b; Patrignani 2015; Pellacchia 2015a. 34 Agathias, Historia 2.16.1. 35 Cosentino 2015a, p. 111, with bibliography. 36 Cosentino 2015a, pp. 105–106 and 116. 37 Baldini, Livadiotti 2015. 38 Livadiotti 2015b, pp. 146–150. 39 Livadiotti 2015b, pp. 150–153. 32 33

Fig. 5. Kos, Early Byzantine Episcopal complex in the second half of the sixth centuy (P. Baronio, in Baldini, Livadiotti 2015).

157

Isabella Baldini

Fig. 6. Kos, reconstruction of Basilica A (P. Baronio, in Baldini, Livadiotti 2017).

in the walls, the intervention appears to have been due to some destructive seismic action.

the district dating to shortly after the mid-sixth century, most likely after the seismic activity of 554. At that time, the episcopal seat moved from its previous position – unfortunately still unknown – which had probably been irreparably damaged.43

Still in use at the end of the fourth century, the Baths experienced a drastic change in function after the midfifth century, when the first Christian building (Basilica A) was built in its northern sector, above the frigidarium.40 It is possible that this change occurred following the earthquake of 469. After this event, the Western Baths could not be maintained and became the property of the local Church.

With access from the decumanus, the new complex included entrance rooms,44 an apsidal room with a monumental nymphaeum,45 private baths (fig. 7),46 a baptistery (fig. 8–9),47 two churches (one of which was built from the restructuring of Basilica A),48 a triclinium,49 and a second entrance accessible from the cardo.

Basilica A (fig. 6) was smaller than the other religious buildings established in Kos some decades before (the Basilicas of the Harbour41 and Epta Vimata42). Its religious role was probably linked to the neighbourhood. Its acquisition of episcopal status seems to correspond to a subsequent reorganisation of the spaces and paths within

The palace, with its impressive mosaic pavements, once again returned to its former role as an urban epicentre of Baldini 2015a, pp. 21–22; 2015b, pp. 155–157. Baldini 2015b, p. 193–195. 45 Baldini 2015b, p. 156; Pellacchia 2015b, p. 191–192. 46 Baldini 2015b, p. 156; Baronio 2015; Pellacchia 2015b, pp. 184–191. 47 Baldini 2015b, pp. 165–166; Boccardi 2015; Giletti 2015, pp. 201–202. 48 Baldini 2015b, pp. 154–155; Marsili 2015. 49 Baldini 2015b, pp. 166–167; Marsili 2015, pp. 229–230. 43 44

Baldini 2015b, pp. 154–155; Marsili 2015. Pellacchia 2015a, pp. 38–45. 42 Orlandi 2015a. 40 41

158

The Episcopal Palace in Early Byzantium

Fig. 7. Kos, Early Byzantine baths of the Episcopal palace, second half of the sixth century (P. Baronio, D. Pellacchia, in Baldini, Livadiotti 2015).

Fig. 8. Kos, Episcopal palace, baptistery (room 28) (Archive SAIA).

be observed in Basilica B. The new episcopal complex was conceived according to a monumentality that corresponded to the importance of the city and its bishopric.

primary importance, and was destined to maintain this importance until at least the end of the seventh or the eighth century, at which point new building activities can 159

Isabella Baldini

Fig. 9. Kos, Episcopal palace, baptistery (room 28), reconstruction (F. Frasca).

The compactness of the building system can be compared to other eastern and western episcopia in both its planimetric and formal aspects;50 it was a multifunctional ensemble conceived to perform liturgical, representative and residential functions. Spaces were enlarged for religious actions, and new decorations, flooring (mosaic51 and opus sectile52), and architectural components (e.g. through the almost exclusive use of Ionic capitals for the colonnades53) were added. It was a huge financial commitment, which also included the import of pre-worked or partially worked Proconnesian elements. An ambo with an octagonal platform (fig. 10), as was documented by the discovery of some white marble elements, can be attributed to Basilica A.54 It is probable that most of the marble elements came instead from the southern church (Basilica B). The importance of this church is demonstrated by its central location and the connection of the monumental entrance rooms to the baptistery by special pathways paved with mosaic and marble floors. The importance of the building is also evinced by its architectural elements and liturgical furniture. In particular, it is important to note the discovery of a huge ambo in Proconnesian marble (fig. 11–12); its monolithic platform is distinguished by its elegant decor, typical of Constantinople, which also includes a cross and the inscription ‘Light – Life’.55 In the southern sector of the complex, a cover belonging to a black marble reliquary was found (fig. 13),56 perhaps originating from a room adjacent to the baptistery and Basilica B.

Fig. 10. Kos, Castle of Nerantzia, marble slab of the ambo pertaining to Basilica A (Baldini, Livadiotti 2015).

Unfortunately, the central sector of the quarter is lost, and a simple analysis of the different parts does not easily permit one to appreciate completely the combination of 50 Baldini 2015b, pp. 159–164 (e.g. Iustiniana Prima, Loloudies, Heraklea Lyncestis, Philippi, Nea Anchialos, Ephesus, Jerash), with bibliography. For the episcopal palace of Miletus, cf. Niewöhner’s contribution in this volume. 51 Baldini 2015b, p. 164; Frasca 2015. 52 Baldini 2015b, p. 164; Lo Ioco 2015. 53 Mazzilli 2015, pp. 284–291 54 Baldini 2015b, p. 168; Mazzilli 2015, pp. 312–313. 55 Baldini 2015b, pp. 168–169; Mazzilli 2015, pp. 307–312 and 363–370. 56 Baldini 2015b, pp. 170–171; Giletti 2015, pp. 206–207.

Fig. 11. Kos, Castle of Nerantzia, marble platform of the ambo pertaining to Basilica B (Baldini, Livadiotti 2015).

external architectural influences and local traditions. One example is the baptistery, which includes not just a main pool accessible from two opposing stairs for the ritual 160

The Episcopal Palace in Early Byzantium

Fig. 12. Kos, Reconstruction of the ambo pertaining to Basilica B (G. Mazzilli, in Baldini, Livadiotti 2015).

immersion, but also two smaller secondary pools, utilised in relation to baptismal practices undertaken before accessing the sacrament.57

according to the models of hierarchical and representative behaviour acquired by Late Antique clergy through the example set by the high aristocracy. A semicircular table with stibadium left traces of ashes (fig. 14), which are still visible in old photographs taken of the room.58 The northern basilica (or Basilica A) probably had a complementary

One of the indicators of the ‘international’ models adopted by the Koan community is also evinced by the form of the Basilica Thermarum’s banquet hall; it was built 57

Baldini 2015b, pp. 165–166.

58

161

Baldini 2015b, pp. 166–167; Marsili 2015, pp. 229–230.

Isabella Baldini role, as suggested by its location and relationship with the bishop’s dining hall. Basilica B shows evidence of prolonged use; a series of marble slabs found in the area dating from the end of the seventh to the eighth century were decorated with the monograms of important dignitaries (fig. 15–16). The monograms were part of a series of eight marble slabs that were probably placed in the narthex of the church. They have been interpreted as being relevant to members of the laity, not to the clergy: a hypatos (or patrikios) kai chartoularios, and a komēs. As for their chronology, it has been suggested that they date from the second half of the seventh century or, more likely, the first half of the eighth century.59 The possibility of extending the use of the episcopal basilica up to this epoch has greatly contributed to the scholarly reconstruction of the development of the local Church and the entire Koan community, in correspondence with a series of pieces of archaeological evidence proving the continuity of life in the city beyond Late Antiquity. After this period, it becomes difficult to reconstruct the development of the episcopal district over the centuries that followed. It is evident from the ruins that are still visible and the written sources60 that stone material was

Fig. 13. Kos, Castle of Nerantzia, marble cover of a reliquary found in room 26 (Baldini, Livadiotti 2015).

Fig. 14. Kos, Episcopal palace, dining hall (room 6), with traces of burning of the dining table on the mosaic floor (Archive SAIA). 59 Baldini 2015b, p. 171–172; Cosentino 2015b (interpretation and chronology); Mazzilli 2015, pp. 303–305 and 376–379. 60 Baldini 2015a, p. 27, with bibliography.

162

The Episcopal Palace in Early Byzantium

Fig. 15. Kos, marble slabs from Basilica A, now at the Castle of Nerantzia (Archive SAIA).

Fig. 16. Kos, reconstruction of marble slabs from Basilica A: carved monograms of high rank imperial officials, dated between the end of the seventh and the eighth centuries (S. Cosentino, G. Mazzilli, in Baldini, Livadiotti 2015).

163

Isabella Baldini in Italia tra Tardoantico ed Altomedioevo. Atti del IX congresso nazionale di archeologia cristiana, Agrigento, 20–25 novembre 2004, I, Palermo, pp. 227–258.

later removed in order to build the medieval fortifications of the city. This example makes it possible to highlight the problem of the development of episcopia during the ‘Dark Ages’, when episcopal palaces evolved towards different residential and representative forms that were often conservative in nature. At the same time, they adapted to architectural models that definitively abandoned the Late Antique peristyle type, while storage and production facilities increased. In the absence of epigraphic and stratigraphic documentation, spolia, repair works and floor patching have mostly indicated the prolonged use of these spaces, continuing beyond Late Antiquity.

Baronio P., 2015. ‘Ipotesi di restituzione delle terme episcopali’, in Baldini, Livadiotti 2015, pp. 196–198. Bersellini G., 2015. ‘I quartieri orientali’, in Baldini, Livadiotti 2015, pp. 81–85. Bertacchi L., 1985. ‘Contributo allo studio dei palazzi episcopali paleocristiani: i casi di Aquileia, Parenzo e Salona’, in Aquileia Nostra, 56, pp. 361–412. Boccardi L., 2015. ‘Il Battistero’, in Baldini, Livadiotti 2015, pp. 210–213. Brown P., 2002. Poverty and Leadership in the Later Roman Empire, Hanover (NH)/London (The Menahem Stern Jerusalem Lectures), pp. 45–73.

Bibliography Abbreviations:

Caillet J.-P., 2010. ‘Remarques sur la problématique des “palais épiscopaux” à la fin de l’Antiquité’, in N. Duval, V. Popović (eds.), Caričin Grad, III, Rome (Collection de l’École française de Rome, 75/3), pp. 508–523.

FelRav = Felix Ravenna. JÖB = Jahrbuch der Österreichischen Byzantinistik. Asimakopoulou-Atzaka P., Parcharidou-Anagnostou M., 2009. ‘Mosaici con iscrizioni vescovili in Grecia (dal IV al VII secolo)’, in R. Farioli Campanati et al. 2009, pp. 25–44.

Casadei S., 2015. ‘I quartieri occidentali’, in Baldini, Livadiotti 2015, pp. 93–94. Ceylan B., 2007. ‘Episkopeia in Asia Minor’, in Lavan, Özgenel, Sarantis 2007, pp. 169–194.

Baldini I., 2014. ‘Palatia, praetoria ed episcopia: alcune osservazioni’, in Pensabene, Sfameni 2014, pp. 163–170.

Ceylan B., 2014. ‘Episcopeia as a Reflection of the Image of the Bishop’, in Pensabene, Sfameni 2014, pp. 191–196.

Baldini I., 2015a. ‘La città Cristiana’, in Baldini, Livadiotti 2015, pp. 12–27.

Cirelli E., 2008. Ravenna. Archeologia di una città, Borgo San Lorenzo (Contributi di archeologia medievale, 2).

Baldini I., 2015b. ‘Il complesso episcopale. Sviluppo architettonico e modelli’, in Baldini, Livadiotti 2015, pp. 154–172.

Cormack R., 1990. ‘The temple as the cathedral’, in C. Roueché, K. T. Erim (eds.), Aphrodisias Papers, I, Ann Arbor (Journal of Roman Archaeology. Supplementary Series, 1), pp. 75–88.

Baldini I. et alii, 2019. ‘L’età tardoantica (fase 5) / L’età tardoantica : 360-500 (fase 6) / L’età bizantina : dal 500 alle fine del VII secolo (fase 7) / L’età bizantina : VIII-IX secolo (fase 8)’, in E. Lippolis, L. M. Caliò, C. Giatti (eds.), Gortina, VIII/1, Athens (Monografie della Scuola archeologica di Atene e delle Missioni italiane in Oriente, 27), pp. 365-544.

Cormack R., Hawkins E., 1977. ‘The mosaics of St. Sophia at Istanbul: The rooms above the southwest vestibule and ramp’, in Dumbarton Oaks Papers, 31, pp. 175–251. Cosentino S., 2015a. ‘Episcopato e società a Kos tra IV e VIII secolo’, in Baldini, Livadiotti 2015, pp. 105–116.

Baldini I., Livadiotti M. (eds.), 2015. Archeologia protobizantina a Kos. La città e il complesso episcopale, Bologna (Collana DiSCi, 6).

Cosentino S., 2015b. ‘La documentazione epigrafica’, in Baldini, Livadiotti 2015, pp. 240–245.

Baldini Lippolis I., 2001. La domus tardoantica nelle città del Mediterraneo. Forme e rappresentazioni dello spazio domestico nelle città del Mediterraneo, Imola (Università degli studi di Bologna. Studi e scavi, 17).

Dark K., Kostenec J., 2006. ‘The Byzantine Patriarchate in Constantinople and the baptistery of the Church of Hagia Sophia’, in Architectura, 36/2, pp. 113–130. Dark K., Kostenec J., 2014. ‘The Patriarchal Palace at Constantinople in the seventh century: Locating the Thomaites and the Makron’, in Jahrbuch der Österreichischen Byzantinistik, 64, p. 33–40.

Baldini Lippolis I., 2005. L’architettura residenziale nelle città tardoantiche, Rome (Le bussole. Archeologia, 211). Baldini Lippolis I., 2007a. ‘Private space in Late Antique cities: Laws and building procedures’, in Lavan, Ozgenel, Sarantis 2007, pp. 197–238.

Dumbabin K. M., 2003. The Roman Banquet. Images of Conviviality, Cambridge.

Baldini Lippolis I., 2007b. ‘Osservazioni sulla tipologia delle prime residenze episcopali in Italia’, in R. M. Bonacasa Carra, E. Vitale (eds.), La cristianizzazione

Durliat J., 1996. ‘Évêque et administration municipale au VIIe siècle’, in C. Lepelley (ed.), La fin de la cité antique et le début de la cité médiévale, de la fin du 164

The Episcopal Palace in Early Byzantium IIIe siècle à l’avènement de Charlemagne. Actes du colloque tenu à l’Université de Paris X–Nanterre les 1, 2 et 3 avril 1993, Bari (Munera. Studi storici sulla Tarda Antichità, 8), pp. 273–86.

Karolingerzeit. Karl der Große und Papst Leo III. in Paderborn, III, Mainz, pp. 109–122. Malaspina M., 1975. ‘Gli episcopia e le residenze ecclesiastiche nella ‘pars orientalis’ dell’impero romano’, in M. Cagiano de Azevedo (ed.), Contributi dell’Istituto di archeologia, V, Milano (Pubblicazioni della Università cattolica del Sacro Cuore. Scienze storiche, 10), pp. 29–173.

Duval N. (ed.), 1989. Actes du XIe Congrès international d’archéologie chrétienne, Lyon, Vienne, Grenoble, Genève et Aoste (21–28 septembre 1986), Rome/ Vatican City (Collection de l’École française de Rome, 123; Studi di Antichità cristiana, 41).

Mango C., 1959. The Brazen House. A Study of the Vestibule in the Imperial Palace of Constantinople, Copenhagen (Arkaeologisk-kunsthistoriske Meddelelser, 4,4).

Ellis S., 1997. ‘Late-Antique dining: Architecture, furnishings and behaviour’, in R. Laurence, A. Wal­ lace-Hadrill, Domestic Space in the Roman World: Pompeii and Beyond, Portsmouth (Journal of Roman Archaeology. Supplementary Series, 22), pp. 41–51.

Marano Y., 2007. ‘Domus in qua manebat episcopus. Episcopal residences in northern Italy during Late Antiquity (4th to 6th centuries A.D.)’, in Lavan, Özgenel, Sarantis 2007, pp. 97–129.

Fabbri M., 1999. ‘La basilica paleocristiana’, in M. Mazzei (ed.), Siponto antica, Foggia, pp. 179–187.

Marsili G., 2015. ‘Il settore settentrionale e la Basilica A. Fasi e sviluppo’, in Baldini, Livadiotti 2015, pp. 214–234.

Farioli Campanati R., Rizzardi C., Porta P., Augenti A., Baldini Lippolis I. (eds.), 2009. Ideologia e cultura artistica tra Adriatico e Mediterraneo orientale, IV–X secolo. Il ruolo dell’autorità ecclesiastica alla luce di nuovi scavi e ricerche. Atti del Convegno internazionale, Bologna–Ravenna, 26–29 novembre 2007, Bologna (Università degli studi di Bologna. Studi e scavi, 19).

Matejčić I., 1995. ‘The Episcopal Palace at Poreč. Results of a recent exploration and restoration’, in Hortus Artium Medievalium 1, 1995, pp. 84–89. Mauskopf Deliyannis D., 2010. Ravenna in Late Antiquity, Cambridge 2010.

Frasca F., 2015. ‘Mosaici’, in Baldini, Livadiotti 2015, pp. 384–398.

Mazzilli G., 2015. ‘La decorazione architettonica e di arredo liturgico’, in Baldini, Livadiotti 2015, pp. 274–383.

Giletti F., 2015. ‘La strada e gli edifici ad est’, in Baldini, Livadiotti 2015, pp. 173–176.

Miller M. C., 1991–92. ‘The development of the archiepiscopal residence in Ravenna, 300–1330’, in FelRav 141–144, pp. 145–173.

Guilland R., 1956. ‘Le Thomaites et le patriarcat’, in JÖB, 5, pp. 27–40.

Müller-Wiener W., 1984. ‘Riflessioni sulle caratteristiche dei palazzi episcopali’, in FelRav, 125–126, pp. 103–145

Hebert L., 2000. The Temple-Church at Aphrodisias, PhD Dissertation, New York University.

Müller Wiener W., 1989. ‘Bischofsresidenzen des. 4.–7. Jhs. im östlichen Mittelmeer-Raum’, in Duval 1989, pp. 651–709.

Hohlweg A., 1971. ‘Bischof und Stadtherr im frühen Byzanz’, in JÖB, 20, pp. 51–62.

Orlandi L., 2015a. ‘Il complesso di Eptà Vimata’, in Baldini, Livadiotti 2015, pp. 57–78.

Janin R., 1962, ‘Le palais patriarcal de Constantinople’, in Revue des études byzantines, 20, pp. 131–155.

Orlandi L., 2015b. ‘I quartieri meridionali’, in Baldini, Livadiotti 2015, pp. 86–92.

Krsmanović B., Milanović L. (eds.), 2016. Proceedings of the 23rd International Congress of Byzantine Studies, Belgrade, 22–27 August 2016. Round Tables, Belgrade: http://www.byzinst-sasa.rs/srp/uploaded/PDF%20 izdanja/round%20tables.pdf.

Pallas D. I., 1971. ‘Episkopeion’, in K. Wessel (ed.), Reallexikon zur byzantinischen Kunst, II, Stuttgart, cols. 335–371.

Lavan L., Ozgenel L., Sarantis A. (eds.), 2007. Housing in Late Antiquity. From Palaces to Shops, Leiden/Boston (Late Antique Archaeology, 3/2).

Patrignani R., 2015. ‘Edilizia residenziale tardoantica’, in Baldini, Livadiotti 2015, pp. 79–80. Pellacchia D., 2015a. ‘Il quartiere del porto’, in Baldini, Livadiotti 2015, pp. 35–53.

Livadiotti M., 2015a. ‘Le aree pubbliche’, in Baldini, Livadiotti 2015, pp. 28–34.

Pellacchia D., 2015b. ‘Il settore meridionale. Fasi e sviluppo’, in Baldini, Livadiotti 2015, pp. 182–195.

Livadiotti M., 2015b. ‘Il complesso termale’, in Baldini, Livadiotti 2015, pp. 146–153.

Pensabene P., Sfameni C. (eds.), 2014. La villa restaurata e i nuovi studi sull’edilizia tardoantica. Atti del convegno internazionale del Centro interuniversitario di studi sull’edilizia abitativa tardoantica nel Mediterraneo, CISEM, Piazza Armerina 7–10 novembre 2012, Bari (Insulae Diomedeae, 23).

Lo Ioco S., 2015. ‘Percorsi e pavimentazioni. Sectilia’, in Baldini, Livadiotti 2015, pp. 399–403. Luchterhandt M., 1999. ‘Päpstlicher Palastbau und höfisches Zeremoniell unter Leo III’, in C. Stiegemann, M. Wemhoff (eds.), 799. Kunst und Kultur der 165

Isabella Baldini imperial – The Village, the Industries, the Imperial Property, Bari (Insulae Diomedeae, 17), pp. 345–458.

Rapp C., 2000. ‘The elite status of bishops in Late Antiquity in ecclesiastical, spiritual, and social contexts’, in Arethusa, 33, pp. 379–399.

Volpe G., Annese C., Favia P., 2007. ‘Terme e complessi religiosi paleocristiani: il caso di San Giusto’, in M. Guérin-Beavois, J.-M. Martin (eds.), Bains curatifs et bains hygiéniques en Italie de l’Antiquité au Moyen Âge. Actes du colloque réuni à Rome, 22, 23 mars 2004, Rome (Collection de l’École française de Rome, 383), pp. 217–261.

Rapp C., 2004. ‘Bishops in Late Antiquity: A new social and urban elite?’, in J. Haldon, L. Conrad (eds.), The Byzantine and Early Islamic Near East, VI, Princeton (Studies in Late Antiquity and Early Islam, 1), pp. 149–178. Rapp C., 2005. Holy Bishops in Late Antiquity. The Nature of Christian Leadership in a Time of Transition, Berkeley/Los Angeles/London (The Transformation of the Classical Heritage, 37; The Joan Palevsky Imprint in Classical Literature).

Volpe G., Biffino A., Giuliani R., 2001. ‘Il Battistero del complesso paleocristiano di San Giusto (Lucera)’, in L’edificio battesimale in Italia. Aspetti e problem. Atti dell’VIII Congresso nazionale di archeologia cristiana, Genova, Sarzana, Albenga, Finale Ligure, Ventimiglia, 21–26 settembre 1998, Bordighera (Istituto internazionale di studi liguri. Atti dei convegni, 5), pp. 1089–1130.

Real U., 2003. ‘Die Bischofsresidenz in der spätantiken Stadt’, in G. Brands, H.-G. Severin (eds.), Die spätantike Stadt und ihre Christianisierung. Symposion vom 14. bis 16. Februar 2000 in Halle-Saale, Wiesbaden (Spätantike, frühes Christentum, Byzanz. Reihe B, 11), pp. 219–237.

Vroom J., 2007a. ‘The archaeology of Late Antique dining habits in the eastern Mediterranean: A pre­ liminary study of the evidence’, in L. Lavan, E. Swift, T. Putzeys (eds.), Objects in Context. Objects in Use Material Spatiality in Late Antiquity, Leiden/Boston (Late Antique Archaeology, 5), pp. 313–361.

Rebillard É., Sotinel C. (eds.), 1998. L’évêque dans la cité du IVe au Ve siècle. Image et autorité. Actes de la table ronde organisée par l’Istituto patristico Augustinianum et l’École française de Rome (Rome, 1er et 2 décembre 1995), Rome (Collection de l’École française de Rome, 248).

Vroom J., 2007b. ‘The changing dining habits at Christ’s table’, in L. Brubaker, K. Linardou (eds.), Eat, Drink, and be Merry (Luke 12:19): Food and Wine in Byzantium. Papers of the 37th annual Spring Sympo­ sium of Byzantine Studies, in Honour of Professor A. A. M. Bryer, Aldershot/Burlington (VT) (Society for the Promotion of Byzantine Studies. Publications, 13), pp. 191–222.

Rizzardi C., 1989. ‘Note sull’antico episcopio di Ravenna: formazione e sviluppo’, in Duval 1989, pp. 711–731. Santangeli Valenzani R., 2011. Edilizia residenziale in Italia nell’altomedioevo, Rome (Studi superiori. Archeologia, 615). Seeliger H. R., Krumeich K., 2007. Archäologie der antiken Bischofssitze, I, Wiesbaden (Sprachen und Kulturen des christlichen Orients, 15). Sodini J.-P., 1989. Les groupes épiscopaux de Turquie, à l’exception de la Cilicie, in Duval 1989, pp. 405–426. Uytterhoeven I., 2007. ‘Housing in Late Antiquity: Thematic perspectives’, in Lavan, Özgenel, Sarantis 2007, pp. 25–66. Volpe G., 2007. ‘Architecture and Church power in Late Antiquity: Canosa and San Giusto (Apulia)’, in Lavan, Özgenel, Sarantis 2007, pp. 131–168. Volpe G., 2008. ‘Vescovi rurali e chiese nelle campagne dell’Apulia e dell’Italia meridionale fra Tardoantico e Altomedioevo’, in Hortus Artium Medievalium, 14, pp. 31–47. Volpe G., 2009. ‘L’iniziativa vescovile nella trasformazione dei paesaggi urbani e rurali in Apulia: i casi di Canusium e di San Giusto’, in R. Farioli Campanati et al. 2009, pp. 405–423. Volpe G., 2011. ‘Vagnari nel contesto dei paesaggi rurali dellʼApulia romana e tardoantica’, in A. M. Small (ed.), Vagnari. Il villaggio, l’artigianato, la proprietà

166

12 The Episcopal Palaces: Worldly Splendour in the Style of the Ruling Class versus Monastic Virtues Helen Saradi Πανεπιστήμιο Πελοποννήσου / University of the Peloponnes (Greece) This paper examines the episcopal palaces of the Empire of the East in the context of the contrast between the splendour of the upper class and the ideal of the monastic life. In order to evaluate the palatial style of the bishops’ residences, one should take into consideration the fact that from the time of Constantine I, many members of the curial and senatorial class were appointed bishops. There are various causes for this development: by joining the clergy, the curiales avoided their fiscal charges; in the large cities of the Empire, the office of the bishop offered great social prestige, wealth, contact with the emperor and the leading class, and an affluent life like that of the aristocrats. By contrast, texts attest that in the small cities of the provinces, many bishops led an ascetic life. Some bishops resided in monasteries, while others assumed real civil power in the cities, with secular activities in construction and in protecting the poor, prisoners, captives etc. The dichotomy between these two models pervades the texts that describe the aristocratic affluence in which certain bishops lived, the ascetic virtues of others, and the combination of secular activities of some bishops with the Christian virtues of humility and temperance. Cet article examine les palais épiscopaux de l’Empire d’Orient en s’intéressant au contraste entre la splendeur de la classe dirigeante et l’idéal de la vie ascétique. Pour évaluer le style palatial de la résidence épiscopale, il faut considérer que de nombreux membres de la classe curiale et sénatoriale furent investis de la dignité d’évêque dès le temps de Constantin Ier. Les raisons en sont diverses. En rejoignant le clergé, les curiales évitaient leurs charges fiscales. Dans les grandes villes de l’Empire, la charge épiscopale offrait un grand prestige social, des richesses, des contacts avec l’empereur et la classe dirigeante, ainsi qu’une vie d’aisance à l’image de celles des aristocrates. Dans les petites villes des provinces, au contraire, les textes attestent l’existence de plusieurs évêques qui menaient une vie ascétique. Certains évêques résidaient dans des monastères, tandis que d’autres exerçaient un véritable pouvoir civil dans les villes, avec des activités liées à la vie séculière telles que la construction, la protection des pauvres, des prisonniers et des captifs, etc. La dichotomie entre ces deux modèles imprègne les textes qui décrivent l’aisance aristocratique dans laquelle vivaient certains évêques, les vertus ascétiques d’autres évêques, ou encore la combinaison chez quelques-uns de fonctions séculières et des vertus chrétiennes d’humilité et de tempérance. In diesem Aufsatz werden die Bischofspaläste im Oströmischen Reich in den Blick genommen, um sie als Ausdruck von Luxus der Eliten im Gegensatz zu den monastischen Idealen zu betrachten. Man muss dabei beachten, dass die Bischöfe seit Konstantin dem Großen in der Regel Mitglieder der curiae oder des Senats waren. Dies hat unterschiedliche Gründe: Der Titel eines Bischofs war – zumindest in den großen Städten – mit Steuerprivilegien, hohem sozialen Ansehen, Reichtum und Kontakt mit den obersten Schichten bzw. mit dem Kaiser selbst verbunden und bot einen hohen, entsprechend angemessen Lebensstandard. Im Gegensatz dazu sollen die Bischöfe kleinerer Städte nach den Quellenangaben ein eher asketisches Leben geführt haben, manche sogar in Klöstern. Die Bischöfe in den größeren Zentren hatten lokale politische Macht, stifteten Gebäude, unterstützten die Versorgung von Armen, Gefangenen und Sklaven. Diese gegensätzlichen Lebensmodelle werden auch in den Quellen beschrieben, dass also manche Bischöfe im großen Reichtum lebten und andere den asketischen Weg bevorzugten. Die Wohltätigkeitsaktivitäten mancher Bischöfe befolgten zugleich die christlichen Tugenden von Humanität und Bescheidenheit.

167

Helen Saradi Il presente contributo si propone di esaminare i palazzi episcopali nel contesto del contrasto tra lo splendore della classe dirigente e l’ideale della vita ascetica nell’Impero d’Oriente. Per valutare il carattere palaziale delle residenze episcopali, si deve considerare il fatto che a partire dall’età di Costantino I numerosi individui di rango curiale e senatoriale furono investiti della dignità vescovile. Le ragioni sono molteplici. Accedendo alle file del clero, i curiali si sottraevano ai loro obblighi fiscali. Nelle grandi città dell’Impero l’episcopato garantiva un grande prestigio sociale, ricchezza, vicinanza all’imperatore e alla classe dirigente, così come una vita di agi aristocratici. Al contrario, nelle piccole città provinciali le fonti attestano la presenza di numerosi vescovi che seguivano uno stile di vita ascetico. Alcuni vescovi risiedevano nei monasteri, mentre altri esercitavano un potere propriamente civile nelle città, assolvendo a funzioni secolari nell’ambito dell’edilizia, della protezione dei poveri, dei carcerati, dei prigionieri, ecc. La dicotomia tra questi due modelli impregna i testi che descrivono gli agi aristocratici che circondavano alcuni vescovi, le virtù ascetiche di altri o, infine, la combinazione di alcune funzioni secolari con le virtù cristiane dell’umiltà e della temperanza. Some bishops are presented in the texts as being able to combine ascetic values with worldly authority. In the fifth century, the famous bishop of Cyrrhus, Theodoretus († c. 460), a prolific author of many theological and historical works, is the best example. He was born into a wealthy family in Antioch, received a classical education, and abandoned the monastic life to accept the bishopric of Cyrrhus in Syria, near the Euphrates river (423). Theodoretus describes his spiritual virtues and his worldly achievements for the benefit of his flock. Neither he nor any member of his clergy had ever been accused of any incorrect act. He never received a penny or a garment, except for the rags he wore. Those who lived in his house never received anything, not even a loaf of bread or an egg. From the revenues of the Church he built public porticoes, two bridges, public baths and an aqueduct to provide water to the city, and he converted several heretical villages to the Orthodox faith.5 Porphyry too, as Bishop of Gaza, continued to practice asceticism. The Life of a lesser known saint, St Theognis (d. c. 522), illustrates the activities of an ascetic bishop. He abandoned his ascetic life in the desert to become bishop of the small town of Betylion in Palestina Prima at the invitation of the Patriarch of Jerusalem.6 Although he spent his time as a bishop between his bishopric and the desert, he also dealt with secular matters. The account of his visit to Emperor Anastasius, at the demand of the city’s landowners, focuses on the simplicity of his behaviour, appropriate to a modest monk. When he offered the emperor three eulogiai (gifts of blessed bread) in a handkerchief, the emperor, amused by the bishop’s simplicity, asked him if he was offering the handkerchief too, laughed and kissed him, and solved the problem for the sake of which Theognis had visited him.7 In order to solve another Church issue, he met with the duke of his city, who had a reputation for being intimidating towards holy men. He dined with him in the antique Greco-Roman style, as they reclined on either side of the stibas.8 In the early seventh century,

In explaining the magnificence of the architecture and decor of episcopal palaces in Late Antiquity, two issues should be taken into consideration: first, the class origin of the bishops, and second, the social, political and economic power of the office. When in the fourth century bishops emerged as new religious and social authorities, there was a constant tension between the simplicity and humility required by the Church, and the authority and worldly magnificence with which the bishops’ office was invested. The Christianisation of the Empire took place at different social levels, and the motivations that led to a religious shift were many, both spiritual and material. The urban aristocracy was gradually realigned with the imperial policy established by Constantine the Great, and it was converted to the new religion. It is from this class of local notables that most of the bishops came. They had prestige and dignity, power and wealth, and classical education, which were all qualities of the upper class. As the cities’ bouleutai were losing power, many of their members were attracted to a career in the Church to avoid the economic burdens of their office, and this trend continued until the sixth century. Members of the senatorial class were also attracted to the episcopal office.1 Since one of the arguments in favour of some candidates for the episcopal see was their illustrious descent (ὅτι γένους ἐστὶ λαμπροῦ),2 it is obvious that the bishop’s office was perceived as belonging to the elite. From the fourth to the fifth century, bishops rose to positions of leadership in the cities and assumed a variety of administrative and secular duties.3 Major Christian figures, such as John Chrysostom in his On the Priesthood and Augustine in his On the Duties of the Clergy, were concerned that the secular tasks of the bishops were an obstacle to Christian virtues and ascetic comportment.4 1 See, for example, Gilliard 1984, pp. 154–157; Rapp 2000, pp. 379–399, esp. pp. 385 ff.; Rapp 2005, pp. 183–195. Curiales rushed into the clergy early on: Codex Theodosianus, XVI, ii, 3 (320) and 6 (326); Novellae Justiniani, CXXIII, i and iv (546). 2 John Chrysostom, De sacerdotio 3.9.16–17 (ed. and transl. Malingrey 1980). Cf. also Gryson 1979, pp. 301–345. 3 See Hohlweg 1971, pp. 51–62; Brown 1992, pp. 76–158; Liebeschuetz 2003, pp. 110–112, 137–155 and passim; Rapp 2005, pp. 155–171. 4 Rapp 2005, pp. 47–49 and passim. Cf. also Sterk 2004.

5 Theodoretus of Cyrrhus, Epistulae 81 (ed. and transl. Azéma 1982, p. 1969–24). 6 Acta Sancti Theognii (ed. Van den Gheyn 1891, pp. 88–89). 7 Acta Sancti Theognii (ed. Van den Gheyn 1891, pp. 9011–918). On bishops and ascetic life, cf. Sterk 2004; Rapp 2005, pp. 149–152 (focusing on the western clergy), p. 297. 8 Acta Sancti Theognii (ed. Van den Gheyn 1891, p. 10211–13).

168

The Episcopal Palaces another major figure, St John the Almsgiver, demonstrated abilities similar to those of Theodoretus, Bishop of Cyrrhus. He was the son of the governor of Cyprus. In 610, the patrician Nicetas, Heraclius’ cousin, who helped Heraclius to overthrow Phocas, attacked and conquered Egypt and became governor. St John was chosen by the emperor to become the patriarch of Alexandria, at Nicetas’ instigation. His Life, written by Leontios of Neapolis, presents him as applying monastic austerity in his daily life, while at the same time being actively involved in Alexandria’s urban life; he built seven hospitals, provided food for the poor and immigrants, conducted all sorts of business transactions for his wealthy Church that owned trading ships, workshops, and estates, imposed the same standards for measures of weight across the entire city, and judged legal cases in front of the Church. His Life presents him as having close contacts with the patrician Nicetas.9

lived in moderation as pure and reverent men.15 John Chrysostom, in his third Homily In Acta Apostolorum, describes a similar social reality: ‘Prefects and local magistrates do not enjoy such honour as the magistrates of the Church. For if a bishop enters the palace, who ranks the highest, or among the matrons, or among the houses of the great? No one is honoured before him.’16 John Chrysostom, in his On the Priesthood, which establishes the ideal qualities of priests and bishops, explains that the office was so attractive that it led to payment of fees for ordination and consecration.17 The practice was forbidden by several Church Councils and Novel 123 of Justinian, which also regulated bishops’ fees upon enthronement. The bishops’ desire for luxury is variously attested in the texts and condemned in accounts of corruption.18 Some bishops appropriated church funds and properties for personal use, a practice condemned by ecclesiastical canons. Thus, in 451, the office of oikonomos was established by the Oecumenical Council of Chalcedon (canon 26) to administer church finances, so that the administration of the Church would be controlled, and the dignity of the clerics would be protected.19 Sophronius, Bishop of Pompeiopolis, was enriched by offerings to the Church.20 Bishop Antoninus of Ephesus was charged with various offences, most of them financial. He melted down sacred vessels for coins. He removed marble slabs from the entrance of the baptistery and used them to adorn his own baths. He removed church columns that had been lying in situ for many years and set them up in his own triclinium. He sold villages donated to the Church by the mother of Emperor Julian and kept the profit for himself. He retained as a law and a fixed belief the sale of the ordination of bishops for profit.21 But those striving for the bishop’s office had other motivations: they declared that they paid consecration fees to Antoninus in order to escape their curial obligations. John Chrysostom, who investigated the case, gave them back their money and promised to persuade the emperor to free them from their curial charges.

As prominent figures in the cities, bishops established contacts with the Empire’s upper class. From the reign of Constantine the Great onward, they had close contacts with the emperor and could even dine with him.10 They exercised influence on emperors,11 and received gifts, privileges and honours. They played influential roles in the cities in periods of crisis, such as enemy invasions and famines, undertook construction works, were active in almsgiving, established charitable institutions, acted as judges in judicial matters (audientia episcopalis), and from the fifth century, participated in the election of civic magistrates. Their pre-eminence in the cities brought them into close contact with the urban aristocracy. Theodoretus of Cyrrhus corresponded with numerous high officials. His letters reveal the spiritual and secular issues with which he dealt.12 Bishops invited members of the upper class to their episcopal residences to establish and strengthen contacts with them and to receive donations for the Church. Pulcheria and her sisters used to dine at the Patriarchal Palace on Sundays following communion.13 In contrast to the ascetic bishops, many others desired the ambience of the upper-class lifestyle. Ammianus Marcellinus relates the contest between Damasus and Ursinus for the bishopric of Rome in 366. Their desire to seize the bishopric caused bloody conflicts between their supporters. Ammianus Marcellinus describes a bishop’s luxurious life in the city: ‘for when they attain it, they will be so free from care that they are enriched from the offerings of matrons, ride seated in carriages, wearing clothing chosen with care, and serve banquets so lavish that their entertainments outdo the tables of kings’.14 Marcellinus contrasts these bishops with provincial bishops, who

Most bishops desired a luxurious residence to match the prestige of their office.22 Thus, building a small episkopeion was praised as an example of modesty.23 Akakios, Bishop of Beroea, while passing through Constantinople, got very angry at the patriarch, John Chrysostom, for not offering him a nice house (καταγωγίου καλοῦ), and felt humiliated by him.24 John Chrysostom was renowned for Ammianus Marcellinus, 27.3.15 (ed. and transl. Marié 1984). John Chrysostom, In Acta Apostolorum 3.5 (PG 60, col. 41). 17 John Chrysostom, De sacerdotio 3.11 (ed. and transl. Malingrey 1980). 18 Rapp 2005, pp. 217–219. 19 ACO 2.1.2, p. 163. 20 Sozomenus, 4.24.14 (ed. Bidez, Hansen 1960). 21 Palladius, Dialogus de vita Sancti Joannis Chrysostomi, 13.161–174 (ed. and transl. Malingrey 1988). The removal of marble slabs from churches to decorate episcopal triclinia is attested through to the end of the Byzantine Empire. Cf. Magdalino 1978, p. 31132–34. 22 Cf. Marano 2007. 23 Mark the Deacon, Vita Porphyrii 18.12 (ed. and transl. Grégoire, Kugener 1930): τὸ παρ’ αὐτοῦ κτισθὲν μικρότατον ἐπισκοπεῖον. 24 Palladius, Dialogus de vita Sancti Joannis Chrysostomi 6.8–11 (ed. and transl. Malingrey 1988). 15 16

Leontios of Neapolis, Vita Sancti Ioannis Eleemosynarii (ed. Festugière 1974, pp. 257–637). 10 E.g. Eusebius, Vita Constantini 3.15 (ed. Winkelmann 1975). Cf. also Drake 2000. 11 On holy men’s access to emperors, cf. Rapp 2005, pp. 260–273. 12 Theodoretus of Cyrrhus, Epistulae 46–56 (ed. and transl. Azéma 1982). 13 Holum 1982, pp. 144 and 153. 14 Ammianus Marcellinus, 27.3.14 (ed. and transl. Marié 1984). 9

169

Helen Saradi palaces. Following the architecture of the Roman domus, some rooms were in use during different seasons of the year. For example, a winter sekreton of the Patriarchate of Constantinople is mentioned in the acta of the Council of 536.33

his fervent denouncement of luxury and vices. When, in 398, he became Patriarch of Constantinople, he introduced changes in his Church and its bishopric. He looked into the books of the Church’s financial office and asked to put an end to expenses from which the Church did not benefit. He found that the expenses of the episkopeion were high, and ordered that the expenses for luxury be transferred to the hospital.25 John Chrysostom wanted to set an example for other bishops by preaching, like Paul, that a cleric ought to live at his own expense (ἐν ἰδίῳ μισθώματι, Acts 28.30). He ended the frequent banquets and never used resources from the Church. He received only his daily meal and appeared ashamed to get even this.26 He attacked the sin of greed (πλεονεξία), and the lifestyle of the Church leaders, asking them ‘not to look for the odour of savoury meat of the rich’, but to be satisfied with their own food.27 It is important to stress, however, that during Chrysostom’s bishopric the Church of Constantinople increased its wealth in an unprecedented way. The most impressive donation recorded was that of Olympias.28

The location of the episcopal residence in the cities was central, and proclaimed the bishop’s authority. In accordance with Roman villa architecture, some episcopal palaces opened onto spectacular views of the surrounding nature, especially the sea. Today one can experience the splendid view from the apsidal central hall of the two-story Episcopal palace built in the sixth century in Poreč, next to the Euphrasius Basilica. It is still standing because it was continuously used during the Middle Ages; it is situated above the defensive sea wall of the city.34 At Porphyrion in Israel, the bishop’s palace had a view of gardens, orchards and the sea.35 The House of Ekdikos in Nicopolis in Greece, which may have been the episcopal palace at some point,36 had a splendid view over the Ambracian gulf.

The story of how Cyril of Alexandria succeeded in regaining his see after he had been removed from it and exiled by Theodosius II provides one further insight into the habits of the time. Cyril offered treasures he had taken from the episcopal treasury of Alexandria: more than half a tonne of gold, and numerous precious items, such as carpets, curtains, wall hangings, cushions, tablecloths, stools and benches.29

An episcopal palace on the city’s acropolis proclaimed the bishop’s leadership. From the late fifth century, some episcopal palaces were built on the acropolis of cities/ kastra, when historical circumstances weakened the antique cities. Justiniana Prima (Caričin Grad) is a wellknown example.37 The case of Louloudies of Kitros in Pieria, north-east of Mount Olympus in Greece, is less well known. It was built by the bishop and residents of the ancient city of Pydna, who moved there in 479, when Pydna was handed over to the Goths by Emperor Zeno. The upper part of it, in the form of a quadriburgium (80 x 90 m, 0.72 h), was dominated by a large basilica and the episkopeion, and included residential quarters for workers and workshops. The bishop’s palace had an apsidal triclinium and several rooms. It was lavishly adorned with mosaic pavements, wall paintings, marble revetment, sculpted plaster ornaments, and probably a coffered ceiling. The apse of the triclinium was decorated with a mosaic pavement depicting a kantharos from which vine branches grew.38

Texts offer scattered information about the episkopeia30 and the bishops’ daily activities there. Archaeological excavations reveal, though in a fragmentary and often uncertain picture,31 the architecture and decor of the episkopeia. The episcopal residence, with various adjacent buildings, was the administrative centre for all the needs of the Church. In a letter to the governor of Cappadocia, Basil of Caesarea refers to the annexes of the episcopal residence (οἴκησιν) and their uses in the following ways: residences for the clergy, which could also be used by the governor’s secular administrators, a hospital for foreigners and the sick, and workshops.32 The main building of the episcopal residence included an apsidal audience hall, rooms for archives, a space for secretarial work and a library, a cubiculum, rooms for the servants, a kitchen and storage rooms. The prestige and power of the episcopal office required residences to be built in the style of palatial architecture. Apsidal audience rooms in episcopal residences copied the most spectacular features of secular

The size and decoration of the triclinium of bishops’ palaces varied. That of the patriarchal palace in Constantinople was very large; 40 bishops, a number with symbolic significance, could be accommodated.39 The large hall (10 x 20 m) of the patriarchate accommodated 168 bishops for the Council of Constantinople in 553, while in the seventh century the patriarch Thomas (607–610) constructed a very large hall, called the Makron.40 Respect and dignity deriving from the episcopal office corresponded to the organisation of the space of the bishop’s residence.

Palladius, Dialogus de vita Sancti Joannis Chrysostomi 5.128–133 (ed. and transl. Malingrey 1988). 26 Palladius, Dialogus de vita Sancti Joannis Chrysostomi 12.7–29 and 17.206–212 (ed. and transl. Malingrey 1988). 27 Palladius, Dialogus de vita Sancti Joannis Chrysostomi 5.121–123 (ed. and transl. Malingrey 1988): ἀρκεῖσθαι τοῖς ἰδίοις ὀψωνίοις καὶ μὴ διώκειν τὰς κνίσας τῶν πλουσίων. 28 Dagron 1974, pp. 498–506. 29 ACO 1.4, p. 22328ff. On bribery in Late Antiquity, cf. MacMullen 1988, pp. 148–167. 30 Pallas 1971. 31 Cf. Caillet 2012. 32 Basil of Caesarea, Epistulae 94 (ed. Courtonne 1957, pp. 205–206). 25

ACO 3, p. 10010. On this palace, cf. Matejčić 1995. 35 Littlewood et al. 2002, p. 216. 36 Pavlidis 2015, pp. 21, 43. 37 Caillet 2012, pp. 149–153. 38 Poulter 1998; Marki 2001, pp. 79–89. 39 Palladios, Dialogus de vita Sancti Joannis Chrysostomi 8.92 (ed. and transl. Malingrey 1988). 40 Dark, Kostenec 2014, pp. 34–35. 33 34

170

The Episcopal Palaces In the First Oration for Bishop Marcian of Gaza, Choricius of Gaza referred to the hall of salutation located near the entrance to the Church of St Sergius. Outside it, people offered the obligatory address (εἰς πρόσρησιν) to the bishop, an obligation that was not just a tradition, but was due to the excellence of his manners.41

and spinning instruments.48 The later Novel 123.29 of Justinian forbade bishops to live in the same house with women, lest they be deposed. Although from the sixth century onwards Byzantine legislation increased the scale of the secular affairs in which bishops were involved, in Byzantium bishops never received as much secular power as they did in the West.49 With the shrinking of the cities and of the economy that began in the last quarter of the sixth century, the bishop as a powerful urban figure in the Late Antique style ended. Episcopal palaces lost the splendour of their Late Antique predecessors, although some continued to be used for centuries.50 Their size was reduced, their architectural appearance more modest, their decoration simpler. Nevertheless, they continued to be constructed with a display of marble and frescoes and included baths. In a letter addressed to the Metropolitan of Thessaloniki, the Metropolitan of Naupaktos John Apokaukos (AD 1199/1200–1232) compares the two episcopal residences. They were both built of white marble. The episkopeion of Naupaktos was like a small palace (συνεπτυγμένον παλάτιον), while that of Thessaloniki was larger, with peristyles. They both had baths built entirely of marble and adorned with paintings.51 The episcopal palace of Thessaloniki was located near the metropolitan church of Ayia Sophia.52 It had a revered tradition, because it was believed that it was the house where St Demetrius had lived.53

The episkopeion was a busy place. Sources allow us to get glimpses of the bishops’ daily activities there. The most comprehensive picture is found in the Life of St John the Almsgiver. He held meetings with the clergy in the sekreton or in his sacristy (τῷ κειμηλιαρχείῳ) to settle civic matters, read edifying stories of holy fathers, dealt with theological issues, and instructed his clergy in decent behavior.42 He placed the gold entrusted to him by an individual under the holy table in the oratory of his own bed-chamber for security (ὑποκάτω τῆς ἁγίας τραπέζης … τῆς οὔσης ἐν τῷ εὐκτηρίῳ τοῦ κουβουκλίου αὐτοῦ).43 He kept the chest containing the money of the Church under his own bed to keep it safe. When the Duke of Alexandria, Nicetas, visited him and asked for the money of the Church because the state was in need, he told him to take the money from the chest under his lowly bed (ἀλλ’ ἰδού, ὑποκάτω τοῦ ταπεινοῦ χαλαδρίου μού ἐστιν ἡ ἀποθήκη τοῦ Χριστοῦ). Nicetas ordered his servants and soldiers to take the money and leave only 100 pounds. As they descended the stairs, some men were coming up carrying small vases sent from Africa to the patriarch. They bore the inscriptions: ‘first quality honey’, ‘second quality honey’ and ‘unsmoked honey’. The inscriptions were a ruse, because the vases were full of coins, and they thus escaped Nicetas’ attention (καὶ ὡς ἔτι οἱ βαστάζοντες κεράμια μικρὰ πεμφθέντα ἀπὸ Ἀφρικῆς τῷ πάπᾳ, ἐπιγράφοντα τὰ μὲν ‘μέλι πρωτεῖον’, τὰ δὲ ‘μέλι δευτέριον’, ἄλλα δὲ ‘μέλι ἀκάπνιστον’).44

In the texts, the references to bishops’ palaces interweave ideas of ascetic modesty and theological spirituality with those of power and secular concerns. All are reflected in the palaces’ architectural remains, as the tension between the two traditions was never resolved.

From the middle of the fourth century, quarters for the clergy were attached to some episkopeia, so that they formed monasteries.45 In other cases, the bishop’s residence was in a monastery, as with the monastery of St Porphyry in Scythopolis.46 But the antithesis between Christian virtue and unethical habits pervades most of the accounts of bishops and their residences. In contrast to the holy ambience of bishops like St John the Almsgiver, in other holy residences the practice of syneisaktoi was deplorable. It was the first bad habit of holy men that John Chrysostom attempted to correct.47 Some were bringing into their house maidens to take care of them in their old age. Even if the men preserved the young women’s virginity, the appearance created by their residence was unacceptable: one could see women’s shoes, belts, headbands, baskets

Bibliography Abbreviations: ACO = Schwartz E., Straub J., Schieffer R., Riedinger R., 1914–1984, Acta conciliorum oecumenicorum [1st series], Strasbourg/Berlin/Leipzig. JÖBG = Jahrbuch der Österreichischen byzantinischen Gesellschaft. PG = Migne J.-P. (ed.), 1857–66. Patrologiae cursus completus. Series Graecae, Paris. Azéma Y. (ed. and transl.), 1982. Théodoret de Cyr, Correspondance, I, 2nd ed., Paris (Sources chrétiennes, 40).

Choricius of Gaza, Oratio, 1.22 (ed. Foerster, Richtsteig 1929, p. 813–16). Leontios of Neapolis, Vita Sancti Ioannis Eleemosynarii 1.16 and 50 (ed. Festugière 1974, p. 347, 364 and 400). 43 Leontios of Neapolis, Vita Sancti Ioannis Eleemosynarii 25 (ed. Festugière 1974, p. 376). 44 Leontios of Neapolis, Vita Sancti Ioannis Eleemosynarii 10 (ed. Festugière 1974, p. 356). 45 Sozomenus, 6.31.6–11 (ed. Bidez , Hansen 1960). 46 Schwartz 1939, p. 180.5. 47 Palladios, Dialogus de vita Sancti Joannis Chrysostomi 5.103–106 (ed. and transl. Malingrey 1988). Against this practice, cf. Dumortier 1955. 41 42

John Chrysostom, Homilae in Genesim 58 (ed. PG 47, col. 508). Liebeschuetz 2003, pp. 151, 154–155. 50 Niewӧhner 2015, p. 32. 51 John Apokaukos, Epistulae 67 (ed. Bees 1971–74, pp. 122–123). 52 Stavridou–Zafraka 2001, pp. 557–559. 53 Eustathius of Thessaloniki, De Thessalonica capta 106.20 ff. (ed. and transl. Kyriakidis, Rotolo 1961). 48 49

171

Helen Saradi Holum K. G., 1982. Theodosian Empresses. Women and Imperial Dominion in Late Antiquity, Berkeley (The Transformation of the Classical Heritage, 3).

Bees N. (ed.), 1971–74. ‘Unedierte Schriftstücke aus der Kanzlei des Johannes Apokaukos des Metropoliten von Naupaktos (in Aetolien)’, in Byzantinisch-neugrichische Jahrbücher, 21, pp. 55–160.

Kyriakidis S. (ed.), Rotolo V. (transl.), 1961. Eustazio di Tessalonica, La espugnazione di Tessalonica, Palermo (Istituto siciliano di studi bizantini e neoellenici. Testi, 5).

Bidez J., Hansen G. C. (ed.), 1960. Sozomenus, Kirchengeschichte, Berlin (Die Griechischen Christlichen Schriftsteller der ersten Jahrhunderte, 50). Brown P., 1992. Power and Persuasion in Late Antiquity. Towards a Christian Empire, Madison (The Curti Lectures).

Liebeschuetz J. H. W. G., 2003. The Decline and Fall of the Roman City, Oxford. Littlewood A. R., Maguire H., Wolschke-Bulmahn J. (eds.), 2002. Byzantine Garden Culture, Washington.

Caillet J.-P., 2012. ‘Le cas de Caričin Grad (Serbie) et le problème de l’identification de certains “palais épiscopaux” de l’Antiquité tardive’, in S. BalconBerry, F. Baratte, J.-P. Caillet, D. Sandron (eds.), 2012. Des domus ecclesiae aux palais épiscopaux. Actes du colloque tenu à Autun du 26 au 28 novembre 2009, Turnhout (Bibliothèque de l’Antiquité tardive, 23), pp. 147–162.

MacMullen R., 1988. Corruption and the Decline of Rome, New Haven/London. Magdalino P., 1978. ‘Byzantine churches of Selymbria’, in Dumbarton Oaks Papers, 32, 309–318. Malingrey A.-M. (ed. and transl.), 1980. Jean Chrysostome, Sur le Sacerdoce (Dialogue et Homélie), Paris (Sources chrétiennes 272).

Courtonne Y. (ed. and transl.), 1957. Saint Basile, Correspondance, I, Paris (Collection des universités de France. Série grecque, 132).

Malingrey A.-M. (ed. and transl.), 1988. Palladios, Dialogue sur la vie de Jean Chrysostome, I, Paris (Sources chrétiennes 341).

Dagron G., 1974. Naissance d’une capitale. Constantinople et ses institutions de 350 à 451, Paris (Bibliothèque byzantine. Études, 7).

Marano Y., 2007. ‘Domus in qua manebat episcopus. Episcopal residences in northern Italy during Late Antiquity (4th to 6th Centuries A.D.)’, in L. Lavan, L. Ozgenel, A. Sarantis (eds.), 2007. Housing in Late Antiquity. From Palaces to Shops, Leiden/Boston (Late Antique Archaeology, 3/2), pp. 97–129.

Dark K., Kostenec J., 2014. ‘The Patriarchal Palace at Constantinople in the seventh century: Locating the Thomaites and the Makron’, in JÖBG, 64, pp. 33–40. Drake H. A., 2000. Constantine and the Bishops. The Politics of Intolerance, Baltimore (Ancient Society and History).

Marié M.-A. (ed. and transl.), 1984. Ammien Marcellin, Histoires, V, Paris (Collection des universités de France. Série latine, 271).

Dumortier J. (ed. and transl.), 1955. Saint Jean Chrysostome, Les cohabitations suspectes – Comment observer la virginité, Paris (Nouvelle collection de textes et documents).

Marki E., 2001. Κίτρος μια πόλη κάστρο της βυζαντινής περιφέρειας. Aρχαιολογική και ιστορική προσέγγιση, Τhessaloniki.

Festugière A. J. (ed. and transl.), 1974. Léontios de Néapolis, Vie de Syméon le Fou et Vie de Jean de Chypre, Paris (Bibliothèque archéologique et historique, 95).

Matejčić I., 1995. ‘The Episcopal Palace at Poreč. Results of a recent exploration and restoration’, in Hortus Artium Medievalium, 1, pp. 84–89.

Foerster R., Richtsteig E. (ed.), 1929. Choricii Gazaei Opera, Leipzig (Bibliotheca scriptorum graecorum et romanorum Teubneriana).

Niewӧhner Ph., 2015. ‘The Late Antique origins of Byzantine palace architecture’, in M. Featherstone, J.-M. Spieser, G. Tanman, U. Wulf-Rheidt (eds.), The Emperor’s House. Palaces from Augustus to the Age of Absolutism, Berlin/Boston (Urban Space, 4), pp. 31–52.

Gilliard F. D., 1984. ‘Senatorial bishops in the fourth century’ in, Harvard Theological Review, 77, pp. 153– 157.

Pallas D. I., 1971. ‘Episkopion’, in K. Wessel (ed.), Reallexikon zur byzantinischen Kunst, II, Stuttgart, cols. 335–371.

Grégoire H., Kugener M.-A. (ed. and transl.), 1930. Marc le Diacre, Vie de Porphyre évêque de Gaza, Paris (Collection byzantine).

Pavlidis E. A., 2015. Νικόπολη. Ο οίκος του εκδίκου Γεωργίου – Nicopolis. The Domus of the Ekdikos Georgios, Athens (Μνημεία της Νικόπολης – Monuments of Nicopolis, 5).

Gryson R., 1979. ‘Les élections épiscopales en Orient au IVe siècle’, in Revue d’histoire ecclésiastique, 74, pp. 301–345.

Poulter, A. G., 1998. ‘Field survey at Louloudies: A new Late Roman fortification in Pieria’, in Annual of the British School at Athens, 93, 463–511.

Hohlweg A., 1971. ‘Bischof und Stadtherr im frühen Byzanz’, in JÖBG, 20, pp. 51–62.

172

The Episcopal Palaces Rapp C., 2000. ‘The elite status of bishops in Late Antiquity in ecclesiastical, spiritual, and social contexts’, in Arethusa, 33/3, pp. 379–399. Rapp C., 2005. Holy Bishops in Late Antiquity. The Nature of Christian Leadership in a Time of Transition, Berkeley/Los Angeles/London (The Transformation of the Classical Heritage, 37; The Joan Palevsky Imprint in Classical Literature). Schwartz E., 1939. Kyrillos von Skythopolis, Leipzig (Texte und Untersuchungen zur Geschichte der altchristlichen Literatur, 49/2). Stavridou-Zafraka A., 2001. ‘Η Αγία Σοφία ως μητροπολιτικός ναός και το επισκοπείο’, in Αφιέρωμα στη μνήμη του Σωτήρη Κίσσα, Thessaloniki, pp. 549– 560. Sterk A., 2004. Renouncing the World, Yet Leading the Church. The Monk-Bishop in Late Antiquity, Cambridge (MA)/London. Van den Gheyn J. (ed.), 1891. ‘Acta Sancti Theognii episcopi Beteliae Paulo Elusensi et Cyrillo Scythopolitano auctoribus […]’ in, Analecta Bollandiana, 10, pp. 73–118. Winkelmann F. (ed.), 1975. Eusebius Werke, I/1, Berlin (Die Griechischen christlichen Schriftsteller der ersten Jahrhunderte, 55).

173

13 The Identification of the Bishop’s Palace at Miletus in Caria (Turkey) Philipp Niewöhner* Georg August-Universität Göttingen (Germany) The Bishop’s Palace in the centre of Miletus in Caria, on the west coast of Asia Minor, dates from the first half of the fifth century AD. It replaced a Late Roman peristyle house from the third century AD that underwent a renovation in the fourth century. The fifth-century building was closely associated with the oratory of St Michael and can thus be identified as the episcopal palace. The oratory flanked the residence to the south, served as a palace chapel, and was rebuilt in the seventh century. The original oratory appears to date from the fifth century, when the palace was built; it had a single nave that consisted of a former temple cella and was too small for a parish church; it can only have served as a palace chapel. In the early seventh century, the temple cella was replaced by a three-aisled basilica with galleries; an inscription identifies the edifice as an oratory (rather than as a parish church) and relates the rebuilding to the patriarch of Constantinople and the local clergy (rather than to a worldly donor). The oratory, a vestibule in the form of an elongated apsidal hall, a relatively large main hall, and the absence of a central peristyle court distinguish the palace from the peristyle houses of Late Antiquity and link it to the Byzantine Palace at Ephesus. The new features may reflect the establishment of a new elite of clerical office-bearers that replaced the leading families of old. Le palais épiscopal situé au centre de Milet, en Carie, sur la côte ouest de l’Asie Mineure, date de la première moitié du Ve siècle ap. J.-C. Il a remplacé une maison à péristyle romaine tardive du IIIe siècle ap. J. C., qui avait subi une rénovation au IVe siècle. Le bâtiment du Ve siècle était étroitement associé à l’oratoire de Saint-Michel et peut donc être identifié comme le palais épiscopal. L’oratoire est flanqué de la résidence au sud, a servi de chapelle de palais et a été reconstruit au VIIe siècle. L’oratoire original semble dater du Ve siècle, quand le palais a été construit. Il n’avait qu’une seule nef, consistant en une ancienne cella du temple, trop petite pour une église paroissiale ; il ne peut avoir servi que de chapelle de palais. Au début du VIIe siècle, la cella du temple fut remplacée par une basilique à trois nefs avec des galeries et une inscription qui identifie l’édifice comme un oratoire (plutôt que comme une église paroissiale), associant la reconstruction au patriarche de Constantinople et au clergé local (plutôt qu’à un mécène laïc). L’oratoire, un vestibule en forme de salle absidale allongée, une salle principale relativement grande et l’absence d’une cour centrale en péristyle distinguent le palais des maisons à péristyle de l’Antiquité tardive, et le relient au palais byzantin d’Éphèse. Les nouvelles caractéristiques pourraient refléter la mise en place d’une nouvelle élite de fonctionaires épiscopaux qui aurait remplacé les anciennes grandes familles. Der Bischofspalast im Stadtzentrum von Milet wurde im frühen 20. Jh. durch Theodor Wiegand entdeckt und in den 1970er Jahren durch Wolfgang Müller-Wiener ausgegraben. Das Gebäude wurde in der ersten Hälfte des 5. Jhs. n. Chr. anstelle eines spätrömischen Peristylhauses des 3. Jhs. n. Chr. errichtet, das im 4. Jh. n. Chr. einen Umbau erfuhr. Über dem zerstörten Palast war noch eine Töpferei aus der Zeit des türkischen Emirats von Menteşe (14./15. Jh.) dokumentiert worden sowie eine noch jüngere osmanische Bebauung. Der Bischofspalast war eng mit der ihm südlich benachbarten Michaelskirche verbunden, die offenbar als bischöfliches Oratorium und Palastkapelle diente. Die Kapelle, ein Vestibül in Form eines langgestreckten Apsidensaals, ein verhältnismäßig großer Hauptsaal sowie der Verzicht auf einen zentralen Peristylhof unterscheiden den Bischofspalast von spätantiken Peristylhäusern und verbinden * I would like to thank Isabella Baldini for inviting me to participate in her roundtable. The excavation and study of the Bishop’s Palace at Miletus was made possible by, and was undertaken within the

institutional framework of, the German Archaeological Institute and many collaborators, for which see Niewöhner 2015. Thanks are also due to the editors of this publication, and to the anonymous referees.

175

Philipp Niewöhner ihn mit dem sog. Byzantinischen Palast von Ephesos. In beiden Fällen geht dies wohl auf die geänderten Repräsentationsbedürfnisse frühbyzantinischer Amtsträger zurück. Il Palazzo Vescovile nel centro di Mileto in Caria, sulla costa occidentale dell’Asia Minore, risale alla prima metà del V secolo d.C. Sostituì una casa in stile tardo romano del III secolo d.C. che aveva subito una ristrutturazione nel IV secolo. L’edificio del V secolo era strettamente associato all’oratorio di San Michele e può quindi essere identificato come il palazzo episcopale. L’oratorio fiancheggiava la residenza a sud, fungeva da cappella del palazzo e fu ricostruito nel VII secolo. L’oratorio originario sembra risalire al V secolo, quando fu costruito il palazzo; aveva un’unica navata che consisteva in una ex cella del tempio ed era troppo piccola per una chiesa parrocchiale; può solo essere servito come cappella del palazzo. All’inizio del VII secolo la cella del tempio fu sostituita da una basilica a tre navate con gallerie e un’iscrizione che identifica l’edificio come un oratorio (piuttosto che come una chiesa parrocchiale) e collega la ricostruzione al patriarca di Costantinopoli e al clero locale (piuttosto che a un donatore laico). L’oratorio, un vestibolo a forma di sala allungata absidata, una sala principale relativamente grande, e l’assenza di una corte centrale peristilio distinguono il palazzo dalle case a peristilio della tarda antichità e lo collegano al palazzo bizantino di Efeso. Le nuove caratteristiche potrebbero riflettere la creazione di una nuova élite di impiegati clericali che sostituiscono le famiglie principali del passato. History of research

(fig. 4).9 This single-nave church looked more like a chapel than a cathedral, and it was only in the seventh century that it was replaced by the basilica. This enlargement seemed too late for its suggested interpretation as cathedral; Early Byzantine Miletus should have had a more substantial main church, certainly by the sixth century, when – as was discovered in the 2000s – a huge transept basilica was built in the necropolis, serving as memorial and cemetery church and attesting to the size and prosperity of the city’s Christian community.10 In fact, the main church of Miletus may have been located in the southern part of the city, where circumstantial evidence that has accumulated over the last 50 years – numerous large Christian column capitals, geophysical evidence, burials in the vicinity, etc. – points to two insulae as the location of a large Early Byzantine basilica.11

The Bishop’s Palace at Miletus, an ancient harbour city in Caria,1 on the west coast of Asia Minor, used to be one of those residences whose episcopal associations were disputable.2 The building was first discovered and partly excavated by Theodor Wiegand in the early twentieth century.3 It occupies the northern half of an insula in the centre of Miletus (fig. 1–3). The southern half of the same insula is taken up by the Church of St Michael that replaced an antae temple of Dionysus.4 The close proximity of palatial mansion and church inspired the interpretation of the buildings as episcopal residence and cathedral. The small size of St Michael’s relative to the main churches of other large cities in the region, e.g. the temple church at the provincial capital Aphrodisias5 or St Mary’s at the regional metropolis of Ephesus,6 was offset by the similarly small size of the only other church in the centre of Miletus, the so-called Great Church,7 which is in fact hardly bigger than St Michael’s. A third church dedicated to St Mary is located on the outskirts of the city, and is even smaller.8

As a result, the argument for an identification of St Michael’s and the neighbouring house with the bishop seemed to dwindle, until renewed excavations in 2013 further clarified the building history and led to a novel understanding of it, not as cathedral with episcopal residence, but as bishop’s palace with chapel.12 In order to make this point, it is necessary first to give a summary outline of the building history as it is understood today, before then turning to the function of the complex once again.

The case for an identification of the buildings as a cathedral with episcopal residence was compromised, however, when Wolfgang Müller-Wiener in the 1960s and 1970s undertook a more detailed investigation of the building history and found that an earlier phase of St Michael’s consisted of no more than the small cella of the former temple of Dionysus with a simple apse added to the east

Building history The Bishop’s Palace incorporates elements of four building phases (fig. 5). The earlier phases have a lower floor level that has been retained in the Northern Hall and the adjacent North Rooms I to IV (fig. 2). Excavations under the higher and later floors of the Main Hall and South Rooms I to V

Kleiner 1968; Niewöhner 2016b. 2 Müller-Wiener 1989, pp. 675–677; Sodini 1989, p. 420; Müller-Wiener 1997, c. 371; Ceylan 2007, pp. 176–178; Caillet 2010, p. 513; Real 2012, pp. 126–128; Ceylan 2014, passim. 3 Wiegand 1911, pp. 34 f. 4 Temple of Dionysus: Müller-Wiener 1977–78; 1979; Emme 2016. Church of St Michael: Niewöhner 2016a, pp. 37–58. 5 Cormack 1990; Hebert 2000. 6 Keil et al. 1932; Karwiese 1989. 7 Niewöhner 2016a, pp. 5–35. 8 Feld 1996. 1

Müller-Wiener 1977–78, p. 101; 1979, pp. 162–164, fig. 2. Niewöhner 2016a, pp. 59–103; 2018. 11 Niewöhner 2016a, pp. 103–107. 12 Niewöhner 2015. 9

10

176

The Identification of the Bishop’s Palace at Miletus in Caria (Turkey)

Fig. 1. Miletus, Bishop’s Palace, from the south-east (P. Niewöhner, 2016). In the foreground, the apse (right), the templon stylobate, and the ambo (left) of St Michael’s (seventh century); in the centre, the Early Byzantine peristyle court (Phase III). The reconstructed walls behind enclose the Northern Hall as well as the North Rooms (right) and the Apsidal Hall (left).

Fig. 2. Miletus, Bishop’s Palace, general plan after the seventh-century re-building of St Michael’s as three-aisled basilica with flanking baptistery (W. Müller-Wiener, P. Niewöhner).

177

Philipp Niewöhner

Fig. 3. Miletus, Bishop’s Palace, reconstructed bird-eye’s view from the south-west. In the foreground to the right, the threeaisled basilica of St Michael with flanking baptistery (seventh century); behind it, the arcaded west front of the Apsidal Hall (left) and the clerestory windows of the Main Hall (fifth century) (S. Gräbener).

Fig. 4. Miletus, Bishop’s Palace, phase plan of the original late Roman peristyle house (Phase I, grey) with Late Antique additions (Phase II, black) and Early Byzantine changes (Phase III, white), flanked by the Temple of Dionysus (right) with the apse of the first temple church (hatched) (W. Müller-Wiener, P. Niewöhner).

(fig. 2), and this appropriation of previously public space indicates elite occupancy with power and influence in the council, as would seem to agree with the central location of the house.

have revealed a courtyard with a well and with surrounding porticoes and rooms on the lower level, so that the earliest building phase can be reconstructed as a conventional peristyle house (fig. 4). Original floor mosaics that have survived in the western part of the Northern Hall and in the adjacent North Rooms III (fig. 6) and IV date the earliest Phase I to the Late Roman period.13 The Late Roman North Rooms were built on top of, and blocked, a former street

The remains of a second building phase comprise the eastern part of the Northern Hall and North Rooms I and II (fig. 4–5). Originally, the Northern Hall was only as long as North Rooms III and IV. The original, Late Roman mosaic floor of the Northern Hall turns southward in front of North Room III and apparently used to continue in a second hall on the east side of the Late Roman courtyard that was excavated under the higher, Byzantine floor of

13 Cf. Knackfuß 1924, pp. 64–69, fig. 52–56, pl. 6–9; von Gerkan 1925, pp. 86–99, pl. 7, foldout 5 f.; Naumann 1980; Filges et al. 2002, pp. 105–118; Scheibelreiter-Gail 2011, pp. 210–212, 308–311, 318–320, fig. 57–64, pp. 394–396 and 414–418. Cf. also Kästner 2016 on another late Roman floor mosaic from Miletus.

178

The Identification of the Bishop’s Palace at Miletus in Caria (Turkey)

Fig. 5. Miletus, Bishop’s Palace, phase plan of mosaic, tile and marble floors; red = late Roman (Phase I), blue = Late Antique (Phase II), green = fifth century (Phase III), yellow = seventh century (Phase IV) (P. Niewöhner).

Room III were blocked with the same mixed masonry that includes layers of bricks and appears to stem from the second building phase (fig. 4 and 7). The use of brick layers and new floor mosaics in the new, eastern part of the Northern Hall (fig. 8) and in North Room I – possibly a triclinium? – establish a fourth-century date for Phase II.14 Building Phase III includes all other rooms of the Bishop’s Palace (fig. 9). The Main Hall and South Rooms I to V were newly laid out on a higher level, c. half a metre above the Northern Hall and the North Rooms from Phase I and II. In addition, an Apsidal Hall was added to the west, partly overbuilding and blocking another ancient street, just as in the case of the Late Roman North Rooms (fig. 2). A new peristyle court between the residence and the temple church appears to have been part of the same building phase and would have replaced the Late Roman courtyard under the Main Hall. The peristyle court encroaches on what used to be the temenos or sacred precinct of Dionysus and connects the residence with the temple cella (fig. 4). The former indicates that the pagan sanctuary had been suspended, and the latter suggests that the cella was now integrated into the residence and, with the addition of the eastern apse, served as palace chapel. Stratigraphic evidence and opus sectile as well as mosaic floors (fig. 10–11) date Phase III to the first half of the fifth century.15 As in the Late Roman Phase I, the appropriation

Fig. 6. Miletus, Bishop’s Palace, dolphin in the eastern border of the late Roman floor mosaic in North Room III (Phase I), looking west (W. Müller-Wiener).

the Main Hall. The eastern back wall of this eastern hall was aligned with the dividing wall between North Rooms II and III, connected to the northern wall of the Northern Hall, and one can see the irregular surface of the masonry, where this connection was broken in Phase II (fig. 7). Only after this change did the Northern Hall extend to the east side of the insula, and North Rooms I and II were built on the same occasion. Their masonry includes layers of bricks, which had not yet been employed in the walls of North Rooms I and II, i.e., the first building phase. Two of the three original southern doors of North

14 Bricks: cf. Deichmann 1956, pp. 19–33; Thür 2009. Mosaics: cf. Scheibelreiter-Gail 2011, pp. 176–178. For triclinia, cf. infra, n. 16. 15 Stratigraphic evidence: Niewöhner 2015, pp. 200–202. Opus sectile: cf. Cottica 2004; Zaccaria Ruggiu 2007, pp. 229–234, fig. 19–22; Raynaud 2009, pp. 60–66 and 87–90; Guiglia Guidobaldi 2011, pp. 413–424.

179

Philipp Niewöhner

Fig. 7. Miletus, Bishop’s Palace, central section of the north wall of the Northern Hall, looking north-west. To the left, Phase I, without bricks save the door of North Room III that was closed with bricks in Phase II; further to the right, the wall surface is broken where the eastern back wall of the former eastern hall used to connect in Phase I; to the right, Phase II with layers of bricks (P. Niewöhner).

opus sectile floor (fig. 9) and marble wall revetment the best that the period could provide. The longitudinal walls were each pierced by a symmetrical set of three doorways and two wide arches that alternated in a rhythmical order, the wide arches each subdivided by a central column. The northern openings all gave onto the Northern Hall and had no specific purpose apart from mirroring the southern wall, which goes to show that symmetry and a decorative effect were paramount. The wide arches in the southern wall each gave onto a room (South Rooms II and IV) that seems to have had no other source of light and would thus have appeared like a deep niche (4.5 m x 5.5 m = 25 m²). This and a mosaic floor with winged personifications of the seasons in South Room II (fig. 13) are consistent with a use as triclinia or dining rooms.16 Thus, the Main Hall and its annexes would have served to receive and entertain guests in an impressive setting. The imposing effect of the Main Hall would have been further enhanced by the addition of the Apsidal Hall, which appears to have served as vestibule. Guests would have arrived from the south, via the street that was blocked by the Apsidal Hall and now ended in front of St Michael’s (fig. 2). The street was distinguished by marble pavement that appears to have been newly added in the Early Byzantine period.17 The mosaic floor of the Apsidal Hall guided the visitor to a nodal point in front of the wide main door to the Main Hall, where a large circular ornament would have arrested movement along the Apsidal Hall and marked the turning point towards the Main Hall (fig. 12).

Fig. 8. Miletus, Bishop’s Palace, tree and jumping cat in the centre of a Late Antique floor mosaic in the eastern part of the Northern Hall (Phase II), looking east (W. MüllerWiener).

of the previously public street to the west and of the former sanctuary of Dionysus to the south points to a powerful and influential member of the city’s Early Byzantine elite. Building Phase IV consists of repair work, in particular the northern part of the Apsidal Hall, where some wild hunting scenes and a paradisiacal meeting of animals at a fountain were inserted into the formerly purely geometric floor mosaic (fig. 12). This and other repairs appear to be contemporary with the building of the basilica church of St Michael in the early seventh century.

Triclinia in general: Dunbabin 1991; 1996. The mosaics: cf. Hachlili 2009, pp. 184–191; Decriaud 2011, pp. 309–332. Cf. also a potential triclinium at Caesarea Maritima and the Mosaic Room 2 in the so-called Odeion-Hanghaus at Ephesus with similar mosaics and busts: Spiro 1992, pp. 250, 254, fig. 12–14; Strocka 1995, pp. 82–86, fig. 4 f.; ScheibelreiterGail 2011, p. 253, fig. 194. Examples of episcopal triclinia: the triclinium of the orthodox bishop’s palace at Ravenna, the so-called domus quinque agubitas that Agnellus describes with five niches: Deichmann 1958–89, II/1, pp. 194–197; the triclinium major and the so-called aula concilii, with three and eleven niches, respectively, that Pope Leo III (795–816) built in his Lateran Palace: Luchterhandt 1999. 17 Niewöhner 2015, pp. 202–205. 16

Function The fifth-century re-configuration of the Bishop’s Palace (Phase III) was geared towards reception and representation. The new Main Hall was exceptionally large (11.75 m x 21.45 m = 252 m²), and its sumptuous 180

The Identification of the Bishop’s Palace at Miletus in Caria (Turkey)

Fig. 9. Miletus, Bishop’s Palace, ground plan with all floors as excavated (Phase IV) (W. Müller-Wiener, P. Niewöhner).

As to the question of which powerful and influential member of the Early Byzantine urban elite could have appropriated a street and the former sanctuary of Dionysus and turned them into a palatial mansion with chapel and opulent reception facilities, the closest approximation to a straight answer is provided by the building inscription of the basilica church that replaced the earlier chapel and temple cella in the seventh century (fig. 14).18 The inscription names the patriarch at Constantinople and other high clergy, but makes no mention of any lay donor, which would appear strange in the case of any other local church, but makes sense, if the bishop had his palace chapel rebuilt. This interpretation appears to be confirmed by the word eukterion, which the inscription uses in reference to St Michael, as this term was commonly employed for oratories or chapels rather than for parish churches or cathedrals.19 A similar appropriation of formerly public urban streets is attested for the Late Antique episcopium of Parentium/Poreč in Croatia.20 Fig. 10 . Miletus, Bishop’s Palace, Apsidal Hall (fifth century, Phase III), south-eastern corner with geometric mosaic floor and a marble frame for a drain (left), looking north (W. Müller-Wiener).

18 Gregoire 1922, pp. 71 f., cat. 225; Rehm et al. 1997–2006, II, pp. 144 f., cat. 1007, pl. 56, 343; Günther 2017, p. 153, s.v. Γεώργιος. 19 Beck 1959, pp. 83–86; Thomas 1987, passim; Magdalino 1991. 20 Chevalier, Matejčić 2012; Sturm 2017.

181

Philipp Niewöhner

Fig. 11. Miletus, Bishop’s Palace, colour scheme of the geometric mosaic floor at the southern end of the Apsidal Hall (fifth century, Phase III), looking north (T. Wiegand).

Given the substantial dimensions of the seventh-century basilica at Miletus, which included a baptistery (fig. 3), it seems outsized in comparison with chapels of more ordinary private houses,21 while other episcopal residences are known to have included sizable churches.22 Also, the fifth-century re-configuration and enlargement of the residence (Phase III) as well as the seventh-century rebuilding of the palace chapel (Phase IV) came at a time

Fig. 12. Miletus, Bishop’s Palace, drawing of the mosaic floor in the northern half of the Apsidal Hall, looking north. A large circular ornament (bottom) marks the turning point for the central door of the Main Hall. At the top, in front of the apse, wild hunting scenes as well as a paradisiacal meeting of animals at a fountain have been inserted later (Phase IV, seventh century?) into the originally purely geometrical pattern (Phase III, fifth century) that continues to the left and to the right of the later insertion (W. MüllerWiener).

21 Cf. the private chapel under SS Giovanni e Paolo at Rome: Rossiter 1989, pp. 103–105; Stupperich 1994, pp. 160–163; Brenk 2003, pp. 98– 105; or at Sinekkale in Cilicia: Dagron, Callot 1988, pp. 58–61; Eichner 2008; Ead. 2011, pp. 287–313. 22 Cf. Real 2012, and Brandenburg 2015, cols. 760–778, for general overviews, and cf. for example the capella arcivescovile that Bishop Peter II (494–520) added to his residence at Ravenna: Deichmann 1958– 89, II/1, pp. 198–204; Miller 1991–92; Rizzardi 2004. An oratory in the residence of Bishop Lawrence (489–510/512) at Milan: Marano 2007, p. 111. Oratories dedicated to Silvester and Lawrence in the Lateran Palace at Rome: Luchterhandt 1999, pp. 113–115; Real 2004. Various older oratories around the baptistery at the Lateran: Brandt 2004; Brandt, Guidobaldi 2008. An oratory of St Stephen in the episcopal residence of Hippo Regius in Africa that was built in 424 and described by Augustin: Real 2003, p. 222. An oratory in the residence of Patriarch John the Merciful (610–619) of Alexandria: Krumeich, Seeliger 2007, p. 25. Cf. also the episcopal(?) complexes at Philippi in Greece and Apamea in Syria with their large churches: Müller-Wiener 1989, pp. 659–664 and 688–690; Caillet 2010, pp. 510 f., 514 f.

when most other urban elite houses in Asia Minor were starting to deteriorate, were downgraded, subdivided and eventually given up.23 The latter can be linked to the Late Antique ‘flight of the curiales’, the disappearance of the ancient urban elites, and their replacement with Byzantine notables, above all the bishop.24 Short of the provincial governor, who has been credited with the building of a comparable fifth-century residence at Ephesus,25 the Zaccaria Ruggiu 2007; Rose 2011, pp. 161 f.; Niewöhner 2017, passim. Brandes, Haldon 2000; Liebeschuetz 2001; Laniado 2002; Saradi 2006, pp. 148–185. 25 Pülz 2010, pp. 554–556. 23 24

182

The Identification of the Bishop’s Palace at Miletus in Caria (Turkey)

Fig. 13. Miletus, Bishop’s Palace, drawing of the mosaic floor in the north-western corner of South Room II, which may have served as triclinium or dining room, showing a personification of a season(?), looking south-east (T. Wiegand).

Fig. 14. Miletus, Bishop’s Palace, inscription relating the seventh-century (re-)building of the palace chapel, the eukterion or oratory of St Michael (Miletarchiv, Hamburg).

bishop was the most likely candidate, when it came to constructing a new and exceptional urban mansion in the Byzantine period,26 and his professional requirements may also explain the focus on reception rooms rather than living quarters.

Minor, and in particular figural scenes are seldom attested and typically restricted to particular spaces,27 for example the atrium and the baptistery but not the nave and aisles of the Great Church at Miletus.28 At Constantinople, where walls and vaults of Early Christian buildings are also preserved, figural images were not customary before the age of Iconoclasm.29 This distinguishes Asia Minor from other provinces with a rich array of Early Christian

The absence of Christian motifs in the floor mosaics of the Bishop’s Palace is of no concern. Such motifs were the exception rather than the rule in Early Christian Asia

Scheibelreiter-Gail 2011, pp. 111–115. Niewöhner 2016a, pp. 13–38. 29 Niewöhner 2014, pp. 263–269. 27 28

26

Pallas 1971; Malaspina 1975; Real 2003; Sterk 2004; Rapp 2005.

183

Philipp Niewöhner figures, for examples Ravenna and Mount Sinai.30 The restraint may have been connected with the many Jewish communities of Asia Minor,31 who would have established a high threshold for religious imagery. Personifications of seasons, as in South Room II (fig. 13), or hunting scenes, as in the Apsidal Hall (fig. 12), were a different matter; such allegorical images were equally common in houses, churches and synagogues and are thus not indicative of any particular milieu.32

Brandt O., 2004. ‘L’oratorio della Santa Croce’, in MEFRA, 116, pp. 79–93. Brandt O., Guidobaldi F., 2008. ‘Il battistero lateranense. Nuove interpretazione delle fasi strutturali’, in Rivista di archeologia cristiana, 84, pp. 189–287. Brenk B., 2003. Die Christianisierung der spätrömischen Welt. Stadt, Land, Haus, Kirche und Kloster in frühchristlicher Zeit, Wiesbaden (Spätantike – frühes Christentum – Byzanz. Reihe B: Studien und Perspektiven, 10).

Bibliography

Caillet J.-P., 2010. ‘Remarques sur la problématique des “palais épiscopaux” à la fin de l’Antiquité’, in N. Duval, V. Popović (eds.), Caričin Grad, III, Rome (Collection de l’École française de Rome, 75/3), pp. 508–523.

Abbreviations: AA = Archäologischer Anzeiger. Ist. Mitt. = Istanbuler Mitteilungen.

Ceylan B., 2007. ‘Episkopeia in Asia Minor’, in Lavan, Özgenel, Sarantis 2007, pp. 169–194.

JRA = Journal of Roman Archaeology. MEFRA = Mélanges de l’École française de Rome. Antiquité.

Ceylan B., 2014. ‘Episcopeia as a reflection of the image of the bishop’, in P. Pensabene, C. Sfameni (eds.), 2014. La villa restaurata e i nuovi studi sull’edilizia tardoantica. Atti del convegno internazionale del Centro interuniversitario di studi sull’edilizia abitativa tardoantica nel Mediterraneo, CISEM, Piazza Armerina 7–10 novembre 2012, Bari (Insulae Diomedeae, 23), pp. 191–196.

RbK = K. Wessel, M. Restle (ed.), since 1963. Reallexikon zur byzantinischen Kunst, Stuttgart. Andreescu-Treadgold I., Treadgold W., 1997. ‘Procopius and the Imperial panels of S. Vitale’, in Art Bulletin, 79, pp. 708–723.

Chevalier P., Matejčić I., 2012. ‘The Episcopium of Poreč’, in S. Balcon-Berry, F. Baratte, J.-P. Caillet, D. Sandron (eds.), 2012. Des domus ecclesiae aux palais épiscopaux. Actes du colloque tenu à Autun du 26 au 28 novembre 2009, Turnhout (Bibliothèque de l’Antiquité tardive, 23), pp. 163–172.

Andreopoulos A., 2002. ‘The mosaic of the Transfiguration in St. Catherine’s Monastery on Mount Sinai’, in Byzantion, 72, pp. 9–37. Angiolini Martinelli P. (ed.), 1997. La basilica di San Vitale in Ravenna – The Basilica of San Vitale, Ravenna, Modena (Mirabilia Italiae, 6).

Cormack R., 1990. ‘The temple as the cathedral’, in C. Roueché, K. T. Erim (eds.), Aphrodisias Papers, I, Ann Arbor (JRA. Supplementary Series, 1), pp. 75–88.

Beck H.-G., 1959. Kirche und theologische Literatur im byzantinischen Reich, Munich (Hanbuch der Altertumswissenschaften, 12/2/1).

Cottica D., 2004. ‘Pavimenti in opus sectile dall’insula 104 a Hierapolis di Frigia’, in Rivista di archeologia, 28, pp. 89–106.

Brandenburg H., 2015. ‘Palast’, in G. Schöllgen, H. Brakmann, S. de Blaauw, T. Fuhrer, H. Leppin, W. Löhr, H.-G. Nesselrath (ed.), Reallexikon für Antike und Christentum. Sachwörterbuch zur Auseinandersetzung des Christentums mit der antiken Welt, 26, Stuttgart, cols. 699–785.

Dagron G., Callot O., 1988. ‘Les bâtisseurs isauriens chez eux’, in I. Ševčenko, I. Hutter (eds.), ΑΕΤΟΣ. Studies in Honour of Cyril Mango, Presented to Him on April 14, 1998, Stuttgart/Leipzig, pp. 55–70.

Brandes W., Haldon J., 2000. ‘Towns, tax, and transformation. State, cities and their hinterlands in the East Roman world, c. 500–800’, in G. P. Brogiolo, N. Christi, N. Gauthier (eds.), Towns and Their Territories Between Late Antiquity and the Early Middle Ages, Leiden (The Transformation of the Roman World, 9), pp. 141–172.

Decriaud A.-S., 2011. ‘Les saisons personnifiées sur les mosaïques romaines tardives (4e–6e siècles) de la partie orientale du bassin méditerranéen (Turquie, Syrie, Liban, Israël, Jordanie)’, in Şahin 2011, pp. 309–332. Deichmann F.-W., 1956. Studien zur Architektur Konstantinopels im 5. und 6. Jahrhundert nach Christus, Baden-Baden (Deutsche Beiträge zur Altertumswissenschaft, 4).

30 Cf. for example the opulent imagery in the Justinianic church of San Vitale at Ravenna and the figural mosaics of St Catherine of Mount Sinai that were sponsored by the same emperor, on the one hand, and, on the other hand, the contemporary, but aniconic mosaics of Justinian’s St Sophia at Constantinople: Whittemore 1933; Forsyth, Weitzmann 1973, pp. 11–18; Deichmann 1958–89, II/2, pp. 163 f., 177 f.; Angiolini Martinelli 1997; Andreescu-Treadgold, Treadgold 1997; Andreopoulos 2002. 31 Herz, Kobes 1998. 32 Hachlili 2009, pp. 163 f., and 184–191.

Deichmann F.-W., 1958–89. Ravenna. Hauptstadt des spätantiken Abendlandes, Stuttgart/Wiesbaden. Dunbabin K., 1991. ‘Triclinium and stibadium’, in W. J. Slater (ed.), Dining in a Classical Context, Ann Arbor, pp. 121–148. 184

The Identification of the Bishop’s Palace at Miletus in Caria (Turkey) Kleiner G., 1968. Die Ruinen von Milet, Berlin.

Dunbabin K. M. D., 1996. ‘Convivial spaces: Dining and entertainment in the Roman villa’, in JRA, 9, pp. 66–80.

Knackfuß H., 1924. Der Südmarkt und die benachbarten Bauanlagen, Berlin (Milet. Ergebnisse des Ausgrabungen und Untersuchungen seit dem Jahre 1899, 1/7)

Duval N. (ed.), 1989. Actes du XIe Congrès international d’archéologie chrétienne, Lyon, Vienne, Grenoble, Genève et Aoste (21–28 septembre 1986), Rome/ Vatican City (Collection de l’École française de Rome, 123; Studi di Antichità cristiana, 41).

Krumeich K., Seeliger H. R. (eds.), 2007. Archäologie der antiken Bischofssitze, I, Wiesbaden (Sprachen und Kulturen des christlichen Orients, 15).

Eichner I., 2008. ‘Sinekkale – Herberge, Kloster oder Gutshof?’, in Olba, 16, pp. 337–360.

Laniado A., 2002. Recherches sur les notables municipaux dans l’Empire protobyzantin, Paris (Travaux et mémoires du Centre de recherche d’histoire et civilisation de Byzance. Monographies, 13).

Eichner I., 2011. Frühbyzantinische Wohnhäuser in Kilikien. Baugeschichtliche Untersuchung zu den Wohnformen in der Region um Seleukeia am Kalykadnos, Tübingen (Istanbuler Forschungen, 52).

Lavan L., Ozgenel L., Sarantis A. (eds.), 2007. Housing in Late Antiquity. From Palaces to Shops, Leiden/Boston (Late Antique Archaeology, 3/2).

Emme B., 2016. ‘Zur Chronologie des DionysosHeiligtums von Milet’, in Ist. Mitt., 66, pp. 355–362.

Liebeschuetz J. H. W. G., 2001. The Decline and Fall of the Roman City, Oxford/New York.

Feld O., 1996. ‘Eine Kirche für Maria in Miletos’, in C. L. Striker (ed.), Architectural Studies in Memory of Richard Krautheimer, Mainz, pp. 67–70.

Luchterhandt M., 1999. ‘Päpstlicher Palastbau und höfisches Zeremoniell unter Leo III’, in C. Stiegemann, M. Wemhoff (eds.), 799. Kunst und Kultur der Karolingerzeit. Karl der Große und Papst Leo III. in Paderborn, III, Mainz, pp. 109–122.

Filges A., Bumke H., Röver E., Stümpel H., 2002. ‘Didyma. Bericht über die Arbeiten 2000’, in AA, 2002/1, pp. 79–118. Forsyth G. H., Weitzmann K., 1973. The Monastery of Saint Catherine at Mount Sinai. The Church and Fortress of Justinian, Ann Arbor.

Magdalino P, 1991. ‘Eukterion’, in A. P. Kazhdan (dir.), The Oxford Dictionary of Byzantium, New York/ Oxford, p. 745.

Gregoire H., 1922. Recueil des inscriptions grecques chrétiennes d’Asie Mineure, Paris.

Malaspina M., 1975. ‘Gli episcopia e le residenze ecclesiastiche nella ‘pars orientalis’ dell’impero romano’, in M. Cagiano de Azevedo (ed.), Contributi dell’Istituto di archeologia, V, Milano (Pubblicazioni della Università cattolica del Sacro Cuore. Scienze storiche, 10), pp. 29–173.

Guiglia Guidobaldi A., 2011. ‘The marble floor decoration in Constantinople’, in Şahin 2011, pp. 413–436. Günther W., 2017. Inschriften von Milet, IV, Berlin (Milet. Ergebnisse des Ausgrabungen und Untersuchungen seit dem Jahre 1899, VI/4).

Marano Y., 2007. ‘Domus in qua manebat episcopus. Episcopal residences in northern Italy during Late Antiquity (4th to 6th centuries AD)’, in Lavan, Özgenel, Sarantis 2007, pp. 97–129.

Hachlili R., 2009. Ancient Mosaic Pavements. Themes, Issues, and Trends. Selected Studies, Leiden/Boston. Hebert L., 2000. The Temple-Church at Aphrodisias, PhD dissertation, New York University.

Miller M. C., 1991–92. ‘The development of the archiepiscopal residence in Ravenna, 300–1300’, in Felix Ravenna, 141–144, pp. 145–173.

Herz P., Kobes J. (eds.), 1998. Ethnische und religiöse Minderheiten in Kleinasien, von der hellenistischen Antike bis in das byzantinische Mittelalter, Wiesbaden (Mainzer Veröffentlichungen zur Byzantinistik, 2).

Müller-Wiener W., 1977–78. ‘Michaelskirche und Dionysos-Tempel [von Milet] – Baubeschreibung und Phasengliederung’, in Ist. Mitt., 27–28, pp. 95–103.

Karwiese S., 1989. Die Marienkirche in Ephesos. Erster vorläufiger Grabungsbericht 1984–1986. Erster vorläufiger Gesamtbericht über die Wiederaufnahme der archäologischen Untersuchung der Marienkirche in Ephesos, Vienna (Österreichische Akademie der Wissenschaften, Philosophisch-historische Klasse. Denkschriften, 200).

Müller-Wiener W., 1979. ‘Untersuchungen im DionysosHeiligtum [von Milet]’, in Ist. Mitt., 29, pp. 162–169. Müller-Wiener W., 1989. ‘Bischofsresidenzen des. 4.–7. Jhs. im östlichen Mittelmeer-Raum’, in Duval 1989, pp. 651–709. Müller-Wiener W., 1997. ‘Milet’, in RbK, VI, cols. 362– 377.

Kästner V., 2016. ‘Das Orpheusmosaik aus Milet’, in H. Schwarzer, H.-H. Nieswandt (eds.), ‘Man kann es sich nicht prächtig genug vorstellen’. Festschrift für Dieter Salzmann zum 65. Geburtstag, Marsberg, pp. 535–543.

Naumann R., 1980. ‘Die Ausgrabungen bei den Thermen in Didyma’, in Ist. Mitt., 30, pp. 177–189.

Keil J., Knoll F., Reisch E., 1932. Die Marienkirche in Ephesos, Vienna (Forschungen in Ephesos, 4/1).

