Proxeny and Polis: Institutional Networks in the Ancient Greek World 019871386X, 9780198713869

Known from ancient authors such as Herodotus, Thucydides, and Plato, and more than 2,500 inscriptions, proxeny (a form o

663 50 4MB

English Pages 410 [431] Year 2015

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Proxeny and Polis: Institutional Networks in the Ancient Greek World
 019871386X, 9780198713869

Table of contents :
Cover
Proxeny and Polis: Institutional Networks in theAncient Greek World
Copyright
Dedication
Preface
Table of Contents
List of Maps
List of Figures
List of Tables
Abbreviations
Introduction
I.1. Scholarship on Proxenia
I.2. The Sources for Proxenia
I.3. Selective Inscription
I.4. Approaching Proxenia
I.5. Architecture of the Argument
1: The Anatomy of an Ancient Institution
1.1. Formulaic Descriptions of Proxenoi
1.1.1. Charmion, son of Eumaridas
1.1.2. The defining characteristics of the proxenos-paradigm
1.1.3. The proxenos as euergetēs
1.1.4. The specificity of the proxenos-paradigm
1.2. The Services Associated with Proxenia
1.2.1. Generic descriptions of the services performed by proxenoi
1.2.2. The communal contexts of proxenoi
1.2.3. The occupations, activities,and milieux of proxenoi
1.2.4. Services associated with proxenia
1.2.5. Services for poleis, their representatives and citizens
1.2.6. Apollodoros´ speech against Kallippos
1.3. The Proxenos-Paradigm in the Longue Durée
1.3.1. The longevity of the proxenos-paradigm
1.3.2. The stability of the proxenos-paradigm
2: Proxeny in its Political Contexts
2.1. Proxenia in the Granting Polis
2.1.1. The proposal and contestation of grants
2.1.2. Prestige and plurality: proposers of proxenoi at Oropos
2.2. Proxenoi at Home and at Court
2.2.1. Proxenoi at home
2.2.2. The perils of proxeny
2.2.3. Proxenoi at court
2.3. Poleis and their Foreign Proxenoi
2.3.1. `And all other honours given to proxenoi´
2.3.2. The ideology of the proxenos
2.3.3. Herakleodoros of Olynthos, proxenos of Thasos
2.3.4. The limits of identification
2.4. Miscarriage and Manipulation: Alcibiades the Proxenos
2.5. Conclusion
3: Proxeny Networks
3.1. Sources for Proxeny Networks
3.1.1. Case study: the chronological proxeny list of Histiaia
3.2. Proxeny Networks Compared
3.2.1. The size of proxeny networks
3.2.2. Rates and rhythms of granting
3.2.3. Diachronic trends in proxeny networking
3.3. Regional Patterns of Distribution: the Horizons of Poleis
3.4. Case Study: The Proxeny Networks of Karthaia
3.5. Conclusion
4: Proxeny and the Performance of Polis Identity
4.1. Theoretical Frameworks: State Identity and Anarchy
4.1.1. State identity/polis identity
4.1.2. Interstate anarchy
4.2. Inter-Polis Institutions and the Performance of Polis Identity
4.2.1. Proxenia within the systemof inter-polis institutions
4.2.2. Interstate institutions and the performance of polis identity
4.3. The Nuances of Proxenia in Performance
4.3.1. Isopoliteia and federation: proxenia as a performance of separateness
4.3.2. Polis performance, proxenia, and dependency
4.3.3. Proxenia and polis recognition
4.4. Inter-Polis Institutions and State Identity: Non-Polis Actors
4.5. Conclusion
5: The Disappearance of Proxeny and the Domination of Rome
5.1. The Disappearance of Proxenia
5.1.1. Corroborating the decline of proxenia
5.1.2 Functional obsolescence?
5.2. Changes in Inter-Polis Interaction
5.2.1. Theoria and theorodokia
5.2.2. Treaties and statuses
5.2.3. Interstate arbitration
5.2.4. Foreign judges
5.3. The Transformation of the Greek World
5.3.1. The end of peer-polity interaction?
5.4. Conclusion: The End of the Age of Proxeny
Conclusion
Appendix Extant Lists of Proxenoi
Introduction
1. Aitolia
1.1-8. Eight limestone stelae bearing twenty-four chronological lists of proxenoi appointed by the Aitolian League (third t...
2. Anaphe
2.1-3. Three chronological proxeny lists (fourth to second centuries bc)
3. Astypalaia
3. A catalogue of proxenoi (c.150 bc)
4. Chios
4.1. Catalogue of proxenoi (fourth or early third century bc)
4.2. Supplementary catalogue of proxenoi (early third century bc)
4.3. List of proxenoi appointed in a particular year (?) (second century bc)
4.4. Unpublished fragment of a list of proxenoi
5. Delphi
5. A chronological list of proxenoi appointed over twenty-two years (197/6-175/4 bc)
6. Epidauros
6.1-4. Four chronological lists of proxenoi (late fourth(?) to late third century bc)
7. Eresos
7. A catalogue of proxenoi (c.230-200 bc)
8. Histiaia (Oreos)
8. A list recording the grant of proxenia to thirty-two recipients (264/3 bc)
9. Kalchedon
9. A catalogue or chronological list (?) of proxenoi (Hellenistic?)
10. Karthaia
10.1. The proxeny catalogue of Karthaia (shortly after 363/2 bc)
10.2. A chronological proxeny list (mid-third century bc)
11. Kleitor
11. A catalogue (?) of proxenoi (second half of the third century bc)
12. Lousoi
12. A series of three chronological proxeny lists? (second half of the fifth century bc)
13. Narthakion
13. A reverse proxeny catalogue (?) from Narthakion (before or after ? c.140 bc)
14. Samothrace
14. 1-13. Thirteen blocks inscribed with sixteen lists of theoroi sent by different poleis and named proxenoi (second to fi...
15. Tenos
15.1. A fragment of a catalogue of proxenoi(?) (second half of the third century bc)
15.2. A list of fourteen proxenoi appointed at the same time (second half of the third century bc)
16. Thera
16.1-4. Chronological lists of proxenoi appointed in seven different years (?) (late second and early first centuries bc)
17. Polis Unknown (Boiotia)
17. A list of theoroi sent to games held at a polis in Boiotia named proxenoi (c.80 bc)
18. Polis Unknown
18. A probable list of proxenoi appointed in a single year (c.130 bc)
Photographs and Scans
Index of Polis Ethnics in Lists of Proxenoi
Index of Regional Ethnics (Including References to Ethnē, Koina, and Kings)
Bibliography
Index of Sources
General Index

Citation preview

OXFORD STUDIES IN ANCIENT DOCUMENTS General Editors Alan Bowman Alison Cooley

OXFORD STUDIES IN ANCIENT DOCUMENTS This innovative new series offers unique perspectives on the political, cultural, social, and economic history of the ancient world. Exploiting the latest technological advances in imaging, decipherment, and interpretation, the volumes cover a wide range of documentary sources, including inscriptions, papyri, and wooden tablets.

Proxeny and Polis Institutional Networks in the Ancient Greek World

WILLIAM MACK

1

3

Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries # William Mack 2015 The moral rights of the author have been asserted First Edition published in 2015 Impression: 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2014956596 ISBN 978–0–19–871386–9 Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

For my parents and for Naomi

Preface The Greek cities (poleis) of the ancient world constitute the most densely populated state system in recorded history, as the work of the Copenhagen Polis Centre has shown with its headline figure of one thousand poleis. The present work, the revised and expanded version of a doctoral thesis, is a study in the structural dynamics of this city-state culture. It focuses on a particularly characteristic interstate institution within this world, proxenia. Proxenia was an honorific status granted to foreign benefactors which permitted poleis to establish permanent networks of these individuals (proxenoi), connecting them as communities to large numbers of other poleis and facilitating interactions between the citizens of different cities. The importance of proxenia is evident in the vast quantity of material which we possess for it, in the form of references in literary sources and, especially, in the epigraphic record comprising more than three thousand inscriptions relating to proxenia. This book is not, however, intended to be a comprehensive manual on the Greek institution of proxenia (although material for many of the questions asked of such a manual can be found using the index). Instead, proxeny is used here as a case study to explore how inter-polis institutions functioned in shaping the behaviour of both individuals and communities in the ancient world, and how this changed with the establishment of the Roman Empire. As a result, this is a book of two halves. The first (Chapters 1 and 2) concentrates on understanding how proxenia operated as an institution invested with a particular meaning, and how both groups and individuals were co-opted to perform the different functions which it required. The second (Chapters 4 and 5) examines how proxenia functioned within a wider system of inter-polis institutions in constructing relations between different poleis. The division between these two halves is bridged by the third chapter, which reconstructs the parameters and scope of the networks of proxenoi possessed by different poleis on the basis of a series of documents—proxeny lists— which provide us with representative samples of data. (The information contained in these texts is presented in the appendix and references are made to individual documents in the form ‘Appendix no. 3’). The study of these networks lies at the heart of this work both because

viii

Preface

they provide concrete evidence for individual cities’ patterns of interaction with other communities and also because, in revealing the desire of Greek poleis to forge and advertise links of particular kinds, they show what was characteristic of the Greek world before it came to be integrated within the Roman Empire. In researching and writing this book I have accumulated many debts of different kinds and would thus like to inscribe a number of euergetai here. First of all, I would like to thank the supervisors of my thesis, Charles Crowther and John Ma, for their encouragement and guidance throughout and since. It was in fact John, as my undergraduate tutor in Ancient History, who set me on this path by introducing me to the history of the ancient world. Their friendship and help have been a great support and their influence will, I trust, be apparent on every page. Patrice Hamon and Robert Parker examined the thesis with great care and gave both useful suggestions and warm encouragement for publication. Patrice Hamon, in particular, proposed and commented on a draft of the appendix. Stephen Lambert and Peter Rhodes read the manuscript for the press and made valuable corrections and suggestions. Cédric Brélaz and Georgy Kantor were kind enough to read and comment on Chapter 5. As ever, neither they, nor any of my other readers, are responsible for any errors or omissions which remain. I would also like to thank Lisa Kallet who supervised my M.Phil. thesis and helped to shape the initial project which became this book, as well as the editors of this series, Alan Bowman and Alison Cooley, for accepting this work and their helpful comments on it. At OUP, Hilary O’Shea, her successor, Charlotte Loveridge, and their colleagues, Nicholas Bromley, Tom Chandler, Annie Rose, and Kizzy Taylor-Richelieu, were extremely helpful and accommodating. This work has benefited from comments from and discussion with many in Oxford and elsewhere, as well as from access to unpublished work. I would especially like to express gratitude to Cordula Bachmann, David Blackman, Ewen Bowie, Boris Chrubasik, Simon Day, Lisa Eberle, Aneurin Ellis-Evans, Roberta Fabiani, Ben Gray, Graham Oliver, Jonathan Prag, Gary Reger, Ian Rutherford, Rosalind Thomas, Peter Thonemann, and Guy Westwood. I regret that I can no longer thank Simon Price who gave encouragement and advice at an important stage of the thesis. Klaus Hallof, Daniela Summa, and Jamie Curberra facilitated access to squeezes in the collection of Inscriptiones Graecae in Berlin, and have been extremely generous in taking photographs for me and allowing me to publish them. At Athens, Nassos Themos of the Epigraphic Museum gave invaluable guidance

Preface

ix

and Tania Gerousi of the British School of Athens and Kalliopi Christophi of the Ecole Française d’Athènes were of great help in enabling me to obtain photographs of inscriptions in Athens and permission to publish them. The munificence of the William Edwards Trust, the Classics Faculty at Oxford University, and the Arts and Humanities Research Council enabled me to complete my graduate work free from financial worries. A scholarship funded jointly by the Gerda Henkel Stiftung and the Elise and Annemarie Jacobi Stiftung permitted me to break the back of the appendix in the ideal working conditions of the Kommission für Alte Geschichte und Epigraphik in Munich, and I am grateful to Christoph Schuler and Rudolf Haensch for electing me to this scholarship as well as Bart Grass and Salvo Vacante for making my time at the Kommission so pleasant. This research could not have been completed without the resources of the Sackler and Bodleian libraries in Oxford. For nine years, Corpus Christi College provided me with a second home in Oxford; I remain indebted to its staff, students, and fellows. Wadham College, where I made the final adjustments to the manuscript, could not have provided me with a warmer welcome, especially Steve Heyworth, my understanding colleague there. My greatest debts are to my family. My father, Peter, read the whole thesis and gave much shrewd advice. He and my mother, Vicki, have been unfailing in their love and support. My sisters, Johanna, Emily, and Rosy, and extended family, especially Bill, helped me in too many ways to list. Finally, and most of all, I must thank my wife, Naomi, who has read this work in its various incarnations far too many times: for her love. William Mack University of Birmingham

An online database entitled Proxeny Networks of the Ancient World will accompany this book, hosted on the website of the Centre for the Study of Ancient Documents (). It is intended to serve as a comprehensive catalogue of all surviving attestations of proxeny in the literary and the epigraphic sources. The aim is partly to make the material which underlies the graphs and statistics

x

Preface

presented in this book transparent to other scholars. It is also meant, however, to make this material more widely accessible, and to enable this evidence of connections between different civic communities in the ancient world to be explored in a variety of different ways.

Table of Contents List of Maps List of Figures List of Tables Abbreviations Introduction I.1. I.2. I.3. I.4. I.5.

Scholarship on Proxenia The Sources for Proxenia Selective Inscription Approaching Proxenia Architecture of the Argument

1. The Anatomy of an Ancient Institution 1.1. Formulaic Descriptions of Proxenoi 1.2. The Services Associated with Proxenia 1.3. The Proxenos-Paradigm in the Longue Durée

2. Proxeny in its Political Contexts 2.1. 2.2. 2.3. 2.4.

Proxenia in the Granting Polis Proxenoi at Home and at Court Poleis and their Foreign Proxenoi Miscarriage and Manipulation: Alcibiades the Proxenos

3. Proxeny Networks 3.1. Sources for Proxeny Networks 3.2. Proxeny Networks Compared 3.3. Regional Patterns of Distribution: The Horizons of Poleis 3.4. Case Study: The Proxeny Networks of Karthaia

4. Proxeny and the Performance of Polis Identity 4.1. Theoretical Frameworks: State Identity and Anarchy 4.2. Inter-Polis Institutions and the Performance of Polis Identity 4.3. The Nuances of Proxenia in Performance

xiii xiv xvi xvii 1 4 9 13 17 19 22 27 48 81 90 91 104 122 142 148 152 164 173 182 190 191 201 207

xii

Table of Contents 4.4. Inter-Polis Institutions and State Identity: Non-Polis Actors

5. The Disappearance of Proxeny and the Domination of Rome

228

5.1. The Disappearance of Proxenia 5.2. Changes in Inter-Polis Interaction 5.3. The Transformation of the Greek World

233 234 254 270

Conclusion

282

Appendix Bibliography Index of Sources General Index

286 363 391 402

List of Maps 1. The proxenoi appointed by the Histiaians, 264/3 bc (Appendix no. 8)

157

2. The proxeny catalogue of Astypalaia, c.150 bc (Appendix no. 3)

176

3. The proxeny catalogue of Eresos, c.230–200 bc (Appendix no. 7)

177

4. The proxenoi appointed by the Delphians, 197/6 to 175/4 bc (Appendix no. 5)

178

5. The proxeny catalogue of Karthaia, shortly after 363/2 bc (Appendix no. 10.1)

183

6. Proxenoi appointed by the Karthaians, mid-third century bc (Appendix no. 10.2)

187

7. The proxeny lists of Aitolia, third to early second century bc (Appendix no. 1)

291

8. The proxeny lists of Anaphe, fourth to second centuries bc (Appendix no. 2)

293

9. The proxeny lists of Chios, fourth to second centuries bc (Appendix no. 4)

305

10. The proxeny lists of Epidauros, late fourth(?) to late third century bc (Appendix no. 6)

313

11. The proxeny catalogue of Kleitor, second half of the third century bc (Appendix no. 11).

326

12. The proxeny lists of Samothrace, second to first centuries bc (Appendix no. 14)

333

13. The proxeny lists of Tenos, second half of the third century bc (Appendix no. 15)

335

14. The proxeny lists of Thera, late second and early first centuries bc (Appendix no. 16)

337

15. The proxeny list of an unknown city, perhaps Boiotian, c.80 bc (Appendix no. 17)

339

16. The proxeny list of an unknown city, c.130 bc (Appendix no. 18)

343

17. A combined view of coverage of the proxeny networks partially attested by Appendix nos. 1–18 (240 different poleis)

344

List of Figures 5.1. Chronological distribution of surviving proxeny decrees per fifty-year period

236

5.2. Number of communities attested inscribing proxeny decrees per fifty-year period

237

5.3. Comparison of media used for the inscription of proxeny decrees at Oropos during different periods (stelae versus statue-bases)

239

5.4. Comparison of the chronological distribution of proxeny decrees for Oropos with the rest of Boiotia

240

5.5. Comparison of the chronological distribution of Athenian inscribed decrees for internal honorands (citizens) and external honorands (xenoi)

241

5.6. Comparison of the chronological distribution of Athenian decrees granting proxenia with those granting citizenship

241

5.7. Attestations of the personal name ‘Proxenos’, per century

248

5.8. Attestations of the personal name ‘Theōros’, per century

259

5.9. Chronological distribution of decrees honouring foreign judges 268 A.10. Astypalaia, catalogue of proxenoi (Appendix no. 3, c.150 bc)

345

A.11. Astypalaia, catalogue of proxenoi (Appendix no. 3, c.150 bc)

346

A.12. Astypalaia, catalogue of proxenoi (Appendix no. 3, c.150 bc)

347

A.13. Chios, catalogue of proxenoi (Appendix no. 4.1; fourth or early third century bc)

347

A.14. Chios, supplementary catalogue of proxenoi (Appendix no. 4.2, early third century bc)

348

A.15. Chios, supplementary catalogue of proxenoi (Appendix no. 4.2, early third century bc)

348

A.16. Chios, supplementary catalogue of proxenoi (Appendix no. 4.2, early third century bc)

348

A.17. Chios, list of proxenoi appointed in a particular year (?) (Appendix 4.3, second century bc)

349

A.18. Eresos, catalogue of proxenoi (Appendix no. 7, c.230–200 bc)

350

A.19. Karthaia, catalogue of proxenoi (Appendix no. 10.1, shortly after 363/2 bc)

351

List of Figures

xv

A.20. Karthaia, catalogue of proxenoi (Appendix no. 10.1, shortly after 363/2 bc)

351

A.21. Narthakion, reverse catalogue of proxenoi (?) (Appendix no. 13, before or after ?c.140 bc)

352

A.22. Unknown polis, chronological list of proxenoi (?) (Appendix no. 18, c.130 bc)

353

A.23. Unknown polis, chronological list of proxenoi (?) (Appendix no. 18, c.130 bc)

353

List of Tables 1. The proxeny lists of Aitolia

289

2. The proxeny lists of Anaphe

292

3. The proxeny lists of Astypalaia

300

4. The proxeny lists of Chios

304

5. The proxeny list of Delphi

307

6. The proxeny lists of Epidauros

311

7. The proxeny catalogue of Eresos

316

8. The proxeny catalogue of Karthaia (mid-fourth century bc)

322

9. The proxeny list of Karthaia (mid-third century bc)

324

10. The proxeny list of Kleitor

325

11. The proxeny lists of Samothrace

331

12. The proxeny lists of Thera

336

Abbreviations For abbreviations of literary sources, I follow or expand on LSJ. All abbreviations of journals follow L’année philologique.1 Achaïe III Agora APF BCH BE CAH CEG CIG CIL CIRB DGE FD FGrHist GIBM Epigrafia greca HCP HCT

IACP

A. D. Rizakis, Achaïe III: Les cités achéennes. Athens, 2008. The Athenian Agora: Results of Excavations Conducted by the American School of Classical Studies at Athens. J. K. Davies, Athenian Propertied Families, 600–300 B.C. Oxford, 1971. Bulletin de correspondance hellénique. Bulletin Épigraphique in Revue des Études Grecques. The Cambridge Ancient History. P. A. Hansen, Carmina Epigraphica Graeca (2 vols.). Berlin, 1983–9. Corpus Inscriptionum Graecarum. Corpus Inscriptionum Latinarum. V. Struve, Corpus Inscriptionum Regni Bosporani. Moscow, 1965. E. Schwyzer, Dialectorum Graecarum Exempla Epigraphica. Leipzig, 1923. Fouilles de Delphes. F. Jacoby, Die Fragmente der griechischen Historiker. Berlin and Leiden, 1923–62. The Collection of Ancient Greek Inscriptions in the British Museum (4 vols.). London, 1874–1916. M. Guarducci, Epigrafia greca (4 vols.). Rome, 1967–1978. F. W. Walbank, A Historical Commentary on Polybius (3 vols.). Oxford, 1957–79. A. W. Gomme, A. Andrewes, and K. J. Dover, A Historical Commentary on Thucydides (5 vols.). Oxford, 1945–81. M. H. Hansen and T. H. Nielson (eds). An Inventory of Archaic and Classical Poleis. Oxford, 2004.

1 (accessed 31 October 2012).

xviii IAphr2007 IC ID IG IGBulg IGR IGUR IosPE I2 IPArk IvO I.Assos I.Beroia

I.Byzantion I.Eph I.Erythrai I.Gonnoi I.Histriae

I.Iasos I.Ilion I.Kalch I.Kaunos I.Kios I.Knidos

Abbreviations J. Reynolds, C. Roueché, and G. Bodard, Inscriptions of Aphrodisias, 2007: http://insaph.kcl.ac.uk/iaph2007. M. Guarducci, Inscriptiones Creticae. Rome, 1935–50. F. Durrbach, Inscriptions de Délos. Paris, 1926–37. Inscriptiones Graecae. G. Mikhailov, Inscriptiones Graecae in Bulgaria repertae, 2nd edn. (5 vols.). Sofia, 1970. R. Cagnat, Inscriptiones Graecae ad Res Romanas Pertinentes (3 vols.). Paris, 1906–27. L. Moretti, Inscriptiones Graecae Urbis Romae (4 vols.). Rome, 1968–90. V. Latyshev, Inscriptiones antiquae orae septentrionalis Pontis Euxini Graecae. Leningrad, 1885–1901. G. Thür and H. Taeuber. Prozessrechtliche Inschriften der griechischen Poleis: Arkadien. Vienna 1994. W. Dittenberger and K. Purgold, Die Inschriften von Olympia. Berlin, 1896. R. Merkelbach, Die Inschriften von Assos. IGSK 4. Bonn, 1976. L. Gounaropoulou and M. Hatzopoulos, ’¯ تæÆçb ˚ø ÆŒ Æ. `΄ ’¯ تæÆçb ´æ Æ. Athens, 1998. A. Łajtar, Die Inschriften von Byzantion. IGSK 58. Bonn, 2000. H. Wankel, R. Merkelbach et al., Die Inschriften von Ephesos (8 vols.). IGSK 11–17. Bonn, 1979–81. H. Engelmann and R. Merkelbach, Die Inschriften von Erythrai (2 vols.). IGSK 1–2. Bonn, 1972–3. B. Helly, Gonnoi II: Les Inscriptions. Amsterdam, 1973. Inscriptiones Daciae et Scythiae Minoris Antiquae, Series Altera. Inscriptiones Scythiae Minoris Graecae et Latinae, I. Inscriptiones Histriae et Viciniae. Bucharest, 1983. W. Blümel, Die Inschriften von Iasos (2 vols.). IGSK 28. Bonn, 1985. P. Frisch, Die Inschriften von Ilion. IGSK 3. Bonn, 1975. R. Merkelbach, Die Inschriften von Kalchedon. IGSK 20. Bonn, 1980. C. Marek, Die Inschriften von Kaunos. Munich, 2006. T. Corsten, Die Inschriften von Kios. IGSK 29. Bonn, 1985. W. Blümel, Die Inschriften von Knidos. IGSK 41. Bonn, 1992.

Abbreviations I.Labraunda

I.Lampsakos I.Lindos I.Magnesia I.Mylasa I.Perge I.Priene I.Stratonikeia I.Thasos I.Thess.Enipeus I.Thrac.Aeg.

I.Tralleis LGPN LSAG LSJ

McCabe, Chios Michel Milet ML Nomima

xix

J. Crampa, Labraunda, Swedish Excavations and Researches III: Greek Inscriptions (2 vols.). Lund, 1969, and Stockholm, 1972. P. Frisch, Die Inschriften von Lampsakos. IGSK 6. Bonn, 1978. C. Blinkenberg, Lindos: Fouilles et recherches, 1902–1914, ii. Inscriptions. Berlin, 1941. O. Kern, Die Inschriften von Magnesia am Maeander. Berlin, 1900. W. Blümel, Die Inschriften von Mylasa (2 vols.). IGSK 34–5. Bonn, 1987–8. S. Şahin, Die Inschriften von Perge (2 vols). IGSK 54 and 61. Bonn, 1999–2004. F. Frhr. Hiller von Gaertringen, Die Inschriften von Priene. Berlin, 1906. M. Ç. Şahin, Die Inschriften von Stratonikeia. IGSK 21–2. Bonn, 1981–90. J. Pouilloux and C. Dunant, Recherches sur l’histoire et les cultes de Thasos vol. 2. Paris, 1954. J.-C. Decourt, Inscriptions de Thessalie I: Les cités de la vallée de l’Énipeus. Athens, 1995. L. D. Loukopoulou, M.-G. Parissaki, S. Psoma and A. Zournatzi, ’¯تæÆçb B ¨æŒÅ  F `NªÆ ı. Athens 2005. F. B. Poljakov, Die Inschriften von Tralleis und Nysa I. IGSK 36.1. Bonn, 1989. A Lexicon of Greek Personal Names. L. H. Jeffery, Local Scripts of Archaic Greece, 2nd edn. Oxford, 1990. H. G. Liddell and R. Scott, A Greek-English Lexicon, 9th edn. revised by H. Stuart Jones and R. McKenzie. Oxford, 1996. D. F. McCabe and J. V. Brownson, Chios Inscriptions: Texts and List. Princeton, 1986. C. Michel, Recueil d’inscriptions grecques. Brussels, 1900–27. Milet: Ergebnisse der Ausgrabungen und Untersuchungen seit dem Jahre 1899. R. Meiggs and D. Lewis, A Selection of Greek Historical Inscriptions, rev. edn. Oxford, 1988. H. van Effenterre and F. Ruzé, Nomima: Recueil d’inscriptions politiques et juridiques de l’archaïsme grec (2 vols.). Rome, 1994.

xx OGIS OMS P.Cair.Zen.

P.Col.Zen. I

P.Mich. RC RDGE

RE RO SGDI SEG StV II2 StV III Syll.3 TAM Tit.Cal. Tit.Cam. Tod

Abbreviations W. Dittenberger, Orientis Graeci Inscriptiones Selectae (2 vols.). Leipzig, 1903–5. L. Robert, Opera Minora Selecta, I–VII. Amsterdam, 1969–1990. C. C. Edgar (ed.), Catalogue général des antiquités égyptiennes du Musée du Caire, Zenon Papyri (4 vols.). Cairo, 1925–31. Vol. 5, O. Guéraud and P. Jouguet (eds.), Cairo, 1940. W. L. Westermann and E. S. Hasenoehrl, Columbia Papyri III, Zenon Papyri: Business Papers of the Third Century B.C. dealing with Palestine and Egypt I. New York 1934. Michigan Papyri. C. B. Welles, Royal Correspondence in the Hellenistic Period. New Haven, 1934. R. K. Sherk, Roman Documents from the Greek East. Senatus Consulta and Epistulae to the Age of Augustus. Baltimore, 1969. Paulys Real-Encyclopädie der classischen Altertumswissenschaft. P. J. Rhodes and R. Osborne, Greek Historical Inscription, 404–323 B.C. Oxford, 2003. F. Bechtel, H. Collitz et al., Sammlung der griechischen Dialekt-Inschriften (4 vols.). Göttingen, 1884–1915. Supplementum Epigraphicum Graecum. H. Bengtson, Die Staatsverträge des Altertums, 2nd edn. Munich, 1975. H. H. Schmitt, Die Staatsverträge des Altertums. Munich, 1969. W. Dittenberger, Sylloge Inscriptionum Graecarum, third edition (4 vols.). Leipzig, 1915–24. Tituli Asiae Minoris. M. Segre, ‘Tituli Calymnii’, ASAtene 22–23, 1944–1945: 1–248. M. Segre and G. Pugliese Carratelli, ‘Tituli Camirenses’, ASAtene 27–29, 1949–1951: 141–318. M. N. Tod, A Selection of Greek Historical Inscriptions (2 vols.). Oxford, 1933–48.

Introduction The history of ancient institutions is the study of the formal structures within which lives were lived in ancient societies. Interstate institutions, in particular, allow us to explore how relations were constructed between different political communities. This book is intended as a study of the system of interstate institutions which characterized the Ancient Greek world—the functions which it performed, the ways in which it worked, and the reasons why it was fundamentally transformed by the Roman Empire—but it proceeds by beginning with one particular interstate institution, proxenia. Proxeny, by allowing us to examine how honorific institutions functioned at the level of both the individual and the political community, offers historians of the ancient world a privileged insight into this phenomenon of polis-networking more broadly. The system of interstate institutions in the Ancient Greek world, however, also has a strong claim on the interest of students of interstate relations. The interstate society of the Ancient Greek world was arguably the most populous in recorded history, consisting of more than one thousand city-state actors at any time until the late Hellenistic period, and the system of interstate institutions which these communal actors evolved for promoting cooperation and mediating their interactions was correspondingly complex and highly developed.1 Proxenia was an honorific status bestowed by poleis on noncitizens who thereby became their proxenoi. The grant of proxenia expressed a formal relationship of friendship between polis and proxenos. This relationship was based on the gratitude of the polis for the intermediary services the proxenos performed for its citizens within

1

Hansen (2006), 31.

2

Proxeny and Polis

his own political community, services which facilitated their access to local civic institutions and networks there. A relief sculpture from a fifth-century Athenian decree (IG I3 91) provides a visual representation of this institution which also emphasizes how proxenoi functioned conceptually as links to other communities. Proxenides, the honorand named proxenos by this decree, is depicted on the relief, a smaller figure standing between two female deities—Aphrodite, the patron deity of his polis of Knidos, and Athena, representing Athens, at whom Proxenides gazes, raising his left arm in greeting or adoration. Aphrodite, on the left, with her hand on the head of her citizen, entrusts him to the care of Athena, who reaches out her own hand towards Proxenides. The composition of the relief draws on the iconography of guest-friendship (xenia), and, in particular, scenes of foster-parenthood. Here, this imagery, which emphasizes xenia as the etymological root of proxenia, works on two levels, representing both the identity of Proxenides the proxenos as the public friend (xenos) of the Athenians, but also the friendship of Athena and Aphrodite (and thus Athens and Knidos). The nature of Proxenides’ role as proxenos, as an intermediary connecting Knidos and Athens, is also reflected symbolically in his position between and visually connected to both goddesses.2 The importance of proxenia as an interstate institution is reflected in the unusually rich and varied dataset that we possess for it. Comprising a wide range of literary references and more than two and a half thousand inscriptions, this material documents the use of proxeny by poleis throughout the Greek world from the Black Sea to Sicily, from the late seventh century bc until the Roman Imperial period. This wealth of material allows us to reconstruct this particular institution in great detail as it was used by far flung poleis. Proxeny emerges as a fundamental means by which the Greek city states constructed and understood their links with each other. Examination of what proxenia was, as well as how and why it worked, also has the potential to shed light on the functions performed by honorific language and interstate institutions more widely. 2 On the similarities between this representation of proxenia and depictions of xenia, see Herman (1987), 22–6 with fig.1a–d and 137 fig. 14. In general on this relief, see Lawton (1995), 115–16 (no. 68). Lawton rightly dismisses the suggestion that this relief depicts dexiosis between Athena and Proxenides. A photograph of this relief, which is now in the Acropolis Museum, is reproduced on the front cover of the print edition of this book.

Introduction

3

In particular, because it constituted a relationship between a polis and an individual non-citizen, proxenia allows us to explore how both these actors’ motivations and their means of pursuing them were shaped by institutions at a variety of different levels. This study argues that the social dynamics which underlay the operation of civic euergetism of the sort delineated by Veyne (1990) and Gauthier (1985) within the Greek cities are crucial for understanding how institutions like proxenia operated between poleis. Proxenia was, in fact, probably the most widespread and frequently used element of a broader system of interstate institutions which included festival networks, the dispatch and receipt of foreign judges, declarations of inviolacy and inter-polis treaties, and privilege grants of various kinds. Many of the other elements of this system of institutions have received recent study under the impetus of a renewed interest in the history and plurality of the Greek poleis, and, in particular, the emphasis which has been placed on their vitality in the Hellenistic period against earlier models stressing post-Classical decline.3 Although some of the other elements of this system appear to have had a much greater political significance when they were used, proxenia, as a case study, allows us to explore this system as a whole. Grants of proxenia reflected and facilitated the full range of different sorts of inter-polis interaction, public and private, from diplomatic negotiations to economic transactions and inter-city pilgrimage. The very substantial networks, comprising hundreds of proxenoi, which even small poleis routinely maintained, reveal what polis vitality actually meant (see the maps of polis networks and the fragmentary proxeny lists on which they are based, collected in the Appendix). They also, however, allow us to explore the reasons for this vitality in both the Classical and Hellenistic periods as a cultural

3 On this system of institutions in general, see Chapter 5; on festival networks and inter-polis pilgrimage (theōrodokia and theōria): Perlman (2000) and, now, Rutherford (2013); on foreign judges, see the works of Crowther (e.g. Crowther 1992, 1993); on declarations of inviolacy (asylia), see Rigsby (1996). The rude health of studies of the Greek poleis is demonstrated not only by two major international projects—the Copenhagen Polis Project (directed by Mogens Herman Hansen, see esp. Hansen and Nielsen 2004) and, more recently, Die Polis als Lebensform (directed by Martin Zimmerman, e.g. Matthaei and Zimmermann 2008)—but also by the raft of individual studies which have been undertaken. The origins of the vitalist paradigm in relation to the Hellenistic poleis lie in the works of Louis Robert, but have been elaborated by his student, Philippe Gauthier, and others (see especially Gauthier 1985 and Ma 2002).

4

Proxeny and Polis

phenomenon. Proxenia in particular emerges as the characteristically Greek form of inter-polis networking and was, I argue, closely bound up with ideas of what it meant to be, and to be seen to be, a polis. This study examines the role of proxenia, and other inter-polis institutions, during the roughly five hundred years encompassed by the Classical and Hellenistic periods (c.500–1 bc). It argues that, despite changing trends in the production and survival of different sorts of source material—which has led to continuing controversy over the function and development of proxenia—the evidence produces a strikingly homogeneous picture of this institution and the way in which it was used and understood throughout this period. One of the central contentions of this book is that this homogeneity was not just an incidental artefact, but central to the functions which proxenia and other inter-polis institutions were used to perform. It seeks to identify the reasons for this homogeneity, and the strong continuity in interstate dynamic that persisted until the end of the first century bc, which characterizes it as an age of inter-polis networking—of proxenia.

I.1. SCHOLARSHIP ON PROXENIA ‘L’étude? Ou je me trompe, ou il n’y a rien d’intéressant à trouver de nouveau sur la proxénie. C’est une question réglée, et, Deo gratias, nous savons ce que c’est, nous pouvons l’expliquer aux débutants. Si le livre de Monceaux est épuisé, il n’y a qu’à le reproduire photographiquement. À mon avis, il n’y a, pour la proxénie, que la place d’un chapitre dans un manuel d’institutions grecques.’ ‘The study? If I am not mistaken there is nothing new of interest to discover concerning proxeny. It is a question which has been solved and, thank goodness, we know what it is, we can explain it to novices. If Monceaux’s monograph is out of print it can be reprinted lithographically. In my opinion there is no place for proxenia other than as a chapter in a handbook on Greek institutions.’ Louis Robert to Biagio Virgilio, 1 March 1970 (reprinted Virgilio 2011, 269–70)4 4 Robert published similar comments on Virgilio’s planned but abandoned project (a corpus and introductory study of all extant proxeny decrees) in BE (1970), no. 114.

Introduction

5

The monograph by Paul Monceaux, Les proxénies grecques, published in 1886, represented the culmination of a series of nineteenth-century works on proxeny.5 The earliest of these, by Franz Wolfgang Ullrich (1822), established the lexicographical interest found in much later scholarship, exploring, in particular, the relation of proxenia (public hospitality) to other forms of xenia (hospitality). The German philologist, Moritz Meier, wrote the first major study (1843), which laid the foundation for later work on proxenia in its comprehensive collation of literary testimonia and systematic consideration of the epigraphic documents that August Boeckh’s Corpus Inscriptionum Graecarum (first volume 1828) had begun to make accessible. Charles Tissot consolidated Meier’s work in a systematic treatment (1863) that explored the limits of the analogy, made at least as early as Doukas’s commentary on Thucydides, between proxenia and the modern institution of the consulship (Tissot himself was the French consul at Edirne).6 Robert concentrates on Monceaux in his letter to Biagio Virgilio partly because, at the time at which he was writing, it constituted the most systematic study, treating, in thirty-eight brief chapters, a very wide range of regional and thematic case studies of proxenia. Robert’s emphasis, however, also reflects his dissatisfaction with the central concerns of subsequent scholarship. In the eighty years that intervened between Monceaux’s account and Robert’s letter to Virgilio, the study of proxeny had been characterized by increasing scepticism concerning the functional nature of proxenia. Scholars, beginning with Emile Szanto (1892), began to see a contradiction between the understanding of proxenia as a quasi-contractual office, involving an obligation on the proxenos to perform particular duties, with the honorific aspect of this status which seemed prominent in the rapidly expanding epigraphic record of proxeny decrees for the Hellenistic period. In its most influential form, this was constructed as a narrative of the functional decline of proxenia which was part of a wider historiographic trend, the more general contemporary belief in the decline of the Greek polis at the end of the Classical period. According to Walther Schwahn (1931), Hans Schaefer (1932), and Günther Klaffenbach (1966), among others, proxenia, as their nineteenthcentury predecessors had understood it, as a consul-like institution, 5 For a lucid account of the earlier, more isolated scholarship on proxenia, see Virgilio (1969). 6 Doukas (1805) on Thuc. 2.29; see Tissot (1863), 101.

6

Proxeny and Polis

could only be identified in the sources, especially literary, for proxeny in the Classical period.7 Thereafter, as early as the fourth century bc, as proxenia came increasingly to be granted in purely honorific contexts to multiple individuals within the same polis and even, on occasion, to women, minors, whole communities, and metics, it lost its functional content altogether, becoming an empty honour devoid of practical significance.8 Later in this letter, Robert signals his own rejection of this simplistic narrative of post-Classical decline (as he repeatedly rejected the wider historiographic belief in the ‘death’ of the post-Chaironeian polis) by citing with approval one work produced in this period— Adolf Wilhelm’s lengthy article of 1942, ‘Proxenie und Euergesie’. In this extremely learned but difficult work Wilhelm sought to undermine the arguments of Schwahn and Schaefer by adducing numerous examples, extending to the late Hellenistic period, of proxenoi still apparently performing services as they had in the Classical period. However, in spite of the evidence marshalled by Wilhelm, narratives of proxeny decline remained dominant.9 Ironically, in the period immediately following Robert’s dismissive statement, research on proxenia entered a far more active phase. A continuous stream of important articles, extended studies, and monographs on both proxenia more broadly and particular sets of material began and has continued to this day.10 Many of the most 7

Schwahn (1931), 108–12; Schaefer (1932), 28; Klaffenbach (1966), 83–5. On these narratives, see also Marek (1984), 2–3 and Gauthier (1985), 131–6; Klaffenbach (1966), 83–4, for references to these instances. Marek (1985), 129–30, collects the isolated grants for women and children. On grants for women, see Mack (2014) and, this volume, Chapter 1, n. 8. 9 So Klaffenbach (1966), 83–5; Ehrenberg (1960), 104–5; Guarducci (1969), 29–30; Walbank (1981), 73. Virgilio (1969), 498, although accepting Wilhelm’s assertion that proxenoi continued to serve a precise function as fundamentally correct, found his examples unconvincing. 10 Major works: Gschnitzer (1973); Marek (1984). Gauthier (1985) contains a substantial section on proxenia. On early forms of and evidence for proxenia, see Wallace (1970), Gauthier (1972), 26–61, but also Zelnick-Abramovitz (2004). On the proxeny decrees of Athens, see Walbank (1978) and Reiter (1991) in addition to Meiggs (1949) on the fifth-century material, and Culasso-Gastaldi (2004) and (2005) on the fourth century. See Knoepfler (2001a) on the decrees from Eretria; Fabiani (forthcoming a) and (forthcoming b) on those from Iasos; Gavazzi (1951) on the material from Thessaly; Huybrechs (1959) and Mosley (1971b) on proxeny at Sparta. Habicht (2002) provides an important discussion of the package of honours granted to proxenoi. Moggi (1995) and Perlman (1958) provide good discussions of the political importance of proxenia in relation to the literary evidence for the Classical period. 8

Introduction

7

important of these works, however, despite Wilhelm’s article and the criticisms of Robert, have continued to be influenced by, or at least framed in response to, the same narrative of decline. In particular, Fritz Gschnitzer’s hundred-column article on proxenia in RE, published in 1973, proposed another version, albeit more nuanced, of the proxeny decline narrative. In it he identified honour and function as two different components of proxenia, present from the start but subject to different trajectories of decline.11 According to Gschnitzer, the practical aspect of proxenia disappeared first, partly as a result of the development in the Hellenistic period of other institutions which usurped the services performed by proxenoi (e.g. the development of inns obviated the need for proxenoi to provide hospitality), with the result that proxenia came to serve principally as an honour without these sorts of practical implications. Subsequently, Gschnitzer argued, by the later Hellenistic period even this honorific function came to be compromised as proxenia became devalued as a result of having been granted too widely and disappeared entirely. Phillipe Gauthier, in his 1985 monograph on euergetism and the Greek polis, followed in the footsteps of Wilhelm by offering a lengthy and persuasive rebuttal of these sorts of narrative of decline in general and Gschnitzer’s account in particular. Gauthier firmly restated the inherently functional nature of proxenia. In particular, he pointed out that, as poleis are frequently attested as possessing multiple proxenoi in the same polis even in the high Classical period, this phenomenon can hardly be taken as decisive evidence of subsequent decline. However, at the same time, Gauthier also allowed more latitude as to precisely what this function consisted of, and emphasized its ability to evolve as the conditions of the Greek world did.12 Christian Marek, conversely, in the most recent major monograph devoted to proxenia (derived from his doctoral thesis and published in 1984), sought to circumvent entirely what he saw as an unprofitable deadlock between scholarship emphasizing honour (and decline) and scholarship stressing function. Instead, in his comprehensive collection of the epigraphic material he adopted an approach that consisted of categorizing the recipients of proxenia as far as possible into different types according to their known activities—as kings and

11 12

Gschnitzer (1973), cols. 643–63. Gauthier (1985), 131–49, esp. 136–45.

8

Proxeny and Polis

dynasts, courtiers, military officers, federal officials, ship-owners, bankers, foreign judges, doctors, and poets.13 On the basis of this analysis, Marek identified a number of different regional and diachronic trends in the uses to which proxenia was put. However, the result of this approach, which, rather than considering these individuals as proxenoi, recognized a series of widely varying sorts of category, was the loss of a sense of proxenia as a specific institution. Instead of moving beyond the dichotomy between honour and function, Marek’s monograph thus served to reinforce the understanding of proxenia as an inherently flexible and thus essentially empty honour.14 The opposition between narratives of decline and scholarship emphasizing the continuing functional nature of proxenia has been fuelled in no small part by the very substantial but seemingly contradictory historical record for proxenia—capable of supplying examples which support the argument of either side. This study argues that a more nuanced analysis of the trends and biases of this record is necessary to understand the evidence for proxenia which it provides. On the basis of this, I propose a new model for interpreting proxenia and understanding the complex relationship between proxeny’s honorific and symbolic functions and the practical services associated with it. This model is based on the identification of a coherent and stable set of expectations that informed the way in which contemporaries conceived and made use of this role—a ‘proxenos-paradigm’. This builds on the work of Wilhelm and Gauthier, who argued that utility was central to proxenia, but shows that this utility was expected to manifest itself in particular ways, especially in intermediary services performed by proxenoi to facilitate interaction between different poleis. It was thus, in fact, proxeny’s highly specific content, not its absence, which invested it with symbolic meaning and made it useful in other instances for constructing relations, on occasion, with individuals and even groups who were atypical proxenoi.

13

Marek (1984), 3–4 and 333–85. See also Culasso Gastaldi (2005), 75, where proxenia is described, for similar reasons, as ‘senza personalità sua propria’. 14

Introduction

9

I.2. THE SOURCES FOR PROXENIA ‘L’importance des proxénies grecques est une des révélations de l’épigraphie.’15

There can be little doubt of the importance of proxeny as an epigraphic phenomenon and, since Monceaux wrote, the rich epigraphic record on which he based his study has grown ever greater. The vast majority of this record consists of proxeny decrees, inscribed texts recording, in concise and formulaic language, the decree by which a particular polis bestowed on a non-citizen the formal status of proxenos in recognition of his services. It is scarcely an exaggeration to say that each year has brought to light new proxeny decrees and there is no sign of this steady stream of material drying up. This study is based on a database of two and a half thousand securely identified proxeny decrees and there are perhaps a thousand other partially preserved inscriptions of which the majority are very probably also proxeny decrees—making proxenia by far the best attested of all inter-polis institutions. These inscriptions also represent a very wide cross section of the poleis of the Greek world, and a substantial proportion of the one thousand poleis identified by the Copenhagen Polis Project.16 The decrees these inscriptions contain were promulgated by more than one hundred and eighty different poleis, stretching from the Black Sea to Sicily.17 In addition, although proxenia was, and, as this study shows, was conceived of as, an institution of the polis, a number of other sorts of authority are represented by grants of proxenia which they made and inscribed on stone. Supra-polis entities like federations are particularly well attested, but we also possess proxeny decrees from petty kings and dynasts as well as associations of merchants and Dionysiac artists.18 The proxeny decree, however, is not the only context in which proxenia is attested epigraphically. In particular, in this study I argue that as important for understanding this institution and how it was 15

Monceaux (1886), v. Hansen and Nielsen (2004), with 1035 entries for certain or probable poleis in the Archaic and Classical periods; Hansen (2008) now identifies 1040. 17 This number includes not only poleis which are represented by the decrees they themselves inscribed (168 poleis) but also poleis whose decrees of proxenia were inscribed at other poleis (as in the honorific dossier recording all decrees passed in honour of Nikomedes of Kos, which was inscribed at his home polis, IG XII 4 129–30). 18 On these, see Chapter 4, Section 4.4. 16

10

Proxeny and Polis

used are the much less numerous inscribed lists and catalogues of proxenoi with which poleis advertised their proxeny networks en masse. In addition, a wide range of other, more marginal types of epigraphic monument give us a sense of the significance and pervasiveness of this ancient institution, illustrating facets of particular proxeny grants and relationships with great vividness: funerary monuments erected for and by proxenoi;19 inscriptions collecting the different grants of proxenia bestowed on a particular individual;20 a probable reverse proxeny catalogue from Narthakion, listing all the proxenoi other cities recognized there;21 stone theatre seats inscribed as belonging to specific proxenoi;22 dedications made by proxenoi;23 references to proxenoi giving or lending money to the poleis to which they were linked;24 a unique inscription which identified the house of a Pergamene physical trainer and proxenos of the Thessalian

19

See Chapter 1, n. 123. So-called Ehrentafeln; see Chapter 2, n. 65–8. 21 22 Appendix no. 13. See Chapter 2, n. 144. 23 SEG 19 595 (Thasos, mid-fourth century bc, on this dedication of a tower, statue, and exedra to all the gods by a proxenos from Olynthos, see Chapter 2, Section 2.3.3); 50 543 (Dodona, c.334–330 bc, a bronze tablet dedicated by Agathon of Zakynthos who claims that his family have been proxenoi of the Molossians and their allies for thirty generations, since Cassandra of Troy, on which see Fraser 2003); 14 455 (Delphi, fourth century bc); IG II2 3882 (Athens, beginning of the first century bc); IG XII 5 917 (Tenos, before 46 bc); Maiuri (1925), no. 8 (Rhodes, late second to early first century bc); SEG 28 657 may well be another example (Olbia, 475–450 bc). A notable cluster of examples comes from Rhodes: IG XII 1 25 and 32; I.Lindos 130 (c.220 bc; cf. no. 165 in which a Rhodian proxenos from Carthage also figures, but apparently as the maker of the thank-offering dedicated). A further set of dedications, the fourth-century Athenian phialai inscriptions, mention proxenoi alongside Athenian citizens and metics with isoteleia (IG II2 1570 l.22; SEG 18 36 B ll. 16–21), though whether because they manumitted the dedicators (e.g. Lewis 1959, recently restated by Vlassopoulos 2011), or unsuccessfully prosecuted them for non-payment of the metoikion (so Meyer 2010, cautiously accepted by Papazarkadas 2012), is disputed. 24 On financial gifts, see Chapter 1, n. 142. For the public loans from pre-existing proxenoi, see Migeotte (1984): no. 24 (= IG V 1 1146, Gytheum, 71/70 bc—the interest rate charged by these individuals, the Cloatii brothers, was 48 per cent despite their status as proxenoi); 70 (= IG XII 9 900A c, Chalkis, second century bc); 117 (= IG XI 4 1049, first third of the third century bc); as well as, perhaps, 61 (IG XII 5 112, Paros, fourth century bc); to these add, SEG 43 448 (Philippi, c.180–170 bc?). Public loans made by individuals are also cited on a number of occasions in the list of reasons given for granting them proxenia, so Migeotte (1984), nos. 10, 25, 48, 56, and no. 9 (I.Oropos 303, 240–230 bc, a decree promising proxenia to anyone who lent a talent or more for the rebuilding of the city wall). To these now add Horos 22–5 (2013), 453–4. 20

Introduction

11

koinon;25 and an oracular consultation concerning an individual’s status as proxenos.26 In addition, we also possess a rich onomastic record for proxenia, of individuals given the name ‘Proxenos’ (Proxenides of Knidos was in fact the son of a man named Proxenos) or called after the polis from which their father had received proxeny (e.g. ‘Samios’ or ‘Korinthios’). These forms of proxeny attestation are important for nuancing and enlivening our understanding of what proxenia meant to individuals and how it was exploited by them. They require careful interpretation to understand in what sense these isolated and less obviously uniform attestations are more broadly significant and how they are exceptional—but, as I argue, similar questions also need to be asked of the record of proxeny decrees which is too often taken as straightforwardly representative. Alongside the substantial epigraphic record for proxenia, it is perhaps unsurprising that the literary evidence, on which the earliest commentators relied heavily, has been emphasized less in some treatments.27 However, even accepting its Athenian bias, the literary record is still remarkably rich and diverse and offers an essential complementary perspective to the view of proxenia provided by our documentary evidence. Whereas proxeny decrees, by their very nature, deal with proxenoi only at the point at which they are named as such, our historiographical sources principally present us with proxenoi in action. The Attic orators, conversely, allow us to explore examples of how proxenoi could be presented in political discourse—crucial given the fact that proxenoi were, in essence, the creations of political speech-act. Attestations of proxenia in other sources—such as the plays of the Athenian dramatists or the odes of Pindar—help to reveal the significance of this institution before we possess widespread epigraphic testimonia of it, and its importance as a source of prestige. These different types of material, literary and epigraphic, present us with contrasting perspectives on proxeny, illuminating different themes and, to an extent, different periods. The interpretation of these differences has been central to the persistence of narratives of the decline of proxenia which have drawn much of their force from 25 26

Praktika (1930), 35 (with ph. 34; second or first century bc). Lhôte (2006), no.15 (SEG 56 663, 300–167 bc) with Mack (2014) and Chapter 2,

n. 12. 27

Marek (1984), esp. 482.

12

Proxeny and Polis

the contrast which exists between the material available for the Classical period and that available for the Hellenistic period. For the Classical period, in addition to the limited epigraphic material, focused on Athens (whose interest in the use of proxenia as a practical tool of imperial control seemed clear),28 historians have been confronted with literary accounts of proxenoi in which the importance of the relationship of proxenia in shaping the actions of individuals was constantly stressed. For the Hellenistic period, conversely, in considering proxeny, scholars have been principally faced with an expanding number of more or less concise, highly monotonous and formulaic decrees in which individuation (the reference to particular concrete deeds) is relatively rare. For historians hoping to learn the reasons why a particular grant of proxenia was made to a specific individual, the formulaic language, which the framers of these decrees used, unsurprisingly rings hollow and empty without the reassurance of strong contemporary literary representations of functional proxenoi.29 Gauthier, in his rebuttal of Gschnitzer’s argument, revealed some of the flaws in this interpretation and stressed, especially in relation to the literary sources, the importance of taking the nature of the evidence into account. He pointed out that the principal reason why we do not have many post-Classical accounts of active proxenoi is that the volume of surviving contemporary historiography declines dramatically.30 One of the central aims of this study is to arrive at a similarly nuanced understanding of the much larger and more complex epigraphic record of proxeny decrees. Because of the numbers in which they survive and have been published, proxeny decrees represent an excellent means of exploring the functions which honorific epigraphic monuments played and the significance of the recurrent formulae which the proposers and composers of decrees felt that 28

Meiggs (1949). e.g. Klaffenbach (1966), 84–5, noting a particularly striking example—an inscribed Delian decree in which the names of the original proposer and honorand were erased and overwritten, but the formulaic description of the honorand left unchanged (IG XI 4 777, Delos, beginning of the second century bc). 30 Gauthier (1985), 136. The focus of Polybius, part of whose work does survive, is very different from his predecessors—the domination of Rome. Polybius’ somewhat contemptuous aside ‘and indeed it was Timaeus who discovered the stelae within temples and the proxenies (inscribed) on the door posts of temples’ (ŒÆd c › a OØŁ ı  ºÆ ŒÆd a K ÆE çºØÆE H  H æ Æ K ıæÅŒg ÆØ  KØ, Plb. 12.11) suggests that Timaeus at least made far more use of this sort of material. 29

Introduction

13

it was important to include. This richness also makes proxenia an important case study for analysing and interpreting trends and trajectories in the epigraphic record and their significance for the institutions which this record attests. A fundamental requirement for all of these studies, however, is an understanding that the inscription of decrees was always partial—that processes of selection were involved in singling out particular decrees and particular proxeny decrees for inscription, the biases of which have left us with what is, despite its reassuring size, a very unrepresentative record of proxenoi.

I.3. SELECTIVE INSCRIPTION The fact that Greek cities did not inscribe all of their decrees has long been remarked upon in scholarship on the ancient world.31 The publication clauses contained by so many of the extant texts attest to the necessity of taking a positive decision to inscribe a particular decree—and at almost all poleis, only a small proportion of the decrees which were passed would ever have been committed to stone. Proxeny decrees, for all that they constituted the single largest category of inscribed decree, were no exception to this general principle of selective inscription. Proxeny decrees were more likely to be inscribed than many other sorts of decree, since one of the most important functions which the act of inscription served was honorific, and proxeny decrees are likely to have been the most frequently promulgated form of honorific decree. However, in only a handful of cases is it at all likely that poleis routinely inscribed all or even most of their decrees granting proxenia over a prolonged period of time. Third-century Delphi, late third-century Oropos, independent Delos, and, perhaps, fourth-century Athens are the only likely candidates

31 Wilhelm (1909), 271–5; Klaffenbach (1966), 70–1; Rhodes with Lewis (1997), 525. This view has recently been challenged, at least in relation to the Athenian material, by M. J. Osborne, but the arguments he adduces, which rest on assertions concerning the necessity of inscription for publicity (Osborne 2012, esp. 48–52), are unconvincing and can account for neither the changing trends in inscription visible at Athens itself (see Chapter 5, Section 5.1 with figs. 3–6) nor the epigraphic practices of other poleis.

14

Proxeny and Polis

and these four poleis together account for more than half of the proxeny decrees which have been published.32 Elsewhere, however, at almost all poleis most of the time, inscription remained very much the exception and there was never any necessity to inscribe grants of proxenia. We can get a sense of how unrepresentative our data collectively are likely to be by working from our evidence of complete proxeny networks. Lists and catalogues of proxenoi, for all that they have been neglected in previous studies of proxenia,33 provide us with invaluable information concerning the size and distribution of poleis’ networks of proxenoi at particular moments in time. The rich, contextualized datasets which they represent clearly highlight the shortcomings of the epigraphic record of individually inscribed proxeny decrees. From these texts even insignificant poleis emerge as being represented by much more substantial networks of proxenoi at particular moments in time than we could possibly guess from the evidence of individually inscribed decrees alone. For instance, Karthaia, one of four minor poleis on the Aegean island of Keos, is attested as having possessed a proxeny network comprising in excess of 86 proxenoi at a particular moment in the first half of the fourth century bc.34 Collectively, this material indicates that regular grants of proxenia and large networks of proxenoi were an absolutely normal feature of life in the Greek poleis. Most poleis during most years probably appointed at least one proxenos (our evidence indicates that higher rates of granting were normal), and it is partly for this reason that making grants of proxenia came to be identified with acting as a polis. By contrast, fewer than 20 per cent of the poleis attested in our dataset as inscribing proxeny decrees are represented by more than ten proxeny decrees. If we use Karthaia’s network—which emerges as small on the basis of comparisons with other proxeny networks of minor poleis—to

32 Total number of proxeny decrees preserved for each polis: Delphi, 735 decrees; Oropos, 207 decrees (excluding decrees of the Boiotian koinon erected at Oropos); Delos, 239 decrees; Athens, 151 decrees (111 other decrees grant citizenship but the identity of the honour granted to non-citizens in a further 174 cases is uncertain). These numbers are drawn from the accompanying database of securely preserved proxeny decrees, on which see Chapter 5, Section 5.1. 33 Marek devotes only four pages to this important category of inscription (Marek 1984, 134–7); Gschnitzer offers the most detailed treatment (Gschnitzer 1973, cols. 695–700). 34 Appendix no. 10.1.

Introduction

15

calculate crudely the scale of proxeny networking, we arrive at a very conservative estimate of 1.2 million for the total number of grants made during the five hundred years that proxeny thrived, of which our sample of nearly two and a half thousand individually inscribed decrees would represent 0.2 per cent.35 Our sample of inscribed proxeny decrees is clearly partial, but it is also unrepresentative for reasons that have nothing to do with low rates of archaeological recovery. This is because the inscription of a particular proxenos’ decree was usually an additional honour, over and above the particular package of honours this status conveyed, involving the polis in certain expense (which few of the other honours granted did).36 The inscription of proxeny decrees was thus not merely partial, it was also usually selective—an honour more likely to be granted to precisely those honorands who were already exceptional for some reason or of particularly high status. In aggregate, this bias towards unusual proxenoi has the effect of seriously distorting the material presented by individually inscribed decrees on which, for instance, Marek draws in his prosopographical analysis of decrees. Moreover, the c.750 individuals Marek was able to categorize by profession, status, or occupation, which he used as the basis of his analysis, represent a sample that is unrepresentative even of these selectively inscribed decrees in that the individuals he was able to pigeonhole were those sufficiently important to be attested elsewhere, or those who had their status and/or actions detailed in the decrees honouring them (in contrast with the lack of individuation which is otherwise the rule). The proxeny decrees of Priene vividly illustrate both how selective the decision to inscribe a proxeny decree could be and the effects of this on our sample of material. Despite the richness of the public epigraphy of Priene in many other areas, we possess only five proxeny decrees. All five honour proxenoi who are obviously exceptional or 35 In this calculation, I assume an average network size of 80 proxenoi for the c.1,000 poleis counted by the Copenhagen Polis Centre, renewed three times a century from 500 to 1 bc (a rate of 1.2 proxeny grants per polis per year). Clearly this calculation glosses over an enormous degree of variation, but, as comparative analysis of proxeny networks in Chapter 3 shows, even allowing for this, these are very conservative figures. Many, much larger networks would have been renewed more frequently than this. See Knoepfler (2001a), 425–30, for a valuable discussion of these difficulties in relation to the surviving proxeny decrees of Eretria. 36 See Chapter 2, Section 2.3.1.

16

Proxeny and Polis

unusual: one is for Alexander the Great’s general, the future king, Antigonos Monophthalmos (I.Priene 2); another includes the grant of an honorific statue—an honour which very few proxenoi ever received (I.Priene 3); a third gives the honorand a plot of land along with proxenia—a gift with only one parallel in our record of inscribed decrees (I.Priene 6);37 another is not properly a proxeny decree at all, but rather a reinscription of a proxeny decree for an individual who was important enough for the original monument to be targeted for destruction by the tyrant who reigned briefly at Priene (I.Priene 12);38 the last of the known proxeny decrees does not honour individuals so much as Priene’s connections with other poleis, granting proxenia to foreign judges from Phocaia, Astypalaia, and Nisyros (I. Priene 8). These would in no sense have been typical of the proxeny decrees which Priene would have passed, year on year. They were inscribed precisely because the honorands in question, especially in the first four cases, merited exceptional honours. In these decrees proxenia played an important role in defining the position of these honorands, but these particular decrees were not really about proxenia in any narrow sense. The Prienean material represents a particularly good case study because the markers of exceptional status in these decrees are so clearly delineated. In many other proxeny decrees, particularly important individuals, or honours granted for particularly exceptional services, are not necessarily so clearly marked, except, perhaps, by the decision to inscribe the decree in question in a context in which few others were ever inscribed. More than 70 per cent of poleis attested as inscribing proxeny decrees are, like Priene, represented by only five or fewer proxeny decrees throughout their entire histories, and 60 per cent are represented by fewer than three. Wide variations in the archaeological visibility of different epigraphic practices also have an important role to play. The all but complete absence of inscribed proxeny decrees from the Greek cities on Sicily and in Magna Graecia (only four survive)39 is not evidence 37

IG XII 9 196, with Knoepfler (2001a), 180–1. On this decree, see Chapter 2, p. 95 with n. 18. 39 IG XIV 612 (Rhegion, c.100 bc); IGUR 2 (Akragas, c.100 bc); SEG 30 1120–1 (Entella, 254–241 bc?). See also IGUR 3 (Malta, c.100 bc). The reference in the document recording an embassy from Centuripae to Lanuvium to proxenia between the two communities, restored when it was first published (SEG 42 837 l. 12), has been rejected in subsequent editions (e.g. SEG 56 1083). 38

Introduction

17

for the absence of proxenia in the west, and nor is it simply accidental. In comparison with stone inscriptions, texts on bronze, the preferred medium for honorific decrees in this region, survive only in exceptional circumstances.40 By contrast, literary references to grants of proxenia by these poleis, the well-formed formulae of the decrees that do survive, and the fact that proxenia is comparatively well attested in use by associations of technitai in Sicily and Magna Graecia, all suggest that the use of this institution was as common and widespread there as it was in Mainland Greece.41 How common the use of proxeny was in other areas where proxeny decrees are unattested and for which other material is lacking (especially among the Hellenistic poleis of inland Asia Minor and the Middle East) we can only guess, though the elaborate and sophisticated response given to the asylia campaign of Magnesia on the Maiander by Antiocheia in Persis (and seconded by Seleukeia on the Tigris, Apameia on the Seleias, Seleukeia by the Red Sea, Seleukeia on the Eulaeus—Susa—and Seleukeia on the Hedypon) should warn us against assuming that absence of evidence for proxenia can be taken as evidence of its actual absence here as well.42

I.4. APPROACHING PROXENIA The question, then, is how to proceed to analyse the complex record for proxenia which is unrepresentative in these crucial respects. 40

On this, see Prag (2002), 24. Cicero, in particular, describes grants of proxeny (translated as publicum hospitium) made by the Syracusans for himself and his cousin, Lucius, of which they were given bronze copies—clearly presented here as an additional honour (Cic. Ver. 2.4.145). The existence of networks of proxenoi between Greek cities in Italy is also attested by Dionysius of Halikarnassos (Dion. Hal. 15.5.2 records an incident involving a number of Neaopolitan proxenoi at Tarentum). More decrees of proxenia made by associations of technitai survive from the west than the Greek east (Le Guen 2001, vol. I, nos. 72, and 75–7), see Chapter 4, n. 133. The chance survival of a pair of bronze stelae from Akragas and Malta granting proxenia to a Syracusan active in Rome c.100 bc are a strong indication of how commonly this institution was used in the west (IGUR 2–3). 42 Rigsby (1996), no. 111. For similar reasons, I find it hard to be confident that the absence of inscribed decrees from Macedon should necessarily be taken as evidence that proxenia was a prerogative of the central authorities there, contra Hatzopoulos (1996), vol. 1, 366–9. Livy 21.12–13 is a probable example of proxenia in Spain, see Chapter 2, n. 164. 41

18

Proxeny and Polis

Central to the approach taken in this study is the analysis of material—in particular the surviving lists and catalogues of proxenoi— which allows us to arrive at a more accurate understanding of the parameters of ancient proxeny networking. This material provides a crucial background for exploring poleis’ patterns of interaction and contextualizing other material, to explore the place which proxenia had in the life of poleis, and the broader functions which this and other institutions played. It is also possible to identify ways in which even the record of individually inscribed decrees, for all its biases, can be used to illuminate proxenia more broadly. In particular, the first chapter concentrates on precisely that aspect of proxeny decrees which has been ignored because it fails to individuate and is non-specific—the recurrent, highly stereotyped formulae used in decrees to describe the recipients of grants of proxenia. In this, the epigraphic record of individually inscribed decrees provides an unrivalled resource for understanding how an ancient institution could be invested with a particular content and meaning, a paradigm of proper proxenos—and polis—behaviour. At the same time this study is also interested in exploring the trends which our record of proxeny decrees reveals. The basis for this is a database of nearly two and a half thousand securely identified proxeny decrees, assigned to different periods using quantitative probability distribution, which allows us to directly compare inscriptions dated with varying degrees of precision. The aim in doing so is to understand the complex relationship between the practice of inscribing proxeny decrees, or a selection of them, which arose and spread very widely, and the institution of proxenia more broadly. Literary sources also provide an important alternative perspective on this paradigm, the way it functioned in practice, and their interpretation is, in turn, informed by our study of proxeny networks. The quantitative indications which we glean from analysis of lists and catalogues of proxenoi reveal how common grants of proxenia were, so that it becomes likely that the majority of individuals prominent enough to be mentioned particularly in our historiographical sources were proxenoi of one or more polis. The fact that these links are, by comparison, so infrequently mentioned by authors confirms the sense we get from analysing their occurrence in context—that literary references to specific links of proxenia are never merely incidental, that this relationship is central to the interpretation of the passages in question, and thus that these passages are important evidence

Introduction

19

which need to be taken into account in any attempt to understand this institution. To interpret this material in general, which is not new but has never been examined systematically in these ways before, and to explore what it can tell us both about proxenia more narrowly and the world of Greek poleis in which this and other institutions were used, I have found it helpful to draw on a number of complementary theoretical frameworks developed in other fields. In particular, models for understanding institutional innovation and adoption created by New Institutionalist scholars in the field of Organizational Studies are used in the first chapter for understanding the spread and stability of proxenia, and directing our attention to what this says about the organizations (poleis) responsible.43 Similarly, terminology used in Social Network Theory is helpful both in describing the features of ancient networks of proxenoi and characterizing them more generally.44 Finally, social theories of interstate relations, especially those developed by Alexander Wendt, provide a useful framework in the fourth chapter for interpreting the motivations of Greek poleis in engaging in interactions with each other through a specific set of inter-polis institutions including proxenia.45

I.5. ARCHITECTURE OF THE ARGUMENT Each chapter of this book explores a different aspect of proxenia, beginning with an examination of the way in which proxenia was understood (Chapter 1) and concluding with a study of the manner of and reasons for its demise (Chapter 5). At the same time, however, these chapters are also intended to build a progressive argument concerning the role performed by interstate institutions, by examining the way in which proxenia functioned at a succession of different levels—conceptually (Chapter 1), within individual poleis as political communities (Chapter 2), from the perspective of poleis in relation to their networks of proxenoi (Chapter 3), and within the context of the wider system of interstate institutions and inter-polis society more generally (Chapters 4–5). 43 45

e.g. March and Olsen (1989). Wendt (1992), (1999).

44

e.g. Kadushin (2012).

20

Proxeny and Polis

The first chapter begins by developing a specific interpretative model for understanding how proxenia, as an institution, was invested with meaning, by examining the narrow range of terms with which the recipients of proxenia were described to argue that these expressed a specific paradigm of expected proxenos behaviour. This paradigm is then compared with the evidence we possess for the specific activities of proxenoi from literary and epigraphic sources in order to explain how the more general descriptions of proxenoi relate in the proxenos-paradigm to the performance of particular intermediary services. I then explore what the stability of the central elements of this paradigm tells us about proxenia and the importance of the functions which it performed. The second chapter then examines how this abstract paradigm of proxenia, with its idealizing assumptions, played out in relation to individual proxenoi in the reality of factional politics in the Greek city. In particular, it explores a wide range of different sorts of material, especially from the epigraphic record and Attic oratory, to understand what motivated individuals to perform the functions associated with proxenia, arguing that it was the prestige value of this institution in the context of elite competition within their own poleis. In this we see the essential symbiosis of honour and function which was the foundation of proxenia and other inter-polis institutions—with individual elite members needing to perform publicly as proxenoi in order to capitalize on the prestige which this honorific connection to an external polis conveyed. The third chapter broadens out this study, and explores how poleis viewed their proxenoi en masse, as geographically distributed networks. Using proxeny lists to illuminate the potential range of sizes of proxeny networks, and their rhythms and rates of granting, it exploits this material to illuminate different patterns of inter-polis interaction and explore how proxeny grants functioned on a symbolic level in constructing the position of a polis relative to its peers. The fourth chapter then develops this insight into the symbolic function of proxenia, and uses this institution as a case study to explore the functions performed by the wider system of inter-polis institutions, including asylia, syngeneia, and theōria. It builds on John Ma’s seminal article on peer-polity interaction (Ma 2003) by using constructivist models from International Relations Studies to explain why poleis invested time and institutional effort in these sorts of interactions. I argue that so much effort was expended on engaging

Introduction

21

with these different international institutions within the anarchical system of the Classical and Hellenistic world because this was the way in which poleis constructed their identity in relation to each other as polis-actors. The final chapter of the book explores the importance of the role played by proxenia within the Greek world by considering the reasons for its disappearance around the time of Augustus in terms of the structural changes which the dynamic of the Greek world underwent at that time. It argues that, although the massive quantitative reduction evident in our record of individually inscribed decrees cannot be taken as direct evidence that proxenia ceased to be granted, this reduced monumental emphasis reflects a more general shift, which is also apparent in our evidence for other institutions during this period. However, this chapter shows that rather than representing the effectiveness of Rome as a central authority in repressing these sorts of activity, this is a product of a wider, polis-centred process of reorientation in which relations with the Roman authorities at the centre replaced inter-polis connections as the source of communal identity and prestige.

1 The Anatomy of an Ancient Institution The Proxenos-Paradigm

When . . . was phrouros, in the month Anthesterion . . . , when Autonomos was prytanis, in the duly constituted assembly Metagenes, son of Metagenes, proposed that: since Adolos, son of Adolos, of Sigeion is [a good man] and zealous (prothumos) towards the polis, he and his offspring be named proxenoi and euergetai of the polis (of the Kians), and that the following rights be given to them the right to possess real estate at Kios (enktēsis), and the right to travel to and from Kios by sea and land without threat of seizure and without the need for a treaty, both in war and peace; and that they also be given preferential seating at festivals (proedria) and freedom from tax (ateleia) on goods which they import and export for their own use . . . and this proxenia is to be inscribed on a stele which the hieropoioi are to erect . . . in the sanctuary of Athena . . . A proxeny decree from Kios, Asia Minor (fourth century bc).1 Paramonos son of Boethos proposed the motion that: since Simias lives his life well-disposed to the polis of the Oropians, performing services for those who need (his aid) both on their own private business and in the public affairs of Oropos, and because he always says and does what is in the interests of the Oropians, be it decreed by the demos of the Oropians that Simias, son of Tattabos, of Lampsakos be proxenos of the polis of the Oropians, both himself and his offspring, and that he should have the right to possess real estate at Oropos 1

I.Kios 1.

The Anatomy of an Ancient Institution

23

(enktēsis) and the tax status of a citizen (isoteleia) and freedom from seizure (asylia) both in war and peace, by land and by sea, and all the other rights which the other proxenoi possess. A proxeny decree from Oropos, central Greece (first half of the third century bc).2

Institutions are historically contingent social constructions which shape human action and behaviour. How they do this depends on their precise content and meaning in context—the reasons why they are relevant to individual actors in a particular society and persist in a particular form. The institution of proxenia was no exception. However, what proxenia meant, how far and when its meaning changed, and what functions it therefore performed, are issues which remain contested today.3 This chapter therefore proposes a new model for understanding how proxenia was used and understood, and how it could both be honorific and at the same time oblige the recipient to perform certain sorts of service for the granting polis.4 It argues that proxenia, as an institution, derived meaning from a widespread and stable understanding of what proxenoi should be and do, a model of proper proxenos behaviour which I term the proxenos-paradigm. This approach draws on modern theories developed under a renewed interest in institutions—New Institutionalism—which offer ways of explaining how institutions operate, as well as the dynamics underlying their evolution, stability, and spread. New Institutionalism arose, in part, as a reaction against the extremes of rational-actor theory and its emphasis on the selfish pursuit of maximal gain. Instead it concentrates on the role played by institutions in shaping actors’ motivations and directing their pursuit of them.5 New Institutionalism is represented in a diverse range of fields, of which the most influential strand for the study of ancient history has so far been New Institutional Economics.6 This chapter, however, makes use of 2

3 I.Oropos 10. See Introduction, Section I.1. Thus Marek (1984), 143–6, rightly dismisses the suggestion of Monceaux (1886), 13–14, that proxenoi were contractually obliged to perform particular duties, but, rather than seeking alternative mechanisms by which this could work, rejects the idea that there could be particular functions associated with proxenia at all. 5 DiMaggio and Powell (1991); Keohane (1988), 382. 6 Douglass North is the central figure in New Institutional Economics, see North (1990); for an overview of the recent work in this field, see Brousseau and Glachant, eds. (2008). Bresson (2007–8) is the fullest treatment of these themes in relation to the ancient world, but see also Ober (2010). 4

24

Proxeny and Polis

theoretical tools developed by New Institutionalist scholars in Organizational Studies, especially the concept of the ‘logic of appropriateness’. Expounded most influentially by March and Olsen, the ‘logic of appropriateness’ emphasizes the way in which individuals, while participating in different sorts of institutions, self-consciously perform socially defined roles and make decisions by using as a primary criterion the appropriateness of a particular action to the role or roles in question (e.g. as jurors, members of a Parent-Teacher Association, or, indeed, as ‘customers’).7 The concept of the ‘logic of appropriateness’ is useful in explaining how proxenoi could be expected to perform particular sorts of function without the necessity of invoking quasicontractual constraints. Proxenia was defined by a set of assumptions which both poleis and proxenoi shared with regard to this role—the proxenos-paradigm. Social expectations rather than legal compulsion underwrote the obligations which proxenoi were felt to have. The central characteristics of this proxenos-paradigm can be identified in the inscribed proxeny decrees in the descriptions which they give of the individuals honoured and their actions, as in the two examples which are used as the epigraph of this chapter—the reasons why a grant of proxenia was made. These descriptions, however, are usually very concise and highly formulaic. Their function was not to identify or differentiate honorands. Instead, they evoke the proxenosparadigm as a stereotype to justify the grant of proxenia to particular recipients. In these decrees, proxenoi are appointed on the basis that they behave as proxenoi should. They reveal that the proxenos was understood as the non-citizen friend of the polis in an external community who proved his affection for it over a prolonged period by performing services for its citizens. Although the services which proxenoi could be expected to perform are not precisely defined in the concise descriptions found in proxeny decrees, the structural function of proxenia is made clear in these texts. The proxenos is defined as an intermediary figure who could be expected to help visitors from the granting city negotiate any difficulties or barriers in his own political community which their status there as strangers and non-citizens might cause them. Proxenia derived its meaning from this proxenos-paradigm, and from proxenoi who acted in accordance with it. However, because it

7

March and Olsen (1989), 160–2; (2006).

The Anatomy of an Ancient Institution

25

constituted the most widely granted and understood form of privileged status, proxenia could also on occasion be used more broadly to express relations of friendship with atypical honorands. These atypical grants occurred on a broad spectrum, from instances where the honorific element was simply more prominent than the practical services associated with the institution, to a handful of rare cases in which the performance of these practical functions does not seem to have been at issue at all (in particular, in the small number of grants that were made to women).8 The most extreme examples of this are provided by the occasional collective grants of proxenia made by poleis under great strain to large groups of individuals en masse, extending and adapting their pre-existing repertoire of honorific symbols to express their gratitude to, and continuing reliance on, groups of mercenaries for their defence, or even entire communities.9 The generic descriptions in proxeny decrees reveal, however, that proxenoi in general took their character from recipients who did conform to the proxenos-paradigm, performing specific sorts of intermediary service in the context of their own communities. 8 Eight decrees granting proxenia to women are known: Cabanes (1976), 565 n. 43; CEG 877 C (Delphic decree for Archon, Alexander’s governor of Babylon, and his mother and brothers, after 321 bc); SEG 38 464 (decree of the Prasaibians, after c.170 bc); IG IX 2 62 and FD III 3 145 (Decrees of the Aitolians and Chaleians for Aristodama, the poet, from Smyrna, 218/217 bc); FD III 3 249–50 (Delphic decree for Polygnota, a harpist from Thebes, 86 bc); Syll.3 689 (Delphic decree for a harpist from Kyme, 134 bc); IG II2 1136 (Delphic decree for Chrysis, priestess of Athena Polias at Athens); IG XII 6 471 (Samos, late first century bc or early first century ad): this inscription, strictly speaking, honours a woman who is already a proxenos rather than granting proxenia. On proxeny grants for women, see Marek (1984), 129–30 with n. 55 (but note that IG IX 2 458 and XII 5 812 do not involve grants of proxenia; I3 178 may or may not grant proxenia to the wife of the honorand). On the honorific and symbolic functions performed by these different grants, see Mack (2014). It is important to note, however, that in the first four of these cases, the grant to a woman was made only in association with a primary, male, recipient; in all but one of the other cases (Syll.3 689) we know that a male relative, who could have fulfilled the services associated with this role, was simultaneously named proxenos. 9 The cardinal example of this is the mass grant made by the polis of Lilaia to nearly three hundred mercenaries sent for their defence by King Attalus (FD III 4 132–5). Similarly striking manipulations of this symbolic language include the Delphic decree declaring that the city of Delphi would collectively act as proxenos for the Sardians (Syll.3 548; but note that the Sardians themselves are not granted proxenia), the Molossian grant of proxenia to the Akragantines (Syll.3 942), the grant made by the koinon of the Aterargoi to the Pergamioi (SEG 15 411), and the famous decree of Mausolus declaring all the Knidians his proxenoi (RO 55). On these and other examples see Gschnitzer (1973), cols. 672–6, with the addition of SEG 44 1148 (Arykanda and Tragalassos, c.200 bc).

26

Proxeny and Polis

None of the elements of the proxenos-paradigm delineated here would have surprised Wilhelm who wrote on proxenia in the 1940s, or even Meier, writing in the 1840s. This approach, however, allows us to explore the functions which the language found in proxeny decrees performed and the relation of the stereotypical descriptions they contained to depictions of proxenoi in the literary sources. Although honorific language has, in recent years, received more attention than formerly, comparatively little work has been done on the stereotypical descriptions of proxenoi that are the focus of this chapter.10 By concentrating on the language of proxeny decrees, we can explore and compare representations of proxenoi from a large number of poleis during different periods. This comparison illuminates the intended meaning and significance of the potentially ambiguous words and phrases used, and allows us to get a much more detailed and accurate understanding of how proxenoi were conceived. What is striking is that the same basic proxenos characteristics, and the same services associated with proxenia, define the use of this institution wherever it is found. The first inscribed decrees we possess, from fifth-century Athens, reveal essentially the same proxenosparadigm as the decrees we possess from the period when proxeny inscription was much more widespread—from the fourth to first centuries bc, from the Black Sea to the western Mediterranean. When this precise proxenos-paradigm arose is unclear, but literary sources from the fifth century bc, some of which predate the earliest inscribed decrees, suggest that it was then fully developed and widespread.11 The reasons for this extraordinary homogeneity and stability shed light on the structure of this world of poleis. Proxenia thus represents an important case study for exploring the importance of the honorific system of inter-polis intercourse that characterized the Greek world. It also allows us to explore the fundamental change in dynamic which this world underwent when it came firmly under the control of the Roman emperor, and proxenia disappeared. 10 These terms receive only very brief mention in Marek (1984), 333–4. Much of the work which has been done has focussed on Attic oratory (e.g. Dover 1974; Ober 1989, 248–92) or the language of Athenian inscriptions (Whitehead 1983; 1993; Veligianni-Terzi 1997; Kralli 2000). See now also, on the decrees of Iasos, Fabiani (forthcoming b). Gauthier (1985), esp. 140–5, takes a broader, comparative approach, which I develop here. 11 On the earlier literary attestations, see Wallace (1970); on attempts to identify distinct uses of the term proxenos, see below, n. 204.

The Anatomy of an Ancient Institution

27

This chapter therefore begins by analysing the way in which those appointed as proxenoi are described, the connection between this description and communal expectations of their future behaviour, and the central characteristics and specificity of this proxenos-paradigm. The second part then seeks to identify the services which were part of this paradigm, arguing that they need to be understood in the context of the proxenos’ assumed context for action—his own polis or another political community external to the granting polis in which he fulfilled an intermediary role. The final part then argues that the essentials of this proxenos-paradigm can be identified as remaining consistent and stable until the end of the first century bc and draws on theories of institutional isomorphism developed by New Institutionalist scholars to understand this stability and what it says both about proxenia and the world in which it operated.

1.1. FORMULAIC DESCRIPTIONS OF PROXENOI The highly stereotyped language of decrees granting proxenia is familiar to anyone who has ever leafed through an epigraphic corpus. Although the rigidity of these formulae, even within the decrees of a single city, should not be overstated, they are, because of this repetitive quality, among the most easily and confidently restored of inscriptions. It is perhaps because of this apparent familiarity that the significance of these descriptions of honorands has been largely overlooked. As authoritative representations of proxenoi, describing the qualities and the activities that made an individual a suitable candidate for the receipt of proxenia, they provide our best evidence for the conception and function of this institution. It is clear that the inscription of decrees was highly selective—and therefore that our surviving sample of proxenoi is biased—but, because they construct a stereotype, these formulaic elements of proxenos-descriptions can nonetheless be taken as representative evidence of the way in which proxenia was conceived.12 After invocation, heading, prescript, and enactment formula, but before the award of proxenia and whatever attendant honours are

12

See Introduction, Sections I.3 and I.4.

28

Proxeny and Polis

granted, in all but the most concise decrees, we find a motivating clause setting out the reasons for which a specific motion was proposed and why, therefore, it was passed.13 In longer examples, these may be both ‘retrospective’ and ‘prospective’, setting out what the honorand had done, did, and would continue to do for this nomination to be appropriate (introduced by KØ, epeidē, ‘since’ or ‘because’) as well as what those responsible hoped to achieve by this decree (introduced by ¥ Æ, hina, or ‹ø, hopōs, ‘so that’ or ‘in order that’). This first type of clause, introduced by epeidē, can be more or less elaborate, sometimes incorporating detailed descriptions of the deeds of the honorand. The basic, invariably present element, however, is a general, stereotypical description of how the honorand behaved towards the city in question and what he did for it. Generic descriptive elements which continually recur in the decrees of different poleis, expressed in similar or equivalent phrases, have usually been ignored because they fail to individuate or explain what those commenting on inscriptions are often most eager to learn—the reasons why a particular grant was made at a particular time. In recent scholarship they have sometimes been treated as an empty sham, recording ‘virtual’ (non-existent) benefactions honoured in an attempt to provoke a ‘real’ euergetical response.14 In fact, they point to the existence of a ‘proxenos-paradigm’, a set of general and specific expectations of this role in the context of which a given honorand’s actions were analysed. It was important to inscribe these phrases because they reflected, or rather condensed, the language used to discuss and think about this institution in public contexts. These were the terms, as their occurrence in motivation formulae makes clear, in which the proposer of a proxenos was expected to discuss his nominee, albeit with greater elaboration and delineation of examples, and it was against the proxenos-paradigm, which these terms evoked, that a candidate was judged.15

13

See Rhodes with Lewis (1997), 4–5, on this structure. Domingo Gygax (2009), 180–3. See similarly, albeit in relation to royal philoi, Mitchell (2009), 22: ‘no more than fishing trips for potential benefactors’. Veyne (1990), 107, discussing the honorific and didactic function of these conventional phrases, is closer to the mark. 15 No such speeches survive in our literary corpus despite the frequency with which they must have been made, but we do possess a model in Hyperides’ parody of the traditional elements of this genre. In the surviving fragments of his speech against Demades for proposing an unsuitable candidate for proxenia (Hyp. fr. 76), he lists the 14

The Anatomy of an Ancient Institution

29

This language reflected and reinforced citizens’ expectations of their city’s proxenoi, but it also shaped their understanding of the relationship they had with any cities which had named them proxenoi and, indeed, informed them of the attitude and actions appropriate if they had any ambitions to become proxenoi of another community. Expressed in the continuous present, this paradigmatic description of the individual named proxenos also reached into the future. The honorand was named proxenos because he did and would continue to behave like one. The connection between this descriptive language and the expectations attached to the role of proxenos is made particularly clear in an early second-century Athenian decree bestowing proxenia on Charmion, from Kydonia in Crete, which also clearly illustrates many of the recurrent characteristics of the proxenos-paradigm.16

1.1.1. Charmion, son of Eumaridas En route to Delphi, on an official delegation from Kydonia, Charmion and his fellow theōroi disembarked at the Piraeus with the intention of continuing their journey by land.17 While in Athens, however, Charmion took the opportunity of contacting the civic authorities about obtaining for himself the honorific position of proxenos enjoyed by his father:18 ‘services’ against the interests of Athens which Demades should have mentioned in his proposal: ‘[Demades] proposes that it be decreed that he (Euthykrates) be proxenos because he speaks and acts in the interests of Philip and as cavalry commander betrayed the Olynthian cavalry to him and through this was responsible for the destruction of the Chalcidians, because after the capture of Olynthus he set the ransoms of the prisoners, because he worked against our city’s interests concerning Delos and, when the city was defeated at Chaironeia, neither buried any dead nor ransomed any prisoners’ (åŁÆØ ª æ çÅ Ø ÆP e r ÆØ æ  , ‹ Ø a غø fi ıçæ  Æ ŒÆd ºªØ ŒÆd  ØE, ‹ Ø ª  ¥ Ææå  f  OºıŁø ƒÆ æ høŒ غø fi , ‹ Ø F æ Æ ÆY Ø  F ƺŒØø Bæ  OºŁæ ı, ‹ Ø ±º  Å  OºŁ ı ØÅ c Kª H ÆNåƺ ø, ‹ Ø I æÆ  B fi ºØ æd F ƒæ F F ˜Åºø, ‹ Ø B ºø æd ÆØæØÆ  ÅŁ Å h  ŁÆł H Ł ø Øa h  H ±º ø PÆ Kº Æ ). 16 IG II3 1137(= II2 844, 193/2 bc). 17 On relations between Athens and Kydonia, attested by a remarkable number of texts, see Papazarkadas and Thonemann (2008), re-editing an Athenian decree recognizing syngeneia between these two poleis. 18 That he is explicitly said to have given an account of his own conduct (I ººªØ ÆØ, l. 56) suggests that the initiative was his. On grants of proxenia to the descendants of proxenoi, despite the hereditary understanding of this relationship, see Chapter 3, n. 40. (I ººªØ ÆØ, IG II3 1137 l. 56)

30

Proxeny and Polis

KØc Ææø e J ¯PÆæ ı F ˚ıøØ ı æ  ı ŒÆd Pæª ı Z  ¼ Ø  Æı e ÆæÆ Œı Çø H[]  ø HØ Æ æd ÆP F ØH ŒÆ ƺ Æ N e —ØæÆØA Łøæe I ƺ  N ˜ºç f e H  ºØ H æ ƪƪ ø ÆP e H æÆ ÅªH I ººªØ ÆØ E  ıºE   Æ æØŒc h ØÆ ŒÆd a ç’ Æı F æƪÆ ŒÆd N e º Øe Kƪªºº ÆØ åæ Ø    ŁÆØ ŒÆd NÆØ ŒÆd Œ ØE E ÆæƪØ  Ø N ˚æ Å ŁÅÆø· Since Charmion, as the son of our proxenos and benefactor Eumaridas, who proves himself worthy of the honours given to his father, having sailed into the Piraeus after being sent as theōros to Delphi by his fellow citizens, has, having been introduced by the generals, delivered an account of his inherited regard (for Athens) and of the services (for Athenians) which he has performed and swears that he will also be of service to those of the Athenians who come to Crete whether on public or on private business . . . 19

Charmion demonstrated that he was worthy of the honorific position enjoyed by his father, Eumaridas, in two ways. Firstly, he delivered an account of the regard for Athens which he had inherited from his father (presumably listing Eumaridas’ benefactions) and of the services that he had himself rendered to the Athenians. It is the second part that is most striking, however, a declaration exactly corresponding with descriptions of those granted proxenia in other inscriptions, yet placed in the mouth of the would-be-honorand (ll. 57–9).20 It shows clearly the connection, present but not so explicit elsewhere, between the way in which the honorand is described as behaving in the motivating clause and the way in which the polis expects and/or hopes he will continue to behave after he is named proxenos. This declaration also shows an awareness on the part of the would-be honorand that this sort of practical assistance to private citizens and official delegations from the granting polis in the context of his own city was the basis of proxenia. Charmion received the proxenia he sought and the way in which the grant was made and monumentalized further reinforces these themes:

19

IG II3 1137 ll. 51–9 (= II2 844). Other examples of declarations include: SEG 38 662 (Potidaia, 400–350 bc); I.Byzantion 1 (Byzantium, 175–171 bc); and I.Histriae 32 (Istria, second century bc). 20

The Anatomy of an Ancient Institution

31

. . .  æåØ ’ ÆP HØ ŒÆd c Æ æØŒc []æ Æ ŒÆd E KŒª Ø· r ÆØ ÆP HØ ŒÆd N e º Øe B[] [N]Æ ƃæ ø  Ø ıøØ c IØ Ø æ ŁÆØ IªÆŁe []Ææa B  ıºB ŒÆd F  ı ‹ ı i çÆÅ ÆØ ¼ Ø  r ÆØ. . . . He is to have the proxenia which his father had—and his offspring are to have it too—and in the future as he makes demonstrations of his personal predisposition (towards Athens), he is to obtain whatever benefit from the council and people he appears worthy of.21

Again we see the expectation clearly expressed, this time by the civic authorities, that Charmion will perform further benefactions as proxenos, but it is expressed in the context of a future need to reward him in response to these predicted services. In this the identity of proxenia as a euergetical institution is clear, involving the creation and manipulation of reciprocal but asymmetric relationships, in which symbols of honour and gratitude answer benefaction. The way in which benefaction is expressed is also important. Here—as we shall see, typically—it is the personal Æ¥æ Ø (hairesis), or preference of the proxenos for the honouring city, which is emphasized. This affection is represented as motivating him to perform services for the honouring city and it is this which qualifies him to be named as proxenos by the polis. The services he performs are also presented as important, however, as constituting proofs of his preference (l. 70) and actually provoking honorific responses on the part of the polis. One of the most striking features of this monument is the prominence of the father of the honorand, Eumaridas, within it. Charmion is introduced in the epeidē clause as the son of his father, the proxenos and euergetēs (52–3). It is Charmion’s ‘paternal proxenia’ ( c Æ æØŒc []æ Æ; ll. 67–8) which is sought and given and his ‘inherited regard’ (   Æ æØŒc h ØÆ; l. 56) which takes pride of place in the account delivered by Charmion justifying his claim to proxenia. The Athenian authorities accepted this emphasis. They took the extraordinary decision (though it is not recorded in Charmion’s decree) of reinscribing two decrees in honour of Eumaridas, the father, above that of Charmion, his son. In part this relates to the way in which proxenia itself was conceived as an inheritable bond, calqued in this respect on xenia, with which etymological links continued to be drawn.22 These decrees must have been inscribed to 21

IG II3 1137 ll. 67–70 (= IG II2 844).

22

Herman (1987), 132–8.

32

Proxeny and Polis

honour Charmion by honouring his father, especially in the case of the first, in which Eumaridas is given very high honours (praise, a gold crown, and bronze statue) for significant services.23 They also, however, recorded a lofty example for the son to live up to and, at the same time, clearly displayed the potential rewards for so doing. The decrees on this monument thus publicized a model of the hereditary proxenos relationship—to Charmion of benefaction rewarded in relation to his father, and to other proxenoi or would-be-proxenoi of a family of faithful proxenoi. To a wider audience, this stele would have been a symbol of the strength of ties binding the Athenians to their friends, of repeated honorific interchanges, and of the importance of Athens measured in terms of the prominence and resources of its foreign benefactors.

1.1.2. The defining characteristics of the proxenos-paradigm From these stereotyped proxenos descriptions, a set of recurrent, overlapping characteristics can be identified which collectively defined the role of proxenos: the personal preference of the proxenos for the polis in question; the manifestation of this preference over an extended period of time; the non-citizen status of the proxenos (i.e. their active membership of a different political community); and, finally, the continuing utility of the proxenos to the granting city as its benefactor, providing services in the context of his own political community. Analysis of the various ways in which these central concepts were expressed in the decrees of different poleis reveals how they interrelated, allowing us a much more nuanced appreciation of how this role was understood.24 Conversely, close consideration of the language used in decrees in which proxenia was not granted reveals the specificity of the proxenos-paradigm to proxenia. The different characteristic elements of these descriptions can be clearly identified in a typical Athenian motivation clause for a recipient of proxenia:

23

On these services, see below, p. 66–7. For considerations of this language in general Dover (1974) is fundamental; Whitehead (1993) is important for focusing on epigraphic examples; Veligianni-Terzi (1997) is a good guide to the Athenian epigraphic material of the Classical period. 24

The Anatomy of an Ancient Institution

33

KØc ¨ ªÅ › ˝ÆıŒæÆ  Å Icæ IªÆŁ K Ø æd e B  e ŁÅÆø ŒÆd  E ‹ Ø Æ ÆØ IªÆŁe ŒÆd NÆØ f IçØŒ ı ı ŒÆd Å ÆØ ŒÆd F ŒÆd K HØ æ Ł åæøØ ŒÆd ÆP e ŒÆd ƒ æª  Ø ÆP F . . . Since Theogenes of Naukratis is a good man in relation to the demos of the Athenians and does whatever good he can to those (of the Athenians) who come to him (at Naukratis) whether on private or public business, now as formerly, as his forefathers did before him . . . 25

Theogenes’ particular attachment to Athens is asserted (Icæ IªÆŁ K Ø æd e B  e ŁÅÆø) and its chronological continuation is expressed by the present tenses (K Ø,  E), as is its personal and hereditary basis (ŒÆd ÆP e ŒÆd ƒ æª  Ø ÆP F). Though the reference to services performed is general ( E ‹ Ø Æ ÆØ IªÆŁe), the context for the performance of these services is specifically designated—his own community of Naukratis when Athenians arrive both on public and on private business. Some or all of these elements of the proxenos-paradigm occur in all but the most abbreviated of inscribed decrees. The precise terms and formulations used to describe proxenoi varied quite widely, as did the degree of detail which these proxenos descriptions went into (in only the most concise texts are these descriptions omitted entirely).26 Analysis of the different terms used, however, makes it clear that this variation in phraseology was not indicative of local divergent proxenos-paradigms, or distinct traditions determining how proxenia was conceived. The same characteristics and patterns of behaviour are continually evoked. Overlapping in meaning, different terms were used as equivalents, neither more nor less suitable than the alternatives for invoking a shared proxenos-paradigm.27 The choice of which of the appropriate terms to inscribe on stone, how much detail to go into, or whether to dispense with a motivation clause at all, was in most cases largely determined by local epigraphic traditions dictating authoritative verbal formulations which the individual drafter of the IG II3 294 ll. 7–13 (= II2 206, 349/8 bc). The account that follows does not attempt to document all the variants in formula attested. Instead it is intended to identify the key words, constructions, and their interrelations. For omission of proxenos description, see the abbreviated decrees of Gonnoi in general and many of the decrees of Delphi (e.g. FD III 1 18, c.254 bc). 27 For this idea, see Dover’s comments on synonymy in a similar context, Dover (1974), 62–4 and esp. 71: ‘The speaker may look at the same thing in different ways and in a great many contexts it does not actually matter which he chooses.’ 25 26

34

Proxeny and Polis

final inscription usually followed closely, but could, on occasion, vary. Close reading of the different terms attested, which could be used as alternatives but which were not semantically co-extensive, and exploration of the ways in which they were used in combination or to the exclusion of other terms, tells us a great deal about how the qualities which they invoked interrelated. The role of the proxenos emerges as something coherent and organic, in which the different characteristics I delineate for the purposes of analysis are not in fact easily separable. The personal disposition of the proxenos with respect to the polis is the most regularly described of all the characteristics. It is, in effect, the basic element of the epeidē clause, present even in decrees in which other characteristics were omitted for brevity. A fairly narrow range of adjectives and adjective combinations is attested introducing the description of the honorand in by far the most common form— either with the present copula (eimi), or a present participle dependent on the verb diateleō (the present participle hyparchō is sometimes also found, as are perfect forms of gignomai, with the same continuous present sense).28 In rough order of frequency, the principle adjectives used in this position are: agathos (often in the conjunction kalos kagathos);29 eunous; prothumos; chrēsimos; philos; euchrēstos; and, less commonly, philotimos. That this ‘well-disposedness’ (eunoia), or ‘regard’ (philia), ‘eagerness’ (prothumia), or even ‘usefulness’ (signalled by chrēsimos/euchrēstos), was directed specifically towards the polis in question is almost always explicitly marked with the placement of a reference to the granting city (either polis, demos, or a collective ethnic) immediately afterwards, either in the dative or in a prepositional phrase (usually following peri or pros).30 This clarifies what is meant by the otherwise seemingly vague description of the proxenos as ‘good’ (agathos). It was not the general demeanour or moral quality of the honorand which mattered for the grant of proxenia—it was his specific partiality for the polis in 28 In the case of gignomai, of course, a perfect tense is logically equivalent with eimi (e.g. I.Iasos 53, 315–285 bc). 29 kalos is not attested in this context in the absence of agathos (even as an intensifier in conjunction with a different adjective; cf. Dover 1974, 73), though agathos frequently occurs on its own. 30 e.g. Ic[æ Iª]ÆŁ [K Ø æd e B]  e ŁÅÆø[], IG II2 132 ll. 3–5 (Athens, 355/4 bc); Icæ IªÆŁ K Ø ŒÆd h[ ı] HØ øØ HØ Åø, IG XII 5 825 ll. 5–6 (Tenos, first half of the second century bc).