Niewöhner P., 2014. ‘Historisch-topographische Überlegungen zum Trierer Prozessionselfenbein, dem 185

Philipp Niewöhner Christusbild an der Chalke, Kaiserin Irenes Triumph im Bilderstreit und der Euphemiakirche am Hippodrom’, in Millennium, 11, pp. 261–288.

tardo antico all’alto medioevo’, in Atti e memorie della deputazione di storia Patria per le provincie di Romagna, 55, pp. 147–175.

Niewöhner P., 2015. ‘Der Bischofspalast von Milet. Spätrömisches Peristylhaus und frühbyzantinische Residenz’, in AA, 2015/2, pp. 181–273.

Rose B., 2011. ‘Troy and the Granicus River Valley in Late Antiquity’, in O. Dally, C. J. Ratté (eds.), Archaeology and the Cities of Asia Minor in Late Antiquity, Ann Arbor (Kelsey Museum Publication, 6), pp. 151–171.

Niewöhner P., 2016a. Die byzantinischen Basiliken von Milet, Berlin (Milet. Ergebnisse des Ausgrabungen und Untersuchungen seit dem Jahre 1899, I/11).

Rossiter J. J., 1989. ‘Roman villas of the Greek East and the villa in Gregory of Nyssa Ep. 20’, in JRA, 2, pp. 101–110.

Niewöhner P. (ed.), 2016b. Miletus / Balat. Urbanism and Monuments from the Archaic to the Ottoman Periods, Istanbul.

Şahin M. (ed.), 2011. XI. Uluslararasi Antik Mozaik Sempozyumu 16–20 Ekim 2009 Bursa, Türkiye. Türkiye Mozaikleri ve Antik Dönemden Ortaçağ Dünyasina Diğer Mozaiklerle Paralel Gelişimi. Mozaiklerin Başlangicindan Geç Bizans Çağina Kadar Ikonografi, Stil ve Teknik Üzerine Sorular – 11th International Colloquium on Ancient Mosaics, October 16th–20th, 2009, Bursa, Turkey. Mosaics of Turkey and Parallel Developments in the Rest of the Ancient and Medieval World. Questions of Iconography, Style and Technique from the Beginnings of Mosaic until the Late Byzantine Era, Istanbul (Uladağ University Mosaic Research Center Series, 1).

Niewöhner P. (ed.), 2017. The Archaeology of Byzantine Anatolia. From the End of Late Antiquity until the Coming of the Turks, New York. Niewöhner P., 2018. ‘The Late Antique “area” of Miletus and other Christian cemeteries in the Aegean’, in M. Korres, S. Mamaloukos, K. Zampas, F. MallouchouTufano (eds.), Ήρως Κτίστης. Μνήμη Χαράλαμπου Μπούρα, Athens, pp. 263–272. Pallas D. I., 1971. ‘Episkopion’, in RbK, II, cols. 335–371. Pülz A., 2010. ‘Zum Stadtbild von Ephesos in byzantinischer Zeit’, in F. Daim, J. Drauschke (eds.), Byzanz – Das Römerreich im Mittelalter / Byzantium – The Roman Empire in the Middle Ages / Byzance – L’Empire romain au Moyen Âge, II/2, Mainz (RömischGermanisches Zentralmuseum. Monographien, 84/2), pp. 541–571.

Saradi H. G., 2006. The Byzantine City in the Sixth Century. Literary Images and Historical Reality, Athens. Scheibelreiter-Gail V., 2011. Die Mosaiken Westkleinasiens. Tessellate des 2. Jahrhunderts v. Chr. bis Anfang des 7. Jahrhunderts, Vienna (Sonderschriften des Österreichischen Archäologischen Institutes in Wien, 46).

Rapp C., 2005. Holy Bishops in Late Antiquity. The Nature of Christian Leadership in a Time of Transition, Berkeley/Los Angeles/London (The Transformation of the Classical Heritage, 37; The Joan Palevsky Imprint in Classical Literature).

Sodini J.-P., 1989. ‘Les groupes épiscopaux de Turquie’, in Duval 1989, pp. 405–426. Spiro M., 1992. ‘Some Byzantine mosaics from Ceasarea’, in R. L. Vann (ed.), Caesarea Papers. Straton’s Tower, Herod’s Harbour, and Roman and Byzantine Caesarea. Including the Papers Given at a Symposium Held at the University of Maryland, the Smithsonian Institution, and the Jewish Community Center of Greater Washington on 25–28 March, 1988, Ann Arbor (JRA. Supplementary Series, 5), pp. 245–259.

Raynaud M.-P., 2009. Corpus of the Mosaics of Turkey, I, Bursa (Uladağ University Mosaic Research Center Series, 2). Real U., 2003. ‘Die Bischofsresidenz in der spätantiken Stadt’, in G. Brands, H.-G. Severin (eds.), Die spätantike Stadt und ihre Christianisierung. Symposion vom 14. bis 16. Februar 2000 in Halle-Saale, Wiesbaden (Spätantike – frühes Christentum – Byzanz. Reihe B: Studien und Perspektiven, 11), pp. 219–237.

Sterk A., 2004. Renouncing the World, Yet Leading the Church. The Monk-Bishop in Late Antiquity, Cambridge (MA)/London. Strocka V. M., 1995. ‘Tetrarchische Wandmalereien in Ephesos’, in Antiquité tardive, 3, pp. 77–89.

Real U., 2004. ‘La residenza lateranense dall’età di Giustiniano all’inizio dell’epoca carolingia’, in MEFRA, 116, pp. 95–108.

Stupperich R., 1994. ‘Zur Beschreibung einer galatischen Villa im 20. Brief Gregors von Nyssa’, in E. Schwertheim (ed.), Forschungen in Galatien, Bonn (Asia Minor Studien, 12), pp. 157–169.

Real U., 2012. Bischofsresidenzen in der Spätantike, PhD Dissertation, Universität Münster. Rehm A., Dessau H., Hermann P., Günther W., Norbert E. (ed.), 1997–2006. Inschriften von Milet, Berlin (Milet. Ergebnisse des Ausgrabungen und Untersuchungen seit dem Jahre 1899, VI/1–3).

Sturm J., 2017. ‘Late Antique episcopal complexes: Bishop Eufrasius and his residence at Porec (Croatia)’, in D. W. Rollason (ed.), Princes of the Church. Bishops and Their Palaces, London (The Society for Medieval Archaeology Monographs, 39), pp. 23–33.

Rizzardi C., 2004. ‘L’episcopio di Ravenna nell’ambitio dell’edilizia religiosa occidentale e orientale dal 186

The Identification of the Bishop’s Palace at Miletus in Caria (Turkey) Thomas J. P., 1987. Private Religious Foundations in the Byzantine Empire, Washington (Dumbarton Oaks Studies, 24). Thür H., 2009. ‘Ziegelmauerwerk in Ephesos’, in M. Bachmann (ed.), Bautechnik im antiken und vorantiken Kleinasien. Internationale Konferenz, 13.–16. Juni 2007 in Istanbul, Istanbul (Byzas, 9), pp. 483–496. von Gerkan A., 1925. Kalabaktepe, Athenatempel und Umgebung, Berlin (Milet. Ergebnisse des Ausgrabungen und Untersuchungen seit dem Jahre 1899, I/8). Whittemore T., 1933. The Mosaics of St Sophia in Istanbul. Preliminary Report on the First Year of Work, 1931– 1932. The Mosaics of the Narthex, Oxford/Paris. Wiegand T., 1911. ‘Siebenter vorläufiger Bericht über die von den Königlichen Museen in Milet und Didyma unternommenen Ausgrabungen’, in Abhandlungen der Königlich Preussischen Akademie der Wissenschaften. Philosophisch-historische Classe, 1911, Anhang, I. Zaccaria Ruggiu A., 2007. ‘Regio VIII, insula 104. Le strutture abitative. Fasi e trasformazioni’, in F. D’Andria, M. Piera Caggia (eds.), Hierapolis di Frigia I. Le attività delle campagne di scavo e restauro 2000– 2003, Istanbul (Hierapolis di Frigia, 1), pp. 211–256.

187

14 Episcopal Basilica of Dion, Greece: The Triconch Building and the Location of the Bishop’s Residence Kyriakos Fragoulis* University of Birmingham (United Kingdom) The Episcopal Basilica of Dion is one of the most important monuments of the city, and occupies a central position in the urban grid. Over its lifetime, the Basilica underwent two major building phases that correspond to two successive churches, conventionally called Basilica A and Basilica B. The first church was a three-aisled timber-roofed basilica that dates, most probably, to the early fifth century AD. After a period of operation, the time limits of which cannot be easily determined, Basilica A was replaced by Basilica B. The new church was equipped with various annexes, including a triconch building situated to its south. The main question concerning the triconch of Dion is that of its function. Although its identification as a martyrium has been proposed, there are several indications that drive us to a new interpretation. Given its position, articulation and relation to the surrounding buildings, as well as the absence of finds typical of a cult building, we consider it quite likely that the triconch of Basilica B was a triclinium. If this is the case, it could only belong to the bishop’s residence, the rest of which must be situated in the unexcavated area to the south of the atrium. This hypothesis gives a new turn to the issue of the overall shape and organisation of the episcopal complex of Dion, and will hopefully contribute to the discussion on the identification of the Early Byzantine episcopal residences of Greece. La basilique épiscopale de Dion est parmi les monuments les plus importants de la ville et occupe une place centrale dans le réseau urbain. Elle a subi deux grandes phases de construction qui correspondent à deux églises successives, appelées conventionnellement basilique A et basilique B. La première église était une basilique charpentée à trois nefs qui date, très probablement, du début du Ve siècle ap. J. C. Après une période d’opération dont les limites ne peuvent pas être facilement déterminées, la basilique B a remplacé la basilique A. La nouvelle église était équipée de diverses annexes, dont un bâtiment triconque situé au sud. La question principale concernant le triconque de Dion concerne sa fonction. Bien qu’on ait proposé de l’identifier comme un martyrium, plusieurs indices nous conduisent à une nouvelle interprétation. Sa position, son articulation, sa relation avec les bâtiments voisins, ainsi que l’absence de découvertes typiques d’un édifice de culte nous invitent à y voir un triclinium. Si tel est le cas, il ne pouvait appartenir qu’à la résidence de l’évêque, le reste devant être situé dans la zone non fouillée au sud de l’atrium. Cette hypothèse ouvre de nouvelles perspectives sur la forme et l’organisation du complexe épiscopal de Dion et pourra contribuer à la discussion sur l’identification des résidences épiscopales protobyzantines de Grèce. Die Bischofsbasilika von Dion gehört zu den wichtigsten Denkmälern der Stadt und hat eine zentrale Lage in der Stadt. Die Basilika hat zwei Hauptbauphasen, das heißt zwei aufeinanderfolgende Kirchenbauten, die Basilika A und B genannt werden. Basilika A war eine dreischiffige mit Holzschindel bedeckte Basilika, die wohl Anfang des 5. Jhs. n. Chr. errichtet wurde. Nach einer längeren Nutzungszeit wurde sie durch die Basilika B ersetzt. Die Basilika B zeigt im Westen ein großes Atrium mit Anbauten an drei Seiten, später wurde ein trilaterales Baptisterium auf der Westseite noch hinzugefügt. Im Süden der Kirche wurde ein Dreikonchenbau ausgegraben. Die wichtigste Frage bezüglich des Dreikonchenbaus von Dion ist die Bestimmung seiner Funktion. Er wurde zwar als ein Martyrium gedeutet, es gibt aber Hinweise, die eine neue Interpretation ermöglichen. Aufgrund seiner Lage, seiner Erscheinung * This paper comprises a revised version of a chapter from Fragoulis 2014, the Master’s thesis that I completed as part of my postgraduate studies at the Aristotle University of Thessaloniki, with the support of the

Alexander S. Onassis Public Benefit Foundation. At this point, I would like to express my gratitude to my supervisor, Prof. Aristotelis Mentzos, for his help, guidance and support throughout my project.

189

Kyriakos Fragoulis und seines Verhältnisses zu den umliegenden Gebäuden sowie aufgrund des Fehlens der für ein Kultgebäude typischen Befunde gehen wir davon aus, dass der Dreikonchenbau der Basilika B ein Triclinium war. Wenn dies stimmt, müsste es zu einer Bischofsresidenz gehört haben, deren restliche Bereiche im ausgegrabenen Gebiet südlich des Atriums zu erwarten wären. Diese Hypothese gibt auch der Diskussion darüber wie Form und Organisation der bischöflichen Komplexe von Dion im Allgemeinen aussahen, eine neue Wendung und könnte dazu beitragen über die Identifizierung der frühbyzantinischen bischöflichen Residenzen in Griechenland zu diskutieren. La Basilica Episcopale di Dion è tra i più importanti monumenti della città ed occupa una posizione centrale nell’impianto urbanistico. Nel corso della sua esistenza, la Basilica ha avuto due principali fasi costruttive che corrispondono a due chiese che si susseguono l’una sull’altra, convenzionalmente chiamate Basilica A e B. La prima chiesa era una basilica a tre navate con tetto ligneo che risale, probabilmente, ai primi anni del V secolo d.C. Dopo un primo periodo di utilizzo, i cui limiti cronologici non possono essere facilmente determinati, la Basilica B sostituisce la Basilica A. La nuova chiesa viene dotata di vari annessi, tra cui un edificio triconco sul lato sud. La questione principale relativa all’edificio triconco di Dion è l’identificazione della sua funzione. Nonostante sia stato proposto di identificarlo con un martyrium, ci sono diversi motivi che ci portano a una nuova interpretazione. In base alla sua posizione, alla sua articolazione, e alla sua relazione con gli edifici circostanti, nonché per l’assenza di reperti tipici di un edificio di culto, riteniamo molto probabile che il triconco della Basilica B fosse un triclinium. Se questo è il caso, esso potrebbe appartenere solo alla residenza del vescovo, che a sua volta deve essere situata nella zona non scavata a sud dell’atrio. L’ipotesi di cui sopra dà una nuova svolta alla questione della forma complessiva e dell’organizzazione del complesso episcopale di Dion e, si spera che possa contribuire al dibattito sull’individuazione delle residenze episcopali protobizantine di Grecia. roofed basilica that dates, most probably, to the early fifth century AD. At the east side, the church ended in a broad apse equipped with a built synthronon. Floor mosaics with geometric motifs are preserved in the nave and the narthex, while the bema floor was laid with marble opus sectile. The area of the sanctuary was also separated from the nave by a Π-shaped chancel screen. The interior walls of Basilica A were covered in murals – imitating colourful marble incrustations – supplemented with stucco decorations.

Dion, one of the most historic sites in Macedonia, is situated at a strategic location in the north-eastern foothills of Mount Olympus – relatively close to the sea – overlooking the passage between Macedonia and Thessaly (fig. 1). In the Classical and Hellenistic periods, it served as the sacred city of the Macedonians, while the sanctuary of Olympian Zeus, located outside and south of the defensive wall, functioned as the centre of the religious life of their kingdom. Following the Roman conquest in the second century BC, Dion entered a new phase with the foundation of a colony there, most likely by Octavian (later Augustus) soon after his victory at Actium in 32/31 BC. During Late Antiquity, it remained the major urban settlement in the ancient region of Pieria, and became the seat of a bishopric. The central basilica, which can be quite securely identified as the cathedral of Dion,1 is among the most important surviving buildings of the city, occupying a prominent site in the vicinity of the Roman Forum (fig. 2). Even though it was excavated during three periods, in the 1920s,2 1960s3 and 1980s–90s, it remains a little-known monument, and many of its features are still unspecified.

After a period of operation, the time limits of which cannot be easily determined, Basilica A was replaced by Basilica B.5 The new church resulted from the reconstruction of Basilica A, which was then covered with an artificial fill about 2 m high. Most of the walls of Basilica B stand directly upon those of its predecessor, maintained to a certain height as the foundation of the overlying basilica. Consequently, the two buildings have a common outline, except for the east end of Basilica B, which was extended by 5 m, forming a new apse. Basilica B also includes a large atrium, surrounded on three sides by porticos with annexes, and a trilateral baptistery later added to the west side of the atrium (fig. 3–4).

In the course of its lifetime, the Episcopal Basilica of Dion underwent two major building phases that correspond to two successive churches, conventionally called Basilica A and Basilica B (fig. 3–4). Basilica A4 was a three-aisled timber-

One of the most interesting parts of the Basilica B complex, however, is a triconch building located a few metres south of the church (fig. 3–5). Specifically, its west side was an extension of the east wall of the narthex’s southern annex. The triconch has the same orientation as the atrium, although it is considered a slightly later

This identification is inferred from the central location of the basilica, its size and decoration, and the repeated reconstructions and extensions it underwent, as well as by the number and also the type of its annexes (Sotiriadis 1932a, pp. 450–451; Mentzos 1997, p. 154; 2001, p. 8). 2 Sotiriadis 1928, pp. 86–90; 1929, pp. 72–73; 1930, pp. 49–50. 3 Pelekanidis 1964, 1965 and 1966. 4 See Fragoulis 2014, pp. 99–156. 1

5

190

Fragoulis 2014, pp. 159–232.

Episcopal Basilica of Dion, Greece

Fig. 1. Primary and secondary network of settlements in the province of Macedonia Prima during the Early Byzantine period (Drakoulis 2012, map 6 – reproduced courtesy of the author).

Fig. 2. Topographic map of the archaeological site of Dion and detail of the Episcopal Basilica area (University Excavation of Dion, Archives).

191

Kyriakos Fragoulis

Fig. 3. Orthophotograph of the episcopal complex of Dion (University Excavation of Dion, Archives).

Fig. 4. Plan of Basilica A and Basilica B (University Excavation of Dion, Archives).

192

Episcopal Basilica of Dion, Greece

Fig. 5. Episcopal Basilica of Dion, triconch (picture taken from the south-west).

addition to the complex.6 It consists of a central square chamber, measuring 6.50 × 6.50 m, with semi-circular apses of equal size (4.60 m wide – 2.30 m deep) on the north, south and east sides. The central square was most probably timber-roofed, while the apses must have been covered by semi-domes. Two equal openings on the west facade provided access to the interior. Although the area in front of the triconch has only been partly excavated, there is evidence to indicate that it opened directly onto a courtyard or a peristyle. In addition, two adjoining rooms were attached between the basilica and the triconch (fig. 7). The latter communicated with the southern room through a narrow opening in the north apse (fig. 6-7).7

of approximately 0.80 m above the estimated floor level, as no traces of a floor have survived. They are made primarily of rubble masonry bonded with lime mortar (fig. 5–6). Between the stone layers, single rows of brick extending over a limited length are inserted sporadically. The corners were built with large rubble stones, ashlars, and marble blocks, mostly spolia, completed and levelled with brick, while the doorposts were made almost entirely of brick. Poor restoration work has altered the original appearance of the masonry. Inside the building, the excavation reached the base of the foundation, revealing that the triconch was set on earlier walls. As is well known, the architectural form of the triconch spread widely during Late Antiquity.9 Because of its impressive spatial effect and strong symbolism, it was used for various types of buildings with special functions, both in secular and religious architecture, like banquet or audience halls, mausoleums, martyria, and baptisteries. The main question concerning the triconch of Dion is that of the determination of its function. Due to the absence of relevant excavation data, we have to rely mostly on planimetric information. Given its position adjacent to the cathedral and its architectural form, it would be reasonable to assume that it was used as a baptistery or a martyrium.10 Even so, none of these possible explanations

At the time of their discovery,8 the walls of the triconch, which are 1 m thick, were preserved to a maximum height 6 According to the excavation records for the year 1984, the west wall of the triconch does not bond with the masonry of the narthex’s southern annex, indicating that the triconch was added later. Unfortunately, the application of cement-mortar external fillings to sections of the ancient walls, as part of a restoration effort in the 1990s, has made it impossible for us to trace the original joints and further investigate this issue. 7 The walls of these rooms survive today on the level of their foundation zone. The opening in the north apse is also lost. We know of its existence thanks to an unpublished photograph from the 1929 excavation (fig. 6) and a rough ground plan published in Sotiriou 1929, fig. 11. 8 Excavation work in the area of the triconch was carried out for the first time in 1929, bringing to light a portion of the north apse and the two rooms attached to the north side of the building. These remains were misinterpreted as belonging to a circular structure that was identified as a possible baptistery (Sotiriou 1929, p. 180; Sotiriadis 1930, p. 50; 1932b, p. 252). Excavation was resumed in 1984 and again in 1993, revealing the actual form of the building (Pantermalis 1984, p. 73; 1993, p. 196).

See Kurtović-Folić 1997, pp. 473–478; Gattiglia 1998, pp. 190–192. The ground plan and superstructure of the building would be the same in either case.

9

10

193

Kyriakos Fragoulis

Fig. 6. Unpublished photograph from the 1929 excavation. The north apse and the adjoining rooms of the triconch (from the south) (©The Archaeological Society at Athens).

seems convincing enough. In the first case,11 the absence of a font or any hydraulic system, as well as the lack of direct connection to the basilica, make this identification very doubtful. On the other hand, the possibility of a martyrium could be considered more likely, as triconch and tetraconch martyria are relatively common.12 But, apart from typology, there is no additional evidence to support this hypothesis. The lack of direct communication with the main church is again problematic.13 No burials have been found inside or around it, nor a subterranean relic installation or other reliquary arrangement of any kind. There have also been 11 Triconch baptisteries are very rare in Greece and the Balkans. Some examples are the baptisteries of the Ohrid Tetraconch (BitrakovaGrozdanova 1975, pp. 31–33, plan II) and the Uvarov Basilica in Chersonesos at Crimea, the dating of which is particularly problematic (Biernacki et al. [ed.] 2004, pp. 194–195, 201). Tetraconch baptisteries are, however, more common. Cf., e.g., the baptistery of the North Basilica at Stobi (Aleksova 1986, pp. 41, 45–48), and that of the cathedral in Caričin Grad–Iustiniana Prima (Popović et al. 2010). 12 Based on this view, Varalis identifies the triconch of Dion as a martyrium (Varalis 1999, p. 195, n. 3). This identification is also adopted by Baldini, Mazzilli 2011, p. 155. Some examples of triconch martyria from Greece are the triconch of the Kraneion Basilica in Corinth (Laskaris 2000, pp. 373–375), the triconch church at Akrini, Kozani (Stollmayer 1999, pp. 118–120) and the first building phase of the St Gabriel church on Kos (Baldini, Mazzilli 2011, pp. 152–159, fig. ΙΙΙ.3.2). 13 In most cases, memorial chapels that are associated with a basilica are found attached to it. Cf., e.g., a group of (possible) triconch martyria from Lycia in Asia Minor and nearby Rhodes annexed to the east side of the main church (Sodini 2009, pp. 19, 21–23). For a similar articulation, cf. also the well-known example of the cella trichora to the east of the Euphrasius Basilica in Poreč, which was initially built as a free-standing building but in such a way that it communicated directly with the church (Gollubits 2008, pp. 78–98, with bibliography).

Fig. 7. Plan of the triconch of the Episcopal Basilica and its surroundings.

no finds of sculpture typical of a cult building. Moreover, the triconch is not preceded by a vestibule, which would be necessary if it functioned as a martyrium. On the contrary, there are several indications that lead us to a new interpretation. Given its position, articulation, and relation to the surrounding buildings, along with its 194

Episcopal Basilica of Dion, Greece trefoil plan, we consider it quite likely that the triconch of Basilica B was a triclinium. If this is the case, it could only belong to the bishop’s residence, the rest of which must be situated in the unexcavated area to the south of the atrium (fig. 7).

implies concerning the size and wealth of the episcopal complex. Besides the advantage of accommodating more stibadia and hence a larger number of guests,24 triconch and other multi-apsidal triclinia were elaborate, costly structures and therefore symbols of social status. As a result, their number was considerably smaller in comparison to those fitted with a single apse.25 Triconch triclinia were quite popular in the West, while in the East they seem to have been relatively rare.26 For example, many are found in Italy, the best-known being the Sicilian villas at Piazza Armerina27 and Patti Marina,28 the villa found in Desenzano29 on the shores of Lake Garda, and Theodoric’s Palace in Ravenna.30 As for Greece, no such building had been found and investigated until recently. But the excavation work carried out during the erection of the new Acropolis Museum in Athens has changed this picture, showing that the progress of archaeological investigation could lead in time to a re-evaluation of the popularity of the triconch triclinium in this particular geographic area, as more examples will possibly come to light. Specifically, the excavation on the site of the Museum uncovered remains of a quarter of the ancient city that correspond to successive phases of occupation. Among many other finds, a small triconch has been discovered, belonging to an impressive residential complex (building E) that dates, according to the excavators, to the late sixth or early seventh century AD (fig. 8). This triconch is attached to a large rectangular hall, most probably the official reception room, and has been identified as a private triclinium meant for selected guests.31

Episcopia14 were similar in form and layout to the aristocratic residences of Late Antiquity – their size and luxury matching the status of their owner – and were thus designed to suit both the private and public life of their occupants.15 As in the case of the residences of state officials, however, an episcopium must have been suited for broader public use.16 Apart from other necessary facilities, a bishop’s residence was equipped with at least one ceremonial hall that could serve as an audience, reception and banquet chamber.17 In the late third to early fourth century AD, a new form of reception room emerged, the apsidal triclinium, which became the customary triclinium of Late Antiquity.18 This was associated with the emergence of the apse as one of the most prominent architectural elements of this period, in parallel with an important change of fashion in dining, the introduction of a single semi-circular platform couch known as the stibadium.19 Apart from the simple apsidal form, triclinia were also built on a triconch plan, which probably resulted naturally from the replacement of the traditional three rectangular couches with three stibadia.20 Furthermore, there are examples of grand dining halls belonging to palaces and great luxury villas equipped with several apses.21 In the majority of private dwellings, the triclinium, usually the largest and most luxurious room in the house, was used as both a formal dining room and a place for the reception of guests.22 Nevertheless, in the richest Late Antique houses there were separate halls for each of these functions.23 This scheme also seems to apply to Dion, where the triconch must have been primarily used for banquets, since the relatively small dimensions make it rather unsuitable for use as an audience or reception room. Consequently, we consider quite possible the existence of another official hall, still undiscovered, with all that that

As for the existence of three-apsed triclinia in episcopal residences, the best example is that in the episcopal complex of Side32 in Asia Minor. The so-called Bishop’s Palace at Aphrodisias33 in Caria and the episcopium of Hippo34 in North Africa, also equipped with triconch triclinia, were most likely lavish private houses – or, in the first case, the residence of a state official – depending on the interpretation of modern scholarship. Another interesting case is that of the remains of a triconch building that have been discovered directly to the south-east of the Uvarov Basilica, probably the cathedral of Chersonesos at Crimea. According to one interpretation, this triconch, which seems to belong to the first building phase of the episcopal complex, was part of the original episcopium and was used

For a discussion of episcopal residences in Late Antiquity and relevant bibliography, see Baldini Lippolis 2005, pp. 102–136. 15 Brands, Rutgers 2005, p. 816; Rapp 2005, p. 210; Ceylan 2007, pp. 184–185; Baldini 2014, p. 165; Caillet 2012, pp. 161–162. 16 Ellis 2007, p. 9. 17 Miller 2000, p. 39; Brands, Rutgers 2005, p. 815; Rapp 2005, p. 210; Ceylan 2007, p. 172. 18 During the period under examination, the apsed triclinium became the most typical architectural characteristic of urban villas in Greece (Petridis 2008, p. 250; for examples, see Said 2002, pp. 158–160). Nevertheless, the traditional rectangular triclinium continued to be used (Said 2002, pp. 168–175; Karagianni 2012, p. 83). 19 Dunbabin 1991, p. 128ff.; 2003, pp. 191–202; Morvillez 1996, pp. 120–125. 20 Ellis 1988, pp. 571–572; 1991, pp. 119–120. 21 For example, the multi-conchal triclinia in the so-called palaces of Lausus (Bardill 1997, pp. 69–86) and Antiochos near the hippodrome of Constantinople (Baldini Lippolis 1994, pp. 298–302). 22 Lavin 1962, p. 5; Rossiter 1991, pp. 201–202; Dunbabin 2003, pp. 171–172. 23 Polci 2003, pp. 83–88; Sfameni 2004, p. 366; Brands, Rutgers 2005, p. 792; Habermehl 2013, p. 117. 14

Rossiter 1991, p. 203; Dunbabin 1996, p. 77; Polci 2003, pp. 81–82. Lavin 1962, p. 5; Ellis 1991, p. 120; Morvillez 1995, pp. 16, 25. 26 Lavin 1962, pp. 4–5; Morvillez 1995, pp. 16–17; Dunbabin 2003, p. 172; Berenfeld 2009, pp. 222–223. 27 For the extended bibliography on this monument, see Bowes et al. 2011, p. 423 n. 1–2. 28 Sfameni 2004, pp. 340–343. 29 Sfameni 2004, pp. 343–344. 30 Herrin 2015, pp. 58–59 and n. 38. 31 Eleftheratou (ed.) 2006, pp. 18–19; 2009, pp. 8–10; Pantermalis et al. 2016, pp. 30, 32 and fig. 10, 16. 32 Brands, Rutgers 2005, p. 817; Sodini 1989, p. 417; Ceylan 2007, p. 174. 33 Berenfeld 2002, pp. 114–181. Cf. also Baldini 2014, pp.167–168. 34 Baldini Lippolis 2001, pp. 215–216; Barrate 2012, pp. 178–179; Caillet 2012, p. 160; Février 1972, fig. 2. 24 25

195

Kyriakos Fragoulis cases the chord of the apse measured even less than 4 m,43 like the triconch of the fourth-century villa at Littlecote44 in Wiltshire, England. It has been estimated that an average seven-person stibadium demanded an apse at least 6 m in diameter.45 By extension, the triconch of Dion, which is fitted with apses 4.60  m wide, would have held smaller stibadia for five or six diners.46 Beyond design and typological characteristics, the identification of the Dion triconch as a triclinium is strongly supported by its arrangement and placement in the complex. Although attached to the Episcopal Basilica, it retained a certain amount of privacy, as it communicated only with the atrium, through a passage on the south side of it (fig. 7). The evidence for the existence of an openair court immediately to the west of the triconch is another important argument, as triclinia were usually situated on one side of a peristyle.47 Respectively, the two adjoining rooms attached to the north side of the building are considered a direct indication of its use as a triclinium. Similar spaces are usually found around all forms of triclinia.48 They are mostly identified as auxiliary rooms, mainly used by servants during banquets.49 In particular, the triclinia in the Triconch Palace at Butrint50 and the so-called Episcopal Palace of Aphrodisias51 communicated directly with such rooms through door openings along the apses.52 A similar opening in the north apse connected the triconch of Dion with its southern auxiliary room (fig. 6–7).

Fig. 8. The small triconch at the archaeological site on the base of the new Acropolis Museum in Athens.

as a hall for the official reception of guests and clergy by the bishop.35 Nevertheless, the great similarities between this building and one of the city’s churches, the Kruze Basilica,36 with which it shares an almost identical trefoil layout, as well as the absence of adequate excavation data (it has been preserved in extremely poor condition), may hinder the acceptance of this view. Furthermore, written sources inform us that in the middle of the fifth century, Bishop Neon (450–452) built within the episcopium of Ravenna a triclinium with five apses.37 In the early sixth century, Bishop Peter II added a domus tricoli, which, according to one interpretation, would have been a threeapsed hall.38 Finally, it is worth mentioning two late examples of triconch triclinia from the Episcopal Palace at the Lateran, dating from the ninth century: the triclinium founded by Pope Leo III39 (795–816) and that of Pope Gregory IV (828–844).40

Taking into consideration all the above arguments, we conclude, based on the information available so far, that the identification of the Dion triconch as a triclinium of the bishop’s residence is the most likely possibility, though this needs to be tested through further excavation. Nevertheless, it gives a new turn to the issue of the overall shape and organisation of the episcopal complex of the city of Dion and, in the event of confirmation, will enrich the very limited corpus of similar buildings in Greece and the Balkans. In addition, it could significantly contribute to the study of the episcopal residences of Late Antiquity in Greece,53 which, despite the progress achieved in recent Morvillez 1996, p. 158; Mulvin 2002, p. 119; Said 2002, p. 165 and n. 471. The use of these buildings as triclinia is evinced by the presence of semi-circular panels of plain mosaic on the apse floor, marking the position where the stibadia were placed (Morvillez 1995, p. 24). 44 This triconch is fitted with apses 3.50 m wide and 2.50 m deep (Ellis 1995, p. 173; Mulvin 2002, 49). 45 Rossiter 1991, p. 203; Morvillez 1995, pp. 23–24; Polci 2003, pp. 81–82; Hudson 2006, pp. 230–236, fig. 174–177; Bowden 2011, p. 296. 46 Approximately 3.50 m wide and 2.30 m deep. These dimensions would also allow the existence of a narrow free space 0.50 m in width between the couch and the wall of the apse, used mainly by servants to offer the guests food and wine or to wash hands between meals. 47 McKay 1975, p. 220; Özgenel 2007, pp. 269–270; Karagianni 2012, p. 73. 48 Petridis 2008, pp. 250–251. 49 Apart from having a support role in association with reception functions, they could also be used for housing guests; see Morvillez 2002, p. 234ff. 50 Bowden 2011, fig. 8.14; Bowden et al. 2011, p. 47. 51 Özgenel 2007, p. 254; Berenfeld 2009, pp. 205, 214, fig. 6. 52 For a similar articulation at the single-apsed triclinium of a house in Thessaloniki, cf. Karydas 2009, pp. 132–133. 53 For the most likely cases among the various buildings in Greece that have been identified, from time to time, as Late Antique episcopia and for relevant bibliography, cf. Baldini 2015, p. 154ff. 43

Triconch triclinia were very diverse in size, generally corresponding to the overall scale of the house.41 Leaving aside the monumental structures of the largest aristocratic houses and imperial palaces, the majority were equipped with apses as much as 6 or 7 m wide.42 There were also several examples of smaller dimensions, while in some Klenina 2008, p. 66ff.; 2013, pp. 919–920, fig. 5. See Biernacki et al. (ed.) 2004, pp. 195, 201. 37 Rizzardi 2007. 38 Miller 2000, pp. 30–31. 39 Herklotz 1995, p. 175 n. 1, for relevant bibliography. 40 Miller 2000, p. 59; Liverani 2012, pp. 127–128. 41 Morvillez 1995, pp. 20–21. 42 Karydas 2009 p. 139; Hudson 2010, pp. 688–689. 35 36

196

Episcopal Basilica of Dion, Greece years, still depends mostly on particularly fragmentary and problematic information.

Bardill J., 1997. ‘The Palace of Lausus and nearby monuments in Constantinople: A topographical study’, in AJA, 101/1, pp. 67–95.

Bibliography

Barrate F., 2012. ‘Les domus ecclesiae: le cas de l’Afrique romaine vandale et byzantine’, in Balcon-Berry et al. 2012, pp. 173–184.

Abbreviations: ΑΔ = Αρχαιολογικόν Δελτίον.

Berenfeld M. L., 2002. The Bishop’s Palace at Aphrodisias. A Late Roman Townhouse in the Center of the City, PhD dissertation, New York University.

AJA = American Journal of Archaeology. ΔΧΑΕ = Δελτίον της Χριστιανικής Αρχαιολογικής Εταιρείας.

Berenfeld M. L., 2009. ‘The Triconch House and the predecessors of the Bishop’s Palace at Aphrodisias’, in AJA, 113/2, pp. 203–229.

JRA = Journal of Roman Archaeology. ΠΑΕ = Πρακτικά της εν Aθήναις Aρχαιολογικής Eταιρείας.

Biernacki A. B., Klenina E. J., Ryžov St. G. (eds.), 2004. Architektura wczesnobizantyjskich budowli sakralnych Chersonezu Taurydzkiego – The Early-Byzantine Church Architecture of Chersonesus Taurica, I, Poznań.

Aleksova B., 1986. ‘The Early Christian basilicas at Stobi’, in R. Fariolli (ed.), XXXIII Corso di cultura sull’arte ravennate e bizantina. Seminario internazionale di studi ‘La Macedonia iugoslavia’, Ravenna 15–22 marzo 1986, Ravenna, 1986, pp. 13–81.

Bitrakova-Grozdanova V., 1975. Monuments Paléochrétiens de la Région d’Ohrid – Starohristijanski spomenici vo Ohridsko, Ohrid.

Balcon-Berry S., Baratte F., Caillet J.-P., Sandron D. (eds.), 2012. Des domus ecclesiae aux palais épiscopaux. Actes du colloque tenu à Autun du 26 au 28 novembre 2009, Turnhout (Bibliothèque de l’Antiquité tardive, 23).

Bowden W., 2011. ‘The domus and Triconch Palace as aristocratic residences’, in Bowden, Hodges 2011, pp. 277–302. Bowden W., Francis K., Gilkes O., Lako K., 2011. ‘The domus and Triconch Palace’, in Bowden, Hodges 2011, pp. 11–55.

Baldini I., 2014. ‘Palatia, praetoria ed episcopia: alcune osservazioni pensabene’, in P. Pensabene, C. Sfameni (eds.), 2014. La villa restaurata e i nuovi studi sull’edilizia tardoantica. Atti del convegno internazionale del Centro interuniversitario di studi sull’edilizia abitativa tardoantica nel Mediterraneo, CISEM, Piazza Armerina 7–10 novembre 2012, Bari (Insulae Diomedeae, 23), pp. 163–170.

Bowden W., Hodges R. (ed.), 2011. Butrint 3. Excavations at the Triconch Palace, Oxford/Oakville (Butrint Archaeological Monographs, 3). Bowes, K., Ghisleni M., La Torre F., Vaccaro E., 2011. ‘Preliminary report on Sofiana/mansion Philosophiana in the hinterland of Piazza Armerina’, in JRA, 24, pp. 423–449.

Baldini I., 2015. ‘Il complesso episcopale. Sviluppo architettonico e modelli’, in I. Baldini, M. Livadiotii (eds.), 2015. Archeologia protobizantina a Kos. La città e il complesso episcopale, Bologna (Collana DiSCi, 6), pp. 154–172.

Brands G., Rutgers L. V., 2005. ‘Κατοίκηση στην ύστερη αρχαιότητα’, in W. Hoepfner (ed.), Ιστορία της Κατοικίας 5000 π.Χ.–500 μ.Χ. Προϊστορία – Πρώιμη ιστορία – Αρχαιότητα, Thessaloniki, pp. 767–827.

Baldini I., Mazzilli G., 2011. ‘Le fasi e la tipologia della basilica’, in I. Baldini, M. Livadiotti (eds.), Archeologia Protobizantina a Kos: La Basilica di S. Gabriele (Studi e Scavi, nuova serie 28), Bologna, pp. 151–176.

Caillet J.-P., 2012. ‘Le cas de Caričin Grad (Serbie) et le problématique de l’identification de certains “palais épiscopaux” de l’Antiquité Tardive’, in Balcon-Berry et al. 2012, pp. 147–162.

Baldini Lippolis I., 1994. ‘Case e palazzi a Constantinopoli tra IV e VI secolo’, in XLI Corso di cultura sull’arte ravennate e bizantina. Seminario internazionale di studi su: Ravenna, Costantinopoli, Vicino Oriente, Ravenna, 12–16 settembre 1994. In memoria del Prof. F. W. Deichmann, Ravenna, 1994, pp. 279–311.

Ceylan B., 2007. ‘Episkopeia in Asia Minor’, in Lavan, Özgenel, Sarantis 2007, pp. 169–194. Drakoulis D. P., 2012. ‘The historical–geographic dimension of Macedonia in Late Antiquity: administrative and spatial transformations’, in D. P. Drakoulis, G. P. Tsotsos (eds.), Ιστορική γεωγραφία της Ελλάδος και της ανατολικής μεσογείου, Thessaloniki, pp. 79–106.

Baldini Lippolis I., 2001. La domus tardoantica nelle città del Mediterraneo. Forme e rappresentazioni dello spazio domestico nelle città del Mediterraneo, Imola (Università degli studi di Bologna. Studi e scavi, 17).

Dunbabin K. M. D., 1991. ‘Triclinium and stibadium’, in Slater 1991, pp. 121–148.

Baldini Lippolis I., 2005. L’architettura residenziale nelle città tardoantiche, Rome (Le bussole. Archeologia, 211).

Dunbabin K. M. D., 1996. ‘Convivial spaces: Dining and entertainment in the Roman villa’, in JRA, 9, pp. 66–80. 197

Kyriakos Fragoulis Dunbabin K. M. D., 2003. The Roman Banquet. Images of Conviviality, Cambridge.

the Age of Absolutism, Berlin/Boston (Urban Spaces, 4), pp. 53–62.

Eleftheratou St. (ed.), 2006. Το Μουσείο και η Ανασκαφή. Ευρήματα από τον χώρο ανέγερσης του νέου μουσείου της Ακρόπολης, Athens.

Hudson N. F., 2006. Dining in the Late Roman East, PhD dissertation, University of Minnesota. Hudson N. F., 2010. ‘Changing places: The archaeology of the Roman convivium’, in AJA, 114, pp. 664–695.

Eleftheratou St. 2009. ‘Η αρχαιολογική ανασκαφή στη βάση του Μουσείου’, in Ανθέμιον, 20, pp. 6–10.

Karagianni F., 2012. ‘The urban house in Macedonia in Late Antiquity (4th–6th c.)’, in M. Tiverios, P. Nigdelis, P. Adam-Veleni (eds.), Θρεπτήρια. Μελέτες για την αρχαία Μακεδονία – Threpteria. Studies on Ancient Macedonia, Thessaloniki, pp. 65–98.

Ellis S. P., 1988. ‘The end of the Roman house’, in AJA, 92/4, pp. 565–576. Ellis S. P., 1991. ‘Power, architecture, and décor: How the late Roman aristocrat appeared to his guests’, in E. K. Gazda (ed.), Roman Art in the Private Sphere. New Perspectives on the Architecture and Decor of the Domus, Villa and Insula, Ann Arbor, pp. 117–134.

Karydas N., 2009. ‘Παλαιοχριστιανικές οικίες με τρικλίνιο στη Θεσσαλονίκη μέρος ΙΙ: δέκα χρόνια μετά’, in P. Adam-Veleni, K. Tzianavari (ed.) 20 Χρόνια. Το Αρχαιολογικό Έργο στη Μακεδονία και τη Θράκη, Επετειακός Τόμος, Thessaloniki, pp. 127–141.

Ellis S. P., 1995. ‘Classical reception rooms in RomanoBritish houses’, in Britannia, 26, pp. 163–178.

Klenina E. J., 2013. ‘The bishopric and Early Chistian architecture in Chersonesos in Taurica’, in O. Brandt, S. Cresci, J. López Quiroga, C. Pappalardo (eds.), Acta XV Congressus internationalis archaeologiae Christianae, Toleti (8–12.9.2008). Episcopus, civitas, territorium, Vatican City (Studi di Antichità cristiana, 65), Città del Vaticano, pp. 915–928.

Ellis S. P., 2007. ‘Late Antique housing and the uses of residential buildings: An overview’, in Lavan, Özgenel, Sarantis 2007, pp. 1–22. Février P.-A., 1972. ‘Les sources épigraphiques et archéologiques et l’histoire religieuse des provinces orientales de l’Afrique antique’, in XIX Corso di cultura sull’arte ravennate e bizantina, Ravenna, 16–29 Aprile 1972, Ravenna, p. 131–158.

Kurtović-Folić N., 1997. ‘Triconch – Its origin and place in the development of architectural forms’, in Facta Universitatis. Series: Architecture and Civil Engineering, 1/4, pp. 473–482.

Fragoulis K., 2014. Η επισκοπική βασιλική της πρωτοβυζαντινής πόλης του Δίου – The Episcopal Basilica of the Early Byzantine City of Dion, Master’s thesis, Αριστοτέλειο Πανεπιστήμιο Θεσσαλονίκης: http://ikee.lib.auth.gr/record/286906.

Laskaris N. G., 2000. Monuments funéraires paléochrétiens (et Byzantins) de Grèce, Athens. Lavan L., Ozgenel L., Sarantis A. (eds.), 2007. Housing in Late Antiquity. From Palaces to Shops, Leiden/Boston (Late Antique Archaeology, 3/2).

Gattiglia A., 1998. ‘Architettura simbolica di età giustinianea nei Balcani: la tricora’, in N. Cambi, E. Marin (eds.), Radovi XIII međunarodnog kongresa za starokršćansku arheoligiju, Split-Poreč, 25.9.– 1.10.1994 – Acta XIII congressus internationalis archaelogiae christianae, Split-Poreč, 25.9.–1.10.1994, II, Vatican City/Split (Studi di Antichità cristiana, 54/2; Vjesnik za arheologiju i historiju dalmatinsku. Suplement, 88), pp. 189–206.

Lavin I., 1962. ‘The House of the Lord: Aspects of the role of palace triclinia in the architecture of Late Antiquity and the Early Middle Ages’, in The Art Bulletin, 44/1, pp. 1–27. Liverani P., 2012. ‘L’episcopio lateranense dalle origini all’alto Medioevo’, in Balcon-Berry et al. 2012, pp. 119–131.

Gollubits J., 2008. Studien zur Baugeschichte der EufrasiusBasilika in Poreč, Master’s thesis, Universität Wien.

McKay A. G., 1975. Houses, Villas and Palaces in the Roman World, London/Ithaca/New York (Aspects of Greek and Roman Life).

Habermehl D., 2013. Settling in a Changing World. Villa Development in the Northern Provinces of the Roman Empire, Amsterdam (Amsterdam Archaeological Studies, 19).

Mentzos Α., 1997. ‘Η εκκλησιαστική αρχιτεκτονική της Πιερίας στην πρώιμη βυζαντινή περίοδο’, in Α. Agelopoulos, Α. Ε. Karathanasis (eds.), Η Πιερία στα Βυζαντινά και Νεώτερα Χρόνια. Επιστημονικό Συνέδριο, Thessaloniki (Εκδόσεις Ιδρύματος Εθνικού και Θρησκευτικού Προβληματισμού, 5), pp. 153–167.

Herklotz I., 1995. ‘Francesco Barberini, Nicolὸ Alemanni, and the Lateran Triclinium of Leo III: An Episode in Restoration and Seicento Medieval Studies’, in Memoirs of the American Academy in Rome, 40, pp. 175–196.

Mentzos A., 2001. ‘Early Byzantine ecclesiastical architecture in Pieria’, in J. Burke, R. Scott (eds.), Byzantine Macedonia. Art, Architecture, Music and Hagiography. Papers from the Melbourne Conference, July 1995, Melbourne, pp. 7–12.

Herrin J., 2015. ‘The Imperial Palace of Ravenna’, in M. Featherstone, J.-M. Spieser, G. Tanman, U. Wulf-Rheidt (eds.), The Emperor’s House. Palaces from Augustus to

198

Episcopal Basilica of Dion, Greece Miller M. C., 2000. The Bishop’s Palace. Architecture and Authority in Medieval Italy. Ithaca/London (Conjunctions of Religion & Power in the Medieval Past).

religiosa occidentale ed orientale dal tardoantico all’alto medioevo’, in J.-P. Caillet, M. Sot (eds.), L’audience. Rituels et cadres spatiaux dans l’Antiquité et le haut Moyen Âge, Paris (Textes, images et monuments de l’Antiquité au haut Moyen Âge), pp. 221–239.

Morvillez E., 1995. ‘Les salles de réception triconques dans l’architecture domestique de l’Antiquité tardive en Occident’, in Histoire de l’art, 31, pp. 15–26.

Rossiter J., 1991. ‘Convivium and villa in Late Antiquity’, in Slater 1991, pp. 199–214.

Morvillez E., 1996. ‘Sur les installations de lits de table en sigma dans l’architecture domestique du Haut et du Bas-Empire’, in Pallas, 44, pp. 119–158.

Said W. B., 2002. Τυπολογία της κατοικίας της ρωμαϊκής περιόδου στην ανατολική Μεσόγειο, PhD dissertation, Εθνικό Μετσόβιο Πολυτεχνείο.

Morvillez E., 2002. ‘Les appartements d’hôtes dans les demeures de l’Antiquité tardive: mode occidental et mode orientale’, in Pallas, 60, pp. 231–245

Sfameni C., 2004. ‘Residential villas in late antique Italy: Continuity and change’, in W. Bowden, L. Lavan, C. Machado (eds.), Recent Research on the Late Antique Countryside, Leiden/Boston (Late Antique Archaeology, 2), pp. 335–375.

Mulvin L., 2002. Late Roman Villas in the Danube-Balkan Region, Oxford (BAR International Series, 1064).

Slater W. J. (ed.), 1991. Dining in a Classical Context, Ann Arbor.

Özgenel L., 2007. ‘Public use and privacy in late antique houses in Asia Minor: The architecture of spatial control’, in Lavan, Özgenel, Sarantis 2007, pp. 239–280.

Pantermalis, D., 1993. ‘Η ανασκαφή του Δίου κατά το 1993 και η χάλκινη διόπτρα’, in Το Αρχαιολογικό Έργο στη Μακεδονία και τη Θράκη, 7, pp. 195–199.

Sodini J.-P., 1989. ‘Les groupes épiscopaux de Turquie (à l’exception de la Cilicie)’, in N. Duval (ed.), 1989. Actes du XIe Congrès international d’archéologie chrétienne, Lyon, Vienne, Grenoble, Genève et Aoste (21–28 septembre 1986), Rome/Vatican City (Collection de l’École française de Rome, 123; Studi di Antichità cristiana, 41), pp. 405–427.

Pantermalis D., Eleftheratou St., Vlassopoulou Ch., 2016. Μουσείο Ακρόπολης. Οδηγός, Athens.

Sodini J.-P., 2009. ‘Rhodes et Lycie: le tétraconque d’Arnitha’, in ΔΧΑΕ, 30, pp. 19–24.

Pelekanidis St., 1964. ‘Ἀνασκαφικαὶ ἔρευναι εἰς τὴν χριστιανικὴν βασιλικὴν Δίου Πιερίας’, in ΑΔ, 19 – Χρονικά-Μέρος Β΄3, p. 384.

Sotiriadis G., 1928. ‘Ἀνασκαφὴ Δίου Μακεδονίας’, in ΠΑΕ, 83, pp. 59–95.

Pantermalis D., 1984. ‘Ανασκαφή Δίου’, in ΠΑΕ, 139, pp. 71–74.

Sotiriadis G., 1929. ‘Ἀνασκαφὴ Δίου Μακεδονίας’, in ΠΑΕ, 84, pp. 69–82.

Pelekanidis St., 1965. ‘Ἀνασκαφικαὶ ἔρευναι εἰς Δῖον Πιερίας’, in ΑΔ, 20, Χρονικά-Μέρος Β΄3 – Χρονικά, pp. 477–479.

Sotiriadis G., 1930. ‘Ἀνασκαφαὶ Δίου Μακεδονίας’, in ΠΑΕ, 85, pp. 36–51.

Pelekanidis St., 1966. ‘Ἀνασκαφὴ βασιλικῆς Δίου’, in ΑΔ, 21, Χρονικά-Μέρος Β΄2, pp. 371–376.

Sotiriadis G., 1932a. ‘Τὸ μακεδονικὸν Δῖον καὶ ἡ ἐγκατάστασις τῶν Γότθων παρὰ τὸν Ὄλυμπον’, in Ἡμερολόγιον τῆς Μεγάλης Ἑλλάδος, 11, pp. 445–471.

Petridis P., 2008. ‘Observations sur les villes et les villas urbaines de la Grèce pendant l’Antiquité tardive’, in ΔΧΑΕ, 29, pp. 247–258.

Sotiriadis G., 1932b. ‘Ἡ Βασιλικὴ τοῦ Δίου’, in A. K. Orlandos (ed.), IIIème Congrès international des Études byzantines, Athènes 1930. Compte rendu, Athènes, pp. 251–252.

Polci B., 2003. ‘Some aspects of the transformation of the Roman domus between Late Antiquity and the Early Middle Ages’, in L. Lavan, W. Bowden (eds.), Theory and Practice in Late Antique Archaeology, Leiden/ Boston (Late Antique Archaeology, 1), pp. 79–109.

Sotiriou G. Α., 1929. ‘Αἱ παλαιοχριστιανικαὶ βασιλικαὶ τῆς Ἑλλάδος’, in Αρχαιολογική Εφημερίς, 68, pp. 161–248. Stollmayer I., 1999. ‘Spätantike Trikonchoskirchen – ein Baukonzept?’, in Jahrbuch für Antike und Christentum, 42, pp. 116–157.

Popović V., Duval N., Caillet J.-P., Reddé M., Baratte F., Jeremić M., 2010. ‘Le baptistère’, in N. Duval, V. Popović (eds.), Caričin Grad, III, Rome (Collection de l’École française de Rome, 75/3), pp. 201–263.

Varalis Y. D., 1999. ‘Deux églises à chœur triconque de l’Illyricum oriental. Observations sur leur type architectural’, in Bulletin de correspondance hellénique, 123/1, pp. 195–225.

Rapp C., 2005. Holy Bishops in Late Antiquity. The Nature of Christian Leadership in a Time of Transition, Berkeley/Los Angeles/London (The Transformation of the Classical Heritage, 37; The Joan Palevsky Imprint in Classical Literature). Rizzardi Cl., 2007. ‘Le sale di rappresentanza dell’episcopio di Ravenna nell’ambito dell’edilizia 199

15 Les espaces domestiques et économiques du groupe épiscopal protobyzantin de Byllis (Albanie) Nicolas Beaudry Université du Québec à Rimouski (Canada)

Pascale Chevalier Université Clermont Auvergne, UMR 6298–ArTeHiS (France) Dans son état du VIe siècle, le groupe épiscopal de Byllis comprenait un secteur où se concentraient les activités économiques et agricoles de l’évêché. Cette insula episcopalis a pris forme autour d’une cour formée par l’arasement, vers la fin du Ve siècle, de bâtiments qui bordaient au sud le decumanus desservant les annexes liturgiques de la cathédrale ; à leur place a été édifié un grand bâtiment pourvu d’un étage consacré à l’habitat. Du côté nord, la cour était bordée par un entrepôt et par un chai dont la façade prolongeait celle du groupe cathédral. Appartenant à l’ultime phase, un bâtiment irrégulier a fermé la cour au sud-ouest, ne ménageant qu’un passage vers la cathédrale. Voué au service de l’évêque, ce secteur constituait un ensemble urbain cohérent, dépendant de l’évêché mais distinct de l’espace liturgique, coupé de l’espace public mais possédant son propre accès direct au cardo longé par le rempart protobyzantin. In its sixth-century state, the episcopal complex of Byllis included an area where the bishop’s economic and agricultural activities were concentrated. This insula episcopalis took shape around a courtyard formed by the levelling, towards the end of the fifth century, of buildings on the south side of the decumanus that gave access to the liturgical spaces of the cathedral. In their place was built a large building with a floor dedicated to habitation. To the north, the courtyard was bordered by a warehouse and a winery whose facade extended that of the cathedral complex. Belonging to the last phase, an irregular building closed the courtyard to the southwest, leaving only a passage to the cathedral. Dedicated to the service of the bishop, this sector constituted a coherent urban unit, dependent on the episcopal see but distinct from the liturgical space, cut off from the public space but possessing its own direct access to the cardo bordering on the early Byzantine ramparts. Während des 6. Jhs. umfasste der Bischofskomplex von Byllis ein Gebiet, auf dem sich wirtschaftliche und landwirtschaftliche Aktivitäten konzentrierten. Bereits gegen Ende des 5. Jhs. formte sich die insula episcopalis um einen planierten Hof herum. Die Gebäude grenzten südlich an den decumanus, wodurch der direkte Zugang zu den liturgischen Räumen der Basilika möglich war. An ihrer Stelle wurde ein großes Gebäude mit Wohnbereich errichtet. Im Norden war der Hof durch einen Weinkeller und ein Warenhaus begrenzt, deren Fassaden den Kirchenkomplex erweiterten. Während der letzten Phase schloss ein Gebäude mit irregulärem Grundriss den Hof ab. Es blieb nur ein Durchgang zur Kirche hin. Dem bischöflichen Betrieb gewidmet, bildete dieser Teil eine zusammenhängende städtische Einheit, die vom Bistum abhängig, aber deutlich vom liturgischen und öffentlichen Bereich getrennt blieb und ihren eigenen direkten Zugang zum cardo besaß, der entlang des frühbyzantinischen Walles verlief. Nel suo stato del VI secolo, il gruppo episcopale di Byllis comprendeva un’area in cui si concentravano le attività economiche e agricole del vescovo. Questa insula episcopalis prese forma attorno a un cortile formato del livellamento, verso la fine del V secolo, di edifici che confinavano a sud con il decumanus che serviva gli annessi liturgici della cattedrale; al loro posto fu costruito un grande edificio con un piano dedicato alle abitazioni. Sul lato nord, il cortile era delimitato da un magazzino e da una cantina la cui facciata prolungava quella del complesso

201

Nicolas Beaudry & Pascale Chevalier della cattedrale. Un edificio di forma irregolare, appartenente all’ultima fase, ha chiuso il cortile a sud-ovest, risparmiando solo un passaggio verso la cattedrale. Dedicato al servizio del vescovo, questo settore costituiva un insieme urbano coerente, dipendente dal vescovato ma distinto dallo spazio liturgico, tagliato fuori dallo spazio pubblico ma dotato di un proprio accesso diretto al cardo lungo il muro protobizantino.

Les vestiges de la cité antique de Byllis occupent le sommet de la colline d’Hekal, en Albanie méridionale. Culminant à 524 m, la colline constitue l’extrémité orientale de la chaîne de Mallakastër, une région accidentée, mais dont les terrasses et les vallées sont propices à l’agriculture et à l’élevage. Plus basses à l’ouest, les autres collines dégagent la vue jusqu’à la mer et sur le cours inférieur de la Vjosa, l’antique Aôos. Fondée au IVe siècle av. J.-C., la capitale de la tribu illyrienne des Bylliones a connu un essor important à l’époque hellénistique.1 Intégrée à la province de Macédoine en 148 av. J.-C., puis colonie romaine après les Guerres civiles, elle assurait le contrôle de l’arrièrepays d’Apollonia et de l’itinéraire qui reliait la côte à l’Épire à la Macédoine par la vallée de l’Aôos.2 Siège épiscopal indépendant depuis le milieu du Ve siècle,3 elle est mentionnée dans le premier tiers du VIe parmi les plus importantes cités de la province de Nouvelle Épire, après Dyrrachion (Durrës), Scampis (Elbasan) et Apollonia mais devant six autres villes.4 Malgré un dernier effort défensif sous Justinien, Byllis a été désertée au début du VIIe siècle devant la pression des migrations slaves. Dépourvu de sources d’eau, le site n’a jamais réoccupé, sinon brièvement par des troupes serbes au XIVe siècle et austro-hongroises pendant la Première Guerre mondiale. Il a essentiellement servi de pacage aux bêtes des habitants du village d’Hekal, établi sur une pente de la colline ; la récupération de matériaux et des labours récents n’ont causé que des dommages mineurs aux vestiges de la cité protobyzantine. Après une brève campagne menée par Camilo Prashniker en 1917-1918,5 Byllis a fait l’objet de fouilles systématiques menées de 1978 à 1991 par le Centre de recherches archéologiques de Tirana, devenu l’Institut archéologique de Tirana.6 Privilégiant les origines illyriennes de la cité et son urbanisme à l’époque hellénistique, les travaux de Neritan Ceka se sont concentrés sur les vestiges classiques, notamment le mur d’enceinte, les monuments civiques,

Fig. 1. Byllis : plan du site (M. Wurch-Kozelj).

l’habitat et les nécropoles, donnant à la partie sud du site sa physionomie actuelle (fig. 1).7 Menés en parallèle, les travaux de Skënder Muçaj se sont intéressés à l’Antiquité tardive et ont porté sur les remparts, les églises, ainsi que d’autres bâtiments publics et privés. L’interruption prématurée des travaux ayant laissé ces vestiges pratiquement inédits, une Mission archéologique francoalbanaise a été créée en 1998 sous l’égide de l’Institut archéologique de Tirana et de l’École française d’Athènes avec pour objectif d’assurer l’étude et la publication des structures de l’Antiquité tardive dégagées avant 1991 et dont la fouille était considérée comme terminée. À partir

Ceka 1984; Ceka, Muçaj 2005; Ceka 2012. Pline l’Ancien, Naturalis Historia, 4.10; Ceka 1987; Shpuza 2006; Deniaux 2008. 3 Alors qu’Apollonia et Byllis sont représentés par un même évêque au concile d’Éphèse de 431, deux évêques différents figurent en 458 parmi les signataires d’une lettre adressée à l’empereur Léon Ier par le synode épiscopal d’Epirus Nova (Farlati, Coleti 1817, p. 396; de Thallóczy et al. 1913, pp. 3-4, nos 14 et 2). Cf. Haxhimihali 2004. 4 Hieroclès, Synekdèmos 653.1-654.1b (éd. Honigmann 1939, p. 20). Cf. Haxhimihali 2004. 5 Praschniker, Schöber 1919; Praschniker 1922. 6 Ces travaux ont donné lieu à des comptes rendus annuels publiés dans la revue Iliria ; quand ils ont pris fin, les vestiges dégagés représentaient environ 25 % de la cité intra muros. 1 2

Cf. notamment Ceka 1984; 1986; 1990; 1992; Ceka, Muçaj 2005, pp. 24-32; Ceka, Ceka 2018.

7

202

Les espaces domestiques et économiques du groupe épiscopal protobyzantin de Byllis (Albanie)

Fig. 2. Vue aérienne de la cathédrale et du quartier épiscopal vers le nord, mai 2007 (A. Islami).

de 2000 et jusqu’en 2012, elle s’est aussi consacrée à la fouille et à l’étude d’un quartier attenant à la cathédrale paléochrétienne.8

en évidence dans le quartier de la cathédrale (fig. 2). Cet article présente ce qui s’est révélé être le cœur économique du domaine épiscopal de Byllis et le cadre de vie d’une communauté au service de l’évêque.

Aujourd’hui un parc archéologique protégé, Byllis est un observatoire privilégié des bouleversements et des processus de mutation à l’œuvre dans les villes tardoantiques des Balkans occidentaux et d’ailleurs : l’abandon des monuments civiques, l’implantation d’églises, l’empiètement de l’espace privé sur l’espace public, le remaniement des dispositifs défensifs, la transformation de l’habitat urbain, la ruralisation et la rétractation de la cité intra muros, etc.9 Si ces processus sont observables à l’échelle de la cité,10 ils ont particulièrement été mis

Le contexte urbain La matrice urbaine de Byllis est l’héritage d’une longue histoire. La cité hellénistique s’étendait au sommet de la colline, de part et d’autre d’une crête calcaire, à l’intérieur d’une enceinte qui suit le contour des plateaux sommitaux en dessinant un vaste triangle de 30 ha (fig. 1). Le cœur civique de la cité était situé au sud-ouest, d’où il dominait la vallée de l’Aôos et les collines de Mallakastër jusqu’à la mer. Un théâtre, aménagé dans le flanc de la colline à son point le plus élevé, était prolongé vers l’ouest par une stoa qui bordait l’à-pic surplombant la vallée. Au nord du théâtre, deux stoai en équerre encadraient une grande agora, ouverte à l’ouest sur les gradins d’un stade dont la piste couvrait la voûte d’une immense citerne. À l’ouest de l’agora s’élevaient aussi un prytanée, un arsenal, ainsi qu’un gymnase converti en thermes à l’époque romaine.11 Le sud-est du plateau ne semble pas avoir été occupé, sinon par d’importantes carrières de calcaire.12 Le reste de la cité intra muros, consacré à l’habitat, était ordonné par un réseau urbain orthogonal comprenant de larges voies nordsud et des voies secondaires transversales plus étroites ; les

La mission archéologique franco-albanaise a été créée et codirigée jusqu’en 2003 par Skënder Muçaj et Jean-Pierre Sodini ; la co-direction française a été assumée à partir de 2004 par P. Chevalier. Les travaux dans le quartier épiscopal ont été dirigés par N. Beaudry. Le financement des travaux et des études a été porté principalement par l’École française d’Athènes et le Ministère français des Affaires étrangères et européennes, avec des contributions de l’Institut archéologique de Tirana, du Centre d’études albanologiques, de l’Institut des Monuments culturels de Tirana et du Fonds québécois pour la recherche sur la société et la culture. La Mission a publié des rapports dans la chronique annuelle du Bulletin de correspondance hellénique (Beaudry et al. 2002; 2003; 2007; 2008; 2009; 2013a; 2013b; et Chevalier et al. 2005; 2006; 2008) et, à partir de 2008, dans le Bulletin du Centre d’études médiévales d’Auxerre (Chevalier et al. 2008; 2009; et Beaudry et al. 2010; 2011; 2013c). Un premier volume sur les fortifications et les basiliques A, C et D est sous presse (Muçaj et al., sous presse) ; un deuxième portera sur la basilique B et ses annexes (Muçaj et al., à paraître). 9 Wiseman 1984; Bowden 2003, pp. 85-103; Kirilov 2005; Sodini 2007; 2013; Bavant 2007; Poulter 2007b, pp. 15-27; Snively 2009; etc. 10 Chevalier, Beaudry 2018. 8

11 12

203

Ceka, Muçaj, 2005; Ceka 2018. Kozelj, Wurch-Kozelj 2012, 2018.

Nicolas Beaudry & Pascale Chevalier insulae déterminées par ces voies (env. 134 × 69 m) étaient divisés en six lots d’environ 30 m de côté, séparés par des ruelles et destinés chacun à une maison à péristyle.13 Le cadre urbain ne semble pas avoir connu de changement majeur avant les invasions gothiques de la fin du IVe siècle.14

Le rempart hellénistique connaît une première campagne de réfection importante au Ve siècle, sans modification majeure de son tracé ni du dispositif défensif de la cité. Au VIe siècle, en revanche, le périmètre défensif est considérablement réduit par l’érection d’un nouveau mur qui traverse le site du nord au sud (fig. 1), et seule la partie occidentale des anciennes murailles est consolidée.18 Implanté sur la crête de la colline jusqu’à son sommet en surplomb du théâtre, ce nouveau rempart comprend aussi bien des pierres extraites de part et d’autre que des spolia, dont la plupart des gradins du théâtre. La ville intra muros, repliée sur la partie occidentale du plateau, est ainsi réduite des deux tiers (11 ha) ; le cardo qui suivait la crête dessert désormais la partie haute de la ville réduite et le rempart lui-même. L’évêché demeure dans le périmètre défendu (fig.  2, 3), tandis que deux églises (basiliques D et E) semblent desservir des quartiers encore habités, mais désormais extra muros. Quatre inscriptions retrouvées à Ballsh célèbrent le maître d’œuvre de ces travaux, Viktôrinos,19 agissant sur ordre de Justinien et connu aussi par une dédicace à la

Le paysage monumental de Byllis est profondément remanié au Ve siècle par la christianisation de la cité. Cinq églises connues et fouillées (basiliques A à E, fig. 1) ont été implantées dans le tissu urbain au cours des Ve et VIe siècles aux dépens d’îlots d’habitats ou de bâtiments publics. Toutes sont des basiliques orientées à trois nefs, couvertes de charpente, pavées de mosaïques parfois figurées15 et dont le décor présente une remarquable unité de style16. On ignore si la stoa nord de l’ancienne agora était déjà inutilisée ou si elle a été désaffectée pour laisser la place à la basilique A, mais l’agora cesse manifestement d’être le centre politique de la cité tandis que la cathédrale (basilique B)17 et l’évêché s’implantent plus au nord, au cœur de l’habitat urbain et à l’écart des anciens monuments civiques.

Fig. 3. Vue aérienne de la cathédrale et du quartier épiscopal vers le sud-ouest, mai 2007 ; au premier plan à gauche, le rempart de Viktôrinos (A. Islami). 13 Ceka 1992; Ceka, Mucaj 2005, pp. 30-31; Cella, D’Ercole 2012 pp. 5860; Muçaj 2018, p. 454. 14 Muçaj et al. 2014; Muçaj 2018. 15 Muçaj 1993b; Muçaj, Raynaud 2005. 16 Muçaj 1993a; Chevalier et al. 2003; Muçaj 2004; Chevalier 2004; Sodini 2004; Vanderheyde 2004; Ceka, Muçaj 2005, pp. 63-79; Muçaj et al., sous presse. 17 La cathédrale a été reconnue à ses grandes dimensions, à son baptistère et au vaste complexe qui lui est associé.

Muçaj 1990; 2008, pp. 381-382; Muçaj et al., sous presse. Une première inscription, aujourd’hui perdue, a été publiée par Praschniker (1922, col.  194-195 et fig.  111) ; les trois autres ont été découvertes en 1983 par Skënder Anamali (1987).

18 19

204

Les espaces domestiques et économiques du groupe épiscopal protobyzantin de Byllis (Albanie) forteresse de Corinthe20 d’après le dossier épigraphique, ils auraient été menés entre 548 et 554, peu après les premières invasions slaves21.

magasin (fig. 4, no 33) qui contenait plus d’une quinzaine de pithoi étanchés au bitume.24 À partir du tournant du VIe siècle, l’évêché s’approprie cette voie pour en faire à la fois l’axe et le pivot de son expansion dans la cité. Un nouveau quartier prend ainsi forme vers l’est et vers le sud, à la faveur de l’arasement et du remblaiement d’habitats détruits vers la fin du Ve siècle. Jusqu’à cet épisode de destruction, probablement violente, les bâtiments du quartier ne semblaient pas empiéter de façon significative sur la voie publique et l’alignement de leurs murs restait celui de la grille urbaine hellénistique. Une construction abritant un chai prolonge vers l’est la façade méridionale du groupe cathédral (fig. 4, N2 à N6), mais face à lui, un grand bâtiment (K-P) pourvu d’un étage est élevé en retrait de l’emprise du decumanus et dans un alignement nettement différent. Cet édifice ferme au sud et domine par son volume un espace qui deviendra une grande cour (L) ; le décor crucifère des fenêtres de son étage répond à celui du groupe cathédral et affirme l’autorité de l’évêché sur le quartier (fig. 5).25

Le quartier épiscopal L’insula episcopalis a été implantée dans le quadrant nord-ouest du plateau, en forte pente vers l’ouest, dans un quartier d’habitat hellénistique remanié à l’époque romaine (fig. 4).22 Elle comprend la cathédrale, une grande basilique pourvue d’une abside saillante, d’un transept bas et d’un narthex.23 Construite au Ve siècle, elle a été prolongée à l’ouest d’un exonarthex surmonté d’une tribune et des trois portiques d’un atrium qui, implantés dans la pente et en contrebas, contribuaient à la monumentalité de la façade occidentale. Le complexe de la cathédrale est bordé au sud par un long passage, un decumanus hérité de la voirie hellénistique, à partir duquel des portes et des couloirs desservent l’atrium, l’exonarthex, un ensemble baptismal et ses latrines, une petite cour pourvue d’une citerne et, à la hauteur du transept et de l’abside, un

Fig. 4. Plan schématique de la cathédrale et du quartier épiscopal (adapté de T. Kozelj et M. Wurch-Kozelj). Gregory 1993, pp. 136-140; Feissel 2000, p. 92. Feissel 1988. Byllis pourrait être Bulpiansus, qui figure chez Procope en tête de liste des fortifications de Nouvelle Épire (De aedificiis 4.4; cf. Haxhimihali 2004, p. 465). Les inscriptions n’ayant pas été trouvées in situ, W. Bowden (2006) n’exclut pas qu’elles concernent une réfection tardive du mur et non sa construction. Les analyses du liant de la maçonnerie du rempart de Viktôrinos attestent cependant qu’il correspond à l’une des phases les plus tardives de l’aménagement de la basilique B au VIe siècle (Beaudry et al. 2013b, pp. 924-925; 2013c, § 25). 22 Beaudry et al. 2013a, pp. 726-728. 23 Cf. Muçaj 1986; Ceka, Muçaj 2005, pp. 68-75; Beaudry et al. 2002, pp. 666-669; 2003, p. 630-636; 2004, pp. 1190-1203; Muçaj et al., à paraître. 20 21

24 Le bitume, qui servait également à étanchéifier les cruches et les amphores, était préféré à la poix en raison de la proximité de gisements d’asphalte d’excellente qualité dans les vallées de la Vjosë et de la Gjanicë. Cf. Cerova et al. 2005; Morris 2006. 25 Beaudry, Chevalier 2014, p. 209, fig. 4.

205

Nicolas Beaudry & Pascale Chevalier Au sud, un axe de circulation transversal (N6-P2) est fermé par un mur coudé (fig. 4). L’implantation diagonale du mur dans sa partie orientale semble chercher à relier deux points au moindre coût : à l’est, en contrehaut, l’embouchure d’une grande citerne taillée dans le rocher et des bâtiments en façade du cardo ; à l’opposé, la rencontre de murs arasés, dont l’un fournit une assise au mur coudé tout en dégageant un passage devant la façade méridionale du bâtiment (K-P).28 Bien qu’il ne soit pas attesté sur toute la longueur de la façade, ce mur constitue de facto la limite sud-est du domaine épiscopal dans son dernier état. Sa limite nord reste à identifier. Appartenant à l’ultime phase du quartier, un bâtiment à étage tardif (D) a été partiellement dégagé par S. Muçaj entre 1987 et 1991. Il s’appuie sur le grand bâtiment (K-P) et ferme la cour à l’ouest en empiétant largement sur l’emprise du decumanus (fig.  7).29 Une porte dans le prolongement de son mur oriental, dont subsiste un des montants rapportés, achève de contrôler l’accès au cœur économique de l’évêché et de distinguer un espace liturgique centré sur la cathédrale d’un espace profane centré sur la cour. Le cœur économique de l’évêché Ouvert de plain-pied sur la cour, le bâtiment du chai a livré tous les éléments d’une installation vinicole de la seconde moitié du VIe siècle, des fouloirs aux celliers (fig. 8).30 Dans son angle sud-est, un fouloir rectangulaire (fig. 9), peut-être associé à une presse, se déverse dans une baignoire trapézoïdale où le moût pouvait décanter avant d’être transvasé dans quatre pithoi étanchés au bitume, alignés contre le mur oriental. Dans l’angle nord-est, un fouloir plus profond, muni d’un escalier extérieur, se déverse en cascade dans un bassin de recueil (fig. 10) puis, à travers un mur de refend, dans une cuve de vinification (fig.  11) ; cette dernière, plus profonde, était pourvue d’un escalier intérieur31 et fermée par un couvercle de bois. Les autres pièces, dont l’une communique avec le magasin qui occupe l’angle sud-est du complexe de la cathédrale, servaient à l’entreposage du vin dans de grands conteneurs. Les fouloirs, le bassin de recueil et la cuve ont été rechapés à plusieurs reprises. Dans son dernier état, le bassin de recueil servait à la préparation du bitume qui étanchait les conteneurs.

Fig. 5. Meneau et imposte trouvés dans K1 et provenant de l’étage (E. Hobdari).

Plus haut vers l’est, le decumanus débouche toujours sur la voie qui suit la crête de la colline, mais dans son dernier état, l’accès au quartier épiscopal est contrôlé par un portail déterminé par quatre piédroits maçonnés (fig.  4, M16), dont l’un est lié au mur d’un bâtiment (M15) en façade du cardo.26 En contrebas, une construction barlongue (N7), adossée à la paroi du rocher et postérieure au portail, empiète largement sur l’emprise du decumanus pour former un vestibule (M14) à la suite du portail. Au sud-est, un mur de soutènement prolonge vers l’ouest la terrasse en façade du cardo (pièces M11 et M12) ; au pied de ce mur, dans la cour, une banquette maçonnée recueillait les eaux ruisselant des toitures en contrehaut pour les diriger vers l’embouchure d’une citerne. Des traces d’extraction de pierres, semblables à celles des carrières de la crête de la colline, indiquent qu’au moins certains des bâtiments à l’est de la cour sont postérieurs au rempart de Viktôrinos.27

Face au chai, le rez-de-chaussée du bâtiment méridional présentait à l’origine cinq pièces ouvertes sur la cour, grandes et profondes (fig.  12).32 La pièce occidentale (K5) a été fouillée par S. Muçaj entre 1987 et 1991; trois autres ont été fouillées à partir de 2001, la pièce centrale (K3) ayant été réservée. La pièce orientale (K1) était une étable, pourvue d’une mangeoire formée d’une Beaudry et al. 2013a, pp. 731-733. Beaudry et al. 2013b, pp. 919-924; 2013c, § 20. 30 Beaudry 2010; Chevalier et al. 2011, pp. 701-702; Beaudry et al. 2013d, pp. 1273-1274; Beaudry, Chevalier 2014, pp. 210-211. 31 Cf. Brun 1993. 32 Beaudry, Chevalier 2014, pp. 211-212. 28 29

Chevalier et al. 2005, pp. 1204-1207. Cf. Chevalier et al. 2006, pp. 843-845; Beaudry et al. 2007, p. 1102; Kozelj, Wurch-Kozelj 2012; 2018. 26 27

206

Les espaces domestiques et économiques du groupe épiscopal protobyzantin de Byllis (Albanie)

Fig. 6. L’accès oriental au quartier épiscopal, vers le sud-est, mai 2004 (N. Beaudry).

Fig. 7. Le bâtiment tardif D vu de la cour et le passage qui dessert les annexes de la cathédrale, mai 2012 (N. Beaudry).

207

Nicolas Beaudry & Pascale Chevalier

Fig. 8. Le chai, vers le nord, mai 2008 (N. Beaudry).

Fig. 9. Le fouloir sud-est et la baignoire dans laquelle il se déverse, vers le sud-est, mai 2008 (N. Beaudry).

208

Les espaces domestiques et économiques du groupe épiscopal protobyzantin de Byllis (Albanie)

Fig. 10. Le fouloir nord-est, à droite, et son bassin de recueil, vers le nord, mai 2008 (N. Beaudry).

banquette maçonnée dans laquelle étaient encastrées les cloisons des stalles (fig. 13).33 La construction barlongue (N7), pourvue d’une banquette comparable, pourrait aussi avoir été affectée à l’entretien des bêtes.34 La cour était en outre pourvue d’un abreuvoir taillé dans un grand bloc de calcaire, adossé contre le mur du magasin à côté d’une gouttière et de la bouche d’un égout (fig.  7). Les deux autres pièces fouillées après 2001 ont livré d’abondants restes fauniques dont certains, découpés et en connexion anatomique, suggèrent l’existence d’une boucherie. La pièce voisine (K2) a livré, remisée contre son mur de fond, la cuve de pierre d’un trapetum qui servait au broyage des olives avant leur pressurage (fig. 13, 14).35 Le bâtiment a aussi livré des poids de presse36 et des assemblages carpologiques dominés par l’olive (Olea Europaea) qui témoignent de l’existence d’une industrie oléicole. À moins que le fouloir sud-est du chai n’ait été pourvu d’une presse, les parties non fouillées du quartier épiscopal pourraient donc encore receler les vestiges d’un pressoir. Les espaces domestiques Le grand bâtiment présente au sud deux pièces d’angle (P3, P4) surmontées d’un étage, flanquant un espace Beaudry et al. 2002, p. 678. Chevalier et al. 2006, pp. 843-846. 35 Frankel 1999, type T331. Cf. Brun 2003, pp. 249-250; Thurmond 2006, pp. 87-92. Une meule hémisphérique de trapetum (orbis) conservée à Apollonia proviendrait de Byllis. 36 Beaudry, Chevalier 2014, p. 212, fig. 12. 33 34

Fig. 11. La cuve de vinification en cours de restauration, vers le sud, mai 2011 (N. Beaudry).

209

Nicolas Beaudry & Pascale Chevalier

Fig. 12. Le grand bâtiment K-P, vers l’ouest, juin 2006 (N. Beaudry).

Fig. 13. Les pièces K1 et K2 du grand bâtiment K-P, vers le sud, juin 2006 (N. Beaudry).

210

Les espaces domestiques et économiques du groupe épiscopal protobyzantin de Byllis (Albanie) central (P1, fig.  15), large, mais peu profond, accessible de plain-pied par une double baie centrale et probablement découvert à l’origine ; les volumes de cette façade devaient lui conférer une certaine monumentalité malgré l’absence de décor architectural. Un escalier dans l’angle nord-ouest de l’espace central donnait accès à l’étage : la fouille a livré des indices que ce dernier était consacré à l’habitat, dont des meules rotatives, des céramiques culinaires et des cruches, de petits objets personnels, ainsi que les briques d’un foyer et des tuiles de toiture percées de cheminées.37

la circulation vers l’escalier, la double baie centrale a été condamnée et remplacée par une porte dans l’angle sudouest, au plus loin de la source de pollution. C’est aussi par le sud-ouest que les eaux usées étaient évacuées vers un égout à l’extérieur du bâtiment. À l’est des foyers, une porte ouvrant sur la pièce d’angle a été bouchée et un mur de partition a été élevé sans fondation sur un niveau indistinct de sédiment noir. Ce nouveau mur détermine une petite annexe couverte (P1D, fig.  15) destinée à la conservation des denrées et à la préparation des aliments ; elle a livré quelques échantillons particulièrement riches en restes aviaires et ichtyologiques. La cuisine (P1) et son annexe (P1D) ont aussi livré un riche mobilier calcaire comprenant un bassin rectangulaire pourvu d’un exutoire, neuf tronçons de colonne remployés en mortier à pilon, en lavabo, en poids de presse (inachevé) ou, calés sur des fragments de tuileau, en piètement pour des plans de travail, ainsi que deux sekômata taillés dans des blocs de la frise du proskênon du théâtre (fig. 18, 19).41

L’espace central barlong (P1) était aussi une cuisine, dont l’aménagement a été considérablement remanié tout au long du VIe siècle. Elle était pourvue d’une batterie de foyers38 adossés au mur de fond, périodiquement remis en état ou arasés pour être remplacés (fig.  17)39. Autour d’eux se sont accumulées plusieurs dizaines de centimètres d’un sédiment noir et gras, particulièrement riche en restes botaniques carbonisés ainsi qu’en restes fauniques brûlés et fragmentés par le piétinement ; son mobilier céramique est caractérisé par la rareté du matériel résiduel, par la forte proportion des céramiques culinaires et par la sous-représentation des fonds, évacués avec les vidanges de foyer.40 L’accumulation de ce sédiment gênant

L’édifice tardif (D) a été implanté en équerre autour de l’angle nord-ouest du grand bâtiment (K-P), après l’obturation des portes de ses deux pièces occidentales (K5, K4) et le surhaussement de leurs sols42 ; son plan, irrégulier

Fig. 14. Trapetum (inv. 1837) contre le mur de fond de la pièce K2, juin 2006 (N. Beaudry). Chevalier et al. 2004, pp. 1210-1211, fig. 16. C’est le cas aussi dans le quartier épiscopal de Novae (Bulgarie). Cf. Biernacki 2005. 39 Beaudry et al. 2007, pp. 1103-1105 et 1108-1109; 2008, pp. 930-931. 40 Beaudry et al. 2007, pp. 1108-1109; 2008, pp. 944-954; 2009, pp. 749750; 2013c. 37 38

Pitarakis 2012, pp. 411-412. La pièce occidentale (K5) n’a plus été accessible que par le haut, si elle n’était pas simplement condamnée, tandis qu’un accès a été aménagé pour sa voisine à partir de la cuisine (P1). 41 42

211

Nicolas Beaudry & Pascale Chevalier

Fig. 15. L’espace P1 vers le nord-est, juin 2005 ; au premier plan, une pièce P1D déterminée par un mur tardif (N. Beaudry).

Fig. 16. Meule rotative (inv. 418) trouvée dans K1 et provenant de l’étage (N. Beaudry).

et remanié à plusieurs reprises, cherche à réconcilier l’alignement du bâtiment voisin avec l’implantation orthogonale du decumanus et des annexes de la cathédrale (fig.  4, 5, 7). Son étage était pavé d’une mosaïque polychrome grossière (fig.  20) et ses fenêtres affichaient un décor crucifère proche de celui du grand bâtiment (KP), ce qui devait lui conférer une certaine dignité malgré

la maladresse de son plan et des maçonneries montées hâtivement à la terre.43 Ouvert sur le passage qui dessert les annexes de la cathédrale, mais présentant à la cour un mur aveugle, le nouvel édifice appartient à l’espace liturgique plus qu’à l’espace profane de la cour, ce qui 43

212

Beaudry et al. 2013b, pp. 919-924; 2013c, § 12-20.

Les espaces domestiques et économiques du groupe épiscopal protobyzantin de Byllis (Albanie)

Fig. 17. Batterie de foyers dans l’espace P1, vers le nord, mai 2007 (N. Beaudry).

Fig. 18. Sekôma (inv. 764) provenant de l’espace P1 (N. Beaudry).

213

Nicolas Beaudry & Pascale Chevalier

Fig. 19. Sekôma (inv. 1070) provenant de la pièce P1D (N. Beaudry).

la cathédrale invite à les chercher plutôt au nord-est ou à l’est, ou encore au sud si sa résidence a été déplacée à la faveur de l’expansion du domaine épiscopal dans cette direction ; dans ce cas, l’édifice tardif (D) pourrait y être associé et elle pourrait s’être étendue à l’étage du grand bâtiment (K-P) de la cour. Conclusions Par la cour transitaient les matières premières et les produits des industries vinicole et oléicole de l’évêché, leurs produits, ainsi que toutes les denrées qui séjournaient dans les magasins de l’évêché ; elle permettait aussi l’entretien et la circulation des bêtes. Autour d’elle a pris forme un ensemble urbain cohérent, dépendant de l’évêché, mais distinct de l’espace liturgique, coupé du domaine public, mais bénéficiant, à l’opposé de la façade monumentale de la cathédrale, d’un accès direct au cardo du rempart. Son parallèle le plus proche, entièrement fouillé et plus riche de certains composants, est sans doute celui qu’offre le groupe épiscopal de Sidi Jdidi (Tunisie) dans son état du Ve siècle : entre les deux basiliques cathédrales, un îlot concentré autour d’une petite cour a livré un moulin à olives, un moulin à céréales, un pétrin et un four à pain, tandis qu’une grande maison (habitée par l’évêque ?) comportait un étage orné de stucs et d’enduits peints au-dessus d’un pressoir et d’un cellier, d’une citerne et d’une écurie.44

Fig. 20. Fragment de mosaïque provenant de l’étage du bâtiment D (N. Beaudry).

n’empêche pas que son étage ait pu communiquer avec celui du bâtiment voisin. On n’a pas encore identifié la résidence de l’évêque de Byllis ni les espaces dédiés aux fonctions protocolaires et administratives de sa charge. Le fort dénivelé au nord de

44

214

Ben Abed-Ben Khader et al. 2011, pp. 225-279.

Les espaces domestiques et économiques du groupe épiscopal protobyzantin de Byllis (Albanie) Le quartier épiscopal de Byllis offre un tableau condensé, mais relativement complet des processus de mutation à l’œuvre dans la ville tardo-antique, jusqu’à sa rétractation et à son abandon final.45 La dissolution des autorités municipales a favorisé l’empiètement transgressif des constructions dans un espace public pourtant protégé par des lois ; elle a aussi favorisé le renforcement de l’autorité de l’évêque, qui prenait part lui-même à la privatisation de l’espace public. L’implantation des églises dans la trame urbaine a profondément remanié le paysage et la géographie de la cité, tandis que les anciens lieux et monuments de la vie civique étaient désaffectés et spoliés. Et ainsi qu’on l’a noté ailleurs, le déplacement de l’habitat à l’étage est concomitant avec le déclin de l’entretien des systèmes d’égout et de drainage comme avec l’apparition en ville d’activités agricoles.46

Anamali S., 1987. ‘Katër mbishkrime ndërtimi nga Bylisi’, in Monumentet, 33/1, pp. 63-72. Bavant  B., 2007. ‘Caričin Grad and the changes in the nature of urbanism in the central Balkans in the sixth century’, in Poulter 2007a, pp. 337-374. Beaudry N., 2010. ‘Një punishte për prodhimin e verës në Bylisi’, in Monumentet, 28, pp. 41-50. Beaudry N., P. Chevalier (2014), ‘Le cœur économique du complexe épiscopal de Byllis (Albanie) au VIe siècle’, in HAM, 20, pp. 206-215. Beaudry N., Blanc P., Bonifay M., Cerova Y., Chevalier P., Haxhimihali M., Hobdari E., Islami A., Kozelj T., Muçaj S., Nallbani E., Raynaud M.-P., Savard M., Sodini J.-P., Tassignon I., Vanderheyde C., WurchKozelj M., 2002. ‘Byllis (Albanie)’, in BCH, 126/2, pp. 659-684.

Dans le contexte des migrations avares et slaves, l’insécurité des campagnes a certainement contribué à la ruralisation de l’espace urbain, autant par l’apport de populations rurales cherchant refuge47 que par le repli d’activités agricoles derrière les murs de la cité.48 L’activité charitable de l’évêché doit aussi avoir joué un rôle dans la concentration de ses moyens de production au cœur même de la ville ; elle invite à voir dans la cour le centre du dispositif de distribution des denrées à la communauté.49

Beaudry N., Blanc P., Bonifay M., Cerova Y., Chevalier P., Haxhimihali M., Hobdari E., Islami A., Kozelj T., Muçaj S., Nallbani E., Sodini J.-P., Tassignon I., Vanderheyde C., Wurch-Kozelj M., 2003. ‘Byllis (Albanie)’, in BCH, 127/2, pp. 622-639. Beaudry N., Berthon A., Bitri E., Chevalier P., Hobdari E., Islami A., Kozelj T., Muçaj S., Nallbani E., Savard M., Wurch‑Kozelj M., 2007. ‘Byllis’, in BCH, 131/2, pp. 1098-1113.

Le modèle bylliote nous rappelle que l’évêché est l’un des principaux acteurs économiques de la cité tardoantique. Il est à la tête d’un domaine foncier qu’il a les moyens de bâtir, d’aménager et d’exploiter50 ; il commande une production agricole et il possède les moyens de la transformer et d’en entreposer les produits ; il est aussi, par son activité charitable, le moteur d’une économie de redistribution. Byllis invite ainsi à envisager l’institution épiscopale en des termes non seulement religieux, administratifs ou urbanistiques, mais aussi économiques, et à situer son activité dans le contexte des mutations qui affectent les villes et leur relation avec les campagnes.

Beaudry N., Berthon A., Bonifay M., Cantuel J., Chevalier P., Kozelj T., Raynaud M.‑P., Wurch‑Kozelj M., 2008. ‘Byllis. Le quartier épiscopal, la Basilique E et les carrières’, in BCH, 132/2, pp. 923-954. Beaudry, N., Bonifay M., Büttner S., Chevalier P., Gagné C., Kozelj T., Savard M., Wurch-Kozelj M., 2009, ‘Byllis. Le quartier épiscopal’, in BCH, 133/2, pp. 735-754. Beaudry N., Chevalier P., Muçaj S., 2010. ‘Le quartier épiscopal, campagne 2009, Byllis (Albanie)’, in BUCEMA, 14, pp. 57-60. Beaudry N., Chevalier P., Muçaj S., 2011. ‘Le quartier épiscopal, campagne 2010, Byllis (Albanie)’, in BUCEMA, 15, pp. 91-95.

Bibliographie Abbreviations :

Beaudry N., Cantuel J., Chevalier P., Hobdari E., Kozelj T., Muçaj S., Raynaud M.‑P., Savard M., Sodini J.‑P., Vinos-Poyo C., Viriot J., Wurch-Kozelj M., 2013a. ‘Travaux de l’École française d’Athènes en 2011. Byllis’, in BCH, 136-137/2, pp. 723-742.

BCH = Bulletin de correspondance hellénique. BUCEMA = Bulletin du Centre d’études médiévales d’Auxerre. HAM = Hortus Artium Medievalium.

Beaudry N., Büttner S., Chevalier P., Kozelj T., Muçaj S., Wurch-Kozelj M., 2013b, ‘Travaux de l’École française d’Athènes en 2012. Byllis’, in BCH, 136137/2, pp. 917-927.

Chevalier, Beaudry 2018, pp. 446-447. Polci 2003, pp. 105-106. 47 Maraković, Turković 2006. 48 C’est ainsi qu’est expliquée la conversion d’une aile du groupe épiscopal de Bargala (Macédoine du Nord) en grenier et en unité de production vinicole, vers le milieu du VIe siècle. Cf. Nacev 2011, 2016. 49 C’est ce qui a été proposé dans le cas de Sidi Jdidi. Cf. Ben Abed-Ben Khader et al. 2011, pp. 354-355. On peut penser que les sekômata de Byllis ont joué un rôle dans les distributions. 50 Sodini 2013. 45 46

Beaudry, N., Chevalier P., Kozelj T., Muçaj T., WurchKozelj, M., 2013c. ‘Le quartier épiscopal de Byllis (Albanie). Les campagnes de fouilles 2011-2012’, in BUCEMA, 17/2: https://journals.openedition.org/ cem/13268. 215

Nicolas Beaudry & Pascale Chevalier della città in area adriatica, Oxford (BAR International Series, 2419), pp. 59-64.

Beaudry N., Chevalier P., Muçaj S., 2013d. ‘Les coulisses d’une cathédrale : le quartier épiscopal de Byllis (Albanie) au VIe siècle’, in Brandt et al. 2013, pp. 12691277.

Ceka O., 2018. ‘Prytaneu e Bylisit’, in Candavia, 7, pp. 185-202.

Bejko L., Hodges R. (éd.), 2006. New Directions in Albanian Archaeology. Studies Presented to Muzafer Korkuti, Tirana (International Centre for Albanian Archaeology Monograph Series, 1).

Cella E., D’Ercole V., 2012. ‘Byllis. La trasformazione dall’impianto ellenistico alla città basilicale’, in Forma Urbis, 17/7-8, pp. 57-64. Cerova Y., Bonifay M., Capelli C., 2005. ‘Amphores épirotes à corps globulaire du VIe s. découvertes à Byllis (Albanie)’, in J. M. Gurt i Esparraguera, J. Buxeda i Garrigos, M. A. Cau Ontiveros (éd.), LRCW, I, Oxford (BAR International Series, 1340), pp. 537-546.

Ben Abed-Ben Khader A., Fixot M., Roucoule S., 2011. Sidi Jdidi, II, Rome (Collection de l’École française de Rome, 451). Biernacki A. J., 2005, ‘A city of Christians: Novae in the 5th and 6th c. AD’, in Archaeologica Bulgarica, 9, pp. 53-74.

Chevalier P., 2004. ‘L’ambon, l’autel et le baptistère dans les églises paléochrétiennes de Byllis’, in Cabanes, Lamboley 2004, pp. 447-453.

Bowden W., 2003. Epirus Vetus. The Archaeology of a Late Antique Province, Londres. Bowden W., 2006. ‘Procopius’ Buildings and the Late Antique fortifications of Albania’, in Bejko, Hodges 2006, pp. 277-286.

Chevalier P., Beaudry N., 2018. ‘Une ville du VIe siècle retournant à la ruralité : désurbanisation et abandon du siège épiscopal de Byllis’, in Lamboley et al. 2018, pp. 435-448.

Brandt O., Cresci S., López Quiroga J., Pappalardo C. (éd.), 2013. Acta XV Congressus internationalis archaeologiae christianae, Toleti (8-12.9.2008). Episcopus, ciuitas, territorium, Cité du Vatican (Studi di Antichità cristiana, 65).

Chevalier P., Raynaud M.-P., Vanderheyde C., WurchKozelj M., Beaudry N., Muçaj S., Sodini J.-P., 2003. ‘Trois basiliques et un groupe épiscopal des Ve-VIe siècles réétudiés à Byllis (Albanie)’, in HAM, 9, pp. 155-165.

Brun J.‑P., 1993. ‘La discrimination entre les installations oléicoles et vinicoles’, in M.-C. Amouretti, J.-P. Brun (éd.), La production du vin et de l’huile en Méditerranée – Oil and Wine Production in the Mediterranean Area, Athènes (BCH, Supplément, 26), pp. 509-537.

Chevalier P., Beaudry N., Büttner S., Cerova Y., Delouis O., Haxhimihali M., Hobdari E., Islami A., Kozelj T., Muçaj S., Nallbani E., Wurch‑Kozelj M., 2005. ‘Byllis’, in BCH, 128-129/2.1, pp. 1186-1225. Chevalier P., Beaudry N., Bonifay M., Cerova Y., Delouis O., Foy D., Hobdari E., Islami A., Kozelj T., Muçaj S., Nallbani E., Savard M., Wurch‑Kozelj M., 2006. ‘Byllis’, in BCH, 130/2, pp. 834-869.

Brun J.-P., 2003. Le vin et l’huile dans la Méditerranée antique. Viticulture, oléiculture et procédés de transformation, Paris (Collection des Hespérides). Cabanes P., Lamboley J.‑L., (éd.), 2004. L’Illyrie méridionale et l’Épire dans l’Antiquité, IV, Paris.

Chevalier P., Muçaj S., Beaudry N., Kozelj T., WurchKozelj M., 2008. ‘Byllis (Albanie), campagne 2007 : le quartier épiscopal, la Basilique E et les carrières’, in BUCEMA, 12, pp. 63-68.

Ceka N., 1984. ‘Koinoni i Bylinëve’, in Iliria, 14/2, pp. 61-89.

Chevalier P., Beaudry N., Muçaj S. 2009. ‘Le quartier épiscopal, campagne 2008, Byllis (Albanie)’, in BUCEMA, 13, pp. 73-76.

Ceka N., 1986. ‘Face à Apollonia : la ville de Byllis’, in Les Dossiers d’archéologie, 111, pp. 64-67. Ceka N., 1990. ‘Fortifikimet antike të bashkësisë byline’, in Iliria, 20/1, pp. 99-146.

Chevalier P., et al., 2011. ‘Byllis’, in BCH, 135/2, pp. 697702.

Ceka N., 1992. ‘Urbanistika dhe banesat në Bylis’, in Iliria, 22/1-2, pp. 73-96.

Deniaux É., 2008. ‘Travaux publics et évergétisme en Albanie à l’époque romaine’, in C. Berrendonner, M. Cébeillac-Gervasoni, L. Lamoine (éd.), Le quotidien municipal dans l’Occident romain, Clermont-Ferrand (Histoires croisées), pp. 431-441.

Ceka N., Ceka O., 2018. ‘The treatment of public space in the Preroman cities of southern Illyria and Epirus (VthIst centuries B.C.)’, in Lamboley et al. 2018, pp. 977990.

Farioli Campanati R. (éd.), 1993. XL Corso di cultura sull’arte ravennate e bizantina. Seminario internazionale di studi, « L’Albania dal Tardoantico al Medioevo, aspetti e problemi di archeologia e storia dell’arte », Ravenna 29 aprile-5 maggio 1993. 1° Colloquio dell’Associazione italiana per lo studio e la conservazione del mosaico, Ravenne.

Ceka N., Muçaj S., 2005. Byllis. Its History and Monuments, Tirana. Ceka O., 2012. ‘Il koinon e la città: l’esempio di Byllis’, in G. de Marinis, G. M. Fabrini, G. Paci, R. Perna, M. Silvestrini (éd.), I processi formativi ed evolutivi

216

Les espaces domestiques et économiques du groupe épiscopal protobyzantin de Byllis (Albanie) Farlati D., Coleti J., 1817. Illyricum sacrum, VII, Venise.

Muçaj S., 1986. ‘Dy monumente të antikitetit të vonë në Bylis’, in Iliria, 16/1, pp. 321-327.

Feissel D., 1988. ‘L’architecte Viktôrinos et les fortifications de Justinien dans les provinces balkaniques’, in Bulletin de la Société nationale des antiquaires de France, pp. 136-146.

Muçaj S., 1987. ‘Bazilika A e Bylisit’, in Iliria, 17/1, pp. 167-202. Muçaj S., 1990. ‘Sistemi fortifikues i qytetit të Bylisit në antikitetin e vonë’, in Iliria, 20, pp. 169-200.

Feissel D., 2000. ‘Les édifices de Justinien au témoignage de Procope et de l’épigraphie’, in Antiquité tardive, 8, pp. 81-104.

Muçaj S., 1993a. ‘Les basiliques de Bylis et leur architecture’, in Farioli Campanati 1993, pp. 569-583.

Frankel R., 1999. Wine and Oil Production in Antiquity in Israel and other Mediterranean Countries, Sheffield (JSOT/ASOR Monograph Series, 10).

Muçaj S., 1993b. ‘Les mosaïques de Bylis et leur place en Épire’, in Farioli Campanati 1993, pp. 585-605. Muçaj S., 2004. ‘Le synthronon, le transept et le chancel du sanctuaire dans les églises de Byllis’, in Cabanes, Lamboley 2004, pp. 417-429.

Fritzilas S., 2012. ‘ΑΜΦΟΡΕΥΣ ΜΕΓΑΛΟΠΟΛΙΤΟΝ. Un sèkôma en marbre de Mégalopolis’, in V. Chankowski, P. Karvonis (éd.), Tout vendre, tout acheter. Structures et équipements des marchés antiques. Actes du colloque d’Athènes, 16-19 juin 2009, Bordeaux/Athènes (Scripta antiqua, 42), pp. 319-331.

Muçaj S., 2008. ‘Aperçu des données archéologiques dans le territoire albanais (IVe-VIIe siècles)’, in Mélanges de l’École française de Rome. Moyen Âge, 120/2, pp. 377384.

Gregory T. E., 1993. Isthmia, V, Princeton.

Muçaj S., 2018. ‘L’évolution urbaine de Byllis de sa fondation jusqu’à l’abandon de la ville’, in Lamboley et al. 2018, pp. 449-466.

Haxhimihali M., 2004. ‘Byllis et sa région à la lumière des sources écrites du VIe siècle’, in Cabanes, Lamboley 2004, pp. 463-466.

Muçaj S., Raynaud M.-P., 2005. ‘Les mosaïques des églises protobyzantines de Byllis (Albanie) : un atelier’, in H. Morlier (éd.), La mosaïque gréco-romaine, IX. Actes du IXe Colloque international pour l’étude de la mosaïque antique et médiévale, Rome, 5-10 novembre 2001, Rome (Collection de l’École française de Rome, 352), pp. 383-398.

Honigmann E. (éd.), 1939. Le Synekdèmos d’Hiéroklès et l’opuscule géographique de Georges de Chypre, Bruxelles (Corpus Bruxellense historiae Byzantinae, Forma Imperii Byzantini, 1). Kirilov C., 2005. ‘The reduction of the fortified city area in Late Antiquity: some reflections on the end of the “Antique city” in the lands of the Eastern Roman Empire’, in J. Henning (éd.), Post-Roman Towns, Trade and Settlement in Europe and Byzantium, II, Berlin/New York (Millennium-Studien zu Kultur und Geschichte des ersten Jahrtausends n. Chr. – Millennium Studies in the Culture and History of the First Millennium C.E., 5/2), pp. 3-24.

Muçaj S., Chevalier P., Bushi S., Xhyheri S., 2014. ‘Monumenti i Terentianusit (Bylis). Raport i gërmimeve 2007-2011’, in Candavia, 4, pp. 181-203. Muçaj S., Sodini J.‑P., Chevalier P., Raynaud M.‑P. (dir.), sous presse. Byllis : les fortifications et les basiliques A, C et D, Athènes/Rome (Recherches archéologiques franco-albanaises, 2).

Kozelj T., Wurch-Kozelj M., 2012, ‘Gray limestone quarries of Byllis (Albania)’, in A. G. Garcia-More, P. Lapuente Mercadal, I. Rodà de Llanza (éd.), Interdisciplinary Studies on Ancient Stone. Proceedings of the IX Association for the Study of Marbles and Other Stones in Antiquity (ASMOSIA) Conference, Tarragona, 2009, Tarragone (Institut català d’arqueologia clàssica. Documenta, 23), pp. 619-627.

Muçaj S. et alii (dir.), à paraître. Byllis : la basilique B et ses annexes, Athènes/Rome (Recherches archéologiques franco-albanaises). Nacev T., 2011. ‘The winery of the Late Antique city of Bargala’, in E. Maneva (éd.), Stephanos archaeologicos in honorem Professoris Ivani Mikulčić, Skopje (Folia Archaeologica Balkanica, 2), pp. 413-424.

Kozelj T., Wurch-Kozelj M., 2018. ‘Les carrières calcaires de Byllis (Albanie)’, in Lamboley et al. 2018, pp. 467479.

Nacev T., 2016. Bargala, Skopje. Pitarakis B., 2012. ‘Daily life at the marketplace in Late Antiquity and Byzantium’, in C. Morrisson (éd.), Trade and Markets in Byzantium (Dumbarton Oaks Byzantine Symposia and Colloquia), pp. 399-426.

Lamboley J.-L., Përzhita L., Skenderaj A. (éd.), 2018. L’Illyrie méridionale et l’Épire dans l’Antiquité, VI, Tirana/Paris.

Polci B., 2003. ‘Aspects of the transformation of the Roman domus between Late Antiquity and the Early Middle Ages’, in L. Lavan, W. Bowden (éd.), Theory and Practice in Late Antique archaeology, Leyde/ Boston (Late Antique Archaeology, 1), pp. 79-109.

Maraković N., Turković T., 2006. ‘Social changes and the idea of urbanity between Late Antiquity and the Early Middle Ages’, in HAM, 12, pp. 91-99. Morris S., 2006. ‘Illyrica pix: the exploitation of bitumen in ancient Albania’, in Bejko, Hodges 2006, pp. 94-106. 217

Nicolas Beaudry & Pascale Chevalier Poulter A. G. (éd.), 2007a. The Transition to Late Antiquity on the Danube and Beyond, Oxford (Proceedings of the British Academy, 141). Poulter A. G., 2007b. ‘The transition to Late Antiquity’, in Poulter 2007a, pp. 1-50. Praschniker C., 1922. ‘Muzakhia und Malakastra. Archäologische Untersuchungen in Mittelalbanien’, in Jahreshefte des Österreichischen Archäologischen Institutes in Wien, 21-22 – Beiblatt, col. 5-224. Praschniker  C., Schöber  A., 1919. Archäologische Forschungen in Albanien und Montenegro, Vienne (Schriften der Balkankommission, Antiquarische Abteilung, 8). Shpuza S., 2006. ‘The Roman colonies of South Illyria: a review’, in Bejko, Hodges 2006, pp. 164-168. Snively C. S., 2009. ‘The fates of Balkan cities in Late Antiquity: recent views’, in Ниш и Византија – Niš and Byzantium, 7, pp. 38-44. Sodini J.-P., 2004. ‘Les annexes liturgiques des basiliques de Byllis’, in Cabanes, Lamboley 2004, pp. 431-446. Sodini J.-P., 2007. ‘The transformation of cities in Late Antiquity within the provinces of Macedonia and Epirus’, in Poulter 2007a, pp. 311-336. Sodini J.-P., 2013. ‘L’activité architecturale et urbanistique des évêques dans les préfectures du prétoire d’Illyricum et d’Orient’, in Brandt et al. 2013, pp. 835-880. de Thallóczy L., Jireček C., de Sufflay E., 1913. Acta et diplomata res Albaniae Mediae Aetatis illustrantia, I, Vienne. Thurmond D. L., 2006. A Handbook of Food Processing in Classical Rome: For Her Bounty no Winter, Leyde/ Boston (Technology and Change in History, 9). Vanderheyde C., 2004. ‘Le tribèlon et les plaques de parapet des églises de Byllis’, in Cabanes, Lamboley 2004, pp. 455-461. Wiseman J., 1984, ‘The city in Macedonia Secunda’, in Villes et peuplement dans l’Illyricum protobyzantin. Actes du colloque de Rome (12-14 mai 1982), Rome (Collection de l’École française de Rome, 77), pp. 289313.

218

Part 3 Architectural Sculpture

16 Marble Masons on Stage: Two Case Studies in the Working Procedures of the Byzantine Sculptural Ateliers Andrea Paribeni Università degli Studi di Urbino Carlo Bo (Italy) The massive production of carved marble in the early centuries of the Byzantine Empire and especially during the golden age of Justinian required the development of a rigorous planning system for the various stages of extraction, working and distribution of materials destined not only for the major and minor centres of the Empire but also for all territories bordering the Mediterranean Sea. In this chapter, we first consider the mode of maritime transport from marble production centres to their destination, especially in the light of the first results of the most recent excavations of the well-known Marzamemi II wreck (the so-called ‘Basilica church wreck’). In the last part of the chapter we present the results of a recent though partial examination of the marble revetments in the spandrels of the arches in Hagia Sophia at Constantinople, which revealed the systematic use of ‘utility marks’ for the assembly of marble elements. La production massive de marbre sculpté dans les premiers siècles de l’Empire byzantin, en particulier pendant l’âge d’or de l’empereur Justinien, a nécessité le développement d’un système de planification rigoureux des différentes étapes d’extraction, de travail et de distribution des matériaux qui étaient destinés non seulement aux centres majeurs et mineurs de l‘Empire, mais aussi à tous les territoires bordant la Méditerranée. Dans cette contribution, nous envisagerons en premier lieu le mode de transport par voie maritime des marbres, des centres de production aux lieux de destination, à la lumière des premiers résultats des fouilles récentes de l’épave bien connue de Marzamemi II (dite « Basilica Church Wreck »). Dans la dernière partie de la contribution, nous présenterons les résultats d’un examen récent, bien que partiel, du revêtement en marbre des arcs du naos de Sainte-Sophie de Constantinople, qui a révélé l’utilisation systématique de « marques utilitaires » pour l’assemblage des éléments de marbre. Die Massenproduktion geschnitzten Marmors in den ersten Jahrhunderten des Byzantinischen Reiches und vor allem während der Blütezeit unter Kaiser Justinian verlangte nach einem genau getakteten System, um die verschiedenen Phasen der Gewinnung, Bearbeitung und Verteilung des Marmors zu koordinieren. Sie waren nämlich nicht nur für die großen und kleinen Zentren des Imperiums, sondern auch für alle an das Mittelmeer angrenzenden Gebiete bestimmt. In diesem Beitrag werden zuerst die Seetransportarten betrachtet, von den Marmorproduktionszentren zu ihren Bestimmungsorten, auch aufgrund der ersten, bei den Ausgrabungen des bekannten Marzamemi-II-Wracks (des sogenannten „Basilica Church Wreck“), erzielten Ergebnisse. Im anschließenden Teil des Beitrags werden die Resultate einer neueren Teiluntersuchung an der Marmorverkleidung der Bögen im Naos der Hagia Sophia von Konstantinopel vorgestellt. Die Marmorverkleidung zeigte die systematische Verwendung von „Gebrauchsspuren“ für die Montage der einzelnen Elemente. La massiccia produzione di sculture in marmo nei primi secoli dell’impero bizantino e soprattutto durante l’età d’oro di Giustiniano richiese lo sviluppo di una rigorosa pianificazione delle diverse fasi di estrazione, lavorazione e distribuzione dei materiali che erano destinati non solo ai centri maggiori e minori dell’impero ma a tutti i territori che si affacciavano sulle rive del Mediterraneo. In questo contributo considereremo in primo luogo le modalità del trasporto marittimo dai centri di produzione dei marmi fino ai luoghi di destinazione, anche alla luce dei primi risultati raccolti negli scavi più recenti del ben noto relitto di Marzamemi II (la cosiddetta “Basilica Church Wreck”). Nella parte conclusiva del contributo presenteremo i risultati di una recente seppur parziale ricognizione dei rivestimenti marmorei dei pennacchi nelle arcate in Santa Sofia di Costantinopoli, che hanno rivelato un uso sistematico di “marchi utilitari” per l’assemblaggio degli elementi marmorei. 221

Andrea Paribeni The massive production and distribution of marble in the Early Byzantine era was fuelled by the increasing demand for architectural items and furniture for the ambitious building programmes promoted during that period by civil and ecclesiastical elites. From the extraction of marble from the quarries to the shipping of the artefacts and their dissemination into all the provinces of the Byzantine Empire, such a challenging goal required a complex series of decisions and well-structured working procedures. In this chapter, I will focus on two distinct topics: 1) the shipping of marbles (finished or roughly worked) from workshop to destination; 2) the organisation of the marmorarii teams at work in the building, as is revealed by the communication system of the signa lapidaria. Despite the discoveries made during the last few decades in the Aegean, Marmara and Mediterranean Seas,1 the first question, i.e., that concerning the marble trade network, is unavoidably bound to the Marzamemi wreck, which still today provides the most valuable archaeological evidence for fully understanding the phenomenon of the production and circulation of marble artefacts in the Mediterranean basin during the fifth and sixth centuries. The discovery and the archaeological excavation of the wreck by Gehrard Kapitän almost sixty years ago2 had such great significance that the so-called Basilica Church Wreck is consistently mentioned in studies of Early Christian and Byzantine sculptural production but, despite its widely recognised importance, it has not yet been the subject of a systematic and comprehensive study. In recent years, many contributions have focused on the wreck and its marbles,3 and this renewed interest is confirmed by two recent exhibitions held in Amsterdam and Oxford, dedicated to the rich heritage of underwater archaeology in Sicily, in which a great deal of space was devoted to the Marzamemi pieces.4 In the meantime, the Marzamemi Maritime Heritage Project, conducted in partnership between Stanford University and the Soprintendenza del Mare, began in 2014 a new series of surveys and archaeological investigations in the area of the wreck, which has allowed the collection of valuable new data about the ship’s additional cargo, personal items, hull remains and marble items that Kapitän could not find in his excavations, due to the continuous shifting of the sand and environmental disturbance of the sea bed.5

Fig. 1. Syracuse, Latomia del Paradiso, marble cargo from Marzamemi shipwreck, general view (A. Paribeni).

Considering all these new elements, the time is ripe for a rehabilitation of the Marzamemi marbles, which had the misfortune to spend 1,500 years under the sea, and the next 60 at the bottom of a latomia (fig. 1). A first step will be the reconstruction of the history, still quite unknown, of the first sightings of the wreck, which, according to some of our preliminary archival researches, date back at least 50 years before the official start of Kapitän’s excavation.6 A more precise determination of these attempts to recover the wreck, planned from 1913 on at least three occasions but perhaps never brought to completion, will not only fill in a gap in the history of the beginnings of underwater archaeology but will also provide valuable data to evaluate the possible impact of partial tampering with the archaeological environment before the systematic excavation of the 1960s.7 Some clues in this direction may be suggested by those marble items belonging to the cargo, which in the recent archaeological investigations directed by Justin Leidwanger and Sebastiano Tusa were found at some distance from the site,8 and also by the two fragments

1 Short reports about Early Byzantine marble wrecks detected along the Mediterranean coast appear in Pulak, Rogers 1994, p. 17; Günsenin 1998, p. 309; Dennert, Westphalen 2004. 2 Kapitän 1961; Kapitän 1980. 3 Bohne 1998; Castagnino Berlinghieri, Paribeni 2011; 2015a; 2015b; Castagnino Berlinghieri 2017. For further bibliography, see the following footnotes. 4 For the Amsterdam exhibition, see Burgersdijk et al. 2015; for the Oxford exhibition (‘Storms, war and shipwrecks. Treasures from the Sicilian Sea’, Oxford, Ashmolean Museum 21 June–25 September 2016), no catalogue has been published. 5 Preliminary reports on the archaeological investigations carried out at Marzamemi by Stanford University and the Soprintendenza del Mare in Leidwanger, Bruno 2013; Leidwanger, Tusa 2015; Leidwanger 2016; Leidwanger, Tusa 2016; Leidwanger, Greene 2017; Leidwanger, Tusa 2017.

6 Relevant documents about this precocious and until now unnoticed investigation of the Marzamemi II wreck are kept in the Archivio Centrale dello Stato in Rome and in the Archivio della Soprintendenza Archeologica di Siracusa. See the forthcoming paper ‘Alcune colonne di un imprecisabile metallo. Avvistamenti e progetti di recupero del relitto di Marzamemi II nella prima metà del secolo scorso’ by E. F. Castagnino Berlinghieri and A. Paribeni. 7 For a rapid and largely incomplete overview of the contents of these documents, see Barsanti, Paribeni 2018, p. 47. 8 Leidwanger, Tusa 2016.

222

Marble Masons on Stage of marble slabs, located some years ago by Elena Flavia Castagnino Berlinghieri and Lorenzo Guzzardi among the materials of unknown provenance kept in the storage rooms of the Archaeological Museum of Syracuse, and recognised as belonging to the set of parapets of the choir screen in the Marzamemi cargo.9

paired with others in marble from Thasos island, one of the most active white marble quarries in Late Antiquity, which was working at the same time in competition and in synergy with the most famous quarries of the Marmara Sea.13 In particular, a study by Donato Attanasio’s team revealed that seven out of the nine Latrun screens sampled have a provenance from Thasos, suggesting that the use of different materials could be related to issues of work organisation and specialisation in different marble production centres.14 Such work organisation could have required the transfer of raw materials to specialised workshops as well as the moving of skilled workers nearer to the quarries.15

If we glance at the marble items that were displayed in the Oxford exhibition (fig. 2), we cannot fail to appreciate the work of cleaning and restoration carried out for this occasion, and to hope that a similar operation will be extended in the near future to the most relevant marbles of the cargo.10 Should such a cleaning campaign be carried out, it would also offer the opportunity to verify by scientific analysis the provenance of the pieces in white marble (columns, capitals, bases, slabs, posts) that until now have been judged by eye (and with good reason) to be of Proconnesian origin.11 We have to keep in mind that recent archaeometric studies conducted on marble furnishings quite similar to ours, such as those of Latrun, Libya,12 traditionally considered to be all of Proconnesian marble, have instead shown a mixed composition, in which elements of Proconnesian were

A close inspection of the Marzamemi marbles offers the opportunity to detect with more precision to what extent any single piece was worked before shipping and what kind of refinement these marbles needed, once they arrived at their destination, from local artisans or from stonemasons travelling with the cargo. Originally, the columns were quite accomplished (except for their terminal collar, whose primary function was to protect the shaft in case of accident and to allow it to be safely moved

Fig. 2. Marble items from Marzamemi shipwreck at the Oxford exhibition ‘Storms, War and Shipwrecks’, held in 2016 (A. Paribeni). Castagnino Berlinghieri, Guzzardi 2014. Cleaning operations on the pieces that were chosen for the exhibition were carried out by Francesco Mannino, under the direction of Rosalba Panvini, Soprintendente per i Beni Culturali e Ambientali di Siracusa. 11 A first attempt to detect with stable isotopes analysis the provenance of the marbles rescued in the more recent excavations has been delivered by Scott Pike of Williamette University; see Leidwanger, Tusa 2016, p. 138. 12 Attanasio et al. 2008. 9

10

Kožely, Wurch-Kožely 2005, with previous bibliography. Attanasio et al. 2008, pp. 1047–1048. 15 The sculptural pieces from the Episcopal Basilica of Stobi, made of Prilep marble, as recent archaeometrical studies have recently shown, are a good example of this second solution; see Niewöhner, Prochaska 2011, pp. 434–436; Audley-Miller et al. 2013, pp. 134–135. 13 14

223

Andrea Paribeni left aside. But a fine example, still existing even if heavily restored, is offered by the sixth-century ambo from the Beyazit Basilica A, now kept in the Ayasofya Müzesi, that was made in pavonazzetto.19 In the capital, we find two fragments of parapets in green Thessalian breccia coming from the church of St John Prodromos at Hebdomon.20 Such a peculiarity, combined with the impressive dimensions of the basilica, indicated by two rows of fourteen columns (or maybe more, given the new capitals found in the last years excavations), leads us to suppose that the marbles sunk off Marzamemi were allotted for the construction of a very distinguished church. Fig. 3. Column base from Marzamemi shipwreck at the Oxford exhibition ‘Storms, War and Shipwrecks’, held in 2016 (A. Paribeni).

With its nuanced picture of sophisticated liturgical furnishings alongside standardised architectural sculpture, the Marzamemi basilica shipwreck may offer us some clues about the destination and patronage of the cargo. In this respect, the absent items could perhaps also be revealing; in the cargo we do not find, e.g. columns, capitals or bases of dimensions suitable for a gallery superimposed on the nave, which is evidence that the church for which those marble items were destined had no galleries at all.21 We also find none of the pulvini so frequently used at Ravenna and also in the western basilica of Latrun, in Cyrenaica, whose architectural sculptures have often been compared to those found at Marzamemi.22 Furthermore, the cargo does not include revetment slabs for walls, or paving slabs, cornices or revetments for spandrels. We might explain this idiosyncrasy by a distribution of tasks between sending structural items first and decorative elements later with another cargo23 or, more probably, by the fact that the church was designed to receive another kind of decoration such as mosaic floors and stucco revetment for the spandrels and the walls above the arches of the nave. Epigraphic data concerning the patronage, which we often find as monograms carved on slabs and capitals, are also lacking.

with a lifting rope), but now they show a corroded surface due to their long immersion in the sea. The thin and regular striation makes it possible, in fact, to distinguish the side that was lying on the sand from other parts more damaged by direct exposure to aggressive environmental action. The column bases seem instead to have been left at a simpler stage of shaping (fig. 3); this cursory treatment, however, is not unusual for the bases that were exported from the Proconnesian quarries in the Early Byzantine age, as several examples from various sites in the empire and also in Constantinople itself show.16 Notwithstanding the damage caused by calcareous concretion, the Kautzsch VII type capitals show a state of fairly complete finishing (fig. 4), except for the upper part of the kalathos, which appears roughly worked.17 The marble slabs of the choir screen, even if in a fragmentary state, show a complete execution of the decorative motifs (fig. 5), as comparison with slabs of the same typology can confirm; likewise, the elements of the imposing ambo in Thessalian breccia (fig. 6) appear completely worked as well.

Despite all these speculations, we are far from being able to suggest a name for the location of the church to which the marble cargo found off Marzamemi was directed – Africa or Sicily are the most credible hypotheses24 – nor to draw an identikit of the patronage milieu that sponsored

It should be emphasised that the Marzamemi ambo (see fig. 1) shares the use of a peculiar kind of polychrome marble with a restricted range of monuments. Paul the Silentiary’s Description, in which the ambo in Hagia Sophia appears to us like a monumental structure made with a lot of luxurious polychrome marbles carefully described by the poet and only partially echoed by miniatures,18 may be

See Fıratlı 1951; Flaminio 2010, pp. 69–72, fig. 64 and 66–67. See Flaminio 2010, pp. 73–75. At Thessaloniki, ambos made with the same marble, albeit differently shaped, are attested in the churches of Hagia Sophia (now in the Archaeological Museum at Istanbul: Karagiorgou 2004, pp. 197, 204, fig. 12) and St Menas. 21 Galleries provided with columns, ionic impost capitals and parapet slabs are attested in the aforementioned Basilica B at Latrun; cf. Michel 2011, p. 30. 22 On the ecclesiastical complex of Latrun, see Widrig 1978; Michel 2011. 23 See, however, Leidwanger, Greene 2017, p. 11, who report the finding of a ‘chunk of amber glass and small slabs of polished marble’ suitable for decorative purposes. 24 Most scholars believe that the ship was directed towards the north African coast (e.g. Kapitän 1980, p. 129; Sodini 1989, p. 167) but the hypothesis of a Sicilian destination should not be discarded (Deichmann 1976, p. 227; Castagnino Berlinghieri, Paribeni 2011, pp. 71–72; 2015b, p. 398); the alternative proposal of Ravenna put forward by the archaeologist of the Marzamemi Maritime Heritage Project is unlikely (Leidwanger, Pike, Donnelly 2018). 19 20

Russo 2006, p. 245; 2010, p. 38. For an overview of these simplified Attic bases, see Barsanti 2013, pp. 490–491, with previous bibliography. 17 Patrizio Pensabene and Eugenio Russo assume that the decoration of the capitals was to be completed once the pieces arrived at their destination (Pensabene 1986, p. 348; Russo 2006, p. 245; 2010, p. 37). More recently the archaeologists of the Marzamemi Heritage Project have stressed the ‘quarry state’ of many of the pieces in the cargo (Leidwanger, Greene 2017, p. 11). For a correct evaluation of the level of finishing of the Marzamemi capitals, see Barsanti, Paribeni 2018, p. 28. 18 For an Italian translation of Paul’s ekphrasis and an archaeological commentary on the parts of the poem where structural, decorative and liturgical items are described, see Fobelli 2005. Some evocative representations of the Hagia Sophia ambo are in the miniatures of the Menologion of Basil II, in which Byzantine appreciation for the sensorial qualities of polychrome marbles reaches its peak (on the peculiar attitude of Byzantine artists and poets towards the description and the representation of marble, see Paribeni 1990; Kiilerich 2012; Liverani 2016). 16

224

Marble Masons on Stage

Fig. 4. Syracuse, Latomia del Paradiso, marble capitals from Marzamemi shipwreck (A. Paribeni).

signs used by Early Byzantine skilled workers for the identification and installation of architectural pieces in the building process. Signs of this kind – different from the identity marks that single out a worker or a group of workers responsible for the realisation of a marble item – can be categorised in the ‘construction mark’ group, according to the typological classification recently proposed by Nicholas Reveyron.25 In a recent paper, Jean-Claude Bessac has stressed the extreme rarity of signa lapidaria in the Near East during the Early Byzantine era,26 but this negative picture can be nuanced with some evidence offered by a number of military and religious buildings promoted by Justinian along the SyroMesopotamian Limes, and in some cases realised by renowned mechanikoi coming from the capital.27 Above all, the Belgian scholar does not take into account in his review the numerous pieces of evidence offered by items produced by marmorarii workshops and disseminated into all regions of the Byzantine Empire, where identity, construction and assemblage marks are widespread.28 Regarding this last class of masons’ marks, I would add, to those already known, some specimina that I was lucky enough to find on the occasion of a survey made in Ayasofya seven years ago, together with Claudia Barsanti, Alessandra Guiglia and Silvia Pedone.29 In that year, the installation of scaffolding on the northern colonnade of the naos had reached the level of the capitals and arches; thanks to the permission of the directors of the Museum, we had the privilege of close contact with the monumental Kesselkapitelle (fig. 7) and their numerous mason marks incised on the tiny space left by the openwork decoration. We could also inspect the openwork decoration of arches and spandrels, elegantly detached from the bottom, which still retain considerable traces of a lively blue colour.30

Fig. 5. Marble slab from Marzamemi shipwreck at the Oxford exhibition ‘Storms, War and Shipwrecks’, held in 2016 (A. Paribeni).

Reveyron 2003. Bessac 2015. Cf. also Bessac 2011, pp. 407–413. 27 Examples of blocks carrying signa lapidaria are attested e.g. in various buildings at Resafa, and also at Qal’at Sim’an and at Qasr al Hayr al Sharqi (Hof 2009, pp. 816–817). 28 In recent times, special attention has been paid to the recognition and the study of masons’ marks incised upon marble artefacts: see in particular Paribeni 2004 for the large collection of marbles from Hagia Sophia, and Marsili 2015; and 2019 for other peculiar cases, both with previous bibliography. 29 Preliminary reports about this survey in Paribeni 2017; Paribeni 2019; Barsanti, Paribeni 2018, pp. 33–34. 30 On the capitals of Hagia Sophia, see now Russo 2012; 2017. For a good example of the use of polychromy in the decoration at Hagia Sophia, see instead Guidobaldi, Pedone 2011, p. 153, fig. 107. 25 26

Fig. 6. Fragment of the staircase parapet from the ambo of the Marzamemi shipwreck at the Oxford exhibition ‘Storms, War and Shipwrecks’, held in 2016 (A. Paribeni).

this luxurious decorative set (imperial entourage? bishop? civil administrator? wealthy dominus?). I will conclude this chapter with few remarks about the system of alphabetical, numerical or simply graphic 225

Andrea Paribeni

Fig. 7. Istanbul, Hagia Sophia, capital of the northern colonnade (A. Paribeni).

Fig. 9. Istanbul, Hagia Sophia, northern colonnade, marble revetment of the second arcade (A. Paribeni).

Fig. 8. Istanbul, Hagia Sophia, nave exedra, arch (Dumbarton Oaks, Image Collections and Fieldwork Archives).

Fig. 10. Istanbul, Hagia Sophia, northern colonnade, marble revetment of the third arcade (A. Paribeni).

The arcades are decorated by two friezes – a branch of acanthus concluded in a key arc by a medallion with a cross; a sequence of five-pointed leaves – bordered by fillets (fig. 8). The arch’s revetment is composed of segments whose exact matching is determined by pairs of Greek letters incised close to the edge of the fillets. As is possible to see in examples from the Roman era, the letters are paired following alphabetical order; not all the letters can easily be read, but the starting point of the numbering is visible in the element above the first column from the west, where, engraved on the fillet, one can clearly distinguish an Α. Continuing east, matching letters identify the point of contact between the twelfth and thirteenth segments (both with Η; see fig. 9) between the thirteenth and fourteenth segments (both with Θ; see fig. 10), between the fourteenth and fifteenth segments (both with Ι). Moving forward, from the twenty-third to the twenty-sixth segment it is easy to recognise a succession of pairs of markings: Ξ, Ο, Π, Ρ (fig. 11–12).

Fig. 11. Istanbul, Hagia Sophia, northern colonnade, marble revetment of the fourth arcade (A. Paribeni).

planning of the work processes within the Justinianic architectural workshop and confirm what Procopius of Caesarea, in his De aedificiis, listed as the chief merits of

Even though the sequence of arches that we were able to inspect was limited, these examples stress the rational 226

Marble Masons on Stage Bohne A., 1998. ‘Das Kirchenwrack von Marzamemi. Handel mit Architekturteilen in frühbyzantinischer Zeit’, in Skyllis, 1/1, pp. 6–17. Burgersdijk D., Calis R., Kelder J., Sofroniew A., Tusa S., van Beek R. (eds.), 2015. Sicily and the Sea, Amsterdam (Allard Pierson Museum Series, 6). Castagnino Berlinghieri E. F., 2017. ‘Marzamemi II: Church wreck’, in P. C. Finney (ed.), The Eerdmans Encyclopedia of Early Christian Art and Archaeology, Grand Rapids, pp. 117–120. Castagnino Berlinghieri E. F., Guzzardi L., 2014. ‘“Nuovi” elementi architettonici della “basilica” a bordo della navis lapidaria di Marzamemi’, in D. Leone, M. Turchiano, G. Volpe (eds.), Atti del III Convegno di Archeologia subacquea, Manfredonia 4–6 ottobre 2007, Bari (Insulae Diomedeae, 24), pp. 49–56.

Fig. 12. Istanbul, Hagia Sophia, northern colonnade, marble revetment of the fourth arcade (A. Paribeni).

Castagnino Berlinghieri E. F., Paribeni A., 2011. ‘Byzantine merchant ships and marble trade: New data from the central Mediterranean’, in Skyllis, 11, pp. 64–75.

the two architects, Anthemius and Isidore: that is, that they were able to carefully adjust the tasks of the various trades involved.31

Castagnino Berlinghieri E. F., Paribeni A., 2015a. ‘Marble production and marble trade along the Mediterranean coast in Early Byzantine age (5th–6th centuries): Data from quarries, shipwrecks and monuments’, in P. M. Militello, H. Öniz (eds.), SOMA 2011. Proceedings of the 15th Symposium on Mediterranean Archaeology, Held at the University of Catania 3–5 March 2011, Oxford (BAR International Series, 2695), pp. 1033–1041.

Bibliography Abbreviations: AMM = Archaeologia Maritima Mediterranea. INAQ = The INA Quarterly. Attanasio D., Brilli M., Rocchi P., 2008. ‘The marbles of two Early Christian churches at Latrun (Cyrenaica, Lybia)’, in Journal of Archaeological Science, 35, pp. 1040–1048.

Castagnino Berlinghieri E. F., Paribeni A., 2015b. ‘Produzione e commercio del marmo lungo le rotte del Mediterraneo: evidenze dai depositi e dai relitti navali delle coste italiane’, in R. Martorelli, A. Piras, P. G. Spanu (eds.), Isole e terraferma nel primo cristianesimo. Identità locale ed interscambi culturali, religiosi e produttivi. Atti del XI congresso nazionale di archeologia cristiana, Cagliari, Dipartimento di Storia, Beni Culturali e Territorio – Sant’Antioco, Sala Consiliare del Comune, 23–27 settembre 2014, Cagliari (Studi e ricerche di cultura religiosa. N.S., 8), pp. 395–402.

Audley-Miller L., Niewöhner P., Prochaska W., 2013. ‘Marbles, quarries and workshops on the Highlands of Northern Macedonia’, in AA, 2013/1, pp. 95–145. Barsanti C., 2013. ‘Una ricerca sulle sculture in opera nelle cisterne bizantine di Istanbul: la Ipek bodrum sarnıcı (la cisterna n. 10)’, in A. Rigo, A. Babuin, M. Trizio (eds.), Vie per Bisanzio. VII Congresso Nazionale dell’Associazione Italiana di Studi Bizantini, Venezia, 25–28 novembre 2009, Bari (Due punti, 25), pp. 477–508.

Deichmann F. W., 1976. Ravenna. Hauptstadt des spätantiken Abendlandes, II/2, Wiesbaden.

Barsanti C., Paribeni A., 2018. ‘La scultura in funzione architettonica a Costantinopoli tra V e VI secolo: aspetti tecnici, tipologici e stilistici’, in Acta ad archaeologiam et artium historiam pertinentia, 30 – N.S., 16, pp. 22–72.

Dennert M., Westphalen S., 2004. ‘Säulen aus Konstantinopel – ein Schiffsfund im antiken Hafen von Amrit’, in Damaszener Mitteilungen, 14, pp. 183–195. Fıratlı N., 1951. ‘Découverte de trois églises byzantines à Istanbul’, in Cahiers archéologiques, 5, pp. 163–178.

Bessac J.-C., 2011. ‘Observations sur la construction monumentale dans le nord-ouest de la République d’Arménie’, in Syria, 88, pp. 379–415.

Fobelli M. L., 2005. Un tempio per Giustiniano, Santa Sofia di Costantinopoli e la Descrizione di Paolo Silenziario, Rome (I libri di Viella. Arte).

Bessac J.-C., 2015. ‘Les marques lapidaires du ProcheOrient: état des recherches’, in R. Romero Medina (ed.), Signum lapidarium. Estudios sobre gliptografia en Europa, América y Oriente próximo, Madrid, pp. 573–601.

Guidobaldi F., Pedone S., 2011. ‘Il viraggio delle scelte decorative nei rivestimenti pavimentali e parietali in età costantiniana: dagli antefatti agli esiti’, in Musiva & Sectilia, 8, pp. 35–163.

31 Procopius of Caesarea, De aedificiis 1.1.24; Paribeni 2004, p. 716; Marsili 2016, p. 243.

227

Andrea Paribeni Leidwanger J., Tusa S., 2016. ‘Marzamemi II ‘church wreck’ excavation: 2015 field season’, in AMM, 13, pp. 129–143.

Günsenin N., 1998. ‘Récentes découvertes sur l’île de Marmara (Proconnèse) à l’époque byzantine: épaves et lieux de chargement’, in Archaeonautica, 14, pp. 309– 316.

Leidwanger J., Tusa S., 2017. ‘Marzamemi II ‘church wreck’ excavation: 2016 field season’, in AMM, 14, pp. 105–122.

Hof C., 2009. ‘Masonry techniques of the early sixth century city wall of Resafa, Syria’, in K.-E. Kurrer, W. Lorenz, V. Wetzk (eds.), Proceedings of the Third International Congress on Construction History, Brandenburg University of Technology Cottbus, Germany, 20th–24th May 2009, Brandenburg, pp. 813–820.

Leidwanger, J., Pike S., Donnelly A. J., 2018. ‘Revisiting the origin and destination of the Late Antique Marzamemi “church wreck” cargo’, in D. Matetić Poljak, K. Marasović (eds.), ASMOSIA XI. Interdisciplinary Studies on Ancient Stone, Proceedings of the XI International Conference of ASMOSIA, Split, 18–22 May 2015, Split, pp. 291–300

Kapitän G., 1961. ‘Schisffsfrachten antiker Baugesteine und Architekturteile vor den Küsten Ostsiziliens. Ergebnisse der archäologischen Taucharbeiten bis 1960’, in Klio, 39, pp. 276–318.

Liverani P., 2016. ‘Il significato del colore nei marmi in età tardoantica’, in P. A. Andreuccetti, D. Bindani (eds.), Il colore nel Medioevo. Arte, simbolo e tecnica. Tra materiali costitutivi e colori aggiunti: mosaici, intarsi e plastica lapidea. Atti delle giornate di studi, Lucca 24–26 ottobre 2013, Lucca (Collana di studi sul colore, 5), pp. 23–40.

Kapitän G., 1980. ‘Elementi architettonici per una basilica dal relitto navale del VI secolo di Marzamemi (Siracusa)’, in W. Gaddoni Frattini (ed.), XXVII Corso di cultura sull’arte ravennate e bizantina: Ravenna, 9–18 Marzo 1980, Ravenna, pp. 71–136. Karagiorgou O., 2004. ‘“  ... και Ατρακίς οππόσα (μάρμαρα) λευροίς χθων πεδίοις ελόχευσε ...  ’: το λατομείο του Ομορφοχωρίου Λαρίσης και η συμβολή του στη βυζαντινή τέχνη’, in Αρχαιολογικά τεκμήρια βιοτεχνικών εγκαταστάσεων κατά τη Βυζαντινή εποχή 5ος–15ος αιώνας. Ειδικό θέμα του 22ου συμποσίου βυζαντινής και μεταβυζαντινής αρχαιολογίας και τέχνης, Αθήνα, 17–18 Μαΐου 2002, Athens, pp. 183–219 and 385–386.

Marsili G., 2015. ‘Il cantiere e i marchi dei marmorari’, in I. Baldini, M. Livadiotti (eds.), Archeologia protobizantina a Kos. La città e il complesso episcopale, Bologna (DISCI. Archeologia, 6), pp. 246–273. Marsili G., 2016. ‘La formazione dell’architetto in età protobizantina: il caso degli Isidori’, in V. Neri, B. Girotti (eds.), La famiglia tardoantica. Società, diritto, religione, (Quaderni di Erga-logoi, 5), pp. 239-255.

Kiilerich B., 2012. ‘The aesthetic viewing of marble in Byzantium: From global impression to focal attention’, in Arte medievale. S. IV, 2, pp. 9–28.

Marsili G., 2019 Archeologia del cantiere protobizantino: cave, maestranze e committenti attraverso i marchi dei marmorari, Bologna (DISCI. Archeologia, 24).

Kožely T., Wurch-Kožely M., 2005. ‘Les carrières de marbre à Thasos à l’époque proto-byzantine. Extraction et production’, in F. Baratte, V. Déroche, C. JolivetLévy, B. Pitarakis (eds.), Mélanges Jean-Pierre Sodini, Paris (Travaux et mémoires, 15), pp. 465–486.

Michel V., 2011. ‘Nouvelles recherches en Cyrenaïque (Lybie): le site d’Érythron-Latrun’, in Les nouvelles de l’archéologie, 123, pp. 28–34. Niewöhner P., Prochaska W., 2011. ‘Konstantinopolitanisches Formenrepertoire in Mazedonien. Zur Bischofskirche von Stobi und den Marmorbrüchen von Prilep’, in Istanbuler Mitteilungen, 61, pp. 433–439.

Leidwanger J., 2016. ‘I relitti di Marzamemi’, in F. Agneto, A. Fresina, F. Oliveri, F. Sgroi, S. Tusa (eds.), Mirabilia Maris. Tesori dai mari di Sicilia. Palermo – Palazzo reale, Sale Duca di Montalto, 6 novembre 2016–6 marzo 2017, Palermo, pp. 185–187.

Paribeni A., 1990. ‘L’uso e il gusto del marmo in età bizantina attraverso le descrizioni e le rappresentazioni antiche’, in E. Dolci (ed.), Il Marmo nella Civiltà Romana. La produzione e il commercio. Mostra/ seminario, Carrara, maggio–giugno 1989, Museo del marmo. Atti del seminario, Lucca, pp. 163–183.

Leidwanger, J., 2018. ‘New investigations of the 6th c. AD “church wreck” at Marzamemi, Sicily’, in Journal of Roman Archaeology, 31, pp. 339–356. Leidwanger J., Bruno N., 2013. ‘Marzamemi II ‘church wreck’ excavation: 2013 field season’, in AMM, 10, pp. 191–198.

Paribeni A., 2004. ‘Le sigle dei marmorari e l’organizzazione del cantiere’, in A. Guiglia Guidobaldi, C. Barsanti (eds.), Santa Sofia di Costantinopoli. L’arredo marmoreo della Grande Chiesa giustinianea, Vatican City (Studi di Antichità cristiana, 60), pp. 649–734.

Leidwanger, J., Greene E. S., 2017. ‘The Marzamemi shipwreck excavation: A Late Antique church under the sea,’ in INAQ, 43/3–4, pp. 8–13. Leidwanger J., Tusa S., 2015. ‘Marzamemi II ‘church wreck’ excavation: 2014 field season’, in AMM, 12, pp. 103–115.

Paribeni A., 2017. ‘Modalità di assemblaggio e messa in opera delle membrature architettoniche nei cantieri romani e bizantini (I–VI sec. d.C.): stato della questione 228

Marble Masons on Stage e nuovi contributi’, in P. Pensabene, M. Milella, F, Caprioli (eds.), DECOR. Decorazione e architettura nel mondo romano, Atti del convegno internazionale, Roma 21–24 maggio 2014, Rome (Thiasos. Monografie, 10), pp. 569–577. Paribeni A., 2019. ‘Riflessi dell’organizzazione del cantiere nel decoro scolpito della Santa Sofia di Costantinopoli’, in S. Cosentino, M. E. Pomero, G. Vespignani (eds.), Dialoghi con Bisanzio. Spazi di discussione, percorsi di ricerca. Atti dell’VIII Congresso dell’Associazione Italiana di Studi Bizantini, Ravenna, 22–25 settembre 2015, Spoleto (Quaderni della Rivista di Bizantinistica, 20), pp. 761–769. Pensabene P., 1986. ‘La decorazione architettonica, l’impiego del marmo e l’importazione di manufatti orientali a Roma, in Italia e in Africa (II–VI d.C.)’, in A. Giardina (ed.), Società romana e impero tardoantico, III, Rome (Collezione storica), pp. 285–429. Pulak C., Rogers E., 1994. ‘The 1993–1994 Turkish shipwreck surveys’, in INAQ, 21.4, pp. 17–21. Reveyron N., 2003. ‘Marques lapidaires: The state of the question’, in Gesta, 42/2, pp. 161–170. Russo E., 2006. ‘La presenza degli artefici grecocostantinopolitani a Roma nel VI secolo’, in Jahreshefte des Österreichischen Archäologischen Instituts in Wien, 75, pp. 243–297. Russo E., 2010. ‘La circolazione degli artefici e del marmo nel VI secolo’, in Bizantinistica. S. II, 12, pp. 31–55. Russo E., 2012. ‘Introduzione ai capitelli di S. Sofia di Costantinopoli’, in Rivista dell’Istituto nazionale d’archeologia e storia dell’arte, 67 – S. III, 35, pp. 95– 72. Russo E., 2017. ‘Sulla lavorazione dei capitelli di S. Sofia di Costantinopoli’, in Bizantinistica. S. II, 18, pp. 45–113. Sodini J.-P., 1989. ‘Le commerce des marbres à l’époque protobyzantine’, in C. Morrisson, J. Lefort (eds.), Hommes et richesses dans l’Empire byzantin, I, Paris (Réalités byzantines), pp. 163–186. Widrig R., 1978. ‘Two churches at Latrun in Cirenaica’, in Papers of the British School at Rome, 46, pp. 94–125.

229

17 New Evidence for Byzantine Sculptures from the Basilica of St Philip at Hierapolis Silvia Pedone Accademia Nazionale dei Lincei (Italy) The purpose of this chapter is to present the preliminary results of research on the Byzantine sculpture of the Church of St Philip at Hierapolis, which was discovered in 2012 by Francesco D’Andria and the Italian Archaeological Team of the Università del Salento (Lecce). The analysis and the study of several sculptures with architectural and liturgical functions were made possible thanks to the FIRB Project Marmora Phrygia (Futuro in ricerca 2012) and a research grant; the latter was coordinated by Manuela De Giorgi and completed in March 2016. The relevance of the sculptures of St Philip is due to the noteworthy quantity and quality of the pieces, as well as to the archaeological context in which the material was found. This allows us to connect the sculptures to the different chronological phases of the monument, and thus to fix an important reference point for the dating of the sculptural production in Hierapolis as well as in the Phrygian region. Furthermore, such a result helps in the reconsideration and better understanding of the role of local workshops and influences from other areas. The present paper focuses on specific aspects of the different sculptural techniques used in the Church of St Philip, in particular the use of colour and its role within the sacred space, in order to reconstruct a ‘visual hierarchy’ of sculptural decoration in the complex design of the building. Cet article présente les résultats préliminaires de la recherche sur la sculpture byzantine de l’église Saint-Philippe à Hiérapolis, qui a été découverte en 2012 par Francesco D’Andria et la Mission archéologique italienne de l’Université du Salento (Lecce). L’analyse et l’étude de plusieurs sculptures dotées de fonctions architecturales et liturgiques, coordonnée par Manuela De Giorgi, ont été menées en mars 2016 grâce au FIRB Projet Marmora Phrygia (Futuro in ricerca 2012) et à une bourse de recherche. L’intérêt des sculptures de Saint-Philippe tient à leur nombre, à la qualité des pièces et au contexte archéologique dans lequel elles ont été trouvées. Ceci permet de faire le lien entre les sculptures et les différentes phases de la chronologie du monument, et, surtout, d’établir une référence importante pour la datation de la production sculpturale à Hiérapolis et dans la région de la Phrygie. De plus, ces résultats aident à réexaminer et à mieux comprendre le rôle des ateliers locaux et des influences provenant d’autres régions. La présente étude se concentre sur les aspects spécifiques des différentes techniques sculpturales employées dans l’église Saint-Philippe. Une attention particulière est consacrée à l’usage de la couleur et à son rôle dans l’espace sacré, afin de reconstruire une « hiérarchie visuelle » du décor sculpté dans la complexité de l’édifice. Dieser Beitrag legt die Ergebnisse einer größeren Untersuchung an den byzantinischen Bauplastiken der Kirche des Apostels Philippus in Hierapolis vor, die 2012 durch Francesco D‘Andria und seinem italienischen Team von der Universität von Salento (Lecce) entdeckt worden sind. Die Analyse und Untersuchung der zahlreichen Stücke als Architektur- oder Liturgieelemente konnte in den letzten Jahren mit der Unterstützung von FIRB, dem Marmora Phrygia Projekt (Futuro in ricerca 2012), durchgeführt werden. Das Projekt endete in der ersten Jahreshälfte 2016. Die Bedeutung der Bauplastiken in Hierapolis liegt in ihrem quantitativen und qualitativen Reichtum sowie im außergewöhnlichen archäologischen Kontext, aus dem sie stammen. Das ermöglicht die Bauplastiken zu den einzelnen chronologischen Phasen des Gebäudes zuzuordnen und liefert zugleich einen wichtigen Anhaltspunkt für die Datierung der Bauplastikproduktion in Hierapolis und in Phrygien. Darüber hinaus helfen die Ergebnisse die Rolle der lokalen Werkstätten und die äußeren Einflüsse zu überprüfen und besser zu verstehen. Die vorliegende Arbeit konzentriert sich auf die Fragen der verschiedenen Handwerkstechniken in der Kirche des Heiligen Philippus. Insbesondere werden die Verwendung von Farben und ihre Bedeutung innerhalb der heiligen

231

Silvia Pedone Räume analysiert, um eine Art „visuelle Hierarchie“ der Bauplastik im gesamten Gebäude erstellen zu können. Il presente articolo intende presentare i risultati di una più ampia ricerca condotta in tandem con Manuela De Giorgi sulla scultura bizantina proveniente dalla chiesa di San Filippo a Hierapolis, la cui eccezionale scoperta, avvenuta nel 2012, si deve a Francesco D’Andria e al team della Missione italiana dell’Università del Salento (Lecce). L’analisi e lo studio del ricco numero di sculture in funzione architettonica e liturgica è stato condotto negli ultimi anni grazie al progetto FIRB Marmora Phrygia (Futuro in ricerca 2012) che si è concluso nella prima metà del 2016. L’importanza delle sculture del San Filippo riguarda la ricchezza in termini quantitativi e qualitativi, nonché il contesto archeologico in cui i materiali sono stati trovati; dato che permette di ancorare cronologicamente la scultura alle fasi del monumento fornendo così un importante punto di riferimento cronologico per la produzione della scultura bizantina della Frigia, e contribuendo alla riconsiderazione e alla comprensione del ruolo delle botteghe locali e delle influenze da altri ambiti geografici. Focus della presente relazione saranno gli aspetti legati alle differenti tecniche esecutive della scultura del San Filippo, in particolare l’uso del colore e il suo ruolo all’interno dello spazio sacro allo scopo di rintracciare una sorta di «gerarchia visiva» della decorazione scultorea all’interno del già complesso e articolato contesto dell’edificio. In this chapter I will present some new finds from the excavation campaigns at the Church of St Philip at Hierapolis (Pamukkale) that may enhance our knowledge – at least, so I hope – of the rich field of sculptural production in Asia Minor, and particularly in the Phrygia region.1 Between 2010 and 2012 in the ancient Hellenistic, Roman and then Byzantine city of Hierapolis in Phrygia2 – well-known for its suggestive white limestone formations and pools – the remains of a three-aisled basilica were excavated on the eastern side of the site (fig. 1).3 The church, with a wide narthex, incorporates a monumental Roman tomb (fig. 2), identified as the ancient resting place of the Apostle St Philip, who was a martyr at Hierapolis in AD 80, as we know from the so-called Acta Philippi.4

Christian East. Thanks to the work carried out by the Italian archaeological team of the University of Salento, led by Francesco D’Andria, the whole hill – partly rising above the remains of the Roman necropolis – is now markedly characterised by such an important discovery as

The building is connected via a staircase with the great octagonal Martyrion, also devoted to St Philip, the original structures of which, still visible today, have been dated to the fifth century by Paolo Verzone,5 on the basis of similarities to other important martyria of the This essay is a preliminary presentation of a wider research project on the Byzantine sculptures of St Philip at Hierapolis that I am carrying out with Manuela De Giorgi (Università del Salento). Some of the results have been presented at several conferences over the last two years. Cf. De Giorgi 2016; Pedone 2016; De Giorgi 2018; De Giorgi, Pedone 2019 (and their forthcoming book La scultura bizantina del San Filippo di Hierapolis: il catalogo, planned for inclusion in the ‘Missione archeologica italiana a Hierapolis di Frigia’ series). The literature on Byzantine sculpture in Phrygia has considerably increased in recent years, also thanks to new archaeological discoveries, so it is not possible to give a complete bibliography here. Nonetheless, I want to mention the still pivotal contributions of Barsanti 1988 and 2007 for the Middle Byzantine period, and the catalogue of Parman 2002. For more recent work on the archaeology of Anatolia, cf. Niewöhner 2017. 2 On the city, see D’Andria 2003; D’Andria et al. 2008; Scardozzi 2015 (with the previous bibliography). 3 D’Andria 2011–12; Caggia 2016a; 2016b. 4 Bovon et al. 1996; Huttner 2013 (cf. chapter 5). On epigraphic, numismatic and literary Christian sources, see Ritti 1985; Silvestrelli 2000, pp. 378–382. On the inscriptions found in the Basilica of St Philip, see Ritti 2011–12, pp. 53–61; also, on the north-west necropolis, see the recent volume of Brandt et al. 2017, especially the essays by D’Andria in the same volume (D’Andria 2017b). 5 Verzone 1960. On the Martyrium, see also Arthur 2006, pp. 154–158; and more recently Gümgüm 2012; D’Andria 2016. 1

Fig. 1. Plan of the St Philip hill with the main monuments dedicated to the memory of the Apostle: the Martyrion, the Basilica, the aghiasma and the staircase (Scardozzi 2015, pl. 21).

232

New Evidence for Byzantine Sculptures from the Basilica of St Philip at Hierapolis

Fig. 2. Hierapolis (Pamukkale), view of the tomb of St Philip (S. Pedone).

a monumental site and an ancient place of worship.6 This area of the town, outside the city walls, was linked with the city centre via a street or pathway ascending toward the religious monuments devoted to the memory of the Apostle, which soon began, early in the Christian era, to become a place of pilgrimage.7 Relevant, newly found archaeological and epigraphic evidence has allowed the sure identification of the tomb with that of the saint.8

A catalogue of this material will be published in a volume of the series devoted to Hierapolis by the Italian archaeological mission. For this purpose, close cooperation among archaeologists, topographers, and the geologist of the Italian team10 has greatly contributed to a deeper understanding of the site and of the different phases of the history of the complex, in its close relationship with the monumental Martyrion.11 Some specific functional aspects of both buildings and their liturgical use are still to be clarified. If it is probable that the Martyrion was a religious healing site in which incubation rites were performed,12 the specific rites carried out in the Church of St Philip are not yet clear. The full understanding of the complex water supply system brought to light by the excavations needs additional information. In particular, the presence of a water collecting basin under the main altar could be connected with the ancient practice of the so-called ‘thalassa’, that is, the rite of the washing of the liturgical objects after the Eucharist, as documented in some Early Christian basilicas in Greece.13

Alongside the archaeological and architectonic analysis of the building, Manuela De Giorgi and I have begun to study the rich sculptural material that is the topic of the present chapter. The quantity and the quality of the records are outstanding material for a better knowledge of the sculpture of Byzantine Asia Minor, also taking into account the complex chronological phases of the monument, spanning the period from the fourth/fifth century to the twelfth/ thirteenth century, when the structures were given new functions and then were gradually abandoned.9

During the excavation campaigns of 2011 and 2012, a great number of fragmentary sculptures with architectural or liturgical functions and belonging to the decoration

D’Andria 2011–12; Scardozzi 2015. D’Andria 2017b. 8 Many graffiti were discovered at St Philip’s, in some of which appear invocations to the Apostle Philip: Guizzi, Nocita 2016; D’Andria 2017a, fig. 11. 9 The archaeological phases of the building are well summarised by Francesco D’Andria (2011–12) and partially confirmed by Caggia in her excavation reports (2016a and b). It is worth adding that after the battle of Manzikert (1071) the central and western regions of Asia Minor began to be threatened by the expansionistic drive of the Turkish-Seljuk army. Although the Emperors Alexios I Komnenos (1081–1118), John II (1118–1143) and Manuel (1143–1180) had reconquered many territories lost after the defeat of Manzikert, the continuous enemy raids caused a final capitulation in 1210. Cf. also Peirano 2011. On the last phases of the history of Byzantine Hierapolis, see Verzone 1978. It seems plausible that the increasing abandonment of the city was complete after the earthquake of 1354 (Arthur 2006). 6 7

10 A similar multidisciplinary approach has been adopted for the realisation of the database Marmora Phrygiae, which gathers the results of the surveys on marble and stone typologies of the area (quarries, working areas etc.). The database also records the samples taken from the materials excavated in the quarries and from the pieces still in situ. For the results, see Ismaelli, Scardozzi 2016. 11 See Goar 1647; Pallas 1952; Orlandos 1952–54, pp. 469–471. 12 D’Andria 2003 and 2011, with previous bibliography. 13 Goar 1647, p. 12; Pallas 1952–54, I. I am currently examining such a hypothesis, in the hope that further new evidence will corroborate it.

233

Silvia Pedone

Fig. 3. Hierapolis (Pamukkale), impost capitals, bases and columns of the Basilica of St Philip (S. Pedone, M. De Giorgi).

originally belonging to the templon, and several pieces from an ambo and a ciborium.17

of the church were found in the nave and lateral aisles.14 Among this material are shafts of columns with bases and Ionic impost capitals (fig. 3),15 decorated with simple or encircled crosses, belonging to the upper galleries, as well as columns, fluted capitals16 and fragments of an epistyle

The stylobate or base of the chancel screen, still in situ, and a relevant portion of the epistyle made it possible to carry out anastylosis of the screen during the 2013 excavation campaign.18 The shape of the architrave (fig. 4a), with its peculiar ‘squat’ profile, is a typical feature of the workshops active in the city, as are the decoration of its lower face between the columns, with geometric, vegetal and symbolic motifs, and the carving technique of a ‘flattened’ relief. Similar examples are known in the

14 See D’Andria 2016, p. 17, fig. 23; Caggia 2016a, esp. fig. 21–22, 24, 30–32 and 35–36. 15 Such a typology was known at Hierapolis, albeit with slight variations in size and decoration of the oblique faces and volutes. Stylistic similarities also characterise the impost capitals of the Hierapolis Cathedral (Ciotta, Palmucci Quaglino 2002), the different size and the richer ornamentation notwithstanding. Two other identical pieces are now housed in the storeroom of the Hierapolis Museum (Parman 2002, p. 189, pl. 117, fig. 154–156). One of the two capitals was carved from a block of ancient marble with Greek inscriptions (De Giorgi 2018; De Giorgi, Pedone 2019). It is possible that they come from the site of St Philip’s, and they could perhaps have rolled down along the slope of the hill because of an earthquake. 16 This is a very common typology at Hierapolis, as demonstrated by pieces from other basilicas of the city and of the Phrygian region. The differences concern the size (diameter and height), the flaring of the capitals, and the presence of decorative elements such as crosses between the volutes. Cf. also the example published by Parman 2002, pl. 118, figs. 157a–b. For analogous types, cf. also ibid., pl. 109, fig. 140; pl. 124, fig. 167; pl. 133, fig. 183.

On the ciborium and the ambo, see the survey of De Giorgi 2016 and 2018; De Giorgi, Pedone 2019 (and La scultura… – supra, n. 1). The many fragments from the ambo found at St Philip’s are to be compared with the well-known typology of liturgical furnishings produced during the sixth century. Similar fragments are known also in the cathedral of the city (Peirano 2012, fig. 5.7). For this type of ambo, we can mention, for instance, the famous examples from the Beyazid Basilica at Constantinople, now held in the Hagia Sophia Museum, or the Marzamemi one. Cf. also Peirano, Garberoglio 2010–11. 18 Caggia 2016b, pp. 3–5, fig. 4. 17

234

New Evidence for Byzantine Sculptures from the Basilica of St Philip at Hierapolis The ambo (fig. 5), with double flights of steps, was connected to the solea. Some parts of it are preserved, i.e., fragments of the platform, the steps, the niches and two of the four rectangular plutei flanking the flights, decorated with orb and cross, according to a well-known scheme widespread throughout the Empire.21 The four octagonal bases set into the area of the chancel22 – typologically akin to those from Aphrodisias and that now at Kușadası (fig. 6a–c) – may be related to several fragments of the ciborium. The latter is characterised by large round arches with a rough decoration of their profiles, and two clipei with flowers and whirls.23 The most interesting pieces, however, as far as local sculptural production is concerned, are the several slabs found in the church area, like the great slab belonging to the galleries, with a threefold decoration scheme (fig. 7),24 in which the motif of the quincunx inscribed in a double circle with crosses on lemnisci is associated with lateral discs encircling great flowers with alternate and overlapping petals, creating a distinctive optical effect.25 We find such a motif on many other fragments from St Philip’s,26 as well as in the example from Ephesus (fig. 8a–c).27 An analogous, exceptional, yet fragmentary piece is the slab of Docimian marble (marmor docimenium), originally the front of a Roman sarcophagus, but re-carved during the Byzantine Age28 with a complex scheme of knots based on a round module.29 Using only the frame with its small winding columns and eliminating the central sculpted relief is typical of this kind of sarcophagus (fig. 9). But this is not Cf. supra, n. 14; De Giorgi 2018. The four bases were found in situ and still in their original position in the presbytery, formerly framing the monumental mensa, which was instead discovered in the north aisle of the church. On the mensa is inscribed the name of the donor Theodosios. See Ritti 2011–12, p. 55, fig. 1–2. On the phases of the excavation, see Caggia 2016a and 2016b. 23 For the ciborium of the cathedral see Peirano 2012, pp. 594–596, fig. 6–7; 2018, pp. 25–32, with a hypothetical reconstruction (fig. 7a–b) and extended bibliography. Cf. also De Giorgi 2018. 24 MAIER (Missione Archeologica Italiana a Hierapolis) Depot, inv. no HA 12 946; HA 12 US 463; HA 12 US 463; HA 12 947; HA 12 955; HA 12 US 463; HA 12 894; HA 12 958a; HA 12 958b; HA 12 US 463; HA 12 963a; HA 12 963b; HA 12 945; HA 12 949; HA 12 937; HA 12 US 463; HA 12 960; HA 12 940a; HA 12 940b. Dimensions: H.: max. 53 cm; L. preserved: 204.5 cm; D.: 7 cm. 25 See the photos of the slab during the excavation campaign and the restoration process in D’Andria 2015, fig. 4; Caggia 2016b, fig. 8. 26 Cf. for the decorative motif a fragmentary slab (inv. nos HA 11 536; HA 12 893 and HA 12 939) now in the MAIER Depot and the similar one on the slab reused in a later phase in the prothesis chapel (inv. no HA 12 892). 27 See Büyükkolancı 2001. There were probably close artistic relationships between Ephesus and Hierapolis, as the presence in both cities of craftsmen active in the same workshops seems to show. A clear example of such connection is the particular type of cross I have specifically discussed in my contribution to the Second Annual Conference on Byzantine and Medieval Studies, Nicosia 12–14 January 2018. Cf. Pedone 2018. 28 MAIER Depot, inv. no HA 11 358; HA 11 705; HA 11 953; HA 11 423; HA 11 618; HA 12 sp.; HA 11 288; HA 10 98; HA 12 805; HA 12 931a; HA 12 931b; HA 12 sp. 29 MAIER, Depot, inv. no HA 11 358; HA 11 705; HA 11 953; HA 11 423; HA 11 618; HA 12 sp.; HA 11 288; HA 10 98; HA 12 805; HA 12 931a; HA 12 931b; HA 12 sp. Dimensions: H.: 93 cm; L. preserved: 193,3 cm; D.: 6.7 cm. 21 22

Fig. 4a. Hierapolis (Pamukkale), epistyle of the templon (S. Pedone); b. Hierapolis, epistyle from the cathedral of the city (Peirano 2012); c. Ephesus, St John’s, view of the epistyle (S. Pedone); d. Ephesus, Church of the Council or Theotokos, view of the epistyle (S. Pedone).

Cathedral of Hierapolis19 (fig. 4b) and also from the region along the river Lykos, e.g. pieces from St John of Ephesus (fig. 4c) and the epistyle in the church of the Theotokos at Ephesus (fig. 4d).20 19 20

Peirano 2012, p. 592, fig. 2. Degasperis 2013, pp. 53–66.

235

Silvia Pedone

Fig. 5. Hierapolis (Pamukkale), St Philip’s Basilica, view of the ambo fragments (S. Pedone, M. De Giorgi).

Fig. 6a. Hierapolis (Pamukkale), St. Philip’s Basilica, one of the four bases of the bema (S. Pedone); b. Aphrodisias, marble base (S. Pedone); c. Kușadası, marble base (S. Pedone).

236

New Evidence for Byzantine Sculptures from the Basilica of St Philip at Hierapolis

Fig. 7. Hierapolis, (Pamukkale), MAIER Depot, marble slab (S. Pedone, M. De Giorgi).

Fig. 8. Hieropolis (Pamukkale), MAIER Depot; a. detail of the marble slab (S. Pedone); b. Hierapolis, St Philip’s Basilica, slab now in the prothesis (S. Pedone); c. Ephesus, marble slab (E. Russo).

237

Silvia Pedone

Fig. 9. Hierapolis (Pamukkale), Roman sarcophagus reused as fountain in the narthex of St Philip’s Basilica (S. Pedone).

the only case of re-use in the Church of St Philip, probably due to the plentiful presence of ancient spolia in the city.30 A further instance is the Roman sarcophagus adapted as a fountain in the narthex,31 on which the original classical decoration depicting spear and shield was re-carved with highly stylised geometric and floral motifs, similar to the ones we have just seen in the previous case.32 Then, in a later phase, between the ninth and eleventh centuries, one of the short sides of the sarcophagus-fountain was re-used as part of the screen in the prothesis of the little northern chapel (fig. 8b).

Proconnesian, the so-called Mermer Tepe (fig. 10).33 The decorative pattern imitates the well-known scheme of the Justinianic plutei, with lozenge-shaped moldings, like those in Hagia Sophia or the fragmentary slab from Ephesus.34 But on the slab from Hierapolis the moldings are spaced out by small pseudo-piers decorated with stylised elements. Besides the sculptural material still in situ in the Basilica of St Philip, there is a rich series of marble pieces coming from the church but now in storage at the Italian Mission and belonging to chronologically different decorative phases of the building.

To such a homogeneous group, another, larger slab is to be added, made of a local type of marble similar to the

Another category of material discovered at St Philip’s is that of the cornices en champlevé, of which many fragments from different periods survive: a sign of

A lot of fragments were re-used on the site of St Philip, but in many cases the ancient marbles were re-carved. 31 Caggia 2016b, fig. 16; D’Andria 2017a, fig. 9–10, 12. 32 Another side of the same sarcophagus is now in the MAIER Depot, inv. HA 11 395. 30

33 34

238

Pedone in De Giorgi, Pedone, 2019, fig. 7a–c. Barsanti 2004, pp. 315–474.

New Evidence for Byzantine Sculptures from the Basilica of St Philip at Hierapolis continuity in the use of such a distinctive technique.35 The narrow cornices, originally framing or dividing marble revetments on the walls, belong to the earliest phase. In the same group, there are also larger cornices, later reused in the opus sectile of the church floor,36 featuring a ‘minimal’ champlevé technique, limited to the profiling of the motif. The reconstruction work also allowed the identification of three different large slabs datable to the Middle Byzantine period, made of Hierapolitan marble, and decorated with a complex scheme based on the motif of the central lozenge with a ribbed ribbon and knots (fig. 11).37 This scheme elaborates the different elements in an original way and combines them with the chromatic quality of the so-called Hierapolitan travertine, which probably had an important role in the overall decoration of the building, as other pieces carved in the same stone clearly show. A very interesting fragment, dated to the later phases of the monument, and unique of its kind, is a large window screen (130  cm high) with a lunate profile, decorated with an inscribed cross (fig 12a) – similar to that we have just seen – but cut out along the arms and in the central

Fig. 10. Hieropolis (Pamukkale), MAIER Depot, marble slab (S. Pedone, M. De Giorgi).

Fig. 11. Hieropolis (Pamukkale), MAIER Depot, marble slab (S. Pedone, M. De Giorgi).

disc.38 The fine tracery work is further underlined by the profiling of the design, with evident traces of the original rubrication (red colour). In the figure, a suggestive visual comparison with a painted example from Göreme in

Pedone 2016. Pedone 2016, fig. 8. On the opus sectile pavement of the church, cf. also Caggia 2016b (passim); 2018. 37 The slab was found in several fragments in different excavation campaigns: inv. no S.N (A1: 24.7 cm x 24.8 cm); HA 11 240 (A2: 26.2 cm x 41.7 cm); HA 11 515 (A3: 29.5 cm x 26.1 cm); HA 10 94 (A4: 36 cm x 44 cm); HA 11 610 (A5: 32.5 cm x 25 cm); HA 12 779 (A6: 27.5 cm x 13.3 cm); HA 13 992 (A7: 24 cm x 35.2 cm); HA 11 397 (A8: 13.3 cm x 21 cm); HA 12 US 301. Dimensions: H. preserved: 68.7 cm (hypothetical H.: 101 cm); L.: c. 156 cm; D.: 5/7.5 cm. 35 36

Pedone 2016, pl. 33, 4. Dimensions: H.: 60 cm, L.: 130 cm, D.: 4.5/10.5 cm.

38

239

Silvia Pedone

Fig. 12a. Hieropolis (Pamukkale), MAIER Depot, window transenna (S. Pedone, M. De Giorgi); b. Cappadocia, Göreme, Chapel of St Daniel (C. Bordino).

Cappadocia39 (fig. 12b) may be seen. The iconography and the execution of the cross are comparable with other slab fragments found at Hierapolis and with a wide range of examples of Middle Byzantine sculptures: for instance, the fragmentary epistyle found in the so-called monastery church of Aphrodisias.40

In such a manifold landscape, which we have only sketched here, some aspects are still to be clarified, especially as far as the chronology and the original position of other marble elements in the building are concerned. In particular, it is worth mentioning a fragmentary slab decorated with a threefold scheme,41 a cross on a globe inserted into an arcade and a lozenge, possibly comparable with a slab of unknown provenance now in the archaeological museum

Cf. the frescoes in the Chapel of St Daniel at Göreme (church n. 10) dated to the eleventh century. I wish to express my personal gratitude to Chiara Bordino, for the photograph of the chapel. 40 Pedone 2016, with bibliography. 39

41 MAIER Depot, inv. no HA 12 943a; HA 12 943b; HA 12 944; HA 12 959; dimensions: H.: 89 cm; L.: 129.5 cm; D.: 5.8 cm; cimasa: 18 cm.

240

New Evidence for Byzantine Sculptures from the Basilica of St Philip at Hierapolis of Hierapolis, datable between the sixth and eighth centuries,42 and two other pieces, also fragmentary, with a fine and complex ornament of a cross and a vegetal motif.43

Caggia M. P., 2016b. ‘The marbles of the Church of St Philip in Hierapolis. Phases of construction and opus sectile flooring’, in Ismaelli, Scardozzi 2016, pp. 473–488. Caggia M. P., 2018. ‘Mosaic and opus sectile pavements in the Church of St. Philip in Hierapolis’, in Şimşek, Kaçar 2018, pp. 309–324.

All these elements testify to a lively history of the monument also during the post-Justinianic and the Middle Byzantine period, with different phases between the ninth and the beginning of the thirteenth century, when the monument was finally abandoned. Needless to say, this is simply a brief survey of the very rich sculptural material of the site, and only a deeper and wider analysis will do it the justice it deserves.

Ciotta G., Palmucci Quaglino L., 2002. ‘La cattedrale di Hierapolis’, in D. De Bernardi Ferrero (ed.), Saggi in onore di Paolo Verzone, Rome (Archaeologica, 137; Hierapolis. Scavi e ricerche, 4), pp. 179–216. D’Andria F., 2003. Hierapolis di Frigia (Pamukkale). Guida Archeologica, Istanbul.

Bibliography

D’Andria F., 2011. ‘Conversion, crucifixion and celebration. St Philip’s Martyrium at Hierapolis’, in Biblical Archaeology Review, 37, pp. 34–46.

Abbreviation: RPAA = Rendiconti della Pontificia Accademia romana di archeologia.

D’Andria F., 2011–12. ‘Il Santuario e la tomba dell’Apostolo Filippo a Hierapolis di Frigia’, in RPAA, 84, pp. 1–52.

Arthur P., 2006. Hierapolis (Pamukkale) Bizantina e Turca. Guida archeologica, Istanbul.

D’Andria F., 2015. ‘Phrygia Hierapolis’i (Pamukkale) 2014 Yilı Kazı ve Restorasyon Çalıșmaları’, in A. Özme (ed.), 37. Kazi Sonuçlari Toplantisi, 11–15 Mayis 2015, Erzurum, II, Ankara (T.C. Kültür ve Turizm Bakanlığı, 3474/2; Kültür Varlıkları ve Müzeler Genel Müdürlüğü, 171/2), pp. 157–174.

Barsanti C., 1988. ‘Scultura anatolica di epoca medio bizantina’, in C. Barsanti, A. Guiglia Guidobaldi, A. Iacobini (eds.), Gruppo nazionale di coordinamento C.N.R. ‘Storia dell’arte e della cultura artistica bizantina’. Atti della Giornata di Studio, Roma, 4 dicembre 1986, Rome (Milion. Studi e ricerche d’arte bizantina, 1), pp. 275–306.

D’Andria F., 2017a. ‘Phrygia Hierapolis’i (Pamukkale) 2015 Yilı Kazı ve Restorasyon Çalıșmaları’, in A. Özme (ed.), 38. Kazi Sonuçlari Toplantisi, 23–27 Mayis 2016, Edirne, I, Ankara (T.C. Kültür ve Turizm Bakanlığı, 3490/1; Kültür Varlıkları ve Müzeler Genel Müdürlüğü, 175/1), pp. 353–370.

Barsanti C., 2004. ‘I plutei degli intercolumni delle gallerie. Le cimase, Le transenne’, in A. Guiglia Guidobaldi, C. Barsanti (eds.), Santa Sofia di Costantinopoli. L’arredo marmoreo della Grande Chiesa giustinianea, Vatican City (Studi di Antichità cristiana, 60), pp. 315–529.

D’Andria F., 2017b, ‘The Sanctuary of St Philip in Hierapolis and the tombs of saints in Anatolian cities’, in Brandt et al., pp. 3–18.

Barsanti C., 2007. ‘La scultura mediobizantina fra tradizione e innovazione’, in F. Conca, G. Fiaccadori (eds.), Bisanzio nell’età dei Macedoni: forme della produzione letteraria e artistica. VIII Giornata di studi bizantini (Milano, 15–16 marzo 2005), Milano (Quaderni di Acme, 87), pp. 5–49.

D’Andria F., Scardozzi G., Spanò A., 2008. Atlante di Hierapolis di Frigia, Istanbul (Missione archeologica italiana a Hierapolis di Frigia, 2).

Bovon F., Bouvier B., Amsler F. (transl.), 1996. Actes de l’apôtre Philippe, Turnhout (Apocryphes, 8).

D’Andria R., 2016. ‘Interventi di scavo e restauro nel Martyrion di San Filippo’, in R. D’Andria et al. 2016, pp. 767–773.

Brandt J. R., Hagelberg E., Bjørnstad G., Ahrens S. (eds.), 2017. Life and Death in Asia Minor in Hellenistic, Roman and Byzantine Times. Studies on Archaeology and Bioarchaeology, Oxford (Studies in Funerary Archaeology, 10).

Degasperi A., 2013. Die Marienkirche von Ephesos. Die Bauskulptur aus frühchristlicher und byzantinischer Zeit, Vienna (Ergänzungshefte zu den Jahresheften des Österreichischen Archäologischen Institutes in Wien, 14).

Büyükkolancı M., 2001. St Jean. Hayatı ve Anitı, Selçuk.

De Giorgi M., 2016. ‘Materials, forms and models in the St Philip Church of Hierapolis: Byzantine architectural elements’, in Ismaelli, Scardozzi 2016, pp. 489–500.

Caggia M. P., 2016a. ‘Prime indagini sul terrazzo dell’Aghiasma: la chiesa di San Filippo’, in F. D’Andria, M. P. Caggia, T. Ismaelli T. (eds.), Hierapolis di Frigia, VIII, Istanbul (Missione archeologica italiana a Hierapolis di Frigia, 8)., pp. 729–765. 42 43

De Giorgi M., 2018. ‘Divine liturgy and human skills in the architectural sculpture from the Church of the Apostle in Hierapolis (Phrygia)’, in Şimşek, Kaçar 2018, pp. 291–308. De Giorgi M., Pedone S., 2019. ‘La scultura della chiesa di San Filippo a Hierapolis. Nuovi dati e il progetto

Parman 2002, p. 167, pl. 93, fig. 116. De Giorgi, Pedone 2019 (and La scultura… – supra, n. 1).

241

Silvia Pedone delle città. Strenna dell’Associazione Storia delle Città, 6, pp. 327–334.

Marmora Phrygiae’, in S. Cosentino, M. E. Pomero, G. Vespignani (eds.), Dialoghi con Bisanzio. Spazi di discussione, percorsi di ricerca. Atti dell’VIII Congresso dell’Associazione Italiana di Studi Bizantini, Ravenna, 22–25 settembre 2015, Spoleto (Quaderni della Rivista di Bizantinistica, 20), pp. 367–387.

Periano D., 2012. ‘Scavi e ricerche nel cantiere della Cattedrale’, in F. D’Andria, M. P. Caggia, T. Ismaelli (eds.), Hierapolis di Frigia, V, Istanbul (Missione archeologica italiana a Hierapolis di Frigia, 5), pp. 591– 600.

Goar J. (ed.), 1647. Εὐχολόγιον sive rituale Graecorum complectens ritus et ordines divinae liturgiae, officiorum, sacramentorum, consecrationum, benedictionum, funerum, orationum, & cuilibet personae, statui, vel tempori congruos, juxta usum Orientalis Ecclesiae […], Paris.

Periano D., Garberoglio E., 2010–11. ‘Amboni Iasii: una rilettura con due inediti’, in Atti dell’Accademia delle scienze di Ferrara, 88, pp. 339–357. Ritti T., 1985. Fonti letterarie ed epigrafiche, Rome (Archaeologica, 53; Hierapolis. Scavi e ricerche, 1).

Guizzi F., Nocita M., 2016. ‘Novità epigrafiche da Hierapolis di Frigia (2015)’, in Scienze dell’Antichità, 22/1, pp. 17–24.

Ritti T., 2011–12. ‘Alcune iscrizioni rinvenute nella chiesa di S. Filippo’, in RPAA, 84, pp. 53–61. Scardozzi G., 2015. Nuovo Atlante di Hierapolis di Frigia. Cartografia archeologica della città e delle necropoli, Istanbul (Missione archeologica italiana a Hierapolis di Frigia, 7).

Gümgüm G., 2012. Il Martyrion di Hierapolis di Frigia (Turchia). Analisi archeologica e architettonica, Oxford (BAR Publishing International Series, 2385). Huttner U., 2013. Early Christianity in the Lycus Valley, Leiden/Boston (Ancient Judaism and Early Christianity, 85; Early Christianity in Asia Minor, 1).

Silvestrelli F., 2000. ‘Bibliografia topografica di Hierapolis di Frigia’, in F. D’Andria, F. Silvestrelli (eds.), Ricerche archeologiche turche nella valle del Lykos – Lykos Vadisi Türk arkeoloji araştırmaları, Galatina (Archeologia e storia, 6), pp. 375–423.

Ismaelli T., Scardozzi G. (eds.), 2016. Ancient Quarries and Building Sites in Asia Minor. Research on Hierapolis in Phrygia and Other Cities in South-Western Anatolia: Archaeology, Archaeometry, Conservation, Bari (Bibliotheca archaeologica, 45).

Şimşek C., Kaçar T. (eds.), Geç antik çağda Lykos vadisi ve çevresi – The Lykos Valley and Neighbourhood in Late Antiquity, Denizli (Laodikeia Çalişmalari. Ek yadin Dizisi – Supplementary Series, 1).

Niewöhner P. (ed.), 2017. The Archaeology of Byzantine Anatolia. From the End of Late Antiquity until the Coming of the Turks, New York.

Verzone P., 1960. ‘Il Martyrium ottagono a Hierapolis di Frigia. Relazione preliminare’, in Palladio. Rivista di storia dell’architettura e restauro, 1–2, pp. 1–20.

Orlandos A.K., 1952–54. Η ξυλόστεγος παλαιοχριστιανική βασιλική της μεσογειακής λεκάνης. Μελέτη περί της γενέσεως, της καταγωγής, της αρχιτεκτονικής μορφής και της διακοσμήσεως των χριστιανικών οίκων λατρείας από των Αποστολικών χρόνων μέχρις Ιουστινιανού, Athens (Βιβλιοθήκη της εν Αθήναις Αρχαιολογικής Εταιρείας, 35).

Verzone P., 1978. ‘Ultime fasi vitali di Hierapolis di Frigia’, in E. Akurgal (ed.), The Proceedings of the Xth International Congress of Classical Archaeology, Ankara-Izmir, 23–30/ IX/1973, Ankara/Izmir, pp. 851–855.

Pallas D. I., 1952. Η ‘Θάλασσα’ των εκκλησιών. Συμβολή εις την ιστορίαν του χριστιανικού βωμού και την μορφολογίαν της λειτουργίας – La ‘Thalassa’ dans l’église chrétienne. Contribution à l’histoire de l’autel chrétien et à la morphologie de la liturgie, Athens (Collection de l’Institut français d’Athènes, 68). Parman E., 2002. Ortaçağda Bizans döneminde Frigya (Phrygia) ve Bölge Müzelerindeki Bizans taş eserleri, Eskișehir. Pedone S., 2016. ‘Byzantine sculpture in Hierapolis: Engraving techniques and colour finishes’, in Ismaelli, Scardozzi 2016, pp. 501–509. Pedone S., 2018. ‘Una singolare bottega di lapicidi bizantini a Hierapolis Efeso e Sardi’, in Acta ad archaeologiam et artium historiam pertinentia, 30 – N.S., 16, pp. 217–236. Periano D., 2008–10, ‘La rinascita delle città anatoliche durante il regno di Manuele Comneno’, in Il tesoro 242

18 La basilique du Léchaion (Corinthe) : étude des sculptures architecturales Christina Tsigonaki Πανεπιστήμιο Κρήτης / Université de Crète (Grèce) La basilique du Léchaion a été découverte par Dimitrios Pallas, qui l’a fouillée de 1956 à 1961. Des travaux complémentaires ont été effectués en 1965. Il s’agit de l’un des monuments protobyzantins les plus importants de Grèce. Le décor architectural de la basilique a déjà été largement commenté dans la bibliographie internationale, sans pourtant avoir été publié dans son intégralité. Les questions fondamentales posées par le matériel ont déjà été abordées par les spécialistes. Cependant, ceux-ci en viennent à des conclusions différentes, principalement en ce qui concerne la datation des sculptures. Dans cet article sont présentées les premières conclusions issues de l’étude, reprise en 2014, de la sculpture architecturale de la basilique. Celle-ci vise à étudier les sculptures architecturales dans leur ensemble en se concentrant sur la relation structurelle, une relation de dépendance totale entre cet ensemble et le bâtiment pour lequel il a été créé. L’étude des chapiteaux composites ioniques à feuilles finement dentelées et le réexamen des données archéologiques montrent clairement que les chapiteaux de ce type datent de la seconde moitié du Ve siècle. The basilica of Lechaion is one of the most important Early Byzantine monuments in Greece. It came to light through excavations conducted by the archaeologist Dimitrios Pallas between 1956 and 1961. Excavations were also carried out in 1965. Although the sculptural decoration of the basilica has been discussed by many scholars, it has not yet been fully published. The main aspects of this material have been examined, however its dating still remains a matter of controversial debate among scholars. This paper presents some preliminary results of the systematic study of the sculptural decoration of the basilica that began in 2014. The aim of this study is to examine the material as a whole, focusing on the structural relationship between the marble sculptures and the building itself. Through the examination of composite capitals with fine-toothed acanthus leaves and the re-examination of the excavation data, it is suggested that this specific type of capital dates to the second half of the fifth century. Die Lechaion-Basilika, eines der bedeutendsten frühchristlichen Denkmäler Griechenlands, wurde von Demetrios Pallas entdeckt und von 1956 bis 1961 ausgegraben. Zusätzliche Untersuchungen wurden 1965 unternommen. Die Bauplastik der Basilika wurde in der Forschung mehrmals kommentiert, jedoch bisher noch nicht komplett publiziert. Ebenso wenig hat die Auseinandersetzung der Spezialisten mit den Grundfragen, die das Material aufwirft, zu einheitlichen Schlüssen in Bezug auf die Datierung geführt. In dem vorliegenden Beitrag werden die ersten Ergebnisse der Forschungen vorgestellt, die seit 2014 an den Bauplastiken durchgeführt werden. Die Untersuchungen haben das Ziel, das gesamte Material zu erfassen und fokussieren sich auf das strukturelle Verhältnis zwischen der Bauplastik aus Marmor und dem Gebäude an sich. Aufgrund der Analyse der Kompositkapitelle mit feingezahntem Akanthus sowie der Revision der Ausgrabungsdaten wird eine Datierung dieses Kapitelltyps in die zweite Hälfte des 5. Jhs. angenommen. La basilica di Lechaeion, uno dei più importanti monumenti proto-bizantini della Grecia, fu scoperta da Dimitrios Pallas e scavata dallo stesso dal 1956 al 1961. Ricerche supplementari furono organizzate nel 1965. Aspetti dell’apparato decorativo scultoreo della basilica sono stati ripetutamente oggetto di pubblicazioni internazionali, senza tuttavia una visione complessiva. Sebbene infatti gli esperti abbiano risposto alle principali domande sollevate dallo studio del materiale, essi giungono a risultati diversi per quanto concerne la cronologia delle sculture. L’articolo presenta i risultati preliminari di una ricerca intrapresa nel 2014 e focalizzata 243

Christina Tsigonaki sull’apparato scultoreo della basilica. Tale ricerca ha avuto come obbiettivo l’esame dell’intero materiale relativo alla basilica per l’evidenziazione della relazione strutturale – una relazione di dipendenza assoluta – tra le sculture architettoniche e l’edificio per il quale queste sono state create. L’esame dei capitelli compositi a foglie di acanto dentellato, insieme al riesame del dato archeologico, suggeriscono una datazione per i capitelli alla seconda metà del V secolo.

Historique de la recherche

P. Velissariou. Malheureusement, le décès prématuré de ce dernier n’a pas permis la publication de ses travaux.4 P. Velissariou mentionne 1149 sculptures architecturales. Même si la plupart ne sont conservées que sous une forme très fragmentaire, le nombre impressionne. L’utilisation immodérée de tous ces marbres, principalement blancs, mais aussi polychromes (comme le verde antico ou le cipolin de Karystos), suggère la puissance économique de l’autorité qui a financé la construction de la basilique du Léchaion.

La situation de la basilique dans le port de Corinthe – capitale de la province de l’Hellas ou Achaïe –, ses dimensions ainsi que la richesse de son décor architectural justifient le fait qu’elle soit mentionnée dans la bibliographie comme l’un des plus importants monuments protobyzantins de l’Illyricum. La basilique du Léchaion a été découverte par Dimitrios Pallas et fouillée par lui de 1956 à 1961, avec des travaux complémentaires en 1965. D. Pallas a publié des rapports détaillés des fouilles dans les Praktika de la Société archéologique d’Athènes et plusieurs articles dans lesquels il a présenté l’avancée des recherches, ainsi que ses hypothèses d’interprétation et les arguments concernant la datation, l’architecture et les usages des espaces du complexe.1

Le matériel se distingue principalement en deux groupes majeurs : a) les sculptures qui ont été importées achevées de la capitale ou exécutées par un atelier de la capitale qui se serait implanté au Léchaion ; b) les sculptures qui ont été exécutées sur place par un atelier local.

En combinant les données de fouille avec celles qu’offraient les sculptures architecturales, D. Pallas a distingué les phases de construction suivantes : premièrement, le corps principal de l’église aurait été érigé autour des années 450460 et terminé dans les années 490-500 ; deuxièmement, les deux cours de la basilique auraient été ajoutées sous le règne de l’empereur Justin Ier (518-527), au plus tôt. Le baptistère était probablement antérieur à la basilique. Selon le fouilleur, cette dernière avait sûrement succédé à un édifice chrétien antérieur.2

Dans le premier groupe s’inscrivent les chapiteaux composites ioniques, les chapiteaux ioniques à imposte, les impostes, les colonnes, les bases ioniques et les plaques (figs. 1-9). La majorité de ces sculptures ont été exécutées dans un marbre blanc avec des veines grisâtres, identifié, grâce à des examens macroscopiques, comme étant originaire de Proconnèse. L’existence d’éléments architecturaux inachevés dans le marbre de cette île indique, selon Jean-Pierre Sodini, que des sculpteurs de Constantinople accompagnaient la matière première et achevaient sur place les sculptures du monument.5 Pourtant, il ne fait aucun doute qu’avec eux, dans le grand chantier de la basilique du Léchaion, travaillaient également des artisans locaux.

D. Pallas a été l’un des pionniers de l’archéologie byzantine en Grèce : l’utilisation des données stratigraphiques pour la datation d’une église était une méthode totalement novatrice dans les années 1960. Jusqu’alors, la datation de la plupart des basiliques de Grèce qui avaient été fouillées se fondait sur la datation de la sculpture architecturale et principalement sur celle des chapiteaux pour laquelle, dès 1936, les chercheurs avaient comme précieux, mais unique guide la monographie de Rudolph Kautzsch.3 Cependant, la datation de la basilique proposée par le fouilleur a été contestée ultérieurement, principalement en raison de la longue période – près de 70 ans – écoulée, selon lui, entre le début des travaux de construction de la basilique et la fin des travaux.

Parmi les sculptures attribuées à un atelier local – celles du second groupe – on distingue comme étant les plus caractéristiques les catégories suivantes : les corniches, les impostes-consoles et les impostes de piliers (fig.  1012). En effet, il s’agit d’éléments architecturaux en partie encastrés dans la maçonnerie, qui ont été fabriqués localement, selon les modules architecturaux du bâtiment. Leur attribution à un atelier local ne fait aucun doute : les feuilles d’eau et l’acanthe molle qui ornent leurs surfaces caractérisent, comme l’a montré J.-P. Sodini, la production d’ateliers locaux dans le Péloponnèse, l’Attique et la Béotie.6 La majorité des sculptures du second groupe ont été sculptées dans un marbre blanc à grains fins.

Les sculptures architecturales D.  Pallas avait confié la publication de la sculpture architecturale de la basilique du Léchaion à son collaborateur

Velissariou 1999 ; et 2001. Sodini 1977, pp. 424-426. Cf. Deichmann 1989, pp. 274-275. 6 Sodini 1977.

Pallas 1956 ; 1957 ; 1958 ; 1959 ; 1960 ; 1961 ; et 1965. Pallas 1970 ; et 1977, p. 171. 3 Kautzsch 1936. 1

4

2

5

244

La basilique du Léchaion (Corinthe)

Fig. 1. Basilique du Léchaion, chapiteau composite ionique (série A).

Fig. 4. Basilique du Léchaion, chapiteau composite ionique (série D).

Les différences dans la provenance de la matière première, les thèmes décoratifs et l’exécution de la décoration sculptée ne marquent pas une différenciation chronologique, mais correspondent au mode d’organisation du travail sur le chantier de la basilique du Léchaion. Les artisans de la capitale et les artisans locaux ont travaillé ensemble sur des catégories distinctes d’éléments architecturaux ayant un emplacement différent dans le bâtiment. La datation des sculptures architecturales Dans la discussion sur la datation du matériel dominent deux tendances : certains chercheurs persistent à dater les sculptures essentiellement sur la base de critères stylistiques et morphologiques, tandis que d’autres se fondent également sur les données de fouille.

Fig. 2. Basilique du Léchaion, chapiteau composite ionique (série B).

Ici, nous nous bornerons à commenter les chapiteaux composites ioniques. Ces chapiteaux constituent les pièces les plus caractéristiques du premier groupe et ils ont souvent fait l’objet d’études et de commentaires. Dans le compte rendu des fouilles publié par D. Pallas dans les Praktika de la Société archéologique d’Athènes sont illustrés six exemples de chapiteaux de ce type, presque intacts, provenant des colonnades des nefs et du transept au sous-sol, ainsi que du narthex.7 Les fouilles du monument ont livré de nombreux morceaux de chapiteaux de ce type, dans un état plus ou moins fragmentaire ; ils seront présentés dans la publication finale du matériel. Les tentatives de sériation des chapiteaux composites ioniques se sont appuyées sur le décor, et surtout sur les différents types de la feuille d’acanthe. Ainsi, R. Kautzsch a réparti les chapiteaux composites à acanthe finement dentelée en trois groupes : les chapiteaux à acanthe dont les lobes inférieurs sont entiers ou fractionnés forment respectivement le premier et le deuxième groupe, alors

Fig. 3. Basilique du Léchaion, chapiteau composite ionique (série C).

7 Pallas 1956, pl. 70a ; 1958, pl. 101b ; 1959, pl. 113c, d et 114a ; 1960, pl. 129a et 130b.

245

Christina Tsigonaki

Fig. 5a-b. Basilique du Léchaion, chapiteau ionique à imposte.

Fig. 6a-b. Basilique du Léchaion, imposte.

que les chapiteaux à acanthe double appartiennent à son troisième groupe.8 Le type de feuille d’acanthe, le décor de la collerette et leurs éventuelles combinaisons sont les critères retenus pour la classification typologique appliquée aux chapiteaux composites remployés dans la Grande Mosquée de Kairouan et à l’église Saint-Marc de Venise, les deux monuments où est attestée la plus grande concentration de ce type de chapiteau.9 Les chapiteaux composites ioniques de la basilique du Léchaion présentent tous des feuilles d’acanthe finement dentelée sur le calathos et une collerette décorée de palmettes dressées. Ils peuvent être regroupés en quatre séries majeures, selon les variations des feuilles d’acanthe, mais aussi en fonction de leurs dimensions diverses. 8

Fig. 7. Basilique du Léchaion, base ionique.