The Anatomy of an Ancient Institution

35

question which was important, his goodness to that city, which manifested itself in services which he performed for it and its citizens. A somewhat wider range of terms is used to invoke these qualities in other constructions, using adverbial forms or abstract nouns, almost always after an adjective description with the copula. All of the terms which occur in the adjectival clauses also occur in other constructions. So, for example, we see a proxenos described as being well disposed to a particular polis (Pø ØÆŒ )31 or as continually making demonstrations of his eunoia (KØŒ  c h ØÆ m åø ıªå Ø)32 or prothumia for a city (KØŒ  b c Æ F æe e B  æ ŁıÆ)33. But we also find different terms referencing similar abstract qualities spoudē and ekteneia, parallel with prothumia—the proxenos’ zeal.34 Similarly, we find the honorand’s attachment to the polis described using the abstract nouns hairesis and prohairesis (personal inclination, in the latter case preference verging on devotion)35 in place of eunoia.36 The affection or personal preference of the honorand for the honouring polis—in accordance with the alignment of xenia with proxenia, his friendship for it—is thus asserted as the basis of the permanent relationship which the polis constructed with proxenia.37 It was almost invariably the first element of the epeidē clause (at least after the honorand was named with his ethnic), and often an adjectival

31 e.g. SEG 32 613 l. 15 (Magnesia, 120–115 bc). This particular formulation, though not the most frequently attested in our corpus of proxeny decrees, is reasonably widespread from the second century bc, occurring in decrees from central Greece (including at Delphi with the adverb variant P œŒH), the Aegean, and Asia Minor. 32 IG XII 9 218 (Eretria, c.270 bc). 33 KØŒ  b c Æ F æe e B  æ ŁıÆ, IG XII 6 31 ll. 11–12 (Samos, 306–301 bc). 34 L åØ K[Œ] ØÆ ŒÆd  ıa bæ A ºØ  ±H, FD III 1 153 l. 2 (Delphi, 146/5 bc). The use of ekteneia at Delphi and its adjectival (KŒ ) and adverbial forms (KŒ H) found more widely in central Greece and the Aegean, seems to be a phenomenon of the later Hellenistic period. 35 LSJ. 36 The fact that we never find eunoia and hairesis in conjunction makes it clear that these were thought of as equivalents (an individual displays his eunoia or his hairesis for the polis, never both; similarly he is only ever praised for one or the other) and this is reinforced in other contexts where they are used in parallel (e.g. Milet I 3 139C ll. 51–2, 262–260 bc). 37 For xenia as both the etymological and institutional basis of pro-xenia—not least because it was a relationship of friendship involving xenoi—see Herman (1987), 132–8.

36

Proxeny and Polis

clause, asserting this amity, was all that there was. First and foremost, then, the proxenos was felt to be the formally recognized, non-citizen friend (i.e. xenos) of the polis. The natural and expected result of this emotional basis, however, was that services would be performed by the honorand for the community in question—and thus the performance of services was, and was treated in this explanatory clause as being, evidence of these feelings of friendship. This is what the decrees mean when they talk of the honorand as showing, making clear, or performing demonstrations (apodeixeis) of his eunoia to the city—that he made his affection for the city manifest in the services which he repeatedly performed on its behalf. The emphasis in these documents on repetition, on continuation, makes it clear that this is the way in which the proxenos was expected to behave in the future. This could be expressed with the present tense, or more markedly with the verb diateleō in conjunction with present participles, with temporal phrases (such as K HØ æ æ  åæøØ, ‘previously’, and F, ‘now’), or by stressing that the honorand always performed services for citizens in need, at every opportunity,38 or indeed in the reference to this disposition being inherited.39 The assertion that the honorand had manifested his eunoia continuously was vital in justifying the grant of proxenia, which, like citizenship, constituted a permanent relationship with the polis. It was expected that the honorand would continue to perform services in the future because he was supposed (and in at least one case actively exhorted) to guard or maintain the same feelings of regard for the city from which these actions would naturally spring.40 Utility was thus central to the proxenos-paradigm,41 and this aspect was stressed when, as

38 e.g. åæÆ Ææå Ø Id HØ  øØ H  ºØ A K Æ d ŒÆØæHØ, IG VII 10 ll. 8–10 (Megara, late fourth to early third century bc)—both elements also frequently occur on their own. 39 See below, Chapter 2, pp. 132–3. 40 []åŁÆØ AØ ºØ . . . ÆæƌƺE b ÆP e ŒÆd K e [º] Øe ØÆçı[º ]Ø a ÆP a æ Ææ Ø, FD III 2 89 ll. 8–10 (Delphi, early second century bc). The case, like this exhortation, is unusual—the honorand, an Athenian called Apollodoros, had been brought to Delphi to help with an arbitration that did not happen—and this is the only interaction between this individual and Delphi which the inscription records. Perhaps it was used here as Apollodoros had no proven form in maintaining this crucial attitude of amity towards the polis (thus replacing references to previous, frequently expressed affection for the city—exhorting him to behave thus rather than noting that he already tended to). 41 See, on this, Gauthier (1985), 140–5.

The Anatomy of an Ancient Institution

37

frequently, recipients of proxenia were described as chrēsimos or euchrēstos,42 or, indeed, more commonly, as prothumos—which was roughly equivalent, meaning ‘eager to perform services for the polis’, as we can see in the ways in which it was sometimes expanded with the infinitive43 and used in relation to the other terms.44 Services performed by proxenoi, however, did not just demonstrate the personal emotional commitment of the honorand to the city. These services were also instrumental themselves in establishing or cementing a relationship between the two actors based on reciprocity. The services of the honorand were gifts which poleis were obliged to respond to, if not in kind—the gifts of the polis were above all honorific in character—in due proportion or measure.45 At its heart the relationship implied by proxenia was thus euergetical in character. It involved individual benefactors (euergetai) performing benefactions (euergesiai), which were rewarded, and encouraged, with honours. The benefactions particularly associated with proxenia differ from those associated with citizen or royal euergetism, on which the bulk of modern scholarship has hitherto focussed, in that they were primarily non-financial.46 Gift and counter-gift nonetheless created and expressed ties of obligation which each party was obliged to take into account in future interactions with each other. Recent scholarship has emphasized the way in which each party sought to put the other in its debt, obliging them to reciprocate in turn with greater benefactions.47 However, it is important not to focus too closely on

42 Interestingly, we never find chrēstos in any description. The explanation of Whitehead (1993), 63–4, in relation to the Athenian evidence, suggesting that the ‘class-based connotations . . . proved too large an impediment in the way of its complete democratization’ hardly explains its absence elsewhere, and overlooks the definitely elite connotations of kalos kagathos and philotimia which certainly do occur, so Whitehead (1983). See Dover (1974), 296–7, on the differences between the ways in which chrēsimos and chrēstos are used in our sources. 43 For prothumos with the infinitive of poieō, see, e.g., IG XII 5 111 l. 5 (Paros, late fourth to early third century bc) and I3 110 ll. 9–10 (Athens, 408/7 bc). 44 No honorand is described as both chrēsimos and prothumos in the initial adjectival clause in which most other combinations of the adjectives used are found—which shows that these were felt to be equivalent in function and is the more marked because these decrees do not shy away from redundancy. 45 Poleis are frequently concerned to assert that they grant ‘worthy honours’ (e.g. I. Priene 6 l.24–27, Priene, 330–300 bc; IGUR I 2 ll. 20–1, Akragas, c.100 bc); see Mitchell (1997a), 18–21. 46 See e.g. Bringmann (1993) and Gauthier (1985), 140–5. 47 Mitchell (2009), 21–2.

38

Proxeny and Polis

individual honorific transactions, extrapolating from isolated rallies of honours and benefactions between particular cities and kings and so reduce the reciprocity underlying euergetism to a balance sheet of payments. Benefaction and reciprocation expressed a relationship which, unlike that of creditor and debtor, was not cancelled when a response was made, but deepened by the cumulative history of mutual interactions.

1.1.3. The proxenos as euergetēs The polis’ understanding of proxenia as a euergetical relationship is explicitly signalled by the fact that individuals were almost always formally named euergetai (benefactors) at the same time as they were pronounced proxenoi. This connection with another well-defined polis institution provides an important insight into how proxenia was conceived. However, cases where the title euergetēs was used in isolation, in particular contexts in which proxenia was apparently not appropriate, highlight the particular content of proxenia, the specificity of the proxenos-paradigm. The importance of euergetism for understanding proxenia is emphasized by the fact that it was enough, in some of the more concise decrees, instead of giving a full motivation clause, to simply state that an honorand was granted proxenia for ‘being a euergetēs’ or ‘for a euergesia’ (Pæª ÅØ Z Ø; ŒÆ ’ Pæª Æ).48 On other occasions, individuals who are explicitly named only as proxenoi appear to have been thought of as euergetai as well, when it is decreed that they should receive all of the honours given to the other euergetai.49 The granting of euergetēs as a title, and especially the inscription of it on stone, had the explicit force of asserting the permanence of these different roles—the individual honorand as perpetual benefactor and polis as continually grateful recipient of benefaction.50 It communicated the expectation of an on-going 48 e.g. I.Gonnoi 3 (early third century bc); Milet I 3 104 (first half of the second century bc). cf. IG II3 324 ll. 38–9 (= II2 373 ll. 23–4, 322/1 bc). 49 Thus we sometimes find individuals who are named simply proxenos but granted ‘all else given to euergetai’ (e.g. I.Eph 1459, before 320 bc; IG IX 1 222, Tithronion, Phocis, second century bc). 50 In our literary sources, the act of inscribing someone as an eternal benefactor, clearly expressing this symbolic force, is well attested—Hdt. 8.85; Pl. Grg. 505c

The Anatomy of an Ancient Institution

39

relationship requiring each party to act in accordance with their designated roles—the proxenos to provide benefactions, the polis to express gratitude.51 The use of the title euergetēs to honour proxenoi and encourage continued benefaction in this way sharply differentiated proxenoi from citizens honoured in civic decrees. In contrast with the universal use of the title in relation to proxenoi, in almost all cities in the Classical and Hellenistic periods euergetēs was considered an unsuitable honorific for citizens (excluding outsiders who were granted citizenship at the same time as they were named euergetai).52 The issue here is that, while it was useful to use this title, drawn from interpersonal relations, to establish permanent relationships with outsiders, the permanent obligation on the part of the polis to the individual, which was implied by ‘euergetēs’, did not seem compatible with a citizen’s membership of the community.53 Likewise, while decrees inscribed in honour of citizens do often draw on the same language of praise that we see used in proxeny decrees, generalizing descriptions of the citizen honorand, the norm for proxenoi, tend to be avoided. Most decrees focus instead on describing the specific benefactions performed by the honorand (often his performance of a magistracy).54 Where, in the most elaborate decrees honouring the greatest citizen benefactors—such as Boulagoras of Samos and (ªØ  Pæª Å Ææ’ K d Iƪªæ łfiÅ). In Thuc. 1.129, although the title euergetēs is not explicitly given, the reference to euergesia inscribed for all time is the same (Œ  ÆØ

Ø Pæª Æ K fiH  æfiH YŒø fi K ÆN I ªæÆ ). On these see Gauthier (1985), 20–1. 51 This did not necessarily mean that new honours had to be provided in response to every service performed as, once granted, pre-existing privileges continued to assert the gratitude of the polis, especially when the proxenos made use of them, see Chapter 2, Section 2.3.1. 52 Cf. Gauthier (1985), 10–16, on the almost universal restriction of this title to citizens. A Delian decree of the end of the fourth century bc, ID 77, in which an individual designated as a Delian was made a benefactor of the Delians, looks superficially like an exception (and it has not been noted). In fact, however, it seems probable that the individual in this case was not a Delian by birth. The fact that this ethnic, ‘Delios’, was mentioned in an internal decree, where citizens never usually receive ethnics, suggests that it indicated an honorific citizen status. The individual in question may have been a resident alien (which was perhaps why proxenia, as the usual complement of euergesia, was considered unsuitable here), who was being granted further honours, including ateleia and proedria. 53 Similarly, Engen (2010), 148. 54 e.g. IG II2 487 (304/3 bc, for an official in charge of inscribing the nomoi of Athens) or Agora 15 71 (283/2 bc, an Athenian board of prytaneis).

40

Proxeny and Polis

Protogenes of Olbia—these generalizing descriptions are used, it is striking that rather than using the continuous present of the proxeny decrees, these tend to be firmly placed in the past.55 Like the avoidance of the title, euergetēs, this communicates a desire to avoid constructing a new permanent relation of obligation, elevating one citizen to the position of perpetual benefactor to (and over) the whole community. Although all proxenoi seem to have been at least thought of as euergetai (and most were formally designated as such), not all euergetai were proxenoi. Before exploring the particular reasons for these exceptions—and what they tell us about what was not thought compatible with proxenia—it is important to establish the relation, and difference, between these two titles. Early scholarship proposed that we should understand these two titles as belonging to a rigid hierarchy of honours.56 Wilhelm, conversely, argued that both proxenia and euergesia entitled those who received them to particular privileges but, whereas euergesia represented the recognition of the existence of a service, proxenia also involved the recipient in certain future duties and tasks.57 On the basis of this, Gauthier argued that the significance of the title euergetēs was polisinternal, in contrast with proxenos, which had a primarily external implication, suggesting interactions with citizens of the granting polis when they were abroad.58 This seems likely to be correct but requires some modification. As the power to grant these titles lay with the polis rather than the honorand, it is important to

55 Thus a wide range of decrees in Athens (II2 657, 287/6 bc, for the poet Philippides), Thessaly (IG IX 2 1103, 130–126 bc), Paros (SEG 32 825, after 188 bc), and further afield (including the decrees for Menippos of Colophon, SEG 39 1243, 120/119 bc, and Protogenes of Olbia, IosPE I² 32, 200–150 bc) regularly use the perfect tense of diateleō (ØÆ  ºŒ), which is rare in decrees granting proxenia (e.g. FD III 1 480, Delphic Amphictyons, 50–1 bc; SEG 3 468, Thaumakoi, Thessaly, first century bc; IG XII 7 388, Aigiale, Amorgos, 200–150 bc). The use of present continuous description in the Boulagoras decree (XII 6 11 ll. 49–52, after 243/2 bc), however, shows that this aversion was not as strong or marked as the use of the title euergetēs. 56 Keil (1899), 184–5, proposed reading this as a rigid tripartite system—praise, euergetēs, and proxenos—with the earlier prerequisite for the latter; Francotte (1910), 194–6, likened euergesia and proxenia to two heraldic orders, higher and lower. 57 58 Wilhelm (1942), 36–7; cf. Monceaux (1886), 97. Gauthier (1985), 23.

The Anatomy of an Ancient Institution

41

stress that the decision to name an honorand simply euergetēs, rather than proxenos and euergetēs, reflects its assessment of him and the inappropriateness in particular instances of the specific implications which were conveyed by proxenos as a more specialized subcategory of euergetēs. While both titles anticipated that future euergesiai would be performed by the recipient, the title proxenos in general involved the expectation that these were likely to take certain forms, above all aid to citizens of the granting polis in the context of the polis of the recipient. The well-known and consistent class of exception—of individuals honoured as euergetai but not proxenoi—were the Hellenistic kings.59 In this case the utility for poleis of the title euergetēs was clear. It allowed them to recast the relationship between ruler and ruled in terms which offered advantages to both parties—of perpetual benefactor versus grateful recipient. The king thereby secured consent for his rule, strengthened by the moral framework by which recipient was bound to benefactor. The polis, for its part, was thereby able to maintain an outward show of independence—subservience became gratitude—and was able, using this same framework, to exert moral pressure in return on the king to behave as a benefactor (as gratitude required something to be grateful for).60 The absence of the title proxenos in this context must relate to its specific content as an institution. Marek argued that the relationship of permanence and exclusivity implied by proxenia was inappropriate to the position of the king who did not serve the interests of any particular community but was instead a universal benefactor, and there may be something in this, although individual poleis were keen to refer in their decrees to the particular fondness which individual kings had for them.61 It also seems likely, however, that the sort of political services expected of proxenoi in the context of their own communities, while appropriate enough in relation to civic elites and, by extension, even members of royal courts, seemed ridiculous in connection with a king who was himself the source of political power. In comparison with the gifts a king could grant in his own right, including recognition of status (freedom, autonomy, or asylia), substantial resources, or funds and protection,62 the services expected of and associated with proxenoi 59 60 61

Marek (1984), 335–9, collects grants for lower ranking monarchs and dynasts. Veyne (1990), 102–3; Bringmann (1993), 17–23; Ma (2002), 202–6. 62 Marek (1984), 337. Bringmann (1993).

42

Proxeny and Polis

were decidedly mediocre—at most consisting of help securing these more substantial benefactions from civic or royal authorities. There are other instances also where it seems probable that the title of proxenos was specifically withheld rather than simply not used, where individuals were appointed euergetēs in an epigraphic context in which most individuals were named both proxenos and euergetēs. Surviving proxeny lists from Thera, dating from the late second to early first centuries bc, attest to the fact that euergetēs was regularly granted in its own right as an alternative to the conjunction proxenos and euergetēs, to individuals for whom the title proxenos was apparently considered inappropriate.63 In at least one instance it seems that we can identify geographical factors as playing a significant role behind this decision. In IG XII 3 1299/1300, alongside a series of three lists of proxeny grants, two grants of the title euergetēs are recorded. Unlike the honorands named proxenoi and euergetai, who come from the Aegean world, the individuals singled out as euergetai originate in the west, with one individual from Syracuse and two, if they are read correctly, from Calabria (˚ƺÆ[æ ]).64 It seems likely that the services which they performed for the Therans had little to do with their origins—and it may be that either they themselves had ceased to have much to do with their native communities, or, more probably, that these were places with which Therans had little to do and where, therefore, the specific, future euergesiai expected in the case of proxenia made little sense (a lack of interest in their native political communities may also be implied by the regional ethnic ‘Kalabroi’).65 We can find a likely parallel for these sorts of scruples concerning proxenia in an early Oropian decree which names an individual from the Carian city of Theangela as euergetēs (but not 63 IG XII 3 333/1298 l. 1, ll. 8–9 (two separate instances, Appendix no. 16.2.2); XII 3 1299/1300 A ll. 1–9, ll. 21–3 (two separate grants, Appendix no. 16.4.1). This is made clear by the fact that, whereas these lists record multiple awards of the joint title proxenos and euergetēs (which we would expect to be the more elevated honour) they only ever record a single grant of euergetēs on its own. 64 The only alternative ˚ƺÆ[Æ ØÆ], a minor community in Asia Minor (Zgusta 1984, no. 411), would be overlong; the names also support a western origin. 65 The roughly contemporary grant of proxenia to individuals from the city of Rome, which was in fact more distant, does not contradict this picture, as Theran interests in Rome at the time of these lists (c.100 bc) would be of a very different kind. The reason why Leontios of Knossos was only named euergetēs in IG XII 3 333/1298 is less clear, especially given the numerous Cretans named proxenoi in IG XII 3 1299/ 1300.

The Anatomy of an Ancient Institution

43

proxenos), at roughly the same time as two decrees granting, in contrast, both proxenia and euergesia to two Macedonians.66 In the few other cases that we have, however, it is often difficult to identify the reasons why the title proxenos appeared inappropriate, and, indeed, the rarity of the use of the title euergetēs on its own in honorific decrees suggests that it was something of a marginal phenomenon.67 In other contexts, under strain, cities seem to have little difficulty in making wider symbolic use of proxenia than the narrower proxenos-paradigm, which seems to have impeded its grant in these cases, strictly suggested.68 However, against this more permissive background, it is significant that, on occasion, these sorts of scruples did find expression and other honorific forms were sought and used as alternatives, reflecting the perceived specificity of the proxenosparadigm.

1.1.4. The specificity of the proxenos-paradigm This specificity was strongly expressed through the stereotypical descriptions of proxenoi. Although the component adjectives, nouns, and verbal phrases from which these descriptions were constructed were not restricted to decrees granting proxenia, belonging instead to a wider honorific vocabulary used in decrees for other citizen and non-citizen honorands, the way in which they were used 66 I.Oropos 3 with Knoepfler (2001a), 367–89, dating these to the brief period of Oropian independence, 338–335 bc. To these examples should be added Skirtias, whose name occurs in a curious list of proxenoi at Kydonia (IC II x 1, c.250–200 bc), to whom the title euergetēs is given instead of an ethnic. On Skirtias, see n. 104. IG XII 8 155 II, a Samothracian decree naming an individual from Zone euergetēs below a decree naming another individual proxenos and euergetēs is probably also an example. 67 See also IG II3 352 (= II2 351 + 624), 330/29 bc—it is unclear whether the honorand was not named proxenos because the nature of his benefaction, which was primarily financial and took place within the context of Athens, did not make him the appropriate recipient of this honour, or because, as a Plataian, he already had citizen rights. Whitehead (1977), 29–30, thought him likely to be a metic because of his participation in this epidosis. Other examples may include IG II2 81 (c.387/6 bc) and 845 (end of the third century bc; both heavily restored), as well as XII 4 75 l. 49 (an individual given this title in an epidosis list from Kos, 202/1 bc). IG I³ 1454 (probably dating to the period before 434/3 bc) is a related case (a particular individual, his sons, but also the koinon of the Eteokarpathians, are inscribed as the euergetai of the Athenians). 68 See n. 9.

44

Proxeny and Polis

together in proxeny decrees was in fact highly specific to grants of proxenia. When we examine substantial series of honorific decrees from particular poleis, a very strong link emerges between the description of the particular characteristics which I have delineated—the proxenos-paradigm—and the grant of proxenia. In particular, in the more than sixty relevant honorific decrees from Samos, in only two cases is it clear that the typical Samian proxenos description was used in relation to honorands who were not then named proxenoi (both were in fact granted citizenship).69 In the more than seventy honorific decrees known from Iasos there is only one instance in which the forms of proxenos description used there were applied to an individual who was not granted proxenia and he was probably already an Iasian proxenos.70 At Oropos and Delos this holds over even more substantial sets of data. In only one of the two hundred known honorific decrees of Oropos is an honorand described in proxenos-terms without the grant of proxenia.71 Of the hundreds of honorific decrees from Delos, only five use the full formulaic proxenos description without a concomitant grant of proxenia—and in some of these cases it is possible that the reason why proxenia was not granted was because it was already held.72 In all of these instances, it is difficult to explain the exceptions, which probably relate to particular circumstances, but the rule, the strong correlation between these particular sorts of proxenos-description and the grant of proxenia, is clear. This correlation is further strengthened 69 IG XII 6 24 (after 306 bc) and 43 (after 322 bc); these stand against some 28 decrees where both the grant of proxenia and the proxenos-paradigm are securely attested (from no Samian proxeny decree is this paradigm known to have been absent). The Samian decree for Gyges of Torone (IG XII 6 46, after 322 bc) provides a useful contrasting case. Instead of the generalizing description found in proxeny grants, the specific service performed by Gyges is described (the import of 30,000 medimnoi of corn) and Gyges receives politeia instead of proxenia. 70 At Iasos it is striking that the only foreign individual honoured whose ethnic is certainly omitted is one of only two granted politeia and not proxenia (I.Iasos 47, 315–285 bc). By contrast, while the priest of Labraunda, honoured in Maddoli (2007) no. 20B (230–190 bc), is described in proxenic terms, this is presumably because he already possessed proxenia, so Fröhlich BE (2009), n. 451. 71 I.Oropos 307 (154–150 bc)—the honorand in question may already have been a proxenos. 72 IG XI 4 542 (Damaratos, king of Sparta, 300–181 bc); 559 (Philokles, king of Sidon, c.280 bc); 646 (c.240 bc); 705 (end of the third century bc); 784 (beginning of the third century bc). In IG XI 4 600 (c.250–200 bc) is another example where the reason for the absence is made clear by the publication clause, which references an earlier decree of proxenia.

The Anatomy of an Ancient Institution

45

by considering decrees which did not grant proxenia. Where these texts contain motivation clauses, they tend to avoid generalizing description of the honorand and instead set out in concrete detail the specific benefactions he performed—something proxeny decrees do comparatively rarely and almost never to the exclusion of generalizing description. On the surface, material from Athens and Ephesos appears to complicate this picture. Although at Athens there is a very strong correlation between the use of these specific, formulaic descriptions and the granting of proxenia, elements of these descriptions are sometimes also found in relation to individuals granted Athenian citizenship instead.73 In the material record from Ephesos, moreover, exceptions to this correlation appear to be the rule, with these forms of description instead associated with grants of citizenship: of the sixty honorific decrees which use these sorts of proxenos-descriptions, only five explicitly grant proxenia, the rest are citizenship grants.74 In both of these cases, however, I would argue that proxenia remained the model underlying these grants. At Athens, politeia was granted to particular foreign benefactors as a more marked honour to outsiders who might otherwise receive proxenia. Whereas, at Samos and Iasos, proxenia and politeia were conceived of as complementary means of establishing a permanent relationship and apparently almost always granted together, at Athens there seems to have been a stricter sense of hierarchy. There politeia, granted much more sparingly,75 expressed a stronger relationship between polis and individual, which seems to have been felt to obviate any need for a grant of proxenia because it superseded it (proxenia was never granted at the same time as politeia or to individuals who already possessed it, but politeia was sometimes given to proxenoi).76 e.g. IG II2 495 ll. 10–16 (303/2 bc); 654 ll. 21–5 (289/8 bc). On the Ephesian material, see Walser (2008), esp. 321 n. 1 on the difficulty of identifying why particular recipients received proxenia as well as politeia. 75 Although, after the end of the fourth century, decrees of citizenship are in fact more abundant than those of proxenia in the Athenian epigraphic record, this reflects a tendency for inscription to be granted less frequently, as a more marked honour (thus granted more frequently alongside politeia, the higher honour). See Chapter 5, Section 5.1 with figs. 5.5 and 5.6. 76 There is only one Athenian decree in which both proxenia and politeia were granted, IG II2 19 (394/3 bc), in which an original proposal of proxenia was upgraded in a rider to include politeia. Though Mitchell (1997a), 40, is correct, in reference to this decree, that proxenia was not ‘incompatible with citizenship’, contra Osborne 73 74

46

Proxeny and Polis

In the decrees of Ephesos it is striking that these descriptions are similarly used predominantly in decrees granting politeia as an honour to individuals who were apparently not expected to settle at Ephesos.77 Even though these grants of citizenship were always formally ‘activated’, with tribe and chiliastys automatically assigned, it is clear from the descriptions that these are not grants of citizenship to immigrants but to outsiders expected to continue to be active, and useful, in their own communities—like the two Milesians who are described, in the continuous present, as aiding citizens who come to Miletos78 or the Akarnanian at the court of King Prepelaos specifically praised for aiding a recent embassy there in addition to his general services to citizens.79 In particular, it is striking that the only citizenship decrees for individuals who certainly were expected to settle permanently at Ephesos are also the only decrees that do not describe the honorands in these proxenos terms but, like the motivating clauses of non-proxenic decrees from elsewhere, describe only particular actions of the honorand.80 Was the use of these euergetical terms inappropriate, or at least less desirable, in relation to individuals who were actually expected to become citizens? In this case it is possible that, within this east Greek context, in which proxenia and politeia were often routinely granted together,81 an explicit grant of proxenia came to be seen as redundant. Indeed it may even have been felt in some sense to be implied by the honorific grant of politeia, in much the same way as the title euergetēs often was.82

(1981–3), D 7, given the lack of other Athenian decrees granting both, politeia probably was conceived in this case as superseding rather than augmenting. 77 This is presumably the reason that Marek included these decrees as an addendum in his catalogue of Ephesian proxenoi, Marek (1984), 105. 78 I.Eph 1411 (c.283/2 bc). 79 I.Eph 1449 (302 or, less probably, 295–281 bc). 80 I.Eph 1415 (c.300 bc) for a young metic boxer who proclaimed Ephesos his hometown on winning at the Nemea (N

ºc ŒÆd ŒÆ Ø[ŒH] K ’¯ç øØ) and I.Eph 1420 (before c.320 bc) for two Athenian potters_ _who promised to complete a dedication for the god whose citizenship is explicitly made contingent on their residence at Ephesos and completion of their task. 81 Including, earlier, at Ephesos, I.Eph 1389 (before 322/1 bc). 82 At Ephesos, despite the fact that the title euergetēs is not explicitly granted by these decrees, the honorands are often granted politeia as given to other euergetai (I.Eph 1442, c.305 bc). That proxenia was also implied is perhaps indicated by the use of an analogous phrase ‘what is provided for the other proxenoi’ in a decree granting citizenship but not proxenia (I.Eph 1422, 306–295 bc—the citizenship grant is restored, but securely, on the basis of the closing formula).

The Anatomy of an Ancient Institution

47

Two series of Athenian decrees for particular individuals highlight the ways in which these inscribed descriptions worked—and how they were specifically related to grants of proxenia. The first is a monument for Herakleides of Salamis, comprising five separate decrees in his honour. The four earlier decrees, for all that they grant substantial honours including gold crowns, do not grant proxenia. Instead, like other honorific decrees from elsewhere which do not invoke this permanent honorific relation, they simply describe the specific benefactions of Herakleides which prompted the honours granted in each case—specifically offering grain at a cheaper price despite being the first merchant to bring his ship in (and thus otherwise in the position to extort a high price). It is only in the last of these decrees, which does grant proxenia, that Herakleides is described in more general, proxenos-appropriate terms, as habitually performing services for the Athenians, as ‘living his life eager for honour from the demos of the Athenians and doing them whatever good he can’, with the earlier, specific honorific decrees inscribed beneath to substantiate this assertion.83 This sort of description was felt appropriate for Herakleides only when he was being considered for a grant of proxenia—and this strongly emphasizes the way in which these stereotypical descriptions were conceived, as expressing a paradigm of specifically proxenos behaviour. The second set of honorific decrees from Athens, three decrees for Euenor the Akarnanian doctor, reveals how these proxenos-descriptions could be deployed in relation to a probably atypical proxenos, deliberately downplaying his divergence from the proxenos-paradigm. In the first of these, in which he is named proxenos, Euenor is purposefully described in the stereotypical terms of the proxenosparadigm, which deliberately assimilate him to other proxenoi— ‘since Euenor the Akarnanian is prothumos in relation to the demos of the Athenians and does whatever good he can’ (K[Ø]c ¯P[øæ] ŒÆæa æŁı K Ø[] æd [ ]e B  e ŁÅÆø ŒÆd  E ‹ [Ø] Æ ÆØ IªÆŁ).84 In the next decree in which he is honoured, inscribed under the first, Euenor’s status as proxenos is mentioned and, although there is a reference, unusual for a proxenos, to his taking care of tasks imposed by the demos, there is again no overt 83 ØÆ ºE çغ Ø   æe e B  e ŁÅÆø ŒÆd  ØH ‹ Ø Æ ÆØ IªÆŁ, IG II3 367 ll. 6–8 (= II2 360, RO 95; 325/4 bc). 84 II3 324 ll. 19–21 (= II2 373 ll. 4–6, 337/6 bc).

48

Proxeny and Polis

mention of his profession as a doctor.85 It is only in the last of these decrees (inscribed on a separate stone), in which Euenor is named an Athenian citizen and his proxeny is unrecorded, that he ceases to be described exclusively in these generic, proxenos-appropriate terms and his medical vocation and activities are explicitly described—‘since Euenor, the doctor, makes himself useful to those in need according to his profession’ ([KØc ¯Pøæ › N]Æ æe . . . [åæ Ø  Æı e Æ]æ åÅŒ ŒÆ a c å[Å E   Ø]).86 The first two decrees in this sequence thus appear to represent an attempt to use an honorific status in relation to an honorand, Euenor, whom it did not perfectly fit—presumably because the services for which he was honoured as proxenos differed markedly from those expected of proxenoi, both in their nature (medical) and in the place (i.e. within Athens) where they were performed. The grant of politeia, by contrast, emerges as expressing a less specific sort of relationship—one for which Euenor’s activities at Athens could be more appropriately cited.

1.2. THE SERVICES ASSOCIATED WITH PROXENIA The way in which these stereotypical descriptions were deployed in our corpus of honorific decrees thus strongly supports the suggestion that they were felt to evoke a coherent and specific proxenos-paradigm, emphasizing in particular the emotional basis and content of this relationship. At the same time utility, from the perspective of the polis, the usefulness of the proxenos, was also conceived of as being central.87 Although the descriptions of this usefulness in our decrees appear very general, in fact, when considered in the context of other assumptions concerning the proxenos expressed in our decrees, above all his communal context, it is possible to identify a specific range of types of service particularly associated with proxenia. These collectively amounted to an intermediary role, allowing individuals from one city access to the institutions and networks of their proxenos’ [K]Øc b ¯Pøæ ¯PÅ ı Ø’ Pæª Æ æ [ ] Kª F  ı F ŁÅÆø ŒÆd –Æ Æ ‹ [Æ] æ  Æ  ÆP HØ › B  › ŁÅÆø ŒÆd NÆØ ŒÆd Œ ØE Kغ ÆØ·, IG II3 324 ll. 38–41 (= II2 373 ll. 23–6, 337/6 bc). 86 87 IG II2 374, ll. 4, 6–7 (319/8 bc). Gauthier (1985), 140–5. 85

The Anatomy of an Ancient Institution

49

polis. This section therefore begins by examining the generic descriptions of the ways in which proxenoi were useful, before going on to argue that the ways in which proxenoi were described in decrees conveyed the basic assumption that the proxenos was and would continue to be active in his own polis.

1.2.1. Generic descriptions of the services performed by proxenoi The services performed by proxenoi were usually described in a number of generic ways—the most common type resembling closely the following formula frequently used at Tenos: ØÆ ºE åæÆ Ææå  ŒÆd Œ ØE BØ ºØ ŒÆd NÆØ Åø E K ıå  ı Ø ÆP HØ He lives his life performing services to those of the Tenians who come to meet him whether on public or private business.88

Although the precise wording did vary (for example with a reflexive construction in paraskeuazō or parechō, with an adjective—‘he makes himself useful’—used instead of diateleō with a participle), this conjunction of elements is absolutely characteristic. Moreover, the second part of this phrase—that these services were performed for Tenian citizens ‘who met with him’ regardless of whether they were acting on their own behalf or for the polis of Tenos—is crucial to understanding proxenia. The proxenos is one who, because of his regard for a polis, aids those of its citizens whom he encounters because of their membership of this community. This joint emphasis on interactions between proxenos and polis on both public and private levels is also one of the most potentially interesting issues in the study of proxenia. It hints at the possibility that these official, public relationships may provide us with information about otherwise unattested private relationships and interactions.89 The nature and context of these interactions, public and private, are questions which have been vigorously debated. In particular, it has been doubted, especially in individual cases, how far the interactions in question are likely to have been economic and the extent to 88 89

IG XII 5 798 (300–250 bc); cf. 816 and 819 (both c.300 bc). I explore this issue more fully in Chapter 3.

50

Proxeny and Polis

which the polis indicated by the ethnic of the proxenos can be assumed to represent the context in which these took place. One difficulty here is the generic nature of these descriptions of services rendered. Where the specific services of particular proxenoi are mentioned by decrees, not only is it likely that this is because they are exceptional in kind or degree, they also seem to be ‘public’ in character, performed for the community as a whole (which should not surprise given that these inscriptions were concise public monuments). These formulaic descriptions are intentionally broad. References, which are ubiquitous in our decrees, to the ‘providing services’(åæÆ Ææå ),90 ‘doing whatever good he can’ ( ØH ‹ Ø Æ ÆØ IªÆŁe),91 of ‘forever being the cause of benefits’ (I Ø  I[ªÆŁ F] Ææ[Æ Ø]  ª ),92 and descriptions of the _ _useful to those of the citizens who come to him way he ‘makes himself by working in their interests’ ( E ÆæƪØ  Ø æe ÆP e H  ºØ H åæ Ø  Æ e Ææå ÆØ, æ

ø a ıçæ  Æ)93 are _ range of benefaction on designed to encourage and recognize a broad the part of the proxenos. Even recurrent references to the proxenos ‘saying and doing things in the interest of (the people)’ (ºªø ŒÆd æ ø a ıçæ  Æ)94—apart from emphasizing the importance of speech as a form of service (especially in public contexts)—do little to communicate any expectations which the polis might have of the specific sorts of services which their citizens could expect to receive from their proxenos. This does not, however, mean that there were no such expectations. In fact, their existence is clearly implied by frequent references which occur throughout our epigraphic evidence to appeals made by distressed citizens to their proxenoi, particularly with the verb parakaleō. Thus, those named proxenoi at Delos are frequently described as ‘[doing] whatever good they can for the demos of the Delians collectively and [providing] services for those of the Delians who meet them privately in whatever matters they call on them for aid’ (ŒÆŁ’ ‹ Ø i ÆP e ÆæƌƺH Ø).95 Elsewhere they are described as making themselves useful in relation to whatever individual citizens called on them

90 91 92 93 95

e.g. IG XII 5 798 (Tenos, 300–250 bc). e.g. IG XI 4 639 (Delos, 250–200 bc). e.g. IG VII 223 (Aigosthena, c.192–146 bc). 94 e.g. IG XII 6 23 (Samos, after 306 bc). e.g. I.Oropos 22 (c.250 bc). e.g. IG XI 4 517 (300–250 bc); cf. I.Oropos 147 (Oropos, c.250–230 bc).

The Anatomy of an Ancient Institution

51

(Kd a ÆæƌƺÆ).96 Similar interactions between the honorands in proxeny decrees and the citizens of the poleis honouring them are described using the verb deō—the individual always aiding individuals who ‘make requests’ or are ‘in need of something’ ( E   Ø; Ø NÆØ H  ºØ H  ı åÅØ).97 These requests for aid, though they are set out in similarly non-specific terms, imply more specific notions about what a proxenos could be expected to do, and thus an association between proxenoi and the sorts of service they provided most regularly. Later in this chapter I will explore the evidence we have for the sorts of service particularly associated with the proxenos. However, this depends on the extent to which the proxenos was expected to perform these services in the context of his own community.

1.2.2. The communal contexts of proxenoi In early scholarship, it was generally assumed that the ethnic recorded in the proxeny decree was significant for the grant—in particular that trade links between poleis could be read from these decrees.98 In response to this, it has rightly been pointed out that the primary context for the contact implied by a proxeny grant may often not have been the proxenos’ polis of origin at all.99 There are cases where an honorand receiving proxenia was demonstrably resident in the granting polis at the time of his nomination.100 More frequently decrees themselves make it clear that particular honorands were appointed for services performed (and to perform future services) in the context of a position at a royal court or as commander of a body of soldiers, rather than in their native poleis.101 Individuals were also honoured e.g. FD III 2 91 (Delphi, 167 bc); IG IX 1 103 (Elateia, Phokis, c.150 bc). E   Ø: e.g. IG VII 9 (Megara, late fourth to early third century bc); I. Oropos 51 (Oropos, 240–225 bc); IG XII 8 640 (Peparethos, after 197? bc). Ø NÆØ H  ºØ H  ı åÅØ: e.g. IG XII 6 30 (Samos, 306–301 bc). For citizens ‘in need’, see also the following formulae: E åæÆ å ı Ø, SEG 30 533 (Magnesian koinon, second century bc); ıæª[E  a]  Å æ ŁıÆ K x [ i ıªå ]ø Ø åæÆ å [ ]; IG XI 4 776 (Delos, early second century bc). 98 e.g. Wilhelm (1942), 51; Rostovzeff (1941), 245; Durrbach (1921–2), 57–8; Shear (1978), 30–1 (Delos). 99 So Reger (1994), 67–9; Archibald (2001), 261–3. 100 See below, pp. 56–7. 101 Robert (1963), 66–7: ‘Il faut penser que ces proxènes peurent être des gens vivant loin de leur patrie, et d’abord comme officiers d’une monarchie, et que des 96 97

52

Proxeny and Polis

for professional skills exhibited in the context of the granting polis— as doctors, perhaps, or poets.102 This sort of mobility—which is thought to be more characteristic of the Hellenistic than Classical Greeks—has been woven into narratives about the way in which this institution changed between periods. In particular, it has been suggested that the hereditary proxenoi of the Classical epoch, during which a family of proxenoi might perform services for their granting polis over a number of generations within the same city, became less common in the Hellenistic when continuity of residence could not be assumed.103 Examples to both support and contradict this narrative can be identified in our rich record for proxenia. However, given that, as an institution, proxenia was capable of encompassing atypical cases, what matters is whether there were more specific, expected norms. The evidence overwhelmingly suggests that there were and that proxenia was and continued to be conceived of as a primarily inter-polis institution. The expectation, communicated by the formulaic descriptions of honorands in proxeny inscriptions, seems to have been that the primary context of the proxenos, the place in which he would most likely perform the services referred to and interact with citizens of the granting polis, was the polis designated by his ethnic. That an individual was likely to retain strong ties with his polis of origin—the ethnic of which he used to identify himself—would not be a surprising assumption for a polis to make, especially in relation to the sort of high-profile, influential individual likely to receive proxenia. Although we have to allow for the possibility that the affiliation indicated by an ethnic might not be significant in a particular case, we should certainly not assume that it was not—and nor did poleis. Above all, this expectation, not of immobility, but of primary residence or even simply continued activity by the individual in his native community, is signalled by the importance of the proxenos’ ethnic in the epigraphic monument recording his grant, which is almost invariably featured prominently in the text of the grant

villes honorent non point à cause de leur origine ethnique, mais à cause de leur situation’. 102 IG IX 2 69, a decree for the horse doctor, Metrodoros, is a particularly good example (Lamia, after 146 bc). For collections of examples, see Marek (1984), 359–79. For the material relating to doctors, see now also Samama (2003). 103 Marek (1984), 387–8.

The Anatomy of an Ancient Institution

53

itself.104 The proxenos’ ethnic was often also indicated in a heading inscribed over it or by a relief depicting a symbolic representation of the polis in question.105 Inscriptions, which make reference to the fact that decrees are initiated by the reports of citizens returning from abroad (especially traders and ambassadors)106 or which make provision for the dispatch of ambassadors to deliver the decree to the 104

In our entire corpus of more than 2,500 inscriptions known to concern proxenia there are very few exceptions to this rule (excluding Romans whose origin was frequently identified instead by their tria nomina) and those there were predominately relate to individuals who were well known in their own right: from Iasos, SEG 36 983 for the sons of Peldemis (before or after 361/0 bc, of which 38 1059 is a much later reinscription, see Fabiani 2013); from Eretria IG XII 9 221 (= Knoepfler (2001a), no. VIII, 318–316 bc), for Aristonous, bodyguard of Alexander, and Knoepfler (2001a), no. 15 (c.260–255 bc); IG XII 5 1004 (Ios, for the nesiarch Bacchon, c.290–280 bc); XI 4 613 (Delos, for a Mytilenean peripatetic philosopher, 270–260 bc); Dimitrova (2008), no. 5 ll. 79–80 (Samothrace, for two theōroi from King Attalus; see Appendix no. 14.5.2, c.150 bc); I.Erythrai 6 (Erythrai, for the Athenian general Konon, 394 bc). For IG XII 9 1187 l. 31, Aristoboulos, son of Persaios, probably a Macedonian courtier (Histiaia, Appendix no. 8, 264/3 bc), see Chapter 3, Section 3.1.1 with n. 28. In Tit.Cal. 1A and B (Kalymna, fourth century bc) there are two further examples of individuals who are not given ethnics (in 1B alongside another individual, a Delphian, who is) for reasons which are unclear, but perhaps relate to the relatively early date of these texts in the fourth century bc (on this, and on a similar case in the proxeny list of Lousoi, Arkadia, see Appendix no. 12). The omission of ethnic in I. Oropos 303 (c.240–230 bc) is easier to explain—it relates to the singular nature of this text, a decree offering proxenia with its attendant privileges to any individual willing to loan more than a talent to Oropos for the construction of a defensive wall. In the bizarre Kydonian list of proxenoi granted the use of certain lands, IC II x 1, Skirtias, instead of receiving an ethnic, is identified as a euergetēs—in this context probably indicating that he was not a proxenos at all, reinforcing rather than undermining the significance of the ethnic for proxenoi. I.Thrac.Aeg. 400 (late fourth to early third century bc) is not an example because the ethnic of the honorand was superfluous in this decree set up in his home community, Zone. The omission of the ethnic (or ethnics) in the case of the proxenoi listed in IG IX 12 1 31 f ll. 32–3 (Appendix no. 1.8.1, c.223/2 bc) was probably accidental. See also Knoepfler (2001a), with discussion of this phenomenon at 279–81. 105 Headings: e.g. II2 63; I.Priene 6; and BCH 57 (1933), 492. Relief sculpture (especially civic badges, parasēma): Lawton (1995), on the Athenian examples (her nos. 29, 30, 42, 68, 79, 87, 158, and perhaps 11 and 114) to which now add IG II3 495 (= II2 267) with Mack ap. Lambert (2012), 402. For the material outside Athens, see Ritti (1969); Zagdoun, FD IV 6, pp. 61–75; Lawton (1995), 21–2 n. 89; Knoepfler (2001a), 30, with the addition of SEG 47 1659 and Killen (2008). Antigonus of Karystos, the third-century writer and bronze-worker, introduces the parasēmon of Krannon in Thessaly as something depicted on inscribed proxeny grants (Kd H æ ØH H IƪæÆç ø e Ææ Å  B ºø, ŒÆŁ æ K d ŁØ  A Ø æ ÆæÆ ØŁÆØ,  ªæ ç  ÆØ  ŒæÆŒ . . . , Mir. 15a 1.1). 106 e.g. IG II3 379 (= II2 343, Athens, 323/2 bc); II31291 (= II2 785, Athens, 184/3 bc); FD III 1 152 (Delphi, c.150/49 bc); I.Byzantion 1 (175–171 bc).

54

Proxeny and Polis

honorand at his home polis,107 reinforce this sense of the proxenos as useful in an external context. Other inscriptions construct the proxenoi of a poleis as a geographically distributed network, including decrees appointing proxenoi in relation to pre-existing proxenoi,108 and lists which present an ordered catalogue of all proxenoi organized by city and, usually, by region as well.109 In one unusually forthcoming inscription, the motivation for passing an honorific decree is explicitly given in these terms: ‘ . . . concerning proxenoi, so that there will be, in Kyrene, individuals who will take care of the Athenians who arrive there’.110 The assumed importance of the proxenos’ ethnic is also clearly communicated by the frequent need for additional information to be given where this expectation, that the ethnic indicated the context of the proxenos, was not met. Thus when an individual was honoured for performing the services of a proxenos in a different context—a royal court or military command—this was often explicitly stated: ‘since Alkaios the Ainian, while spending time at the court of King Demetrius, lives his life doing whatever good he can for the Athenians who come . . . ’ ( . . . ØÆ æø Ææa HØ Æ ØºE . . . )111 or ‘having been put in command of a garrison’ ( . . .  ƪ  Kd B çıºÆŒB . . . ).112 Proxenia, primarily an inter-polis institution, had to be explicitly adapted to deal with these different sorts of political community, royal court, and military camp. The intermediary functions performed by the honorands within this community were the same, or sufficiently similar, that the same set of formulaic expressions could be used to describe them, but the ethnic of the honorand, which usually indicated the context in which these services had predominately taken place and would continue to do so, often needed to be

107 e.g. IG XII 7 388 (Minoan decree for a citizen of Aigiale, 200–150 bc); I. Lampsakos 7 (= I.Thasos 171, decree of Lampsakos for Thasian, c.130 bc). See Marek (1984), 372–3 and 376–9. 108  a F  æå   æ  - AØ ºØ K øØ, I.Knidos 603 ll. 9–13 (Knidos, 400–350 bc);  a H []Ææå ø [æ ]ø ŁÅ Ø, IG XII 5 1000 ll. 6–7 (Ios, fourth century bc). 109 For examples, see Appendix nos. 4.1, 7, 10.1, and 15.1. 110 [æd b] æ ø ‹ø i [t Ø K ˚ıæÅØ], ¥ Ø Kغ [  ÆØ ŁÅÆø ]H IçØŒ ø, IG II² 176 ll. 17–19 (377–352 bc). 111 e.g. IG II² 495 ll. 11–12 (Athens, 303/2 bc). 112 e.g. IG XII 6 30 ll. 9–10 (Samos, 306–301 bc).

The Anatomy of an Ancient Institution

55

supplemented.113 In much the same way, in a rather smaller collection of examples, when individuals resident in cities other than their native poleis are named as proxenoi, this is explicitly given as the context for the services they provide.114 There are apparent exceptions, cases where we infer that an honorand was appointed proxenos for reasons that had little to do with the polis to which he belonged although the decree in question contains no supplements of this sort. The most obvious examples are where non-polis, regional ethnics were used of individuals. The best known of these is the regional ethnic of Macedon ( ƌ). Those Macedonians, whose community of origin is not further described,115 frequently receive proxenia explicitly as the courtiers or officers of kings but there is also a significant number for whom no such additional designation is given, where it is unclear in what context they functioned and were expected to function in future as proxenos. Perhaps some of these were atypical proxenoi, for whom such a grant signified a bond of friendship not conceived of as a link to any other community. However, in this concise epigraphic genre it is also probable that in many cases it was not thought necessary to

113 In the case of I.Oropos 175 (Oropos, 221–205 bc) and IG XII 6 118 (Samos, 280–246 bc), the analogous nature of the functions performed in each context is made particularly clear—the honorands in question are described in terms which make it clear they performed as proxenoi at both their home cities and court. On the force of diatribō, see Savalli-Lestrade (1998), 256–7, and Chapter 2, Section 2.2.3. 114 The earliest example may be I.Lindos 16 (411–407 bc; but against the suggestion that it names an Aiginetan proxenos at Naukratis, see Bresson 1980, 302–7); IG XI 4 588 (the Delians honour a Naxian who aids their citizens at Alexandria, 300–250 bc); IC IV 218 D1 l. 3 (the Gortynians honour an individual resident at Messene, first century bc); SEG 2 332 (the Delphians honour four Pergamenes resident in an unknown third polis); IGUR I 2–3 (the Akragantines and Maltese separately honour a Syracusan resident in Rome, c.100 bc). I.Eph 1447 is an analogous example in which the Ephesians honour four Kerameis on Rhodes for providing services to visiting Ephesians, though the grant here is of politeia rather than proxenia (third century bc). Xenophon’s situation, installed in Skillous as proxenos of the Spartans, would be another example (Diog. Laert. 2.51). 115 Like other regional ethnics, individuals bearing this ethnic are often further identified as the members of specific poleis, e.g. as a Macedonian from Amphipolis ( ÆŒg K[ ] çغø, IG XII 6 19), in which case the expected civic context for interactions is even more strongly marked. The contrast is particularly clear in the list of proxenoi appointed in a single year from Histiaia (XII 9 1187, Appendix no. 8), where we find three Macedonian nominees for whom an additional polis designation is given (one from Aigai; two from Thessalonike) and one recorded simply as a Macedonian. On these expanded ethnics in general, see Fraser (2009), 119–42.

56

Proxeny and Polis

give supplementary information—such as a particular Macedonian’s place at the king of Macedon’s court, service in a royal army, or residence in a third city—which was common knowledge in the town making a particular grant.116 In other words, although the regular recurrence in these brief inscriptions of such pieces of supplementary information, where an ethnic was not thought sufficiently informative, does strongly suggest that the community indicated by an ethnic was usually thought important, the absence of this information in other cases where it might be relevant does not support the reverse argument.117 The other class of exception mentioned most frequently in the scholarship which might complicate this picture are metic proxenoi.118 If metics, permanent non-citizen residents in the granting polis, regularly received proxenia, this might make it more difficult to argue that this was conceived of as an intermediary institution enabling interaction with an external community. There is, however, no evidence that proxeny was regularly granted to individuals expected to permanently reside at the granting polis.119 At Athens, the polis where metic-proxenoi have been most confidently identified, there existed another well-defined honorific status for resident foreigners, isoteleia, distinct from proxenia, and with a different, standard set of appropriate accompanying privileges (including the right to march with the citizen rather than alien contingents in the Athenian army).120 The fact that proxenia and isoteleia were granted 116

On the omission of ethnics, see n. 104. It is striking, however, that proxenoi identified simply by the ethnic ÆŒg are outnumbered by more than 200 per cent by those given an extended Macedonian ethnic, or even, simply, their polis-ethnic (95 occurrences of extended or polis-ethincs versus 40 proxenoi simply identified as Macedonians; based on Tataki 1998). 118 Identified by Clerc (1893), 218–20, and accepted, at least in the case of the Akarnanian doctor Euenor, by Wilhelm (1942), 58. On Euenor, IG II3 324 (= II2 373) 2 and II 374 see above, pp. 47–8. 119 The grant of enktēsis, used by Clerc to identify metic proxenoi (1893, 218–20), in fact had no such implication of residence. See Chapter 2, nn. 146 and 148. On the difficulty otherwise of identifying metic proxenoi, see Whitehead (1977), 13, 30, and 63–2. Whitehead positively identified only four proxeny decrees for metics: IG II3 367 (= II2 360, 325/4 bc); 324 (= II2 373, 337–321 bc); II2 786 (after 229 bc); and 835 (after 229 bc). On II3 367 and 324—atypical decrees—see pp. 47–8, and n. 121. The few other clear-cut cases in our corpus of proxeny decrees include: IG XI 4 789 (Delos, c.200 bc) and IG V 1 1146 (two Roman money lenders at Gytheum, 71/0 bc). 120 Those honoured with isoteleia tend to be described in quite different terms— instead of reference to the continual performance of services for individuals they encounter, the emphasis is on ordinary and extraordinary financial contributions (the 117

The Anatomy of an Ancient Institution

57

together only very rarely is evidence of the emphasis of proxenia on external political contexts.121 The few decrees which do grant both are clearly exceptions, probably made either because it was unclear where the individual would settle, or because it seemed likely that they would spend extended periods in both contexts.122 It is striking that, in comparison with the numerous gravestones for isoteleis, only five are known for proxenoi or their families at Athens (and perhaps only two elsewhere).123 Apparently proxenoi, for the most part, lived, and indeed died where they were expected to, in political contexts external to the granting polis. In some decrees, the language used in the generic descriptions of services performed by the honorand often makes it clear that the ethnic of the honorand indicates the context in which these took place. The citizens of the honouring polis whom the honorand is described as helping are frequently described using the participles of payment of eisphorai; participation in epidoseis) as well as marching out in support of the Athenian army. 121 IG II2 83 (c.375 bc) and perhaps 287 and IG II3 493 (= II2 288, mid-fourth century bc) if we are right to restore proxenia in the former case and isoteleia in the latter (it is rightly doubted by Lambert in the new edition). II3 367 (= II2 360) should probably also be taken as an example (although isoteleia is not granted per se, it is presumably encompassed by the rights to march alongside and pay the same eisphorai as the Athenians, which are). 122 Contra Pečírka (1966), 33, who held that isoteleia implied residence, Whitehead (1977), 63 n. 26, pointed to the existence of clauses which limited the grant of isoteleia, requiring residence (e.g. IG II² 287). The dual grants of proxenia and the rights of isoteleia to Herakleides of Salamis on Cyprus, who as a merchant would not have been expected to be resident all year round at Athens, is a particularly clear example of this (IG II3 367). 123 Funerary monuments mentioning isoteleia (29 from Athens, fourth to first centuries bc): IG II2 7862–81, 8652; Agora 18 384–5; SEG 18 112–13; 21 940; 26 311–12; 57 228. Funerary monuments mentioning proxenia at Athens: IG I3 1154 (erected at public expense, 460–450 bc); SEG 56 75 (for a proxenos from Akanthos, c.450–424 bc); 26 352 (origin of proxenos unspecified, Hellenistic; the funerary monument of this individual’s daughter, 26 351, also mentions his proxeny); 29 272 (origin of proxenos unspecified, third century bc); IG II² 9304 (for the wife and daughter of a Megarian proxenos, fourth century bc). Outside Athens: ML 4 (the famous Corcyrean mnēmē of their proxenos lost at sea, 625–600 bc); Clara Rhodos 2 (1932), 219 no. 61 (Rhodes, for a proxenos from Cyrene, third to second century bc). IG XII 8 420 (Thasos, fourth century bc) is probably another example, though it could also be a dedicatory inscription. We also possess two public funerary monuments from Thera, the first honouring an individual from Sparta who had been appointed proxenos in a list from c.100 bc, but without mentioning this status (IG XII 3 1625; cf. 1299/1300 l. 25), and the second honouring the dead wife of a proxenos from Astypalaia (IG XII 3 497).

58

Proxeny and Polis

aphikneomai (IçØŒ  ) and paragignomai (Ææƪ ), as ‘coming to’ the proxenos, interacting with him where he normally resides—the polis indicated by his ethnic, court, or camp if this is explicitly mentioned. Even when the participles of other verbs are used, most commonly entynchanō (K ıªå ø) but also apantaō (IÆ H), which are on the surface more ambiguous about where such interactions have, and are expected to, take place, it is nonetheless often made clear, because the honorand’s place of residence is explicitly mentioned when it differs from the polis of his ethnic, that it was this external political community rather than the granting polis which was at issue.124 It is also significant that the citizens in these meetings are always described as the active party, the subject of the verb or participle—it is they who are said to ‘meet’ or ‘come to meet’ the proxenos, rather than the other way round.125 The picture of the typical proxenos conveyed by these generic descriptions is thus of an individual providing services to the honouring polis, and to citizens from it, in the context of another community—usually assumed to be that indicated by his ethnic. This is vital for understanding how proxenoi in general were conceived. Nonetheless, an important question must be answered before we can consider what specific functions were associated with the proxenos-paradigm. That is, what use should we make of the other information we have concerning the situations and activities of particular proxenoi in reconstructing the role which they performed?

124 e.g. [Ææ’  ]تøØ ŒÆd ˜ÅÅ æøØ Ø[Æ æø å]æ Ø  Æı e Ææå[ ÆØ ŒÆd Œ Ø]BØ BØ ºØ ŒÆd NÆØ [E K ıªå ] ı Ø H  ºØ H (IG XII 6 20, Samos, 314–306 bc); ØÆ æø Ææa Æ ØºE !ŒıŁH ˚ÆØ ÆØ h ı ŒÆd æŁı  Æı e HØ øØ ØÆ ºE [Ææå ] ŒÆd NÆØ E K ıªå  ı Ø ÆP HØ H  ºØ H (IGBulg I² 41 ll. 4–8, late third to early second century bc). 125 This is highlighted in IG XII 7 5 (Arkesine, Amorgos, c.357/6 bc)—a very rare exception to this rule—in which the proxenos is the subject of this verb because the citizens he came across (æØ ıå) and freed had been captured in warfare and therefore could not come to meet him in this active sense. The verb entynchanō is used in this way almost exclusively of non-citizens who are usually to be understood as operating in a context external to the polis, but when, very rarely, it is used of resident foreigners or citizens, its precise significance is unclear—whether indicating the fact that they also operated outside the polis (Knoepfler 2005, 292–4) or emphasizing social distance (Ma 2012b, 156; cf. Reger 2004, 169–70 with n. 88). This verb probably also had overtones of ‘petition’ and ‘appeal’ which are prominent at least in Hellenistic use (e.g. Plb. 4.76.2; in S. fr. 88.8 it seems to mean ‘obtain an audience’, LSJ).

The Anatomy of an Ancient Institution

59

1.2.3. The occupations, activities, and milieux of proxenoi In response to what he saw as an inappropriately juristic model of proxenia in which proxenoi had ‘duties’ or ‘tasks’ which they were expected, almost as if by contract, to fulfil, Marek argued for an approach to interpreting the functions of proxenia based on the collection, categorization, and analysis of proxenoi according to their other known activities.126 The different spheres of activity thus represented— political, cultural/social, and economic—accurately reflect according to Marek the interests of poleis in making specific grants and give us, because a relatively high proportion of honorands can be so categorized (up to a quarter of Marek’s estimate of over 3,000 attested proxenoi), a representative sense of the different sorts of honorands singled out and services sought with the grant of proxenia.127 This sort of approach presents us with long series of different categories, in contrast with the coherent and cohesive inter-communal mediating institution depicted by the language of these decrees. In place of proxenoi we find: kings and dynasts; royal functionaries; military officers; federal officers; political envoys; political partisans and agents; ship-owners and merchants; bankers and lenders; builders; foreign judges; doctors; philosophers, scholars, poets, and artists; theōrodokoi and theōroi; priests, religious donors, and functionaries.128 The diversity thus represented is striking, but it is also potentially deceptive. The suggestion that a royal official was appointed proxenos because of his position, a doctor because of his profession, or a banker because of his financial activities and resources—each for services he performed and was expected to perform in these capacities—may seem plausible.129 However, it obscures what is important about the way in which grants are made and, by implying that, if we had enough information, all proxenoi could be categorized and thus explained, it falsely ascribes an equal significance to these different sorts of category. This sort of analysis also makes the mistake of assuming that the information we happen to know about a particular individual is the relevant fact for explaining a particular

126 127 129

Marek (1984), 143–6, contra Monceaux (1886), 13–14. 128 Marek (1984), 333–4. Marek (1984), 333–81. Gauthier (1985), 143.

60

Proxeny and Polis

grant of proxenia. Herakleitos, son of Asklepiades, may well have been the ‘Halikarnassian guest’ and fellow poet, whose fate Callimachus famously lamented, but his receipt of proxenia from Histiaia and Chios is more likely to relate to his importance in relation to the Ptolemaic court or as a citizen of his native polis.130 There is, of course, a difference between using the information about honorands provided by specific decrees to interpret their grants and drawing prosopographical connections to infer motives about which the decrees themselves remain silent.131 However, even setting this issue to one side, there is good reason to think that these categories in general collect atypical cases. At various points his account relies heavily on material from Delos and Delphi—places likely to be unusual—and both underestimate the typical scale of proxenoi-networks and overestimate the degree to which the material we have is likely to be representative. As I argued in the Introduction, these sorts of proxenoi, whom Marek places at the heart of his interpretative scheme, are probably greatly over-represented in proportion to the whole.132 It is also necessary to distinguish the benefactions which lead to an individual being granted proxenia from the sorts of service which he was expected to provide as proxenos, and the anticipated context for these interactions with individuals from the honouring polis. Activities which were not considered characteristic of proxenoi—like a medic’s period of professional residence at a polis—might nonetheless be interpreted, in the way they were conducted, as revealing the same eunoia for the granting polis which was the fundamental basis of the relationship of proxenia and the predictor of future benefactions.133 130

See Swinnen (1970). Of the thirty-seven proxeny decrees for medics collected by Marek (1984), 372–3, in eleven no specific reference to either their identity as doctors or medical activities is made. 132 See Introduction, Section I.3. 133 Gauthier (1985), 142, discusses one particularly good example of this, a decree for Diokleidas, who delivered a royal letter to Minoa, exhorted the Minoan demos to cease from civil strife and pledged himself in the future to do whatever good he could for them—receiving a grant of proxenia for his pains (IG XII 7 221 b, 250–200 bc). Domingo-Gygax (2009), 179, takes this as an example of proleptic honours according to his model, given in the hope of prompting future benefaction. In fact, Diokleidas’ services—in delivering the letter, exhorting the Minoans to make internal peace, and pledging his own future aid—were, for the Minoans, real, revealing very vividly his own feelings of amity (so Gauthier), and making him in truth a deserving nominee for proxenos. 131

The Anatomy of an Ancient Institution

61

That proxenia was often granted to such individuals at the point of their departure—as they returned to their home community or went on to another—made good honorific sense, marking the close of the particular act for which they were being honoured.134 But it also agreed closely with the way in which proxenia was used to identify and strengthen links with individuals in other communities beyond the honouring polis. This sort of grant is closely paralleled in grants made to the official representatives of other communities—ambassadors (both secular and religious) and arbitrators and foreign judges invited to settle internal disputes. Their enthusiastic participation in these tasks, again, could be represented as an act of benefaction towards the honouring polis, which made plain their friendship for it. The consequent grant of proxenia sought to strengthen this relationship, as a tie with both the individual and the polis of which he was an active member. It asserted the continuation of contact between the two poleis and with the honorand there in particular, and thus of this relationship in the future. The activities of these individuals which brought them into contact with other poleis—professional in the case of the medic, public in the case of the polis’ representatives—far from hindering their position as active members of their own poleis, or other communities, probably augmented or translated into personal prestige. In at least one Samian decree for a Koan doctor it is made clear that an active medical practice could go hand in hand with the activities in which a prominent citizen might engage. Philistos, son of Nikarchos, in two parallel sections of the epeidē clause, is explicitly praised not only for providing his professional services to the citizens of the honouring polis who came to Kos on public and private business and fell sick, but also for providing the same services to these citizens at Kos that any individual might be named proxenos for providing.135 134 FD III 3 119 (c.134 bc), a Delphic honorific decree for a teacher, with its past tenses, implies that this period of residence and service has expired; cf. IG XI 4 666 (239–210 bc). 135 IG XII 4 138 ll. 11–25 (after 241 bc): KØc ºØ  ˝ØŒ æå ı ˚HØ  NÆ æe _ J A Æ åæÆ ØÆ ºE [Ææå ] ŒÆ a c NÆ æØŒc åÅ ŒÆ[d º Æ ]H _  ºØ H KØÅ Æ Æ[ Kª ˚HØ, f]_ b ŒÆ a ŁøæÆ I ƺ[ Æ e ]F  ı, f b ŒÆŁ’ NÆ[ N ˚H Ææƪ]  ı, K  Æ [N Iææø Æ _

62

Proxeny and Polis

It is therefore necessary to distinguish between the different categories used by Marek—in particular those grants which indicate the professional activities of certain individuals, which may or may not be relevant to the context in which they are active, from those made to members of non-polis communities, including members of royal courts and the officers of armies. In the latter case the information given about a particular recipient was clearly intended to differentiate him from the other recipients, who represented the assumed norm, active in their own poleis. However, the position of these individuals was in fact modelled on that of their peers established in poleis and the intermediary services which they are likely to have been called on to perform were probably in many cases very similar, even if the community in question was a court or garrison rather than a polis.136 Moreover, such royal friends or mercenary chiefs were not necessarily thought of as being any more permanently sundered from the poleis to which they belonged by birth than any of the other individuals who moved for the sake of their economic pursuits. In a Samian decree, two royal philoi of Ptolemy are honoured for the services they continually provide for Samians coming to Lampsakos, their native polis, and there is no reason to suppose that this sort of activity could not take place after or before even demanding royal appointments.137 Allowing for mobility in the Hellenistic period ought to mean allowing for the continuation and constant renewal of ties to poleis—and not just movements away from them. Indeed, when the citizens of Paros honoured Apollodoros of Kyzikos for the regard he had shown them as nesiarch of the League of Islands, it was to the city of Kyzikos that the Parians dispatched the decree honouring him, asking for permission to erect a statue of the honorand there. In other words, despite the (extended) royal office which provided the context for these services (and the fact that this individual apparently continued to rent property on Delos), this was still conceived of as being an KØŒØ] ı Ø øØ  [ a  Å çغ Ø]Æ, vac. E  ¼ºº [Ø E _ ÆæƪØ  Ø] H  ºØ H N ˚H [ ººa Ææå ÆØ åæ]Æ, PŁb Kºø _ [æ ŁıÆ a ŒÆºH ]å  Æ ıªŒÆ Æ Œı Ç[Ø ÆP E ŒÆd I]d ÆŒ ø N A æd z ¼ Ø [ÆP e] ÆæƌƺBØ. _ __ 136 For armies as political communities, see Hornblower (2004); these similarities are further brought out in the honorific decrees from Rhamnous. 137 [çº] Ø Z  F Æ Øºø — ºÆ ı,   E æ[ ]æ  åæ Ø s Ø [Œ]Æ _ _ _ _ Z]  [d æŁı Ø ŒÆd] NÆØ E IçØ[Œ] ı[ Ø] [N] ¸[ ][ł]Æ[Œ  åæ Ø Ø _ _ _ _ _ bc). Ø º ı ŒÆd åæ[Æ . . . ], IG XII 6 118 ll._ _ 4–7_ (280–246 __ _ __

The Anatomy of an Ancient Institution

63

inter-polis interaction, and Kyzikos was still the context in which, the Parians felt, Apollodoros would wish to be honoured.138 There is also the difficulty of how we use the information we glean from the identity of the proxenoi to discuss the interests of the honouring polis. Marek argued on the basis of the paucity of traders explicitly attested in proxeny-grants against the then dominant hypothesis that trade was one of the most important factors in the appointment of proxenoi. For the same reason, he also argued that trade was likely to be a less important factor in mobility than others, better epigraphically attested—the movements of armies, artists, theōroi, and judges—suggesting that these motivations should be assumed, before trade, for the interactions with the areas implied by the honorand’s ethnic.139 It seems implausible, however, that a large proportion of movement in the Hellenistic world was not to some degree motivated by economic considerations. Whatever intermediary services proxenoi were expected to provide, citizens travelling to take part in economic interactions (in trade, in contracts, and, as a result, in disputes and difficulties) seem likely to have often been those individuals most in need of them. This need is clearly articulated in two Athenian decrees awarding proxenia where the polisauthority proclaims that a particular award is made because of services either performed for, or reported by, its citizen merchants and ship-owners.140 However, these needs did not disappear even if the grain shortage was eased, which had made them a matter of public concern when the decrees in question were inscribed. These sorts of services would certainly not have required the experience or resources of proxenoi who were themselves heavily involved in trade. Instead, the ability of a proxenos to provide these sorts of assistance would have depended principally on his status and prominence as a citizen of the community in question. It is thus not surprising that merchants should be comparatively invisible as 138 Michel 534 (= CIG 3655, start of the third century bc), recording the text of the answering Kyzikene decree; the Parian copy of the decree honouring Apollodoros is simply to be placed in the public records—N a Å ØÆ ªæ Æ Æ (l. 26). As the Kyzikene decree records only those honours for which the Kyzikenes needed to grant consent, the other honours granted by the Parians are not mentioned (they are likely to have included proxenia). Reger (1991) on the identity of this Apollodoros. 139 Marek (1984), 359. 140 See IG II3 379 (323/2? bc) and 454 (= II² 416 fr. b, 334/3–322 bc). These texts are both to be associated with the grain shortages in Athens in 331–c.320 bc.

64

Proxeny and Polis

recipients of proxenia, except in special cases when they performed particularly urgent services for the polis as a whole which were relevant to their professional activities (e.g. as corn dealers in time of food shortage or as money lenders during a financial crisis). In general, as far as the Greek poleis were concerned, the proxenos provided services to citizens who came to his community on public or on private business (almost all trade would have been resolutely considered a private affair), and a network of proxenoi was sustained and enlarged for this general reason. But this does not mean that, collectively, the city necessarily had any sort of public policy relating to the development of trade,141 except in that its members with interests in trade were aware of and successful in promoting them in the context of the public assembly.

1.2.4. Services associated with proxenia In our corpus of proxeny decrees, in addition to the ubiquitous, generic descriptions of honorands providing services, a fairly wide range of specific sorts of benefaction are also found. It is important, however, to identify which sorts of service were thought characteristic of proxenoi and which were mentioned precisely because they were extraordinary. It is not that any form of benefaction was thought incompatible with proxenia or the proxenos-paradigm. However, whereas some services were regularly recurrent and could therefore be expected (even confidently requested) of any proxenos, others were not, for all that they might be taken as evidence of the attachment of proxenos to polis and thus identify the individual in question as a suitable proxenos. Gifts by proxenoi of money probably fall into this latter category, as do gifts of corn, and professional and cultural contributions to the honouring polis—all are attested of proxenoi, but, comparatively, very infrequently.142 Similarly, when foreign 141

Contra Marek (1985). Financial gifts: IG VII 2418 (Thebes, 355–351 bc); [Dem.] 40.36 (Athens, a sum of three hundred staters collected from Mytilene from ‘your proxenos and the friends of the city’); II3 367 (= II2 360, Athens, 325/4 bc); II2 835 (Athens, 229–201 bc). It is striking how few proxenoi are attested in the epidosis lists (only, as far as I am aware, one, see Migeotte 1992, 158), which suggests that this is an important respect in which ideas of what it was to be a good citizen differed from the behaviour expected of a proxenos. For gifts relating to corn provision, see above. See also IG XII 6 56 for a grant for an Aiginetan actor. 142

The Anatomy of an Ancient Institution

65

judges were granted proxenia it was probably not to secure their services in this regard for the future but the detailed knowledge of the polis and its citizens which they would have gained by performing this role made them particularly suitable proxenoi when they returned to their own cities. The key to understanding these conventional or stereotypical services is the context in which they took place. This was the community, usually the polis, to which the honorand belonged. Citizens of the granting polis who travelled to another polis or community regularly encountered the same obstacles and sought the same aid of their proxenoi in their need. These difficulties were structural, due to the political fragmentation of the Mediterranean world in the Classical and Hellenistic periods. As a result the citizen of one city was at a great disadvantage both legally and practically outside his own community. These needs could be as basic as hospitality or accommodation—the provision of which was fundamental to the relationship of xenia (a private institution which similarly circumvented these inherent limitations on external intercourse) on which proxenia, to an extent, continued to be calqued. The sorts of service, however, which loom largest in our evidence relate to the ability of the proxenos, as citizen, to access and influence local legal and administrative structures, as well as his knowledge of local conditions, his local network of friends, and personal prestige or authority, and, in general, his ability, as the active member of one polis, affiliated by proxenia grant to another, to act as an official connection between the two.

1.2.5. Services for poleis, their representatives and citizens Our evidence for the services which proxenoi habitually performed for poleis and their representatives, consisting of references to the actions of proxenoi in the ancient historians and orators and a select number of more descriptive inscriptions, has not substantially changed since Gschnitzer collected it for his 1973 RE article. It is surprisingly sparse and patchy for such a significant institution, in no small part because so many of the activities of proxenoi were too mundane to be worth specifically describing epigraphically, or to rate much of a mention in our narrative histories. However, from these actions, which were thought sufficiently important or exceptional to

66

Proxeny and Polis

be explicitly recorded, or are mentioned in connection with events that were, it is possible to get a sense of the innumerable habitual services which lie in their shadow. According to our stereotypical descriptions of proxenoi, the proxenos was one who always said and did what was advantageous to the demos (ºªø ŒÆd æ ø Id a ıçæ  Æ HØ øØ),143 who worked to secure its interests. For the most part this seems to have equated to political acts and political speech, working within the power structures of the community to which the individual in question belonged. In poleis this meant both its assemblies and magistrates, through which the collective power of the community was formally mediated and expressed. At the royal court such power technically resided not in the demos but in the person of the king. However, in its ability to shape and direct the decisions of the king, the court served an analogous function to that of the assembly and magistrates in the polis.144 In each case, success in promoting the interests of the granting polis thus depended on the proxenos’ capacity, within his community, to persuade, and thus the political prominence of a proxenos was directly related to his usefulness. Sometimes the proxenos might be able to operate, at least in part, independently, as an important magistrate or military commander. However, such appointments were usually temporary, and ultimately devolved or depended on the authority or resources of a polis or king. Inscribed decrees for Eumaridas the Kydonian and Neoptolemos the Aitolian, both honoured for services they performed as proxenoi, explicitly describe some of these services which, though perhaps exceptional in degree, were surely characteristic of this intermediary role proxenoi in general played, promoting the interests of one community in the context of another.145 Neoptolemos, once he had been apprised by the ambassadors of Erythrai of their mission to overturn 143 e.g. IG XII 9 221 (Eretria, 316–318 bc; the phrase and its equivalents are very widely attested). 144 The explicit reason which Thucydides gives for the nomination of Nymphodorus from Abdera as proxenos of the Athenians, despite the previous hostility towards themselves which the Athenians imputed to him, was his influence at the court of his brother-in-law, the Thracian king Sitalces (Thuc. 2.29). 145 Neoptolemos is explicitly referred to as the proxenos and fellow citizen of the Erythraians (of Ionia), Syll.3 412 (c.260 bc). In contrast, no mention is made in either of the two Athenian honorific inscriptions for Eumaridas of his proxenia, which is mentioned only when it (patrikē proxenia) is renewed for his son. Nonetheless, as he is not nominated as proxenos in either of the two reproduced honorific decrees (which

The Anatomy of an Ancient Institution

67

a fine levied by the Amphictyony against their polis, advised and worked alongside them zealously ( ı ºı  ŒÆd ıæÆ  çغ ø), and suggested, moreover, that they should send an embassy to the Aitolian koinon about this fine.146 Eumaridas of Kydonia is similarly honoured for publicly serving as advocate on behalf of the interests of Athens ( ıŪæÅ  N e   Æ æÆåŁBÆØ a ıçæ  Æ), when the Athenians sent an embassy to re-establish friendly relations and the annulment of any right of plunder (presumably against Athenian property and territory) granted by the polis of Kydonia.147 Eumaridas, moreover, took a personal part in this embassy when it progressed to another city, Knossos, and recruited the support of his friends in a third, Gortyn.148 This is a striking example in which the potential influence of the proxenos, and his personal network of friends, extended beyond the bounds of the polis. But what is described here must also reflect the situation within the polis of the proxenos. A city’s proxenoi ideally threw their personal political weight behind the requests of delegates, and that could include mobilizing the network of friends and supporters from which their political influence in part derived. As the decrees themselves often assert, the proxenos strove zealously alongside the ambassadors sent by the polis.149 The proxenos made himself the eager champion of the interests of the polis in his political arena.150 Whether within his own polis, or at the court of a king, this frequently took the form of making speeches on behalf of the interests of the polis that honoured him.151

confer very substantial honours, rarely given without a previous grant of proxenia), it seems clear that this grant must already have taken place. 146 Syll.3 412 l.5. 147 IG II3 1137 l. 14 (= II2 844). For this interpretation, with bibliography, see Pritchett (1991), 144–7. 148

ıæ ı  b ŒÆd N ˚ø e ŒÆd f ı å ı, øŒ b ŒÆd KØ ºa E æ ı ÆE N — ºæÅÆ æe f çº ı, ‹ø ıæƪÆ ø ÆØ  ’ ÆP H æd H ıçæ ø, IG II3 1137 ll. 15–18 (= I2 844). 149 E [ ]  Ø e F  ı æ []ı ÆE ıƪøØÇ[  KŒ ]H ØÆ ºE, IG II3 1336 ll. 13–15 (= II2 945, 168/7 bc; the end of this decree is lost, but it is probable, given the language used in the motivation clause, that the honorand was granted proxenia or the higher honour, politeia). 150 æŁı  ÆP

Æı e IªøØ a ÆæÆ Œı ÇØ bæ H AØ ºØ

ıçæ ø, FD III 1 152 ll. 5–6 (Delphi, 150/49 bc). 151  ºº[ f ŒÆd] ıçæ  Æ ºª ı  ØE ÆØ bæ B ºø[], I.Oropos 175 ll. 9–10 (c.205 bc).

68

Proxeny and Polis

Numerous examples can be found for this sort of active political support given by proxenoi (or those subsequently nominated proxenoi) to representatives of poleis who sought a range of different concessions from the cities and royal courts where these individuals were active. These could take the form of requests for expressions of friendship or, as described in the Samian exiles’ decrees, advocacy before King Alexander of the Samian right to return to their city.152 We also have at least one attestation of aid given by two individuals (one of whom was a proxenos), probably both at that time members of the Athenian boule, to a panel of ekdikoi (public advocates) from Karthaia, sent to plead on behalf of two Karthaian citizens in the Athenian courts.153 This example should probably stand for many more cases where aid was given to official delegations in the context of local legal institutions—with the aid given, and qualifications needed, probably resembling closely better attested aid in the context of assemblies. In the light of this expectation of aid, it is not surprising that at Klazomenai individuals were forbidden from sitting as jurors on cases which involved citizens of poleis of which they were proxenoi, in much the same way that they were forbidden cases involving their close relatives.154 What is apparent is a fundamental understanding of the proxenos as a conduit, an official link, capable of facilitating interaction not least by being able, officially, to confirm the identity of visitors from the polis to which he was affiliated (either by knowing them personally, or being able to relate them to others he did know). The proxenos, at least in theory, was someone whose word carried weight in both communities, capable of standing surety for the identity of private visitors and, especially, public delegates, and introducing them to the relevant authorities. When, after the battle of Corinth, Agesilaus pointedly ignored the Boiotian ambassadors who had come

152 IG XII 6 28 (Samos, 306–301 bc); XI 4 765 (Delos, beginning of the second century bc) describes, though in general terms, a Pergamene’s efforts to aid Delians who came to obtain concessions at the court of the Attalid king. 153 IG XII 5 528 and 538 + add. pp. 319 and 321, with Mack (2011), 338–9 (303/2 bc). 154 SEG 29 1130bis B ll. 35–41 for the jurors’ oath (first half of the second century bc). A similar prohibition is known from a treaty between Chaleion and Oeantheia in western Greece (IG IX 12 3 717 B ll. 10–11, 500–450 bc): if the xenodikai fail to reach a decision, the xenos may choose the judges he likes other than his proxenoi or personal friends (idioxenoi); for this interpretation, see Zelnick-Abramovitz (2004), 94–5.

The Anatomy of an Ancient Institution

69

to treat with him, ‘even though Pharax, their proxenos [at Sparta], was standing by to introduce them’, what was noteworthy was not the fact that a proxenos attempted to perform this function, but the rebuff which he and the Boiotians received, despite his considerable importance at Sparta.155 This was a service which other proxenoi would have been called on to perform whether at royal courts in relation to kings or at cities in relation to important magistrates and the local council or assembly.156 At the panhellenic sanctuaries of Olympia and Delphi, this role of intermediary between political communities seems to have found expression in a formal ritual function—the performance of sacrifices which foreign delegations wished or needed to make at a shrine (often as a preliminary to consulting the oracle) but which, as non-citizens, they could not perform themselves.157 The proxenos, although ostensibly appointed only by one community to serve as their contact in his own city, was also, in practice, a potentially bi-directional link. Poleis made use of citizens who were the proxenoi of other cities to establish official contacts between the

› b ªÅ ºÆ   ºÆ ªÆº çæø  ı b P' ›æA KŒØ, ŒÆæ  æÆŒ  F æ  ı Ææ ÅŒ  ÆP E, ‹ø æ ƪ ª Ø· (Xen. Hell. 4.5.6). The importance of Pharax is reflected in the numerous other references made to him (Hell. 6.5.33; Hell. Oxy. 2.1; D.S. 14.63.4, 79.4–5). 156 The lexicographical sources refer to this as one of the characteristic functions of the proxenos: Poll. 3.59; b › . . . æ ªØ a æ Æ y  æe e Å Ø  (Ael. Dion. iota.2.2). 157 The necessity of this function emerges in the decree in which the Delphian polis as a whole undertook to perform as proxenos for the city of Sardis, which, because contact between the two communities had been broken, lacked a proxenos at Delphi (Syll.3 548, c.275 bc; on the interpretation of this text see Marek 1984, 169–70; see also Sourvinou-Inwood 1990, 295–7). A similar function seems to be presupposed when we see proxenoi at Olympia instructed to keep individuals or communities away from the altar (so IvO 10 and 13, early fifth century bc). These texts have also been understood as attestations of a separate institution—of magistrate proxenoi (‘la proxénie liturgique’, Monceaux 1886, 6–11; for a recent restatement of this, with bibliography, see Zelnick-Abramovitz 2004, 102–4). This, however, is highly implausible, requiring the use of the same term ‘proxenos’ to refer to two entirely distinct, co-existing institutions—internally polis appointed magistrates versus external polis-friends—separately charged with performing indistinguishable intermediary functions. The prerogative of the Spartan kings described by Herodotus, to appoint citizens as proxenoi for visiting delegations (Hdt. 6.57), has sometimes been used to support this reconstruction of a liturgical form of proxenia (Zelnick-Abramovitz 2004, 103) but is best understood as a system for supplementing the arrangements which states normally made to appoint proxenoi rather than as a separate institution (Mosley 1971b). 155

70

Proxeny and Polis

two, frequently sending them as representatives to the poleis with which they were affiliated.158 Proxenoi, after all, were known at the other community, and possessed both personal contacts and an official status there, qualifications which must have made them appear likely to be more effective as ambassadors and, particularly, more persuasive speakers (in the speeches attributed to proxenoi their proxenia is certainly emphasised in this way).159 The importance of this potential of the proxenos to serve as a trustworthy link between poleis seems to have increased when the two cities themselves were in a state of war and other ties were severed.160 The right of access to the council or assembly widely given to proxenoi by the granting polis (prosodos/ephodos) and the right to enter or leave their city unmolested even in time of war—rights which non-citizens did not possess by default—certainly facilitated this use of proxenoi as a bi-directional link, and were perhaps granted in part with this intention. The provision of hospitality to individuals sent by the granting polis was a practical instantiation of this intermediary function and one which emphasized the nature of proxenia as a public form of xenia—but which surely also served to strengthen the personal ties of the proxenos to leading members of the polis he represented. This function is frequently mentioned in a number of the definitions of proxenia given in the lexicographical sources161 and the evidence for it in other contexts, although sparse, is highly suggestive. When, during the uproar caused by Sphodrias’ attempt on the Piraeus, three Spartan ambassadors were seized, they were staying at the 158 The Rhodian proxenos at Delphi was sent to Rhodes to recruit arbitrators (FD III 3 383, 180–179 bc); Philinos of Byzantium was probably appointed proxenos of Athens (IG II2 76) before being sent there on an embassy (IG II2 41); King Agis of Sparta sent the Byzantine proxenos at Sparta to bring that city over (Xen. Hell. 1.1.35; but see the sequel 1.3.15–9 for the most inappropriate behaviour of this proxenos; cf. Thuc. 8.8, 80; D.S. 14.12); Neapolitan proxenoi at Tarentum were sent (along with the most notable Tarentines) to Naples (Dion. Hal. 15.5.2). See Perlman (1958), 187, but note that some of the references in his n. 1 are wrong. 159 See Chapter 2, pp. 130–1; three of the six references to proxenia in Xenophon are made by individuals seeking to persuade audiences at the polis of which they were proxenoi. 160 For the use of proxenoi in peace negotiations in particular, see Chapter 2, n. 109. For their importance in arranging the ransoming of captives, see below. 161 e.g. Poll. 3.59; Ael. Dion. iota 2.2; schol. Dem. 15.7; see Meier (1843), 2–3 n. 8, for more material. Pollux also makes reference to proxenoi arranging the provision of proedria for ambassadors, cf. Aeschin. 2.110.

The Anatomy of an Ancient Institution

71

house of Kallias, their proxenos (or, more properly, one of their proxenoi).162 That this was a norm, at least at this time, is shown by the defence these ambassadors made of themselves. They argued that if the attack made by Sphodrias had been officially sanctioned, and if they had known about it beforehand, they would hardly have placed themselves in the power of the Athenians, and would certainly not have stayed with their proxenos, implying that this is where they would be expected and easily found. Though we have other evidence for elite members of a polis staying with their proxenos (were they always on public business?),163 we do not know whether less politically prominent traders or tradesmen felt they could expect to receive the hospitality of their proxenos, although it seems probable that many had private arrangements with their friends and formal xenoi at poleis to which they regularly went.164 An even more important and useful aspect of the proxenos’ role as intermediary than the provision of hospitality, was as a source of information for both public and private visitors concerning local conditions, as well as of access to his own network of local friends and contacts. When in Aristophanes’ play, The Birds, the first words out of the mouth of the Athenian episkopos are ‘where are the proxenoi?’ ( F æ  Ø, Av. 1021), he does not just represent the attitudes and needs of a functionary of the Athenian Empire. His need for information on the local political conditions prevailing in Cloudcuckooland—and advice for how to proceed in his mission— was the same that any other representative of a city had, and he, naturally, relied on his city’s proxenoi. This dependence is quite neatly illustrated by a story told by Thucydides, of how Alcibiades purported to give the delegation from Sparta, the polis of which his forebears had been proxenoi, tactical advice in presenting their case before the assembly.165 That Alcibiades in fact sought to sabotage this alliance rather than promote it does not detract from the underlying

162

Xen. Hell. 5.4.22. ‘for you are their proxenos and the most important men amongst them stay at your house’, æ   ’r ŒÆd ŒÆ ª  ÆØ Id Ææa

d ƒ Œæ Ø Ø ÆP H, Xen. Symp. 8.39; Ion of Chios FGrHist 392 fr. 6 (Athenaeus 13. 603e-f ); Fæ’ IçØŒ  Ø Ææa

d ŒÆ ºı , `N åÅ, ŒÆd f æ P Ø, Dem. 18.82 (Aeschines here probably put up the ambassadors as proxenos of Oreos, so Hennig (1997), 355–6; see Chapter 2, n. 58). 164 See Hennig (1997), collecting the material on public hospitality in general. 165 Thuc. 5.43–5. On the interpretation of this incident, see Chapter 2, Section 2.4. 163

72

Proxeny and Polis

assumption that this was precisely the sort of help that delegations were used to receiving, and expected from their polis’ proxenoi.166 Because of their intermediary position between two poleis, proxenoi were also well placed to take care of captives and facilitate their ransom, both those taken as a result of inter-polis public warfare167 and those through private (though perhaps publicly sanctioned) raids.168 Thucydides gives a particularly good example of how proxenoi might be expected to involve themselves in the first of these. The Corcyreans taken prisoner in the sea battle with the Corinthians at Epidamnos are allowed to return under the pretext of a large ransom, for which their proxenoi at Corinth stood surety, but in reality because the Corinthians have succeeded in turning them.169 The way in which this story is framed, as a persuasive ruse, suggests that proxenoi could be expected to play this role, for which they would be admirably suited, enabling prisoners to be freed without the ransom necessarily being paid in advance (which might be difficult, given the large sums of ready cash required). Standing security themselves to their own state for the sum, proxenoi would have had a much better chance of recovering what was in effect a loan to the captives because of their honorary and often explicit legal standing within the polis of the prisoners.170 Walbank has suggested that proxenoi played a similar role when the Aitolian ambassadors agreed to, and stood surety

166 This does not mean, however, that proxenia ever developed into an organized system of intelligence-gathering agents, working to promote the interests of other poleis over what they perceived to be the interests of their own communities, contra Gerolymatos (1986), see Chapter 2, esp. bibliography at n. 190. 167 Pritchett (1991), 245–97, presents a particularly rich collection of literary and epigraphic examples of ransoming. Although Pritchett downplays the prevalence of ransoming, this wealth of material suggests that it was a relatively common practice. Care for the Spartiate captives from Sphacteria was one of the means Alcibiades used in seeking to renew his family’s proxeny of the Spartans (Thuc. 5.43). 168 e.g. IG XI 4 1054 (Delos, 229–204 bc); XII 8 159 (Samothrace, second century bc); SEG 25 539 (Eretria, third century bc). Ransom was to the advantage of both pirates and poleis. The pirates obtaining a much higher price, which reflected the value of the captives as individuals in their community rather than as a commodity; the community receiving its members back in return, see de Souza (1999), 65. 169 Thuc. 3.70. The sum mentioned seems improbably large—800 talents for about 250 men (even allowing for the fact that they were of the first rank)—and so the emendation Oª Œ  Æ, proposed as early as Valla (1452), may well be correct. 170 Gerolymatos (1986), 65–7, wrongly suggests that the proxenoi here are Corinthian proxenoi at Corcyra—rather than the Corcyrean proxenoi at Corinth, on whom Corcyreans would naturally rely for aid.

The Anatomy of an Ancient Institution

73

for, the level of their own ransom at the hands of the Epirotes.171 We also have numerous references to honorands in proxeny decrees having freed, released, and ransomed captives taken in war and Hyperides’ parody of a proposal to confer proxenia makes reference to the fact that, among Euthykrates’ various deeds, rendering him unsuitable for this nomination, was his failure in the aftermath of the Athenian defeat at Chaironeia to either bury any Athenian dead or ransom any Athenian prisoners—suggesting that these were some of the archetypal deeds of the proxenos.172 In parallel with the sort of aid proxenoi were expected to give to official delegations in accessing the political institutions at their poleis, proxenoi probably also gave similar support to private citizens. Although it seems likely that this was among the most important of the functions which proxenoi performed, it is poorly attested in our evidence because such aid was resolutely ‘private’ in character. It is also difficult to reconstruct precisely what would have been involved because of the complexity of the situation in the world of the Greek poleis in relation to access to judicial institutions for non-citizens. In particular, while it is unlikely that non-citizens were, as a rule, completely barred from access to the judicial institutions of other cities,173 our evidence does reveal that there was an extremely wide range of different judicial statuses that might apply to a non-citizen at a particular polis. For non-citizens, access to particular institutions, and the terms which mediated that access, could depend on what rights if any they had been granted, either themselves, or through their family or city (which could take the form of an individual honorific grant or an interstate legal treaty), as well as their polis of residence, gender, and, also, the nature of the legal dispute in 171

Plb. 21.26 with Walbank, HCP ad loc. For Euthykrates, see n. 15 (Hyp. fr. 76). Ransoming: IG XII 7 5 (Arkesine, Amorgos, c.357/6 bc); XII 8 3 (Athenian dēmos on Myrina, 350s bc); II3 430 (= II2 283, c.337 bc) and, perhaps, II3 390 (= II2 284, Athens, 350–40 bc); Daux BCH 52 (1928) 46–50 (= I.Lampsakos 6, c.320 bc; given the reference to progonoi in l. 10, a reference to the honorand’s current status as proxenos should be restored in l. 6, e.g. [æ   B ºø  æå]ø); IG IV 756 (Troizen, 197–195? bc). On all these, see Bielman (1994). Burial: Anth. Pal. 13.12 (an epigram written by Hegesippos in the third century bc for an Abderite shipwrecked on Seriphos cremated and sent home in a bronze urn by his city’s proxenoi there); Herodotus also notes a cenotaph erected for the Aiginetans at Plataia by their proxenos there, Hdt. 9.85. 173 So Gauthier (1982), 103–4; Bresson (2007–8), vol. 2, 106–18. Contra e.g. Whitehead (1977), 97; Cassayre (2010 ), 185–93. 172

74

Proxeny and Polis

question.174 However, regardless of what precise judicial standing particular non-citizens possessed, in virtually all of these cases they had a need which proxenoi would have been well-placed to service: the need to formally establish their identity. The reality of this need is illustrated by the institution of prostatai in relation to resident aliens, metics, at Athens and elsewhere. It has been a source of bemusement that although the Athenian metic was obliged, on pain of enslavement, to officially register a prostatēs, we have no record of a metic ever needing, after this registration, to formally act through one.175 Enrolment, however, was probably the point. The prostatēs was ‘like a guarantor’ (enguētēs);176 he was not responsible for anything the metic might do, but rather he was the formal guarantor of his identity as a free individual, and served, in a similar manner to the proxenos, as a link between the non-citizen and the citizen body.177 The identity of the metic was thus linked to his citizen prostatēs, which is why, according to Isocrates, ‘[we Athenians] judge metics by the prostatai they choose’ and presumably why, in one of Herodas’ mimes, a metic speaking in a law case at Kos against another metic, makes much of their respective prostatai.178 The prostatēs was not necessarily needed to speak on behalf of his metic in law, but he might be called on to bear witness to his identity.179 The problem of identifying particular non-citizen individuals is probably also at issue in the named citizen enguoi linked 174 Thus for Athens, for example, an individual from another city could be a xenos, metoikos, isotelēs, proxenos, euergetēs, or even possess the rights of a citizen, or have rights which derived from their polis of origin (for example, access to the court of the Polemarch, granted to the Phaselites by the Athenians in the fifth century, IG I3 10); conversely access to and standing in the Athenian dikai emporikai, was determined not by citizenship, nationality, or honorific standing, but the nature of the dispute (in this case ‘litigation arising from a written contract providing for transportation of goods to or from Athens’, Cohen 1973, 56–9). 175 Harrison (1968–71), vol. 1, 189–93; Whitehead (1977), 90–2; Todd (1993), 197–8; Meyer (2010), 72. 176 u æ Kªªı Å Z Æ, Lex. Rhet. 201 p. 189, n. 4. 177 This function probably explains why metics who could not produce a prostatēs were enslaved—there was no formal guarantee of their free status. On this penalty, see Harrison (1968–71), vol. 1, 165. 178 ŒÆd f b  Œ ı Ø  ı r ÆØ  Ç , ¥ ı æ i f æ Æ ø Ø (Isoc. 8.53; trans. Whitehead 1977, 52); Herod. 2.10–15. Aristotle treats this requirement for a prostatēs as a widespread condition for metics, 1275a12–14. 179 This is presumably the reason why Aristotle states that metics required a prostatēs in order to go to law (u  ŒÆd ŒÅ åØ ŒÆE ØŒ Ç ŁÆØ . . . Ø I ªŒÅ æ Å, Pol. 1275a7–14). Perhaps this took place during the initial

The Anatomy of an Ancient Institution

75

with grants of proxenia in the decrees of the Aitolians and other western Greek communities. It was these individuals who would be called on to confirm the identity of a particular proxenos (which was eminently practical given how many grants survive).180 The close connection between the role of proxenos and prostatēs is stressed in our lexicographical sources, in which the proxenos is very frequently defined as the prostatēs of a polis.181 We do not know whether a proxenos could initiate a legal case on behalf of a noncitizen who did not formally possess this privilege (though there may have been ways of working round any obstacles that did exist).182 It is, however, clear that it would always have been useful to have an official conduit for the formal identification of individuals—a proxenos who might provide aid in a specific legal case, but, as a politically prominent individual, could also put pressure on a magistrate to recognize the individual he represented and observe any local legal rights or privileges this individual might possess. This was a function that any citizen could potentially fulfil, it was not peculiar to the proxenos, but not all visitors will have had sufficiently close private connections with prominent citizens to be able to induce them to perform these acts on their behalf. By contrast many more of them will have had access to proxenoi of their polis to help them in these circumstances, and it seems likely that this role came to be particularly closely associated with proxenoi. This need is probably also illustrated by a list of at least fourteen individuals named proxenoi at the same time by Tenos, of which at least nine represent different poleis on Crete.183 This list thus seems to anakrisis, which might explain why we do not otherwise hear of it, Harrison (1968– 71), vol.1, 192–3. 180 On the Aitolian enguoi, see Chapter 2, pp. 102–3. The overlap between or even homonymity of the terms enguos and prostatēs is clearly evident in the lists of debtors at Thespiai from the second half of the fourth century bc in which both terms are used to refer to guarantors (prostatai: I.Thespiai 44, 48, 56; enguoi: 50–5, 57; both are used together in 47). 181 This equation is made explicit in the Hellenistic period in Aristophanes of Byzantium’s definition: b ŒÆº F ÆØ ƒ ŒÆ a ªÆ  ºØ ØŒe æ ÆØ ºø ‹ºø ªØ Ø . . . (Ar. Byz., Slater 1986, fr. 301); cf. Hesych. sv ‘æ  Ø’. See Meier (1843), 2–3 n. 7 for other examples. On this connection between prostasia and proxenia, see Monceaux (1886), 12–15; Walbank (1978), 4. 182 See Section 1.2.6, one possible explanation of the actions of Kallippos in [Dem.] 52. 183 IG XII suppl. 304, cf. also XII suppl. 313 (Appendix no. 15.2 and 15.1, both second half of the third century bc).

76

Proxeny and Polis

reflect a conscious attempt to recruit proxenoi and to obtain wide coverage for Tenos’ network of proxenoi on Crete (note especially that only one individual from each polis is represented). The context for this was surely the contemporary Tenian embassy successfully sent to Crete to acquire asylia from its various cities184 and the aim was presumably to enable Tenian citizens, supported by these local intermediaries, to make use of this right of freedom from seizure. The implication is that without these official intermediaries it would have been difficult for Tenians to make effective use of the rights they possessed—both in proving their entitlement and identity, and approaching (not to mention persuading) the relevant magistrates to implement it.185 The issue of identification would also have been of central importance in relations between private individuals as well. Although our sources are meagre, we might expect that proxenoi were regularly called on to formally identify non-citizens when private contracts were struck. Certainly it seems likely that, in parallel with the role they played in arranging the ransom of prisoners, proxenoi also played an important role in facilitating access to credit by noncitizens, either by making loans themselves (probably involving smaller sums in emergencies, as in the case of the ephodia—money to get home—which proxenoi are sometimes attested as providing to captured citizens of the granting polis),186 or by standing surety in relation to loans for larger sums (as Corcyrean proxenoi at Corinth were claimed to have done).187 If proxenoi acted very regularly in these capacities it would help to explain the analogical extension of this term to contexts in which it is often translated as ‘witness’ or ‘guarantor’.188

This context was first proposed in Graindor’s ed. pr. of this text, MB 14 (1910), 233–6 n. 25. The asylia campaign of the third century is principally known from inscriptions created when renewals were sought in the first half of the second century, Rigsby (1996), 154–63, but note the similar prominence of Cretans here (no. 56–60, most of the surviving inscriptions). cf. Brulé (1978), 71; noted by Marek and placed in the context of other grants to Cretans, (1984), 315–16. 185 See also IG XII 3 254—a proxeny decree of Anaphe for a Cretan from Lyttos inscribed on the same stone as, and perhaps connected to, a grant of asylia made by the Cretan koinon, despite the difference of hand; see Robert, OMS 2, 1040–4. 186 This is one of the services performed by Eumaridas when he was already proxenos, IG II3 1137 l. 10 (= II2 844) 187 188 Thuc. 3.70. See p. 72 and n. 169. See n. 204. 184

The Anatomy of an Ancient Institution

77

1.2.6. Apollodoros’ speech against Kallippos A common theme connects the different sorts of service attested in this source material that were expected of proxenoi. Collectively they amounted to an intermediary function—of facilitating interaction between communities, which was fundamental to the conception of proxenia. Unfortunately, many of these sources are relatively brief and disconnected. However, one literary source, Apollodoros’ speech against the proxenos Kallippos, provides a more detailed and contextrich account of the activities of different proxenoi and merits a more extended interpretation.189 The background to this speech is that Apollodoros is being prosecuted by Kallippos, proxenos of the Herakleians, concerning a sum of money which Lykon, a citizen of Herakleia, had left on deposit at the bank of Apollodoros’ father, Pasion.190 The dispute was over what should have happened to the money following the death of Lykon. The usual difficulties pertain in using adversarial oratory historically. Apollodoros’ speech in fact represents an attempted character-assassination of Kallippos as a proxenos. However, careful analysis makes it possible to identify a common conception of proxenia—the proxenos-paradigm—which this speech presupposes was shared by Apollodorus’ audience (we do not have Kallippos’ speech, but he would have made much of this status and certainly Apollodoros in this speech expects him to). Immediately after the death of Lykon (who was fatally wounded by pirates), Kallippos first came to Pasion’s bank. Apollodoros, Pasion’s son, purports to reproduce the exchange between Kallippos and Phormion, the slave who managed Pasion’s bank: æå ÆØ Kd c æ ÇÆ ˚ ººØ   d PŁf Kæø H, ¸ŒøÆ " ˙ æÆŒº Å N ªØª Œ Ø. I ŒæØÆ ı b  æø  ı ıd ‹ Ø ªØª Œ Ø, ‘pæÆ ŒÆd KåæB E’; çÅ ›  æø· ‘Iººa æe  Kæø fi A’; ‘æe ’; çÅ· ‘Kª

Ø KæH. KŒE  b  º ÅŒ, Kªg b æ H ıªå ø H " ˙ æÆŒºø H. I ØH   E Æ  Ø a

189

[Dem.] 52. As Kallippos is linked with Isocrates’ circle (14), which had strong connections with the Black Sea region (Moreno 2007, esp. 175–6), it seems most likely that this was Herakleia Pontika; however, on the difficulties of ambiguous ethnics see Fraser (2009), 179–91. On Apollodoros, son of Pasion, see Trevett (1992). 190

78

Proxeny and Polis

ªæ Æ Æ, ¥ ’ NH Y Ø ŒÆ ƺº Ø IæªæØ · K I ªŒÅ ª æ   K Ø ±  ø " ˙æÆŒºø H KغE ŁÆØ.’ This man Kallippos came to the bank, and asked whether they knew Lykon, the Herakleian. Phormion, who is here present, answered that they knew him. ‘Was he a customer of yours?’ Phormion replied that he was, ‘but why do you ask?’ ‘Why? I will tell you. He is dead and, as it happens, I am proxenos of all the Herakleians. I demand therefore that you show me your books, that I may know whether he has left any money; for I must of necessity look after the affairs of all the men of Herakleia.’191

Kallippos’ description of his moral duty as proxenos of the Herakleians, to take care (epimeleia) of all Herakleians, directly corresponds to the stereotypical descriptions of proxenoi discussed above. The underlying assumption that this was how a proxenos should represent his role, and that this was an appropriate way in which this care should be expressed, chasing up the affairs of a dead citizen of the polis he served, was taken for granted. Phormion, moreover, is represented as accepting without demur the right of Kallippos as proxenos to look into the affairs of Lykon at Pasion’s bank. Kallippos is then said to have been shown a note from Lykon instructing the banker that the deposit should be paid to his business partner, another non-citizen (Kephisiades, an inhabitant of Scyros) whom Pasion must rely on Lykon’s Athenian friend, Archebiades, to identify (note again the issue of identifying non-citizens).192 Kallippos, according to this narrative, returned five months later and, after informing Pasion that Lykon had died without sons, instructed him to disregard Kephisiades’ claim and give him, as proxenos, the sum. Pasion, according to Apollodoros, expressed willingness to do this so long as Lykon’s former Athenian friends, and Kephisiades himself, consented. In the event, Lykon’s friends refused to cooperate and Kallippos later filed suit against Pasion, suing him for damages on the basis that Pasion had paid out Lykon’s deposit to Kephisiades despite having promised Kallippos he would not. As ever it is perilous to accept uncritically the version of events, and motivations, presented by only one forensic speech and, as Kallippos’ speech has not survived, we do not know how large the gulf between 191

[Dem.] 52.5; Loeb trans., adapted. Gauthier (1972), 80–1, on this passage and 76–85 on the issue of identification in general; cf. Lewis (1996), 80–2. 192

The Anatomy of an Ancient Institution

79

the two accounts was.193 We should note, however, that Kallippos had learned of Lykon’s death and its circumstances before Pasion, presumably from other citizens of Heraklea with whom he associated. He also somehow learned not only that Lykon was a customer of Pasion (he asks Pasion for confirmation of this information) but also was later able to find out that Lykon had no sons to inherit. Both facts clearly show how a proxenos could access, and actively seek, information relating to citizens of the city with which he was affiliated. Within a given polis, the proxenos, probably his house, represented a hub about which citizens from a particular city might gather even if they did not necessarily require any particular service of him, both to meet with fellow citizens and as a clearing house for news of the city in question and its citizens, gathered from them as they passed through the city of the proxenos. The use Kallippos made of this information is also important. I would suggest that in the first instance Kallippos’ aim was probably to ensure that Lykon’s money, deposited at Athens, would be inherited by any heirs he might have. This would have been the reason Kallippos sought information about Lykon’s heirs, and how he discovered that Lykon had, as Apollodoros mentions, no sons ([Dem.] 52.9). This, at any rate, is surely how he represented his actions to the Athenian audience. Indeed the function of Apollodoros’ ridiculous caricature of Kallippos the proxenos, puffed up with self-importance and pomposity, was to undermine the defence that he expected Kallippos to make of his actions, as being in accordance with the moral obligations of his position. According to Apollodoros, Kallippos was quite clear about the fact that he was seeking the money for himself, but Kallippos may well have been acting on behalf of other individuals from Herakleia.194 He may well have argued that, although Lykon meant his business partner to have access to the money while he was alive, now he was dead this property should really go to his compatriots (someone at Herakleia surely inherited whatever property Lykon had left there). If this is correct—or was at least how Kallippos justified his own actions—this would be evidence

193

Even if this were not a general rule, Apollodoros’ plea that the jurors should trust the bare facts he presents over the rhetoric he ascribes to Kallippos should give us pause (52.1f.). 194 See, similarly, Welles (1970), 808.

80

Proxeny and Polis

that the proxenos might pursue even long-running cases directly on behalf of non-citizens. There is also another potential example of this sort of service. In Kallippos’ speech to Pasion it is reported that Lykon’s boat was attacked by pirates as he rounded the Gulf of Argos and as a result he was wounded, and he and his cargo were taken to Argos.195 At Argos he went to the house of Strammenos, proxenos of the Herakleians at Argos. On his death, Lykon gave to his host, presumably out of gratitude for his care, the property that was brought in with him.196 As well as being a good example of the sort of aid an individual might receive from his proxenos, this story also poses a difficult question: how was Lykon able to give to his proxenos the property which presumably the pirates held? It is possible that Strammenos, as well as looking after Lykon, managed to regain for him his property, either through the courts or with the help of magistrates—a very useful service for a merchant to be able to call on. Apollodoros, in attempting to demonstrate that Lykon did not make use of Kallippos and thus minimize the importance of the relationship implied by his proxeny, also lists the ways in which a proxenos could be expected to aid citizens from the city which granted him this status.197 Apollodoros alleges that Lykon did not call on Kallippos for aid when he went to law with two Athenians over an aborted business venture he had lent money for—but rather looked to his two Athenian friends, one of whom mediated a settlement in the case.198 He further points out that if Lykon had wanted to give Kallippos the money, he could have deposited it with him from the first, in the certainty that he could recover it from his friend and proxenos on his return,199 and that, moreover, Lykon never even stayed with Kallippos. The impression which Appollodorus thus seeks to create may or may not be correct. Certainly it was not the only side of the story, as Apollodoros states that the only evidence of Lykon’s attachment to himself which Kallippos has not claimed is

195

196 [Dem.] 52.5. [Dem.] 52.9. [Dem.] 52.20. 198 The verb used here, ‘ÆæÆŒ º ’, is significant—it is exactly the same verb used in decrees of proxenia to describe the action of individuals who call on the honorand for aid, which he provides. 199 This argument conveniently overlooks the fact that this sum was at Pasion’s bank in the first place as the balance of their transactions, [Dem.] 52.2. 197

The Anatomy of an Ancient Institution

81

that Lykon stayed with him.200 What is clear, however, is that these sorts of intermediary functions could be performed by individuals other than proxenoi. As a general possibility this rings true. I would note, however, that while at Athens, where Lykon frequently did business, he may have developed and used a network of private friends, at Argos, where the same was probably not true (and where he had not intended to go), wounded and robbed by pirates of his merchandise, it was to the house of his city’s proxenos that he went and received aid.

1.3. THE PR OXENOS-PARADIGM IN THE LONGUE DURÉE The formulaic descriptions of honorands who were named proxenoi reveal how proxenia was conceived both at the level of the community, which judged the merits of particular individuals for this status against the paradigm, and also by individuals, including nominators and would-be nominees. From comparison of these descriptions and the different types of service which they are attested as being expected to perform, we can thus reconstruct a proxenos-paradigm which underpinned and gave meaning to the institution of proxenia. This analysis clearly shows that the same proxenos-paradigm defined proxenia at widely separated poleis. Consideration of these descriptions, however, can also allow us to go further and resolve the longrunning scholarly controversy concerning when and if proxenia changed.

1.3.1. The longevity of the proxenos-paradigm Many of the proxeny decrees considered so far as examples in relation to the proxenos-paradigm derive from the fourth to early second centuries bc, as this is the period from which the bulk of surviving proxeny decrees date. The essential elements of this paradigm are, however, clearly expressed in some of the earliest surviving decrees: those inscribed at Athens in the second half of the fifth century. In 200

[Dem.] 52.22.

82

Proxeny and Polis

one decree for two brothers from Delphi and their father, the honorands are named proxenoi for ‘doing good (for the Athenians), in word and deed, in whatever way they can’ (ŁÆ  P[æª Ð Æ ŒÆd] ºª Ø ŒÆd æ[ª Ø ‹ Ø i ıÆ ]e ’Ø).201 Brief as it is, this captures the key element_ of benefaction and utility central to proxenia, and also the different ways in which these actions were performed by proxenoi—‘in word’ (particularly advice and political speech) and ‘deed’. Similarly, an Athenian decree dated to 421/0 bc, evokes all the central elements of the proxenos-paradigm, in praising Asteas the Aleian ‘because he does good for the Athenians, both privately and publicly, for (any Athenian) who comes (i.e. to Aleia), both now and formerly’ (h Ø s  E ŁÆ  ŒÆd NÆØ ŒÆd  ÆØ e IçØŒ  _ 202 Though very concise, this clearly ŒÆd F ŒÆd K ÐØ æ Ł åæ Ø). communicates an understanding of the role of proxenos as someone who habitually performs services on behalf of all Athenians because of their shared citizen identity regardless of whether they were acting for the community as a whole or as private individuals. These texts reflect an early stage in the development of the practice of inscribing proxeny decrees on stone.203 Nonetheless, they clearly show that the proxenos-paradigm itself was already fully developed, and attestations of proxenoi, both epigraphic and literary, from earlier in the fifth

201 IG I3 27 ll. 11–13; Walbank (1978) no. 13. The dating of the fifth-century series of Athenian proxeny decrees has been controversial, but it is probably true that the inscription of proxeny decrees on stone began or at least only became common at Athens in the 430s bc, see Papazarkadas (2009), 79. This text has recently been downdated to c.430 from 450/49 bc by Mattingly (2007) on compelling grounds. 202 IG I3 80 ll. 9–13; Walbank (1978) no. 49. 203 Outside Athens, very few decrees of proxenia antedate the beginning of the fourth century bc. The earliest of these took the form of highly abbreviated decrees inscribed on bronze plaques of which examples come from Argos (SEG 13 239, a decree for a citizen of the perioikic town of Oios attributed to the first half of the fifth century bc), Lousoi (Appendix no. 12, a disc inscribed with a series of three entries), Pherai in Thessaly (SEG 13 415–16, the earliest in a longer sequence of proxeny decrees, c.450–425 bc), Elis (Minon 2007, no. 24, c.425 bc), and perhaps a newly discovered bronze plaque, now in the museum at Thebes, inscribed with a decree of the Boiotian koinon (we must await the ed. pr. by V. Aravantinos, forthcoming in the proceedings of the Berkeley Boeotia Symposium, ed. N. Papazarkadas). The inscription of proxeny decrees on stone, even at Athens, seems to have been a later phenomenon (for the non-Athenian examples, from Iasos, Rhodes, Lindos, and Eretria, see Chapter 4, n. 121). Decrees, perhaps proxeny decrees, were certainly inscribed on bronze at Athens earlier, see Chapter 2, n. 12.

The Anatomy of an Ancient Institution

83

century and before, suggest that this was far from being a recent phenomenon.204 How long this proxenos-paradigm continued to dictate the use of proxenia is a controversial question. According to the most influential scholarly narratives, the institution of proxenia had lost much of its functional significance by the early Hellenistic period and declined to an empty honour—something held to be evident in the massive inflation of the number of decrees preserved. In these narratives the swift reduction in the number of proxeny decrees known from the mid-second century bc onwards was identified as the last stage of this process, of its final failure as an honour.205 These interpretations can be challenged on a number of different grounds. Detailed consideration of the quantitative trends in the inscription of proxeny decrees makes it clear that the relationship between the number of proxeny decrees inscribed and the number of grants of proxenia that were actually made is much more complicated than this implies (see Chapter 5). The synchronic study presented earlier in this chapter also clearly shows that such narratives of decline and obsolescence are

204

Some of the oldest epigraphic material is difficult to interpret and attests to proxenoi who appear at odds with the proxenos-paradigm. On the basis of this and other literary material, Gauthier (1972), 57–9 identified two distinct meanings of the noun proxenos and an associated verb (proxeneō), which he argued were both functionally distinct and separate etymological developments: proxenos (followed by the gentive case) as patron or protector (prostatēs) of foreigners versus proxenos as trusted intermediary between two strangers (with a verbal construction ‘to introduce’ followed by the accusative and dative). This use of proxenos by the same communities in completely separate senses is difficult to accept, however, and unnecessary, as recent reinterpretation of some of the early material has shown (ZelnickAbramovitz 2004, esp. 103–6). In fact uses of the noun and verb which can be translated as ‘introduce’, ‘procure’, ‘witness’, or ‘guarantee’ (see LSJ) are probably best understood as analogical extensions from the institution of proxenia, deriving from the functions proxenoi performed, see pp. 74–7. The proxenoi referred to in the fourth-century author Aineas Tacticus—the citizen guarantors of particular mercenaries stationed at a polis—are probably another such extension (Aen. Tact. 22.29; for discussion of this passage, see Mataranga 1993). The proxenos as a foreign friend and benefactor of the community as a whole is relatively well attested in our earliest material, in the very earliest attestation of this word, on the inscribed mnēma of Menekrates from Oianathea, proxenos of Corcyra (IG IX 4 882, 625–600 bc (= ML 4): ‘for he was the dear (philos) proxenos of the damos’) and also in the two stone seats at Elis, inscribed in the sixth and fifth centuries as the property of individual Spartan proxenoi, LSAG 199.15 and SEG 26 476 (= Nomima 37–8). For the early literary material, see Wallace (1970). On the attestations of proxeny in the poetry of Pindar, see Chapter 2, pp. 106–7. 205 See Introduction, Section I.1.

84

Proxeny and Polis

without foundation in relation to material from the rest of the Classical period and the Hellenistic period to the end of the first half of the second century bc. In a similar way, close consideration of the descriptions of honorands in later decrees, and the more specific examples of proxenos-appropriate behaviour which they give, makes it clear that the same proxenos-paradigm continued to underpin proxenia, and thus that proxenia continued to be understood and used in more or less the same way, throughout the late Hellenistic period, at least until the end of the first century bc. In the second half of the second century bc a set of four decrees inscribed at Thasos c.130 bc honouring two brothers, Dionysodoros and Hestiaios, clearly show that the same proxenos-paradigm informed the use of this institution then.206 These texts were part of two monuments erected in the southern part of the Thasian agora, which each otherwise comprised a statue of the brothers, set in the middle of a semi-circular stone bench (three decrees for Dionysodoros, two, from Lampsakos and Assos, on stele at either end of the bench, one, from Rhodes, probably inscribed on the statue base; and one decree from Samothrace, on a stele, for Hestiaios). All four texts, using local formulae specific to the polis in question, reference the central characteristics of the proxenos-paradigm: describing Dionysodoros as ‘well disposed to the demos and useful to the polis on all occasions’ (h ı  æåø HØ []øØ ŒÆd ª  håæÅ  BØ [ºØ] Kª ŒÆØæ E IƪŒÆ Ø),207 or as ‘making himself zealous and eager in the affairs of the demos, and rendering assistance to those engaged in private business’ (KŒ B ŒÆd æŁı  Æı e N a F  ı ÆæÆ Œı ÇØ æ ªÆ Æ, ıÆ غÆ  Æ[Ø] b ŒÆd E ŒÆ ’ NÆ Æı F  Ø ı  åæÆ).208 Similarly a third decree, probably from Samothrace, justifies a grant of proxenia to Hestiaios on the basis that he, as ‘a friend of the polis and reverent to the gods, lives his

206 These are I.Thasos 169–72. Recent work on the letter from Rhodes for Dionysodoros has increasingly singled out the late second century bc as the most plausible context for the Roman magistrate, Lucius Aurelius, and activities mentioned in I.Thasos 172, see Kallet-Marx (1995), 232 n. 25; Kreiler (2010); Hamon BE 2011 no. 476. If the reconstruction of this monument followed here is correct, and the orthostat block on which the Rhodian decree was engraved was from the statue base for Dionysodoros, then this text would be a subsequent addition to the monument in his honour, and thus the latest of these decrees. On this monument in general, see Hamon (forthcoming). 207 208 I.Thasos 170 ll. 4–6 (Assos). I.Thasos 171 ll. 14–16 (Lampsakos).

The Anatomy of an Ancient Institution

85

life by making himself useful and zealous at every opportunity towards those of the citizens who come to meet him, assisting them in furthering both their own interests and those of the community’ (çº  K d B ºø ŒÆd æe f Ł f P H åø ØÆ ºE E  ÆæÆ ıªå  ı Ø H  ºØ H håæÅ  Æı e ŒÆd çغ Ø  K Æ d ŒÆØæHØ Ææå ÆØ, ıæ ø a ıçæ  Æ ŒÆd ŒÆ a Œ Øe ŒÆd ŒÆ ’ NÆ Œ øØ).209 The text from Rhodes, again honouring Dionysodoros, makes the connection between the practical role delineated by the proxenos-paradigm and the status of proxenos even more explicit, in describing how Dionysodoros, having received a grant of proxenia ([NºÅçg] b ŒÆd Ææ’ ±E æ Æ; i.e. because he was proxenos),210 rendered assistance of various sorts to a Rhodian embassy en route to the Roman general Lucius Aurelius. The number and range of different communities represented in the decrees at Thasos for individuals from a single family clearly show that the same understanding of proxenia was still prevalent in the Aegean. For the first century bc, despite a striking reduction in the number of decrees inscribed by granting communities (see Chapter 5, figs. 1 and 2), we nonetheless possess a number of decrees which reveal the continuing association of the grant of proxenia with a description of the honorand using phrases typical of the traditional proxenos-paradigm. The fact that these texts make reference to the performance of specific sorts of proxenos service makes it clear that we are not simply dealing here with fossilized archaic language that has become devoid of practical function. From the middle of the first century bc, in particular, two decrees from widely geographically spaced communities both attest to the continuing close connection between the grant of proxenia and the proxenos-paradigm. The first, from Thaumakoi in Thessaly, dating from c.50 bc, explains a grant of proxeny for its honorand, Androsthenes of Gyrton, on the basis that ‘since his earliest youth he has lived his life always being the cause of some good for the polis of the Thaumakoi and for those of its citizens who meet him, lacking no eagerness or zeal on behalf of the interests of the polis of the Thaumakoi’ (‘Ie [ B ]æ Å ºØŒÆ ØÆ ºE I Ø  IªÆŁ F ÆæÆ Ø  ªØ  [ B fi _ ºØ] B fi ¨Æı[ÆŒ]H ŒÆd E K ıå  ı Ø _ 209

I.Thasos 169 ll. 4–9.

210

I.Thasos 172 ll. 5–6.

86

Proxeny and Polis

ŒÆ ’ NÆ H  ºØ H  ıB ŒÆd çØ[º ØÆ] PŁb [Kºº]ø H ıçæ ø B fi ºØ B fi ¨ÆıÆŒH).211 The second comes from Odessos in the Black Sea and, as the prelude to a grant of proxenia, describes the honorand, Zoilos of Heraia, a general of the Thracian king Sardalos, and his actions in very familiar terms: Zoilos served the proxenos-role of intermediate and advocate for the Odessans at the king’s court212 and even went so far as to promise that in the future he would be even more zealous in pursuing the interests of the polis.213 A decree from Thouria in Messene, dating to the Augustan period, is even more forthcoming. It grants proxenia to a Damocharis the Spartan, the descendant of previous proxenoi of Thouria. It describes, specifically, how Damocharis acts as prostatēs and advocate for embassies from their polis, working to achieve what is in its interests.214 It then goes on to further describe how Damocharis made himself useful to private citizens who came to him, as a proxenos should, in speaking both on behalf of their requests (in the assembly?) and acting as their advocate in court cases.215 From the very end of the first century bc, a decree from Methana, in the Troizen, in honour of Lucius Licinnius Anteros from Corinth, completes this set, describing how he welcomes visiting magistrates and citizens from Methana and acts as an intermediary for those who have need of him.216 The fact that these inscriptions were actually inscribed, and at length, in a period in which this was done for few other proxeny decrees of even the briefest kind, suggests that they are unusual and 211

IG IX 2 219, ll. 3–6. Icæ ŒÆºe ŒIªÆŁe  æåø h ı ŒÆd æŁı  Æ e Ææå ÆØ æe a B _ ºØ H ŒÆ a Œ Øe j ŒÆ ’ ºø H æ ªÆ Æ ŒÆd E K ıªå  ı Ø Æ HØ H  _ NÆ åæÆ A Ø ıæªH çغ ø ŒÆd ıæ ªØ  KŒ H  ººa H _

ıçæ ø E ıŒÆ Æ Œı Ç ÆØ, IGBulg I2 43 ll. 6–13 (45/4–42 bc). 213 Kƪºº ÆØ b ŒÆd N e º Øe  ºf æ Łı æ  Æ e Ææ Æ ŁÆØ æe a B ºø H æ ªÆ Æ, IGBulg I2 43 ll. 18–21. 214 ÆæØ

[  ŒÆd] ÆæÆŒºÅ ø Øa Æ e E I

ºº  Ø[ Ææa] A º  ±H æ Çı ÆE ŒÆ æªÆ Ç  [Id a] ıçæ  Æ fi A ºØ, SEG 11 974 ll. 2–5. 215 E  ÆæƪØ  Ø  ’ ÆP e [ º Æ]Ø Ææå  håæÅ

 Æı e   ÆæÆŒº[  Ø] ŒÆd ŒæØ Åæø ÆæÆ  Ø ºªø ŒÆd æ ÆıŁ[ H ‰] bæ Nø  ºØ A, SEG 11 974 ll. 2–5; for the interpretation of the last clause ‘even on behalf of murderers’, see Cartledge and Spawforth (1989), 153–4. 216   Id ¼æå  Æ ŒÆd f ÆæØÆ F Æ   Æ H  ºØ H _  å  çغ çæø ŒÆ[d] e ŒÆŁBŒ   H A Ø ÆæÆŒºÅ ø , ‰ _ æ  K , E ÆP F åæÆ _å ı Ø, IG IV 853, _ll. 14–16. 212

The Anatomy of an Ancient Institution

87

relate to very specific contexts. For instance, the Thouriate decree for Damocharis the Spartan, with its emphasis on legal cases, seems to relate to the situation at Thouria after the battle of Actium, when it was deprived of its autonomy by Augustus and given to the Spartans who probably monopolized certain judicial functions.217 However, what is striking in all of these cases is the extent to which the same sorts of description of the individual honorands, and the same sorts of service and function, continued to be conceived of as central to the grant of proxenia. Collectively they strongly suggest that, at least until the end of the first century bc, proxenia continued to be defined by the familiar proxenos-paradigm of the Classical and earlier Hellenistic periods, that, in essentials it continued to be understood and used in the same way at different ends of the Mediterranean over some five hundred years, despite the significant changes which the Greek world as a whole underwent with the rise of the Hellenistic kings and Rome.

1.3.2. The stability of the proxenos-paradigm Theoretical models for understanding institutional growth and change, developed under the impetus of New Institutionalism, help to explain the mechanisms underlying proxeny’s remarkable stability, especially the concept of institutional isomorphism. This concept was developed by scholars working in Organizational Studies to explain the processes through which particular institutions and structures tended to be replicated throughout an organizational field—for instance how company hierarchies in a particular industry come to be more or less identical. DiMaggio and Powell, in particular, identified what they saw as three different sorts of institutional isomorphism: coercive (externally imposed); mimetic/competitive (copying the institutional innovations of organizations believed to be successful in order to get or make up a competitive advantage); and normative (adopting institutions as indicators of group membership).218 Although our evidence does not reveal the processes by which proxenia (and a shared proxenos-paradigm) came to be universal among Greek poleis because proxeny grants only began to be inscribed 217

So Cartledge and Spawforth (1989), 153–4; Fournier (2010), 496–7. DiMaggio and Powell (1983); for normative isomorphism, DiMaggio and Powell (1983) talk of ‘professionalization’ in relation to individuals, an analogous process. 218

88

Proxeny and Polis

relatively late in the history of proxeny and much of the early evidence is difficult to interpret, two of these forms of isomorphism, mimetic and normative, represent good ways of understanding the processes involved. Proxeny clearly did offer advantages to poleis that adopted it and, at least in time, it also came to be one of the most important ways in which cities asserted in relation to each other that they were poleis (see Chapter 4). New Institutionalist models explaining the inherent conservatism of institutions (‘path dependency’) and the ways in which innovation arises are also useful in understanding the development of proxenia in the long-term—or rather its remarkable resilience to change. In particular, the functions fulfilled by proxenia continued to be much needed in a Mediterranean world that remained politically fragmented, and this political fragmentation also made it difficult for significant innovation to arise or spread at the local level. This was because proxenia was an explicitly inter-polis institution, functioning only in relation to individuals outside the granting city, and it depended for its function on the fact that both polis and proxenos shared the same understanding of what it entailed. Local innovation which would have had any sort of effect on the fundamentals of this institutional relationship would thus have been precluded because it could not benefit the innovating polis. Moreover, the need for such innovation may also have been obviated by the potential flexibility of this institution, which could be used to honour atypical honorands and on occasion express different sorts of relationship by analogy. Perhaps the only form of change which would have been possible would have been alterations imposed from above (coercive isomorphism)—and during this period there was no single, central authority capable of imposing such changes universally, at least until the domination of Rome. Fundamental change was thus rendered unlikely but the degree of conservatism which we see was not the product simply of inertia. On the contrary, both the consistency across space and the stability across time of this institution were the result of a continuous effort, which is reflected in the constant recommunication of the proxenosparadigm attested in our proxeny decrees. In regularly making new grants of proxeny the citizens of poleis not only reminded themselves of the definition and content of this role—they also performed their awareness of it in front of an external audience of their peers. This

The Anatomy of an Ancient Institution

89

was a continuous process of normative isomorphism, of asserting a shared Greek polis-identity. The terms in which poleis described proxenoi made sense and were stable because they and the specific concepts which they referenced were shared, calqued on other shared, stable institutions like xenia and euergetism. But grants of proxenia did not just reflect and rely upon a broader Greek cultural homogeneity—they were one of the instruments of its continuous creation. Proxenia, of course, did eventually disappear (see Chapter 5). The reasons why and the way in which it did disappear illuminate the change the Greek world as a whole underwent at this point, which was more profound than the transition from Classical to Hellenistic, and this helps us to further understand the importance of the role which proxenia—and the proxenos-paradigm—played before.

2 Proxeny in its Political Contexts Competition and Prestige

Proxenia, throughout the five hundred years in which it flourished as an institution, was based on a shared understanding of the proxenos as the official friend of another city who performed certain sorts of intermediary function within the context of his own political community. The existence of this sort of abstract proxenos-paradigm, however, raises the question of how it worked in practice—of how, from the perspective of individual actors, honour and function interrelate. In particular, what motivated individuals within their own poleis, as the proposers of grants of proxenia for outsiders and as proxenoi in relation to external poleis, to participate in this institution? This chapter draws on a wide range of evidence illustrating proxenia in practice and shows that this diverse material can be structured into a coherent account of how proxeny, as an honorific institution, shaped interactions between individuals and between individuals and cities. One particular motivation recurs, despite the different political contexts and historical circumstances involved: the pursuit of prestige and elite status within a polis-based context. The three principal sections of this chapter examine, in turn, the main contexts in which proxenia operated. The first section explores how decrees of proxenia functioned in the context of the poleis in which they were made and how, as an institution, with its claims to construct universal civic ties, it functioned in the highly competitive and often factional political processes frequently attested. The second section considers why this relationship was actively sought, even initiated by individuals whose primary interests were in their own political communities, rather than the poleis which granted them

Proxeny in its Political Contexts

91

proxeny, considering the two structurally distinct cases of proxenoi active in poleis and in royal courts. The third section then concentrates on the position of the proxenos in relation to the external polis which honoured him and explores the function of the different honours and titles which were granted at the same time as proxenia in constructing and encouraging a particular sort of emotional relationship. In all of these different instances, proxenia represents an important case study for understanding the role which civic honours and honorific institutions played among different political communities in the ancient world, both their potential for and their limitations in shaping the actions of honorands in the context of competing personal interests and other institutional roles.

2.1. PROXENIA IN THE GRANTING POLIS Inscribed civic decrees tend to provide us with a very partial view of the decision-making processes of poleis. As a rule they suppress the role of political controversy and contestation which, our other sources reveal, was common both before, and even after, the decisions they represent were made.1 This was, in fact, one of the functions of the inscribed decree: to present the decision of the polis as authoritative, collective, and final, and to emphasize, retrospectively, the order, regularity, and tranquillity of the decision-making process.2 However, in order to understand how proxenia worked as an institution, it is necessary to examine the realities of politics in the poleis and the role played by competitive, factional cliques and individuals. In particular, it is important to ascertain how far internal competition coloured grants of proxenia—theoretically a universal civic tie—given that the ability to successfully propose honours could express personal prestige and both individual and factional influence. If Gabriel Herman is right and ties of personal friendship (especially xenia) usually preceded and motivated the proposal of proxenoi, does

1

Hansen (1974), 62. Thomas (1992), 71; Ma (2002), 20–1. Inscribed decrees gloss over the fact of victory—and defeat—(Loraux 2001, 22) and depoliticize (Osborne 1999, 343). The dedication of decrees to the gods was also one way of asserting their finality as communal decisions, see Bresson (2005), 163–6. 2

92

Proxeny and Polis

this contradict the assertions of the honorific decrees themselves that proxenoi were the friends of the polis as a whole and performed services for all its citizens?3 The central issue here is the relationship of the rhetoric of the proxeny decrees to reality, and the function, from the point of view of proposer and polis, of the decrees that were inscribed. To resolve these questions, it is necessary to ascertain how controversial proposed grants of proxenia were—how forcefully nominations of proxenoi were contested, and how frequently proxeny decrees were repealed, or their validity questioned, after they had been passed.

2.1.1. The proposal and contestation of grants Within Greek poleis, honorific grants were frequently contested because they gave communal validation to the actions of particular, prominent citizens, serving to mark out both the policy of a polis and the ascendancy of certain figures within it. Our evidence concerning the prosecution of individuals at Athens for having proposed unconstitutional decrees (graphē paranomōn) gives us a sense of how liable to contestation honorific decrees were, comparatively, even after they were passed. Of the thirty-three prosecutions attested, for which the content of the original proposal is known, no fewer than twenty (61 per cent) were made in relation to proposals which were honorific.4 Against this background it might not seem surprising if proxeny decrees also regularly served as focal points for similar expressions of bitter factionalism, not least because they could be read as signalling the foreign policy of the polis,5 and drew the political community as a whole into permanent relations of moral obligation with particular outsiders. The surviving Attic oratory yields evidence, in the form of certain recurrent rhetorical tropes, which illustrates the sort of opposition that proposed proxeny grants could meet. In the extant Athenian orations, descriptions of proxenoi as unsuitable or unworthy, or as having been appointed improperly, are relatively frequent. Thus Demosthenes is accused separately by both Deinarchus and Hyperides 3

Herman (1987), 138–42. In six further cases we know nothing of the decree being contested. See Hansen (1974), 62; Johnstone (1999), 39–40; Liddel (2007), 162–3. 5 For an example of this sort of reading of proxeny decrees, see Arnush (2000). 4

Proxeny in its Political Contexts

93

of having been bribed to propose grants of proxenia.6 Demosthenes himself states as a fact that proxenies were procured by bribery7 and that many who were unworthy—rogues (mastigioi) and, surely hyperbolically, slaves (douloi)—had also become proxenoi through the agency of prominent politicians.8 In at least one case we have a much fuller example of this rhetoric in Hyperides’ retrospective prosecution of Demades for proposing an unconstitutional proxeny decree for Euthykrates (a graphē paranomōn). In this, Hyperides parodies the conventional form and content of the proxeny proposal to show that Euthykrates had been made proxenos despite doing the opposite of what a proxenos should.9 It is not clear whether Hyperides publicly opposed this and other grants for Philip’s friends, using this sort of rhetoric when they were first proposed by Demades and others. These, however, are the terms in which proxeny grants are likely to have been opposed in general—with arguments that the individuals in question did not, in fact, fit the paradigm of the proxenos which, here, is given quasijudicial significance. However, evidence of rhetorical tropes used to oppose grants of proxenia before they were made does not prove that these grants were as potentially or frequently controversial as other sorts of decree. It is important that, of the twenty prosecutions known for illegally proposed honorific decrees, Hyperides’ prosecution of Demades is the only example we have in which a proxeny decree was at issue. From this dataset it is clear that it was civic honours for citizens which were most contentious before and after they were granted, constituting fifteen of these twenty instances (75 per cent).10 Moreover, excluding Hyperides’ prosecution of Demades, all of the prosecutions relating to honours for non-citizens concern the granting of citizenship and, in particular, the granting of citizenship to individuals who

6

Dein. 1.45; Hyp. 5.25 with Whitehead (2000), ad loc. Øa f ØŁF a ØÆFÆ ªæç Æ, Dem. 20.132. 8 ŒÆ Ø  ¼ººØ FºØ ŒÆd Æت ÆØ, ¸ıŒ Æ ŒÆd ˜Ø Ø . . . æ Ø  Ø

ººd Øa H ºØ ı ø ª ª ÆØ Ææ’ E ØFØ, z x  KØ › ¸ıŒ Æ, Dem. 20.131–2. 9 Hyp. fr. 76. This should now be read in the context of Hyperides’ assertion, in his newly recovered speech against Diondas, that Philip had secured Athenian honours for a number of his friends who had actively opposed Athenian interests previously, Carey et al. (2008), ll. 24–31. 10 Hansen (1974), 62. 7

94

Proxeny and Polis

were likely to settle at Athens, that is grants which, unlike proxeny decrees, would alter the composition of the Athenian citizen body.11 Our epigraphic material, from Athens and elsewhere, similarly suggests that grants of proxenia were rarely as potentially or actually controversial as honours for citizens or grants of citizenship. The evidence which we have for the subsequent repudiation of grants, especially in the form of attested destructions of the stelae on which proxeny grants were inscribed, suggests that grants of proxenia became sufficiently contentious to be repudiated only in very unusual circumstances—for reasons which related to the subsequent actions of individual proxenoi, or, exceptionally, in the context of political revolution.12 The only instances of destruction clearly attested—of a series of decrees at Athens and of a single stele at Priene—both took place in the context of revolution.13 In the case of the destruction of at least nine stelae inscribed with proxeny decrees at Athens by the Thirty Tyrants in 404/3 bc,14 it has been suggested that what we see is the violent, monumental

11

Hansen (1974), nos. 15 and 16 to individual exiles; 4 and 27 to metics. The only literary example of which I am aware is Arthmios of Zeleia, who regularly recurs in Attic oratory of the late fourth century bc as an example of a traitor punished by the Athenians (Dem. 9.41–3; 19.271; Aeschin. 3.258; Dein. 2.25). He is said to have been declared atimos and an enemy of the Athenians in a decree attributed both to Themistocles and Cimon for distributing Persian gold in the Peloponnese, but only Aeschines records that he was a proxenos of the Athenians. The authenticity of this decree, which Demosthenes claims was published on a bronze stele standing on the Acropolis, has been doubted (Habicht 1961, 23–7), but on insufficient grounds (Meiggs 1972, 508–12). On this decree, see further Wallace (1970), 200–2, and Moggi (1995), 147–8. Bases for triangular bronze stelae have been excavated in the Athenian Agora, Stroud (1963), 143 with pl. 54. In Mack (2014), I have argued that a consultation of the oracle at Dodona (SEG 56 663, 300–167 bc) probably concerns whether or not to rescind a proxenos’ status, presumably as a result of offences he committed. 13 Another instance is implied by an instruction inscribed at Paros that a number of Chians are to be inscribed as proxenoi as they were before (I ƪæłÆØ b ÆPe ŒÆŁ æ æ æ qÆ æ ı, IG XII 5 111 ll. 9–11, fourth century bc). It is tempting to relate this to another Parian text, XII 5 112 (fourth century bc) which, as restored by Hiller, records a public debt owned by Parians to Chian proxenoi. Wilhelm (1939), esp. 9–12, reconstructs a dispute and settlement under the Second Athenian League (RO 29, 372 bc), but caution is advisable, see Migeotte (1984), 213–15. 14 References to the destruction have been preserved in eight decrees from the early fourth century bc, some of which certainly are and others of which probably are reinscribed proxeny grants: IG II² 6; I3 227; II2 52; I³ 229; SEG 14 40 (= II2 66 fr. c); Agora 16 37, 39 and 45. 12

Proxeny in its Political Contexts

95

expression of a new foreign policy, cutting links with certain regimes and certain sorts of individuals.15 It seems likely, however, that oligarchic ideology also played a vital role. As the privileges which proxenoi received were to an extent modelled on the rights citizens possessed (such as enktēsis), it is hard to believe that this action was unrelated to the well-documented activites of the Thirty in restricting citizenship and civic participation, including the targeting of noncitizen residents.16 The example from Priene appears exceptional in other respects. The original inscribed decree of Euandros of Larisa was apparently destroyed during the reign of the tyrant Hieron at Priene in the early third century bc.17 I.Priene 12 is the re-inscription of this grant, preceded by the decree of the re-established demos which authorizes this, granting a further privilege with the addition of an ‘entrenchment clause’, by which any future attempt to rescind Euandros’ honours is declared invalid. In this case it seems very likely that this destruction related to the relationship of Euandros to those exiled under the regime of the tyrant. Certainly, he is significant for reasons other than his proxenia as the fact that an honorific decree was inscribed for him at Priene at all clearly shows.18 One reason for this reticence may have been that many inscribed decrees were usually dedications to gods, erected within particular sanctuaries, and thus protected by their power. However, other material strengthens the suggestion that proxeny was also inherently less controversial. Our other evidence indicates that even after a political revolution, the cancellation of proxenies was an unusual and highly marked act. Because honorific decrees explicitly recorded the existence of an obligation, rescinding a decree could be represented as defaulting on one’s debts.19 Although it was possible to 15 Shear (2011), 172. Honours were also apparently rescinded under the oligarchic government imposed on Athens in 322, but the examples, e.g. IG II3 378 (= II2 448, 322–317 bc), tend to concern citizenship primarily. 16 Xen. Hell. 2.3.18–20; Ath. Pol. 36.2, 37.2. See Gray (2011), 96–106, on the intellectual and ideological justifications underlying these attempts to more narrowly define the citizen body. 17 The original suggestion is Hiller’s and has been accepted as probable by subsequent commentators: Robert, OMS 3, 9–10; Crowther (1996), 205–16; Habicht (1998b), 84. 18 On inscription of proxeny decrees as an exceptional honour at Priene, see Introduction, pp. 15–16. 19 The close connection between financial debts and these debts of honour is emphasized by both Demosthenes, in his speech against Leptines (e.g. Dem. 20.36

96

Proxeny and Polis

argue that if a polis underwent sufficient constitutional change (especially involving widening or narrowing of participation) then the result was a new polis, free from obligations incurred by the government of the old, in practice, given continuity of place and population, this was unsatisfying.20 Indeed, the restored democracy in 403/2 bc famously undertook to repay even the debt the Thirty incurred for the Spartan garrison of Athens to safeguard their oligarchic regime.21 IG I3 98 paints a vivid picture of honorific continuity, despite the political vicissitudes of late fifth-century Athens. Pythophanes, already proxenos of Athens, was recognized by the oligarchic regime of the Four Hundred (or Five Thousand) which granted him additional honours including the inscription of the original decree granting him proxenia, presumably immediately above this decree on the stone. Subsequently, after democratic and then oligarchic rule, in 399/8 the democracy, once again in power, added yet another decree in his praise, which was inscribed below the oligarchic one.22 The almost complete absence of entrenchment clauses from the nearly three thousand inscribed proxeny decrees suggests that proxeny, more generally, was rarely subject to this degree of political, factional controversy. Entrenchment clauses were inserted by the proposer to pre-empt later attempts to rescind a decree, and they suggest that a specific piece of legislation was actually or potentially controversial or contested enough for such attempts at cancellation to be a possibility. Apart from the entrenchment clause contained in the decree of Euandros, exceptional for the reasons discussed above, very few of these clauses occur in contexts relating to proxenia.23 The instances of which I am aware—in two decrees of Thasos,24 a decree of Ilion for four Tenedian brothers,25 and the Athenian decree for Sthorys of Thasos of 394/3 bc26—occur in decrees granting additional, more controversial honours to pre-existing proxenoi and do

and 133–4), and in Antigonos Monophthalmos’ letter to the Teians regarding their projected synoicism with the Lebedians (where provisions regarding the continued recognition of each sort of public debt are juxtaposed, RC 3, ll. 19–24, 306–302 bc). 20 Arist. Pol. 1276a6–1276b13. 21 Ath. Pol. 40.3; Isoc. 7.67–9; Dem. 20.11–12. 22 ML 80 with commentary on the deeper-cut letters of the final decree. 23 As collected by Gschnitzer (1973), col. 705, with the addition of IG II2 17. 24 IG XII 8 267 (beginning of the third century bc); XII suppl. 358 + add. p. 217 (first part of the third century bc). 25 26 I.Ilion 24 (c.300 bc). IG II2 17.

Proxeny in its Political Contexts

97

not relate to the grant of proxenia itself.27 Other related sorts of protective clause suggest that individual proxeny proposals could on occasion be extremely controversial. For instance, a proxeny decree from Pherai contains a heavily (perhaps dubiously) restored clause threatening retribution if anyone should remove or steal the decree.28 Such indications are, however, very unusual and presumably have more to do with the particular circumstances in question than the fact that proxenia was being granted. The surviving lists of proxenoi similarly suggest that it was rare for cities to repudiate individual proxenoi, whether as a result of changing relations with them, or the ascendancy of a faction opposed to their appointment in the first place. In the extensive surviving fragments of proxeny catalogues and lists of proxeny grants, which preserve records of c.1,400 proxenoi in total, there is no sign that the names of particular proxenoi have been erased, as we might expect if grants of proxenia were regularly overturned.29 If the arguments which I have made elsewhere about the context for the erection of the Karthaian catalogue of proxenoi are correct—and it was created in the aftermath of the dissolution of the Keian federation, with the Karthaians keen to

27 The Thasian decrees granted citizenship. The decree of Ilion granted ateleia, the right to conduct sylē from Ilion, and also includes a fine for those who tried to collect taxes despite this grant (suggesting that not all would be pleased). The Athenian decree for Sthorys was probably controversial because it granted him citizenship; for similar controversies surrounding citizenship proposals in this period, see Ath. Pol. 40.2; [Plut.] X Or. 835f–836a. The controversy of this motion perhaps explains a further oddity of it, the fact that the main decree is introduced by a subsequent bouleutic decree which provides for the inscription of the original decree on two separate stelae (the first decree only explicitly allowed for one). Osborne (1981–3), 47, explained this away as a clarification, obtained by Sthorys from the boulē of an ambiguity in the original decree—an explanation which Gauthier found unconvincing (BE 1996, no. 126). I wonder whether the inscription ordered by the original decree might not have been delayed because the motion was controversial (a fact indicated by the entrenchment clause) as a result of which Sthorys brought the matter up with the council, which, out of embarrassment because of this delay, or perhaps out of a desire to emphasize the finality of this grant, reiterated Sthorys’ deserts and ordered that these decrees be inscribed twice. IC II v 35 (Axos, first century bc) is perhaps another example, but this fragmentary decree is very difficult to understand. 28 SEG 23 424 (fourth century bc). 29 The erasure in IG XII 5 542 l. 47 (Appendix no. 10.1), was made to correct an initial misinscription of the first element of a proxenos’ name. The sort of wholesale erasure of previous chronological lists of proxeny grants to make way for new ones that we see at Aitolia (of which the best example is IG IX 12 1 31), is an entirely different phenomenon, see Appendix no. 1.

98

Proxeny and Polis

cooperate with the Athenians—it is significant that this list included individuals who are likely to have received proxenia in the context of the earlier Keian revolt from Athens.30 The absence of erasures in the more than two thousand individually-inscribed proxeny decrees also provides important supporting evidence.31 This failure to cancel honours for outsiders should perhaps not surprise us. If Demosthenes is to be believed, the Chians did not cancel the extraordinary honours they had granted to the Athenian general Konon, even when he later led the armies of Athens against Chios.32 It does, however, contrast with our evidence for the fate of decrees for citizens within the polis, which were vulnerable both to later official repudiation and jealous private attacks—as we see in the speech for Euxitheos in which he accuses his rivals of chiselling out a decree passed in his honour by his fellow demesmen.33 It is striking that the only decree granting proxenia in which an honorand’s name appears to be pointedly erased is a decree of Eretria for foreign judges erected at their home city of Miletos—presumably because the dikastēs in question, in contrast to his fellow honorands, was subject to public disgrace.34 Thus, while there is evidence that nominations of proxenoi were subject to political opposition, it appears that, after proxeny grants were made, it was rare for them to become the focus of serious political controversy. There were, I would suggest, two reasons for this. The first is that honours for outsiders, although they could reflect the prominence or ascendancy of a particular politician or faction, were not as highly charged or potentially as politically involved as honours for citizens often were. The other reason is the plurality that was inherent in, and essential to, this system. Decrees of proxenia 30

Specifically the Knidian, grouped (uniquely) in the section otherwise devoted to Boiotians probably because he was nominated in the context of Epameinondas’ flotilla (the occasion for the Keian revolt), otherwise attested in a recently published Knidian decree of proxenia for Epameinondas himself; see Mack (2011), 328. 31 There are a couple of possibilities. The name, patronymic, and ethnic of the honorand of IosPE I2 21 (Olbia, fourth century bc) are said by the editors to have been chiselled from the stone, but it is striking that part of the name of the granting city of Olbia has also been lost. In the case of FD III 4 377 (Delphi, 346/5 bc) there is a deep erasure—but after the honorand’s name, which makes it look likely that a mistake was being removed. The closest we in fact come is IG XI 4 777 (Delos, early second century bc) from which the proposer and honorand were erased to facilitate the reuse of this stone for a new honorand (with which compare the reuse of stelae by the Aitolian League, n. 29 above). See also SEG 28 465, a similar example from fourth-century Thebes. 32 33 Dem. 20.81. Dem. 57.64; Liddel (2007), 162. 34 Milet I 3 154 l.14–15 (first half of the second century bc).

Proxeny in its Political Contexts

99

were not unimportant, and as we shall see shortly it seems likely that in some cases the ability to successfully propose individuals as proxenoi could serve as a marker of political influence and prestige. However, since grants were continually made, often to multiple individuals from a single polis, very few individual grants were likely to have seemed highly significant politically in themselves. Moreover, while it was for the assembly to decide whether an individual had performed a euergesia for its benefit, none disputed the moral imperative of responding with honours to benefactions that had been performed. The question of whether an individual had performed a benefaction might be subject to dispute, and the rhetoric identified above suggests that on occasion it was, but, when recognized, the decree of honours was an undisputed consequence.35

2.1.2. Prestige and plurality: proposers of proxenoi at Oropos The corpus of proxeny decrees from Oropos provides a unique case study for exploring the importance of political competition and prestige in relation to proposers’ nominations of proxenoi. We possess more than two hundred proxeny decrees from Oropos, the vast majority inscribed on a number of pre-existing statue bases during a fifty-year period (c.250–200 bc). This concentrated sample of relatively lengthy decrees, and the fact that we can understand the spatial relation of these texts, enables us to examine the ideology of proposers in ways that are not possible elsewhere.36 One of the most striking features of this material is the prominence of proposers of individual decrees—the name of the proposer of the 35 Dem. 20.57 on the inability of the polis to choose which benefactors it wanted to honour. The frequent formula in decrees from a wide range of poleis ‘so that the polis may be seen to honour with worthy thanks those who perform benefactions’, is also an expression of this moral obligation to recognize and be seen to recognize benefactions with honours. 36 See Rhodes with Lewis (1997), 491–7, for a comparative study of proposers in decrees. Often the formal proposal of decrees was entrusted to particular officials who would often not have been responsible for the initiative in the first instance; elsewhere the names of the proposers were frequently omitted entirely. The inscription of such a large number of detailed proxeny decrees at Oropos was facilitated by the decision, taken at some point in the mid-third century bc, that pre-existing statue bases could be used as an alternative to purpose-hewn stelae for inscription. On this, see Chapter 5, pp. 238–40, figs. 5.3–4.

100

Proxeny and Polis

proxeny decree is always included in the inscription of it. When, as often, dating formulae are omitted,37 the proposer’s name literally introduces the decree and is arguably given more prominence than the name of the proxenos honoured. It is hard to escape the conclusion that these inscriptions commemorate the successful proposer of the decree as well as the honorand, and this is further reinforced by the way in which these texts are distributed over the monumental topography of the Amphiareion. In particular, decrees, often from different years, were grouped on the same statue base according to the individual who proposed them. One particular statue base, for Ptoion and his wife (I.Oropos 418–19), received three decrees from three different years proposed by Oropodoros, son of Theozotes (I.Oropos 151–2; 174), and three decrees from two different years proposed by Kleomachos, son of Meilichios (I.Oropos 176; 186–7), and there are many other examples.38 Decrees proposed by members of a single family are also grouped in this way. On one statue base five decrees were inscribed proposed by two brothers prominent at Oropos, Python (I.Oropos 111; 114; 117) and Aristandros (I.Oropos 113; 115), the sons of Kalligeiton. These statue bases were used as monuments of the proposers of these decrees, communicating personal and familial prestige within public space. This is particularly clear in the juxtaposition on one statue base of three decrees proposed by two brothers, Hermodorus and Theoteles. All explicitly dated to the year in which their father served as the priest of the Amphiareion, these texts seem intended to mark the importance of this family in this year.39 However, for all that these inscriptions express claims to prestige, apparently pursued in a competitive context, what is also strongly evident is plurality. While there are one or two individuals and families who propose comparatively large numbers of decrees— such as, Oropodoros (5),40 Plutarchides (5),41 Sophilos and his sons

37 Perhaps up to half of the decrees of Oropos are undated, e.g. I.Oropos 113–15; 140–6. 38 Another is inscribed with three different decrees by Ariston, son of Nikostratos (I.Oropos 196; 204–5), and two different decrees, dated to different years, proposed by Paramythos, son of Kratinos (I.Oropos 184; 202); for further examples see I.Oropos 72; 73; 128. 39 On inscribed decrees as monuments for their successful proposers at Athens, see Liddel (2007), 163. 40 41 I.Oropos 69; 135; 151; 152; 174. I.Oropos 112; 156; 168; 169; 170.

Proxeny in its Political Contexts

101

(8),42 and, above all, Python (13)43—of the eighty-three proposers attested for this period more than half (47) propose only one decree. Stretched over a period of up to fifty years, at three surviving per year, it is still not clear that it was possible for just any Oropian to make a successful proposal and have his name inscribed.44 However, a relatively large proportion of citizens at the small polis of Oropos apparently could and, at least on occasion, did propose decrees that were inscribed.45 Pluralism is also evident in the prominence of honorands from particular poleis, above all the nearly fifty Athenians honoured with proxenia within this narrow period, up to a quarter of the overall total. The duplication of proxenoi at cities with which contacts were most frequent probably meant that the theoretical problem of individuals needing the help of a proxenos whose appointment they had opposed rarely arose in practice. It would also have had the effect of reducing the burden of services expected of any given proxenos at those cities. The potential for a polis to possess multiple proxenoi at particular cities, rather than a sign of the degeneration of this institution, thus seems likely to have been important for its function. The material from Oropos also sheds light on the role which private relationships played in the operation of proxenia as an institution. In one case, Kleomachos, the prominent son of Meilichios, proposed in the same year, probably on the same occasion, proxeny decrees for two Elateians, Gennaios and Kallias,46 and, on another occasion, was himself created proxenos by the polis of the Elateians in the year in which the same Gennaios was archon at Elateia.47 It seems reasonable to infer that a personal relationship existed between Gennaios and Kleomachos, perhaps it was even ritual xenia, which, Herman claims, was behind most grants of proxenia.48 Although this is the only case where such reciprocal grants of proxenia can be identified—and, indeed, in very few other cases can 42

I.Oropos 53; 68; 77; 140–3. I.Oropos 97–8; 103–5; 111; 114; 116–17; 160; 164; 192; 208. 44 This sample, of course, despite being very large, is not complete—though whether this is simply because some inscribed decrees have been lost or, as is also possible, not all proxeny decrees were inscribed, is impossible to know. On these issues, and for rates of granting, see Chapter 3, Sections 3.2.1 and 3.2.2. 45 See the earlier analysis of Rhodes with Lewis (1997), 123, of the number of proposers then known. 46 I.Oropos 186–7. 47 IG XI 1 100 (Foucart BCH 1886, 361); cf. Paris (1892), 62 + add. p. 318. 48 Herman (1987), 138–42. 43

102

Proxeny and Polis

we positively identify connections between proposer and nominee outside the decree—this is probably a reflection of the difficulty of proving personal connections rather than the rarity of them in proxeny grants.49 The granting of honours was, and was certainly always constructed as being, reactive, in response to benefactions already performed. Except in cases where the interaction in question occurred in the context of the granting community (e.g. if an individual were honoured for his performance in a public delegation from his own community to the granting polis), the honorand’s contact would have been with individual members of the community initially rather than the community as a corporate whole. The benefactions in question would, in the first instance, have been services performed for individuals (even when they were present representing the polis). In this sense a personal relationship is likely to have motivated the original proposer of the honorific decree and the proposal of public honours would thus be, in part, an expression of the proposer’s personal gratitude. The proxeny networks of poleis were in fact heavily dependent on private networks. In large part, the proxenoi of a polis would have comprised the private connections of its citizens appropriated and transformed into a public network. Individuals, in creating these links and in putting them forward for proxeny, played an essential role that was sometimes recognized in the decrees themselves. They probably also played an important continuing role in maintaining these ties— as the personal contacts of these individuals in the granting city. This seems to be reflected in western Greece in the very widespread practice, when appointing proxenoi, of simultaneously naming enguoi tēs proxenias—‘guarantors of the proxeny’—whose significance is indicated by the fact that they are listed even in extremely concise decrees where no other details are given. The issue here was not 49 Given the apersonal way in which inscribed decrees are invariably framed, we are reliant on rare circumstantial data to suggest personal connections. For example, it seems probable that the individual called Samios who proposed a proxeny decree at Magnesia on the Maiander for a Samian, was himself a proxenos of Samos and the son of another Samian proxenos, and had a personal connection with the honorand (I.Magnesia 6, early third century bc); similarly, it has been suggested that Pandius, an Athenian who proposed an honorific decree for Dionysius I of Syracuse, granting citizenship and a treaty with the Syracusans, was probably himself a proxenos of Syracuse (IG II2 103 and 105, 369/8 and 368/7 bc; Rhodes with Lewis 1997, 28). On the difficulties caused by norms dictating formal proposal by magistrates, see above, n. 36, and Mitchell (1997a), 33, on the prevelance of probouleusis at Athens.

Proxeny in its Political Contexts

103

citizens being liable to their polis for the performance of a proxenos, or even standing surety to a proxenos for the honours their polis granted him.50 Instead it was of establishing or authenticating the identity of the proxenos.51 Although visitors to the polis of the proxenos would have little difficulty in identifying their proxenos by asking his fellow townsmen, the polis as a whole (even more so the koinon) would have been in a much more difficult position in relation to a visiting proxenos. Mutual recognition was, however, a sine qua non for the sort of permanent relationship which was asserted by proxeny and for enacting the privileges promised. The western poleis apparently solved the problem of this structural asymmetry by formally devolving this responsibility on private individuals who knew the proxenoi—who may often have been identical with the proposers of the decrees.52 Elsewhere, other poleis probably also continued to rely on the friends of the proxenos to facilitate this aspect of their relationship with him. In this context the formulaic description of the honorand in the proxeny decree played a crucial role in reconstituting the relationship. It converted what was often, perhaps usually, a fundamentally private relationship into a public one by suppressing specific details of interactions with individuals except as citizens and emphasizing, as the basis of the relationship, the affection of the honorand for the honouring polis as a whole. It is not that this language was a fraud or sham. It was performative instead of simply being descriptive, just as

50 The explanations offered by Meier (1843) and Boeckh (at CIG 1776), which dominated subsequent scholarship (Gschnitzer 1973, col. 689; both were rightly rejected as early as Tissot 1863, 89–90). 51 So Gschnitzer (1973), cols. 690–2, rightly relating enguoi to the tokens (symbola) which private individuals used to establish identity. On symbola, see Herman (1987), 62–7. 52 The proxeny lists of Aitolia, which always list enguoi, provide a good case study. In these lists there is considerable variation in the number of enguoi for each proxenos, presumably reflecting the number of individuals connected or willing to connect themselves to a particular honorand. In the koinon of the Ainianians the magistrates were obliged to fulfil this function (thus IG IX 2 5b ll. 14–18, c.130 bc; the last two enguoi, who were brothers and are not separately named as magistrates, seem likely to have been responsible for the decree). The sporadic attestation of magistrates as enguoi in the Aitolian decrees, by contrast, perhaps reflects their involvement in proposing or supporting particular grants of proxenia (similarly, the fact that in an Opuntian decree of the second half of the second century bc, IG IX 5 1912, Soteles occurs both as archon, dating the decree, and enguos of the proxeny probably indicates that he proposed this grant in his year of office).

104

Proxeny and Polis

the decree of proxenia itself was. Moreover, the package of honours granted with the decree was intended to facilitate and encourage frequent interaction and the attachment of proxenos to polis.53 The success of the attempt which this language represents may be judged from evidence, examined in the third section of this chapter, of proxenoi internalizing this description and ideology. First, however, it is important to consider the reasons why proxenoi participated in this institution—the value of proxenia within the polis of the proxenos.

2.2. PROXENOI AT HOME AND AT COURT Proxenoi performed useful services for cities and their citizens. They did not, however, do this out of narrow self-interest, as part of the symmetrical reciprocity of hospitality that characterized private xenia, serving as xenos of an individual in their own community in return for the other’s services as xenos in the corresponding city. The proxenos was explicitly the friend of an entire city, of individuals most of whom he knew little of, and who perhaps knew personally next to nothing of him.54 Indeed, many of the intermediary services identified in Chapter 1 as particularly associated with the proxenos were of a kind that the proxenos himself had no need of in the city which recognized him as such, where he was not simply an undistinguished xenos but a proxenos who had official status and legal standing. What, then, was in it for the proxenos? Why did individuals not only allow themselves to be appointed as proxenoi with obligations which they might be called on to perform, but even actively seek it, taking the initiative in performing services which they then asked a polis to recognize by decreeing this permanent relationship? Our evidence suggests that the title, proxenos, was sought by individuals because it, and the relationship with an external polis which it signified, was a prestigious symbol in the context of elite citizen-competition within their own poleis. In order to capitalize on 53

See Section 2.3.1. This is stressed in the second part of the definition of proxenia given by the Hellenistic scholar, Aristophanes of Byzantium, where he states that those (sc. of the relevant city) who do not know the proxenos by sight, ask for and make enquiries after him ( . . . o ŒÆd ƒ Iª F   ÆE Zł Ø ØÆ ı Ł  ÆØ ŒÆd K ØÇÅFØ, Ar. Byz., Slater 1986, fr. 301). 54

Proxeny in its Political Contexts

105

this award, which was one of the means of asserting elite status, individuals had to advertise their proxenia to their fellow citizens. This was complicated, however, by the active interest which poleis as political communities had in controlling and monopolizing the expression of honours in civic space.

2.2.1. Proxenoi at home Poleis granted other honours with proxenia—indeed it was normal for a community to define, by law or practice, a specific set of honours and privileges to which proxenoi were entitled by virtue of their proxenia.55 These could include enktēsis (the right to own land in the granting community), proedria (prominent seating at civic performances), ateleia (freedom from taxation), politeia (citizenship), and in certain areas epinomia (rights to pasture on land belonging to the community), as well as, very commonly, expressions of praise. It would be perverse to argue that none of these privileges had an economic value and that the desire for them did not motivate certain individuals to seek proxenia. However, since proxenoi were generally envisaged as being politically active within their own polis, and these grants mostly concern the rights of the proxenos within the honouring polis, they will not on their own explain why proxenoi in general actively cooperated with this institution. There can be no doubt that proxenia was itself both attractive and actively sought. Decrees granting proxenia were always constructed as reactive, in response to benefactions. Individuals knew what sort of services were expected of proxenoi and performed them in anticipation of, and in order to prove themselves worthy for, the award of proxenia, which we know they sometimes then openly requested of the polis in question.56 The desire to acquire such grants, and thus their value, is illustrated in our oratorical sources by accusations of bribery made in connection with them, in particular, the sixty talents which Demosthenes and Demades were each said to have made out of ‘decrees and proxenies’,57 and also in an accusation made of Aeschines and his fellow ambassadors that they abandoned 55

Marek (1984), 142–4; see 129–30. Athenian and non-Athenian examples are collected and discussed in two appendices in Gauthier (1985), 181–95. 57 Hyp. 5.25. 56

106

Proxeny and Polis

their duty and instead set about obtaining proxenies for themselves at Oreos.58 A passage in Aeschines’ speech against Ktesiphon reveals the prestige-value such external appointments had within the context of the proxenos’ native polis, as well as one important example of how grants could be converted into symbolic capital—in the theatre in Athens on the occasion of the tragic contests. Formerly, citizens who had acquired grants of proxeny from foreign poleis had these awards announced on their own initiative by the herald who, with the whole city watching (and, Aeschines explicitly states, individuals from all the other Greek cities), proclaimed how ‘the demos (e.g.) of the Rhodians . . . crowns [x] for his aretē and andragathia’.59 For Aeschines this was the most invidious manifestation of a general tendency to undermine the control of the dēmos and boulē over the expression of honours within civic space, and he applauded the action of a lawgiver who clamped down on this abuse, allowing the proclamation of foreign honours (grants of proxenia and crowns) only when an embassy from the granting polis obtained prior permission from the proper civic authorities.60 This not only illustrates the real prestigevalue of such awards, it also reveals the tension between the interest which individuals had in maximizing this by publicizing awards to an audience of their fellow citizens as well as the interest of the community in controlling such displays, in part because of the phthonos, the envy, which they engendered in the wider citizenry.61 This is brought out in Pindar’s fourth Isthmian ode in a particularly striking way, in the description of the honorand’s family: d b t ¨ÆØØ Ø   IæåAŁ ºª ÆØ

æ  ’IçØŒØ ø Œ ºÆ A ’OæçÆ 

oæØ . . .

58 Aeschin. 2.89; Hennig (1997), 355–6, is surely right that Dem. 18.82 is a reference to actions performed by Aeschines as proxenos of Oreos (Histiaia). 59 Aeschin. 3.42. 60 ‹ ’q K ØçŁ Æ , æ Æ æÅ Ø Ø b K ÆE ø º Ø, Ø æ  I ƪæ ŁÆØ ‹Ø  çÆ E ÆPf › B, N oø  åØ, › H    ø j  ø j ŒÆd ¼ººÅ Ø e º ø Iæ B  ŒÆ ŒÆd I æÆªÆŁ Æ, Aeschin. 3.42. 61 Whitehead (1983), esp. 56–60, on the potential negative connotations of philotimia, ambition for honours; see also Osborne (1999), 355–6, connecting such proclamations with athletic victory; on the similar connection between athletic victory and phthonos in Pindar, Kurke (1991), 195–224.

Proxeny in its Political Contexts

107

But from the beginning they are said to have been honoured in Thebes as proxenoi of their neighbouring peoples and free of loud-voiced arrogance . . . 62

The connection here between the long-term elite status of this family at Thebes and their identity as proxenoi of nearby communities is clearly expressed. The assumption implicit in Pindar’s statement that the Kleonymidai are ‘free of loud-voiced arrogance’ is similarly clear—these symbols of status were things which individuals, under normal circumstances, could be expected to preen themselves on, in ways that might be harmful to civic harmony.63 Decrees of proxenia frequently made provision for public proclamation at festivals in the polis of the proxenos, illustrating the reality of Aeschines’ concern, but this was only one of the ways in which these awards could be publicized within the proxenos’ community.64 One of the means which is now most visible to us was the erection of monuments inscribed with copies of the external honorific decrees a citizen received, so called Ehrentafeln, of which the monuments of Nikomedes of Kos (thirty decrees inscribed 306–301 bc),65 Eudemos of Seleukeia (172 bc),66 Kassandros of Alexandria Troas (c.165),67 and Sosos of Rhodes (post 88 bc)68 are the most impressive examples. However, in the same way that the Athens of Aeschines had restricted public proclamations at the city’s dramatic festivals, poleis in general seem to have been effective in restricting the public inscription of such honours by private individuals for the same reason, the desire to exert authority over honorific expression in public space.69 With the exception of external decrees honouring foreign judges or other official representatives, which poleis did frequently inscribe because they honoured the city itself as much as they

62

Pindar Isth. 4.7–9, trans. Loeb, adapted. On proxeny in Pindar, see Currie (2005), 340–3. On amphictyonies, ‘those dwelling around’, as competitive contexts, see Parker (1998), 22. 64 e.g. in decrees from different cities inscribed at Kos, IG XII 4 139; 141–3; 147 (all second half of the third century bc). 65 66 IG XII 4 129–30. Syll.3 644–5. 67 68 Syll.3 653A (cf. 653B). Maiuri (1925), no. 18. 69 This is made particularly clear in a Koan decree honouring a citizen, which explicitly grants him permission, having inscribed the decree on a stone stele, to set it up in the sanctuary of Asclepius, IG XII 4 59 ll. 43–5 (mid-second century bc); cf. 57 ll. 27–9 (mid-second century bc). 63

108

Proxeny and Polis

honoured particular individuals, very few examples of monuments of this sort are known.70 There were, however, other ways in which individuals could communicate their possession of proxenia within their polis, and thereby derive benefit from its prestige-value, which was essential to the functioning of this institution. One means, which is likely to have been more important than we will ever know, was private display, the placement of visible honorific monuments within the house of the individual citizen. We are told by Aeschines of the retention of civic crowns in the houses of citizens by the recipients and their descendants as memorials (the word used is hypomnēma), and the context dictates that the same was expected of crowns awarded by other states.71 It is likely that the copies (antigrapha) of honorific decrees which cities explicitly ordered to be given to honorands were also frequently displayed in this way, whether the medium was bronze, papyrus, or painted board, and this practice of giving copies seems likely to be much more common than the relatively infrequent epigraphic references would suggest.72 Indeed, the erection of monuments such as Ehrentafeln in the proxenos’ home polis, inscribed with multiple honorific decrees from different cities, which do not themselves explicitly mention the giving of antigrapha, strongly suggests that proxenoi were generally given, or could easily acquire, copies of 70 A few honorific decrees were erected at the explicit request of the granting polis, especially at Delos (perhaps because of the international character of the shrine; e.g. IG XI 4 1054, 229–204 bc, and 1055, c.230–220 bc) and Kos (e.g. XII 4 138–48, second half of the third century bc—mostly for medics and/or officers dispatched by the Koans). The instances that we have, moreover, were not infrequently constructed as interactions between the granting city and that of the honorand (especially as letters between poleis e.g. IX 2 11, 179–146 bc; I.Thasos 169–71, after 130 bc). Where one polis sufficiently dominated another, it could also simply order honorific decrees inscribed at the subordinate polis, as in the case of Delos under the Athenians—e.g. ID 88 (inscribed on Delos, 368 and 362 bc) or IG I3 156 (Athenian copy with provisions for inscription at Halikarnassos, 435/4–428/7 bc). The Lindians could instruct an individual to set up their decree for a resident of Naukratis in the Panhellenion there presumably because of the peculiar political structure of this city, Syll.3 110 no. 4 (c.450–411 bc). 71 Aeschin. 3.47. 72 References to antigrapha in inscribed decrees appear from the second century bc onwards: e.g. I.Gonnoi 40 (first half of the second century bc); IG XIV 612 (Rhegion, second to first centuries bc); IGUR 2–3 (Akragas, c.100 bc); and IG XII 5 599 (Ioulis, second century bc?) with Marek (1984), 141 and n. 116 on this. This does not mean that, formally or even informally, individual proxenoi did not receive, or could not obtain, the text of their decree earlier, see below.

Proxeny in its Political Contexts

109

these texts.73 Private display seems to be explicitly attested in the case of the two bronze tablets, inscribed with proxeny decrees for Diodorus of Syracuse by Akragas and by Malta, which were discovered together in Rome. They were presumably set up together at the house of Diodorus in that city where he welcomed delegations and citizens from these cities, as the Akragantine bronze explicitly states (it also explicitly references itself as a hypomnēma).74 The audience for these monuments, although it was of course much narrower in this domestic context, would also have been more select, likely to consist of members of the social stratum to which proxenoi in general belonged, with whom they most directly competed. Such monuments did not have to be public or permanently visible to continue to communicate and articulate the prestige of the honorand since peers who saw or heard of such monuments would continue to be aware of them. Within their native poleis, proxenoi could also, however, communicate and derive prestige from their honorary position more actively. This was by publicly performing the role of the proxenos, by interceding with magistrates, introducing visiting delegates and private citizens, defending their interests in court, and ostentatiously providing hospitality. This is the face of the proxenos we see above all in Apollodoros’ portrait of Kallippos, proxenos of the Herakleians— arriving at Pasion’s bank and successfully demanding, as proxenos, to be shown the accounts of a deceased Herakleian, and later warning Pasion to beware of injuring so weighty a person as a proxenos.75 The picture may be overblown, as Apollodoros stands to gain from making his opponent ridiculous. However, the responses in this narrative—Phormion submissively showing Kallippos the ledger, Pasion agreeing that he would have to be mad to want to cross a proxenos—reinforce the sense conveyed of the essential prestige 73

One text of the early third century, I.Labraunda 42, even decribes how the son of the proxenos was able to produce the decree of the Plataseis, in this case to prove his own entitlement to ateleia. 74 IGUR 2–3 (c.100 bc). The new bronze inscriptions from Halaisa, two copies of an honorific decree for Nemenios, provide us with a new example of this sort of domestic honorific display (ed. pr. Scibona 2009). The pair were discovered in the destruction layer of a private house clearly belonging to the honorand, in which they had been displayed by being nailed to wooden doors (for a treatment of these texts and the many questions they raise, see SEG 59 1100). 75 Kallippos: ‘if anyone wishes to defraud me, let him know he is defrauding one who is a proxenos’ (ŒÆd Y     º ÆØ Iç ºŁÆØ, æ  Z Æ IçÆØæ ŁB ÆØ, [Dem.] 52.10, Loeb trans.).

110

Proxeny and Polis

inherent in proxenia. It is striking here again how the symbiosis of honour and function was fundamental to the institution of proxenia, as the individual needed to perform the functions associated with proxenia within his own community to capitalize there on the honour of being proxenos. Conversely, proxenia’s dark-side was that the individual exposed himself to shame if, when called on publicly, in the manner which our inscriptions mention and by visitors whose proxenos he was, he did not act as he was supposed to do. The same desire to capitalize on the prestige inherent in proxeny by communicating this status can also be identified in widespread Greek patterns of naming. In particular, the popularity of the use of ‘Proxenos’ as a personal name, attested 134 times in the published volumes of the Lexicon of Greek Personal Names, is likely to have been due to the elite status associated with this honorific role (just as names containing ‘hippos’ had definite aristocratic connotations).76 When a proxenos named his son ‘Proxenos’ this action reflected the conception of the role as hereditary and the aspirations of the father for the social position of his child.77 At the same time, however, it would also have been intended to make claims for the father’s own social position and was another means of publicizing the honorific position he held. Individuals also commemorated and advertised specific links to the poleis of which they were proxenoi in the names of their sons, by naming them after the ethnic of that city. Thus Cimon, the great fifthcentury Athenian statesman, proxenos of the Spartans, famously named his son ‘Lakedaimonios’.78 The large number of names derived from ethnics in the LGPN makes it clear that this was a common means of proclaiming links with external communities, but it is also important to note that these names made most sense in the context of the home city, distinguishing a prominent family from its rivals.79 76

Ar. Nu. 60–74, signalling in this case elite aspirations rather than status; on these names, see Dubois (2000). 77 In a few instances we even find individuals called Proxenos (or their sons) receiving grants of proxenia: IG I3 91 (Athens, 426/5–416/5 bc), 181 (Athens, c.410? bc); I.Oropos 35 (Boiotian League, 240–230 bc); SEG 16 373 (Lamia, first half of the third century bc). SEG 56 1710 (Olbia, fourth century bc) is another example (the published text should be corrected from the photo to read  ˙æ —æ| ı from  ˙æ| › ˛ ı—a fact I am grateful to Charles Crowther for pointing out). 78 Plut. Cim. 16.1; IG I3 364 l.8 (433/2 bc); see also Connor (1967), 67–71. 79 Chalkideus, Eretrieus, Thespieus, Korinthios, Megareus, Messenios, Mantineus, Samios, Thasios, Rhodios, Ephesios, and Milesios are some of the ethnic-names with at least five attestations. Athenaios is attested more than 250 times, but this is

Proxeny in its Political Contexts

111

This emphasis on the native polis as the context in which individual proxenoi were most interested in communicating this award prompts the question of what, from the point of view of the proxenos, was the point of decrees inscribed not in his home city, but in the granting polis (the location of the vast majority of the inscriptions we have).80 Just as monuments in private space could effectively convert external honours into symbolic capital, monuments in external public spaces could perform the same function under certain circumstances. Individuals competing within a polis could have an awareness of the monuments, honouring their peers, in other poleis, though of course this would be most effective when they existed in cities which were comparatively accessible, regularly visited anyway by citizens of the proxenos’ native polis. We can imagine without too much difficulty (we have no explicit sources) visitors proudly pointing out inscriptions erected in honour of themselves or jealously, grudgingly (but not without interest and curiosity) looking to identify monuments they know to have been erected in honour of their neighbours and fellow citizens. In this connection, for the honorand, the utility of a monument erected in a city never or rarely visited by fellow citizens from his polis would have been much reduced. Major cities, panhellenic sanctuaries, those places where we find the largest number of texts inscribed, were the places where such monumentalization would have been most valuable, and they served as external sites in which fellow citizens could compete directly with each other.81 In particular, the Amphiareion at Oropos, even during its independence, seems to have served as a major subsidiary arena for Athenian competition and display. During the second half of the third century upwards of forty Oropian awards of proxenia were inscribed for Athenians, between a quarter and a

complicated by the fact that Athenaios was understood as a theophoric name as well as an ethnic, which is why, unlike other ethnic-derived names, it is attested at the city itself (144 times at Athens). 80 That there was a point is clearly shown by the fact that the honorand sometimes supplemented the public money allocated to ensure a more lavish monument or paid the entire cost themselves—e.g. IG I3 95, 107, 155, 156 (all Athens, fifth century bc); cf. Low (2005), 103; Clinton (1996). For an extended discussion of the (distinct) functions which the inscription of honorific decrees performed from the perspective of the polis that inscribed them, see Lambert (2011). 81 This is what Marek meant when he used the term ‘Repräsentationszentrum’ of Delos in the third century and Delphi under the Aitolians, Marek (1984), 263.

112

Proxeny and Polis

third of the overall total. This fact clearly communicates the desire of Athenians to be honoured in this space in which their peers had previously been commemorated, probably reflected in a greater general eagerness to perform suitable services for Oropians in pursuit of proxenia.82 Delphi seems to have functioned similarly as an external space for competition between members of the same polis, to judge from the way in which decrees were located according to the polis of the honorand on the monumental topography of the shrine. Thus decrees for Megarians are concentrated on the so-called wall of the Megarians,83 and so on for the bases of the Argives,84 Arkadians,85 Aitolians,86 and Boiotians.87 Often the monuments in question had a pre-existing connection with the community of the honorands and, as such, were likely to have been specifically visited by individuals from these communities, as in the case of the treasuries of the Cyreneans.88 In the case of the treasury of the Athenians, the frequent use of internal designations (demotics and phyletics) further emphasizes the functioning of this as an external site for the continuation of internal competition beyond the ability of Athens to regulate or control.89 The numbers of individuals from the same city attested on these monuments draws our attention to an apparent paradox—that the honorific position of proxenos remained prestigious despite being potentially obtainable (and probably held) by a large proportion of the elite. At Classical Athens most members of the political elite (and all prominent politicians) were probably proxenoi of one or more external polis. This is something which only our epigraphic evidence, particularly lists of proxenoi, can give us a sense of because our literary and historical sources only mention an individual’s status as

82 These sorts of competitive dynamics are also probably visible in the proxeny list of Delphi, in the tendency for different individuals from the same polis to be named proxenos shortly after each other, see Chapter 3, n. 66. 83 FD III 1 155–96 (date range: c.320–200 bc), only certainly excepting 158; 164; 171; 176; 178–9; 183; and 195–6. 84 FD III 1 79–88 (date range: c.300–240 bc), only certainly excepting 84–5. 85 FD III 1 12–46 (date range: 300–200 bc). 86 FD III 1 142–54 (date range: c.355–145 bc). 87 FD III 3 79–116 (date range: c.320–250 bc; III 3 82, however, was inscribed in 172/1 bc). 88 FD III 3 158–78 are inscribed on the treasury—nos. 163, 165, 166, 169 are proxeny decrees for Cyreneans (mid-third century bc). 89 FD III 2 71–97 (date range: c.290–140 bc) with only 75, 83–4, 88 representing honours for non-Athenians.

Proxeny in its Political Contexts

113

proxenos when it is relevant to the issue at hand (not least because so many individuals were proxenoi).90 For instance, in its catalogue of proxenoi, the minor polis of Karthaia recognized no fewer than fifteen prominent Athenians as proxenoi at some point in the mid-fourth century. Even allowing for the fact that the ties between Athens and this polis on Keos were particularly strong, this suggests a huge number of proxenoi active at Athens.91 Although it seems probable that the total number of citizens who were proxenoi was lower in other cities, the evidence preserved by the proxeny-lists of Karthaia, Chios, and later Eresos and Astypalaia clearly show that the numbers appointed at even minor communities were likely still to be substantial, especially as expressed as a proportion of the citizen population.92 Proxenia, arguably, remained prestigious not so much because it was particularly difficult for prominent individuals to obtain (the typical services considered above were things influential citizens could do without incurring too much trouble or expense), but because it served as an important marker of status, a means of identifying those belonging to the political elite of a polis.93 Just as the value of inscribed honours varied according to the context of inscription, so too the prestige value of grants of proxenia also varied depending on the importance of the granting polis among a particular civic elite. The proxenos, who, through the practical functions he performed, represented citizens of a particular polis, could come in some contexts to stand symbolically for it. By virtue of his proxenia, the proxenos acquired prestige and importance in 90

See Introduction, pp. 14–15. See Mack (2011), 329–30, on these proxenoi (Appendix no. 10.1.4): all of those whose names can be read are attested as belonging to the liturgical class or are closely related to others who are. 92 See Appendix nos. 10.1 (Karthaia, after 363/2 bc); 4.1–4 (Chios, fourth to third centuries bc); 7 (Eresos, c.230–200 bc); 3 (Astypalaia, c.150 bc). 93 Veyne (1990), 121–2 and 129–31, discusses the function of honours in constructing class distinctions, though this perhaps implies too strong a degree of social stratification in the case of proxenia. Lambert (2011), 202–3, by contrast, argues that the term elite is of ‘questionable applicability in a society [the Classical Athenian democracy] in which there was genuine breadth of political participation’. As the Aeschines anecdote above illustrates, there was certainly a desire on the part of the democracy to limit forms of personal citizen display (Aeschin. 3.42; above, p. 106), but proxenia was of worth in part because it was not controlled by the proxenos’ own polis and, as it was explicitly part of a broader interstate system of values and symbols, the Athenian democrats could make use of it in relation to outsiders but not effectively redefine it on more egalitarian principles. 91

114

Proxeny and Polis

relation to the position and importance of the polis with which he was associated. This association was, on occasion, extended to the point where the stereotyped national character traits of a particular community could be imputed to its proxenos.94 In the same way that the inscriptions created by the polis to honour its proxenoi mapped its relations with other cities onto civic, monumental space, so its citizens who had been appointed proxenoi of other poleis could symbolize, internally, particular external communities. Thus Plato has the Spartan Megillos relate how, as the son of an Athenian proxenos, other boys would blame or praise him for the actions of ‘his’ polis.95 A unique monument from the minor polis of Narthakion, in the Malian Gulf, from the mid-second century bc, represents an attempt to map such foreign connections onto the citizenry of Narthakion, listing the proxenoi, or, less probably, theōrodokoi, whom other communities had nominated among the elite of this community.96 The intention was surely not to keep a public record of these potentially traitorous connections or to inform visitors from these communities of their representatives, for which civic archives would be much better suited.97 It was instead to present a view of the world as connected to this small community, to advertise the number and importance of the communities who thought it worthwhile to maintain links with Narthakion, including Miletos and perhaps Magnesia on the Maiander.

2.2.2. The perils of proxeny Proxenia was sought as a source of personal prestige and the link which it embodied seems to have been emphasized in order to benefit from the status inherent in this role. However, it could also, under certain circumstances, be a political liability for the individual, a basis for his political opponents to attack the motivation of his public actions. In one notable passage, Aeschines, commenting on the disastrous consequences of the alliance with Thebes, represents Demosthenes’ proxenia, his care for the interests of Thebes, as prodosia, a

94

95 Plut. Cim.14. Laws 642 b–c, on which, see pp. 132–4. Appendix no. 13 (before or after c.140 bc). 97 Two of the three reasons suggested by Gschnitzer (1973), 699–700, and accepted by Marek (1984), 136–7. 96

Proxeny in its Political Contexts

115

betrayal of the interests of Athens.98 When the needs of the two cities diverged—or could be represented as diverging—a prominent politician might expect his proxenia of the other community to be brought up against him. The extent to which the institution of proxenia could actually take precedence over the institutional expectations and models of behaviour bound up with citizenship—that proxenoi became, in truth, foreign agents—is questionable.99 It is clear, however, that such accusations were a commonplace and seem to be closely connected to similar tropes of political invective based on bribery and friendship.100 It is for this reason that Demosthenes is keen, in another speech, to stress the fact that he is advocating an alliance with the Rhodians on the basis of the benefits it would bring the Athenians rather than the Rhodians, stating that he has no reason to promote the Rhodians’ interests being neither their proxenos, nor a xenos of any of their citizens privately.101 According to Plutarch, the fifth-century Athenian statesman, Cimon, drew on the connection between xenia, proxenia, and bribery in an even more interesting way. Defending himself in court, probably at his euthunai, against an accusation that he had been bribed by King Alexander to refuse to invade Macedon, he is said to have replied that he did not ‘serve as proxenos of either the rich Ionians or Thessalians, as others did, performing services and taking bribes’. Instead, he was proxenos of the Lacedaimonians, whose moral purity and austerity he imitated, and consequently only took wealth from the enemy and used it to adorn Athens.102 The context is presumably that Cimon was accused of being bribed as the friend, xenos, of Alexander,103 which would make him, because Alexander was a

98  I º  ƃ æ Ø P Ø’ K, Iººa Øa c c æ Æ ŒÆd c æe ¨ÅÆ ı æ Æ , Aeschin. 2.141. Harris’s arguments, that Demosthenes’ proxenia may simply be an invention of Aeschines, cannot stand: Harris, (1995), 199 n. 15. The Athenian audience would have known which communities a prominent politician like Demosthenes was proxenos of. 99 See Section 2.3.3. 100 For the ubiquity of the accusation, see Harvey (1985), esp. 89–102 and 114–17; see also Perlman (1976) and Mitchell (1997a), 181–6. 101 Dem. 15.15. 102 Plut. Cim. 14. It seems likely that the source of this striking statement was contemporary (it would be an odd thing for a later source to make up). 103 For the double meaning of dōrodokia depending on perspective—accepting gifts from a friend and accepting a bribe—see Herman (1987), 73–8; and Mitchell (1997a), 181–6.

116

Proxeny and Polis

king and his friends therefore were the public friends of Macedon, the equivalent of the proxenos of a polis,104 and for this reason he raises the topic of others paid off as proxenoi, in contrast with his own relationship with the Spartans. It is tempting to suppose that when he refers to Athenians who serve as proxenoi of Ionians and Thessalians he has in mind specific candidates who have been attacked with this sort of accusation, whose identities would occur to members of his audience, and that he is not just generally referencing stereotypes (though he is, of course, doing this as well). At the least, this manner of reference suggests that such accusations were a fact of political life, especially as a large proportion of the political elite of Athens would have been proxenoi of one or more polis at any one time. The function of this rhetoric was, in most cases, like accusations of bribery, to cast doubt on one’s opponents’ motives and muddy the water. In some instances, however, we see it developed further, alongside accusations of ‘Lakonizing’, ‘Boiotizing’, or, from other perspectives, ‘Atticizing’, that is in connection with an intensification of the association of the proxenos with a particular polis. In a number of cases, when the political situation was sufficiently charged, the result was exile for the individuals concerned. This was the fate of Cimon (albeit in the moderated form of ostracism) and of at least thirty-four Thasians, including one Athenian proxenos, exiled, so the Athenian inscription holds, ‘for Atticizing’ ([K ’ I]ØŒØHØ; surely a strategic simplification by the Thasians concerned of a complex political event in order to elicit Athenian gratitude).105 In the case of Cimon it is clear from the sayings preserved by Plutarch, in which admiration for Sparta is the most prominent theme, that this association was to a large extent self-constructed. His ostracism, in the aftermath of a Spartan snub of Athenian aid with which Cimon was closely identified, was both a punishment of the man but, more than that, a symbolic internal expression of the Athenian foreign policy response (just as Cimon’s early recall to Athens, was intended to both 104 This is also evident in the treatment of the Thebans after the destruction of Thebes—the friends of Philip and Alexander and proxenoi of the Macedonian cities are spared, illustrating this equivalence (Arr. Anab. 1.9.9). 105 IG II2 33 ll. 6–7 (385 bc); cf. Dem 20.52 (ref to those z  ØØ Øa c æe A h ØÆ æ ÆØ B Ææ ). The proxenos (who is not distinguished on this list) may have been Amyntor whose grant of proxenia was apparently inherited from Apeimantos, his father (this is perhaps why Amyntor was the recipient of a reinscribed proxeny grant with his brothers, IG II2 6, shortly after 403/2 bc).

Proxeny in its Political Contexts

117

promote, and signal, Athenian rapprochement with Sparta).106 Cimon illustrates the dangers of being out of step with the prevailing political mood, allowing oneself to become an internal symbol of an external power which was or had become the focus of hostility. In the same period, the grandfather of Alcibiades sought to avoid a similar fate by publicly renouncing his proxenia with the Spartans, a charged gesture in this same symbolic vocabulary. The attempt failed and the Athenians, rejecting his attempt to disassociate himself from Sparta, decided to ostracize him nonetheless.107 It is thus not surprising that recent analysis of the speeches of Demosthenes has shown how careful he was in his public oratory, despite being (or becoming) proxenos of the Thebans, to reflect wider Athenian hostility towards them (though this, of course, did not stop Aeschines from accusing Demosthenes of placing Theban interests above Athenian).108 Like the repudiation of grants of proxenia, such persecution of citizens serving as proxenoi for other communities was probably comparatively rare, arising in specific political situations in which a proxenos was associated too closely with a polis that was hostile or fell into disfavour. Much more common rhetorical invective signalled the existence of, and attempt to capitalize on, a potential tension between an individual’s duty as a citizen and external identification as a proxenos. However the frequent use of such citizens on sensitive missions to the cities of which they were proxenoi109 and, more generally, the fact that a large proportion of a given city’s elite 106 This, at least, is how Cimon’s return to Athens was later remembered (Andoc. 3.3, Aeschin. 2.172), though the accuracy of this has been doubted. 107 Thuc. 5.43.2 and Hornblower (1991–2008), ad loc. with bibliography. 108 Trevett (1999), 184–202; see above, pp. 114–15. 109 Poleis suspicious of the loyalties of citizens who were proxenoi are hardly likely to have trusted them to look after their interests in the difficult business of negotiating alliances or peace, and yet in our literary sources they frequently recur: Cimon was sent to negotiate with the Spartans (Andoc. 3.3; Aeschin. 2.172); the Spartans twice sent Lichas the proxenos of the Argives to Argos to negotiate peace (Thuc. 5.22 and 76); Kallias the Athenian, proxenos of the Spartans, was sent to them as an ambassador (Xen. Hell. 6.3.2–4); Demosthenes was sent to negotiate an alliance with the Thebans, whose proxenos he was (Aeschin. 2.141–3). Instead cities sought themselves to take advantage of the good relations which these individuals had with the cities they represented. The fact that at Plataia a Spartan proxenos was one of the few to remain until the bitter end of the siege—surely to negotiate if the need came (he was in fact one of the two interlocutors used) Thuc. 3.52—is a case in point, his loyalty to the polis of Plataia first and foremost cannot have been in any doubt. See Perlman (1958), 186. See Chapter 1, n. 158 for the other examples of citizens sent to the poleis of which they were proxenoi.

118

Proxeny and Polis

would have been proxenoi of other poleis indicate that this potential tension was not necessarily perceived as a problem and was unlikely to have serious consequences for the individuals involved. It was, at any rate, insufficient to counteract the allure of proxenia in this competitive context.

2.2.3. Proxenoi at court The ‘friends of the king’, those ‘spending time with the king’, the individuals ‘placed in charge of ’ some military or geographic division—a number of different designations were used in decrees granting proxenia to explicitly identify individual honorands as operating in a non-polis context.110 Although many of these individuals were not necessarily physically ‘at court’, that is, with the king,111 either absent on royal service, or indeed serving or associating with those who were, the court was in a very real sense a political context to which they belonged, a competitive environment, centred on the ruling king, which shaped their actions and aspirations. They were, in short, active members of a wider ‘court society’, perhaps better a ‘court culture’.112 In this context the sort of good relations with Greek cities indicated by grants of civic honours and honorific relations like proxenia could be an asset, easing the performance of their royal duties. However, it is not clear whether, or how, proxenia, along with other civic honours, could usefully serve as objects of prestige in this context, and thus motivate the recipients to participate in these civic institutions. This is because, despite certain superficial similarities, city and court had fundamentally different competitive structural dynamics. The citizens of the polis, who competed amongst themselves for status, were, at least conceptually, equals, and grants of proxenia served as a source of prestige in this competition because they were authoritative, externally derived symbols of status. By

110 On these descriptions, see Chapter 1, pp. 54–5; on court titulature and its development in general, see Savalli-Lestrade (1998), 251–87. 111 For this abstraction of the term, see Plb. 4.77 and 5.81, the king’s aulē on the battlefield, with Herman (1997), 204–5. 112 On the theory of court dynamics, Elias (1983) remains fundamental; for the ancient world see Herman (1980, 1997), Paschidis (2008), and now Ma (2011). On attempts to distinguish ‘inner’ and ‘outer’ court members in ancient courts, not always wholly successful, Spawforth (2007).

Proxeny in its Political Contexts

119

contrast, the court was a rigidly hierarchical institution in which status and the symbols which communicated it were internally derived, a function of proximity and access to, as well as the condescension of, higher powers (the king, above all, but also prominent courtiers).113 Within the competitive context of the court, why, therefore, did individuals such as Nikomedes, son of Aristandros, or Aglaos, son of Theukles, both of Kos, not only act in such a way as to be named proxenoi in the first place (especially by aiding embassies to their kings), but continue to conform to the same paradigm, to perform as proxenoi, as both are recorded as having done?114 The difference between these two political cultures, city and court, however, is significant only if we assume an unwarranted degree of separation between them. Each had an interest in asserting its own claims over individuals, and this often found its expression in official constructions of the identity of individuals: in the case of the court in omitting ethnic and patronymic, often substituting court titulature;115 in the case of the city, in stressing the primacy of these elements, especially the ethnic indicating polis-belonging, over court affiliation which was often indicated using phrases in diatribō, describing current activity rather than identity.116 Both the court and the polis, however, stood to gain by making use of those individuals who were active in both spheres. Although there might be an ongoing tug-of-war between conflicting attempts to claim their loyalties and influence their actions (attempts to de-socialize and re-socialize, to adapt the terminology of Ma, emphasis dependent on perspective),117 their value was precisely as intermediary figures.118 At the interface between these two political cultures, proxenia functioned as a marker for status in part because its recipients often continued to compete with peers in the context of poleis where it conveyed prestige.