9

246

Kautzsch 1936, pp. 125-137. Harrazi 1982, pp. 119-142; Deichmann 1981, nos 8-15, pp. 149-150.

La basilique du Léchaion (Corinthe)

Fig. 9a-b. Basilique du Léchaion, fragment de plaque décorée sur les deux faces.

divisent en trois digitations (fig.  2 et 16b). La digitation centrale ressort en raison de sa plus grande hauteur et de la rangée de trous de trépan qui court tout son long. Le rendu final des feuilles d’acanthe varie entre les faces du chapiteau.11 Sur une troisième variante du type (série C), le lobe inférieur des feuilles finement dentelée se divise en deux digitations (fig.  3 et 16c).12 Dans la quatrième série (série D) s’inscrivent les chapiteaux dont les lobes inférieurs des feuilles sont entiers (fig. 4).13 Fig. 8. Basilique du Léchaion, base ionique et colonne de la nef centrale.

Les chapiteaux à palmettes dressées sur la collerette et à feuilles d’acanthe dont les lobes inférieurs demeurent entiers ou divisés en deux digitations (ce qui correspond aux séries C et D du Léchaion) représentent les séries les plus nombreuses du chapiteau ionique composite, lesquelles ont de plus connu une large diffusion géographique.14 Il est probable que cette large diffusion corresponde à une large période de production, par différents ateliers.

Les chapiteaux à doubles feuilles d’acanthe finement dentelée appartiennent à la série A (fig. 1 et 16a). Plusieurs motifs décoratifs prennent place sous les angles de l’abaque, les plus caractéristiques étant les aigles aux ailes déployées. Le rendu final des feuilles d’acanthe n’est pas obligatoirement identique sur toutes les faces du chapiteau : l’exemple le mieux conservé présente deux faces consécutives où la nervure centrale des feuilles est soulignée par des trous de trépan, qui sont pourtant absents des deux autres faces.10

En revanche, les séries de chapiteaux à feuilles d’acanthe finement dentelée double, d’une part, et à lobes inférieurs qui se divisent en trois digitations (séries A et B du Léchaion), d’autre part, présentent une plus grande homogénéité que Pallas 1960, p. 157, pl. 129a. Dimensions : diam. 0,48 m, haut. 0,56 m, côté de l’abaque 0,75 m. 12 Dimensions : diam. 0,32 m, haut, 0,40 m, côté de l’abaque 0,58 m. 13 Pallas 1960, p. 157, pl. 130b. Dimensions : diam. 0,36 m, haut. 0,44 m, côté de l’abaque 0,60 m. 14 Tsigonaki 2002, pp. 57-61, fig. 4 et 5.

Les chapiteaux de la série B se distinguent par des feuilles d’acanthe finement dentelée dont les lobes inférieurs se

11

10 Pallas 1959, p. 131, pl. 113c, d. Dimensions : diam. 0,565 m, haut. 0,67 m, côté de l’abaque 0,91 m.

247

Christina Tsigonaki sur la collerette, tandis que cette partie sur les chapiteaux de la basilique du Stoudios est ornée de palmettes disposées tête-bêche et reliées par une tige bifide.16 La basilique du Stoudios fait partie du peu de monuments datés avec certitude du milieu du Ve siècle. J. Kramer est le premier à avoir avancé une datation du monument entre 453 et 462-463. Il a notamment interprété les aigles qui décorent les angles de l’abaque des chapiteaux du narthex comme une allusion à la fonction consulaire exercée par Stoudios.17 Trente ans après, le chercheur est revenu sur le sujet en soutenant que les chapiteaux ioniques composites des basiliques du Stoudios, du Léchaion et de l’Acheiropoietos doivent être datés entre 450 et 480, c’està-dire d’une période de trente ans qui correspond, selon lui, au temps d’activité d’un atelier.18 La diffusion très limitée des chapiteaux composites ioniques à doubles feuilles d’acanthe finement dentelée ajoute un argument complémentaire au raisonnement de J. Kramer. La série A du Léchaion trouve des parallèles à Constantinople19, à Thessalonique (basiliques de l’Acheiropoietos20 et de Saint-Démétrius21), à Philippes22 et dans la ville byzantine de Nakoleia en Phrygie (actuelle Seyitgazi)23. Les chapiteaux à feuilles d’acanthe dentelée dont les lobes inférieurs se divisent en trois digitations (ce qui correspond à la série B du Léchaion) ont également connu une diffusion géographique limitée. Mis à part au Léchaion, on les retrouve à Constantinople24, Ravenne25, Kairouan26, Thessalonique27, Thèbes de Phthiotide28 et Nakoleia29.

Fig.10. Basilique du Léchaion, imposte-console.

Plusieurs des exemplaires mentionnés ci-dessus proviennent de monuments bien connus. Les chapiteaux de la lointaine Nakoleia ne détonnent pas au milieu de ceux-ci, au contraire, ils confirment la relation étroite 16 Les deux différents motifs de collerette se retrouvent sur deux chapiteaux de pilastre, découverts lors des fouilles de la Rotonde à Thessalonique : Kramer 1968, pp. 72-73, 88, fig. 30. 17 Kramer 1968, pp. 61-64. Cf. Mango 1978, pp. 115-122. Cet auteur date le monument d’environ 453. Selon U. Peschlow, la date pourrait être reculée à 450 : Peschlow 1982, pp. 431-432. 18 Kramer 1998, p. 56. 19 Musée Archéologique (no d’inv. 2385) : Mendel 1912-14, p. 446, no 1208. 20 Raptis 2016, III, pp. 536-568 (avec toute la bibliographie antérieure). 21 Kautzsch 1936, p. 135, no 433, pl. 27 ; Kramer 1998, p. 56, pl. 13, fig. 20 ; Sotiriou, Sotiriou 1952, p. 164, pl. 39 b, c. 22 Kramer 1998, p. 57, pl. 15, fig. 34. 23 Şuca-eddin-Tekke : Wulzinger 1913, pp. 66-67, fig. 65 ; Kramer 1973, p. 242, fig. 7. 24 1- Musée Archéologique (no d’inv. 3916) : Zollt 1994, p. 215, no 618 ; 2- Musée Archéologique (no d’inv. 4939) : ibid., pp. 215-216, no 620 ; 3- Musée Archéologique (no d’inv. 5075) : ibid., p. 216, no 621 ; 4- Cour Hadim Ibrahim Pasa Camii : ibid., p. 216, no 622. Un exemplaire provient de l’île d’Antigone : ibid., p. 215, no 619. 25 Piazza del Popolo (auparavant Place Victor-Emmanuel) : Laurent 1899, pp. 208-209, fig. 2. 26 Grande Mosquée de Kairouan : Harrazi 1982, p. 136, nos 267-268. 27 Basilique de Saint-Démétrius : Sotiriou, Sotiriou 1952, p. 163, pl. 38 b ; lapidarium de la Rotonde : Mavropoulou-Tsioumi, Papanikola-Bakirtzi 1979, p 33-34, no 23, pl. 9b. 28 Basilique A : Sythiakaki-Kritsimalli 2012, pp. 173-174, pl. 12, fig. 6871. Un exemplaire similaire mais sans tore se trouve actuellement au Musée de Phthiotide à Ypate : ibid., pp. 180-181, 489, no 26, pl. 13, fig. 73. 29 Seytigazi-Tekke, portique  : Wulzinger 1913, fig.  21  ; Kramer 1973, p. 242, no Α2  ; Mango 1984, p.  46, fig.  6  ; Niewöhner 2007, p.  288, no 451, pl. 55.

Fig. 11. Basilique du Léchaion, imposte de pilastre.

les deux précédentes et ont manifestement une diffusion géographique plus limitée. Les chapiteaux composites ioniques de la série A du Léchaion ont été très justement mis en rapport par Joakim Kramer en 1968 avec ceux de deux monuments très connus de l’ère protobyzantine : la basilique de Saint-Jean du Stoudios à Constantinople (fig. 13) et l’Acheiropoietos de Thessalonique (fig. 14).15 Les caractéristiques communes des chapiteaux de ces trois monuments sont les doubles feuilles d’acanthe finement dentelée et les aigles sur les tranches des volutes. Les chapiteaux du Léchaion et de l’Acheiropoietos présentent de plus la même décoration 15

Kramer 1968, pp. 36-61.

248

La basilique du Léchaion (Corinthe)

Fig. 12. Basilique du Léchaion, corniche de l’abside.

Fig. 15. Corinthe, Musée Archéologique : chapiteau composite ionique provenant du Léchaion.

des chapiteaux de ce groupe. Une scholie de la Souda mentionne que Stoudios, le fondateur de l’église de SaintJean à Constantinople, a également construit à Nakoleia une église dédiée à l’archange Michel. Elle a aussi fait mention de l’épigramme gravé dans l’église, lequel, paradoxalement – si l’on excepte la référence à SaintJean –, est le même que celui de l’église de Constantinople. Les deux monuments semblent ainsi avoir en commun – outre l’épigramme – un décor architectural similaire, si l’on en juge par les chapiteaux dispersés auparavant dans les monastères de derviches de Bektaşi, aujourd’hui au musée de Seyitgazi.30

Fig. 13. Constantinople, basilique du Stoudios : chapiteau composite ionique (V. Marinis).

Les chapiteaux dont les lobes inférieurs se divisent en trois digitations ont dû être produits en parallèle des exemplaires à feuilles doubles. La combinaison de ces deux formes de feuilles d’acanthe sur un nombre limité de chapiteaux plaide en faveur de cette hypothèse. Les deux chapiteaux les mieux conservés proviennent du tribèlon de l’Acheropoietos31, tandis qu’un autre provient du Léchaion (fig.  15), sans que l’on soit pour autant en mesure de confirmer qu’il appartenait à la basilique.32 Les données de fouille Nous nous contenterons ici de commenter deux des sondages qui ont livré les monnaies utilisées pour la Kramer 1973, pp. 244-246 ; Mango 1984, pp. 45-47. Raptis 2016, pp. 536-542, fig. 376-384. 32 Kautzsch 1936, pp. 133-134, pl. 26, no 426. 30

Fig. 14. Thessalonique, Acheiropoietos : chapiteau composite ionique.

31

249

Christina Tsigonaki

Fig. 16a-c. Basilique du Léchaion, chapiteaux composites ioniques (séries A, B, C) (I. Mpitis).

liquéfaction du sol ; il s’agit d’un phénomène associé aux séismes qui touche les alluvions côtières et conduit à des fractures, des affaissements et des glissements latéraux. Ainsi, l’instabilité du sous-sol sur lequel les fondations de la basilique sont posées et le phénomène de liquéfaction ont été la cause de la création de petits cratères en différents endroits du sol. Les spécialistes considèrent que l’action de liquéfaction avait débuté avant l’époque de la construction de la basilique et réapparaissait après des événements sismiques majeurs. Ils ont distingué au moins trois événements de liquéfaction, dont l’un est postérieur à la construction de la basilique et probablement lié au tremblement de terre qui a touché Corinthe pendant le règne de Justin Ier (517-528).36 En prenant en compte tout ce qui vient d’être dit, on peut suggérer que la monnaie de Justin Ier trouvée dans le sondage de la nef centrale, dans un remblai artificiel, date probablement une réfection du pavement de la basilique ; elle ne doit donc plus être liée à la datation de la construction de l’édifice. Cette interprétation justifie pleinement le rapport entre les unités stratigraphiques illustrées sur la coupe publiée par D. Pallas.

datation de la basilique. Dans le numéro de 1960 des Praktika de la Société archéologique d’Athènes, D. Pallas décrit le sondage effectué à l’extérieur de l’angle nordouest de l’aile nord du transept. Là, dans un remblai ayant probablement comblé la tranchée de fondation ont été trouvées deux monnaies, l’une de Valentinien III (425455), l’autre de Marcien (450-457).33 En 1965 la fouille a été poursuivie dans la nef centrale, à deux endroits où le pavement était endommagé. Dans le premier, dans une unité stratigraphique d’étendue limitée (US 3 sur le dessin), a été recueillie une monnaie de Justin Ier (518527). Le fouilleur a considéré que la monnaie de Marcien découverte dans la tranchée de fondation témoignait de l’inauguration du projet, tandis que celle de Justin Ier datait la période de nivellement de l’intérieur de la basilique, et donc celle de la fin des travaux de construction.34 Cependant, dans la bibliographie courante, l’opinion qui gagne du terrain est celle qui voit dans la monnaie de Justin Ier un terminus post quem pour la construction de la basilique. Ainsi, la basilique et ses sculptures seraient à dater des premières décennies du VIe siècle.35 Néanmoins, le réexamen des coupes stratigraphiques permet une interprétation différente des données de fouille. Sur la coupe du premier sondage, on reconnaît, sans l’ombre d’un doute, l’image caractéristique d’une tranchée de fondation : les monnaies trouvées dans cette unité stratigraphique offrent un terminus post quem pour la datation de la basilique. Quant à la coupe stratigraphique du deuxième sondage, elle laisse vraiment perplexe : le relevé présente en effet des unités stratigraphiques adjacentes aux parois verticales (US 3 et 6), fait inhabituel.

Conclusions Sur la base des observations ci-dessus, les sculptures architecturales de la basilique du Léchaion doivent être datées de la seconde moitié du Ve et non du VIe siècle. L’étude détaillée du matériel et sa mise en corrélation avec les données de fouille devrait très probablement permettre une future datation plus précise. La datation de l’édifice et de ses éventuelles phases de réparation devrait être réexaminée en tenant compte des nouvelles données apportées par les études géophysiques du sous-sol.

Par ailleurs, les cratères si caractéristiques du pavement de la basilique du Léchaion ne passent certainement pas inaperçus des visiteurs du monument. Cependant, jusqu’à une date récente, il n’existait dans la bibliographie aucune explication convaincante de leur existence. De telles déformations circulaires sont généralement interprétées comme étant le résultat de la chute des éléments architecturaux. Or, selon une étude récente, la forme, la disposition et les dimensions des structures de déformation du sol et du substrat de la basilique sont des indicateurs de

Pour la compréhension des sculptures du Léchaion, il est nécessaire de les étudier dans leur ensemble, en mettant en valeur la relation structurelle – une relation de dépendance totale – entre cet ensemble et le bâtiment pour lequel elles ont été créées. C’est dans la même optique que l’on doit effectuer les comparaisons avec les sculptures de la basilique de l’Acheiropoietos et de la basilique du Stoudios. Par ailleurs, l’étude des sculptures provenant de la basilique du Léchaion va nous permettre de reconsidérer la justesse et les limites des méthodes de datation qui

Pallas 1960, pp. 159-160 et 165-169, fig. 2. Pallas 1965, pp. 157-162, fig. 4. 35 Sythiakaki-Kritsimalli 2012, p. 179. 33 34

36

250

Minos-Minopoulos et al. 2015.

La basilique du Léchaion (Corinthe) Mango C., 1984. ‘Saint Michael and Attis’, in Δελτίον της Χριστιανικής Αρχαιολογικής Εταιρείας, 12, pp. 39-62.

sont généralement employées, ainsi que de contrôler les classifications typologiques qui ont été proposées pour différentes catégories de matériel.

Mavropoulou-Tsioumi C., Papanikola-Bakirtzi D., 1979. ‘Kιονόκρανα της συλλογής της Pοτόντας Θεσσαλονίκης. Mέρος A’. Kορινθιακά κιονόκρανα και παραλλαγές’, in Mακεδονικά, 19, pp. 11-38.

La compréhension des principes de la conception et de la mise en œuvre d’un bâtiment ambitieux qui pouvait rivaliser avec ceux de Constantinople et de Thessalonique est la première étape indispensable à l’intégration précise du bâtiment et de ses sculptures dans leur époque.

Mendel G., 1912-14. Catalogue des sculptures grecques, romaines et byzantines, I-III, Constantinople. Minos-Minopoulos D., Pavlopoulos K., Apostolopoulos G., Lekkas E., Dominey-Howes D., 2015. ‘Liquefaction features at an archaeological site: Investigations of past earthquake events at the Early Christian Basilica, Ancient Lechaion Harbour, Corinth, Greece’, in Tectonophysics, 658, pp. 74–90.

Bibliographie Abbréviations : BCH = Bulletin de correspondance hellénique. ΠΑΕ = Πρακτικά της εν Aθήναις Aρχαιολογικής Eταιρείας.

Niewöhner  P., 2007. Aizanoi, Dokimion und Anatolien. Stadt und Land, Siedlungs- und Steinmetzwesen vom späteren 4. bis ins 6. Jahrhundert n. Chr., Wiesbaden (Archäologische Forschungen, 23; Aizanoi, 1).

Barsanti C., Paribeni A., 2007. ‘Broken Bits of Byzantium: frammenti di un puzzle archeologico nella Costantinopoli di fine Ottocento’, in A. Calzona, R. Campari, M. Mussini (éd.), Immagine e ideologia. Studi in onore di Arturo Carlo Quintavalle, Milan, pp. 550-565.

Pallas  D., 1956. ‘Ανασκαφή βασιλικής εν Λεχαίω’, in ΠΑΕ, 111, pp. 164-178. Pallas D., 1957. ‘Ανασκαφή της βασιλικής του Λεχαίου’, in ΠΑΕ, 112, pp. 95-104.

Deichmann F. W. (éd.), 1981. Corpus der Kapitelle der Kirche von San Marco zu Venedig, Wiesbaden (Forschungen zur Kunstgeschichte und christlichen Archäologie, 12).

Pallas  D., 1958. ‘Ανασκαφή της παλαιοχριστιανικής βασιλικής του Λεχαίου’, in ΠΑΕ, 113, pp. 119-134.

Deichmann F. W., 1989. Ravenna. Hauptstadt des spätantiken Abendlandes, II/3, Wiesbaden/Stuttgart.

Pallas D., 1959. ‘Ανασκαφή βασιλικής Λεχαίου’, in ΠΑΕ, 114, pp. 126-140.

Harrazi N., 1982. Chapiteaux de la grande mosquée de Kairouan, Tunis (Institut national d’archéologie et d’art. Bibliothèque archéologique, 4).

Pallas  D., 1960. ‘Ανασκαφή εν Λεχαίω’, in ΠΑΕ, 115, pp. 144-170. Pallas  D., 1961. ‘Ανασκαφή εν Λεχαίω’, in ΠΑΕ, 116, pp. 137-154.

Kautzsch R., 1936. Kapitellstudien. Beiträge zu einer Geschichte des spätantiken Kapitells im Osten vom vierten bis ins siebente Jahrhundert, Berlin/Leipzig (Studien zur spätantiken Kunstgeschichte, 9).

Pallas  D., 1965. ‘Ανασκαφικαί έρευναι εν Λεχαίω’, in ΠΑΕ, 120, pp. 137-166. Pallas D., 1970. ‘Über der Datierung eines Kapitells der Basilika von Lechaion (Korinth)’, in Byzantinische Zeitschrift, 63, pp. 69-70.

Kramer J., 1968. Skulpturen mit Adlerfiguren an Bauten des 5. Jahrhunderts n. Chr. in Konstantinopel, Cologne. Kramer J., 1973. ‘Architekturteile des SeyitgaziTekke (Vilâyet Eskişehir) und die Michaelskirche von Nakoleia’, in Jahrbuch der Österreichischen Byzantinistik, 22, pp. 241-250.

Pallas D., 1977. Les monuments paléochrétiens de Grèce découverts de 1959 à 1973, Cité du Vatican (Sussidi allo studio delle Antichità cristiane, 5). Peschlow U., 1982. ‘Die Johanneskirche des Studios in İstanbul. Bericht über die jüngsten Untersuchungsergebnisse’, in H. Hunger, W. Hörandner (éd.), XVI. Internationaler Byzantinistenkongress, Wien, 4.-9. Oktober 1981. Akten, II/4 = Jahrbuch der Österreichischen Byzantinistik, 32/4, pp. 429-435.

Kramer J., 1998. ‘Bemerkungen zu den Methoden der Klassifizierung und Datierung frühchristlicher oströmischer Kapitelle’, in U.  Peschlow, S.  Möllers (éd.), Spätantike und byzantinische Bauskulptur. Beiträge eines Symposions in Mainz, Februar 1994, Stuttgart (Forschungen zur Kunstgeschichte und christlichen Archäologie, 19), pp. 45-59.

Raptis K., 2016. Αχειροποίητος Θεσσαλονίκης. Αρχιτεκτονική και γλυπτική, I-III, thèse de doctorat, Αριστοτέλειο Πανεπιστήμιο Θεσσαλονίκης.

Laurent J., 1899. ‘Delphes Chrétien’, in BCH, 23, pp. 206279.

Sodini J.-P., 1977. ‘Remarques sur la sculpture architecturale d’Attique, de Béotie et du Péloponnèse à l’époque paléochrétienne’, in BCH, 101, pp. 425-450.

Mango C., 1978. ‘The Date of the Studius Basilica at Istanbul’, in Byzantine and Modern Greek Studies, 4, pp. 115-122.

251

Christina Tsigonaki Sotiriou G. A., Sotiriou M. G., 1952. H Βασιλική του Αγίου Δημητρίου Θεσσαλονίκης, Athènes (Βιβλιοθήκη της εν Αθήναις Αρχαιολογικής Εταιρείας, 34). Sythiakaki-Kritsimalli V., 2012. Ο ανάγλυφος αρχιτεκτονικός διάκοσμος στη Θεσσαλία και Φθιώτιδα. Παλαιοχριστιανικά και πρώιμα μεσαιωνικά χρόνια, Volos (Μελέτες, 4). Tsigonaki C., 2002. La sculpture architecturale de Crète à l’époque protobyzantine (IVe-VIIe siècles), I-III, thèse de doctorat, Université Paris 1 Panthéon-Sorbonne. Velissariou  P., 1999. ‘Τα γλυπτά της Βασιλικής του Λεχαίου’, in ΠΑΕ, 154, pp. 333-339. Velissariou  P., 2001. ‘Βασιλική του Λεχαίου. Αρχιτεκτονικά γλυπτά (περίοδοι 2000 και 2001)’, in ΠΑΕ, 156, pp. 141-144. Wulzinger K., 1913. Drei Bektaschi-Klöster Phrygiens, Berlin (Beiträge zur Bauwissenschaft, 21). Zollt T., 1994. Kapitellplastik Konstantinopels vom 4. bis 6. Jahrhundert n. Chr. Mit einem Beitrag zur Untersuchung des ionischen Kämpferkapitells, Bonn (Asia-Minor-Studien, 14).

252

19 The Sculptural Decoration of the Acheiropoietos Basilica (Thessalonike), Re-evaluated in Light of a Recent Architectural Analysis of the Monument Konstantinos T. Raptis* Εφορεία Αρχαιοτήτων Πόλης Θεσσαλονίκης / Ephorate of Antiquities of Thessaloniki City (Greece) dedicated to the loving memory of my wife, Maria Kakagia Based on new evidence documented in the light of a recent architectural analysis of the Acheiropoietos Basilica (Thessalonike), this chapter presents various architectural sculptures of the Early Byzantine monument, including: (i) a variety of bases of both the classical and the simplified variations of the Attic-Ionic type, (ii) various monolithic columns, (iii) the largest group of twenty-nine composite capitals with double fine-toothed acanthus leaves – the socalled ‘Theodosian’ capital – from the tribelon, the nave colonnades and the columns of the outer narthex, which beyond their superficial uniformity are divided into four subgroups, (iv) the impost-blocks that crown the composite capitals, (v) various subtypes of Ionic impost capitals in the south gallery colonnade, (vi) the consoles that support the traverse arches of the narthex, as well as (vii) a variety of cornices that adorn the apse and the nave arcades. In addition, several types of Constantinopolitan mason marks that connect the primary architectural sculptures of the main nave with specific imperial sculptural workshops of the sixth century are discussed. À partir des données nouvelles livrées par une analyse récente de l’architecture de la basilique de l’Acheiropoietos (Thessalonique), cette étude présente les différentes sculptures architecturales de ce monument paléochrétien : (i) des bases de type attique-ionique classique et des variations simplifiées de ce type ; (ii) des colonnes monolithes ; (iii) un groupe de vingt-neuf chapiteaux composites à double feuilles d’acanthe finement dentelées (dits « théodosiens ») appartenant au tribélon, aux colonnades des nefs et à celles de l’exonarthex, et qui, au-delà de leur uniformité superficielle, sont divisés en quatre sous-groupes  ; (iv)  divers sous-types de chapiteaux ioniques à imposte de la tribune sud ; (v) les vingt neuf impostes qui couronnent les chapiteaux composites ; (vi) les consoles qui supportent les arcs transversaux du narthex ; (vii) ainsi qu’une variété de corniches qui ornent l’abside et les arcades de la nef centrale. En même temps sont discutés plusieurs types de marques de tâcheron d’ateliers constantinopolitains qui associent les sculptures architecturales de la nef centrale avec les ateliers de sculpture impériaux du VIe siècle. Der vorliegende Beitrag stellt die Erkenntnisse vor, die im Rahmen neuerer bauhistorischer Untersuchungen an den Bauplastiken in der frühbyzantinischen Acheiropoietosbasilika in Thessaloniki erzielt worden sind. Es handelt dabei um (i) eine Vielzahl von Säulenfüßen sowohl der klassischen als auch der vereinfachten Form des attisch-ionischen Typs, (ii) verschiedene monolithische Säulen, (iii) die größte erhaltene Gruppe von neunundzwanzig Kompositkapitellen mit doppeltem fein gezahntem Akanthusmuster, ein sog. theodosianisches Kapitell am Tribelon, im Schiff, und im Säulengang des Exonarthex, die trotz ihrer Gleichförmigkeit in vier Untergruppen unterteilt werden können, (iv) die neunundzwanzig dekorierten Kämpferblöcke, die Kompositkapitelle, (v) verschiedene Untertypen der ionischen Kämpferkapitelle am Säulengang der Südempore, (vi) die Konsolen, welche die Querbögen des Narthex tragen sowie (vii) eine Vielzahl verschiedener Gesimse, welche die Bögen in Apsis und Schiff schmücken. Zugleich werden verschiedene Typen von Werkstattmarkern aus Konstantinopel besprochen, welche die

253

Konstantinos T. Raptis Zuordnung der ursprünglichen, plastisch gestalteten Bauteile im Hauptschiff zu den kaiserlichen Werkstätten des 6. Jhs. ermöglichen. Sulla base di nuove testimonianze documentate alla luce di una recente analisi architettonica della basilica Acheiropoietos (Salonicco), questo articolo presenta le varie sculture architettoniche del monumento paleocristiano come: (i) una varietà di basi nelle varianti sia classiche sia semplificate del tipo attico-ionico, (ii) varie colonne monolitiche, (iii) il più grande gruppo conservatosi di ventinove capitelli compositi con due corone di doppie foglie di acanto finemente dentellato — del tipo cosiddetto Teodosiano (sic)— nel tribelon, nei colonnati delle navate e nelle colonne del nartece esterno che, oltre la loro apparente uniformità, si dividono in quattro sottogruppi, (iv) le ventinove imposte che coronano i capitelli compositi, (v) vari sotto-tipi di capitelli imposta-ionici del colonnato della galleria meridionale, (vi) i modiglioni che sostengono gli archi trasversali del nartece, nonché (vii) una gamma di diversi tipi di cornici che decorano l’abside e le arcate delle navate. Al tempo stesso vengono discussi i numerosi tipi di sigle dei marmorari di Costantinopoli che collegano le sculture architettoniche primarie della navata centrale a specifici laboratori imperiali di scultura del VI sec.

The Acheiropoietos Basilica1 (fig. 1–2), in its present state, constitutes a typical example of the three-aisled, timber-roofed, Hellenistic-type basilica with narthex and galleries2, thus fitting more closely than any other of the still-standing monuments of Thessalonike the standardised Early Byzantine ecclesiastical architecture.3 Based on a recent architectural analysis of the monument, however, it appears that Acheiropoietos was originally erected as a Western-Roman-type basilica, without galleries, during the last decade of the fifth or the first decade of the sixth century,4 being, probably, the first part of a presumably extensive architectural project of Anastasius I in Thessalonike.5 Even though the ground plan of the basilica has remained unaltered through the ages, its upper structure was subsequently remodelled. The galleries, a no longer extant clerestory and extended annexes were added during the first among many structural restorations of the building, which was redesigned and in large extent rebuilt as a result of the series of notorious earthquakes that devastated large parts of the city, probably during the third decade of the seventh century.6 Based on new evidence regarding the building history of the basilica, the present chapter discusses several aspects of its architectural sculptures.

The bases of the nave colonnades (fig. 3) as well as the bases that support the columns of the tribelon (fig. 4) – all made of finely dressed Proconnesian marble – belong to the Attic-Ionic type and present its classic formation with rectangular plinth, torus, scotia and upper torus.7 Nevertheless, they present various morphological differences in the formation of their profile, as well as different levels of sculptural quality. The bases of the tribelon and a few of the south colonnade – mainly those at the axis of the south propylon – are the finest examples, closest to the fifth-century archetypes. The rest seem to be products of lower quality, made by rather mediocre sculptors. In a few of them, instead of the voluminous lower torus of the finest examples, a quarter-round one is formed, while on some bases the upper torus has been replaced by a simplified double ring. The latter seems to be a late fifth- or early sixth-century simplification of the type, which probably sets the date of the installation of this set of ready-made bases.8 The twenty-four monolithic, unfluted shafts of the nave colonnades, made of dressed Proconnesian marble (fig. 3), along with the two analogous shafts of the tribelon (fig. 4), made of thoroughly polished green Thessalian marble, comprise a solid group with similar dimensions and morphological characteristics in the formation of both their apothesis and apophysis.9

The ground floor The most prominent among the architectural sculptures of the basilica furnish the ground floor level and belong to the foundation phase of the monument.

Composite capitals with fine-toothed acanthus leaves – of the type formerly called ‘Theodosian’10 – crown the columns of the tribelon (fig. 4–6), the nave colonnades (fig. 3 and 7–8), and the columns of the exonarthex – i.e. outer narthex – (fig. 9) that once formed the western façade of the Acheiropoietos Basilica.

For a summary of the history of the monument in English, see Raptis 2019b, pp. 115–128.. A detailed revised analysis of the building is offered in Raptis 2016. 2 Οrlandos 1952–54, passim. 3 Krautheimer 1965, pp. 74–75; Mango 1976, pp. 39–40; Ćurčić 2010, pp. 107–109. 4 Raptis 2016, ΙΙΙ, pp. 752–765; 2017; 2019a, pp. 29–60; 2019b, pp. 115–128. 5 Raptis 2017b, p. 46. 6 Raptis 2017c, as well as the same author’s forthcoming paper ‘The seventh century restoration of Acheiropoietos Basilica and its significance for the urban continuity of Thessalonike during the Dark Age’, in the Proceedings of the 46th Spring Symposium of Byzantine Studies ‘Byzantine Greece: Microcosm of Empire?’ (Birmingham, 23–25 March 2013). 1

Kramer 1970, pp. 271–278. Raptis 2016, III, pp. 503–511. 9 Raptis 2016, III, pp. 525–526. 10 Kautzsch 1936, pp. 115 ff.; Orlandos 1952–54, II, pp. 289–294. 7 8

254

The Sculptural Decoration of the Acheiropoietos Basilica (Thessalonike)

Fig. 1. Acheiropoietos Basilica. Plan. (K. T. Raptis, after the Archives of the Ephorate of Antiquities of Thessaloniki City).

Fig. 2. View of the nave (K. T. Raptis).

255

Konstantinos T. Raptis

Fig. 3. (a) The north colonnade from the west; (b) column of the north colonnade (K. T. Raptis).

The capitals that crown the columns of the nave,11 each with two collars of double fine-toothed acanthus leaves12 (fig. 7–8), seem morphologically homogeneous. Based on slight differences in their general design, however, they are classified into two subgroups. The twelve capitals of the south and the four easternmost ones of the north colonnade (fig. 7b and d–e) have the rim of the basket adorned with a bead-and-reel molding that defines the echinus. On the same capitals, the abacus corners are pointed. On the remaining eight capitals of the north colonnade (fig. 7a and c), the rim of the basket is shallow and formed as a narrow, undecorated listel, while the abacus at its corners forms minor facets, to which the schematic laurel-leaf decoration

is applied. Identical within each set even in the details of their baskets, they are distinguished only by the extremely simplified abacus floses and the motifs that adorn the corners of the volutes; these are the only elements of these capitals where the sculptors felt free to improvise, choosing from an extensive, yet predetermined repertoire of mainly vegetal or floral motifs – acanthus leaves (fig. 8c), palmettes, acanthus stems with fruits or floral calyxes, sprouts with lilies or acanthus flowers, interconnected scrolls ending in ivy leaves – and, in three examples, eagles (fig. 7e–8e). The latter motif, applied also in the same positions on the more elaborate capitals from the Stoudios Basilica in Constantinople and Lechaion near Corinth, links these three monuments in terms of the provenance of their capitals from the same Constantinopolitan workshop of the second half of the fifth century.13 But the sculptural treatment of the Acheiropoietos capitals is less elaborate

Kautzsch 1936, pp. 134–135, pl. 26, no 431; Orlandos 1952–54, II, pp. 290–292; Krautheimer 1965, pp. 74–75; Kramer 1968, pp. 58–64, fig. 20–22; Farioli-Olivieri 1964, pp. 154–155; Raptis 2016, III, pp. 543–577. 12 The subtype with double acanthus leaves is rather rare, with no more than fifty-two known preserved examples, twenty-nine of which are found in Acheiropoietos. 11

13

256

Kramer 1968, p. 64.

The Sculptural Decoration of the Acheiropoietos Basilica (Thessalonike)

Fig. 4. The tribelon (K. T. Raptis).

than the capitals of the other two monuments, with the drill used only to define the outline of the leaves (fig. 8a–b and d), leaving either unfinished or intentionally simpler the lace effect of the most elaborate examples of the type.14

applied on all sides of their baskets (fig. 6), and it is awkwardly missing from the most prominent sides in their current position on the tribelon columns.16 Two of the three capitals, found during excavation at the west of the basilica17 and attributed to its exonarthex, belong to the first subgroup of the nave capitals (fig. 9a), of which the basket rim is decorated with bead-and-reel.18 The third (fig. 9b) differs from the rest of the group, resembling in its details the capital from Phthiotic Thebes.19

The two tribelon capitals,15 with baskets composed of one collar of simple five-lobed acanthus leaves and one collar of double leaves (fig. 5), are obviously different from those of the nave colonnades. More elegant than the rest, they present a highly decorative treatment in terms of their details, as for example the elaborate decoration of the abacus with drilled acanthus-rinceau. Apart from the fine execution of their details, e.g. the inverted palmettes of the astragal or the floses of the abacus that give a lacelike general impression, the use of the drill is not evenly

Even though it is clear that all the composite capitals of the basilica were produced by qualified sculptors of the same Constantinopolitan workshop, nevertheless, based on their differences in terms of design, their unfinished Raptis 2016, III, pp. 536–542. Farioli-Olivieri 1964, p. 154, fig. 4; Raptis 2016, III, pp. 557–560 18 Raptis 2016, III, pp. 557–560. 19 Kautzsch 1936, pp. 134, 252, no 428. 16

Raptis 2016, III, pp. 543–557. 15 Kautzsch 1936, pp. 134 ff., no 431; Farioli-Olivieri 1964, 154; Kramer 1968, pp. 52–53, fig. 23; Raptis 2016, III, pp. 536–542. 14

17

257

Konstantinos T. Raptis

Fig. 5. Composite capital with fine-toothed acanthus leaves: (a, c) on the south, and (b, d) on the north column of the tribelon (K. T. Raptis).

details and the lack of the drilling effect on most of them, it seems that they were not custom-designed and sculpted for Acheiropoietos – as has previously been stated20 – as was probably the case for the more manneristic ones in the Stoudios Basilica.21 On the contrary, it seems that the two subgroups, identical within each set, that were used in the nave colonnades, were probably ready-made capitals, produced for unrealised architectural projects and combined afterwards, in order to fill the order for the architectural decoration of the Acheiropoietos Basilica, which needed more column sets than the averagesized basilicas of the period. Apparently, the shipment was not accompanied by sculptors from the capital, so the drilling effect that was usually carried out on the site was never realised. The two obviously different capitals of the tribelon were probably prefabricated stock that remained unsold by the time the order for the Acheiropoietos Basilica was being processed. Thus, they were added to the same shipment and found their way to Thessalonike.22

The twenty-six impost blocks (fig. 10) of the tribelon and the nave colonnades, as well as three more that were found during excavation and attributed to the western façade of the exonarthex, have – with only one exception – similar dimensions and seem to form a homogeneous set made of finely dressed – but not polished – Proconnesian marble. Their fronts, towards the nave, are adorned with schematic wreathed christograms or crosses, flanked by rich syntheses of smooth acanthus leaves, with a great variety of patterns and details (fig. 10a–l). Their secondary faces, towards the lateral aisles, bear plain Latin crosses in relief (fig. 10m). But, despite the vast majority whose decoration points to workshops of the capital, two pieces – of which one is preserved in situ on the westernmost column of the north colonnade (fig. 10l) – are clearly different. Their acanthus leaves are more schematic, imitating insufficiently the Constantinopolitan prototypes, while their wreathed chrisms are turned into massive bosses, on which extremely schematic wreaths enclose crosses executed in low relief. The latter were probably manufactured by local craftsmen in order to supplement the already imported set.23

Kramer 1968, pp. 60–64; Sodini 1977, pp. 424–245; 1984, p. 225. Kautzsch 1936, pp. 135–136, 251, no 434, pl. 27; Kramer 1968, pp. 36 ff., fig. 15–19. 22 Raptis 2016, III, pp. 561–562. 20 21

23 Sodini 1984, p. 269, fig. 29–31; Zollt 1994, pp. 343–344, pl. 49; Raptis 2016, III, pp. 581–595.

258

The Sculptural Decoration of the Acheiropoietos Basilica (Thessalonike)

Fig 6. Composite capital with fine-toothed acanthus leaves on the south column of the tribelon: drilling details (K. T. Raptis).

Fig. 7. Composite capital with fine-toothed acanthus leaves on (a) the seventh, (b) the ninth, and (c) the fifth column of the north colonnade (counting from the west), as well as on (d) the twelfth and (e) the eleventh column of the south colonnade (K. T. Raptis).

259

Konstantinos T. Raptis

Fig. 8. Details of the fine-toothed acanthus leaves on (a) the seventh, and (b) the ninth capital of the north colonnade, and on (d) the sixth capital of the south colonnade. Ornamental motif (c) of the first capital of the north and (e) the eleventh capital of the south colonnade (K. T. Raptis).

Fig. 9. Composite capitals with fine-toothed acanthus leaves attributed to the non-preserved colonnade of the outer narthex (Raptis 2016, III, figs. 422 and 425a, after the Archives of the Ephorate of Antiquities of Thessaloniki City).

260

The Sculptural Decoration of the Acheiropoietos Basilica (Thessalonike)

Fig. 10. Impost blocks of the nave colonnades (K. T. Raptis).

The acanthus of the clearly Constantinopolitan impost blocks, which constitute the majority of the group, present similarities with analogous decoration on the Ionic impost capitals of the Stoudios Basilica24 in Constantinople, on imposts from Lechaion,25 and on analogous sculptures from Basilica A in Phthiotic Thebes.26 But the longitudinal lobes that are branched at the upper corners of the imposts and develop horizontally in order to cover the upper part of the imposts’ surface – apparently due to horror vacui – are analogous to similar acanthus motifs on the capitals in the narthex of Koča Mustafa Paşa Camiı,27 as well as on three Ionic impost capitals, today in the Hagia Sophia Museum (inv. no 170 and 501) and the Archaeological Museum of Istanbul (inv. no 5532), dated to late fifth or early sixth

century,28 pointing to a date around the year 500 for the imposts in the Acheiropoietos Basilica. Several types of mason marks have been found on the architectural sculptures of the basilica. Among them, the small-sized letter Φ (fig. 11d) – found also on the gallery cornice of the Justinianic Hagia Sophia29 – and the incised syllable zω30 (fig. 11a–c), noticed several times on bases, shafts and imposts – but not on the composite capitals – of the Acheiropoietos Basilica,31 indicate their construction by craftsmen of a certain imperial workshop that operated in Constantinople during the first decades of Zollt 1994, pp. 11–13, 344, no 5 and 9–10, pl. 4–5. Deichmann 1976, p. 221; Paribeni 2004, p. 730, no 93. 30 Deichmann 1976, pp. 220–221. 31 Sodini 1977, p. 408, fig. 59–60; 1984, pp. 269–270, fig. 29; Zollt 1994, p. 343; Raptis 1999, p. 232, fig. 6; Paribeni 2004, p. 697, n. 107; Raptis 2016, III, pp. 680–682. 28 29

Ζollt 1994, pp. 9–10, no 1–3, pl. 1–2. Sodini 1984, p. 269; Zollt 1994, 344, pl. 50d. 26 Sodini 1984, pp. 269, 272, fig. 32. 27 Zollt 1994, p. 15, no 16. 24 25

261

Konstantinos T. Raptis

Fig. 11. Mason marks (K. T. Raptis).

the sixth century, providing architectural sculptures for the erection of imperially funded monuments, including St Sergius and Bacchus32 and the Justinianic Hagia Sophia as well, where the zω mark alone is presented 87 times.33 In architectural sculptures that were in column sets, apart from bases identical to those in Acheiropoietos, such as the one in Lechaion,34 the same mason mark has been traced on Lyre Corinthian35 and Roofed-over Corinthian capitals,36 types dated to the late fifth and the early decades of the sixth century. The existence of the mason mark zω on the architectural sculptures of the Acheiropoietos Basilica, along with its presence at Justinianic foundations in Constantinople, was in the past explained as an indication of the continuous operation of that workshop from the seventh decade of the fifth until the fourth decade of the sixth century.37 But the continuous activity of the same workshop under the guidance of a sole sculptor, whose trademark was probably this mason mark, is rather paradoxical, especially since most of the other known mason marks represent a period that does not surpass one generation.

the same workshop that, at the middle of the same century, produced the capitals of the Stoudios Basilica. But the capitals of the type used in the Acheiropoietos Basilica were probably among the latest examples of this limited series of composite capitals with double, fine-toothed acanthus leaves. On the other hand, the bases, the column shafts and the imposts of the same set of architectural sculptures are probably among the earliest works of the workshop with the mason mark zω as its trademark, since the same workshop was later commissioned to produce secondary architectural sculptures for the erection of the Justinianic Hagia Sophia. The latter was probably its last large commission, since its characteristic mason mark does not appear on the mid-sixthcentury monuments in Ravenna.38 The sculptural decoration of the main nave was complemented by a series of cornices made by local workshops. Of these, only the custom-made cornice that follows the curve of the segmented apse and defines the sill of the original five-lobed window of the sanctuary is preserved in situ (fig. 12a).39 Numerous, well-maintained blocks of another, two-faced marble cornice (fig. 12b) were found during excavation. The main side of this cornice, apparently made in Thessalonike of white Thasian marble, was adorned with large five-lobed acanthus leaves that seem to imitate Constantinopolitan prototypes.40 Since the width of the lower surface of the cornice equals the masonry thickness of the nave arcades, it could be hypothetically restored in the longitudinal arcades of the central aisle, and specifically 8 m above the pavement of the nave.41

Therefore, a slight chronological inconsistency between the capitals and the rest of the column components is observable. The composite capitals with fine-toothed acanthus leaves, whose demanding, detailed decoration requires a certain degree of specialisation, were produced by sculptors of Deichmann 1976, pp. 216–217, 221, fig. 50. Deichmann 1976, pp. 216–217, 221, fig. 51, 4a; Butler 1989, p. 148; Paribeni 2004, pp. 694, 696–697, 707, 711, 722, no 33, 33a, 33b, 33c and 33d, fig. 355 and 359. 34 Deichmann 1976, pp. 220–221; Sodini 1977, p. 424, no 5; Paribeni 2004, p. 697, fig. 416. 35 Barsanti 1989, p. 134, no 170; Paribeni 2004, p. 697. 36 Barsanti 1989, p. 117, no 77, fig. 24; Paribeni 2004, p. 696. 37 Sodini 1989, p. 168; Paribeni 2004, p. 702. 32 33

Raptis 2016, III, pp. 688–690. Raptis 2016, III, pp. 623–625. 40 Raptis 2016, III, pp. 327–631. 41 Raptis 2016, III, pp. 631–632. 38 39

262

The Sculptural Decoration of the Acheiropoietos Basilica (Thessalonike)

Fig. 12. Cornices, (a) in situ at the sanctuary apse, and (b) found during excavation; probably from the nave arcades (K. T. Raptis).

The two pairs of consoles decorated in relief, which support the transverse arches of the narthex, are also products of a local workshop.42 A large number of less prominent architectural sculptures, like the imposts of the tribelon and the five-lobed window of the sanctuary apse,43 as well as a series of thresholds,44 were made of marble slabs cut from Attic or local sarcophagi. At the same time, three of the four consoles that protrude from the exterior masonry of the apse are made of a Roman architrave, cut into equal parts.45 The gallery level The architectural sculptures of the gallery level have been recently attributed to the second – seventh-century – phase of the basilica. The twelve bases of the south gallery colonnade (fig. 13) belong to the simplified variation of the Attic-Ionic type with rectangular plinth, latent, either quarter-round or flattish torus and plain ring instead of scotia and upper Raptis 2016, III, pp. 633–647; 2018, pp. 453–465. Raptis 1999, pp. 227–228, fig. 2; 2016, III, pp. 606–607, 612. 44 Raptis 2016, III, p. 652. 45 Raptis 1999, p. 231; 2016, III, pp. 347–349. 42 43

Fig. 13. The south gallery colonnade (K. T. Raptis).

263

Konstantinos T. Raptis

Fig. 14. Impost-ionic capitals: (a) restored on the easternmost column of the north gallery; (b–o) in situ on the columns of the south gallery colonnade; and (q–s) in situ on the columns of the tribelon at the east end of the north aisle (K. T. Raptis).

264

The Sculptural Decoration of the Acheiropoietos Basilica (Thessalonike) Conclusion

torus,46 while the equal number of shafts (fig. 16) are divisible into three subgroups based on the formation of their apophysis. One among them is adorned with a Latin cross in relief.47

The architectural sculptures of the gallery level, which until recently were considered to be part of the original building, have been re-dated and attributed to its second phase. At the same time, several characteristics, unobserved until now, of those at the ground level – apparently belonging to the first phase of sculptural decoration of the basilica – demonstrate that they were not tailor-made exclusively for the late fifth- to early sixth-century basilica; on the contrary, they consist of (i) ready-made architectural sculptures, made of Proconnesian marble in Constantinopolitan workshops, (ii) local copies of the Constantinopolitan sculptures, produced in order to complete the imported set with the necessary missing items, (iii) lower-quality sculptures from local workshops made of white coarse-grained marble from Thasian quarries, and (iv) secondary architectural members such as imposts and thresholds that were cut on the building site from marble sarcophagi, looted from the abandoned Late Roman cemeteries of the city.

Among the Ionic impost capitals of the gallery arcades (fig. 14), only three – one restored on the easternmost column of the north gallery (fig. 14a) and two formerly located in the courtyard of the monument, and therefore none in its original place48 – belong to the group of Ionic impost capitals whose volutes project, even slightly, from the lateral sides of the impost.49 Dated generally from the second half of the fifth to the beginning of the sixth century, they may be attributed to a period synchronous with that of the architectural sculptures of the nave colonnades. Τhe twelve capitals of the south gallery (fig. 14 b–o)50 belong to variations of the second group of Ionic impost capitals with closed contour, in which the Ionic part is diminished, compared to the impost. These capitals, whose ionic part is less than one third of its total height, can be further sorted into different subgroups, characterised by either the mediation or the absence of a listel – i.e., a diminished abacus – between the Ionic part and the impost;51 the former date generally to the first half of the sixth century, and the latter until its last quarter.52 Probably prefabricated in a standardised size, these capitals, as well as the bases and the shafts of the same colonnade, comprise a multifarious, long prefabricated set, used – or perhaps reused – during the seventh-century restoration of the basilica, when the galleries were added to the building.53

These conclusions reveal a significant imbalance in the distribution of costs for the procurement of the architectural sculptures of the basilica. At the same time, they possibly hint at the finances for the erection of the building, with a variety of funds of different scale and origin, and raise the question of whether the choice to use spolia was made for reasons of economy in a luxurious building, or the luxurious Constantinopolitan architectural sculptures were the exception, as a special donation indicative of high-end Constantinopolitan – even imperial – patronage of this major monument outside the capital of the Empire.

Analogous are the two simplified Ionic impost capitals (fig. 14q–s) of the tribelon, opening in the east wall of the north aisle,54 forming the entrance to a mid-Byzantine annex. Set on a mismatched pair of re-used columns, both are characterised by the imbalance of the Ionic part and the impost, as the first corresponds to just one quarter of their total height. The above ratio, in conjunction with the complete absence of the diminished abacus between the Ionic part and the impost, the extremely schematic volutes and the coarsely malformed ovoli that decorate the echinus, suggests a seventh-century date.55

Bibliography Barsanti C., 1989. ‘L’esportazione di marmi dal Proconneso nelle regioni pontiche durante il IV–VI secolo’, in Rivista dell’Istituto nazionale di archeologia e storia dell’arte. S. III, 12, pp. 91–220. Butler L. A, 1999. The Nave Cornices of Hagia Sophia in Istanbul, PhD Dissertation, University of Pennsylvania. Ćurčić S., 2010. Architecture in the Balkans from Diocletian to Süleyman the Magnificent, New Haven/ London.

Raptis 2016, III, pp. 513–514. On the type, see also Kramer 1970, p. 273. Raptis 2016, III, pp. 527–528. 48 Vemi 1989, pp. 18–20, 131–132, no 121–122, 124, pl. 40–41; Raptis 2016, III, pp. 570–573. 49 Zollt 1994, pp. 275–276: Gruppe 1. ‘Kapitelle mit seitlich vorkragenden Voluten’. 50 Vemi 1989, pp. 20, 131, no 123, pl. 41; Raptis 2016, III, pp. 754–580. 51 Zollt 1989, pp. 275–276, 313–320: Gruppe 2. Kapitelle mit geschlossenem Umriß. Einfachere ionische Kämpferkapitelle. Undergruppe 2.2. Kapitelle mit schwach ausgebauchter Trennleiste; 2.5. Kapitelle ohne Trennleiste. 52 It is worth mentioning that, until recently, these capitals were considered to be part of the original building, and consequently have been dated to the middle or the third quarter of the fifth century (Kautzsch 1936, p. 167; Farioli-Olivieri 1964, p. 172; Vemi 1999, pp. 19, 131). 53 Raptis 2016, III, p. 580. 54 Vemi 1989, pp. 18, 132; and K. T. Raptis’ forthcoming paper on ‘The seventh century restoration of Acheiropoietos Basilica’ (cf. supra, n. 6). 55 Raptis 2016, III, pp. 578–579. About a group of sixth and seventh century pilaster capitals at the propylon of the Acheiropoietos Basilica, see Raptis 2018b. 46

47

Deichmann F. W., 1976. Ravenna. Hauptstadt des Spätantiken Abendlandes, ΙΙ/2, Wiesbaden. Farioli-Olivieri R., 1964. ‘I capitelli paleobizantini di Salonicco’, in XI. Corso di cultura sull’arte ravennate e bizantina, Ravenna, 8–21 marzo 1964, Ravenna, pp. 132–177. Kautzsch R., 1936. Kapitellstudien. Beiträge zu einer Geschichte des spätantiken Kapitells im osten von vierten bis ins siebente Jahrhundert, Berlin/Leipzig (Studien zur spätantiken Kunstgeschichte, 9). Kramer J., 1968. Skulpturen mit Adlerfiguren an Bauten des 5. Jahrhunderts nach Chr. in Konstantinopel, Munich. 265

Konstantinos T. Raptis Raptis, K. T., 2018a. ‘Μαρμάρινοι κιλλίβαντες με ανάγλυφο διάκοσμο στα εγκάρσια τόξα του νάρθηκα της Αχειροποιήτου στη Θεσσαλονίκη’, in A. Semoglou, I. P. Arvanitidou, E. G. Gounari (eds.), Λεπέτυμνος. Μελέτες Αρχαιολογίας και Τέχνης στη μνήμη του Γεωργίου Γούναρη. Ύστερη Ρωμαϊκή, Βυζαντινή, Μεταβυζαντινή – Studies in Archaeology and Art in memory of Georgios Gounaris. Late Roman, Byzantine, Postbyzantine Period, Thessaloniki (Βυζαντινός Δόμος, 26), pp. 453–465.

Kramer J., 1970. ‘Attische Säulenbasen des 5. und 6. Jahrhunderts und ihre Rohform’, in Bonner Jahrbücher, 170, pp. 271–278. Krautheimer R., 1965. Early Christian and Byzantine Architecture, Harmondsworth/Baltimore/Ringwood (The Pelican History of Art, Z 24). Mango C., 1976. Byzantine Architecture, New York (History of the World Architecture). Orlandos A.K., 1952–54. Η ξυλόστεγος παλαιοχριστιανική βασιλική της μεσογειακής λεκάνης. Μελέτη περί της γενέσεως, της καταγωγής, της αρχιτεκτονικής μορφής και της διακοσμήσεως των χριστιανικών οίκων λατρείας από των Αποστολικών χρόνων μέχρις Ιουστινιανού, Athens (Βιβλιοθήκη της εν Αθήναις Αρχαιολογικής Εταιρείας, 35).

Raptis K. T., 2018b. ‘The Marble Pilasters and Their Pilaster-Capitals at the Meridional Propylon of the Acheiropoietos Basilica in Thessaloniki’, in F. Coden (ed.), Minima Medievalia, in Atti dell’ Accademia Roveretana degli Agiati. Serie IX, vol. VIII/A, pp. 70–77. Raptis K. T., 2019a. ‘H Αχειροποίητος Θεσσαλονίκης στο πλαίσιο της πρωτοβυζαντινής εκκλησιαστικής αρχιτεκτονικής - The Acheiropoietos Basilica within the Framework of Early Byzantine Ecclesiastical Architecture’, in Δελτίον της Χριστιανικής Αρχαιολογικής Εταιρείας, 40, pp. 29–60.

Paribeni A., 2004. ‘Le sigle dei marmorari e l’organizzazione del cantiere’, in A. Guiglia Guidobaldi, C. Barsanti (eds.), Santa Sofia di Costantinopoli. L’arredo marmoreo della Grande Chiesa giustinianea, Vatican City, 2004 (Studi di Antichità cristiana, 60), pp. 649–734.

Raptis K. T., 2019b. ‘The Building History of Acheiropoietos Basilica Reconsidered’, in Ниш и Византија – Niš and Byzantium, 17, pp. 115–129.

Raptis K. T., 1999. ‘Παρατηρήσεις επί ορισμένων δομικών στοιχείων της Αχειροποιήτου’, in Το Αρχαιολογικό Έργο στη Μακεδονία και στη Θράκη, 13, pp. 219–237.

Sodini J.-P., 1977. ‘Remarques sur la sculpture architecturale d’Attique, de Béotie et du Péloponnèse à l’époque paléochrétienne’, in Bulletin de correspondance hellénique, 101, pp. 423–450.

Raptis K. T., 2016. Αχειροποίητος Θεσσαλονίκης. Αρχιτεκτονική και γλυπτός διάκοσμος, PhD Dissertation, Αριστοτέλειο Πανεπιστήμιο Θεσσαλονίκης.

Sodini J.-P., 1984. ‘La sculpture architecturale à l’époque paléochrétienne en Illyricum’, in Πρακτικά του 10ου διεθνούς συνεδρίου χριστιανικής αρχαιολογίας, Θεσσαλονίκη 28 Σεπτεμβρίου–4 Οκτωβρίου 1980 – Actes du Xe congrès international d’archéologie chrétienne, Thessalonique, 28 septembre–4 octobre 1980, Vatican City/Thessaloniki (Studi di Antichità cristiana, 37; Ελληνικά. Παράρτημα, 26), pp. 31–117.

Raptis K. T., 2017a. ‘Αχειροποίητος Θεσσαλονίκης: επανεξετάζοντας την αρχιτεκτονική και την οικοδομική ιστορία της πρωτοβυζαντινής βασιλικής’, in Τριακοστό ‘Eκτο Συμπόσιο Βυζαντινής και Μεταβυζαντινής Αρχαιολογίας και Τέχνης. Αφιέρωμα στον Χαράλαμπο Μπούρα. Πρόγραμμα και Περιλήψεις Εισηγήσεων και Ανακοινώσεων, Αθήνα, 12, 13, και 14 Μαΐου 2017, Βυζαντινό και Χριστιανικό Μουσείο, Βασ. Σοφίας 22, Αθήνα – Thirty-Seventh Symposium on Byzantine and Post-Byzantine Archaeology and Art. Tribute to Charalambos Bouras. Programme and Abstracts of Major Papers and Communication, Athens, 12, 13 and 14 May 2017, Byzantine and Christiana Museum, Vas. Sofias 22, Athens, Athens, pp. 112–114.

Sodini J.-P., 1989. ‘Le commerce des marbres à l’époque protobyzantine’, in C. Morrisson, J. Lefort (eds.), Hommes et richesses dans l’Empire byzantin, I, Paris (Réalités byzantines), pp. 163–186. Vemi V., 1989. Les chapiteaux ioniques à imposte de Grèce à l’époque paléochrétienne, Athens (Bulletin de correspondance hellénique. Supplément, 17).

Raptis K. T., 2017b. ‘Αρχιτεκτονικά έργα του Αναστασίου Α΄ στη Θεσσαλονίκη’, in Διεθνές συμπόσιο προς τιμήν του ομότιμου καθηγητή Γεωργίου Βελένη, Θεσσαλονίκη, Αμφιθέατρο Αρχαίας Αγοράς, 4–7 Οκτωβρίου 2017. Περιλήψεις ανακοινώσεων – International Symposium in Honour of Professor Emeritus George Velenis, Thessaloniki, Amphitheatre of Ancient Agora, 4–8 October 2017. Abstracts, Thessaloniki, p. 46.

Zollt Th., 1994. Kapitellplastik Konstantinopels vom 4. bis 6. Jahrhundert n. Chr., mit einem Beitrag zur Untersuchung des ionischen Kämpferkapitells, Bonn (Asia Minor Studien, 14).

Raptis K. T., 2017c. ‘Η δομική αποκατάσταση της Αχειροποιήτου κατά τον 7ο αιώνα και η σημασία της για την αστική συνέχεια της Θεσσαλονίκης κατά τους σκοτεινούς αιώνες’, in I. Varalis, F. Karagianni (eds.), Κτίτωρ. Αφιέρωμα στον δάσκαλο Γεώργιο Βελένη, Thessaloniki, pp. 289–306. 266

20 I capitelli imposta del tipo cosiddetto a pannelli nell’ambito della produzione scultorea postgiustinianea Claudia Di Bello Sapienza Università di Roma (Italy) All’indomani del lungo regno di Giustiniano il panorama della scultura bizantina appare alquanto problematico. Nonostante lo scenario dei tre secoli che seguono sia disomogeneo e frammentario, con un esiguo numero di testimonianze datate o databili e per lo più decontestualizzate, è stato possibile evincere che le tipologie di capitelli di tradizione classica furono gradualmente soppiantate da quelli a imposta. In particolare, sembrano aver riscontrato notevole fortuna quelli del tipo cosiddetto a pannelli, categoria già diffusa nel VI secolo e ampiamente documentata nei secoli di transizione da una serie di esemplari. Si tenterà, pertanto, di rintracciare le metamorfosi e le alterazioni che caratterizzarono le forme e i decori di questa categoria di capitelli durante il periodo di transizione, facendo riferimento soprattutto alle testimonianze costantinopolitane, e in particolar modo ad alcuni interessanti esempi inediti che contribuiscono, peraltro, a ridurre la carenza di testimonianze dalla capitale bizantina in questo ambito. Au lendemain du long règne de Justinien, le panorama de la sculpture byzantine s’avère assez problématique. Malgré l’absence d’homogénéité et l’état fragmentaire des connaissances sur les trois siècles suivants, en raison du très petit nombre des témoins datés ou datables, on a pu déduire que les chapiteaux de tradition classique ont été progressivement supplantés par des chapiteaux à imposte. On peut notamment remarquer le succès considérable du type dit « à encadrement », déjà répandu au VIe siècle et bien attesté pendant les siècles de transition par une série d’exemplaires. On essayera donc de tracer les métamorphoses et les modifications caractérisant les formes et décors de ce type de chapiteau entre le VIIIe et le IXe siècle, en faisant référence surtout aux témoignages de Constantinople. Parmi ceux-ci, on examinera d’intéressants exemples inédits qui permettent de combler les lacunes des témoignages provenant de la capitale byzantine dans ce domaine. During the period after Justinian’s long reign, the overview of Byzantine sculpture becomes quite problematic. Knowledge of the following three centuries is uneven and fragmentary, based on a small number of dated or datable pieces, most of which lack context. Despite that difficulty, it has been possible to deduce that the typologies of capitals in the classical tradition were gradually replaced by the impost capital. In particular, this chapter highlights the remarkable fortune of the so-called ‘panelled’ type (or framed type), a category already widespread in the sixth century and widely documented in the transitional period by a series of examples. Starting with these examples, this chapter sets out to track down the metamorphosis and alterations that characterised forms and decorations of this category of capital during the age of transition, referring to the Constantinopolitan evidence and in particular to some interesting unpublished pieces that fill the gap of evidence from the Byzantine capital in this sphere. Für die Zeit nach der langen Regierungszeit Justinians ist die Beurteilung der byzantinischen Bauplastik problematisch. Unser Wissen über die nachfolgenden drei Jahrhunderte ist inhomogen und fragmentarisch und basiert auf einer kleinen Anzahl von datierten oder datierbaren Funden, die meisten ohne Befundkontext. Dennoch lässt sich zeigen, dass ältere Kapitelltypen durch Kämpferkapitelle abgelöst und ersetzt wurden. Eine besondere Beliebtheit scheinen gerahmte Kämpferkapitelle genossen zu haben. Dieser Typ war bereits im 6. Jh. verbreitet, in der Folgezeit wurde er aber besonders häufig verwendet. Dieser Beitrag beschreibt die Form und den Dekor

267

Claudia Di Bello dieses Kapitelltyps und versucht seine Veränderung vom 7. bis zum 9. Jh. zu rekonstruieren. Ein besonderes Augenmerk liegt auf einigen hier erstmals publizierten Kapitellen aus Konstantinopel, wo bisher Kämpferkapitelle dieser Zeit kaum bekannt geworden sind. La scultura architettonica prodotta nell’età successiva al lungo regno di Giustiniano è un campo d’indagine in larga parte non ancora esplorato in modo sistematico, pertanto sfugge la possibilità di metterne a fuoco le ‘eredità’ e di inquadrare gli schemi decorativi e le tecniche che perdurarono in quella che si può definire come età di transizione, dall’exploit produttivo della prima metà del VI secolo alla nascita di nuove forme nei decenni successivi alla fine dell’iconoclastia, nella seconda metà del IX.

Quella a pannelli è appunto la tipologia di capitello prediletta dalla scultura dell’età di transizione. Sfortunatamente, però, le testimonianze sono solo in minima parte datate, mentre il maggior numero di esempi rimane sospeso in un vago ambito cronologico che comprende il VII, l’VIII e il IX secolo6. Al gruppo già noto possono ora essere aggiunti alcuni esemplari in area costantinopolitana che contribuiscono, peraltro, ad arricchire il panorama, in verità piuttosto scarno, della capitale, nonché a delineare meglio il ruolo da essa svolto per la scultura dei secoli di transizione.

Andrebbero, dunque, meglio comprese le ragioni per le quali sembra esaurirsi proprio l’aspetto più eclatante della scultura di età giustinianea, quello che si esprime nelle straordinarie invenzioni per i decori plastici delle chiese costantinopolitane di S. Polieucto, dei SS. Sergio e Bacco e di Santa Sofia, e che è caratterizzato da una esuberante ricchezza del repertorio ornamentale. Del resto, insieme a quest’ultimo già coesisteva nello stesso torno di anni un linguaggio antitetico, basato su forme semplificate e decorazioni ridotte all’essenziale1.

Al modello dell’Hebdomon, caratterizzato da un decoro essenziale, sembra ispirarsi, ad esempio, un capitello rinvenuto nell’area di scavo di Yenikapı, collocato, insieme ad altri materiali provenienti da vari cantieri della città, all’esterno del Museo Archeologico7 (figg. 1-2). Il capitello, che reca i segni evidenti di una lunga permanenza in ambiente marino su metà della superficie, è caratterizzato dalla presenza, su due dei pannelli, di un clipeo con monogramma interpretabile rispettivamente con i nomi di Costantino e Anastasia e, negli altri spazi, di una croce a estremità patenti su globo8. Il medesimo impaginato ricorre nei capitelli del vicino Ospizio di Samson, riconducibili agli ultimi anni del regno di Giustiniano (564-565), dove, però, ai clipei monogrammatici si sostituisce un disco liscio, forse predisposto per la medesima funzione9. L’alternanza di un nome femminile e di un nome maschile nello scioglimento dei monogrammi evoca senza dubbio l’esempio di Giustiniano e Teodora al quale non è escluso si siano riferite anche altre coppie imperiali. Questa suggestione lascerebbe spazio a due possibilità: Anastasia, infatti, è sia il nome acquisito da Ino, consorte di Costantino Tiberio (578-582), dopo la salita al trono del marito che, stando alle fonti, si fece promotore del restauro di molte chiese della capitale10, sia il nome della moglie di Costantino IV (668-685)11.

Quest’ambivalenza lessicale interessò anche il decoro dei capitelli a imposta, una tra le creazioni più originali della prima metà del VI secolo che, come ha difatti cercato di mettere in evidenza Martin Dennert nel suo ampio studio dedicato ai capitelli mediobizantini, fu, tra tutte, la tipologia che riscontrò in seguito maggiore diffusione2. In particolare, proprio nel VI secolo, sembrano aver avuto notevole fortuna i capitelli imposta del tipo cosiddetto a pannelli; anche in questo caso, accanto a una versione dal decoro ricco e variegato3 sembra coesistere e avere maggior seguito un’interpretazione opposta, caratterizzata da un lessico estremamente semplificato, che trova un ideale modello nel ben noto capitello siglato dai monogrammi di Giustiniano e Teodora proveniente dall’area dell’Hebdomon, oggi nel Museo Archeologico4. In genere, nello spazio trapezoidale dei pannelli, incorniciato da lisce modanature o da eleganti serti di alloro, trovano posto solo dischi, gigli, croci, foglie e clipei, con o senza monogrammi5.

Kramer 1988; Zollt 1994, pp. 96-104; Dennert 1997, pp. 39-45. Il numero apposto in fase di scavo è YKM.13.18014, ed è certo riferibile all’area e alla stratigrafia di provenienza. Misure in cm: h 60, l max 68, Ø base 45, Ø clipeo 22. In occasione dello stesso Congresso di Belgrado il capitello è stato segnalato anche dalla prof.ssa Asnu–Bilban Yalçın durante la sua comunicazione dal titolo New Finds in Istanbul: Archaeological Evidence for Sculptural Production and Marble Trade in Early Byzantine Constantinople. 8 Un referente per il monogramma con il nome di Costantino è rintracciabile in Zacos, Veglery 1972, cat. 308, p. 378, tav. 237 n. 296, in cui tuttavia risultano invertiti l’ω e il k, mentre per il monogramma di Anastasia si veda ivi, cat. 1701, p. 978, tav. 229 n. 20, anche se di genere maschile. 9 Tezcan 1989, figg. 150-153; Zollt 1994, nn. pp. 247-249, p. 99, tav. 35. 10 Teofane Confessore, Chronographia, A.M. 6071 (ed. de Boor 1883, pp. 249.26-250.6). Lo stesso Tiberio decise di assumere ufficialmente il nome di Costantino nel 580/81, cf. ibid. AM 6073 (ed. p. 251.16-19). 11 Si veda Teofane Confessore, Chronographia, A.M. 6186 e 6203 (ed. de Boor 1883, pp. 367.20 e 380.19). I nomi di Costantino IV e di Anastasia, inoltre, figurano insieme a quello di Costante II in un sigillo databile al 654-659, nel quale, tuttavia, l’identità dell’imperatrice è incerta, cf. Zacos, Veglery, 1972, n. 19, pp. 20-21, tav. 12. 6 7

1 Sul divario lessicale presente nella scultura giustinianea si è più volte soffermata Claudia Barsanti alla memoria della quale, nel frattempo scomparsa, tengo a dedicare con affetto questo scritto (cf. Barsanti 1989, p. 182; e 1993, p. 205). 2 Dennert 1997, pp. 35-134. 3 Sono esemplificativi i capitelli conservati nel Museo Archeologico di Istanbul per i quali si veda Zollt 1994, nn. 243-244, 246, pp. 97-99, ai quali si deve aggiungere un quarto esemplare segnalato da Barsanti 2006, p. 42, n. 39, studio che inoltre offre una breve collazione dei capitelli a pannelli ornati a giorno, ancora mai sistematicamente censiti. 4 Inv. n. 1239. Misure in cm: h 50; l max 65; Ø 41,5. Cf. Mendel 1914, n. 1240, p. 465; Kautzsch 1936, n. 618, p. 191, tav. 38; Kramer 1988, pp. 183-184, figg. 6-7; Zollt 1994, n. 250, p. 100. 5 Kramer 1988, pp. 182-184; Barsanti 1993, p. 205; Dennert 1997, pp. 39-48, 175.

268

I capitelli imposta del tipo cosiddetto a pannelli nell’ambito della produzione scultorea postgiustinianea

Fig. 1. Istanbul, Museo Archeologico, esterno (Osman Hamdi Bey Yokuşu), capitello a pannelli dagli scavi di Yenikapı da N-E (autore).

Fig. 2. Istanbul, Museo Archeologico, esterno (Osman Hamdi Bey Yokuşu), capitello a pannelli dagli scavi di Yenikapı da S-O (autore).

L’ampiezza dell’intervallo cronologico tra le eventuali identificazioni e le affinità con il modello dell’Hebdomon, permetterebbero di esprimere una preferenza per la prima possibilità, ma, come vedremo, questa variante perdurerà inalterata anche oltre il VII secolo. Ad esempio, la medesima alternanza di motivi caratterizza i due capitelli in opera nelle finestre degli ambienti orientali delle gallerie della Santa Irene (fig. 3)12. La loro possibile appartenenza, sostenuta da Dennert, alla fase di ricostruzione della chiesa promossa da Costantino V in seguito ai danneggiamenti dovuti al terremoto del 740, può essere avvalorata dal confronto tra i clipei presenti nei pannelli laterali, che ospiterebbero i resti di un monogramma13, e quelli della lastra reimpiegata, in due metà, nel pavimento dello stesso edificio, senz’altro riconducibile a quel restauro grazie alla presenza, tra gli altri, del monogramma dello stesso Costantino14. I clipei, difatti, sono analogamente definiti da due listelli concentrici di diverso spessore, e non è da escludere che, così declinati, siano stati scelti per siglare, oltre la lastra, anche altri elementi, come i capitelli, dell’apparato scultoreo della chiesa nella sua ristrutturazione di VIII secolo.

Fig. 3. Istanbul, Santa Irene, capitello a pannelli in opera nell’ambiente orientale della galleria nord (C. Barsanti 2017).

un decoro ridotto all’essenziale e per il ruolo sempre più predominante del clipeo monogrammatico, è testimoniata da un buon numero di esempi, anche fuori dalla capitale: fino alla fine del VII secolo la loro produzione è garantita da alcuni esemplari ben datati, vale a dire il capitello con il monogramma di Giustino II (565-578) a Büyükada15, la maggiore delle Isole dei Principi, il grande capitello

L’ampia diffusione nei secoli di transizione di questa variante dei capitelli a pannelli, che si caratterizza per Dennert 1997, n. 70, pp. 40, e 187, tav. 13. Già Kramer 1988, p. 183, n. 18, aveva segnalato, seppure con incertezza, la presenza dei monogrammi. Più sicuro della loro presenza, avendone rintracciato dei lacerti, è Dennert 1997, p. 40. Da quanto mi è stato possibile constatare recentemente (ottobre 2018), benché da una certa distanza, mi sembra che dei monogrammi non sia rimasta traccia. 14 Ulbert 1969, n. 1, pp. 61-62, tav. 40a-c; 1969-70, n. 38, pp. 349-350, 357, tav. 72; Di Bello 2019, in part. tav. II, fig. 3. 12 13

Feissel 1987, n. 43, p. 427. Büyükada, l’antica Prinkipo, è anche il luogo dove secondo le fonti (cf. Teofane Confessore, Chronographia, A.M. 6062 [ed. de Boor 1883, p. 243]) Giustino II fece edificare una sua residenza, alla quale dunque il capitello può essere plausibilmente ricondotto.

15

269

Claudia Di Bello di marmo di Dokimion nel Museo di Afyonkarahisar in Frigia, da ricondurre all’imperatore Foca (602-610)16, e il capitello frammentario ritrovato durante gli scavi della Kalenderhane Camii, appartenente all’arredo della bemachurch, con monogramma inciso su un disco rilevato17; a questi esemplari datati si può inoltre affiancare un gruppo di altri capitelli costantinopolitani oggi nel Museo Archeologico, che, pur privi di una precisa cronologia, probabilmente estesa in certi casi anche all’VIII secolo, confermano il prolungarsi del successo di questa variante18. Si potrebbe, inoltre, aggiungere al gruppo un capitello di piccole dimensioni conservato nel giardino del Patriarcato Ecumenico di Istanbul, nel quartiere di Fener, con cornici modulate da semplici modanature e croci patenti alternate ai clipei monogrammatici (fig. 4) 19.

Fig. 4. Istanbul, giardino del Patriarcato Ecumenico, capitello a pannelli (A. Guiglia 2010).

A questa variante con modanature lisce se ne affianca un’altra, che utilizza per le incorniciature una più elaborata soluzione decorativa con eleganti rami d’alloro. Essa è testimoniata da quello che possiamo ritenere il primo referente cronologico certo, non solo per i capitelli a pannelli ma per tutta la scultura postgiustinianea, referente rappresentato dai capitelli dell’esedra di Yalova Termal, l’antica Pythia di Bitinia, siglati dai monogrammi dell’imperatore Giustino II e della moglie Sofia, e quindi datati tra il 565 e il 578 (fig. 5)20. I pannelli accolgono una grande foglia di vite a cinque lobi che, su due lati tra loro opposti ospita un clipeo contenente il monogramma con il nome dell’imperatore o dell’imperatrice. Lo schema adottato testimonia il perdurare, oltre il periodo giustinianeo, del tema assai comune della grande foglia di vite pentalobata21, che, almeno per quel che riguarda i capitelli a pannelli, sembra aver trovato nel corso della prima metà del VI secolo particolare successo anche

Fig. 5. Yalova Termal (Pythia di Bitinia), esedra, capitello a pannelli con monogramma di Giustino II (A. Guiglia 1990).

in abbinamento alle cornici con foglie di alloro22. Fa tuttavia eccezione quello che per impaginato può ritenersi il referente più vicino ai capitelli di Yalova, vale a dire uno splendido capitello recentemente acquisito dal Museo Archeologico di Istanbul, nonostante in questo caso le cornici siano decorate con un sinuoso tralcio con foglie cuoriformi contenuto in due sottili listelli, mentre le foglie d’alloro ornano solo il collarino (fig. 6)23. La foglia di quello che evidentemente era il lato principale reca al centro un clipeo contenente un monogramma a legatura cruciforme che potrebbe essere sciolto con il nome di Eugenio24. Per le sue peculiarità stilistiche, contraddistinte

Asutay-Effenberger 2002. Misure in cm: h 100 ca.; l max 110; Ø clipeo 44; l max specchiatura 88. Il capitello fu ritrovato alla metà degli anni ’80 a Şuhut, l’antica Synnada, e si ruppe in due frammenti durante il trasporto al Museo. Anche se danneggiate, le lettere dei monogrammi clipeati al centro dei due pannelli superstiti possono essere sciolte con il nome di Foca. 17 Peschlow 1997, cat. 165, pp. 109-110, 112, tav. 134 ; 2007, cat. 165, pp. 296, 301, 328. Misure del frammento in cm: l 13,5; L 8,5; h 10,5. Il capitello, insieme ad altre sculture ugualmente siglate dal monogramma di Markella, doveva far parte dell’arredo liturgico della prima fase della chiesa, circorscritta alla fine del VII secolo, per la quale cf. Striker, Kuban 1997, pp. 45–58. 18 Cf. Dennert 1997, nn. 78, 86-87, pp. 43-45, 188-189, tavv. 14-15, con bibliografia precedente. Lontano da Istanbul si segnala un capitello dalla terrazza del tempio di Atena a Pergamo: cf. Otten 2010, p. 821, fig. 8. 19 Misure stimate: h 20 cm ca., l max 30 cm ca.. I due monogrammi possono essere sciolti con un’invocazione alla Madre di Dio da parte di un certo Cosma (“Θ[Ε]ΟΤΟΚ[Ε Β]ΟΗΘ[Ε]Ι KOCMA”). L’interpretazione del monogramma si avvale di un esempio analogo in Zacos, Veglery 1972, I, n. 404, p. 415, pl. 237, n. 282, benché K e M siano invertite. 20 Ai quattro capitelli superstiti, uno dei quali frammentario, è dedicato l’articolo di Kramer 1988. Misure in cm: h 54; l max 70; Ø 41. 21 Sul tema e sul suo impiego nelle diverse tipologie di manufatti scultorei, tra cui cornici, imposte, capitelli ionici a imposta e capitelli imposta con pigne angolari (Pinienzapfenkapitelle) si è recentemente soffermata Guiglia Guidobaldi 2010. Per i capitelli con pigne angolari si veda in particolare Krumeich 1997. 16

Si vedano ad esempio i due capitelli del Museo Archeologico di Istanbul: il primo, inedito, è esposto nel cortile, nei pressi della caffetteria, e reca una foglia al centro dei pannelli tranne che in un caso, in cui essa lascia il posto a un clipeo vuoto; per l’altro, caratterizzato da straordinari effetti plastici e oggi nei depositi, si vedano Ebersolt 1921, pp. 30-32, tav. XXIV,2; Tezcan 1989, fig. 480. Non lontano dalla capitale, si segnalano due capitelli a Bursa, l’uno conservato al Museo Archeologico e l’altro reimpiegato nella Türbe di Murad II, per i quali si vedano Barsanti 1995, p. 60, fig. 8; Guiglia Guidobaldi 2010, pp. 114-115, fig. 13. 23 H 62 cm, l max 85 cm; Ø monogramma 13 cm. 24 Cf. l’analogo esempio offerto dal sigillo in Zacos, Veglery 1972, n. 609, p. 499, tavv. 65, 232, n. 137. 22

270

I capitelli imposta del tipo cosiddetto a pannelli nell’ambito della produzione scultorea postgiustinianea da una spiccata resa naturalistica dei morfemi vegetali, sembra verosimile orientarne la datazione alla prima metà del VI secolo. Rispetto a quest’esempio più antico, i capitelli di Yalova mostrano una sensibile distanza dovuta in particolar modo alla marcata rigidità che caratterizza l’intaglio delle foglie e alla grafica frammentazione dei lobi. Tra le sculture di Yalova e i loro modelli può essere inserito il capitello, assai lacunoso, che si trova erratico nell’Hüseyin Ağa Medrese, antistante la chiesa dei SS. Sergio e Bacco, nel quale un processo di semplificazione degli stilemi vegetali sembra già avviato ma non ancora risolto (fig. 7) 25.

Fig. 6. Istanbul, Museo Archeologico, esterno (Osman Hamdi Bey Yokuşu), capitello a pannelli (autore).

La medesima tendenza è rintracciabile anche lontano dalla capitale in alcuni capitelli a pannelli con cornici lisce o decorate coinvolti nell’ampio fenomeno dell’esportazione di manufatti di marmo proconnesio, come ad esempio quello nel Lapidario di El Merj26, in Cirenaica, del tutto simile al frammento presso i SS. Sergio e Bacco, o il piccolo esemplare dalla basilica Γ di Peyia a Cipro datato alla seconda metà del VI secolo27, sovrapponibile al capitello conservato nell’area del monastero cretese di Paljani/Temene28, e quello di Mokva, in Abkhazia29. Attestato il prolungarsi della sua diffusione nella seconda metà del VI secolo, il tema della foglia di vite sembra scomparire dal registro decorativo dei capitelli a pannelli dei secoli successivi e la componente prettamente ornamentale nell’aggiornamento del repertorio si ridurrà, ove presente, al decoro variamente declinato delle cornici e alla presenza sempre più preponderante dei monogrammi.

Fig. 7. Istanbul, Küçük Aya Sofya Camii, esterno (Hüseyin Ağa Medrese), frammento di capitello a pannelli (A. Guiglia 2010).

Può dunque essere esemplificativo il caso di un capitello con due monogrammi conservato nei depositi del Museo Archeologico di Istanbul (inv. n. 5506), menzionato tra le nuove acquisizioni del 1962/63 come proveniente da Sultanahmet (fig. 8)30. I pannelli sono incorniciati da rami di alloro che, sebbene resi in maniera piuttosto schematizzata, non sembrano discostarsi poi molto dai già citati esempi della seconda metà del VI secolo sui quali ci siamo appena soffermati. Grazie al numero 25 Zollt 1994, n. 254, p. 101, che propone una condivisibile datazione al terzo quarto del VI secolo. Misure in cm del frammento: h 45; l max 57. 26 Ward Perkins, Goodchild 2003, p. 261, fig. 204; Guiglia Guidobaldi 2010, p. 115, fig. 14. La provenienza del capitello è ignota, esso è lacunoso su un lato e misura 45 cm di lato all’altezza dell’abaco, e 34 cm di altezza. Al centro dei pannelli conservati si alternano una croce e una foglia di vite. 27 Raptis, Vasiliadou 2005, n. 57, pp. 209, 219, figg. 17-18. Misure in cm: h 40; Ø 28; l max 39. Il capitello è stato ritrovato durante gli scavi della basilica Γ all’altezza del tribelon, ed è oggi esposto nel lapidario dell’area archeologica di Agios Georgios. 28 Sanders 1982, pp. 107-108, tav. 53. 29 Khroushkova 2006, pp. 138-139, tav. 107c, datato al periodo giustinianeo o postgiustinianeo. 30 İstanbul Arkeoloji Müzeleri yıllığı, 11-12, 1964, p. 6. Misure in cm: l 71; h 45, che tuttavia deve essere ritenuta parziale poiché la parte del collarino è lacunosa. Il capitello è stato conservato per diversi anni nel deposito esterno accanto al Çinili Köşk.

Fig. 8. Istanbul, Museo Archeologico, deposito, capitello a pannelli, inv. n. 5506 (Centro Documentazione Storia Arte Bizantina, Sapienza Università di Roma, foto 1982).

inventariale il capitello può essere identificato con quello che Zollt associa, per dimensioni e sovrapponibilità dei monogrammi, alla coppia di esemplari esposta al centro della corte dello stesso Museo (inv. nn. 5447-5448), ritrovata nel 1958/59 nelle vicinanze della Nakılbent Camii, e quindi proveniente dalla stessa area di Sultanhamet 271

Claudia Di Bello (figg. 9-10)31. Ognuno dei tre capitelli è difatti siglato dalla stessa coppia di monogrammi i quali sono disposti simmetricamente al centro di pannelli tra loro opposti. La lettura si ripete dunque identica in ognuno degli esemplari e restituisce al genitivo il nome di un personaggio, Narses (“NΑΡCΟY”), e la sua carica, interpretata da Zollt con il titolo di protospatario (“Α’CΠΑΘΑΡΙΟΥ”), che fornisce allo studioso gli elementi utili per una identificazione con il generale di Giustiniano morto intorno al 574. Gli altri pannelli sono in due casi attraversati verticalmente da una fascia liscia e rilevata32 (inv. nn. 5448, 5506, cf. figg. 8-9), e nell’altro ospitano un clipeo con cristogramma (inv. n. 5447, cf. fig. 10).

Fig. 9. Istanbul, Museo Archeologico, capitello a pannelli, inv. n. 5448 (autore).

Le incorniciature dei due capitelli nella corte offrono un’alternativa all’utilizzo dei rami di alloro: per un esemplare è stato scelto un motivo con nastro intrecciato (inv. n. 5448, cf. fig. 9), piuttosto inusuale nei capitelli prima della fine del VII secolo33 e, per l’altro (inv. n. 5447, cf. fig. 10), una sequenza di orbicoli annodati all’interno dei quali trovano posto fiori a quattro petali e motivi gigliati, un decoro in effetti assai diffuso in età giustinianea, come mostrano il fregio sui frammenti di una cornice dall’area del Grande Palazzo, oggi nel giardino del Museo del Mosaico34, e numerosi esempi offerti dall’arredo scultoreo della Santa Sofia tra i quali basterà ricordare le cornici lavorate a giorno che inquadrano le lastre del rivestimento marmoreo nell’area del bema 35. Martin Dennert, con il proposito di posticipare all’VIII secolo la datazione dei due capitelli (inv. nn. 5447-5448, cf. figg. 9-10), interpreta la varietà del decoro delle cornici, che rende davvero singolare questo gruppo di sculture, come il segnale di un allontanamento da motivi piuttosto consueti nel VI secolo36. Lo studioso, tuttavia, non tiene conto del terzo capitello con serto d’alloro (inv. n. 5506), che tra i tre del gruppo è quello che stilisticamente lascia meno dubbi circa una datazione alla seconda metà del VI secolo, e, pur accettando la lettura dei monogrammi proposta da Zollt, subordina la sua ipotesi di cronologia alla proposta di datazione all’inizio dell’VIII secolo della prima attestazione nelle fonti della carica di protospatario37.

Fig. 10. Istanbul, Museo Archeologico, capitello a pannelli, inv. n. 5447 (autore).

VII secolo38, l’interpretazione del monogramma che fa riferimento alla carica di Narses dovrebbe essere a mio parere ridiscussa poiché leggendo una sola volta l’alfa esso può essere differentemente risolto con la più semplice qualifica di spatario (“CΠΑΘΑΡΙΟΥ”). Questa lettura amplierebbe di nuovo il range cronologico dei tre capitelli al VI secolo39, alla seconda metà del quale essi possono essere attribuiti stilisticamente, e permetterebbe perfino di ipotizzare un’identificazione di Narses con l’omonimo spatario, nonché uomo di fiducia di Giustino II (565-578), menzionato appunto con questo titolo insieme all’imperatore e all’imperatrice Sofia nell’iscrizione celebrativa del restauro delle mura teodosiane, ancora in opera nella Mevlevihane Kapı, l’antica porta di Rhesion40.

A prescindere dal momento in cui venne introdotto il titolo di protospatario, forse da anticipare alla fine del İstanbul Arkeoloji Müzeleri yıllığı, 9, 1960, pp. 7, 60; Zollt 1994, nn. 251-253, pp. 100-101, fig. 7; il capitello inv. n. 5506 è brevemente menzionato in Barsanti 1996, pp. 61-62; Dennert 1997, n. 76, pp. 43, 188, tav. 14. Misure in cm inv. n. 5448: h 58; Ø 45; l max 74. Misure in cm inv. n. 5447: h 57; Ø 45; l max 73. 32 Per la sua funzione si veda Sodini 1982. 33 Un confronto può essere offerto da due piccoli capitelli con monogrammi oggi perduti appartenenti forse all’antica recinzione della Koimesis di Nicea: Dennert 1997, n. 75, pp. 42, 188, tav. 14, con bibliografia precedente. 34 Brett et al. 1947, tav. 10,1-2; Barsanti, Guiglia 2010, pp. 119-121, figg. 137-138. 35 Russo 2011, pp. 41-46, figg. 50-57. Per gli altri esempi offerti dall’arredo scultoreo della Grande Chiesa cf. 2014-15, figg. 2-4, 7, 12, 17-18. 36 Dennert 1997, n. 76, pp. 43, 188, tav. 14. 37 Kazhdan, Cutler 1991. 31

La produzione di capitelli a pannelli con cornici decorate prosegue anche nei secoli successivi, come dimostra un Secondo Ševčenko 1970, p. 5, il primo riscontro storico della carica di protospatario andrebbe rintracciato nella biografia di papa Sergio (687701) del Liber Pontificalis 86.7 (ed. Duchesne 1955, p. 373). 39 Kazhdan 1991a. 40 Oltre alla carica di spatario, a Narses è attribuita nella stessa iscrizione anche quella di sacellario. Sull’iscrizione si veda Ševčenko 1970, che si occupa in particolare della figura di Narses a pp. 4-5. Per la Mevlevihane Kapı, cf. anche Asutay-Effenberger 2007, pp. 35-37, 73-76, figg. 14, 42, 78-81. 38

272

I capitelli imposta del tipo cosiddetto a pannelli nell’ambito della produzione scultorea postgiustinianea esempio che costituisce soprattutto un saldo riferimento cronologico, vale a dire il capitello del Museo Bizantino di Atene proveniente dall’Acropoli e siglato dai quattro monogrammi di una formula invocativa alla Theotókos che vede protagonista l’imperatrice Irene, appellata “Augusta”, e quindi datato intorno al 780, o forse meglio tra 769 e 79741. È caratterizzato da incorniciature decorate con un tralcio di foglie cuoriformi, analogo nella forma a quello del già citato capitello del Museo Archeologico con il monogramma di Eugenio (cf. fig. 6), ma da esso certamente distante nella resa stilistica. Si potrebbe semmai proporre, d’accordo con Dennert, un confronto per il decoro delle cornici con un capitello da Barbaros, in Tracia, dove boccioli trilobati si alternano alle foglie a cuore42.

Fig. 12. Istanbul, Museo Archeologico, frammento di capitello a pannelli, inv. n. 912, particolare dell’incorniciatura (autore).

spiraliformi che terminano con foglie alternatamente a cuore e trilobate, simile a quello che orna le cornici del già menzionato capitello di Barbaros. Il tralcio accenna in quattro punti equamente distanti ad un’estensione in verticale, delineando dunque l’originario impaginato della struttura decorativa a pannelli (fig. 12). Al centro di ogni pannello resta un clipeo monogrammatico sovrastato da un elemento identificabile con il piede e un tratto del braccio verticale di una croce a estremità patenti. In tutti e quattro i lati, il monogramma si scioglie con la parola “BOHΘEI”, unico elemento sopravvissuto della consueta formula invocativa in cui la richiesta di protezione veniva espressa con le parole KYRIE BOHΘEI o THEOTOKE BOHΘEI, appunto in forma monogrammatica, seguite dal nome del personaggio, come nel citato capitello di Irene ad Atene. Una soluzione analoga è offerta da un altro esemplare, nei depositi del Museo Archeologico di Istanbul (inv. n. 5461) e proveniente dall’area dei SS. Sergio e Bacco, con l’invocazione di un certo Costantino patrizio, sul quale viene proposto l’abbinamento tra croci, in secondo piano, e clipei monogrammatici, sovrapposti con effetto prospettico44.

Nell’ambito dei capitelli con cornici decorate mi sembra significativa la testimonianza del frammento di straordinarie dimensioni conservato nella corte del Museo Archeologico di Istanbul (inv. n. 912)43 (fig. 11). Il frammento, può essere a prima vista confuso con un informe elemento lapideo poiché quasi tutta la sua superficie è stata scalpellata e perché la collocazione in effetti non ne favorisce un esame accurato. Tuttavia ciò che resta della decorazione permette di riconoscervi la parte inferiore di un capitello a pannelli. Il collarino è decorato da un tralcio vegetale accentuatamente ondulato, dal quale si dipartono girali

Ma, a differenza di questi esemplari, dove ogni pannello ospita un solo monogramma dei quattro che compongono la formula, nel colossale capitello (inv. n. 912) sembra che l’invocazione si ripetesse su ognuno dei quattro pannelli. Si può ipotizzare che il soggetto dell’invocazione fosse la croce stessa, che emerge appunto al di sopra dei monogrammi, altrimenti si dovrebbe supporre che la formula fosse contenuta in quattro monogrammi disposti, plausibilmente, alle estremità della croce di cui si distingue ancora uno dei bracci (fig. 13).

Fig. 11. Istanbul, Museo Archeologico, frammento di capitello a pannelli, inv. n. 912 (autore).

41 Misure in cm: 33; l max 46,5. Le letture del primo monogramma sono state in passato discordanti, ma un’analisi ravvicinata della scultura mi permette di confermare che l’invocazione è rivolta alla Theotókos [“ΘΕΟΤΟΚΕ ΒΟΗΘΕΙ ΕΙΡΗΝΗ ΑΥΓΟΥΣΤΗ”], così come proposto a suo tempo da Foss 1969, p. 66, che nella stessa occasione corresse una lettura errata di Sotēriou a proposito del quarto monogramma. Di diverso avviso è stato Dennert 1997, n. 77, pp. 43, 188, che ha interpretato il primo monogramma con “KYPIE”, seguito poi da Sklavou Mavroeidi 1999, n. 111, p. 82. Per quel che invece riguarda la cronologia, la datazione proposta da Clive Foss in base al confronto con la monetazione del periodo del regno comune di Irene con il figlio Costantino VI (780797), è stata estesa agli anni di matrimonio con Leone IV da Martin Dennert; la Sklavou Mavroeidi, non tenendo conto dei pareri precedenti, si esprime, invece, a favore di una datazione tra il 797 e l’802, il periodo in cui Irene regnò da sola. 42 Ötüken, Ousterhout 1989, p. 146, tav. XXXVI,a; Dennert 1997, p. 43, nota 334, in cui è inserito in ambito mediobizantino. 43 Le dimensioni del frammento in cm: h 62; Ø max 135; Ø clipeo 22; l cornici 17 cm ca.

D’altronde una soluzione simile fu adottata nelle specchiature della lastra reimpiegata, in due frammenti, nel pavimento della chiesa di Sant’Irene alla quale si è già accennato45. Quello con il nome di Costantino è l’unico ben leggibile di otto monogrammi i quali in origine dovevano comporre, divisi in due gruppi, un’invocazione dal consueto formulario e che erano probabilmente collocati a forma di croce all’estremità di una figura, forse un quadrato disposto diagonalmente. İstanbul Arkeoloji Müzeleri yıllığı, 10, 1962, p. 97, tav. 27.2; Zollt 1994, n. 265, p. 104, tav. 36; Dennert 1997, n. 80, pp. 44, 188-189, tav. 14. 45 Cf. supra, nota 14. 44

273

Claudia Di Bello possibile, impiego del capitello, poiché dopo i primi due decenni di quel secolo sembrano sfumare le attestazioni documentarie circa l’erezione di nuove colonne onorarie49. Per il periodo successivo, si dovrebbe nondimeno tenere conto dell’erezione, testimoniata dalle fonti, di una statua di Giustiniano II (685-695/705-711) su una colonna già esistente nella zona conosciuta come Deutèron, voluta da quell’imperatore per celebrare la riconquista del trono dopo l’esilio50: quest’intervento, che in effetti dimostra l’utilizzo della colonna onoraria come efficace strumento del potere ancora nei primissimi anni dell’VIII secolo, avrebbe potuto richiedere l’impiego di un capitello di proporzioni considerevoli; tuttavia esso, per distanza cronologica, difficilmente avrebbe potuto mostrare caratteristiche lessicali tanto vicine ai modelli di VI secolo come quelle che invece presenta il capitello frammentario del Museo Archeologico (inv. n. 912).

Fig. 13. Ipotesi dello schema decorativo dei pannelli del capitello del Museo Archeologico di Istanbul, inv. n. 912 (elaborazione grafica dello Studio Azimut Roma).