113

Herman (1997), 216. IG XII 4 129 decree V (306–301 bc); ID 1517 (c.154 bc). 115 J. and L. Robert (1983), 114–15. 116 Ma (2002), 210–11; cf. Savalli-Lestrade (1998), 277, distinguishing between the usage of free and subject cities. Ethnics are occasionally omitted for individuals very deeply embedded in court service, see Chapter 1, n. 113. 117 Ma (2002), 206–11. 118 Davies (2002), 11–12, describes them evocatively as ‘the human hinges of Hellenism, not just channels of communication but basic, load-bearing components of the system’. See also Paschidis (2008), esp. 483–6. 114

120

Proxeny and Polis

Recent work has in fact emphasized the extent to which members of courts retained strong links with their poleis of origin, often being simultaneously active (even regularly physically present) in both political contexts.119 Moreover, in the same way that even absent monuments played a part in elite competition within the polis, individuals absent at court or in royal service did not necessarily cease to belong to the elite of their city or to compete for status there. The status at court of citizens, and their importance for their polis as a conduit for relations, would itself have been a source of prestige, which the individuals themselves might attempt to capitalize on, and on which members of their family, still present in their polis, certainly would have done. Like the proxenoi of cities, the citizenphiloi of the king stood for and symbolically represented the external reality of the royal court within the polis.120 Access to royal resources gave increased opportunities for personal aggrandisement and the augmentation of political influence within the polis.121 Access to the court, to the king and influential figures around him, materially enhanced the ability of the individual to perform, and be ostentatiously rewarded for, benefactions to cities. Poleis assumed that individuals in court service continued to maintain strong links with their cities of origin.122 In one instance, we have material which allows us to explore what, for one middle-ranking official in the Ptolemaic hierarchy, the maintenance of these sorts of ties meant. Correspondence surviving from the archive of Zenon the Kaunian, oikonomos in the service of the dioiketes Apollonios, vividly attests to the interest which he maintained in the affairs of his own polis and its neighbours. We possess a letter in which Zenon is sent news of a recent political scandal at Halikarnassos123 and a number of requests for the application of his influence in relation to both the royal administration and the cities of the addressees in question.124 In one of these, a delegation from Kaunos represents it as Zenon’s duty, along with other citizens in important positions outside the polis, to continue to further the interests of his native polis and fellow 119

120 Paschidis (2008). Paschidis (2008), 482–3. Plb. 20.5, for the example of ‘the house of Neon and Brachylles’, strengthened by Antigonid aid, with Paschidis (2008), 477–81. 122 123 See Chapter 1, Section 1.2.2. P.Cair.Zen. I 59037 (258 bc). 124 P.Cair.Zen. III 59341a–b (247 bc) contains two appeals from different citizens of Kalynda—an epigraph by Zenon shows that he was moved to help his cousin in the latter case. See also P.Mich. I 23 (257 bc). 121

Proxeny in its Political Contexts

121

citizens.125 In return the delegates promise to report his aid to the dēmos of Kaunos, that is, presumably, to propose a decree in his honour.126 Zenon, of course, as a citizen of Kaunos, his native polis, could not be its proxenos (although it is striking that, as here, in other honorific decrees for citizen courtiers their services at court on behalf of fellow citizens are described in the same terms as those of proxenoi).127 Other evidence, however, suggests that the assumption made by cities—that courtiers continued to want to compete in their own civic space—was not necessarily unfounded in relation to proxenia. In particular, we possess the striking fourth-century monument of Nikomedes of Kos, consisting of two large opisthographic stelae bearing at least thirty decrees of Greek states, most of which probably conferred proxenia, to judge from the fragments that survive. The erection of this monument in Nikomedes’ home polis stresses that, during and despite his service under the Antigonids, Kos remained the relevant competitive context for these grants.128 A similar dossier of proxeny decrees inscribed for Eudemus, son of Nikon, at his home city of Seleukeia in Kilikia, attests a comparable interest in his social standing there.129 Although these monuments are comparatively unusual, the concerns which they reveal would have been much more generally shared and poleis certainly assumed that they were. Of course, not all courtiers would have been preoccupied to the same extent with their status in a specifically polis-based context when they participated in these polis-based institutions. However, proxenia, and other civic honours, could still have functioned as objects of prestige in these instances, at least below the highest rank of courtiers, because royal courts were so closely intertwined with poleis. Courts would have been substantially composed of individuals for whom such grants were inextricably 125 . . .  Ç   K غº Ø Ø ŒÆŁ æ [E º]Ø E º ÆØ E I e F _  º ı ºØ ı Ø çæ  Ç Ø  ø , P.Col.Zen. I 11 ll. 5–7 (257 bc). 126 _ I  çÆ ØF , P.Col.Zen. I 11 l. 10. The base verb, emphanizō, HØ  [øØ] is frequently used in motivation clauses of decrees to describe the report of behaviour initiating a decree. 127 e.g. IG XII 4 32 (Kos, mid-third century bc, for a citizen doctor at the Ptolemaic court); cf. XII 7 36 (Arkesine, Amorgos, second century bc). 128 IG XII 4 129–30; 1039B and 1040 (both late fourth century) are statue bases from a related family monument, for statues of Nikomedes and his mother respectively. 129 Syll.3 644–5 (172 bc).

122

Proxeny and Polis

linked with prestige, and, hence, proxenia could have come to serve as a desirable indicator of status and influence at court among even those apparently unconnected with poleis.

2.3. POLEIS AND THEIR FOREIGN PROXENOI Poleis actively sought to foster and cement the emotional basis of this relationship by granting, along with proxenia, a concomitant package of honours which collectively articulated the regard of the city for its proxenoi, encouraging them in different ways to form an attachment by defining a privileged status for them at the polis. How effective this attempt was in general is difficult to say but the evidence which we have does allow us to begin to explore the use that proxenoi made of this ideology and how some seem to have internalized it.

2.3.1. ‘And all other honours given to proxenoi’ In the same way that poleis differed in the precise language which they used in decrees to communicate what was essentially the same proxenos-paradigm, the specific content of packages of honours associated with proxenia could vary greatly between poleis. Thus, cities of the Peloponnese and central and northern Greece commonly granted the right of pasturage (epinomia) on their extensive publicly-owned pasturelands—which was elsewhere all but unknown.130 Similarly grants of citizenship (politeia) emerge as a characteristic of proxeny decrees in many of the cities of the eastern Aegean (e.g. Samos and Iasos), but at others they were rarely made, and never became part of this defined package of honours, especially in the west.131 Identifying precisely what constituted the standard set of honours and privileges at a given polis is often difficult. In a number of cases we have references to the fact that this was formally defined, the 130 Chandezon (2003), 351–89, esp. fig. 13 p. 382, for the precise distribution of cities granting epinomia alongside proxenia. 131 Habicht (2002), 13–30. Marek, in his catalogue of proxenies, gives a useful overview of the privileges granted alongside proxenia at different poleis, but does not distinguish between those that were exceptional and those that were typical, see Marek (1984), 8–118.

Proxeny in its Political Contexts

123

subject of a proxenikos nomos or similar regulation, of which we possess the texts for two cities, prefacing catalogues of proxenoi from Karthaia (specifying eisplous/ekplous, prosodos, and either ateleia or proedria) and Kalchedon (politeia, isoteleia, eisplous/ekplous, and inscription on the stele).132 Elsewhere, we possess only vague references to honorands receiving ‘all that is given to proxenoi’, which may imply the existence of either a specific law of this kind or possibly just that a conventional set of honours had arisen in practice. It is often difficult to ascertain which of the honours specifically delineated in a particular decree form part of this package and are therefore conferred even in proxeny decrees which do not specifically mention them, and which were granted in addition, reflecting the polis’ view of the particular merits of individuals, and/or their specific requests. This is particularly true since exceptional decrees granting additional honours (of which inscription was often one) are likely to be over-represented in our evidence.133 Nonetheless, it is possible to distinguish between honours that regularly formed part of this standard set, and those that tended to be distinct, rewarding subsequent or exceptional services, such as crowns (especially of gold), statues, and public maintenance, as well as extremely rare gifts of real property.134 The honours that cities habitually granted to proxenoi had a longterm symbolic value in defining the relationship between polis and proxenos which, from the point of view of the polis, went above and beyond any immediate economic advantages they might confer. A number of the honours most regularly granted to proxenoi represent abstractions of, or were calqued on, rights which citizens exercised by virtue of their citizenship. Indeed, it seems probable that such grants were crucial for the development, communication, and continuing definition of concepts of citizenship and what it entailed. In this category belong enktēsis, access to the boulē and/or assembly (prosodos or ephodos), freedom from taxes which non-citizens were expected to pay (ateleia), or the right to pay at a lower, citizen rate 132 Appendix no 10.1 (Karthaia); 9 (Kalchedon). The Eresian catalogue (Appendix no. 7) may well have been prefaced with a similar decree, see Mack (2012), 220. For the proxenikos nomos as dictating the privileges awarded to proxenoi, see I.Ilion 53 ll. 19–20 (Ilion, second century bc); cf. Milet I 3 140A–B (treaties between Miletos and Knossos and Gortyn respectively, mid-third century bc), where the proxenikos nomos (at Knossos, Gortyn and 23 other Cretan poleis) is compared to the nomos emporikos at Miletos—at issue here are probably rights of access or accelerated rights of access (prodikia) to courts. 133 134 See Introduction, Section I.3. Habicht (2002), 29–30.

124

Proxeny and Polis

(isoteleia),135 and, of course, citizenship itself (politeia). Granted to a proxenos, these privileges assimilated him, at least partially, to the status of the citizen. They allowed him, in the case of enktēsis, the ‘right, which Athenians (i.e. citizens) possess, of owning property, both land and houses’ as our earliest ad hominem grant states,136 or, in combination, to ‘take part in all things other citizens do’, as grants of politeia and isopoliteia frequently assert.137 Some scholars have seen a fundamental contradiction between the activities associated with proxenoi in their own communities, and grants of citizenship and enktēsis in particular, which, it is argued, they could only have made use of (i.e. converting them from potential to actual grants) if they permanently migrated to the city in question, thereby ceasing to function as proxenoi.138 Indeed, the routine grant of politeia with proxenia, which we clearly see in the standard package of honours from Kalchedon in the second century bc, has been taken as evidence of the degeneration of proxenia to an empty title.139 In the first case, the contradiction has been rightly challenged—the honorand was the proxenos of another city in the context of his own, its citizen when he visited the second polis.140 These honours were intended to function under normal circumstances, and not just if an individual was exiled from his native community. In practice, it is not likely that many proxenoi served as magistrates or participated in deciding law cases—two of the central but more onerous features of the Aristotelian definition of citizenship (which, in fact, excludes

135 Marek (1984), 159; on the importance of this as a mark of civic belonging at Athens, see Liddel (2007), 181–2. 136 [ŒÆd ªŒ Ø] r ÆØ ÆPE [ æ Ł Æ Ø, [ŒÆd ª ] ŒÆd NŒ Æ, IG I3 102 ll. 30–1 (410/9 bc). In addition, the honorand is also granted oikēsis (the right of residence), which is not granted later, presumably because it was not thought necessary, as enktēsis became a more familiar concept, suggesting the novelty of enktēsis at this date. 137 e.g. ºØ Æ ŒÆd  ı Æ  ø z ŒÆd BÆæªıºØBÆØ  åıØ , I.Priene 47 ll. 18–19 (c.200 bc); ŁBØ b ÆPHØ ŒÆ[d] ºØ Æ  å Ø  ø z ŒÆd E ¼[º]ºØ º ÆØ  Ø , I.Iasos 51.12–4 (230–190 bc); cf. Szanto (1892), 12. 138 Szanto (1892), 14–17; Francotte (1910), 199–200; Osborne (1981–3), 3.45, ‘proxenia, a privilege clearly incompatible with the citizenship’. On enktēsis: Baslez (1989), 351–2, suggesting that enktēsis indicates that foreigners were coming to reside; for an overview of the evidence for foreigners owning real property, see Hennig (1994). 139 Schwahn (1931), esp. 108–12. 140 Wilhelm (1942), esp. 44–50; 77–8; Marek (1984), 152–3; Mitchell (1997a), 39–40.

Proxeny in its Political Contexts

125

many citizens born and brought up under more oligarchic constitutions).141 There were, however, other ways, perhaps more appealing and certainly more compatible with visiting, in which the inclusion of the honorand in the citizen community would have been expressed.142 In particular, civic feasts and festivals in which the whole population partook, and was, thereby, defined, were probably more effective in promoting a sense of belonging in those proxenoi who came to participate, as many, especially from local poleis, could quite easily have done.143 The significance of these occasions is underlined by the explicit grant of proedria frequently made to proxenoi—preferential seating at performances. Proedria was a privilege the proxenos shared with members of the civic elite and his seat, among the magistrates, important priests, and citizen-benefactors, emphasized the importance of the proxenos within the broader community looking on.144 The symbolic significance of enktēsis grants similarly lay in the ideological importance attached to land ownership in the Greek poleis, which was frequently linked, actually or ideally, to citizen participation.145 The grant of enktēsis by a community to an honorand was a compliment, an unequivocal statement of its estimate of his worth. However, if the proxenos (the most frequent recipient) was in practice excluded from such ownership by the expectations attached to his performance of proxenia, this gesture would have rung rather

141 Aristotle, Pol. 1275a–b. Marek (1984), 155, is quite right to note that just because we have no evidence that proxenoi did participate in these ways, this does not mean they were conscious of being barred. 142 So Marek (1984), 154–5. 143 On ‘religious citizenship’, see de Polignac (1995), 79–8; Parker (2011), 151–2. Participation in magistracies and religious rites were two parallel and complementary methods of constructing citizen-identity as I.Priene 12 makes clear ([ ı Æ ] ƒ æH[ ] ŒÆd Iæå[ ] ø zª ŒÆd —æØÅ E ƒ ¼ºº[Ø], l.22, early third century bc). 144 _ In a_ first-century bc decree from the Messenian city of Thouria this is made explicit: r Æ [ Æ]PfiH K ÆE H ıÅæ ø ±æÆØ æ æ Æ [ŒÆd] æ  Æ  a H ± æø NÆæ Æ ø[ ] (SEG 11 974 ll. 18–21, first century bc). The status implications of proedria are brought out when it occurs in descriptions of courtiers and their specific standing with the king; for examples, see Savalli-Lestrade (1998), 268 nn. 71–2. Proedria is given a particularly concrete meaning in the two seats recovered at Olympia from the sixth and fifth centuries bc which were inscribed as belonging to individual Spartan proxenoi of the Eleans (LSAG 199.15; SEG 26 476). 145 e.g. in the property qualifications which frequently determined the degree of political participation; Politics 1329b 37, for the ideal of the restriction of these most important rights to H B ºØ Æ   å ø ; Migeotte (2009), 78–9.

126

Proxeny and Polis

hollow.146 In fact, there seems little reason to accept that proxenoi were in general excluded in this way. The Attic Stelae, recording the sales of property confiscated from citizens at the end of the fifth century, make it clear that it was theoretically possible to own and derive profit from property overseas—even to farm it directly rather than lease it out—and give us a glimpse of the complexity of property holding possible for the elite.147 The very few extant provisions limiting the exercise of enktēsis to residents either reflect the extraordinary, temporary nature of a grant or—like other limitations on enktēsis especially of value and location—local, often highly specific sensitivities.148 There is no reason, therefore, to assume that this privilege could not be used by a proxenos who remained normally resident and active in his own polis. Indeed, it seems probable that, from the point of view of the polis, a grant of enktēsis more effectively fulfilled its function if the invitation that it constituted was taken up. It involved investment on the part of the individual of his own resources in the community he served as proxenos, in which he 146 The reconstruction by Marek (1984), 158, on the basis of IG XI 4 543 (Delos, early third century bc), of a necessary secondary decree activating a grant of enktēsis and further defining its limits, is also very dubious. This inscription is not an activation of this grant, but a response to a specific request made by this proxenos that any property of his on Delos should not be liable for seizure to repay public debts—perhaps reflecting his own knowledge of the Delian public finances or simply the economic realities of polis life (cf. IG XII 7 67, c.300 bc, in which Praxikles the Naxian is promised as security all property in Arkesine, including that of non-citizens, which he has the right to foreclose on in the event of the city defaulting on its debt). 147 The cardinal example is the property portfolio of Adeimantos, which included land at both Thasos (IG I3 426 ll. 44–50) and Ophryneion (430 ll. 10–12; both 414 bc). The sale of crops from the latter suggests that it was farmed by Adeimantos rather than leased out. The context for these particular texts, of course, was the Athenian Empire of the fifth century bc, but these give us a sense of what might have been possible for individuals who were explicitly granted the right to acquire property in particular cities. One individual otherwise known to have taken advantage of a grant of enktēsis while continuing to live abroad was Leukon, king of the Bosporos, who owned a number of properties at Athens (Dem. 20.40). 148 Other than IG II3 316 (= II2 237, 338/7 bc) to a group of Akarnanians whose exile, it is anticipated, will not be permanent (a special case, see Pečírka 1966, 139–40), IG V 1 4 (Sparta, after 188 bc) is the only example of enktēsis being made conditional on settlement. It is tempting to read this provision, specific but unusual as it is, as evidence that residence was not, in fact, otherwise the rule, and to link it to the emphasis placed in this inscription on the role of the honorand in requesting honours. The only other non-Athenian example of limitations on enktēsis is the Prienean decree for an Ephesian (I.Priene 3, after 334 bc) which limits the value of land he can acquire (though, at 5 talents, it is not very limiting) and specifies that it must not be too close to the border Priene and Ephesos share.

Proxeny in its Political Contexts

127

thereby took a material stake that aligned his interests more closely with that polis, deepening his identification and intensifying his interactions with it. Even privileges like freedom from taxes which non-citizens were liable for had a symbolic significance that was arguably central, especially where freedom from import duty was granted only on goods for personal consumption, and thus involved neither a substantial cost to the polis nor great profit to the proxenos.149 This was a significance which would have been strengthened and regularly renewed by the actual use of the privileges involved. Paying citizen dues on goods, or none at all, explicitly differentiated the proxenos from other foreign visitors, particularly on entering and leaving the city, and as such was a mark of belonging. The importance of what imposts one paid in a specific community for personal identity is communicated particularly clearly by the replacement of ethnics by the designation isotelēs in Athens, even in personal funerary monuments, when the individual in question was fortunate enough to obtain this tax status.150 Similarly, access to faster and preferential legal institutions (e.g. prodikia or, at Athens, the court of the polemarch), as well as having practical advantages for someone not intending to remain at the polis in question for very long, also had the effect of clearly and publicly distinguishing the proxenos from other visiting xenoi.151 Many of these grants, indeed, seem to have been intended to encourage and ease the interactions of the proxenos and the city, to facilitate the habitual and emotionally involved relationship described

As at Odessos (IGBulg I² 42, third or second centuries bc), Eretria (IG XII 9 198, 304–301 bc), Thera (XII 3 1290, third century bc), Samothrace (XII 4 148, inscribed at Kos, second half of the third century bc), Chios (SEG 12 390, c.320 bc), Magnesia (I.Magnesia 6 ll. 20–1, early third century bc), Priene (I.Priene 2, 334 bc), Ephesos (I.Eph 1453 ll. 1–20, 299 bc), Kios (I.Kios 1, fourth century bc); cf. Marek (1984), 157–8. This restriction is not the norm, but rather, like other restrictions, expresses local conditions and concerns. It is phrased with absolute clarity in a decree from Abdera, ‘tax-exemption on goods he imports and exports for personal consumption but not for commercial purposes’ ( N c N Æ åæ Æ ŒÆd c ŒÆ’ K æ Æ , I.Thrac.Aeg. 8 ll. 27–9, second century bc). 150 For references, see Chapter 1, n. 123. 151 As well as Delphi, prodikia is also attested elsewhere in central Greece (e.g. IG IX 1 1, Antikyra, Phokis, second century bc), northern Greece (e.g. SEG 23 420, Pherai, fourth century bc), the Black Sea (e.g. IGBulg I² 37bis, Odessos, third or second century bc), and the Aegean (e.g. IG XII 5 715, Andros, third century bc). 149

128

Proxeny and Polis

by the decrees.152 The right of direct access to the authoritative bodies of the polis of which he was the official friend—prosodos or ephodos— was one expression of this. Nothing, however, communicates this more clearly than the explicit right of travelling to and from the city freely and safely, regardless of whether a state of peace or war existed with his native polis, and without fear of seizure (sylē).153 This is probably the most widely and frequently attested privilege granted to proxenoi, found, in various equivalent formulations, in practically every decree which describes the privileges granted to a proxenos at all (as eisplous/ekplous, eisagōgē/exagōgē, aphixis, asphaleia or asylia, or, most commonly, a combination of these elements). The significance is that the proxenos (unlike other, undistinguished xenoi) was not to be held responsible for any wrongs which his fellow citizens, individually or collectively, perpetrated against citizens of the granting polis, even in war. The proxenos is thus cast as a true intermediary figure, whose individual identity is recognized by the granting polis and whose relationship with it is given precedence, from its perspective, over his membership of his own polis. As well as privileges like these—encouraging the proxenos to visit to strengthen the emotional tie identified by the polis as the basis of his action as proxenos—in many proxeny decrees, poleis themselves expressed a care for the interests and well-being of their proxenos. This was the counterpart to the care the proxenos was expected to exert for the city’s interests and citizens. Given the collective nature of the polis (as contrasted with the proxenos), this care was expressed in the form of instructions to the city’s officers and magistrates—to ensure that an honorand could enjoy his honours and to take more generalized care of him.154 We do in fact have some examples of how this care, which again assimilated the proxenos to the citizen,155 could manifest itself: the releasing without ransom of proxenoi taken in

152

On association (koinonia) as the foundation of friendship, see Mitchell (1997a), 7. This is a paraphrase, but it crystalizes the sense of the many different ways in which this grant was framed. On sylē and asylia in general, with bibliography, see Bravo (1980) and Gauthier’s response (1982) as well as Buraselis (2004) and Lintott (2004) with more recent bibliography. 154 e.g. IG II2 19 ll. 10–11 (Athens, 394/3 bc); XI 4 613 ll. 12–16 (Delos, 270–260 bc); I.Magnesia 48 ll. 29–30 (Eretrian decree, inscribed at Magnesia on the Maiander, 208 bc). 155 cf. Lys. 28.1, ŒÆd ªaæ º Ø æ øŒg çÆ ÆØ, ŒÆd æ ı ŒÆd º Æ  æı M،Ō. 153

Proxeny in its Political Contexts

129

wars with their native poleis;156 delegations seeking the release of proxenoi captured by others or to revenge their deaths;157 and funeral honours and monuments for individuals who died at the cities they served as proxenoi.158 The Athenian dēmos of the fifth century bc expressed this care in an otherwise unparalleled form—so-called harm clauses which state that, if the proxenos is killed in one of the cities controlled by the Athenians, the punishment will be the same as if an Athenian citizen has been killed, involving a fine of five talents levied against the city concerned. Like other honours, this assimilated the proxenos to the citizen, extending a protection enjoyed by the latter. Meiggs saw the collective aspect of this punishment as a ‘sure sign of developed imperialism’.159 In fact, it is an extension of the pre-existing interpolis practice of sylē. What is imperialistic here is the ability to impose a fine, something they could only do, of course, ‘in the cities over which [they had] control’, which is why this phrase occurs in relation to these clauses in proxeny decrees.160 More conventional sanctions, like 156 In both Xenophon (Hell. 7.2.16) and Polybius (5.95) different cities are noted for having freed proxenoi captured in war, ¼ ı º æø , without ransom. The interpretation of e ªaæ — ººÅ Æ æ  has been disputed, however, as Xenophon refers to this as a praiseworthy act—with proxenos interpreted both as a personal name (revived in a recent translation of the Hellenica, Strassler 2009, ad loc), or, emended, as a reference to the fact that the individual was proxenos of their enemies (Knoepfler 2002—in which case freeing him would be an inexplicable act). The difficulty of this phrase has been greatly overstated. It is closely paralleled in Xenophon and elsewhere (e.g. ØÅ Ł  e æÆ Ç Ø æ  Z Æ H ı  Œø , Xen. Anab. 5.4.2). There is no further qualification of whose proxenos the Pellenean is because it is obvious in context. It is the situation of the Phleiasians which make this action notable. In spite of their great financial difficulties (ŒÆ

æ  ø  Æ ØÇ Ø) they did their duty to ‘their Pellenean proxenos’ (so Pritchett 1991, 259 with n. 373). Relevant also is Alexander’s act in sparing his own philoi and the proxenoi of the Macedonians after the sack of Thebes (Arr. Anab. 1.9.9). 157 Demosthenes reports that the Athenians sent three embassies to Philip seeking the release of their proxenos, a Karystian, whom Philip instead killed and even refused to allow them to bury (Dem. 7. 38). Revenge: RO 39 ll. 27–40 (an Antipater condemned to death for killing a proxenos, presumably in the first stage of the Keian revolt against Athens, 363/2 bc). 158 ML 4 (= IG IX 4 882, 625–60 bc); æ Æ Iæ B  åæØ æ ø  ŒÆd ÆP Ð K Ł ŁÅ ÆEØ —ıŁÆªæÅ Ł Æ , IG I3 1154 (460–450 bc); cf. Dem. 7.38 with the preceding note and p. 57 n. 123. 159 Meiggs (1949); de Ste Croix (1961), 268. 160 Other authorities describe the territories which they ruled as spheres in which they could make particular grants in similar ways, especially in relation to grants of ateleia—so Pherai (SEG 23 418 ll. 6–7, fourth century bc; now with Graninger 2011),

130

Proxeny and Polis

sylē, might be expected in cases of citizen and proxenos slaying in cities over which the Athenians did not have this sort of control.

2.3.2. The ideology of the proxenos Civic decrees thus both communicated the polis’ understanding of what proxenia entailed (the proxenos-paradigm) and embodied the active steps which it took to establish the relationship implied. This prompts the question—how did individuals, as proxenoi, respond? The expectations attached to this institution seem to have influenced the behaviour of proxenoi—as proxenoi reproduced ‘the manner of thinking and feeling characteristic’ of the proxenos (the ideology of the proxenos).161 However, although we have some evidence of proxenoi not merely verbalizing this ideology but even internalizing it (or being represented as doing so), proxenoi cannot be understood as the ‘agents’ or ‘henchmen’ of the granting poleis, acting against the interests of their own, as they were accused in ancient oratory. To explain why, it is necessary to explore how this ideology was communicated to proxenoi, as well as how proxenia related to other institutional roles within the city, and the different ways in which individuals possessed agency to pursue their own interests within this framework of institutions. Proxenoi are frequently attested voicing this ideology in interactions with the authoritative bodies of the cities they served. In decrees honouring both new and pre-existing proxenoi we find references not only to the proposer’s description of the honorand’s regard for the city, but the honorand’s own proclamation of his personal disposition towards the polis.162 More striking, however, is the material from our literary sources. Thucydides and Xenophon record no fewer than four speeches given by foreigners before the Spartans in which the speaker draws attention to his identity as and Mausolus (RO 55, mid-350s bc). These phrases serve the same function as clauses in later decrees limiting grants of ateleia to taxes which the authority in question controlled (usually excluding royal taxes which they did not), e.g. I[º] ØÆ Kª ˚ÆøHØ z i ˚Æø[ ]E Œ æØØ tØ , I.Mylasa 942 (Hellenistic period). 161 ‘A system of ideas concerning phenomena, esp. those of social life; the manner of thinking characteristic of a class or individual’, Shorter OED3 sv ‘Ideology’ no. 3; on this use of the term, see Whitehead (1977), 3. 162 e.g. Eumaridas, IG 3 1137 ll. 55–9 (= II2 844, Athens, 193/2 bc).

Proxeny in its Political Contexts

131

proxenos and the conventional emotional dimension of this role. In each of these cases, the function is explicitly persuasive—the Spartans should listen, because as proxenos the speaker is acting out of a desire to benefit, or as Jason of Pherae states, save them (ÆFÆ ! , çÅ, Kªg

æŁıFÆØ, HÆØ A ıº  Ø  c F Ææe çغ Æ

æe A ŒÆd Øa e æ E H ).163 Similarly, in Herodotus, Alexander, king of the Macedonians, is recorded as beginning his speech with a reference to his eunoia for the Athenians, and develops this theme further in the personal entreaty which he delivers when he urges them, for the sake of their own welfare, to accept the offer of Mardonius (Alexander, indeed, is explicitly said to have been sent by Mardonius as a persuasive interlocutor because Mardonius learnt that he was proxenos of the Athenians).164 The use of this rhetoric reveals an awareness of how a proxenos should speak—and feel. In other cases, it does not seem fanciful to talk of proxenoi actually internalizing this ideology. This is certainly how the proxenos, Kallippos, is represented, even by his enemy, Apollodoros, as he is made to say ‘for it is, of necessity, incumbent on me to take care of all citizens of Herakleia’ (K I ªŒÅ ªæ  KØ ±  ø  ˙æÆŒº øH K Ø º EŁÆØ).165 Care, the reference here to epimeleia, corresponds to and reciprocates the care which poleis expressed for their proxenoi, the emphasis of ‘all’ echoing directly the assertion of proxeny decrees that this was a relationship with the entire citizen population. The most striking element of this pronouncement, however, is the forceful character of this duty of care, presented as a strongly felt moral obligation by the phrase ‘of necessity’ (K I ªŒÅ).

163

Xen. Hell. 6.4.24, in addition to 6.1.4 (Polydamas of Pharsalos, warning the Spartans of Jason!) and 6.3.4 (Kallias of Athens). Alcibiades’ presentation of his proxenia in his speech before the Spartans in Thucydides (6.89) is necessarily more complex, see Section 2.4. 164 Hdt. 8.140 ; 136. The Athenians similarly make use of conventional moral language in their reply, in rejecting his advice (‘seeming to render a good service (chrēsta) you advise us to commit a great crime’), and in bidding him be gone (‘for we do not wish you, who are our proxenos and friend (philos), to suffer anything unworthy (acharis; i.e. a mark of ingratitude) at the hands of the Athenians’; Hdt. 8.144). On Alexander’s relations with the Athenians, and the probability that he aided the Athenian delegation to the Persian court of 507 bc, subsequently repudiated, see Badian (1994). Alorcus, a Saguntine proxenos (‘publice Saguntinis amicus atque hospes’, Livy 21.12) begins his speech to the Saguntines attempting to persuade them to accept Hannibal’s terms in the same way (21.13.2–3). See Chapter 4, n. 144. 165 [Dem.] 52.5.

132

Proxeny and Polis

In Plato’s Laws the potential personal emotional basis of this relationship from the point of view of the proxenos is asserted even more clearly and convincingly—expressed as a general rule by the proxenos of Athens, the Spartan Megillos:166 Yø b s ŒÆd AØ E ÆØ , K Øa IŒ øØ ‹Ø Ø  NØ

º ø æ Ø, Æ fiÅ Ø h ØÆ KŒ ø PŁf K  ÆØ "ŒÆ #H H æ ø B fi º Ø, ‰  ıæÆ fi hfiÅ Ææ Ø  a c ÆF ºØ · ŒÆd c ŒÆd Kd F ÆPe F Kªªª . IŒ ø ªaæ H Æ ø PŁ , Y Ø çØ  j ŒÆd K ÆØ E ¸ÆŒ ÆØ ØØ ŁÅ Æ ı, ‰ “  ˙

ºØ H , t ªØºº ,” çÆÆ , “#A P ŒÆºH j ŒÆºH ææ  ” — ÆFÆ c IŒ ø , ŒÆd Æå  æe ÆPa  bæ H I d æe f c

ºØ N łª ¼ª Æ, AÆ h ØÆ å , ŒÆ Ø F X  çø c

æçغc H . . . Probably all children, when they hear that they are the proxenoi of some city, conceive an affection (eunoia) for that state from infancy, as being a second fatherland after their own city. So indeed this very thing has happened to me. For hearing this from the first from other boys, whenever the Lakedaimonians were blaming or praising Athenians, how they said ‘that city of yours, Megillos, has treated us badly’ or ‘has done well’— hearing these things indeed, and constantly fighting on your behalf against those who were blaming your city, I acquired a complete affection for it, and even now your accent is very pleasing to my ears . . . 167

The context is crucial here. This is not a highly charged public oration in which the speaker attempts to capitalize on the conventional emotional content of a relationship he happens to have with the assembly he is seeking to persuade, as is the case in the speeches preserved by our historiographical sources. Instead, Megillos delivers this little monologue in response to the recent characterization of Sparta, and himself, as as being brachylogos, ‘of few words’ (which he comically confirms, remaining all but silent throughout the rest of the dialogue). Megillos reassures the Athenian speaker that he will listen willingly to the rest of his speech because of the affection for all things Athenian (including Attic Greek) which, as hereditary proxenos, he has conceived from boyhood. Moreover, couched as a general rule, surely included in this section of the Laws because of Plato’s 166

For the identification of this Megillos with a Spartan recorded as having been sent to Athens on an embassy along with Endios (FGrHist 324 fr. 44), see Parke (1957). 167 Plat. Laws 642 b–c.

Proxeny in its Political Contexts

133

concentration on education there, it rings true, explaining the role of peers’ projections onto an individual in the formation of personal identity. It illustrates the complexity of the different influences involved. The polis that granted proxenia to an individual communicated its proxenos-paradigm in the decree which honoured him, and with the honours that it granted encouraged the proxenos to engage himself actively and emotionally in this relationship. However, in most cases it was probably the proxenos’ own polis, his interactions with his peers and their assumptions, which most effectively instilled this proxenos-ideology in him and, as here, fostered his emotional attachment. This process surely operated with adults named proxenoi as well. Indeed, the evidence we have for rhetorical attacks based on grants of proxenia, considered above, illustrates its function in ‘grown-up’ civic contexts (Megillos explicitly states that he was continually defending the Athenians, a process that began, but did not conclude with his youth). Nonetheless, stress here on the particularly formative influence of hereditary proxeny is striking, and it is probably for this reason that such emphasis was placed on inherited proxeny in our evidence. This process was probably even more intense when the proxenos in question actually bore the ethnic of the city with which he was connected, and grew up in Sparta being called ‘the Olynthian’, ‘the Samian’, ‘the Chalkidian’, or even ‘the Boiotian’, to mention a few of the examples attested there.168 This proxenos-ideology finds a particularly full expression in the description of the actions taken by one proxenos. Towards the end of his final book, Thucydides narrates a story about his namesake, the Pharsalian proxenos of the Athenians, in which the internalization of proxenos ideology is clearly implied.169 In 411 bc, in the context of mass hysteria, when the older men were only with difficulty restraining the Athenians of the city from attacking the Athenians in the Piraeus, Thucydides the proxenos is credited by the historian with playing a decisive part in preventing the outbreak of civil war: . . . ºØ b H  æ ıæø ØÆŒøºı ø f K fiH ¼ Ø ØÆŁ Æ ŒÆd K d a ‹ ºÆ ç æ ı ŒÆd ¨ıŒı ı F $Ææƺ ı F æ ı B º ø Ææ  ŒÆd æŁ ø K   %ŒØ Olyntheus: Xen. Hell. 6.5.33; Chalkideus: RE ‘Chalkideus’; Samios: Hdt. 3.55.2 and Xen. Hell. 3.1.1; D. S. 14.19.4–5; Boiotios: Xen. Hell. 1.4.2. 169 On the identity of this individual, see Hornblower (1991–2008) on this passage and Thuc. 2.22. 168

134

Proxeny and Polis

ªØª  ı ŒÆd K Øø ı c Kç æ ı ø Kªªf H º  ø I ºÆØ c Ææ Æ, # åÆ  ŒÆd çH ÆPH I å . Only with difficulty were the older restraining those in the city who were running around and going for their arms (to attack those in the Piraeus) when Thucydides of Pharsalos, proxenos of the Athenians, who happened to be present in Athens, eagerly (prothumōs) put himself in the way of each of them and invoked them not, with their enemies lying nearby in wait, to bring about the destruction of the fatherland. The Athenians quietened down and refrained from attacking each other.170

This ‘noble and urgent plea’ of Thucydides the proxenos,171 and the way it is framed by Thucydides the historian, who very deliberately mentions his namesake’s proxenia, emphasizes the relationship between these actions and the proxenos-paradigm. The ideology of the proxenos is evident in the eagerness and energy, the prothumia (a key characteristic of the proxenos) with which he delivers this plea.172 Thucydides does not urge them to save Athens or their fatherland, but the fatherland, in which he himself, as proxenos, is emotionally invested. However, the most striking instance of this sort of emotional investment on the part of a proxenos—and its consequences— is to be found in a fourth-century inscription from Thasos.173

2.3.3. Herakleodoros of Olynthos, proxenos of Thasos The text records a benefaction made by Herakleodoros of Olynthos to the polis of Thasos of which he was proxenos. He paid for the construction of a tower, reinforcing the defences of the closed military harbour of Thasos, along with an exedra and a statue, collectively dedicating these monuments to ‘all the gods’. On even a cursory reading, Herakleodoros’ gift, his contribution to the defence of

170 Thuc. 8.92.8. Thucydides’ precise syntax is difficult to render in English but the narrative force is clear enough: Thucydides saved the day. 171 Rood (1998), 195, n. 6, on the highly marked verb used here. 172 For prothumia in the interests of the granting polis as a hallmark of the proxenos, see Chapter 1, pp. 34–7. 173 ed. pr. Pouilloux (1948) (= SEG 19 595 = Maier 1959, no. 55 = Meier 2012, no. 38). A full re-edition of this text has been prepared by Patrice Hamon and will appear in his corpus of the public inscriptions of Thasos of the fourth century and Hellenistic periods, Hamon (forthcoming). I am very grateful to him for allowing me to see this in advance of publication and for long and fruitful discussion of this text.

Proxeny in its Political Contexts

135

Thasos, emerges as an impressive, concrete assertion of the concern that he had for the town as its proxenos. None of the other gifts or dedications we know were made by proxenoi equal that of Herakleodoros in respect of either their cost or their symbolic force.174 In fact, financial gifts are so rarely attested of proxenoi that it seems likely that they were only very infrequently made, and all indications suggest that they did not feature among the services which proxenoi were expected to perform by virtue of this honorific status.175 However, by reading this rich dedicatory inscription carefully and reconstructing the monument of which it was part, it is possible to understand the circumstances which led Herakleodoros to make his exceptional gift as a proxenos, and to get a sense of the complex negotiation between the roles of proxenos and citizen which it represents. The text reads:  ˙æÆŒº øæ æØ Œı ! Oº ŁØ › æ  e æª ŒÆd c KæÆ Œ[Æd] e I æØ Æ Ł E AØ KŒ B ÆæÆŁŒ[Å] w ŒÆºØ Ææa æå øØ HØ  IØÆ [ı]. _ __ Herakleodoros, son of Aristonikos, the Olynthian, the proxenos, (dedicated) the tower and the exedra and the statue to all the gods from the deposit which he had entrusted to Archedemos son of Histiaios.

The way in which the two individuals who frame the text are named encapsulates its essential incongruity: it is Herakleodoros, emphatically identified as a foreigner, not the citizen Archedemos, who is responsible for the construction of this defensive tower. These references, however, are also an invitation to us (as they would have been to contemporaries) to reconstruct the events surrounding the act of dedication. In particular, in the second half of the fourth century bc, no use of this ethnic Olynthios could fail to remind the audience of the 174 We are not told the cost of this dedication but it must have been considerable. When, in the second century bc, Sotas, son of Europos, came to repair another tower in the wall of Thasos using public revenues, the cost was 7,000 drachmas (IG XII 8 391). It is rare even for groups of wealthy citizens to pay for a tower (IG II2 2331, Athens, 172/1 bc) and there are very few cases of individuals bearing this cost (e.g. IosPE I2 179 from Olbia, late fourth century?, with an elaborate honorific epigram) though we do possess one example from Thasos from approximately this period (Grandjean and Salviat 2000, no. 74). For dedications made by proxenoi, see Introduction, n. 23; for gifts of other kinds see Chapter 1, n. 142. 175 See Chapter 1, Section 1.2.4.

136

Proxeny and Polis

fate of that polis, destroyed by Philip II in 348 bc. By prominently using his Olynthian ethnic, and thus highlighting his lack of a polis, Herakleodoros explains the extraordinary dedication for a proxenos which follows—Herakleodoros had lost the city for which he might more properly make this sort of benefaction. The poignancy of an Olynthian, whose own city had been taken and destroyed, paying for the construction of a defensive tower for another would have been lost on no one. This also explains why Herakleodoros is so unusually keen to explain the precise source of the money used for the dedication. Rather than simply stating that these monuments were constructed from his private resources (KŒ H N ø ), as we would usually expect, Herakleodoros mentions that it was taken from the money he had on deposit with Archedemos in order to honour this individual, who had not taken advantage of Herakleodoros’ plight and sought to keep the money for himself.176 However, perhaps Herakleodoros also gives this detail to emphasize that he himself had not abandoned or betrayed Olynthos during the siege with Philip, stealing away to Thasos with his money to avoid the fate of a captured polis. This description emphasizes Herakleodoros’ relations with Thasos prior to Olynthos’ fall, painting him as a man with business at Thasos and connections there (hence his status as proxenos).177 This detail concerning the origin of the money also makes it very unlikely that the monument as a whole could have been constructed in the last part of the fourth century bc or at the start of the third—the date that Pouilloux gave the inscription on palaeographic grounds.178 Honouring Archedemos’ restraint in this way, while explicable shortly after Herakleodoros came to reclaim his money following the destruction of Olynthos, would become progressively odder as time went by. It seems unlikely that even a decade could have elapsed since the money was in Archedemos’ keeping, much less thirty or fifty years.179 176 Hdt. 6.86, the story of Glaukos the Spartan, who sought to keep such a deposit for himself, illustrates the vulnerability of foreign depositors, a theme which recurs in Apollodoros’ speech against Kallippos, see above, Chapter 1, Section 1.2.6. 177 Pouilloux (1948), 264–6. 178 According to Patrice Hamon, the lettering cannot be placed so confidently at the very end of the century and a date much closer to 348 bc is paleographically possible. 179 Pouilloux (1948), 263. It seems highly unlikely that Herakleodoros would make this dedication after the Thasians joined the league of Corinth in 338 bc (RO 76, b l.5; cf. Dem. 18.197) and it is most plausibly read as the expression of a desire by the Thasians to improve their defences in the context of Philip II’s aggressive campaigns

Proxeny in its Political Contexts

137

Herakleodoros surely did not need to refer to himself in this way, as a proxenos from Olynthos. His munificent gift would have earned him citizenship at Thasos, or any other polis, had he wanted it, especially as, in the period which follows, a number of inscriptions attest to the fact that citizenship could be bought at Thasos (and for far less than the cost of this gift).180 The inscription directs a reading of the monumental complex of which it was part as a monument to and for Herakleodoros, both as a proxenos of Thasos and as an exile from Olynthos. This, for instance, is probably the reason for the universalizing dedication of this complex ‘to all the gods’ (rather than to the polis and/or dēmos of Thasos)—it is a dedication both to the gods of the Thasians, but also, potentially, to the gods of Herakleodoros’ former polis.181 The tower itself which this text mentions has been plausibly identified with the circular tower built to reinforce the pre-existing wall which enclosed Thasos’ port of war (which contained its triereme sheds)—tower G.182 This inscription, written on the outer face of a convex block, was probably inscribed on the tower itself, on the outside of the wall (rather than within the closed harbour where it would have been rendered invisible by a trireme shed), which implies that this outward face was accessible by foot, perhaps along a quayside from the commercial port.183 The epideixis of this text, moreover, its reference to ‘the tower’, ‘the exedra’, and ‘the statue’, strongly suggests that the latter monuments were adjacent to the tower where this text was inscribed, which means that they were almost certainly located at its base. The exedra which this text refers to was probably not a free-standing bench but, in line with contemporary usage elsewhere in our sources, a building with a roof intended to contain a bench—perhaps a lean-to

in the northern Aegean (the dedication of this tower presupposes detailed discussions between Herakleodoros and the Thasians concerning their needs). 180 e.g. IG XII suppl. 355 (the fee is 100 staters per person, beginning of the third century bc). IG XII 8 267 attests to the alternative in the early third century bc—a proxenos of Thasos receiving citizenship, probably after the destruction of his polis of Zone (on this text and its context, see Psoma 2008). 181 Pouilloux (1948). 269–70, on the possible context for this dedication at Thasos, but Maier (1959), 202, is right to stress how unusual this dedication to ‘all the gods’ is. 182 Grandjean (2011), 322–4, dating the tower to the fourth century bc. 183 An impressive platform around the base of this monument (which thus made it accessible by foot) is in fact known archaeologically, Lianos (1999), 262.

138

Proxeny and Polis

against the tower to protect the statue from the elements.184 The most important question is the identity of the subject of the statue. The text provides two possibilities, Herakleodoros himself or Archedemos his faithful banker (we cannot, of course, exclude the possibility that it was neither). Archedemos has usually been championed as Herakleodoros’ benefactor,185 but, given the neutral way in which his benefaction is referenced at the end of the text without direct or explicit praise, it seems rather hard to read this monument as a whole as erected in his honour. Herakleodoros, in fact, explicitly situates himself in this central place in relation to the monument as a whole by dedicating it, and it seems much more likely that it was he who was the subject of this statue. In this case his name, in the nominative case, occupying the first line of the dedicatory text— Herakleodoros, son of Aristonikos, the Olynthian, the proxenos— would thus serve as the label for the statue itself.186 If so, this statue would have communicated clearly the unhappy situation of Herakleodoros, with its back to the defensive tower he constructed at Thasos, standing symbolically outside its walls and looking perpetually westwards over the sea to the modern promontory of Pachys and beyond, to Olynthos, the polis from which he had come and to which he could not return.

2.3.4. The limits of identification There were limits to the extent to which proxenoi in general identified with, and indeed were expected to identify with, the polis with which they were thus affiliated, above all when its interests conflicted with the interests of their native community. In as much as there was a 184 Hellmann (1992), 126–30. The suggestion of Lianos (1999), 262, that by exedra, Herakleodoros meant a raised platform around the base of the tower onto which ships could be hauled, is thus implausible (it is enthusiastically taken up by T. Koželj and M. Wurch-Koželj in Blondé et al. 2000, 3 with an improbable illustration by way of reconstruction on p. 56). 185 So Pouilloux (1948), 268–9; Meier (2012), 310. 186 See Briant (1998) for a discussion of the occurrence of andrias in a contemporary inscription, probably, like this, a statue of the dedicator himself. I wonder whether this statue may not have been decreed by the Thasians to honour Herakleodoros for his promise to pay for the harbour tower. If so (and there is no evidence in the text that we have, which treats it as simply a private dedication), Herakleodoros responded by undertaking the cost of the statue as well (a practice well attested in the Hellenistic period and later, see Ma 2013, 237).

Proxeny in its Political Contexts

139

hierarchy of obligations, the duty of a citizen was never conventionally subordinated to his duty as proxenos. In Plato’s Laws, Megillos states that, although potentially strong, the attachment of the proxenos to the polis which honoured him was as for a second fatherland, after his own (‰  ıæÆ fi hfi Å Ææ Ø  a c ÆF ºØ ). This was also expressed even when poleis honoured their foreign benefactors,187 often by explicitly praising the honorand for acting in the best interests of his own fatherland (usually mentioned first) and the honouring dēmos.188 This is hardly surprising. In making honorific grants communities were communicating values to their members even more than to the honorand, and it was in no polis’ interest to encourage disloyalty in citizens by setting proxenia above politeia. Nonetheless some modern scholars, following ancient rhetoric in which proxenia could be represented as prodosia, have described certain proxenoi as ‘henchmen’, as first and foremost the agents of the poleis with which they were affiliated.189 Such interpretations have been effectively challenged before,190 but it is important to explore instances in which proxenoi appear to be acting against the interests of their own poleis because they reveal how divisive the question was of what constituted the interests of the community. The relationship expressed by proxenia remained fundamentally based on friendship (rather than employment and dependence),191 expressed by foreign cities with members of a civic elite who were primarily motivated by interests which they attributed to the polis and themselves in the context of that polis. The nature of this link did not apparently change, even under the Athenian Empire at the end of the fifth century bc. However, when the communities linked by proxenoi were particularly unequal in terms of strength and resources, or subject to an unequal power relation (e.g. between hegemonic and subordinate poleis), the position of link, of proxenos, became much

187 Thus, the city of Alexandria Troas honoured an Ilian  ıæÆ r [ÆØ] [ Çø _

Æ]æ Æ c # æÆ ºØ (I.Ilion 68 l.3–4) with the Roberts, BE (1976), 566); that this inscription was erected at Ilion suggests that the force of this  ıæÆ was clear. 188 e.g. IG II2 467 ll. 20–22 (Athens, 306/5 bc), º[ªø ŒÆd æø a ıç] æ Æ BØ  ÆF 〚æ〛Æ[æ Ø { Ææ Ø} ŒÆd HØ øØ HØ Ł]Å Æ ø ·, honouring a preexisting proxenos. 189 Papazarkadas (2009), 78–9; Gerolymatos (1986), esp. 93–103. 190 Marek (1988), reviewing Gerolymatos (1986); Reiter (1991) on the Athenian examples, 321–8; Russell (1999), 76–83; Culasso-Gastaldi (2004), 13, with nn. 8–9. 191 See Chapter 1, n. 37.

140

Proxeny and Polis

more highly charged. This was particularly true in the broader context of internal factional and ideological polarization in alignment with wider, binary conflict, which we see in the case of the second Peloponnesian war. Three cases illustrate this particularly well: the proxenoi of Mytilene, who betrayed to the Athenians the plans of their fellow citizens to synoecize Lesbos and rebel;192 Peithias, the etheloproxenos of the Athenians at Corcyra, who sought to bring his polis into alliance with them and was, in the end, assassinated;193 and, from the first years of the fourth century, Xenias of Elis, proxenos of the Spartans and xenos of their king, who took the opportunity of a Spartan-led invasion of the territory of Elis to attempt to eliminate his political rival, Thrasydaios, and thereby gain control of Elis.194 Interpretation, then as now, depended largely on perspective. From the perspective of members of the opposing faction, these actions were surely interpreted (and represented) as a betrayal of the polis and its interests as they conceived them. From the point of view of the dominant polis, actions of this sort performed by their proxenos, especially where successful, were likely to be received as valuable services requiring thanks, as evidence of the regard of their proxenos for their interests. Conversely, as our narrative histories all make clear, from the perspective of the individual proxenos, actions of this sort (as distinct from other specific sorts of service, performed in his capacity as proxenos) were primarily performed for his own political gain, for interests which were his but which coincided, in his view, with the interests of his polis (or what he thought they should be). He was the friend—rather than the agent—of a great external power, a position he sought to manipulate for his own ends. In the case of the Mytilenean informers, Thucydides’ description makes these priorities clear:  ØØ ªaæ Z   ÆPE ØçæØ ŒÆd ÅŁı ÆEØ ŒÆd ÆPH ıØºÅ Æ ø N Æ fi ¼ æ  ŒÆa Ø , æ Ø ŁÅ Æ ø , Å ıÆd ª ª  ÆØ E ŁÅ Æ Ø. For the Tenedians, who were at variance with the Mytileneans, and the Methymnians as well as, privately, some of the Mytileneans themselves

192

193 Thuc. 3.2. Thuc. 3.70. Xen. Hell. 3.2.26–31; Paus. 3.8.3–5 recording, unlike Xenophon, Xenias’ Spartan proxenia and personal xenia with King Agis—dubiously? 194

Proxeny in its Political Contexts

141

who happened to be proxenoi of the Athenians and were motivated by factional strife, all became informants to the Athenians (concerning the activities of the Mytileneans).195

These private Mytileneans were acting, in Thucydides’ estimation, first and foremost as the political opponents of those advocating synoecism and revolt from Athens. Their relationship of proxenia with the Athenians was relevant in that their interests, and probably, according to their sincere assessment, their polis’ interests, were bound up in the continuation of the alliance with Athens. This relationship also probably made them the most persuasive messengers to send, on behalf of what must have been a larger community of opinion at Mytilene (as Thucydides’ reference to stasis, factional strife, implies).196 It is made clear, however, that these beliefs were not simplistically determined by their position as proxenoi by the fact that one of these proxenoi was successfully persuaded to go on a second embassy, on behalf of the dominant party working against the Athenian alliance, to counteract the first (perhaps persuaded that events had now gone too far, Thuc. 3.4). Faction is clearly presented as the decisive factor in the other two narratives also—and each author emphasizes the initiative of the etheloproxenos and proxenos respectively, making opportunistic use of the situations they were presented with and the courses of action proxenia made possible.197 The Athenians themselves were clearly used to local players, both internal polis factions and nearby external enemies (here the Tenedians and Methymnians), attempting to manipulate and channel Athenian resources for their own ends. Here, exhausted by plague and war, the Athenians, recognizing the motives of the informants, chose to disbelieve the substance of their reports.198

195

Thuc. 3.2; Aristotle preserves an alternative, partisan account, presumably derived from his close connections with Lesbos, which identifies personal pique as the motivation of the embassy to Athens, Arist. Pol. 1304a9. 196 Wilhelm (1942), 43. 197 Opportunism is particularly prominent in Xenophon’s account—‘since the territory was being ravaged and an army was at Kyllene, Xenias’ faction wished to make the city over to the Spartans {through their own efforts?{’ (fiÅı Å b B åæÆ, ŒÆd hÅ B æÆØA æd ˚ıºº Å , ıº Ø ƒ æd ˛ Æ . . . . {Ø' ÆPH { æåøæBÆØ E ¸ÆŒ ÆØ Ø, 3.2.27). It is striking that Agis makes no efforts on behalf of Xenias, when his political rival came to terms. Personal initiative is also clearly explicit in the fact of Peithias’s etheloproxenia, Thuc. 3.70. 198 Thuc. 3.3.

142

Proxeny and Polis

Victory in factional stasis was obtained through the dominance of a particular perspective—concerning the identity of the polis, its interests, and therefore the actions of opponents (informed by a different perspective) in opposition to these, which were thereby characterized as betrayal of the polis and punished with death or exile. The prosecution of Peithias the etheloproxenos at Corcyra for trying to enslave Corcyra to the Athenians was nothing less than an attempt by his opponents to make authoritative their own view of his actions. The attempt failed and Peithias responded in kind (the fact that his opponents were prosecuted for other offences does not mean that anyone failed to realize that what was at stake was the political direction of Corcyra—and the political destruction of the opposing faction) and he was successful. Even these authoritative judgements, however, were not received as final, and in the event Peithias was murdered by his opponents. Exiles, even working openly against the interests of their city by encouraging foreign invasion, were able to represent themselves still as true patriots, rejecting the claims of authority and validity made by those in control of their city.199

2.4. MISCARRIAGE AND MANIPULATION: ALCIBIADES THE PROXENOS Proxeny patterned the behaviour of individuals. In pursuit of prestige, individuals conformed to the paradigm of the proxenos, but in ways which were limited, defined by the specific, and not too taxing, services associated with proxenia, and conventionally circumscribed by the greater claims of their native poleis. The relationship which proxenia embodied could, however, also miscarry and fail, despite the efforts of participants. This potential for failure is best illustrated by Thucydides’ account of Alcibiades’ interactions with the Spartans between the Peace of Nicias and the Athenian alliance with the Argives (421/0 bc), which highlights the themes of prestige and elite competition discussed in this chapter. Alcibiades is introduced in a complex passage in Thucydides’ history as a frustrated and embittered would-have-been proxenos of the Spartans: 199

The best example is Alcibiades, Thuc. 6.92.

Proxeny in its Political Contexts

143

ŒÆa ØÆ Å c ØÆçæa Z ø H ¸ÆŒ ÆØ ø æe f ŁÅ Æ ı, ƒ K ÆE Ł ÆØ Æs ıº Ø ºFÆØ a   a PŁf K Œ Ø . qÆ b ¼ººØ  ŒÆd ºŒØØÅ › ˚º Ø ı, I cæ #ºØŒ Æ fi b Ø  J  ‰ K ¼ººfiÅ º Ø, IØÆØ b æª ø Ø · fiz KŒ Ø b ŒÆd ¼ Ø  r ÆØ æe f æª ı Aºº åøæ E , P  Ø Iººa ŒÆd çæ ÆØ çغ ØŒH M Æ ØF, ‹Ø ¸ÆŒ ÆØ ØØ Øa ˝ØŒ ı ŒÆd ¸åÅ  æÆÆ a   , %Æıe ŒÆ  c ÅÆ  æØ   ŒÆd ŒÆa c ƺÆØa æ Æ b sÆ P ØÆ  , m F 

ı I Ø   ÆPe f KŒ B ı ÆPH ÆNåƺı Ł æÆ ø Ø  E I Æ ÆŁÆØ. Since the Spartans were now thus at variance with the Athenians, those at Athens who wished to dissolve the treaty began immediately to exert themselves. Alcibiades, the son of Kleinias, was one of these. He would have been then still a youth in another city, but had a prominent position at Athens because of the reputation of his forebears. To him the alliance with Argos really did seem the better course, but this course of action also agreed with his own emulous pride, because the Spartans had acted through Nicias and Laches in establishing this treaty, and had passed him over because of his youth and had not honoured his ancestral proxenia, which had been renounced by Alcibiades’ grandfather, but which Alcibiades himself had been trying to renew by caring for the Spartan prisoners captured on the island.200

In his own estimation Alcibiades had behaved, especially in relation to the Spartan prisoners taken at Pylos, in the manner expected of a proxenos, out of a desire to renew this connection which had been defunct for a generation (that is, he wished to be formally named Spartan proxenos).201 He was frustrated, however, by the failure of Spartan delegations to treat him as their proxenos and make use of him in conducting negotiations, turning instead to Alcibiades’ competitors within the polis, individuals whom he is later made to describe as his political enemies. The Spartans’ reason for their neglect of him was his youth and Thucydides, in his authorial persona, makes it clear that their assumption that being a neos would render him incompetent in this context was reasonable, even if it did not hold because of the special circumstances prevailing at Athens in Alcibiades’ case. This neglect ruptured the nascent relationship. From Alcibiades’ perspective, it constituted a crucial breakdown in

200 201

Thuc. 5.43–5. Daux (1937) minimizes the importance of this formal act of recognition.

144

Proxeny and Polis

reciprocity. Thucydides nicely observes Alcibiades’ double motivation in preferring an Argive alliance—his apparently sincere assessment of what was best for Athens, which coincided with and, Thucydides’ syntax slyly suggests, was shaped by, his personal pique at this perceived Spartan insult against himself.202 The emphasis on Alcibiades’ pride and competitiveness in this passage highlights the reasons he sought appointment as Spartan proxenos in the first place—because of the prestige within Athens that being the Spartans’ main connection there would bring. In this competitive context, this Spartan failure to reciprocate was felt as a public dishonour (Alcibiades is later made to describe it as a mark of atimia).203 The sequel to this episode illustrates the degree to which this relationship could miscarry (Thuc. 5.45). When the Spartans, made anxious by the proposed Athenian alliance with Argos, dispatched new envoys more amenable to the Athenians and fully empowered to negotiate, Alcibiades intervened. He promised to obtain for them what they most wanted—the return of Pylos—on the condition that they obeyed him in their negotiations (specifically that they did not announce to the assembly that they were empowered to agree terms). In other words, Alcibiades acted ostensibly as their proxenos, seeking to detach them from their pre-existing relationship with Nicias by promising results and offering tactical instruction. On this occasion, unlike the former, the Spartan envoys accepted his performance of the role of proxenos (it may have helped that one of the envoys, Endios, was Alcibiades’ xenos). In the event, however, Alcibiades is described as betraying their trust to devastating effect, denouncing them for bad faith before the assembly for acting as he had instructed them to, and, as a result, derailing the negotiations for peace and cementing the Athenian preference for an Argive alliance.204 Alcibiades—or rather Thucydides’ version of him—is remarkable for the way in which he then represents his actions. In a speech which Alcibiades delivers before the Spartan assembly, he artfully misrepresents the events as they are presented in Thucydides’ direct account. 202

203 Hornblower (1991–2008), ad loc. Thuc. 6.89. Herman’s reading of this episode, Herman (1987), 147–9, takes the xenia between Alcibiades and Endios as the key, arguing that this enabled Alcibiades to persuade the Spartan envoys to betray their city’s interests. However, as an account of the envoys’ motivation it is unconvincing and contradicts Thucydides’ assertion that Alcibiades plotted against and tricked them by promising something he had no interest in delivering. See Hornblower (1991–2008), ad loc. 204

Proxeny in its Political Contexts

145

Like other speakers before the Spartan assembly in our historians (and presumably life), Alcibiades begins by making reference to his claim to proxenia and, hence, the persuasiveness which this, as evidence of his concern for Spartan interests, should confer. Unlike other speakers, however, Alcibiades must deal with the fact that the relationship has hardly been felicitous for the Spartans. I ƪŒÆE æd B KB ØƺB æH K A N E , ¥ Æ c å Eæ a ŒØ a fiH   ø fi ı IŒæÅŁ . H ! KH æª ø c

æ Æ H ŒÆ Ø ªŒºÅÆ I Ø  ø ÆPe Kªg ºØ I ƺÆ ø KŁ æ ı A ¼ººÆ  ŒÆd æd c KŒ — ºı ıçæ . ŒÆd ØÆ ºF  ı æŁ ı  E æe ŁÅ Æ ı ŒÆƺºÆ Ø E b KE KåŁæE  ÆØ Ø' KŒ ø æÆ  , Kd b IØ Æ

æØŁ  . ŒÆd Øa ÆFÆ ØŒÆ ø  ! KF æ  a Æ Ø ø ŒÆd æª ø æÆ  ı ŒÆd ‹Æ ¼ººÆ K Å Ø Å E Kº  Ł · It is necessary first of all to speak to you about the prejudice against me so that you may not give a less favourable hearing to matters of public concern because of your suspicions against me. Seeing that my ancestors had renounced their position as your proxenos on the grounds of some complaint they had, in order to renew this relationship I performed services for you in other matters and especially in regard to your misfortune at Pylos. And during this period when I was continually zealous (prothumos) towards you, you, in making peace with the Athenians through my personal enemies, conferred power upon them and brought dishonour on me. For this reason the harm which you suffered when I promoted the interests of the Mantineans and Argives and in the other affairs which I opposed you, you did not suffer unjustly.205

Alcibiades here paints himself, and his grandfather, as faithful proxenoi, drawing directly on the language of the proxenos-paradigm (especially in the reference to his ‘continual zeal’, ØÆ ºF  ı

æŁ ı) to emphasize his own commitment to the role. He skates over his own betrayal and championing of an alternative Argive alliance, and his grandfather’s renouncement of his proxenia in the anti-Spartan political climate of 462–460 bc, by placing the blame for the rupture in both cases firmly on the Spartans themselves. In his own case he does this by emphasizing the Spartan decision not to negotiate through him, describing it in highly charged terms; in that of his grandfather he does this by referencing a vague (certainly

205

Thuc. 6.89.

146

Proxeny and Polis

fictional) offence of the Spartans as the reason for the rupture.206 In his actions, Alcibiades was not a typical proxenos, but he is important as an example of how the relationship involved could fail and illustrates the agency which individuals retained, working within the institutional structures and language of proxenia.

2.5. CONCLUSION Together these studies—of proxeny grants in poleis, proxenoi in their own poleis, and proxenoi of other poleis—add up to an explanation of how proxenia functioned, of why members of these communities actively participated at different levels in reinforcing and performing the role expected of the proxenos. Proxenia enabled cities to reach beyond the limits of their own citizenry to establish and encourage strong links with outsiders as liminal figures who were simultaneously both (quasi-)citizens with established identities and privileges in relation to the honouring polis and citizens resident and active in their own political communities. In many cases these links were surely derived from interpersonal relations between the honorands and individual citizens, but they were deliberately transformed by the rhetoric of the honouring polis in its authoritative descriptions of the honorands and the privileges it granted them. The transformation was successful, as we can see in a number of cases where individuals retained relations with a succession of different regimes, because the role of the proxenos, the proxenosparadigm, its expectations and obligations, was so well defined and understood. In this, although the decrees of the honouring city had their established part to play in constructing and communicating the relationship, it was the home polis of the proxenos which was most effective in inculcating the ideology of proxeny. It was in his own city that the proxenos (or the individual hopeful of becoming one) came most frequently and directly into contact with ideas about what proxenoi should be like or do as he took part in the assembly in which grants of proxenia were so frequently proposed and debated. Similarly, it was within his own political community that most was 206 ŒÆ Ø ªŒºÅÆ ‘covers this up’ for a Spartan audience, Andrewes, HCT, ad loc.; Hornblower (1991–2008), ad loc.

Proxeny in its Political Contexts

147

probably made of his association with another city as, in the competitive, often factional process of polis politics, proxenoi within a polis could in certain circumstances come to symbolize internally the other poleis with which they were connected. This context for the communication of proxenos ideology is also important for understanding why proxenoi were not the foreign agents they were sometimes accused in hostile speeches of being. Poleis, in making proxeny decrees, were aware that they were communicating primarily internally, and emphasis was therefore frequently placed, in praising proxenoi, on the primacy of the proxenos’ own polis. The attraction of proxenia is also best understood in the context of the proxenos’ home polis, as a source of prestige that marked him as a member of the elite of his city, part of a broader trans-civic elite.207 This prestige, which individuals sought to capitalize on by communicating their position, arose from their status as intermediary figures, individuals formally able to facilitate and mediate interactions between members of different communities because of their standing in each. They derived personal importance and status from being connected to, or rather connecting, different groups and communities, in much the same way as the old, heavily aristocratic xenoi had derived prestige, and the kings’ philoi continued to do so.208

207

See Herman (1987), 162–5.

208

Herman (1987), 34–40.

3 Proxeny Networks The Horizons of Poleis

Individual poleis viewed their proxenoi en masse as geographically distributed networks. This wider perspective is crucial not just as a context for interpreting individual grants but also for understanding how proxenia worked on this inter-polis level, which is the focus of Chapters 3 and 4. The best evidence for this perspective, despite the usually fragmentary state of their preservation, are the inscribed lists in which poleis gathered and presented their proxenoi, and these are used in this chapter to reconstruct the parameters of proxeny networking (the lists themselves are presented and analysed in the Appendix). From these lists, which have not been systematically considered before, it becomes clear that the proxeny networks of even very small poleis could be extremely large and wide-ranging, and that all networks were in constant need of renewal. At the same time, however, we also see an enormous degree of variation in local rhythms of granting and patterns of distribution, which complicate any attempt to read simplistic narratives into this material. Proxeny lists present us with a dynamic picture of what this stereotypical honorific institution meant in the world of the Greek cities, and how it resulted, despite the fixed and narrowly defined content of proxenia, in vibrant, polis-specific networks. The scale of activity these texts reveal, however, also raises questions concerning the broader function of proxeny networking and these forms of inter-polis honorific interchange, which are explored in more detail in Chapter 4. Grants of proxenia were not made in a vacuum. The individual decrees that we possess are always explicitly constructed as responses to services (generally a series of services) rendered by the honorand to

Proxeny Networks

149

the citizens of the granting city while they were engaged in activities of different kinds, typically in the context of the proxenos’ own polis. Collectively, therefore, networks of proxenoi—and the proxeny lists which attest them—can be read as indexes of interaction, vividly illustrating the density and frequency of inter-polis contacts for particular cities. Since proxenoi facilitated a wide range of different sorts of interaction, networks of proxenoi potentially also map economic contacts which citizens from particular cities had engaged in prior to the inscription of the list, as well as those specifically concerned with inter-polis relations. However, although proxeny networks are explicitly constructed as reactive, they were not simply reflective of interaction. They were also the products of self-conscious processes of selection. As the principal institution by which poleis sought to mediate and facilitate interaction with each other, networks of proxenoi were the most important means by which poleis understood and constructed their connectivity, emphasizing their links with particular poleis and regions, and their own position in relation to them. This complicates, but also enriches, our readings of proxeny networks, and especially the way in which they were organized and presented on monumental lists of proxenoi. In the ebb and flow in annual rates of proxeny granting, which some of these proxeny lists reveal, we see both the hum and the buzz of the Mediterranean world with its constant Brownian motion of interaction, and the deliberate efforts of poleis to assert their position within it.1 First, however, it is necessary to explain the theoretical frameworks used here to study ancient networks of proxenoi. In recent years, scholars of the ancient world have increasingly turned to theoretical approaches developed in Social Network Analysis to identify and understand the structural dynamics of networks in the ancient world.2 A number of these concepts and descriptive categories emerge as useful in relation to networks of proxenoi. In particular, Mark Granovetter’s work illuminating the distinction between ‘strong’ and ‘weak’ ties in networks encourages us to concentrate on

1 Hum and buzz: J. and E. Fentress (2001), 217, reviewing Horden and Purcell (2000). Brownian motion: Horden and Purcell (2000), 142 and 150. 2 A range of different approaches can be found in Malkin, Constantakopoulou, and Panagopoulou (2009); see also Ober (2008), 134–51; Malkin (2011); on social networks and religious change, Collar (2013).

150

Proxeny and Polis

the structural role performed by proxenoi.3 In this context, the strength of a tie is understood as a function of the time invested in it, its emotional intensity, its intimacy, and the nature of reciprocal services involved.4 Strong ties are understood as exerting a greater influence on the actors involved, but weak ties are paradoxically more important in spreading information throughout a network in that they act as ‘bridges’ between otherwise distantly connected ‘clusters’. In this scheme, proxenoi represent archetypally weak ties, bridging clusters of individuals bound by mutual strong ties (poleis). Part of the value of this, however, is the contrast which it draws out with the way in which proxeny decrees seek to construct these links—in terms of a history of interaction, emotional content, and reciprocal services—as stereotypically strong ties. There are, however, problems relating both to the nature of our data and the institution of proxenia itself that limit the application of the quantitative methods which are central to Social Network Analysis. The most intractable difficulty is the fact that quantitative analysis requires complete data on the relevant relations of all actors in the network, or at least of a representative sample—a rarity in ancient history.5 For only a handful of poleis can we confidently reconstruct even the basic structure of their networks of proxenoi at a particular moment in time, much less the networks of all relevant poleis at that point (which, given the way in which these networks overlap, would swiftly become all poleis). In addition the asymmetric nature of this relationship—involving two different kinds of actor, poleis and proxenoi—makes it unlikely, even if we had a complete dataset, that many of the more interesting structural properties which quantitative analysis can reveal would emerge in relation to proxeny networks. Analysed as a link between poleis, proxenia was an unidirectional, non-reciprocal tie: the fact that A named a citizen of B its proxenos did not mean that B possessed a link of proxenia to polis A, and neither required nor was likely to provoke polis B to make a corresponding proxeny grant.6 Moreover, this idiosyncratic combination of different types of actors with a highly specific institutional

3

Granovetter (1973); (1983); Kadushin (2012), 30–1. Granovetter (1973), 1361. 5 Wasserman and Faust (1994), 33–5; Kadushin (2012), 17; Malkin (2011), 25; Graham (2006) is an analysis of an unusually complete ancient dataset. 6 Wasserman and Faust (1994), 72–3. 4

Proxeny Networks

151

content (the proxenos-paradigm) meant that it was very unlikely that the proxenoi of different poleis could combine to provide a path between communities which were not directly connected. In other words, unlike links of private friendship, links of proxeny between poleis A and B and B and C did not provide an indirect path from A to C. Thus, the proxeny relations of different poleis are unlikely to reveal the interesting structural properties—in particular, the influence and importance of particular actors in serving as indirect links between different clusters—which the quantitative methods of network analysis are particularly useful for elucidating.7 Although the categories and insights of network theory are valuable in informing the approach adopted for dealing with networks here, and are indirectly important through their influence on other theoretical frameworks which I make use of in the next chapter, the peculiarities of proxenia, and the nature of our evidence for it, oblige us to devise specific analytic tools to explore the structural features of proxeny networks. One analytic category is particularly useful in bringing out the similarities and differences which proxeny networks exhibit as well as their relation with the landscapes in which they were selfconsciously embedded. I label this category the ‘local region of primary interaction’. In relation to proxeny networks, I define this as the region in which either the majority or the largest minority of the proxenoi of a polis were located, that is, the region with which a city is presented as being most densely interconnected by its network of proxenoi. As such, this category is indebted to work which has been done on economic regions as the areas within which the bulk of economic activity is understood as taking place8 and the attempts of scholars to identify particular regions on the basis of the records of economic transactions or the archaeological distribution of certain sorts of coin or pot.9 As a category, however, ‘local region of primary interaction’ allows us to explore and compare the full range of interactions which proxenia reflected and was used to facilitate— both private and public, as the decrees themselves positively assert, and political and cultural, as well as economic interactions. All the

7 For an introduction to different ways of measuring the structural importance of nodes (degree centrality, closeness centrality, betweenness centrality, and eigenvector centrality) see Jackson (2008), 37–43. 8 e.g. Migeotte (2004), 616; Reger (2011), 368. 9 Reger (1994); (2011); cf. Vlassopoulos (2007), 167–8.

152

Proxeny and Polis

proxeny networks I explore in this chapter exhibit a marked local region of primary interaction, but different networks vary significantly in the extent to which they are confined to that region, and the way in which they reach outside it. This category also encourages us to explore the way in which very different regions were shaped both consciously and unconsciously by the actions of individual cities, in response to, but not determined by, their geographical setting, with all the potential for different views of regions which that entailed.10

3.1. SOURCES FOR PROXENY NETWORKS Unlike individually inscribed decrees, which represent a problematic and often unrepresentative sample of the decrees that were made, proxeny lists provide a firm foundation for the study of proxeny networks.11 Whereas individual decrees were inscribed selectively, the rhetoric and raison d’être of proxeny lists was completeness. Thus, even though many lists of proxenoi are fragmentary, after identifying the ordering principles of each inscription, the rich, contextualized datasets which these present can be used to explore different poleis’ networks of proxenoi in great detail. There are two principle types of proxeny list with different principles of layout and organization, which shed complementary light on proxeny networks—proxeny catalogues and chronological lists of proxenoi. The purpose of proxeny catalogues was to list all the proxenoi recognized by a particular polis at a particular time. In other words, they were intended to present a structured account of an entire network of proxenoi. Even though these catalogues are preserved only in fragments, by identifying the organizing principles of the original monuments we can obtain crucial evidence of the size and regional distribution of different cities’ proxeny networks. Thus, in 10 For the importance of the study of processes of region formation, see Vlassopoulos (2007), 166–8, but his interest is primarily in regions as an alternative unit of analysis to the polis, whereas proxeny networks reveal regions as they are constructed by and of poleis. See now Mackil (2013), who argues convincingly these zones created by the mutual interactions of different kinds among a group of poleis were crucial for the emergence of koina like the Boiotian League. 11 See Introduction, Section I.3.

Proxeny Networks

153

the case of the proxeny catalogue of Eresos, we can conservatively estimate the original network at some 175 proxenoi—a far from unparalleled figure—and observe that probably two-fifths of these proxenoi were appointed by the Eresians at other, neighbouring cities on Lesbos.12 The form and ordering principles of catalogues tend to be consistent and clear: the names of proxenoi are always grouped according to their polis of origin, listed together in the nominative case with their patronyms, one after another, usually beneath their ethnic, which is used as a heading in the nominative plural. The organization of these different polis groupings amongst themselves seems usually to be geographic, with poleis of the same region (from the perspective of the granting polis) grouped together. Chronological proxeny lists, by contrast, were intended to collect and present not all the proxenoi of a polis at a particular moment in time, but rather all proxenoi appointed by the polis on a particular occasion or within a particular administrative period, usually a year or six months. In function, they thus closely resemble individual decrees, and this is reflected in the grammatical case often used of the proxenoi listed: the accusative (x is to be proxenos) or dative (the recipient of a grant of proxenia) in contrast with the nominative case typical of proxeny catalogues.13 However, unlike the individually inscribed decrees which they seem intended to replace, chronological lists provide us with complete or near-complete sets of data for the period in question.14 Although chronological lists do not usually allow us to estimate the total size of networks at a given moment, as catalogues do, they nonetheless give us snapshots of proxeny networks in motion. In providing us with accurate data for annual rates 12

Mack (2012), 224. The only certain exceptions to this usage are the lists of theōroi from different cities who became proxenoi for whom the nominative is used: e.g. Appendix nos. 14.1–13 (Samothrace) and 17 (Boiotia). 14 Where chronological lists purport to present the proxeny grants made on a particular occasion, such as the theōroi lists of Samothrace (Appendix no. 14), or a particular year, as in the case of the Histiaia list (Appendix no. 8), we should expect these to be complete. By contrast errors and omissions are more likely to affect continuous series of chronological proxeny lists spanning a number of years, such as the proxeny lists of Epidauros (Appendix no. 6.1–3) or Delphi (Appendix no. 5). The occurrence of a handful of individual proxeny decrees at Delphi from the period covered by the proxeny lists probably reflects, in part, inconsistencies of administrative practice, but, in contrast to the large numbers of decrees inscribed individually at Delphi prior to this point, it emphasizes rather than contradicts the point that individual decrees were replaced by this proxeny list. 13

154

Proxeny and Polis

of granting, chronological lists reveal the rhythms of proxeny networks, and their inherent year-on-year variability. Chronological lists, unlike most catalogues, often also allow us to distinguish between grants given to individuals and grants made to groups, and thus to begin to unpick the different sorts of interaction which contributed to the continual renewal of proxeny networks. For instance, in the great chronological lists of Delphi, which give us twenty-two years’ worth of chronological lists divided by half year, it is possible to identify a number of groups of theōroi dispatched by their poleis to Delphi and honoured there with proxenia. Although the distinction between these two categories, chronological list and catalogue, is useful, there was often a significant overlap. In order to remain complete, catalogues had to be updated over time, and this could take the form of a subsequent chronological list (as at Karthaia), or the insertion of additional proxenoi in spaces left in the original text of the catalogue. The catalogues of Chios, and especially Eresos, take on a palimpsest-like quality as a result of these successive phases of inscription. The latest insertions in the Eresian proxeny list, indeed, in the accusative case of the proxeny decree rather than the nominative of the catalogue, reflect the transformation of this monument from ordered catalogue to a place simply to record new appointments of proxenoi.15 Similarly, a sequence of chronological lists maintained for a sufficient period of time (i.e. a generation or more) would come to represent, like catalogues, a virtually complete record of a polis’ network of proxenoi. Proxeny lists, both catalogues and chronological lists, illuminate the proxeny networks of their respective poleis. The question, however, is how far these networks of proxenoi were representative of the proxeny networks of poleis in general. Was the decision to inscribe proxeny lists of these kinds itself an indication that the network in question was unusual (i.e. exceptionally large) in a way which should prevent us from using it as the basis for a discussion of proxeny networks more generally? The decision to construct and inscribe a catalogue at a moment in time was a marked and context-specific act. This is true, albeit to a lesser extent, even of chronological lists, for all that they represented a cheaper form of publication than individual decrees. However, the

15

Appendix no. 7; see Mack (2012), 218.

Proxeny Networks

155

significant number of poleis which are attested as having inscribed proxeny lists clearly shows that this was not an exceptional practice in itself, especially given how much more vulnerable these individual monuments would have been, collectively, to loss and destruction than the far more numerous individually inscribed decrees. In particular, fragments of fully-fledged proxeny catalogues are preserved from Astypalaia, Chios, Eresos, Karthaia, Kleitor, and perhaps Tenos and Kalchedon.16 Chronological lists are known for Aitolia, Anaphe, Delphi, Epidauros, Lousoi, Tenos, Thera, and Samothrace, as well as two other unknown cities.17 In addition, references in other decrees attest to the existence of lists (probably catalogues, being subsequently supplemented) at Megara, Lakonian Epidauros Limera, Kyzikos, Assos, Gryneion, and Bargylia.18 The very wide range of poleis represented here—in terms of material resources and circumstances, as well as size—should give us confidence that this material does at least reflect the range of possibilities open to different poleis. In any case, ‘the average proxeny network’ is likely to prove just as elusive and hard to identify as ‘the average polis’. Nonetheless, at the same time as reading these lists of proxenoi for the quantitative information they contain, it is important to be aware that they operate as statements rather than simply as records—a function which other sorts of document performed. In particular, 16

Appendix nos. 3, 4.1, 7, 9, 10.1, 11, and 15.1. Appendix nos. 1.1–8, 2.1–3, 5, 6.1–4, 8, 10.2, 12, 14.1–13, 15.2, 16.1–4, 17, and 18. Appendix no. 9, the proxeny list from Kleitor, may conceivably have been a chronological list; no. 4.3, from Chios, is likely to be another. 18 IG VII 5 (Megara, c.300); IG V 1 932 (Epidauros Limera; undated); Syll.3 644–5 no. VII (Kyzikos, part of the dossier of decrees for Nikon at Seleukeia, c.172 bc); I. Thasos 170 ll. 13–16 (Assos, inscribed at Thasos c.130 bc); IG XII 4 129 XI ll. 80–1 (Gryneion, part of the dossier for Nikomedes at Kos, 306–301 bc); I.Iasos 608 ll. 29–30 (Bargylia, inscribed at Teos, 270–260 bc) and I.Priene 47 (Bargylia, inscribed at Priene, c.200 bc; this clause is also plausibly restored in another decree of Bargylia, found there, I.Iasos 612, c.127 bc). There are also references to a proxeny list at Eresos from the latter part of the second century bc, which attests a subsequent list, IG XII suppl. 139 ll. 92–4 (inscribed at Miletos, mid-second century bc) and also for Chios, SEG 12 390 (c.320 bc). As one might expect given the fact that proxeny lists usually seem to replace rather than supplement the inscription of individual decrees, these references tend to occur in texts of decrees inscribed by other communities. Other references to the archiving of proxeny decrees and decrees in general include IG I3 155 (Athens, late 430s–420s bc) and XII 4 129–30 no. XII l. 95 (Phokaia, from the dossier of Nikomedes at Kos, 306–301 bc), both specifying whitened boards, and perhaps I.Magnesia 34 (Phokian League, inscribed at Magnesia on the Maiander, 208 bc, though in this case the focus of the order to archive may be the broader asylia decree). 17

156

Proxeny and Polis

the very fact that catalogues of proxenoi could be assembled at poleis, for which we see little or no evidence of systematic inscription of individual decrees, clearly shows that archival records of proxeny grants were regularly maintained and accessible, and probably that some record was maintained of which proxenoi were still alive and active. Although the particular circumstances in which each monument was erected are obscure, their symbolic force would have been clear. Such monuments asserted the importance of the polis which produced them, by stressing the number of important individuals who had sought to befriend and aid the city, as well as the cities to which these proxenoi linked the granting polis. In creating these inscribed monuments, cities were self-consciously mapping their social world, but, most of all, emphasizing the central place they occupied within it.

3.1.1. Case study: the chronological proxeny list of Histiaia The chronological list of Histiaia, which records thirty-two grants of proxenia made in the year 264/3 bc, represents a compact case study, a sort of proxeny network in miniature.19 It illustrates how these lists of proxenoi can be read, in context, to reveal the communal horizons of poleis and the different interests and relations which cities had with communities in distinct regions. It also makes it clear, however, that networks of proxenoi and the inscriptions which document them were the product of both general, recurrent patterns of interaction and specific historical events. What is immediately striking about the proxenoi collected in the text from Histiaia, is their wide geographical distribution (see Map 1). The poleis that are represented—Syracuse and Tarentum in the west, Kalchedon in the north-east, Sidon in the east, and Cyrene to the south—define the outer edges of the Aegean world to which Histiaia belonged, and, altogether, attest to a remarkable range of connections for this polis on the northern tip of Euboia to have forged in a single year. At the same time, it is notable that, even within this far-reaching set of proxeny grants, almost two-thirds of the proxenoi listed (up to 19

Appendix no. 8 (= IG XII 9 1187).

Proxeny Networks

157

Map 1. Thirty-two proxenoi appointed by Histiaia in the year 264/3 bc (Appendix no. 8). Proxenoi apparently appointed individually are represented by dots; proxenoi appointed as part of a group, according to my analysis, represented by joined squares. The inset map shows Histiaia’s region of primary interaction (central and northern Greece).

twenty of these thirty-two) came from a relatively narrow geographic area, roughly comprising central and northern Greece (see inset map). This area thus seems to have constituted Histiaia’s local region of primary interaction, the area within which most of its interactions with other poleis occurred. This regional differentiation of interactions, however, is not just visible in the clustering of proxenoi in this list. It is also reflected in the sort of grants which were made: the relative distribution of ‘individual’ to ‘group grants’. The majority of proxeny grants in general were made to individuals for reasons which it can be hard to identify with certainty. Indeed, the specific services which prompted a proxeny grant are usually represented as being only the last in a series of deeds which together made an individual worthy of this honorific status. A significant minority of grants, however, were made collectively, to groups of individuals

158

Proxeny and Polis

usually from the same community.20 While some of the individual recipients of these group grants may have been personally responsible for services in the past, the immediate reason for the grant was their participation in an embassy or mission from their home polis, or their support, within their own city, for a public mission from the granting polis. Group grants thus indicate a particularly intense form of public contact between two communities. In other proxeny lists group grants are often indicated by the use of a single, collective ethnic in the plural for multiple proxenoi.21 In the Histiaian text, by contrast, all the recipients of proxenia are listed individually with their own ethnic in the singular. However, careful reading allows us to identify a number of probable group grants. In particular, since it can be observed that grants in this list have not been arranged by ethnic but in the order in which they were made (or at least with individuals appointed at the same time grouped together), it becomes highly likely that individuals from the same polis (or other political organization) who are juxtaposed within it were appointed together.22 Working on this principal, it is possible to identify five collective grants in this list, made to different groups: a group of three citizens from Echinos (a polis on the mainland, opposite Histiaia);23 a group of three Amphictyonic hieromnēmones in their year of office;24 a group of two Athenians;25 and two different groups of Macedonians, one of two Macedonians from Thessalonike (perhaps including another Macedonian, juxtaposed on this list but lacking an additional polis ethnic), the other probably comprising a Macedonian from Aigai and an individual, without an ethnic, 20 Here it is important to distinguish ‘group grants’ made to groups of unrelated individuals from the same community or political organization for specific reasons (usually relating to inter-polis relations) from grants made to multiple members of the same family simultaneously which are not functionally different from individual grants. 21 See Appendix no. 10.1 (Karthaia). 22 The Aitolian, Locrian, and Herakleian juxtaposed in ll. 21–3 were clearly appointed at the same time in the context of their service as hieromnēmones of the Delphic Amphictyony in 264/3 bc, which thus gives us the precise date of this list. The fact that the first two of these individuals are identified with only their federal ethnics (without a further polis designation) underlines the fact that these individuals were appointed proxenoi in the context of their service on the Amphictyony, as representatives of federal states. 23 ll. 8–10; note that an individual grant of proxenia was thus also made to a separate citizen of Echinos in the same year, l. 28. 24 25 ll. 21–3, see n. 22. ll. 12–13.

Proxeny Networks

159

identified only as Aristoboulos, son of Persaios.26 The omission of Aristoboulos’ ethnic seems unlikely to be a mason’s error.27 Instead, it was probably felt that the name of Aristoboulos’ father more clearly communicated the political context in which Aristoboulos was appointed proxenos and expected to be active—if, as seems highly likely, Persaios was the influential courtier of the same name at Antigonos Gonatas’ court.28 Since Aigai was one of the most important royal centres of the Macedonian kingdom, this pair of proxenoi thus reflects direct contact with the Antigonid court, as the proxeny grants to the two Macedonians from Thessalonike probably also do (specifically inter-polis relations between Histaia and Thessalonike would be a possible but less likely context for this grant). The first thing to note is that all five of these group grants were made within Histiaia’s local region of primary interaction. They thus seem to illustrate the relatively restricted geographic range of these sorts of intense political contact in contrast with the breadth of Histiaia’s wider horizons. At the same time, this intriguing combination of inter-communal contacts is an invitation to attempt to reconstruct the complexities of the Histaians’ foreign relations in this particular year. As we have no continuous narrative source for the early third century bc, we are reliant on scattered epigraphic documents and literary anecdotes to reconstruct the eventful history of mainland Greece during this time. This was a period dominated by the establishment of Antigonos Gonatas, son of Demetrios Poliorketes, as king of Macedon, but also by the challenges to his authority which he faced, especially in the form of an invasion by Pyrrhus of Epirus (274 bc), and the Chremonidean war (268–263/2 bc) in which Athens, 26

ll. 37–8. (the Macedonian without a civic ethnic is l. 39); ll. 30–1. Knoepfler (2001a), 280 n. 58 suggests that this entry in l. 31 may have been a mis-inscription of a similar name in l. 29 (in which case, why was it not erased?), but this is unnecessary. 28 On Persaios of Kition, the remarkable philosopher-courtier, see RE Persaios (1) and Erskine (2010). For Persaios’ influence at court (and interest in Euboia) at about this time see Diog. Laert. 2.143, where he is credited with dissuading the king from restoring democracy at Eretria. Persaios’ permanent position at court clearly made him too famous for his now irrelevant ethnic (which would not, however, have been the most distant listed here) to be thought either necessary or helpful in this reference to his son. For discussion of the other examples of the omission of ethnics in proxeny grants when the individual was involved in court service or too famous to require it for identification, see Chapter 1, 53 n. 104. 27

160

Proxeny and Polis

Sparta, and the Ptolemaic kings formed the core of a coalition of Greek cities against Antigonos.29 In an article published in 1995, Denis Knoepfler attempted to reconstruct how three Euboian poleis, Eretria, Chalkis, and Histiaia, responded to these events by reexamining the records of the Delphic hieromnēmones who were sent by different poleis to sit on the Amphictyonic council. By working on the principle, established by Beloch, that the Aitolians, who controlled the Delphic Amphictyony in this period, did not permit cities dominated by the Macedonian king to contribute hieromnēmones to it, Knoepfler used a reanalysis of the sequence of Amphictyonic decrees to reconstruct the chronology of these cities’ relations with Macedon.30 According to this, Eretria, Chalkis, and Histiaia were all liberated from Antigonid control in the second half of the 270s bc, but while both Eretria and Chalkis were subjugated once more by the Macedonians in or before the spring of 267 bc, at the start of the Chremonidean war, the city of Histiaia remained independent and continued to fill the Euboian seat on the Amphictyonic council until it was garrisoned by the Macedonians in 251/0 bc with only a brief interruption in 261/0 bc.31 The group grants which can be isolated in the proxeny list of Histiaia appear to shed further light on these events from the perspective of this Euboian polis. First of all it is interesting that in the year 264/3 bc, shortly before the capitulation of the Athenians to Antigonos, when the situation for the Athenians must already have appeared dire, the Histiaians nonetheless engaged in some form of official contact with them, and even named two Athenians who were involved in this contact as their proxenoi. Perhaps these individuals represent an Athenian delegation to Histiaia, though they could also have helped a Histiaian delegation to Athens. In any case, this grant probably reflects the desire of the Athenians and Histiaians, as two poleis still trying to assert their independence from Macedon, to maintain relations with each other, and perhaps efforts to cooperate militarily were involved as well. In this context it is not unlikely that another recipient who occurs on this list, Herakleitos of Halikarnassos,

29 On these events, see Will (1979–82), vol. 1, 212–33; for a more up to date account of the Chremonidean war, especially as regards the Athenian part in it, see Habicht (1997), 142–9. 30 Beloch (1927), 385–426. For the example of Athens, see Habicht (1997), 126–7. 31 Knoepfler (1995).

Proxeny Networks

161

was appointed as a result of having served as an ambassador from King Ptolemy, the ally of the Athenians and Spartans in the Chremonidean war.32 The grants of proxeny for the two Macedonian delegations, however, suggest that the reason the Histaians were able to continue to participate in the Delphic Amphictyony in subsequent years was that they managed to reach some sort of accommodation with the Macedonian king, which left them with a form of independence (presumably, unlike the cities of Eretria, Chalkis, and Athens, they were not forced to receive a Macedonian garrison). Their continued participation in the Amphictyony may also, however, have had something to do with direct contact between Histiaia and that body implied by the group grant of proxenia to three hieromnēmones in this list (is it an accident that the Aitolian heads this list of three?). Other interpretations of these group grants are possible, and, without further information, certainty is impossible—and much depends on whether or not these group grants were listed in the order in which they were made. In any case, the proxenoi in this inscription depict the Histiaians successfully negotiating a position for themselves which allowed for some freedom of action between Athens, the kings of Macedon and Egypt, and the Delphic Amphictyony (more probably, the Aitolian League which controlled it). The proxenoi appointed in this list, however, also reflect wider, probably less highly-charged interactions which the Histiaians underwent in other contexts as well. In a well-known article, Louis Robert compared the wide distribution of Histiaia’s proxenoi with the distribution of its coinage, arguing that both reflected the reach and interest of the Histiaians in trade.33 Although the comparison between coins and proxenoi is not convincing,34 the distribution of proxenoi along the major sea route from the Black Sea to Cyrene, as 32

This suggestion was made by Swinnen (1970), though without the Chremonidean context. Herakleitos was honoured with a statue at Oropos, dedicated by his brother, a rare honour (I.Oropos 415). This Herakleitos also occurs in a contemporary proxeny catalogue from Chios (Appendix no. 4.2 b l.11–13), though whether he was appointed in relation to intermediary services at the Ptolemaic court or at Halikarnassos is unclear. Herakleitos was probably the poet and Halikarnassian xenos at whose death Callimachus wept, C. Ep. 36. 33 Robert (1951). 34 Not because of the geographical mismatch between the distribution of proxenoi and tretrobols in coin hoards, contra Marek (1977), but because the large issues of tetrobols which Robert studied were minted nearly a century later, to pay soldiers in the context of the Third Macedonian War, reflecting the military rather than

162

Proxeny and Polis

well as at the major ports of Syracuse and Tarentum, does strongly suggest that long distance trade was a decisive factor in the appointment of many of these proxenoi, and it seems likely that these individuals were mostly appointed for services that they performed for Histiaians in the context of their own cities. This is the implication of the two grants of proxenia to citizens of Tenedos. If the interpretation of the organization of the Histiaian proxeny list offered here is correct, the isolation of these grants from each other (in ll. 18 and 33) indicates that they were made separately. These grants were thus probably made in different contexts and, since relatively few interactions between citizens and non-citizens would have resulted in services for which the award of proxenia was appropriate (especially as individual grants tend to imply earlier services as well), the two Tenedians on this list probably indicate a high volume of Histiaian activity at Tenedos. If these grants concerned almost any other distant island polis of the Aegean, we might be inclined to doubt this conclusion, but at Tenedos these contacts make good sense. Tenedos served as a gateway for ships sailing into Propontis and beyond since ships were obliged to wait in the harbour of the island for months at a time for the right combination of winds and currents to sail the Hellespont.35 Consequently, it is likely that the island polis was important as a commercial centre in its own right, as a place where connections were made and contracts could be struck between merchants of different cities.36 Strong Histiaian economic interests in the Black Sea are, in fact, communicated by a roughly contemporary decree which the Histiaians made in the context of a public delegation from Sinope in the Black Sea. In it, the Histiaians reciprocated a grant of the economic privilege, isoteleia, which their citizens had been given in their dealings at Sinope, adding numerous other honours and privileges (including asphaleia) which collectively constructed a close relationship of friendship between these two distant cities.37

commercial importance of Histiaia at that time—so Ashton (2000), 103–4; see already the doubts of Wallace (1962), 21–2. 35 On Tenedos and its importance as the entry point to the Hellespont, see Barnes (2006). 36 This economic importance is also reflected in no fewer than four separate proxenoi at Tenedos in the Karthaian catalogue, the second highest total known for any polis in this list after Athens. See Appendix no. 10.1 and below, Section 3.4. 37 IG XII 9 1186. The inscription has been lost.

Proxeny Networks

163

The best-documented reason for the commercial importance of both Tenedos and the Black Sea region beyond during this period was the grain trade.38 This comparatively high visibility is unsurprising, given that food supply is the economic area which most frequently became a matter of public concern.39 However, even if it was, as seems likely, the main driver of economic exchange with other regions, there will have necessarily been other goods that were significant (because they were exchanged for corn), and secondary patterns of exchange which developed as a result. This case study shows how the detailed reading of a proxeny list can illuminate a particular network of proxenoi, although elements necessarily remain tentative or obscure (such as the nature of Histiaian contacts with proxenoi in cities other than Tenedos beyond its local region, even if many of them were likely to have an economic dimension). But this also reveals what we lose by examining networks in isolation. Paradoxically, by studying this text on its own we lose sight of its extraordinariness—of the exceptional activity of the Histiaians which it represents in this year. The thirty-two individuals it collects in at least twenty-one different places do not just represent a large number of proxenoi for a polis to appoint in a particular year— this is the highest rate of granting that we know of for any polis in any year. Similarly, the degree to which the Histiaians reached outside their local region in this year, to express ties with a wide range of distant communities, is brought into sharp relief when we compare this to other attested patterns of distribution. The different intercommunal contacts which this document depicts suggest that this was a particularly fraught year for the Histiaians, and from the historical context the peril which they were probably in is readily apparent. Perhaps the number of grants they made, even in contexts that did not primarily concern inter-polis affairs, reflects the fact that they were particularly keen to make connections with other communities as a result. In any case, it is tempting to read this document as a 38 The importance of the grain trade from the Black Sea is clearly illustrated by the dossier of material collected in Moreno (2007), 337–8, in which Tenedos frequently recurs. See also Oliver (2007), 18–19 (and pp. 71–2 on Tenedos). On this trade with the Black Sea in general, see now Müller (2010), 219–70. 39 The reciprocal decrees between Histiaia and Sinope may well have been made in the context of a food shortage at Histiaia (as the reference to sōtēria in l. 7 should probably be taken to imply), even though more long-term economic relations are envisaged as well.

164

Proxeny and Polis

memorial of the range of contacts they made in a single year, and, in particular, the success of their diplomacy in securing their position.

3.2. PROXENY NETWORKS COMPARED The proxeny list of Histiaia illustrates the importance and potential interest of proxeny networks but also the limitations inherent in analysing them in isolation. In order to interpret networks of proxenoi and understand the functions which they performed, it is therefore necessary to compare all available testimonia to establish common parameters and the degree to which there was variation—in terms both of the scale of networks of proxenoi and their patterns of regional distribution. These comparisons reveal that the size of a polis’ network, and, more importantly, its reach, were not necessarily correlated to its size or material resources but rather reflected the horizons of the community and interests of its citizens. They also make clear, however, that networks of proxenoi were never static, but needed to be constantly renewed or recreated in response to the continual attrition of proxenoi, as individuals died off. This is because, although most grants of proxenia were theoretically hereditary, in practice, like links of xenia, they needed to be renewed at each generation to effectively maintain this reciprocal relationship.40 As a consequence, a polis’ network of proxenoi would have reflected its interactions within the generation before the inscription of a catalogue of proxenoi (though connections with individual families could stretch much further back if they were regularly renewed). 40 The case of Alcibiades, Chapter 2, Section 2.4, underlines this need for reciprocity and illustrates the potential consequences of its failure; cf. Marek (1984), 161; Herman (1987), 58–72, on renewals of xenia, and 138, proxenia. This is made explicit in the decrees which describe the grant of proxenia as a renewal, or which mention, in the context of a grant of proxenia to an individual, family members who served as proxenoi in previous generations. Athens (fourth to late third centuries bc): IG II2 13; 86; 172; II3 294 (= II2 206); 375 (= II2 365); 1137 (= II2 844); Agora 16 59. Delphi (third century bc): FD III 1 86; 359; 2 193; 4 20; 225; 403; 406; 414; SGDI 2600; BCH 82 (1958), 88. See also: IG IX 12 1 71 c (Aitolia, 169–167 bc); IG XI 4 593 and 887 (Delos, third century bc); 1049 (an unknown polis, inscribed at Delos, first half of the fourth century bc); SEG 11 974 (Thouria, first century bc). Peek (1969) no. 62 is perhaps another example. On renewals at Delphi, see Bousquet (1958) 85–90 n. 1. On the similar case of patroneia—theoretically hereditary but in practice in need of renewal in each generation—see Eilers (2002), 81–3.

Proxeny Networks

165

3.2.1. The size of proxeny networks Catalogues of proxenoi make it clear that proxeny networks in general were large and that even small poleis could possess substantial networks. This is clear from our earliest catalogue of proxenoi, from Karthaia. This catalogue reveals that the Karthaians, despite being only one of four poleis on the small and far from wealthy Cycladic island of Keos, possessed a network of at least eighty-six proxenoi.41 A fragment of another catalogue illuminates the network of Eresos, a larger, but still definitely mid-ranking polis more than a century later. On the basis of this fragment, Eresos, one of five poleis on the larger island of Lesbos, can be conservatively estimated as possessing a network of c.175–250 proxenoi when this catalogue was first inscribed in c.230 bc.42 The surviving fragments of mid-second century proxeny catalogue of Astypalaia, an island-polis in the Dodecanese, suggests that this city maintained a network of c. 270–450 proxenoi then.43 None of these were particularly large or important poleis and comparison of their estimated territorial extent suggests that the size of their networks was not simply determined by their resources: Karthaia’s territory was doubtless the smallest, at c.67 km2, as indeed was its network, but in the case of Eresos and Astypalaia, the relationship between network size (175–250 proxenoi: 257–450 proxenoi) and territorial extent (c.225 km2: 97 km2) was inverted.44 All three of these were, of course, island poleis, but the proxeny catalogue of Kleitor clearly shows that even an entirely landlocked polis in Arkadia was capable of accumulating a substantial network of proxenoi: more than ninety-eight proxenoi, though not all from the same phase of inscription, can be identified on the three partially preserved sides of this monument, originally inscribed on all four sides.45 Extended series of chronological lists, such as we possess for Delphi and Epidauros, can provide us with further points of comparison.46

41 Keos as a whole is attested as paying four talents in tribute to the Athenians in the mid-fifth century bc (IG I3 263 IV l. 21). However, to judge from the fact that when, in 451/0, the Koressians separately paid 2.25 talents, the other three poleis of Keos, including Karthaia, paid 1.2 talents (IG I3 262 I l. 21; V l. 21), Karthaia’s material resources were only a relatively small fraction of what this would imply for Keos as a whole, at least at this time. 42 43 Mack (2012), 223–4. Appendix no. 3. 44 All estimates of territorial extent taken from Hansen and Nielsen (2004). 45 46 Appendix no. 11. Appendix no. 6.1–4.

166

Proxeny and Polis

Although these do not present us with complete networks, it is possible, given the continual need of poleis to renew their networks, to get some sense of the overall size of the networks which these rates of granting would have sustained. Since most proxenoi would have been appointed in maturity, proxeny networks probably needed to be renewed generationally, and the total size of a network is thus unlikely to have exceeded the number of grants made over twentyfive to thirty-five years.47 Thus, if we can suppose that the fifty-two grants which a list at Epidauros collects over a fifteen-year period constitute a complete record of the grants made, these would have supported a network of c.85–120 proxenoi.48 Similarly, the twentytwo years of proxeny grants recorded by the chronological lists of Delphi (121 individual proxenoi) for 197/6–175/4 bc would have sustained a proxeny network of approximately 135–189 proxenoi.49 Given the international importance of Delphi, and the degree to which it was connected, institutionally, with a wide range of other Greek cities,50 this total is low in comparison with the networks of other apparently much more insignificant poleis which we have noted. It is of course true that larger cities, and those more interested in maritime trade, are more likely to have had larger proxeny networks. These examples clearly show, however, that the networks created by small communities were not necessarily small, even by the standards of much larger and more important cities. The comparatively equal institutional ability of poleis large and small to create and maintain substantial proxeny networks which this attests to is also reflected in the rates of granting discussed in the following section. Altogether this indicates a remarkable intensity of proxeny networking in a world which was largely made up of small poleis, and an interest in expressing these sorts of links which does not simply reflect the practical functions they served.

3.2.2. Rates and rhythms of granting When we consider the maximum rates of granting to which chronological lists attest, the total size of these networks is not surprising. The proxeny list of Histiaia, which names thirty-two proxenoi 47 50

48 49 See n. 40. Appendix no. 6.1. Appendix no. 5. See Plassart (1921) for Delphi’s very large network of theōrodokoi.

Proxeny Networks

167

appointed in a single year, provides us with the highest annual total which we know for a polis, but it is itself eclipsed by a chronological list of the Aitolian League, coincidentally dating from within the same decade as the Histiaian list (c.270–260 bc), which collects grants for no fewer than seventy-four individuals.51 Other poleis also come very close to the Histiaian total: the Epidaurians appointed twenty-seven proxenoi in a single year;52 the Therans, at least twenty-six;53 a new text from central Greece lists twenty-three proxenoi;54 in two separate years the Samothracians named thirty-one and twenty-nine individuals proxenoi;55 in one particular year, the Tenians appointed fourteen proxenoi in different poleis.56 Moreover, we possess lists of foreigners from Chios and an unknown city which, if they are chronological lists of proxenoi, attest to the appointment of sixteen proxenoi in a single year and a number which would have originally been significantly higher than the thirty proxenoi now preserved.57 The fact that even relatively minor poleis could name substantial numbers of proxenoi in individual years highlights the reason why networks were not necessarily linked to polis-size. The institutional ‘cost’ of granting proxeny (e.g. in terms of assembly time) was more or less the same regardless of polis-size. If these rates of granting were at all close to the average, we would expect proxeny networks to be even larger than they appear above. However, where we have sufficiently long sequences of lists to calculate reliable averages, these tend to be much lower. In fact, the range and variability of rates of granting are just as striking as these maximal rates and, for our purposes, even more important. Thus, within the five-year sequence of lists in which the Epidaurians are attested as appointing twenty-seven proxenoi in one year, in the preceding year they appointed only one, and the average rate of granting (inflated by the particularly active last year) was eleven proxenoi.58 At Delphi, conversely, just over five proxenoi were appointed per year on average, or 3.5 distinct grants (counting individually grants made at the same time to families and groups from the same city), but the numbers of proxenoi appointed in a particular year varied enormously, ranging from none in 193/2 to eighteen in 51 53 55 57

52 Appendix no. 1.2.1. Appendix no. 6.3.6. 54 Appendix no. 16.4.3. Appendix no. 17. 56 Appendix no. 14.5.2 and 14.9. Appendix no. 15.2. 58 Appendix no. 18. Appendix no. 6.3.

168

Proxeny and Polis

188/7 bc.59 Although our evidence elsewhere is less rich, everything suggests that this degree of fluctuation in annual rates of granting was characteristic of proxeny networks more broadly. Within this variation it is possible to identify certain relatively regular rhythms of granting, particularly in relation to the festivals which cities held at fixed intervals, inviting official delegations from other poleis to attend. Group-grants of proxenia were frequently made to visiting representatives (theōroi) and these contributed to particularly high totals of proxenoi which certain poleis named in certain years. In fact, the Samothracian lists, recording as many as thirty grants in a single year, comprised only theōroi (they were invariably introduced with the heading ‘the following became proxenoi of the polis when they visited as theōroi’).60 The new first-century list from central Greece describes the circumstances for the appointment of twenty-three proxenoi from eleven cities in similar terms, as delegates sent to the first celebration of a contest and sacrifice (i.e. a festival) of Herakles.61 Similarly, in the long sequence of chronological lists from Delphi a number of group grants can be identified—surely reflecting theōroi—which were made during the year in which the Pythian games took place, and this must also be the explanation of some of the group grants in the Epidaurian lists.62 Regular theoric rhythms were thus significant, but they seem to have caused, and allow us to account for, only part of the year on year variation in proxeny granting. In particular, theōria does not seem to have any detectable role to play in the proxeny list of Histiaia, unlike other lists which name nearly as many proxenoi, and this emphasizes 59

Appendix no. 5. ¥  æØ B[ º]ø Kª   Łø[æd Æ]æƪ Ø, Dimitrova (2008), no. 6 ll. 19–21 (= Appendix no. 14.6.2). Presumably they visited a particular festival, but what festival is unclear (the substantial numbers of poleis listed at a time are otherwise hard to explain). It is clear that, alongside these appointments of groups of theōroi, the Samothracians also made grants to individual proxenoi which make it explicit that they had the same functional expectations of them that any polis had. See Appendix no. 14. 61 [ a ˙挺ØÆ or H ˙æÆŒºø e] æH , æØ ŒÆd Pæª [ ÆØ KÆ ƺ ]  Kd e IªHÆ ŒÆd c Ł[ıÆ ‹Ø], Appendix no. 17 ll. 2–4 . 62 See Appendix nos. 5 and 6. It is also tempting to suppose that theōria also underlay the large group of twenty-six proxenoi appointed by the Therans in a single year (Appendix no. 16.4.3), given the relatively small number of poleis represented and the partial overlap between this and a smaller list which immediately precedes it on the stone (16.4.2—appointed in the context of the announcement of the festival?), but this is very uncertain. 60

Proxeny Networks

169

the exceptional scale and range of contacts of this polis. Perhaps surprisingly, festival theōria also seems to play a very modest role in the proxeny lists at Delphi. In the twenty-two years covered by the Delphian lists, only eight delegations of theōroi from poleis can be confidently identified (not all of them in years in which Pythian games were held), surely far fewer than would actually have been sent by poleis during this period.63 This presumably reflects a conscious selection on the part of the Delphians, though on what basis we know not. This does, however, throw the other sorts of grants in these lists into greater relief. The largest group grants were not made to groups of theōroi coming to the Pythian games from other poleis, but rather to a group of thirteen Alexandrians in 188/7 bc, presumably a delegation sent by Ptolemy Epiphanes (responsible for the spike in this year of eighteen proxenoi),64 and to nine Rhodians in 180/79 bc, requested by the Delphians to arbitrate a boundary dispute with their neighbours. Definite public initiatives, by Delphi and other state actors in relation to Delphi, are thus also responsible for some of the variation which we see. As another example of this, it is also worth mentioning the list of fourteen grants made in a single year by the Tenians, which seem to reflect a desire to improve the coverage of its proxeny network in relation to Crete in this year.65 We also, however, should not discount the possibility of private initiatives and motivations, both on the part of citizens of the granting city and would-be recipients, in causing this variation in rates of granting. It is possible that we can isolate these sorts of private initiatives in the proxeny lists of Delphi, in the marked tendency for individuals to be appointed proxenoi shortly after the appointment of others from the same polis or region.66

63 Massalia (four Massilotes, Appendix no. 5 196/5 II), Lamia (three Lamians, 196/ 5 II), Lebadeia (twice: 186/5 I, three Lebadeians; and 185/4 I, two), Koroneia (twice: 182/1 II, three; and 178/7, two) and Sikyon (176/5 II, three). The group of three Thespians appointed proxenoi in 189/8 II are likely to be a panel of foreign judges (so Habicht 1987, identifying this group with the honorands of SEG 1 132). 64 65 Appendix no. 5 188/7 I and 180/79 I. Appendix no 15.2. 66 Thus, we see separate grants in consecutive years for Naupaktians (Appendix no. 5 188/7 I and 187/6 I), Elateians (three grants, three years: 177/6 I, 176/5 II and 175/4 I), Lebadeians (186/5 I and 185/4 I), and Chians (three grants, three years: 186/5 II, 185/4 II, and 184/3 II), and grants for two or more individuals from Corinth (181/0 ΙΙ and 178/7 I) and Pontic Chersonese (195/4 I and 192/1 I) within three years of each other—constituting the vast majority of repeated grants for individuals from particular poleis in this list. A similar phenomenon is evident in the proxeny lists of Epidauros, see Appendix no. 6.

170

Proxeny and Polis

A possible explanation of these clusters is that the grant of proxenia to one individual in a polis made his peers there more likely to actively seek appointment themselves, in competition with their fellows. In the variation of proxeny granting we thus see some structures— in particular, a regular, but sometimes faint, theoric rhythm. Private and economic activities are also likely to have contributed to this clustering of proxeny granting in specific years, as relations between individuals based on interactions over a longer period reached a particular point, resulting in a proposal of proxenia, or especially marked services provoked the proposal of grants more quickly. But the inter-annual variability in rates of granting is probably a complex function of other factors as well: the highly variable rate of inter-polis interactions, at least from the perspective of individual poleis (as in the case of the Histiaians who clearly experienced a particularly eventful year of diplomacy in 264/3 bc) as well as the inconsistency of particular poleis’ responses to them. It seems likely that in some years poleis were, collectively, more interested in emphasizing the links which they had with other cities, but also that internal political competition between citizens proposing the appointment of proxenoi could result in more grants in specific years.

3.2.3. Diachronic trends in proxeny networking Despite the considerable fluctuation which characterized rates of proxeny-granting year by year, it seems that we can detect certain long term trends. Our evidence seems to suggest that from the late Classical period onwards proxeny networks were more likely to incorporate larger multiples of proxenoi at particular poleis and be larger as a result. This trend can be crudely represented by the proxeny catalogues of Karthaia (mid-fourth century bc), Eresos (mid to late third century bc), and Astypalaia (late second century bc). The first of these catalogues includes many fewer proxenoi than the last, and they are, in general, spread much more thinly across different poleis (Eresos represents something of a mid-point in terms of both total size and relative distribution). Thus, where Karthaia seems to have recognized relatively few proxenoi at any given polis (in many cases probably only one), most of the cities encompassed by the Astypalaian network seem to be represented by at least six proxenoi (eleven to seventeen

Proxeny Networks

171

in the case of larger poleis). These networks are surely a reflection of the differing situations and outlooks of Karthaia and Astypalaia, but this trend towards appointing substantial groups of proxenoi at particular poleis, seems to be well represented in our other evidence. One of the main components of this trend seems to be a tendency to make more group grants, in particular to panels of theōroi and foreign judges, which seems to become more common from the late third century onwards, though this may in itself be part of a more general cultural shift among poleis towards granting proxenia more frequently. However, although we can probably thus identify an overall increase in the total average size of proxeny networks, it is clear that the local trends which fed into this picture were far more complex. Two third-century chronological lists from Epidauros, separated by perhaps fifty years, seem to show a particularly swift increase in the average number of proxenoi appointed. The first, a record of fifteen years between c.260 and 40 bc, yields an average of 3.7 proxeny grants per year.67 The second, a record of five years between c.220 and 200 bc, yields an average of approximately eleven grants—or 7.5 even if we exclude the last year in which twenty-seven proxenoi were appointed.68 This large and relatively sudden increase—200 to 300 per cent in the space of perhaps fifty years—illustrates how polisspecific individual patterns of granting are likely to be. The material from Delphi complicates this narrative still further. The complete list of proxenoi that we have for the first part of the second century bc attests to an average rate of just over five grants per year (5.4), whereas the 180 individual proxeny decrees which we possess for the thirty-five year period 315–280 bc attest to a rate which is only slightly lower (5.14).69 Since it is extremely unlikely that we possess as complete a sample of individually inscribed decrees for 315–280 bc as we do in our list for 197/6–175/4 bc, it appears that, after a period which witnessed a large increase in proxeny granting at Epidauros, the rate of granting at Delphi was, if anything, lower than it had been before. The importance of the overall quantitative trend that can be observed—towards larger proxeny networks composed of larger 67

68 Appendix no. 6.1. Appendix no. 6.3.2–6. The numbers of individually inscribed decrees are drawn from Marek’s study of the Delphic decrees, and rely on his dating, Marek (1984), 165–7; see 217–46 for his complete chronology. 69

172

Proxeny and Polis

groups of proxenoi at particular poleis—should not be overstated. It seems highly unlikely that contemporaries were aware of the general trend among poleis for increasing networks of proxenoi, and proxenia continued to be understood as an institution associated with particular sorts of practical function, nonetheless. Indeed, although the appointment by cities of multiple proxenoi at particular poleis and the simultaneous appointment of groups of individuals have both been taken as signs of the degeneration of this institution in the Hellenistic period,70 both practices can be identified much earlier. The recognition by a polis of multiple proxenoi at another polis is in fact well attested in the fifth century bc,71 and this potential for multiplicity was probably important for the function of proxeny.72 Similarly, although grants of proxenia to groups became more common in the later Hellenistic period, it is a practice that can be traced as early as the mid-fourth century bc. In particular, although the proxeny network of Karthaia is much more thinly scattered in general than that of Astypalaia, this early network illustrates both phenomena: in addition to four proxenoi at Tenedos, who apparently received their grants of proxenia individually, we also see a group of at least two proxenoi appointed at the same time at Kios.73 The biggest concentration attested in the Karthaian network, at Athens, however, consisted of no fewer than fifteen proxenoi, of whom six were appointed as a group at the same time, making it almost as large as the largest group attested in the Astypalaian network (seventeen proxenoi at Halikarnassos). In this case the strength and intensity of the interactions attested between Keos and Athens,74 and the high status of the individuals appointed proxenoi there,75 provides us with 70

71 Gschnitzer (1973), cols. 676–7, 657. Gauthier (1985), 137–40. See Chapter 2, Section 2.1.2. 73 Appendix no. 10.1.6 (IG XII 5 542 ll. 50–3 and 58); see Mack (2011). 74 All of the different cities of Keos belonged to both the Delian League in the fifth century and the Second Athenian League in the fourth (RO 22). The settlements imposed after their attempted secession from the Second Athenian League in 364/3 bc, shortly before this text, are documented in RO 39. From about the same period we also possess a set of decrees, RO 40, which the Athenians persuaded at least three of the cities on Keos to make (Karthaia, Koresia, and Ioulis), instituting an Athenian monopoly on the export of miltos (ochre or ruddle) from the island—attesting to the economic importance of this island for the Athenians. The strength of this connection, however, is probably most eloquently revealed in the large number of Keians attested in inscriptions at Athens (more than twenty, excluding the numerous Keians mentioned in RO 39–40), for which see Mendoni (2007b). 75 Mack (2011), 329–30. 72

Proxeny Networks

173

the context we lack in other cases for understanding how so many proxenoi at a particular place could be useful.76 In general, the potential for the possession of groups of proxenoi at particular poleis will have lightened the load for the individual proxenos and also created incentives in the form of competition between the proxenoi of a city.77 In fact, the desire to provoke this sort of competition—to encourage philotimia among foreigners—may have been partially responsible for the overall increase in the number of grants made, as cities competed with each other to advertise, by publicly granting honours, the fact that they rewarded foreigners for the services they rendered.78 The grant of proxenia to groups of theōroi illustrates that proxenia was being used to perform other sorts of function in connection with this as well, to do with the construction of inter-polis relations, which are explored in greater detail in the next chapter. While long-term trends can plausibly be identified, they represent more of an amplification of certain possibilities which were already, and probably always, available to poleis in using proxenia rather than any sort of fundamental change. Given the more or less equivalent capacity of poleis to grant proxenia, what mattered more than the absolute size of a given polis’ network of proxenoi was the relative distribution of its grants, both chronologically (marking the distinction between particularly proactive and quiet years) and geographically.

3.3. REGIONAL PATTERNS OF DISTRIBUTION: THE HORIZONS OF POLEIS Absolute figures for the size of proxeny networks and the variation in annual rates of granting present us with a general vision of proxeny as both vital and dynamic. However, for all that this view draws our 76 It seems clear that the prominent politician Aristophon, son of Aristoboulos (along with five other Athenians) was appointed when he was sent to Keos as stratēgos in 363/2 bc after an attempt to overturn Chabrias’ settlement (Chabrias may himself have been appointed proxenos the previous year); the decree which Aristophon proposed after his return explicitly praised the Karthaians (RO 39, l.54–5). See Mack (2011), 329. 77 The Alcibiades narrative also illustrates this dynamic, see Chapter 2, Section 2.4. 78 On the encouragement of philotimia—‘honour seeking behaviour’—see Lambert (2011), 194–7.

174

Proxeny and Polis

attention to the possibility of chronological variation at particular poleis, it elides the differences between different networks and poleis. The fact that even small poleis were able to make large numbers of proxeny grants does not mean that all poleis’ networks were the same, or that in their relations with each other they were equal. The proxeny networks of different cities—or rather the different patterns of relative distribution of proxenoi within them—reveal distinct, polisspecific patterns of interaction and engagement with other poleis and regions. Proxeny networks, and especially the proxeny catalogues into which they were self-consciously organized, in delineating the horizons of a polis and its position in relation to a range of other poleis, defined that polis in an important way. The identity and nature of a polis’ local region of primary interaction is particularly important for understanding its regional engagement. Some of the local regions that emerge from proxeny networks as the origin of the majority or largest minority of a particular polis’ proxenoi, appear to be relatively straightforward geographical units of the sort which we tend to use unquestioningly (of which the network of Kleitor, considered below, provides an example). Other examples, however, show that while these networks of interactions were constructed in response to geography, they were not straightforwardly determined by it. These regions instead represent the perspective of a particular polis, its pattern of interactions and the choices it made about which connections to privilege and honour. Indeed different cities could have different, even contradictory ways of constructing the local region of which they were part. The local region of Chios consisted of the poleis strung out along the coast of Asia Minor and the nearby islands, to judge from a fragment of its catalogue from the mid-third century bc, whereas the catalogue of Eresos—a polis which would probably have fallen within Chios’ local region—reveals a much more narrowly defined local region, excluding Chios. It apparently consisted of just the four other poleis on the island of Lesbos on which it was situated.79 By contrast, the proxeny catalogue of Karthaia, which, like Eresos, was located on an island with three other poleis, reflects a much larger local region (and indeed seems to omit the other three poleis on Keos entirely).80

79 80

Appendix no. 7; cf. Mack (2012), 224. Appendix no. 10.1; cf. Mack (2011), 330. See Section 3.4 and Map 5.

Proxeny Networks

175

Proxeny networks also differed greatly and defined their respective poleis by the degree to which they were limited to this local region of primary interaction. The extent to which a polis recognized proxenoi only within this zone, or reached beyond it to express relations with individuals in poleis outside it, reveals how wide that city’s horizons were, how interested its citizens were, collectively, in interacting and establishing permanent connections with poleis which lay beyond its immediate circle of proximate poleis. Although the precise distribution of each polis’ network of proxenoi was unique and subject to continual evolution, it is possible to identify a set of types which define the spectrum of different proxeny networks which poleis are attested possessing. The first type is the particularly narrow network, encompassing relatively few proxenoi outside a polis’ local region of interaction. A good example of this is the network of Kleitor. Of the ninetyeight proxenoi whose ethnics can be read from the catalogue no fewer than fifty-nine are from Peloponnesian poleis, with a further twentythree from mainland Greece. In this case, it is tempting to relate this narrow focus, and especially the lack of maritime poleis, to the landlocked position of Kleitor, located in the centre of Arkadia. The network of Epidauros, which was on the seaboard, however, appears even more parochial: setting aside Cretans (of whom there are fifteen), of the 136 proxenoi whose ethnics are known from these different chronological lists only about 10 per cent originated in poleis outside central Greece and the Peloponnese, and the only Aegean islands represented are Kos and Astypalaia. It is striking that, of the networks we can analyse in detail, one of the most heavily restricted to its local region was that of Astypalaia, an island (see Map 2). Of the cities confidently identified in the alphabetical catalogue, all but one come from within a relatively narrow radius of the island and it is noticeable that the only polis attested outside this radius—Aspendos—is represented by only two proxenoi, in comparison with the much larger multiples at other poleis. Neither of these civic networks was entirely restricted in its appointment of proxenoi to cities within its local region of primary interaction. However, in comparison with the networks of other poleis, it is striking how circumscribed and exclusive they are, despite the considerable effort these cities put into proxeny networking. This tells us a great deal both about the interactions and interrelations of these particular poleis, and the comparatively narrow collective limits of their communal

176

Proxeny and Polis

Map 2. The proxenoi of Astypalaia, c.150 bc (Appendix no. 3). The proxenoi depicted come from two blocks, preserving groups from poleis whose names begin with ‘A-’ (Arkesine, Aigiale, Anaphe, Aspendos, and Halikarnassos) and ‘K-’ (Knidos and Kos). The numbers indicate the numbers of proxenoi at each polis.

horizons—of the interactions and relations their citizens had and were keen to express. The second type of proxeny network, in addition to a well-defined local region, also reflects a particularly strong engagement with a well-defined secondary region represented by significant numbers of proxenoi from a coherent cluster of poleis. This pattern can be clearly seen in relation to the fourth-century bc network of Karthaia on Keos which, in addition to incorporating proxenoi at a substantial range of poleis within its local radius, also recognized a large and distinct group of at least sixteen proxenoi around the Hellespont and Propontis, more than the Karthaians possessed in the Cycladic Islands (see Map 5, Section 3.4). The relations implied by these proxenoi also appear to have been remarkably stable and long-lasting, as we can see from the fact that a substantial proportion of the proxeny grants attested on a stone from the mid-third century bc were also made to individuals from that area (three of fourteen; see

Proxeny Networks

177

Map 3. The proxenoi of Eresos, c.230–200 bc (Appendix no. 7). The inset map depicts proxenoi attested on Lesbos. This map omits two proxenoi with the Roman tria nomina, including one Roman. The numbers indicate the number of proxenoi attested at each polis.

Map 6, Section 3.4). To judge from its third-century bc proxeny catalogue, Eresos possessed a comparable, long-distance connection with Karystos, Megara, and Aigina on the Greek mainland which seem to have continued, and been renewed, over a period of perhaps some thirty years (grants were made to Karystians on no fewer than four separate occasions (see Map 3).81 Whereas local regions seem to reflect a wide range of different sorts of interaction between citizens of the polis in question and individuals at other communities within this region—on political, economic, religious, and even social levels—these more distant clusters probably reflect a narrower range of interaction. The most plausible interpretation of these sorts of stable, long-term relationships is that they relate to established links based on long-distance trade, which regularly brought Karthaians to the Hellespont and Eresians to central Greece. 81

Mack (2012), 222–3.

178

Proxeny and Polis

Map 4. Poleis at which proxenoi were appointed by Delphi from 197/6 to175/4 bc (Appendix no. 5).

Our material reveals a third type of proxeny distribution— networks which reached well beyond the polis’ local region of primary interaction and, instead of focusing on a secondary region, articulated links with a much wider geographic range of poleis. This is the pattern which we see in the long chronological lists from Delphi and, particularly dramatically, in the list of proxenoi appointed in a single year from Histiaia. Of the 121 proxenoi appointed by Delphi in seventyeight separate grants, just over half (sixty-two proxenoi; thirty-one grants) were made outside its local region (consisting of central Greece, including Athens and Euboia and the northern but not the southern Peloponnese), to individuals from Alexandria, Asia Minor, the Propontis, and western Greece (see Map 4 for the distribution of poleis represented). In this context, the reach of Histiaia is even more extraordinary. Of the twenty-four grants made in a single year (thirtytwo proxenoi), thirteen were for individuals who came from poleis beyond its local region and represent all of these far-flung corners of the Greek world (Map 1). Moreover, while in the case of the Delphic

Proxeny Networks

179

list it seems likely that a substantial proportion of those honoured had come to Delphi, the pattern of distribution which we see in the case of Histiaia suggests Histiaian contacts abroad.82 Both the desire and the capacity to articulate connections—to network—seem to have been shared by all poleis, but these were not channelled in the same way in each case. None of these different patterns of distribution represented the norm for networks of proxenoi. Instead they constitute a set of tendencies that contributed, in different measures, towards the shape of particular poleis’ networks. By recognizing the range of sizes and patterns of distribution attested in these fragmentary lists of proxenoi, it is possible to begin to imaginatively repopulate the Greek world with its vibrant proxeny networks; to visualize the tens or even hundreds of thousands of unidirectional proxeny links which would have criss-crossed the Mediterranean at any one time, connecting poleis to their peers; and then to look beyond them, to appreciate the intensity and frequency of inter-polis interaction which these links both implied and facilitated.83 The variables involved in this sort of modelling are complex, and admit too many unknowns and uncertainties to be extrapolated mechanically to produce a ‘connectivity map’ for the world of poleis as a whole, which we might then explore using the powerful quantitative tools of Social Network Analysis. Nevertheless, we can still do much to understand the fabric and structural dynamics of these networks en masse by considering what the recurrent patterns which we can identify at the level of the individual polis network would mean in aggregate. In particular, where poleis’ local regions of primary interaction coincided with each other to a greater extent, the result will have been particularly dense, structurally cohesive clusters consisting of numerous and reciprocal proxeny links between most of

82 For most of the Delphic proxenoi we simply cannot tell what the context for contact was. A rare exception is provided by the grant of proxenia for Titus Quinctius Flamininus (Appendix no. 5), which, because it is dated to 189/8 bc, the year of his censorship in Rome, must relate to his activities there (Syll.3 585 no. 46). Presumably he aided the Delphians in their petition before the Senate in that year for freedom, autonomia, ateleia, and asylia (Syll.3 612)—in other words performing in much the same way as proxenoi were generally expected to. 83 A very low estimate of average proxeny network size, say fifty, over the 1,000 poleis identified by the Copenhagen Polis Centre would yield 50,000 proxeny links between poleis at any given moment.

180

Proxeny and Polis

the actors involved. Thessaly, the Peloponnese, Lesbos and, perhaps, Crete seem likely to have been good examples of this phenomenon.84 But in all of these instances the formation of regional cliques will have been moderated and blurred by the considerable ties which many of the poleis concerned will have maintained with others outside—and especially by poleis whose interactions and visions of their relations cut across regions as they were conventionally understood or experienced (as in the case of Karthaia, below). At the same time, the increased likelihood that poleis would appoint proxenoi—and more proxenoi—at poleis they deemed more important would have contributed to a much more uneven, variegated structure both within these more densely networked regional clusters and between them. We see this tendency clearly in the catalogues of Karthaia and Astypalaia in the comparatively high concentrations which these poleis had at more significant but distant poleis (the Karthaians had fifteen proxenoi at Athens, more than in all the islands of the Cyclades; the Astypalaians had approximately six proxenoi at local island poleis, but eleven to seventeen at more major cities like Kos, Knidos, and Halikarnassos). These concentrations of proxenoi—and thus the two poleis’ estimation of the importance of a link—were not necessarily reciprocal. The Athenians will not have had fifteen proxenoi at Karthaia; they might, however, have placed a similar emphasis on proxenoi at Sparta as the Spartans did at Athens, at least in the context of their overall network of proxenoi. But, because these sorts of value judgements of different poleis will often have coincided, the result will have been that these asymmetries coalesced, constructing hierarchies in the connectivity of the Mediterranean despite the relative equality of participation in proxeny networking, singling out particular poleis to which the others were particularly keen to connect (e.g. Athens, Byzantium, Rhodes, or, as we have seen, Tenedos). In Social Network Analysis terms poleis, at least potentially, despite their differences in size, often had a similar out-degree centrality in relation to other poleis (the number of proxeny grants they made), i.e. they could be equally gregarious. However, they differed greatly in their in-degree centrality (the 84

It is precisely these sorts of polis-cluster that are likely to have given rise to regional koina (so Mackil 2013). Koina are, indeed, known for Thessaly, as well as groups of poleis within the Peloponnese, and on Lesbos and Crete. The creation of koina, however, did in some cases complicate the granting of proxenia (see Chapter 4).

Proxeny Networks

181

number of proxeny grants other poleis made to their citizens, especially other poleis with highly scored in-centrality)—that is in their popularity or status within the network.85 The inscription from Narthakion, probably a catalogue of other poleis’ proxenoi there, offers us the rare opportunity, despite the impossibility of near complete proxeny data for any period, of seeing how these structural asymmetries played out for a minor Greek polis on the Malian Gulf.86 Narthakion was linked-to by more than twentyseven poleis, a reasonably respectable number for its importance at this point in the mid-second century bc. However, in this age of plural proxenoi, only one of these poleis apparently set sufficient store by its links to Narthakion to appoint more than one proxenos there.87 In visualizing networks of proxenoi in this way, as links within a network of poleis, we get a sense of the multiplicity of this world, the way in which it was understood as being composed of many polis actors interacting with a large number of their peers, and the hubs and hierarchies that resulted. This emphasis on establishing links with particular poleis, especially through the appointment of multiple proxenoi including group grants to visiting delegations from these communities, also draws our attention to functions which grants of proxeny performed between poleis, which we explore in Chapter 4. However, while this sort of synthetic analysis of the information contained in proxeny lists does allow us to get some sense of the variation between proxeny networks, and the overall network dynamics which resulted, the cost of accommodating multiple polisperspectives in this way is to lose sight of the point of view of the individual city in relation to its network of proxenoi, and its selfconscious construction with it of its own particular position in relation to other poleis. Fortunately, one comparatively well-preserved proxeny catalogue from the minor polis of Karthaia allows us a corrective case study. The Karthaian proxeny network, and the way 85 Scott (2000), 69; Kadushin (2012), 31–2. Instead of ‘popularity’ Wasserman and Faust (1994), 174–5, speak instead of ‘prestige’ and ‘status’. 86 Appendix no. 13. 87 Appendix no. 13 ll. 17–18, excluding brothers (ll. 15, 22, and 28), who surely reflect single acts of granting (probably to a father). Since the ethnics have almost all been lost we cannot exclude the possibility that some ethnics occurred twice—though this is only likely in the limited number of cases of individuals who were proxenos of more than one place (like Menes in ll. 12–13). On the possible significance of Narthakion on the Malian Gulf in relation to routes inland, see Monceaux (1886), 180.

182

Proxeny and Polis

in which it was carefully presented in this inscription, reminds us that, while we can reconstruct rather crudely the rough outlines of proxeny networking, we need to be aware that we have almost completely lost not only the nuances of nearly all poleis’ regional engagement, but also the idiosyncratic ways in which they used their proxeny networks, collectively, to represent and understand their connections and social position.

3.4. CASE STUDY: THE PROXENY NETWORKS OF KARTHAIA The proxeny catalogue of Karthaia is not only our earliest substantial proxeny list, dating from the first half of the fourth century bc.88 It is also the inscription that allows us to reconstruct a network of proxenoi in its entirety with greatest precision. This is because, although we only possess half the width of the stele, the proxenoi which it lists were organized into a number of self-contained regional groupings which we can identify and accurately quantify. In order of their appearance on the stone, these were: Euboia (ten proxenoi); the eastern Peloponnese (eighteen proxenoi); Boiotia and the Malian Gulf (seventeen proxenoi); Athens (fifteen proxenoi); the Cycladic islands (ten proxenoi); the Hellespont/Propontis (sixteen or more proxenoi). The result is a map of Karthaian orientations, of the intensity of Karthaian contact, constructed by the Karthaians themselves, in the first instance by the appointment of proxenoi, and, secondarily, on our stele, by listing them deliberately in this way (see Map 5). This comparative index of interest and interaction paints an intriguing picture of Karthaia’s relations with nearby poleis. In particular, the relative unimportance of the Cycladic Islands is striking, with the Karthaians recognizing only as many proxenoi there as they did on Euboia, and far fewer than in any of the other defined regions (far too few, in fact, to have achieved anything like comprehensive coverage of Cycladic poleis). Instead, despite its position on the eastern coast of Keos, Karthaia looked predominately west, first and foremost to

88

Appendix no. 10.1.

Proxeny Networks

183

Map 5. The fourth-century proxeny network of Karthaia, 86+ proxenoi, shortly after 363/2 bc (Appendix no. 10.1). Karthaia is marked with a cross and known proxenoi with dots; regions into which the network is grouped on the catalogue are indicated with circles and numbers denoting the probable total size of each regional group.

Athens, but also to Euboia and central Greece in the north, and in the south to Sparta and Corinth and the eastern coast of the Peloponnese.89 In some of these cases, external evidence for contact with these different regions emphasizes the connection between the regional links revealed in this network and particular historical events. Two reciprocal extensions of citizenship which survive between the Keian federation and the Euboian cities of Eretria and Histiaia attest to a very close relationship indeed with Euboea in the period just prior to the creation of this catalogue. These treaties permitted any citizens who wished, excluding exiles, to migrate from one community and take up

89 Étienne and Dourlot (1996), 24–5, note the general and, in the context of the Cyclades, unusual, orientation of Karthaia towards the mainland.

184

Proxeny and Polis

citizenship at the other.90 Similarly, the Keians entered into an alliance with the Boiotians in the same period, defecting from the Second Athenian League when Epameinondas of Thebes launched the naval flotilla which challenged the supremacy of the Athenian fleet at sea. In fact, this particular incident, Epameinondas’ naval expedition, seems to be reflected in the proxeny catalogue of Karthaia by the proxenos from Knidos which it lists, out of place, among the entries for cities in Boiotia.91 The different links attested here between the Karthaians and Spartans (and Spartan dominated communities in the Peloponnese) are also partially understandable in these terms. In the period before the battle of Leuctra especially (371 bc), Sparta had been one of the two principal power centres in the western Aegean and the Spartans themselves had been active, operating a fleet off the island of Aigina to harry Athenian territory and shipping in the 380s and 370s.92 Finally, we have already seen how close political relations between Athens and the Keians were in this period, and that events immediately prior to the erection of this catalogue (Athenian military activity on the island of Keos) led directly to the nomination of a group of five Athenians as Karthaian proxenoi, including the Athenian general Aristophon, as well as, perhaps, the general Chabrias, by a grant the previous year.93 Because of the rich external material which we have for the fourth century bc, the eventful political history of this period, and the place of Karthaia (and Keos) within it, provides a rich context for interpreting Karthaia’s proxeny network—but one that is, ultimately, onedimensional.94 The proxeny network of Karthaia was also clearly rooted in broader patterns of political, economic, and cultural interaction (as were, of course, these particular political relationships and events), of which we can get only a limited sense from our sources. References to Keians (and, specifically, Karthaians) in documents at the panhellenic shrines of Delos and especially Delphi, suggest

90 StV II2 232 (Eretria) and 287 (Histiaia). Although the editors of these texts are inclined to date them differently, their close similarities in content surely mean that they belong in the same context, probably the revolt of Keos from Athens in the context of Epameinondas’ naval flotilla, on which see below. 91 On this, see Appendix no. 10.1.3 with n. 42, and Mack (2011). 92 Xen. Hell. 5.1.1–29; 5.4.61 (with specific reference to their presence around Keos). 93 See n. 76. 94 For an excellent overview of the evidence for the history of Keos in this period, see Brun (1989).

Proxeny Networks

185

strong connections on both official and private levels with these sanctuaries.95 Institutional similarities between the Keian federation and the city of Eretria are also suggestive of the strength of the links between the two.96 The decrees restricting the export of Keian miltos (ochre) to Athens, which three Keian cities including Karthaia were persuaded to make, emphasize the importance of commercial exchange, while an Athenian naval list from 405 bc attests to a contingent of at least eleven Keians (the largest group of xenoi from any community in this list by far) who gained employment in the Athenian fleet.97 Together these texts illustrate the economic opportunities which were surely taken up by other Karthaians at Athens. Similarly, although it is unusual to have external reference to Spartan perioikic communities, it seems likely that the occurrence of at least Kyphanta and Epidauros Limera in this catalogue has less to do with historic political relations between Sparta and Karthaia than the location of these communities on the eastern seaboard of the Peloponnese and their consequent importance as stopping places and even ports of trade for Karthaian shipping. The economic dimension is brought out most clearly, however, in the strong concentration of Karthaian proxenoi in the Propontic region of sixteen or more proxenoi, which is very striking, especially in comparison, again, with the rather smaller Cycladic cluster. The number of distinct individual grants which this implies (including no fewer than four at the important economic gateway of Tenedos), suggests a well-established connection for which we otherwise have only very meagre evidence in the form of a single honorific decree, and it is best understood as reflecting patterns of long distance trade and, probably,

95 Delos: ID 98 A l.12 (= Migeotte 1984, no. 45; a debt of four talents owed by the Keians in the accounts of the Delian Amphictyons, 377/6–374/3 bc). Delphi: CID 2.6, 12, 13, 15, 17, and 18 (on these, as evidence of Keian theōria in the mid-fourth century bc, including a number of Karthaians, see Rutherford 2004); SEG 31 536 (a Delphic proxeny decree for a Keian, late fourth century bc). Although no Delphians are known in the catalogue, it is possible (I think likely) that Delphic proxenoi would originally have been present, probably in the section for central Greece. 96 For these, and the limitations of this sort of evidence, see Lewis (1962). 97 RO 40 (miltos was an important colourant composed of iron oxide, and the red miltos of Keos was particularly prized, see Theophr. Lap. 52–3); IG I3 1032 VI l.72–82. These economic connections are further underlined by the discovery of a number of Keian coins during the excavations of the Athenian Agora, see Kroll, (1993), 251–2 no. 831–40, and Papageorgiadou-Banis (1997), 67.

186

Proxeny and Polis

Karthaians regularly present in the region.98 The parameters of Karthaia’s trade with the Black Sea (as, indeed, Karthaia’s economic relations with other poleis) can only be conjectured. It seems likely that Karthaian interest mirrored that attested for other poleis—the grain trade—but what cargo the Karthaians brought with them is unclear. The most famous export of Keos, miltos, may have figured prominently (the later Athenian attempt to obtain an export monopoly clearly shows that there were other markets), but there may well have been other local components as well (Karthaian coinage, stamped with grapes and transport amphorae suggests that wine may also have been a possibility).99 It is probably best not to fixate on individual products, but recognize that these cargoes could have been very mixed and need not have consisted predominantly, or even largely, of specifically Karthaian or Keian products. As the history of Aigina makes clear, large margins could be made on the transport trade, especially when long distances were involved. The way in which this fourth-century catalogue was organized allows us to reconstruct, with particular precision, the proxeny network of Karthaia, and at least to begin to explore the historical events and patterns of regional interaction and engagement of which it was the product. The inscription, however, also reveals the remarkable extent to which these networks could change over time. Comparison between the comprehensive view of Karthaia’s regional engagement presented by the fourth-century catalogue and two fragments of a chronological list from a century later reveals how Karthaia responded to the changing dynamic in the Mediterranean between the Classical and Hellenistic periods (see Map 6).100 The only region well represented in both is the Propontis—suggesting that the longdistance economic interests of Karthaians in that zone were particularly durable. Otherwise, as comparison of these two figures shows, the Karthaians reacted to the new realities of the Hellenistic world. Karthaia’s horizons remained broad, and indeed broadened, but, if this effectively random sample of fourteen proxenoi is representative of its network in general, as we would expect given the way 98 In Mendoni’s prosopography of Keians abroad (2007b), no. 108 is a dedication of a Keian, who may have been proxenos of the Olbiapolitans, to Apollo Iatros (SEG 28 657 = IosPE I² 164; 475–450 bc). 99 On the coinage of Keos, see Papageorgiadou-Banis (1997), esp. 34–42 on the coin types of Karthaia. 100 Appendix no. 10.2.

Proxeny Networks

187

Map 6. The Hellenistic proxenoi of Karthaia, 14 proxenoi of the mid-third century bc (Appendix no. 10.2). Contrast distribution with Map 5.

in which networks seem to have been continuously renewed, its interests, and the emphasis it placed on particular relations, shifted dramatically. Other than the grants to the Hellespont and a solitary grant to an inhabitant of Chalkis on Euboia, these grants suggest that Karthaian interests in the areas that dominated their fourth-century network had dwindled considerably. By contrast, the appearance of Macedon and Thessaly, both previously absent, surely reflects their new importance as centres of political gravity, and it is tempting to read the grant to a Chalkidian as a reflection of the new importance of this northern route. Similarly, the attestation of poleis on the western and eastern sea routes (especially to southern Asia Minor), again both apparently neglected in the previous network, reveals the increased significance of these routes for the third-century Karthaians. The fourth-century catalogue of Karthaia, however, is perhaps most important for the attitude to and use of the polis’ proxeny network which it reveals. It was erected at some point in the midfourth century bc, probably shortly after the Athenians quelled the

188

Proxeny and Polis

abortive revolt of Keos and enforced a break-up of the island’s federation.101 The monumental function of this catalogue, quite apart from defining the privileges which proxeny conveyed, seems likely to have been to advertise the size and extent of the proxeny network of this polis—to assert the importance of Karthaia by listing its prominent, official friends.102 In this context, the unusual way in which these proxenoi are listed seems significant. Unlike the catalogues of other poleis, which tend to begin by collecting the proxenoi from their local region of primary interaction,103 the Karthaian catalogue is not obviously ordered in a way that associates this city with any particular region. Instead, by ordering the regional groups of proxenoi on two axes which meet at Keos—north–south (Euboia, the Peloponnese) and then west–east (Boiotia, Athens, the Cyclades, then the Hellespont)—it seems to construct Karthaia as a cross-roads between distinct regions rather than a part of one region. In this it reflects the broader function which proxeny networks in general served, and the catalogues in which they were presented: to emphasize the centrality of the polis in question which thereby advertised its links with external communities where individuals had thought it important enough to actively aid its interests. The singular ordering of the Karthaian catalogue reflects the idiosyncratic perspective of the Karthaians on this function—emphasizing the central, connected position which they occupied in relation to particular other poleis and regions.

3.5. CONCLUSION This material shows that the networks of proxenoi of even minor poleis were as capable of being large, widely spaced, and vital as those of more historically important poleis. Analysis also allows us to paint a vivid picture of inter-polis interaction, in all its different kinds, as continual and intense from the point of view of the Mediterranean as a whole, but, from the perspective of individual poleis, subject to 101

Mack (2011), 332–3. On this as one of the functions of inscribed proxeny decrees in general, see Lambert (2011), 203. 103 See, for example, Appendix no. 7 (Eresos). 102

Proxeny Networks

189

sometimes extreme fluctuations as well as regular rhythms. These studies also suggest that although proxeny networks had a functional basis and were rooted in patterns of regional interaction, they were also the product of a desire to establish connections with other poleis and assert a central position and identity in relation to them.

4 Proxeny and the Performance of Polis Identity Proxeny networks both reflected and facilitated interaction between poleis, and inscribed lists of proxenoi therefore allow us to explore the intensity of the contacts which particular poleis had with poleis in different regions. At the same time, however, this material also shows that, on its own, the practical functions which proxenoi performed are an insufficient explanation for poleis’ use of proxenia. In particular, the large multiples of proxenoi that cities appointed at certain poleis and, especially, the group grants they made to delegations from poleis suggest that proxenia also played a broader, symbolic role which is crucial for understanding the importance of this institution. In exploring what interstate anarchy meant for the Greek cities, and the specific functions which proxenia and other inter-polis institutions performed as a result, this chapter draws on two different conceptual frameworks. In particular, theories that have been developed within the field of modern International Relations Studies are well equipped for exploring issues of authority and state identity which the use of these institutions raises. At the same time, this chapter also builds on work done by John Ma in demonstrating the usefulness of the concept of peer-polity interaction for the history and institutions of the Greek city.1 Ma shows convincingly that the Greek poleis should be understood as a self-supporting network of competing, equipollent actors, which continually constructed their interrelations through a series of concrete and symbolic interactions. This chapter, therefore, explores how proxenia operated at this level, as a 1 Ma (2003); this concept was developed within the study of archaeology, see Renfew (1986) and also Arafat and Morgan (1994).

Proxeny and the Performance of Polis Identity

191

central element within a wider system of inter-polis institutions which, together, defined the fabric of the Mediterranean world as an interstate society composed of poleis. Although proxenia is all but absent from Ma’s account, it was in fact one of the most important institutions which poleis used to communicate their status and membership of the anarchical interstate society which they constructed through their interactions.

4.1. THEORETICAL FRAMEWORKS: STATE I DENTITY AND ANARCHY Modern theories of state identity and interstate anarchy offer a starting point for any attempt to reconstruct the world of the ancient Greek polis. Although the applicability of particular concepts and terms may be open to challenge, as involving expectations or assumptions which are inadmissible in relation to the conditions of the ancient world,2 International Relations Studies represents the most fully developed engagement with ideas of what it means to function as a state actor in the absence of a single over-arching authority.

4.1.1. State identity/polis identity The issue of sovereignty has been central to debates concerning the extent to which modern ideas of statehood, and therefore theories of state interaction and identity relying on them, are useful for the ancient world.3 This debate has been bound up with how we understand ancient references to autonomia in relation to poleis—variously understood as autonomy, sovereignty, and independence—and the extent to which we understand autonomia as a necessary condition for polis status. One of the most important recent contributions, that of Mogens Herman Hansen, makes a distinction between internal sovereignty (equated with the Greek term kyrios), which he argues was integral to the concept of the polis, with external sovereignty 2 On sovereignty: Hinsley (1986), 27–36; Davies (1994); Ober (1996), 30 and 107–22. On state: Osborne (1985), 6–10. 3 e.g. Giovannini (1994), 269.

192

Proxeny and Polis

(identified with autonomia) which he argues should be dissociated from it.4 There is a problem, however, with the sort of maximal definition which Hansen adopts for autonomia as ‘an unqualified independence in one’s own affairs’.5 It obscures the ambiguity inherent in this term, which was central to the way in which it was used to both perform and negotiate power relations.6 Absolute independence, of course, could not have been fundamental to the conception of the polis. Such a requirement would exclude, as Hansen rightly points out, poleis that were part of federations or hegemonic alliances, which clearly were thought of as such. However, Hansen overlooks the fact that there was a degree of dependence—a lack of autonomia—which was thought incompatible with polis identity, and which suggests that polis status was somehow bound up with state identity.7 Two contrasting theories of state identity from international law, developed to deal with similar difficulties of identity in relation to modern states, provide a more profitable approach than questions of sovereignty to the issue of the relationship between polis status and state identity. The first is constitutive. The theory of recognition maintains that states become persons under international law if, and only if, they are formally recognized as such by other states.8 The second is declarative. This definition of statehood, as it is inscribed in the Montevideo Convention on the Rights and Duties of States (1933), asserts that ‘the political existence of the state is independent of recognition by the other states’, and gives instead as its definition the following:9

4

Hansen (1998), 52–83. Hansen opposes to this, the ‘negative and restricted sense [of autonomia] of “selfgovernment which is willing to accept subordination to superior powers in some matters” ’ (quoting Ostwald 1982, 29, and citing Bickerman 1958, 327). 6 Goldhill (1996) sums this up brilliantly in a review of an early volume from the Copenhagen Polis Centre: ‘In particular, recent work on ancient and modern political language has made it much harder to avoid the way in which each and every use of “autonomia” by a Greek writer is not just describing how things are, but it is also staking out a political position and negotiating a set of power relations . . . Autonomia functions not just as the description of an objective state of affairs but also as a performative in the agonistic system of power plays that makes up Greek political life and its rhetoric’. 7 Thebes on Mykale was one such dependent, non-polis community—so Mack (forthcoming 2015); see also Fröhlich (2010), 660–7. 8 9 Grant (1999), 2–4. Grant (1999), 5–8. 5

Proxeny and the Performance of Polis Identity

193

The state as a person of international law should possess the following qualifications: (a) a permanent population; (b) a defined territory; (c) government; and (d) capacity to enter into relations with the other states.

Neither of these definitions is wholly satisfactory or complete in itself. One troubling consequence of constitutive theory is that it places the power to grant statehood in the hands of other states, resulting in the apparent contradiction that some states (e.g. Northern Cyprus) are recognized as such by some but not all states.10 Conversely, classic declarative theory overlooks the existence of other sorts of state actor—such as governments-in-exile in the Second World War or modern Palestine—which, although they do not meet its criteria for statehood, are capable of acting as states in certain external contexts because they are treated as such. What is striking in both of these approaches, however, is their common ground. In each account states are defined as entities which are treated as such by other states. State identity thus appears to be externally constructed through states’ interactions and relations with other states, whether this is simply a necessary potential capacity (as in the case of declarative theory) or the sufficient condition (as under constitutive theory). The use of these two theories in combination, without attempting to iron out their contradictions—which reflect the inherent contestability of state identity11—provides a productive approach to the question of the external construction of polis identity in the ancient world.12 In this combined model statehood is understood as being constituted within a particular state system, with both an underlying, characteristic state type, but also the potential for other forms of state actor to be recognized and thus perform as such.13 According to this understanding, despite the diversity of state and quasi-state actors in the ancient system, including kings, federations, leagues, ethnē, associations, and amphictyonies, which have been stressed in recent accounts,14 the Mediterranean world before the domination of Rome should be understood as a state system constituted primarily by 10

11 On this see Craven (2010), 243–6. Vincent (1987), 42–3. For a theoretically nuanced approach to the internal status of the polis, see Anderson (2009). 13 Wendt (1999), 11–12; see also Reus-Smit (2012), 70–1: ‘A deep politics of identity . . . undergirds international society, determining its membership.’ 14 Brock and Hodkinson (2000), 21–9; Low (2007), 256; and Vlassopoulos (2007), 55, among others. 12

194

Proxeny and Polis

poleis (in contrast to nation states which characterize the state system of the modern world). Hansen’s Inventory of Archaic and Classical Greek Poleis collects more than a thousand poleis, which gives it a strong claim to be considered the most populous state system in recorded history, dwarfing the barely two hundred states that comprise the modern global state system.15 The vast majority of interstate interactions which occurred were thus carried out between poleis and these actors in fact assumed that they were primarily dealing with other poleis.16 As a result, the highly developed system of interstate language and institutions, which was evolved to express relations between state actors, was heavily polis-orientated. An important question remains, however—did polis identity in this ancient system play a comparable role to state identity in the modern system? Within the ancient world the formal recognition of other states, or rather poleis, did not develop as a widespread institution, even a ‘duty’ of the state, as it has been framed by certain theorists of International Relations.17 The closest we come to this is the granting of polis status or autonomia, which, represented as a gift (usually by a king), was instead about enforcing a hierarchical relation between the two.18 There are, however, instances in which we can clearly see that being a polis mattered because it meant statehood within the interstate system of the ancient world. This emerges most clearly when this status was denied in our inscriptions: when an actor, rather than referring to a particular community using their ethnic, adopted a circumlocution to avoid the implication that they accepted it as a legitimate polis. Thus, a Gortynian decree from the second century bc, in imposing terms on a dependent community, refers to the members of this community consistently not as ‘the Kaudians’ (which would emphasize their existence as a political community)

15 Hansen (2006), 31. The authoritative list of UN member states is reproduced online: (accessed 12 July 2012). 16 See, for example, the prospectus for the Second Athenian League (RO 22, 378/7 bc), with a list of subscribers headed ‘these poleis were allied to the Athenians’ (l. 78) despite the fact that it includes, in addition to forty-four poleis, three rulers, two federations, as well as one ‘splinter community’. On this document and other examples of this phenomenon (such as Thuc. 2.9.1 similarly purporting to list allied poleis but also including other sorts of state-actor), see Dreher (1995). 17 Lauterpacht (1947); on recognition in general, see Talmon (1998). 18 e.g. SEG 47 1745 (Letters of Eumenes II to Toriaion, 188–157 bc).

Proxeny and the Performance of Polis Identity

195

but simply as ‘those inhabiting Kaudos’.19 In a similar way, as Angelos Chaniotis has shown, the Nagideis, a polis in southern Asia Minor, chose to refer to a community at Arsinoe as ‘the inhabitants’ of the land in question rather than using their ethnic, the Arsinoeis. It was only after they formally granted the second community the territory on which it stood that the Nagideis switched to using the Arsinoeis’ polis ethnic to refer to them.20 Viewed in the context of the declarative theory of state identity, polis ethnics were important in that they encapsulated the claims of the community to be a polis/state, which in fact closely correspond to the qualifications of the state as set down in the Montevideo Convention. The polis ethnic asserted the existence of a political community—both (a) a permanent population and (c) a government—which was rooted in and derived its identity from (b), a particular territory. In both of the two examples considered above, the deliberate non-use of ethnics constituted a studied denial of polis identity—of their statehood: in relation to Kaudos this was, perhaps, because this status would have conflicted with the Gortynian desire to control this community that was clearly expressed by this document; in relation to Arsinoe it was because, for Nagidos, this would have meant renouncing claims to own its territory (as in fact the Nagideis did before using the ethnic Arsinoeis).21 Conversely, in the ancient world we also see polis ethnics used to assert polis/state identity in contexts where the actor in question did not meet the criteria of declarative theory. The most obvious example of this is ‘poleis’, which other poleis continued to recognize in particular contexts in relation to international institutions despite the fact that they had lost their territory and been physically destroyed—poleis-in-exile.22  KåæÅÆ ƒ ˆ æ Ø Ø  E a ˚Æ[F] ϝ ØŒ Ø, IC IV 184 with Chaniotis (1996), 407–20 (late third to early second century, bc). 20 Chaniotis (1993). For a similar shift to the use of an ethnic, this time after the formal grant of polis status, see SEG 47 1745 (Toriaion) with Savalli-Lestrade (2008), 10–15. 21 Chaniotis (1993), 36–7. 22 Two examples of continued recognition, as polis actors, of exiled groups: the ‘dēmos of the Zakynthians in the Nellos’, who were admitted to the Second Athenian League of the fourth century bc, RO 22 l.131–4, were probably one such exiled group, so Seibert (1979), 117, and Gehrke (1985), 198; citizens of Phthiotic Thebes were also permitted to serve as hieromnēmones in 204/3 and 202/1 bc, during which period they were probably in exile, Walsh (1993), 41–5. For discussion of other 19

196

Proxeny and Polis

4.1.2. Interstate anarchy The use of these theories of statehood allows us to shed light on the nature and importance of polis identity in the pre-Roman world. International Relations Studies, however, also possesses theories concerning the structural properties of the sort of world which these polis-states inhabited, which help us focus on how this shaped their actions, relations, and engagement with interstate institutions. In International Relations terms, before the domination of Rome, the interstate system of the ancient Mediterranean was characterized by anarchy. This is defined as the absence of a single, central authority capable of enforcing norms and monopolizing the exercise of legitimate force within the system, in contrast to the internal organization of the states that made it up.23 One ancient historian, Arthur Eckstein, has drawn extensively on neo-realist theories of anarchy in an attempt to illuminate the rise of Rome in this context.24 Neo-realist theories, formulated most influentially by Kenneth Waltz, attribute a quasi-determinative significance to interstate anarchy. States, according to this understanding, are primarily selfregarding. Although states may have different aims and motivations, these differences disappear as important factors, edged out by the aggressive pursuit of survival against potential threats (‘self-help’), which the inherent structural dynamic of anarchy selects in favour of and inculcates.25 The only important variable allowed for in Waltz’s account is the relative distribution of ‘capabilities’, by which he means power resources (especially military and economic). In this model, ‘capabilities’ define the positions that different actors occupy within the system and their motivations, and thus shape the outcomes which result.26 The most important consequence of all this is an international system in which war is normal and cooperation made difficult and rare as a result of pessimism and ignorance. This interpretative framework claims to be based on unchanging human drives and systemic dynamics and thus to be atemporal (indeed it takes Thucydides’ history as its foundational text). According to Eckstein,

examples of exiled communities acting and being accepted as poleis, see Gray (2011), 180–226, esp. 196–211. 23 24 Bull (2002), 44. Eckstein (2006); (2008). 25 26 Waltz (1979), 91–2. Waltz (1979), 97–9.

Proxeny and the Performance of Polis Identity

197

‘the systems of warlike and aggressive states that existed in the ancient Mediterranean conformed from the beginning to the grimmest and most unforgiving of Realist paradigms’.27 The flaw in this sort of account, however, is that it attaches little importance to the existence of specific interstate institutions or norms at all and, concentrating narrowly on power and war, can do little to explain the effort expended on them or the emphasis which they received.28 As Eckstein comments, and others have noted before, declarations of asylia, inviolacy, offered little practical protection, even for shrines—so why, to cite one spectacular example, did the city of Magnesia on the Maiander dispatch at least twenty teams of theōroi to Greek cities across the Mediterranean to obtain this grant, inscribing the obliging responses of more than two-hundred poleis to their request?29 Similarly, proxeny networks clearly show that constant contact and interaction—the desire to network—was one of the most characteristic features of the interstate system of poleis in the ancient Mediterranean. Eckstein, in discounting the importance of these institutions, overlooks this evidence of ties to the extent that he describes the ‘limited institutional ability and even limited desire of ancient states to communicate continually with each other’.30 Recent scholarship on International Relations, in particular the work of Alexander Wendt, has questioned the assumptions on which this realist paradigm was constructed, offering a much more productive approach to the structural dynamics of interstate anarchy in the ancient world. In place of the determinism of the neo-realist view, Wendt stresses the role of discourse and identity in the construction of state interests and interactions to argue that ‘anarchy is what you make of it’.31 In his model, motivations are not simply ahistorical material givens but culturally specific; interests are ideas, often responding to specific material conditions, but constituted within a particular interstate system on the basis of its history of previous interactions.32 ‘Self-help’ exists, but as a cultural construct, 27

Eckstein (2006), 10. Eckstein (2006), 8, quoting and dismissing as inappropriate for the Hellenistic world, Bull (2002), 13; see also on specific institutions, Eckstein (2006), 59 n. 96 and 99 (stating ‘it is unlikely that our Hellenistic sources have passed over many successful negotiations that by true mutual compromise defused crises between significant states . . . ’ sv index ‘Proxenoi, ineffectiveness of ’). 29 30 Eckstein (2006), 80; Rigsby (1996), 13–29. Eckstein (2008), 13. 31 32 Wendt (1992); (1999), 92–138. Wendt (1999), 92–138. 28

198

Proxeny and Polis

as one, but only one, of the possible discourses which may shape deliberation and action within a state of anarchy. Wendt’s account, as a consequence, recognizes a much broader range of state motivation: ‘self-esteem’, the ‘need to feel good about [oneself ]’ within a social context, the need for ‘honor, glory, achievement, recognition . . . , power, group membership’.33 This framework allows us to take account of the interest which emerged in relation to proxeny networks—the interest poleis had in their social position in relation to other poleis. For Classical Greece, Polly Low’s recent monograph provides a useful starting point for understanding what the Greek poleis made of anarchy. Drawing heavily on work done within the so called ‘English School’ in International Relations, Low argues forcefully that the dominant discourse in interstate relations had as its basis reciprocity, and, as such, was an extension of the discourse used of social relations between individuals.34 Similarly, in the Hellenistic period in the vast number of epigraphic monuments which were inscribed, poleis explicitly construct the world they inhabit as an ‘anarchic society’, a society of peer-polities, sharing common norms, institutions, and concerns.35 Relations with individuals and communities external to the polis are represented as permanent, as based on the concepts of friendship, kinship, mutual regard, and mutual moral ‘goodness’, as well as the recognition of benefaction and mutual obligation.36 Epigraphic texts, moreover, are explicitly framed as deliberate performances of these qualities in front of an interstate audience. Decrees honouring non-citizens and other states are inscribed ‘in order that others may know that the dēmos knows how to give thanks to its benefactors’.37 In other texts these motivation clauses are dissociated from publication clauses, and the decreeing of honours itself is presented as a performance of gratitude before a wide audience: ‘in order that it is clear to all that the dēmos of the Akragantines 33

After Wendt (1999), 132. Low (2007), esp. 129–74; see also, Mitchell (1997b). 35 Bull (2002), 13; Ma (2003); for this concept of ‘a society of poleis’, see already Giovannini (1994), 284 and now (2008) passim. 36 On the use of these terms, see Section 4. 2.1. 37 ‹ø [i] ŒÆd ƒ ¼ºº [Ø NHØ ‹Ø ›] B [ ] K Æ[Æ]Ø å[æØÆ I Ø ÆØ  E Pæª][ÆØ ], IG II3 516 ll. 16–8(= II2 579; Athens, second half of the fourth century bc); XII 5 653 ll. 56–8 (Syros, first century bc?); XII 6 150 (Samos, late fourth century bc). 34

Proxeny and the Performance of Polis Identity

199

knows how to give worthy thanks to those disposed to perform benefactions for them’;38 ‘in order that all may know that the polis of the Eretrians always takes great care for its friends . . . ’;39 or, ‘in order that the city may be seen to give worthy thanks to its benefactors’.40 In communicating the need to be seen to comply with interstate norms of reciprocity, these recurrent phrases reveal the existence of a broader social context and a concern for interstate opinion—that is the standing of the polis within it. Furthermore, at the same time as these honorific acts were conceived of as performances in front of an external, even universal audience, they would also have served as performances before the internal citizen body of the polis. Played out in the physical context of the city’s theatre where so many assemblies met, the proposal and granting of decrees of proxenia were ostentatious performances of conformity, on behalf of the whole community, to the norms of inter-polis society which they thus collectively recognized.41 Similarly, at the same time as the publication of decrees on stone and the announcement of honours at major civic festivals could be constructed as performances in front of a wider audience—‘so that all may know that the dēmos of the Prieneans knows how to give thanks to those who do well by it’—the primary and most attentive audience would have been the citizens of the city concerned, keen to see their polis being seen to do the gracious and grateful thing.42 These are the social dynamics responsible for the ‘normative institutional isomorphism’ identified in the first chapter as responsible for the remarkable stability of the institution of proxenia over five hundred years—the desire, even need, to be seen within the broader society of poleis to conform to its norms.

38 ‹ø< > AØ çÆ æe fi q ‹Ø › A H ŒæÆªÆ  ø K ÆÆØ åæØÆ I Ø ŒÆÆ Æ  E PæªE æ ÆØæ ı Ø ÆP , IGUR I 2 (c.100 bc). 39 ‹ø i   NHØ ‹Ø  ºØ  ’¯ ææØø Id  ººc æ ØÆ åØ bæ H ÆB ç ºø , IG XII 9 211 ll. 3–6 (late fourth century bc). 40 Zøæ b ŒÆd I ºæ ŒÆÆ ÆØæ çÆ ÆÆØ åæØæ I Æ ØHÆ  Eæ ÆPAæ PæªÆØæ, IvO 39 (Elis, c.200 bc). 41 Kolb (1981), esp. 88–91, but see Hansen and Fischer-Hansen (1994), 48–53; on the theatricality of public life in the Hellenistic period, see Chaniotis (1997). 42 a b ø[æa ] a   Æ ÆPHØ I Æ[ªª]EºÆØ eª Œæı[ŒÆ  E ] ˜Ø ı Ø , ‹ø NHØ [ ] ‹Ø  ÆÆØ › [B ] › —æØÅ ø I []Ø ÆØ åæØ[Æ ]  E s _  Ø FØ [ÆP ·], I.Priene 3 ll. 24–7 (Priene, second half of the fourth century bc). The polis/dēmos as eucharistos: IG XI 4 772 ll. 10–12 (Delos, 250–219 bc); XII suppl. 137 (Erythrai, before 167 bc).

200

Proxeny and Polis

Realist critics can, of course, point to examples of the flagrant violation of these norms, not least the surprise attack on Apollonia which the Kydonians carried out, entirely obliterating this community despite the treaty of philia and sympoliteia which bound the two poleis.43 Such readings, however, overlook the moral framework in which such acts were set by other communities—Polybius describes this as ‘a shocking act of treachery universally condemned’—and thus the importance of these acts themselves in the restatement of interpolis norms.44 These non-realist inter-poleis ties could also have an important practical influence on communal actors. In the same way that proxenoi were induced to perform services for the poleis which had granted them this status for the sake of the prestige and social capital which accrued to them from the performance of proxenia, poleis can be seen engaging in certain institutions (the sending of theōroi or dikasts) for the same reason, social position within the polis system. The honorific decrees which the Samians and Entellans inscribed after their return—in gratitude to the individuals and communities who had preserved them in exile and worked for the refoundation of their cities—illustrates how important these sorts of ideas could be within polis society for the actual survival of polis communities.45 The best example, however, is the action of the polis of Stymphalos, which took in the citizens of Elateia after the destruction of their polis, providing them with farmland on favourable terms and a long sequence of other services, both in terms of resources and diplomatic initiatives pursued on the Elateians’ behalf. In performing these extraordinarily generous acts of benefaction, the citizens of Stymphalos were apparently influenced by a mythological connection, which they believed placed them in a kinship relation with the Elateians, since the eponymous founder of Stymphalos was understood to be the son of the eponymous founder of Elateia, Elatos.46

43

Plb. 28.14.4–5; so Eckstein (2006), 102. æAªÆ Ø e ŒÆd Ææ  › º ª ı ø , Plb. 28. 14.1 (Loeb trans.). 45 IG XII 6 17–40 (Samos, 322–300 bc); SEG 30 1117–1118 (Entella, 254–241 bc?). 46 Stymphalos and Elateia (SEG 25 445, 187/6 bc; cf. Paus. 8.4.5–6; 10.34.2) with Habicht (1998b), 67–9. The genealogical relations underlying this kinship claim are attested as early as the mid-fourth century bc (FD III 1 3–11—a statue group—with Paus. 10.9.6). 44

Proxeny and the Performance of Polis Identity

201

4.2. INTER-POLIS INSTITUTIONS AND THE PERFORMANCE OF POLIS IDENTITY This understanding of the way in which poleis conceived of the interstate world of which they were part—as a society of poleis—is vital for interpreting the symbolic importance attached to the use of inter-polis institutions in general and proxenia in particular. Proxenia and the other inter-polis institutions were underpinned by a set of shared concepts and values which stressed the communality and homogeneity of the world inhabited by Greek poleis, and these institutions constituted the primary means for the expression of intercommunal relations within it. Consequently, participation within this broader system of inter-polis institutions was one of the most important ways in which communities were able to assert their identity as poleis and claim a position within polis society.

4.2.1. Proxenia within the system of inter-polis institutions The Greek poleis possessed a wide range of concepts and institutions, including proxenia, for expressing different sorts of relationship with each other. Collectively these constituted a remarkably stable, selfsupporting system which needs to be analysed as a whole to understand the place of individual institutional elements within it. Proxenia was the most commonly granted status that we know of. It was, of course, a status granted primarily to individual non-citizens, but proxenoi were conceived of as links to communities. Its use as an explicitly inter-polis institution is apparent in other ways also. Particular grants of proxenia to public delegations from poleis, or their magistrates, were made to express relations with these communities as much as, or more than, with the individual honorands themselves.47 Thus, when the Delphians granted proxenia to Chrysis on the occasion of the great Athenian procession to Delphi, they did so to express relations with, and to honour, the Athenian polis as a

47 The granting of proxenia to multiple groups of theōroi who had assembled for a specific festival is a particularly clear example of this phenomenon, see above, Chapter 3, pp. 168–9.

202

Proxeny and Polis

whole, which she, as the priestess of Athena Polias, represented.48 On other occasions, poleis granted proxenia in exceptional circumstances to whole communities rather than individuals, signalling their gratitude, but also emphasizing the permanence of the relationship and implying continued service and goodwill.49 However, while proxenia could communicate close relations, because it was a grant exclusively made to non-citizens, it could also stress the distinct political identity of the polis that made the grant from the polis of the recipient. Proxenia did not exist in a conceptual vacuum. One of the reasons for its stability was that it remained closely related to another ancient institution, xenia, which was itself a complex function of concepts of friendship (philia) and kinship (syngeneia).50 Both friendship and kinship (syngeneia, as well as other, more specific familial terms, including mētropolis and oikeiotēs) are also found playing central roles in the vocabulary of terms with which poleis constructed their mutual relations, as indeed are concepts of eunoia, aretē, and euergesia, which are familiar from the study of honorific language in decrees of proxenia. Some of these constituted formal statuses in their own right, which cities could mutually grant and recognize (philos, syngenēs, euergetēs).51 Collectively these were also used as the basis for the exchange of other honours and of honorific status, either as status formally recognized by earlier decree, or simply as terms used in descriptions justifying the worth of the honorands in question for the honours or practical assistance they were given. Other honours and honorific statuses, which were explicitly located within this same conceptual framework, were commonly granted to whole poleis as well as to individuals. These included, of course, citizenship (politeia), which was often granted to entire communities in the form isopoliteia (clearly showing that this was primarily a IG II2 1136 (106/5 bc). Examples include: the famous decree of Mausolus declaring all the Knidians his proxenoi (RO 55, mid-350s bc?); the Molossian grant of to the Akragantines (Syll.3 942, after 338 bc); the grant made by the koinon of the Aterargoi to the Pergamioi (SEG 15 411, after 264 bc); the grant of Delphians to all the Dionysiotechnitai in Athens (FD III 2 48, 97 bc); the grant of the Cretan koinon to the dēmos of Mylasa (I.Mylasa 643, second century bc). See Gschnitzer (1973), cols. 672–6. 50 Herman (1987) 16–31 and 132–42. 51 philia: Mitchell (1997b) and Panessa (1999); syngeneia: Curty (1995) and Jones (1999). For a useful summary of recent work on syngeneia and oikeiotēs and the distinction between them (a fact of blood-relation versus a bond, albeit a close one, based on association), see Saba (2010), 275–9. 48 49

Proxeny and the Performance of Polis Identity

203

symbolic gesture and not seriously expected to lead to a mass influx of immigrants).52 Grants of politeia and isopoliteia expressed a particularly strong claim to a close relationship between the two communities. Unlike grants of proxenia, they deliberately elided the distinction between the two as political communities, with implications of a quasi-kinship relation, given the popular understanding of the city as a descent group.53 Other grants of rights which citizens by default possessed (e.g. ateleia, enktēsis, asylia, and epigamia),54 fulfilled a similar symbolic function, asserting the existence of strong inter-polis relations to partially overcome the ‘identity-gulf ’ which separated the members of these different communities.55 One of the most distinctive institutional features of the Greek interstate world was the dispatch and receipt of official delegations of different sorts. In addition to ambassadors sent to make requests and negotiate treaties, representatives were also sent to announce or visit particular festivals or seek recognition of the inviolacy or sacredness of a shrine, to reconcile and judge cases which local legal systems had proved unable to cope with, and arbitrate inter-polis disputes. At first sight these may appear qualitatively distinct from the other elements of the institutional system outlined above. However, such delegations represented the most direct possible form of contact between different poleis (short, at least, of warfare) and, as such, were of particular importance in constructing relations between them. The dispatch or arrival of a delegation constituted a regular and highly charged context for the deployment of the language and institutions of inter-polis relations, and represent the occasions on which we can most clearly see how these were used to create links on different levels with delegations as both individuals and as the physical embodiment of the poleis they represented. Thus, when a panel of dikasts, dispatched at the request of one city, successfully resolved disputes there, this was treated as a service which simultaneously revealed the eunoia of both the dikasts and the polis which sent them. Both were praised and honoured—the dikasts with proxenia, the polis with a crown.56 Similarly, when a city dispatched a theōria to 52

On isopoliteia, see Gawantka (1975), esp. 81–91, see also Chapter 5, Section 5.2.2. Gawantka (1975), 110–13, rightly stressing the conceptual relation between isopoliteia and syngeneia; cf. Loraux (2001), 197–213. 54 55 Chapter 2, Section 2.3.1. Chapter 1, pp. 75–6. 56 e.g. I.Erythrai 122 (a decree honouring the demos of Erythrai and dikasts it had sent, after 167 bc). 53

204

Proxeny and Polis

announce their festival, it might honour both the individuals who offered hospitality and support to their delegation in other cities (with theōrodokia and/or proxenia) and the theōroi who were sent by these cities to their festival as a result (with proxenia).57 Delegates sent out by poleis to ask other cities to recognize their shrines as holy and inviolate (hieros kai asylos) might be themselves honoured as proxenoi, praised along with the city which sent them for the goodwill which they showed to the honouring city in their decision to seek its assent and for the nice things they said about the relations between the two cities in making their request.58

4.2.2. Interstate institutions and the performance of polis identity These examples involving the use of delegations bring out the transactive quality of the inter-polis system of institutions. Participation in this common institutional framework involved engaging in collaborative performances which permitted poleis to both assert claims to polis status and have their claims recognized by other polis actors. However, at the same time as asserting a community’s membership of polis society, this institutional framework could be used competitively, to differentiate a polis from its peers, establishing its particular social position within it. By participating in this system of inter-polis institutions, by being seen to do what poleis could and were meant to do (not least by granting proxenia), poleis performed their identity as fully-fledged polis states. This was not, of course, the only means by which state identity could be performed in interstate relations. It was not even the most emphatic, which was surely war, especially border wars waged against neighbours that forcefully expressed the territorial identity of poleis. However, participation in this institutional system did allow poleis to express relations with, and assert and have their identity confirmed in relation to, a much larger number and much wider geographic range of other poleis. In particular, by making grants of proxenia, by sending out delegations to other poleis, inviting them to 57

e.g. Syll.3 604 (Delphic decree in response to two theōroi sent by Pontic Chersonese, 192 bc). 58 e.g. I.Magnesia 36 (Decree of Ithaka in response to Magnesian asylia campaign, 208 bc).

Proxeny and the Performance of Polis Identity

205

recognize or send representation to particular festivals, or create treaties or exchange honours with them, poleis declared they were polis states. Conversely, by receiving and responding appropriately to delegations from other states—which by no means meant necessarily acceding to their requests—they were not only themselves recognized as state actors, they also explicitly recognized the claims and status of the cities which had sent these delegations out, reaffirming the ability of both to do what poleis could do, and thus confirming their mutual membership of polis society. As such, this institutional framework, especially proxenia, served both declarative and constitutive functions within polis society. Naturally, these functions of interstate institutions were more prominent at some times than at others. In relation to most grants of proxenia to individuals, for instance, the desire to perform as a polis probably took second place to considerations of the particular merits of a given honorand (although in expressing and acting in accordance with an obligation to honour benefit, political communities drew directly on ideas of how poleis should behave in front of an inter-polis audience). The close connection between polis status and the granting of proxenia came to the fore, however, when this institution was used in certain more highly charged contexts (explored below) and on the frequent occasions when it was used to respond appropriately to inter-polis delegations and express relations with the communities from which they came. The importance of this function of interstate institutions is highlighted in the case of communities whose ability to participate in this system was specifically limited. In particular, in the provisions which Antigonos Monophthalmos sent to Teos and Lebedos to regulate their proposed synoecism, the Lebedians were to be permitted to continue to send a theōros to the Panionion, and thus to be seen to continue to participate as a community in this festival, which was central to the construction of polis identity in Ionia.59 The restrictions imposed on this performance of identity, however, were stringent and precise. The Lebedian theōros was to act alongside and share accommodation with the Teians and, in fact to be formally identified as a Teian.60 Together, these would have constituted an unambiguous RC 3 l. 2 (306–302 bc). ‘TØ]ŁÆ E [æØ  Æ a Œ ]Ø a e Y åæ , ŒÅ F b  F ŒÆd Æ ÅªıæÇØ a H Ææ [H Iƺ] ø ŒÆd ŒÆºEŁÆØ Å ’, RC 3 ll. 2–4; see also Rutherford (2013), 219–20. 59 60

206

Proxeny and Polis

performance of the new status of Lebedos, as an internal subdivision of Teos, no longer able to perform independently as a polis in its own right. The attention given to these details, which occur at the start of this document, before more practical matters like housing arrangements, stresses the importance invested in such performances of polis identity. Although both the actors, and therefore the status relations, were theoretically permanent (especially kinship)—in much the same way that proxeny grants were theoretically hereditary—in practice, interpolis relations needed to be regularly renewed. Collectively, the use of these institutions continually recreated this society of states, its norms, and the individual and general relations which made it up. The basic external equality of communal status—‘equipollence’ to use the language applied by Ma61—was expressed through the more or less equivalent capacity poleis had to engage in this institutional system. This broader system of inter-polis institutions, however, also had a potentially competitive function within this society of poleis. Participation asserted and involved recognition of a community’s claims to being a polis state, but, at the same time, these sorts of relations and connections could also be engaged in by communities to distinguish themselves from each other, as a source of prestige—to compete. Thus it seems that the remarkable asylia campaign of Magnesia on the Maiander, mentioned above, and the decision to publish it extensively on stone, represented a claim to self-importance which must, at least in part, be seen in its local context as a response to a similar campaign waged by the Milesians (who themselves were perhaps responding to—even as they apparently denied—an earlier, more restricted Magnesian campaign).62 Similarly, the decision the Karthaians took to inscribe a complete catalogue of their proxenoi, which offers a view of a world literally centred on Karthaia, is probably best understood as an expression of the communal importance invested in these links collectively in the aftermath of the break-up of the Keian federation.63

61 62 63

Ma (2003), 15. See Thonemann (2007), unpicking the events and dynamics. Mack (2011), 332–3 (Appendix no. 10.1).

Proxeny and the Performance of Polis Identity

207

4.3. THE NUANCES OF PROXENIA IN PERFORMANCE By using the various different elements which made up this system of interstate institutions, communities performed their polis status before an external audience of their peer poleis. Proxenia, as the rates of granting in our inscriptions attest, was the most frequently used element of this system. Proposals for proxeny grants consequently constituted the most regular occasions on which the civic community collectively considered and expressed links with the outside world. In a sense, proxeny stands for the broader institutional system of which it was part, representing the Greek desire to network in its most general form, to construct links with other poleis. The wealth of material which survives for proxenia allows us to explore in detail the specific symbolic significance of this institution and its capacity for the performance of different sorts of inter-communal relations. In a series of case studies this section examines how proxenia was used in marginal situations that complicated or increased the significance of the claims to separate polis identity which it communicated, and considers what it meant when proxenia was not granted to citizens of a particular polis or poleis. Some of the sensitivities which poleis exhibited in granting proxenia were general, reflecting the inappropriateness of elements of this institution in particular situations. We see this in the widespread reluctance of poleis to grant proxenia to the adjudicators who arbitrated their disputes with other poleis. In this case, the partiality which was the basis for proxenia would have conflicted with the impartiality demanded of the arbitrators, potentially undermining the authority of their verdict.64 In other instances, the decision on the part of the polis to grant proxenia seems to have been influenced by the actions of other nearby actors, in particular those communities which it interacted with most regularly, which constituted its local region of

64 It is striking that the only cases where proxenia was granted were when the arbitration was unsuccessful or, as in the case of the Delphian honours for would-be Rhodian adjudicators in the early second century bc, when it failed to take place (FD III 3 383). The arbitration was between Delphi and Amphissa, and the profuse expressions of gratitude made by the Delphians have lead one modern commentator to wonder how impartial the Rhodian panel would have been had the Amphissans gone ahead (as it was they did not show up), Ager (1996), 314–17 no. 117.

208

Proxeny and Polis

primary interaction. For example, it is notable in the inscribed responses to the Magnesian asylia campaign that, while virtually all of the attested cities of western Greece granted proxenia to the Magnesian theōroi, very few others did.65 This distribution reflects the fact that these performances using institutional symbols were competitive, about asserting social identity in relation to particular circles of peers. The granting of proxenia had clearly become a part of the appropriate response for poleis in this area to the Magnesian request for recognition of the asylia of their shrine. These dynamics are important for our understanding of the way in which grants of proxenia—and participation more broadly in the system of inter-polis institutions—functioned. The case studies which follow, however, concentrate on exploring more marginal, ambiguous contexts, especially where relations between poleis were particularly close-knit or intense, in which the use of proxenia was more highly charged and different communities took idiosyncratic decisions about whether and how to grant proxenia.

4.3.1. Isopoliteia and federation: proxenia as a performance of separateness Grants of isopoliteia expressed a particularly close relationship between poleis, which seems to have been felt to be incompatible in certain contexts with granting proxenia between them to individual citizens. In particular, Denis Knoepfler has argued that the absence of proxeny decrees directly linking Eretria and Oropos, despite their proximity and extensive epigraphic records, was a result of the fact that these two poleis were bound together by a treaty of isopoliteia.66 By way of contrast—to show how marked this absence in fact is—in the more than two hundred Oropian decrees, there are at least nine proxeny decrees for citizens of Chalkis, which was much less strongly connected than Eretria with Oropos, and more than fifty for Athenians. 65 Rigsby (1996), 179–279 nos. 77, 81–2, 84–6, 95–6. It is possible that proxeny grants were also made by island states of the Aegean—both Eretria (no. 98) and Paros (no. 100) certainly granted proxenia, but Chalkis (no. 99) did not and it is unclear if the other eighteen cities inscribed beneath the Parian decree as having passed similar decrees made grants of proxenia as well. 66 Knoepfler (2001a), 23 and 251.

Proxeny and the Performance of Polis Identity

209

This phenomenon of a striking absence of proxeny decrees between closely linked cities which otherwise possess rich epigraphic records can also be clearly identified within certain federations. In particular, among the numerous proxeny decrees of the Boiotian cities we possess only two grants made between member poleis of the league and these, because of this otherwise marked absence, are in fact attributed to the period after the dissolution of the Boiotian federation in 171 bc.67 Similarly, the probable absence of citizens from other poleis on Keos from the Karthaian catalogue of proxenoi is likely to be another instance of this phenomenon, given the fact that this monument was constructed in the aftermath of an extended period of Keian federation.68 By contrast, the proxeny catalogue of Eresos shows how numerous grants to citizens of other poleis sharing the same island could be—up to two fifths of the perhaps 175 proxenoi recognized by the Eresians in this list were from the other cities on Lesbos.69 This contrast suggests that cities faced a choice in how they constructed their relations with cities to which they were particularly close—to make a comparatively large number of grants or avoid granting altogether. These examples seem to suggest that there was a degree of closeness between cities which was thought incompatible with grants of proxenia between them, especially where the cities in question symbolically shared a single citizen identity. This was because proxenia, as it was defined by the honorific language used and the concomitant package of honours and privileges, was a grant of quasi-citizen or, in some cases, explicitly full-citizen status. Decrees of proxenia thus emphasized the foreign status of the community from which the recipients came. They implied an ‘othering’ of the recipient community, or rather made a distinction between two states that seems sometimes to have appeared incompatible with particularly close relations, especially where the citizen bodies of two states were either co-extensive (i.e. where they had granted citizenship en masse to the other community) or shared a federal citizen identity. Proxenia, which functioned as a declaration of individual and independent 67 Perdrizet BCH 23 (1899), 94 no. IV.1; Migeotte (1984), no. 10. On these texts see Müller (2005), 99–105 with discussion and bibliography. Appendix no. 17 is probably a rather later example. On the issue of proxeny granting and federations in general, see Gschnitzer (1973), cols. 667–70, and Gauthier (1985), 145 n. 42bis. 68 Appendix no. 10.1; see Mack (2011), 332–3. 69 Appendix no. 7; see Mack (2012), 223–4.

210

Proxeny and Polis

state identity, was inappropriate in this context. It would have been a performance of separateness to an audience with whom they held themselves inextricably connected, and whom they did not consider, in this sense, a wholly separate polis community. However, although this phenomenon is quite clearly evident in these particular instances, some poleis do not seem to have observed this practice or the scruples underlying it. Instead they are attested granting proxenia to individuals from cities to which they were linked by mutual isopoliteia treaty or grant, or even federated. Thus, the Karthaian proxeny catalogue lists proxenoi in Eretria, to which, as members of the Keian federation, they would have been joined by treaties of isopoliteia prior to the construction of this catalogue.70 We even have instances in which poleis named individuals as proxenoi at the same time as they made mass grants of politeia or isopoliteia to their poleis of origin.71 We can now also adduce the evidence from a recently published cache of eighteen proxeny decrees from Kallipolis in Aitolia from the first half of the second century bc.72 This prominent player within the Aitolian League is revealed as having made numerous grants of proxenia to the citizens of other poleis belonging to the league. Indeed, all but two of the decrees for which the honorands’ ethnics are known (or plausibly restored) honour individuals from other cities that are known to have been members of the Aitolian League.73 Examination of these apparent exceptions reveals the very real differences in the ways in which these relationships were perceived and allows us to nuance our understanding of the symbolic content of proxenia. The first thing to note is that grants of politeia and isopoliteia can serve very different functions depending on the context in which they are made. Between geographically close, especially territorially contiguous poleis, which already had an extremely high rate of interaction, such grants and their exploitation blurred the boundaries

70 IG XII 5 542 l.8 (Appendix no. 10.1); StV II2 232 (isopoliteia with Eretria; first half of the fourth century bc). 71 e.g. Rigsby (1996), nos. 155, 156, 159, and 160 (Teos, first half of the second century bc). 72 ed. pr. Rousset (2006) (= SEG 56 581–98). 73 SEG 56 584 (for a Chaironeian) and 585 (for an individual from the Achaian city of Ascheion) are the only certain exceptions.

Proxeny and the Performance of Polis Identity

211

between different poleis, rendering the use of proxenia inappropriate.74 Xenophon describes the processes involved. His speaker warns the Spartans that the result of citizenship barriers being lowered between cities in the Chalkidike, notably concerning the rights to own real property within the territory of another city and of intermarriage, will be that it will soon be impossible to separate these cities again.75 Although not territorially contiguous, Eretria and Oropos were separated only by the narrow Euboian Gulf and their strong ties apparently dated from the foundation of Oropos by Eretria.76 Other mass grants of politeia and isopoliteia, conversely, lacked this sort of context which lent these grants particular importance and consequently eroded, or rather purposefully elided, the distinctions between polis-states. In these cases, even where the actual assumption of citizenship was automatic and not regulated, as was often the case in ways which maintained the distinction in practice between the two poleis, grants of isopoliteia or politeia asserted a closeness which was not matched by the frequency of contact or strength of connections between them. This was a strong gesture in a symbolic language, but it did not necessarily reflect or therefore always signal a degree of mutual identification which would jar with the granting of proxenia.77 Our evidence for proxeny-granting within the Aitolian League is, on the face of it, more surprising and problematic. The Aitolian League had a concept of shared citizenship as well as the institutional apparatus, a common assembly in addition to magistracies, to grant this Aitolian citizenship (alongside proxenia) to outsiders. Nonetheless, it was appropriate for the citizens of Kallipolis, one of the cities that had always been at the heart of the league, to make grants of proxeny to the citizens of fellow league members (including one Trichonian, from one of the other core cities).78 The reasons for this may partially relate to the geographic size and structure of the Aitolian League, not to mention its significant expansion—in contrast with the relative stability and homogeneity of the Boiotian and Keian federations.79 It is 74

See Gawantka (1975), 81–91, on this distinction between sympolity and isopolity. Xen. Hell. 5.2.19. 76 On which, see Knoepfler (2001a), 251–2 n. 976. 77 Gawantka (1975), 92–164. 78 ed. pr. Rousset (2006), 381–434 (SEG 56 581–98). 79 The maps provided by Scholten (2000), maps 3, 4, and 8, illustrate this remarkable expansion. Although in the early third century bc this expansion seems not to have necessarily involved full integration (a number of citizens of Amphissa in 75

212

Proxeny and Polis

nonetheless striking that, as in this case, federal structures did not necessarily monopolize inter-polis relations within a league and that, independent of these, it was possible for at least some poleis to express relations with other individuals and cities belonging to it in the same way that poleis continued to do with other cities which lay outside their league. The use of interstate institutions by these different actors could, of course, be complementary, with honorific initiatives and grants that were made at the polis level being reinforced at the level of the federation. Thus the Keian League responded to a specifically Naupaktian grant of citizenship by granting politeia to the entire league of the Aitolians, which then reciprocated by granting Aitolian asylia to the Keians.80 Similarly, if it is permissible to compare the geographical distribution of the third-century Aitolian grants of proxenia with those surviving from Kallipolis in the early second century, it appears that the two are largely operating in different spheres. The Kallipolitan decrees operated at a very local level (largely Aitolian-internal) in comparison to the Aitolian decrees which, made to honorands from Sicily and Italy to the Black Sea and Syria, as well as individuals in neighbouring regions, represent a performance of identity in front of a wider audience. Nonetheless, although the participation of poleis and federations in interstate institutions could be complementary, it is important to recognize that they did not need to be.81 This potential inconsistency in practice highlights the tension inherent in an interstate system in which both federal leagues and the poleis which made them up were able to assert claims to being equivalent state actors by using the same international institutions. Hansen sought to iron out this conceptual contradiction between polis status and federal membership by minimizing the importance of external sovereignty in his definition

Ozolian Lokris are granted isopoliteia by the Aitolian koinon after they are known to have joined it, IG IX 12 1 12a), by the end of the century and before our decrees, the extension of federal citizen rights seems to have become the norm, see Scholten (2000), 55 and Mackil (2013), 360–1. 80 StV III 508 (223/2 bc?); see also IG V 2 367 (Kleitor, 168–146 bc) in which a federal proxeny grant reinforces a polis’ awards to foreign judges. 81 For instance, the recipients of proxeny decrees enacted by individual Boiotian poleis do not appear to have been any more geographically restricted than those honoured by the Boiotian koinon.

Proxeny and the Performance of Polis Identity

213

of the polis.82 I think it better, however, to recognize the reality of this paradox—that the theoretical equality of status between actors within anarchical interstate society could sit uneasily with the participation of poleis in larger, hierarchical federal structures which themselves asserted corporate membership of interstate society. This paradox finds its expression in the disputes over membership and authority which took place between cities and federal structures in ancient leagues, as well as the rhetoric of autonomia denied, invoked both by external powers seeking to break up federations but also by dissatisfied members of them.83 Whatever efforts may have been made to gloss over this contradiction by the actors involved, the activities and motivations of poleis and koina in participating within this system of interstate institutions did overlap, creating the potential for tension and conflict between the two.84

4.3.2. Polis performance, proxenia, and dependency Evidence for the restricted granting of proxenia between closelylinked cities and within leagues illuminates the connection between the use of this and other interstate institutions and the performance of separate polis identity. However, this function of proxenia in particular, and participation in the system of inter-polis institutions in general, emerges more clearly in contexts where close relations between political communities were more heavily asymmetric and characterized by domination and dependence. In the most extreme cases, by arrogating to themselves the exclusive right to engage in honorific interactions with external states and especially by explicitly restricting their dependencies from these performances of statehood, dominant communities asserted their own status as poleis and also their control of these communities as sub-polis entities. Such restrictions were not the only way in which these institutions were used to 82

Hansen (1998), 46–7. In relation to the Boiotian League, see Xen. Hell. 3.18; 4.8.15; 5.1.32; on attempts to secede from the Achaian federation in the second century bc, Plb. 23.9.12. 84 This tension brought out particularly clearly by a dispute between the Achaian polis Helike and the Achaian koinon. The Achaians had granted a request made by a delegation from the Ionian cities for access to altars over which the Helikaians claimed exclusive control (the sources are Heraclid. Pont. fr. 46a; and D.S. 15.49; Mackil discusses this incident, 2013, 393–4). 83

214

Proxeny and Polis

signal relations of dependence, however. In other cases, where these institutions were less strongly monopolized, dependent communities could themselves use and adapt honorific grants of proxenia in relation to individual honorands from the dominant polis to acknowledge the claims of that community at the same time as asserting their own polis status. Unfortunately, cases where we can see evidence for the restriction of dependent communities from participating in this system of interstate institutions are difficult to identify. This is not least because such restrictions would mostly mean a complete absence of epigraphic evidence, and epigraphic silence is notoriously difficult to interrogate. Nonetheless, some correlations between the inscription of proxeny decrees and periods of independence seem particularly suggestive, especially at poleis at which large numbers of texts were inscribed. In particular, the correlation of the decrees of Oropos granting proxenia with its fourth-century periods of independence, is especially suggestive—with a small cluster of decrees in the early second half of the century separated by a gap from a sequence of texts which began after Oropos’ lengthy period of independence from 322 bc. It is certainly tempting to read the inscription of two proxeny decrees from the earlier cluster, which establish links with very prominent Macedonians, as the confident expression of its capacity as a newly independent polis to conduct relations with foreign powers (and, perhaps, to visibly reward individuals thought to have played an important role in bringing about that independence from Thebes).85 This interpretation is reasonably likely, but it is worth noting that the dating of these decrees, and this sequence of events, present some problems of circularity in proving the connection between polis status and proxeny granting, since it depends in part on assumptions about what it meant to be part of, and independent from, the Boiotian League at this time.86 Moreover, given the general paucity of the

85 I.Oropos 1–2; I.Oropos 3, a decree naming a citizen of Theangela euergetēs probably dates from the same period. For the attribution of this cluster of decrees to a period of independence for the Oropians after their liberation from the Thebans under Philip II in 338 bc and before Oropos was granted to the Athenians by Alexander in 335 bc, see Knoepfler (2001a), 367–89, esp. 373. 86 A parallel instance would be provided by IG XII 8 2, a proxeny decree of Myrina on Lemnos, dated to the period of independence of this community from Athens between 404–387 bc. The reasoning here is also dangerously circular, however, as Graham (1964), 186, saw, and, in fact, the grant of ateleia in this text qualified by

Proxeny and the Performance of Polis Identity

215

epigraphic record in general for proxenia before the second half of the fourth century, we should certainly be cautious about accepting Knoepfler’s broader thesis that the general absence of evidence for the inscription of proxeny decrees by other members of the Boiotian League at this time is an indication that league membership—i.e. Theban hegemony—precluded the granting by individual poleis of proxenia, even if we accept that the inscription of these decrees of Oropos was an expression of its brief period of liberation.87 In the case of the disappearance of proxeny inscription at Delos, which had previously had a strong tradition of inscribing proxeny decrees, after Delos’ loss of independence to Athens in 167 bc, the connection between proxeny inscription, proxeny decreeing, and independence seems much stronger. However, given the context for this disappearance— the change of political population implied by the imposition of an Athenian cleruchy—this looks to be a different sort of phenomenon. In one broader set of material, absence can probably be taken to be reasonably secure evidence of these sorts of restrictions—in the decrees of demes, that is political communities which were fully integrated parts of larger poleis, from which they derived their identity in external contexts (e.g. as Rhodians or Athenians). In all our evidence demes never grant proxenia or politeia and only rarely honour non-members of the community at all.88 On the few occasions on which they do honour outsiders (including fellow citizens belonging to other demes), they do so in relation to the activities of those individuals within the deme—as the inhabitants of the Athenian deme of Eleusis do in honouring two resident Thebans, or similarly Brykos, one of the demes of Rhodes in the Hellenistic period, in honouring a Samian doctor for his services performed amongst them.89 This makes these the most inward looking, and thus least outwardly assertive or performative, of decrees for outsiders. In the case of the decree of Brykos, moreover, the subordinate

± ø z ıæØ ÆE Ø Œ æØ

NØ, ll. 10–1, clearly shows that there was some sort of _ higher power involved—probably the fifth-century Athenians (and their post 413 bc harbour tax?), see Salomon (1997), 63–6. 87 The attestation of the personal name Thespieus in Athens during the fifth century bc (e.g. IG I3 32) is perhaps evidence that other Boiotian poleis were in the habit of granting proxenia earlier at least. 88 Rhodes (1995), 102–9; (1997), 500. 89 IG II2 1186 (mid-fourth century bc); XII 1 1032 (second century bc).