Dei precedenti che invece attestino l’inserimento dei clipei con monogramma all’interno del corpo della croce sono rintracciabili nel decoro della lastra della seconda metà del VI secolo reimpiegata come paliotto d’altare nel battistero di Grado46, e, a Costantinopoli, sui due lati di un pluteo proveniente dall’arredo della bema-church nel complesso della Kalenderhane Camii47, della fine del VII secolo, nei quali, tuttavia, il clipeo si inserisce all’incrocio dei bracci e non alle estremità. Pertanto, per trovare un confronto più aderente si deve guardare oltre l’ambito scultoreo, e più precisamente a uno dei raffinatissimi polycandela argentei del Tesoro di Sion lavorati a giorno con i monogrammi del vescovo Eutichiano intorno al 565. In particolare, il punto focale del lampadario di forma circolare, nel quale figurano delfini e ornati vegetali stilizzati, è appunto costituito dalla figura di una croce con bracci patenti che si uniscono alle estremità con dei clipei vuoti per l’inserimento delle lampade, ai quali, poi, si affiancano verso l’esterno i clipei con i monogrammi del vescovo48.

In base alle poche notizie che abbiamo sull’esiguo numero di monumenti onorari attestati nei primi decenni del VII secolo, si può semmai tenere in considerazione un’eventuale attribuzione del capitello alla colonna onoraria eretta a est della chiesa dei Quaranta Martiri presso il Tetrapylon bronzeo, iniziata per volere di Foca (602-610) ma terminata da Eraclio (610-641), il quale nel 612 vi fece erigere una croce51. Proprio questo simbolo avrebbe pertanto potuto rappresentare un tema decisamente appropriato per il decoro del capitello che doveva concretamente sostenere la croce e a questa ipotetica scelta potrebbe in effetti rimandare la presenza reiterata di questo simbolo nei pannelli del nostro capitello frammentario (cf. fig. 13). Dunque, i capitelli imposta del tipo a pannelli mantennero, anche nel periodo postgiustinianeo, quel gusto ornamentale sobrio e austero di cui si fecero fin dall’inizio perfetti interpreti. Bisognerà infatti attendere le soglie del IX secolo per veder riemergere sui capitelli delle trifore della Fatih Camii di Trilye52, in Bitinia, le caratteristiche, seppur reinterpretate, di quell’esuberanza ornamentale sperimentata nei capitelli della prima metà del VI secolo nella variante più raffinata lavorata a giorno.

Le dimensioni davvero fuori dal comune del capitello frammentario del Museo Archeologico, che si stima raggiungesse circa 1,5 m di altezza, lasciano spazio a poche ipotesi sull’originaria destinazione del pezzo, per la quale infatti sembra plausibile l’impiego in un monumento onorario, stesso utilizzo al quale, d’altro canto, sembra essere stato destinato il capitello di Foca già menzionato.

Bibliografia Abbreviazione: Ist. Mitt. = Istanbuler Mitteilungen.

Nella forma del frammento sono chiaramente rintracciabili i segnali di quel graduale ‘rigonfiamento’ dei profili che, in larga misura, ricorre nei capitelli a pannelli nel periodo postgiustinianeo. Inoltre, alla luce dell’influenza ancora sensibile dei modelli di VI secolo e del ritmo elegante e perfettamente modulato dei girali nelle cornici, mi sembra plausibile proporre per il capitello frammentario del Museo Archeologico di Istanbul (inv. n. 912) una datazione entro la metà del VII secolo. Una conferma, in tal senso, può essere suggerita dall’originario,

Asutay-Effenberger N., 2002. ‘Ein unbekanntes Ehrenmonument des Kaisers Phokas aus Synada bei Akronion (Afyon)’, in Byzantinische Zeitschrift, 95, pp. 417-421. 49 Il silenzio delle fonti al riguardo è stato interpretato come il segnale dell’eclissi della colonna onoraria come tipologia architettonica da Mango 1993, X, p. 16. Cf. anche Taddei 2014, p. 543. 50 Patria, III.79 (ed. Preger 1907, p. 244). Cf. Guilland 1969, II, p. 64; Peschlow 1986, p. 30, nota 32. 51 L’analisi più completa delle testimonianze sulla colonna è proposta da Mango 1993, X, pp. 14-17. Cf. anche Jordan-Ruwe 1995, pp. 188-189; Taddei 2014, p. 539, il quale suggerisce la possibilità che il fusto della colonna fosse articolato in rocchi marmorei. 52 Dennert 1997, n. 170, pp. 78-79, 198, tav. 30; Di Bello, 2018, pp. 1315, figg. 3-4, con bibliografia precedente.

Tagliaferri 1981, n. 647, pp. 414-415, tav. CCXXXVI. Peschlow 1997, cat. 118, pp. 109-110, 112, tavv. 132-133; 2007, cat. 118, pp. 299, 322. 48 Boyd 1992, n. 25, pp. 9-11, 24. 46 47

274

I capitelli imposta del tipo cosiddetto a pannelli nell’ambito della produzione scultorea postgiustinianea Asutay-Effenberger N., 2007. Die Landmauer von Konstantinopel-İstanbul. Historisch-topographische und baugeschichtliche Untersuchungen, Berlino/New York (Millennium-Studien zu Kultur und Geschichte des ersten Jahrtausends n. Ch., 18).

22–25 settembre 2015, Spoleto (Quaderni della Rivista di Bizantinistica, 20), t. I, pp. 423-438. Duchesne L. (ed.), 1955. Le Liber pontificalis, I, rist. anast., con nuova pref. di J. Bayet, Parigi (Bibliothèque des Écoles françaises de Rome et d’Athènes. 2e série [: Registres et lettres des papes du XIIIe siècle, III/1-3]).

Barsanti C., 1989. ‘L’esportazione di marmi dal Proconneso nelle regioni pontiche durante il IV-VI secolo’, in Rivista dell’Istituto nazionale d’archeologia e storia dell’arte. III serie, 12, pp. 91-220.

Ebersolt J., 1921. Mission Constantinople, Parigi.

archéologique

de

Feissel D., 1987. ‘De Chalcédoine à Nicomédie. Quelques inscriptions négligées’, in Travaux et mémoires, 10, pp. 405-436.

Barsanti C., 1993. ‘Capitello, area bizantina e area islamica’, in Enciclopedia dell’arte medievale, IV, Roma, pp. 200-211.

Foss C., 1969. ‘Historical Note on the Church at Sige’, in H. Buchwald, The Church of the Archangels in Sige near Mudania, Vienna/Colonia/Graz (Byzantina Vindobonensia, 4), pp. 64-71.

Barsanti C., 1995. ‘Materiali bizantini poco noti o inediti della Bitinia’, in XLII Corso di cultura sull’arte ravennate e bizantina. Seminario internazionale sul tema Ricerche di archeologia cristiana e bizantina, Ravenna, 14–19 maggio 1995, Ravenna, 1995, pp. 47-69.

Guiglia Guidobaldi A.., 2010. ‘Da Costantinopoli a Genova e a Varese: insolito itinerario di una scultura bizantina del VI secolo’, in A. Armati, M. Cerasoli, C. Luciani (edd.), «Alle gentili arti ammaestra». Studi in onore di Alkistis Proiou, Roma (Testi e Studi bizantinoneoellenici, 18), pp. 97-124.

Barsanti C., 2006. ‘I “Catini d’oro” di Padova: spoglie costantinopolitane di VI secolo’, in G. Trovabene (ed.), Florilegium artium. Scritti in memoria di Renato Polacco, Padova (Facoltà di lettere e filosofia dell’Università di Venezia. Miscellanea, 8), pp. 37-48.

Guilland R., 1969. Études de topographie de Constantinople byzantine, Berlino/Amsterdam (Berliner byzantinistiche Arbeiten, 37).

Barsanti C., Guiglia A. (ed.), 2010. The Sculptures of the Ayasofya Müzesi in Istanbul. A Short Guide, Istanbul. Boyd S. A., 1992. ‘A “metropolitan” treasure from a church in the provinces: an introduction to the study of the Sion Treasure’, in S. A. Boyd, M. Mundell Mango (edd.), Ecclesiastical silver plate in sixth-century Byzantium. Papers of the Symposium Held May 16-18, 1986, at the Walters Art Gallery, Baltimore and Dumbarton Oaks, Washington, D.C., Washington.

Jordan-Ruwe M., 1995. Das Säulenmonument. Zur Geschichte der erhöhten Aufstellung antiker Porträtstatuen, Bonn (Asia Minor Studien, 19). Kautzsch R., 1936. Kapitellstudien. Beiträge zu einer Geschichte des spätantiken Kapitells im Osten vom 4. bis ins 7. Jahrhundert, Berlino/Lipsia (Studien zur spätantiken Kunstgeschichte, 9).

Brett G., Martiny G., Stevenson R. B. K., 1947. The Great Palace of the Byzantine Emperors. Being a first report on the excavations carried out in Istanbul on behalf of the Walker Trust (the University of St. Andrews) 19351938, Oxford.

Kazhdan A. P., 1991a. ‘Spatharios’, in Kazhdan 1991b, pp. 1935-1936. Kazhdan A. P. (dir.), 1991b. The Oxford Dictionary of Byzantium, New York/Oxford.

de Boor C. (ed.), 1883. Theophanis Chronographia, I, Lipsia.

Kazhdan A. P., Cutler A., 1991. ‘Protospatharios’, in Kazhdan 1991b, p. 1748.

Dennert M., 1997. Mittelbyzantinische Kapitelle. Studien zu Typologie und Chronologie, Bonn (Asia Minor Studien, 25).

Khroushkova L., 2006. Les monuments chrétiens de la côte orientale de la Mer Noire. Abkhazie, IVe-XIVe siècles, Turnhout (Bibliothèque de l’Antiquité tardive, 9).

Di Bello C., 2018. ‘Le sculture della Fatih Camii a Trilye: una tappa significativa nel processo di transizione verso la plastica mediobizantina’, in C. Di Bello, R. Gandolfi, M. Latella (edd.), In corso d’opera. Ricerche dei dottorandi in Storia dell’Arte della Sapienza, II, Roma, pp. 11-18.

Kramer J., 1988. ‘Kämpferkapitelle mit den Monogrammen Kaiser Justinus’ II. und seiner Gemahlin, der Kaiserin Sophia in Yalova Kaplıcaları (Termal)’, in M. Restle (ed.), Festschrift für Klaus Wessel zum 70. Geburtstag. In memoriam, München (Münchener Arbeiten zur Kunstgeschichte und Archäologie, 2), pp. 175-190.

Di Bello C., 2019. ‘La scultura postgiustinianea tra modelli e nuove formule: il caso della lastra di Costantino V a Sant’Irene’, in S. Cosentino, M. E. Pomero, G. Vespignani (edd.), Dialoghi con Bisanzio. Spazi di discussione, percorsi di ricerca. Atti dell’VIII Congresso dell’Associazione Italiana di Studi Bizantini, Ravenna,

Krumeich K., 1997. ‘Spätantike Kämpferkapitelle mit Weinblatt- und Pinienzapfen-Dekor’, in Ist. Mitt., 47, pp. 277-314. Mango C., 1972. The Art of the Byzantine Empire, 3121453. Sources and Documents, Englewood Cliffs (Sources and Documents in the History of Art Series). 275

Claudia Di Bello Mango C., 1993. Studies on Constantinople, Aldershot (Variorum Reprints. Collected Studies Series, 394).

Striker, C. L., Kuban Y. D. (edd.), 1997. Kalenderhane in Istanbul. The Buildings, their History, Architecture, and Decoration. Final Reports on the Archaeological Exploration and Restoration at Kalenderhane Camii 1966-1978, I, Magonza.

Mendel G., 1914. Catalogue des sculptures grecques, romaines et byzantines, III, Constantinople. Otten T., 2010. ‘Das byzantinische Pergamon – ein Überblick zu Forschungsstand und Quellenlage’, in F. Daim, J. Drauschke (edd.), Byzanz – Das Rômerreich im Mittelalter / Byzantium –The Roman Empire in the Middle Ages / Byzance – L’Empire romain au Moyen Âge, II/2, Magonza (Monographien des RömischGermanischen Zentralmuseums, 84/2/2).

Taddei A., 2014. ‘Smaragdos «patrikios», la colonna dell’imperatore Foca e la Chiesa di Roma’, in M. Gianandrea, F. Gangemi, C. Costantini (edd.), Il potere dell’arte nel Medioevo. Studi in onore di Mario D’Onofrio, Roma (Saggi di storia dell’arte, 40), pp. 531-550. Tagliaferri A., 1981. Le diocesi di Aquileia e Grado, Spoleto (Corpus della scultura Altomedievale, 10).

Ötüken Y., Ousterhout R., 1989. ‘Notes on the Monuments of Turkish Thrace’, in Anatolian Studies, 39, pp. 121-149.

Tezcan H., 1989. Topkapı Sarayı ve çevresinin Bizans devri arkeolojisi, Istanbul.

Peschlow U., 1986. ‘Eine wiedergewonnene byzantinische Ehrensäule in Istanbul’, in O. Feld, U. Peschlow (edd.), Studien zur Spätantiken und Byzantinischen Kunst. Friedrich Wilhelm Deichmann Gewidmet, I, Bonn (Römisch-Germanisches Zentralmuseum. Monographien, 10/1), pp. 21-33.

Ulbert T., 1969. Studien zur dekorativen Reliefplastik des östlichen Mittelmeerraumes (Schrankenplatten des 4.10. Jahrhunderts), München (Miscellanea Byzantina Monacensia, 10).

Peschlow U., 1997. ‘Architectural sculpture’, in Striker, Kuban 1997, pp. 101-111.

Ulbert T., 1969-70. ‘Untersuchungen zu den byzantinischen Reliefplatten des 6. bis 8. Jahrhunderts’, in Ist. Mitt., 19, pp. 339-357.

Peschlow U., 2007. ‘Byzantine Architectural Sculpture’, in C. L. Striker, Y. D. Kuban (edd.), Kalenderhane in Istanbul. The Excavations. Final Reports on the Archaeological Exploration and Restoration at Kalenderhane Camii 1966-1978, II, Magonza, pp. 295-342.

Ward Perkins J. B., Goodchild R. G., 2003. Christian Monuments in Cyrenaica, ed. J. Reynolds, London (Society for Libyan Studies. Monographs, 4). Zacos G., Veglery A., 1972. Byzantine Lead Seals, Basilea.

Preger T. (ed.), 1907. Scriptores originum Constantinopolitanarum, II, Lipsia (Bibliotheca scriptorum Graecorum et Romanorum Teubneriana).

Zollt T., 1994. Kapitellplastik Konstantinopels vom 4. bis 6. Jahrhundert n. Chr.. Mit einem Beitrag zur Untersuchung des ionischen Kämpferkapitells, Bonn (Asia Minor Studien, 14).

Raptis K. T., Vasiliadou S., 2005. ‘Διαχρονική χρήση, διαδοχικές θέσεις και απόπειρα επανένταξης των μαρμάρινων αρχιτεκτονικών μελών των βασιλικών Α, Β, Γ Αγίου Γεωργίου Πέγειας (Πάφος): από την παραγωγή στα παλαιοχριστιανικά λατομεία της πρωτεύουσας στην διαμόρφωση του αρχαιολογικού χώρου’, in Επιστημονική Επετηρίδα του Τμήματος Αρχαιοτήτων Κύπρου, 43, pp. 199-224. Russo E., 2011. Le decorazioni di Isidoro il Giovane per S. Sofia di Costantinopoli, Roma (I Libri di Viella. Arte). Russo E., 2014-15. ‘Sulla cornice della cupola di S. Sofia di Costantinopoli’, in Bizantinistica. Serie seconda, 16, pp. 1-23. Sanders I. F., 1982. Roman Crete. An Archaeological Survey and Gazetteer of Late Hellenistic, Roman and Early Byzantine Crete, Warminster (Classical Studies). Sklavou Mavroeidi M., 1999. Γλυπτά του βυζαντινού Μουσείου Αθηνών, Atene. Sodini J.-P., 1982. ‘Un chapiteau «mixte» d’époque paléochrétienne à Delphes’, in L. HadermannMisguich, G. Raepsaet (edd.), Rayonnement grec. Hommages à Charles Delvoye, Bruxelles (Travaux de la Faculté de Philosophie et Lettres de l’Université de Bruxelles, 83), pp. 325-340. 276

21 Chapiteaux corinthiens inédits d’Apamée (Syrie) Catherine Vanderheyde Université de Strasbourg, UMR 7044–ARCHIMÈDE (France) et Université libre de Bruxelles (Belgique) Cet article présente les résultats partiels d’une recherche menée sur les sculptures architecturales protobyzantines issues des fouilles archéologiques réalisées à Apamée depuis 1965. Étant donné la tourmente politique qui secoue la Syrie depuis plusieurs années, il était important, lors du 23e Congrès international des Études byzantines à Belgrade, de documenter ce matériel archéologique encore mal connu et d’attirer l’attention de la communauté scientifique sur ces pièces appartenant au patrimoine byzantin qui sont inaccessibles à l’heure actuelle et dont une bonne partie a probablement été détruite ou volée lors des pillages dont a malheureusement été victime le site d’Apamée depuis 2011. Parmi les sculptures répertoriées, les chapiteaux sont les mieux conservés et proviennent à la fois d’édifices religieux et de résidences privées. Plusieurs chapiteaux corinthiens en marbre ont été trouvés lors des fouilles de la maison aux colonnes bilobées menées en 1970 et 1978 dans le quartier nord-ouest de la ville. Si les informations consignées dans les carnets de fouilles sont lacunaires, plusieurs de ces éléments architecturaux présentent une typologie comparable à celle caractérisant d’autres exemplaires datés du Ve ou du VIe siècle conservés en Syrie, mais aussi à Constantinople ainsi qu’à Chypre. This paper is the first presentation of research begun in 2009 on the Early Byzantine sculpture discovered during the excavations at Apamea in Syria conducted since 1965. Given the political turmoil in Syria over the last several years, it seemed particularly important at the 23rd International Congress of Byzantine Studies in Belgrade to document this still poorly known archaeological material and to draw the attention of the scientific community to this Byzantine heritage that may have been destroyed or stolen during the looting of the site of Apamea since 2011. The items in the best state of preservation are the capitals that come from religious buildings and private residences. Several marble Corinthian capitals were found during excavations of the house called ‘aux colonnes bilobées’, conducted in the northwest area of the town. The information recorded in the excavation notebooks does not help to clarify the chronology of these carved pieces, but many of the capitals are very similar to capitals found in Syria and also in Constantinople and in Cyprus, which can be dated from the fifth or sixth century. In vorliegendem Aufsatz werden die Teilergebnisse einer Untersuchung über frühbyzantinische Bauplastiken vorgestellt, die aus seit 1965 in Apamea durchgeführten Ausgrabungen stammen. Angesichts der politischen Unruhen, die Syrien seit einigen Jahren erschüttern, war es auf dem 23. Internationalen Kongress byzantinischer Studien in Belgrad wichtig, dieses wenig bekannte archäologische Material vorzustellen und so die Aufmerksamkeit der Forschung auf diesen Teil des byzantinischen Erbes zu lenken. Es ist derzeit nicht zugänglich und viele seiner Teile wurden seit 2011 durch die Plünderung der Apamea-Stätte wahrscheinlich zerstört oder entwendet. Unter den best erhaltenen Kapitellen finden sich sowohl welche, die aus religiösen Gebäuden zum Vorschein kamen als auch welche, die aus privaten Residenzen stammen. Mehrere korinthische Kapitelle aus Marmor wurden bei Ausgrabungen des Hauses « aux colonnes bilobées » entdeckt, die 1970 und 1978 im nordwestlichen Teil der Stadt stattfanden. Obwohl die erhaltene Ausgrabungsdokumentation unvollständig ist, weisen viele dieser architektonischen Elemente eine Typologie auf, die mit anderen Beispielen aus dem 5. oder 6. Jh. vergleichbar ist, die aus Syrien, aber auch aus Konstantinopel und von der Insel Zypern stammen. Questo articolo presenta i risultati parziali di una ricerca sulle sculture protobizantine provenienti dagli scavi archeologici effettuati ad Apamea da 1965. Dati i disordini politici che hanno scosso 277

Catherine Vanderheyde la Siria per diversi anni, è stato importante per il 23° Congresso Internazionale di Studi Bizantini a Belgrado documentare questo materiale archeologico poco noto, anche per attirare l’attenzione della comunità scientifica su queste testimonianze dell’eredità bizantina che sono attualmente inaccessibili e molte delle quali sono state probabilmente distrutte o rubate durante il saccheggio che ha devastato il sito di Apamea sin dal 2011. Tra le sculture repertoriate, i capitelli sono quelli meglio conservati e provengono sia da edifici religiosi che da residenze private. Diversi capitelli corinzi marmorei sono stati rinvenuti durante gli scavi della casa “aux colonnes bilobées” nel 1970 e nel 1978 nel quartiere nord-occidentale della città. Sebbene le informazioni contenute nelle note di scavo siano incomplete, molti di questi elementi architettonici hanno una tipologia paragonabile a quella di altri esempi datati al V o VI secolo, conservati in Siria, ma anche a Costantinopoli e a Cipro. Cet article présente les résultats partiels d’une recherche menée sur les sculptures architecturales protobyzantines issues des fouilles archéologiques réalisées à Apamée entre 1965 et 2009.1 Étant donné la tourmente politique qui secoue la Syrie depuis plusieurs années, il était important, lors du 23e Congrès international des Études byzantines à Belgrade, de documenter ce matériel archéologique encore mal connu et d’attirer l’attention de la communauté scientifique sur ces pièces appartenant au patrimoine byzantin qui sont inaccessibles à l’heure actuelle et dont une bonne partie a probablement été détruite ou volée lors des pillages dont a malheureusement été victime le site d’Apamée depuis 2011. Avant de présenter le corpus de chapiteaux qui fera l’objet de cette étude, une rapide mise en contexte de ces éléments sculptés dans l’environnement urbain dont ils proviennent s’impose.

révélé la présence d’une aristocratie riche.4 Après les tremblements de terre successifs de 458, 526 et 528, des changements urbanistiques affectent la Grande Colonnade bordant le cardo : les dalles polygonales initiales de cette large voie sont alors recouvertes d’un pavement plus régulier en dalles de calcaire, tandis que des trottoirs de 3,50 m de large sont aménagés le long des portiques.5 Cette allée centrale est par endroits complètement séparée du trafic, désormais orienté ouest-est, grâce à l’installation de séries de trois à cinq marches au nord du tycheion et à hauteur de l’église à atrium. Des échoppes ont été peu à peu installées en bordure des édifices publics, créant ainsi de nouvelles zones commerciales. Les premiers empiétements sur l’alignement des rues et les structures de l’urbanisme classique apparaissent aussi durant cette période et ne feront que s’accentuer aux VIIe et VIIIe siècles.6

De la cité hellénistique à la ville romaine puis byzantine : un bref aperçu

La pénétration du christianisme à Apamée est attestée par plusieurs sources hagiographiques : le récit relatant le martyre de saint Maurice et de ses soixantedix compagnons-soldats persécutés lors du règne de Maximien, celui consacré à saint Antonin, ancien tailleur de pierre, et celui relatif à saint Marcel, évêque de la ville.7 Ces trois saints hommes se heurtent aux païens qui défendent fermement leurs dieux et leurs idoles. La renommée de certains saints d’Apamée semble par ailleurs avoir dépassé les frontières de la cité comme le suggère l’oratoire contenant le tombeau du martyr Thomas, apocrisiaire d’un monastère apaméen décédé vers 545, situé dans le cimetière d’Antioche, à l’extérieur de la porte de Daphné.8 D’après les fouilles menées jusqu’ici par la mission archéologique belge, c’est dans le quartier sudest de la ville d’Apamée que paraissent se concentrer la plupart des monuments chrétiens, comme en témoignent les vestiges de la Rotonde, de l’église à atrium, de la cathédrale de l’Est9 et de l’église du secteur VIII.1710 (fig.

Lors de sa fondation par Séleucos Ier en 300 avant J.-C., la ville d’Apamée fut progressivement protégée par une enceinte de 6,3 km de long, ponctuée d’une cinquantaine de bastions et percée de trois portes au nord, au sud et à l’est, englobant une superficie de 250 hectares quadrillée par des rues se croisant à angle droit qui délimitaient des ensembles réguliers de maisons ou insulae. Le cardo, large de 20 m et long de 1,85 km, et le decumanus, long de 1,23 km, se croisaient aux deux-tiers inférieurs de la superficie de la ville et constituaient probablement dès cette époque les axes de communication principaux.2 De cette cité hellénistique, bien peu de choses sont parvenues jusqu’à nous, car elle fut détruite par un séisme en 115 de notre ère et totalement reconstruite au cours du IIe siècle. Aux périodes romaine et byzantine, Apamée connut une grande prospérité, comme l’attestent ses monuments richement ornés de sculptures, d’opus sectile et de mosaïques.3 Le cardo, bordé de portiques et pavé de dalles en calcaire, donnait accès aux édifices majeurs de la cité : l’agora, les thermes, une fontaine monumentale, une grande latrine, et, le decumanus, au théâtre (fig. 1). Les imposantes villae exhumées dans la partie nord-est de la ville ont par ailleurs

4 Les aspects de l’architecture domestique à Apamée ont fait l’objet de plusieurs études rassemblées dans les actes du colloque d’Apamée qui s’est déroulé à Bruxelles du 29 au 30 mai 1980. Voir Balty 1984. 5 Balty 1969, p. 41. 6 Balty 1969, p. 42. 7 Delehaye 1935, pp. 225-237. 8 Maraval 1985, pp. 337 et 342. 9 Balty 1969, pp. 39-40, fig. 2-3. 10 Les résultats de la campagne de fouille que j’ai menée avec Jean et Janine Balty dans ce secteur situé à l’est de l’église à atrium ont fait l’objet d’une communication lors du colloque sur Apamée de Syrie. Recherches archéologiques 1980-2004, dont les actes sont sous presse.

Je remercie Jean Balty ainsi que Didier Viviers, directeurs successifs de la mission archéologique belge d’Apamée de m’avoir confié l’étude de ces sculptures architecturales. 2 Balty 1969, pp. 33-34, n. 1. 3 Balty 1981, pp. 46-155. 1

278

Chapiteaux corinthiens inédits d’Apamée (Syrie)

Fig. 1. Plan topographique d’Apamée (Mission archéologique belge à Apamée de Syrie).

1). Ces édifices devaient attirer de nombreux pèlerins durant la période byzantine, car plusieurs reliquaires y ont été retrouvés, notamment dans le martyrion de l’église à atrium et dans la cathédrale de l’Est.11 Ces

écrins présentent une forme de petit sarcophage et la plupart d’entre eux conservent un dispositif permettant de verser et de recueillir dans une coupelle sculptée dans la masse de la cuve du reliquaire l’huile ayant imprégné les reliques. Ce myron sanctifié par ces dernières était ensuite recueilli dans des ampoules en verre afin d’être

Mayence 1935a, p. 7; 1935b, p. 202, pl. XX; Napoleone-Lemaire, Balty 1969, pp. 57-64, fig. 13 et 15-16, pl. LI.1-2 et LII.1-2, et pp. 34-42, fig. 9 a-c; Verhoogen 1964, no 18; Vanderheyde 2003, pp. 63-69 et 75-76, fig. 1-8.

11

279

Catherine Vanderheyde conservé à des fins curatives dans des eulogies.12 Trois des reliquaires exhumés portent des inscriptions indiquant qu’ils contenaient des reliques de saints dont le culte était répandu à Constantinople, tels Callinikos ou saint Jean le Soldat, ou des restes de saints populaires dans les régions orientales de l’Empire byzantin, comme saints Cosme et Damien ou saint Théodore, ou encore des reliques de divers saints non nommés.13 Plusieurs autres reliquaires anépigraphes ont aussi été retrouvés à Apamée. L’un d’eux, provenant dans l’église à atrium de la grande colonnade14 et dont le couvercle est conservé aux Musées Royaux d’art et d’histoire à Bruxelles15, a été sculpté dans du marbre vert issu des carrières de Thessalie. Une concentration de sept reliquaires dépourvus d’inscriptions a par ailleurs été observée lors de la fouille de l’église située à l’extérieur des remparts, près de la porte nord16.

effet aussi par un emploi varié de matériaux, de formes architecturales, de décors sculptés et de techniques. Cette hétérogénéité intentionnelle caractérise l’ordonnance des supports et le décor architectural de Saint-Démétrios à Thessalonique où, comme l’a bien observé Jean-Michel Spieser, « la richesse se fondait sur la discontinuité, la variation et non sur la répétition et la régularité ».21 Il semble que ce principe esthétique, recourant ou non à l’utilisation de remplois, ait pu concerner l’ensemble des églises d’une même ville tant dans les régions orientales que dans les régions balkaniques de l’Empire. En Syrie, particulièrement nombreuses sont les variantes du chapiteau corinthien sculptées dans la pierre calcaire locale durant le VIe siècle.22 Ce matériau, parfois peint en rouge, a aussi été largement utilisé pour les sculptures retrouvées dans la « maison des chapiteaux à consoles » ainsi que celles découvertes dans l’église à atrium et dans la cathédrale d’Apamée. Trois chapiteaux du Musée de Qalaa’t el Mudiq proviennent des fouilles de la « maison des chapiteaux à consoles » réalisées en 1977 et 1979.23 Cette demeure, située dans la partie sud-est du quartier central de la ville, date du IIe siècle, mais a continué à être occupée durant la période byzantine. François Baratte, qui a fouillé ce secteur, signale un chapiteau à protomés d’animaux (fig. 2) semblable à l’un trouvé dans la cathédrale et demeuré inédit, suggérant que cette pièce en proviendrait.24 Des formes de chapiteaux répandues dans d’autres sites byzantins de Syrie, tel Qasr ibn Wardan,25 sont aussi présentes à Apamée. Lors des fouilles menées à l’église à atrium de la grande colonnade ont été exhumés de beaux chapiteaux corinthiens à médaillon dotés d’une ornementation sculptée variée et de bonne facture (fig. 3). Certains exemplaires se caractérisent par une qualité d’exécution exceptionnelle, tel l’un des chapiteaux en marbre rose de la cathédrale de l’Est sur lequel est sculptée une inscription mentionnant l’évêque Paul (fig.  4). Un

Omniprésence des chapiteaux Parmi les sculptures architecturales étudiées ci-dessous, les chapiteaux sont les mieux conservés et ont généralement été trouvés dans les édifices religieux, mais aussi dans les résidences privées. Ce type d’élément architectural, à moins qu’il ne soit décoré de croix, ne fournit guère d’information sur la fonction du bâtiment auquel il appartenait, ce qui constitue une sérieuse difficulté dans la détermination de la provenance initiale de ce matériel. Dans l’architecture religieuse byzantine, le chapiteau occupe néanmoins un rôle important, car il constitue l’unique élément porteur d’une ornementation sculptée, étant donné la standardisation de plus en plus poussée des bases et des fûts de colonnes. Il connaîtra une évolution formelle intéressante et deviendra le réceptacle de décors très variés. Par rapport au chapiteau antique intégré à un ensemble régi par les ordres architecturaux, le chapiteau byzantin apparaît plus autonome.17 La variété des matériaux, des décors et des procédés techniques dont témoignent les chapiteaux d’Apamée démontre la créativité des ateliers locaux de sculpteurs chargés de la réalisation du décor architectural des édifices durant la période protobyzantine. Une telle diversité ne se rencontre pas seulement à Apamée et dans la région du Massif calcaire où ont certainement œuvré les mêmes ateliers de sculpteurs aux côtés d’équipes locales d’artisans. La sculpture architecturale de plusieurs édifices religieux situés dans les villes balkaniques, telles par exemple Caričin Grad18, Stobi19 et Sandanski20, se distingue en 12 La description de ce dispositif a été clairement énoncée par Fernand Mayence dès leur découverte. Cf. Mayence 1935a, p. 4. Au sujet des pratiques rituelles associées à ce type de reliquaire, voir Delehaye 1935, pp. 240-244; Gessel 1988, pp. 183-202. 13 Delehaye 1935, pp. 237-240; Vanderheyde 2003, pp. 67-68, nos 1 et 2, pp. 74-75, fig. 1 et 3-4; Comte 2012, pp. 99-100. 14 Mayence 1935a, p. 4; Napoleone-Lemaire, Balty 1969, pp. 58-62, fig. 14 et 15, pl. XXX.1 et LIII.1-5. 15 Vanderheyde 2003, no 3, pp. 75-76, fig. 5-6. 16 Comte 2012, pp. 368-372. Voir aussi le compte rendu dans Vanderheyde 2015. 17 Peschlow 2001-04, col. 118. 18 Nikolajević-Stojković 1957, pp. 51-60. 19 Niewöhner 2010, pp. 440-444. 20 Vaklinova 1980, pp. 641-649.

Fig. 2. Chapiteau à double zone de la « Maison des chapiteaux à consoles », Musée de Qal’at el-Mudiq. (C. Vanderheyde ; détourage A. Stoll). Spieser 1984, pp. 199-211 (en particulier 211). Strube 1983, p. 65. 23 Baratte 1972, p. 119, pl. XVLI.1 et XLVI.2. 24 Baratte 1972, p. 119. 25 Strube 1983, p. 62, et pl. 9 a et 15 a. 21 22

280

Chapiteaux corinthiens inédits d’Apamée (Syrie) chapiteau corinthien à deux rangs de feuilles d’acanthe se détachant de la corbeille et présentant des lobes sommitaux particulièrement protubérants (fig. 5 a-b).26 Le marbre : une denrée rare en Syrie Cette étude a plus particulièrement trait à six chapiteaux corinthiens qui se distinguent des autres éléments sculptés par leur matériau – un marbre blanc-jaunâtre veiné de gris. Les caractéristiques externes de ce matériau font penser au marbre issu des carrières de l’île de Proconnèse, mais aucune analyse archéométrique permettant de confirmer cette hypothèse n’a été menée. Les connaissances relatives à l’exploitation de carrières locales de marbre blanc en Syrie restent limitées. Il a été suggéré que le marbre issu des carrières de Qassabiye, exploitées durant l’Antiquité, aurait été utilisé pour la réalisation du reliquaire de Huarte.27 Des recherches plus approfondies devraient être entreprises, lorsque la situation politique de cette région le permettra, afin de déterminer l’ampleur de cette production locale. Celle-ci ne couvrait manifestement pas tous les besoins, car des marbres blancs et colorés étaient importés des autres régions de l’Empire. Le coût de l’acheminement de ce matériau devait être conséquent et son emploi était certainement réservé à des éléments considérés comme précieux, destinés à prendre place dans les espaces cultuels ou dans de riches demeures privées. Dès le Ve siècle, diverses pièces sculptées en marbre appartenant au mobilier liturgique, tels des plaques et des petits piliers de chancel, des plateaux et des supports de table d’autel, ainsi que des reliquaires, ont été retrouvés dans les églises syriennes dont la construction avait été promue par de riches donateurs, notamment à Esraʽ et Bosrà. Des placages en marbre de Proconnèse ornaient les murs intérieurs de l’église de Qal’at Sem’an, tandis que dans l’église Sainte-Croix à Resafa ces mêmes placages

Fig. 3. Chapiteau corinthien à médaillon, église à atrium (C. Vanderheyde ; détourage A. Stoll).

Fig. 4. Chapiteau corinthien à médaillon, cathédrale de l’Est (C. Vanderheyde).

Fig. 5a-b. Chapiteau corinthien peint en rouge, cathédrale de l’Est (C. Vanderheyde ; détourage A. Stoll).

autre chapiteau inédit, en calcaire peint en rouge et probablement inachevé, provient du même ensemble architectural : il s’agit d’une variante très réussie du

Chapiteau en deux fragments jointifs recollés : haut. 39 cm ; lit d’attente 46 x 38 cm ; diam. reconstitué du lit de pose 28 cm. 27 Comte 2012, p. 383. 26

281

Catherine Vanderheyde

Fig. 6 a-d. Chapiteau corinthien, « Maison aux colonnes bilobées » (A. Stoll et C. Vanderheyde).

« salle A ») où se trouvaient des éléments sculptés (dalle de chancel, chapiteau avec croix, chrisme entouré de l’alpha et de l’oméga) provenant probablement d’une église des alentours et réutilisés ultérieurement dans cette demeure.33 L’un de ces chapiteaux (cat. no 1, fig. 6 a-c), en marbre blanc-jaunâtre veiné de gris, est comparable aux chapiteaux corinthiens constantinopolitains à feuilles d’acanthe épineuse à échancrure en « V » sous le bouton d’abaque, datés entre la deuxième moitié et le milieu du Ve siècle et dont la typologie a été établie par Rudolf Kautzsch (type V et VI)34, puis par Annie Pralong (type IVb)35. Ce type de chapiteau était produit dans les carrières de marbre de Proconnèse, comme l’avait déjà supposé Josef Strzygowski.36 Il est donc présent à Constantinople,37 mais il est aussi très diffusé dans l’ensemble du Bassin méditerranéen, jusqu’en Afrique du Nord, et sur les côtes de la Mer Noire.38 Des importations constantinopolitaines de ce même type de chapiteaux corinthiens se retrouvent par exemple sur le littoral bulgare (fig. 7), en Abkhazie

voisinent avec d’autres, réalisés en brèche verte, de Thessalie.28 À Apamée, le marbre blanc a été utilisé non seulement pour les chapiteaux que nous allons examiner, mais aussi pour des placages sculptés en champlevé, des reliquaires, des plaques, des mortiers,29 tandis qu’une table en sigma découverte dans la « maison du Cerf » et un reliquaire de l’église à atrium sont en marbre vert de Thessalie.30 Des chapiteaux en marbre dans la maison dite « aux colonnes bilobées » Plusieurs chapiteaux corinthiens en marbre ont été mis au jour lors des fouilles menées par Frank Van Wonterghem en 1970 et en 1978 dans le quartier nord-ouest d’Apamée, à l’endroit de la maison dite « aux colonnes bilobées », partiellement exhumée (fig. 1). Cette riche demeure, qui occupe le secteur V.10 sur le plan topographique établi par Henri Lacoste,31 avait déjà été repérée en 1934, lors du dégagement d’une colonne bilobée haute de 5 m reposant sur un pavement en mosaïque aux motifs géométriques multicolores et variés.32 La campagne de fouilles de 1970 a permis de dégager une grande salle mosaïquée (appelée

Balty, Balty 1972, p. 21. Kautzsch 1936, pp. 59-61, nos 194-195, pl. 14, et pp. 75-78, nos 226 et 235, pl. 16. 35 Pralong 2000, pp. 85-89, fig. 6 et 7 b; Pralong 1993, p. 139, fig. 9 : chapiteau du Musée archéologique d’Istanbul, no inv. 3181. 36 Kautzsch 1936, p. 60. 37 Ce type de chapiteau est souvent réutilisé dans les églises ou les citernes de la capitale. Cf. Zollt, pp. 176 et 197, nos 471 et 489. 38 Barsanti, Guiglia Guidobaldi, Sodini, 1998, pp. 320-321; Sodini 2000, pp. 429-430, fig. 9. 33 34

Naccache, Sodini 1989, p. 488. Vanderheyde 2003, pp. 75-77, fig. 3-4 et 9-10, et pp. 78-82, fig. 13-23. 30 Vanderheyde 2003, pp. 75-76, fig. 5-6. 31 Balty 1969, plan I. 32 Balty 1984, p. 13, fig. 2. 28 29

282

Chapiteaux corinthiens inédits d’Apamée (Syrie)

Fig. 8. Chapiteau corinthien fragmentaire, « Maison aux colonnes bilobées » (Mission archéologique belge à Apamée de Syrie ; détourage A. Stoll). Fig. 7. Chapiteau corinthien, Musée archéologique de Nessebar (C. Vanderheyde).

nombreuses marques de ce type qui nous informent sur le mode d’organisation des équipes de sculpteurs sur ce chantier de construction exceptionnel.43 La présence d’un tel signe sur le décor végétal du chapiteau no 3 intrigue, car aucune marque similaire n’a été repérée sur les autres chapiteaux d’Apamée. La position du «  ε  » suggère que le sculpteur a gravé cette lettre lors qu’il taillait ce chapiteau. D’autres lettres renversées repérées sur d’autres chapiteaux, notamment « KY » pour « Κυριέ », indiquent en effet que ces derniers étaient posés à l’envers lorsqu’ils étaient produits en série par les marbriers dans les carrières de Proconnèse.44 Néanmoins, la place discrète de cette marque et son incision délicate pourraient suggérer distribution de ce chapiteau en vue de son exportation.45

ou dans d’autres grandes villes de l’Empire, comme Thessalonique.39 D’autres chapiteaux en marbre à deux rangs de feuilles d’acanthe épineuse, retrouvés à Antioche et à Daphné, constituent des illustrations supplémentaires d’importations constantinopolitaines en Syrie.40 Une série de petits chapiteaux en marbre à un seul rang de feuilles d’acanthe a également été découverte lors des fouilles de la maison aux colonnes bilobées réalisées en septembre-octobre 1978. Le carnet de fouilles de cette campagne indique que les travaux se sont poursuivis au sud de la zone exhumée en 1970. L’un des chapiteaux (no  3, fig. 9 a-d) a été trouvé le premier jour de la campagne dans le secteur « B » au sud-est de la salle A mise au jour lors de la précédente campagne.41 La marque «  ε  », visible à la base de la côte centrale de l’une des feuilles d’acanthe (fig. 9 d) apparaît à l’endroit lorsque le chapiteau repose sur son lit d’attente. S’agit-il d’une marque de tâcheron ? Divers éléments architecturaux (bases, colonnes, chapiteaux, plaques de chancels et de revêtement) produits dans les carrières de Proconnèse et exportés ensuite dans les régions de l’Empire en vue de servir à l’aménagement d’édifices religieux, présentent généralement des lettres grecques incisées à un endroit bien visible, tels le bourrelet d’une colonne, la plinthe d’une base, l’abaque ou la corbeille d’un chapiteau.42 L’étude du décor sculpté de Sainte-Sophie a révélé de

Un deuxième chapiteau en marbre à un seul rang de feuilles d’acanthe (cat. no 4, fig. 10 a-f) a été trouvé dans la salle « B », au sud-est de la maison aux colonnes bilobées,46 tandis qu’une troisième pièce (cat. no 5, fig. 11 a-c) a été découverte dans la salle « C2 », au sud-ouest de cette zone.47 Il convient d’ajouter à cette série un petit chapiteau en marbre aux dimensions similaires dont la provenance n’a malheureusement pas été renseignée (cat. no 6, fig. 12). Les informations consignées dans les carnets de fouilles sont le plus souvent lacunaires et ne fournissent malheureusement pas d’indices susceptibles d’éclairer la fonction et la datation de ces chapiteaux. Les deux chapiteaux trouvés lors de la première campagne de fouilles sont certainement des remplois issus de l’église partiellement exhumée située au nord de cette maison (fig. 6 a-c et 8). Trois des petits chapiteaux en marbre découverts lors de la seconde mission (cat. nos 3, 5 et 6 ; fig. 9, 11 et 12) s’apparentent à des exemplaires protobyzantins provenant de l’étage des portiques de l’atrium de la basilique

Au sujet des importations de sculptures architecturales en marbre de Proconnèse sur le littoral bulgare, voir Barsanti 1989 ; Vanderheyde 2012, pp. 6-9. L’église de Caïči en Abkhazie conserve par exemple deux chapiteaux du même type. Cf. Khroushkova 2006, pp. 138, pl. 106 a et c. Parmi les éléments architecturaux réutilisés au XVIe siècle dans la mosquée de Hamza Bey à Thessalonique, on remarque sept chapiteaux du même type, tandis que d’autres exemplaires analogues ont été repérés à Saint-Dimitrios et dans la collection de sculptures qu’abrite la Rotonde. Cf. Raptis, Basileiadou 2013, pp. 61-63, KP/2-4, 6, 8, 10-11, fig. 7. 40 Stillwell 1941, p. 151, nos 5-8, pl. 33, et pp. 156-157, nos 65 et 68, pl. 32. 41 Van Wonterghem 1978, p. 1. 42 Plusieurs de ces marques de tâcherons ont été repérées et étudiées par Friedrich Wilhelm Deichmann. Voir Deichmann 1976, pp. 206-230. De tels signes sont aussi très diffusés à Istanbul et en Grèce. Cf. Sodini 1987, pp. 503-510; 1989, pp. 163-186; 2000, pp. 423-448. 39

Paribeni 2004, pp. 649-734; 2010, pp. 113-117. Sodini 1987, p. 512. 45 Je remercie Andrea Paribeni pour ses pertinentes remarques à ce sujet. 46 Van Wonterghem 1978, p. 6. 47 Van Wonterghem 1978, p. 9. 43 44

283

Catherine Vanderheyde

Fig. 9a-d. Chapiteau corinthien, « Maison aux colonnes bilobées » (C. Vanderheyde ; détourage A. Stoll).

épiscopale de Kourion à Chypre.48 Deux des chapiteaux réutilisés dans le templon de l’église nord du monastère de Koutsovendis situé sur la même île présentent aussi des similarités avec ces mêmes chapiteaux.49 On peut dès lors supposer que ces chapiteaux sont issus d’un édifice religieux et qu’ils surmontaient peut-être les piliers-colonnettes d’une clôture de sanctuaire, comme le suggère leur module. Ils ont sans doute été remployés par la suite dans une demeure proche. Certains ont pu par exemple servir à la réparation du portique de l’étage donnant sur la cour à péristyle de la « maison aux colonnes bilobées » ou à d’autres installations. Ils semblent donc bien appartenir à la période protobyzantine. Cependant, comme les chapiteaux en marbre à un rang de feuilles d’acanthe trouvés à Antioche présentant des dimensions analogues, la fonction initiale des 3 chapiteaux réutilisés dans la maison aux colonnes bilobées d’Apamée reste difficile à déterminer.50 Le chapiteau no 4 (fig. 10) présente un rang de feuilles d’acanthe molle dont les nervures sont

exécutées suivant la technique du « trépan courant ».51 Ses caractéristiques techniques et décoratives sont différentes des trois autres pièces répertoriées et invitent à le dater d’une période plus haute. Conclusions La concentration de ces chapiteaux en marbre dans la « maison aux colonnes bilobées » ne manque pas d’étonner. Bien que la fouille de cette demeure n’ait pas été terminée, les vestiges du pavement en mosaïque et les éléments architecturaux sculptés dénotent un luxe qui laisse supposer la richesse et probablement le statut important de ses propriétaires successifs. S’ils ne sont pas exactement identiques, les petits chapiteaux à un rang de feuilles d’acanthe provenant du même endroit sont intéressants, car ils sont plus rarement documentés que les autres types de chapiteaux. Le fait qu’ils soient en marbre et que l’un d’eux présente une lettre incisée indique qu’ils faisaient partie d’éléments architecturaux importés. La variété et la qualité d’exécution de leur décor montrent que

Megaw 2007, pp. 189-191. Mango 1990, p. 71, chapiteau C et D, fig. 42 et 43. 50 Stillwell 1941, pp. 152-153, no 27, pl. 34, et no 31, pl. 35. 48

Au sujet de cette technique utilisée depuis l’époque archaïque, voir Braunstein 2010, pp. 71-96.

49

51

284

Chapiteaux corinthiens inédits d’Apamée (Syrie)

Fig. 10a-f. Chapiteau corinthien, « Maison aux colonnes bilobées » (C. Vanderheyde ; détourage A. Stoll).

Annexe : Catalogue

l’on a affaire à une production plus soignée que celle qui caractérise le chapiteau corinthien no 1 à feuilles d’acanthe épineuse à échancrure en « V » sous le bouton d’abaque (fig. 6 a-c) dont le type est largement diffusé dans les diverses régions de l’Empire. D’une manière générale, la majorité des chapiteaux en marbre étudiés témoigne de l’empreinte de l’esthétique constantinopolitaine dans l’aménagement liturgique des édifices religieux situés dans les provinces orientales de l’Empire. Ces chapiteaux furent manifestement encore considérés comme précieux et décoratifs bien après leur usage initial comme l’atteste leur réutilisation au sein d’une riche demeure d’Apamée.

Chapiteau corinthien (fig. 6 a-c) No inv. AP. 70. V.6.54 Trouvé le 05/09/1970 dans la salle A. Marbre blanc-jaunâtre avec d’épaisses veines gris clair. État de conservation : plus de la moitié du chapiteau est conservée ; cassé sur une face et demie et au lit d’attente (cassure transversale se prolongeant jusqu’à 8,5 cm du lit de pose) ; ébréché au lit d’attente.

285

Catherine Vanderheyde

Fig. 11a-c. Chapiteau corinthien, « Maison aux colonnes bilobées » (C. Vanderheyde ; détourage A. Stoll).

Fig. 12a-c. Chapiteau corinthien, « Maison aux colonnes bilobées » (C. Vanderheyde ; détourage A. Stoll).

286

Chapiteaux corinthiens inédits d’Apamée (Syrie) Haut. 33,5 cm ; diam. du lit de pose 32 cm ; trou de goujon (diam. 4 cm) avec coulée de plomb encore en place. Lit d’attente : dim. max. conservée de bouton d’abaque à bouton d’abaque 48 cm. Lit d’attente et lit de pose piquetés à la pointe (traits de 1 à 3 cm).

rainuré de l’abaque (haut. bandeau inf. 3 cm). L’exécution de l’ensemble est soignée : les nervures des feuilles sont précisément incisées, mais la forme générale des feuilles d’acanthe demeure schématique. Lit d’attente présentant un tracé préparatoire circulaire et des traces d’utilisation de pointe fine.

Chapiteau corinthien à deux rangs de feuilles d’acanthe. La partie inférieure comporte une couronne de quatre feuilles épineuses déployées en largeur, dont les lobes sommitaux à digitation «  en fleuron  »52 sont recourbés vers l’avant et sculptés en haut relief. Leur épaisse nervure centrale est délimitée par deux sillons presque parallèles, tandis que les feuilles sont composées de cinq lobes à deux et cinq digitations marqués par un sillon central. À la rencontre des digitations, triangle, trapèze et carré sur la pointe sculptés en creux se superposent. La couronne d’acanthe supérieure comporte quatre feuilles d’angle à trois lobes. Leur nervure centrale est délimitée par deux sillons presque parallèles. Leurs premiers lobes latéraux sont schématiques et présentent une forme semicirculaire. Les deuxièmes lobes latéraux comportent deux digitations, tandis que le lobe sommital en relief présente une digitation « en fleuron ». Un bandeau plat (larg. 1 cm) longe les feuilles d’angle et décrit une échancrure en « V » ou « en lyre » sous le bouton d’abaque. Il se termine par deux petites volutes et devait comporter un motif différent sur chaque face. Deux des motifs sont conservés : une croix latine cassée à la partie supérieure (9 × 11 cm) se prolongeant sur le bouton d’abaque et se détachant en relief (environ 5 mm) du fond lisse de la corbeille ; un motif végétal couvrant le bouton d’abaque, constitué d’une tige sinueuse d’où émergent des feuilles et des bourgeons. Abaque (haut. 5,5 cm) rainuré à deux bandeaux de 2 et 3 cm de large. Astragale constitué d’un listel d’1 cm de large.

3. Chapiteau (fig. 9 a-d) No inv. AP 78. V.1 Trouvé le 20/09/1978, secteur B, à 40 cm de profondeur, dans une couche de remblai. Marbre blanc-jaunâtre avec quelques veines grisâtres. État de conservation : en deux fragments jointifs ; un angle et une face cassés. Haut. 15,5 cm ; diam. max. du lit de pose 22 cm ; diam. de la surface circulaire en saillie d’environ 4 mm correspondant à la section de la colonne supportant le chapiteau : 16 cm. Lit d’attente piqueté : 26 × 27 cm. Chapiteau corinthien à une couronne de feuilles d’acanthe à cinq lobes comportant deux et trois digitations situées aux angles du chapiteau. La jonction des digitations décrit un rectangle auquel se superpose une forme trapézoïdale. La tige centrale de chacune des feuilles est sculptée en relief et présente un lobe sommital en forme d’amande. Une marque gravée «  ε  » est visible sur la tige de l’une des feuilles d’acanthe. L’abaque est divisé en deux parties égales par une rainure, se prolongeant sur le bouton central dont la partie inférieure entame la corbeille. L’exécution du décor végétal est précise et soignée : la partie centrale de chacune des digitations des feuilles d’acanthe est marquée d’une fine et profonde incision. 4. Chapiteau (fig. 10 a-f) No inv. AP.78.V.14 Trouvé le 03/10/1978, secteur B, dans une couche avec traces de calcination, à 3 m de profondeur. Marbre blanc friable à gros grains. État de conservation: un angle et deux boutons d’abaque cassés. Lit d’attente fendu. Haut. 17,5 cm.

Chapiteau (fig. 8) No inv. AP. 70. V.6.55 Trouvé le 05/09/1970 dans la salle A. Marbre blanc-jaunâtre parcouru de veines grisâtres. État de conservation : seule la partie supérieure subsiste ; cassé en bas, à l’arrière et sur les côtés ; boutons d’abaque cassés. Haut. max. conservée 23,5 cm. Diam. max. lit d’attente 51 × 21,5 cm ; diam. reconstitué environ 39 cm.

Diamètre max. du lit de pose 19 cm. Surface circulaire (diam. 15,5 cm) en saillie de 5 mm correspondant au diamètre de la colonnette supportant ce chapiteau. Trou de goujon circulaire (diam. 4 cm ; prof. 5 cm). Traces de ciseau grain d’orge et tracé préparatoire radian.

Chapiteau corinthien présentant à l’origine deux rangs de feuilles d’acanthe. Le rang supérieur comportait quatre feuilles à trois lobes présentant deux et trois digitations. Les feuilles au lobe sommital recourbé sont placées de part et d’autre d’une croix grecque (11 × 11 cm) sculptée en relief (saillie du relief par rapport au fond : 5 mm). Celleci était surmontée d’une lunule interrompant le bandeau

Lit d’attente (28,5 × 28 cm) avec traces de ciseau grain d’orge. Quatre lignes préparatoires à l’incision, partant d’un point central, sont également visibles. Chapiteau corinthien à une couronne de feuilles d’acanthe. Chaque feuille comporte cinq lobes à trois et sept digitations. Des nervures profondes incisées au trépan sont situées entre les lobes. Le rendu des feuilles d’acanthe apparaît ici plus abouti que sur les autres chapiteaux. À la jonction des digitations, deux petits triangles et un trapèze

52 En analysant la composition des feuilles d’acanthe, Annie Pralong a distingué divers types de lobes sommitaux dont elle offre une terminologie commode à utiliser. Voir Pralong 1993, pp. 137-139, fig. 7.3.

287

Catherine Vanderheyde Balty J. (éd.), 1984. Apamée de Syrie. Bilan des recherches archéologiques, 1973-1979. Aspects de l’architecture domestique d’Apamée. Actes du colloque tenu à Bruxelles les 29, 30 et 31 mai 1980, Bruxelles (Fouilles d’Apamée de Syrie. Miscellanea, 13).

surmontés d’une feuille triangulaire située sous le bouton d’abaque. De la pointe de cette dernière partent deux bandeaux (larg. 1,5 cm) qui longent l’abaque (haut. 2 cm) et se terminent par une volute flanquant le lobe sommital de chacune des feuilles d’acanthe du chapiteau.

Balty J., Balty J.-C. (éd.), 1972. Apamée de Syrie. Bilan des recherches archéologiques, 1969-1971. Actes du colloque tenu à Bruxelles les 15, 17 et 18 avril 1972, Bruxelles (Fouilles Apamée de Syrie. Miscellanea, 7), pp. 187-205.

5. Chapiteau (fig. 11a-c) No inv. AP.78. V.24 Trouvé le 09/10/1978, secteur C2, à 2,80 m de profondeur. Marbre blanc avec veines grisâtres de couleur claire et plus foncée. État de conservation : Une moitié et la partie inférieure de l’une des faces sont cassées ; abaque ébréché. Haut. 16 cm. Diamètre restitué du lit de pose 20 cm (diam. cons. 16 cm) ; trou de goujon à moitié conservé (diam. environ 3 cm ; prof. 5,5 cm). Sur le lit de pose sont tracées des lignes préparatoires à l’incision. Lit d’attente 22,5 × 11,5 cm.

Balty J.-C., 1981. Guide d’Apamée, Bruxelles/Paris. Balty J.-C., 1989. ‘Apamée au VIe siècle. Témoignages archéologiques de la richesse d’une ville’, in Morrisson, Lefort 1989, pp. 79-96. Baratte F., ‘La maison des chapiteaux à consoles. Bilan des trois premières campagnes de fouilles’, in Balty, Balty 1972, pp. 107-125. Barsanti C., 1989. ‘L’esportazione di marmi dal Proconneso nelle regioni pontiche durante il IV-VI secolo’, in Rivista dell’Istituto nazionale di archeologia e storia dell’arte. III Serie, 12, pp. 91-220.

Chapiteau corinthien à quatre feuilles d’acanthe angulaires dont deux sont conservées. Chacune des feuilles présente cinq lobes et une ou trois digitations. La partie supérieure du lobe central est collée à la corbeille. À la jonction des digitations des feuilles, petit trapèze et feuille lisse à arête médiane saillante dont la pointe triangulaire remonte sur le bouton d’abaque. L’abaque (haut. 3,5 cm) est lisse.

Barsanti C., 1995. ‘Alcune riflessioni sulla diffusione dei materiali di marmo proconnesio’, in Akten des XII. Internationalen Kongresses fur̈ christliche Archaölogie, Bonn, 22.-28. September 1991, Münster (JbAC. Ergänzungsband, 20/2), pp. 515-523.

6. Chapiteau (fig. 12 a-d)

Barsanti C., Guiglia Guidobaldi A., Sodini J.-P., 1998, ‘La sculpture architecturale en marbre au VIe siècle à Constantinople et dans les régions sous influence constantinopolitaine’, in N. Cambi, E. Marin (éd.), Radovi XIII međunarodnog kongresa za starokršćansku arheoligiju, Split-Poreč, 25.9.-1.10.1994 – Acta XIII congressus internationalis archaelogiae christianae, Split-Poreč, 25.9.-1.10.1994, II, Cité du Vatican/ Split (Studi di Antichità cristiana, 54/2; Vjesnik za arheologiju i historiju dalmatinsku. Suplement, 88), pp. 301-376.

No inv. ? Provenance : maison aux colonnes bilobées (?) Marbre blanc-jaunâtre avec veines grisâtres. État de conservation : boutons d’abaque cassés ; ébréché sur l’une des faces ; un des angles ébréché. Haut. 16 cm. Diamètre du lit de pose : 19 cm. Lit d’attente travaillé à la pointe : 27 × 28 cm. Feuilles d’acanthe angulaires à cinq lobes comportant une à deux digitations. À la jonction des digitations : losange et petit trapèze d’où émerge une feuille lisse à arête centrale se prolongeant sur le bouton d’abaque. À la partie supérieure du lobe central de chacune des feuilles est sculptée une proéminence ovoïdale. L’abaque (haut. 3,5 cm) est rainuré et constitué de deux bandeaux (larg. 1,5 et 2 cm).

Braunstein D., 2010. ‘L’emploi du trépan dans la sculpture archaïque : la technique du trépan courant’, in Bulletin de correspondance hellénique, 134, pp. 71-96. Comte M.-C., 2012. Les reliquaires du Proche-Orient et de Chypre à la période protobyzantine (IVe-VIIIe siècles). Formes, emplacements, fonctions et cultes, Turnhout (Bibliothèque de l’Antiquité tardive 20).

Bibliographie

Deichmann F. W., 1976. Ravenna. Hauptstadt des spätantiken Abendlandes, II/2, Wiesbaden.

Abbréviations : BMRAH = Bulletin des Musées royaux d’art et d’histoire.

Delehaye H., 1935. ‘Saints et reliquaires d’Apamée’, in Analecta Bollandiana, 53, pp. 225-244.

JbAC = Jahrbuch für Antike und Christentum.

Gessel W., 1988. ‘Das Öl der Märtyrer. Zur Funktion und Interpretation der Ölsarkophage von Apamea’, in Oriens Christianus, 72, pp. 183-202.

Balty J. (éd.), 1969. Apamée de Syrie. Bilan des recherches archéologiques, 1965-1968. Actes du colloque tenu à Bruxelles les 29 et 30 avril 1969, Bruxelles (Fouilles d’Apamée de Syrie. Miscellanea, 6).

Kautzsch R., 1936. Kapitellstudien. Beiträge zu einer Geschichte des spätantiken Kapitells im Osten vom 288

Chapiteaux corinthiens inédits d’Apamée (Syrie) vierten bis ins siebente Jahrhundert, Berlin/Leipzig (Studien zur spätantiken Kunstgeschichte, 9).

Pralong A., 1993. ‘Remarques sur les chapiteaux corinthiens tardifs en marbre de Proconnèse’, in L. Pressouyre (dir.) L’acanthe dans la sculpture monumentale de l’Antiquité à la Renaissance, Paris (Mémoires de la Section d’archéologie et d’histoire de l’art, 4; Histoire de l’art, 6), pp. 133-146.

Khroushkova L., 2006. Les monuments chrétiens de la côte orientale de la mer Noire. Abkhazie, IVe–XIVe siècles (Bibliothèque de l’Antiquité tardive 9), Turnhout. Maraval P., 1985. Lieux saints et pèlerinages d’Orient. Histoire et géographie des origines à la conquête arabe, Paris (Cerf. Histoire).

Pralong A., 2000. ‘La typologie des chapiteaux corinthiens tardifs en marbre de Proconnèse et la production d’Alexandrie’, in Revue archéologique, 2000/1, pp. 81-101.

Mayence F., 1935a. ‘La quatrième campagne de fouille à Apamée’, in BMRAH, 7/1, pp. 2-10.

Raptis K. Th., Basileiadou, St. D., 2013. ‘Αρχιτεκτονικά γλυπτά της ύστερης αρχαιότητας και των βυζαντινών χρόνων στο Χαμζά Μπέη τζαμί στη Θεσσαλονίκη’, in Δελτίον τῆς Χριστιανικῆς Ἀρχαιολογικῆς Ἑταιρείας, 34, pp. 53-65.

Mayence F., 1935b. ‘La quatrième campagne de fouille à Apamée (rapport sommaire)’, in L’Antiquité classique, 4, pp. 199-204. Megaw A. H. S, 2007. Kourion. Excavations in the Episcopal Precinct, Washington (Dumbarton Oaks Research Collection, 38).

Sodini, J.-P. 1987. ‘Marques de tâcheron inédites à Istanbul et en Grèce’, in X. Barral i Altet (éd.), Artistes, artisans et production artistique au Moyen Âge, II, Paris, pp. 503-510.

Morrisson C., Lefort J. (éd.), 1989. Hommes et richesses dans l’Empire byzantin, I, Paris (Réalités byzantines, 1).

Sodini J.-P., 1989. ‘Le commerce des marbres à l’époque protobyzantine’, in Morisson, Lefort 1989, pp. 163-186.

Naccache A., Sodini J.-P. 1989. ‘Le décor architectural de Syrie byzantine’, in J.‑M. Dentzer, W. Orthmann (éd.), Archéologie et histoire de la Syrie, II, Sarrebruck (Schriften zur vorderasiatischen Archäologie, 1/2), pp. 447-490.

Sodini J.-P., 2000. ‘Le commerce des marbres en Méditerranée (IVe-VIIe s.)’, in J. M. Gurt, N. Tena (éd.), V Reunió d’Arqueologia Cristiana Hispànica, Cartagena, 16-19 d’abril de 1998 – V Reunión de Arqueología Cristiana Hispánica, Cartagena, 16-19 de abril 1998, Barcelone (Monografíes de la secció histórico-arqueológica, 7), pp. 423-448.

Napoleone-Lemaire J., Balty J.-C. (dir.), 1969. Fouilles d’Apamée de Syrie, I/1, Bruxelles. Niewöhner Ph., 2010. ‘Der frühbyzantinische Rundbau beim Myrelaion in Konstantinopel. Kapitelle, Mosaiken und Ziegelstempel’, in Istanbuler Mitteilungen, 60, pp. 411-459.

Spieser J.-M., 1984. Thessalonique et ses monuments du IVe au VIe siècle. Contribution à l’étude d’une ville paléochrétienne, Paris (Bibliothèque des Écoles françaises d’Athènes et de Rome, 254).

Nikolajević-Stojković I., 1957. Рановизантиска архитектонска декоративна пластика у Македонији, Србији и Црној Гори. Дисертација – La décoration architecturale sculptée de l’époque basromaine en Macédoine, en Serbie et au Monténegro, Belgrade.

Stillwell R. (dir.), 1941. Antioch on-The-Orontes, III, Princeton (Publications of the Committee for the Excavation of Antioch and its Vicinity). Strube C., 1983. ‘Die Kapitelle von Qasr Ibn Wardan. Antiochia und Konstantinopel im 6. Jahrhundert’, in JbAC, 26, pp. 59-106.

Paribeni A., 2004. ‘Le sigle dei Marmorari e l’organizzazione del cantiere’, in A. Guiglia Guidobaldi, C. Barsanti (éd.), Santa Sofia di Costantinopoli. L’arredo marmoreo della Grande Chiesa giustinianea, Cité du Vatican (Studi di Antichità cristiana, 60), pp. 649-734.

Vaklinova M., 1985. ‘Ateliers de décoration architecturale au Ve et VIe siècle dans la région de Nicopolis ad Nestum (Bulgarie)’, in Πρακτικά του 10ου διεθνούς συνεδρίου χριστιανικής αρχαιολογίας, Θεσσαλονίκη 28 Σεπτεμβρίου – 4 Οκτωβρίου 1980 – Actes du Xe congrès international d’archéologie chrétienne, Thessalonique, 28 septembre – 4 octobre 1980, Cité du Vatican/Thessalonique (Studi di Antichità cristiana, 37; Ελληνικά. Περιοδικόν Σύγγραμμα Εταιρείας Μακεδονικών Σπουδών. Παράρτημα, 26), pp. 641-649.

Paribeni A., 2010. ‘The artist’s signature: marble masons’ marks’, in C. Barsanti, A. Guiglia (éd.), The Sculptures of the Ayasofya Müzesi in Istanbul. A Short Guide, Istanbul, pp. 113-118. Peschlow U., 2001-04. ‘Kapitelle’, in G. Schöllgen, H. Brakmann, A. Dihle, J. Engemann, K. Hoheisel, W. Speyer, K. Thraede (dir.), Reallexikon für Antike und Christentum. Sachwörterbuch zur Auseinandersetzung des Christentums mit der antiken Welt, XX, Stuttgart, col. 7-123.

Vanderheyde C., 2003. ‘Objets et éléments décoratifs en pierre issus d’Apamée’, in BMRAH, 74, pp. 63-106. Vanderheyde C., 2012. ‘The architectural decoration of the early Byzantine churches on the west coast of the Black Sea’, in Проблеми на изкуството, 2, pp. 6-9.

289

Catherine Vanderheyde Vanderheyde C., 2015. ‘Compte rendu : M.-C. Comte, Les reliquaires du Proche-Orient et le culte des martyrs’, in Journal of Roman Archaeology, 28, pp. 941-945. Van Wonterghem F., 1978. Maison à la colonne bilobée, carnet de fouilles conservé au Centre belge de recherches archéologiques à Apamée de Syrie (Musée Arts & Histoire, Bruxelles). Verhoogen V., 1964. Apamée de Syrie aux Musées royaux d’art et d’histoire, Bruxelles. Zollt T., 1994. Kapitellplastik Konstantinopels vom 4. bis 6. Jahrhundert n. Chr. Mit einem Beitrag zur Untersuchung des ionischen Kämpferkapitells, Bonn (Asia-Minor-Studien, 14).

290

22 Byzantine Capitals of the Architectural Complex of the ‘Basilica of 1935’ in Chersonesos in Crimea Liudmila G. Khrushkova* Моско́вский госуда́рственный университе́т и́мени М. В. Ломоно́сова / Lomonosov Moscow State University (Russia) Most of the Early Byzantine basilicas of Chersonesos in the Crimea were excavated in the late nineteenth to early twentieth century. The only exception was the so-called ‘Basilica of 1935’ investigated by the archeologist Grigory D. Belov (1898–1979) in collaboration with Anatoly L. Yakobson (1906–1984). It is a complex of four successive structures. First founded as a small synagogue, it was subsequently replaced by a сhurch, which was later rebuilt and expanded. The latter, a three-nave basilica with narthex and additional rooms to the south, was named the ‘Basilica of 1935’. In the late Middle Ages, a chapel was built on the ruins of the structure. It underwent restorations in 1936, 1956 and to some extent again in the 2000s. Today, the basilica does not contain only the capitals that were discovered in the initial investigation. The 1935–1936 excavations found a series of early Byzantine capitals of three types: 1) composite capitals with two rows of openwork leaves of fine-toothed acanthus (the so-called ‘Theodosian’ capitals). Seven examples of this type are known in Chersonesos, four of which have been proven to originate from the basilica and are dated to the second half of the fifth century; 2) Corinthian capitals of the second half of the fifth century, three of which are of the common type with two rows of leaves of acanthus and one of which, the lost capital, belongs to the Corinthian ‘V-shaped’ subtype (according to Rudolf Kautzsch); 3) impost capitals, one of which has survived to the present day and is analogous to the capitals of the church of St John Stoudios in Constantinople (453). According to the prevailing opinion, the capitals and the other elements of the marble decoration of the complex are from two different structures. In my opinion, all the capitals belong to the same ‘Basilica of 1935’, which was built in the last quarter of the fifth century. À Chersonèse en Crimée, la plupart des basiliques protobyzantines ont été fouillées à la fin du XIXe et au début du XXe siècle. La seule exception est la « Basilique de 1935 », étudiée par Grigorij D. Belov (1898–1979), en collaboration d’Anatolij L. Yakobson (1906–1984). Le complexe est constitué de quatre édifices successifs. Le plus ancien est une petite synagogue, remplacée par une église à nef unique qui, à son tour, a été transformée en basilique à trois nefs, dotée d’un narthex ainsi que d’annexes du côté sud et appelée « Basilique de 1935 ». Une petite chapelle a été bâtie sur ses ruines vers la fin de l’époque médiévale. Des travaux de restauration ont été ménés en 1936, en 1956 et au début des années 2000. Aujourd’hui on voit dans la basilique non seulement des chapiteaux découverts durant les fouilles de 1935-1936, mais aussi d’autres qui proviennent d’autres sites de Chersonèse. Les fouilles de G. Belov ont livré des chapiteaux de trois types : 1) des chapiteaux composites à deux couronnes de feuilles d’acanthes finément dentelées (dits chapiteaux « théodosiens ») : sept spécimens datant de la seconde moitié du Ve siècle sont connus à Chersonèse, dont quatre proviennent de la « Basilique de 1935 » ; 2) des chapiteaux corinthiens de la seconde moitié du Ve siècle, dont trois du type courant, avec deux rangées de feuilles, et un autre appartenant au type « en V » (selon Rudolf Kautszch) ; 3) un chapiteau à imposte qui trouve un parallèle proche dans les chapiteaux de l’église du monastère de Stoudios à Constantinople (453). Selon l’opinion admise par les chercheurs de Chersonèse, les chapiteaux et d’autres éléments de décor en marbre de ce complexe appartiennent à deux édifices distincts, l’un du Ve et l’autre du VIe siècle. À mon avis, tous les chapiteaux découverts dans les fouilles proviennent de la « Basilique de 1935 » qui a été bâtie dans le dernier quart du Ve siècle.

*

Translation from the Russian by Kesha Gelbak.

291

Liudmila G. Khrushkova Ein Großteil der frühbyzantinischen Basiliken der Krim wurden an der Wende vom 19. zum 20. Jh. ausgegraben. Die von Grigory Below (1898–1979) und Anatoly L. Yakobson (1906–1984) erforschte sogenannte „Basilika von 1935“ bildet eine Ausnahme. Dieses Ensemble bestand aus vier aufeinander folgenden Bauphasen: Zuerst entstand eine kleine Synagoge, die von einer Kirche abgelöst wurde. Darauf folgte eine dreischiffige Basilika mit Narthex und Nebenräumen auf der Südseite mit dem Namen „Basilika von 1935“. Im Spätmittelalter wurde auf den Ruinen der Basilika eine Kapelle errichtet. In den Jahren 1936, 1956 und während der 2000er wurden Restaurierungsarbeiten durchgeführt. Nicht alle Kapitelle der Säulen wurden bei den Ausgrabungen entdeckt. Bei den Ausgrabungen zwischen 1935–1936 wurden drei Typen von Kapitellen registriert: 1) Kompositkapitelle mit feinem Akanthus-Durchbruchmuster (sog. theodosianische Kapitelle). Sieben solche Stücke sind aus Chersones insgesamt bekannt, darunter vier aus der „Basilika von 1935“. Sie datieren in die zweite Hälfte des 5. Jhs. 2) Korinthische Kapitelle aus der zweiten Hälfte des 5. Jhs. Drei davon sind vom klassischen Typ mit liegenden Akanthusreihen und eine, heute verschollene, gehört zu der V-geformten Untergruppe nach Rudolf Kautzsch. 3) Ein Kämpferkapitell. Seine Analogie ist aus der Studionbasilika in Konstantinopel (453) bekannt. Es wird angenommen, dass die Kapitelle und die anderen Elemente der Marmordekoration verschiedenen Bauphasen angehören, die wiederum auch unterschiedlich datiert werden. Ich vertrete die Meinung, dass alle Kapitelle der dreischiffigen „Basilika von 1935“ zuzuschreiben sind, die im letzten Viertel des 5. Jhs. entstand. La maggior parte delle basiliche paleocristiane del Chersoneso in Crimea sono state scavate alla fine del XIX – inizio del XX secolo. Un’eccezione è costituita dalla cosiddetta “Basilica del 1935”, studiata da Grigorij Belov (1898-1979) in collaborazione con Anatoly L. Yakobson (1906-1984). Questo complesso è formato da quattro edifici, che si sono avvicendati l’uno dopo l’altro. Il più antico era una piccola sinagoga che fu sostituita da una chiesa mononavata; più tardi quest’ultima fu trasformata in una grande chiesa di tipo basilicale a tre navate con nartece e ambienti sussidiarii sul lato sud. Questa è stata denominata “Basilica del 1935”. In epoca tardomedievale sui suoi resti fu costruita una piccola cappella. I restauri sono stati condotti nel 1936, 1956 e, parzialmente, negli anni 2000. Attualmente nella basilica si trovano diversi capitelli, non tutti rinvenuti negli scavi ma provenienti anche da altri siti del Chersoneso. Gli scavi di G. Belov hanno riportato alla luce una serie di capitelli di tre tipi: 1. Capitelli compositi con due corone di foglie di acanto finemente dentellato (cosiddetti «capitelli teodosiani»). Dei sette esemplari di questo tipo, ritrovati nel Chersoneso, quattro provengono dalla «Basilica del 1935» e sono datati alla seconda metà del V secolo. 2. Capitelli corinzi della seconda metà del V secolo, tra i quali tre di tipo «normale», e uno, perduto, del tipo a “V” (5° gruppo della classificazione di Rudolf Kautzsch). 3. Capitelli imposta, uno dei quali è ben conservato e presenta strette analogie con quelli della chiesa di S. Giovanni di Studio a Costantinopoli del 453. Secondo l’opinione più diffusa tra gli studiosi del Chersoneso i capitelli e gli altri elementi della decorazione marmorea del complesso appartengono a due diversi edifici. A mio parere invece tutti i capitelli appartengono soltanto alla «Basilica del 1935», che è stata costruita nell’ultimo quarto del V secolo. More than a dozen large churches, mostly basilicas, were built at Chersonesos in Crimea in the fifth–sixth century. Byzantine marble was widely used. Many structural and decorative elements of these churches, i.e., their liturgical furnishings and pavements, were made of marble brought from the workshops of Proconnesos. The extensive use of marble is comparable to that at Ravenna and Ephesus. Apart from those in the State Historical and Archaeological Museum-Preserve ‘Tauric Chersonese’,1 collections of marbles from Chersonesos are stored in the Hermitage, in the State Historical Museum in Moscow and in the Historical Museum of the city of Odessa (Ukraine). Some of the capitals in the Hermitage Museum and the Odessa Museum have been published, but the more extensive Moscow collection remains almost unknown.

Most of the marbles were discovered during archaeological works of the last third of the nineteenth century and the beginning of the twentieth century, when most of the basilicas of Chersonesos were excavated. Unfortunately, due to the imperfect excavation methods of that era, we do not now know in which churches these marbles were found.2 In Crimean archaeology, the study of marble has never been a priority. I will mention the most important works. Anatoly L. Yakobson (1906– 1984) in his book Early-Medieval Chersonesos, devotes a chapter to the marble.3 Half a century later, Andrzej B. Biernacki published a catalogue of the marbles of Chersonesos.4 The Crimean marble is known to Western readers thanks to the extensive work of Claudia Barsanti

1 For the sake of brevity, I will refer to the State Historical and Archaeological Museum-Preserve ‘Tauric Chersonese’ using its traditional name: the Chersonesos Museum.

2

Khrushkova 2012b, pp. 232–240; 2016c, pp. 9–46; 2017a, pp. 856–872. Yakobson 1959, pp. 130–152. 4 Biernacki 2009. 3

292

Byzantine Capitals of the Architectural Complex of the ‘Basilica of 1935’ in Chersonesos in Crimea on marbles of the Pontic area,5 as well as some more recent publications.6

The so-called ‘Basilica of 1935’ is the only large complex in Chersonesos excavated under the Soviet government. The excavations of the coastal area in which the basilica is located were conducted by Grigory D. Belov (1898– 1979), and we have a detailed published report. In this report, Belov gives most of his attention to the monuments of the Roman period; archaeologists of that period did not focus on the marble decoration of the basilica.

The aim of this paper is to examine the marble capitals from the so-called ‘Basilica of 1935’. This is important for the study of the building phases of this large complex, the chronology of which has long been debated.  The ‘Basilica of 1935’

The basilica was twice restored, at the end of the excavation, in 1936 (when three whole columns were put together out of fragments, and three capitals were placed on them; fig. 1)8 and then a second time in the 1950s. The latter works gave the basilica its final appearance (fig. 2).9 During both interventions, some architectural elements from other monuments were moved here. Our goal is to identify the capitals that we may reliably conclude were found in the ‘Basilica of 1935’.

Only in rare cases can we associate the marbles with the buildings in which they were found. Among these buildings is the Episcopal Basilica of Chersonesos (also known as the ‘Uvarov’ Basilica, or Basilica No 23). Its marble and mosaic decor has recently been studied in its architectural context.7 The ‘Basilica of 1935’, like the Episcopal church, was one of the most important religious centres of the city. Both complexes occupy a privileged location in the oldest part of the city; it is very likely that these complexes were connected by the processions of the stational liturgy.

The expression ‘Basilica of 1935’ is a conditional term adopted to describe a complex of several succeeding

Fig. 1. ‘Basilica of 1935’, view to the west (Belov 1938, fig. 71). Belov 1938, p. 108, figs. 70–71. Zherebtsov 2009, p. 147; Khrushkova 2013b, fig. 3; Klenina 2016, fig. 3. Biernacki’s catalogue lists the capitals currently in the basilica. In some cases it is stated that the object comes from the ‘Basilica of 1935’, but there is no source for that information: Biernacki 2009, pp. 106–107, 195 212, 223, 225, 226, 233, 234, 239, 240, 247, 256, 257, 258, etc. 8

Reproductions of several capitals from Chersonesos: Barsanti 1989, fig. 40, 44, 70, 73, 76, 79, 83, 88, 96, 99 and 146. 6 Khrushkova 2012a, pp. 129–139; 2013a, pp. 387–402; 2013c, pp. 85– 94. 7 Khrushkova 2016a, pp. 327–435; 2017b. 5

9

293

Liudmila G. Khrushkova

Fig. 2. ‘Basilica of 1935’, view to the west (L. G. Khrushkova).

structures. The nature of these transformations has long been debated. The early phase of the complex remains opaque, particularly as relates to the possible presence of a synagogue.10 In my opinion, the most reasonable point of view is as follows. Initially, a small synagogue was built on this site. Then it was rebuilt into a Christian basilica. This latter building phase quickly collapsed, and was replaced by another, better-preserved one. Finally, during the Middle Ages, in the twelfth to thirteenth centuries, a small chapel was constructed right on the marble floor of the nave of the Basilica.

(Khrushkova 2013b, pp. 264–272). Echoes of these events are felt in the pages of the Belov report, which was written precisely in the darkest year of 1937 and published in 1938. In it, the monuments of Roman antiquity are described in great detail, while the Byzantine ones are treated in a brief and restrained manner. Nevertheless, Belov’s report is an important source of evidence, giving information about the place of discovery of the capitals, their number and the nature of the restoration work. In addition, the report is interesting as a document of an era that was extremely harsh and hostile to Byzantine studies in the USSR.

The basilica was richly decorated with Proconnesian marble. The present chapter deals with the capitals, the most reliable source for which is Belov’s report. Regrettably, it has few illustrations, and it deals very summarily with the marble elements. There is an explanation for this fact. In the 1930s, the Soviet government had a very negative attitude toward Byzantine studies in the Soviet Union, in the context of its consistent policy of compulsory and aggressive atheism. In 1927 came the closure of the only publication of Soviet Byzantinists, the journal Vizantijskij Vremennik (Byzantina Chronika), founded in 1894. The 1930s saw the beginning of open repression against Byzantinists, including historians, philologists and art historians. First came dismissals, arrests, interrogations, imprisonment and exile, then, in 1937, executions

Most of the capitals were discovered in 1935. In the following year, excavations continued in dwellings adjacent to the basilica, to the east. Another three capitals were found. The capitals of the basilica complex ‘of 1935’ are of three types: the Composite, with a fine-toothed acanthus (the so-called ‘Theodosian’), the Corinthian ‘normal’ type, and the impost. Ionic impost capitals are notably absent, although this type is the most common in Chersonesos, where up to 40 per cent of the surviving capitals belong to that type (47 capitals).11 Composite capitals with fine-toothed acanthus Capital 1 (fig. 3) is on display in the medieval history section of the Chersonesos Museum. It was found on the

Khroushkova 2008, pp. 155–156; Jastrzȩbowska 2011–12, pp. 61–73; Zolotarev, Korobkov et al. 2013.

10

11

294

Biernacki 2009, p. 45.

Byzantine Capitals of the Architectural Complex of the ‘Basilica of 1935’ in Chersonesos in Crimea

Fig. 3. Composite capitals with fine-toothed acanthus 1 (L. G. Khrushkova).

Fig. 4. Composite capitals with fine-toothed acanthus 2 (L. G. Khrushkova).

mosaic floor of the south aisle of the Basilica. The capital is fairly well preserved; only one volute is broken off. Its height is 44 cm, its lower diameter 38 cm. In the base of the capital, an opening retains a fragment of an iron rod held in lead. This is probably the best known of all the capitals of Chersonesos.12 All its features are fully consistent with the type of the ‘so-called Theodosian capital, the Composite capital with fine-toothed acanthus’.13 On the astragal there is a row of oblique acanthus leaves, eight in each crown, strongly bent forward. The spaces between the volutes contain four small palmettes, some completed with halfpalmettes at the edges. These palmettes are strongly bent forward from the echinus, forming an openwork crown with triangular and circular holes. The curls of the volutes are decorated with small acanthus teeth, while the edges of the volutes at the corners of the capital are covered with embossed floral motifs. There are clearly cut chevrons on the abacus; their corners are emphasised with drilled holes. In the middle of the abacus there is a strongly protruding cone, also with drilled holes.

type of ‘double’ leaf seen in the most perfect examples of the ‘Theodosian capital’. Capital 2 is on display in the Chersonesos Museum (fig. 4). The capital was found in the masonry of the northern wall of the chapel. Our capital was in the bottom course of the masonry; its abacus faced the interior of the chapel.15 Its height is 43.5 cm, its base diameter 38 cm, and the abacus 53 x 54.5 cm.16 This capital is very similar to Capital 1; Biernacki calls Capitals 1 and 2 ‘practically twins’.17 However, there are differences between these capitals. They are seen in the design of the echinus; on Capital 2 there are three palmettes between the volutes, supplemented with two half-palmettes at the edges, near the volutes. These palmettes also form an openwork crown, but here the holes are shaped differently than in Capital 1; they are rhomboid, oval and triangular. On the front side in the centre of the echinus there is a small ‘Latin’ cross. An important difference from Capital 1 is that the leaf veins of the lower crown are highlighted with a series of round holes. In his time, Guillaume de Jerphanion identified two subtypes of the ‘so-called Theodosian capitals’ on the basis of the presence of such drilled holes.18

Our capital has a notable feature. The leaves of its top crown are very similar to ‘double’ leaves; here the inner part of the leaf, consisting of five lobes of acanthus, is enclosed in a kind of border consisting of small lobes of acanthus. This border is clearly separated from the central portion of the leaf by a series of triangular holes. The ‘double’ leaves of our capital are simplified, not as delicate, and with a smaller number of lobes of acanthus compared with the examples in the Church of St John Stoudios in Constantinople, Hagios Demetrios in Thessaloniki or the Leonidas Basilica in Lechaion.14 Nevertheless, this is the

Capital 3, broken on one side, is in Room 10 of the depot of the Chersonesos Museum (fig. 5). This capital was found during excavation of the north aisle of the ‘Basilica of 1935’.19 The height of the capital is 39.5 cm, its lower diameter 37 cm, and the abacus measures c. 55 x 56 cm.20 Inv. No 203/36983. Belov 1938, pp. 81–82, and 129, fig. 49 and 82. Khroushkova 2008, fig. 33; 2011, fig. 2; 2012a, fig. 4; 2017a, fig. 19. 17 Biernacki 2009, p. 281, pl. 62, 555/973. Biernacki mistakenly reports that the place of discovery of the capital is unknown. 18 de Jerphanion 1930, pp. 96–119. 19 Inv. No 21/35673. Belov 1938, p. 82, fig. 50. 20 Biernacki 2009, p. 294, pl. 63, 603/973. Biernacki does not refer to Belov’s report of 1938, which confirms the origin of the capital from the ‘Basilica of 1935’. Cf. Khroushkova 2008, fig. 32; 2011, fig. 3; 2012a, fig. 5. 15 16

12 Inv. No 203/36983. Belov 1938, pp. 80–81, fig. 48; Yakobson 1959, p. 134, fig. 45,1; Barsanti 1989, fig. 70; Biernacki 2009, p. 328, pl. 62, 773/973; Khrushkova 2011, fig. 1; 2012a, fig. 3. 13 Kautzsch 1936, p. 115: ‘Die schönste Schöpfung, die eigentliche Grosstat des fünften Jahrhunderts…’ (‘The most beautiful creation, the real grandeur of the fifth century ...’). 14 Kramer 1998, p. 56, pl. 13.20, 21, 23; and 14.26–29.

295

Liudmila G. Khrushkova residential complexes of the twelfth to fifteenth centuries were excavated here. A Composite capital with finetoothed acanthus was found in ‘Room 5’, which had served as a courtyard. The capital had been used as a seat. It was damaged by fire and severely cracked.21 This capital could not be identified in the collection of the Chersonesos Museum; in all likelihood it is lost. This specimen was similar to Capital 3; there are drilled round holes on the bottom crown leaves. The volute curls were smooth; there were palmettes on the echinus.

This sample differs in some details from those described above. Capital 3 is slightly smaller; the curls of its volutes are conventionally smooth. On the echinus there are three palmettes with two half-palmettes at the edges. The palmettes form an openwork crown, with triangular holes of different sizes. The junctions of the volutes are covered with floral motifs; in one case there is a plastically interpreted bunch of grapes. Acanthus leaves of the lower crown meet to form a motif of small arches, into which two lobes of adjacent leaves are inscribed. These arches and lines connecting adjacent leaves are emphasised with round holes. The holes do not emphasise leaf veins; instead they form abstract geometric motifs. The abacus is decorated with a motif of scales.

These four capitals, definitely found in the ‘Basilica of 1935’, form a homogeneous group. Capital 5, in the lapidarium of the Chersonesos Museum, is similar to the ones described above (fig. 7). Its origin is unknown. The height of the capital is 41.8 cm; the abacus measures 61 x 59 cm.22 The bent parts of the leaves are lost, as is a volute; the surface of the marble is eroded. The veins of the leaves forming the bottom crown are emphasised with a drill, as in Capital 2. The acanthus leaves of the lower crown join to form arches. The volute curls are smooth, the connections of the volutes are decorated with a floral motif, and there is a motif of scales on the abacus. These features are similar to those of Capital 3. Between the volutes of Capital 5 are five small palmettes, and there are two half-palmettes at the edges, as in Capital 1. Y. Zherebtsov, who excavated and restored the basilica

Capital 4 (fig. 6): In 1936, excavations continued on the site adjacent to the basilica, to the east. Medieval

Fig. 5. Composite capitals with fine-toothed acanthus 3 (L. G. Khrushkova).

Fig. 7. Composite capitals with fine-toothed acanthus 5 (L. G. Khrushkova). Belov 1938, p. 301, fig. 132. Biernacki 2009, p. 177, pl. 63, 55/973; Khrushkova 2011, fig. 5; 2012a, fig. 6. 21

Fig. 6. Composite capitals with fine-toothed acanthus 4 (Belov 1938, fig. 132).

22

296

Byzantine Capitals of the Architectural Complex of the ‘Basilica of 1935’ in Chersonesos in Crimea chancel barrier.29 The total number of surviving ‘Theodosian’ capitals suggests that they all could have decorated a single church. The ‘Theodosian’ capitals decorated the first basilica, built in the second half of the fifth century. This dating is confirmed by numismatic evidence: a coin of Leo I (457–474) and its stratigraphic context.30 Impost capitals We do not know how many Impost capitals were found in the excavations of the ‘Basilica of 1935’.31 At present there are two imposts on columns, but they were placed there in the course of restoration works in the twentieth century. Belov reports that in the narthex, near Grave 26, there lay an unusually large marble impost with a cross on the narrow side. Its length is 1 m, its width 0.85 m, and its height 0.3 m. In 1936, this impost was mounted on the capital in the southern row of columns. It has likely been lost; its photograph is not published.32 All the imposts now kept in the Museum of Chersonesos are significantly smaller than the one found by Belov. According to Yakobson, who took part in the excavations of 1935–1936, several ‘trapezoidal impost capitals with a simple cross on the short side’ and a column base, made of local limestone, were found. These also remain unidentified and unpublished.33

Fig. 8. Composite capitals with fine-toothed acanthus. The State Historical Museum, Moscow (Khrushkova 2016b, fig. 9).

in 1956, confidently writes of five whole ‘Theodosian’ capitals. Was this capital the fifth example? They do make up one series. We know of another two, almost whole Composite capitals with two crowns of fine-toothed acanthus leaves. The place of their discovery is unknown. One of these is stored in the Chersonesos Museum;23 the other is in the State Historical Museum in Moscow (fig. 8).24 Both are smaller than the capitals described above; both present a simplified technique of execution. Biernacki includes the lower parts of three capitals in the Chersonesos Museum among the ‘Composite-Theodosian’ capitals of the ‘Basilica of 1935’.25 We do not, however, know the source of that information. In addition, these three fragments may have been parts of two-zone capitals. There are such capitals in Chersonesos.26 Therefore, I do not group these fragments with the capitals from the ‘Basilica of 1935’. Four fragments of volutes derive from Composite capitals with fine-toothed acanthus. The place of discovery of one of them is not known, and the other three are, presumably, related to the ‘Basilica of 1935’.27 The volutes derive from two different capitals.

Thus, we now know only one impost reliably found in the ‘Basilica of 1935’, in its north aisle. Belov connected it with the first basilica of the fifth century34. The well-preserved impost is displayed in the Museum of Chersonesos; its dimensions are 86 x 54.6/55.7  cm, height 29.5  cm (fig. 9–10).35 Its narrow base is 5 cm in height; it is completed

According to my calculations, seven whole Composite capitals with fine-toothed acanthus were found in Chersonesos; five of these are in the Chersonesos Museum, one in Moscow, and one we know from Belov’s photos. Biernacki’s count has thirteen ‘Composite-Theodosian’ capitals, as he calls them, in the Chersonesos Museum.28 Three small four-leaf capitals with one crown of fine-toothed acanthus leaves could have belonged to the entablature of a

Fig. 9. Impost capital, the front side (L. G. Khrushkova). 29 Biernacki calls them ‘eight-leafed’: Biernacki 2009, pp. 37–39; Khrushkova 2011, fig. 10 and 12; 2012a, fig.–17; 2017a, fig. 10. 30 Belov 1938, pp. 81–82. 31 Biernacki speaks of five imposts from the ‘Basilica of 1935’: Biernacki 2009, p.107, but elsewhere in his book we find alternative information. In point of fact, the place of origin of three out of these five imposts is unknown: ibid., pl. 57/973, p.177, 58/973, p.177 and 584/973, and p. 289. Another impost is said to come from the ‘Basilica of 1935’, but the source of this information is not given: ibid., pl. 139/973, and p. 195. 32 Belov 1938, p. 93, fig. 70. 33 Yakobson 1959, p. 132. 34 Belov 1938, p. 82, fig. 51–52. 35 Inv. No 22/35673; Yakobson 1959, p. 140, fig. 49, 3; Barsanti 1989, fig. 88; Biernacki 2009, p. 279–280, pl. 102, 549/973; Biernacki dates the impost to the second half of the sixth century.

23 Biernacki 2009, p. 295, pl. 66, 606/973; Khrushkova 2011, fig. 4; 2012a, fig. 9. 24 Khrushkova 2016b, fig. 9. 25 Biernacki 2009, pl. 514/973, 610/973 and 620/973 26 Biernacki 2009, pl. 554/973; Khrushkova 2011, fig. 13–14. 27 Biernacki 2009, pp. 296, 303–304, pl. 64, 609 and 643–645/973. 28 Biernacki 2009, pp. 40–42. This number is excessive. One Corinthian capital is included in this group by mistake; it lacks the Ionic element: Khrushkova 2011, fig. 9; 2012a, fig. 14.

297

Liudmila G. Khrushkova

Fig. 10. Impost capital, the back side (L. G. Khrushkova).

by a smooth strip of 6.5 cm in height. The impost is only decorated on the two narrow sides. There is a stonemason’s mark, a И, on the lateral side, near the top. On the front side, there is a ‘Latin’ cross with an elongated lower arm; the ends of the arms widen. It is surrounded by symmetrically arranged soft acanthus leaves, which fill the entire space. Two large leaves incline to the upper arm of the cross; they are supplemented by two long shoots along the edges and by two small trilobed acanthus leaves below. This is the well-known scheme of the ‘acanthus-cross’ (or the ‘acanthus-chrismon’), common in the second half of the fifth and in the sixth century, both on imposts and on Ionic impost capitals.36 On the back side of the impost, there is a composition consisting of three gadroons rounded on one side and supplemented with semi-lunulae on the other.37 Two trilobed soft acanthus leaves are arranged near the edges of the impost.

Fig. 11. Corinthian capital 1 (L. G. Khrushkova).

the central part being wider and more massive, and small leaves stem from the base. The leaves are strongly bent forward, treated somewhat plastically, and the veins are emphasised. Neighbouring leaves meet to form geometric shapes, ovals, rhombuses, trapezoids; this is the so-called ‘acanthus-mask’ technique. In helixes, the curl is well expressed, but the helix itself is plotted more graphically, with a line that delimits a triangular, slightly protruding field. On one side under the abacus, there is the mark of the stonemason: ΘΕ.

An important analogy for the front side of our impost is an impost from the Stoudios Basilica of 453.38 As with the Composite capitals with fine-toothed acanthus, the Stoudios provides a precisely dated analogy. Our impost capital can be dated to the same time as the ‘Theodosian’ capitals from the first ‘Basilica of 1935’.

Capital 2 (fig. 12) was found in the nave of the basilica, in front of the bema. Its height is 59.3 cm, and its abacus measures 68.3 x 68.3 cm.39 The corners of the capitals are partially broken off. In its features, it is very similar to Capital 1. On one side, under the abacus, the same ΘΕ mark is placed.40 It is possible that Capitals 1 and 2 were made by the same craftsman.

Corinthian capitals Belov’s report mentions the Corinthian capitals very briefly; they are barely discernible in photographs of the general view. Capitals 1, 2 and 3 are very similar in size; they are identical in type, technique and style.