216

Proxeny and Polis

authority of this community, a former polis on the island of Karpathos, now part of the Rhodian polis, is strongly stressed by the referral of the award of the gold crown and the question of inscription in the shared sanctuary to the ‘whole demos’.90 Although demes were not typical of the sorts of community which the term ‘dependent communities’ is usually used to refer to, their rights and activities were an important model for poleis in constructing their relations with dominated communities. Demes thus represent one extreme of this discourse of state identity, the tension, in relations between states, between separation and integration, which we have already considered in the context of the relation of poleis within federations. The inability of demes to make grants of proxenia illustrates the identification of the use of this institution with polis identity. However, the material which best illustrates the desire of dominant communities to restrict the participation of their dependencies in this system of inter-polis institutions does not relate directly to proxenia. It does allow us, nonetheless, to see how the same interest which the dominant polis had in restricting unmediated interactions with other poleis worked in the case of related institutional elements. In particular, we are fortunate in possessing inscriptions from Megara and Mantinea, which provide us with rare glimpses of dominant communities structuring their relationships with heavily dependent communities. A degree from Megara, IG VII 1, illustrates directly these sorts of restrictions on the institutional capacity of dependent non-poleis to use inter-polis institutions to interact directly with representatives from external communities, in this case in relation to politeia, which represents a close model for proxenia. This decree, attributed to 306 bc, records a request made by the Aigosthenitans that the Megarians honour Zoilos, son of Kelainos, a Boiotian officer of Demetrios Poliorketes placed in charge of the garrison at Aigosthena, for his own good conduct and for keeping his troops under control.91

90 › b ƃæŁd ÆN[]Å[]Łø K HØ  Æ [Ø ]øØ a Ø  F ç ı ŒÆd _ ƒæe , IG XII 1 1032 ll. 32–4. Gabrielsen (1994) is surely right u I ÆŁØ N e that Rhodian demes and associations were competent to pass internal enactments as long as they contravened no polis law without specific ratification from the polis authorities (contra Swoboda 1890, 21)—but, by concentrating on the necessity of the polis authorizing inscription in centrally administered shrines, he downplays central control over the granting of crowns, which is very well attested in these texts. 91 On the identity of this individual see Marasco (1983), 221–2.

Proxeny and the Performance of Polis Identity

217

The Megarians duly acceded to the request, naming him ‘a citizen of the polis of the Megarians’, granting him ‘proedria in all the contests which the polis puts on’, and ordering that the decree be inscribed, ‘so that all may know that the dēmos of the Megarians honours those who do good in speech or act on behalf of the polis or on behalf of the kōmai’.92 In the subsequent history of this community the Aigosthenitans are attested as a polis in both the Boiotian and Achaian leagues, granting proxenia to the citizens of other poleis and establishing relations directly with them.93 Indeed, in one of the earliest of these decrees, the Aigosthenitans use the Boiotian dialect which seems calculated, as Peter Liddel has recently suggested, to stress Aigosthena’s new identity as a polis, separate from Megara.94 By contrast, at the date of the Megarian decree for Zoilos, although constituting, as ‘the Aigosthenitans’( d `Nª  EÆ[Ø]), a political community capable of making the decision to initiate this request—presumably by passing a decree themselves—the people of Aigosthena were aware that they could not or, perhaps, had better not attempt to establish a permanent honorific relationship with Zoilos themselves. In this they cooperated with the Megarian construction of Aigosthenitan communal identity, as simply one of the Megarian kōmai, making clear the association between the inability to grant politeia (and also, it is probably safe to assume, proxenia), and this sub-polis status. In fact, I would argue, this is something which the Megarians are particularly keen to emphasize in inscribing this decree, making a great show of their own ability, as a (repeatedly) self-proclaimed polis, to perform as an interstate actor on behalf of their territorial possession, which could not.95 It is also important to note, in relation to this subordinate 92 ‹ø NH Ø   ‹Ø ›  [› ]ªÆæø ØB fi  f IªÆŁ Ø æ Æ j ºªøØ j æªøØ bæ A ºØ j bæ Aª ŒøA , ll. 15–7. 93 IG VII 207, 208, 213, 219, 223; SEG 49 500 (Aigosthena, second half of the third century to early first century bc). Four other, newly discovered decrees from Aigosthena will be published by Yannis Kalliontzis. 94 IG VII 207 with Liddel (2009), 427. 95 The Megarians refer to themselves as a polis on no fewer than four occasions in this brief text (IG VII 1 ll. 9, 12, 14) making the contrast explicit at the last of these, bæ A ºØ j bæ Aª ŒøA , l.17. This desire to stress the subordinate position of the Aigosthenitans may owe something to the fact that Aigosthena looked very like a polis at this period (so [Skylax] 39.2) and had a remarkably impressive defensive wall which Megara is unlikely to have been responsible for building, whether it should be attributed to Athens (Ober 1983, 391–2) or the Boiotian League (Cooper 2009,

218

Proxeny and Polis

status, that the interaction which led to the Aigosthenitan request for this grant, like those which prompted the deme decrees I discussed above, took place at Aigosthena, which similarly illustrates the limited extent to which this community was able to be active within interstate society at this point in time. If Louis Robert is right in his reconstruction of a later document found at Pagai, Megara retained these sensitivities.96 As restored, this document is a decree of the Megarians, issuing honours, in this case proxenia, on behalf of one of its other kōmai, Pagai, rewarding advocates who came from Sicyon, a fellow member of the Achaian League to which the Megarians belonged, and served as advocates on behalf of Pagai in a territorial dispute. What is striking here, however, is that the Pagaians, despite the fact that they are unable to act as a polis in honouring their benefactors, are nonetheless able to act as contestants in this dispute—a curious mixture of recognition of Pagai as a political community (with a territory) and a determination to exert control over it. If this reconstruction is correct, it may be that what is going on here relates to the identity of the other protagonist in this dispute, Aigosthena. Perhaps the Megarians’ treatment of this as a territorial dispute, not between itself and another polis but as between two kōmai, represents a studied contempt of the new polis status of Aigosthena, formerly a Megarian dependency. The famous treaty between Mantinea and Helisson is even more open in its interest in polis (and non-polis) status, and connects the ability to participate in this system of interstate institutions with polis status in an even more explicit way. [Ł] ·  åÆ [IªÆ]Ł. [ ] ŁØ Æ[ ]Ø F[Ø] ŒÆd  EºØϝÆ Ø [N] ¼Æ[Æ] _ _ __ [] Æ.    E Æ Ø FØ ŒÆd  E  EºØϝÆ Ø[ ]· e [ E]º[Ø]_ []ϝØ Æ Ø Æ q ÆØ ϝE ŒÆd P E , Œ[ ]Ø Ç Æ[  ]ø ‹ø ŒÆd ƒ Æ Ø B , çæ[ ] Æ a åæÆ ŒÆd a [ºØ ] N Æ Ø []Æ N e e e Æ Ø ø , Ø  Æ A []_ [ ] ÆºØ H  EºØϝÆ ø uæ å[Ø] N  Æ åæ , ŒÆ _ _ 161–3 and 181–3 with bibliography on this question). See Flensted-Jensen and Hansen (2007), 210, plausibly emending this corrupt passage of Pseudo-Skylax so that Aigosthena is the only polis listed within the section followed by ‘Pegai (a fort) and Geraneia (a mountain)’, overlooked by Shipley (2011), 118. 96 SEG 13 327 (c.192 bc). For an alternative interpretation, arguing that this text concerned instead a dispute between Kassope and Thyrreion, see Wacker (1996).

Proxeny and the Performance of Polis Identity

219

Æ e  EºØϝÆ H Æ Ø ø — ŁÆæe q ÆØ K  EºØ[ ]Ø ŒÆæ K ÆE ¼ººÆØ ºØØ_ — a Łı Æ Ł ŁÆØ _a N  E_ _ ºØ Ø ŒÆd a ŁÆæ Æ ŒŁÆØ Œa a æØÆ. God. Good fortune. Agreement between the Mantineans and Heliswasians, for all days. It seemed good to the Mantineans and the Heliswasians: the Heliswasians are to be Mantineans, equal and the same, sharing in all things even as the Mantineans, bringing their territory and polis to Mantinea to the laws of the Mantineans, and, while the polis of the Heliswasians is to remain as it is for all time, they are to be a kōmē of the Mantineans; there shall be a theōros from Helisson as from the other poleis; they are to make sacrifices in Helisson and receive theōriai in accordance with their traditions.97

Collectively this ‘agreement’ amounted to a sort of ‘gobbling up’ sympolity—the Heliswasians were to be wholly absorbed as a political community by the Mantineans, losing the use of their own laws even amongst themselves.98 The question which this document raises, therefore, is—in what sense were the Heliswasians to remain a polis (‘for all time’, as the agreement emphatically states) despite becoming a kōmē, an internal civic subdivision of the polis of Mantinea, both in theory and, by becoming Mantineans, in fact? For Mogens Herman Hansen, this is evidence that there was no contradiction in poleis being dependent—in this most literal case lacking autonomia and being a kōmē.99 Peter Rhodes, by contrast, argues that the word polis is being used in two different, distinct senses here, referring in the first instance to Helisson as a political unit but, in the second clause, committing to its continued existence only as an urban centre.100 Given how interested this inscription is, however, in ideas of what it means to be a polis, it is difficult to accept that these sorts of distinctions were really intended, especially given the pains which are taken in the syntax of this inscription to juxtapose and contrast polis and kome with the insertion of the genitive absolute phrase: Ø  Æ A []ºØ H  EºØϝÆ ø uæ å[Ø] N  Æ åæ (‘the polis of_ the Heliswasians is to remain as it is for all time’). RO 14 (= SEG 37 340 ll. 1–10, early fourth century bc). Contra Hansen (1995a), 39; the clear implication of the clause which sets out that contracts agreed amongst the Heliswasians before the sympolity are to continue ‘to be valid according to the laws which they had when they entered Mantinea’ (Œa e  e ÆP d qå ‹ ºøŒ N Æ Ø Æ , ll. 13–16) is that in other cases the Heliswasians were to be obliged to use Mantinean laws even between each other. 99 100 Hansen (1995a), 39. Rhodes (1995), 96–7. 97 98

220

Proxeny and Polis

In fact, this text reads as a carefully negotiated surrender on the part of the Heliswasians. It represents a strictly defined compromise—a fudge—between the Heliswasians’ desire to retain the status of and continue, in some contexts, to assert their identity as, a polis, and the Mantineans’ very definite wish that they be absorbed and become a kōmē of Mantinea.101 In the clauses which immediately follow the declaration that the polis of the Heliswasians will remain, the sense in which they will be allowed to retain polis identity is set out: they will be permitted to contribute a theōros ‘as the other poleis do’ (a phrase which explicitly marks the connection between this action and the retention of polis status) and, secondly, to make sacrifices and receive theōriai as they previously had. The function of the second of these rights is clear enough. The Heliswasians are to continue to perform their festivals and, as part of this, to receive formal delegations from other poleis to witness and participate in them. In other words, they are to be permitted to continue to perform as a polis in this specific institutional inter-polis context. At first sight it seems probable that the theōros which the Heliswasians are permitted to contribute should be understood similarly, as a delegate sent out to represent the polis of Helisson at other poleis’ festivals, to perform their polis identity in front of an audience of their peers. The explanation which has been preferred, given the existence of a specific board of officials at Mantinea and other poleis called theōroi, attested in Thucydides administering an oath, is that the Heliswasians were to be represented in this magistracy.102 However, juxtaposed as they are with a reference to theōria, it seems perverse to dissociate the theōroi here completely from a theoric function.103 This clause probably constituted a limited concession—allowing the Heliswasians to continue to perform as a polis, in a sense, at external festivals, but perhaps as part of a wider Mantinean delegation, a phenomenon which would be closely paralleled by the continued but limited participation envisaged for the Lebedians after their proposed synoecism with Teos discussed above.104 101

Cf. Fröhlich (2010), 663–4. Thuc. 5.47; see Nielsen (2002), 359–61, for this interpretation, especially at n. 251 collecting the Arkadian evidence for this magistracy. 103 So Thür and Taeuber ad loc., IPArk 9. 104 There is a final sting in the tail. Given the importance in this document of the precise terms used to refer to Helisson, it cannot be an accident that, in the clauses detailing the sense in which Heliswasians were to be permitted to assert polis identity, 102

Proxeny and the Performance of Polis Identity

221

As the case of Helisson reveals, although for dominant poleis limiting the ability of their dependent communities to interact with other poleis through inter-polis institutions was an abiding concern, the restrictions which they placed on the use of these institutions were not necessarily absolute. In other examples it becomes apparent that certain dependent communities felt able to make honorific grants to individuals, especially officials, from the poleis which dominated them, using them to express a complex, nuanced relationship between the two. In particular, two proxeny decrees from fourth-century Delos, then under the Athenians, assert the ability of the community to act at as a polis at the same time as they recognize particularly close ties with the city of Athens and the citizens who made it up. They do this by using demotics—an almost exclusively internal means of designation—to refer to the Athenians they honour, in one case in addition to the individual’s ethnic (ID 75), in the other, to the exclusion of it (ID 74). This use of an explicitly internal designation blurs the boundaries between these two communities in a particularly explicit way, representing Delos as, in a sense, part of Athens, at the same time as these grants of proxenia (like others made to non-Athenians, e.g. ID 76, for an individual from Ios) stress the polis status of this community.105 It is also striking that at least one of these Athenians is praised in terms, andragathia and dikaiosunē (the virtues of an internal public official), which make it clear that he was acting with authority as an Athenian magistrate within Delos.106 It is tempting to compare this tendency, to honour the officials sent by the dominant polis as a way of constructing relations, with the proxeny decrees made on a number of occasions by the Delphians to the epimelētai sent to them by the Aitolians.107 By Helisson is referenced not with an ethnic but with a toponym—ŁÆæe q ÆØ K  EºØ[ ]Ø . . . a Łı Æ Ł ŁÆØ a N  EºØ Ø . . . (ll. 8–10)—a reference more _ _ appropriate to a kōmē than a polis. _ 105 For a nuanced account of the relations between Delos and Athens (which treated the former administratively as an Athenian shrine), see Chankowski (2008), but note that she probably underrates the significance of the two anti-Athenian episodes attested: violence against the Athenian amphictyons in 377 bc (ID 98 ll. 24–30); and a Delian appeal for freedom against Athenian control in 345/4 or 343/2 bc (Plut. X Or. 840e, 850a). 106 ID 75; ID 79, which also refers to andragathia and dikaiosunē, should certainly be taken as another Athenian with this ethnic, or an Athenian demotic, restored in line 3. 107 FD III 4 175 (early fourth century bc); Syll.³ 534B (224–220 bc); 534 (218/17 bc); FD 1 451 (200/199 bc).

222

Proxeny and Polis

constructing these relations with, and asserting their own status with regard to, these official delegates, these communities were simultaneously negotiating more broadly their status in relation to the communities these individuals stood for and represented.108 Honorific decrees from the minor communities of Olymos and Myus in Asia Minor seem to have functioned in a similar way in constructing relations within sympolities in which each was the junior party. From Olymos, in particular, a number of decrees honour external individuals for whom no external ethnics are given, probably in token of the relationship of sympolity which existed between the two poleis.109 At the same time, however, by pointedly granting politeia and ‘the right to participate in all things on an equal footing with the other Olymians’, these decrees clearly assert the continuing status and political identity of Olymos as a polis. Similarly, after their sympolity with the Milesians, the Myousians are found collectively honouring one Apollodorus, son of Metrophanes, on the occasion of his dedication of four silver phialai to Apollo Termintheus.110 The honorand is probably the Milesian stephanephoros of the year 212/1 bc,111 making an official dedication in his year of office.112 Omitting his ethnic, which suggests that he was a citizen of the community, but honouring him in terms which make it clear that he is not a member of the Myousian community and which would probably have been followed by a grant of proxenia or politeia.113 the Myousians both acknowledge their closer relationship with Miletos at the same time as

108 For a similar use of civic honours in relation to royal officers, ‘to proclaim distance’, see Ma (2002), 207. 109 Blümel (2000) nos. 1–3 (= SEG 50 1121–2). For the interpretation of these texts and their relation with the sympolity between Mylasa and Olymos which follows, see Reger (2004), 164–8. 110 Milet VI 3 1029. 111 Milet I 3 124.27, but see Wörrle (1988) for the chronology. 112 So J. and L. Robert, BE (1966), n. 375. Gauthier BE (1989), n. 255 further identified this individual as likely to be the Apollodoros, son of Metrophanes, from Magnesia on the Maiander who had been granted the citizenship of Miletos in Milet VI 3 1060 col. II l. 2. If so, his accession to the Milesian stephanephorate within twenty years of becoming a Milesian citizen marks him out as one of the most prominent and politically successful recipients of the citizenship of another polis we know of in the Hellenistic period. 113 See Mackil (2004), 494–7; ‘[KØc]  ººøæ Åæ ç ı æ ÆØæ[  . . . HØ ]øØ    E ææ åæ Ø h ı[ Æıe Ø][º]Ø Ææå[] Œ Ø BØ  ŒÆd N ÆØ  E IÆ HØ ÆPHØ ıÅ ø ’ (Milet VI 3 1029 ll. 1–4) would be a typical example of the use of the proxenos-paradigm.

Proxeny and the Performance of Polis Identity

223

they stress their own, distinct polis status. Unfortunately, we do not know in either case whether the Olymians or the Myousians felt able to grant honorific status more widely to individuals from other poleis.114 It is, however, probably significant that the grants which each chose to inscribe related to individuals from the dominant poleis—in relation to whom these inscribed texts continued to assert an independent political existence, in front of audiences of Olymians and Myousians respectively.115 Relations between cleruchies (communities of citizens established on foreign soil) and the poleis which sent them out, although different in many respects from these cases, involved similar issues of integration and separation, and proxenia appears to have been used to perform analogous functions in this context as well.116 Two cleruchies are attested granting proxenia—the fourth-century Athenian cleruchy at Myrina on Lemnos and the second-century Samian cleruchy at Minoa on Amorgos.117 The extant proxeny decrees of the Samian dēmos at Minoa correspond closely to the sorts of patterns of behaviour observed in the case of dependent communities. Both concern individuals honoured for services performed at Minoa itself, and the latter, strikingly, names Ouliades of Samos as proxenos of the Samian dēmos at Minoa. From the effort devoted to describing Ouliades’ service as a doctor, it is clear that the primary aim of this decree was honouring him for his professional care. At the same time, however, in relation to an individual who was, on one level, a fellow citizen from the polis to which the cleruchy as a whole had come, the use of this inter-polis institution and its established decree formulations, which imply separate polis status, cannot have been wholly naïve.118

114 The Olymians were responsible for a grant of politeia for an individual from the unidentified city of Terssogassa (I.Mylasa 866), but, as Reger (2004), 165–6, convincingly shows, this dates to the period before its sympolity with Mylasa. 115 For the ed.pr, see. Herrmann (1965). A pierre errante, the stone itself was found at Miletos, the block on which it was inscribed transported from Myus. On the other fragments of the temple of Apollo Termintheus, see Koenigs (1981). 116 Cleruchies are included in Hansen’s typology of dependent poleis, Hansen and Nielsen (2004), 87 no. 5. 117 The decrees are: IG XII 8 3 (Athenian demos at Myrina) and IG XII 7 226 and 231 (Samian dēmos at Minoa). 118 Most explicit is the clause granting ‘all the other honours which proxenoi and euergetai of the polis receive’ which, along with the specific privilege of prosodos that follows, explicitly makes the claim that the Samian dēmos on Minoa was a separate polis ([åøæd ]b  ø  [Æ ‹Æ æåØ  E] æ  Ø [ŒÆd Pæª]ÆØ B

224

Proxeny and Polis

That proxenia could be used as a way of performing relations of dependency may appear to jar with the declarative function explored earlier—grants of proxenia staking claims for the identity of a community as a polis, a state actor—which seem to be supported by restrictions limiting the actions of certain dependent communities. However, it was because proxenia and, in a similar way, politeia, had these associations of statehood and state identity that it could be used in this fashion to negotiate the complexities of real relations between political communities. In this context, it is worth considering one other evidentiary lacuna which has been thought highly significant by a modern scholar in relation to dependency. This is the complete absence of honorific decrees for Athenians from poleis of the Athenian Empire in the fifth century bc, in contrast to the more than ninety Athenian grants (attested in both literature and epigraphy) collected by Walbank.119 Polly Low sees this absence not as the result of these communities’ loss of liberty, but rather of an Athenian desire for a one-way flow of honours (and thus obligations), honours which the Athenians gave rather than received.120 The evidence, however, cannot support this sort of argument. Our sample of inscribed honorific decrees granting proxenia from the mid to late fifth century bc from outside Athens is exiguous—with only Lousoi, Iasos, Lindos (and, after synoecism, Rhodes), Eretria, and perhaps Myrina, of the hundreds of poleis listed on the tribute lists positively attested inscribing grants of proxenia on stone during the second half of the fifth century (and none of these are represented by more than a couple of inscribed grants).121 It is hardly surprising if a sample this small fails to register any Athenians among the known honorands. This is simply a reflection of the fact that the inscription of proxeny decrees had not really

ºø [ŒÆd æ  ] K[d]  ı[ºc Œ]Æd B æø[Ø ][a a ƒæ], IG XII 7 231 ll. 14–16). The decree itself was to be sent to Samos (ll. 42–3). 119 120 Walbank (1978). Low (2007), 242–8. 121 Appendix no. 12 (Lousoi); SEG 36 982 (Iasos), placed by their original editors in the first half of the fifth century bc—certainly too early—contra Frei and Marek (1997), 56 n. 134 who down-date this text to the beginning of the fourth (possible, but perhaps a little late); I.Lindos 15 (Lindos), 16 (Rhodes); IG XII suppl. 549A and XII 9 187A (Eretria). On XII 8 2 (Myrina), see n. 86. See Lambert (2011), 203–4, similarly pointing out the precocity of Athenian epigraphic practice (and relating it specifically to the architectural development of the Acropolis in this period).

Proxeny and the Performance of Polis Identity

225

become common on stone outside Athens before the late fifth century, not that grants of proxenia, or grants of proxenia to Athenians, were rare under the fifth-century empire. A passage in the Old Oligarch in fact presupposes that Athenians were frequently honoured—both representatives sent out by the Athenians (the examples given are generals, trierarchs, and ambassadors), and others for their services in the contexts of lawsuits at Athens.122 Although we have little positive evidence, the way in which all of our literary sources treat references to proxenia suggests that, as an institution, it was probably already as universal and nearly as frequently used, in the early fifth century as it is later positively attested as being.123

4.3.3. Proxenia and polis recognition So far I have emphasized the declarative function which proxenia played, as a way of performing and, with a constant stream of grants, continued to perform the status of the granting community as a polis with links to others. It is also, however, worth briefly considering what a decree of proxenia signalled on the part of the granting polis about the status of the community referenced with the ethnic of the honorand. Proxenoi were conceived of as links to other poleis. To what extent, therefore, did grants of proxenia to individuals perform a constitutive function, recognizing the honorand’s community as a polis? Apart from cases where proxenia was granted to official representatives sent by another polis (where, as a reciprocating symbol in this inter-polis transaction, recognition is clearly implied), this sort of function is often quite hard to pin down. This is, not least, because, as these grants established relations with individuals in the first instance, except where grants were made to inter-polis delegations it is usually difficult to be sure how far a particular decree deliberately expressed relations with, and thus recognized the status of, the polis of the individual’s city ethnic as well. The city ethnic was, after all, part of the identity of the individual proxenos and the proxenos was thus likely to have a greater investment in this precise designation than the 122

[Xen.] 1.18. The frequent references to proxenia in Pindar, especially Isth. 4.7–9, are particularly suggestive, see Currie (2005), 340–3. Wallace (1970) collects and discusses the early material which gives the impression that proxenia was very common and widespread before it became well attested epigraphically. 123

226

Proxeny and Polis

honouring polis. However, examples where the same individual is designated in different ways in different communities—as an Ioulietan in one polis (his polis-ethnic) and a Keian in another (his island ethnic)—suggest that the use of ethnics is more complex than this would imply.124 Nonetheless, when we see no fewer than three of the Spartan perioikoi communities attested in the fourth-century Karthaian catalogue of proxenoi, this probably does suggest that the Karthaians did not view these communities as being sufficiently dissimilar from the other poleis to which they were connected by their proxeny network to make the use of proxenia inappropriate in these cases as well.125 The context strongly suggests that these grants were not marked assertions that these communities were poleis, challenging the hegemonic claims of post-Leuktran Sparta. Consequently, it seems that what is unusual about a comparable fifthcentury proxeny grant made by Argos for another perioikos, Gnosstas of Oinous, is the survival of the bronze plaque on which it was inscribed.126 The grant is more likely to reflect the sort of regular interactions which otherwise characterized the relations between nearby communities than an attempt to interfere with the perioikoi more broadly, and we are certainly not justified in seeing the use of this individual’s ethnic (Oinountios rather than Lakedaimonios) ‘as a deliberate affront to Sparta’.127 However, in a few instances it appears that something stronger was being asserted by the use of both proxenia and polis ethnics. Grants of proxenia, which symbolically established a permanent link between two poleis, were particularly pointed when they were made to individuals whose poleis had been physically destroyed. Such grants arguably signalled recognition by the granting polis of the continuing existence of the destroyed polis as an interstate polis actor. Thus, the probable Athenian decree of proxenia for a Plataian, after the destruction of Plataia by the Spartans and Thebans in 427 bc but before its refoundation in 386 bc, seems to represent a clear assertion that Plataia,

124 IG XI 4 693 contrast XII 5 820 (both second half of the third century bc). The fact that this individual was a doctor, and thus the services for which he was honoured took place at the polis in question, helps to explain why the distinction between the polis and island ethnic was less important for the granting poleis. 125 Appendix no. 10.1 (= IG XII 5 542 l.20–3). 126 SEG 13 239 (first half of the fifth century bc). 127 Cartledge (1979), 216.

Proxeny and the Performance of Polis Identity

227

from the Athenian perspective, continued to exist as a polis.128 Similarly, in 324/3 bc the Delphians issued a renewal of proxenia for a family of Thebans after the destruction of that polis by Alexander in 335 bc.129 The later physical refoundation of both of these cities is a testament to the potential practical consequences of this sort of continued recognition. To these two examples we can now add a third, an Akraiphian grant of proxenia to a citizen of Haliartos. Since both of these cities were members of the Boiotian League, this text dates to after 171 bc and the dissolution of the league, and, thus, also to after the destruction of Haliartos at the hands of the Romans (in fact, as Habicht shows, it can be placed between c.140 and 120 bc).130 Although there was little prospect of refoundation in this case, this text perhaps constitutes an act of communal solidarity by the Akraiphians for this lost polis, long after its destruction. In other instances, although physical destruction is not at issue, we see proxenia granted to an individual whose polis, as an independent actor capable of granting proxeny in its own right, has ceased to exist. Thus the Athenians made a grant of proxenia to an Ialysian after that community synoecized into Rhodes (something which the Athenians may not have welcomed, given their general approach to disaggregating local power blocs in the Aegean).131 Similarly, the Tenians appear to have made a grant of proxenia to a citizen of Minoa while Minoa was under Samian control.132 In these cases, both in the decision to name these particular individuals as proxenoi and in the public inscription of the ethnics which they still claimed, the honouring polis cooperated with the honorand in asserting the continuing existence of the political community in question as a legitimate polis, despite its incorporation into, or subjugation to, another polis. 128 IG II2 2 with Fossey (1985) for the restoration of the ethnic. See SEG 32 38 and Tracey (2003), 353, on the standard date (403/2 bc), contra Walbank (1982). The other possible archon years similarly fall in the period of the Plataians’ exile, see Gray (2011), 207 n. 135. 129 FD III 1 356. See Daux (1943), 16; Marek (1984), 219. See Arnush (2000), 302 with n. 32, on the interpretation of this anti-Macedonian gesture in the context of a number of others from this period. 130 Perdrizet, BCH 22 (1899), 94 no. IV.1; cf. Feyel (1936), 27, attributing this text to c.150 bc and Habicht (1993), 39–43, placing them, more firmly, in 140–20 bc on prosopographical grounds. See also Müller (2005), 99–101, and Ma (2005), 173–5. 131 SEG 28 48 (c.403 or c.294 bc); on the interest of the Athenians in encouraging the segmentation and disaggregation of their allies, see Ma (2009), 134–5. 132 IG XII 5 821 (second century bc) with Reger (1992), 381 n. 89.

228

Proxeny and Polis 4.4. INTER-POLIS INSTITUTIONS AND STATE IDENTITY: NON-POLIS ACTORS

Proxenia, as one of the most frequent and important ways in which cities interacted with external individuals and communities, was thus central to the assertion of external polis status within an interstate system which was conceived of as a society of poleis. In fact, polis identity was so bound up with the idea of state identity in this system, and with performance as a legitimate interstate actor on this world stage, that a number of non-polis state and quasi-state actors can be observed adopting proxenia and other institutions to assert their status within interstate society. We see this in the use of proxenia by Greek associations, like the association of Dionysiac artists, and even in the much more widespread use of this polis-based institution by Greek federations and ethnē. However, this is brought out even more clearly when we also see proxenia used in this way by nonGreek actors, as well as dynasts and petty kings. Given the restrictions placed on the ability of dependent communities to participate in this system of inter-polis institutions, it may seem surprising that such performances by definitely non-polis actors, like associations and synods, were not apparently contested. Indeed, on a number of occasions, Greek cities were perfectly willing to deal with the Dionysiac artists as though they constituted a polis, even seeking, in the case of the Magnesians, their recognition of asylia and the status of a festival.133 In the case of a synod of merchants and ship-owners, the Athenian boulē even granted them permission to erect an honorific image that they had voted for their proxenos, a portrait of him painted on a shield.134 This active cooperation contrasts starkly with the restrictions clearly imposed on the external 133

Rigsby (1996), 245–7 no. 103. For the proxeny decrees of the Dionysiac artists, see Marek (1984), 118 and Le Guen (2001), vol. I, nos. 30, 37 (technītai of Dionysus, Isthmia, and Nemea, second and first centuries bc), 61 (technītai of Dionysius, Egypt, before 246 bc), 72 (technītai of Dionysus, Rhegion, second half of the second century bc), 75 (technītai of Dionysus, Syracuse, c.100 bc), 76 and 77 (technītai of Aphrodite of Syracuse, 47–45 bc). The technītai of Dionysus at Athens who participated in the Pythia were also the recipients of a number of collective grants of honours made by the Delphians, which included proxenia (e.g. Le Guen 2001 no. 13). A synod of ‘the victors in sacred games from the inhabited world’ (H Ie B NŒ [ı] Å ƒæ ØŒH , OGIS 494, first or second centuries ad) is also attested making a similar use of proxenia. 134 IG II2 1012 (112/11 bc).

Proxeny and the Performance of Polis Identity

229

relations of dependent communities and their frequent incapacity, which must in some sense have been enforced from above, to grant statuses like proxenia and politeia. The context for the use of proxenia in these cases is clearly vital. In relation to heavily dependent communities, the unrestricted use of inter-polis institutions to create connections with other poleis was problematic because this use asserted claims to polis status that conflicted with the control which dominant communities asserted over them. By contrast, when these institutions were used by synods or associations, whose activities did not constitute a challenge of this sort to particular political communities, these issues probably did not arise.135 The symbolic importance of proxenia emerges even more clearly when we see it being adopted by non-Greek communities and actors as a means of engaging in interactions within, and asserting membership of, the wider community of states dominated by poleis. This is striking when we see obviously non-Greek communities making use of this polis-based institution—for instance in the bilingual proxeny decree of Kaunos from the late fourth or early third century bc,136 or the proxeny decree inscribed on bronze from Malta from the second or first century bc.137 These sorts of performances are even more remarkable, however, when they involve non-communal actors attempting to pass themselves off as poleis not only by granting proxenia, but in the decree-like form in which they make these grants. In particular, a number of dynasts and petty kings are attested granting proxenia: Mausolus, Satrap of Caria; the Bosporan Kings;138 the dynast of Paphlagonian Korylas;139 and Korris, the priest-prince of Labraunda.

135 For a discussion of the organization and activities of associations of Dionysiac artists in general, see Le Guen (2001), vol. II, 77–82, who draws out the ways in which, in participating in inter-polis institutions (especially in sending and receiving embassies and expressing links of friendship and syngeneia with poleis), associations acted as poleis in miniature. See also Aneziri (2003), 21–124, for a detailed regional and diachronic study of these associations. 136 I.Kaunos 1 (late fourth century to early third century bc); cf. I.Labraunda 42, a proxeny decree of sophisticated formulation of the fourth century bc (though inscribed in the third) from the Karian community the Plataseis. 137 IGUR I 3 (c.100 bc). Material from Malta reveals evidence of the influence of its Phoenician heritage until the late republic. For an overview, see Bonanno (2005), 51–123 and 137–63, as well as Coleiro (1971) and Perassi and Novarese (2006) on the numismatic material. 138 139 CIRB 1–5 (mid to late fourth centuy bc). Xen. Anab. 5.6.11.

230

Proxeny and Polis

In the decrees of Mausolus of Caria and Korris this imitation is particularly clear, involving the adaptation of formulae developed for polis institutions to very different structures. Thus the decree of Mausolus is prefaced ‘it was decided by Mausolus and Artemisia’ ([]  ÆıººøØ ŒÆd [æ]Ø ÅØ·).140 The grant made by _ begins similarly: ‘It was decided by the Korris, priest at Labraunda, priest Korris and the relations of Korris’ (åŁÆØ HØ ƒ[æ]E ˚ææØØ ŒÆd  E [ı]ªª [Ø  F ˚ææ]Ø ; also set up at Labraunda).141 _ _ Both mimic the standard bipartite structure of the Greek decree formula (‘it was decided by the dēmos and the boulē’). What is remarkable in these cases is that these individuals, by granting proxenia in these terms, opted for this civic status rather than the alternative, more kingly, model of royal friends, which might, on the face of it, have appeared more immediately appropriate (this is particularly striking in the case of the decree of Korris, produced in the last half of the third century bc, by which point systems of royal friends were becoming highly developed and widespread, although perhaps the usage of Mausolus, commemorated at Labraunda, was also important here). This clearly shows the dominance of the polis model, and the institutional system which went with it, for the performance of state identity in the interstate society in which they sought to assert their position.142 Certain state actors did not, of course, participate in this polisorientated system of institutions in the same way that others did. In particular, Hellenistic kings of the first rank neither granted nor received proxenia. The explanation in this case seems to be that they had little interest in asserting polis-like status within the interstate system. Instead of equality with poleis, they insisted on recognition of their superiority. This does not mean, however, that they were entirely able to avoid the influence of this polis-based and -biased system of institutions. As recent studies have clearly shown, in their interactions with cities, kings had little choice but to draw on the language of this institutional system.143 Rome appears on first glance to have been another exception. However, the Roman I.Labraunda 40 (= RO 55, 360s bc?). I.Labraunda 11–12 (c.250–225 bc). 142 The evidence for the Hekatomnid dynasty’s multifaceted engagement with the Greek polis-culture of the Aegean world is examined in Hornblower’s classic work (Hornblower 1982). 143 Ma (2002); (2003), 30; Billows (2008), 303–6. 140 141

Proxeny and the Performance of Polis Identity

231

institution of hospitium publicum seems likely to have been developed in imitation of proxenia—or even as a translation of it (in the works of Livy and Cicero, for instance, hospitium publicum is in fact the regular means of rendering grants of proxenia in Latin and is functionally difficult to distinguish from it).144 It is striking in particular that in its first attested occurrence, in 394 bc, this institution was used in relation to a Greek magistrate at the polis of Lipara, who, in working on behalf of Roman ambassadors captured by the Liparans on their way to Delphi, acted in a manner which could have been calculated to procure a grant of proxenia from a Greek polis.145 It is not difficult to see why the Romans might have adopted this institution, which was, within the anarchical interstate society defined by poleis they inhabited, both the normal means of establishing links with outsiders and a potent symbol for the construction of state identity.146

4.5. CONCLUSION Even in relation to non-polis actors, the structure of this Greek Mediterranean world was defined by the ubiquitous and incessant interactions which poleis engaged in with each other. These studies of proxenia and other elements of the specific institutional framework through which poleis constructed their interrelations clearly reveal that this was a world of peer-polity interaction, a society of networking poleis. What was distinctive about this system, however, was the way in which peer-polity interaction was bound up with questions of state identity. Interstate anarchy invested these general relationships, and thus the institutional framework through which they were expressed, with particular importance because authority was distributed throughout the interstate system, even after the Hellenistic kings 144 Cic. Ver. 2.4.145; Livy 21.12–13 (a probable example of proxenia granted by Saguntum in Spain). Nichols (2011), 424, is right to recognize the similarities between proxenia and hospitium publicum but wrong to argue that ‘the Romans tended to treat this kind of relationship more formally and legally than did the Greeks’. In Appian, the ambassador of Perseus to the Senate speaks of ‘your philos and proxenos, Erennius of Brundisium’ (Hist. Mac. 11.7). 145 Livy 5.28.4–5 with Ogilvie (1965), 690; cf. Diodorus Siculus (D.S. 14.93) who, unlike Appian, interestingly avoids retranslating hospitium publicum as proxenia. 146 For the wider context of the Roman use of Greek institutions, see Gruen (1984).

232

Proxeny and Polis

intruded upon it. By making use of elements of this framework, constructing relations with other communities, and reproducing the shared norms in which they were embedded, poleis and the other actors who imitated them were engaging in performances before their peers, asserting their own identity and competing for status. With the domination of Rome, however, and the consolidation of power in the hands of an emperor, interstate anarchy became hierarchy. Chapter 5 therefore explores what the consequent changes in this system of interstate institutions tell us about how poleis’ perception of their relations with each other, their reasons for engaging in interactions, changed as a result, and seeks to unpick the relation between the symbolic and practical functions which participation in these interpolis institutional systems served.

5 The Disappearance of Proxeny and the Domination of Rome Interstate Hierarchy from Anarchy

Grants of proxenia were one of the most important means by which Greek cities sought to establish their identity in relation to each other as poleis, as full participants in this international society of city states. There were other related institutions which served a similar function, but proxeny decrees, because of the frequency with which they were made and the numbers in which they were inscribed, represent the best attested element of this institutional system. Proxenia, indeed, stands more broadly for this system of honorific intercourse, which shaped how Greek cities thought about, and sought to construct interaction with, each other. However, to understand the significance of the role of proxenia, it is necessary to consider the counterfactual case: what would the absence of proxenia imply about relations between polis actors? In fact, we can cast this question in historical terms: Why, after five hundred years of relative stability, did proxenia apparently disappear during the early Roman principate? The answer, I argue, relates to the profound transformation which the world of the Greek cities underwent in the late Republic and early Empire from interstate anarchy to imperial hierarchy. In examining the particular reasons for the disappearance of proxenia and other institutions, we not only get a better sense of the significance of proxenia and what it was used to express, we are also forced to confront what empire meant for the Greek cities individually and collectively—how it changed their capacity and motivations for entering into relations with each other. As a result, the evidence for inter-polis institutions offers us a differently periodized perspective on Antiquity from the

234

Proxeny and Polis

point of view of the Greek city: an age of polis-networking, of proxenia, encompassing traditional periods—late Archaic, Classical, and Hellenistic (both High and Low), and the ‘long fourth century’—and then passing and disappearing into the age of empire, during the course of the first century bc.1 Before examining these broader questions, however, it is necessary to begin by considering how, methodologically, we can go about identifying decline or disappearance. In particular, a recurrent theme in this chapter is the relation between the epigraphic record, its trends and trajectories, and the fortunes of the institutions it attests. In this context, even when we have apparently substantial sets of data, how far can we take a reduction in the quantity of epigraphic material (i.e. relative absence of evidence) as evidence of institutional decline or disappearance? With its particularly rich source material, proxenia is a useful case study for exploring how far, and in what particular ways, trends in the epigraphic record can illuminate the institutions they relate to. More clearly than any other element of this institutional framework, proxenia reveals the effect of the transformation of the Mediterranean from an anarchical interstate society to hierarchy.

5.1. THE DISAPPEARANCE OF PROXENIA When dealing with numbers of inscriptions as large as we have for proxenia, bar graphs represent the best way to convey an accurate sense of the quantitative trends in the epigraphic record. But, in order for conclusions drawn from these graphs to be valid, we need to ensure that two conditions in particular are met. The first is that all of the different objects which are counted relate to the same phenomenon under study and the second is that we can usefully date these different objects in relation to each other. The difficulty is that proxeny is attested epigraphically in a number of different types of inscription and these objects are dated with widely varying degrees of precision. In order to enable comparison between the quantities of texts inscribed at different periods, texts have therefore been 1 On ‘High’ and ‘Low’ periods in Hellenistic history, see first Robert, OMS 2, 841; for the concept of the ‘long fourth century,’ see Ma (2000a); Gray (2011), 6 n. 4.

Disappearance of Proxeny

235

categorized by fifty-year period, using quantitative probability distribution2 and the decision has been taken to concentrate on one particular type of epigraphic monument—proxeny decrees inscribed by the communities which made these grants. This focus has meant excluding some classes of material, in particular ‘travelling decrees’ (decrees inscribed by another community or at sites beyond the administrative control of the granting community), and ‘reinscriptions’ (decrees clearly inscribed sometime after they were formally passed, especially to replace monuments destroyed in the meantime), all of which represent distinct epigraphic phenomena. Lists of proxenoi inscribed at the granting polis have been included, but counted in terms of the number of separate acts of inscription which they represent in order to avoid distorting the dataset. Thus, catalogues of proxenoi are counted the same as a single inscribed decree, whereas, for chronological lists of proxenoi, each distinct chronological section has been counted individually (i.e. each year for which a new list is added is treated as a discrete act of inscription).3 Incomplete inscriptions not preserving direct reference to the grant of proxenia have been included only where the restoration of proxenia appears certain or very probable. As a result of the application of this rule, the dataset underlying Figures 5.1 and 5.2 in this chapter, at 2,470 items, is smaller by perhaps a third than it would have been had all potential or even likely proxeny decrees been included and my figures for particular, individual sites, such as Delos, are substantially lower than others’.4 It should, however, represent a 2 On the usefulness of quantitative probability distribution, which enables the comparison of objects assigned a wide range of more and less precise dates, see Wilson (2009), 219–27. Objects with dates which range across the periods used (e.g. ‘fourth century bc’, ‘375–325 bc’, or ‘c.350 bc’ if the categories in question are ‘399–350 bc’ and ‘349–300 bc’) are divided fractionally between them rather than assigned according to the mid-point of the dating range. 3 Thus, for the great proxeny-list of Delphi (Appendix no. 5), I count the thirtyfour subsections in which the proxenoi are divided, representing the appointments for particular half-year periods, rather than the seventy-seven distinct grants actually made. 4 For example, this dataset includes only the 239 decrees for Delos in which the grant of proxenia is securely preserved. By contrast, Marek (1984), 247, talks of more than 500 Delian proxeny decrees, but this includes a significant number of texts which certainly do not include a grant of proxenia. Habicht’s figure of 467 Delian proxeny decrees (Habicht 2002, 15) excludes these texts, but still relies on taking all other Delian decrees for foreigners (IG XI 4 nos. 510–1021 and 1349) as proxeny decrees, including a large number of highly fragmentary texts.

236

Proxeny and Polis

more methodologically secure basis for conclusions concerning proxeny inscription and we probably have enough data to be confident that it presents a reasonably representative sample for this phenomenon overall—not of grants made, but grants inscribed. In any case, since the material which has been excluded from this particular dataset closely follows the same overarching chronological trajectories, the picture presented by it is not distorted by its principles of selection. The dataset on which these graphs are based will be presented online, within a wider database of proxenies created to accompany this volume which will be hosted by the Centre for the Study of Ancient Documents (). The epigraphic record of proxeny inscription in the Mediterranean world, thus represented, reveals a striking decline, with the inscription of proxeny decrees by the granting polis all but disappearing in the first century bc (Figure 5.1). From a peak in the second half of the third century bc, the epigraphic record for proxeny inscription apparently enters a steep decline. By the first half of the first century bc the number of proxeny decrees inscribed has fallen to almost nothing outside Delphi. Later, apart from a small trickle of material from Delphi (and a few decrees from the Black Sea), the inscription of proxeny decrees appears to be more or less extinct. This picture is 700 600 500 400 300 200 100

0 –1 14 50 9– 10 0 99 –5 0 49 –0 1– 5 51 0 –1 10 00 1– 1 15 50 1– 20 0

19 9

20

0 24

9–

25

0 29

9–

30

0 34

9–

35

0 39

9–

40

0 44

9–

45 9–

49

54

9–

50

0

0

Fig. 5.1. Graph showing the chronological distribution of proxeny decrees per fifty-year period (excluding those which were not inscribed by the polis responsible for the decree).

Disappearance of Proxeny

237

90 80 70 60 50 40 30 20 10

0 20 19 0 9– 15 14 0 9– 10 0 99 –5 0 49 –0 1– 5 51 0 –1 10 00 1– 1 15 50 1– 20 0 9–

24

0

25 9–

29

0

30 9–

34

0

35 9–

39

0

40

45

9– 44

50 9–

49

54

9–

0

0

Fig. 5.2. Graph showing the number of communities attested inscribing their own grants of proxenia during each period. This graph, derived from the same underlying dataset as Figure 5.1 shows the number of communities represented by at least one inscribed proxeny decree in the fifty-year period in question to show how widespread proxeny inscription was at any given time. In cases where a community only registers 0.5 of a decree in a particular period in my dataset, as a result of the method of quantitative probability distribution adopted, that community receives a similar weighting (0.5) for each period in this graph. No community is counted as more than 1 for any period for which an inscribed proxeny decree is known. 186 different political communities are represented (166 poleis and 20 non-polis actors including federations, associations, and dynasts).

corroborated, and refined, if we examine the same dataset from the point of view of the number of communities attested as inscribing a grant in each fifty-year period (Figure 5.2). Whereas the picture presented by the first graph (Figure 5.1) is heavily dominated by a relatively small number of prolific poleis, the second (Figure 5.2) allows us to chart how widespread the inscription of proxenia was at a particular moment in time. By comparing the two, we can see that, although the absolute peak in the number of proxeny decrees inscribed occurred in the second half of the third century before decreasing significantly in the first half of the second, the inscription of proxeny grants appears more widespread in the first half of the second century bc. Although the absolute volume of material decreased during the second century,

238

Proxeny and Polis

it was only going into the first century bc that the number of poleis inscribing proxeny decrees seems to have declined, and dramatically at that. Given the large quantity of material that we have for proxenia, and the number of communities from which it comes, it is tempting to believe that it can be taken as a representative sample of the decrees passed, and therefore that this downward trajectory in inscribing proxeny decrees reflects a similar decline in the number of grants of proxenia that were actually made. In effect, this decline in granting proxenia would have meant the reduction in size of networks of proxenoi within the Greek world, and the cessation of granting, their disappearance, since proxeny networks required constant renewal.5 Even if, as argued in the Introduction, proxeny inscription was inherently selective and only a small proportion of the grants made were inscribed on stone, as long as that proportion remained roughly constant then the near complete disappearance of inscribed proxeny grants would still be a strong indicator that grants of proxenia were greatly reduced. However, at the level of the polis or even region, this assumption that the proportion of proxeny decrees inscribed remained constant can be proved false. When we analyse local epigraphic traditions it swiftly becomes apparent that they can arise, spread, change, and disappear for reasons which have little to do with actual rates of granting and which may seem historically trivial. Nowhere is this more clearly demonstrated than at Oropos where a tenfold increase in the number of proxeny decrees inscribed from the first to the second half of the third century bc correlates with (and clearly depended upon) the Oropians identifying a new, more cost-effective means of publication—pre-existing statue bases rather than purpose-hewn stelae (Figure 5.3). Moreover, if we set this material in its regional context, it becomes clear that both the sudden increase in proxeny decrees inscribed and their subsequent rapid decline are paralleled in the epigraphic tradition for Boiotia as a whole. It is apparent that in the second half of the fourth century it became more important for Boiotian poleis to publish decrees granting proxeny on stone—an epigraphical fashion, which at some other poleis was similarly facilitated by inscription on

5

See Chapter 3, p.164.

Disappearance of Proxeny

239

160 140

Decrees inscribed on stelae (new)

120

Decrees inscribed on statue bases (reused)

100 80 60 40 20 0 349–300

299–250

249–200

199–150

149–100

Fig. 5.3. Graph comparing the differing quantities of proxeny decrees at Oropos inscribed on purpose-hewn stelae and pre-existing statue bases for each period.

pre-existing surfaces.6 It is not at all clear, however, that this expression of a desire to publicize one’s proxenoi in any way correlates with, or can be used as evidence for, actual rates of granting (Figure 5.4). In any case, the first two graphs depicting overall trends in the inscription of proxeny decrees (Figures 5.1 and 5.2) do not give an accurate picture of early patterns of proxeny granting. The epigraphic habit, or rather specific epigraphic traditions involving the regular inscription of proxeny decrees, did not originate at the same time as proxenia, but were a later, distinct development. Although the practice may have been foreshadowed by earlier inscription of proxeny decrees on bronze, the inscription of significant numbers of proxeny decrees on stone appears to have been an Athenian innovation of the last half of the fifth century bc.7 The decline of proxeny inscription, by contrast, is arguably a more revealing phenomenon, at least in relation to the fortunes of proxenia 6

As at Thespiai or Tanagra (the only other poleis attested represented by more than ten proxeny decrees in this period). 7 For the inscription of proxeny decrees on bronze, see Chapter 1, n. 203; for the earliest proxeny decrees on stone, see Chapter 4, n. 121.

240

Proxeny and Polis

180 Oropos 160

Other Boiotian poleis

140 120 100 80 60 40 20 0 399–350 349–300 299–250 249–200

199–150

149–100

99–50

Fig. 5.4. Graph comparing the number of proxeny decrees surviving for Oropos with those preserved for the rest of Boiotia.

as an institution. Ceasing to inscribe grants of proxenia on stone is likely to have been more significant than the failure of communities to inscribe before inscription became an issue. However, epigraphic traditions at particular communities can have their own internal trajectories that may reveal relatively little about the fortunes of that particular institution there. At Athens, where we have enough material to be able to trace clear trends over centuries, the long-term decline in the inscription of decrees granting proxeny seems to reflect changing attitudes to inscription rather than a dramatic reduction in the number of proxeny grants made. After the end of the fourth century, many fewer honorific decrees in general were inscribed and, of those that were, decrees honouring citizens constituted the largest proportion—a marked break with earlier practice (Figure 5.5). Even of the decrees honouring non-citizens which were inscribed subsequently, decrees in which proxenia was granted constitute a much smaller proportion of the total than those granting citizenship (Figure 5.6). Instead of indicating a decline in the use of proxenia as an institution this trend, and an increase in the length of individual texts, seems likely to reflect, at least in part, a tendency for inscription itself to be considered a more marked and substantial honour, granted less frequently and more commonly to individuals receiving grants of citizenship, which had a higher symbolic value at Athens than grants of proxenia.

Disappearance of Proxeny

241

250 Honours for citizens Honours for non-citizens

200

150

100

50

0 9– 2 19 00 9– 1 14 50 9– 10 0 99 –5 0 49 –0 1– 5 51 0 –1 10 00 1– 1 15 50 1– 2 20 00 1– 25 0

0

25

24

29

9–

30

0 34

9–

0 9–

35

0

40 39

44

9–

45

9–

49

54

9–

50

0

0

Fig. 5.5. Graph comparing the chronological distribution of Athenian inscribed decrees for internal honorands (citizens) and external honorands (xenoi).

60

Grants of proxenia Grants of politeia

50 40 30 20 10

50 1–

0

–0 49

–5 99

–1 00

15 0

14 9

0 20

19 9–

0 24 9–

25 29 9–

0 30

0 34 9–

35 39 9–

0 40

44 9–

49 9–

45

0

0

Fig. 5.6. Graph comparing the chronological distribution of inscribed Athenian decrees for foreigners granting proxenia with those granting politeia.

242

Proxeny and Polis

Mediterranean-wide trends in inscription are likely to be more significant than the local trajectories of particular poleis, especially the widespread shift which we see away from the inscription of proxeny decrees. Figures 5.1 and 5.2 illustrate how the choice to inscribe a decree of proxenia came to be taken increasingly infrequently by fewer communities until it ceased to be a realistic possibility at all. Care, however, is required here in weighing the significance of this wider epigraphic phenomenon. On the basis of the first graph (Figure 5.1), which shows that by the first half of the first century bc proxeny decrees had all but ceased to be inscribed by the poleis which made them, we might expect proxeny to be moribund in this period. In fact, other types of inscription reveal that it was still a vital and strongly functional institution at this time. The first chapter demonstrated how a number of proxeny texts from the first century bc closely adhere to the proxenos-paradigm delineated in that chapter and refer to specific intermediary services characteristic of proxenia. At the same time, lists of proxenoi which survive into the first century bc show that, alongside this continuing functionality, practices involving the regular appointment of substantial numbers of proxenoi also continued despite the dearth of individually inscribed decrees.8 The marked reduction which we see, both in terms of the numbers of decrees inscribed and the numbers of communities inscribing decrees, cannot therefore be interpreted directly as evidence for the degeneration of proxenia as an institution. Inscribing proxeny decrees was probably always a relatively marginal activity in the sense that in most communities, most of the time, proxeny decrees would have been inscribed only as the result of a deliberate decision by the authority responsible, singling out the honorands of particular decrees for this marked honour. Instead, this reduction seems to

8 Although many of the Samothracian lists of proxenoi can only be approximately dated to the first or second centuries bc (Appendix no. 14), the continuation of the practice of proxeny-granting to theōroi, which they reflect to the end of the first century bc, is attested by a Samothracian decree inscribed at Kaunos in the early first century ad (I.Kaunos 28). For other early examples, see also Appendix no. 16 (Thera) and 17 (unknown polis, Boiotia). It is also significant that at Gortyn, where the cost of inscribing individual decrees of proxenia was very low (consisting of the name and ethnic of the proxenos and a very abbreviated decree formula inscribed on a preexisting surface) up to half of the forty-six decrees known are attributed to the first half of the first century bc.

Disappearance of Proxeny

243

reflect the changing monumental priorities of Greek communities. However great the variation or narrow the selection of decrees inscribed, the epigraphic record—which ought to give us a representative sample of the decrees inscribed, if not of the grants of proxeny which were made—clearly reveals that the decision to inscribe was regularly made by a large number of communities, and therefore open to many more, until the second half of the second century bc, although there is already clear evidence of a steep decline in the number of texts actually inscribed. By the end of the first century bc, the inscription of decrees granting proxenia was apparently no longer a realistic possibility practically anywhere. Cities were no longer interested in the same way in inscribing this sort of monument for outsiders. This reduced interest in monumentalizing proxeny grants is evidence of how proxeny as an institution had come to be viewed by poleis, along with the relationships and links with other poleis which it embodied.

5.1.1. Corroborating the decline of proxenia The disappearance from the epigraphic record of proxeny decrees cannot be taken as direct evidence for the disappearance of this institution as it was understood by both poleis and proxenoi in the Classical and Hellenistic periods. We do, however, have other evidence that proxenia did indeed disappear at some point in the early Imperial period, or, at least, that it changed out of all recognition, losing the specific, functional paradigm which had constituted a stable basis for proxenia for a 500-year period previously. At Delphi, a continuous record of proxeny decrees runs into the second century ad. However, at some point in the early first century ad the format of these decrees changes in a way which clearly shows that proxenia has ceased to be a distinct and well-defined institution. In the traditional format, exhibited in decrees up to this point (the latest dated in Marek’s sequence is attributed to c. ad 23), proxenia is presented as defining the relationship between the honorand and Delphi. Proxeny comes first in the list of honours granted and is presented as determining the package of privileges which the individual received in addition to the honours specifically listed in the standard phrase ‘and he is to have whatever other honours and privileges the city gives to proxenoi and euergetai’ (ŒÆd ¼ººÆ ‹Æ

244

Proxeny and Polis

[ŒÆd] E ¼ººØ æ Ø ŒÆd P æª[]ÆØ ±  ºØ çغ ŁæøÆ ŒÆd [] ØÆ  øØ ).9 In the new format, conversely, the grant of proxenia occurs as an undistinguished, probably antiquarian, honorific title without particular content or significance. Proxenia is simply an extra item in the list of honours, occurring not only after politeia but even promanteia or proedria, a sure sign that it was no longer understood as signifying a particular relationship. In these later decrees, proxenia is also no longer presented as a status defining a specific package of honours to which the honorand is entitled as a proxenos. The honorand, instead, is simply to be given ‘all other honours which are given to good men’ (ŒÆd pººÆ 

ØÆ ‹Æ E ŒÆºE ŒIªÆŁE I æ Ø  ÆØ).10 Proxenia here seems only to be granted at all because of its historic identification with honours granted to outsiders and it is striking that the individuals honoured with it in these decrees are overwhelmingly identified as philosophers and artists, rewarded for their activities at Delphi rather than for playing the role of proxenos in their own communities.11 The reduced significance of proxenia, now devoid of the specific content which had previously defined its use, is apparent in a Delphian decree where it apparently serves as a means of designating honours for outsiders. In order to emphasize the high honours it granted, this text from the mid-first century ad asserts that the honorand in question was not to receive ‘the common form of politeia, of the sort others customarily receive from the Delphians because of proxeny or honour’ (º ØÆ . . . c ŒØ  , æ Æ j Ø[B] ¥ Œ m ¼[ Ø H ¼ºº]ø < >ØÆÆ º Ø Ææa ˜ ºçH ) but a more

9 e.g. FD III 2 160 (c. ad 23); 4 59 (beginning of the first century ad); see Marek (1984), 245. 10 e.g. FD III 4 34 (ad 79). These elements of the new format are found in all decrees subsequent to FD III 2 160 in Marek’s chronology (to which add the absent FD III 4 35 and 74 and BCH 70 (1946), 253 no. 7, as other decrees in this form; correct 3 107 to 4 107 on p. 246)—see Marek (1984), 194 and 245–6. See also Wilhelm (1942), 58–9, and Gschnitzer (1973), col. 662. The fragmentary FD III 3 105, with proxenia in second place in the list of honours, but retaining the traditional catch-all formulation (¼ººØ æ Ø), perhaps represents a transitional example between the two formats. 11 Contrast Puech (2012): Puech raises the possibility that these late recipients may have gone on to act as proxenoi traditionally had, but, though possible, this seems implausible and misses the point that the Delphians themselves seem no longer to have conceived of them as such.

Disappearance of Proxeny

245

exclusive form, the status of a damiourgos, and right to participate in cults restricted to the elite.12 A similar transformation in the signification of proxenia is visible in decrees from the only other region that continued to inscribe references to this institution, the Black Sea. From the late first and second centuries ad, we see a string of grants of proxenias politeia made by the Pontic city of Chersonese.13 Proxenia here in the genitive is clearly dependent on politeia. It no longer references the distinct institution of the Classical and Hellenistic periods and only serves to further qualify or explain the grant of citizenship, which is the substantive concern of these decrees. Following Gschnitzer, we should probably understand something like what we see in the last Delphic case, as politeia ‘because of proxenia’ (æ Æ ¥ Œ )—as a way of stressing the explicitly honorific character of these grants of citizenship.14 This is presumably the sort of proxeny, in essence a means of designating citizenship granted in an honorific context, which we are dealing with in Dio of Prusa (38.22) when, in c. ad 100, he mentions the proxenies (rather than proxenoi) which linked the two propontic cities of Nikomedia and Nikaia in the context of stressing their various common ties (ŒÆd æ Æ b å ). There may be further reasons why the reference to proxenia was retained in these specific contexts. In particular, Rudolf Haensch has argued that the Black Sea grants of proxenias politeia should be understood, in general, as an example of a phenomenon called ‘two-level sovereignty’ by Fergus Millar, by which communities, without necessarily challenging the overall framework of the Roman Empire and authority of the emperor, laid claim to ‘diplomatic consideration and respect, to the operation of internal self-government, and to self-representation in a manner which implied a degree of independence’.15 This sort of idea does seem to be explicit in the grant these proxeny decrees typically made to their honorands of the right ‘to sail to (eisplous) and from (ekplous) the city, in peace and in war, without the need for a treaty and without being liable for seizure’. This privilege, which was closely associated with proxenia, occurring 12

FD III 4 442. Decrees of proxenias politeia (first to second centuries ad): IosPE I² 357–359; 364 (SEG 56 876); 365; 697; SEG 48 999 (in ll. 20–1 æ [Æ º Ø]

Æ should be restored in place of æ [Æ , º Ø]

Æ ); 56 874 (cf. 56 871–7). 14 Gschnitzer (1973), col. 721, contra Wilhelm (1942), 49; cf. Marek (1984), 154. 15 Haensch (2009), 218–20; Millar (1996), esp. 165. See also Haensch (2005). 13

246

Proxeny and Polis

almost exclusively in proxeny decrees, clearly expresses a claim on the part of these cities to formally act as states and control access to their territory and engage in interstate relations involving war, peace, and interstate treaties. The Black Sea poleis may have been making a point of the fact that, since they were outside the Roman Empire, they still formally retained these powers unlike cities within Roman provincial boundaries, even though these were not abilities which they were able to exercise in practice, especially in relation to Rome or Roman citizens. As the continued granting of proxenia at Delphi shows, however, proxeny itself could be used in other ways, even by cities within Roman provincial boundaries. If valid, what this interpretation of this material from the Black Sea illustrates is the way in which the use of proxenia fragmented when it lost its institutional content, and different aspects of the meaning with which this content and typical use had endowed it in the Classical and Hellenistic periods were developed in different contexts. In an honorific decree from Byzantium, dated to the mid-first century ad (inscribed at the honorand’s polis of Olbia), we are perhaps presented with an intermediary stage in this disappearance of the institutional content of proxenia. In IosPE I² 79, Orontas, son of Ababos, is honoured for services which would be typical of the proxenos-paradigm, although using formulae that are very different from anything we see in our earlier decrees:16 ººa b ŒÆd BıÇÆ  ø  º Ø ŒÆ  a Æ Æ åæÆ ŒÆd H N e K  æØ º ø æÆ Æ ª  ı Tç º ı, ŒÆd ÆPe u æ a ºØa F Ææ , oø a d e A h ØÆ ŒÆd æ Æ ØÆ   , e  ø b Ææıæ EÆØ H

Nº ø N e —  º ØA Kd çØºÆ Łæø ÆØ ŒÆd   ÆØ . . . . . . since (he performs) many services for the polis of the Byzantines according to its public needs and is useful in the care which he gives to those (of the Byzantines) who sail to the emporion, and just as he has inherited everything else from his father, so he has inherited his father’s eunoia and proxenia for the dēmos, and all of the citizens sailing to the Pontus bear witness to his benevolence and care . . . 17 16 17

For bibliography on this decree, see I.Byzantion 3. IosPE I² 79 ll. 7–15.

Disappearance of Proxeny

247

Proxenia here is still in a sense connected with the performance of particular services. However, referenced in parallel with eunoia (a d e A h ØÆ ŒÆd æ Æ ØÆ   ), it is distanced from the formally granted title of proxenos and instead presented as a personal attribute motivating Orontas’ actions. Similarly, the emphasis on inheritance in this text seems to relate to the earlier tradition of proxeny as a hereditary institution, but again it is striking that Orontas is named citizen rather than proxenos. Earlier it would have been normal for the son of a proxenos who was himself honoured by the city in question to be named proxenos in his own right, or at least for the grant to be ‘renewed’ (even in cases where proxenia was explicitly granted to the father as a hereditary office). In fact, as Gschnitzer argues, it is entirely possible that the father had not been formally named proxenos in the proper sense of the term either. Proxenia here may simply be a short-hand, encapsulating the services described in detail in the case of the father in relation to his son, and thus a reason for granting honours, but no longer an honorific status in its own right.18 It is possible that we possess important supporting evidence for the change which proxenia underwent at the end of the first century bc in the onomastic record. If we analyse the 136 occurrences of the name ‘Proxenos’ which occur in the database of the Lexicon of Greek Personal Names (volumes I–Va) and, to correct for the different sample-sizes, compare them as a proportion of the total number of names for each period, the graph which results looks remarkably similar to that which we have for the inscription of proxeny decrees (Figure 5.7). After the beginning of the phenomenon we find centuries of comparably high levels of activity, declining to practically nothing by the start of the first century ad. What significance should we attach to this evidence? Although Greek naming traditions tended to be conservative in the short and medium term, with sons named after grandfathers and family names in some cases persisting for generations, it is striking that in the longer term names do fall out of use in connection with the fortunes of the concepts, divinities, and, in this case, institutions, with which they are connected. If a name loses the meaning and associations which caused it to be chosen in the first place, and does not gain other

18

Gschnitzer (1973), cols. 662–3.

248

Proxeny and Polis 0.18%

60

Number of attestations of the name proxenos 0.16%

50

Proportion of total number of names attested 0.14%

40

0.12% 0.10%

30

0.08%

20

0.06% 0.04%

10

0.02%

0 0 50

0 40

40 1–

0 30

30 1–

0 20 1–

20

10 0

10 1–

1–

–1 99

–1 00

0

19 9

20

0 30

9– 29

9– 39

49

9–

40

0

0.00%

Fig. 5.7. Graph showing the number of attestations of the personal name ‘Proxenos’ per century, both in terms of absolute numbers and as expressed as a proportion of the total number of named individuals dated by the LGPN within each century (except for the fourth and fifth centuries ad where the total sample size is too exiguous for this proportion to have any validity). Source: Lexicon of Greek Personal Names, I–Va (136 attestations of the name ‘Proxenos’ in total).

associations, over time it is likely to be selected against and decline if not disappear. ‘Proxenos’ seems to have been used relatively commonly as a name in the Classical and Hellenistic periods because of the prestige and status associated with proxenia, the reason individuals actively sought to become proxenoi.19 The near-complete disappearance of this name from our onomastic record between 100 bc and ad 100, in parallel with a similar decline in the epigraphic record, may well reflect a significant perceived decline in the importance of and status associated with this institution. This varied material presents us with a complex, nuanced picture of the decline of proxenia. The different pieces of evidence do not precisely coincide because they illustrate different but related phenomena. In the case of the epigraphic record, what we see appears to be a general, long-term shift by poleis, beginning in the late Hellenistic period, away from commemorating monumentally these links with other polis communities (a shift which is apparent for other institutions

19

See Chapter 2, Section 2.2.

Disappearance of Proxeny

249

in the epigraphic record). The onomastic material, conversely, would present us with the perspective of individuals, their reaction to this communal reorientation visible in the monumental evidence. The reduction in the use of ‘Proxenos’ names arguably reflects an awareness of the reduced honorific emphasis placed on proxeny and thus a similar reduction in the claims to prestige and status capable of being communicated by this sort of reference to it. Finally, the material from Delphi and the Black Sea shows proxenia’s eventual loss of integrity as an institution—its transformation from well-defined status-relation to an honorific title, largely empty of specific content but worth making at Delphi and in the Black Sea because of the long association between the title of proxenos and honours for outsiders and, perhaps in the latter case, the implications of communal status which granting it had earlier conveyed. The fact that this final transformation of proxenia lagged behind the other changes which we can observe is not surprising. It is an expression of institutional conservatism, of the same forces which contributed to the stability of this institution over its lifetime. Proxenia depended for its function on universal compatibility, on participants in widely dispersed communities understanding the expectations involved and sharing a common set of assumptions, and this made proxenia resistant to change. It was only after this institution had ceased to be used in practice by most poleis that innovation could arise at a local level, reshaping the remnants of proxenia according to its now limited role in particular communities, and it took time for these changes to find expression in inscribed decrees in the face of the cumulative weight of traditional, highly conservative formulae.

5.1.2 Functional obsolescence? The picture we are presented with by our evidence for the decline of proxenia is thus unusually rich and informative. The question remains, however, of why proxenia declined at all. The timing of the different processes which I have identified as contributing to this development clearly correlate with the transformation of the Greek world effected by Rome. The Roman Empire, however, is not a sufficient explanation in itself for the decline of proxenia. What was it about Rome, or the structural changes which it caused in the world of the poleis, which brought about proxeny’s demise? Since, as I have argued, the practical functions which proxenoi fulfilled were an important basis for the

250

Proxeny and Polis

long-term stability of this institution, it makes sense to consider first what difference Roman imperial administration and institutions would actually have made to the sorts of intermediary services performed by proxenoi. Like other services performed by proxenoi on behalf of private individuals, aid given in enabling access to local legal institutions and in providing direct legal support (perhaps as advocate) is rarely directly attested.20 However, the law would certainly have been one of the contexts in which the services of a proxenos were most important. Jurisdiction was also probably the area of Greek civic life in which the Roman Empire intervened most directly.21 The Roman governor constituted the highest legal authority within his province, determining jurisdiction, establishing courts, and hearing cases. In addition to a patchwork of specific rights of jurisdiction granted to particular communities and individuals, which expressed the authority of Rome over jurisdiction,22 for some provinces there were also more comprehensive regulations which might have effectively eliminated the utility of proxenoi in this context. In particular, the provisions of the lex Rupilia entrust jurisdiction in Sicily to the legal institutions of cities over cases restricted to co-citizens, but refer cases arising between the citizens of different cities to an external panel, composed probably of Sicilians.23 Cases arising between Sicilians and non-Sicilians without Roman citizenship are not explicitly mentioned, but are presumably among the ‘other cases’ in which ‘it is customary to select judges from the conventus of Roman citizens’.24 In each instance, the concern appears to be for establishing neutral courts for both litigants: in the first case, a panel of Sicilians heard the case (rather than the court of one of the litigants’ cities) to circumvent any bias on the part of the jurors or legal institutions themselves in favour of citizens; in the second, a panel of Roman citizens was used presumably to avoid any preference Sicilians might have in favour of fellow islanders against IG XII 5 528 (Karthaia 303/2 bc) is the only explicit example. Fournier (2010) and Kantor (2010) are the most recent significant works on the problem of jurisdiction under the Roman Empire in the eastern provinces and Sicily. 22 Grants to communities: e.g. SEG 39 1243–4 (Colophon, after 120/19 bc); IAphr2007 8.27 (Aphrodisias, document 39/8 bc, inscribed second to third centuries bc). Grants to individuals: e.g. CIL VI.viii.3 40890 (78 bc). 23 Cic. Ver. 2.2.32 with Kantor (2010), 188. 24 ‘Ceterarum rerum selecti iudices ex conuentu ciuium Romanorum proponi solent’, Cic. Ver. 2.2.32. 20 21

Disappearance of Proxeny

251

an outsider. With such a system there would have been no need for proxenoi to facilitate access to local courts for outsiders, Sicilian or nonSicilian, and arguably these probably less biased courts, backed by Roman authority, would have represented a more attractive prospect than the local judicial systems proxenoi may have facilitated access to. How far the provisions of the lex Rupilia were typical of the judicial regulations of other provinces is difficult to say. Cicero as governor of Kilikia seems to have had more scope than this would have allowed to permit the use of panels of foreign judges (presumably in deciding cases arising within cities), following the precedent of the famous jurist, Scaevola.25 Georgy Kantor is probably right to stress the untidiness in general of the situation outside Sicily in the Late Republic, with a range of potentially conflicting judicial privileges granted to different individuals and communities—an untidiness which would have had the effect of intensifying Roman intervention, as the imperial power was forced to rule on conflicts between grants.26 Nonetheless, it would not be surprising if outsiders regularly sought to circumvent local—biased—legal systems entirely by seeking access to external (and more authoritative) Roman courts, whether or not they possessed proxenoi. In the area of jurisdiction, it is particularly easy to see how the role performed by the proxenos in facilitating local judicial access for individuals from the city to which he was affiliated could have diminished or in some areas been largely superseded under the Roman Empire. Some other contexts for proxenos activity, on behalf of another polis, would have been eliminated entirely—in particular warfare between poleis which might involve proxenoi in facilitating the ransom of prisoners, ensuring burial for the dead, or acting as knowledgeable (and persuasive) envoys to the cities to which they were connected in order to negotiate the cessation of hostilities. In other contexts, however, most of the functions which proxenoi tended to perform were probably comparatively undisturbed by the domination of Rome, especially the role of proxenoi in facilitating economic contacts between the citizens of different poleis. This is not to say that nothing changed. The arrival of Italian negotiatores in the eastern Mediterranean, first attested in the late third century and 25 Cic. Att. VI.1 l. 15. On the interpretation of this passage I follow Larsen (1948); see Kantor (2010), 195 n. 24 for bibliography. 26 Kantor (2010) 194–200.

252

Proxeny and Polis

present in very large numbers by the early first century bc, must have had an important impact on the economic landscape of the Greek cities.27 However, for all that these new economic agents would have been, on one level, incompatible with proxenia—though capable of being named proxenoi themselves, they did not belong to a polis able to assist them with a network of its own—their appearance would not have altered the conditions under which proxenoi were useful in facilitating interaction. In particular, there is no reason to believe that private (especially economic) interaction between the citizens of Greek cities ceased to be important or very frequent, and the problems of establishing identity, and the usefulness to visitors of local knowledge, would have persisted. There is also no evidence to suggest that the provision of hospitality ceased to be an issue in the Roman period any more than it had in the Hellenistic.28 In IG IV 853, probably the last proxeny decree we have clearly expressing the proxenos-paradigm inscribed by a granting polis (c.1 bc), the hospitality of Lucius Licinius Anteros, the Corinthian, to magistrates and ordinary citizens visiting from Troizen is listed prominently among the deeds which made him worthy of being named proxenos. Hospitality is also one of the most prominent services performed by Iunia Theodora, a resident of the Roman colony at Corinth, for officials and private citizens of both the Lycian league and its constituent poleis (three of the decrees on this monument at Corinth mention her services in this regard).29 Evidently this need had not disappeared by the mid-first century ad when this inscription is best placed.30 The services for which the honorand of 27 The classic works on the Italian negotiatores in the Republican period are Hatzfeld (1919) and Wilson (1966), now partially superseded by a recent collection of regional studies edited by Hasenohr and Müller (2002). Probably the earliest negotiator attested epigraphically in the eastern Aegean is Publius Arellius, son of Quintus, inscribed towards the end of the third century bc on the Eresian catalogue of proxenoi; presumably the forerunner of a family of negotiatores of that name later well attested throughout Greece and the Aegean (Appendix no. 7 l. 48; Mack 2012, 219). 28 On hospitality in general and its chronological development, see Hennig (1997) who flags up the fact that hospitality to Roman officials could be particularly burdensome (Hennig 1997, 363–8). 29 SEG 18 143, esp. ll. 27–8, 58–9 and 74–6 with Hennig (1997), 357. 30 Behrwald (2000), 120–8, argued for a date shortly after the foundation of the colony at Corinth in the mid-first century bc, but the mid-first century ad date is more persuasive (Jones 2001, 166–7). Gschnitzer (1973), col. 663, on this omission (and the disappearance of proxenia). On this text in general, see Robert, OMS 2, 840–58.

Disappearance of Proxeny

253

the ‘transitional’ proxeny decree of Byzantium, discussed above, is honoured, though different from those we find in our earlier decrees, strongly suggest a continuity of functional need, referencing ideas of utility (both on public and private levels) and an intermediary function which was earlier central to the proxenos-paradigm (here made particularly explicit with the use of the term prostasia).31 Proxenia was not simply superseded, as Paul Monceaux suggested, by the institution of patronage, adopted by the Greek cities from Rome.32 Patroneia, as it was cast in Greek, was used only in relation to Rome and represented an attempt to co-opt a Roman institutional language which could communicate more effectively to particular Roman addressees the sense of permanent relationship, involving mutual obligation, inherent in Greek proxenia. Between Greek cities, however, nothing replaced proxenia and, among the Greek cities of the east, patroneia itself did not long survive proxenia and arguably declined as the result of the same loss of plurality, with the establishment of an emperor.33 Other institutions were probably used in different contexts to perform private services analogous to those of civic networks of proxenoi during and after the decline of this institution. In particular, it seems likely that, in the economic sphere, private associations operated to lower transaction costs between individuals from different communities in a similar way, sharing local knowledge and personal networks (functions which private networks of friends, where individuals possessed them, had always continued to perform alongside proxenoi).34 These arrangements may have become more important after the decline of proxenia. However, like earlier judicial symbola or the Athenian dikai emporikai, which existed alongside proxenia, associations did not supersede it (indeed associations are attested as making use of proxenia themselves).35 Instead, both the use of associations in this context and proxenia reflected structural needs within the urbanized Greek world which persisted. As an explanation of the decline of proxenia, the obsolescence of its functions is thus insufficient, at least in relation to private contexts for 31

IosPE I2 79 ll. 7–10. Monceaux (1886), 315. On ‘patronage’, see Eilers (2002); Ferrary (1997b). 33 Eilers (2002), 161–81. 34 On the economic role of associations in the Hellenistic period, see Gabrielsen (2001); (2007); on those active at Hellenistic and Roman Delos, Rauh (1993). 35 See Chapter 4, Section 4.4. 32

254

Proxeny and Polis

the services of proxenoi to individuals. It is thus tempting to connect the decline of proxenia with its functions in the context of public inter-polis contacts. Such a connection would, on one level, be unsurprising. Proxeny decrees themselves explicitly focus on the services of proxenoi for the polis as a whole in facilitating public interaction, especially in aiding official delegations sent by the granting polis, or for their own performance on delegations sent to it.36 The decline in proxenia and its commemoration may be related to a reduced need for the functions of proxenoi in these public contexts, and thus also reduced opportunities for individuals to perform services of this kind, if the number of delegations sent—the amount of official inter-polis interaction—declined. In our evidence for the first century bc we do see a remarkable shift, in parallel with the disappearance of proxenia, in which the other institutional and honorific forms of interaction between poleis which characterized the pre-Roman Greek world vanish from our sources. It is not, however, that proxenia disappeared simply because these other institutions did. Rather the disappearance of proxenia needs to be understood in the context of the degeneration of this wider self-supporting system of institutions as a whole, which indicates a profound change in the structural dynamic of the world of the Greek poleis at this time.

5.2. CHANGES IN INTER-POLIS INTERACTION Proxenia was part of a broader system of interstate institutions.37 By the end of the first century bc, however, the principal elements of this well-developed system, after being highly visible in the epigraphic record of the Classical and Hellenistic periods, more or less completely disappear. This section examines the evidence for a number of these different elements—theōria and theōrodokia, the exchange of treaties and formal statuses (including isopoliteia and asylia), as well as interstate arbitrations and foreign judges. None of these elements

36

See Chapter 1, Section 1.2.4; on the function of the suppression of details of specific private interactions, see Chapter 2, pp. 103–4. 37 For an overview of the scholarship on many of these institutions over the last twenty years, see Saba (2010).

Disappearance of Proxeny

255

can lay claim to a material record as rich as that of proxenia, which also stands for and represents honorific interaction in general to a far greater extent. However, the explanations which emerged from considering the more copious evidence for proxenia highlight and invite us to ask similar questions of the evidence for changes to these other institutions. Precisely because they are more specifically focused, the other parts of this system illuminate more brightly particular aspects of the change which the Greek world of poleis underwent as a whole.

5.2.1. Theōria and theōrodokia One of the most important means by which poleis expressed and symbolically maintained their connections with each other was the institution of theōria—the official delegations, composed of theōroi, which poleis despatched to each other to announce or attend festivals, consult oracles, or visit and make dedications at particular shrines.38 Sending a delegation of theōroi to the festival of another polis, especially in response to an invitation issued by a representative of that city, performed a similar constitutive function to the inter-polis interchanges discussed in the previous chapter, as each polis recognized the other and in turn had its own identity as a polis recognized. However, as well as these bilateral exchanges, the institution of theōria allowed for more multilateral performances of polis identity, when multiple theōriai representing different poleis converged at the festival of the same polis, whether in front of a very wide audience (as at a panhellenic shrine such as Delphi) or a more local one. It is thus not surprising that in these contexts we see concerns expressed over how more marginal communities are allowed to identify themselves. A particularly good example of this is the Lebedians, considered above, who, when synoecized into the polis of Teos, were required by the regulations set out in a letter from King Antigonos to be part of the Teian delegation and identify themselves as Teians when participating in the festival of the Panionion.39 The importance and frequency of theōria in the Classical and Hellenistic periods is probably difficult to overestimate. Like internal 38 39

Rutherford (2013), 5–6, for an exhaustive list. Rutherford (2013), 219–20; RC 3 ll. 1–4. See Chapter 4, pp. 205–6.

256

Proxeny and Polis

civic festivals, these external festivals, both at major panhellenic and regional shrines, and on a more parochial level, between neighbouring communities, are likely to have been the major events of the polis calendar. Some sense of the scale of this phenomenon is reflected in proxeny-granting, certainly in the great lists of theōroi named proxenoi at Samothrace, but also in the patterns of grants inscribed at Delphi and Epidauros.40 The sheer volume of proxenoi at local cities known from Astypalaia or Eresos probably also reflects the receipt of delegations, of which a significant proportion are likely to have been theōroi. For particular sanctuary-orientated poleis, the reach and variation of these theoric connections can also be traced through inscribed records of their networks of theōrodokoi. Theōrodokia as an institution was a very close parallel to proxenia, except that the context for the intermediary services implied here was much narrower, limited to visits of festival-announcing public delegations.41 At some poleis, these two civic networks—of proxenoi and theōrodokoi—seem to have been separately maintained, in parallel (as, apparently, at Delphi, though there was some overlap even here);42 at other cities, these institutions were used together (for example, at Epidauros).43 Given that the functions associated with theōrodokia could easily be encompassed within the broader institution of proxenia, it is difficult to explain why the former developed at all, leading to apparent institutional redundancy. However, it surely persisted because it had simply become de rigueur for communities possessing significant shrines to maintain networks of both proxenoi and theōrodokoi and this highlights the importance of their symbolic value. After relatively high visibility from the Classical period onwards, like proxenia, theōria, as a form of direct official contact between communities in cultic contexts, declines sharply in the epigraphic record in the early first century bc. In particular, in the case of theōrodokia, although the inscriptions relating to this institution are not sufficiently homogeneous in type or numerous enough to produce a useful graph as we can for proxenia, a similar pattern in inscribed grants is clearly visible. The last known grants of theōrodokia

40

Appendix no. 14, 5, and 4; see Chapter 3, pp. 168–70. 42 Perlman (2000), 18–22, with bibliography. Perlman (2000), 28–9. 43 For the Delphic network of theōrodokoi, see Plassart (1921) with Robert, OMS 1, 327–44 on lines 58–77. On the overlapping use of proxenia and theōrodokia at Epidauros, see Appendix no. 6. 41

Disappearance of Proxeny

257

date from the second century bc and the very last reference we have to this specific institution was inscribed on a stone dating from the mid-first century bc.44 Because we possess no later evidence relating to theōrodokia, we cannot positively prove that it disappeared, or, like proxenia, lost its content and function as an institution. However, it seems unlikely, given that proxenia did disappear, that the much more specialized theōrodokia continued, as it was never granted by nearly as many poleis and was closely tied up with proxenia. If theōrodokia had continued in any significant way to be granted or even if it retained honorific value much beyond the middle of the first century bc, we might expect to see some trace of it in the honorific decrees of Delphi, which continued throughout this period. The disappearance of theōrodokia indicates a waning of interest on the part of significant shrines in maintaining civic networks with explicitly religious connotations and fits into a wider picture of theoric decline, although the evidence here is somewhat more ambiguous. Given how deeply embedded particular shrines were, and continued to be, in Greek religious culture, it would be very surprising if these sorts of polis delegations disappeared entirely. It is striking, however, how little material we have relating to theōria from the Imperial period, especially as, in other respects, Greek festival life appears to have flourished.45 The post-Augustan material for theōria contributes only six entries in Ian Rutherford’s compendious catalogues (excluding the material from Claros).46 Unsurprisingly a significant proportion of the examples we do have reflect the continuation of very long established theoric links—between Athens and Delphi, Athens and Delos, and Keos and Delos.47 It is noticeable also that, in our later epigraphic record, the attestations we do have mention theōria on monuments erected to honour one’s own citizens rather than to emphasize links between cities.48 Where we do have particularly rich epigraphic sources, these seem to reflect a very

44

45 Perlman (2000), 26–7. See, for example, IAph2007 15.330 (ad 180s). Rutherford (2013), Appendix G, 435–8. 47 Athens and Delphi: FD III 2 65–6 (end of the first century ad). Athens and Delos: ID 2535, 2536 and 2538 (second century ad). Keos and Delos: ID 2539 (second century ad). 48 Thus I.Eph. 891–6 (second to third centuries ad), honours for a series of female theōroi to the Olympic Games (on which see Robert, OMS 5, 669–74); IG XII 5 946 (the third century ad) notes that the honorand served as architheōros four times without mentioning any of the external poleis to which he was sent. 46

258

Proxeny and Polis

different sort of phenomenon, as at Claros.49 The way in which these interrelations were restructured under the Roman Empire emerges particularly clearly in the case of the civic delegations sent to Ephesos (attested by dedications they made there), for which the context was the first neōkoreia of that city—that is a provincially organized, imperial cult.50 As in the case of proxeny inscription, the epigraphic record for honorific decrees is unlikely to map exactly onto patterns of the use of the institution. It is striking, for instance, that the last of the lists of theōroi inscribed at Samothrace as receiving proxenia are dated to the first century bc, whereas we find a later example of a Samothracian grant of proxenia to Kaunian theōroi inscribed at Kaunos in the early twenties ad.51 What the epigraphic record directly attests is that the decision to inscribe public decrees attesting to links of theōria was much less frequently taken from the first century bc onwards. However, as with proxenia, it seems highly likely that this reduced monumental emphasis corresponded to a reduction in the effort devoted to the institution of theōria and the formal links which it constructed.52 This is partially corroborated by the fact that theōroi also disappear from our epigraphical record of inscriptions of other sorts, including, notably, dedications, which unlike decrees honouring visiting theōroi reflect the commitment and interest of the cities despatching theōroi and, perhaps, the theōroi themselves. Thus, again at Samothrace, although dedicatory inscriptions and inscribed records of initiates do not disappear, theōroi cease to be recorded in either capacity after the end of the first century bc. This reduced monumental and honorific emphasis on theōria is perhaps also reflected in trends in the attestation of names related to this institution. Although we possess a sample that is significantly smaller than that which we have for the name ‘Proxenos’ and one which is also much more distorted, heavily dominated in particular by Athens (from which about two thirds of the sample comes), the disappearance of ‘Theoros’ as a personal name after the end of the first century bc is similarly marked, and probably reflects the 49 On the material from Claros, see Robert, OMS 6, 540–8; (1969), 299–312; J and L Robert (1989), 3–6. For an overview and detailed chronology of these texts, many as yet unpublished, see Ferrary (2005). In particular, the absence of older poleis and the prominence of coloniae is striking. 50 I.Eph. 232–42; 1498; 2048; on these dedications made between ad 88 and 91, see Heller (2006), 245–54 and 163–237. 51 52 I.Kaunos 28. Rutherford (2013), 47–50 and 355–60.

Disappearance of Proxeny

259

20 18 16 14 12 10 8 6 4 2

0 50

0 40

40 1–

0 30 1–

0

30 20 1–

20 10 1–

10 0 1–

–1 99

–1 00

0

19 9

20

0 9– 29

30 9– 39

49

9–

40

0

0

Fig. 5.8. Graph showing the total number of attestations of the personal name ‘Theoros’ per century. Source: Lexicon of Greek Personal Names, I–Va (45 attestations of the name ‘Theoros’ in total).

response of individuals to the reduction in prestige associated with links between communities of this kind (Figure 5.8).

5.2.2. Treaties and statuses In the Greek world, cities constructed relations with each other directly by agreeing treaties, engaging in the mutual exchange of honours and privileges, and granting or recognizing particular statuses which other cities laid claim to (e.g. asylia). This overlapping set of institutions represented a series of gestures which, if they were not necessarily utilized as frequently as delegations of theōroi or grants of proxenia, were nonetheless more highly marked individually and played a central role within the broader system of interstate institutions. Even allowing for this, however, and partial as it is, our evidence paints a very vivid picture of interstate activity. The standard modern collections, which draw on literary as well as epigraphic sources and thus privilege the more powerful states, collect some 435 treaties attested for the ancient world for the five hundred year period between 700 and 200 bc (so far omitting 200–100 bc, a

260

Proxeny and Polis

productive period), of which the vast majority involve at least one Greek polis (the majority were between two).53 ‘Treaty’ does not really do justice to this range of gestures, and is sometimes stretched to breaking point in these collections—as in the case of StV III 508 which, rather than being a bilateral agreement, is really a series of reciprocal privilege grants, a very common form of interstate interaction otherwise uncollected in these volumes. Specialized works, such as Gawantka (1975) on treaties and inter-communal grants of isopoliteia and Rigsby (1996) on asylia, contribute to filling this gap, and show (especially in the case of Rigsby’s work) how numerous and widespread the interactions which a single community might enter into could be.54 The familiar trend recurs here as well. During the first century bc, practically all of these institutional forms of contact, by which poleis sought to construct relations with each other, disappear from our sources. In the case of relations such as alliances (symmachia), this disappearance, at least under the Principate, is hardly surprising. Under the Pax Romana, interstate warfare was no longer among the competencies of the Greek cities and therefore military alliances (symmachia) would have been frowned upon.55 It is also possible that the disappearance of grants of isopoliteia to entire communities, whether mutual or one-directional, may reflect a Roman desire to divide those they ruled.56 Although these sorts of grants, especially between non-contiguous communities, did not result in the creation

53 StV II2 and III; Chaniotis (1996) collects a further thirty-eight treaties for Crete alone in the second century bc. 54 Gawantka (1975), esp. catalogue, 206–20. 55 See Brélaz (2008); Plutarch explicitly states that the affairs of the poleis no longer encompassed ‘acts of alliance’ (ıÆåØŒa æ  Ø, Praec. Ger. Reip. 805 a). R. KalletMarx is surely correct that earlier, under the late Republic, the power vacuum which the Romans left in the east made life more uncertain for Greek poleis, but even in this context, as he points out, the treaties which poleis were keenest to acquire and inscribe in this period were treaties with Rome (Kallet-Marx 1995, 192–3). 56 Jones (1940), 172; Gawantka (1975), 199; Jones (1978), 83–4 and 88–9 clearly shows that homonoia coinages under the Roman Empire do not reflect either alliances or treaties of isopoliteia. Some sympolities (the merging of contiguous communities), conversely, may well have gone ahead in Lykia under the influence of Rome, perhaps for administrative reasons: Zimmermann (1992), 140; Schuler (2010), 396. A similar process presumably underlies the reduction in the number of Cretan poleis in the Roman period. For Augustus’ promotion of koina as intermediary units to facilitate administration, see Bowersock (1965), 91–100.

Disappearance of Proxeny

261

of larger poleis, the claims to closeness, even unity, which they expressed may have been perceived by the Greek cities as being incompatible with Roman insistence in other contexts on maintaining the distinctions between cities: in Bithynia, there is evidence for this in the Pompeian prohibition on citizenship grants to citizens of other Bithynian poleis; in relation to Macedonia, we know of the decision to partition the kingdom into four separate administrative zones in the aftermath of the third Macedonian War, the boundaries between which were reinforced by a prohibition on intermarriage between regions and the ownership of property in more than one.57 Alongside, or even discounting this sort of evidence of Roman opposition, we should also allow for the probability that, for the cities themselves, expressing links like this, or commemorating them on stone, ceased to seem worthwhile. In other cases, as when the diplomatic language, based on concepts from personal relations, of friendship (philia) and kinship (syngeneia), all but vanishes, the disappearance is more apparent than real. It is not that these concepts ceased to be meaningful—these were still the terms which communities used to think about their relations with each other. Instead, this was a secondary phenomenon, merely a function of the marked disappearance of records of the other interstate institutions in relation to which this language was used.58 More revealing is the evolution of the Greek institution of asylia, involving delegations sent out to obtain recognition of the sacredness or inviolacy of a city’s shrine. These theōroi, often dispatched to distant and widely spaced cities and kings, were not, for the most part, primarily aimed at securing shrines from pillaging by pirates or in war.59 Rather they reflect the prestige value associated by these communities with obtaining extensive recognition, by other poleis, and royal and imperial powers, of the status of their shrine and perhaps also of an associated competition.60 As I argued in the

57 Jones (1940), 172; the Pompeian law was, however, a dead letter by Pliny’s day (was it ever enforced?): Pliny, ep. 10.114–15; Livy 45.29. See also Dio 38 and 41 with Jones (1978). 58 Jones (1999), 106–21; for examples from the Roman Imperial period, see Curty (1995), nos. 69–88. 59 Rigsby (1996), 13–25, esp. 22–5. See, however, the case of the Tenian asylia campaign, Chapter 1, pp. 75–6 with Appendix no. 15.2. 60 The Magnesians explicitly ask for their festival—and thus a victory there—to be recognized as of equal standing to the Pythian Games, resulting in the same awards of

262

Proxeny and Polis

previous chapter, asylia campaigns very clearly show how important a community’s social position was in relation to its peer poleis in motivating its engagement in interactions with them. The role of asylia, and the way in which it was sought, continued despite Roman domination in the early second century bc, at least initially, without significant change. Rome’s acknowledgement of the asylia of Stratonikeia’s shrines may have been of most value in c.80 and c.40 bc, but it was still worth obtaining and inscribing the acquiescence of at least seventy other cities in the first instance (a fragmentary list with seven surviving ethnics perhaps relates to the second).61 Before long, however, this ceased, and asylia came to be something that only Rome could meaningfully give or recognize. Under the influence of Rome, this nebulous symbol became a concrete privilege, the right of asylum from Roman law, and granting it also became a Roman monopoly.62 Asylia continued to be prestigious within the context of competition between cities. Indeed, like other titles which came to be granted by the Romans, its prestige probably increased precisely because it was more rigidly controlled (it seems likely, that after the claims to asylia were reviewed by the Senate in ad 22, very few further grants, if any, of this status were made).63 This development, however, reveals the fundamental change of the system—from one in which prestige originated in interactions between peers, to one in which it derived from and continued to be dependent on hierarchical relations with the imperial power, and, ultimately, the person of the emperor.

5.2.3. Interstate arbitration Interstate arbitration and foreign judges were two distinct but overlapping institutions aimed at enabling the resolution of disputes by

civic honours for citizen victors as these other festivals. By granting these honours to citizen victors, poleis signalled the degree to which they collectively set store by a citizen victory at a particular set of games and thus its importance. On stephanitic status, see Slater and Summa (2006). 61 Rigsby (1996), 418–27. 62 On asylum under the Roman Empire, see Gamauf (1999). 63 Rigsby (1996), 29, suggests that a formal prohibition on further grants of this status was in fact instituted in this year.

Disappearance of Proxeny

263

facilitating the provision of neutral, external judges for adjudication or, preferably, mediation. Both are extremely well attested, especially in the Hellenistic period, and have been extensively studied.64 Explicitly intended to circumvent recourse to violence at the different levels at which they operated, they represent two of the most important institutions which defined the fabric of the Hellenistic world. Interstate arbitration, by which two states seek mediation and/or arbitration of a dispute from a third state actor, appears to be culturally non-specific and the motivations of disputes certainly are (territory, transport routes, economic resources, and the control of shrines).65 In the Greek world, however, the practice seems to have been especially widespread with highly developed procedures from an early period: our earliest documentary source dates from the mid-fifth century bc and the historical references, both literary and epigraphic, have been collected in catalogues of sixty-one items down to 338 bc and 171 items for 337–90 bc.66 In our historians, moreover, we see mediation embedded, from an early stage, in moral language concerning the act of going to war. Peace treaties frequently established arbitration as the necessary means of dispute resolution and offers to go to arbitration (sometimes explicit challenges to accept arbitration) between would-be belligerents clearly represented an important preliminary to war, at the least an effort to displace responsibility for beginning hostilities, if not always a serious attempt to avert them.67 What is particularly striking in our evidence for the Greek world, however, is the eagerness of third parties to arbitrate and/or mediate the disputes of others. We find numerous unsolicited offers, which on at least one occasion had seriously adverse consequences for the 64 Interstate arbitration: Tod (1913) is foundational; see Piccirilli (1973) on the Archaic and Classical examples, down to 338 bc, continued by Magnetto (1997) to 200 bc and Ager (1996) to 90 bc respectively; for a study of twelve cases of arbitration between Greek cities involving Roman intervention, see Camia (2009); Magnetto (2008) provides a very full treatment of the arbitration between Samos and Priene. Foreign Judges: Robert, OMS 5, 136–54; Gauthier (1972); Crowther (1992); (1993); (1995); (1999); Cassayre (2010) 127–75. 65 Tod (1913), 169–72; Ager (1996), 7. See Roebuck (2001) for the comparative perspective. 66 Piccirilli (1973); Ager (1996). 67 Thuc. 1.145 (cf. 140.2); see Low (2007), 105–8. The arbitration provisions in treaties were not always particularly detailed, apparently assuming goodwill, but some were more so than others, see Ager (1996), 7–8.

264

Proxeny and Polis

would-be mediator, Rhodes,68 and even where arbitration was sought by disputing parties it involved prominent citizens from the arbitrating polis spending a possibly prolonged (and certainly unremunerated) period away from home. Furthermore, they were unlikely to be compensated for their efforts with civic honours which would have a prestige value in their own community, the usual reward of external benefactors. The losing city would be unlikely to reward judges who had rejected their claims, and the city which benefited most would hardly wish to cast doubt on the impartiality and thus authority of the verdict, leaving it vulnerable to future challenge, by granting honours which asserted the partiality of the judges. Such honours were, indeed, only granted when the attempt to arbitrate failed.69 The most likely explanation is that performing these functions was a source of prestige in itself, both for the poleis which sent representatives to arbitrate a dispute and the arbitrators themselves. For a Greek city being known as having rendered an impartial verdict in a dispute between other poleis—especially being known to have been asked to perform this function—was of value in itself, important for the construction of its identity in this social world of cities. The precise value, of course, depended on the status of the states in dispute. The more exalted they were, the greater the prestige to be gained from successfully mediating between them (one of the reasons, perhaps, why the Rhodians were particularly keen to intercede between the Romans and Antiochus III). For the individual arbitrator, successful participation in this activity which brought honour to his own city is likely to have been also served as a source of prestige for him within it. When states sought an arbitrator, conversely, the most important criterion was neutrality, but there also seems to have been a preference expressed for arbitrators who were powerful in their own right, whose position would lend most authority to the eventual judgement and leave it least open to challenge. In the Hellenistic period this often meant kings. After the domination of Rome, from c.200 bc, it was the Senate which was most frequently requested to mediate, with

68

Plb. 21.29.10–19. Would-be Rhodian arbitrators were honoured by only the Delphians, after the Amphissans failed to turn up, FD III 3 383 (= Ager 1996, no. 117, 179 bc); Magnesians were honoured by both parties after another failed attempt at arbitration, I.Magnesia 65a +75 and 65b + 76 (= Ager 1996, no. 127, before 167? bc). 69

Disappearance of Proxeny

265

all the most intractable cases reaching Rome.70 This is important for understanding how the interstate structure of the Mediterranean world changed in this period. The processes involved were not simply top-down, but instead reflected the initiative of the Greek cities, in this case in seeking to involve the Romans in their disputes.71 However, even if the Roman magistrates and Senate rarely intervened directly in cases themselves, at least in the early period, and instead delegated the task of arbitration to other communities, they nonetheless participated in having their authority, and their dominant position, recognized.72 Roman domination did not preclude other instances of interstate arbitration which were not remitted to Rome. From the epigraphic point of view, the second century bc constituted a high point in the number of arbitrations attested in general. Even provincialization did not eliminate the use of Greek poleis in settling disputes, as the arbitration of Erythrai and Sardis in a dispute between Priene and Miletos in the early first century bc shows.73 As in the case of other institutions in this study, however, the first century bc does appear to have been a turning point after which interstate arbitration rapidly ceases to be attested, and it is clear that the practice itself was transformed. Territorial disputes between cities continued under the Roman Empire and they continued to be resolved. On occasion a third community might even be called on to do the legwork.74 However, as the Roman presence became permanent, Roman governors or officers sent specially by the emperor increasingly took on the task of resolving these disputes themselves.75 There came a point when it was no longer a question of cities agreeing on a third community to settle their differences—probably when the use of violence to attempt to resolve them ceased to be a realistic possibility.76 Authority in these 70 The ‘repeaters’—Melitaia and Narthakion, Samos and Priene, and Sparta and both Messene and Megalopolis—were finally settled by Rome, Ager (1996), 28. 71 So Gruen (1984), 97–131. 72 Kallet-Marx (1995), 148–83, for this important nuance. 73 Ager (1996), no. 171 (= I.Priene 111, 120, 121). 74 e.g. I.Thess.Enipeus 13 (cf. Helly BE 1997 no. 298 and SEG 45 610, Thessaly, ad 15–35). 75 Ager (1996), 28 n. 73. 76 When Roman authority broke down, conversely, the temptation to attempt to settle disputes violently could prove hard to resist, see Burton (2000) and Brélaz (2008), 182–4.

266

Proxeny and Polis

disputes no longer resided with the communities themselves, to be delegated to a third community for resolution, it belonged to Rome (as did the decision that arbitration of a dispute was required). By the time we see examples of the resolution of territorial disputes in the High Imperial period, international arbitration—‘voluntary submission to the verdict of a neutral tribunal’—is no longer at issue, but dispute resolution.77 Dispute resolution was one of the functions performed by the imperial hierarchy, and it was probably performed quite well, but it represents another way in which the world of the Greek poleis was transformed.

5.2.4. Foreign judges In the Hellenistic period the institution of foreign judges similarly represented a solution to the problem of finding neutral means of settling disputes. In this instance, the issue was finding courts that were equally acceptable to individuals—whether because one litigant was a non-citizen, or because the city in question was too small to be able to empanel a jury that would not contain individuals bound by ties of obligation to one of the citizen litigants. For this problem foreign judges represented an ingenious and distinctly Greek solution. Brought in from another polis, often one at some remove from the particular city or cities, and thus unconnected to any of the litigants in question, a relatively small panel could work through even a large judicial backlog.78 The use of this institution, which developed towards the end of the fourth century bc, did present some ideological difficulties as, ideally, jurisdiction within a polis was a citizen monopoly.79 This is reflected in the emphasis which the surviving texts frequently place on reconciliation as the primary function of foreign judges, with actual adjudication represented as a regrettable recourse to be pursued only when reaching a negotiated settlement agreeable to both parties proved impossible.80 Despite 77 Tod (1913), 182. On the resolution of territorial disputes in the Roman provinces, see Burton (2000). 78 Such backlogs often reflected, and constituted, social crises, Robert, OMS, 5, 146–7. 79 So Aristotle, Pol. 1275a19–b22. 80 IG XII 6 95 (Samos, c.280 bc) is a particularly good example in which reconciliation is much more prominent than judgement. See Robert, OMS 5, 145.

Disappearance of Proxeny

267

these ideological problems, which perhaps diminished with the passage of time, because of its effectiveness this institution was very widely used indeed, as the substantial epigraphic record, which incorporates 250 decrees honouring foreign judges, shows.81 As in the case of proxenia, the institution of foreign judges could only fulfil this useful function because the cities involved were successful in co-opting outsiders to perform the roles set out for them. The inducement was the same, prestige, but, in contrast with proxenia, it functioned at two levels—both at the level of individual foreign judges and at the level of the polis which responded to the request of another city and chose and dispatched a panel of its citizens there. This dual level at which prestige operated is clearly expressed by the honorific decrees which the city receiving foreign judges passed in honour firstly of the city which dispatched them, and then of the judges themselves. The importance which the dispatching cities invested in these marks of honour from other cities is evident in the fact that they so frequently published these decrees honouring themselves. More than half of the decrees for foreign judges (138 of 250) were inscribed by the home polis of the judges rather than the polis which received them and passed the decree in their honour. By contrast, only a handful of the more than two and a half thousand decrees granting proxenia (excluding foreign judges decrees) were inscribed at the polis of the proxenos.82 Like the other institutions which we have considered by which poleis constructed honorific links with each other, the epigraphic record for foreign judges decrees attests to a now familiar trend— the quantity of material declines sharply in the first century bc (Figure 5.9). This reduction is even more marked in the case of foreign judges decrees because these reach their peak in the second century bc and are, in fact, far better attested in even the second half of that century than for the fourth and third centuries bc combined.

81 On the development of this institution, see Crowther (1992), 39–40; Cassayre (2010), 131–54 provides a helpful table, containing the vast majority of the texts, and 155–61 for an overview of the picture which this material paints for different cities and regions; Fournier (2010), 607–9 collects the later material and, at 537–9, provides a good treatment of it. I am very grateful to Charles Crowther who supplied the comprehensive catalogue of foreign judge inscriptions on which the analysis and graph in this section are based. 82 See Chapter 2, pp. 107–8 with n. 69.

268

Proxeny and Polis

70 Decrees inscribed by cities receiving judges 60 Decrees inscribed by home-cities of judges 50 40 30 20 10

15 0 10 1–

50

–1 00 51

1–

–0 49

0 –5 99

–1 00

–1 50

14 9

20

0

19 9

0 25

9– 24

0 30

9– 29

9– 34

39

9–

35

0

0

Fig. 5.9. Graph showing the chronological distribution of decrees honouring foreign judges, inscribed by the polis granting the honours (which received the judges) and the judges’ own city respectively.

It is important to reiterate again that this is direct evidence that the number of decrees inscribed declined rather than for a reduction in the quantity or frequency of the use of this institution. Given the importance for the functioning of this institution of prestige, however, as for the other institutions studied, the fact that inscription was curtailed, which was one of the primary means by which prestige was expressed, is likely to be important for understanding the institution and its evolution. We do, in fact, have more evidence for the use of foreign judges in the Roman period than for almost any other Hellenistic institution. Julien Fournier’s catalogue of material for foreign judges in this period lists some twenty-eight separate items, although this includes no honorific decrees and no fewer than fifteen of these relate to Mylasa and take the form of dedications made by panels of foreign judges there (ad 100–200, with a further funeral monument known at

Disappearance of Proxeny

269

Paros for an individual who also served as a judge there).83 The provision of neutral courts clearly remained an issue in the Roman period, perhaps even beyond the limited areas represented in this material (Lykia and Karia, Thessaly and the Peloponnese). It should not surprise us that, rather than bearing the administrative and financial burden of meeting that need themselves, some later Roman governors who had responsibility for jurisdiction in their provinces made use of this institution, or at least allowed the use of it, in the same way Cicero as governor in Kilikia had—surely for reasons of practicality as well as its avowed popularity with the Greeks.84 The different sorts of monument which attest foreign judges in the Roman period, however, reveal the way in which this institution was transformed in its use within the Roman system. There is no longer a question of emphasizing links between cities or the position of one polis in relation to other poleis. In all but the dedicatory texts, the purpose of these monuments is to honour individuals within their own cities who derive prestige from the performance of different duties in relation to this institution—not because of honorific links they thereby had with other communities, but as though for (and in parallel with) their performance of other, polis-internal liturgies.85 In these texts, honour and prestige are apparently derived by elites not from constituting links to other poleis, but from fulfilling functions and liturgies within their own polis.86 83 Fournier (2010), 607–9; Kantor (2011) adds SEG 54 1103 (another Mylasan dedication) and now, per ep., F.Xanthos VII 86 H l. 5 (Roman Imperial period) as well as, more tentatively, I.Perge 323 (first half of the third century ad?). 84 ‘graeci vero exsultant quod peregrinis iudicibus utuntur’, Cic. Att. VI. 1 l. 15. On the interpretation of this, see n. 25. 85 It is noticeable that the majority of honorands attested in this connection are honoured for their services in relation to foreign judges coming to the city—as dikastagōgoi, envoys sent to bring back judges (Sparta, first half of the second century ad: IG V I 39; 819; SEG 11 491; 493; 496) or epistatai, responsible for caring for the judges in the city (Lykia, first and second centuries ad: TAM II 420; 583; and 915) or, in one case, for paying them (Stratonikeia, second or third centuries ad?: I.Stratonikeia 229). Only in two career inscriptions is actual service as a foreign judge mentioned (SEG 11 491, alongside and after service as dikastagogos, and IG XII 5 305, the Parian funeral epigram, first century ad). 86 This is signalled clearly in the Peloponnesian texts: two specify the cities to which the individuals were sent to bring back jurors, but, indicating that this information was optional, and relatively unimportant, the other three simply specify ‘Asia’ (SEG 11 493; IG V 1 39; presumably the Roman province), and, probably, ‘Thessaly’ (IG V 1 819).

270

Proxeny and Polis 5.3. THE TRANSFORMATION OF THE GREEK WORLD

Together, the evidence for these institutions reveals a striking shift away from regularly monumentalizing links with other communities—to the extent that most of the interstate institutions which defined the Greek world of poleis vanish completely from the epigraphic record during the first century bc. Although the epigraphic traditions and trajectories of individual communities, or even institutions, may represent an insecure basis for interpretation, as being vulnerable to the vagaries of archaeological recovery, the overall picture here is clear and secure—not only because of the amount of material of different kinds which contributes to it, the wide range of different poleis from which it comes, but also because of the way in which the trends exhibited by these different elements coincide. From the mid-second century bc onwards, the ‘Low Hellenistic’ identified by Robert as a distinct period, the shift away from inscribing this institutional framework begins.87 This is indicative of an increasing reduction in emphasis placed on inter-polis connections which was probably reflected in a broader reduction in the frequency of inter-polis interactions of these kinds. After this period of transition our evidence for these institutions shows that, relatively early on in the principate, the decline of some of these institutional elements and the evolution of others was complete. The dissolution of a system of interstate institutions which had, for hundreds of years, shaped the way in which poleis constructed relations with each other, reflects a profound transformation, a change in orientation of this interstate system from anarchy to hierarchy, from a society of Greek poleis to a Roman world-empire. It is not enough, however, to simply point to the ‘unipolarity’ of this system to understand these changes.88 Instead, it is necessary to explore what this structural transformation meant for individual actors: how the position of poleis altered along with their capacity for interacting with each other and motivations for doing so and what their role was in the processes which brought this change about. One important question must be considered first. Did the Greeks cities’ ability to use these institutions change at any point as a result of 87

Robert, OMS 2, 841.

88

Wiemer (2010).

Disappearance of Proxeny

271

Roman intervention, either at the point at which these institutions effectively cease to be attested in the epigraphic record, or by the point at which they seem to disappear entirely during the first half of the first century ad? The wider epigraphic trend probably begins too early and is too gradual to represent the consequence of direct Roman intervention. However, by the time the institutional elements themselves either seem to have been forgotten or transformed, Roman control had long been permanent and invasive. It is important, therefore, to ask whether the Romans restricted the Greek cities’ ability to use these institutions by intervening directly in their internal political organization or by regulating their activities. Under the empire, Romans, beginning with Augustus, certainly showed a great willingness to intervene, even in the affairs of those cities to which they had granted the title of ‘free’.89 We also have examples in which Roman authorities intervened directly to restrict at least one element which had been central to the system of interstate institutions—the granting of citizenship—attesting to both direct intervention in a particular case (Augustus’ prohibition on the Athenians selling citizenship) and the promulgation of a general restrictive regulation (Pompey’s law prohibiting the cities in Bithynia from granting citizenship to each other’s citizens).90 In addition to this sort of specific measure, restricting, in these instances, the range of institutions which cities could use to construct relations with each other, it is worth considering a more general development thought to have taken place under Roman hegemony: the domination of authority within poleis by the boulē. The Romans were predisposed to favour more oligarchic constitutional arrangements, which entrusted the greater part of political power to wealthier elements of the state. As a result, it is argued, they therefore encouraged (or enforced) the establishment within poleis of permanent boulai which, on the model of the Roman Senate, had their basis in a property qualification, and increasingly came to dominate political decision making within the polis at the expense of popular assemblies.91 These developments have been represented as bringing about ‘the end of democracy’, 89 Bowersock (1965), 88, on Augustus’ interference in Athens, Sparta, Thessaly, Kyzikos, Tyre, and Sidon. 90 Dio 54.7.1; Pliny ep. 10.114; but see IosPE I2 4 with SEG 47 1196 (both early third century ad). 91 So de Ste. Croix (1981), 518–37, for the material; Jones (1940), 170–2 and 180–91; Magie (1950), esp. 114–15.

272

Proxeny and Polis

introducing an age of political exclusion and apathy.92 They have also been held responsible for the reduction in inscribed decrees as well as a decrease in inter-polis diplomacy as a result of a ‘loss of communal consciousness’.93 However, although the general trends underlying this thesis can be plausibly identified in our evidence, we should be careful about extrapolating too far and assuming that the Romans had an active interest in suppressing this sort of inter-polis activity. We do see more evidence for the permanent appointment of bouleutai in Greek cities from the first century bc onwards, but in some cities annual or even bi-annual appointment remained the norm.94 Roman political preferences might be written into local constitutions as a punitive measure, but we have no reason to think they were systematically imposed.95 The Greek cities were too useful in reducing the administrative burden of empire for the Romans to want to attempt to impose constitutions which the Greeks would not accept, or which might impair the institutional efficiency on which that usefulness depended.96 Moreover, many of the processes involved in the elite capture of the boulē, although intensified by the structural dynamics of Roman domination and Roman ideological preferences, had their genesis in the earlier Hellenistic polis.97 In recent scholarship it has

92 e.g. Lintott (1993), 147, ‘By the end of the Roman republic it must be doubtful if a city assembly remained which was still democratic by classical standards’. But see Rhodes with Lewis (1997), 531–6, for the flexibility of the term democracy, and 548–9, for its continuation. 93 Jones (1999), 112–13. 94 For nuanced accounts of the processes involved in the Low Hellenistic period, see Hamon (2005) and Heller (2009). Karystos retained a boulē selected by lot (IG XII 9 11, Hadrianic Period); two colleges of bouleutai per year remain the norm at Rhodes (e.g. AD (1967) no. 18 A, mid-first century ad), but see Heller (2009), 353–4, arguing that these institutional features can be reconciled with elite control of the boule in these poleis as well. 95 Rhodes with Lewis (1997), 548. See also S. Dmitriev’s wide-ranging survey of the institutions of Greek city administration more broadly, which stresses (albeit perhaps too strongly) the continuities between the Hellenistic and Roman periods and the relative lack of Roman interference (Dmitriev 2005, esp. 306–12). 96 Bowersock (1965), 87; even according to more pessimistic accounts, citizenship in a polis continued to involve considerably more participation than in a Roman colonia or municipium: Lintott (1993), 147; cf. Rhodes with Lewis (1997), 549. For a judicious account of popular participation in civic life in the west, see Jacques (1984), 379–425. 97 See, inter alios, Veyne (1990); Gauthier (1985); Quass (1993); and Hamon (2005).

Disappearance of Proxeny

273

also become increasingly apparent that the domination of the polis by the boulē has been overstated.98 Although rumoured in relation to the poleis allied with Antony after the battle of Actium, we have little evidence for the abolition of public assemblies.99 As far as we can see assemblies continued to represent the norm and, rather than simply rubber-stamping decisions of the boulē, it is clear in some of our limited sources that they remained a context in which issues of importance to those involved were debated and decided.100 In fact, even the attested instances of Roman intervention in elements of the Greek system of interstate institutions—in particular the granting of citizenship—illuminate what continued, and demonstrate the limits of active Roman efforts to regulate in this area. The intervention of Augustus at Athens, couched as a prohibition specifically on the sale of citizenship rather than on granting in general, shows not only the expectation that citizenship, as an honour, had continued hitherto and would continue to be granted, it also clearly reveals that it continued to be perceived as valuable. The Pompeian law relating to the Bithynian cities is even more striking. Again, grants of citizenship in general are still permitted (and expected), it is simply grants between Bithynian cities which are prohibited.101 Moreover, the fact that, by the time Pliny mentions this legislation (and probably long before), it had become a dead letter shows the limitations on the ability of successive governors, and probably their relative lack of desire, to enforce this sort of regulation in the face of Greek resistance. In short, we have no evidence that any deliberate action taken by the Romans was separately responsible for the dissolution of this system of interstate institutions after its prolonged decline in the 98 See, in general, Fernoux (2011); de Ste. Croix (1981), 518–37, provides a wealth of material but draws the wrong conclusions; Kuhn (2008), 115–28, explores mechanisms of communal control over the elite and, at 145–6, representations of the demos and boule as equal partners, capable of conflict; Zuiderhoek (2008) reconciles the evidence for increasingly oligarchic boulai with an active assembly. On the picture of active Greek political life painted by the surviving speeches of Dio, especially with regards to the assembly, see Ma (2000b); Vujčić (2009); and especially Kuhn (2008). See also, already, Jones (1971), 111 and (1978), 91–2. 99 Dio 51.2.1 with Bowersock (1965), 85–6. 100 Rhodes with Lewis (1997), 549; Vujčić (2009), esp. 168; Fernoux (2011), 251–322, gives a comprehensive discussion of the different sorts of business they transacted. 101 Pliny, ep. 10.114.

274

Proxeny and Polis

epigraphic record. Consideration of a partial parallel for the situation in which the Greek cities found themselves, dominated by Rome, may help illuminate why the Romans were not concerned to intervene heavily in this area. In her book, Hellenistic Democracies, Susanne Carlsson suggests a comparison with poleis which lost their political independence and came to be absorbed by other poleis. The example given is Kalymna, which ceased to engage in certain sorts of interstate activity after being integrated, as a deme, within the polis of Kos.102 As Carlsson acknowledges, there are significant differences. Above all, poleis did not apparently lose their polis status as a result of their subordination under the Roman Empire. The comparison is nonetheless interesting in that it forces us to explore the contrast between the roles which these sorts of inter-polis interactions play in each case. The crucial difference is that, in the Classical and Hellenistic periods, dominant poleis had a strong stake in restricting their dependent communities from using this institutional language of inter-polis interaction because it implied a status which conflicted with the dominant polis’ claims to own and control. Conversely, within a Roman Empire in which the Greek poleis were by that time firmly established as subordinate, the claims which most of these institutions made, above all to polis status, were not capable of conflicting with Roman claims to rule (with the possible exception of symmachia and international arbitration). It is in fact possible that it was because this institutional language lost this force and potency that it ceased to be used by poleis, now that even polis status ultimately depended more on guarantees from Rome than interactions with other poleis. It is striking that when we do see evidence for the extension of Roman influence over the different facets of political life in the Greek poleis, the initiative often seems to arise from Greek individuals and cities themselves, underlining their agency in this process. The logic which led Greek cities to seek Roman arbitration, and Greek individuals to seek Roman adjudication—the desire for the most authoritative decisions—also led certain poleis on occasion to refer their communal decrees to their Roman governors for confirmation (epikyrōsis). That this confirmation was not, for the most part, a Roman-imposed requirement is made clear by the fact that the inscribers of most of the surviving decrees saw no need to include

102

Carlsson (2010), 284.

Disappearance of Proxeny

275

any mention of Roman approval.103 On one occasion, the governor’s response to a request for his approval explicitly makes this clear as well. When asked by the city of Sidyma to ratify their decree to establish a gerousia, Pomponius Bassos replied that it was his praise that was called for rather than his ratification, since their own decision was authoritative.104 It is an indication of the degree of autonomy that this city still possessed that it was told that it had the authority to make this sort of major constitutional change. It is revealing, however, that the authorities of Sidyma thought it desirable to get the governor to say as much and took the decision to have this confirmation of their decision (which may have been internally contentious), and recognition of their authority, inscribed on stone. Plutarch knew well and aptly described the dynamics of the phenomenon involved, and its effect on the political institutions of the Greek cities: ØF Æ  Ø ŒÆd Ææå Æ E ŒæÆFØ P ØŁB c Ææ Æ  E c æ ŒÆ Ø F , Åb F Œºı    ı æı ºº Ø ŒÆd e æ åź , u æ  ØØ, ŒÆd ØŒæa ŒÆd 

Çø çæ   Kd f ª  Æ K Ø ÇıØ c ıº

Æ , Aºº  ‹ºø c ºØ

Æ I ÆØæFØ, ŒÆƺBªÆ ŒÆd  æØ A ŒÆd  ø ¼Œıæ ØF  . u æ ªaæ ƒ åøæd NÆæF   Ø E  º ŁÆØ ı ŁØŁ   P ‹  çØ  øØ åæH ÆØ fiH ªØÆ Ø , oø ƒ Æ d  ªÆØ ŒÆd ı æ ø fi ŒÆd å æØØ ŒÆd ØØŒ Ø æ ª   ª  ØŒc Œæ Ø I ƪŒ ÇıØ ÆıH Aºº j º ÆØ   Æ r ÆØ f ªı ı. ÆN Æ b ı  ºØÆ º   Æ ŒÆd çغ ØŒ Æ H æø · j ªaæ K x  º ıØ f Kº  Æ KŒØ Ç ÆØ ç ª Ø c  ºØ j  æd z ØÆçæ ÆØ æe Iºººı PŒ IØF   K E º ÆØ å Ø ºÆ K ª ÆØ f Œæ

 Æ· KŒ ı b ŒÆd ıºc ŒÆd B ŒÆd ØŒÆæØÆ ŒÆd Iæåc AÆ c Kı Æ I ººıØ. However, the statesman, while making his native state readily obedient to its sovereigns, must not further humble it; nor, when the leg has been fettered, go on and subject the neck to the yoke, as some do who, by constantly referring everything, great or small, to the sovereigns, bring the reproach of slavery upon their country, or rather wholly destroy its constitutional government, making it dazed, timid, and powerless in everything. For just as those who have become accustomed neither to dine nor to bathe except by the physician’s orders do not even enjoy that degree of health which nature grants them, so those who invite the sovereign’s decision on every decree, meeting of a council, granting of an honour, or administrative measure, force their sovereign to be their 103 104

Rhodes with Lewis (1997), 547; on these in general, see Oliver (1954). TAM II 175 (reign of Commodus).

276

Proxeny and Polis

master more than he desires. And the cause of this is chiefly the greed and contentiousness of the foremost citizens; for either in those cases, in which they are injuring their inferiors, they force them into exile from the polis, or, in matters concerning which they differ among themselves, since they are unwilling to occupy an inferior position among their fellow-citizens, they call in those who are mightier; and, as a result, the council (boulē), popular assembly (dēmos), and the entire local government lose their authority.105

As we saw in the case of asylia campaigns, the function performed by titles and status-designations was transformed in a similar way. Rome’s recognition of particular civic titles and statuses came to be sought first as being the most authoritative and prestigious, but later, as Rome’s position became more absolute, theirs became the only declaration that was required. It was no longer the case that cities were given Roman acknowledgement of the terms they used to describe themselves—the titles granted by the Roman authorities were ones which cities could not otherwise lay claim to. Alongside the evolution of asylia into a specifically defined right of asylum, under the Roman principate a new set of honorific titles were developed. Although some, like mētropolis, were derived from terms earlier used by cities to construct relations between each other, these titles, along with others created under the Roman Empire, came to make sense only within the Roman provincial system. Cities sought to obtain the designation of mētropolis or prōtē polis of the province, or to be neōkoros, the province’s centre of imperial cult.106 Cities also frequently emphasized the authority lying behind these titles and their success in obtaining them by explicitly stating that a particular title was conferred ‘in accordance with the decrees of the Senate’ or ‘in accordance with the judgements of emperors’.107 Moreover, when these titles were granted it was not just the right to use them with reference to the imperial hierarchy that was at issue—cities were now obliged to use these honorific titles in reference to each other.108 In 105

Plutarch Praec. Ger. Reip. 814e–815a; Loeb trans., adapted. Robert OMS 6, 228; on ‘mētropolis’ in civic titulature, Bowersock (1985); on prōteia, Robert (1969), 287–8; on neokoreia, see Burrell (2004); for a detailed account of the significance of these titles, see Heller (2006), 283–338. 107 Jones (1999), 118; on the cult of homonoia, see Thériault (1996). 108 We possess a letter from Emperor Antoninus Pius to Ephesos explaining that the Smyrnians’ omission of these titles must have been a mistake—which it surely was not, Syll.3 849 (ad 140–144). 106

Disappearance of Proxeny

277

Dio Chrysostom’s speech, addressed to Nikomedia on concord with Nikaia, we see how bitterly such titulature could be disputed, but this episode also makes clear that these titles depended only on relations with the Roman emperor.109 The Nikomedians asserted the sole right to be inscribed as prōtoi, a title which they received from Domitian in addition to their earlier title of mētropolis.110 It was only, however, when the Nikaians backed Septimius Severus’ unsuccessful competitor for the imperial power that the Nikomedians got their wish and the title prōtoi was erased from the titulature of their rival city, Nikaia, and, literally, from its civic monuments.111

5.3.1. The end of peer-polity interaction? When we consider both the epigraphic disappearance and subsequent dissolution of the Greek system of interstate institutions, we are dealing with a Greek response—first to the domination of Rome, and then to the establishment of the Roman emperor as an absolute authority. Various elements of these institutions had performed valuable functions, but as a whole they relied on the fact that prestige and status, both for individuals within the polis and poleis amongst themselves, was generated in this interstate system by the creation of honorific connections and relationships between different communities. It was a pluralistic system in which authority was dispersed, albeit never evenly, between the state-actors within it, and which had persisted throughout the Hellenistic period, despite the appearance of much more powerful kings, because of the absence of a single dominant power. Within this system, it was worthwhile, important even, for a polis to seek recognition of the asylia of its shrine from distant and even minor poleis, or incorporate apparently inconsequential communities within its network of theōrodokoi or proxenoi. Similarly, for poleis and individuals within them, it made sense to perform both concrete and symbolic services for other communities 109 Other examples of such disputes: Aristid. Or. 23; on inter-civic disputes in general, see Heller (2006); Kuhn (2008), 138–44. 110 Dio 38.28 with Jones (1978), 84. Cf. Beroia, which petitioned for the right to be the only neocorate polis in its province (bæ F  Å ÆPc å Ø c øŒæ Æ ), _ I.Beroia 117 ll. 5–6 (end of the first century ad). 111 Jones (1978), 87; Robert, OMS 6, 14–15; texts: Schneider (1938), 45 no. 11 and 46 nos. 14–15.

278

Proxeny and Polis

and enter into interactions involving the expenditure of time and resources (e.g. proxenia or theōria) because standing in polis society was to a significant extent dependent on these links and acts. Rome’s rise to a position of domination first reduced the importance for poleis of advertising or commemorating these social networks in that it constituted a higher and more final authority. Eventually, by monopolizing authority within the interstate system in ways which had been beyond the Hellenistic kings, and thereby monopolizing also symbols of prestige, it undermined the euergetical basis of these institutions, breaking the bond between honour and function on which they had relied. In part the process which we see in our evidence was one of displacement. Rome took over, and was asked to assume, some of the functions which these institutions had fulfilled. Rome constituted a new, more authoritative source of prestige and political identity, granting titles and other marks of status and enforcing their use between communities. Some of the institutional effort which cities had previously placed into establishing and maintaining links with peer-polities was now redirected to Rome. This is clearly reflected in the very frequent, sometimes more than annual, embassies sent by individual poleis to the emperor there as well as the development of the Roman imperial cult.112 However, when we consider the frequency and intensity of institutional interpolis interaction in the Hellenistic period, it is difficult to deny that this process also entailed some sort of absolute decline. Did the Greek Mediterranean cease to be a peer-polity world, becoming one in which inter-polis interaction no longer mattered? Given the various different ways which research of the last twenty years has clearly demonstrated that particular forms of civic culture continued to flourish—festivals, koina, and new forms of ritual interaction, as at Claros—this might seem somewhat difficult to maintain.113 112

For a catalogue of the evidence, see Souris (1982); see also Millar (1977), 410–34. Some frequent ambassadors to emperors: OGIS 494 (first or second century ad); IGR III 628 (Xanthos, mid-second century ad); 982 (Kition). The importance of these embassies in protecting a polis’ territory and status are underlined in a decree from Maroneia (SEG 53 659, ad 41–54) which makes extraordinary provisions in the form of an ‘eternal decree’ for the dispatch of an embassy at very short notice in case of need. On the Roman imperial cult (and the agency of Greek communities in its development), see Price (1984). 113 Festivals: on the more high profile festivals, see now SEG 56 1359, the new letters of Hadrian to the Dionysiac artists inscribed at Alexandria Troas; for the result of a profusion of festivals at the local level, see IAph2007 15.330 (180s ad). Ma (2003), 37, asserts the continuation of peer-polity interaction into the Imperial period.

Disappearance of Proxeny

279

Above all, the works of Dio Chrysostom arguing in favour of the establishment of homonoia between cities, and the disputes over relative status and precedence which they presuppose, clearly show the continued importance of relations and rivalries between poleis, and the various homonoia coinages allow us to see these processes taking place more broadly.114 The competitive context of the Greek polis remained other poleis, the peers against which each continued to compare and assert itself. The emphasis, however, shifted fundamentally. The symbols from which cities derived status were no longer generated by or obtained from links within the network of poleis. Instead, they were obtained from outside, and were dependent on continuing relations with the Roman authorities, and ultimately the Roman emperor, who enforced their use between cities. This dependence is exemplified by the well-known delegations sent by cities to each new emperor, on his accession, to obtain assurance of the continuation of their status and privileges.115 The titles and privileges themselves, moreover, made sense only within an imperial, Roman conception of the world, where each city’s ‘region of primary interaction’, its primary context for competitive display, which cities had earlier defined for themselves as a result of their own interactions, were now administratively defined provinces. This was no longer a world constructed by the honorific inter-relations of poleis. The age of proxeny had passed, and, though poleis continued to inhabit this world, it was defined by Rome and empire. A letter sent by the emperor Hadrian to Naryx in ad 137/8 shows clearly what had stayed the same, and what had changed: ªÆŁB fi ❦ åfiÅ· `PŒæ øæ ˚ÆEÆæ Ł F æÆœÆ F ❦ —ÆæŁØŒF ıƒ , Ł F ˝æÆ ıƒø , æÆœÆ e AæØÆ e  Æ , IæåØ æ f ªØ, ÅÆæåØŒB Kı Æ · e Œ´ · ÆPŒæ øæ e · ´·, oÆ e ª´, Æcæ Ææ  ❦vacat ˝ÆæıŒ

Ø ❦ vacat åÆ æ Ø · vacat PŒ r ÆØ IçØÇÅ Ø Ø a  ºØ A å Ø ŒÆd  º ø  ŒÆØÆ, ›  ŒÆd

N e ŒØ e H çØŒı ø ı  º E ŒÆd N e ŒØ e H BØøH ŒÆd BØø æåÅ Ææå  ŒÆd —Æ ººÅ Æ Æƒæ EŁ

114

Jones (1999), 112.

115

Millar (1977), 410–18.

280

Proxeny and Polis

ŒÆd Ł ÅŒ º   , ŒÆd KØ E ŒÆd ıºc ŒÆd ¼æå   ŒÆd ƒ æ E ŒÆd çıºÆd  EººÅ ØŒÆd ŒÆd Ø  Oı  ø , ŒÆd  º E ç æ  a H åÆØH · K ŁÅÆ b H ŒÆd ØÅÆ Ø  H K ºªØø ø ŒÆd  øÆEØ ŒÆd  ‚ººÅ  ‰ ˝ÆæıŒ

ø O  ÇıØ b ŒÆd H æø Ø a KŒ B  º ø B  æÆ ›æÅŁ Æ· ø  ŒÆ,

N ŒÆd Ææƺº ØÆØ E e ªæ ç Ø æe ___ f ÆPŒæ æÆ ŒÆd Ææ[– – –_ c.12 – – –] __  Ø ºÆ [ Ø – – – – – – – – – – – – – – – –] _ __ With good fortune. Emperor Caesar, son of god Traianus Parthicus, grandson of god Nerva, Traianus Hadrianus Augustus, pontifex maximus, with tribunician power for the twenty-second time, imperator for the second time, consul for the third time, father of his country, to the Narykians, greetings. I do not think that anyone will dispute that you possess a polis and the rights of a polis, seeing that you contribute to the Amphictyonic League and, to the Boiotian koinon, you provide a boiotarch, you choose a panhellene, you send a theokolos, and you have a boulē, magistrates, priests, Greek tribes, and the laws of the Opuntians, and you pay tribute along with the Achaeans. You have also been mentioned by certain of the most famous poets, both Roman and Greek, as ‘Narykians’, and they also name certain of the heroes as having started from your polis. For these reasons, even if you have omitted writing to the emperors . . . 116

Hadrian reassures the Narykeians that no-one ought to be able to doubt their status as a polis given all of the ways in which they participate, as a polis, alongside other communities with polis status: contributing towards the Amphictyony, sending a boiotarch, choosing a panhellene, and sending their own theokolos. They also look like a polis, in terms of their political institutions, and are recognized as such administratively by the Romans in the way they are obliged to pay tribute. It is significant, however, that in this exhaustive list of ‘proofs’ of polis status (which the Narykeians themselves probably rehearsed to Hadrian) the emphasis is on the involvement of Naryx in various supra-polis organizations and there are none of the interstate institutions which I have discussed above, through which cities had previously in the Classical and Hellenistic periods performed their 116

IG IX 12 5 2018; trans. Jones (2006), 152, adapted.

Disappearance of Proxeny

281

identity as poleis. Even the theokolos mentioned, which seems to resemble the theōros of previous inter-polis relations, is sent not to another polis or Greek shrine, but the Panhellenion established under Hadrian himself. The most significant point, however, is that none of these ‘proofs’ or performances of polis status was sufficient for the Narykeians (or perhaps for others who may have doubted their status).117 What mattered, what they sought, and got, and then commemorated in bronze, was their emperor’s explicit recognition of their status and standing.

5.4. CONCLUSION: THE END OF THE AGE OF PROXENY As an institution proxenia derived its meaning from practical services that defined the proxenos as an intermediary, a link with other poleis. When proxenia did decline, however, under the Roman Empire, it was not because these practical services had ceased to be useful. Rather, it was because the interstate system in which it operated had shifted as a whole so that Rome rather than networks between poleis became the focus for authority, prestige, and identity. Proxenia was not the only institution to be affected by this change in how poleis thought about each other, but in our evidence it is the one which appears to undergo the most profound transformation. What had been a universal institution among the Greek poleis, founded on a shared and uniform proxenos-paradigm, all but disappeared. The later evidence suggests that, where it survived, proxenia lost its specific content and became fragmented, developing parochial functions within locally defined honorific systems. Proxenia was more deeply affected than other elements because it stood for polis networking in its broadest sense.

117 For the possibility that the Narykeians also had more tangible benefactions from the emperor in view, see Jones (2006), 161–2.

Conclusion This book is intended to be both a study of the ancient Greek institution of proxenia and a contribution to the study of institutional history and theory more broadly. Having traced the decline and disappearance of proxenia in the last chapter, here, at the end of the book, it may be helpful to revisit some of its central arguments and themes. In this book, I tried to develop a new model for understanding how proxenia was invested with meaning, arguing that proxenia was defined by a coherent set of expectations that Greeks had of their proxenoi—the proxenos-paradigm—in which the honorific value of ‘proxenos’ as a title was directly dependent on its identity as a practical intermediary role. This ideal conception of proxenia, which appears to have been both universal and stable over five hundred years, operated effectively despite the changing political realities of the Greek cities. Building on this understanding of proxenia, as an institution which connected cities, this study explored the scale of the phenomenon and its potential as a source of information for regional and inter-regional interaction. The patterns of proxeny networking which this study has revealed emphasize the symbolic importance of proxenia in the context of a wider system of inter-polis institutions. Proxenia played a particularly important role in enabling communities to maintain relations with each other, to assert, and have their status recognized, as poleis, within an interstate society which consisted of other poleis. Ultimately, proxenia disappeared not because the intermediary services which formed part of the proxenosparadigm had lost their utility, but because, with the disappearance of the anarchic conditions which had previously characterized the Greek world, poleis no longer thought about or valued their relations with each other in the same way.

Conclusion

283

At the same time as offering a new account of proxenia and its functions, this study was intended to bring out a number of themes that are important to illustrating the role of institutions more broadly in shaping human behaviour at a variety of levels. The first of these themes is the role that institutions play in defining the interests which motivate both individuals and groups to act. For instance, I argue in Chapter 2 that proxenia was desirable and sought by individuals as an object in its own right because it functioned as a status symbol, signalling membership of the elite. In Chapter 3, we saw a similar phenomenon in relation to poleis, as collective actors, in their need to possess substantial networks of proxenoi to communicate their links with, and status in relation to, other polis communities. Proxenia, like other institutions, worked by giving specific shape to the social desires and drives which actors, individually and communally, possessed. Institutions also shape the means by which interests are pursued. The proxenos-paradigm defined a highly specific role, a set of responses, both verbal and practical, which individuals who pursued proxenia were expected to perform and, similarly, which proxenoi, if they wanted to advertise their possession of this status, conformed to. Moreover, we possess such extensive epigraphic evidence of the proxenos-paradigm because reproducing it, and being seen to reproduce it, was important so that communities could signal that they shared the norms of inter-polis society. However, while this book does bring out the various ways in which institutions could direct actors’ outlook and activity, it also makes clear that it would be wrong to assign a deterministic significance to them. Proxenia functioned effectively as an institution because, as well as a generalized attitude of friendship towards the granting polis, it involved a highly specific and narrow range of practical services unlikely to conflict with individuals’ other interests. Where it could be represented as conflicting with other institutionalized roles, particularly that of citizen, these often took precedence over proxenia. The institutional role of the citizen, in particular, permitted far more latitude in interpretation, for example in deciding precisely what the interests of the community were, so that even here individual agency remained. Proxenia was also capable of miscarrying, as the example of Alcibiades, the would-be proxenos of the Spartans, discussed in Chapter 2, made clear. The case studies explored in Chapter 4 demonstrated that proxenia could be used in idiosyncratic ways by poleis to construct specific sorts of relations with particular individuals and

284

Proxeny and Polis

communities (e.g. with dominant poleis by honouring their officers). Proxenia, as an institution, constrained actions, directing them down specific lines in particular situations, but it also created possibilities which both individuals and cities exploited. The approaches taken to proxenia in this book could usefully be applied to understanding the evolution, adoption, and continuation of others sorts of institution as well—including apparently internal institutions, such as citizenship or particular sorts of magistracy. Recent interpretations of institutional innovation have tended to concentrate on the pursuit of performance, the lowering of transaction costs, and efficiency gains.1 In contrast, the normative model of institutional isomorphism explored in this book would encourage explanations of the adoption and use of particular sorts of civic forms, as well as the persistence of local idiosyncratic institutions, in social terms, as communal performances of legitimacy and identity, particularly within specific local or regional contexts. This book, however, is also about Greek epigraphy and the interrogation of epigraphic records. In particular, it represents a prolonged engagement with the issue of how to interpret substantial quantities of epigraphic material produced by a large number of different communities while allowing for the processes of selection, which shape and distort all samples of epigraphic material. This has involved developing a number of complementary methodologies for interpreting epigraphic traditions: a detailed analysis of how the formulaic language of decrees works (Chapter 1); an examination of surviving material which provides a representative quantitative overview of the phenomenon (Chapter 3); and a method of representing, using quantitative probability distribution, the chronological development of specific epigraphic traditions and interpreting the significance of the trends which they reveal (Chapter 5). Throughout this study, the difficulty of talking about the polis has been repeatedly underlined—for example in the multiplicity of perspectives which the epigraphic record yields, and, indeed, in the very nature of proxenia as an inter-polis institution. The difficulty identified here, however, is not the one which Kostas Vlassopoulos recently argued has a ‘pernicious effect on the study of ancient history’, namely the failure to recognize that small to middling and hegemonic

1

Ober (2010), 276–8; Bresson (2007–8).

Conclusion

285

poleis are two completely distinct types of polity.2 Rather, it is the artificiality of considering a single polis in isolation. This emerges clearly in Polybius’ attempt to do just that, when he describes, in order to explain the basic dynamics of constitutional change, the lifecycle of a sort of petri-dish polis apart, arising from a conglomeration of primitive humans, and, after running the full gamut of constitutional forms, dissolving once more into savage individualists.3 In this book, I have argued that ‘the polis’ was a system phenomenon—or, better, that we should think in terms of a self-supporting network of poleis. To be a polis was to be one among other poleis. To be a polis did not require complete external independence or even necessarily full internal autonomy, but it did require the ability, which, as I showed in Chapter 4, not all urbanized political communities possessed, to enter into relations with other poleis. In short, to be a polis, at least before the Roman emperors, was to engage in inter-polis interaction. Throughout their history, impotence and vulnerability to oppression and exploitation was the base state of most Greek urbanized communities.4 However, as poleis, they constituted the characteristic state type in probably the most populous state system in recorded history. Even under the Persian Empire and the later Hellenistic kingdoms, poleis defined the interstate fabric of the Mediterranean world as a society of poleis. The system of inter-polis concepts and institutions, including proxenia, which they evolved as a result of continual interaction represented the dominant discourse which would-be interstate actors of other kinds (ethnē, federations, nonGreek cities, and even dynasts) adopted to assert their legitimacy, and into which even definitely non-polis actors (Hellenistic kings and their courts) were drawn. This illustrates the importance of institutions in shaping the way in which individuals and communities think, as well as their aims and their methods of pursuing them. Proxenia was both central to, and embodied, a highly developed, self-perpetuating system of institutions which only the profoundest of interstate transformations unseated.

2 4

Vlassopoulos (2007), 92–6. Gauthier (1984), 85–7.

3

Plb. 6.5–9.

APPENDIX

Extant Lists of Proxenoi Introduction This appendix collects and presents the different lists of proxenoi that form the basis of Chapter 3. Its purpose is to make the data they contain relating to the proxeny networks of particular poleis more easily accessible and comparable. Its interest is thus primarily in identifying the different poleis and regions with which particular inscriptions attest relations—in other words the ethnics of proxenoi that they preserve—rather than the prosopography of the individuals in question.1 This appendix is intended to gather all the inscriptions which are certain or likely to be lists of proxenoi, whether chronological lists or catalogues of proxenoi.2 In relation to catalogues of proxenoi, a fairly narrow epigraphic genre, this is relatively straightforward. The term chronological list, by contrast, necessarily encompasses a wider range of epigraphic phenomena because, in function, these inscriptions were much closer to individually inscribed decrees, and it is thus harder to draw strict distinctions. Ideally this term refers to lists collecting all the discrete proxeny grants made by a community within a certain, fixed period of time, in which the individual entries are not logically complete decrees in themselves. In practice, however, it is unsatisfactory to exclude completely lists collecting proxenoi appointed at a particular moment in time (such as no. 6.1–4, the Epidaurian lists, and no. 13.1–14, the Samothracian lists), and it is difficult to draw the line between chronological lists of proxenoi and sequences of abbreviated decrees granting proxenia inscribed on the same stone. Since these different types of inscription need to be interpreted on their own terms, this does not create great difficulties. I have, however, excluded a number of texts which seem to depart too far from the idea of a chronological list: a sequence of highly abbreviated decrees from Gortyn, inscribed individually at different times on the same surface;3 a set of full decrees from Akraiphia which share an archon-

1 For the limitations of the systematic prosopographical method in relation to the study of proxenia, see Introduction, pp. 15–16; on the identification of particular individuals, see the earlier editions and bibliography cited for each text. 2 On these different sorts of monument, see Chapter 3, Section 3.1. 3 e.g. IC IV 206 a–m.

Appendix

287

date heading;4 and a number of texts from Miletos with headings which seem to suggest that that they are chronological lists of proxenoi, but which seem in practice to represent grants to individuals or groups from the same polis.5 Within the Appendix proxeny lists are grouped into sections according to the different communities from which they come, which are then organized alphabetically (the proxeny lists of uncertain origin, nos. 17 and 18, are placed at the end). The entries themselves present a mixture of full texts and tables summarizing the information contained in particular lists. The choice of which texts to give in full has been determined by considerations of practicality and utility. For proxeny catalogues, where the structure of the text is often complex and important for the interpretation of the evidence contained, full texts have generally been given. By contrast, chronological lists which are either numerous (like those of Samothrace, no. 14.1–13) or very long (like that of Delphi, no. 5) are only presented in summary. Each entry consists of a discussion of the inscription followed by a text, in which the ethnics of the proxenoi have been underlined, or a table, which lists these ethnics and is presented at the end of the relevant section. Help in interpreting these different ethnics and cross-referencing the different lists of proxenoi in which they occur is given in the form of a consolidated annotated index of ethnics at the very end of the Appendix. The information for the proxeny network of each polis which can be gleaned from these texts is presented in the form of a map either at the end of the relevant entry (Maps 7–16) or in Chapter 3 (Maps 1–6, for which cross references are given in the relevant entries, nos. 3, 5, 7, 8, and 10). (Where the ethnics of fewer than five proxenoi are securely preserved, no map has been given.) Photographs of some these inscriptions (or of squeezes taken from them) are presented separately, in some cases for the first time, after entry no. 18, before the index of ethnics (Figures 10–23). These lists, and especially the ethnics which they contain, present us with a series of particular, polis perspectives on interactions and networks within the Greek world, which are the more valuable for being preserved only in a fragmentary state. Incomplete and partial in various ways, as they are, collectively these fragmentary inscriptions from eighteen different political communities name nearly 1400 proxenoi and preserve 240 different polis ethnics (and sixteen nonpolis designations; the poleis attested are presented together, on Map 17). Together they provide a good sense of the scale—and ubiquity—of proxeny networks. BCH 22 (1899), 90–4, no. III (Akraiphia). The first of these decrees is not a proxeny decree. 5 e.g. Milet I 3 96, 99, 115, 1067, and 1069. I.Tralleis 33, contra Marek (1984), 110, with its high concentrations of Thessalians and Macedonians without polis specification, looks more like a list of mercenaries than of proxenoi; see also Jones (1992) eliminating another candidate from Rhodes. Achaïe III 134, a highly fragmentary text, has been identified as a catalogue of proxenoi, but the use of the dative case makes this unlikely. IvO 31, a decree of the Arkadian koinon granting proxeny to five individuals from Magnesia, Thebes, Syracuse, Sikyon, and Argos, ‘providing money to the sanctuary of Zeus’ on an occasion when the Olympia was under Arkadian control, has also been omitted from this appendix. 4

OUP CORRECTED PROOF – FINAL, 5/2/2015, SPi

288

Appendix 1. Aitolia

1.1–8. Eight limestone stelae bearing twenty-four chronological lists of proxenoi appointed by the Aitolian League (third to early second century bc) IG IX 12 1 13, 17, 21, 24, 25, 29, 30, and 31; cf. Fossey (1996).

The earliest of these lists, no. 1, is in the form of a series of abbreviated decrees dating from the same year and inscribed one after another on a single stele. The other lists in this series, by contrast, are differentiated from the individual decrees granting proxenia (and politeia) that were also sometimes inscribed at Thermon during this period by a distinctive structure: a heading, dating the list by magistrates (especially stratēgos), and proclaiming something like ‘the Aitolians gave proxenia in accordance with the law to the following people and their descendants’, which is then followed by a list of recipients in the dative (in the form name, patronym, and ethnic). The name of each proxenos entry in the dative is also then usually followed by the name of an Aitolian enguos in the nominative (giving his name, patronym, and city ethnic). These lists of recipients of proxenia are inscribed on the same stelae as decrees granting politeia and in a number of cases, because the initial heading has been lost, it has been thought uncertain whether or not a particular text lists recipients of proxenia or politeia (this is true for the largest of the lists of proxenoi, no. 2.1).6 In fact, the texts analysed here can be confidently identified as lists of proxenoi on the basis of the occurrence of enguoi, even where the initial heading has been lost or is incomplete.7 One thing that makes the interpretation of these lists of proxenoi more difficult is that the limestone stelae on which they were inscribed seem to have been viewed by the Aitolians as a reusable resource. In the case of the larger stelae, in particular, earlier texts were chiselled out to receive new lists of proxenoi. The best example of this is no. 1.8, a stele which stood nearly two 6

Fossey (1996). Within the Aitolian corpus of honorific decrees for external citizens (IG IX I2 5–50, including individually inscribed decrees as well as lists), the appointment of a citizen enguos is clearly directly linked with the granting of proxenia rather than politeia. In all the grants of proxenia that the Aitolians inscribed there is only one recipient for whom no enguos is inscribed (12 ll. 25–35). By contrast, of all the decrees which grant politeia without proxenia, for only two are enguoi listed (9, an unusual grant for a woman, and 30 ll. 10–17; 17 ll. 131–4 should probably be restored as a proxeny grant and [æ][]Æ[] would be a perfect fit for the traces reported in l.131). This is in comparison with the 12 other known grants of citizenship for which enguoi are omitted, often inscribed alongside proxeny decrees where they are present. Where both proxenia and politeia are granted together, we often see the further specification ªªı A æÆ , which can leave no room for ambiguity concerning their connection with proxenia. On the likely functions which these citizen enguoi performed, see Chapter 2, p. 103 n. 52. 7

Appendix

289

metres in height, on which a number of original text or texts were erased and replaced with a succession of fifteen different chronological lists of proxenoi inscribed over a period of fifty years (in addition to a further three texts granting politeia).8 This process of erasure and over-writing, wholesale as it was, is unlikely to reflect attitudes to individual proxenoi or a rescinding of grants, even though in certain cases it could apparently occur within a single generation of the original text (see Klaffenbach on no. 1.4.1–2 (IG IX 12 24 b and a)). Table 1. The proxeny lists of Aitolia Text (IG IX 12 1)

Date

Number Ethnics appointed

1.1 (13)

c.271/0

14

1.2.1 (17 ll. 1–96)10 Before 262 bc

74

ŁÅÆøØ; ªÆºº ÆØ; _ ªÆºº ÆØ; ˆºÆıŒ øØ; ØƺE; ˚ØøØ; ˆæªŁØ[]E; ¨ƺHØ _ _ _ (3);9  I ØÆE; ƺøæØÆ AØ; ˜ıÆØ_[ ] _ __ _  I ØÆE ŁÅÆøØ;  ¯ æøØ; ºØ ÆØE; _ _ _ _ÆøØ; ¸ÆæØÆøØ; [ ] __ ªÆºº  ÆØ (2); ØŒıøøØ; _ _ ˚Ææı øØ;__ çغ Ø; Ø ıºÅÆøØ; _ _ ÆŒØ; ¨ƺHØ Kª ¸ÆæÅ ; —ººÆøØ; [ ÆŒ]Ø (2);  ¯ åØÆøØ; _ _ _ —ÆæÅŁØ (2); —ÆæÅŁøØ; _ _ —ÆæÅŁøØ;  ¯ æ æØE;   øØ; _   øØ;  øØ; æŁıE; ƺŒØE; _ ƺŒØE; ˚ƺåÆ øØ; æÆŒØ ÅØ; [ƺŒ]ØE; ŁÅÆø_Ø;_ [Ø]ŒıøøØ; _ øÆ_[]øØ; _ __ _ __ _ [- -]ŁøøØ;  — ººÆE; _ _ _ __ _ continued

8 31 t, a list of grants of politeia dating to before 238 bc, is the oldest text on this stele though it occurs in what was probably the last space to be inscribed on it. 9 This proxeny decree, dated to the generalship of Polykritos of Kallion, interrupts a list otherwise devoted to proxeny decrees made during the year of Skopas of Trichonion. Presumably their generalships were consecutive and this list of proxeny decrees was inscribed early in the year of Polykritos. 10 This list of 74 proxenoi was inscribed in two columns (ll. 3–50 and ll. 51–96) beneath a heading of which the first line and the first part of the second line have been lost, which would have read something like [Kd  æÆ Æª  nomen, ƒÆæå  nomen, ªæÆÆ   nomen, | æÆ `N øºd øŒÆ] ÆP E ŒÆd KŒª[Ø ] [E]. _ _ individual _ __ _ _this _ _ list, to separate Particular care was taken, presumably because of the length_of entries with dashes (a practice also seen in no. 1.6), in addition to which paragraphoi are also used, especially in the second column, to highlight the beginning of some new entries. Four entries are also highlighted by monograms which were inscribed on the left of the second column. These monograms were composed of the initial letter or two letters of the ethnic of the proxenos which are inscribed directly to the left—l. 62 (ÆøØ), l. 73 (ƺŒØE), l. 85 (ØŒıøØ ), and l. 88 ˚æ(Å ). The reason for the selection of these particular entries is unclear. On these monograms, see Ritti (1969).

290

Appendix

Table 1. Continued Text (IG IX 12 1)

Date

Number Ethnics appointed

1.2.2 (17 l.97–101)11 1.2.3 (17 l.102–126?)

c.262 bc

3+

mid-third century bc

13+

1.3 (21) 1.4.1 (24 b)12

c.245 bc c.250 bc

4+ 10

1.4.2 (24 a)

250–200 bc

6

1.5.1 (25 a)

c.246–236 bc

20

1.5.2 (25 d)

8

1.6 (29)

c.246–236 bc (after 25a) 210/9 bc

11+

1.7 (30) 1.8.1 (31f)13 1.8.2 (31g)

c.196/5 bc c.223/2 bc c.223/2 bc

5+ 11 6

1.8.3 (31h)

c.223/2 bc

3

åÆØHØ KŒ ¸[- -]; ˚ƺåÆøØ; _ [ºÆ ÅØ(?)]; _ _ ¯ ºÆ ÅØ; _ _ _ ¯ ¨ÆøØ; _ ÆøØ; _ Øæ ÆØ; ˚ıÇØŒÅHØ; _ ıæÆŒØ (2);  ¯ æ æØFØ (2); _ _ _ —ººÆøØ; ÆŒØ —ººÆøØ; KŒ —ººÅ ; ƺŒØE;  Æ øØ;_ _ ; ¸ÆŒÆØøØ;  Æ øØ _ ØŒıøøØ; ¨ÆØ ; ˚æØŁøØ; __ _ _ ØŒıø[]øØ; ˚æÅ [d] K ØŒıøøØ; _ _ __  ¯ ºıŁ[æ]Æ ; ŁÅÆøØ; ŁÅÆøØ; _ (2); [æÆŒ]Ø ÆØ [ Œæ]ƪÆ Ø _ [—]_ — ººÆ E_; [][º]ıŒE Ie _ __ _ _ _ _ _  Oæå[]øØ;  Oæå[]Ø[ ] (2); __ _ ª_Å_ [Ø KŒ] ˜Å _ Å æØ  ; _  ¯_ ç_ øØ; _ _ø[Ø]; _ _ __ _ _ _  _ _ _ —ÆæÅŁ[Ø ] [ ]Ł[Å] _ __ _ _ _[øØ] BıÇÆ øØ; BØø øØ ŁÆ _ ˚_æØŁØ (4); ØŒıøøØ;_ Ø ıºÅÆø fi ; _ < >ªÅ Ø Ie ÆØ æ _ _ [ı] _ æØØ ; ØøøØ; ÆæÆ Ø [ ] (2); _ __ ŁÅÆØ; ˚ıÇØŒÅHØ _ _ _ æØFØ (2);  ¯ æ æØE; æªØ (2);  ¯ æ

_ _  ¯ æ æØE;  ¯ æ æØE;  ¯ æ æØFØ; _ ØøE; Ø ıºÅÆøØ; —æªÆøØ;  ¯ æ æØFØ (2); Ø ıºÅÆøØ; _  Oæå øØ; ªÆæE; ŁÅÆøØ; _ _ _ [ªÆæ]E; ŁÅÆøØ; _ _ ZÆŒıŁØ _ ø[Ø]; ŁÅÆøØ;  EæØE; _ _ ŁÅÆØ (3) _ _ ºE; `NªØ ÆØ; ¸ÆŒÆØ[øØ]; _ _ _ _ _ŁÅÆøØ; _ _ _ _ ı_ æ__Æ_ŒøØ; ˚æø[E]; _ _ ¨Æ[ºE] _ _ _ ; _`NªØE _ º[]E; _ _ æªøØ_ æª[]øØ; æªøØ; _ _ÆØ _ _ _ _ _(4); Bı øØ; ˚ÆıØ (2) _ _ ÆŒº ÆØ KŒ F — ı (2); _  ˙æ _ _ ø_ E _ (2); _ _ _ _˚æÅ

_ _ _ d_  ¯ ºıŁæÆØ; _¨æ __ _ _ _ __ ˚æøE ˚æøØ (2); ¨ØE _ _

11 This list of proxenoi named in the second generalship of Polykritos of Kallion seems likely to have continued further, and may well have included the proxenoi inscribed on Face B or C (collected together here under 1.2.3). 12 This stele seems to have originally been filled with a list of individuals designated by name and patronym but not ethnic (of which part survives as 24 c, l. 19f.), which then saw two subsequent phases of erasure and inscription with lists of proxenoi (24 b and, finally, 24 a). On this, see Klaffenbach, ad loc. 13 Four individuals (three in a row) lack their ethnics in this list, probably as a result of careless drafting (ll. 32–3).

Appendix 1.8.4 (31s)

214/3 bc

11

1.8.5 (31i) 1.8.6 (31p) 1.8.7 (31q)

208/7 bc 205/4 bc 225–200 bc

4+ 2+ 5

1.8.8 (31a) 1.8.9 (31k)

225–200 bc 225–200 bc

5+ 10

1.8.10 (31l) 1.8.11 (31b) 1.8.12 (31c) 1.8.13 (31d-e) 1.8.14 (31r)

225–200 bc 196/5 bc c. 196/5 bc 192/1 bc 175/4 bc

6 1 1 5 1+

291

ƺŒ[ØE]; Ø; [ƺŒØ]E; _ ˚æŒıæ ÆøØ; ˆ_ æıØ_ (=ˆæ ıøØ); _ _ _ _ _ _`NªØØ [ ] ØåE; (2);_ åÆ[ØE K] _ __ _ `Nªı;  I ØÆØE; ŁÅÆØ; ˚[-] (2); [ƺŒØ]FØ (2) _ ÆØ BØø E _ (2) ¨Å _ ƺÆØ; [.]øØ Øæ ÆØ (2); [¯P]æøøØ;_ _ ¯ ªåÅ HØ Øæ ÆØ _ º[çHØ]; ˚Ææåı E (?) ˜ºçHØ_; ˜ _ _ _ ÆæÆ øØ; _ _º_º ÆØ _ __ ªÆ (2); — ƺE; _ __ _ (1 + 2); _ ˜ºçHØ; —ºıæÅøØ ¨ºçıøØ —æØ (6?); (—ºıæÅØ (1 + 2))14 __ ¸ÆæØÆøØ ¨ıæØøØ_ _ _ _ øØ; ˚ƺıøØ; —æå"Ø[ ] (3) Æ _ _ _ _ __ ºÆæE __

Map 7. The proxenoi lists of Aitolia, no. 1.1.1–8.14 (third to early second century bc) 14 This individual, Kleodoros, son of Kleodamos, along with his two sons, are inscribed here for the second time, surely by mistake. The drafting of these entries is particularly error-prone, see Klaffenbach’s note on l. 96 where Lykiskos from Kalydon, clearly the enguos of the previous proxenos, is inscribed in the dative as though he were a proxenos himself.

292

Appendix 2. Anaphe

2.1–3. Three chronological proxeny lists (fourth to second centuries bc) Three stones (2.1: a block of blue marble; 2.2: a column of white marble; 2.3: a stele of blue marble), each inscribed with the names of a number of different proxenoi added in different hands, apparently as they were appointed. The entries are made in an identical format, despite the differences in the dating of these stones: the name of the proxenos, comprising of the standard elements (name, patronym, ethnic), followed by the phrase ‘proxenos of the Anaphaians, himself and his offspring’ (æ ÆçÆø ÆP e ŒÆd ªªØ)—a sort of abbreviated decree formula.

Table 2. The proxeny lists of Anaphe Text

Date

Number Ethnics appointed

2.1 (IG XII 3 251 + XII 3 suppl. p. 279)

350–300 bc

8

¨ƺe KŒ Ææ ºı; ıŒØ ; ˚Ø ; — æØØ (2) E ; []º[]Å[]

325–275 2.2 (IG XII 3 250 bc + XII 3 suppl. p. 279)15 200–150 2.3 (IG XII 3 252 16 bc + XII 3 suppl. p. 279)

4

 Oº[] Ø[] ; ç[Ø ]

9

  ØØ (2); ¨Å[æ]ÆE[ ]; #.... ; [˚]Ø ; [- -]غ (?); [º][ª ] [æ]Ø ; . . . Ø ; ¨ÅæÆE

15 ed. pr. Legrand BCH 16 (1892), 146 n. 36 cf. Robert (1969), 382–91 (on the restoration of the ethnics in this list). 16 One difference that may be significant in this text is that, whereas in the earlier inscriptions all grants seem to be explicitly stated to be hereditary (using the formula ÆP e ŒÆd ªªØ), here only some are. This apparent distinction may relate to a more general trend in the late Hellenistic period, evident in a more extreme form in the catalogue from Astypalaia, to designate only a selection of proxenies hereditary.

Appendix

293

Map 8. The proxeny lists of Anaphe, no. 2.1–3 (fourth to second centuries bc)

3. Astypalaia 3. A catalogue of proxenoi (c.150 bc) Block 1 (‘A’): ed. pr. Legand BCH 16 (1892), no. 23, p. 138 (Michel 416); Hiller IG XII 3 168, XII suppl. p. 79 and Syll.3 722; Rouse JHS (1906), 178; Peek (1969), no. 82. Block 2 (‘K’): Peek (1969), no. 97. Block 3 (supplements): Peek (1969), no. 98. Dimensions: Block 1 (0.4m high; 0.9m wide; 0.215m thick); Block 2 (0.23m high; 0.68m wide; 0.19m thick); Block 3 (0.225m high; 0.395m wide; 0.17m thick). cf. Robert BE (1971) 477, 485; Gschnitzer (1973), cols. 698–9; Hitchman and Mack (in preparation). See Map 2 (Chapter 3, Section 3.3) and, for photographs of squeezes, Figures A.10–A.12. Three blocks of blueish limestone from a monumental wall bearing a catalogue of proxenoi. The names of the proxenoi are inscribed in a series of columns, grouped according to the polis from which they came, and these polis groups were then further organized alphabetically (albeit by first letter only). The catalogue as a whole was prefaced with a decree, the bottom of which is preserved on Block 1 and concerns the inscription of future proxenoi. Block 1 also preserves five columns of proxenoi from the beginning

294

Appendix

of the catalogue (at least 6 groups of proxenoi from poleis beginning ‘A-’). Block 2 preserves four columns from a later section of the same catalogue including two groups of proxenoi from poleis beginning with the letter ‘K’. The third block is inscribed in a variety of different hands, apparently with the supplements to the catalogue provided for in the decree, but much about this stone is unclear (especially the apparent absence of polis ethnics).17 The text of the decree, and thus also the columns of proxenoi continuing on ll. 39–42 and 61–76 of the extant text, seem to have begun on a block of the wall immediately above Block 1 (whether the entire catalogue was inscribed across a height of two blocks is uncertain, and Block 2 perhaps tells against this). What the earlier part of the decree treated is unclear, except that it contained provisions concerning the inscription of the catalogue as a whole, but it is not unlikely that it may also have defined the package of privileges granted to the proxenoi of Astypalaia. The alphabetical organization of this catalogue is striking—especially in comparison to the other catalogues which we possess in which the proxenoi, grouped by polis, are organized by region (see nos. 4.1, 7, and 10.1). Probably in the case of Astypalaia, which seems to have had an extremely dense network of proxenoi within its local region and fewer connections beyond it, regional organization would have been of little help in offering a structured view of its network of proxenoi. The entries which have survived from this catalogue are enough to give a rough impression of its original total size. Block 1 preserves the names of some forty-eight proxenoi from poleis beginning ‘A’ (including fifteen from unknown poleis); Block 2 would have been inscribed with fifty-one proxenoi

17 The identification of this as a list of proxenoi was made by Peek, on the basis of the attestation of the phrase ‘ŒÆd ŒªØ’ col I, l. 3. We may also have, in col. I, l. 8, and col. II, l. 5, the names of demiourgoi in the genitive case preceded by the preposition K, to indicate that it was in this year that the proxenos was appointed. However, only one potential proxenos ethnic can be identified in the list, ˝ Ø (col. II, l. 12), which may perhaps be a personal name. Perhaps loose drafting is at fault here. The decree states that individuals are to be inscribed ‘on the wall in the same manner’, i.e. on the wall on which the catalogue was inscribed in a similar form (N e Eå ŒÆ a a ÆP

). However, the format apparently decreed for the catalogue—proxenoi grouped by polis beneath a polis ethnic—was clearly inappropriate for subsequent supplements made each year, and perhaps polis ethnics were omitted in these cases as a result of this confusion. The two cases on the second column in which we have individuals listed without even their patronyms (l. 8 and l. 12) are harder to explain in this way as the decree would certainly have specified the inscription of a patronym per proxenos—could this be an alternative means of stating the demiourgos in which a particular grant was made, distinguishing them from the proxenoi who follow who are given their patronyms?

Appendix

295

from poleis with ethnics beginning ‘K’ and possibly also ‘I’ (including an estimated twenty-one proxenoi for Columns I and II). Given the uneven distribution of polis names throughout the alphabet, it seems reasonable to suppose that the eighty-nine proxenoi for these two or three letters constitute a fraction somewhere between a third and a fifth of the original network, which we would thus estimate in the range of 270–450 proxenoi. By extrapolating the pattern of distribution evident in the extant sample of the catalogue we can get a sense of how this could work—270 proxenoi would mean that the Astypalaians recognized groups of proxenoi at something like ten further small poleis and ten further larger poleis.18 It would not be difficult, by allowing for somewhat greater coverage (especially on Crete), to account for a network closer to the middle or even upper range of this estimate. Dating : Despite the large number of names attested in this catalogue of proxenoi, relatively few prosopographical conections can be made. The most useful, in terms of dating, are a number of Koans on the second block: most importantly, Aglaos, son of Theukles (Block 2, col. III, l.6), apparently an important officer of Ptolemy VI Philometer, attested as active in the mid to late second century bc by two honorific decrees by the Parians (SEG 33 682) and the Cretan koinon (ID 1517);19 and Mandrogenes, son of Mandris (Block 2, col. III, l. 14), who also occurs in TCal. 85 l. 38, a subscription list dated to c.200 bc on the basis of an overlap with individuals contained in IG XII 4 75. Philistos, son of Nikarchos (Block 2, col. III, l. 10), identified by Habicht, ap. Robert BE 1973 no. 125, with the doctor honoured by the Samians in a decree to be dated after 241 bc, must be a relative in a subsequent generation rather than the man himself (so Rigsby, IG XII 4 138, ad loc.).20 The text presented of block ‘A’ is that of Hiller supplemented by the readings of Peek (which I do not, for the most part, note separately). I have preserved the line numbering of the original editions. I present scans of the Berlin squeezes of these texts, which Jaime Curbera kindly brought to Oxford and Maggy Sasanow scanned. For the geographical distribution of these proxenoi, see Map 3 (Chapter 3, Section 3.3).

18 This reconstruction is based on average concentrations of six proxenoi at smaller poleis (ten likely candidates: Andros, Delos, Thera, Ios, Mykonos, Naxos, Paros, Sikinos, Syros, Tenos) and twelve proxenoi at larger poleis (ten further possibilities: Chios, Gortyn, Ephesos, Iasos, Kaunos, Miletos, Mylasa, Priene, Rhodes, Samos). 19 On Aglaos, son of Theukles, see also: Robert BE (1967) n. 441 (on the Parian inscription); Bagnall (1976), no. 150; Savalli-Lestrade Chiron 26 (1996) 152–3, n. 9 and 12; Höghammar (1996), 350–2; and, Habicht (2000), 297–8. 20 Contra Sherwin-White (1978), 270 and also Höghammar (1996), 350–2, who would update the Astypalaia catalogue on this basis.

Block 1 (‘A’)

Block 2 (‘K’) Col. I

[ ][]ºº[]øæ _ ¨Ø ı _ —  ÆØ 4 ˜Øıı vacat

Col. II ———— ˜ØØ غ[]Œæ ı [. .]Ø [— — — —] _  I [Ø]Æı Zøº[] 5 ———— _ _ [. .]Æ[. . ]øØŒæ Å _ _— — —— ———— [ . . . .]Æ []Ææغ ı 10 — — — — _ _ _ [.][ . . . .][]ı [ ]æ[Œ]ºÆ _ _ [ . . . .]ı ŒıºÆ _ ˜[Ø][Ø ˚]ıØŒºF _ ¯PæÆ  —ºº ı 15 ÆƪæÆ ºø ˜æ Œø —æªı —æªE ˜æ Œ  ˛ø —ºıŒæ [ı ] __

Col. III [˚H]ØØ `[ . . . . . . ]Å ˜æ Œ  #[. 5-6 .] غı _ 6-7 .] Åı —[. 5 [. 4-5 .] ˜Æª" ı @ªºÆ ¨ıŒºF ºŁÆØÅ  ˙æØÆ Œ"æÆ  غ ı _ _ __ ˝æÆ ˝ØŒÆ 10 ºØ  ˝ØŒ æåı —æÆºÆ Å Æ ˚ºE `Nåºı ˚ºåØ ˚ºØƪ[æ]Æ ÆæªÅ [æØ] 15 غŒ å[- -] vacat

Col. IV ˚ØØ æØ Œæ Å æØ "ı ºØ 5 % ÆØŒºF ØÆE ØÆı ŒºB ˚ƺºı  ˙æÆø ˚ºø 10  æÆ  øØŁı ¨ÆÅ  ºÆæÆ  ¯ ª 15  تı % Æ挺B  Iæø  ¯  Œ  ˚ºø []ºø ºø

298

Appendix

Block 3 (supplements?) Col. I [. 5-6 .] . # . [.] . . [- -]  ¯ [Øç]ÆÆ ¸Æ [ı] _ _ _ [. 3-4 .].[. .]ı ŒÆd ŒªØ [ . . . .] ˚º  ı 5 øÆe —æÆ ºÆ _ _ __ [.]ıæÆ[.] øÆF _ [][Ø]Æ ¸[] ı [Kd]  ˙åı ¯P []æı [. .]Å ØŁı (?) 10 [. .] —æø Æ _ [¯P]ƪæÆ ¯PŒº ı [¸]ıƪæÆ —ÆæÆF ˚ÆæE ¸ÆŁı __ æø ˚ºÆæ ı 15 [ ]æø º[ æı?] _

Col. II ———— `[-] _ ŒºÅ[ -] _ _ __ ¸ıØÆ[ -] vac 5 K[d] Ł Ø _ ¸[. .]Å [-] __ _ [. 6-7 .]ºØ[-] [ . . . .] %[. .]. Ø __ 10 Ø[]  ØØ[-] __ _ Z Øå ¨æ[ı] ˝ Ø ¯hÆ [ . . . ] . . . [-]

Block 1 (‘A’): 5–6 Peek divides incorrectly here (IÆ|ªæÆça) but rightly notes ÆP HØ, earlier omitted. 6 Rouse incorrectly reports ÆÆ for ºÆ. 8 Peek omits at the end of the line. 12 Peek omits at the end of the line. Col. I 16–17 Inserted line read by Peek. Col. II 20–3 From the cramped spacing of these entries, this group of two proxenoi from Aspendos appears to have been inscribed after the other entries had already been completed (this is particularly clear in l. 24, see ph.). However, from the format (a catalogue entry—it is not inscribed beneath the name of a demiourgos) and lettering it is clear that this was not a subsequent supplement, but contemporary with the main catalogue, if reflecting a second thought concerning its layout. For lettering which is similarly cramped in comparison with the main text, but nonetheless still part of the catalogue, see Block 2 Col. IV 29–34. Col. III 29-30 The occurrence at the head of this column of Archonidas son of Sosikleus, otherwise attested on a funerary stele from Anaphe (IG XII 3 292), makes it clear that it continues the list of proxenoi at Anaphe from the previous column. The lacuna between the two columns is likely to have contained the names of up to two further proxenoi. Col. IV 41 The suggestion made by Marek (1984), 291, that this individual could be identical with the man of the same name in IG VII 4131 ll. 6–7, a foreign judge from Larissa, is excluded by the name of that man’s polis (which does not, of course, begin with an ‘A’). Col. V 61–86 The interpretation of this final column is difficult. The lettering here is significantly more cramped than the earlier columns in order to fit it on to this block without overrunning onto the next. This need not, however, indicate that it was (or that all of these entries were) inscribed subsequently—the lettering of column IV on

Appendix

299

Block 2 is similarly cramped in comparison with the rest of the inscription (and names and patronyms are placed in individual lines there as for Block 1 ll. 61–76). 77–86 If these lines are contemporary with the rest of this inscription (and not subsequently inscribed proxenoi headed by the name of the demiourgos), then it is tempting to read ll. 77 and 83 as ethnics (e.g.  ¯  ’`[- -], possible given the way in which the genitive of the toponym is used as a heading in l. 43; for l. 83, Peek proposes @[ØØ], but Hiller thought he could read letter traces before the A). Block 2 (‘K’): Cols. I–II On the identification of the polis from which these proxenoi came, see Hitchman and Mack (in preparation). Peek recognized the names of a number of the proxenoi in Col. II as Cretan, suggesting that perhaps it was originally headed by the ethnic ˚æB  , on the model of Cols. III and IV; although this method of referring to Cretans is paralleled in the catalogue from Kleitor, given how much closer Astypalaia was to Crete and evidence of other interactions with specific Cretan poleis, Hitchman’s suggestion, grounded in onomastic analysis, that they may have been  IæÆ ØØ (which would work in terms of the alphabetical organization of this list) should be taken seriously. Col. III On the prosopography of the Koans, see above. 5 Probably to be corrected to the common name ˜Æ ªÅ ı (so Peek). Col IV 14–15 This individual may be a relative of a man of the same name, attested among other Knidians in an inscription dated c.280 bc, in which they are thanked for making a loan to the Milesians (Milet I 3 138). Block 3 (supplements): Col. II 12 ˝ Ø taken by Peek as the ethnic of the individual in l. 11, inviting the question where the ethnics for the other individuals are (see above, n. 17). 14 Peek reads a kappa.

Translation (Block 1, decree): [The secretary then in office is to record] those appointed proxenoi according to the law in their month (of office). [The secretary] is to have them inscribed on the wall according to the same regulations, inscribing also the damiourgos in whose period of office each was appointed. The tamiai are to give him the payment for the inscription, paying out not more than half a drachma for each proxenos. If the secretary in office does not see to the inscription of (the proxenoi) appointed after the damiourgos Prytanis, he is to pay a fine for every proxenos he fails to have inscribed of one hundred drachmas. Let anyone who wishes denounce (such a failure) to the logistai in return for half the fine. And whoever obtains the contract for inscribing the proxenoi (sc. of the catalogue) is to inscribe this decree at the start. Agreed unanimously.

300

Appendix

Table 3. The proxeny lists of Astypalaia Region

Number of Proxenoi

Local Region Pamphylia Troad?

64 2 3

Surviving Ethnics Arkesine (5+);21 Anaphe (6+); Aigiale (3+);22 Halikarnassos (17); Kos (14); Knidos (11);23 [Hierapytnioi? Or Cretans?] (8+);24 Aspendos (2) Assos? (3)

4. Chios 4.1. Catalogue of proxenoi (fourth or early third century bc) ed. pr. Zolotas, Athena (1908), 214–16, no. 13 (with corrections by Wilamowitz (1909), 21 n. 1; (McCabe Chios 51); Vanseveren (1937), 328–30. For a photo of this fragment, see Figure A.13.

Two marble fragments bearing two columns of names, grouped beneath foreign ethnics. The identification of this as a list of proxenoi, first proposed by Wilamowitz, was persuasively argued by Vanseveran who produced a cursory new text for the first of these stones, or, rather only a fragment of it (the height she gives, of 0.29m, is less than half the 0.75m recorded by Zolotas; the rest of this stone and the other fragment remain lost to this day). This fragment (Chios Museum inv. 522) was badly worn and difficult to read in Vanseveran’s day and is no easier now (see Figure A.13). Proper republication of even this fragment lies beyond the scope of this appendix. However, from the photograph taken by Charles Crowther, it is clear that the structure of this list is more complex than existing accounts suggest and that it shares important structural similarities with the supplementary list of proxenoi (no. 4.2), especially the use of plural ethnics as headings for columns of proxenoi, which were offset to the right beneath, and also evidence of different strata of inscription. In particular, it is clear that a group of two proxenoi from Rhodes were subsequently inserted at the top of the second column in the space left between the end of one group of proxenoi and the start of a group from 21 The names of at least five proxenoi survive in the first column. Probably between one and three more were obliterated with the destruction of the inscribed surface in the bottom left of the stone, giving us a maximum of eight proxenoi in this group. 22 This group of proxenoi probably continued on the block above Block 1, in the fourth column (but it seems extremely unlikely that the proxenoi in ll. 39–42 in that column are a continuation of that group rather than a new group). 23 The absence of a vacat at the end of this column means that we cannot be sure that this list of proxenoi at Knidos is complete. 24 See on Block 2, cols. I–II.

Appendix

301

Karystos. If this group in particular is omitted, clear evidence of regional organization can be detected in the ethnics identified by Wilamowitz for the first part of this column, comprising groups from Karystos, Keos, Peparethos, and Herakleia Trachinia (all poleis on or near Euboia). This evidence implies the systematic organization and inscription of a large number of proxenoi from different places and suggests that we are dealing with a catalogue of proxenoi that collects and presents a complete network of proxenoi for the first time and not a supplementary catalogue, like no. 4.2. A Chian decree of c.320 bc instructs that the names of certain foreign judges be inscribed on ‘the proxenic stele’ ( c  "ºÅ c æ[ØŒc]), quite probably this inscription.25 The following groups of proxenoi can be identified (with varying plausibility) from Wilamowitz’s corrections to Zolotas’ text: two from Rhodes; four from Karystos; three from Keos (with one further identified as having come from Koresia and two from Ioulis); one from Peparethos; between two and five from Herakleia Trachinia; perhaps up to seven proxenoi from Aitolia; two Dolopians; up to four Thessalians; and up to twelve Macedonians.

4.2. Supplementary catalogue of proxenoi (early third century bc) ed. pr. Vanseveren (1937), 325–32, no. 6, ph. of squeezes. Dimensions: 0.335m high; 0.235m wide; 0.23m maximum thickness. For photographs of the stones, see Figures A.14–A.16.

Two fragments of a block of marble inscribed on two contiguous faces (Chios Museum inv. 618 and 1010). Two phases of inscription can be identified (signalled below by different font sizes). The first phase on both faces involved the inscription of polis ethnics in the plural as headings, with the names of proxenoi inscribed beneath them, offset to the right. Subsequently, the names of further proxenoi were added in any available space. This inscription does not seem to have been a complete catalogue of proxenoi at the point when it was originally inscribed because most of the polis groups which are preserved for the first phase consist of one proxenos each and no effort to organize these groups by region can be detected (see Face A). Instead, it seems that this is a supplementary catalogue, probably begun because the space for inscribing new supplements had been exhausted on an earlier catalogue of proxenoi (perhaps no. 4.1). The best criterion for dating this inscription is the occurrence within it of Kallias, son of Kallias, the younger, of Erythrai (Face B ll. 13–14), who is inscribed as a supplement to this list and is otherwise attested at Erythrai in a document concerning the sale of priesthoods, which has been securely dated to the first half of the third century bc (I.Erythrai 201 d ll. 4-6).

25

SEG 12 390 ll. 30–1.

302

Appendix

Face A Fr 1

Face A 6 Vanseveran Øæı. 9 ˇ%%¸: surely the shared patronym of the two preceding proxenoi (perhaps to be understood as ƒ %¸). 10  ¯ ºÆØØ here functions as an ethnic heading for [․2–3․]ŒæÆ Ø[—]  ˙ª[ .], cf. ll. 11–12.

Face B Fr. 2

4 Fr. 1 5

.. Ἀπολλ[- -] [—]ΟΙΙΣ Βυζάν[τιος] Μητρ[—] Μητρ[—] [—].ος Σαμοθ[ράιξ.]

[—]ΑΜΟΣ [—]

Letter traces [— ο]υ Σῶσος Χαρμ[ —] —] [—]χου̣ Ἀπολλωνι[ Ἑρμογένου̣ [—]χου Ἡράκλειτο[ς] [—]ου Ἀσκληπιάδο[υ] Ἁλικαρνασσεύς. [—].ου

Appendix 10

15

20

303

[—]θηναιος Τήϊος [—]ος Ψύλλο[ς] Βοηθ[ - -] [—]Ἀρχαγόρης [—Ν]ικήτης Καλλίας Καλλίου ὁ νεώτερος [—]ας Ἐρυθραῖος [—]σιλας [-ο]υ Καύνιος Ἀσκληπιαδη[ς] [- -]. Ἑρμίου [-Διο]νυσίου Φιλομήλου [-] [Φιλό]μηλος [Κο?]ρ̣ίνθιος ΙΣΙ////ΙΗΟΣ///// Λου Ἐλαι[ίτης]

[- -] Ἀπολ[λο]δωρ[ο͎ [-] [—]ΙΕΡΜ[—]

Face B 20 [—]æŁØ is also a possibility. Vanseveran proposed [—Æ]æ" ŁØ tentatively, __ but the iota and nu are clear from the photographs.

4.3. List of proxenoi appointed in a particular year (?) (second century bc) ed. pr. Zolotas, Athena (1908), 213–14, no. 12; Vanseveren (1937), 327–8. Chios Museum inv. 205. Dimensions: none yet published. For a photo of this fragment, see Figure A.17. A table summarizing all of the ethnics preserved for Chian proxenoi in lists 4.1–3 (Table 4) and a map showing their distribution (Map 9) are presented at the end of the section on Chios.

A stele, the heading of which has been lost, bearing the names of of sixteen individuals in the nominative from different poleis, mostly from coastal Asia Minor and inland Lydia and Caria. It has been suggested that this is a list of mercenaries, but, given the fact that each polis is only represented by one individual, this is rather unlikely. Since poleis found here are represented in 4.2, the earlier catalogue of proxenoi (Gargara, Elaia, and Halikarnassos), it seems likely that this is also a list of proxenoi—probably appointed within a single year— in which case it would indicate closer contacts with inland Asia Minor in this period than we can observe from our earlier (admittedly highly fragmentary) lists. The order in which the different poleis are listed is certainly not geographic: if it is not random, it may reflect the order in which these individuals were appointed. [․2–3․]ŒºÅ — æØ ˚ºÆ[ÇØ ] [․]ÅŒÆ ˆæªı ºÆÆ[ ] [ ]Łf [․1–2․]ı ¸[Ø ] [․․․]øæ åø ÆæØÆ 5 [․4–5․]Ø  ˙æÆŒº[ ı] "œ [ ]Œæ Å  EæÆŒ 

F  EæÆŒ  AºØŒÆæÆ  EæªÅ  I æåı BÆæªıºØ" Å   æ Ø  ¯ ºÆØ Å 10 ÅÆå ººæı ˝ıÆ[ ] çæø ˚æÆ æF ˚Øıæ [Å ] ¨  —ıŁø ¨ıÆ ØæÅ ÆæÆ Łø  OæŁøØ ¨çغ  ¯ تı 

æØ

304

Appendix

15  ˙æ ŒºØ  źçı ˆÆæªÆæ çæø —Æ æŒºı ıÆ Ææ Å ææı ¸ÆØŒ .

4.4. Unpublished fragment of a list of proxenoi A new fragment of a list of proxenoi from Chios will be published in due course by Dr G. Malouchou. Table 4. The proxeny lists of Chios Text

Number of Proxenoi

Regions/Poleis Attested

4.1 (fourth or early third century bc)

40+ (?)

4.2 (early third century bc)

46+

Mainland Greece: Herakleia Trachinia (2–5); Aitolia (7?) Thessaly (4?); Dolopia (2?); Macedon (12?) Aegean Islands: Rhodes (2); Karystos (4); Keos (3: 2 Ioulietans and 1 Koresian); Peparethos (1) Asia Minor: Gargara (1); Elaia (1); Teos (1); Pergamum (1); Phokaia (1) (Halikarnassos (3); Kaunos (1); Propontis/Black Sea: Byzantium (2); Kyzikos (2); Pantikapeion (1); Lampsakos (1); Elaious (1); Sinope (2) Aegean Islands: Mytilene (1); Astypalaia (1); Tenos (1); Samothrace (1) Mainland Greece: Amphipolis (1); ethnos of Malis (2); koinon of Boiotia (1); Plataia (1); Corinth (1)

4.3 (second century bc)

17

Asia Minor: Klazomenai (1); Alabanda (1); Lebedos (1); Sardis (1); Teos (1); Halikarnassos (1); Bargylia (1); Elaia (1); Nysa (1); Kibyra (1); Thyateira (1); Orthosia (1); Antandros (1); Gargara (1); Synnada (1); Laodikeia (1)

5. Delphi 5. A chronological list of proxenoi appointed over twenty-two years (197/6–175/4 bc) Pomtow Syll.3 585; cf. Daux (1936), 17–46 and 421–4; Gschnitzer (1973), col. 698; Marek (1984), 168. See Map 4 (Chapter 3, Section 3.3).

A chronological list of proxenoi inscribed on sections C–D of the great polygonal retaining wall below the temple of Apollo and apparently updated regularly by different letter-cutters with new proxenoi. The main phase of inscription collected proxenoi from a twenty-two year period (197/ 6–175/4 bc) until the free space was exhausted, but a small number of other additions were made irregularly in the years that followed, the last in

Appendix

305

Map 9. The proxeny lists of Chios, no. 4.1–4 (fourth to second centuries bc) 149/8 bc. The text begins with the simple heading  ˜ºçH æØ. Within the list the names of the proxenoi who have been appointed (given in the usual form of name, patronym, and ethnic) are then preceded by a formula giving the names of the eponymous archon and the bouleutai and specifying, within the year that these names indicated, the particular halfyear in which the decree was passed. In some cases the proxenoi appointed within a particular period are grouped together under a single heading, although this is usually restricted to groups who were appointed by the same decree, especially families and, for example, groups of theōroi. In most cases, however, the name of each individual proxenos is preceded by the entire dating formula, which separates, on occasion, the names even of proxenoi who were clearly appointed in the same context. These intrusive dating formulae reflect a tension, which we can observe in relation to other chronological lists, between their function as a register of proxenoi appointed by a polis in a particular period and the function of the entries within it as highly abbreviated individual decrees granting proxenia to particular honorands. In the case of this chronological list from Delphi, the (usually) separate dating formulae highlight the fact that the proxenoi listed are the product of individual decrees. At the same time, however, the identity of this as a coherent monument is asserted by the fact that the entries, as

306

Appendix

decrees, are incomplete as they stand. Individually they lack the decretic formula by which proxeny was granted and therefore depend, syntactically, upon the initial heading, with which the names of the recipients agree (  ˜ºçH æØ). The fact that the proxenoi in this text are not gathered together under dating headings prompts the question how far it can be taken as a complete record of all proxenoi appointed by the Delphians in the period 197/6–175/4 bc. One important indication that it should be taken as an effectively comprehensive record is that the inscription of individual proxeny decrees all but completely ceases in the period covered by this text, whereas earlier the inscription of individual proxeny decrees at Delphi, more than at any other polis, seems to be the norm.26 Another significant phenomenon is the fact that some proxenos entries are inscribed out of chronological sequence.27 This suggests two things: firstly, that the administrative processes of collating and inscribing the names of proxenoi appointed in particular years were not foolproof; and, secondly, that the correction of such omissions subsequently seemed important, and thus that completeness was the aim. Some of the impetus for these corrections may also have come from the proxenoi themselves. In particular, it is striking that, where these corrections were made, they were sometimes inserted in the context of later grants made to individuals from the same community.28 For the distribution of these proxenoi, see Map 4.

26 In the five cases where full decrees for honorands who occur in this list are separately inscribed, it is usually because these also relate to relations between the Delphians and the polis from which the individuals in question came (as in the case of the decree granting proxenia to the would-be arbitrators from Rhodes, FD III 3 383, to the theōroi from Pontic Chersonese, Syll.3 604, or to the theōros from Velia, BCH (1944/5) 102 no. 11). The other cases seem to reflect the importance of the individual in question (thus Marcus Aemilius Lepidus received a separate inscribed decree for his substantial services, FD III 4 427 B I; Kassandros of Alexandria Troas also had a record of his Delphic proxeny inscribed as part of the Ehrentafel erected in his honour, Syll3 653B; cap of Ehrentafel). The other side of this phenomenon is individuals known to have been named proxenoi in this period but who are omitted from the list of proxenoi. Only two proxeny decrees whose honorands are not included in this list certainly, or almost certainly, date from this period—FD III 2 89 and possibly 4 430 (contra Marek (1984) 409 n. 17, it is not clear that 1 453 is a grant of proxenia). The most likely explanation for the omission of these decrees from the proxeny list and their separate inscription is probably administrative oversight. 27 For examples, see the following entries: 196/5 II; 191/0 II; and 187/6 II. 28 This is the case in relation to the proxenia of Botrus, son of Timotheus, of Alexandria, inscribed after two other proxenoi from Alexandria appointed in the first half of 185/4 bc whose grant is dated by archon to 191/0 bc.

Appendix

307

Table 5. The proxeny list of Delphi Year

197/6 196/5 195/4 194/3 [193/2] 192/1 191/0 190/89 189/8 188/7

187/6 186/5 185/4 184/3 183/2 182/1 181/0 180/79 179/8 178/7 177/6 176/5 175/4

Half- Number of Ethnics year Proxenoi I II II I II I II

1 1 4 3 3 4 2

¸ØºÆØ ŁÅÆE ÆƺØH ÆØ (4); (¸ÆØE (3))29 æÆ Æ ; —Æ ØŒÆÆØ ; ˚æŁØ ˚ÆıE ;   øÆE ; `NªØ ŒÆæA (3); ˚HØ ºÆæf KŒ A æøØ  ; E

I I II

3 1 4

I II I II

8 3 1 8

I

16

II I II I II I II II I II II I II I I II I II II I

1 2 4 4 1 5 1 2 1 4 2 9 1 2 2 3 2 1 5 1

æÆE ÆØ KŒ F — ı (2); ˜ÅÅ æØ ŁÅÆE æªıæØÆ ; Bæ E ;  ˙æÆŒºØ Æ ; ( ºÆæ )30 ¨ÅæÆE ;  ŪE ;   øÆEØ (5); ŒæƪÆ E ººøØ Æ ; ¸ıŒ Ø ; —ººÆE ŁÅÆE ¨ØE (3); Œ ıÆE ;  øÆE ;   øÆE ;  øÆE ; ÆæÆ E  ¯ º Æ ;  ¯ º Æ ; ºÆæf (1+1); ºÆæE (4); ºÆæ ; ºÆæ ; ºÆæ ; ºÆæ ; ºÆæ ; ºÆæ ; ˝Æı Œ Ø ; ºÆæ ; ºø åØ æØå ; ˝Æı Œ Ø ’#æØ ; æ ªØ ; A  ºØŒÆæÆ ; (˚ÆÆæ )31 ¸ÆE (3); ¨Ø E ¸ÆE (2); ºÆÆ ; ºÆæ ; ºÆæ ; E EØ (2) ŁÅÆE æÆŒØ Æ ; ˚æøE (3)32 ˚ƺåÅØ ; ˚æŁØ   ØØ (9) (¨ÅÆE )33  ¯ º Æ ; ƺŒØ ZçıæØ Å ; ˚æŁØ ˚æøE (3)34 BæØ ;  ¯ ºÆ  ; % ºØ ŒıØØ (3); — çØ ;  ¯ ºÆ   ¯ ºÆ 

Inserted after the entries for 194/3 bc. On the subsequent inscription of this proxenos, see n. 28. 31 Inserted in the entries for 186/5 I. 32 The first of these, clearly the architheōros of a theōria from Koroneia, occurs in these lists as being appointed as proxenos again in 178/7 bc, again as architheōros. 33 This proxenos, dated to the year 180/79 bc, was inserted for reasons of space below the entries for 188/7 bc. 34 See n. 32. 29 30

308

Appendix 6. Epidauros

6.1–4. Four chronological lists of proxenoi (late fourth(?) to late third century bc) Four limestone stelae found at Epidauros preserve what seem to be a series of chronological proxeny lists from different years. Texts 6.1–3 collect a series of different decrees, apparently from a sequence of consecutive years, which name groups of individuals proxenoi and theōrodokoi (these are identified by name, patronym, and ethnic). These groups can be quite large—one of these lists (no. 6.3.6/VI) contains twenty-seven individuals. It is striking, however, that the headings which subdivide these texts present these groups of individuals as the proxenoi and euergetai appointed by the Epidaurians through a single decree, which is always made at the same assembly on the same day of the year—the fourth of the month Apellaios. (The different years can be distinguished by the eponymous magistrates, the katalogos of the boulē, named in the heading of each text, though, curiously, in some years more than one decree appointing proxenoi and theōrodokoi was made at the same assembly.) The reason why such potentially large groups of proxenoi and theōrodokoi were appointed on the same day each year—and why no other appointments made at different times of the year are noted—is probably that this was the annual assembly at which all proposals of proxenia made during the rest of the year had to be resubmitted for formal ratification,35 which is why the fourth of Apellaios, or rather the assembly held on it, is described as teleia in all these texts, meaning ‘final’ or ‘authoritative’ (Apellaios would also make sense as the final month in the calendar).36 These texts probably thus

35 For other two-stage processes for decreeing honours attested elsewhere, see Rhodes with Lewis (1997), 497–8 and Gauthier (2001), 212–19. Often these processes involved higher honours but they are also attested in relation to proxenia. In particular, we possess a set of four decrees from Rhodes from the early second century bc all concerning the grant and subsequent ratification of proxenia for Eudemos of Seleukeia (Syll.3 644–5 decrees a–d), with the form of decrees b and especially c (the subsequent confirmation) very closely resembling our Epidaurian texts. If the Epidaurian agonistic festival of Asklepios was indeed held in or before the month Apellaios, then it is possible that the reason for ratifying dates on this particular date related to the occurrence of this festival every five years (and a possible ‘lesser’ version of it in off-years). However, contra Perlman (2000), 91–3, who does not consider the possibility of ratification, the connection is unlikely to be that these decrees honour only theōroi and individuals attending these games, but rather that these decrees were ratified so that they could be proclaimed at these, the most prominent civic festivals of the honouring polis, as honorific decrees often were. 36 The usual introductory formula, Åe ººÆı 

æ ÆØ, ºÆØ, leaves the precise reference of teleia ambiguous, but this is made clear in no. 6.1.1/II where it qualifies the noun for assembly, [±ºØ]ÆÆ ºÆ (Perlman 2000, E. 3 l.3).