Capital 3 (fig. 13), on which now rests an impost, was found standing in the corner of the medieval chapel. The height of the capital is 58.5 cm, and its abacus measures 68.2 x 68.2 cm; its base diameter is 47.5 cm.41 This capital, in contrast to the two previous ones, has eight leaves of acanthus in each crown. This gives the geometric forms

Capital 1 (fig. 11) was found in the eastern corner of the nave. Its height is 59 cm, and its abacus measures 68 x 68 cm. The massive abacus is provided with deep profiling; in the middle there is a strongly protruding cone decorated with stylised leaves of acanthus. In the upper crown there are eight leaves of acanthus, while in the lower crown there are seven leaves. The leaves are three-lobed,

Belov, p. 92–93, fig. 55; Khrushkova 2012a, fig. 84; Biernacki 2009, p. 326, pl. 47, 768/973. Biernacki makes no reference to Belov’s report, which mentions this capital. In his descriptions of this and other Corinthian capitals, Biernacki regularly uses the term ‘pendentive’ to refer to the element between the helices. This architectural term seems inappropriate in this context. 40 Belov 1938, pp. 92–93, fig. 56; Biernacki 2009, p. 319, pl. 47, 731/973. Here too Biernacki omits all reference to Belov’s report. 41 Belov 1938, pp. 92–93, and 129–130, fig. 81; Biernacki 2009, p. 309, pl. 48, 675/973. Here too Biernacki omits all reference to Belov’s report. 39

Kramer 1994, pp. 9–15, pl. 1–5; Zollt 1994, p. 345. There are a number of analogies for this motif in the Pontic area: Barsanti 1989, fig. 100–106. 38 Kautzsch 1936, p. 167, pl. 33, 540 a–b. 36 37

298

Byzantine Capitals of the Architectural Complex of the ‘Basilica of 1935’ in Chersonesos in Crimea

Fig. 14. Corinthian capital 4 (Belov 1938, fig. 138). Fig. 12. Corinthian capital 2 (L. G. Khrushkova).

Capital 4: I will mention one other Corinthian capital, found in 1936 in the masonry of the wall of the building adjacent to the basilica (fig. 14).44 I was unable to identify it in the collection of the Chersonesos Museum. Its height is about 40 cm, and the length of the abacus is approximately 60 cm; it is possible that the lower part of the capital is broken off. It is a large, four-leafed capital; large leaves of soft acanthus are placed at the corners. The leaves meet to form a wideopen V-shaped motif. The abacus is double profiled; the cone in the middle of the abacus is very large and strongly protruding. This is Kautzsch’s Type 5, which, alongside the similar Type 6 (the ‘lyre’ capital), was common between the end of the fifth and the middle of the sixth century. The Chersonesos collection of Corinthian capitals includes 30 or 31 samples,45 of which only four are ‘standard’ whole capitals, and all of them are now in the ‘Basilica of 1935’. The chronology of the basilica complex ‘of 1935’ remains debatable. This, however, applies to the entire group of Early Byzantine basilicas in the Crimea,46 including the best-known one – the Bishop’s (also known as Uvarov’s) Basilica.47 As for the marble, it has had low priority as an object of study since the very beginning of the study of Chersonesos.48 According to the prevailing opinion, the capitals and the other elements of the marble decoration of the complex are from two different structures. The Composite capitals with fine-toothed acanthus and the Impost capital are related to the first basilica, built in the second half of the fifth century. The second basilica, decorated with Corinthian capitals, is dated differently by different authors, either to the Justinianic era or to the end of the sixth century. In my opinion, all the capitals belong to the same three-aisle ‘Basilica of 1935’, which was built in the last quarter of the fifth century. The first church, built on the site of the synagogue, had a single nave, and the capitals could not have been used in it.

Fig. 13. Corinthian capital 3 (L. G. Khrushkova).

formed by the leaves at their contact points a regular order. Under a helix, there is the mark of the stonemason, the letters CA in ligature. Our Corinthian capitals are of Type 7, according to the classification of Rudolf Kautzsch, who divided Corinthian capitals into eight types, arranged chronologically. Type 7 corresponds to the end of the fifth and the first half of the sixth century.42 Such standard (‘normal’, ‘regular’) Corinthian capitals abound in the Pontic regions.43 Our capitals show the evolution of the Corinthian capital by the end of the Late Antique era. In particular, we are talking about the degradation of helices and the decrease in the number of leaves in the lower crown on the kalathos. 42 43

Belov 1938, p. 309, fig. 138. Yakobson 1959, p. 136; Biernacki 2009, p. 34. 46 Khrushkova 2017a, pp. 856–872. 47 Khrushkova 2017b, pp. 27–78. 48 Khrushkova 2017c, pp. 311–341; Khrushkova 2017d, pp. 147–168. 44 45

Kautzsch 1936, pp. 65–67; Sodini 1984, pp. 35–36. Barsanti 1989, pp. 111–125.

299

Liudmila G. Khrushkova vierten bis ins siebente Jahrhundert, Berlin/Leipzig (Studien zur spätantiken Kunstgeschichte, 9).

Due to repeated alterations and repairs, the stratigraphic picture of the architectural complex is not clear. The auditing excavations, conducted by S. F. Strzheletsky in 1949–1950, did not bring clarity, but, on the contrary, caused new discussions. Among the materials obtained by excavations from different years, the capitals are an important chronological reference point. There is no reason to date the group of Corinthian capitals to Justinian I, determining on this basis the time of the capital reconstruction of the first basilica. The ‘Basilica of 1935’ was built no later than the last decades of the fifth century. Its construction was part of the ‘building programme’ of the architectural design of the coastal zone of Chersonesos, a zone that was prominent in the topography of the city. The earliest building here was the episcopal (‘Uvarov’) basilica, built around the middle of the fifth century. The development continued to the west, along the coastline. The construction of the ‘Basilica of 1935’ was an important stage. Finally, the formation of the coastal ensemble ended in the Justinian era with the construction of the large complex of the ‘Western’ Basilica. Its territory incorporated the eastern section of the ‘Western’ cemetery.

Khroushkova L., 2008. ‘Chersonesos in the Crimea: The first Christian buildings (4th–5th Centuries)’, in Antiquité tardive, 16, pp. 141–158. Khrushkova L. G., 2011. ‘Проконнесский мрамор в Херсонесе Таврическом: капители с тонким зубчатым аканфом’, in Византийский Временник – Byzantina Chronika 70 (95), pp. 1740–191. Khrushkova L. G., 2012a. ‘Chersonesus in the Crimea: Early Byzantine capitals with fine-toothed acanthus leaves’, in G. R. Tsetskhladze (ed.), The Black Sea, Paphlagonia, Pontus and Phrygia in Antiquity. Aspects of Archaeology and Ancient History, Oxford (BAR International Series, 2432), pp. 129–140. Khrushkova L. G., 2012b. ‘Geschichte der Christlichen Archäologie in Russland vom 18. bis ins 20. Jahrhundert (3. Folge)’, in RQ, 107/3–4, pp. 202–248. Khrushkova L. G., 2013a. ‘Chersonèse en Crimée aux IVe– VIe siècles: topographie et chronologie’, in O. Brandt, S. Cresci, J. López Quiroga, C. Pappalardo (eds.), Acta XV Congressus internationalis archaeologiae Christianae, Toleti (8–12.9.2008). Episcopus, civitas, territorium, Vatican City (Studi di Antichità cristiana, 65), pp. 387–402.

Establishing the chronology of the basilica complex of ‘of 1935’ requires an examination of all the data: the architectural remains, stratigraphic evidence and archaeological materials, as well as elements such as the marble, mosaic and pictorial decor. We intend to examine the complete construction history of the ‘Basilica of 1935’ in the context of the topography of the Early Byzantine Chersonesos in another study.

Khrushkova L. G., 2013b. ‘Geschichte der Christlichen Archäologie in Russland vom 18. bis ins 20. Jahrhundert (4. Folge)’, in RQ, 108/3–4, pp. 254–287. Khrushkova L. G., 2013c. ‘Quelques chapiteaux protobyzantins inédits du Sud-Ouest de la Crimée’, in C. Blondeau, B. Boissavit-Camus, V. Boucherat, P. Volti (eds.), Ars auro gemmisque prior. Mélanges en hommage à Jean-Pierre Caillet, Zagreb (Studies in Early Christian and Medieval Art History and Archaeology. Dissertationes et monographiae, 6), pp. 85–94.

Bibliography Abbreviation: RQ = Römische Quartalschrift für Altertumskunde und Kirchengeschichte.

Christliche

Barsanti C., 1989. ‘L’esportazione di marmi dal Proconneso nelle regioni pontiche durante il IV–VI secolo’, in Rivista dell’Istituto nazionale di archeologia e storia dell’arte. S. III, 12, pp. 91–220.

Khrushkova L. G., 2016a. ‘Епископская базилика Херсонеса Таврического: методы изучения, результаты, современный взгляд’, in V. V. Majko, T. Ju. Jašaeva (eds.), Владимирский Сборник. Материалы международных научных конференций ‘I и II Свято-Владимирские Чтения’ – St Vladimir’s Readings. Materials of the First and Second International Scholarly Conferences ‘St Vladimir’s Reading’, Kaliningrad, pp. 327–435.

Belov G. D., 1938. Отчет о раскопках в Херсонесе за 1935–36 гг., Simferopol. Biernacki A. B., 2009. Wczesnobizantyjskie elementy i detale architektoniczne Chersonezu Taurydzkiego – The Early-Byzantine Architectural Elements and Details of Chersonesus Taurica, Poznań (Architektura wczesnobizantyjskich budowli sakralnych Chersonezu Taurydzkiego, 2).

Khrushkova L. G., 2016b. ‘Ранневизантийские капители и другие элементы архитектурного декора из ЮгоЗападного Крыма’, in Ученые Записки Крымского Федерального Университета им. В. И. Вернадского. История. Исторические науки, 2 (68)/2, pp. 137– 162.

de Jerphanion G., 1930. La voix des monuments. Notes et études d’archéologie chrétienne, Paris/Brussels. Jastrzȩbowska E., 2011–12. ‘The church atop a synagogue in Chersonesos?’, in Archeologia, 62–63, pp. 61–73.

Khrushkova L. G., 2016c. ‘Византийская архитектура Херсонеса Таврического: история изучения, методы и результаты’, in Вопросы всеобщей истории архитектуры, 6/1, pp. 9–46.

Kautzsch R., 1936. Kapitellstudien. Beiträge zu einer geschichte des spätantiken Kapitells im osten von 300

Byzantine Capitals of the Architectural Complex of the ‘Basilica of 1935’ in Chersonesos in Crimea по истории христианского Херсонеса, I, Saint Petersburg, pp. 139–149.

Khrushkova L. G., 2017a. ‘The study of the Early Byzantine architecture of Chersonesus in the Crimea: Progress or dead end?’, in Hortus Artium Medievalium, 23, pp. 856–872.

Zollt T., 1994. Kapitellplastik Konstantinopels vom 4. bis 6. Jahrhundert n. Chr. : mit einem Beitrag zur Untersuchung des ionischen Kämpferkapitells, Bonn (Asia-Minor-Studien, 14).

Khrushkova L. G., 2017b. ‘The Bishop’s Basilica (“Uvarov’s”) of Chersonesos in the Crimea. The modern view after a century and half of study’, in Archaeologia Bulgarica, 21/2, pp. 27–78.

Zolotarev M. I., Korobkov D. Ju., Ušakov S. V., MacLennan R. S., Overman J. A., Olive J., Edwards D. R., Lindstrom G., 2013. Древняя синагога в Херсонесе Таврическом. Материалы и исследования Причерноморского проекта 1994–1998 гг., I, Moscow (Μυριοβιβλιον. Византия и ее окружение. Серия Исследования, 1).

Khrushkova L. G., 2017c. ‘Византийский мрамор Херсонеса Таврического: начало изучения’, in Материалы по археологии и истории древнего и средневекового Крыма, 9, pp. 311–341. Khrushkova L. G., 2017d. ‘Об одном маленьком юбилее: изучение византийского мрамора в Херсонесе Таврическом’, in Ученые Записки Крымского Федерального Университета им. В. И. Вернадского. История. Исторические науки, 3 (69)/2, pp. 147–168. Khrushkova L. G., 2018. ‘Notes on the study of Byzantine marbles of Tauric Chersonesos’, in S. Pedone, A. Paribeni (eds.), ‘Di Bisanzio dirai ciò che è passato, che passa e che sarà’. Scritti in onore di Alessandra Guiglia, Rome, 2018, pp. 51–66. Klenina E., 2016. ‘The Early-Christian churches architecture on Chersonesos Taurica’, in O. Brandt, G. Castiglia (eds.), Acta XVI Congressus internationalis archaeologiae christianae, Romae (22–28.9.2013). Costantino e i Costantinidi – L’innovazione costantiniana, le sue radici e i suoi sviluppi, Vatican City (Studi di Antichità cristiana, 66), pp. 2255–2280. Kramer J., 1994. Korinthische Pilasterkapitelle in Kleinasien und Konstantinopel. Antike und spätantike Werkstattgruppen, Tübingen (Istanbuler Mitteilungen. Beiheft, 39). Kramer J., 1998. ‘Bemerkungen zu den Methoden der Klassifizierung und Datierung frühchristlicher oströmischer Kapitelle’, in U. Peschlow, S. Möllers (eds.), Spätantike und byzantinische Bauskulptur. Beiträge eines Symposions in Mainz, Februar 1994, Stuttgart (Forschungen zur Kunstgeschichte und christlichen Archäologie, 19), pp. 43–58. Sodini J.-P., 1984. ‘La sculpture architecturale à l’époque paléochrétienne en Illyricum’, in Πρακτικά του 10ου διεθνούς συνεδρίου χριστιανικής αρχαιολογίας, Θεσσαλονίκη 28 Σεπτεμβρίου–4 Οκτωβρίου 1980 – Actes du Xe congrès international d’archéologie chrétienne, Thessalonique, 28 septembre–4 octobre 1980, Vatican City/Thessaloniki (Studi di Antichità cristiana, 37; Ελληνικά. Παράρτημα, 26), pp. 31–117. Yakobson A. L., 1959. Раннесредневековый Херсонес. Очерки истории материальной культуры, Moscow/Leningrad (Материалы и исследования по археологии СССР, 63). Zherebtsov E. N., 2009. ‘Раскопки базилики 1935 г. в Херсонесе’, in S. A. Belyev (ed.), Очерки 301

23 Transennae from Nikopolis Eugenia Chalkia* Βυζαντινό και Χριστιανικό Μουσείο / Byzantine and Christian Museum (Greece) This paper analyses published and unpublished examples of openwork chancel screens (transennae) found in Nikopolis and other minor centres in Epirus. The typological and decorative similarities between some Nikopolis pieces and well-known transennae from Constantinople and Ravenna point to the workshops of the capital as their place of production. On the other hand, the plainer examples from Nikopolis and the other minor centres of the region indicate that local workshops imitated Constantinopolitan models and supplied minor centres. Cet article présente des plaques de chancel ajourées (transennes), publiées et inédites, trouvées à Nicopolis et dans d’autres centres mineurs d’Épire. L’affinité typologique et décorative de certaines pièces de Nicopolis avec des transennes bien connues de Constantinople et de Ravenne nous amène à penser qu’elles ont été produites dans des ateliers de la capitale. Par ailleurs, les exemples les plus simples, provenant à la fois de Nicopolis et des centres mineurs de la région, indiquent probablement que des ateliers locaux imitaient les modèles constantinopolitains et fournissaient les centres mineurs. In diesem Artikel werden sowohl publizierte als auch noch unbekannte Fragmente von Durchbruchschrankenplatten (transennae) besprochen, die aus Nikopolis und Umgebung stammen. Einige von ihnen weisen Typologie- und Dekorationsmerkmale auf, die gewisse Ähnlichkeiten zu den bekannten Durchbruchschrankenplatten von Konstantinopel und Ravenna besitzen. Man kann annehmen, dass sie ebenfalls aus Werkstätten der Hauptstadt stammen. Dennoch finden sich auch einfacher gearbeitete Exemplare unter den Fragmenten von Nikopolis und Umgebung. Das könnte darauf hinweisen, dass aus Konstantinopel importierte Platten in den Werkstätten von Nikopolis nachgeahmt und in regionale Zentren geliefert wurden.  Nell’articolo vengono presentate le cornici di numerose transenne, in parte pubblicate e in parte inedite, rinvenute a Nicopoli e in altri centri minori dell’ Epiro. L’affinità tipologica e decorativa di alcuni frammenti di Nicopoli con le transenne ben note di Costantinopoli e di Ravenna ci inducono a considerare che si tratti di opere prodotte in botteghe della capitale. D’altro canto, i frammenti più semplici, provenienti sia da Nicopoli che da altri centri minori della regione, rivelano che probabilmente i modelli costantinopolitani venivano copiati da botteghe locali, le quali rifornivano anche i piccoli centri.

In my earlier publications on the sculptures of Nikopolis, brief mention was made of parts of openwork screens (transennae) that had been found in Basilicas D and B.1 The reason for my return to this subject was new excavation finds, in Nikopolis itself and in other regions of Epirus, as well as the exhibition of some transennae pieces in the recently established Archaeological Museum of Nikopolis.2 These new data not only permit a typological classification of all transenna frames known to date and the reconstruction

of one of these; they also allow us some observations concerning the dissemination of types in the wider region of Epirus, of which Nikopolis was the artistic centre in the Early Christian period.3 To date, transennae have been found in two of the basilicas at Nikopolis: Basilica B, the city’s oldest and grandest and its cathedral church,4 and the funerary Basilica D, outside the city walls.5 Given that most of the surviving pieces have been found in the latter, we shall begin with a presentation of them.

* I would like to express my acknowledgments to the Byzantine and Christian Museum, Athens, and to the Ephorates of Ioannina, Thesprotia and Preveza for making available the related photos from their Archives. 1 Konstantios, Chalkia 1987, pp. 318–319; Chalkia 2006, pp. 282–283. 2 Riginos, Katerini 2009, p. 85.

For Early Christian Nikopolis, see Chrysos 1981, pp. 22–27 and passim; Pallas 1979, 1987; Bowden 2007; Chalkia 2013. 4 For Basilica B, see Papadopoulou 2015, with relevant bibliography. 5 For Basilica D, see Chalkia 2015, with relevant bibliography. 3

303

Eugenia Chalkia Transennae from Basilica D

lower one is decorated with a Lesbian cymation which forms lanceolate leaves with a deep grooved rib in their centre. The tips of similar leaves appear in the intervening spaces. A bead and reel (astragal) intervenes between this final zone and the openwork interior decoration. Though not high, the relief is distinguished by its plasticity, which is achieved by chiaroscuro created with the use of a drill, as well as by precise execution. Another five similar pieces from frames (inv. no 24, 107, 322, 727 and 2612) found during both the first and second excavation periods10 led to the reconstruction of the frame as it is displayed today in the Archaeological Museum of Nikopolis (fig. 1).11 Our recent investigation in the Museum’s storerooms allowed the identification of another three pieces (inv. no 123, 281 and 1780)12 and a more complete artist’s reconstruction of the

A considerable number of transenna fragments came to light during Anastasios Orlandos’ excavations of the basilica in the late 1950s to early 1960s,6 and during more recent ones that I conducted in the first years of the present century.7 The most important example of openwork screens from Basilica D is the greater part of the corner of a frame (inv. no 11) found during the first period of excavations,8 which preserves a few traces of the inner openwork decoration of the slab.9 Made of white marble, it consists of two parallel decorated zones. Between them is a thin, flat band; a similar, slightly wider band defines the outer edge of the frame. It is 0.68 m long and 0.335 m wide, with a thickness of 0.065–0.070 m. Its upper zone is decorated by a relief rinceau, while the

Fig. 1. Transenna from Basilica D, Nikopolis, in the Archaeological Museum of Nikopolis (Archive of the Ephorate of Antiquities, Preveza). 10 The pieces with inv. no 24, 107, 727 and 2612 had been found during the first period of excavations (Orlandos 1956a, p. 152, pl. 55e; Konstantios, Chalkia 1987, p. 318, fig. 2), while the piece with inv. no 322 was found in 1999, in the north aisle of the transept. 11 See note 2. 12 The pieces with inv. no 123 and 1780 had been found during the first period of excavations (Orlandos 1956a, p. 152, pl. 55e). Fragment no 281 was found in 1999 in the north aisle of the transept.

For the excavation of A. Orlandos, see Orlandos 1956a, 1956b, 1959a, 1959b, 1961a, 1961b, 1961–62. 7 Chalkia 2007. 8 Orlandos 1961a, p. 101, pl. 59a, 1961b, pp. 111–113, fig. 110, Konstantios, Chalkia 1987, p. 318, fig. 1, Papadopoulou 2012, pp. 458– 460, no 95. 9 See fig. 1–2. 6

304

Transennae from Nikopolis

Fig. 2. Artist’s reconstruction of the transenna from Basilica D, Nikopolis (E. Apostolopoulou).

frame (fig. 2). There was also an attempt at a hypothetical completion of the interior openwork decoration on the basis of related fragments that have occasionally been found, the largest of which form crosswise-disposed palmettes or branches (fig. 3).13 Our transenna when completed has a height of 0.98 m and a length of 1.16 m.14 As regards the frame type, its structure and decoration are common to quite a few transennae, chiefly in Ravenna15 (fig. 4) and in Constantinople (fig. 5).16 The same characteristics also appear in some small pieces of transenna frames attributed to workshops in the capital, like the piece found in the

Basilica of the Heraion on Samos17 and that reused in the masonry of the Church of St Catherine in Thessaloniki.18 In both cases, the relief rinceau of the upper zone is high, and executed by the use of a drill. Transenna frames of similar technique and decoration have also been found in the Episcopal Basilica at Stobi (fig. 6).19 This important monument has provided us a significant number of transenna frames with varied and original decorative compositions20 which, as has been observed, reveal, despite their provincial style, a connection with and knowledge of trends prevailing in workshops in the capital.21 At Nikopolis, this iconographic type of decoration is represented only by this fragmentary transenna. There are two other, similar transenna frames, however, in the greater Epirus region. The first, which comes from a large

For the artist’s reconstruction of the transenna Ι would like to thank my friend, the sculptor Eleni Apostolopoulou. The transenna from Sant’Apollinare Nuovo in Ravenna, on which similar interior openwork ornaments are found (see Angiolini Martinelli 1968, p. 75, no 131), was taken as our model. Comparable ornaments are also found in two transennae of Hagia Sophia in Constantinople: Barsanti 2004, pp. 499– 502, no TL01, TL02. 14 The Ravenna transenna no 131 has approximately the same dimensions. 15 Cf. the transenna from San Vitale, in the National Museum of Ravenna: Angiolini Martinelli 1968, p. 73, no 124. 16 Cf. the transenna in Hagia Sophia and that in the Church of SS Sergius and Bacchus, the latter reused in the mihrab of Küçük Ayasofya Camii: Barsanti 2004, pp. 501–502, no TL02, p. 514, fig. 287, 288. 13

Westphallen 1994, pp. 322–324, fig. 3, pl. 79,4. Barsanti 2004, p. 522, fig. 298. 19 I would like to express my acknowledgments to my colleague Ljubinka Džidrova, to whom I owe the photo. 20 Egger 1929, p. 69, pls. 68–71; Hoddinot 1963, pl. 40a. 21 Kitzinger 1946, 107, Barsanti 2004, p. 525. 17 18

305

Eugenia Chalkia

Fig. 3. Inner decoration of the transenna from Basilica D, Nikopolis (Archive of the Ephorate of Antiquities, Preveza).

transenna from Basilica D, that is, an outer zone with a relief rinceau and cross, and an inner zone with a Lesbian cymation, although without the pointed tips between its leaves. The astragal (bead and reel) is also absent, and the relief is fairly ‘loose’ and far more carelessly executed

three-aisled basilica with beautiful mosaic pavements, was discovered fairly recently at the site of Opagia, Doliana, in the prefecture of Ioannina; it is currently in the Byzantine Museum of Ioannina (inv. no 765).22 It displays the same structure and decoration as the frame of the

Fig. 4. Transenna in the Archaeological Museum of Ravenna, inv. no 124 (P. Angiolini Martinelli). 22 Papadopoulou 2012, pp. 463–464, no 98, fig. 98. For the Basilica, see Karamperidi 2004, 2007.

306

Transennae from Nikopolis

Fig. 5. Transenna in Hagia Sophia, Constantinople (C. Barsanti).

Fig. 6. Frame of transenna from the Episcopal Basilica in Stobi (L. Džidrova).

(fig. 7). The second frame fragment, found during the excavation of a three-aisled Basilica at Krystallopigi, Thesprotia23 (fig. 8) and now in the Archaeological Museum of Igoumenitsa (inv. no 120), is identical to that from Basilica D. The details of the rendering of the Lesbian cymation with grooved rib, pointed tips between leaves, astragal and cross in the middle of the upper side of the slab do not differ from the frame at Nikopolis. There is also a similarity in the inner decoration of the two transennae.24 They differ noticeably, however, as regards the rendering of decorative motifs, both those of the frame and of the inner zone’s decoration. Fig. 7. Frame of transenna from the Basilica of Opagia, Doliana (inv. no 765) (Archive of the Ephorate of Antiquities, Ioannina).

Vasilikou 2009, pp. 202–203, fig. 5–6 For the inner openwork decoration of the Nikopolis screen, see fig 3. For that of the Krystallopigi screen, see Vasilikou 2009, fig. 6.

23 24

307

Eugenia Chalkia

Fig. 8. Frame of transenna from the Basilica of Kristallopigi (inv. no 120) (Archive of the Ephorate of Antiquities, Thesprotia).

In contrast with the unique example of the first type at Nikopolis, there are more examples of another type of frame, which appear to belong to four different transennae of white marble, also found during the first25 and second periods of excavations. Although they present the same structure as the frame of the transenna described above, they differ in the decorative theme of their outer zones, which display slight variations. In the first variant, to which three transenna frames belong (inv. no 108, 109, 338 and 334),26 the outer zone is decorated by alternating upright and inverted trefoils. Their leaves, which are of unequal length, cluster at the point where the two longer ones converge, and a spherical knob is formed (fig. 9–10). Small deviations from the first transenna are also observable on the Lesbian cymation of the inner zone. The lanceolate leaves carry a relief rib in their centre and are fairly far apart. In the second variant, to which one frame fragment (inv. no 255)27 belongs, the trefoils are clumsier. The spherical knob is missing at the point where the leaves join, and there are small drilled holes both in the depth and the surface of the leaves (fig. 11). On all the examples of frames of these two variants, the relief is high with pronounced chiaroscuro, but the execution is more careless than that seen in the transenna of the first type. I was unable to locate any frames of openwork transennae of precisely similar form. In iconographic terms, the closest example with comparable decoration is that found in the Basilica at Stobi (fig. 12).28 It consists of three parallel decorative zones, the central one of which is decorated with trefoils arranged approximately like those found on the frames of transennae from Basilica D at Nikopolis.

Fig. 9. Frame of transenna from Basilica D, Nikopolis (inv. no 109) (Archive of the Ephorate of Antiquities, Preveza).

Finally, part of a rather carelessly executed frame of a third, simpler type has been found in Basilica D (inv. no 61). 29 It consists of a wide, undecorated band and zone of Lesbian cymation of the same form as that of the frames of the second type (fig. 13). Notably, frames of exactly this type have been found at other places in Epirus. Three such fragments (probably from the same screen) have been found at Ladochori, Thesprotia, the site of an important Roman settlement whose life continued into Early Christian times.30 All are in the Archaeological Museum of Igoumenitsa (inv. no 55, 77 and 95).31 Here too we observe the wide, flat undecorated zone followed by a Lesbian cymation (fig. 14) of exactly the same type as that in the second group from Basilica D. Another two pieces from frames of a similar type are in the Byzantine Museum of Ioannina. The first, inv. no 132932 (fig. 15), is from the Basilica of Opagia in Doliana, mentioned above, while

Konstantios, Chalkia 1987, p. 318, fig. 3. The two last fragments (no 338 and 344) belong to the same transenna. The dimensions of the fragments are: no 108: L.: 0.17 m, W.: 0.21 m, Th.: 0.075 m, no 109: L.: 0.11 m, W.: 0.21 m, Th.: 0.075 m, no 338: L.: 0.11 m, W.: 0.13 m, Th.: 0.081 m, no 344: L.: 0.063 m, W.: 0.211 m, Th.: 0.08 m. 27 L.: 0.20 m, W.: 0.24 m, Th.: 0.075 m. 28 Egger 1929, p. 70, fig. 72. 25

29 The dimensions of the fragment are: L.: 0.15 m, W.: 0.21 m, Th.: 0.07 m. 30 Papadopoulou 2012, pp. 91–98, with relevant bibliography; Papadopoulou 2016. 31 Papadopoulou 2012, pl. 67 c. 32 Papadopoulou 2012, pp. 463–464, no 98.

26

308

Transennae from Nikopolis

Fig. 10. Frame of transenna from Basilica D, Nikopolis (inv. no 338 and 344) (Archive of the Ephorate of Antiquities, Preveza).

Fig. 12. Frame of transenna from the Episcopal Basilica at Stobi (R. Egger).

the second, inv. no 586, is of unknown provenance.33 The latter, in two joining pieces, preserves outside the frame a considerable part of its openwork fish-scale decoration (fig. 16), a common motif in less-embellished examples of the type.34

Papadopoulou 2012, pp. 462–463, no 97. See for example the openwork screen from the Basilica of Topkapı Sarayı in Constantinople (Tezcan 1989, p. 66, fig. 66) and those from the Basilica of Karkour in Negev (Figueras 1998, pp. 270–271, fig. 5–6), the Basilica of St Sophia in Mytica (Vocotopoulos 1984, p. 129, pl. 108a), and that in the Museum of Thasos (Orlandos 1951, pp. 66–67, fig .6). 33 34

Fig. 11. Frame of transenna from Basilica D, Nikopolis (inv. no 255) (Archive of the Ephorate of Antiquities, Preveza).

309

Eugenia Chalkia

Fig. 16. Transenna in the Byzantine Museum of Ioannina (inv. no 586) (Archive of the Ephorate of Antiquities, Ioannina).

Transennae from Basilica B The transennae from Basilica B were found during excavations by Georgios Sotiriou and Anastasios Orlandos in the 1920s and 1930s.35 They were more numerous than those in Basilica D, and displayed a greater variety. Unfortunately, only a few examples survive today. The others had probably been transferred to the museum that Alexandros Philadelpheus founded in the mosque at Preveza,36 which was destroyed in a World War II bombing raid.37 They are known, however, from photographs in the Sotiriou Archive, which is now in the Byzantine and Christian Museum, Athens. From these transennae too, apart from a single instance, only the frames are preserved, which may be classified into three types. The structure of the first type, sub-divided into two variations, follows the same disposition as those from Basilica D: two parallel decorated zones separated by a narrow band and defined on the upper side by a corresponding but wider band. The inner zone is also adorned with a Lesbian cymation, while the outer one presents small differences. The fragments preserved today (inv. no ΑΛ 3122a–c), published by Barbara Papadopoulou,38 plus another two joining fragments (inv. no 2624 and 1499)39 (fig. 17) belong to the first variant, whose outer zone is decorated by a rinceau with trefoils in the spaces between the curves. In the second variant, to which belong the fragments of frames no longer surviving, a vine tendril forms the decoration of the corresponding zone, with grape clusters, leaves, spirals and crosses in the gaps (fig. 18). In both groups, the Lesbian cymation carries relief ribs in the centre of its leaves, between which pointed

Fig. 13. Frame of transenna from Basilica D, Nikopolis (inv. no 61) (Archive of the Ephorate of Antiquities, Preveza).

Fig. 14. Frame of transenna (inv. no 95) from the Basilica of Ladochori (Archive of the Ephorate of Antiquities, Thesprotia).

35 For the excavations of G. Sotiriou and A. Orlandos in Basilica B, see Sotiriou, Orlandos 1929, 1930; Sotiriou 1938. 36 Philadelpheus 1922. 37 Konstantaki 2013, pp. 363–364 38 Papadopoulou, Konstantaki 2007, p. 648, fig. 26, Papadopoulou 2012, p. 461, no 96, fig. 96a–b. 39 Dimensions of no 2624: L.: 0.26 m, W.: 0.28 m, Th.: 0.033 m; of no 1499: L.: 0.135 m, W.: 0.28 m, Th.: 0.033 m.

Fig. 15. Frame of transenna from the Basilica of Opagia, Doliana (inv. no 1329) (Archive of the Ephorate of Antiquities, Ioannina).

310

Transennae from Nikopolis

Fig. 17. Frame of transenna from Basilica B, Nikopolis (inv. no 2624, 1499) (Archive of the Ephorate of Antiquities, Preveza).

Fig. 18. Middle and upper zone: frame of transenna from Basilica B, Nikopolis (Archive of the Byzantine and Christian Museum, Athens).

The next two types of transenna frames are of simpler construction, coarser execution, and greater thickness, at least as appears from photographs. The first includes only a piece of a frame formed by a wide, molded band, a zone with Lesbian cymation and an astragal (fig. 20), while the second belongs to an entirely different type of transenna which also preserves part of its inner openwork decoration (fig. 21). The frame, which is entirely different from the preceding ones, is decorated by a zone containing a rinceau and another, narrower zone, with a string course motif. A

leaf-tips emerge, as in the case of the first transenna from Basilica D. The bead and reel (astragal) is missing from the frames of the latter group. The morphological, iconographic, and stylistic relationship displayed by the frames of both variants to the transennae in Hagia Sophia in Constantinople, specifically those in the sultan’s gallery (fig. 19), is evident.40 For the frames of the first variant, see Barsanti 2004, pp. 501–502, no TL.02; for those of the second, ibid., pp. 499–500, no TL.01. 40

311

Eugenia Chalkia

Fig. 19. Transenna in Hagia Sophia, Constantinople (C. Barsanti).

Fig. 20. Lower left corner: frame of transenna from Basilica B, Nikopolis (Archive of the Byzantine and Christian Museum, Athens).

312

Transennae from Nikopolis

Fig. 21. Frame of transenna from Basilica B, Nikopolis (Archive of the Byzantine and Christian Museum, Athens).

similar motif also surrounds the interlocking circles of the inner decoration, which recalls that on the transenna from the Kalenderhane Camii in Constantinople.41

the Topkapı Sarayı Basilica,45 may enclose exceptionally refined openwork inner decoration. In any case, the frames of the transennae at Nikopolis in the first category belong to two related types. The first type, to which the frames of most of the transennae from Basilica B and the first frame from Basilica D belong,46 displays a close morphological, iconographic and stylistic affinity with the transennae of Constantinople and Ravenna. The second, which differs from the first only with respect to the decoration of its outer zone, includes the second group of frames from Basilica D,47 for which no iconographic parallels have been identified. The second category includes only one type of the simpler frames with Lesbian cymation, like that from Basilica D, in addition to more complex ones, like that from Basilica B,48 while the last transenna with interlocking circles forms a special type.49

Conclusions All the examples presented here indicate that there were a significant number of transennae and a variety of types in Nikopolis. None has been preserved intact so that we can know their inner decoration, but we can surmise the style of some of them from surviving fragments. Moreover, as Claudia Barsanti has observed, to date no transennae displaying exactly the same inner decoration have been identified.42 We can, however, classify the frames of transennae generally into two categories. The first includes the frames with composite decoration that enclosed elaborate openwork inner decoration, as attested by surviving fragments. The second category includes frames of simpler form and coarser construction. In the latter case, there may have been comparable corresponding decoration, as happens in most examples of openwork transennae with simple frames.43 This was not a rule, however. Frames of simple type, for example that of the transenna from San Vitale in the Archaeological Museum of Ravenna44 and that from part of a transenna in

We thus conclude that some of the transennae from both basilicas, at least those which display a close resemblance to examples from Constantinople and Ravenna, were imported, as were many other sculptures at Nikopolis.50 But what about the others, as for example those of the second type in the first category, for which no parallels Tezcan 1989, p. 66, fig. 67. See fig. 1, 2, 17, 18. 47 See fig. 9, 10, 11. 48 See fig. 13, 20. 49 See fig. 21. 50 Chalkia 2006; Papadopoulou, Konstantakii 2007.

Peschlow 1997, pp. 107, 110, pls. 113–115; Barsanti 2004, p. 515, fig. 289. 42 Barsanti 2004, p. 513. 43 There are many examples. See, indicatively, no 286 in the Museum of Ioannina (fig. 16) and those mentioned in note 34 above. 44 Angiolini Martinelli 1968, p. 74, no 128.

45

41

46

313

Eugenia Chalkia have been found in Constantinople51 or Ravenna?52 Their relation to the sculptural repertoire of the capital results indirectly from the iconographic affinity they display with the transennae of the Episcopal Basilica at Stobi.53 On the other hand, we also observed in the last transenna from Basilica B54 a direct iconographic relation to sculpture in the capital. Probably these, along with those of the simpler type in the second category, with frames decorated solely by a Lesbian cymation, were manufactured in local workshops, in accordance with the models provided by workshops in the capital.

Bowden W., 2007, ‘Nicopolis – The ideology of the Late Antique city’, in Nicopolis II, pp. 135–149. Chalkia E., 2006, ‘La scultura paleocristiana di Nicopoli. Alcune considerazioni’, in R. Harreither, Ph. Pergola, R. Pillinger, A. Pülz (eds.), Acta congressus internationalis XIV archaeologiae christianae, Vindobonae 19.–26. 9. 1999 – Akten des XIV. internationalen Kongresses für christliche Archaölogie, Wien 19.–26. 9. 1999. Frühes Christentum zwischen Rom und Konstantinopel, Vatican City/Vienna (Studi di Antichità cristiana, 62; Österreichische Akademie der Wissenschaften. Philosophish-historische Klasse, Archäologische Forschungen, 14), pp. 281–287.

Although the Nikopolis sculptures have not been systematically studied, the existence of local workshops in this important Early Christian centre in Epirus has been confirmed,55 and one may reasonably suppose that they also furnished smaller, largely inland centres in the region, which did not enjoy easy communication with Constantinople. We may also attribute the transennae from Doliana, Krystallopigi and Ladochori56 to these workshops, which imitate both the transennae of the first category at Nikopolis and those of the second, although without attaining their level of artistic quality.

Chalkia E., 2007, ‘Συμπληρωματική ανασκαφή στη βασιλική Δ της Νικόπολης’, in Nicopolis II, pp. 659– 666. Chalkia E., 2013, ‘Nikopolis’, in J. Albani, E. Chalkia (eds.), Heaven and Earth. Cities and Countryside in Byzantine Greece, Athens, pp. 140–155. Chalkia E., 2015, Η βασιλική Δ της Νικόπολης – Basilica D of Nicopolis (Monuments of Nicopolis, 9), Athens. Chrysos E., 1981, ‘Συμβολή στὴν ἱστορία τῆς Ἠπείρου κατὰ τὴν Πρωτοβυζαντινὴ περίοδο’, in Epeirotika Chronika, 23, pp. 9–111.

Bibliography Abbreviations:

Egger R., 1929, ‘Die stätliche Kirche von Stobi’, in Jahreshefte des Österreichischen Archäologischen Institutes in Wien, 24, p. 42–87.

Ergon = Τὸ Ἒργον τῆς ἐν Ἀθήναις Ἀρχαιολογικῆς Ἑταιρείας. Nicopolis I = Chrysos, E. (ed.), 1987, Νικόπολις Πρακτικά του πρώτου Διεθνούς Συμποσίου για Νικόπολη, 23–29 Σεπτεμβρίου 1984 – Nicopolis Proceedings of the First International Symposium Nicopolis, 23–29 September 1984, Preveza.

1. τη 1. on

Farioli R., 1983, ‘Ravenna, Costantinopoli. Considerazioni sulla scultura del VI secolo’, in XXX Corso di cultura sull’arte ravennate e bizantina. Seminario Giustinianeo, Ravenna, 6–14 Marzo 1983, Ravenna, pp. 205–253.

Nicopolis II = Zachos, K. L. (ed.), 2007, Νικόπολις Β’. Πρακτικά του Δευτέρου Διεθνούς Συμποσίου για τη Νικόπολη (11–15 Σεπτεμβρίου 2002) –Nicopolis B. Proceedings of the Second International Symposium (11–15 September 2002), Preveza.

Figueras P., 1998, ‘L’église byzantine de Karkour dans le Néguev’, in N. Cambi, E. Marin (eds.), Radovi XIII međunarodnog kongresa za starokršćansku arheoligiju, Split-Poreč, 25.9.–1.10.1994 – Acta XIII congressus internationalis archaelogiae christianae, Split-Poreč, 25.9.–1.10.1994, III, Vatican City/Split, 1998 (Studi di Antichità cristiana, 54/2; Vjesnik za arheologiju i historiju dalmatinsku. Suplement, 88), pp. 261–272.

ΠΑΕ = Πρακτικὰ τῆς ἐν Ἀθήναις Ἀρχαιολογικῆς Ἑταιρείας. Angiolini Martinelli P., 1968, ‘Corpus’ della scultura paleocristiana, bizantina ed altomedievale di Ravenna. I, Rome.

Hoddinott R. H., 1963, Early Byzantine Churches in Macedonia and Southern Serbia. A Study of the Origins and the Initial Development of East Christian Art, London/New York.

Barsanti C., 2004, ‘Le transenne’, in A. Guiglia Guidobaldi, C. Barsanti (eds.), Santa Sofia di Costantinopoli. L’arredo marmoreo della Grande Chiesa giustinianea, Vatican City (Studi di Antichità cristiana, 60), pp. 489–529.

Karamperidi A., 2004, ‘Παλαιοχριστιανική βασιλική στον Άγιο Γεώργιο Δολιανών’, in K. Konstantinidis (ed.), Διαλέξεις, ΙΙ, Ioannina, pp. 33–47. Karamperidi A., 2007, ‘Tα ψηφιδωτά δάπεδα της βασιλικής των Δολιανών και η σχέση τους με τα ψηφιδωτά των παλαιοχριστιανικών βασιλικών της Νικόπολης’, in Nicopolis II, pp. 667–680.

For the transennae found in Constantinopole, cf. Barsanti 2004, pp. 494–517. 52 It is well known that its (Ravenna’s) sculptures form a basic tool for the fuller study of sculpture in the capital during the sixth century. On this, cf. Farioli 1983. 53 See above, p. 4, fig. 12 54 See fig. 21 55 See note 50. 56 See fig. 7, 8, 14, 15. 51

Kitzinger E., 1946, ‘A survey in the Early Christian town of Stobi’, in Dumbarton Oaks Papers, 3, pp. 81–162.

314

Transennae from Nikopolis Sotiriou G., 1938, ‘Ἀνασκαφαὶ Νικοπόλεως’, in ΠΑΕ, 93, pp. 112–117.

Konstantaki A., 2013, ‘Ο αρχαιολογικός χώρος της Νικόπολης μετά την απελευθέρωση της Πρέβεζας’, in Prevezanika Chronika, 49–50, pp. 339–382.

Sotiriou G., Orlandos A., 1929, ‘Ἀνασκαφαὶ Νικοπόλεως’, in ΠΑΕ, 84, pp. 83–86

Konstantios D., Chalkia E., 1987, ‘Παλαιοχριστιανικά γλυπτά Νικοπόλεως’, in Nicopolis I, pp. 317–325.

Sotiriou G., Orlandos A., 1930, ‘Ἀνασκαφαὶ Νικοπόλεως’, in ΠΑΕ, 85, pp. 79–80.

Orlandos A. K., 1951, ‘Παλαιοχριστιανικὰ γλυπτὰ τοῦ Λιμένος Θάσου’, in Ἀρχεῖον Βυζαντινῶν Μνημείων τῆς Ἑλλάδος, 7, pp. 62–71.

Tezcan H., 1989, Topkapı Sarayı ve çevresinin Bizans devri arkeolojisi, Istanbul.

Orlandos A. K., 1956a, ‘Ἀνασκαφή Νικοπόλεως’, in PΑΕ, pp. 149–153.

Vasilikou N., 2009, ‘Παλαιοχριστιανική βασιλική στην Κρυσταλλοπηγή Παραμυθιάς’, in Thesprotia Expedition, I, Helsinki (Papers and Monographs of the Finnish Institute at Athens, 15), pp. 197–206.

Orlandos A. K., 1956b, ‘Νικόπολις’, in Ergon, pp. 60–63. Orlandos A. K., 1959a, ‘Ἀνασκαφή βασιλικῆς Δ Νικοπόλεως’, in PΑΕ, pp. 90–97.

Vocotopoulos P., 1984, ‘Ἀνασκαφὴ βασιλικῆς Ἁγίας Σοφίας Μύτικα’, in ΠΑΕ, 139, pp. 129–130.

Orlandos A. K., 1959b, ‘Νικόπολις’, in Ergon, pp. 67–72.

Westphallen S., 1994, ‘Die frühchristliche Basilika im Heraion von Samos und ihre Ausstattung’, in Athenische Mitteilungen, 109, p. 301–335.

Orlandos A. K., 1961a, ‘Ἀνασκαφὴ Νικοπόλεως’, in PΑΕ, pp. 98–107. Orlandos A. K., 1961b, ‘Νικόπολις’, in Ergon, pp. 107– 113. Orlandos A. K., 1961–62, ‘Ἀνασκαφὴ Νικοπόλεως’, in Ἀρχαιολογικὸν Δελτίον, 17, pp. 199–201. Pallas D., 1979, ‘Corinthe et Nikopolis pendant le Bas Moyen Âge’, in Felix Ravenna, 118, pp. 93–142. Pallas D., 1987, ‘Οι χαρακτήρες και η ακτινοβολία της εκκλησιαστικής αρχιτεκτονικής της Νικόπολης’, in Nicopolis I, pp. 225–239. Papadopoulou B., 2012, Παλαιοχριστιανική Ήπειρος (4ος– 7ος αιώνας). Η μαρτυρία της γλυπτικής, PhD dissertation, Πανεπιστήμιο Ιωαννίνων. Papadopoulou B., 2015, Η βασιλική Β της Νικόπολης – Basilica B of Nicopolis (Monuments of Nicopolis, 8), Athens. Papadopoulou B., 2016, ‘Ladochori (Sybota?). An Early Christian settlement in the inner part of Igoumenitsa gulf’, in Abstracts of 1st International Conference on the Archaeology & History of Thesprotia (Igoumenitsa, December 8th–11th 2016), Ioannina, pp. 36–37. Papadopoulou B., Konstantaki A., 2007, ‘Ανάγλυφα αρχιτεκτονικά μέλη βασιλικής Β (Αλκίσωνος). Παρατηρήσεις και συμπεράσματα’, in Nicopolis II, pp. 637–657. Peschlow U., 1997, ‘Architectural sculpture in Kalenderhane in Istanbul. The buildings, their history and decoration’, in C. L. Striker, Y. D. Kuban (eds.), Kalenderhane in Istanbul Final Reports on the Archaeological Exploration and Restoration at Kalenderhane Camii, 1966–1978, I, Mainz, pp. 101– 111. Philadelpheus A., 1922, ‘Μουσεῖον Νικοπόλεως’, in Ἀρχαιολογικὴ Ἐφημερὶς, 61, pp. 66–79. Riginos G., Katerini E. (ed.), 2009, Νικόπολη. Οδηγός Αρχαιολογικού Μουσείου, Athens. 315

24 Il rivestimento marmoreo del fonte del battistero di Nocera Superiore: nuove considerazioni Alessandra Avagliano Sapienza Università di Roma (Italy) Il rivestimento marmoreo del fonte del battistero di Nocera Superiore è composto da nove lastre realizzate mediante la ben nota tecnica dello champlevé. Tali lastre costituiscono un unicum non solo in Campania, ma anche in tutta la penisola italiana. Questa peculiarità non è stata finora sottolineata, in quanto i confronti che sono stati suggeriti riguardavano unicamente i motivi iconografici, ma non la tecnica di esecuzione. È dunque necessario inquadrare le lastre di Nocera nell’ambito della ricca produzione di scultura a champlevé d’area bizantina, perché esse, oltre a fornire una testimonianza della diffusione di tale tecnica anche nell’Italia meridionale, potrebbero offrire utili indizi per la datazione del battistero stesso. Le revêtement en marbre de la cuve baptismale de Nocera Superiore est constitué de neuf plaques réalisées dans la technique bien connue du champlevé. Ces plaques sont un unicum non seulement en Campanie, mais aussi dans toute l’Italie. Cette particularité n’a pas été soulignée jusqu’à présent, parce que les parallèles suggérées ne concernaient que les motifs iconographiques, mais pas la technique d’exécution. Inclure les plaques de Nocera dans le contexte de la production byzantine de champlevés offre un témoin de la diffusion de cette technique en Italie méridionale et pourrait livrer des donnés utiles pour la datation du baptistère lui-même. The marble revetment of the Baptistery font in Nocera Superiore is composed of nine slabs using the well-known champlevé technique. These slabs are unique not only in Campania but also in the whole Italian peninsula. This peculiarity has not been highlighted, because the suggested analogues provide parallels only for the iconographic motifs and not the technique of execution. It is important to analyse the slabs of Nocera in the context of the rich production of champlevé sculpture in the Byzantine area, because they attest to the diffusion of this technique in Southern Italy and thus can help to date the Baptistery itself. Die Marmorverkleidung des Taufbeckens von Nocera Superiore besteht aus neun Marmorplatten mit Champlevé-Dekor. Die Platten sind nicht nur für Kampanien einzigartig, sondern im gesamten Italien. Ihre Besonderheit wurde bisher nicht erkannt, weil die Vergleichsanalysen allein die Ikonographie beachteten, nicht aber die Herstellungstechnik. Die Platten von Nocera müssen in den Kontext der reichen Champlevé-Relief-Produktion im byzantinischen Bereich eingeordnet werden. Die Platten sind Zeugnisse für die Verbreitung dieser Technik in Süditalien, und ihre byzantinischen Parallelen liefern Hinweise für die Datierung des Baptisteriums von Nocera Superiore.

Il battistero di Santa Maria Maggiore a Nocera Superiore è indubbiamente uno dei monumenti più interessanti della Campania, che appare ancora di dibattuta cronologia. In un primo tempo, la sua costruzione era stata collocata nel corso del IV secolo1, poi, a seguito delle ricerche di Michael

Stettler2, la maggior parte degli studiosi si è orientata per una datazione nel periodo immediatamente successivo alla conquista bizantina della città nel 5553, finché, di recente, è stata avanzata la proposta di anticiparla al 498, ponendola Stettler 1940, pp. 134-142 (trad. it. pp. 21-59). Testini et al. 1989, p. 98; Scognamillo 1994, p. 121; Cirillo, 1999, p. 183; Fortunato, Santangelo 2006, p. 88; Brandt 2006-07, p. 191; 2012, pp. 183-185; 2016, pp. 278-279. Di recente, Pasquale Natella ha suggerito di retrodatare la fondazione del battistero a prima del 552-553. Natella 2010, pp. 17-21.

2

Una datazione nel corso del IV secolo era stata proposta da Bilotti (1924, p. 161), poi ripresa anche da D’Alessio (1935a, p. 204). Una datazione al IV-V secolo era, invece, stata proposta da De Angelis (De Angelis, 1926, in part. p. 41) e successivamente da Gino Chierici. Chierici 1934, p. 214; D’Alessio 1935b, p. 227. 1

3

317

Alessandra Avagliano

Fig. 1. Nocera Superiore, Battistero di Santa Maria Maggiore, planimetria (Santangelo 2014).

battesimale non risiede unicamente nelle sue notevoli dimensioni, ma è accentuata dal rivestimento costituito da lastre marmoree appositamente realizzate per essa, delle quali solo nove sopravvivono. Il primo ad aver posto l’attenzione sulle lastre è stato Stettler, il quale, nel suo fondamentale studio sul battistero7, ne aveva analizzato il decoro proponendo dei confronti con i motivi decorativi che ornano una serie di manufatti databili tra il V e il VI secolo, soffermandosi, in particolare, sulla ricorrenza delle croci e delle losanghe8. Gli studiosi, in seguito, si sono concentrati soprattutto sugli aspetti architettonici, storici ed archeologici del monumento, mentre per quel che riguarda le lastre marmoree sono state sostanzialmente riprese le riflessioni dello studioso svizzero9. È, dunque, necessario soffermarsi su tali manufatti individuandone

in relazione con l’episcopato di Lorenzo, antagonista di Simmaco per il soglio pontificio4. Il battistero, di dimensioni notevoli, ha pianta circolare e copertura cupolata sorretta da quindici coppie di colonne (fig. 1), le quali, come del resto tutti gli altri elementi del decoro architettonico dell’edificio, sono spoglie probabilmente provenienti dall’antica Nuceria Alfaterna5. La parte centrale è occupata dalla monumentale vasca battesimale (fig. 2), di forma circolare all’interno e ottagonale con i vertici smussati all’esterno, la quale, con il suo diametro di 7 metri è seconda per grandezza soltanto al fonte battesimale di San Giovanni in Laterano a Roma. Sulla vasca sono collocate 5 delle 8 originarie colonne, di cui solo due sono ancora provviste di capitelli, mentre rimangono tutte le otto basi. Intorno ad essa si sviluppa un deambulatorio voltato a botte, delimitato da un colonnato. La piscina non ha alcuna apertura né scale per l’ingresso, né tantomeno strutture per il sistema idrico, se si eccettua un foro per lo scolo dell’acqua, che doveva essere versata dall’alto6. L’eccezionalità della vasca

L’importanza di tale studio viene ribadita anche da Brandt, il quale attribuisce giustamente a Stettler il merito di aver gettato le basi per ogni tipo di discussione moderna. Cf. Brandt 2006-07, pp. 189-193. 8 Stettler propone numerosi confronti per i motivi decorativi: innanzi tutto con le lastre marmoree delle chiese romane di San Clemente, Santa Cecilia in Trastevere e Santa Maria in Cosmedin (senza ulteriori precisazioni) poi con quelle delle chiese di Santa Sofia di Costantinopoli e della Koimesis di Nicea per i motivi a losanghe, per i quali ricorda anche le tarsie della Basilica Eufrasiana di Parenzo. Per la particolare tipologia delle croci con terminazione a goccia ricorda i plinti della galleria occidentale di San Lorenzo fuori le mura a Roma o le croci a mosaico del nartece della Santa Sofia costantinopolitana. Cf. Stettler 1940, pp. 130-134 (trad. it. pp. 51-53). 9 Brandt 2012, pp. 183-185; Santangelo 2014, p. 31. 7

4 Santangelo 2014, p. 31. Per la figura di Simmaco si veda Guiglia 2016, pp. 109-116, per un approfondimento sullo scisma si veda soprattutto p. 110, n. 4, con bibliografia precedente. 5 Mariani 1987, pp. 7-58; Pensabene 2005, pp. 69-143. 6 Stettler 1940, pp. 114-125 (trad. it. pp. 40-49); Scognamillo 1994, p. 121; Brandt 2012, pp. 183-185; Santangelo 2014, pp. 26-27.

318

Il rivestimento marmoreo del fonte del battistero di Nocera Superiore

Fig. 2. Nocera Superiore, Battistero di Santa Maria Maggiore, interno, vasca battesimale.

le peculiarità e soprattutto considerandoli da un punto di vista non esclusivamente iconografico.

procedimento consiste, come è noto, nella realizzazione del decoro mediante l’abbassamento del piano di fondo, che veniva riempito con paste colorate, in modo da simulare l’effetto visivo dell’opus sectile, utilizzando una tecnica senza dubbio più economica e meno preziosa14.

Come già accennato, rimangono solo nove delle sedici lastre che dovevano rivestire la vasca battesimale, una delle quali, in precedenza reimpiegata nel pavimento, è stata ricollocata in situ durante i restauri10 che hanno coinvolto l’edificio dopo il crollo della cupola a seguito dell’eruzione del Vesuvio del 194411.

Va sottolineato che, purtroppo, l’impianto generale della decorazione è soltanto ipotizzabile a causa delle manomissioni che sono intervenute nel corso dei secoli, ma è possibile affermare che la composizione doveva essere caratterizzata da un sicuro gusto per la simmetria e dall’enfasi conferita al simbolo della croce: larghe fasce campite unicamente da essa si alternavano, infatti, ai disegni geometrici più complessi delle piccole lastre angolari. Non si può, peraltro, escludere che originariamente alcune parti del decoro potessero essere

Si possono distinguere sostanzialmente due diverse tipologie: cinque lastre sono rettangolari e sono campite al centro da una semplice croce; altre quattro, che rivestono gli angoli smussati dell’ottagono, sono di dimensioni inferiori e di forma quadrangolare e mostrano una decorazione più ricca e complessa. Tutto il decoro delle lastre è stato realizzato con la tecnica detta a champlevé12 o anche a niello o “a incrostazione di mastice”13. Tale

Boyd 1982, pp. 313-325; Barsanti, Pedone 2005, p. 413; Coden 2006, pp. 25-55; 2007, pp. 304-311. Lo champlevé era utilizzato per il decoro delle strutture interne degli edifici – quali capitelli, pilastri, cornici di porta o finestra, lastre di varie tipologie – già dal III secolo nelle regioni del Mediterraneo e del Mar Nero e ha trovato poi una grande diffusione soprattutto nel periodo protobizantino. La maggior concentrazione di testimonianze di questo tipo di decoro è individuabile in Siria, in particolare ad Antiochia, dove basterà ricordare i materiali dal Martyrium di Seleucia Pieria, datati in larga parte ai primi decenni del VI secolo e ornati sia da scene figurate sia da motivi geometrici (Stillwell 1941), ma anche a Cipro, dove riveste particolare importanza il cospicuo gruppo di rivestimenti marmorei che decoravano il complesso episcopale di Kourion, la maggior parte dei quali sono collocabili nel primo terzo del V secolo (Boyd 2007).

14

10 La disposizione delle lastre non è quella originaria, poiché dopo il crollo della volta nel corso dell’Ottocento, alcune sono andate distrutte e durante i restauri degli anni Trenta, condotti da Chierici, le lastre che erano state reimpiegate nel pavimento sono state appunto ricollocate nella vasca battesimale: Gambardella 2007, p. 79. Per la storia dei numerosi interventi di restauro si veda: Friello 2010, pp. 111-136. 11 Brandt 2006-07, p. 193. 12 Il parallelismo tra la lavorazione degli smalti su metallo e il mastice dei rivestimenti marmorei è stato suggerito per la prima volta da Bégule 1905, p. 3. 13 Per una discussione sulla questione terminologica si veda: Coden 2004, pp. 69-94.

319

Alessandra Avagliano Un confronto forse più convincente può essere individuato in altri esempi geograficamente più vicini ai nostri esemplari, tra i quali il più significativo è un sarcofago romano del III secolo, rilavorato nel corso del VI secolo nella parte posteriore, attualmente esposto nel Museo Archeologico della stessa Nocera18, per il quale ho potuto accertare senza alcun dubbio tracce di policromia19. Resta invece dubbia la presenza della policromia nella croce, assai simile alla nostra, sulla lastra che reca l’epitaffio dell’abate Pascasio conservata a Cava dei Tirreni, ma proveniente da Monte Sant’Angelo in Puglia e per la quale si possiede fortunatamente una datazione ad annum e cioè al 55420. Anche per l’Alfa e l’Omega non mancano confronti in aree limitrofe a Nocera, quali la lastra reimpiegata come paliotto d’altare nell’antica chiesa di Santa Maria a Capaccio, oggi Santuario della Vergine del Granato21, quella proveniente dalla chiesa di Santa Maria a Vico di Giffoni, oggi nell’atrio della Pinacoteca di Salerno22 e il paliotto d’altare della cripta della Cattedrale di Nola23, tutte attribuibili al VI secolo. È noto che la presenza delle lettere apocalittiche ai lati della croce è assai diffusa soprattutto in ambito sepolcrale con riferimento al passo dell’Apocalisse in cui Dio viene definito come principio e fine dell’universo24, ma va ricordato che esse, spesso, costituiscono un semplice riferimento a Cristo25, probabilmente, anche in relazione al Carme sul valore dell’Alfa e dell’Omega, composto da Paolino da Nola all’inizio del V secolo26.

Fig. 3. Nocera Superiore, Battistero di Santa Maria Maggiore, planimetria della vasca battesimale (Stettler 1940).

enfatizzate dall’uso dell’oro, di cui purtroppo non rimane alcuna traccia15. Per facilitare la comprensione dell’attuale distribuzione delle lastre, si è ritenuto opportuno numerarle iniziando da quella che si trova quasi di fronte all’ingresso orientale e proseguendo in senso orario (fig. 3). Le lastre nn. 2, 4, 6, 9 sono decorate da una croce latina con bracci patenti terminanti con piccoli peducci arcuati e con un bottone profilato al centro (fig. 4); nella lastra n. 8, invece, dai bracci della croce pendono le lettere apocalittiche Alfa e Omega (fig. 5), mentre il centro è occupato da un quadrato diviso da una crocetta in quattro quadratini minori. La croce, delineata da una sottile incisione, presenta la superficie ruvida destinata ad accogliere il mastice policromo: nella prima lastra (n. 2) restano, infatti, tracce di colore rosso (fig. 6). L’impiego di mastici policromi, come già evidenziato, assicurava un forte impatto visivo16, enfatizzando il tema simbolico della croce che riveste un ruolo di primaria importanza nel contesto della decorazione della vasca battesimale.

18 Il sarcofago, pubblicato per la prima volta da Romito 2005, pp. 18-19 e 23-24, che si occupa sostanzialmente dell’analisi della fronte romana, era già stato segnalato da Samaritani 1991, p. 24. È stato, successivamente, oggetto di una più lunga analisi da parte di Chiara Lambert (cf. 2005, p. 44; 2008, pp. 74-76, figg. 29a-e). 19 Con l’aiuto del dott. Vittorio Barra, al quale sono grata per la sua disponibilità e la sua professionalità, ho potuto condurre analisi non invasive mediante metodologie di imaging multispettrale, con particolare riferimento alla tecnica della luminescenza infrarossa indotta da irraggiamento visibile, che però non hanno restituito informazioni sui pigmenti, ma solo una lieve luminescenza, indice della presenza di materiale organico. Grazie all’analisi fotografica nel visibile è stato possibile individuare dei pigmenti di colore verde. Si ringraziano, inoltre, la dott.ssa Silvia Pacifico e la dott.ssa Matilde Romito per avermi accordato il permesso per condurre tali indagini e il personale del Museo Archeologico Provinciale dell’Agro Nocerino per la gentile collaborazione. 20 Lambert 2005, p. 48 e 51, fig. 17. 21 Calvino 1973, pp. 291-294, tav. I-II; Peduto 1976, pp. 39-40, tav. XIV, fig. 1; Mello 2001, pp. 69-91; Lambert 2005, p. 47-48, fig. 12; Ebanista 2013, p. 529, fig. 3c. 22 Mallardo 1938, pp. 271-291, tav. XXX; Carillo 2002, p. 102, fig. 6, p. 103, fig. 8, p. 110, n. 59; Peduto 2007, p. 22, n. 8; Lambert 2005, pp. 47-48, fig. 13. 23 Datata al V secolo da Mallardo 1938, pp. 271-291, tav. XXX, la lastra è stata successivamente ascritta al VI secolo. Cf. Ebanista 2013, p. 529 e 541, fig. 3b. 24 Apocalisse, 1.8; 21.6; 22.13. 25 La presenza delle lettere apocalittiche era stata interpretata anche come una polemica antiariana, ma tale tesi non ha riscontrato molti consensi. Per un approfondimento sulla questione, si veda Ferrua 1991, pp. 48-50, con bibliografia precedente. 26 Paolino, infatti, interpreta le due lettere come l’espressione della Trinità divina, dal momento che entrambe sono costituite da tre linee e come rappresentazione della passione redentrice di Cristo, in quanto inizio della vita e fine dei mali (Paolino da Nola, Carmina, 19.643-647 [ed. von Hartel 1884, p. 140]): «Et quia morte crucis cunctis Deus omnia Christus / extat in exortum vitae finemque malorum, / alpha crucem circumstat et ω, tribus utraque virgis / littera diversam trina ratione figuram / perficiens, quia perfectum est mens una, triplex vis.» Cf. Guarducci 1958, p.59; 1964, p. 469.

La resa delle croci rivela una certa analogia con gli esemplari di ambito ravennate, in particolare con quella sul parapetto curvilineo dell’ambone in opera nel battistero della cattedrale, per il quale, tuttavia, gli studiosi hanno proposto la presenza di un inserto metallico17, che nel caso nocerino sarebbe da escludere, poiché non sembra possibile rintracciare fori per l’incasso né si rileva una sufficiente profondità dell’incavo. Va ricordato che, in alcuni casi, le sculture a champlevé erano provviste anche di doratura, che assicurava indubbiamente una significativa enfasi coloristica. Metzger 1980, p. 545 e 561; Boyd 1999, p. 51 n. 5. Resti di doratura sono conservati, ad esempio, in una lastra di Amathonte. Pralong 1990, p. 994, figg. 9-10. 16 Lo spessore delle lastre non è rilevabile. Le misure sono le seguenti: n.1 cm 60,5x52,5; n. 2 cm 64x195; n. 3 cm 65x70; n. 4 cm 62x207; n. 5 cm 64x64; n. 6 cm 65x216; n. 7 cm 63x50; n. 8 cm 64x197; n. 9 cm 64x207. Per una trattazione sull’uso del colore nella scultura a champlevé si veda: Pedone 2012. 17 Non si conosce l’epoca in cui l’ambone, datato al terzo quarto del V secolo, sia stato reimpiegato nella vasca e se sia parte dell’antico arredo marmoreo del battistero o se provenga da edifici limitrofi. Cf. Novara 2005, p. 307, fig. 17. 15

320

Il rivestimento marmoreo del fonte del battistero di Nocera Superiore

Fig. 4. Nocera Superiore, Battistero di Santa Maria Maggiore, lastre della vasca battesimale nn. 2, 4, 6, 9.

Se, quindi, il repertorio delle lastre che rivestono i lati lunghi dell’ottagono della vasca è molto semplificato e si apparenta alle decorazioni di altri manufatti campani, quello presente sugli angoli è indubbiamente più articolato, essendo costituito da motivi geometrici di diversa natura,

anche complessi. Le quattro lastre (nn. 1, 3, 5, 7) hanno tutte, come si è detto, una forma approssimativamente quadrangolare e sono caratterizzate da un impiego variato della tecnica a champlevé, che talora viene utilizzata per evidenziare il fondo, talora per dare maggior risalto alla 321

Alessandra Avagliano

Fig. 5. Nocera Superiore, Battistero di Santa Maria Maggiore, lastra della vasca battesimale n. 8.

Fig. 6. Nocera Superiore, Battistero di Santa Maria Maggiore, particolare della lastra n. 2.

figura, in composizioni nelle quali vi è una predominanza di motivi circolari e a losanga e uno spiccato gusto per la simmetria (fig. 7).

maniera diversa in ciascuna lastra e vi è la ricorrenza di un inconsueto motivo che potremmo quasi definire a tratteggio.

Due delle quattro lastre (nn. 1 e 3) sono ornate da uno schema decorativo più semplificato nel quale predomina l’elemento circolare. Le altre due (nn. 5 e 7), invece, sono molto più elaborate e sono caratterizzate dalla presenza della croce, inserita in una ricca composizione geometrica. È interessante notare che le incorniciature sono rese in

Se l’aspetto generale è, comunque, riconducibile a un repertorio iconografico ben noto nell’ambito della produzione scultorea a champlevé, va sottolineato che non sembra possibile individuare un riscontro palmare per il decoro delle singole lastre, che si distinguono, quindi, per una estrema originalità. 322

Il rivestimento marmoreo del fonte del battistero di Nocera Superiore

Fig. 7. Nocera Superiore, Battistero di Santa Maria Maggiore, lastre della vasca battesimale nn. 1, 3, 5, 7.

Si possono, comunque, segnalare dei confronti interessanti soprattutto nell’area del Mediterraneo orientale e del Mar Nero, datati tra V e VI secolo. In particolare, il motivo che decora la seconda lastra (n. 3) è molto simile a quello di un frammento di lastra di Seleucia Pieria presso Antiochia27 e di un altro conservato invece nel lapidario di Santo Stefano a Mesembria28, in cui il sistema di pieni e vuoti si articola in maniera opposta rispetto a quello nocerino. Questo tipo di schema decorativo, che non è particolarmente complesso, campisce anche l’intera superficie di una lastra rinvenuta nello scavo della chiesa ravennate di Sant’Apollinare in Classe, oggi conservata nel Museo Nazionale29.

Per quel che riguarda la terza lastra (n. 5)30, inoltre, va sottolineato che il gusto per la decorazione geometrica e lo sviluppo circolare della composizione è rintracciabile anche in un gruppo di frammenti provenienti da Cipro31, da Antiochia32 e in una lastra a Khirbat al-Minya in Galilea33. In alcuni di questi manufatti è individuabile anche il motivo Va segnalato che lo Stettler aveva messo, convincentemente, in relazione questa lastra con un alcuni dischi in opus sectile, che decorano le esedre della Santa Sofia costantinopolitana. Anche in questo caso, un clipeo caratterizzato dalla successione di figure triangolari racchiude al suo interno una croce. Cf. Stettler 1940, p. 132 (trad. it. p. 52). Per l’immagine si veda Antoniades 1908, p. 267, fig. 340 e più recentemente Russo 2012, tav. I, fig. 1. 31 Pralong 1998, pl. IX, fig. 9; Boyd 2007, pp. 271-273, pl. 6.13, nn. 117118. 32 Stillwell 1941, p. 134, pl. 29, nn. 510-511; Djobadze 1986, tav. 18, fig. 75, n. I 74, e p. 49. 33 La lastra è stata trovata negli scavi dell’area del Palazzo omayyade e con tutta probabilità si tratta di un manufatto bizantino reimpiegato nel periodo islamico: Rosen Ayalon 1974, pl. 51A; Ritter 2012, p. 129, fig. 4a. 30

Il frammento, datato al tardo V secolo, fa parte di un gruppo di quattordici frammenti decorati con scene del Nuovo Testamento: Stillwell 1941, p. 129, pl. 24, n. 447. 28 Barsanti 1989, p. 206, e 205, fig. 167. 29 La lastra, di cui non è ancora stata pubblicata una foto, è citata in Boyd 2007, p. 299; Barsanti 2008, p. 521, n. 20. 27

323

Alessandra Avagliano della croce centrale racchiusa da un clipeo, che però è resa a rilievo pieno.

di restaurazione dell’Impero avviato da Giustiniano alla fine della guerra greco-gotica.

Un’ultima considerazione riguarda la predominanza delle figure triangolari, soprattutto nell’ultima lastra (n. 7): tale motivo, infatti, è largamente utilizzato nei decori di tipo geometrico anche come protagonista delle composizioni sia di epoca bizantina – si veda ad esempio il fregio rinvenuto nell’area della Basilica Campanopetra a Salamina di Cipro e datato entro l’inizio del VI secolo34-, sia ancora in epoca omayyade, forse eco di più antiche testimonianze della zona, quale, ad esempio, quello recuperato a Khirbat alMinya35, nei quali il decoro è costituito sostanzialmente da una successione di triangoli bianchi che si stagliano su fondo rosso. Non mi sembra trascurabile, infine, il confronto offerto da una base di colonna di Altıntaş presso Küthaya in Frigia datata alla prima età bizantina, sulla quale la croce presenta al suo interno una sequenza di triangoli di analogo gusto decorativo36.

Bibliografia Abbreviazione: ASPS = Archivio storico per la provincia di Salerno. Antoniades E. M., 1908. Έκφρασις της Αγίας Σοφίας, ήτοι μελέτη συνθετική και αναλυτική υπό έποψιν αρχιτεκτονικήν, αρχαιολογικήν και ιστορικήν του πολυθρυλήτου τεμένους Κωνσταντινουπόλεως, II, Atene. Barsanti C., 1989. ‘L’esportazione di marmi dal Proconneso nelle regioni pontiche durante il IV-VI secolo’, in Rivista dell’Istituto Nazionale di Archeologia e Storia dell’Arte, 3, pp. 91-220. Barsanti C., 2008. ‘Una nota sulla diffusione della scultura a incrostazione nelle regioni adriatiche del Meridione d’Italia tra XI e XIII secolo’, in C. Pennas, C. Vanderheyde (edd.) La sculpture byzantine VIIeXIIe siècles. Actes du colloque international organisé par la 2e Éphorie des antiquités byzantines et l’École française d’Athènes (6-8 septembre 2000), Atene (Bulletin de correspondance hellénique. Supplément, 49), pp. 515-557.

Questa breve rassegna di esempi che possono essere, in qualche modo, avvicinati alle lastre del battistero di Nocera lascia spazio ad alcune considerazioni. In primo luogo, ritengo che il decoro della vasca battesimale può essere considerato un unicum nel panorama della scultura decorativa del VI secolo, dal momento che non sembra possibile individuare una struttura, di dimensioni così notevoli, che abbia conservato un rivestimento marmoreo realizzato con la tecnica a champlevé ancora in opera.

Barsanti C., Pedone S., 2005. ‘Una nota sulla scultura ad incrostazione e il templon della Panaghia Episcopi di Santorini’, in F. Baratte, V. Déroche, C. Jolivet-Lévy, B. Pitarakis (edd.), Mélanges Jean-Pierre Sodini = Travaux et mémoires, 15, pp. 407-425.

Inoltre, va ricordato che in area occidentale, le attestazioni di decoro a incrostazione di mastice di epoca così precoce, vale a dire tra V e VI secolo37, sono relativamente scarse e concentrate in area medio e altoadriatica: esempi significativi sono individuabili a Ravenna38, Parenzo39 e Gata presso Spalato40. Il rivestimento del fonte battesimale di Nocera resta dunque un esempio isolato nell’Italia centro meridionale41 e semmai trova confronti se pur non palmari, come si è visto, in sculture della medesima tecnica nell’area bizantina, il che potrebbe confermare, quindi, l’ipotesi che vede l’inserimento dell’edificio nel disegno

Bégule L., 1905. Les incrustations décoratives des Cathédrales de Lyon et de Vienne. Recherche sur une décoration d’origine orientale et sur son développement dans l’art occidental du Moyen Âge, Lione. Bilotti P. E., 1924. ‘Il Tempio di S. Maria Maggiore in Nocera. Dagli atti della Commissione provinciale per la conservazione dei documenti e delle opere d’arte’, in ASPS, 4/3-4, pp. 161-162. Boyd S., 1982. ‘A little-known technique of architectural sculpture: champlevé reliefs from Cyprus’, in H. Hunger, W. Hörandner (edd.), XVI. Internationaler Byzantinistenkongress, Wien, 4.-9. Oktober 1981. Akten, II/5 = Jahrbuch der Österreichischen Byzantinistik, 32/5, pp. 313-325.

Roux 1998, pp. 214-215, AR 821. 35 Ritter 2012, p. 134, fig. 15. 36 Niewöhner 2006, p. 437, fig. 16, cat. 39. 37 Semmai dobbiamo segnalare alcuni esemplari nei quali l’uso del mastice non è destinato a campiture di una certa ampiezza, ma a linee incise che definiscono il disegno, con un risultato che si avvicina più al niello che all’opus sectile. Si pensi, ad esempio, alla già citata lastra di Santa Maria del Granato a Capaccio, al retro del sarcofago del Museo Archeologico di Nocera Inferiore e al paliotto della Cattedrale di Nola (cf. supra). Manufatti di questo tipo sono diffusi anche nel nord Italia e Oltralpe: per una breve rassegna si veda Coden 2016, pp. 235-239, con bibliografia precedente. 38 Si pensi, ad esempio, alla cornice lavorata a champlevé della chiesa di San Vitale: Novara 1998, p. 89, fig. 13. 39 Si tratta di due piccoli capitellini di parasta, collocati nell’abside e datati alla metà del VI secolo. Russo 1991, pp. 105-108, figg. 80, 82, nn. 57, 60. 40 Nel museo di Almissa è, infatti, conservata una piccola lunetta con due colombe affrontate, datata alla metà del VI secolo: Coden 2004, pp. 82 e 84, fig. 20. 41 In Campania, infatti, sculture a champlevé trovano una certa diffusione soltanto in epoca successiva. Cf. Coden 2006, pp. 409-411. 34

Boyd S., 1999. ‘Champlevé production in Early Byzantine Cyprus’, in N. Patterson Ševčenko, C. Moss (edd.), Medieval Cyprus. Studies in Art, Architecture and History in Memory of Doula Mouriki, Princeton, pp. 49-70. Boyd S., 2007. ‘The champlevé revetments’, in A. H. S. Megaw (dir.), Kourion. Excavations in the Episcopal Precinct, Washington/Carmbidge (MA) (Dumbarton Oaks Studies, 38), pp. 235-320.

324

Il rivestimento marmoreo del fonte del battistero di Nocera Superiore Brandt O., 2006-07. ‘Osservazioni sul battistero paleocristiano di Nocera Superiore’, in Opuscola Romana, 31-32, pp. 189-202

D’Alessio G., 1935a. ‘Il Tempio Battesimale di S. M. Maggiore in Nocera Superiore. Note da documenti inediti e da tradizioni’, in ASPS. Nuova Serie, 3/3, pp. 202-208.

Brandt O., 2012. Battisteri oltre la pianta. Gli alzati di nove battisteri paleocristiani in Italia, Città del Vaticano (Studi di Antichità cristiana, 64).

D’Alessio G., 1935b. ‘Il Tempio Battesimale di S. M. Maggiore in Nocera Superiore (Continuazione)’, in ASPS. Nuova Serie, 3/4, pp. 227-244.

Brandt O., 2016. ‘Tra Napoli, Cimitile e Nocera Superiore: nuovi dati sull’orizzonte architettonico e cronologico del Battistero di S. Giovanni in Fonte’, in M. Rotili, C. Ebanista (edd.), Territorio, insediamenti e necropoli fra tarda Antichità e Alto Medioevo. Atti del Convegno internazionale di studi Territorio e insediamenti fra tarda Antichità e Alto Medioevo, Cimitile-Santa Maria Capua Vetere, 13-14 giugno 2013 – Atti del Convegno internazionale di studi Luoghi di culto, necropoli e prassi funeraria fra tarda Antichità e Medioevo, Cimitile-Santa Maria Capua Vetere, 19-20 giugno 2014, Napoli (Giornate sulla Tarda Antichità e il Medioevo, 7), pp. 271-284.

De Angelis M., 1926. ‘Basiliche vecchie, civiltà nuova. Le origini della nuova civiltà italica’, in ASPS, 5/1, pp. 3-90. Djobadze W. (dir.), 1986. Archeological Investigations in the Region West of Antioch on-the-Orontes, Stoccarda (Forschungen zur Kunstgeschichte und christlichen Archäologie, 13). Ebanista C., 2013. ‘Lastre con decorazione incisa dalla catacomba di S. Gennaro a Napoli’, in F. Bisconti, M. Braconi (edd.), Incisioni figurate della tarda antichità, Atti del Convegno di Studi, Roma, Palazzo Massimo, 22-23 marzo 2012, Città del Vaticano (Sussidi allo studio delle Antichità cristiane, 25), pp. 527-546.

Calvino R., 1973. ‘Una lastra paleocristiana sconosciuta a Capaccio’, in Campania sacra, 4, pp. 291-294.

Ferrua A., 1991. La polemica antiariana nei monumenti paleocristiani, Città del Vaticano (Studi di antichità cristiana, 43).

Carillo S., 2002. ‘Tutela e restauro dei monumenti nella provincia di Salerno: gli atti della commissione archeologica (1873-1874)’, in Apollo, 17, pp. 97-141.

Fortunato T., Santangelo G., 2006. ‘Il Battistero paleocristiano di Santa Maria Maggiore (Nocera Superiore)’, in T. Fortunato (ed.), Nuceria. Scritti in onore di Raffaele Pucci, Postiglione, pp. 88-92.

Chierici G., 1934. ‘L’elemento romano nell’architettura paleocristiana della Campania’, in C. Galassi Paluzzi (ed.), Atti del III Congresso nazionale di studi romani (Roma, 23-27 aprile 1933), I, Bologna, pp. 207-214.

Friello A., 2010. ‘Il Battistero paleocristiano di Santa Maria Maggiore di Nocera Superiore: progetti e “restauri” nella prima metà del XIX secolo’, in Archivio storico per le province napoletane, 128, pp. 111-136.

Cirillo O., 1999. Dalla dissoluzione del linguaggio classico alla renovatio carolingia e ottoniana, Napoli (Architettura in Italia. Storia, temi e caratteri, 2), pp. 183-190.

Gambardella A., 2007. ‘Considerazioni di uno studioso restauratore del monumento per un decennio’, in Pappalardo 2007, pp. 73-80.

Coden F., 2004. ‘Da Bisanzio a Venezia: niello o champlevé? Questioni critiche sulla scultura ad incrostazione di mastice’, in Venezia, la IV Crociata e le arti = Saggi e memorie di storia dell’Arte, 28, pp. 6994.

Guarducci M., 1958. I graffiti sotto la confessione di San Pietro in Vaticano, I, Città del Vaticano.

Coden F., 2006. Corpus della scultura ad incrostazione di mastice nella penisola italiana (XI-XIII sec.), Padova (Humanitas, 3).

Guarducci M., 1964. ‘Il fenomeno orientale del simbolismo alfabetico e i suoi sviluppi nel mondo cristiano d’occidente’, in Atti del convegno internazionale sul tema, l’oriente cristiano nella storia della civiltà, Roma 31 marzo-3 aprile 1963, Firenze 4 aprile 1963, Roma (Problemi attuali di Scienza e di Cultura, 62), pp. 467-497.

Coden F., 2007. ‘Scultura ad incrostazione di mastice: confronti tra la tecnica orientale e quella occidentale’, in A. C. Quintavalle (ed.), Medioevo Mediterraneo: l’Occidente, Bisanzio e l’Islam. Atti del convegno internazionale di studi, Parma, 21-25 settembre 2004, Milano (I Convegni di Parma, 7), pp. 304-311.

Guiglia A., 2016. ‘Il VI secolo: da Simmaco (498-514) a Gregorio Magno (590-604)’, in M. D’Onofrio (ed.), La committenza artistica dei papi a Roma nel Medioevo, Roma (I Libri di Viella. Arte), pp. 109-144.

Coden F., 2016. ‘Nuove considerazioni sulla scultura ad incrostazione di mastice (dal Tardoantico alla fine del Medioevo)’, in P. A. Andreuccetti, D. Bindani (edd.), Il colore nel Medioevo. Arte, simbolo, tecnica. Tra materiali costitutivi e colori aggiunti: mosaici, intarsi e plastica lapidea. Atti delle giornate di studi, Lucca, 2425-26 ottobre 2013, Lucca (Collana di studi sul colore, 5), pp. 233-260.

Lambert C., 2005. ‘Un prezioso anello di congiunzione tra Tarda Antichità ed Altomedioevo nel Museo di Nocera’, in Apollo, 21, pp. 44-58. Lambert C., 2008. Studi di epigrafia tardoantica e medievale in Campania, I, Borgo S. Lorenzo (Medioevo scavato, 3). 325

Alessandra Avagliano paléochrétienne’, in Bulletin de correspondance hellénique, 114, pp. 994-996.

Mallardo D., 1938. ‘Una fronte d’altare nolana della fine del sec. V’, in Campania Romana. Studi e materiali, I, Napoli, pp. 271-291.

Pralong A., 1998, ‘À propos d’un bloc de marbre d’Iznik’, in ΕΥΨΥΧΙΑ. Mélanges offerts à Hélène Ahrweiler, II, Parigi (Byzantina Sorbonensia, 16/2), pp. 603-609.

Mariani A., 1987. ‘La decorazione in marmo scolpito nel territorio dell’antica Nuceria’, in Rassegna storica salernitana, 4, pp. 7-58.

Ritter M., 2012. ‘Umayyadisches Ornament und christliche Motive: Marmorrelieffriese (Champlevé) im Palast von Hirbat al-Minya’, in L. Korn, A. Heidenreich (edd.), In memoriam Marianne Barrucand, Wiesbaden (Beiträge zur Islamischen Kunst und Archäologie, 3), pp. 113-137.

Mello M., 2001. Studi paleocristiani, Salerno. Metzger C., 1980. ‘Exemples d’iconographie de mosaïque appliquée à la sculpture. À propos de deux plaques à décor « champlevé » du Musée du Louvre’, in Mélanges de l’École française de Rome. Antiquité, 92, pp. 545-561.

Romito M., 2005. Vecchi scavi, nuovi studi. Museo Archeologico Provinciale dell’Agro Nocerino nel Convento di Sant’Antonio a Nocera Inferiore, Salerno.

Natella P., 2010. ‘Catelia diaconissa di VI secolo e la fondazione del Battistero di S. Maria Maggiore in Nocera Campana’, in Rassegna del Centro di cultura e storia amalfitana, 39-40, pp. 7-29.

Rosen-Ayalon M., 1974. ‘Notes on a particular technique of architectural decoration’, in Israel Exploration Journal, 24, pp. 232-236.

Niewöhner P., 2006. ‘Frühbyzantinische Steinmetzarbeiten in Küthaya. Zu Topographie, Steinmetzwesen, und Siedlungsgeschichte einer zentralanatolischen Region’, in Istanbuler Mitteilungen, 56, pp. 407-473.

Roux G., 1998. Salamine de Chypre, XV, Parigi. Russo E., 1991. Sculture del complesso eufrasiano di Parenzo, Napoli (Pubblicazioni dell’Universita degli studi di Cassino. Sezione di studi filologici, letterari, storici, artistici e geografici, 1).

Novara P., 1998. ‘Il rivestimento parietale nella chiesa di San Vitale a Ravenna’, in Ravenna. Studi e ricerche, 5/2, pp. 61-118.

Russo E., 2012. ‘Preliminari sull’opus sectile nell’estradosso delle arcate delle gallerie a S. Sofia di Costantinopoli’, in Bizantinistica. Serie seconda, 14, pp. 45-52.

Novara P., 2005. ‘Considerazioni su alcune sculture note e poco note conservate in Ravenna’, in M. Tagliaferri (ed.), La Chiesa metropolitana ravennate e i suoi rapporti con la costa adriatica orientale. Atti del XXVII Convegno del Centro studi e ricerche antica provincia ecclesiastica ravennate, Ravenna 29-31 maggio 2003, Imola, pp. 287-310.

Samaritani C., 1991. Guida al Museo di Nocera Inferiore, Napoli. Santangelo G., 2014. ‘Nocera Superiore (Sa). Il battistero paleocristiano di Santa Maria Maggiore’, in Association pour l’Antiquité tardive. Bulletin, 23, pp. 25-39.

Pappalardo U. (ed.), 2007. Il Battistero di Nocera Superiore. Un capolavoro dell’architettura paleocristiana in Campania, Napoli (Grand tour, 1).

Scognamillo E., 1994. ‘Struttura e restauri del battistero di Nocera’, in A. Pecoraro (ed.), Nuceria Alfaterna e il suo territorio. Dalla fondazione ai longobardi, Nocera Inferiore, pp. 121-131.

Pedone S., 2012. ‘Il colore scolpito. Raffinatezze cromatiche nella scultura ad incrostazione del Medioevo mediterraneo’, in A. Acconcia Longo, G. Cavallo, A. Guiglia, A. Iacobini (edd.), La Sapienza bizantina. Un secolo di ricerche sulla civiltà di Bisanzio all’Università di Roma, Roma (Milion. Studi e ricerche d’arte bizantina, 8), pp. 203-224.

Stettler, M., 1940. ‘Das Baptisterium zu Nocera Superiore’, in Rivista di archeologia cristiana, 17, pp. 83-142 (trad. it. in Pappalardo 2007, pp. 21-59). Stillwell R. (dir.), 1941. Antioch on-the-Orontes, III, Princenton (Publications of the Committee for the Excavation of Antioch and its Vicinity).

Peduto P., 1976. ‘Aspetti urbanistici e caratteri architettonici di Capaccio vecchia’, in P. Delogu, G. Maetzke, P. Natella, P. Peduto, E. Tabaczynska, S. Tabaczynski (edd.), Caputaquis medievale, I, Salerno, pp. 33-46.

Testini P., Cantino Wataghin G., Pani Ermini L., 1989. ‘La cattedrale in Italia’, in N. Duval (ed.), 1989. Actes du XIe Congrès international d’archéologie chrétienne, Lyon, Vienne, Grenoble, Genève et Aoste (21–28 septembre 1986), Roma/Città del Vaticano (Collection de l’École française de Rome, 123; Studi di Antichità cristiana, 41), pp. 5-232.

Peduto P., 2007. ‘Arechi II a Salerno: continuità e rinnovamento’, in L. Sinisi (ed.), Presenze longobarde in Italia. Il caso della Puglia, Ravenna (Quaderni del S.A.G.E.O. Anno II, 2), pp. 19-30.

von Hartel W. (ed.), 1884. S. Pontii Meropii Paulini Nolani Opera, II, Vienna/Praga/Lipsia (Corpus scriptorum ecclesiasticorum Latinorum, 30).

Pensabene P., 2005. ‘Marmi e committenza negli edifici di spettacolo in Campania’, in Marmora, 1, pp. 69-143. Pralong A., 1990. ‘Rapport sur les travaux de l’École française à Amathonte de Chypre en 1989. La basilique

326

25 The Riddle of the Anatolian Cross Stones: Press Weights for Church or Monastic Estates? Philipp Niewöhner* Georg August-Universität Göttingen (Germany) Anatolia has preserved numerous large marble or stone blocks that are beautifully sculpted with crosses and other Christian symbols. Some have previously been attributed to the liturgical furniture of Byzantine churches, but technical features identify them as screw weights for large oil or wine presses, the so-called lever and screw presses. Screw weights are well known from other regions of the Byzantine Empire, where they normally remained undecorated. This paper presents an exemplary selection of decorated screw weights from Anatolia and gives an overview of their formal repertoire. The riddle of these cross stones is their decoration: why should press weights have been decorated, why with Christian symbols, why in Anatolia and not elsewhere, and why only a few, while most screw weights remained undecorated even in Anatolia? The answer may possibly be sought in Christian ownership, if the cross stones were to indicate that the presses belonged to monasteries or to the church. L’Anatolie a conservé de nombreux blocs de marbre ou de pierre de grande taille, magnifiquement sculptés de croix et d’autres symboles chrétiens. Certains ont déjà été attribués au mobilier liturgique des églises byzantines, mais leurs caractéristiques techniques les identifient comme des poids à vis pour de grandes presses à huile ou à vin, dites à levier et à vis. Les poids à vis sont bien connus dans d’autres régions de l’Empire byzantin, où ils restaient normalement sans décor. Cet article présente une sélection exemplaire de poids à vis décorés d’Anatolie et donne un aperçu de leur répertoire formel. L’énigme de ces pierres à croix est leur décor : pourquoi des poids de presse auraient-ils été décorés, pourquoi avec des symboles chrétiens, pourquoi en Anatolie et pas ailleurs, et pourquoi seulement quelques-uns, alors que la plupart des poids à vis sont restés sans décor, même en Anatolie ? La réponse doit peut-être être recherchée du côté de leur propriétaire chrétien, si les croix avaient pour objet d’indiquer que les presses appartenaient à des monastères ou à l’Église. In Kleinasien haben sich zahlreiche große Marmor- oder Steinblöcke erhalten, die mit Kreuzen und anderen christlichen Symbolen kunstvoll versehen sind. Einige hat man deshalb für die liturgische Ausstattung byzantinischer Kirchen gehalten, aber technische Merkmale identifizieren sie als Gewichte für große Öl- oder Weinpressen, die sogenannten Spindelpressen. Solche Gewichte sind auch aus anderen Regionen des byzantinischen Reiches bekannt, wo sie aber normalerweise nicht verziert waren. Anhand ausgewählter Beispiele gibt dieser Beitrag einen Überblick über die verzierten Pressgewichte Kleinasiens und ihr Formenrepertoire. Das Rätsel dieser Kreuzsteine besteht aus den Fragen, warum Preßgewichte dekoriert wurden, warum mit christlichen Symbolen, warum ausschließlich in Kleinasien und nicht anderswo, und warum nur in wenigen Einzelfällen, während das Gros der Preßgewichte auch in Kleinasien undekoriert blieb. Die Antwort könnte darin zu suchen sein, dass die Kreuze kirchlichen Besitz anzeigten, die solchermaßen dekorierten Pressen also Klöstern bzw. der Kirche angehörten. L‘Anatolia ha conservato numerosi grandi blocchi di marmo o pietra splendidamente scolpiti con croci e altri simboli cristiani. Alcuni di essi sono stati in precedenza attribuiti agli arredi liturgici delle chiese bizantine, ma le caratteristiche tecniche suggeriscono piuttosto di identificarli come pesi a vite per grandi presse da olio o da vino, le cosiddette presse a leva e a vite. I pesi a vite sono * I would like to thank Alessandra Guiglia for inviting me to participate in the thematic session that she co-convened with Claudia Barsanti, who, alas, is no longer with us. I would also like to remember Urs Peschlow and Joachim Kramer. The former discussed and disagreed with my interpretation and thus helped to improve my argument. Both

generously provided photographs and a line drawing as indicated; their photographs are archived at the Johannes Gutenberg University at Mainz (U. Peschlow) and at the German Archaeological Institute in Berlin (J. Kramer). Finally, thanks are due to the editors of this publication, and to the anonymous referees.

327

Philipp Niewöhner ben noti in altre regioni dell’impero bizantino, dove normalmente rimanevano senza decorazione. Questo contributo presenta una selezione di esempi di pesi a vite decorati dall’Anatolia e offre una panoramica del repertorio formale. L’enigma di queste pietre a croci è la loro decorazione: perché i pesi da pressa dovrebbero essere decorati, perché con simboli cristiani, perché in Anatolia e non altrove, e perché solo pochi di essi, mentre la maggior parte dei pesi a vite è rimasta senza decorazione anche nella stessa Anatolia? La risposta potrebbe essere ricercata nella proprietà cristiana, se le pietre con croci dovessero indicare che le presse appartenevano ai monasteri o alla chiesa. Anatolia has preserved numerous large marble or stone blocks that are beautifully sculpted with crosses and other Christian symbols. Some have previously been identified as parts of columns, others as altars. This paper discusses these and other possibilities one by one – columns, altars, cross monuments, mortars, and, lastly, screw weights for oil or wine presses – thereby giving an overview of the Anatolian cross stones and their decorative repertoire. As it turns out, only the last function as weights can explain the technical features of the stones and must therefore be the correct identification. Such screw weights are wellknown from other regions of the Byzantine empire, where they normally remained undecorated. This leads to the paper’s final question: why were some Anatolian weights decorated, and why with crosses? Columns Two cross stones near the Turkish capital and Byzantine metropolis of Ankara in Galatia led S. Mitchell to identify their find spot as the location of the memorial church for St Theodotus of Ancyra in the village of Malos.1 The two stones found by Mitchell are 65 and 75 cm high, one is heptagonal, the other octagonal, and some sides are decorated with crosses, others with plants.2 Each block also has a keyhole-shaped slot in the upper side. In addition, the octagonal block bears a short inscription invoking the help of St Theodotus for one Antoninos, son of Theodotos. Mitchell proposed that the keyhole-shaped slots served for the attachment of columns, and that the cross stones were part of the memorial church.

Fig. 1. Ankara in Galatia, Archaeological Museum Inv. 9019, octagonal cross stone with a keyhole-shaped cavity and dovetailed sockets on top; the three sides to the left are decorated with crosses (J. Kramer).

Octagonal or more frequently rectangular (but not heptagonal) column bases were indeed commonly employed in the Early Byzantine churches of Anatolia,3 and some such bases (elsewhere) are also decorated with crosses.4 The peculiar keyhole-shaped slots, however, which Mitchell does not depict, but which are known from other cross stones (fig. 1–4), do not appear to be dowel holes for the attachment of columns. While Mitchell’s understanding of the blocks as column bases and parts of a church would therefore seem to be amiss, his further conclusions about the location of the memorial complex of St Theodotus may still stand confirmed, as will transpire at the end of this paper. Mitchell 1982, pp. 99 f., pl. 25b and 26a. Mitchell notes a third block as also being octagonal, but in this case he does not mention any decoration or keyhole-shaped cavity. 3 For example, Niewöhner 2007, p. 172 (bibliography); 2016, pp. 16–18. 4 For example, at the Episcopal Basilica of Stobi in North Macedonia: Audley-Miller et al. 2013, pp. 113, 132, cat. 34. 1 2

Fig. 2. Ankara, Archaeological Museum, as fig. 1: top view of keyhole-shaped cavity with dovetailed sockets (J. Kramer).

328

The Riddle of the Anatolian Cross Stones: Press Weights for Church or Monastic Estates? Altars Other, similar cross stones in the north-western Anatolian province of Bithynia have been identified as altars and their keyhole-shaped cavities as sepulchres for the deposition of relics.5 The deposition of relics began to be considered indispensable for the consecration of altars in the Early Byzantine period.6 A monolithic round altar in the archaeological museum at Kayseri is indeed decorated with large crosses as well as other, figural reliefs, but does not contain a cavity for relics.7 Four octagonal blocks or pedestals in the nave of the East Church at Alahan Manastır in Isauria are also decorated with one or several crosses, but their function remains to be established.8 An octagonal block in the crypt of the church of Panagia at Gereme south of Kayseri is decorated with a cross and has a cross-shaped cavity with an additional, deeper round hole in the centre.9 The cavity has rabbeted edges and pinholes (?), may possibly have been covered by a (metallic?) lid, and could conceivably have served or been re-used as a sepulchre for relics.10

Fig. 3. Boğazköy in Galatia, Museum, octagonal cross stone with a keyhole-shaped cavity and dovetailed sockets on top (U. Peschlow).

The cross stones in question do not, however, contain cross-shaped sepulchres but keyhole-shaped cavities without fixtures for lids (fig. 1–4). Moreover, the ‘bits’ of the ‘keyholes’ have dovetailed cross-sections and look like sockets for the fastening of wedge-shaped anchors (fig. 2). Such dovetails are a characteristic feature of all cross stones under consideration, but could take different shapes. Alternatively to the keyhole-shaped combination with the central hole, the dovetails also occur as separate grooves on two opposite sides of the block, again and always with 5 Dörner 1941; 1952, pp. 66 f., cat. 197–202, pl. 29 f.; Peschlow et al. 2002, pp. 514–516, cat. 98, 191. An octagonal cross stone at Amasya is 69 cm high, has a round cavity, is decorated with grape vines, and is broken around the upper perimeter (Keskin 2014), where it would have been hollowed out by the dovetail, all of which suggests a reconstruction as a press weight rather than as an altar (cf. infra). Altın, Şahin 2018 hypothesise that a 64 cm high press weight that they found at Orhaniye near Iznik could be re-used and may originally have served as Christian altar or grave marker, because it has a cross monogram engraved on its front, but the evidence of the other cross stones under discussion here leaves no doubt that Altın and Şahin are in fact dealing with an original press weight that received its Christian decoration at the outset. 6 Troianos 2000, p. 383; Botte, Brakmann 2004, cc. 1150, 1168. 7 Originally from Pusatlı near Tomarza in Cappadocia: Kollwitz 1950; Restle 1979, I, p. 165; II, fig. 110 and 112; Lemaigre Demesnil, 2002, pp. 41 f.; Işler, Kadiroğlu, Peker 2010, pp. 100 and 114, fig. 3. 8 Gough 1985, pp. 115 f., fig. 53, pl. 48. Cf. two square blocks, one with a cross, that formed the ends of the templon in the transept basilica extra muros at Corycus in Cilicia: Herzfeld, Guyer 1930, pp. 116–118, fig. 116. 9 Çorağan Karakaya 2010, p. 182, fig. 4. 10 Cf. the basilica at the Letoon near Xanthos in Lycia, where a re-used fluted column drum may have served as an altar, with a hole on top as sepulchre: Metzger 1966, p. 109, fig. 9. Or a re-used round statue base in the apse of a cemetery church at the neighbouring city of Patara, also in Lycia: Peschlow 2017, p. 287. H. Hörmann’s suggestion that an octagonal column base should have been re-used as an altar at Ephesus, because it has a 14 cm wide and 8 cm deep hole on top, carries little conviction: Keil et al. 1951, p. 144, fig. 32, 1. Massive block-shaped altars do not appear to have been widespread in Anatolia, whilst remains of numerous altar tables have survived; cf. Niewöhner 2007, p. 181, n. 1008 (bibliography). Cf. also cross-shaped reliquaries and sepulchres with rabbeted edges in the base slabs of altar tables: Buschhausen 1971; Peschlow 2006; Kalinowski 2011; Kazan 2015. In light of what follows below, however, the block at Gereme may in fact be a Late Antique press weight that was later brought into the Byzantine church and possibly slightly altered in order to serve as an altar and sepulchre.

Fig. 4. Boğazköy, Museum, as fig. 3: front with cross (below) and top side with keyhole-shaped cavity and dovetailed sockets (U. Peschlow, 1983).

329

Philipp Niewöhner

Fig. 5. Çorum, Archaeological Museum, rectangular cross stone with dovetailed sockets on opposite sides and a cupshaped cavity on top (P. Niewöhner, 2012).

Fig. 8. Çorum, Archaeological Museum, rectangular cross stone with a long dovetailed channel and a cup-shaped cavity on top (P. Niewöhner, 2012).

by a larger, or containing a smaller, central hole (fig. 8).12 In one shape or another, both the dovetails and the central holes appear to have been essential features of the cross stones. They suggest the need for strong joints that were focussed centrally on the blocks. All of this does not agree with a reconstruction as altars, and most cross stones would also appear unpractically low for that purpose, as seems to be confirmed by the greater height of surviving altars from the Early Byzantine period.13 Cross monuments Fig. 6. Günyüzü near Germia in Galatia, octagonal cross stone (H.: 75 cm) with dovetailed sockets (H.: 50 cm, W. at bottom: 23 cm, at top: 14 cm, D.: 11 cm) on opposite sides and a cup-shaped cavity (Ø: 48 cm) on top, Ø cross medallion: 52 cm, total W.: 90 cm (P. Niewöhner, 2009).

The decoration with crosses may suggest that the stones served as stands for cross monuments, rather than as column bases or altars. The so-called four-sided stelae or pre-altar crosses inside and outside of early Armenian and Georgian churches come to mind.14 These tall and heavy monuments would have required massive stands, and it seems possible that some Caucasian blocks with crosses not unlike the Anatolian cross stones, albeit without dovetails, may have served this purpose.15 Prealtar crosses supposedly originated in Jerusalem,16 and it appears conceivable that some such custom was also practised in Anatolia, halfway between the Caucasus and the Holy Land. Large processional crosses that could have required massive stands are attested at Constantinople.17 A cross monument on the Forum of Constantine must Frankel 1997, type 3; 1999, p. 113. Cf. for example Terry 1988, pp. 43 f., 46–48, cat. 49 f. with references to more altars and bibliography; Angiolini Martinelli 1968, pp. 15–22; Deichmann 1958–1989, II/2,3, pp. 329 f. (bibliography); Peschlow 2006, pp. 182–185. 14 Mikeladze-Andreasen 2010. Cf. also http://georgiske-kors.dk/ ; Dümler 2011; Tchakerian 2016. 15 For information about and photographs of such cross stones at Chwilischa, Jeli, Kumurdo, Manglisi, Seda-Tmogwi, and Speti, I am indebted to Brigitta and Rolf Schrade. 16 Schrade 2004. Some such crosses and their stands seem to be attested in Syria: Mundell Mango 1986, 195. 17 Cotsonis 1994. 12

Fig. 7. Iznik in Bithynia, Archaeological Museum, rectangular cross stone with dovetailed sockets on opposite sides (U. Peschlow, 1991).

13

a central cup-shaped cavity on top (fig. 5–7).11 A third type has a single long dovetailed channel across the width of the block as if to create a sliding joint, always interrupted For the find spot and historical context of the cross stone at Günyüzü near Germia (fig. 6), see Niewöhner 2013a.

11

330

The Riddle of the Anatolian Cross Stones: Press Weights for Church or Monastic Estates?

Fig. 10. Istanbul, Archaeological Museum Inv. 7456, octagonal cross stone, broken on top, with remains of a cup-shaped cavity and rectangular (dowel?) holes (not visible in this figure); two sides missing completely, the other six decorated with aediculae that contain crosses on globes (P. Niewöhner, 2007).

blocks without any decoration. Peschlow reconstructs a wooden frame that is anchored in the dovetails and serves as a mount for a pestle, with the central hole as mortar.21 In support of his hypothesis, Peschlow claims that the central holes are abraded from their use as mortars, but this is often not the case (fig. 10). Moreover, Peschlow’s hypothesis fails to account for keyhole-shaped holes that also occur on blocks without decoration,22 but cannot be reconstructed as mortars (fig. 1–4). The decoration with crosses and other Christian symbols is not addressed at all. One may also question the practicability of the proposed reconstruction of a mechanical pestle that could add leverage, but which seems out of proportion, considering that the presumed mortar holes are hardly larger than many handheld mortars.23

Fig. 9. Istanbul, Archaeological Museum Inv. 5885, crossshaped marble gravestone (P. Niewöhner, 2007). The inscription reads † Ενθάδε κατάκιτε Λαπετία γυνή Θεόδορος κουβίτορος ετελιόθη μη(νί) Δεκεβρίου ινδ(ικτιώνος) α †, Here rests Lapetia, wife of Theodor the surveyor(?); she died in the month of December in the first indiction.

also have been relatively large,18 and a marble cross at the Hebdomon outside the capital would presumably have required some sort of stand, too.19 Other, similar marble crosses that served as grave markers would likewise have benefited from stands (fig. 9).20 None of these cross monuments, however, appears to relate in any way to the peculiar dovetails of the Anatolian cross stones, and there seems to be no obvious way that any of the crosses could have been affixed to the stones under consideration.

Press weights From a technical point of view, only a use as weights for lever and screw presses for oil or wine appears to offer satisfactory explanations for all the different cavities and dovetails of the various blocks. J. G. C. Anderson made this abundantly clear as early as 1903, when he published a comprehensive selection of undecorated but equally dimensioned and dressed stones in the north Anatolian province of Pontus.24 In the Early Byzantine period olives and grape vines were cultivated throughout Anatolia.25

Mortars Alternatively, U. Peschlow ignores the cross stones and instead tries to explain the central cup-shaped cavities in combination with lateral dovetails that also occur on Parastaseis syntomoi chronikai 30: 10–31; Patria Constantinopolis 160: 15–20. Cf. Klein 2004, p. 51. For more references to early Byzantine cross monuments, cf. Bouras, Cutler 1991; Heid 2002, pp. 227–237. 19 Downey 1954, p. 280. 20 Cf. Öğüt-Polat, Şahin 1986, pp. 125 f., cat. 102, pl. 13, with more examples and bibliography. Cf. Altın, Şahin 2018, and supra, n. 5, for a misinterpretation of a cross stone as Christian grave marker. 18

Peschlow 1993. Anderson 1903, pp. 14–16. 23 Korkut 2002; Vanderheyde 2003, pp. 69–70 and 76–78. 24 Anderson 1903, pp. 14–16. 25 Vermoere 2004, pp. 164, and 157, fig. 4.21; Izdebski 2013. 21 22

331

Philipp Niewöhner With regard to Peschlow’s observation of abraded holes that he explains by the stones’ use as mortars, it should be noted that in a press weight the shape and condition of the central cavity depends on the screw, whether it rotated inside the hole, and whether the cavity was padded with leather, wood, or metal in order to minimise friction.27 Only if the screw rotated inside a hole without padding would this lead to abrasion.28 This most basic arrangement appears to have been the rule at the city of Amorium in Phrygia, where C. Lightfoot observes that ‘all of the Amorium examples display clear signs of rotary wear in the round socket.’29 Use as weights for lever and screw presses could also explain the polygonal shapes of some cross stones (fig. 1–4, 6 and 10–13), because polygonal blocks minimised the danger of sliding or tilting in case the suspension was not exactly centred. In addition, polygonal weights had the advantage that they would protrude minimally beyond the width of the lever and would therefore not be in the way when it came to turning the screw in order to lift or lower the lever. In Palestine and elsewhere, similarly large and heavy screw weights were more often cylindrical, which made no difference as to their use and function, but the cylindrical weights were normally plain, without any decoration.30 In Anatolia, polygonal weights may thus have been preferred not least because they were better suited for relief decoration.

Fig. 11. Istanbul, Archaeological Museum, octagonal cross stone, two fragments, broken on top; decorated with arcades that contain a vine, date palms (2x), a XP-medallion, a cross on a globe, a stylite’s column, among others (P. Niewöhner, 2007).

Why the decoration, and why with crosses? The decoration constitutes the riddle of the Anatolian cross stones. It seems to be a regional peculiarity. The blocks consist of whatever stone happened to be locally available. The extensively published screw weights of other provinces do not include any cross stones,31 which goes some way towards explaining why the Anatolian cross stones were not recognised as press weights and instead mistaken for column bases and for altars. Their decoration is typically centred on one (fig. 3–8 and 11) or several crosses (fig. 1, 10 and 13), which suggests that the Anatolian fashion of decorating screw weights originated after Christianisation, i.e., in Late Antiquity or the Early Byzantine period.32 Some other decorated press weights that are hardly older consist of re-used marbles of various types, for example a weight with an ancient frieze of oak leaves re-used from the Temple of Augustus at Ankara33 or two weights in the shape of inverted ancient column capitals.34

Fig. 12. Istanbul, Archaeological Museum, as fig. 11: top view of mortised centre part with keyhole-shaped cavity, dovetailed sockets, and four (dowel?) holes (P. Niewöhner, 2007).

Frankel 1997, p. 81; 1999, pp. 107–111. Cf. the abraded cavity of a modern press weight in Portugal: Amouretti, Comet, Ney, Paillet 1984, pp. 394–397, fig. 5–7. 29 Lightfoot 2003, pp. 74 f. and in particular p. 77, cat. 3, pl. 5, 3 f. 30 Frankel 1997, p. 81. 31 Frankel 1997; 1999; Ayalon, Frankel, Kloner 2009; Caillet, Popović, Vasić 2010, pp. 490 f., fig. 6, 23 f. 32 Cf. supra, n. 1 and 5, as well as Pralong 2003, pp. 246, 266, cat. 71; Belke 1996, fig. 48, 72, 74 f.; Lightfoot 2003, p. 78, pl. 5, 10; Sayar 2004, p. 228, fig. 12; Laflι, Zäh 2008, p. 711, cat. 4.1.9; Laflı 2012, pp. 270 f., 275, fig. 13a, and p. 278, fig. 24c. 33 Peschlow 2015, p. 33, pl. 10, fig. 28 f. 34 Asgari 1985, pp. 77 f. fig. 10–13; Peschlow 2015, p. 135, pl. 70, fig. 255; Russo 2011, p. 53, fig. 68. 27 28

Anderson also points out that ‘where there are no mortised sockets [i.e. dovetails] in the sides the stone was evidently set in a regular frame’. R. Frankel supplements this with a technical explanation for the small, additional (dowel?) holes that sometimes surround the central cavity (fig. 11–12).26 26

Frankel 1999, p. 119.

332

The Riddle of the Anatolian Cross Stones: Press Weights for Church or Monastic Estates? that these substances would gain through contact with relics or at the Eucharist. Alternatively, the cross decoration may conceivably have had a more specific meaning, and one possibility is considered below; it lacks definite proof, but can highlight the need for more information about the original context of the Anatolian cross stones and serve as a working hypothesis for future research. If, then, the cross stones could be taken to relate certain presses to the church or to monasteries, this would explain why screw weights began to be decorated in Late Antiquity or the Early Byzantine period. The regional limitation of cross stones to Anatolia is in accordance with other regional traits of the Late Antique or Early Byzantine church, such as architecture,38 liturgy,39 liturgical furniture,40 and the use of, or abstinence from, figural representation,41 all of which could vary greatly between the many different provinces of the vast empire. Thus, the Anatolian cross stones may conceivably have designated presses that belonged to the church or to monasteries. It goes without saying that larger monasteries in particular would have consumed copious amounts of oil and wine,42 and archaeological evidence shows that some monasteries operated their own presses.43 Alternatively, presses with cross stones may have belonged to estates or villages that were in the possession of monasteries or the church.44

Fig. 13. Istanbul, Archaeological Museum Inv. 1236, octagonal cross stone, broken on top, with remains of a cup-shaped cavity and four rectangular (dowel?) holes; all eight sides are decorated with crosses, every second of which stands on a globe (U. Peschlow).

If a lever and screw press was to be decorated, the screw weight was an obvious candidate, for it had to be cut anyway, so decorating it required little additional labour, and it was in full view of anybody who had dealings with the press. Whoever wanted to access the press bed in order to fill in olives or grapes or to take out oil or juice would normally pass the weight. This would entail considerable publicity, for the cross stones are among the heaviest screw weights and imply long levers and big presses that served large estates or several smaller producers.35 Thus, over time, a good number of people would have busied themselves around the press while their goods were being processed, as is common procedure at comparably sized presses to this day.

This hypothesis could be checked, if cross stones, which are so far known only as museum pieces or stray finds, were to be found and documented in their original contexts – which brings the argument back full circle to the beginning of this paper and S. Mitchell’s observation of the two blocks outside Ankara, the only contextualised cross stones to date. Unfortunately, the identification of the find context, in this case the memorial complex of St Theodotus at Malos, is based on the cross stones themselves, which poses the problem of a circular argument. There is, however, other, independent evidence that identifies the find spot, if not with St Theodotus, then at least with the village of Malos, which was also known for its wine. In addition, Mitchell’s argument is further strengthened by the invocation of St Theodotus on one cross stone, for such invocations were typically placed in the sanctuaries of the invoked saints, for example on liturgical furniture.45 It does, therefore, seem conceivable and even likely that Mitchell’s cross stones belonged to presses that were part of, or associated with, the memorial complex of St Theodotus.

It remains to be asked, however, why oil or wine presses were decorated at all, why this became customary in Late Antiquity or the Early Byzantine period, and why the decoration centred on crosses. Most screw weights did, after all, remain without decoration, even in Byzantine Anatolia.36 Thus, the few decorated blocks might have merely been an inconsequential fashion, the crosses to be understood only in general terms as apotropaic symbols, such as occur on many Late Antique and Byzantine buildings, inscriptions, and even on spoons.37 In the case of oil or wine presses, such symbolism could, perhaps, have been inspired by the special spiritual significance

Krautheimer 1986; Loosley 2012. Taft 1992; Parenti 2010. 40 Loosley 2012; Niewöhner 2013b, pp. 236–245. 41 Niewöhner 2014, pp. 263–268; Campagnolo et al. 2015. 42 Talbot 2002; Koder 2017. 43 Martini, Steckner 1993, pp. 143–149; Frankel 1997, p. 82; Lewit 2012, p. 146, n. 116. 44 Talbot, Kazhdan 1991. 45 Cf. for example Feissel 2006. 38

Lewit 2012. 36 For example, five screw weights without decoration in the vicinity of the cross stone at Günyüzü (fig. 6), one at Atlas, one at Ayvalı, and three at Gecek (cf. Niewöhner 2013a). For more undecorated screw weights, cf. Anderson 1903; Peschlow 1993; Lightfoot 2003; Aydınoğlu 2009; Keskin 2013; Keskin 2015, pp. 155–158. 37 Dinkler, Dinkler-von Schubert 1995, cc. 138–142; Hauser 1992; Taft 1996.

39

35

333

Philipp Niewöhner Bibliography

Popović (eds.), Caričin Grad, III, Rome (Collection de l’École française de Rome, 75/3), pp. 451–508.

Abbreviations:

Campagnolo M., Magdalino P., Martiniani-Reber M., Rey A.-L. (eds.), 2015. L’aniconisme dans l’art religieux byzantin. Actes du colloque de Genève (1–3 octobre 2009), Geneva.

AA = Archäologischer Anzeiger. AS = Anatolian Studies. AST = Araştırma Sonuçları Toplantısı.

Çorağan Karakaya N., 2010. ‘Gereme Panagia Kilisesi’, in K. Pektaş, S. Cirtil, S. Özgün Cirtil, G. K. Öztaşkin, H. Özdemir, E. Aktuğ, E. Uykur (eds.), XIII. Ortaçağ ve Türk Dönemi Kazıları ve Sanat Tarihi Araştırmaları Sempozyumu bildirileri, 14–16 Ekim 2009 – Proceedings of the XIIIth Symposium of Medieval and Turkish Period Excavations and Art Historical Researches, 14–16 October 2009, Denizli (Pamukkale Üniversitesi. Fen-Edebiyat Fakültesi. Sanat Tarihi Bölümü Yayınları, 1), pp. 175–183.

BZ = Byzantinische Zeitschrift. DOP = Dumbarton Oaks Papers. Ist. Mitt. = Istanbuler Mitteilungen. Altın A. A., Şahin M., 2018. ‘Nikaia’nın Antik Dönemdeki Tarımsal Üretimine Dair İzler: Nikaia’dan Bir Pres Taşı’, in Uludağ Üniversitesi Fen-Edebiyat Fakültesi Sosyal Bilimler Dergisi, 19, pp. 815–827. Amouretti M.-C., Comet C., Ney C., Paillet J.-L., 1984. ‘À propos du pressoir à huile’, in Mélanges de l’École française de Rome. Antiquité, 96, pp. 379–421.

Cotsonis J. A., 1994. Byzantine Figural Processional Crosses. Catalogue of an Exhibition at Dumbarton Oaks, 23 September 1994 through 29 January 1995, Washington (Dumbarton Oaks. Byzantine Collection. Publications, 10).

Anderson J. G. C., 1903. A Journey of Exploration in Pontus, Brussels (Studia Pontica, 1).

Deichmann F.-W., 1958–89. Ravenna. Hauptstadt des spätantiken Abendlandes, Stuttgart/Wiesbaden.

Angiolini Martinelli P., 1968. ‘Corpus’ della scultura paleocristiana bizantina ed altomedioevale di Ravenna, I, Rome.

Dinkler E., Dinkler-von Schubert E., 1995. ‘Kreuz’, in K. Wessel, M. Restle (ed.), Reallexikon zur byzantinischen Kunst, V, Stuttgart, cols. 1–219.

Asgari N., 1985. ‘Istanbul Temel Kazılarından Haberler – 1984’, in AST, 3, pp. 75–89.

Dörner F. K., 1941. ‘Zwei Reliquienaltäre von der bithynischen Halbinsel’, in BZ, 41, pp. 165–168.

Audley-Miller L., Niewöhner P., Prochaska W., 2013. ‘Marbles, quarries and workshops on the Highlands of Northern Macedonia’, in AA, 2013/1, pp. 95–145.

Dörner F. K., 1952. Bericht über eine Reise in Bithynien. Ausgeführt im Jahre 1948 im Auftrage der Österreichischen Akademie der Wissenschaften, Vienna (Österreichische Akademie der Wissenschaften. Philosophisch-historische Klasse. Denkschriften, 75/1).

Ayalon E., Frankel R., Kloner A. (eds.), 2009. Oil and Wine Presses in Israel from the Hellenistic, Roman and Byzantine Periods, Oxford (BAR Publishing International Series, 1972).

Downey G., 1954. ‘A processional cross’, in The Metropolitan Museum of Art Bulletin, 12, pp. 276–280.

Aydınoğlu Ü., 2009. Dağlık Kilikia Bölgesinde Antik Çağda Zeytinyağı ve Şarap Üretimi: Üretimin Arkeolojik Kanıtları, Istanbul.

Dümler B., 2011. ‘Zeichen allumfassenden Heils. Die Vielschichtigkeit des Programms armenischer Kreuzsteine unter Rückgriff auf Elemente spätantiker und byzantinischer Ikonographie’, in D. Bumazhnov, E. Grypeou, T. B. Sailors, A. Toepel (eds.), Bibel, Byzanz und christlicher Orient. Festschrift für Stephen Gerö zum 65. Geburtstag, Leuven/Paris/Walpole (Orientalia Lovaniensia Analecta, 187), pp. 413–459.

Belke K., 1996. Paphlagonien und Honorias, Vienna (Österreichische Akademie der Wissenschaften. Philosophisch-historische Klasse. Denkschriften, 249; Tabula Imperii Byzantini, 9). Botte B., Brakmann H., 2004. ‘Kirchweihe’, in E. Dassmann, G. Schöllgen (ed.), Reallexikon für Antike und Christentum, Sachwörterbuch zur Auseinandersetzung des Christentums mit der antiken Welt, XX, Stuttgart, cols. 1139–1169.

Feissel D., 2006. ‘Christliche und byzantinische Inschriften’, in P. Herrmann, W. Günther, N. Ehrhardt (eds.), Inschriften von Milet, III, Berlin (Milet. Ergebnisse der Ausgrabungen und Untersuchungen seit dem Jahre 1899, VI/3), pp. 289–296.

Bouras L. Ph., Cutler, A., 1991. ‘Cross, Processional’, in Kazhdan 1991, p. 553. Buschhausen H., 1971. Die spätrömischen Metallscrinia und frühchristlichen Reliquiare, I, Vienna (Wiener byzantinistische Studien, 9).

Frankel R., 1997. ‘Presses for oil and wine in the southern Levant in the Byzantine period’, in DOP, 51, pp. 73–84. Frankel R., 1999. Wine and Oil Production in Antiquity in Israel and other Mediterranean Countries, Sheffield (JSOT/ASOR Monograph Series, 10).

Caillet J.-P., Popović V., Vasić Č., 2010. ‘Le complexe 4, appelé couramment “palais épiscopal”’, in N. Duval, V.

334

The Riddle of the Anatolian Cross Stones: Press Weights for Church or Monastic Estates? Gough M. (ed.), 1985. Alahan. An Early Christian Monastery in Southern Turkey, Based on the Work of Michael Gough, Toronto (Pontifical Institute of Mediaeval Studies. Studies and Texts, 73).

in Asia Minor and the Stoneworks of Byzantine Period, Ankara. Klein H. A., 2004. Byzanz, der Westen und das ‘wahre’ Kreuz. Die Geschichte einer Reliquie und ihrer künstlerischen Fassung in Byzanz und im Abendland, Wiesbaden (Spätantike – frühes Christentum – Byzanz. Reihe B: Studien und Perspektiven, 17).

Hauser S. R., 1992. Spätantike und frühbyzantinische Silberlöffel. Bermerkungen zur Produktion von Luxusgütern im 5. bis 7. Jahrhundert, Münster (Jahrbuch für Antike und Christentum. Ergänzungsband, 19).

Koder J., 2017. ‘Byzantinisches Mönchtum und die Umwelt’, in H. Baron, F. Daim (eds.), A Most Pleasant Scene and an Inexhaustible Resource. Steps Towards a Byzantine Environmental History. Interdisciplinary Conference November 17th and 18th 2011 in Mainz, Mainz (Byzanz zwischen Orient und Okzident, 6), pp. 217–240.

Heid S., 2002. ‘Vexillum crucis’, in Rivista di archeologia cristiana, 78, pp. 191–259. Herzfeld E., Guyer S., 1930. Meriamlik und Korykos, zwei christliche Ruinenstätten des rauhen Kilikiens, Manchester (Monumenta Asiae Minoris antiqua, 2). Işler B., Kadiroğlu M., Peker N., 2010. ‘The iconography of the scene of ascension of Elijah in Cappadocian and Tao-Klardjetian wall paintings’, in S. Y. Aydın (ed.), İslâm ve Hıristiyan Sanatında Melekler, Peygamberler ve Azizler – Angels, Prophets and Saints in Islamic and Christian Art, Istanbul, pp. 97–122.

Kollwitz J., 1950. ‘Ein Altar im Museum von Kayseri’, in W. Tack (ed.), Festgabe für Alois Fuchs zum 70. Geburtstage am 19. Juni 1947, Paderborn, pp. 15–21. Korkut T., 2002. ‘Steinerne Mörserschalen aus Patara’, in AA, 2002/1, pp. 233–245.

Izdebski A., 2013. Rural Economy in Transition. Asia Minor from Late Antiquity into the Early Middle Ages, Warsaw (The Journal of Juristic Papyrology. Supplement, 18).

Krautheimer R., 1986. Early Christian and Byzantine Architecture, 4th ed., New Haven/London (The Pelican History of Art).

Kalinowski A., 2011. Frühchristliche Reliquiare im Kontext von Kultstrategien, Heilserwartung und sozialer Selbstdarstellung, Wiesbaden (Spätantike – frühes Christentum – Byzanz. Reihe B: Studien und Perspektiven, 32).

Laflı, E., 2012. ‘Archäologische Evidenz zum Weinanbau im südwestlichen Paphlagonien in römischer und frühbyzantinischer Zeit’, in E. Ohlshausen, V. Sauer (eds.), Die Schätze der Erde. Natürliche Resourcen in der antiken Welt, Stuttgart (Geographica Historica, 28), pp. 261-279.

Kazan G., 2015. ‘Arks of Constantinople, the New Jerusalem. The origins of the Byzantine sarcophagus reliquary’, in Byzantion, 85, pp. 77–125.

Laflı E., Zäh A., 2008. ‘Archäologische Forschungen im byzantinischen Hadrianopolis in Paphlagonien’, in BZ, 101/2, pp. 681–713.

Kazhdan A. P. (ed.), 1991. The Oxford Dictionary of Byzantium, New York/Oxford.

Lemaigre Demesnil N., 2002. ‘Sculptures figurées d’époque paléochrétienne en Cappadoce. À propos d’un chapiteau historié inédit’, in Cahiers archéologiques, 50, pp. 41–50.

Keil J., Soteriou G. A., Hörmann H., Miltner F., 1951. Die Johanneskirche, Vienna (Forschungen in Ephesos, IV/3).

Lewit, T., 2012. ‘Oil and wine press technology in its economic context’, in Antiquité tardive, 20, pp. 137–149.

Keskin E., 2013. ‘Çorum Ili ve Çevresinde Bulunan, Zeytinyağı ve Şarap Üretiminin Bizans Dönemine Ait Arkeolojik Kanıtları’, in N. Türker, G. Köroğlu, Ö. Deniz (eds.), I. Uluslararası Karadeniz Bölgesi Kültür Kongresi, 06–09 Ekim 2011, Sinop, Karabük, pp. 372–386.

Lightfoot C. S., 2003. ‘Stone screw press weights’, in C. Lightfoot (ed.), Amorium Reports, II, Oxford (BAR Publishing International Series, 1170), pp. 73–79. Loosley E., 2012. The Architecture and Liturgy of the Bema in Fourth- to Sixth-Century Syrian Churches, Leiden/ Boston (Texts and Studies in Eastern Christianity, 1).

Keskin E., 2014. ‘Amasya Müzesi’nde Bulunan Bir Altar Masası’, in M. Acara Eser, E. Bilget Fataha, G. Koyun (eds.), Uluslararası Katılımlı XVI. OrtaçağTürk Dönemi Kazıları ve Sanat Tarihi Araştırmaları Sempozyumu, Cumhuriyet Üniversitesi – Sivas, 18–20 Ekim 2012 / Proceedings of the XVIth International Symposium of Medieval-Turkish Era Excavations and Art History Researches, Cumhuriyet University – Sivas, 18–20th October 2012, II, Sivas, pp. 473–482.

Martini W., Steckner C., 1993. Das Gymnasium von Samos. Das frühbyzantinische Klostergut, Bonn (Samos, 17). Metzger H., 1966. ‘Fouilles du Létoon de Xanthos (1962– 1965)’, in Revue archéologique, 1966/1, pp. 101–112. Mikeladze-Andreasen N., 2010. O sztuce gruzińskiej czasów królowej Tamary: krzyż z Kacchi, Warsaw. Mitchell S., 1982. ‘The life of Saint Theodotus of Ancyra’, in AS, 32, pp. 93–113.

Keskin E., 2015. Küçük Asya’da Kutsal Kent Euchaita ve Bizans Dönemi Taş Eserleri – The Holy City Euchatia

335

Philipp Niewöhner Mundell Mango M., 1986. Silver from Early Byzantium: the Kaper Koraon and Related Treasures, Baltimore.

Russo E., 2011. ‘I capitelli del quadriportico di S. Giovanni a Efeso’, in Bizantinistica, 13, pp. 23–60.

Niewöhner P., 2007. Aizanoi, Dokimion und Anatolien, Wiesbaden.

Sayar M. H., 2004. ‘Kilikya Yüzey Araştırmaları 2003’, in AST, 22/2, pp. 219–228.

Niewöhner P., 2013a. ‘Bronze Age hüyüks, Iron Age hill top forts, Roman poleis, and Byzantine pilgrimage in Germia and its vicinity’, in AS, 63, pp. 97–136.

Schrade B., 2004. ‘Ad crucem. Zum Jerusalemer Ursprung der georgischen Voraltarkreuze anhand früher Beispiele aus Svanetien’, in R. Gordeziani (ed.), ქრისტიანობის 20 საუკუნე საქართველოში – 20 Centuries of Christianity in Georgia, Tbilisi, pp. 308–326.

Niewöhner P., 2013b. ‘Phrygian marble and stonemasonry as markers of regional distinctiveness in Late Antiquity’, in P. Thonemann (ed.), Roman Phrygia. Culture and Society, Cambridge (Greek Culture in the Roman World), pp. 215–248.

Taft R. F., 1992. The Byzantine Rite. A Short History, Collegeville (MN). Taft R. F., 1996. ‘Byzantine communion spoons’, in DOP, 50, pp. 209–238.

Niewöhner P., 2014. ‘Historisch-topographische Überlegungen zum Trierer Prozessionselfenbein, dem Christusbild an der Chalke, Kaiserin Irenes Triumph im Bilderstreit und der Euphemiakirche am Hippodrom’, in Millennium, 11, pp. 261–288.

Talbot A. M., Kazhdan A. P., 1991. ‘Athos, Mount’, in Kazhdan 1991, pp. 224–226. Talbot A.-M., 2002. ‘Byzantine monastic horticulture. The textual evidence’, in A. R. Littlewood, H. Maguire, J. Wolschke-Bulmahn (eds.), Byzantine Garden Culture, Washington, pp. 37–67.

Niewöhner P., 2016. Die byzantinischen Basiliken von Milet, Berlin. Öğüt-Polat S., Şahin S., 1986. ‘Katalog der bithynischen Inschriften im Archäologischen Museum von Istanbul’, in Epigraphica Anatolica, 8, pp. 109–128.

Tchakerian S., 2016. ‘Toward a detailed typology. Foursided stelae in Early Christian South Caucasus’, in I. Foletti, E. Thunø (eds.), The Medieval South Caucasus. Artistic Cultures of Albania, Armenia and Georgia, Turnhout (Convivium. Supplementum, 1), pp. 124–143.

Parenti S., 2010. ‘Towards a regional history of the Byzantine euchology of the sacraments’, in Ecclesia Orans, 27, pp. 109–121.

Terry A., 1988. ‘The sculpture at the Cathedral of Eufrasius in Poreč’, in DOP, 42, pp. 13–64.

Peschlow U., 1993. ‘Byzantinische Mörser’, in Ist. Mitt., 43, pp. 487–493.

Troianos S. N., 2000. ‘Die Einweihung und Entweihung der Kirchengebäude nach orthodoxem Kirchenrecht’, in C. Scholz, G. Makris (eds.), ΠΟΛΥΠΛΕΥΡΟΣ ΝΟΥΣ. Miscellanea für Peter Schreiner zu seinem 60. Geburtstag, Munich (Byzantinisches Archiv, 19), pp. 382–394.

Peschlow U., 2006. ‘Altar und Reliquie. Form und Nutzung des frühbyzantinischen Reliquienaltars in Konstantinopel’, in M. Altripp, C. Nauert (eds.), Architektur und Liturgie. Akten des Kolloquiums vom 25. bis 27. Juli 2003 in Greifswald, Wiesbaden (Spätantike – frühes Christentum – Byzanz. Reihe B: Studien und Perspektiven, 21), pp. 175–202.

Vanderheyde C., 2003. ‘Objets et éléments décoratifs en pierre issus d’Apamée’, in Bulletin des musées royaux d’art et d’histoire, 74, pp. 63–106.

Peschlow U., 2015. Ankara. Die bauarchäologischen Hinterlassenschaften aus römischer und byzantinischer Zeit, Vienna.

Vermoere M., 2004. Holocene Vegetation History in the Territory of Sagalassos (Southwest Turkey). A Palynological Approach, Turnhout (Studies in Eastern Mediterranean Archaeology, 6).

Peschlow U., 2017. ‘Patara’, in P. Niewöhner (ed.), The Archaeology of Byzantine Anatolia. From the End of Late Antiquity until the Coming of the Turks, New York, pp. 280–290. Peschlow U., Peschlow-Bindokat A., Wörrle M., 2002. ‘Die Sammlung Turan Beler in Kumbaba bei Şile (2)’, in Ist. Mitt., 52, pp. 429–522. Pralong A., 2003. ‘Matériel archéologique errant’, in B. Geyer, J. Lefort (eds.), La Bithynie au Moyen Âge, Paris (Réalités byzantines, 9), pp. 225–286. Restle M., 1979. Studien zur frühbyzantinischen Architektur Kappadokiens, Vienna (Österreichische Akademie der Wissenschaften. Philosophisch-historische Klasse. Denkschriften, 138; Veröffentlichungen der Kommission für die Tabula Imperii Byzantini, 3).

336

Part 4 Decoration and Small Objects

26 The Representations of Dolphins on the Inlaid opus sectile Panels above the Imperial Door in the Church of Hagia Sophia in Constantinople and their Meaning Anđela Gavrilović* Универзитет у Београду / University of Belgrade (Serbia) This chapter deals with two opus sectile panels placed on the west wall above the Imperial door in the Church of Hagia Sophia in Constantinople (532–537). They bear representations of eight dolphins devouring octopuses. We argue that they do not represent a simple ornament of the building but, on the contrary, bear deep symbolism, and have been placed in this strategic, highly significant position in the building due to their meaning. Our research has led to the conclusion that the motif of a dolphin devouring an octopus in Hagia Sophia represents a marine variant of the theme of calcatio, serving as a powerful Christian apotropaic and having the function to repel and ‘draw away’ threatening evil forces, namely evil demons and Satan. Its strongly imperial connotations are also covered in the paper. L’article traite de deux panneaux d’opus sectile placés sur le mur ouest, au-dessus de la porte impériale de l’église Sainte-Sophie de Constantinople (532–537). Ils portent des représentations de huit dauphins dévorant des pieuvres. Nous soutenons dans cet article qu’ils ne représentent pas un simple ornement du bâtiment, mais qu’ils sont porteurs d’un symbolisme profond et qu’ils sont placés en un point stratégiquement important et hautement signifiatif du bâtiment en raison de ce symbolisme. On conclut que le motif du dauphin dévorant une pieuvre est à Sainte-Sophie une variante marine du thème de la calcatio, une puissante image apotropaïque chrétienne dont la fonction est de repousser et d’éloigner les forces du mal, c’est-à-dire les démons et Satan. L’article traite aussi des fortes connotations impériales de ce motif. Der Aufsatz befasst sich mit zwei opus sectile Tafeln, die an der Westwand über dem Kaiserportal der Hagia  Sophia in Konstantinopel (532–537) angebracht sind, und acht, je einen Oktopus verschlingenden, Delfinen abbilden. Hier wird die Ansicht vertreten, dass sie nicht bloß als Ornamentelemente des Gebäudes anzusehen sind. Vielmehr haben sie eine tiefere Symbolik und wurden aufgrund dieser Bedeutung an dieser strategisch wichtigen Stelle angebracht. In der Studie wird gefolgert, dass das in der Hagia Sophia abgebildete Motiv des Delfins, der einen Oktopus verschlingt, eine maritime Variante des Themas ‘calcatio’ darstellt und somit als mächtiger christlicher Apotropaion gegen drohende böse Mächte, nämlich gegen böse Dämonen und Satan, diente. Die starken imperialen Konnotationen des Motivs werden ebenfalls im Artikel behandelt. L’articolo tratta di due pannelli in opus sectile posti sulla parete ovest al di sopra della Porta imperiale nella chiesa di Santa Sofia a Costantinopoli (532-537), che raffigurano otto delfini intenti a divorare altrettanti polipi (fig. 1). Nell’articolo si sostiene che essi non rappresentano un semplice ornamento dell’edificio ma che, al contrario, rivestono un profondo significato simbolico e perciò sono collocati in questo punto strategico e molto importante dell’edificio. Si conclude che il tema dei delfini che divorano i polipi in Santa Sofia rappresenta una variazione marina del tema della ‘calcatio’, con forte valenza apotropaica cristiana e ha la funzione di respingere e allontanare le forze del male, cioè i demoni del male e Satana. Nell’articolo vengono anche analizzate le connotazioni in chiave imperiale della raffigurazione.

* This chapter was written as a result of research conducted within the project Serbian Medieval Art and Its European Context (no 177036) of the Ministry of Education, Science and Technological Development of

the Republic of Serbia, and it was delivered in an abbreviated version on the occasion of the 23rd International Congress of Byzantine Studies held in Belgrade between 22nd and 27th of August 2016.

339

Anđela Gavrilović

Fig. 1. Dolphins chasing and devouring octopuses, Hagia Sophia, Constantinople, naos, west wall, 532–537 AD (D. Osseman).

Guidobaldi notices the symmetrical configuration of the elements in the opus sectile panels above the Imperial Door and the ‘great fineness of execution’ of the dolphin panels, considering the opinions of Cyril Mango and Kristin Ann Kelly.4

In the Church of Hagia Sophia in Constantinople (532– 537), on the west wall above the Imperial door and on the other side of the famous mosaic featuring Byzantine Emperor Leo the Wise (886–912), two opus sectile panels bear representations of eight dolphins chasing or devouring octopuses around two porphyry discs (fig. 1). These motifs of dolphins have drawn the attention of very few scholars, but are minutely described by Paul Underwood in his report on the works in Hagia Sophia in 1958.1 In her doctoral thesis on opus sectile, Kristin Ann Kelly studies the motifs of dolphins in Hagia Sophia on the west wall above the Imperial door and in the south nave.2 She analyses the motif of the dolphin in Greco-Roman art in its pagan milieu and in its Christian context in general, quoting examples. She expresses the opinion that the presence of dolphins at Hagia Sophia in Constantinople at both locations can be associated with the emperor, and that the dolphin, the trident and the anchor all symbolise Christ’s crucifixion. Although her conclusions are correct, when it comes to the Church of Hagia Sophia, they are still very general. Besides, she does not mention the octopuses at all. According to Cyril Mango, the panels of inlay (with paired dolphins) may represent ‘older, re-used pieces’, and the arrangement of the marble revetment on the west wall as a whole may be Justinianic.3 Alessandra Guiglia

As we can see from this summary, the meaning of the motifs of dolphins chasing or devouring octopuses and the reasons for their positioning above the Imperial door have remained unclear and have not been properly covered by scholars. Therefore, we will focus on them here. Judging from preserved examples, representations of dolphins were very numerous in ancient Greek and Roman art and in the art of the Early Christian period, while Justinian’s rule (527–565) appears to be the last period in which dolphins regularly occur in Byzantine art. From that time on, they almost disappear, or appear mainly in miniature paintings as a part of furniture ‘decoration’ following traditional ancient patterns of depicting this motif. In ancient and Early Byzantine art, dolphins could be represented in different iconographical contexts: as independent motifs in the scope of larger units, or in semantic connection to other motifs in the scope of larger units. Thus, they could be depicted swimming in pairs or independently; ridden by Amor; wrapped around an anchor

1 Lethaby, Swainson 1894, pp. 243–246, fig. 49; Dalton 1911, p. 698; Underwood 1960, p. 208, fig. 4; Majeska 1971, pp. 290–291, pl. I–II; Rickert 1998, p.172. 2 Kelly 1987, pp.175–187. 3 Mango 1997, p. 66.

4

340

Guiglia Guidobaldi 2007, pp. 165–166.

The Representations of Dolphins on the Inlaid opus sectile Panels above the Imperial Door in the Church... or a trident; next to an anchor, a trident, a cross, a shell, Christ’s bust, an apostle’s bust, Α and Ω, or a Chi and Rho monogram with cross; or they could – in a stylised manner and symbolically – form various floral ‘ornaments’.5 Sometimes they were depicted next to octopuses or squid, or they were depicted chasing or devouring octopuses, as is the case with their representation above the Imperial door in Hagia Sophia in Constantinople.6 Sometimes dolphins are depicted in pairs, of which one dolphin is devouring an octopus and the other a fish, which is an iconographic variant of the same motif.7

the background of the marble panels is whitish with purple veining, as is the inner ring encompassing the porphyry discs. Although dolphins are very frequently represented in ancient art, the motif of a dolphin devouring an octopus (or another cephalopod) occurs quite rarely in ancient and even more rarely in Byzantine art. Together with the individual motif of the dolphin, the representation of dolphins devouring octopuses can be traced back to the distant past. This theme occurs as early as Minoan art,9 and from the Christian period it can be traced back to catacomb paintings (beginning of the third century AD).10 The motif is found on ancient sarcophagi (the crypt of St Maximin, Trier).11 One is surprised by the relatively large number of depicted and preserved representations of dolphins devouring octopuses on the floor mosaics in ancient Ostia.12 This circumstance can be explained firstly by the town’s location by the sea, and secondly by the fact that it represented the central port of the Roman Empire, which gave Rome access to the Mediterranean, and was thus the main port of the ancient world. This fact is not underscored without a reason. The vicinity of imperial Rome as the capital of the Empire and the central source of artistic inspiration of the highest level speaks to the significance of this motif and the iconographic subtlety of the representations cited, as well as the quality of the execution of the mosaics and other themes depicted on them. Despite the factors mentioned above, one is surprised by the relatively large number of representations depicted, since they are more generally very rare, although continuously present from the Minoan period onwards. One should especially note the presence of this motif in the Church of Santa Constanza in Rome, whose patron was Constantia, the daughter of Constantine the Great (324–337).13 That circumstance leads to the assumption that the motifs of dolphins eating octopuses could have already been present in the Church of Hagia Sophia on the same spot at the time of its founding under Emperors Constantine the Great and Constantius (324–337; 337– 361). Apart from the representations in Santa Constanza in Rome and in Hagia Sophia in Constantinople (532–537), in the Middle Ages we find the motif of a dolphin eating an octopus only on the ring of Adhemar, French bishop of Angoulême (1076–1101).14

The two opus sectile panels with representations of dolphins in Hagia Sophia are found immediately above the bronze lintel of the Imperial doors. A large piece of verde antique marble, surmounted by an opus sectile panel featuring the aedicule housing the jewelled cross, now stands between the two panels with dolphins. Both panels depict four dolphins heraldically arranged in pairs above and below the central discs of red porphyry. The composition of the plaques with dolphins is very elaborately designed, and the position of the dolphins, the whole composition of the panels, as well as the adjacent iconographical details, most directly and in multiple ways rely on the artistic tradition of ancient Greece and Rome. Here we have in mind the dolphins’ features, their colours, the shapes of their head and tail, their poses, and adjacent iconographical motifs (those of a trident and a plant or an anchor), as well as the specific iconographical context in which dolphins are depicted chasing and devouring cephalopods. In both panels with a yellow background, green dolphins are represented in pairs above and below the porphyry discs surrounded by two rings. They are heraldically arranged and depicted with their tails tied to the trident with richly inlaid ribbon on the right panel, and to a plant or an anchor in the left panel.8 The dolphins are depicted in a stylised form, following the structure of the disc; the heads of the ones above are facing down, and the heads of the ones below are facing up. The caudal fins are tripartite and highly stylised, resembling floral leaves. The dolphins have dorsal and ventral fins, pointed white teeth, and large eyes, behind which small stylised fins emerge. Each dolphin is represented devouring or preparing to devour one octopus. Thus, octopuses are also depicted in pairs. They are shown in the middle of the panels between the dolphins’ beaks, with their oval white bodies and tentacles. Below the upper and above the lower pairs of dolphins,

Due to the fact that the dolphins are not mentioned in Holy Scripture, they play only a minor role as symbols in the works of the holy fathers,15 who are as a rule transmitting data on dolphins already known from ancient authors.16

5 Compare for example Dölger 1922, pl. LXXVI.1; Deichmann 1958, fig. 311–313; Germain 1973, pl. IV.8, XI.22, XIV.27, XX.56, XXXI.79, LII.161, LIII.162, LIV.162 and LXVIII.193; Dunbabin 1978, pl. II.3, VIII.16, IX.17, XXIV.59, XXXVI.93, XLVIII.122, XLIX.124, L.126, LVII.145–146, LVIII.147 and LXXIII.188; Lidov 2013, fig. 25; Ringbom 2003, fig. 8 and 11; Ringbom 2018, pp. 115–145. 6 See Germain 1973, pl. XXII.58; Dunbabin 1978, pl. XXXVI.93, L.129, LVI.142–143, LVIII.148–149, LIX.151 amd LXI.154. 7 Compare Wilpert 1932, pl. 244/2. 8 Other differences exist between these panels as well; compare Underwood 1960, p. 208.

Evans 1928, p. 502, fig. 306. Wilpert 1903, pl. 49. 11 Wilpert 1932, pl. 244/2. 12 Compare, for example, Becatti 1961, pl. XCVII/359, CXLIV/276, CXLV/211, CLXI/45, CLIV/271, CLXX/271, 361 and 211. 13 Ringbom 2003, p. 25, fig.14. 14 De Rossi 1870, pp. 49–73, pl. IV. n. 6; Leclercq 1920, p. 292, fig. 3615. 15 Wehrhahn-Stauch 1968, p. 503. 16 See for example PG, 29, col. 152 A; ibid., 30, 21 A–B; ibid., 36, col. 618 D; ibid., 37, cols. 1507 А and 1538 A; ibid., 44, col. 265 B. 9

10

341

Anđela Gavrilović images were the holy rider, Nemesis and the Eye itself.22 The dolphin devouring an octopus actually represented a marine variant of the iconographic theme calcatio, the image of the prostrate enemy trampled underfoot, which served as a sign of a victory.23 Thus, the plaques with dolphins devouring octopuses were set in Hagia Sophia above the Imperial door as powerful apotropaia, having the same function as apotropaic inscriptions and images, to repel and ‘draw away’ (ἀποτροπεῖν) the threatening evil forces, namely evil demons and Satan. The position of the dolphins next to a trident and floral ornaments and below the image of the Holy Cross would reinforce their semantic connection to the Holy Cross and emphasise their apotropaic and prophylactic function. In such a way, analogies for the meaning of the motifs of dolphins devouring octopuses and their position in the church can be found, for example, in the mosaic featuring Christ treading upon a lion and a snake over the doorway in the Archbishop’s Chapel in Ravenna (beginning of the sixth century AD) or over the gate of the city of Ravenna, as can be seen in the mosaic of St Apollinare Nuovo in the same city (first quarter of the sixth century AD).24

These data are generally very brief. As specific written sources concerning the meaning of the motif of a dolphin eating an octopus have not been preserved, we will focus on the rare, relevant preserved visual analogies in order to decipher that meaning. We believe that in any case the dolphins in Hagia Sophia are not mere ornaments and space-fillers, as has been suggested in previous literature.17 Since the dolphin was not a mere ornament in ancient art, since there exists a continuity in representing the motif of a dolphin eating an octopus, and since Hagia Sophia represents the most significant and the greatest church of the Byzantine capital, it would be very hard to believe that dolphins were represented there without reference to a wider context. Besides, they are represented in a very prominent place in the church – above the main, central, Imperial door connecting the naos with the narthex. Among all the representations of dolphins eating octopuses, there is one mosaic from Ostia bearing an inscription that is very significant in the context of the meaning of our specific dolphin image.18 The presence of an inscription makes it unique in Greek, Roman, Late Antique and Byzantine art. On the fragment of this floor mosaic on the south side of the decumanus in a fishmonger’s shop, above and below the dolphin with an octopus in its beak, runs the inscription: Inbide calco te (‘Envious one, I tread on you’), threatening envy and envious ones with an unpleasant fate.19 Such inscriptions are frequently placed at doorways or on thresholds, and have a very precise aim: to exclude from the building the harmful manifestations of envy, perceived as the greatest evil, and in particular the power of the Evil Eye – βάσκανος ὀφθαλμός or oculus invidiosus.20

As has already been noted, the decorations on the west wall of the nave in Hagia Sophia are repeated nowhere else in the building.25 This decoration with dolphins has been executed exactly between the Imperial door and the Imperial gynaeceum, which brings it into close connection with the Emperor and his family. The use of the central west door in Hagia Sophia in Constantinople was restricted to the imperial entourage on special occasions and was therefore held in particular reverence.26 The imperial nature of the decoration of this section of the wall is indicated by several means: by the central element of the two pairs of opus sectile panels – discs of red porphyry, the traditional imperial stone (rotae porfiricae), as well as by the dolphins in the lower pair of the opus sectile panels. The artist could have used some other variant on the same theme, but has intentionally chosen the one with a dolphin.27 The motif of the dolphin eating an octopus as an extra special marine variant of the calcatio theme is thus here also suggestive of the imperial triumph, emphasising in a specific way total victory over the enemy.28 One should also note that Hagia Sophia in Constantinople is positioned in the immediate vicinity of the sea, which makes the presence of this motif quite natural. The same goes for the other monuments where this motif occurs, sometimes even in particularly large numbers.29

In Hagia Sophia, the artist has transferred the image with the dolphin from the floor – where it is found in Ostia – to the space above the Imperial door, depriving it of the inscription, obviously well-known to the Emperors, learned people and believers who entered the Great Church and understood the visual message of the panels. The artist also increased the number of dolphins, featuring not one, but eight of them, obviously due to the fact that dolphins were symbolically related to the image surmounted by the cross which once stood between the panels with dolphins.21 The dolphin plaques in Hagia Sophia, like the mosaic in Ostia, are set at the entrance, and to be observed from inside the building. No matter how rare the preserved motif of a dolphin devouring an octopus may be, it is certain that this motif entered the range of familiar and, to judge by its presence in Hagia Sophia, obviously highly sophisticated apotropaic images with specific connotations. Other similar apotropaic

Dauterman Maguire et al. 1989, pp. 22–28; Dunbabin 1991, p. 34. On the term and ritual of calcatio and on this iconographic theme, cf. Dinkler-Von Schubert 1970, pp. 67–69; McCormick 1990, pp. 58.76, 97, 144 and 161ff.; Dunbabin 1991, pp. 30–34. 24 Compare Deichmann 1958, fig. 109 and 217; Deichmann 1974, pp. 145 and 203; cf. for example Kessler 2009, pp. 123–126, fig. 7–8; cf. also the previous footnote in this paper. 25 Majeska 1971, p. 290. 26 Majeska 1971, p. 291. 27 See above. 28 Cf. supra, n. 23. The presence of the motifs of dolphins in the imperial metatorion in the south-eastern exedra of the church would also speak to the imperial connotation and use of the motif; cf. Majeska 1971, p. 291. 29 Compare above. 22 23

For the dolphins in Hagia Sophia, see Lethaby, Swainson 1894, pp. 243–246, fig. 49; Dalton 1911, p. 698; Rickert 1998, p. 172; for other representations of dolphins, cf. for example Wulff 1914, p. 316; Leclercq 1920, p. 289; Stebbins 1929, pp. 129–130; Diez 1957, p. 678. 18 Becatti 1961, CLXX/361. 19 Dunbabin 1991, p .26. 20 Donceel-Voûte 2014, pp. 347–400. On the Evil Eye, see Elliott 2015– 17. 21 See below. 17

342

The Representations of Dolphins on the Inlaid opus sectile Panels above the Imperial Door in the Church... Concerning the meaning of dolphins and their function above the western door in Hagia Sophia, it is very important to point out that the image of the Saviour once formed part of the decoration of this space. Surmounted by the cross above the aedicule, it was once displayed between the panels with dolphins where the green marble slab is now visible. It was a copy of the icon of the Chalke Saviour mentioned in the writings of medieval Russian travellers to Constantinople.30 The icon itself had typical imperial connotations, placed at the entrance to the Imperial palace in Constantinople and being one of the city’s major religious symbols.31 It is difficult to say when this icon was placed on the west wall of the church, but based on the overall context of its decoration it may be assumed that an icon of Christ could have stood there as early as the time of Justinian (527–565), when the present church was built (532–537). The previous literature suggests that the Saviour icon could have been placed on the west wall ‘at the time of the first or second restorations of icons (787 or 843), but probably at the institution of the feast of the “Triumph of Orthodoxy” in 843’, signifying clearly the new imperial policy of protecting the cult of icons.32 The triumphal connotations of the representations of dolphins devouring octopuses must have perfectly fitted the idea of the institutionalisation of the feast of the ‘Triumph of Orthodoxy’. The preserved representations of dolphins chasing and devouring octopuses next to the figure of Christ on the ceiling of the Catacomb of Praetextatus (beginning of third century AD), as well as the dolphins surrounding Christ’s bust on the triumphal arch in the Church of San Vitale in Ravenna (546–547 AD), would support the assumption that the plaques with dolphins in Hagia Sophia could have stood next to the Saviour’s icon on its west wall at the time of Emperor Justinian, and maybe even at the time of Constantine the Great and Constantius (324–337; 337–361).33

to St Augustine, ‘He was not stripped of the flesh by any obligation to any power whatsoever, but willed His own death, for He who could not die unless He willed doubtless died because He willed; and therefore He openly exposed the principalities and the powers, confidently triumphing over them in Himself.’35 Cyril of Jerusalem, in praising the power of the cross, says that the cross is a terror to devils, for He ‘displayed them openly, leading them away in triumph by the force of it’.36 Alluding to this, Eusebius of Alexandria ends one of his homilies with the observation that ‘you have a phylactery which brings victory over the βασκανία of the devil in everything, His Precious Cross; and the devil can do to you no harm; the power of the Holy Cross is invincible, through our God and Lord, Jesus Christ.’37 Thus, when considering the meaning of the representations of dolphins above the Imperial door, one may firmly say that their position and meaning fits the decoration on the west wall in meticulous detail. One may also observe that the motifs of dolphins eating octopuses above the Imperial door are in line with the words of imperial acclamations and Byzantine attitudes about dangerous supernatural powers threatening the emperor and the Empire itself that are reflected in imperial ceremonies. Namely, in imperial acclamations it is said: ‘… May envy be absent from his reign’; ‘May God protect your power. May God grant peace to your reign …’; or ‘… May phthonos be far away from Your majesty.’38 Such acclamations, containing wishes for divine protection, have in themselves a form of apotropaic formulas that forbid an undesired event. Given the position of the dolphins in the main church of the Byzantine Empire, it is also important to note that in the coronation ceremonies the protective acclamations were directed towards the citizenry of Constantinople and the entire Empire governed by the emperor. The newly chosen emperor was regarded as a blessing for the Empire – by his pious life he was protecting the Empire threatened by phthonos as well. 39 One should here be reminded as well that in the south nave of Hagia Sophia, two panels with heraldically arranged pairs of dolphins were also found: one with four dolphins and a trident placed on the southeast pier and one with four dolphins chasing fish placed on the opposite half-pier.40 The presence of dolphins, most probably bearing a meaning and role similar to that of the dolphins on the west wall above the Imperial door, strongly supports the assumption that the metatorion – the imperial chamber – was located in the area where the

Taking into account what has been said above, the marine iconographic theme of calcatio above the entrance to the Church of Hagia Sophia, dedicated to Christ’s Holy Wisdom, simultaneously alludes to Christ’s total victory over Satan, and – in the context of its imperial nature and of the decoration of the west wall – to Christ’s triumphal procession, according to the following verses: ‘Therefore it says, when He (Jesus Christ) ascended on high (to the third heaven), He (Jesus Christ) led captive an army of captives’ (Ephesians 4:8) and ‘When He (God the Father) disarmed the (angelic) rulers and (angelic) authorities, He (God the Father) made public display of them (Satan’s angelic armies), having trampled over them through Him (Jesus Christ, by the cross) [Colossians 2:15].’ Specifically, Paul wrote Ephesians 4:8 having seen Jesus’s everlasting triumph in battle over Satan and death as the picture of a victorious king who has overthrown his foes.34 According

PL, 42, col. 899. PG, 33, col. 816 A–B; cf. also ibid., 62, col. 340. 37 PG, 86/1, col. 356 D; on the Cross as apotropaic when placed above the door in the Early Christian period, cf. Dauterman Maguire et al. 1989, pp. 18–22. 38 Hinterberger 2013, pp. 52–53 and 59 (with sources and bibliography). 39 Hinterberger 2013, pp. 59–61. 40 Cf. supra, n. 28, as well as Matthews, Kelly 1987, pp. 178–179, 184–185, pl. 257–258; for the panel with dolphins and tridents on the south-east pier, see also Mainstone 1988, p. 184, fig. 211; for the panel with dolphins chasing fishes on the opposite half-pier, see also http:// atom.doaks.org/atom/index.php/southeast-exedra-looking-southwest-tosoutheast-buttress-and-south-gallery (accessed: 17 March 2018). 35 36

Majeska 1971, pp. 284–295; Majeska 1984, pp. 136–137, 207 and 209–212ff. 31 Cameron, Herrin 1984, p. 175. 32 Majeska 1971, p. 293. 33 Compare Wilpert 1903, pl. 49; Deichmann 1958, fig. 311. 34 Phillips 1993, pp. 115–117. 30

343

Anđela Gavrilović Cutler A., 1991. ‘Metatorion’, in A. Kazdhan (dir.), Oxford Dictionary of Byzantium, New York/Oxford, p. 1353.

dolphins are positioned.41 It was the room reserved for emperors. Although the accounts regarding the function of this chamber vary, one can be sure that it was meant for imperial use, which can be supported by the presence of the dolphin motif.42

Dalton O., 1911. Byzantine Art and Archaeology, Oxford. Dauterman Maguire E., Maguire H. P., Duncan-Flowers M. J., 1989. Art and Holy Powers in the Early Christian House. Exhibition, Krannert Art Museum of the University of Illinois at Urbana-Champaign, 25 August to 1 October 1989, and Kesley Museum of Archaeology of the University of Michigan in Ann Arbor, 27 October 1989 to 29 April 1990, Champaign/Urbana/Chicago.

The most learned Byzantine spiritual men, as artistic advisers of the emperors, used the image of dolphins devouring octopuses as a refined, marine iconographic theme of calcatio to indicate the image of total victory over the enemy and to allude to Christ’s victory over phthonos/Satan. This originally pagan motif was accepted by Christianity as the nuances of its meaning fit Christian purposes. In Hagia Sophia it served as a powerful apotropaic, meant to ‘drive away’ threatening evil forces from the holy place itself, from the life of the Emperors who were crowned in it and from the life of the entire Empire governed by the emperor. It stood on the crossroads between the ancient, Late Antique and Early Byzantine periods, being in its form and use much closer to the initial centuries of the Byzantine Empire than the Middle Byzantine period. It is highly probable that the panels with dolphins placed above the Imperial door at the time of Justinian between 532 and 537 AD represent older, re-used pieces, as previously supposed.43 We believe that these panels were spared the fate that befell the icon of Christ, surviving till the present day due to the fact that they were not explicitly connected to the image of Christ by their appearance. As iconoclastic rulers shared the same ideas about the notion of phthonos, they left the motifs of dolphins unharmed to serve their own purpose.44

Deichmann W. F., 1958. Ravenna Hauptstadt des spätantiken Abendlandes, [III], Wiesbaden. Deichmann W. F., 1974. Ravenna, Hauptstadt des spätantiken Abendlandes. Kommentar, II/1, Wiesbaden. De Rossi B. G., 1870. ‘L’anello trovato nel sepolcro di Ademaro Vescovo di Angoulême ed il delfino simbolo di Cristo Salvatore’, in Bolletino di archeologia Cristiana. Ser. 2, 1, pp. 49–73. Diez E., 1957. ‘Delphin’, in RAC, III, cols. 667–682. Dinkler-Von Schubert E., 1970. ‘Fußtritt’, in LCI, cols. 67–69. Dölger F., 1922. ΙΧΘΥΣ. Der heilige Fisch in den antiken Christentum, III, Münster. Doncel-Voûte P., 2014. ‘Barrer la route au Malin: Une typologie des stratégies utilisées. Images et signes à fonctionnement sécuritaire sur support fixe dans l’Antiquité tardive’, in Aitken B. E., Fossey J. M. (eds.), The Levant, Crossroads of Late Antiquity. History, Religion and Archaeology – Le Levant, carrefour de l’Antiquité tardive. Histoire, religion et archéologie, Leiden/Boston (Monographies en archéologie et histoire classiques de l’Université McGill, 22), pp. 347–400.

Bibliography Abbreviations: LCI = E. Kirschbaum, W. Braunfels (ed.), Lexikon der christlichen Ikonographie, Freiburg im Breisgau, 1968–76.

Dunbabin K. M. D., 1978. The Mosaics of Roman North Africa. Studies in Iconography and Patronage, Oxford (Oxford Monographs on Classical Archaeology).

RAC = Klauser Th., Dassmann E., Schöligen G. (ed.), since 1950. Reallexikon für Antike und Christentum, Sachwörterbuch zur Auseinandersetzung des Christentums mit der antiken Welt, Stuttgart.

Dunbabin K. M. D., 1991. ‘Inbide, calco te … Trampling upon the Envious’, in E. Dassmann, K. Thraede (eds.), Tesserae. Festschrift für Josef Engemann, Münster (Jahrbuch für Antike und Christentum. Ergänzungsband, 18), pp. 26–35.

PG = Migne J.-P. (ed.), 1857–66. Patrologiae cursus completus. Series Graecae, Paris. PL = Migne J.-P. (ed.), 1844–55 and 1862–65. Patrologiae cursus completus. Series Latinae, Paris.

Elliott H. J., 2015–17. Beware the Evil Eye, I–IV, Eugene (OR).

Becatti G., 1961. Scavi di Ostia, IV, Rome.

Evans A., 1928. The Palace of Minos at Knossos. A Comparative Study of the Successive Stages of the Early Cretan Civilization as Illustrated by the Discoveries, II/2, London.

Cameron A., Herrin J. (ed. and transl.), 1984. Constantinople in the Early Eighth Century: The Parastaseis Syntomoi Chronikai, Leiden (Columbia Studies in the Classical Tradition, 10).

Germain S., 1973. Les mosaïques de Timgad. Études descriptive et analytique, 2nd ed., Paris (Études d’antiquités africaines).

For the metatorion, see Cutler 1991, p. 1353. Cutler 1991, p. 1353. 43 Cf. supra, n. 3. 44 Cf. the article cited supra, n. 39. 41

Guiglia Guidobaldi, A., 2007. ‘I marmi di Giustiniano: sectilia parietali nella Santa Sofia di Costantinopoli’,

42

344

The Representations of Dolphins on the Inlaid opus sectile Panels above the Imperial Door in the Church... in A. C. Quintavalle (ed.), Medioevo mediterraneo: l’Occidente, Bisanzio e l’Islam, Atti del Convegno internazionale di studi, Parma, 21–25 settembre 2004, Milano, 2007 (I Convegni di Parma, 7), pp. 160–174.

A. Landen (eds.), Ecce leones! Om djur och odjur i bildkonsten, Lund, pp. 115–145. Sabeti A. S. J., 1890. ‘The dolphin in Christian symbolism’, in American Ecclesiastical Review, 2, pp. 112–120.

Hinterberger M., 2013. ‘Phthonos: A pagan relic in Byzantine Imperial acclamations?’, in A. Beihammer, St. Constantinou, M. Parani (eds.), Court Ceremonies and Rituals of Power in Byzantium and the Medieval Mediterranean. Comparative Perspectives, Leiden/ Boston (The Medieval Mediterranean, 98), pp. 51–66.

Stebbins B. E., 1929. The Dolphin in the Literature and Art of Greece and Rome. A Dissertation Submitted to the Board of University Studies of the Johns Hopkins University in Conformity with the Requirements for the Degree of Doctor of Philosophy 1927, Menasha. Underwood A. P., 1960. ‘Notes on the work of the Byzantine Institute in Istanbul: 1957–1959’, in Dumbarton Oaks Papers, 14, pp. 205–222.

Kelly A. K., 1987. Motifs in opus sectile and its Painted Imitation from the Tetrarchy to Justinian, PhD Dissertation, Columbia University.

Wehrhahn-Stauch L., 1968. ‘Delphin’, in LCI, I, pp. 503– 504.

Kessler L. K., 2009. ‘Christ the Magic Dragon’, in Gesta, 48/2, pp. 119–134.

Wilpert G., 1932. I sarcofagi cristiani antichi, I/2, Rome (Monumenti dell’Anticà cristiana, 1/1).

Leclercq H., 1920. ‘Dauphin’, dans F. Cabrol, H. Leclercq (ed.), Dictionnaire d’archéologie chrétienne et de liturgie, IV/1, Paris, pp. 283–295.

Wilpert J., 1903. Die Malereien der Katakomben Roms. Tafelband, Freiburg im Breisgau.

Lethaby R. W., Swainson H., 1894. The Church of Sancta Sophia Constantinople. A Study of Byzantine Building, London/New York.

Wulff O., 1914. Altchristliche und byzantinische Kunst, I, Berlin/Neubabelsberg (Handbuch der Kunstwissenschaft).

Lidov K. A., 2013. ‘ “Сияющий диск и вращающийся храм”: икона света в византийской культуре’, in Византийский Временник – Byzantina Chronika, 72 (97), pp. 277–292. Mainstone J. R., 1988. Hagia Sophia. Architecture, Structure and Liturgy of Justinian’s Great Church, London. Majeska P. G., 1971. ‘The Image of the Chalke Savior in Saint Sophia’, in Byzantinoslavica, 32, pp. 284–295. Majeska P. G., 1984. Russian Travelers to Constantinople in the Fourteenth and Fifteenth Centuries, Washington (Dumbarton Oaks studies, 19). Mango C., 1997. Hagia Sophia. A Vision for Empires. Essays, Istanbul. McCormick M., 1990. Eternal Victory. Triumphal Rulership in Late Antiquity, Byzantium and the Early Medieval West, Cambridge/New York/Paris (Past and Present Publications, 24). Phillips J., 1993. Exploring Ephesians and Philippians. An Expository Commentary, Grand Rapids (MI) (The John Phillips Commentary Series). Rickert F., 1998. ‘Inkrustation’, in RAC, XVIII, cols. 160– 179. Ringbom Å., 2003. ‘Dolphins and mortar dating – Santa Constanza reconsidered’, in R. Suominen-Kokkonen (ed.), Songs of Ossian. Festschrift in Honour of Bo Ossian Lindberg, Helsinki (Taidehistoriallisia Tutkimuksia – Konsthistoriska Studier, 27), pp. 23–38. Ringbom Å., 2018. ‘On the dolphin in classical art. A symbol of light and the afterlife’, in L. Berggren,

345

27 Early Byzantine Silver Chalices from Viminacium: Parts of a Tableware Set or Church Vessels? Ivana Popović* Археолошки институт Београд / Institute of Archaeology, Belgrade (Serbia) Five silver footed chalices, four of them with lids (two lids went missing during the First World War) have, since their publication in 1903, become the subject of various attempts to solve the problems of their function and time of manufacture. Their dating in the existing literature spans the period from AD 250 to the fifth or sixth century, and the explanation of their purpose varies from votive gifts to pagan gods to liturgical vessels. The chalices were discovered in Kostolac (Viminacium), in a niche underneath the foundations of a building that was never explored. The chalices from Viminacium have hemispherical receptacles shaped into a rather deep bowl and a groove for the lid under the rim. They have a high foot of trumpet shape with a ring-like enlargement at the top. The lids are slightly concave, and have conical hoop-shaped handles. The closest analogues for these vessels are chalices from hoards discovered in Carthage and Canoscio. The hoard from Viminacium, which in addition to the footed chalices contained other silver objects, represents the treasure of a single man, a military or civilian official. While the objects were made in the second half of the fourth or at the beginning of the fifth century, the treasure was most probably hidden before the Hunnic destruction of the city in 441. The treasure from Viminacium consists of typologically atypical footed chalices with lids, possibly of local manufacture, like the other objects from this hoard, and thus indicates that the workshops in Byzantine Illyricum maintained close contacts with toreutic production centres in the western part of the empire. Cinq coupes sur pied en argent, quatre d’entre elles munies d’un couvercle (dont deux ont disparu durant la Première Guerre mondiale), ont fait l’objet de diverses études relatives à leur fonction et à leur fabrication, depuis leur publication en 1903. Dans la littérature existante, leur datation varie de 250 au VIe siècle et leur interprétation fonctionnnelle va des offrandes votives pour des dieux païens à des vases liturgiques. Les coupes ont été découvertes en 1899 à Kostolac (Viminacium), dans une niche au-dessus des fondations d’un édifice dont l’emplacement exact n’a jamais été déterminé. Ces coupes sont hémisphériques, avec un pied haut en forme de trompette et un couvercle concave muni d’une anse conique. Les exemples connus les plus proches sont des coupes appartenant à des trésors trouvés à Carthage et à Canoscio. Celui de Viminacium, qui comprenait d’autres objets en argent, était la propriété d’un riche citoyen. Alors que les objets ont été produits vers la fin du IVe ou au début du Ve siècle, le trésor a peut-être été dissimulé avant la destruction de la cité par les Huns en 441. Les coupes sur pied de Viminacium sont atypiques et possiblement de facture locale. Avec d’autres objets du trésor, elles indiquent que des liens commerciaux et artistiques étroits existaient entre les ateliers toreutiques de l’Illyricum byzantin et ceux des parties occidentales de l’Empire. Bereits seit ihrer Erstveröffentlichung im Jahre 1903 wird über Funktion und Herstellungszeit der Viminaciumer Silberbecher mit Fuß, davon vier mit Deckeln (heute sind nur zwei davon erhalten, zwei gingen während des Ersten Weltkrieges verloren) diskutiert. In der Forschung werden sie zwischen dem Jahr 250 und dem 6. Jh. eingeordnet, funktional spricht man von Votivgeschenken an heidnische Gottheiten bis hin zu liturgischen Gefäßen. Die Becher wurden 1899 in Kostolac/ Viminacium entdeckt, in einer Nische unter dem Fundament eines Gebäudes, dessen Standort unbekannt ist. Die Körper der Becher haben eine halbkugelförmige Form bzw. sind sie geformt wie eine tiefe Schale und haben eine Rille für die Deckel am Rand. Sie haben einen * This article is the result of the project Romanisation, Urbanisation and Transformation of Urban Centres of Civil, Military and Residential Character in Roman Provinces in the Territory of Serbia (no 177007),

funded by the Ministry of Education, Science and Technological Development of the Republic of Serbia.

347

Ivana Popović trompetenförmigen, nach oben hin ringförmig erweiterten Fuß. Die Deckel sind leicht konkav und haben konisch bügelförmige Griffe. Die nächste Parallele dieser Gefäße sind Becher aus Hortfunden aus Karthago und Canoscio. Der Hortfund aus Viminacium, der neben den Bechern auch weitere Silbergegenstände enthielt, ist ein Schatzfund, der einst wohl einem reichen Offizier oder zivilen Beamten gehörte. Obwohl die Gegenstände wohl erst während der zweiten Hälfte des 4. Jhs. oder zur Beginn des 5. Jhs. hergestellt worden sind, dürfte der Schatzfund vermutlich erst während der hunnischen Zerstörung der Stadt im Jahre 441 verborgen worden sein. Der Schatzfund von Viminacium zeigt eine typologisch untypische Fußform der Becher mit Deckel, wohl aus lokaler Herstellung, wie die weiteren Objekte auch. Sie verdeutlichen zudem, dass die byzantinisch-illyrischen Werkstätten enge Kontakte zu den Gefäßmanufakturen im westlichen Teil des Reiches unterhielten. Cinque calici argentei, con piede, quattro dei quali con coperchi (oggi ne sono rimasti soltanto due con coperchi, mentre due sono scomparsi durante la Prima Guerra mondiale), già dal momento in cui furono pubblicati per la prima volta nel 1903, divennero oggetto di varie interpretazioni, a riguardo della loro funzione e del periodo in cui furono prodotti. Nella letteratura esistente, la loro datazione comprende l’intervallo tra l’anno 250 e il VI secolo, e la loro funzione varia tra offerte votive alle divinità pagane e suppellettile liturgica. I calici furono scoperti nel 1899 a Kostolac (Viminacium), in una nicchia sotto le fondazioni di un edificio la cui posizione esatta non fu mai scoperta. I calici hanno la forma semisferica, con un piede in forma di tromba ed un coperchio concavo munito di una maniglia conica. Forme analoghe a quelle di questi calici si possono trovare nei tesori di Cartagine e di Canoscio. Il tesoro di Viminacium che, oltre ai calici, conteneva anche altri oggetti argentei, doveva essere proprietà di un cittadino ricco. Il tesoro fu realizzato alla fine del IV o all’inizio del V secolo, e fu probabilmente sepolto nel quarto decennio del V secolo, quando le città dell’Illirico correvano il pericolo di essere devastate dagli Unni. Tenendo presente il fatto che i calici con piede di Viminacium rappresentano una variazione atipica di questo gruppo di manufatti, è possibile che si trattasse di prodotti locali. Questi calici, come gli altri oggetti trovati a Viminacium, indicano legami forti tra le officine toreutiche dell’Illirico bizantino e quelle delle parti occidentali dell’Impero. Five silver chalices on a foot or stand from the National Museum in Belgrade (fig. 1 a–e), since the time of publishing in 1903,1 have become the subject of various approaches to solving the problems of their function and time of manufacture. Their dating in the existing literature spans the period from the year 2502 to the fifth or sixth century,3 and explanations of their purpose have varied from votive gifts to pagan gods4 to liturgical vessels.5 The chalices were discovered in Kostolac (Viminacium), in a niche underneath the foundations of some building, which was never explored. Together with the chalices were also discovered a cylindrical silver object decorated with a series of ellipsoid fields with motifs of impressed gems (fig. 6), and two small silver plates with representations of running animals and masks of Maenad and Satyr on the horizontal rims (fig. 4).6 These objects went missing during the evacuation of the treasures of the National Museum in Belgrade during the First World War.

slightly concave and have conical hoop-shaped handles. Despite being almost identical, they differ slightly in size. Three of them are 13 cm high, with a rim diameter of 17 cm, and foot diameter of 8 cm, while two other chalices are 12 cm high, with a rim diameter of 15.50 cm, and foot diameter of 7 cm. As was noticed already in the mid-1980s, the closest analogies for these vessels are chalices from a hoard discovered in Carthage (North Africa). At that time, the chalices from Viminacium were vaguely dated to the fourth century, with the remark that related vessel types are scattered over a vast territory from Nubia to North Africa during that entire century.7 It was also mentioned even then that related vessels were also recorded in a hoard from Canoscio in Umbria (Italy). Dating of the Viminacium chalices to the second half of the fourth century is accepted also in later texts.8 After more recent publications and a detailed analysis of the finds from Carthage9 and from Canoscio,10 it is also possible to draw more precise conclusions about the chalices from Viminacium.

The chalices from Viminacium have hemispherical receptacles shaped into a rather deep bowl, and a groove for the lid under the rim. They have a high foot of trumpet shape, with a ring-like enlargement at the top. The lids are

It is obvious already at first sight that the chalices from Viminacium and from Carthage (fig. 2) are of almost identical height (c. 13 cm), while those from the Canoscio treasure are somewhat shorter (10.8–8.4 cm). The

Vassits 1903, pp. 23–25. Vassits 1903, p. 30. 3 Kondić 1995, pp. 184–185. 4 Vassits 1903, p. 28. 5 Kondić 1995, p. 185. 6 Vassits 1903, p. 17. 1 2

Baratte 1975, p. 38. Popović 1994, p. 112–113, cat. 277–281. 9 Baratte et al. 2002, with quoted earlier literature. 10 Aimone 2015, with quoted earlier literature. 7 8

348

Early Byzantine Silver Chalices from Viminacium: Parts of a Tableware Set or Church Vessels?

Fig. 1 a–e. Silver chalices on a foot, two examples with lids, Viminacium (documentation of the National Museum in Belgrade).

receptacles of these analogous chalices are somewhat deeper than the Viminacium vessels, and the foot is conical, and thicker in comparison with the Viminacium specimens of trumpet shape. Typologically close to the chalices from Carthage and Canoscio are examples from the sites of Arten (Veneto), Reggio Emilia (south Italy), Canicattini Bagni (Sicily), from grave 3 at the site of Ballana (Sudan), as well as a few examples of Sassanid provenance. These vessels are dated to the fifth to sixth century, and the Persian specimens to the fifth to seventh century.11 Chalices from Carthage, Canoscio, Reggio

Emilia and Canicattini are of approximately the same size as the chalices from Viminacium. Nevertheless, on the basis of the shape of the receptacle and its decoration, the chalices are classified into three variants, while examples from Viminacium – despite belonging to the same group of vessels – reveal different typological characteristics and therefore represent a distinct variant of the footed chalices.12 In the hoard from Viminacium, there are five footed chalices. Four had lids, of which two are preserved

Fig. 2. Silver chalices on a foot with lid and the lid used as a shallow plate, Carthage Treasure, Louvre Museum (Baratte et al. 2002, fig. 28 and 31). 11

Baratte et al. 2002, pp. 40–46; Aimone 2015, pp. 132, 134.

12

349

Aimone 2015, pp. 129–130.

Ivana Popović handles, they could have been used as coasters or shallow plates (fig. 2). A dual function of some parts of tableware sets is confirmed sporadically also in the sets from the Early Empire.18

while two went missing during the First World War. Five chalices were also discovered in the Carthage treasure (three are in the British Museum in London, one is lost and one is housed in the Louvre Museum in Paris). The chalices from Viminacium with preserved lids weigh 795– 796 g, so that the total weight of all five specimens, with lids, would have been 3980 g, while the total weight of the chalices from Carthage – on the basis of dimensions of the examples from the British Museum – is 2344 g and, given the weight of the Louvre example, five chalices of that type would weigh 3172 g.13 Thus we came to the conclusion that the chalices from Viminacium were heavier than those from Carthage, of which one example was bigger and heavier than the others. In the treasure from Canoscio, there were also five footed chalices of various dimensions, also with lids.14 These data suggest the conclusion that footed chalices with lids were produced in groups of five, not always of identical size, and that a set of five chalices was part of a larger set of tableware. The hypothesis was proposed that the Viminacium chalices were almost certainly used for drinking, because they could be raised with one hand,15 while chalices in the treasures from Carthage and Canoscio could not have been lifted with one hand because of the thicker conical foot. Therefore, because two hands were needed, they were not suitable for drinking, but were vessels used for storing liquid foods or sauces,16 and the lids were needed to keep their contents warm. This would have been an innovation in Early Roman tableware sets, where vessels with lids have not been recorded.17 Taking into account, however, that under the rim of the Viminacium chalices was a groove for securing the lid, the question remains open as to whether these vessels were also suitable for drinking or whether they too were used only for storing warm liquid or halfliquid food. On the other hand, the lids of the chalices from all three hoards had a multi-functional character, because when turned upside down and placed on their conical

The contents of all three mentioned hoards are heterogeneous. Besides footed chalices and other silver vessels, the treasure from Carthage also contained gold jewellery, while in the treasure from Canoscio were various silver vessels and spoons in addition to the five footed chalices. The Viminacium hoard also contained two small silver plates, 16 cm in diameter, but lost today (fig. 4). Their horizontal rims were decorated with representations in relief of running animals, which were separated by Bacchic masks;19 that was a common type of decoration for vessels originating from the second half to the end of the fourth century. This type of decoration of the horizontal rims of small plates with diameters between 15.2 and 20.6 cm was recorded on two plates from the

Fig. 4. Small silver plate with decorated rim, Viminacium (Vassits 1903, fig. 19).

Fig. 3. Silver chalice on a foot (a) with the lid (b), Canoscio Treasure (Aimone 2015, fig. 65 and 67). Baratte et al. 2002, p. 40. Aimone 2015, pp. 82–86, no 8, 9–10, 12–13. 15 Baratte et al. 2002, p. 45. 16 Aimone 2015, p. 131. 17 Baratte et al. 2002, p. 45. 13 14

Baratte et al. 2002, p. 45. Vassits 1903, p. 26, fig. 19; Popović 1994, p. 358, cat. 338; 1997, pp. 83–84, fig. 10. 18 19

350

Early Byzantine Silver Chalices from Viminacium: Parts of a Tableware Set or Church Vessels?

Fig. 5. Small silver plate with decorated rim, Carthage Treasure, British Museum (Baratte et al. 2002, fig. 8).

Carthage treasure (fig. 5), on one example, probably from Šabac, and today in the Art History Museum in Vienna, as well as on five plates from the Mildenhall treasure.20 In the centre of the small plates from Viminacium was depicted a mask of a Maenad in relief, and on other examples at the same spot were vegetal or pastoral motifs, mythological or genre scenes. Besides the mentioned vessels, the Viminacium hoard included yet another cylindrical object, now lost, which we assume was part of an object which would also have stood on the table set for a meal. This silver cylinder, 39 cm high and 6.5 cm in diameter, was decorated with seven rows of ellipsoid fields within which were motifs usually depicted on gems: animals, birds, vegetal ornaments and symbols like an amphora, basket, snake and vase, dolphin and anchor, snake and anchor, and the like (fig. 6).21 When this hoard was published for the first time, this cylindrical object was explained as a box, i.e., an étui, where manuscripts or small instruments could have been kept.22 The ornamental design of this object finds an analogy only in the ornamental design of a silver amphora discovered in the sea near Porto Baratti (Tuscany). Its entire surface is decorated with ellipsoid fields with diverse mythological scenes and motifs in relief (fig. 7). The amphora was produced in the final decade of the fourth century, probably in one of the workshops in the East, and it reached the sea near Porto Baratti much later, probably at the time of the Crusades.23 The ornamental designs used on the amphora from Porto Baratti as well as on the cylinder from Viminacium reveal the aspiration to imitate decoration from earlier Hellenistic-Roman objects

Fig. 6. Silver cylindrical object decorated with pseudo-gems, Viminacium (Vassits 1903, fig. 18).

when, according to the writings of antique authors, various vessels and other objects were decorated with inlaid gems of semi-precious stones or coloured glass.24 In our opinion, the cylinder from Viminacium, to judge by its size, is not a box but a part of some object of which the top and foot are missing. It may have been the shaft of candelabrum with a receptacle shaped as a vase on its top.25 Candlesticks of that type appear together with tableware already in finds from the second half of the fourth century, as in the Kaiseraugst treasure,26 and they continued to be used in the Early Byzantine period as well, also as part of tableware sets.27 A candelabrum from Sadovec in Bulgaria, dated to the sixth century, is of approximately the same size (height

Baratte et al. 2002, p. 28. Vassits 1903, p. 25, fig. 18; Popović 1994, p. 358, cat. 339; 1997, pp. 87–90, fig. 16–17. 22 Vassits 1903, p. 25. 23 Arias 1986; 1987; Popović 1997, pp. 87–89, fig. 19; Baratte 2000, pp. 502–503, cat. 126. 20 21

Popović 1997, pp. 88–89. Popović 1997, fig. 18. 26 Cahn, Kaufmann-Heinimann 1984, cat. 42. 27 Vrom 2007, p. 204 24 25

351

Ivana Popović workshops, but also the result of various contacts between Mediterranean and Sassanid cultural circles.33 Examples from the Viminacium hoard, in contrast to the chalices mentioned above, had smoothly executed surfaces of receptacles and lids. A thinner, trumpet-shaped foot with ring-like ends made them more elegant than the North African and South Italian chalices. It is an open question whether the chalices from Viminacium, typologically distinguished as a distinct variant, are products of a local workshop. This possibility is also suggested by a marble funerary monument mentioning the freedman Eutychus faber argentarius, to whom the monument was erected by Refidius, a military official of the Fourth Flavian legion. The monument was found in Smederevo,34 a town on the Danube, around 20 km upstream from Viminacium. For construction of Smederevo in the medieval period, stone monuments brought from nearby Viminacium were used to a considerable extent. The name of the freedman who was producing silver objects, recorded on the funerary monument, is of Greek origin. He was in the service of a member of the Fourth Flavian legion, which was stationed in Singidunum. Origins in the Greek-speaking provinces of the Empire and status of a freedman in the service of a member of the armed forces are common characteristics for artisans who were engaged in silver working,35 and who, as the monuments reveal, carried out their activities in large urban centres along the Danube. The settling of veterans of the IV Flavian and VII Claudian legion around the military camp in Viminacium had become particularly intensive and was supported by the emperor after AD 238, when the city received the status of a colony and when its rapid economic development had begun.36 Therefore, it is not surprising that an official of the IV Flavian legion erected in that city a monument to the artisan who produced silver objects. We do not know the scope and quality of production of those local craftsmen, nor to what extent they had contacts with colleagues in other workshops, but they certainly produced silverware not only for military personnel but also for civilians, who need not have been exceptionally rich, but wanted to show off their status with silver vessels or to provide for their status in the future.37 Gold jewellery38 discovered in the graves of the Viminacium necropoleis had been used for the same purpose; of the very decorative wall-paintings that embellished the tombs, the most impressive is one depicting a young woman wearing gold jewellery typical of the Late Antique period.39 This is the evidence that there was a relatively affluent class of Viminacium citizens in the fourth century; among them we should look for the owner of silver objects, whom we discussed above.

Fig. 7. Silver amphora, Porto Baratti (Arias 1987).

38.5 cm, 6 cm in diameter)28 as the cylindrical object from Viminacium when the missing receptacle on the top is added. If the cylindrical object from Viminacium was in fact a box – an identification that cannot be confirmed today, as it is lost – it could have been used for keeping silver instruments for cleaning teeth and ears. Such boxes, as confirmed by the hoards of Late Antique silverware from Kaiseraugst,29 Hoxne30 and Vienna,31 in Gaul and Britannia, were common components of tableware sets. One example of such a dual-purpose instrument from an unknown site in Serbia is housed today in the National Museum in Belgrade.32 Analysis of footed chalices from the treasure at Carthage and from Canoscio has revealed that North African and South Italian vessels of this type are typologically very close, indicating the conclusion that they were products of one workshop or of a few production centres, which maintained close contacts. The receptacles of the chalices and the lid surfaces of the vessels from Carthage, Canoscio and Canicattini Bagni were decorated with shallow channels, which are considered to have been a decorative characteristic of west Mediterranean

Perhaps his identity is indicated by the inscriptions on some objects from this discovery. The inscription TVR,40 Aimone 2015, p. 134. Mirković 1986, pp. 121–122, no 93. 35 Mirković 1986, p. 122. 36 Mirković 1986, p. 59. 37 Baratte 1996, p. 79. 38 Popović 2001, cat. 13, 17, 24, 31–35, 39, 41 and 43. 39 Korać 1991, p. 118, fig. 11–12; 2005, pp. 169–171, fig. 2; 1997. 40 Vassits 1903, pp. 24–25, no XXI–XXIII. 33 34

Werner 1986, p. 14, fig. 2. Martin 1984, pp. 122–132, cat. 38–40. 30 Bland, Johns 1993, pp. 28, cat. 145–153. 31 Baratte et al. 1990, pp. 81–83, no 21. 32 Popović 1997, pp. 80–81, fig. 15. 28 29

352

Early Byzantine Silver Chalices from Viminacium: Parts of a Tableware Set or Church Vessels? denoting the owner of the silver objects, is engraved on the back side of the foot of two chalices and on the base of the cylindrical object. An identical inscription appears as a cognomen in the Spanish provinces and, as a gentile name, also appears once each in the Spanish provinces, Gaul and Germania, and four times in Dalmatia.41 The inscription should most probably be read as Turus. This name does not appear in the inscriptions from Moesia, but it is recorded four times in northern Italy, once in the Spanish provinces, and even 17 times in Dalmatia.42 On the other hand, on the back side of the foot of two chalices and on one lid from this find was engraved the inscription Matrona or Matro,43 leading to the conclusion that these were dedications to the goddess Dea Matrona, the Celtic Great Mother of the gods venerated in parts of Germania, eastern Gaul and North Italy, and whose cult the Roman army spread to other parts of the Empire.44 These data suggest the conclusion that the owner of the silverware set originated from the western provinces of the Empire, Dalmatia, North Italy, Gaul or Germania. He was in Viminacium holding a high military or civilian office, and for him objects of silver served as confirmation of his rank in society.

the fourth century.46 Given the date of manufacture of the amphora from Porto Baratti,47 cylindrical objects decorated with motifs of gem impressions were also made at approximately the same time. Therefore, we assume that the footed chalices may also be dated to the period around the end of the fourth and the very beginning of the fifth century, and we could consider them the prototypes of such vessels, since the most recent analyses have dated the production of somewhat analogous chalices from the hoard at Carthage to the end of the fifth or beginning of the sixth century,48 and that was also the date of manufacture of the chalices from Canoscio.49 All the hoards of silver objects mentioned had been deposited in moments of crisis and the imminent threat of war. But, while the treasures from Carthage and Canoscio were the result of many years of collecting valuables, the one from Carthage was the treasure gathered over the course of many decades by the Cresconii family, which had a prominent position in African society at the end of the fourth century;50 the hoard from Viminacium, by contrast, represents the treasure of one man, a military or civilian official, whose name was Turus, whose origin was in the western provinces of the Empire. The treasure from Viminacium was probably hidden before the Hunnic destruction of the city in 441, when Viminacium was razed to the ground, as recorded in written sources.51 Later material from the time of Gothic domination and Germanic settlement during the Early Byzantine restoration of the city does not indicate the existence of valuable objects of Roman provenance.52 Deposition of the treasure from Carthage is related to the invasion of the Vandals in the mid-fifth century but, as recent analyses revealed, objects – including also those from earlier periods – were sometimes hidden later, at the beginning of the sixth century.53 The deposition of Umbrian and South Italian silver chalices and other objects is related to the crisis provoked by the Ostrogothic conquests in 489– 493 and especially to the exceptionally unstable situation during the Gothic–Byzantine war between AD 535 and 554.54 So, different historical circumstances brought about the hiding of silver objects in different parts of the Empire. But the contents of those treasures bear witness that in the period from the fourth to the seventh century we can speak about a cultural koiné, which included not only the coasts of the Black and Mediterranean Seas55 but also the Balkan provinces as the close neighbourhood of that world. Nevertheless, cultural connections with the Roman-Mediterranean world had been interrupted in this part of the Empire in the mid-fifth century, after Hunnic destruction of the Balkan cities and the later short-lived dominance of the Goths and the Gepids. Despite the fact that the northern section of Moesia Prima became once

The treasure from Viminacium consists of typologically atypical footed chalices with lids, possibly of local manufacture, and two small plates with scenes of running animals and Bacchic masks on the horizontal rims; in contrast to the North African, British and Pannonian vessels of that type, the plates did not have a border consisting of silver beads, i.e., they did not have a beaded edge. It is not impossible that these plates are a local variant of vessels inspired by the shape and decoration of small plates produced in western workshops. The cylindrical object decorated with ellipsoid fields with impressed motifs from gems was by all appearances made in some eastern workshop, or its decoration was inspired by some object originating from those production centres. Therefore, the contents of the hoard indicate that the silver objects may have been products of different workshops. Nevertheless, they were all part of a tableware set, and nothing suggests that they had been in use in the Christian cult. The footed chalices with lids have their best analogies among finds from Carthage as well as those from South Italy that were also part of tableware sets. The cross engraved on the back side of one lid from Canoscio and one from Canicattini Bagni does not automatically mean that they were altar vessels, although we cannot exclude the possibility of their use for paraliturgical purposes, that is, their use for taking in the church bread and wine for the Eucharist.45 But, as no Christian symbols were recorded on the chalices from Viminacium, their use for secular purposes seems certain. Small plates with representations of running animals and Bacchic masks on the horizontal rims may be dated relatively reliably to the second half or the very end of

Cf. supra, n. 20. Cf. supra, n. 23. 48 Baratte et al., 2002, p. 109. 49 Aimone 2015, p. 163. 50 Baratte et al., 2002, p. 109. 51 Mirković 1986, pp. 25–27. 52 Ivanišević, Kazanski, Mastykova 2006. 53 Baratte et al., 2002, p. 109. 54 Aimone 2015, p. 171. 55 Bianchi Bandinelli 1965; Kitzinger 1969, Aimone 2015, p. 172. 46 47

OPEL IV, 2002, p. 133. OPEL IV, p. 134; Alföldy 1969, p. 315. 43 Vassits 1903, pp. 24 and 28. 44 Garmon 2008; Lindow 2001, p. 224. 45 Aimone 2015, p. 132, fig. 61–62 and 138 b. 41 42

353

Ivana Popović again part of the Byzantine Empire during the second decade of the sixth century, during the reign of Anastasius I, stronger cultural relations were not re-established before Justinian’s restoration of the Danube limes, fortifications and cities in its hinterland. That restoration, as revealed by archaeology and written sources, took place after the middle of the fourth decade of the sixth century,56 although cultural impulses from the Byzantine world reached Balkan provinces also during the reigns of his predecessors Anastasius I and Justin I.

Cahn H. A., Kaufmann-Heinimann A. (eds.), 1984. Der spätrömische Silberschatz von Kaiseraugst, Derendingen (Basler Beiträge zur Ur- und Frühgeschichte, 9). Garman A. G., 2008. The Cult of the Matronae in the Roman Rhineland: An Historical Evaluation of the Archaeological Evidence, Lewiston (NY). Ivanišević V., Kazanski M., Mastykova A., 2006. Les nécropoles de Viminacium à l’époque des Grandes Migrations, Paris (Centre de recherche d’histoire et civilisation de Byzance. Monographies, 22).

Bibliography

Kitzinger E., 1989. L’arte bizantina. Correnti stilistiche nell’arte mediterranea dal III al VII secolo, Milan (La cultura).

Abbreviation: OPEL IV = Lőrinz B., (ed.), 2002. Onomasticon provinciarum Europae Latinarum, IV, Vienna.

Kondić J., 1995. ‘Касноантичко и рановизантијско сребрно посуђе из Србије’, in I. Popović, T. Cvjetićanin, B. Borić-Brešković (eds.), Радионице и ковнице сребра. Акта научног скупа одржаног од 15. до 18. новембра 1994. године у Народном музеју у Београду – Silver Workshops and Mints. Symposium Acta, November 15–18 1994, National Museum Belgrade, Belgrade (Народни Музеј. Монографије, 9), pp. 181–187.

Aimone M., 2015. Il Tesoro di Canoscio, Rome (Monumenti antichi. Series monografica 72 / Series miscellanea, 18). Alföldy G., 1969. Die Personennamen der römischen Provinz Dalmatia, Heidelberg (Beiträge zur Namenforschung. Beiheft, 4). Arias P. E., 1986. L’anfora argentea di Porto Baratti, Rome (Bolletino d’arte. Monografia).

Korać M., 1991. ‘Late Roman tomb with frescoes from Viminacium’, in Starinar. N.S., 42, pp. 107–122.

Arias P. E., 1987. L’anfora di Baratti. Guida alla mostra, Firenze.

Коrać M., 1995. ‘The paintings in the Late Classical tombs in Viminacium’, in D. Srejović (ed.), The Age of Tetrarchs. A Symposium held from the 4th to the 9th october 1993, Belgrade (Српска академија наука и уметности. Одељење историјских наука. Научни скупови, 24), pp. 169–183.

Baratte F., 1975. ‘À propos de l’argenterie romaine des provinces danubiennes’, in Starinar. N.S., 26, pp. 33–41. Baratte F., 1996. ‘Les observations sur la vaisselle d’argent et ses possesseurs à la fin de l’Antiquité’, in Annals de l’Institut d’Estudis Gironins, 36, pp. 63–79.

Korać M., 2007. Сликарство Виминацијума – Die Malkunst Viminaciums – The Paintings of Viminacium, Belgrade.

Baratte F., 2000. ‘Anfora con decorazione figurata’, in S. Ensoli, E. La Rocca (eds.), Aurea Roma. Dalla città pagana alla città cristiana, Rome, pp. 502–503.

Lindow J., 2001. Norse Mythology. A Guide to Gods, Heroes, Rituals and Beliefs, Oxford.

Baratte F., Le Bot Helly A., Depassiot M. C., Lauglet V., 1990. Le trésor de la place Camille-Jouffray à Vienne (Isère), Paris (Gallia. Supplément, 50).

Martin M., 1984. ‘Weinsiebchen und Toilettgerät’, in Cahn, Kaufmann-Heinimann 1984, pp. 122–132.

Baratte F., Lang J., La Niece S., Metzger C., 2002. Le trésor de Carthage: contribution à l’étude de l’orfèvrerie de l’Antiquité tardive, Paris (Études d’antiquités africaines).

Mirković M. (ed.), 1986. Inscriptions de la Mésie Supérieure, II, Beograd. Popović I., 1994. ‘Silver production in the Early Empire period: Local products and imports, Cat. 277–281, 338– 339’, in I. Popović (ed.), Античко сребро у Србији – Antique Silver from Serbia, Belgrade, pp. 107–115, 330–332 and 358.

Bianchi Bandinelli R., 1965. ‘Naissance et dissociation de la koiné héllenistico-romaine’, in Le rayonnement des civilisations grecque et romaine sur les cultures périphériques. Huitième congrès international d’archéologie classique, Paris, 1963, Paris, 1965, pp. 441–463.

Popović I., 1997. ‘Miscellanea argentea’, in Starinar. N.S., 48, pp. 73–90.

Bland R., Johns C., 1993. The Hoxne Treasure. An Illustrated Introduction, London.

Popović I., 2001. Касноантички и рановизантијски накит од злата у Народном музеју у Београду – Late Roman and Early Byzantine Gold Jewelry in National Museum in Belgrade, Belgrade (Народни Музеј. Антика, 8).

Mirković 1986, pp. 25–26; Ivanišević, Kazanski, Mastykova 2006, pp. 131–133.

56

354

Early Byzantine Silver Chalices from Viminacium: Parts of a Tableware Set or Church Vessels? Vassits M., 1903. ‘La vaiselle d’argent du Musée national de Belgrade’, in Revue archéologique, 4/1, pp. 17–30. Vrom J., 2007. ‘The changing dining habits at Christ’s table’, in L. Brubaker, K. Kindarou (eds.), Eat, Drink, and be Merry (Luke 12:19). Food and Wine in Byzantium. Papers of the 37th Annual Spring Symposium of Byzantine Studies, in Honour of Professor A. A. M. Bryer, Aldershot/Burlington (Publications, 13), pp. 191–222. Werner J., 1986. Der Schatzfund von Vrap in Albanien. Beiträge zur Archäologie der Awarenzeit im mittleren Donauraum, Vienna (Österreichische Akademie der Wissenschaften. Philosophisch-historische Klasse. Denkschriften, 184).

355

28 Late Antique Bronze Crosses Found in Punta de l’Illa, Cullera, Valencia, Spain: The Archaeological Remains of Emperor Justinian’s Dream? María J. S. Vicent Universidad de Valencia (Spain) Punta de l’Illa, in Cullera, Spain, was archaeologically examined for the first time during the summer of 1955, with excavations supported by the Bryan Foundation and managed by M. Tarradell and E. Pla. Several structures were documented; among them was one in which the archaeological remains consisted of a large quantity of Late Roman unguentaria and a collection of coins, including several Justinianic nummi and two bronze crosses. These last form the subject of this paper. These archaeological remains are unrelated to other sites nearby and are regarded as very singular finds. The largest cross has overall dimensions of 22.5  cm with a thickness of 1 mm. Although a fragment of one arm was loose, the cross appears to have arms of equal length. Tree-shaped motifs may be distinguished on each arm, which this chapter examines. The decoration was made by small cross-shaped patterns engraved in the bronze. The smaller cross, intact and also with arms of equal length, is less than 5 cm across. This cross could be considered a pendant cross, because it has three hanging elements. The aim of this paper is to analyse available archaeological information and to search for remains with a similar subject, shape, manufacture and function, mainly in Byzantine ecclesiastical objects and in the Early Byzantine city of Caričin Grad. Another research question relates to the cultural context in which the Christian meaning of the Tree of Life may be understood and identified. In addition, these various aspects are linked to the geographical features of the archaeological site and to one of the aims of Justinian’s policies, according to Procopius: expanding the Empire ‘to the farthest shores of the ocean’. La Punta de l’Illa de Cullera a été archéologiquement étudiée pour la première fois à l’été 1955, à l’occasion de fouilles dirigées par M. Tarradell et E. Pla, avec l’appui de la Fondation Bryan. Plusieurs structures ont été décrites, dont l’une a livré de nombreux unguentaria romains tardifs, une collection de monnaies comprenant plusieurs nummi de Justinien ainsi que deux croix de bronze qui font l’objet du présent article. Ces vestiges archéologiques ne sont pas liés à d’autres sites à proximité. Aussi, sont-ils considérés comme des trouvailles très singulières. La plus grande croix mesure 22,5 cm pour une épaisseur de 1 mm. Bien qu’une branche horizontale soit perdue, il semble que les deux aient été de même longueur. On distingue sur celles-ci des motifs d’arbre qui seront examinés dans cet article. Le décor est composé de petits motifs cruciformes gravés dans le bronze. La plus petite croix, d’une largeur de moins de 5 cm, a été trouvé complète et a aussi des branches de même longueur. Elle pourrait être considérée comme un pendentif, puisqu’elle comporte trois éléments suspendus. Cet article analysera les données archéologiques disponibles et cherchera des objets de forme, de facture et de fonctino similaires, principalement parmi les objets ecclésiastiques byzantins et dans la cité protobyzantine de Caričin Grad. Il s’intéressera également au contexte culturel dans lequel l’Arbre de Vie aurait pu être identifié et compris. De plus, ces différentes questions seront liées aux caractéristiques géographiques du site archéologique et à l’un des objectifs des politiques de Justinien, selon Procope : l’expansion de l’Empire « jusqu’aux rivages les plus lointains de l’océan ». Das sogenannte Gelände in Punta de l’Illa in Cullera wurde im Sommer 1955 erstmals archäologisch untersucht. Die Ausgrabungen führte M. Tarradell und E. Pla mit Unterstützung der Bryan Foundation durch. Es wurden mehrere Gebäude untersucht, in einem davon kamen viele spätrömische Unguentaria, zahlreiche Münzen darunter justinianische Nummi und zwei Kreuze zum Vorschein. Im Folgenden werden hier die Kreuze behandelt. Sie zeigen keine Beziehung zu anderen Funden in diesem Gebiet, folglich handelt es sich um besondere Entdeckungen. Das größere Kreuz ist 22,5 cm lang und 1 mm dick. Obwohl einer der Arme fehlt, dürfte es sich um ein Kreuz mit vier gleichlangen Armen handeln. Baummotive sind auf jedem Arm des Kreuzes zu 357

María J. S. Vicent erkennen, welche in dieser Studie näher untersucht werden. Diese Verzierung wurde durch kleine kreuzförmige Punzen in die Bronze eingebracht. Das kleinere Kreuz hat ebenfalls vier gleichgroße Arme und weniger als 5 cm Durchmesser. Dieses Stück muss als Anhänger gedient haben, wie die drei Aufhängerelemente es verdeutlichen. Das Ziel dieser Studie ist einerseits die Analyse dieser Objekte mit der Suche nach Parallelen in Form, Herstellungsweise und Funktion unter frühbyzantinischer Kirchenausstattung, insbesondere in der Stadt von Caričin Grad. Andererseits werden die kulturellen Hintergründe für die christliche Bedeutung des Motivs des Lebensbaumes untersucht. Anschließend werden diese Aspekte mit den geographischen Gegebenheiten dieser archäologischen Stätte ebenso in Verbindung gebracht wie mit einem der Ziele Iustinians Politik und zwar laut Prokop: „zu den entferntesten Küsten des Ozeans“. Il sito di Punta de l’Illa di Cullera fu indagato archeologicamente per prima volta durante l’estate del 1955. Gli scavi furono sostenuti dalla Bryan Foundation e condotti da Tarradell e E. Pla. Vennero documentate diverse strutture architettoniche e tra gli oggetti trovati all’interno di una di esse si contano, principalmente, un grande numero di unguentari tardo romani, una collezione di monete, inclusi diversi nummi giustinianei, e due croci in bronzo che sono il soggetto di questo articolo. Questi ritrovamenti non sono confrontabili con altri nei dintorni e vengono quindi considerati scoperte molto particolari. La croce più grande ha bracci di 22,5 cm e uno spessore di 1 mm. Anche se uno dei bracci è perduto, si potrebbe ipotizzare una croce a bracci uguali decorati ciascuno da motivi in forma di albero che vengono analizzati in questo saggio. La croce minore non raggiunge 5 cm, ma fu rinvenuta completa ed è ugualmente una croce greca. Questa andrebbe ritenuta una croce pendente in ragione dei tre elementi sospesi. Lo scopo di questo articolo è, a partire dai dati archeologici, l’individuazione di altri manufatti simili per motivi, forma, tecniche e funzione tra gli oggetti ecclesiastici bizantini come quelli rinvenuti nella città di Caricin Grad. Viene ugualmente evidenziato il contesto culturale nel quale il significato cristiano dell’Albero della Vita potrebbe essere identificato e compreso. Inoltre, i diversi aspetti accennati vanno vincolati alle circostanze geografiche del sito archeologico e, seguendo le parole di Procopio, a uno degli obiettivi della politica di Giustiniano: l’espansione dell’Impero “fino alle sponde più lontane dell’oceano”. Introduction

objects of the early Christian and Byzantine periods). Finally, the religious and political context in Valentia and the Iberian Peninsula will inform a possible interpretation of the archaeological site and the finds.

Ships sailed across the sandbanks of Hispania’s eastern coast, while marshes and low shores unfolded from the inland of the Gulf of Valentia. Forty-five kilometres south of the city of Valentia, a mountain rose up from the flatness and fell away down to the sea, forming the only bay where a ship could drop anchor in 200 km of coast. The bay was bordered on its northern side by a small island (fig. 1), known as Punta de l’Illa, which was joined to the land in the eighteenth century (fig. 2). Amadeo Serra Desfilis directed my attention to this place, and to the unique remains of two bronze crosses found there as I began my research.

Numerous studies have attempted to explain the overall archaeological site of Punta de l’Illa. Most of them study the bronze crosses in the context of the broader set of remains found there. The first study that focused on the crosses was written in 1992: Enric A. Llobregat’s article ‘Las cruces de la Punta de l’Illa (Cullera)’,1 published in Estudios de arqueología ibérica y romana: homenaje a Enrique Pla Ballester. This research made an important contribution to knowledge on the crosses, comparing them with others of varying origins and varied functions, though the author suggests that only the smaller cross could have been part of a censer or similar liturgical device.

The aim of this study is to review previous research into the bronze crosses found on Punta de l’Illa. Until now, they have been analysed mainly through archaeological methods. This paper goes a step further, aiming to shed new light on the bronze crosses through a stylistic and historical analysis of them as decorative metalwork elements of lighting fixtures in early Christian churches. This re-examination will provide a fresh interpretation of these exceptional archaeological remains.

Research into the Punta de l’Illa bronze crosses began by expanding the work carried out by E. Llobregat with new publications that included more recent archaeological findings, in addition to reviewing some of the references cited in the 1992 article.

The chapter begins by reviewing the main existing studies on the crosses found on Punta de l’Illa and explaining the archaeological material found at the site more generally. It will then continue with a detailed account of the bronze crosses, considering earlier research on the subject, on their shape, manufacture and function (mainly as ecclesiastical

Similar crosses have been sought both in Hispania and in collections of crosses with origins in the Mediterranean basin. E. Llobregat analysed the larger cross as a processional cross. As we will see below, parallel examples 1

358

Llobregat 1992, pp. 663–670.

Late Antique Bronze Crosses Found in Punta de l’Illa, Cullera, Valencia, Spain

Fig. 1. Aerial view of Cullera Bay.

of processional crosses have been sought for the large cross, but are not a satisfactory basis for study.

which we are concerned. A handicraft context has been found through the connection of our bronze crosses with other contemporary metal liturgical objects.

However, an examination of the bibliography concerning the hanging elements of both lighting fixtures and censers has provided examples closer to the typology of the crosses that form the subject of our study. The reference objects found via this method of study, as we will see, are formally and functionally more similar to those with

At this stage of research, several papers are essential. One of these is M. Xanthopoulou’s Les lampes en bronze à l’époque paléochrétienne: in its pages, we can find significant parallels for the two crosses in question. It is based on a thorough analysis and personal examination

Fig. 2. View of the Punta de l’Illa.

359

María J. S. Vicent The smaller cross with hanging device (fig. 4), which is intact, is less than 5 cm across, and is also an equal-armed cross, made up of horizontal and vertical bars. The hanging device is formed by three metal strips.

of objects in Greek, French and English museums, a knowledge of copper alloys and the techniques used in their manufacture, and a reading of Late Antique and early Byzantine sources dealing with lighting. Above all, it offers one of the first typology studies relating to the metallic lamps and luminaires of this period, digging up new clues about their origin, their dating and the exchanges or relationships that give testimony to them.

Judging from E. Llobregat’s research, several crosses in the Mediterranean world can be compared with the larger one from Late Antiquity. Examples from Hispania can be found in the National Archaeological Museum of Spain, where the Guarrazar Hoard (621–672), the cross of Burguillos and the cross of Torre Don Jimeno are housed. A few interesting examples are catalogued in the book Byzantine Figural Processional Crosses, by J. A. Cotsonis.5 In its list of crosses, four dated between the sixth and seventh centuries are worth mentioning.6 They are three crosses with figures and votive inscriptions, and a centre fragment of a cross with figures. All of them are cast-bronze crosses with flaring arms that end in pairs of circular flat serifs, and the figures and the inscriptions are engraved into the cast bronze. The figures of the cross belonging to the Royal Ontario Museum (item 10) are rendered schematically, and the figures engraved on the centre fragment of one cross (item 11) are described as very schematised drawings. We can find the same sort of very schematic depiction in the trees engraved in the larger cross from Punta de l’Illa. In this case, however, the tree design is made of repeated small crosses, becoming a symbolic and simplified tree as a combination pattern carved into the bronze. As with the other crosses mentioned above, solid casting was the method of manufacture.

It has also been necessary to consult literature about archaeological sites in the Eastern Mediterranean and the Balkan Peninsula, in order to achieve satisfactory results in the search for parallels. Early Christian metal crosses from (mainly) Stobi, Caričin Grad/Justiniana Prima and Scythia Minor have proved to be the most similar to the crosses studied. The bronze crosses found in Punta de l’Illa Two excavation campaigns were carried out: the Bryan Foundation, under the direction of M. Tarradell and E. Pla Ballester, supported the first in 1955–56, and the Valencian Museum of Prehistory2 supported the second in 1966, under the direction of E. Pla Ballester. Several architectural contexts were documented: among them was one in which the archaeological remains consisted of a large number of late Roman unguentaria and an interesting collection of coins, along with the two bronze crosses that constitute the subject of this paper. The excavation revealed an architectural complex made up of several masonry wall enclosures. Archaeologists documented three different excavation areas: Zonas I, II, and III.3 Zona I is a rectangular building, of which only the enclosing wall was found. The building measures 14.75 m in length and 8.30 m in width, with the short sides presenting a north–south orientation. The bronze crosses were found along the west wall, together with a Roman terracotta lamp and a bronze statue. The different types of finds suggest that the site had been occupied for several centuries since Antiquity.

With the above in mind, I would like to underline the important differences between the Punta de l’Illa crosses and others found elsewhere. One difference is the tree motif engraved on each arm of the larger cross. E. Llobregat interpreted it as the Tree of Life.7 Within the research available on the use of this motif in liturgical objects, I would highlight S. Zwirn’s comment about a book cover that is part of the Sion Treasure housed in the Early Byzantine Dumbarton Collection.8 The book cover bears a cross flanked by cypress trees. In the Christian tradition of interpretation (exegesis), the cross symbolised the presence of Christ and could stand for the Tree of Life, in the midst of the garden (Genesis 2.9). The death of Christ on the cross is the central image in Christian art and the visual focus of Christian contemplation. What is more, the liturgical objects were placed on the altar, where these pieces of religious imagery depicted a belief in a heavenly paradise. The cross surrounded or flanked by trees was not a symbol frequently used, nor was it a common decorative motif in Hispania.

The larger cross (fig. 3) measures 22.5 cm across and is 1 mm thick; one arm has been detached from the cross, and the end of another arm is broken. Nevertheless, it appears to be an equal-armed cross, and tree motifs can be distinguished on each of its arms. The tree motifs are a unique decorative element for a cross, with small crossshaped patterns engraved in the bronze (fig. 3). The arms end in a ‘v’ shape, which sometimes appears on sixth- and seventh-century pilgrims’ ampullae from the Holy Land,4 like the silver ampulla belonging to the Monza Duomo treasure.

Another difference from the reference crosses is that these were manufactured in one piece, while the two

I would like to express my gratitude to Ángel Sánchez and Manuel Gozalbes from the Valencian Museum of Prehistory for the useful information they provided on archaeological context. In addition, I would like to thank Silvana Blazevska from the Stobi archaeological site (North Macedonia) for helping me in my research. 3 Cf. the diary of the excavation in 1955 (Museu de Prehistòria de València), pp. 86–87. 4 Cotsonis 1994, p. 41. 2

Catalogue of the exhibition held at Dumbarton Oaks from 23 September 1994 to 29 January 1995. 6 Items 8–11. Cf. Cotsonis 1994, p. 96. 7 Llobregat 1992, p. 664. 8 Cf. http://museum.doaks.org/Obj27043?sid=23667&x=90079&port=2607. 5

360

Late Antique Bronze Crosses Found in Punta de l’Illa, Cullera, Valencia, Spain

Fig. 3. The larger bronze cross.

crosses under study were made by overlaying one bronze assembly sheet over another. This suggests that they were made by a metallurgist with limited resources and skills, not by a skilled craftsman of metal liturgical objects9 or a goldsmith. This peculiarity is also observable in a Byzantine metalwork copper-alloy cross belonging to the Metropolitan Museum of Art (accession number: 1999.519.4). It dates from the sixth century or later and has no engraving. Like the larger cross from Punta de l’Illa, it was assembled from two sheets, not by solid casting. Lamp and censer holders of various types are frequent discoveries in early Byzantine provinces. A significant number of pieces have been brought to light on the Balkan Peninsula. An interesting volume on the subject was recently published: Studies in the Antiquities of Stobi.10 It contains a chapter, ‘A corpus of early Christian metal crosses from Stobi’, that compiles a set of different types of crosses found in Stobi between the years 1940 and 2009. In this chapter, we find four categories of cross, organised according to their use. The most relevant category for us is ‘4. Crosses that are part of interior lighting fixtures’, because it contains several bronze crosses. These finds are the most similar to our crosses, as they are characteristic examples of artefacts associated with hanging lighting. Examples 8–11 may be compared to our larger cross, while 6 and 7 may be compared to the smaller cross. Information on working with precious metals can be found in: de Palol 1968, p. 214 and fig. 121–122; Barroso-Cabrera 1990, p. 85. On precious metal objects being manufactured through smelting, cf. Balmaseda Muncharaz, Papí Rodes 1998, p. 121. 10 Blaževska 2018. 9

Fig. 4. The smaller cross with the hanging elements.

361

María J. S. Vicent Minor. In figure 2 of the publication, we can see a hanging light fixture from Libida. In this case, a smaller cross is formed of two metal strips and three-hooked arms on the lower side of the suspended fixture.

The examples numbered 8–11 form a group named ‘bronze chandelier segment’ and all of them date from the sixth century. These crosses were all formed in the same manner: by riveting together vertical and horizontal bars. Numbers 9–11 also have pairs of hoops for holding the glass lamps arranged in them. In addition, most of them have circular perforations at the lower and upper ends of the vertical bar, for hanging wires. In this corpus, the example closest to our smaller cross is number 6. This cross is in the central decorative area of the bronze chandelier segment and is made of a horizontal and vertical bar with rounded ends. As with the rest of the crosses formed by two bars, both bars are joined perpendicularly by a rivet in the middle.

The holes, located at each end of the arms in the larger Punta de l’Illa cross, could be functional perforations where hoops or additional holders were attached to configure the hanging lighting device. There are also similarities with the smaller Punta de l’Illa cross in the size and the hanging elements of the crosses from the early Byzantine city of Caričin Grad.16 N. Duval and M. Jeremic discovered cross number 137-53 (5.9 cm) in the single-nave church; others were discovered in the Acropolis basilica, such as cross number 12-57 (3.5 cm) and a fragment of cross number 20-57 (5 cm). The piece most similar to the smaller cross is the one found in the double church (cross number 17-65 (12.9 cm)), because it was made by assembling two bronze sheets. This type of cross is frequently found in sixth-century votive crowns, hanging light fixtures and censers, and appears on many pieces from this period. Other, smaller reference crosses from Caričin Grad can be found in the article by M. Milinković, in particular, a smaller cross presented in figure 13 of his 2011 paper, ‘Прилог проучавању тзв. ранохришћанских полијелеја у Србији’ (‘A contribution to the study of socalled Early Christian chandeliers in Serbia’).

According to Davča Spasova, the role of these bronze crosses was to serve as holders of glass lamps, as is demonstrated by the banded hoops that act as additional holders of glass lamps. Another shared characteristic between this group and the Punta de l’Illa crosses is the use of plain wires for suspension, instead of chains. For the purpose of this study, the George Marcopoli collection,11 housed in the British Museum, is highly relevant. This collection is a group of objects that includes early Byzantine Syrian metal liturgical finds. These objects were acquired by Georges Marcopoli at the end of the nineteenth century, when he was residing in Aleppo, Syria, as Spanish consul in the city. Since the outbreak of war between Italy and the Ottoman Empire, the collection has remained in Europe.

Discussion Taken together, the manufacturing method, the treeshaped monogram on the large cross, the absence of an identified bishop or donor, and the smaller reference crosses found mostly outside of Hispania would suggest that the architectural site of Punta de l’Illa was not settled by the Visigoths. Moreover, there may have been an explicit purpose for crosses of slightly different sizes and manufacture, reflecting the relative importance of the church where they were housed.17

Two of the objects contained in this collection offer Middle Eastern parallels to the Punta de l’Illa crosses. One is a suspension element12 made up of a hook with a cross inserted. This cross was made in the same way as the smaller Punta de l’Illa cross. The other is a copper-alloy polycandelon,13 of which there are several in the collection, but this one consists of a pierced disk, with round perforations alternating with heart-shaped ones. The suspension unit, made up of three strips with hoops in both ends, is similar to the suspension device joined to our smaller cross.

Their features more closely resemble the usual characteristics of Byzantine plain crosses, in terms of the cheaper metal employed (bronze), the dimensions (following the same pattern), and the figures (very schematised drawings). It seems that the trees engraved in the arms of the cross could have been intended to add power to the sign of the cross.18

Two fifth/sixth century bronze crosses, also from Syria, are included in the exhibition catalogue Die Kunst der frühen Christen in Syrien. They are number 6 and 8, both from Hama Museum.14 They are made from sheet metal, like the smaller Punta de l’Illa cross.

The coins found on Punta de l’Illa include an interesting set of small Vandal and Byzantine nummi, the patterns on which are consistent with those found at Byzantine archaeological sites.19 In the archaeological sites of North Africa, Vandal nummi are usually found together with Justinianic nummi. Scholars speculate that the dissemination of the coins may have been linked to Byzantine control of the territory. On

A series of suspended light fixtures15 were discovered in early Byzantine settlements from the province of Scythia 11 The Marcopoli family ran an antiques business and played an important diplomatic role in Aleppo. George Marcopoli seems to have been a leader in the antiques business (Jenkins-Madina 2006, p. 26). 12 Museum no 1994.0610.24. 13 Museum no 1994.0610.1. 14 Fansa, Bollman 2008, pp. 144–145. 15 Petcu, Nuțu, Nastasi 2014, p. 358.

Duval, Vladislav 1984, p. 131, fig 129, and p. 138, fig. 141. Milliken 1954, pp. 190–192 and 195. 18 Like inscriptions and images on other crosses. Cf. Cotsonis 1994, p. 40. 19 Marot, Llorens 1996, p. 161. 16 17

362

Late Antique Bronze Crosses Found in Punta de l’Illa, Cullera, Valencia, Spain Punta de l’Illa, the small nummi with the monogram of Christ from the first years of Byzantine control in Carthage are the most abundant.

The size and layout of the architectural spaces, together with epigraphical evidence that mentions a martyr’s sanctuary for St Vincent, point to the possibility that they could have been established as a monastery.21

The Visigoths had already converted to Christianity when they broke from the Roman Empire, but they were Arian, a sect that denied the divine nature of Christ. The Roman Church and its bishops suffered religious persecution by the Arian Visigoths, which was especially intense before the Visigoths joined the Catholic faith at the Council of Toledo in the sixth century. In the latter half of this century, Valentia was the focus of the fight between Arians and Catholics, and until the conversion of King Reccared at the Council of Toledo in 589, there were two co-existing bishops: Ubiligisclus (of Arian faith) and Celsinus (of Catholic faith).

Martyria, relics and holy images were honoured by light, according to Greco-Roman tradition. Representations of the caves of the Nativity and the Anastasis show kandēlai and, more rarely, lamps hanging or held by candelabra at the entrance to or inside these monuments. During her visit to the tombs of Christ and Job, Egeria noticed the lamps that burned there day and night. In addition, the geographical location of Punta de l’Illa, at the limes of the land held by the Byzantine Empire in Hispania, could indicate that the site was occupied by Byzantine military forces following the same strategy Diehl describes in his book, Byzantine Africa. According to Diehl, Byzantine troops established various strongholds near cities they intended to conquer.22 Diehl cites several examples of Catholic architectural structures related to reoccupation, like the preliminary positions where the Byzantine military forces were established to invade important cities.23 In the case at hand, there could also have been motivation to protect the Catholic shrine from Visigoths intent on prevailing and promoting Arianism.24

One inscription20 found in the city of Valentia, now missing, but the words of which were transcribed by St Isidore, alluded to the fact that Bishop Justinian of Valentia (531–546) was the head of the Catholic faith in the city, and that he venerated St Vincent the Martyr. In his book, De viris illustribus, St Isidore says of him: Pius preclarus doctor alacer facundus Justinianus caelebs pontifex sacerdos noba templa construens uetustaque restaurans ornabit festa dictis predicans in populis uirgenes instituens monascosque gubernans scripsit plura posteris profutura seclis hic miro maris insolam munimene sepsit in qua maris circumfluentibus undis sílice disrupto predulcem repperit limfam hic Vicentium gloriosum martirem XPi sat pio quem coluit moderamine uiuens hunc deuotus miriens reliquit eredem undecim presentis quinquennia uite quattor lustris uisque quaternis mensibus connumerandus scis ministrabit antetis.

Conclusions The monastery of Punta de l’Illa may have been under close episcopal supervision,25 and therefore built to worship St Vincent. In this case, it would be an important site for the Catholic Hispano-Roman population. The characteristics of the crosses, together with the geographical setting of the archaeological site where they were found, indicate that they may have come from outside the Iberian Peninsula. The closest parallels to these crosses come from sixth-century AD contexts across the Byzantine world.26 It should be emphasised that the basic Christian symbol, the cross, used in lighting fixtures, was standardised, and that local, regional or supra-regional workshops served civilian, religious and military demands.

Pious, agile, eloquent, illustrious doctor, Celibate, priest, Bishop Justinian He built a new church and restored the antiquities He adorned the celebrations with preaching to the people He founded a nunnery and governed monks He wrote many works that will be used well into the future centuries He encircled with a marvellous fortification an island of the sea On which, surrounded by the waves of the sea Digging the limestone ground, he found a fresh spring Here, the glorious martyr Vincent, Whom he so venerated in life with his holy rule, When he died he appointed as his heir His life was twelve quinquennia And he was bishop for four five-year periods and twenty months We must number him among the group of saints. 20

Nothing can give us absolute certainty that the bronze crosses of Punta de l’Illa belonged to, were made for, or were used to celebrate liturgy at a provisional Byzantine settlement in Punta de l’Illa. Nevertheless, the archaeological context and the geographical features of this coast lead us to think that the Byzantine troops may have been at this site to defend the Orthodox faith at the hypothetical monastery located in Punta de l’Illa, and that, to achieve their goal, they needed to arrange and decorate

Llobregat 1977, p. 26. Diehl 1896, p. 142. 23 Diehl 1896, p. 180. 24 Diehl 1896, p. 408. 25 Fernández 2013, p. 237. 26 Xanthopoulou 2010, p. 69. 21 22

Llobregat, 1986, p 71.

363

María J. S. Vicent Goubert P., 1946. ‘Administration de 1’Espagne byzantine (suite) II: les provinces’, in REB, 4, pp. 71–110.

the interior of the church, from which only two bronze crosses have come down to us today.

Héron de Villefosse A., 1914. ‘La croix byzantine de Bulla Regia’, in Comptes rendus des séances de l’Académie des Inscriptions et Belles-Lettres, 58/8, pp. 697–702.

Bibliography Abbreviations:

Jenkins-Madina M., 2006. Raqqa Revisited: Ceramics of Ayyubid Syria, New York.

AJA = American Journal of Archaeology. BMAN = Boletín del Museo arqueológico nacional.

Llobregat E. A., 1977. La primitiva cristiandat valenciana, segles IV al VIII, Valencia.

REB = Études byzantines/Revue des études byzantines.

Llobregat E. A., 1992. ‘Las cruces de Punta de l’Illa (Cullera)’, in J. J. Cabanilles (ed.), Estudios de arqueología ibérica y romana: homenaje a Enrique Pla Ballester, Valencia (Museo de Prehistoria de Valencia. Trabajos Varios, 89), pp. 663–668.

Balmaseda Muncharaz L. J., Papí Rodes C., 1998. ‘Cruces, incensarios y otros objetos litúrgicos de épocas paleocristiana y visigoda en el Museo Arqueológico Nacional’, in BMAN, 16, p. 119–142. Barroso-Cabrera R., 1990. ‘Dos joyas de orfebrería Hispanovisigoda procedentes de Huete (Cuenca)’, in BMAN, 8, pp. 83–90.

Llobregat E. A., Yvars J. F. (ed.), 1986. Història de l’art al País Valencià, I, València (Biblioteca d’estudis i investigacions, 10).

Cotsonis J., 1994. Byzantine Figural Processional Crosses. Catalogue of an Exhibition at Dumbarton Oaks, 23 September 1994 through 29 January 1995, Washington (Dumbarton Oaks Byzantine Collection Publications, 10).

Marot T., Llorens M., 1996. ‘La circulación monetària en el siglo VI dC en la costa mediterrània: la Punta de l´Illa de Cullera (Valencia)’, in Revista d´Arqueologia de Ponent, 6, pp. 151–180. Mathews T. F., 2003. The Clash of Gods. A Reinterpretation of Early Christian Art, rev. and exp. ed., Princeton.

de Palol P., 1968. Arte Hispánico de la época visigoda, Barcelona (Biblioteca de arte hispánico).

Milinković M., 2011. ‘Прилог проучавању тзв. ранохришћанских полијелеја у Србији’, in Ниш и Византија – Niš and Byzantium, 9, pp. 73–84.

Diehl C., 1896. L’Afrique byzantine. Histoire de la domination byzantine en Afrique, Paris (Description de l’Afrique du Nord).

Milliken W. M., 1954. ‘Byzantine goldsmith work’, in The Bulletin of the Cleveland Museum of Art, 4118, pp. 190–192 and 195.

Dodd E. C., 1987. ‘Three Early Byzantine silver crosses’, in W. Tronzo, I. Lavin (eds.), Studies on Art and Archaeology in Honor of Ernst Kitzinger on his Seventy-fifth Birthday = Dumbarton Oaks Papers, 41, pp. 165–179.

Muñoz-Amilibia A., 1982. ‘Cruz de bronce monogramática procedente de Cehegín (Murcia)’, in P. de Palol (ed.), II Reunió de arqueología paleocristiana hispánica. IX symposium de Prehistòria i arqueologia peninsular, Montserrat, 2–5 novembre 1978, Barcelona, pp. 265– 275.

Duval N., 1984. ‘L’architecture religieuse de Tsaritchin Grad dans le cadre de l’Illyricum oriental au VIe siècle’, in Villes et peuplement dans l’Illyricum protobyzantin. Actes du colloque de Rome (12–14 mai 1982), Rome (Collection de l’École française de Rome, 77), pp. 399–481.

Petcu R., Nuțu G., Nastasi I., 2014. ‘Suspending light devices from Scythia Minor’, in Studia antiqua et archaeologica, 20, pp. 353–368.

Duval N., Vladislav P. (eds.), 1984. Caričin Grad, I, Rome (Collection de l’École française de Rome, 75/1).

Spasova D., 2018. ‘A corpus of Early Christian metal crosses from Stobi’, in S. Blaz̆ evska (ed.). Студии за старините на Стоби – Studies in the Antiquities of Stobi, IV, Stobi, pp. 393–404.

Fansa M., Bollmann B. (eds.), 2008. Die Kunst der frühen Christen in Syrien. Zeichen, Bilder und Symbole, 4. bis 7. Jahrhundert. Begleitband zur Sonderausstellung im Landesmuseum Natur und Mensch Oldenburg, Mainz am Rhein (Schriftenreihe des Landesmuseums für Natur und Mensch – Publications from the State Museum of Nature and Mankind, 60).

Wiseman J., Mano-Zissi D., 1971. ‘Excavations at Stobi, 1970’, in AJA, 75/4, pp. 395–411. Wiseman J., Mano-Zissi D., 1972. ‘Excavations at Stobi, 1971’, in AJA, 76/4, pp. 407–424.

Fernández D., 2013. ‘City and countryside in Late Antique Iberia’, in Antiquité tardive, 21, pp. 233–241.

Xanthopoulou M., 2010. Les lampes en bronze à l’époque paléochrétienne, Turnhout (Bibliothèque de l’Antiquité tardive, 16).

Goubert P., 1944. ‘Byzance et 1’Espagne wisigothique (554–711)’, in REB, 2, pp. 5–78. Goubert P., 1945. ‘L’administration de 1’Espagne byzantine, I: les gouverneurs de l’Espagne byzantine’, in REB, 3, pp. 127–142. 364

Part 5 Restoration and Conservation

29 Nouvelle intervention sur la mosaïque du sanctuaire de Mar Gabriel (Turquie orientale) Patrick Blanc Musée départemental Arles antique (France)

Véronique Blanc-Bijon Aix Marseille Université, CNRS, UMR 7299 – Centre Camille Jullian, Aix-en-Provence (France)

Sébastien de Courtois Institut français d’Ankara (France/Turquie) À la demande de l’évêque diocésain du Tur Abdin qui réside dans le monastère, nous avons entrepris à partir de 2011 le nettoyage et la consolidation des mosaïques du sanctuaire de l’église de Mar Gabriel, non loin de Midyat, en Turquie du Sud-Est. Ce programme de conservation soutenu par la Fondation Leventis s’accompagne de la révision de l’étude des mosaïques qui sont datées par des manuscrits de 512, de la fin du règne d’Anastase, une datation confirmée par les analyses des tesselles effectuées durant nos recherches. At the request of the diocesan bishop of Tur Abdin, who dwells in the monastery, we have since 2011 been carrying out a campaign of cleaning and consolidating the ancient wall mosaics in the sanctuary of the Mar Gabriel church in Qartmin, not far from Midyat, in south-eastern Turkey. This conservation programme, supported by the Leventis Foundation, is linked to an updating of the research into those mosaics, which are dated to 512, at the end of the reign of Anastasios, with reference to manuscripts as well as analyses of glass tesserae conducted during our new examination. Auf Ersuchen des im Kloster ansässigen Diözesanbischofs von Tur Abdin erfolgt seit 2011 die Reinigung und Sicherung der alten Wandmosaiken im Chorraum der Mor Gabriel Kirche in Qartmin, nicht weit von Midyat, im Südosten der Türkei. Dieses von der Leventis-Stiftung geförderte Konservierungsprogramm ist mit der erneuten Untersuchung der Mosaiken verknüpft. Die Mosaiken werden an das Ende der Regierungszeit von Anastasios auf 512 datiert und zwar anhand von Schriftquellen sowie aufgrund der Analysen von Glastesserae, die im Rahmen dieser neuen Untersuchung analysiert werden. Su richiesta del vescovo diocesano del Tur Abdin, che risiede nel monastero, dal 2011 abbiamo intrapreso la pulitura e il consolidamento degli antichi mosaici delle pareti e della volta nel santuario della chiesa di Mar Gabriel a Qartmin, non lontano da Midyat, nella Turchia sudorientale. Questo programma di conservazione sostenuto dalla Fondazione Leventis è collegato a una revisione dello studio di quei mosaici, che sono datati al 512, alla fine del regno di Anastasio, da manoscritti e ora anche dall’analisi delle tessere di vetro condotta durante la nostra nuova indagine.

des mosaïques antiques du sanctuaire de l’église de Mar Gabriel à Qartmin, non loin de Midyat, dans le Sud-Est de la Turquie. Ce programme de conservation est lié à

À la demande de l’évêque diocésain du Tur Abdin, Mgr Samuel Atkas, qui réside dans le monastère, ont été entrepris depuis 2011 le nettoyage et la consolidation 367

Patrick Blanc, Véronique Blanc-Bijon & Sébastien de Courtois de Berlin9, description commentée par l’abbé Leroy en 1956 : « Le sol du sanctuaire est couvert de mosaïques en marbre blanc, noir, jaune, pourpre violet, lie-de-vin, avec des figures variées. Ses murs, disposés en cercle, des plaques de marbre les recouvrent et, au-dessus, sur la voûte, des mosaïques de cubes dorés »10. Ce n’est qu’en 1972 qu’une équipe de Dumbarton Oaks11 put examiner les mosaïques d’un point de vue tant iconographique que technique. Marlia Mundell publia dans les Dumbarton Oaks Papers de 197312 une étude détaillée de ce monastère et de son décor, accompagnée d’une très riche illustration photographique et d’un plan révisé (fig. 1)13.

la constitution du corpus des inscriptions syriaques de la région et à une révision de l’étude des mosaïques.1 Fondé en 397 par Samuel, originaire de Mardin et fils spirituel de l’évêque martyr Karpos tué au cours d’un raid des Perses contre Nisibe au milieu du IVe siècle, le monastère Mar Gabriel est au cœur de l’histoire du Tur Abdin, haut-lieu de l’implantation du christianisme et de la culture syriaque au Proche-Orient. L’histoire du Tur Abdin, de ses monuments, de ses manuscrits, de ses nombreux personnages saints, ascètes, moines, évêques, scribes, écrivains, est une des pièces maîtresses de l’histoire de la Syrie et de la Mésopotamie du IIIe siècle à nos jours.

Avec quelques divergences, la plupart des chercheurs s’accordent pour dater le décor de la phase de construction de l’église sur les bases de deux inscriptions publiées par H. Pognon14 et des manuscrits syriaques évoqués cidessus, soit de « la 21e année [du] règne [du] victorieux et miséricordieux Anastase le Christophore, qui avait revêtu la robe de la foi orthodoxe des chrétiens »,15 correspondant à l’année 512 de notre ère.

À partir du milieu du XIXe siècle, plusieurs visiteurs se rendirent au monastère, important centre religieux de la communauté syriaque,2 en particulier Henri Pognon en 1891 et 1905,3 Conrad Preusser en 1909, Gertrude Lowthian Bell en 1909 et 1911. Puis, de 1918 à 1954, le Tur Abdin devint zone militaire inaccessible ; l’abbé Jules Leroy fut le premier à pouvoir y retourner en 1954.4 Ces visiteurs signalent que des vestiges du décor antique perdurent dans le sanctuaire de l’église principale : il reste une voûte en berceau ornée ainsi que ses deux lunettes de mosaïques et un sol en opus sectile. C. Preusser évoque le pavement en 1911.5 G. L. Bell photographia ce décor en 1909 et le décrivit également en 19116 ; elle publia un plan des bâtiments avec deux dessins de la mosaïque de la voûte réalisés à partir de ses photographies et de croquis levés sur place.7

En 2006, lors de notre premier séjour effectué pour établir un constat d’état de conservation des mosaïques (fig. 2), nous avons observé, outre le très fort encrassement général de la voûte souvent noté par les premiers visiteurs, que la mosaïque présentait de très nombreux points de détachement de son support. Déjà G. Bell notait que « the vault is much blackened by smoke; if it were cleaned every detail would be visible »,16 ce qui lui faisait douter de la richesse de la palette.17 Cette constatation était répétée un demi-siècle après par l’abbé Leroy  : «  la mosaïque de la voûte est difficile à lire à cause de la saleté ».18 Dix ans plus tard, en 1968, alertant la communauté scientifique sur « l’état présent des monuments chrétiens du sud-est de la Turquie (Tur Abdin et environs) », le même abbé Leroy dressait un triste état des mosaïques : les tableaux des deux lunettes étaient ‘détruits’, écrivait-il, et « de larges pans [de la voûte] menacent de tomber d’un moment à l’autre ».19 Cependant Ernest J. W. Hawkins, qui a pu faire des observations précises sur place en novembre 1972, donne un état

Le décor du sanctuaire en était déjà connu grâce à une description transmise par un manuscrit syriaque du British Museum, l’Additionalis 17265 daté par Wright du XIIIe siècle8, complété par le manuscrit 221 de la Bibliothèque 1 Dirigée par Sébastien de Courtois, Alain Desreumaux et Patrick Blanc, la mission compte des restaurateurs spécialisés travaillant pour les musées et les monuments nationaux français, et des spécialistes de la mosaïque. L’initiative de cette intervention remonte à un projet de S. de Courtois, alors doctorant à l’EHESS, et d’A. Desreumaux (chercheur au CNRS, Laboratoire des études sémitiques anciennes de l’UMR 8167 Orient-Méditerranée, Institut d’études sémitiques du Collège de France), soutenu par le Ministère des Affaires étrangères, et notamment par M. Michel Pierre. À l’issue d’une première mission exploratoire effectuée en octobre 2006, un rapport initial sur l’état de conservation a été publié en 2009 (Blanc et al. 2009, pp. 5-19). Aujourd’hui, le programme de conservation et d’étude est supporté par la Fondation Leventis que nous tenons à remercier vivement pour son soutien, ainsi que Charalambos Bakirtzis et la Société d’études syriaques. Une présentation préliminaire de ces travaux a été donnée par P. Blanc et V. Blanc-Bijon, avec une introduction de S. de Courtois, lors de l’assemblée générale de la Société d’études syriaques, à Paris, le 1er mars 2013, présidée par A. Desreumaux. 2 Voir Leroy 1967. Le premier voyage relaté semble être celui du R. P. Badger, en 1842 (Hawkins et al. 1973, p. 282, n. 36). 3 Pognon 1907, p. 39. 4 Leroy 1956. 5 Preusser 1911, pp. 31-34. 6 Bell 1911, pp. 314-316; 1913, fig. 7-8; 1982, pp. 31-35, fig. 19 et 21-22 (avec notes de M. Mundell Mango, pp. 137-138). Déposées à l’Université de Newcastle-upon-Tyne, les archives de G. L. Bell conservent ses journaux en partie encore inédits et des photographies (concernant Mar Gabriel : M 151-152, 155-157, R 198-200), tandis que ses carnets sont à la Royal Geographical Society de Londres. 7 Bell 1982, p. 33. 8 Wright 1872, p. 1140, no DCCCCLXII.

Sachau 1899, pp. 581-587, no 179 (Sachau 221). Leroy 1956, avec traduction du texte syriaque pp. 77-78. 11 E. J. W. Hawkins examina les mosaïques et prit des clichés lors d’un court séjour au monastère en novembre 1972, alors que M. Mundell et C. Mango (certains clichés sont redevables à ce dernier) y ont séjourné en août de la même année. Deux autres chercheurs du Dumbarton Oaks ont été de l’un ou l’autre voyage et prirent des photographies, Robert Van Nice et Richard Anderson. 12 Hawkins et al. 1973. 13 Alors que notre mission de restauration se déroulait, et avec des clichés de ce travail, W. Jobst a présenté une nouvelle fois cet ensemble lors du colloque international de l’AIEMA à Venise en 2012, une communication qui a été publiée hors des actes du colloque : Jobst 2014. 14 Pognon 1907, no 13, p. 42, et no 14, p. 44. 15 Leroy 1956, p. 77. 16 Bell 1982, p. 33. 17 Bell 1982, p. 34 : « I do not think that the range of colours is large ». 18 Leroy 1956, p. 80. 19 Leroy 1968, p. 483. 9

10

368

Nouvelle intervention sur la mosaïque du sanctuaire de Mar Gabriel (Turquie orientale)

Fig. 1. Plan de l’église principale avec localisation du sanctuaire conservant les décors antiques (Hawkins et al. 1973, fig. 3).

de conservation des mosaïques plus précis,20 analysant également les vestiges de mortier antique.

nous avons procédé à des injections et des infiltrations de coulis de mortiers et à la pose de presses (fig. 5). Puis les lacunes ont été comblées par des mortiers de chaux teintés (fig. 6).

En 1997, l’intérieur de l’église a été « restauré » : les murs ont été mis à nu et les vestiges des anciens enduits pariétaux retirés. Les pierres ont alors été rejointoyées au mortier de ciment blanc. Les lacunes présentes à la retombée des mosaïques de la voûte ont été bloquées par un mortier beige, soulignant l’arrêt inférieur de la voûte. Antérieurement à notre venue avaient été mis en place des entoilages sur les zones les plus fragiles, avec poses ponctuelles d’une couche de gaze fixée à la mosaïque par un adhésif. Très détériorées, ces protections étaient encore en place lorsque débuta notre intervention en 2011 (fig. 3).

Le nettoyage a révélé de nombreux indices architecturaux : la présence d’accroches métalliques tant pour maintenir la mosaïque que pour suspendre des luminaires ; un orifice communiquant avec la nef centrale constituant un pot acoustique, et qui avait été installé dès la réalisation de la mosaïque ; etc. Ont pu également être observées de nombreuses traces du travail des mosaïstes : usage de mortiers allégés ; traitement en arêtes de poisson du support pour l’accroche des couches successives de mortier ; arrêts de mortier correspondant probablement à des journées de travail ; tracés peints préparatoires (fig.  7) dessinant les motifs soit à l’ocre rouge, soit en brun… La qualité technique des mosaïstes se révèle aussi par les jeux d’ombres et de lumière, apportés comme on le sait par les espaces entre les files de tesselles dans les fonds dorés, mais également, par exemple, par la disposition polychromique autour des étoiles où se juxtaposent des files de tesselles blanches,

Nous avons alors entrepris un nettoyage tesselle à tesselle qui a permis de retirer les siècles de concrétions, de poussières et de noirs de fumée (fig. 4). Sont alors réapparus pour la première fois dans leurs couleurs d’origine les détails de l’iconographie. Ces nettoyages ont été accompagnés de consolidations. Pour rétablir l’adhésion de la mosaïque et de son support à la paroi, 20

Hawkins et al. 1973, p. 283.

369

Patrick Blanc, Véronique Blanc-Bijon & Sébastien de Courtois

Fig. 2. Examen de l’état général de la voûte avant intervention (S. de Courtois).

argentées et noires, accentuant l’effet d’irisation de la figure ; usage de tesselles circulaires ou en gouttes ; effets de perspective ; etc.

des musées de France (C2RMF), dans le cadre d’un programme européen CHARISMA portant sur les liens entre l’or des monnaies et l’or des tesselles, et sur le verre utilisé pour ces tesselles, un programme mené par I. Biron, M. F. Guerra, M. Verità et E. Neri, laquelle a participé à l’une de nos missions21.

Les tesselles sont en pierre locale (blanc, crème, rosé, gris) ou en verre (verre opaque de couleurs très variées et vives, verre à feuille d’or ou d’argent dont le support en verre translucide offre également une large gamme de tonalités qui devaient contribuer à donner des reflets différents). Dans les lunettes, les tesselles d’or servant pour les fonds sont disposées en lignes régulières horizontales, très espacées, la face des tesselles légèrement basculée vers le bas et se détachant sur un bain de pose rehaussé de rouge foncé. Ainsi décalées et mises en exergue sur le bain de pose, les tesselles offrent un reflet plus accentué et accrochent mieux la lumière fournie par les luminaires. Plus prosaïquement, l’espacement des lignes permettait aussi d’économiser des tesselles. Ailleurs, et en particulier sur la voûte, les fonds dorés ne sont pas organisés de la sorte.

Selon l’étude des manuscrits proposée par François Nau, le sanctuaire aurait été « bâti et terminé et tous ces ornements admirables […] vinrent de la ville impériale l’an 823 (512) en laquelle Sévère fut fait patriarche d’Antioche ».22 Outre de l’or et des architectes, Anastase aurait également envoyé au monastère « des fabricants (ouvriers) de mosaïques ».23 Aussi, nous a-t-il semblé utile de tenter de déterminer l’origine exacte de ces mosaïstes et

21 Dans le cadre de son projet « AGLAOS – Analysis of Ancient Gold Leaf Tesserae and Gold Coins », alors qu’elle était en contrat postdoctoral au Labex ResMed, rattachée à l’UMR 8167 – Orient & Méditerranée. Voir Neri, Verità, Biron, Guerra, 2016. Voir aussi Neri, Verità 2013. 22 Nau 1907, p. 59. 23 Traduction de Leroy 1956, p. 77.

Un échantillonnage de ces tesselles a pu être analysé grâce aux moyens du Centre de recherches et de restauration 370

Nouvelle intervention sur la mosaïque du sanctuaire de Mar Gabriel (Turquie orientale)

Fig. 3. En 2006, des entoilages nombreux tentaient de prévenir l’effondrement des tesselles (P. Blanc).

Fig. 4. La voûte en cours de nettoyage (P. Blanc).

371

Patrick Blanc, Véronique Blanc-Bijon & Sébastien de Courtois des matériaux également pour confronter les hypothèses de datation. Ces analyses, effectuées au C2RMF et au laboratoire MONARIS de l’Université Paris VI avec les méthodes SEM/EDS et PIXE/PIGE, et l’étude en cours d’interprétation montrent que l’essentiel des tesselles en verre relève de technique au natron de type Levantin I, caractéristique des verres des Ve-VIe siècles en Orient, et l’on comparera utilement ces matériaux avec les résultats obtenus en particulier à Saint-Philippe de Hiérapolis.26 De plus, on observe que la recette du verre utilisée pour les tesselles diffère de celle en usage par les ateliers d’Antioche aux IIe-Ve siècles ; la réalisation de ce décor de Mar Gabriel ne semble donc pas devoir être mise en relation avec l’élection du patriarche Sévère comme on a pu l’écrire. Il faut toutefois signaler que le verre diffère également de celui utilisé au VIe siècle à Constantinople pour les tesselles de Saint-Polyeucte par exemple. À Mar Gabriel, l’analyse des feuilles d’or permet de proposer une datation de l’or entre 491 et 568, par comparaison avec les alliages des monnaies.27 Enfin, on voit une importante restauration très maladroite qui a obturé tardivement une large lacune dans la lunette sud. L’étude de certaines tesselles en verre a montré qu’une restauration a impliqué également des réfections des zones mosaïquées vers le milieu du XIe siècle, plus précisément vers 1040-1050. « Moi je suis le cep de la vigne, et vous en êtes les branches. Celui qui demeure en moi et en qui je demeure porte beaucoup de fruits ; car vous ne pouvez rien faire sans moi » (Jean 15.5). La voûte (fig.  8) est ornée d’un vaste rinceau de vigne issu de quatre grands vases d’angle, vases à décor de godrons et à doubles anses en S portés par un petit pied triangulaire, tous quatre différents (fig. 9). L’ensemble est limité par trois bordures distinctes : une ligne de grands et petits carrés sur pointe – faisant effet de ligne gemmée –, une ligne d’étoiles à huit pointes et une bande de chevrons emboîtés. Presque rigide et systématique vu de près, le dessin du rinceau offre au regard, vu de loin, un accroissement de l’élévation de la voûte réelle,28 confortée aussi par la bordure d’étoiles où la perspective visuelle ne permet pas de percevoir, depuis le sol, les importantes variations de modules mises en œuvre : en haut de la voûte les étoiles mesurent environ 35 cm, alors qu’elles ont 53 à 55 cm en bas. Il en est de même pour la bordure de chevrons.

Fig. 5. Pour rétablir le lien entre la mosaïque et le bâti, des infiltrations de mortier ont été nécessaires, maintenues par des presses le temps du séchage (P. Blanc).

de leurs matériaux24 : étaient-ils venus de Constantinople ? d’Antioche ? Par ailleurs, la datation même de la mosaïque a pu être discutée,25 d’où l’intérêt de procéder à des analyses

Cette voûte comporte trois médaillons ornés chacun d’une croix différente. Au centre de la voûte, au-dessus de l’autel perdu, mais décrit dans le manuscrit de Londres,

Au Centre Camille Jullian (UMR 7299), nous menons un programme de recherche sur « les mosaïstes et leurs matériaux », qui a été initié par un projet de recherche amorce lié aux Ateliers thématiques de recherche interdisciplinaires (ATRI) du LabexMed d’Aix Marseille Université, en 2014-2015. 25 Par exemple, dans le compte rendu de Leroy 1956 que l’on peut lire en Stern 1957, pp. 385-386, et dans lequel Henri Stern rapprochait le motif de piques décrit par Jules Leroy – qui l’attribue par erreur au sol, alors qu’il marque chaque étoile de la bordure de la voûte – des mosaïques du Dôme du Rocher à Jérusalem, plus tardives de trois siècles. 24

Neri, Biron, Verità, 2016. Cf. le rapport de mission produit par E. Neri en 2016: ‘Mar Gabriel (Turkey). A preliminary report about the analysis of glass tesserae by PIXE/PIGE’, 26 pp. Cette étude des tesselles en verre due à E. Neri sera publiée dans la monographie sur l’ensemble de l’intervention à Mar Gabriel que nous préparons. 28 À son point maximal, la voûte s’élève à 5,36 m au-dessus du sol. 26 27

372

Nouvelle intervention sur la mosaïque du sanctuaire de Mar Gabriel (Turquie orientale)

Fig. 6. Les petites lacunes de la voûte ont été comblées par des mortiers teintés (P. Blanc).

Fig. 7. L’examen détaillé de la mosaïque a permis de révéler diverses traces du travail des mosaïstes : ici, des tracés préparatoires en noir, brun et ocre rouge (V. Blanc-Bijon).

373

Patrick Blanc, Véronique Blanc-Bijon & Sébastien de Courtois

Fig. 8. La voûte, ici en cours de restauration, présente un vaste rinceau de vigne interrompu par trois médaillons (P. Blanc).

Fig. 9. Le rinceau jaillit de vases gemmés placés aux quatre angles de la voûte (V. Blanc-Bijon).

374

Nouvelle intervention sur la mosaïque du sanctuaire de Mar Gabriel (Turquie orientale)

Fig. 10. Le médaillon situé à l’est (P. Blanc).

une grande croix dorée et gemmée se détache sur fond de ciel étoilé irradiant de rayons argentés, bordé par une ligne ondée de calices trifides rehaussée de tesselles à feuilles d’or et d’argent, bordure très commune dans les mosaïques de pavement des églises de Syrie et de Palestine aux Ve-VIIe siècles. Cette croix sur fond de rinceau de vigne serait le prototype de la croix triomphale de la Passion.29

d’églises en Syrie. Ces ciboria abritant souvent la croix peuvent être clos par des rideaux, des chancels, des portes mêmes.30 Ici point de fermeture, les entrecolonnements sont ouverts. Aux chapiteaux latéraux de chaque édicule sont suspendus des luminaires globuleux. Sous le ciborium sud est conservé un caisson richement décoré de gemmes, peut-être le tabernacle, sur lequel reposent les éléments de l’Eucharistie : deux calices entourant une panière de pain (fig.  11). Au bas de cette même lunette sud, l’inscription publiée par Cyril Mango a pu être retrouvée. Elle est écrite en caractères grecs, élément confirmant une datation au VIe siècle au plus tard. Cyril Mango y voyait les débuts de la signature des mosaïstes (έγένετο ϊ μουσομᾳ[---).31 En 2013, elle a été relue par Ergün Lafli et Hadrien Bru qui la rapprochent de la citation de Luc 22.19 : « Et Jésus dit: ceci est mon corps...  » (έγένετο Ι(ησοῦς) μου σῶμᾳ[--- ).32 Ce commentaire – qui n’est, rappelons-le, que le début d’une ligne de texte dont il manque la plus grande partie, près

À l’ouest et à l’est, deux médaillons plus petits : à l’ouest, la croix sur fond rouge est posée sur trois degrés et encadrée par une paire d’ailes bleues, croix de l’Incarnation. À l’est, le médaillon à fond blanc est incomplet, et toute une partie des tesselles s’est détachée, montrant les vestiges d’un tracé préparatoire qui paraît attester du podium à degrés sur lequel s’élevait la croix ; sous le bras gauche de la croix est encore visible le dessin de suspensions de perles (fig. 10). Dans les lunettes nord et sud (la lunette nord est malheureusement considérablement détruite) prend place un tout autre décor : dans des jardins boisés, paradisiaques – le sanctuaire est image du Paradis –, est figuré un édicule dont le dôme est très nettement différencié d’une lunette à l’autre par le traitement de sa couverture, à nervures bleu foncé et argenté ou en écailles dorées. De semblables édicules se retrouvent dans une petite série de pavements 29

30 On renverra à l’étude d’un panneau en mosaïque du musée de Hama (Paribeni 1993, avec parallèles), et l’on rappellera aussi le fragment entré en 2000 dans les collections du Musée du Louvre, Département des antiquités grecques, étrusques et romaines (inv. MNE 1187 = MA 5093), connu depuis 1969 (Metzger 2004-05, pp. 154-156, fig. 1; cf. aussi Giroire 2012, pp. 302 et 307, et fig. 296 ; nous tenons à remercier vivement Catherine Metzger de nous avoir rappelé cet exemple). 31 Hawkins et al. 1973, p. 281, fig. 39, et p. 296. 32 Lafli, Bru 2013, pp.  301-303. Voir aussi Bru 2013 (nous remercions vivement l’auteur qui a bien voulu nous adresser copie de son article).

Thierry 1972, p. 248.

375

Patrick Blanc, Véronique Blanc-Bijon & Sébastien de Courtois

Fig. 11. La lunette sud, en cours de restauration (P. Blanc).

des trois-quarts – est ainsi en accord parfait avec l’image de l’Eucharistie qu’il souligne.33

entre les terres de l’empereur et le monde extérieur à l’empire.

Pour conclure momentanément, les travaux de nettoyage ont donc permis de redonner son lustre à cet ensemble ornant la voûte du sanctuaire de l’église principale de Mar Gabriel, mais aussi de préciser un certain nombre de points tant techniques qu’iconographiques.

Ces travaux sont encore en cours et nous procédons depuis 2018 au traitement du sol en opus sectile. Une fois les missions terminées paraîtra une publication nouvelle, monographique, actuellement en préparation par notre équipe.

À partir des éléments de chronologie réunis par l’étude de M. Mundell Mango, principalement le rapprochement du manuscrit de Londres mentionnant les dons faits au monastère monophysite par l’empereur Anastase qui finança, dit le texte, la construction de cette église et envoya au monastère de l’or ainsi que ses meilleurs architectes et artisans, orfèvres, sculpteurs, peintres et mosaïstes, la datation habituellement retenue était de 512.

Bibliographie Abbréviations : CA = Cahiers archéologiques. CRAI = Comptes rendus des séances de l’Académie des Inscriptions et Belles-Lettres. JAS = Journal of Archaeological Science.

Grâce aux études techniques nouvelles, cette datation paraît confortée, l’or mis en œuvre dans la majorité des tesselles examinées pouvant correspondre à celui des monnaies émises sous le règne d’Anastase. La question toutefois de l’origine de l’or utilisé pour les tesselles se pose de manière particulière dans le cas qui nous concerne, puisque c’est l’empereur lui-même qui offre son or et envoie ses orfèvres et ses mosaïstes au monastère. Si le texte doit être pris à la lettre, il faudrait alors comprendre que les matériaux expédiés depuis Constantinople ont été travaillés sur place et qu’il ne serait peut-être pas question de refonte de monnaies.

Bell G., 1910. ‘The churches and monasteries of the Tur’Abdin’, in M. van Berchem, J. Strzygowski, G. L. Bell, Amida, Heidelberg/Paris, pp. 224-262. Bell G. L., 1911. Amurath to Amurath, Londres/New York. Bell G. L., 1913. Churches and Monasteries of the Tur-’Abdin and Neighbouring Districts, Heidelberg (Zeitschriften für Geschichte der Architektur. Beiheft, 9). Bell G. L., 1982. The Churches and Monasteries of the Tur ’Abdin, introd. de M. M. Mango, Londres. Blanc P., Desreumaux A., de Courtois S., 2009. ‘Report on the state of preservation of the Byzantine mosaics of the Saint Gabriel monastery of Qartamin, Tur Abdin (South-West Turkey. October 10th-14th, 2006’, in Hugoye. Journal of Syriac Studies, 12/1, pp. 5-19.

Par ailleurs, il n’est pas inutile de rappeler la situation même de ce monastère, sur la frontière orientale de l’empire ; le monastère et son décor constituent alors un véritable manifeste impérial destiné à marquer les limites

Bru H., 2013. ‘La mosaïque de l’eucharistie au monastère Mor Gabriel de Qartamin (Tur ‘Abdin)’, in R. Lebrun,

33 Bru 2013. Après un réexamen qu’il nous a été possible de faire lors d’une récente mission, cette inscription sera revue par A. Desreumaux dans la publication en préparation.

376

Nouvelle intervention sur la mosaïque du sanctuaire de Mar Gabriel (Turquie orientale) A. Degrève (éd.), « Deus medicus ». Actes du colloque organisé à Louvain-la-Neuve, 15-16 juin 2012, Turnhout (Homo Religiosus, Série II, 12), pp. 9-19.

Paribeni A., 1993. ‘Un mosaico con rappresentazione architettonica nel museo di Hama’, in R. Farioli Campanati (éd.), XL Corso di cultura sull’arte ravennate e bizantina. Seminario internazionale di studi, « L’Albania dal Tardoantico al Medioevo, aspetti e problemi di archeologia e storia dell’arte », Ravenna 29 aprile-5 maggio 1993. 1° Colloquio dell’Associazione italiana per lo studio e la conservazione del mosaico, Ravenne, pp. 615-641.

de Courtois S. (éd.), 2004. Les derniers Araméens. Le peuple oublié de Jésus, Paris. de Courtois S., 2017 (éd.). ‘Sauvetage d’une mosaïque byzantine’, in Le Monde de la Bible, 221, pp. 74-77. Giroire C., 2012. ‘Émergence et diffusion du christianisme’, in N. Bel, C. Giroire, F. Gombert-Meurice, M.H. Rutschowscaya (dir.), L’Orient romain et byzantin au Louvre, Arles/Paris, pp. 300-307.

Pognon H. (éd.), 1907. Inscriptions sémitiques de la Syrie, de la Mésopotamie et de la région de Mossoul, Paris. Preusser C., 1911. Nordmesopotamische Baudenkmäler altchristlicher und islamischer Zeit, Leipzig (Wissenschaftliche Veröffentlichung der Deutschen Orient-Gesellschaft, 17).

Grabar A., 1956. ‘Quelques observations sur le décor de l’église de Qartamin’, in CA, 8, pp. 83-91. Hawkins E. J. W., Mundell M. C., Mango C., 1973. ‘The mosaics of the monastery of Mār Samuel, Mār Simeon, and Mār Gabriel near Kartmin with a note on the Greek inscription’, in Dumbarton Oaks Papers, 27, pp. 279-294.

Sachau E., 1899. Verzeichniss der syrischen Handschriften der Königlichen Bibliothek zu Berlin, Berlin (Die Handschriften-Verzeichnisse der Königlichen Bibliothek zu Berlin, 23).

Jobst W., 2014. ‘Wandmosaik und Opus Sectile-Paviment der Grossen Klosterkirche von Mor Gabriel (Tur Abdin/ Türkei)’, in C.  Şimşek, B.  Duman, E.  Konakçı (éd.), Mustafa Büyükkolancı’ya Armağan. Essays in Honour of Mustafa Büyükkolancı, Istanbul, pp. 345-360.

Stern H., 1957. ‘Compte rendu : Cahiers archéologiques, 8, 1956’, in Syria, 34/3-4, pp. 385-387. Thierry N., 1972. ‘Art byzantin du Haut Moyen Âge en Cappadoce : l’église no 3 de Mavrucan’, in Journal des Savants, 1972/4, pp. 233-269.

Lafli E., Bru H., 2013. ‘Chronique d’Orient. Chronique 2013’, in Dialogues d’histoire ancienne, 39/2, pp. 265-315.

Wright W., 1872. Catalogue of the Syriac Manuscripts in the British Museum acquired since the Year 1838, III, Londres.

Leroy J., 1956. ‘Le décor de l’église du monastère de Qartamin d’après un texte syriaque’, in CA, 8, pp. 75-81. Leroy J., 1967. ‘Recherches archéologiques sur les églises du Tur Abdin’, in CRAI, 111/2, pp. 324-333. Leroy J., 1968. ‘L’état présent des monuments chrétiens du Sud-Est de la Turquie (Tur Abdin et environs)’, in CRAI, 112/4, pp. 478-493. Metzger C., 2004-05. ‘Deux panneaux de mosaïque d’église du Musée du Louvre’, in Bulletin de la Société nationale des Antiquaires de France, pp. 154-158. Nau F., 1907. ‘Notice historique sur le monastère de Qartamin, suivie d’une note sur le monastère de Qennesré’, in Actes du XIVe Congrès international des Orientalistes, Alger 1905, II, Paris, pp. 37-135. Neri E., Biron I., Verità M., 2017. ‘New insights into Byzantine glass technology from loose mosaic tesserae from Hierapolis (Turkey): PIXE/PIGE and EPMA analyses’, in Archaeological and Anthropological Sciences, 10/7, pp. 1751-1768. Neri E., Verità M., 2013. ‘Glass and metal analyses of gold leaf tesserae from 1st to 9th century mosaics. A contribution to technological and chronological knowledge’, in JAS, 40/12, pp. 4596-4606. Neri E., Verità M., Biron I., Guerra M. F., 2016. ‘Glass and gold: analyses of 4th-12th centuries Levantine mosaic tesserae. A contribution to technological and chronological knowledge’, in JAS, 70, pp. 158-171. 377

30 Recent Restoration Work in the Hagia Sophia Museum Asnu-Bilban Yalçın* İstanbul Üniversitesi / Istanbul University (Turkey) (with the collaboration of Bilgen Deveci, Defne Tekay and Miraycan Kirkoçoğlu)** Many archaeological works, due first to the new metro lines, and then to the ‘urban renovation’ of the old city buildings, are ongoing in Istanbul. Alongside these, there are many restorations carried out on the historical buildings, where, in most cases, archaeological investigation also takes place, which yields surprising discoveries. Such is the case of Hagia Sophia, one of the most important monuments of the world’s cultural heritage, thanks to various repair works carried out during different periods throughout its long history. The latest works began in 2012 and are ongoing. In the first stage, works were planned for the western façade, the decorated surfaces of the interior and the galleries. They have led, so far, to archaeological finds in the main pavement and in that of the galleries. This paper gives a short presentation of the latest restorations and the archaeological investigations carried out alongside them. De nombreux travaux archéologiques se poursuivent à Istanbul, dus en premier lieu aux nouvelles lignes de métro et à la rénovation des bâtiments de la vieille ville. En parallèle, de nombreux bâtiments historiques font l’objet de restaurations auxquelles s’ajoutent, dans la plupart des cas, des travaux archéologiques intérieurs ou extérieurs qui permettent des découvertes surprenantes. Tel est le cas de Sainte-Sophie, l’un des monuments les plus importants du patrimoine culturel mondial, qui a été préservé jusqu’à aujourd’hui grâce à diverses réparations effectuées tout au long de sa longue histoire. Les travaux les plus récents ont débuté en 2012 et se poursuivent toujours. Prévus dans un premier temps sur la façade ouest, sur les surfaces décorées de l’intérieur et dans les galeries, ils ont mené à des découvertes archéologiques dans le pavement principal et dans celui des galeries. Cet article présente brièvement les dernières restaurations et les travaux archéologiques en cours. Viele archäologische Arbeiten finden in Istanbul statt, vor allem im Zuge des Baus der neuen U-Bahnlinien aber auch im Zuge der Stadtrenovierung der älteren Bausubstanzen. Begleitet wird dies durch Restaurierungsarbeiten an den historischen Gebäuden. In den meisten Fällen findet auch eine archäologische Begleitung der Arbeiten in und um die Bauten statt, was zu interessanten Entdeckungen führen kann, wie auch im Falle der Hagia Sophia, einem der wichtigsten Denkmäler des Weltkulturerbes. Die über die Jahrhunderte entstandenen Schichten von Neuerungen und Reparaturen dieses Baus werden bis heute erforscht. Die neusten Arbeiten begannen 2012 und dauern noch an. Im ersten Schritt wurden Eingriffe an der Westfassade, an den Verzierungen im Gebäudeinneren und in den Galerien vorgenommen. Diese führten zu archäologischen Funden in der Hauptpflasterung und in den Galerien. Der vorliegende Beitrag gibt eine kurze Vorstellung dieser Neufunde im Zuge der aktuellen Restaurierungs- und Ausgrabungsarbeiten. Diverse indagini archeologiche, dovute in primo luogo ai lavori delle nuove linee della metropolitana, ma anche al “rinnovamento urbano” degli edifici della vecchia città, sono attualmente in corso ad Istanbul. Accanto ad esse, anche i numerosi restauri degli edifici storici, ai quali, nella maggior parte dei casi, si aggiungono saggi di scavo all’interno o all’esterno, conducono a scoperte sorprendenti. Questo è il caso di Santa Sofia, uno dei monumenti più importanti del patrimonio culturale mondiale, che ha potuto giungere fino ai giorni nostri grazie a vari lavori di manutenzione effettuati in diversi periodi nel corso della sua lunga storia. I lavori più recenti sono iniziati nel 2012 e sono attualmente in corso. In primo luogo, sono stati programmati I am indebted to H. Cengiz, Director of the Hagia Sophia Museum, who permitted the presentation and the publication of this preliminary report on the ongoing restorations.

*

Ayasofya Müzesi Müdürlüğü / Directorate of the Hagia Sophia Museum.

**

379

Asnu-Bilban Yalçın lavori sulla facciata occidentale, sulle superfici decorate dell’interno e nelle gallerie. Lavori che hanno portato, per ora, a importanti ritrovamenti nella pavimentazione principale e nelle gallerie. Questo articolo offre una breve presentazione degli ultimi restauri e delle indagini archeologiche all’interno del monumento. At present, the main maintenance and repair works are being carried out on the top lead covering, exterior façades, interior decorated surfaces, marble surfaces, and mosaic ornaments (fig. 1). The latest restorations began in 2012. In the first stage, works were planned for the western façade, the decorated surfaces of the interior, and the galleries.

Any archaeological work carried out in Istanbul is an event important not only for the specialist in the field but also for the worldwide public. Recently, Istanbul has been the scene of many archaeological activities, due first to the new metro lines, and then to the ‘urban renovation’ of old city buildings. All of these are rescue excavations carried out by the Archaeological Museums of Istanbul. Only two excavations have been scientifically undertaken by academic institutions: the Yoros Fortress excavation by Istanbul University and the Küçükçekmece Lake Basin excavation by Kocaeli University; both are discovering interesting finds from the Byzantine period.

The western façade One of the priorities of the Hagia Sophia Scientific Committee is the improvement of the façades. The present external plaster coating is from different periods; much of it is a cement-based plaster made in 1955–1956. Many repairs intervened after that, until 1986, when the entire exterior was coloured as it is today.

In addition to this field work, many restorations have taken place in historical buildings. In most cases, when the structural consolidation of walls and pavements becomes necessary, archaeological investigation is carried out in and around those walls and floors. Thus, the opportunity arises to discover original levels and the material used in them.

According to a decision made in 1993 during scientific consultation, the damaging plaster was to be removed and replaced with an original type of mortar or with plaster, depending on the condition of the masonry surface.5 The first step towards the restoration work on the western façade was to remove the cement-based plaster coating. Between 2007 and 2008, the original surface was uncovered. The entire western façade was scanned with a 3D laser device, and a survey was completed for the following restoration projects, which began in 2012 (fig. 2).6

Such is the case of Hagia Sophia, one of the most important monuments of the world’s cultural heritage, in which various repair works have been carried out during different periods throughout its long history.1 Since 1993, a scientific committee has been working for the Hagia Sophia Museum in order to evaluate the opinions and proposals that will form the basis of restoration work to be done within an interdisciplinary scientific framework.2

During work on the western façade, an interesting piece of evidence was discovered: the iron beam that had been placed by the Fossati brothers during the restorations between 1847 and 1849 (fig. 3). The Fossatis encircled the dome with a large iron cable in order to strengthen it; they then removed the small flying buttresses over the squinchlike arches between the buttress arms and the base of the dome.7 These small buttresses were probably constructed in the early Ottoman period, as visible in the drawings of Melchior Lorichs dated c.1559.8

Of the works undertaken in recent decades, those carried out between 1993 and 2010 on the main dome mosaics are of particular importance.3 During the restoration work on the north-east semi-dome, the angel figure in the northeast pendentive, whose face had been covered during the Fossati restorations (1847–1849), was also conserved.4 The works on the dome and the angel mosaic were completed in 2010. Within the large bibliography existing on Hagia Sophia, few items deal specifically with the restoration works. For a general view on the repairs and additions during Byzantine and Ottoman times, see Mainstone 1988. On the late Ottoman and the Republican-era restorations, a recent doctoral thesis has been published, based on the archive documents; see Diker 2016. 2 According to the decision of the Turkish Ministry of Culture of 3 May 1993, no 1711. The academic members are: Mustafa Erdik (Engineer, Bosporus University), Zeynep Ahunbay (Architect, Istanbul Technical University), Nevzat İlhan (Architect, Istanbul Yildiz University), AsnuBilban Yalçın (Byzantine art historian, Istanbul University), Selçuk Mülayim (Ottoman art historian, Marmara University), Ahmet Güleç (Restorator, Istanbul University), Hayrullah Cengiz (Director of the Hagia Sophia Museum), Salman Ünlügedik (Directorate of Surveying and Monuments of Istanbul), Ali Osman Avşar (Directorate of Central and Regional Laboratory for Restoration and Conservation of Istanbul). 3 The work was completed in five stages: north-east dome (1993–1998), north-west dome (1998–2002), south-west dome (2003–2004), southeast dome (2004–2009), and north-east dome (2009–2010). For the dome restorations, see Ozil 2000, pp. 66–69. 4 Schlüter 1999; 2000, pp. 62–65. 1

The uncovered iron beam, measuring c. 8 x 8 cm, continues on the upper level of the base of the dome. After rust was cleaned from the surfaces of the beam, protecting material was laid, and it was re-covered with mortar of the original type. In the gallery level and on the walls of the narthex, a Byzantine type of mortar was applied, while in the lower Ahunbay Z. 2010, p. 62. According to these principles, improvement works were carried out on the surfaces of the south-west corner and on the eastern side, under the dome structure. 6 We are indebted to the architect M. E. Çoşkun, responsible for the works on the western façade, in an account of the Gökalp project company by the permission of the Governorate of Istanbul, for the information. 7 Mainstone 1988, p. 103. 8 Oberhummer 1902, pl. VI. Mainstone describes them as ‘erected at an indeterminate date’, while according to Necipoğlu a large structural intervention was made by Selim II between 1572 and 1574. See Necipoğlu 1992, pp. 205–207. 5

380

Recent Restoration Work in the Hagia Sophia Museum

Fig. 1. General view of the upper levels, from the south-east (Archive of the Ayasofya Müzesi).

Fig. 2. Survey of the western façade (Archive of the Ayasofya Müzesi).

381

Asnu-Bilban Yalçın

Fig. 3. Western façade, Fossati’s iron beam (Archive of the Ayasofya Müzesi).

parts, on the Ottoman additions and on the buttresses, the traditional mortar was replaced.9

joins were carried out, as well as in the marble window grilles and the damaged stone and brick wall structure. The restoration on the façades of the building is ongoing, with the preparation of projects for the remaining south, north and east sides.12

Another work on the western façade was the restoration of the brick dogtooth frieze course. This frieze had been largely re-created with cement at various unknown times; in fact, it was a recreation of what had probably been made by the Fossatis (fig. 4). The decision of the Scientific Committee was to remove the cement-based dogtooth course, replacing it with one made of brick (fig. 5), thus keeping the one frieze course probably of the Fossatis’ time.

The interior: The north-west half of the tympanum wall The works on the northern tympanum began in 2012 with the construction of scaffolding, followed by 3D laser scanning of the western half of the wall (fig. 7).13 The documentation with the survey revealed the material of the wall structure14 and made possible further analysis of damage.

As the most visible and therefore representative part, the western façade was originally covered with marble slabs,10 some of which are still in situ. As part of the project, the surviving slabs were removed for conservation and then replaced in their original position (fig. 6).11 In addition to the applications mentioned on the western façade, repairs to the Ottoman period lead coverings and in the wooden

On the first part of the tympanum wall (area no 1), the cement-based plasters were scraped off, after a square-bysquare mapping of the surface (fig. 8), for a more detailed documentation, in order to prepare for a second survey

Although, at the beginning of the projects, the principle adopted was to leave visible the original brick wall masonry, this idea was abandoned because of the greater damage caused by the south-west wind that often blows in the city. See Ahunbay M. 2010, p. 87. For this reason, after the consolidation around the dome and the works on the masonry, the surface around the dome was again covered with mortar. 10 Mango 1974, p. 68. 11 The marble slabs were removed in 2016, and work on their conservation is ongoing.

When this article was submitted, the 3D laser scanning and surveying of the remaining façades had been completed. 13 We thank the architect-restorator K. Sevim (responsible from the company Aslı Project, by the permission of the Governorate of Istanbul) for the detailed information. 14 Identification of stone material and analyses of different plaster layers and types of wood were made by the Central and Regional Laboratory for Restoration and Conservation of Istanbul; Chemical and Petro-Chemical Report no 861 of 5 September 2014.

9

12

382

Recent Restoration Work in the Hagia Sophia Museum

Fig. 4. Dogtooth course frieze (Antoniades 1907).

(fig. 9). The detailed scanning with 3D made possible the laboratory analysis of material samples (mortar, plaster, wood15 and metal). The different scales and types of damage were included in the survey.

works on the marbles of the main floor began in the southeast part and are still continuing as of 2017. The works are proceeding systematically, with the museum visitor’s route being taken into consideration. Detailed documentation work was completed before the consolidation of the marbles in the floor. The documentation indicates approximately 3,000 marble floor slabs, including broken and cracked pieces.

The scraping also made visible the reworking of the pointed arched windows of the Ottoman period, as the red painted fillet on the limestone blocks and the wooden beams fitted into them were preserved in situ. Mosaic remnants were found on the lower levels of the tympanum wall and, after being documented, were protected by a facing application (fig. 10).

Given the results of this documentation, the protection and consolidation works were started in 2016; these include renovation of the joints, the removal and replacement of unsteady plaques for correction of the floor level, reinforcement of the immovable plates with mortar injection, and integration of missing parts (fig. 11).

As of April 2017, works in the interior are continuing; after decisions have been made about intervention, restoration work will be carried out on the other half of the northern tympanum wall.

The differences and varying formations visible at various points in the main floor required further archaeological investigations16 within the Museum. In fact, during these consolidation works, in the north-east area, the removal of a circular porphyry plaque, numbered 1934 in the survey, showed a tiny dark soil layer. As research continued, numerous glass tesserae were found mixed with the soil (fig. 12). At a deeper level, glass fragments, glazed

Consolidation of the marble floor of the main nave and archaeological investigations Among the interior repairs is also included the consolidation of the marbles of the floor in the main nave. Strengthening 15 Dendrochronological studies and analyses of different types of wood were carried out by Istanbul University, Faculty of Forestry, Department of Forest Botanics. A first preliminary report was presented to the Scientific Committee on 29 July 2016.

The work was carried out by the archaeologist D.Tekay of the Hagia Sophia Museum.

16

383

Asnu-Bilban Yalçın

Fig. 5. Brick dogtooth course frieze, after restoration (Archive of the Ayasofya Müzesi).

Fig. 6. Western façade, marble revetment before and after removal (Archive of the Ayasofya Müzesi).

384

Recent Restoration Work in the Hagia Sophia Museum

Fig. 7. Interior, northern tympanum wall survey before rasping (Archive of the Ayasofya Müzesi).

potsherds, a glass lamp handle, some lead fragments, fine bone fragments, pearl fragments, a piece of white marble and a small amount of charcoal came to light (fig. 13). The circular space, with a compressed terracotta base, has a diameter of c. 26–28 cm, a depth of 26.5 cm, and a thickness of 5.5 cm. After its cleaning, it became clear that the space was a specially prepared structure in the form of a terracotta pot.

found inside it. The second terracotta feature in this area, between marble fragments numbered 2275–2274–2276, measures 25 cm x 26 cm. The depth is 24.5 cm. A large number of glass tesserae were found inside it as well. These features show some small differences: the first one was made of terracotta, specifically to fit in that space. By contrast, the two in the south-east, in the mihrab, have an irregular circular shape made with mortar. Except that they all contained a large number of glass tesserae, there are not many common features among the three opened areas.

Because this was the first such discovery inside Hagia Sophia, we did not know its specific function. It became necessary to search for other possible examples of such terracotta features beneath the pavement. For this purpose, further investigation was carried out on the marble floor in the southern part of the mihrab, where two round marble plaques were selected. The first of these two terracotta features has a diameter of 28–30 cm and a depth of 23.5 cm (fig. 14). It was located between the marble fragments numbered 2264–2265–2266 in the survey. After the thin soil layer inside the area was removed, it was seen that the irregular space had been partly cut into the rock and partly formed with white mortar. Numerous glass tesserae were

Again, during the same consolidation work on the marble pavement, some glass tesserae were noticed in the mortar between marble plaques near the south-east mahfil of the muezzin, specifically between the marble fragments numbered 2751–2752–2753. After removal of marble pieces no 2752 and 2753 (fig. 15), the investigation advanced when another circular area was found beneath the pavement. Here a cylinder-shaped, glazed terracotta structure, with a diameter of 27 cm and a depth of 28 cm, contained small marble fragments and soil. This terracotta ‘pot’ had been placed inside a bed of solid mortar. When 385

Asnu-Bilban Yalçın

Fig. 8. Interior, northern tympanum wall, mapping of area no 1 (Archive of the Ayasofya Müzesi).

the soil around this ‘pot’ was removed, several pieces of worked marble blocks came to light; perhaps they had been reused as part of the substructure of the floor (fig. 16). Many glass tesserae, two small metal fragments, small bone fragments and glazed potsherds were also found in the compressed soil.

archeological investigations inside, on the pavement. For this reason, the discoveries of 2016 described above are important and relevant for the history of the building. The galleries

It is very difficult to determine the exact function of these features,17 given that we do not have much information about archeological investigations inside the building after it was transformed into a museum. The excavations of A. M. Schneider and of M. Ramazanoğlu outside the building are well known,18 but no documentation exists about

The projects on the gallery floor and the south-west ramp began in 2014,19 with the preparation of a survey after 3D laser scanning. The survey project included detailed analysis of materials, description of damage and differentiation of chronological periods. Groundpenetrating radar (Georadar) scanning was carried out on the walls (fig. 17) and on the floor (fig. 18), in order to

17 The member of the Scientific Committee of the Hagia Sophia Museum, engineer M. Erdik, suggests that these areas may be special spaces for the accumulation of water in the plaster. However, their function remains unknown, pending further discoveries. 18 Schneider 1941; Ramazanoğlu 1946.

19 We thank the architect A. C. Aydın (responsible from the company Akademi Project, by the permission of the Governorate of Istanbul) for the information on the project. During the preparations, because of the extensive deterioration of the marble plaques of the pavement in the galleries, a temporary route was created for visitors.

386

Recent Restoration Work in the Hagia Sophia Museum

Fig. 9. Interior, northern tympanum wall, area no 1 after rasping, survey of material (Archive of the Ayasofya Müzesi).

Fig. 10. Interior, northern tympanum wall, area no 1, mosaic remnants (Archive of the Ayasofya Müzesi).

387

Asnu-Bilban Yalçın

Fig. 11. Main nave floor, reinforcement of marble plaques (Archive of the Ayasofya Müzesi).

388

Recent Restoration Work in the Hagia Sophia Museum

Fig. 12. Main nave floor, north-east area, research under marble plaque no 1934 (Archive of the Ayasofya Müzesi).

Fig. 13. Main nave floor, north-east area, research under marble plaque no 1934, finds (Archive of the Ayasofya Müzesi).

389

Asnu-Bilban Yalçın

Fig. 14. Main nave floor, south-east area, research under marble plaques no 2264–2265–2266 (Archive of the Ayasofya Müzesi).

determine risky areas and damage caused by humidity. Based on these results, the Scientific Committee decided to make further investigations under the floor of the galleries (fig. 19–20). These and consolidation works in the galleries began recently and are ongoing as of April 2017. These conservation efforts – just begun and still ongoing – are of considerable importance because they have resulted in the discovery of many unknown parts of this exceptional building, such as the ‘rooms’ under the gallery pavement, still under excavation, and the works in the apse. Since Hagia Sophia is the primary monument in Turkey that is part of the world’s heritage, its study and preservation deserve a special importance and care. We cannot exclude the importance of the data obtained from these investigations that occur in an occasional and surprising way; the archaeological material helps not only in the dating but also in understanding the context of the discovery. In a complex building like Hagia Sophia, it is not always easy to understand the stratifications. Because various interventions were made in different periods, we are sometimes creating a reverse chronology, beginning

Fig. 15. Main nave floor, area of the south-east mahfil of the muezzin, removal of the marble plaques (Archive of the Ayasofya Müzesi).

390

Recent Restoration Work in the Hagia Sophia Museum

Fig. 16. Main nave floor, area of the south-east mahfil of the muezzin, after the removal of the marble plaques (Archive of the Ayasofya Müzesi).

Fig. 17. Western gallery, georadar investigation on the walls and risk map (Archive of the Ayasofya Müzesi).

391

Asnu-Bilban Yalçın

Fig. 18. Southern gallery, georadar investigation on the floor (Archive of the Ayasofya Müzesi).

Fig. 19. North gallery, underground research spot (no 7) (Archive of the Ayasofya Müzesi).

392

Recent Restoration Work in the Hagia Sophia Museum

Fig. 20. South gallery, underground research spot (no 8) (Archive of the Ayasofya Müzesi).

1999, Mantua Casa del Mantegna 14 Kasm–31 Aralk 1999, Istanbul.

with our recent projects and continuing back through the centuries to the original construction of the building. For now, however, the sub-structure of the sixth-century floor remains an important discovery for our understanding of the raising of the floor after the destruction of the second church, whose level coincides with that of the monumental entrance excavated by Schneider in 1935.

Mainstone R., 1988. Hagia Sophia. Architecture, Structure and Liturgy of Justinian’s Great Church, London. Mango C., 1974. Architettura Bizantina, Milan (Storia universale dell’architettura). Necipoğlu G., 1992. ‘The life on an Imperial monument: Hagia Sophia after Byzantium’, in R. Mark and A. Ş. Çakmak (eds.), Hagia Sophia from the Age of Justinian to the Present, Cambridge, pp. 195–225.

Bibliography Abbreviation: AMY = Ayasofya Müzesi Yıllıkları – Annual of Ayasofya Museum.

Oberhummer, E. (ed.), 1902. Konstantinopel unter Sultan Suleiman dem Grossen. aufgenommen im Jahre 1559 durch Melchior Lorichs aus Flensbur, Munich.

Ahunbay M., 2010. ‘Ayasofya Kubbesinin Kasnak Doğu Cephesi’, in AMY, 13 (= Ayasofya Müsezi Yayınları, 16), pp. 79–100.

Ozil R., 2000. ‘Ayasofya’nin kubbe mozaiklerinde koruma ve onarim uygulamalari’, in Hoffmann, Schlüter, Kangal 2000, pp. 66–69.

Ahunbay Z., 2010. ‘Ayasofya’nın Dünya Mirası Değerlerinin Korunması ve Yönetimi’, in AMY, 13 (= Ayasofya Müsezi Yayınları, 16), pp. 57–78.

Ramazanoğlu M., 1946. Sentiren ve Ayasofyalar Manzumesi – L’ensemble Ste Irène et les diverses Ste Sophie, Istanbul.

Antoniades E. M., 1907. Έκφρασις της Αγίας Σοφίας, ήτοι μελέτη συνθετική και αναλυτική υπό έποψιν αρχιτεκτονικήν, αρχαιολογικήν και ιστορικήν του πολυθρυλήτου τεμένους Κωνσταντινουπόλεως, I, Paris/ Athens (Βιβλιοθήκη Μαρασλή).

Schlüter S., 1999. Gaspare Fossatis Restaurierung der Hagia Sophia in Istanbul 1847–49, Bern (Neue Berner Schriften zur Kunst, 6). Schlüter S., 2000. ‘Gaspare Fossati’nin Ayasofya onarımları (1847–49)’, in Hoffmann, Schlüter, Kangal 2000, pp. 62–65.

Diker H. F., 2016. Ayasofya ve Onarımlari, Istanbul (Fatih Sultan Mehmet Vakıf Üniversitesi Yayınları, 21). Hoffmann V., Schlüter S., Kangal S. (eds.), 600 Yıllık Ayasofya görünümleri ve 1847–49 Fossati restorasyonu, Bern, Historisches Museum 12 Mays–11 Haziran 1999, Stendal, Winckelmann-Museum 24 Haziran–26 Eylül

Schneider A. M., 1941. Die Grabung im Westhof der Sophienkirche zu Istanbul, Berlin (Istanbuler Forschungen, 12).

393

B A R I N T E R NAT I O NA L S E R I E S 2 9 7 3

2020

Published in memory of Prof. Claudia Barsanti, Archaeology of a World of Changes provides a selection of papers presented in sessions on Late Roman and Early Byzantine archaeology, architecture, sculpture and landscapes of the 23rd International Congress of Byzantine Studies, “Byzantium – A World of Changes” (Belgrade, 22-27 August 2016). The variety of topics in archaeology and art history that are discussed in this volume illustrates the richness of material culture in the Roman East and the Eastern Mediterranean during the transition to the Middle Ages, especially in Greece and the Balkans. Christian buildings, not only churches but also episcopal palaces, along with their architecture and decoration, receive special attention. Indeed, the volume includes the complete proceedings of a round table on the historical development, the architectural typologies and the domestic spaces of bishops’ residences which took place at the Congress.

‘I believe the collection of papers will make a significant contribution to this field… As a multi-authored work it will be of interest to a range of European and North Atlantic readers and libraries.’ Professor Jim Crow, University of Edinburgh

Edited by: Dominic Moreau, Carolyn S. Snively, Alessandra Guiglia, Isabella Baldini, Ljubomir Milanović, Ivana Popović, Nicolas Beaudry, Orsolya Heinrich-Tamáska Contributors: Sophia Akrivopoulou, Georgi Atanasov, Alessandra Avagliano, Isabella Baldini, Nicolas Beaudry, Patrick Blanc, Véronique Blanc-Bijon, Skënder Bushi, Eugenia Chalkia, Pascale Chevalier, Sébastien de Courtois, Claudia Di Bello, Nihat Erdoğan, Mikhail V. Fomin, Kyriakos Fragoulis, Anđela Gavrilović, Dejan Gjorgjievski, Alessandra Guiglia, Liudmila G. Khrushkova, Kosta Lako, Anaïs Lamesa, Ljubomir Maksimović, Dominic Moreau, Skënder Muçaj, Philipp Niewöhner, Andrea Paribeni, Silvia Pedone, Diego Peirano, Ivana Popović, Konstantinos T. Raptis, Konstantinos Z. Roussos, Goran Sanev, Helen Saradi, Carolyn S. Snively, Christina Tsigonaki, Catherine Vanderheyde, María J.S. Vicent, Ariadna Voronova, Suela Xhyheri, Asnu-Bilban Yalçın

Printed in England