Appendix

309

provide us with a comprehensive view of the proxenoi and theōrodokoi appointed in these years.37 No. 6.4, to judge from the script and the fact that it is inscribed stoichedon, is likely to be the earliest of these inscriptions (late fourth or early third century bc), and it seems to have been structured in a different way. It simply lists a series of individuals in the form name, patronym, and ethnic, below the headings ‘—æ[Ø]’ and, probably, ‘[¨ÆæŒØ]’. Given the strong concentration of individuals from a single polis (three certainly from Phlius with another two possibilities), it seems likely that this text corresponds in some sense to one of the individual decrees contained in texts no. 6.1–3 (perhaps an earlier prototype), recording the names of the individuals appointed in a single year. These texts appoint individuals to two different honorific statuses, proxenia and theōrodokia, but the relationship between them is difficult to grasp. Within these lists for the most part it appears to be normal for individuals to be named either ‘proxenos and theōrodokos’ or simply ‘proxenos’, which suggests that proxenia was the more basic honour, with theōrodokia (which was explicitly identified with the temple of Apollo) being an optional extra, perhaps a mark of greater distinction. In some cases, however, proxenia seems to be omitted and the lists of honorands are headed separately by ‘Theorodokoi’ versus ‘Proxenoi’. The reason for this apparent inconsistency is not clear (it occurs within the decrees inscribed on a single stele; contrast e.g. 6.3.3/III with 6.3.6/VI). One possibility is that because this hierarchy of honours was so well understood, theōrodokia, at Epidauros, was understood to imply proxenia, and what we see is the result of inconsistent drafting. In any case, for all that these inscriptions are keen to assert a distinction between these two statuses, the way in which they are referenced in these texts asserts that they are closely connected, and performed similar functions and could be used, symbolically, to assert and reinforce a particular sort of symbolic relationship between polis and honorand. The regional distribution of the honorands in these lists appears remarkably narrow. In contrast to the very wide ranging fourth-century network of theōrodokoi attested by two inscribed catalogues (Perlman (2000) E. 1–2), which illustrate links with a very large range of poleis in Italy and Magna Graecia as well as western and northern Greece, the proxenoi/theōrodokoi appointed in these lists come predominantly from the Peloponnese. Where communities from outside the Peloponnese are attested, it is striking that

37 One indication that strengthens this suggestion—that, as in the case of the Delphic proxeny list, these lists became the primary means of proxeny decree publication—is that only one of the twelve proxeny decrees inscribed on individual stelae can be attributed to the same half century as these lists (Perlman 2000, E.14, and that on insecure grounds), in contrast to the periods both before and after.

310

Appendix

they are concentrated in specific years, apparently reflecting in these cases the attendance of theōroi from particular communities at the penteteric festival of Asklepius at Epidauros. There is also likely to be a competitive regional element to this attendance at particular festivals as well. Thus, it is hardly likely to be accidental that all the Boiotians represented in these lists are honoured in a single decree list no. 6.3.6/VI, and that many of them occur alongside groups from the same polis: one Theban, two Chaironeans, four Thespians, and two Tanagrans are honoured as proxenoi and theōrodokoi in this year, and a further Plataian is named as simply a proxenos (the regional dynamic here is further emphasized by the two Delphians and the Aitolian also honoured in the same year). The strong concentration of Cretans in 6.3.1/I–6.3.2/II may well be another instance of the same sort of phenomenon. For full texts and bibliography, I refer the reader to the excellent edition of these texts in Perlman (2000), which is the basis of the table I present here. Sample extract: Perlman (2000) E. 3, decree III, ll. 9–14 (= no. 1.2/III, below). Åe ººÆı 

æ ÆØ, ºÆØ, Kd ŒÆ ƺªø ıºA ºª , [ ]æ[ı , ¸ ]10 ŒæØ ,  ¯ ØŒæ ı ,  ıºAØ ŒÆd  øØ H  ¯ ØÆıæø æı r [ ŒÆd] Pæª Æ A ºØ H  ¯ ØÆ[ı]æø ŒÆd ŁÆæ[Œı] F ººø ŒÆ[d

F ]ŒºÆØF Aª"Ææå æØ ı[ ] ˚ıÆE, ˚[æ ]Ø[] ı _ ˚ıøØ Æ, `[挺]B  ¯ [Ø]ªı ƺÆ æØ, øØ [˜]ÆŒı º[Æ]æB, ¯høæ ˜æŒ[Æ] ¸ÆŒ[ÆØ]_ Ø· —æ ˜Æ æØ ªÆŒºÆ æª[E]. vacat _

Translation:

On the fourth of the month Apellaios, at the authoritative (assembly), in the time when Andromenes, Lakris, and Epikrates of (the deme vel sim.) Selegeis were katalogoi of the boulē, it was decided by the boulē and dēmos of the Epidaurians that the following men were to be proxenoi and euergetai of the polis of the Epidaurians and theōrodokoi of Apollo and Asclepius: Agesarchos, son of Aristomenes of Kyme; Kratippos, son of Sosos, of Kydonia; Androkles, son of Epigenes, of Phalasarna; Sosibios, son of Damonikos, of Alexandria; Eudoros, son of Derkidas, of Lakedaimon. To be a proxenos: Demetrios, son of Megakleidas, of Argos.

Appendix

311

Table 6. The proxeny lists of Epidauros Year/ Decree

Type of appointment Number Ethnics (‘proxenos and theōrodokos’ appointed versus ‘proxenos’)

1. Perlman (2000) E. 3 (c.260–240 bc) 1/I proxenos 1 1/II theōrodokos 1 proxenos 2 2/III proxenos and theōrodokos 5 proxenos proxenos and theōrodokos proxenos and theōrodokos

1 2 5

proxenos

5

13/XVI

proxenos and theōrodokos proxenos and theōrodokos proxenos proxenos proxenos proxenos proxenos and theōrodokos proxenos and theōrodokos proxenos proxenos and theōrodokos proxenos and theōrodokos proxenos proxenos and theōrodokos

2 2 1 1 1 3 2 1 1 2 2 1 5

13/XVII

proxenos and theōrodokos

6

2/IV 3/V

4/VI 5/VII 6/VIII 6/IX 7/X 8/XI 9/XII 10/XIII 11/XIV 12/XV

14/XVIII proxenos

5

13/XIX 15/XX

3 1

proxenos proxenos

? [ªÆº]º Æ _ __ Ø [ Æ]; ŁÅÆE ˚ıÆE; ˚ıøØ Æ; ƺÆ æØ; º[Æ]æB; ¸ÆŒ[ÆØ]Ø æªE `NªØB; `NªØB BæøÆE; ª Æ; (`NªØB;)38 ºÆæB; ÆıåæØ æªE ; æªE ; `NªØ ; ˚Ø { }; ˚Ø ¸ÆŒÆØØ; æªE ªÆº[]º Æ ;  ˙æÆ ˚æ" ˚Ø ªÆºº Æ  OæåØ ;  ˙æÆ ; ˚ƺÆıæ Æ ºE; ˚ºØ æØ  Æ —ººÆB [- -]E; æªE ? [˚]HØ —ººÆB; [—æªÆ]Å;  Ø; [ æª]E ˜æØ[B] ([—æªÆ]Å); [¸]ÆŒÆØØ; (  Ø; —ººÆB); ºØ Ø; (˜æØB)39 ÆŒØ; ªÆºº [ÆØ]; ªÆºº ÆØ; ˚ıÆæØØFØ (2) ¸ÆŒÆØØ; ª Æ;  Æ ?Ø Æ?/?ı Æ? continued

38 For some reason this individual, ª Ø  ¯ æÆı, is listed in two different years (here and in 6.1.2/IV), an error of the stonemason perhaps. 39 Those ethnics enclosed in round brackets represent individuals repeated from the previous decree—if a mason’s error rather than a product of bad drafting, it is a remarkable one, especially as the repeated individuals are ordered differently and interspersed with those inscribed for the first time.

312

Appendix

Table 6. Continued Year/ Decree

Type of appointment Number Ethnics (‘proxenos and theōrodokos’ appointed versus ‘proxenos’)

2. Perlman (2000), E. 4 (c.240–200 bc) 1/I theōrodokos c.9

2/II

proxenos

c.9–18?

theōrodokos ?

c.10?

3/III ? ? 3. Perlman (2000), E. 5 (c.220–200 bc) 1/I ? 7+ 2(?)/II

theōrodokos

10

3/III

proxenos theōrodokos proxenos

3 5 6

4/IV

theōrodokos

4

5/V 6/VI

proxenos proxenos and theōrodokos proxenos and theōrodokos

1 1 20

proxenos

7

4. Perlman (2000), E. 6 (350–275(?) bc) 1/I theōrodokos 11 proxenos 17

[ ]æªE[ ]; ¨ÅæÆE ; `N øº ; [- -]Ø ; [- -]Ø ; ¸ÆŒÆØØ ; ¸ÆŒÆØØ[ ]; [- -]Ø  Ø; ¸ÆŒÆØ[Ø]; [˜ı]ÆE;  ˙æÆ[ØÆ]; `NªØ[ Æ] _ ? [ ]æªE ; æªE ; æª[E] ; æªE ; ªÆæ ; ª Æ ; ˚æc ˚Ø ˚HØ ; ˚HØ ; ˚HØ ; ØŒ ; ˚æc ¸ÆÆE ; ˚ÆçıE ; ª Æ ; _ ˚æc [—ºı]ææ"Ø[ ]; ˚æc ˚ıøØ Æ ; ˚æc ˚ıøØ Æ ºØ Ø ; []ºØ Ø ; æªE ˚æØŁØ ;  Ø ; [- -]  ıç ºØ ;  [Ø] ; ªÆæ ; ˚æ[]ŁØ ;  æÆE ; [`] [æÆ]E[ ] ª Æ ; ª Æ ; ˚ƺåÆØ ; ª Æ Æ Ø ªÆºº Æ ¸ÆŒÆØØ ; æªE ; ¨ÅÆE ; ªÆæ ; æªE ; ÆØæ[ø] ; ÆØæø ; ¨Ø ; ¨Ø ; ¨Ø (2); ÆƪæÆE ; ÆƪæÆE ; ˚Ø ; ˜ºç ; ˜ºç ; ŁÅÆE ; `NªØ Æ ; `NªØ Æ ; æ Ø —ºÆ ÆØ ; [ ]Æ Ø ; `N øº KŒ ˚ƺºı; ªÆæ ; ¸ÆŒÆØØ ; ªÆæ ; ˚æB ? ˚[æ]ŁØ[ ]; [- -]Ø ; K[ . . . . .]Ø[- -]; º[Ø Ø] ; [º]Ø [Ø ]; ºØ [Ø ]; [ ˙æ]ÆØ ; ˚[æ][ŁØ] ; Æ Ø[ ]; æ[ªE] _

Appendix

313

Map 10. The proxeny lists of Epidauros, no. 6.1–4 (late fourth (?) to late third century bc)

7. Eresos 7. A catalogue of proxenoi (c.230–200 bc) ed. pr. Hiller IG XII suppl. 127 (with the help of a copy by Peek and notes by Klaffenbach); Hodot (1976), pp. 60–5 with pl. 12; Mack (2012) with table of proxenoi as well as text and facing ph. of the Berlin squeezes at pp. 228–9 (reproduced below); cf. Robert (1966), 115–21 (with excellent emendations for many of the personal names). For a photograph of squeezes taken from this inscription, see Figure A.18. See also Map 3 (Chapter 3, Section 3.3).

A fragment of a large stele, broken on three sides (top, bottom, and left hand side), inscribed on three surviving faces: A (0.27m high; 0.25m wide); B (0.365m high; 0.18m wide); C (0.365m high; 0.305m wide). It was initially inscribed in letters of c.11mm in height as a catalogue of all extant proxenoi who were grouped according to their polis (with their polis ethnic used as a heading in the nominative plural), with these poleis groups, in turn, organized by region (inscription phase (a)). Subsequently appointed proxenoi were then inscribed in two phases: phase (b), during which proxenoi were inscribed as they were

314

Appendix

appointed in the spaces left below the polis group to which they belonged or, when this space was exhausted, in spaces below other groups with the addition of their polis ethnic, again used as a heading; finally, in phase (c), after these spaces had been exhausted, proxenoi were inscribed in much smaller letters wherever there was room, including in the space between lines, without regard to the earlier geographic order, and with their ethnics following their names. The way in which the proxenoi inscribed on Face A (evidently the front face) are spaced suggests that this catalogue of proxenoi was prefixed with an inscribed decree (as in the case of texts nos. 3, 9, and 10.1) presumably setting out the privileges they were to receive and/or the procedure for the inscription of new proxenoi. Directly below this, the catalogue began by listing proxenoi recognized by the Eresians at other cities on Lesbos (including Mytilene), continuing, on the reverse of the stone (Face C), with the proxenoi at other cities on Lesbos (including an unknown polis, probably Methymna, and then Pyrrha). Other polis groups of proxenoi then followed: those on nearby islands and Asia Minor (to judge from the group of proxenoi attested at Miletos by the second column of Face C). Proxenoi in other, more distant regions were apparently then inscribed on the remaining space, including groups of proxenoi in central Greece (Face B). On the basis of this reconstruction, it is possible to give a rough estimate of the proxeny network as a whole, given the density of proxenoi inscribed on its extant faces (see Mack 2012). A fairly conservative estimate for the original height of this stele of one metre would yield a total for the number of proxenoi inscribed on it in all three phases of 350 proxenoi, with c.175 belonging to the first phase of inscription, i.e. the initial network. Given the layout of this text, approximately two-fifths of these (c.70 proxenoi) are likely to have come from the other poleis on Lesbos. For the dating of this monument (phase (a) to c.230 bc and phase (c) to c.200 bc), see Mack (2011). For the relative geographic distribution of this network, see Map 3 (Chapter 3, Section 3.3). Face B 22 This name can be securely restored on the basis of the attestation of this individual, from Aigina, in the contemporary lists of proxenoi and theōrodokoi from Epidauros (text 6.3.6/VI = Perlman (2000), E.5 l.036). 32 The ethnic Ø (Hiller, invoking, erroneously, Dion on Euboia for which the proper ethnic was ˜Øı ) can be corrected to Ø because this individual is known from contemporary Rhodian documents 28 ¯hıº : given (Tit.Cam. 41; SEG 41 663, a statue in his honour). the absence of an ethnic, this proxenos must be taken as a Karystian, like the brothers [˛ ]ŁØ and —º [æÆ]  (ll.31 and 32) and []ÆØæÆ , appointed after the main catalogue was inscribed and added in the spaces which had been left. 38 ¸ #[-]: omitted Hiller, Robert, Hodot; likely, _ ethnic, to have also been of Selymbria. given the absence of space for an

c

bc

316

Appendix

Face C 41–50 This column of proxenoi (and the five others added to this list subsequently, ll. 51–2 and 54–5) were identified on the basis of onomastics as from Assos (see note on l. 56); a polis on Lesbos (either Antissa or Methymna, as the groups for the other two poleis are preserved) is even more likely on these grounds, and necessary given the order (see Mack 2012, 218–19). 41 [ ¯ ]ºçÆ[ -]: the only attested male name which can fit the initial lacuna; the _ attestation of a general from Methymna with this as a patronym (IG XII suppl. 114 1. 3), makes the identification of this as the likely polis of this group of proxenoi very likely. 56 ¸ı: an abbreviation of ¸ıØÆå (Hodot). æØÆØ : the patronymic of ºÆªå[æ ] (l. 55). `[-]: interpreted by Robert as the ethnic of the proxenoi in ll. 51–2 and 54–5, restored by him as @[ØØ], but it is instead the name of a proxenos (or an ethnic heading) in column II. Table 7. The proxeny catalogue of Eresos Inscription Phase

Total Number of Proxenoi

Regions/Poleis Attested

7.1: Phase (a)

33

Lesbos (18): Mytilene (4+); Pyrrha (7+); [Methymna] (7+) Saronic Gulf/Euboia (7): Aigina (3); Karystos (2); Megara (2) Elsewhere (3): Chersonesos (1+); Miletos (2+)

7.2: Phase (b)

20

Lesbos (7): Mytilene (1+); [Methymna] (6) Saronic Gulf/Euboia (5): Aigina (1); Karystos (3); Athens (1) Elsewhere (3): Byzantium (1); Selymbria (2)

7.3: Phase (c)

18

Lesbos (4): Mytilene (3); Methymna (1) Saronic Gulf/Euboia (1): Karystos (1) Elsewhere (7): Rhodes (1); Rome (2); Lysimacheia (2) Hymesseis (1); Chios (1)

8. Histiaia (Oreos) 8. A list recording the grant of proxenia to thirty-two recipients (264/3 bc) ed. pr. Couve BCH 15 (1891), 412–15; Wilhelm AEM 15 (1892), 111–22; (Michel 65); Ziebarth IG XII 9 1187 and Syll.3 492; cf. IG XII suppl. p. 198; Robert (1951); Swinnen (1970); Gschnitzer (1973), cols. 695–6; Marek (1977); Knoepfler (1995), 157 (on the dating of the hieromnēmones). Dimensions: 1.1m high; 0.47–0.5m wide; 0.23m thick See Map 1 (Chapter 3, Section 3.1.1).

Stele of white marble with a pediment, broken at bottom. The form of this text, with the initial dating and decretic formulae followed by a list of recipients of proxenia in the dative, is closely paralleled by the contemporary Aitolian texts

Appendix

317

(the Aitolian proxeny grants to two Histiaians in c.271/0 bc, suggests a connection). Dittenberger was inclined to doubt whether so many proxenoi could really have been appointed in the same year, but could not deny that this is what the text itself asserts,40 and evidence of similar rates of granting from other poleis in particular years places the matter beyond doubt (see Chapter 3, Section 3.2.2). On this inscription in general (including its date), the reconstruction of its internal structuring principles, and the identification of proxenoi appointed as groups within it, see Chapter 3, Section 3.1.1. For the geographic distribution of this network, see Map 1.

5

10

15

20

25

30

IªÆŁBØ åÅØ. K’ Iæå ø  ø KŒ —æø ı, غ æ ı ¯NºıØø , ¯PçæÆ ı ˜Øø , æØ ı K @ø ºç[ı], غ Ææı K % æ ı, ¸Œø ¯NæØø , ƒæŁ ı ØÅØŁı, E øŒ › B æÆ ÆP E ŒÆd KŒªØ ŒÆ a e ·  ”ÆØ ˜æøºı  ¯ åØÆøØ ÆÆŒØ ØÆÆ  ¯ åØÆøØ  (æºÆØ % æÆ  ¯ åØÆøØ ¨ıæøØ ˜Øıı ØøøØ غ [ø]Ø Åغı ŁÅÆøØ ¯PŁıŒæ[ ]øØ ¯PŁıŒæ ı ŁÅÆøØ ¸ÆŒº[Æ]Ø @åø øŒE æåøØ ¯Pı ıæÆŒøØ æ ØæøØ ºÆ Æź ÅØ ŒºÅ[]Ø ÅØ  ˙æ ı ÆøØ ˜ØçºøØ —ºıæı øØ ¸ Ø —Æ ƺ  ÆæÆ øØ غå æØ `P Œºı ˚ıæÅÆøØ ¯PæÆØ  æÆ Œı `N øºHØ ˝ØŒçH Ø æØ º ı ¸ŒæHØ ¨ıæøØ ˜Æı  ˙æÆŒº ÅØ —ÆıØ åøØ —[æ]ø ı AºØŒÆæÆE ¸ıŒçæØ ˚Øı  ˙æÆŒº ÅØ  ˙æÆŒº øØ ŒºÅØ ı AºØŒÆæÆE ªÆŁ æåøØ ¯Pç ı ˚ı ØØE ˜æŒÆØ ¯Påæı  ¯ åØÆøØ æØ ºøØ —Øغ ı ˚ƺåÅøØ  ÆØ ø ÆŒØ K `Nªø æØ ºøØ —æÆı ººæøØ  ˙æÆŒºı ˚ıÇØŒÅHØ [˚]ƺºÆØ  EæÆçºı øØ [ ª]ÆŁ æåøØ ˜ÅŒæ ı åÆØHØ K `NªæÆ

40

Dittenberger ap. Ziebarth Syll.3 492.

318

Appendix

35 [% ]Ø ˜ÅŒºı  ¯ æıŁæÆøØ [. .c.6. .]øØ ª åı åÆØHØ Kª ¸ÆæÅ [. .c.7 . . . ]øØ æå"ı ÆŒØ KŒ ¨

ƺŒÅ[ ] [ . . . c.9 . . . ]ø ÆŒØ KŒ ¨

ƺŒÅ [ . . . .c.12 . . . .] ı ÆŒØ. _

Translation (l.1–7): With good fortune, when the archons were Phyton from Proteion, Philostratos of Eilymnion, Euphrantides of Dion, Aristomenes from Anolophos, and Philetairos from Eiristos, Lykon from Eirieis, and Timesitheos was hierothytē, the dēmos granted proxenia to the following, both to themselves and their offspring in accordance with the law . . .

9. Kalchedon 9. A catalogue or chronological list (?) of proxenoi (Hellenistic?) ed. pr. Kubitschek (1894), 63–4 (on the basis of a drawing originally made by Kemplen, an Austrian diplomat who saw and copied the stone at the site of Kalchedon in 1741); (Michel 540); Merkelbach I.Kalch 4 (with a reproduction of the manuscript drawing). cf. Hanell (1934), 151 n. 6; Wilhelm (1942), I 644 (discussion).

This inscription is only known from the copy made by Kempel of which the extant version seems to be a subsequent copy (the manuscript collection of inscriptions, n. 8640 in the Austrian National Library, is in three different hands and corrected by a fourth, see Kubitschek, pp. 47–8). It is evident that the stone was at that time broken at the bottom (where the catalogue of proxenoi should have continued). From the restorations it appears that the left and right hand sides were complete and the top may also have been (nothing is required by the sense, though the lack of any preamble or heading seems a bit odd). This inscription begins, like the catalogue from Karthaia, by defining the privileges which proxenoi received by virtue of this honorific status. Unlike the Karthaian inscription, this text also makes explicit provision for the inscription of proxenoi, specifying the information to be given (name, patronym, and polis ethnic). There are two ways of understanding this clause— as ordering the inscription of a catalogue of all proxenoi currently recognized or establishing that individuals named proxenoi in the future are to be inscribed on this stele. The context of this clause and the way it is framed mean that the latter is almost certainly correct. In the former case, this clause would simply function as a publication clause, ordering the inscription of the catalogue, but in this case we would also expect to find specific instructions given for the inscription of the text of the decree which precedes it.

Appendix

319

These provisions relating to the privileges to be given to proxenoi and the inscription of their names are probably just excerpts from a longer decree, chosen for inscription because they continued to have a significant function to play, and that is why it begins with the dative E æØ which serves also as a heading. In the same way, in the decree prefixing the Karthaian catalogue of proxenoi (no. 10.1), there is no reference to the order to inscribe this monument (this was clearly implied by the fact that it was inscribed), but merely of the privileges to be granted, because these continued to be valid and important. There is a direct parallel for this concern with ensuring the inscription of future proxenoi, expressed in very elaborate provisions in the decree prefixing the Astypalaian proxenoi (see no. 3). It is not even clear that what followed was a catalogue of current proxenoi. The use of singular ethnics rather than plural headings (attested in the Karthaia catalogue but not elsewhere) and the juxtaposition of the two which are preserved— [ OºØÆe H æe ] ὝÆØ and —ÆæØÆ[e ]—which does not seem to support any suggestion of regional organization, makes it possible that what followed was a chronological list of proxenoi subsequently appointed. Since no prosopographical connections can be drawn and we cannot even judge the lettering of this text, its dating is highly problematic. It is most likely that it was produced in the Hellenistic period, if only because that is when public epigraphy at Kalchedon is better attested. [ E æ]Ø o {o } ŒÆ "Å ÆØ [› A][ ]ŁÆØ ºØ Æ ŒÆd N[ ][ºØ]Æ[ ŒÆd] Yºı ŒÆd Œºı Œ[Æd ][ºı ŒÆ]d æ Æ ıºd ŒÆd I[]· 5 [ŒÆd ]e[ ƒæ] [Æ] f æı[ ] [ f ªØ]ı Iƪæ [ç]Ø []N

[ºÆ] [ºØŁÆ]   ZÆ ÆP [F ŒÆd] F [Æ][ æ]e [ŒÆd A] ºØ[ · e b I] [ºøÆ Ø][ f NŒ][ı ] Œa[ e] []. _ 10 æ[]Ø· [ – – – – – – – – – – – ]ŒØ[ – – – ] [ – – – – – – – ]Œ[ – – ]ÆıØ[ – – ][ – – ] _ [ OºØÆe H æe ] ὝÆØ [ – – – – – – ººø(?)]ı —ÆæØÆ[e ] 15 [ – – – – – – – – – – – – – ]Ø [ ] –––––

1 Kubitschek; Kempel copied ]ΟΕΕΝΟΙΣ (probably his mistake rather than an error on the stone; see similarly ll. 3 and 5). 3 Kubitschek; Kempel copied ΣΙΣΠΛΟΥΝ. 5 Kempel: ΠΡΟΣΕΝΟΥ. 6 [ f ªØ]ı : Kubitschek; Michel and Merkelbach: [ f ª]ı . 7–8 Wilhelm’s restorations, reported by Kubitschek. 8–9 Wilhelm (ap. Kubitschek) proposed  b  ]Æ[F Æ I |ºøÆ Ø ø] ƒ [ ÆÆØ]; the present restoration is Hanell’s. 10 Wilhelm’s supplement (ap. Kubitschek). 13 Restoration Kubitschek.

320

Appendix

Translation: For the proxenoi whom the demos appoints, citizenship is to be given and isoteleia and the right to sail to and from (Chalchedon) in war and peace without being liable to seizure and without the need for a treaty; and the hieromnēmōn is to write up on a stone stele those who become proxenoi—their name, their father’s name, and their polis—and the oikonomoi are to pay out the cost (of inscription) according to the law. Proxenoi . . .

10. Karthaia 10.1. The proxeny catalogue of Karthaia (shortly after 363/2 bc) ed. pr. Koehler (1884); Hiller IG XII 5 542; Mendoni (2007a), 44–5 (with a small photograph of Face A); Mack (2011), 320–33 (the text followed here; larger photos of the front face of the stone and of photos of the Berlin squeezes of both faces A and B). Cf. Pridik (1892), 74–8, 153 n. 10; A. R. D. Sutton in IG XII Suppl. p. 113 (restorations); Marek (1984), 135 (inaccurately identifying this as a chronological list of proxenoi). On ll. 5–13 see now Knoepfler BE (2013), n. 200 See Map 5 (Chapter 3, Section 3.4) and Figures A.19–A.20.

The left-hand side of a stele of white marble topped by a pediment and broken at the bottom, 0.87 m in height, between 0.175 and 0.192 m in width, and 0.075 m in depth. Brought from Keos to Attica and held first in the Marathon Museum and now in the Epigraphic Museum (EM 13491). Face A preserves a decree setting out the privileges granted to proxenoi by virtue of this honorific status below which a catalogue was inscribed at the same time of all current proxenoi, who were grouped by polis and organized by region (they were listed in the form name, patronym, and ethnic, each element separated by a two-dot interpunct). The preserved height of the stele would originally have held c.86 proxenoi. On the right-hand face of this stone, Face B, and perhaps below the catalogue on Face A, the names of proxenoi named after the erection of this catalogue were inscribed in different hands, apparently as they were appointed. I reproduce the text of the decree and the first two regional groupings (Euboia and the Eastern Peloponnese) to illustrate the text. Context and dating: the catalogue was inscribed after, but probably not long after, 363/2 bc. In 364, the Keians, as a federal community, attempted to secede from the Athenian alliance under the influence of the fleet that Epameinondas launched in that year to challenge Athenian dominance at sea. The revolt was quashed by the Athenian general, Chabrias, who imposed a settlement on the island which involved the dissolution of the Keian federation into its constituent poleis. Further trouble erupted the following year at Ioulis (which had been the centre of the federation), and Athenian control was reimposed by Aristophon, son of Aristophanes, who praised the Karthaians in a decree he proposed, which documents these events (RO 39

Appendix

321

ll.54–5) and is recorded in this list as having received proxenia from them at the same time as up to six other Athenians.41 The table below gives the size and ethnics preserved for each regional grouping of proxenoi. Plural ethnics indicate that more than one proxenos was appointed at a particular moment in time. In the case of the proxenoi identified at Athens, I give their names instead. For discussion of these texts, see Chapter 3, Section 3.4 and Maps 5 and 6. For the detail of the reconstruction of this monument, see Mack (2011). Face A [¨ ¯ ] ˇ %. [—  ˇ ˛ ]¯ ˝ ˇ %. _ [ BØ ıºBØ ŒÆd HØ "]øØ·  ̑ ÆØ E æØ [Y][ºı ŒÆd Œºı Iıºd ŒÆd I]d ŒÆd K Næ["]ÅØ Œ[Æd K] __ 5 [ºøØ, ŒÆd .c.7–8 ., ŒÆd æ] æe c [ıºc] [ŒÆd e] [B æ Ø  a a ƒæ . c.5–7 .]ø : ˚çƺø[ . . . . . . . .] [ . . . . . . . . . . . c.25–30 . . . . . . . . . . . . .]Ø : ˆ ŁØ [ . . . . . . . . . .] [ . . . . . . . . . . . c.25–30 . . . . . . . . . . . . .]Æ :  ¯ æ æØ : ˆ.ª[ . . . ] [ . . . . . . . . . . . c.25–30 . . . . . . . . . . . . .]æø : æØ [ç]H[  ] _ 10 [ . . . . . . . . . . . c.25–30 . . . . . . . . . . . ¨]æÆıÆ : ƺŒ[Ø . . . .] [ . . . . . . . . . . . c.25–30 . . . . . . . . . . . ] : ˚ƺºØŒæÆ Ø[ . . . . . . . .] [ . . . . . . . . . . . c.23–28 . . . . . . . . . . ˛]çH  [ . . . . . . . . . . .] [ . . . . . . . . . . . c.25–30 . . . . . . . . . . . .]Ø : Ø[ . . . .][ . . . . . ] [ . . . . . . . . . . . c.25–30 . . . . . . . . . . . ]ı : ˚æŁØ [ . . . .]Œ[ . . . ] 15 [ . . . . . . . . . . . c.25–30 . . . . . . . . . . .]å : ¯P æå[ı :  ¯ ]ØÆ[][æØ . . . . . . . . . c.20–25 . . . . . .  ¯ ]ØÆæØ : [.]%¸[ . . . .]([. .] [ . . . . . . . . . . . c.23–28 . . . . . . . . . . ¯ ]ØŒæı : [ . . . . . . . . . . .] [ . . . . . . . . . . . c.21–26 . . . . . . . . . ¸ÆŒ]ÆØØ[ . . . . . . . . . ] [ . . . . . . . . . . . c.25–30 . . . . . . . . . . .] : —æç[  ı . . . . . . . . . .] 20 [ . . . . . . . . . . . c.25–30 . . . . . . . . . . .] : —ºº[Æ]f : K[Œ B ¸ÆŒø][ØŒB . . . . . . . c.16–21 . . . . . ˚ıçÆ] Æ[ : B]æø[ . . . . .] [ . . . . . . . . . . . c.21–26 . . . . . . . . ¯ Ø]ÆæØ : KŒ [B] ¸ÆŒ[øØ][ŒB . . . . . . . . c.22–27 . . . . . . . .] —Łø :  ˙æÆŒºı[ . . . ] [ . . . . . . . . . . . c.25–30 . . . . . . . .]. : `NªØ" Å : etc.

3 Sutton proposes [ ˚ø BØ ıºBØ ŒÆd HØ "]øØ making this a catalogue of federal Keos—extremely unlikely on the basis of phraseology and given the historical context of this monument, see Mack (2011), 325. The discrepancy in the number of letters restored here (24 in comparison with the 28 which appear normal elsewhere) is probably not an issue. An alternative, longer restoration [nomen (c.11 letters) r   HØ "]øØ, is possible, but we would expect a decree like this to be made with probouleusis. 3–4 Hiller restores [I |ºØÆ   ø ŒÆd KºØ ŒÆd KŒº]Ø ŒÆd K Næ["]ÅØ Œ[Æd K| ºøØ]; the reference to peace and war implies a personal

41

On these events, see Mack (2011).

322

Appendix

grant of asylia instead (which the surviving letter traces fit better). 5 Hiller restores [ . . . ŒÆd ªB ªŒ ÅØ . . . ] to fill out the restoration in this line, a very unlikely formulation. Sutton proposes [ . . . Iıºd ŒÆd Id . . . ] which would be a peculiar qualification for a grant of ateleia (which is to be omitted anyway, see 3–4). As these restorations are probably overlong, I propose instead restoring a single shorter honour, probably either [ . . . ŒÆd ææÆ . . . ] or, less likely, [ . . . ŒÆd I ºØÆ . . . ]. 6 Hiller, albeit very doubtfully, restored the first proxenos as [ . . . ˚ÅçØç]H ˚çƺø[ ŁÅÆE ], which would be completely contrary to the ordering principles of this inscription. 10 Keil proposed ƺŒ[ÅØ ]; given the structuring principals of this text Koehler’s restoration, followed here, is to be preferred in this section on proxenoi from Euboea. 23 Koehler restored —Łø:  ˙æÆŒºı[ `Y|Ø ] because he saw in the juxtaposition of these two names a connection with the brothers Python and Herakleides of Ainos who slew the Thracian king Kotys, and restored their ethnic accordingly. Both names are, however, far too common for this connection to be plausible, especially as it is excluded by the regional ordering of this text. Translation: Gods. Proxenoi. Decided by the boulē and dēmos, proxenoi are to be able to sail from and to (Karthaia) during peace and war, without being liable to seizure and without the need for a peace treaty, and they are to have proedria (or ateleia) and the right of access to the dēmos and the boulē after religious matters . . .

Table 8. The proxeny catalogue of Karthaia (mid-fourth century bc) Region

Number of Proxenoi

Surviving Ethnics (in the case of Athens, names)

10.1.1 Euboia (l.1–10) 10.1.2 The Eastern Peloponnese (l.11–28)

10 18

10.1.3 Central Greece (l.24–35)

17

 ¯ æ æØ (l. 8); ƺŒ[Ø ] (l. 10) ˚æŁØ (l. 14); [ ¯ ]ØÆ[æØ ] (l. 15); [ ¯ ]ØÆæØ (l. 16); [¸ÆŒ]ÆØØ[ ] (l. 18); —ºº[Æ]f K[Œ B ¸ÆŒøØŒB ] (l. 20–1) [˚ıçÆ] Æ[ ] (l. 21); [ ¯ Ø]ÆæØ : KŒ

[B] ¸ÆŒ[øØŒB ] (l. 22); `NªØ" Å (l. 24) ¸Æ[ ] (= Lebedeia; l. 25); ˚Ø (l. 26);42 [¨]ÅÆE (l. 27); [Å]º[Ø ] (l. 30);

42 This individual is the only one who violates the otherwise strict regional ordering of this text. The most plausible explanation of why this individual from Knidos in southern Asia Minor was listed in the region devoted to Boiotia is that he was appointed in the context of Epameinondas’ flotilla (for which Knidian support is well attested in the form of a proxeny decree for Epameinondas; SEG 44 901) and placed here because there were no other proxenoi in his region along with whom he could be better placed (cf. Koehler (1884), 276 suggesting that Knidian might be a Boiotian metic). On this, see Mack (2011), 328.

Appendix

10.1.4 Athens (ll. 35–44) 15

10.1.5 Cyclades (ll. 44–9) 10 10.1.6 The Hellespont (ll. 50–60) 10.1.7 Subsequently added proxenoi (Face B ll. 1–25)

16+

323

[¸Æ]æØØÆE : KŒ ŁØ[ Ø ] (l. 32); [K  ˙æ]ÆŒºÆ : B [æÆåØÆ ] (l. 33) [¨Ç] Å : ˝ØŒ[ æ ı] (l. 35);43  Iæı[ -] (l. 36); [˜Å]Œæ Å : [ı] (l. 39);44 [ÆæÆ (?): ˚ Åı(?)] (ll. 40–1);45 [-Ø] :  I[Œ]æ [ ı ]; [˝ØŒÅ ]: ¯PŒ Æı; [ç]Æ[æ]f[ ]: % Œæ ı (?)] æØ çH: æØ ç [ı ŁÅÆEØ] (ll. 41–4)46 [-] æØ (l. 44); "Ø (l. 45); æØØ (l. 46); []æçØ (l. 48); ˜"ºØ (l. 49) []Ø (l. 50); [Ø ] (l. 51);47 [Ø ] (ll. 51–2); [][]Ø (l. 52); ˚ıÇ،Š(l. 54); [˚ı]Ç،Š(ll. 55); —æŒ["Ø ] (ll. 57–8); ˚Ø[Æ] (l. 58) [ Ææ]ø Å (l. 2); @æØ (ll. 6–7); `NªØ" Å _ _ (ll. 21); [`N]ªØ" Å[ ] (l. 24)48 _ _

43 On this individual’s politically active grandfather and service as choregos before 348 bc, see Dem. 20.59 and Davies APF 6915. 44 Prytanis of the tribe Oineis in 360/59 bc (II2 1745 ll. 54) and father of a liturgist attested c.330 bc (II3 550 l. 23 = II2 417). 45 The identification of this individual as the Athenian general Chabrias is not as hopelessly optimistic as it looks, given the fact that this grant is juxtaposed with a group of Athenians, at least one of whom was sent to Keos as a general the year after Chabrias himself had been. On this, see Mack (2011), 329–30. 46 This group of up to six Athenians were appointed together as the fact that they are not listed with individual ethnics makes clear. On Aristophon, son of Aristophanes (prosecuted unsuccessfully seventy-five times by graphe paranomon), see above. On [˝ØŒÅ ] ¯PŒ Æı (not, as Davies rightly shows, ˝ØŒÆ , an ephebe in 334/3 bc) who was of the liturgical class, see Davies APF nos. 24 and 10814. Hiller’s restoration here, of Aphareus, son of the famous Athenian orator Isocrates, is very dubious. 47 The ethnic in the nominative singular can be restored with confidence here, given the three other secure attestations (one before; two after) that define this group. 48 The juxtaposition of two individuals from Aigina in this subsequent, chronological list, inscribed in the same hand, clearly indicates that they were appointed proxenoi at the same time (and presumably for the same reason) despite the individual ethnics which they are given. For these readings, see Mack (2011), 331–2.

324

Appendix

10.2. A chronological proxeny list (mid-third century bc) Fragment a: ed. pr. Hiller IG XII 5 1073;49 fragment b: IG II2 2455; cf. Mitsos AEph (1950–1), 45–6 (publishing a decree of Karthaia inscribed on the other side of this fragment). Mendoni (1990) argued convincingly that these fragments are likely to have come from the same stele (see also Matthaiou Horos (1990–1), 207–14). See Map 6 (Chapter 3, Section 3.4).

This subsequent, chronological list of proxenoi from Karthaia, attributed to the third century bc (IG XII 5 1073 + IG II2 2455), offers a striking contrast, in terms of distribution, to the proxeny network presented by the fourth-century catalogue. In addition to the ethnics from the list proper, I also give the ethnics of the subjects of two honorific decrees preserved on the reverse of one of these fragments (IG XII 5 1072).50 Table 9. The proxeny list of Karthaia (mid-third century bc) Region

Number of Proxenoi

10.2.1 Euboia 10.2.2 Northern Greece

1 4

10.2.3 Aegean Islands 10.2.4 Asia Minor

1 2

10.2.5 Hellespont

3

10.2.6 Corinthian Gulf 10.2.7 Western Greece and Magna Graecia

1 2

Ethnics (IG XII 5 1073) ƺŒØ (l. 10) [ ÆŒ] (l. 5); [¨]

ƺe KŒ 挌Š(l. 8); _ ¨Æ[ºe KŒ . . . ] (l. 12); [¨

ƺe KŒ ¸Æ]æÅ (IG II2 2455 l. 10) [˚æc ]_ K @[ı] (l. 1) _ (l. 11); ˚[Ø ] (IG XII 5 1072 A  ºØŒÆæÆ[ ] ll. 12–13); [ ]ıÅ (l. 3); [¸Æ]łÆŒÅ (l. 6); ˚ıÇØŒ[Åe ] (IG XII 5 1072 ll. 1–2) `NªÆØf (l. 4) []ıæÆŒØ (l. 7); [ æªE] K çغåø[] (l. 9)

11. Kleitor 11. A catalogue (?) of proxenoi (second half of the third century bc) ed. pr. Milchhöfer MDAI(A) 6 (1881) 303–4; (SGDI 1189); Hiller IG V 2 368 + add. p. 146. cf. Gschnitzer (1973), cols. 697–8; Habicht (1998a), 490–1.

49

Curiously the ed. pr. of this text which Graindor had prepared and to which he alluded in his edition of the text on the other side of this fragment never appeared (Graindor BCH 1906, 101–2). 50 On this fragmentary list, see Mack (2011), 333–7 with photographs on 334.

Appendix

325

A free-standing stele 1.6m in height, originally inscribed on all four faces though only three are even partly legible. The faces of this stone are inscribed with lists of groups of individuals (name and patronym) each headed by a polis-ethnic in the nominative plural. Given this layout, as well as the number and range of different groups the list contains and the fact that during a late stage of the use of this stele it was inscribed with proxeny decrees, it is unlikely that it is not a list of proxenoi. The interpretation of this text is, nonetheless, difficult. The format (proxenoi grouped together according to the polis from which they came) would suggest that it was a catalogue of proxenoi. However, the fact that multiple groups of proxenoi from the same poleis were inscribed on the stele makes it look more like this was some sort of chronological list. Certainly there were multiple phases of inscription (could they have involved deletion, as at Aitolia?), but the badly worn surface does not allow us to unpick and understand these different layers as we can for other lists (although the abbreviated decrees inscribed on the right-hand face, ll. 146–77, surely represent the last phase of inscription). The best we can do is to examine the distribution of these proxenoi collectively. In 221/0 bc Mantinea became Antigoneia; the occurrence of two groups of Mantineans on the left-hand face indicates that this phase of inscription was completed prior to this point; the appointment of an individual from Antigoneia on the right-hand face (l. 69), in one of the abbreviated decrees, places this later phase of the inscription after this date. Table 10. The proxeny list of Kleitor Region

Number of proxenoi

11.1 Peloponnese

59

11.2 Western Greece 11.3 Central Greece 11.4 Aegean 11.5 Black Sea and Propontis 11.6 Sicily

9 14 11 3 2

Polis ethnics [ªA ÆØ]51 (3+);  ØØ (3?); ºEØ (4); —Æ æB (2); []ªA ÆØ (4); Æ Ø (1);  ØØ (5); —ººÆE (21); Æ ØB (5); ºEØ (3);  O ØØ (2); [ºØ ?]ØØ (3);  ˙æÆF[Ø] (2); [ ] تB (1); ØŒı[Ø] (1) _ (1); [ ]æÆŒØH ÆØ (8) [ ]ºıÇÆEØ _ ƺŒØE (3); [ Ł]ÅÆEØ (10); [Ł]ÅÆE (1); ˚æB  (4); ˚æB  (7); ØøE (1); _˝ØŒ[]Å[f ] (1); __ ˝ØŒÅ[[]]f (1); ÆæÆ EØ (2+)

51 Restored on the basis of a prosopographical identification (IG V add. p. 146; IG V 2 106).

326

Appendix

Map 11. The proxeny catalogue of Kleitor, no. 11 (second half of the third century bc)

12. Lousoi 12. A series of three chronological proxeny lists? (second half of the fifth century bc) ed. pr. Wilhelm JÖAI 4 (1901), 78–80, ph.; Hiller IG V 2 387; (Schwyzer 669; Jacobsthal Diskoi (1933), 29 no. 1 with excellent ph., p. 28; Guarducci Epigrafia Graeca I, 122–3, n. 3; Dubois (1986), 215–17). cf. Gschnitzer (1973), cols. 695, 701.

A bronze disk from the second half of the fifth century bc, now in Berlin, 0.19m in diameter and complete with nail holes for display; it is inscribed with a series of proxeny grants and is by far the earliest list we possess. It is inscribed, in different hands, with what appear to be a series of three chronological lists of proxeny grants. Whether they are in fact comprehensive records of the proxenoi appointed within a particular period of time is unclear, but the plural heading, æØ ¸ıØÆ A, suggests that the first of these is a genuine list, collecting three proxenoi not necessarily appointed by the same decree, while the third, which names an individual proxenos, and the second, which may do so, could simply be an abbreviated decree (see

Appendix

327

Gschnitzer). The spacing of the last two entries, with vacats left before, and in the case of the second inscription, after the names of the proxenos, was presumably designed to give them more prominence. The omission of the ethnics of these proxenoi (as well as their patronyms) is striking, especially given the emphasis later placed on them in proxeny inscriptions. It is surely a reflection of the early date of this text (as well as its conciseness), before the inscription of proxeny decrees became widespread and conventions relating to what was inscribed had developed. It cannot, however, be taken as an indication that the poleis of these proxenoi were unimportant to the Lousiatans: the absence of the ethnic, like the absence of the patronyms, merely shows the identity and circumstances of these men was felt to be well known at Lousoi. (I) æØ ¸ıØÆ A· æ Bº Æ 5 ºŒÆ  ÆP d ŒÆd ª . (II) æ ¸ıÆ A· vacat —Æ̑ , æØ vacat 10 ÆP e ŒÆd ª . (III) æ ¸ıÆ A· vacat Bæıå̄ Æ ŒÆd ª .

9 æØ: it is uncertain whether this is the name of a new proxenos or the patronym of —Æ̑ (though as patronyms are otherwise absent, the former should probably be preferred).

13. Narthakion 13. A reverse proxeny catalogue (?) from Narthakion (before or after ? c.140 bc) ed. pr. Latycheff (1882), 580 no. 2; emendations by Lolling MDAI(A) 10 (1885), 284–5; (Michel 657); Kern IG IX 2 90 (+ corrigenda p. ix, n. 90). Cf. Monceaux (1886), 180; Gschnitzer (1973) 699–700; Marek (1984), 136–7 (arguing that this document is more likely to be a catalogue of citizens who served as theōrodokoi). For a photograph, see Figure A.21.

328

Appendix

A fragment of yellow-grey stone, measuring 0.72m by 0.32m and 0.43m thick, which was recut for architectural reuse (this involved reshaping the stone so that it was roughly semi-circular and chiselling away the left-hand side of the inscribed face of the stone to create a new margin). It was found on the site of the church of St John, close to Limogardi in Lamia. It is a list in which the names of citizens of Narthakion follow polis ethnics in the genitive plural, clearly connecting these individuals with particular, foreign poleis. To judge from the restoration in l. 19 (which is secure) and the photo of the stone taken by Y. Béquignon, held at the French School at Athens (reproduced here), the vacat on the left hand side consisted of c.7–8 letters. The date of this text can be established only very approximately in relation to the well-known senatus consultum concerning Narthakion, dated tentatively by Sherk to 140 bc (RDGE 9), which it either pre- or post-dates: Agelaos, son of Glauketas, in l. 11 of this catalogue, is likely to be either the father or the son of Glauketas, son of Agelaos, one of the three archons of Narthakion mentioned in RDGE 9 (the latter may also be attested in the citizen list, IG IX 2 91 ll. 63–4). A number of other prosopographical connections can be drawn with a citizen list from Narthakion—Nikolaos, son of Dikaiarchos, and Philetairos, son of Philonis, occur in both (though they may refer to grandfathers and grandsons rather than the same individuals) whereas [Kle]ainetos, son of Kleostratos, in the catalogue is likely to be either the father or the son of Kleostratos, son of Kleainetos, in the citizen list. This text lists twenty-seven cities with which individual citizens are connected. Two Narthakians are connected with multiple poleis: Menes, son of Nikatas, is connected with three poleis, including Athens (ll. 12–13); Aischinas is connected with two, including Miletos (l. 29). Given the absence of space for an initial ethnic in l. 18, this individual must have been connected with the same polis as Kleainetos in the previous line. For discussion of this text, see Chapter 2, p.114, and Chapter 3, p.181. [— — — — — — —]Ł[— — — — — — —] _ [— — — — — — —] º[— — — — — —] _ [ . . . c. 7–8 . . . ] øÆ[ — — — — —] __ _ [ . . . c. 7–8 . . . ]ı. vacat 5 [. . c. 5–6 . .  ]æÆ ø ˚ºÆ.[— — — —] _ [. . c. 5–6 . . ]ººøÆ º[— —] _ [ . . . c. 7–8 . . . ] ¯hØŒ ¯PŒ[ı — —] _ [ ƪ" ø] H Ie ÆØ æ[ı — — —] _ [ . . . c. 7–8 . . . ]æı. vacat 10 [. . c. 6–7 . . ˝]ØŒºÆ ˜ØŒÆØ æå[ı — — —] _ [ . . . c. 7–8 . . . ] ªºÆ ˆºÆıŒ Æ [— — —] _ _ [. . c. 6–7 . . ø]· Å ˝ØŒ Æ· › ÆP [e —] _ [ . . . c. 7–8 . . . ]ø· › ÆP e ŁÅÆø. [ . . . c. 7–8 . . . ] ˛ŒæØ  ª Łø . 15 [ . . . c. 7–8 . . . ]ø· ¨æÆÆå źÆ ƒ `[— —] _

Appendix

329

[ . . . c. 7–8 . . . ] vacat [. . c. 4–5 . .˚º]Æ  ˚º æ ı. [ . . . c. 7–8 . . . ] ºÆŁÆ. vacat _ [ åÆØH K] `Nªø[Ø]· ØÆ ˝ØŒ æåı. _ 20 [ . . . c. 7–8 . . . ]·  æÆ ØŒ  æÆ º [ı]· _ [ø] º ._ [ . . . c. 7–8 . . . ]ø· ¯hºØ _ [ . . . c. 7–8 . . . ] A· ØŒŒH  Ø æı. [. . c. 5–6 . . . ·  ]ºÆØŒ ŒÆd  Ø ƒ  ت[ ]. [ . . . c. 7–8 . . . ]A· غ ÆØæ غøÆ. 25 [. . c. 6–7 . . ]º ŁØ ¯PıºÆ. _ __ [. . c. 6–7 . . ]æØ Å ¨æÆ. __ __ _ [ . . . c. 7–8 . . . ]ø· æÅ ¨ ı. _ [. . c. 5–6 . . . · ¯P]ı  Ø æı. [ . . . c. 7–8 . . . ] ŒÆd ºÆŁÆ ƒ ˚æÆ ı. 30 [ . . . c. 7–8 . . . ]ø ŒÆd غÅø· `NåÆ Æıæ[Œı?] [ . . . c. 7–8 . . . ]º ˜ÆÆØ ı. _ vacat

3 Latycheff restored [ . . . ø ]H I[e . . . ]. 6 Latycheff restored [˜ºçø· ]ººøÆ º[ÆŁÆ] on the basis of the second-century catalogue of theōrodokoi (then thought to be a catalogue of proxenoi) in which an individual of this name was identified as the theōrodokos in Narthakion.52 At the same time, he restored l. 17 to refer to this same individual (see note ad loc.). Neither of these restorations is warranted. At l. 6 there is no reason to connect Apollonidas (a common name) with the individual attested at Delphi. The occurrence of the same individual in two places would also violate the fundamental ordering principal of this text and, although Altheas is attested only once outside Narthakion, it was evidently a reasonably common name there (it is attested also in l. 28). 8 Latycheff ’s restoration, presumably on the basis of the fact that the Magnesians on the Maiander sent delegates to help arbitrate a boundary dispute between Narthakion and its neighbour, Meliteia (Sherk RDGE 9); [  Øåø] H Ie ÆØ æ[ı] would also fit _ the patronym of the lacuna. 16 vacat: photo, presumably the result of the two brothers in l. 15 running over into this line as there is hardly room for a new name and patronym, even of an individual connected to the same polis as the brothers in l. 15. 17 Latycheff [¯P]Æ  ˚º æ ı, but this restoration is to be preferred on the basis of the occurrence of these names in the citizen list of Narthakion, IG IX 2 91 ll. 51–2. 18 Lolling, [ ººøÆ] ºÆŁÆ, see above on l. 6. 19 Kern, IG IX 2 corrigenda, p. ix, restores [¨ÅÆø] erroneously (he used the name of the individual in this line to restore the name of a magistrate in a text from Phthiotic Thebes, IG IX 2 132). 22 ØŒŒH: photograph. ØŒŒø: Lolling. ŒŒ : Latycheff. __ 52

BCH 45 (1921), 29 V C (b), l. 10; cf. pp. 42–3.

330

Appendix 14. Samothrace

14. 1–13. Thirteen blocks inscribed with sixteen lists of theōroi sent by different poleis and named proxenoi (second to first centuries bc) These thirteen stones preserve parts of at least sixteen lists of proxenoi. Eleven (probably twelve) were blocks of the same height (c.0.35m) and would originally have been part of the same wall, though the identity and location of the monument of which it was part is not known. The other stone inscribed with a proxeny list, no. 14.4, was part of a monument base. Each inscribed list seems to have been introduced by a similar heading: ‘during the year in which X was Basileus, the following became proxenoi, having come as theōroi’. These texts were then broken down into a number of shorter lists of the individuals from particular cities whose names are given beneath their ethnic used as a heading in either the nominative or genitive plural. I reproduce the text of the first of these below as an example.53 The festival to which these theōroi came was probably the Samothracian Dionysia.54 It is striking that the majority of these theōroi tended to come from coastal Asia Minor, but this probably reflects less Samothrace’s local region of primary interaction than the regional dynamics influencing the different communities which thought it worthwhile to send theōroi. In relation to these theōroi it is interesting to note that a number of family groups can be identified, suggesting that a personal interest in visiting Samothrace and its festival could play an important role in the decision of a polis to send a delegation of theōroi.55 Other decrees relating to proxenia inscribed at Samothrace (IG XII 8 151 I–III; 152; 154; 155 I; 157; 159), Kos (IG XII 4 148), and Thasos (I.Thasos 169) respectively make it clear that proxenia was not granted only to theōroi, but was understood in terms of the same proxenos-paradigm, as a practical, intermediary role in relation to a foreign city, as elsewhere (the Koan example, for instance, also makes it clear that individuals who already had a connection with Samothrace in the form of the status of proxenos could subsequently be chosen to act as theōroi by that community). The Samothracian lists of proxenoi are to be placed in the second and first centuries bc

53 For a number of other texts recording individuals from different poleis in a similar format though on stelae and bases rather than the wall-blocks of this monument, see Dimitrova (2008), 51–70; sometimes the individuals in question are referred to as theōroi, but no reference to proxenia is made. 54 See Dimitrova (2008), 72–4. 55 e.g. Dimitrova (2008), nos. 1 ll. 9–10 and 8 ll. 7–8

Appendix

331

without exception,56 but their relative chronology cannot be accurately reconstructed, much less their precise dating. The last datable attestation of the grant of proxenia to a group of theōroi by the Samothracians comes not from Samothrace itself, but an inscription at Kaunos from the early Imperial period (I.Kaunos 28). In the table below I use Dimitrova’s numbering and direct the reader to her for a full critical edition of these texts. Sample text: ed. pr. Fraser (1960) no. 23; Dimitrova (2008) n. 1 (ph.).

Table 11. The proxeny lists of Samothrace Inscription no. (Dimitrova 2008)

Date

Probable number of proxenoi

Polis-groups preserved and their size

14.1 (1)

2nd century bc(?)

10+

14.2 (2)

2nd–1st centuries bc

¸ÆæØÆEØ (4); ˚ıÆEØ (3+); _  ¯ çØØ (3+)

7

14.3 (3)57 14.4.1 (4.1) 14.4.2 (4.2)

mid-2nd century bc 2nd or 1st centuries bc 2nd or 1st centuries bc

6 ? 11+

A  ºØŒÆæÆE (3); ˚ºçØØ (2);  ¯ çØØ (2) —æØÅE (2);  ØØ (4) ?  ¯ çØØ (4); ˚ºÆÇØØ (2); ºÆÆE (1+);  ıƺ[ÆØE ] (3); ˚HØØ (1+) continued

56 Dimitrova (2008), no. 4 (and p. 18), states that her no. 4, the lists inscribed on the monumental base, was attributed on the basis of letter-forms of the first to second century ad by Friedrich, IG XII 8 168. Friedrich’s date, however, ‘litteris a. 100 insculptis’, should in fact be understood as c.100 bc (which is where these plain letterforms belong)—and this is how Dimitrova herself has understood the equivalent phrase ‘litteris a. 100 incisis’ (IG XII suppl. 171) for her no. 6, plausibly preferring in this case a second-century bc date. 57 Inscription known only from a fifteenth-century copy made by Cyriacus of Ancona.

332

Appendix

Table 11. Continued Inscription no. (Dimitrova 2008)

Date

Probable number of proxenoi

Polis-groups preserved and their size

14.5.1 (5.1)

c.150 bc

18+

14.5.2 (5.2)

c.150 bc

31

14.6.1 (6.1) 14.6.2 (6.2)

2nd century bc(?) 2nd century bc(?)

6+? 7+

14.7 (7)

2nd century bc(?)

11+

14.8 (8)

2nd century bc(?)

17+

14.9 (9)

2nd century bc(?)

29

14.10 (10)

2nd–1st centuries bc

25+

14.11 (11)

2nd century bc (?)

11+

14.12 (12) 14.13 (13)

Early 2nd century bc (?) 5+ 2nd century bc (?) 13+

ºÆÆE (2); ˚ıÆEØ (2); ı غÅÆEØ (2); BÆæªıºØB ÆØ (2); ˝ ØØ (2+) ÆæøØ H (2); —æØÅø (2); ˚Æıø (3); ÅæØ H (2); Æø (3); ˚Øø (3);  []ø (3); % Æø (2);  æÆ ØŒø (3); Ææa Æغø

ºı (2); —ÆæØÆH (3);  ¯ çø (3) ıæØÆø (2) Æ[ø] (3); ˚Øø (2);  Å ø (1+) [-]Æø (5); ºÆ[Æø] (2); _ æÆ  _ _ (2+) [ØŒø] _ (2); A [—æØ]ÅE  ºØŒÆæÆE (3); _ ˚ºÆÇØ (2); ¨ ØØ (2); ÆæøE ÆØ (2); ÅæE ÆØ (2) øŒÆØE (2); ˜ÆæÆE (2); `NªÆØE (2); [ıæ]ØÆEØ (2); __ ˚ıÇ،Š(2);_ A  ºØŒÆæÆE _ (3 or 12);  ¯ æØØ (2); (4); EØ ˝ıÆØE (1) ˚ıÇ،Š(2);  ¯ æØØ (2); ˚ºçØØ (2); ˚ıÆEØ (2); "ØØ (2); —æØÅ (1); F ŒØF H [æd e ˜Øı]

åØ H H [Ie % øÆ ] ŒÆd  EººÅ [ı] (2) ˚Øø (2); BÆæªıºØÅ H (1); ˚ºÆÇø (2); F ŒØF

H [æd e ˜Øı]

åØ H [-] (2); ÆæÆ E (1) ıºÆE (2); ˚ıÇØŒÅH (1+) ˚ıÇ،Š(2); A  ºØŒÆæÆE (2); [˚]ÆØØ (3); ºÆÆE (2); ˜ÆæÆE (2)

Appendix

333

Map 12. The proxeny lists of Samothrace, no. 14.1–13 (second to first centuries bc)

15. Tenos 15.1. A fragment of a catalogue of proxenoi(?) (second half of the third century bc) ed. pr. Graindor MB (1910), 45, 23; IG XII suppl. 313. Dimensions: 0.16m (height); 0.14m (width); 0.05m (thickness).

A fragment of a stele of white marble, broken on all four sides. The most probable suggestion (made by Graindor) is that this fragment came from a catalogue of proxenoi organized by region, in a similar format to the catalogue of Karthaia (i.e. with proxenoi listed continuously, with their names and patronyms followed by an ethnic, without the single proxenos per line which we see in other catalogues). If so, we possess part of the section listing Tenos’ proxenoi in Magna Graecia and Sicily, as it contains the ethnics of Tarentum, Kroton, Rhegium, and Syracuse (for other examples of close connections between the Tenians and Greeks in the west, see IG XII 5 816 and 843, Tenian decrees which certainly and, in the latter case, probably, grant proxenia to individuals from Syracuse and Naples respectively).

334

Appendix

[ – – – – ].??.¸[ – – – – – – – – ] _ [ – – – – ]Å ¯PªØ [ – – – – – – ] [ – – – ] ÆæÆ E[Ø – – – – – – ] [ – – – ]ø æØ   [ºı – – – ] __ 5 [ – – – ]º åı, Øå [æÅ –––] _ [ – – – ]ºÅ ˚æ øØA[ ÆØ – – – – ] [ – –   ]ŪEØ, æåº[Ø – – ] [ – – – ]ı , ¸ıŒ"[Å – – – – ] [ – – ¸] Å غŒ[ı – – – ] 10 [ – – – ]ı, ˚º"Å [ – – – – ] [ – – – ?ıæÆŒ][]ØØ [ – – – – – ]

The plural ethnics restored in l. 3 (ÆæÆ E[Ø]) and l. 6 (˚æ øØA[ ÆØ]) could equally be singular.

15.2. A list of fourteen proxenoi appointed at the same time (second half of the third century bc) ed. pr. Graindor MB 14 (1910), 233–6 n. 25; IG XII suppl. 304. Dimensions: 0.38m (height); 0.32m (width); 0.09m (thickness).

White marble, broken at top and bottom. On the plausible context for this grant of proxenia (particularly given the strong concentration of at least nine Cretans in this text)—the third-century asylia campaign of the Tenians—see Chapter 1, pp. 75–6. An additional mark of the exceptional nature of this set of grants is the gift of a crown to all of these proxenoi (see ll. 14–15), which is comparatively rare as an honour granted at the same time as proxenia, and is never likely to have been a standard accompaniment of it. [ – – c.21 letters – – ]Œºı . . . . . . . . . . .. æØ  ˜ÆØŁ..ı ¸

Ø, [ø]_ __ Æ ¸Ø _  ¯ ºıŁæÆE, ˚Æ  Eæ æåı ˚Ø, —[Ø]Ø Å ææı ˆæ [Ø]5 , ººB çØŒºı ¸ÆÆE, º[º]Ø ˝Œø %  æØ, —Ł ŁÅŒø[ ] [ ]ƺºÆE, øÆ ØŒæ[ ] ı ˚ıøØ [ Å], . . . ..Ø .ºø[  I]æÆ Ø, ˝ŒÆØ  . . . . _ [ – – c.20 letters – – ]ªæ[ – – c.10 letters – – ] _ 10 [ – – – – – – c.32–4 letters – – – – – ] [ – – – – – – c.32–4 letters – – – – – ] . . . Ø ˝ØŒı AºØŒÆæÆÆ, ºØ ¸ø.. [.. ]åÆØ, øتÅ øتı ˚Hœ Œ[Æd] [ ç]ÆHÆØ HØ KŒ F ı  ç øØ I[æ]15 [ B ]Œª ŒÆd çغ ØÆ B N c [][ºØ ŒÆ]d IƪæFÆØ ÆP E e  çÆ [ e ¼æå] Æ c  çÆÅçæ Iæåc  []

Appendix

335

[ HØ ƒæHØ F —]ØH ŒÆd B çØ æ Å [ŒÆd K HØ Ł æøØ —Ø]ø(!) ŒÆd ˜Øıø 20 [ HØ IªHØ H æƪøØH]· r ÆØ b ÆP [f] [ŒÆd KŒªı æı ŒÆd P]æª Æ[ ] B[ ] [ºø – – – c. 27 letters – – – – ]

Translation of lines 13–22: . . . and crown them with the crown set out by the law for their excellence and philotimia towards the polis and the magistrate holding the stephanephoric office is to announce the crown for them in the [temple of Pose]idon and Amphitrite and [in the theatre during the tragic contest of Poseid]on and Dionysus and they are to be [along with their offspring proxenoi and eue]rgetai of the [polis . . . ]

Map 13. The chronological proxeny list and proxeny catalogue (?) of Tenos, no. 15.1–2 (second half of the third century bc)

16. Thera 16.1–4. Chronological lists of proxenoi appointed in seven different years (?) (late second and early first centuries bc) These stones preserve a set of chronological lists of proxenoi, each inscribed beneath a heading which states that the following are those inscribed as proxenoi by a particular secretary (grammateus) or pair of secretaries, whose

336

Appendix

Table 12. The proxeny lists of Thera Text

Date

Number of proxenoi

Ethnics surviving

16.1 (IG XII 3 332 + XII 3 suppl. 1297)

125–75 bc

4

[ŁÅÆE(?)](4);58

16.2.1–2 (IG XII 3 333 + XII 3 suppl. 1298)59 16.2.1 (ll .1–5) 125–75 bc 2 16.2.2 (ll. 7–12) 125–75 bc (shortly 2 after 16.2) 16.3 (IG XII 3 334) 125–75 bc 3+? 16.4.1–3 (IG XII 3 1299/300) 16.4.1 (A ll. 10–32) c.100 bc 16.4.2 (B ll. 32–9) c.100 bc

8+ 6

16.4.3 (B ll. 40–69)

24

غ["]Ø[ ]; [ ]غ["]Ø[] ŁÅÆE ; [Ł]ÅÆE [ Æ]Œ[]; [ غ]"Ø (?)

60

c.100 bc

[¸ÆŒÆØØ ]61 [ ]ªÆæ ; ŁÅÆE[ ]; ºØ ; "Ø ; % Æ[ ]; ˆæ Ø ˚HØ ;  ¯ ØÆæØ[ ]; ˚Ø ; æŒØ[ ]; % Æ ; æŒØ[ ]; æŒØ ; æŒØ ; `NªØƺ[ ]; ºØ ;   øÆEØ; ˚ıÇ،Š; ˆæ Ø ; ˝ Ø ; ˚ıøØ Æ ; ˆæ Ø ; ˆæ Ø ; ˆæ Ø ; ˆæ Ø ; ˆæ Ø ; ºØ ; ˚HØ ; [˚]Ø ; ˚HØ

names are given (e.g. ¥  Iªæ ç )e ªæ[Æ]Æ ø ŁÅÆı F ˜Øª[ı ] . . . æØ ŒÆd Pæª ÆØ·). The reference to the name of the secretary or secretaries responsible was presumably intended to date these

[ŁÅÆE] (restored dubiously once and three further times with no basis at all). The second of these entries also contains an individual from Knossos who is pointedly named euergetēs but not proxenos. 60 Three different chronological lists of proxenoi inscribed on two doorposts. It is tempting to believe that lists 16.4.2 and 16.4.3, which occur one after another on the same door post, were inscribed in consecutive years, and thus that the overlap in the poleis represented in each points to a relationship between the two (perhaps a festival announced in the first year, leading to grants of proxenia for those supporting the Theran theōria, followed by delegations of theōroi sent, and honoured, in the second). List 16.4.1 occurs below an earlier text simply awarding the title of euergetēs, in a gymnastic contest to judge from the presence of the Theran gymnasiarch; 16.4.1 itself begins by naming two ‘˚ƺÆ[æ]’ simply euergetai before going on to name those who are both proxenoi and euergetai. On this distinction, see Chapter 1, n. 52. 61 Ethnic securely restored on the basis of an honorific sepulchral monument for this individual erected for him by the demos of Thera, IG XII 3 1625 (though this does not mention his status as proxenos). 58 59

Appendix

337

Map 14. The chronological proxeny lists of Thera, no. 16.1–4.3 (late second and early first centuries bc) lists, and thus they surely correspond to administrative periods (perhaps to years) in which proxenoi were appointed, but the secretaries in these cases may also have been responsible for identifying the proxenoi in question (like the enguoi of proxenies in western Greece).

17. Polis Unknown (Boiotia) 17. A list of theōroi sent to games held at a polis in Boiotia named proxenoi (c.80 bc) ed. pr. V. N. Bardani, Horos 5 (1987), 75–7 with ph.; (SEG 37 388); cf. Gauthier, BE (1989), no. 249; Chaniotis, Kernos 5 (1992), 269, n. 10; Bardani, Horos 8–9 (1992), 205–6; Knoepfler, Chiron 22 (1992), 476–7 no. 121 (date); Habicht, Historia 55 (2006), 161. Dimensions: 0.45m high; 0.395m wide; 0.045m thick

Plaque of white stone in the Museum in Thebes. In format, this text is very close to the Samothracian lists of proxenoi named theōroi. Bardani suggested that the festival in question may well have been the Theban Herakleia, but the prosopographical identification on which this was based is too uncertain

338

Appendix

for this to be anything but a conjecture. Knoepfler plausibly suggests that ‘æH ’ (l.3) should be taken to indicate not that this was the first time this festival was celebrated ever (Chaniotis) nor that it was the first time Athenades had served as agōnothētē for it, but that it was the first time this festival was celebrated following a war which had interrupted it (probably the Mithridatic). Habicht collects the evidence for the interruption of games and festivals. [..]ºø ¼æå [ - - - - - - - - - - - - -] _ [ Ł]Å ı IªøŁ F  [- - - - - - -] _ [..] æH , æØ ŒÆd Pæª[ ÆØ - - - -] 5  HºEØ 20 [. . .]æø Æı [¸] Øå æı  ØØ [. . .]çH vv ¸çH  _ 10 [. .]ŒØÆ æı 25 _ ƺŒØE [˜Å]" æØ æ åı [. . .]º Zø ºı  çÆE ÆØ 30 15 [. .]ø  ˜Øıæı _ [. . .]Å ˜Øıæı [. . .]Ø ºÅ ø æı _ []ÆƪæÆEØ [- - - - - - - ] æhıi _ --------------35

¨[- - - -] ¯hÅ [- - - - - - - - - - - - ] ˜ÆŒæØ [ - - - -] _  ¯ å[ØÆEØ ?] _  I ØÆE [- - - - - -] æØ Œº[B - - - - -] _ —[- - - - -] غØ ø[- - - - - -] ªÆ [- - - - - -] _  OÆ [ - - - - - - - - - - - ]  O[ ØØ ?] _ ˚Æçø[- - - - - - - - - - - - - -] AæØ[ - - - - - - ] _ `N [ªŁE ÆØ ?] _ ÆæØŒºB[ - - - - - - - - - - - - -] Ø[ - - - - - - - - - - - - ] Z œº [- - - - - - -] _ [ªÆæE ?] _ ---------------

1–4 Bardani: [. . .]ºø ¼æå [ , ˝ØŒ åı F | Ł]Å ı _ IªøŁ F  [ a  ˙æ ŒºØÆ or H  ˙æÆŒºø | e] æH , æØ _ ŒÆd Pæª[ ÆØ KÆ Æ|º]  . 2–3 Knoepfler: IªøŁ F  [name of games I F º|ı] æH . 3–4 Gauthier, rightly, compares these lines with the headings of the Samothracian lists of proxenoi. 14  çÆE ÆØ: this polis may be Stephane in Phokis (so Bardani) but Stephanopolis in Epirus and a Boiotian polis temporarily renamed after the Mithridatic war are also possibilities (Knoepfler).

Appendix

339

Map 15. The chronological proxeny list of an unknown city, perhaps Boiotian, no. 17 (c.80 bc)

18. Polis Unknown 18. A probable list of proxenoi appointed in a single year (c.130 bc) ed. pr. Combe (1815), pl. 41; (CIG 1936); Marshall GIBM 1154a; (SGDI II 2291); Ziebarth IG XII suppl. p. 198, on 9 1187; cf. Bechtel Beiträge zur Kunde der indogermanischen Sprachen 20 (1894), 239–41 (squeeze); Smith JHS (1916), 82-5 (on the recutting of the stone and the relief sculpted on it); Roussel REG (1917), 410; Robert OMS 2, 1093; Habicht, Chiron 2 (1972), 103–4 (with Robert BE (1972) no. 37). In accordance with the reconstruction of the original monument offered here, the order of the two faces has been reversed and continuous line numbering abandoned. This edition is based on squeezes of Faces B and C held in Oxford and an autopsy of Face C (Face B is currently inaccessible). Dimensions (upright stele): 1.08m (height); 0.66–0.65m (width, bottom to top); 0.24–0.22m (thickness, bottom to top). For scans of squeezes taken from this stone, see Figures A.22 and A.23.

The stone on which this text was inscribed was subsequently reused in the late Hellenistic period for a rectangular archaizing votive relief. It depicts, in the centre, a warrior standing, with downturned spear, before a trophy (a helmet, breastplate, and greaves on the trunk of a tree); to the left, on the other side of

340

Appendix

the trophy, a female figure stands, pouring out a libation into a bowl from which a snake drinks; behind the warrior, on the far right, a horse and groom stand. To prepare for this relief, the stone, which was originally an upright stele, was rotated through ninety degrees clockwise and recut, correcting the original taper of the stele (preserved on Face C, the new top face) to enable it to stand squarely on its new base, to which it was attached by two new dowel holes in Face B (the originals are still visible on the old bottom face, now the left hand face; see Smith on all this). As a result, only the top thirteen centimetres of inscribed text on Face B survive with a lacuna in ll. 5–7 corresponding to the new dowel hole. More importantly, the relief sculpted on Face A, the front face of the original stele, has completely obliterated the start and main part of the text which would have been inscribed there. The preserved text consists of a list of the names of foreigners in the dative case, in the form: name, patronym, and ethnic. As earlier editors noted, this form, along with the broad geographical range of the cities identified, makes it highly likely that what we are dealing with is a list of proxenoi. Given the dative case, which would identify these individuals as recipients of proxenia, it seems likely that this was a chronological list of proxenoi appointed in a single year, like the earlier lists from Aitolia or Histiaia which are also in the dative case. This would have been made clear by a decree which would have been inscribed on the front face of the stele (Face A) above the start of the list of individuals. This inscription, which is unprovenanced, is usually attributed to a polis on Euboia or in its general vicinity on the basis of the fact that the majority of proxenoi preserved on Face C originate in this region (this is Face A in the editions of Marshall et al.). However, since Face C was, to judge from the vacat which occurs less than halfway down, the last to be inscribed, it probably collects the proxenoi most distant from the granting polis. (In other geographically organized lists, proxenoi from a polis’ local region of primary interaction are inscribed first, in this case probably on Face A below the text of the decree). This is corroborated by the fact that groups of different, unrelated individuals are attested at individual poleis on Face B (from the Propontis), whereas the poleis from the Peloponnese and central Greece on Face C are represented only by individuals (or at most groups of brothers, reflecting a single proxeny grant). On the basis of this ordering, a polis in the eastern Aegean would be a far more likely candidate for appointing this sort of list of proxenoi, and the wide geographic range and number of proxenoi apparently appointed within a single year make it likely that we are dealing with one of the more significant poleis, of the level of Chios or even Rhodes. This is because, in addition to the twenty-two proxenoi listed over the twenty-eight lines of Face C, and eight proxenoi extant on Face B, Face B is likely to have contained up to a further forty proxenoi (allowing for a similar density of proxenoi as Face C) and perhaps thirty-five proxenoi on Face A (allowing, generously, for the decree

Appendix

341

to take up half of this face), which would make it the largest number of proxenoi appointed in any year that we know of (the next largest total, for the Aitolian league, is 74, no. 1.2.1). The text can be securely dated by the occurrence of Aiakides and Charephanes of Delphi, the sons of Babulos, who are active at their polis in either 134/3 or 130/29 bc (Habicht). It was probably reused for the votive relief within one hundred years of its inscription. The iconography—the warrior, the horse and servants (common attributes of the heroized dead), the snake receiving cult, and the trophy—draws on common themes found in earlier burial and victory monuments (see Smith), and suggests that it was erected (by public decree?) after some sort of successful military action. Face A (now relief sculpture) Up to c. 68 lines, probably comprising the text of the decree and the start of the subsequent list. Face B (now lower face) [one line missing] [—]ºıæfi ø ººø[ı] _ ƺåÅØ ,

5

[ ]ºı" ø fi ºı" ı ŒÆ[d] [B]Æ

Œø fi ººA ŒÆd æ[ . . . ], ÆæÆ ŒÆ[d . 3–4 .]ø fi ˝ØŒŁH[ ] _ BıÇ[Æ Ø] , Zø º[ø fi . ]¸ % . [ . ]Æ ıø [ . . . ] [ . . . 6–7 . . . ]øæø fi ºº[F] _ lacuna of up to sixty lines

Face C (now upper face) one line missing ºØçÅæE, _ _ fi ˆæªı ˚ÆçıØE, [ˆ]æªø ºÆ fi ˜ÆŒæ ı ª fiÅ غŒøØ ˚ıØ åı 5 æØÇÅøØ, ˚ÆçØæø fi ÆØ å[ı] ¸ÆE, ØŒæ Ø ¸Œø ¸ÆæıÆø fi , 10 `NÆŒfiÅ ŒÆd Ææç fiÅ [E ] Bƺı ˜ºçE , _ øç  fi Å ŒÆd ªº øØ E[ ] ºø  ¯ ºÆ FØ, ˚ºøøØ ø Æ[ø fi ], 15 % æøØ çı  Oı øØ, —ºıfiÅ ø ŒÆæçE,

342

Appendix

˚ƺºØŁfiÅ ç  ¯ åØÆ[ø fi ], _ _ Æø fi ˇNºŒı Ææƺ[ø fi ], 20 ˜ÆŒø fi ˜ÆŒı Œ ıÆøØ, — æ åø fi [æ]Ø ı ªª fiÅ, _ _ `Nå fi Å ŒÆd æŒø fi E __ ˜ÆŁı Å æº [ÆØ ], 25 ˝ØŒÆ fi ƺ Œæı æØŒÆ[ø fi ],  Øø fi æåº ı ƺÆÆøØ vacat

Face B 5 ÆæÆ ŒÆ[d]: squeeze; Marshall, Ziebarth etc. have Ææ[ı ŒÆd]. Robert tentatively _ suggested [˜]øØ but this is probably not long enough. 7 Zø º[ø fi .]¸ % . [.]Æ ı: squeeze; Marshall et al. have  ¯ Ø øØ  ¯ Ø ı. ø [- -]: if an ethnic, since this section of proxenoi is from the Propontis, perhaps the ethnic of Neon Teichos, a fortified town on the European side near Perinthus (one of Alcibiades’ strongholds, Nepos Alc. 7.4), e.g. ˝ø [Øå fiÅ]. This seems too long, however, and it is probably better to restore  [æ ] or F ø [æı], and assume that this individual, and the individual in l. 8 (whose name would then be preceded by ŒÆd), shared an ethnic in the plural in l. 9. 8 ]øæø fi : squeeze (]ºøæø fi is _ but unattested_ name also possible); Marshall et al. have ]ºŁø fi (an all termination). Face C 10–11 [E ] | Bƺı: Bechtel, Habicht; cf. Marshall: [F | ˝]ƺı. 12 E[ ]:_ squeeze; Marshall, Ziebarth: F. 18 ç: çÆØ is clear on the stone, and is probably a mason’s or compositor’s mistake for the genitive of the patronym which is printed, without note or comment, by Marshall et al. (the alternative, two proxenos names juxtaposed without patronyms and without ŒÆd, would be entirely unparalleled in this inscriptionl. 19 Æø fi : Bechtel, Habicht; Marshall has ÆÆø fi . 22 ªª fiÅ: Bechtel, Robert; Marshall and Ziebarth have `ª fiÅ (‘contre l'estampage (Bechtel), contre le facsimilé _ l’usage’, Robert). même de Marshall et contre 24 ˜ÆŁı: Habicht and Bechtel. Marshall and Ziebarth have ÆŁı. 25 æØŒÆ [ø fi ]: æØŒÆ[ø fi ] is visible on the Oxford squeeze and stone.

Appendix

Map 16. The proxeny list of an unknown city, no. 18 (c.130 bc)

343

Map 17. The 240 different poleis attested across all lists of proxenoi, nos. 1–18.

Photographs and Scans Fig. A.10. Astypalaia, catalogue of proxenoi, Block 1 (Appendix no. 3, c.150 bc), Berlin squeeze (archive of IG), scan M. Sasanow

Fig. A.11. Astypalaia, catalogue of proxenoi, Block 2 (Appendix no. 3, c.150 bc), Berlin squeeze (archive of IG) (2 separate squeezes), scan M. Sasanow

Appendix

347

Fig. A.12. Astypalaia, catalogue of proxenoi, Block 3 (Appendix no. 3, c.150 bc), Berlin squeeze (archive of IG), scan M. Sasanow

Fig. A.13. Chios, catalogue of proxenoi (Appendix no. 4.1; fourth or early third century bc), photo C. Crowther

Fig. A.16. Chios, supplementary catalogue of proxenoi, fr. 2 Face B (Appendix no. 4.2; early third century bc), photo C. Crowther

Fig. A.15. Chios, supplementary catalogue of proxenoi, fr. 1 Face B (Appendix no. 4.2; early third century bc), photo C. Crowther

Fig. A.14. Chios, supplementary catalogue of proxenoi, fr. 1 Face A (Appendix no. 4.2; early third century bc), photo C. Crowther

Fig. A.17. Chios, list of proxenoi appointed in a particular year (?) (Appendix 4.3; second century bc), photo C. Crowther

Fig. A.18. Eresos, catalogue of proxenoi, Faces A–C (Appendix no. 7; c.230–200 bc), Berlin squeezes (archive of IG) 2 squeezes, photos J. Curbera, K. Hallof, and Daniella Summa

Appendix

351

Fig. A.20. Karthaia, catalogue of proxenoi, Face B (Appendix no. 10.1; shortly after 363/2 bc), Berlin squeeze (archive of IG), photo author

Fig. A.19. Karthaia, catalogue of proxenoi, Face A (Appendix no. 10.1; shortly after 363/2 bc), Berlin squeeze (archive of IG), photo author

352

Appendix

Fig. A.21. Narthakion, reverse catalogue of proxenoi (?) (Appendix no. 13; before or after ?c.140 bc), photo Y. Béquignon (# l’Ecole française d’Athènes)

Appendix

353

Fig. A.22. Unknown polis, chronological list of proxenoi (?), Face B (Appendix no. 18; c.130 bc), Oxford squeeze (CSAD), scan M. Sasanow

Fig. A.23. Unknown polis, chronological list of proxenoi (?), Face C (Appendix no. 18; c.130 bc), Oxford squeeze (CSAD), scan M. Sasanow

354

Appendix Index of Polis Ethnics in Lists of Proxenoi

Æ (Abai, Phokis, IACP no. 169): 18 Åæ Å (Abdera, Thrace, IACP no. 640): 14.5.2, 14.8 ıÅ (Abydos, Troad, IACP no. 765): 10.2.5 ªª Å (Angeia, Dolopia, IACP no. 418): 18 ŁÅÆE (Athens, IACP no. 361): 1.1, 1.2.1, 1.4.2, 1.5.1, 1.5.2, 1.6, 1.8.4, 5.197/6, 5.191/0, 5.189/8, 5.183/2, 6.1.1, 6.3.6, 7.2, 8, 10.1.4, 11.3, 16.2.2, 16.4.2 `NªÆØ (Aigion, Achaia, IACP no. 231): 1.6 (= 5.195/4), 1.8.4, 5.195/4 (= 1.6), 6.1.2, 10.2.6 `NªÆØ (Aigai, Aiolis, IACP no. 801): 14.9 `Nªø, ÆŒ K (Aigeai, Macedonia, IACP no. 529): 8 `NªØƺ (Aigiale, Amorgos, IACP no. 471): 3, 16.4.3 `NªØ" Å , `NªØ Æ (Aigina, IACP no. 358): 1.6, 6.2.2, 6.3.6, 7.1, 7.2, 10.1.2, 10.1.7 `NªæÆ , åÆØ K (Aigeira, Achaia, IACP no. 230): 8 `NªŁE ÆØ (Aigosthena, Megarid, IACP no. 224): 17? ŒæƪÆ E (Akragas/Agrigentum, Sicily, IACP no. 9): 1.2.1, 5.190/89 ºÆÆ (Alabanda, Karia, IACP no. 870): 4.4, 14.4.2, 14.5.1, 14.7, 14.12 ºE (Alea, Arkadia, IACP no. 265): 6.1.8 ºE , see  HºE (Elis) ºÆæ (Alexandria, Egypt): 1.8.14 (=5.188/7), 5.191/0, 5.188/7, 6.1.2, 6.1.3 ºÆæ KŒ A æøØ  (Alexandria Troas, Troad): 5.194/3 AºØŒÆæÆ (Halikarnassos, Karia, IACP no. 886): 3, 4.1, 4.2, 4.4, 5.187/6, 8, 10.2.4, 14.2, 14.8, 14.9, 14.12, 15.2 ºØçÅæ (Alipheira, Arkadia, IACP no. 266): 18 ºıÇÆE (Alyzeia, Akarnania, IACP no. 112): 11.2 æÆŒØ Å (Ambrakia, Akarnania, IACP no. 113): 1.2.1, 5.182/1, 11.2 çغ Å (Amphipolis, Thrace, IACP no. 553): 1.2.1, 4.2 ÆçÆE (Anaphe, Cycladic Islands, IACP no. 474): 3 @æØ (Andros, Cycladic Islands, IACP no. 475): 10.1.7 ŁÅØ , BØ Ø ŁÅØ (Anthedon, Boiotia, IACP no. 200): 1.2.3, 1.3 

æØ (Antandros, Troad, IACP no. 767): 4.4  ت , see Æ Ø  Øå (which Antiocheia is unclear): 1.8.4 @ı, ˚æc K (Axos, Crete, IACP no. 950): 10.2.3 ººøØ Æ (Apollonia, Illyria, IACP no. 77): 5.190/89  æÆE (Aptera, Crete, IACP no. 947): 6.3.3

Appendix

355

æªE (Argos, IACP no. 347): 1.5.1, 1.7, 6.1.2, 6.1.3, 6.1.4, 6.1.11, 6.1.13, 6.2.1, 6.3.1, 6.3.2, 6.3.6, 6.4 æªE K çغåø (Amphilochian Argos, Akarnania, IACP no. 115): 10.2.7 æªıæØÆ (Arpi, Italy): 5.191/0 æŒØ (Arkesine, Amorgos, IACP no. 472): 3, 16.4.3 æØ (perhaps Arsinoe/Konope, Akarnania, or Arsione/Methana, Argolid IACP no. 352, or Arsinoe/Koresia, Keos, IACP no. 493): 1.4.2 Ø , çØ (Aspendos, Pamphylia, IACP no. 1001): 2.2, 3 @Ø (Assos, Troad, IACP no. 769): 3?  ıƺÆØ (Astypalaia, IACP no. 476): 4.2, 14.4.2 åØ (Ascheion, Achaia, IACP no. 232): 5.188/7 æ ªØ (Atrax, Thessaly, IACP no. 395): 5.187/6 [.]ø Øæ Æ (Otherwise unattested, Epirus): 1.8.7 BÆæªıºØ" Å (Bargylia, Karia, IACP no. 879): 4.4, 14.5.1, 14.11 BæøÆ (Beroia, Macedonia, IACP no. 533): 6.1.3 BæØ (Boura, Achaia, IACP no. 233): 5.177/6 B Ø (Bouttos, West Lokris): 1.8.1 Bæ E (Brentesion/Brundisium, Italy, IACP no. 78): 5.191/0 BıÇ  Ø (Byzantium, Propontis, IACP no. 674): 1.3, 4.2, 7.2, 18 ˆÆæªÆæ (Gargara, Troad, IACP no. 775): 4.2, 4.4 ˆæªŁØ (Gergis, Troad, IACP no. 777): 1.1 ˆºÆıŒÆ(E) (Glauke, Ionia, IACP p. 1059): 1.1 ˆæ Ø (Gortyn, Crete, IACP no. 960): 1.8.4, 15.2, 16.4.2, 16.4.3 ˜ÆæÆ (Dardanos, Troad, IACP no. 774): 14.9, 14.12 ˜ºç (Delphi, IACP no. 177): 1.8.8, 1.8.9, 6.3.6, 18 ˜æØ (Derion, Akarnania, IACP no. 117): 6.1.13 ˜"ºØ (Delos, IACP no. 478): 10.1.5 ˜ÅÅ æØ , ªÅ KŒ ˜ÅÅ æØ  (Demetrias, Magnesia): 1.2.3, 5.192/1 ˜ıÆE (Dyme, Achaia, IACP no. 234): 1.1, 6.2.2  ¯ ªåÅ  Øæ Å (Unknown, Epirus): 1.8.7  ¯ ºÆØØ (Elaious, Thracian Chersonese, IACP no. 663; or, perhaps, less probably, Ionia, IACP no. 843): 4.2  ¯ ºÆØ Å , also  ¯ ºÆ Å (Elaia, Aiolis, IACP no. 807): 1.2.1, 4.2, 4.3  ¯ º Æ (Hyele/Elea/Velia, Italy, IACP no. 54): 5.188/7, 5.179/8  ¯ ºÆ  (Elateia, Phokis, IACP no. 180): 5.177/6, 5.176/5, 5.175/4, 18  ¯ ºıŁæÆE , ˚æc K  ¯ ºıŁæÆ (Eleutherna, Crete, IACP no. 957): 1.2.1, 1.8.2, 15.2  ¯ ØÆæØ (Epidauros, IACP no. 348): 10.1.2, 16.4.3  ¯ ØÆæØ KŒ B ¸ÆŒøØŒB (Epidauros Limera, Lakonia, IACP no. 329): 10.1.2  ¯ æØ (Eresos, Lesbos, IACP no. 796): 1.2.1, 14.9, 14.10  ¯ æ æØ (Eretria, Euboia, IACP no. 370): 1.2.1, 1.5.1, 10.1.1

356

Appendix

 EæØ (Hermion, Argolid, IACP no. 350): 1.5.2  ¯ æıŁæÆE (Erythrai, Ionia, IACP no. 845): 4.2, 8 ¯PæØ (Europos, Macedonia, IACP no. 536): 1.8.7  ¯ ç (Ephesos, Ionia, IACP no. 844): 1.2.3, 14.1, 14.2, 14.4.2, 14.5.2  ¯ åØÆ (Echinos, Malian Gulf, IACP no. 429): 1.2.1, 8, 17?, 18 ZÆŒŁØ (Zakynthos, Ionian Sea, IACP no. 141): 1.5.2 ZçıæØ Å (Zephyrion, Kilikia, IACP p. 1214): 5.178/7  HºE , also ºE (Elis, IACP no. 251): 11.1, 17  ˙æÆ (Heraia, Arkadia, IACP no. 274): 6.1.5, 6.1.7, 6.2.2, 6.4, 11.1?  ˙æÆŒº Å KŒ F — ı (Herakleia Pontica, Black Sea, IACP no. 715): _ 1.8.2  ˙æÆŒº Å , also K  ˙æÆŒºÆ B æÆåØÆ (Herakleia Trachinia, Malian Gulf, IACP no. 430): 4.1, 5.191/0, 8, 10.1.3 ¨ Ø (Thasos, Northern Aegean, IACP no. 526): 1.2.1, 14.8 ¨ºçØ (Thelphousa, Arkadia, IACP no. 300): 1.8.9 ¨Ø (Thespiai, Boiotia, IACP no. 222): 1.8.3, 5.189/8, 6.3.6 ¨

ƺŒÅ , ÆŒ KŒ (Thessalonike, Macedonia): 8 ¨ÅÆE (Thebes, Boiotia, IACP no. 221): 1.8.6, 6.3.6 ¨ÅæÆE (Thera, Cycladic Islands, IACP no. 527): 2.3, 5.190/89, 5.180/79, 6.2.1, 10.1.2 ¨Ø (Thisbe, Boiotia, IACP no. 223): 5.186/5 ¨ıæ (Thurioi, Italy, IACP no. 74): 1.8.12 ¨ıÆ ØæÅ (Thyateira, Lydia): 4.3 % Æ (Iasos, Karia, IACP no. 891): 14.5.2 % Æ (Ios, Cycladic Islands, IACP no. 484): 16.4.2, 16.4.3  IæÆ Ø (Hierapytna, Crete, IACP no. 962): 3?, 15.2 % ºØ (Ilion, Troad, IACP no. 779): 5.177/6 % ıºØ" Å (Ioulis, Keos, Cycladic Islands, IACP no. 491): 4.1  I ØÆØ (Histiaia/Oreos, Euboia, IACP no. 372): 1.1, 1.8.4 %  æØ (Istron, Crete, IACP no. 964): 15.2 ˚ƺÆıæ Æ (Kalaureia, Saronic Gulf, IACP no. 360): 6.1.7 ˚ƺºı, `N øº KŒ (Kallion/Kallipolis, Aitolia, IACP no. 147): 6.3.6 ˚ƺı (Kalydon, Aitolia, IACP no. 148): 1.8.13 ˚ƺåÅØ , ˚ƺåÆØ , ƺåÅØ (Kalchedon, Propontis, IACP no. 743): 1.2.1, 5.181/0, 6.3.4, 8, 18 ˚ÆıE (Canusium, Italy): 5.195/4 ˚Ææ Ø (Karystos, Euboia, IACP no. 373): 1.2.1, 4.1, 7.1, 7.2, 7.3 ˚Ææåı (?) (Otherwise unattested): 1.8.8 ˚ÆÆæ (Kassandreia/Potideia, Macedonia, IACP no. 598): 5.187/6 ˚ÆØ (Kaunos, Karia, IACP no. 898): 1.8.1, 4.2, 14.5.2, 14.12 ˚ÆçıE (Kaphyai?, Arkadia, IACP no. 275): 6.3.2 ˚ØÆ (Kios, Propontis, IACP no. 745): 10.1.6 ˚Øıæ Å (Kibyra, Lykia): 4.3

Appendix

357

˚ºÆÇØ (Klazomenai, Ionia, IACP no. 847): 4.3, 14.4.2, 14.8, 14.10 ˚ºØ æØ (Kleitor, Arkadia, IACP no. 276): 6.1.8 ˚Ø (Knidos, Karia, IACP no. 903): 1.1, 2.1, 2.3, 3, 6.3.6, 10.1.2, 10.2.4, 16.4.3 ˚Ø , ˚æc ˚Ø (Knossos, Crete, IACP no. 967): 6.1.3, 6.1.6, 6.3.1, 15.2 ˚ºçØ (Kolophon, Ionia, no. 848): 14.2, 14.10 ˚æ"Ø (Koresia/Arsinoe, Keos, Cycladic Islands, IACP no. 493): 4.1 ˚æŁØ (Corinth, IACP no. 227): 1.2.1, 1.4.1, 4.2, 5.195/4, 5.181/0, _ 5.178/7, 6.3.3, 6.4, 10.1.2 ˚æŒıæÆE (Korkyra, Ionian Sea, IACP no. 123): 1.8.4 ˚æø (Koroneia, Boiotia, IACP no. 210): 1.6, 1.8.2, 1.8.3, 5.182/1, 5.178/7 ˚æ øØ Æ (Kroton, Italy, IACP no. 56): 15.1 ˚ıøØ Å , ˚æc ˚ıøØ Æ (Kydonia, Crete, IACP no. 968): 6.1.2, 6.3.2, 15.2, 16.4.3 ˚ıÇ،Š(Kyzikos, Propontis, IACP no. 747): 1.2.1, 1.4.2, 8, 10.1.6, 10.4.2, 14.9, 14.10, 14.12, 14.13, 16.4.3 ˚ıÆE (Kyme, Aiolis, IACP no. 817): 6.1.2, 14.1, 14.5.1, 14.10 ˚ıÆæØØ (Kyparissos, Messenia, IACP no. 317): 6.1.14 ˚ıæÅÆE (Kyrene, Libya, IACP no. 1028): 8 ˚ı ØØ (Kytinion, Doris, IACP no. 392): 8 ˚ıçÆ Æ (Kyphanta, Lakonia, IACP no. 335): 10.1.2 ˚HØ (Kos, Sporades, IACP no. 497): 3, 5.194/3, 6.1.12, 6.3.2, 14.4.2, 14.5.2, 14.6.2, 14.11, 15.2, 16.4.3 ¸ÆŒÆØØ (Sparta, IACP no. 345): 1.2.1, 1.6, 6.1.2, 6.1.4, 6.1.13, 6.2.1, 6.2.2, 6.3.6, 10.1.2, 16.4.1 ¸ÆłÆŒÅ (Lampsakos, Hellespont, IACP no. 748): 4.2, 10.2.5 ¸ÆØŒ (Laodikeia on the Lykos, Lydia?): 4.3 ¸ÆÆE , ˚æc ¸ÆÆE (Lappa, Crete, IACP no. 970): 6.3.2, 15.2 ¸ÆæØÆE , ¨ƺ Kª ¸ÆæÅ , ¨

ƺ KŒ ¸ÆæÅ (Larisa, Thessaly, _ IACP no. 401): 1.2.1, 1.8.11, 10.2.2 ¸ÆæØÆE (Larisa, Thessaly, IACP no. 401; Larisa, Troad, IACP no. 784; or Laris(s)a, Aiolis, IACP no. 818): 14.1 ¸ÆæØØÆE KŒ ŁØ Ø , åÆØ Kª ¸ÆæÅ (Larisa Kremaste, Achaia Phthiotis, IACP no. 437): 8, 10.1.3 ¸ÆæıÆE (Larymna, East Lokris, IACP no. 383): 18 ¸Æ , ¸Æ (Lebadeia, Boiotia, IACP no. 211): 5.186/5, 5.185/4, 10.1.3, 18 ¸Ø (Lebedos, Ionia, IACP no. 850): 4.3 ¸ıŒ Ø (Leukas, Akarnania, IACP no. 126): 5.190/89 ¸ØºÆØ (Lilia, Phokis, IACP no. 185): 5.196/7 ¸ıØÆå (Lysimacheia, Thracian Chersonesos): 7.3

358

Appendix

¸

Ø (Lyttos/Lyktos, Crete, IACP no. 974): 15.2 ªÅ Ie ÆØ æı (Magnesia on the Maiander, Ionia, IACP no. 852): 1.4.1 ºØ (Melos, Cycladic Islands, IACP no. 505): 16.4.2, 16.4.3 ƺºÆE (Malla, Crete, IACP no. 975): 15.2 Æ Ø ,  ت (Mantinea before 221 bc, Antigoneia thereafter, Arkadia, IACP no. 281): 6.3.4, 6.3.6, 6.4, 11.1 Ææø Å (Maroneia, Thrace, IACP no. 646): 10.1.7, 14.5.2, 14.8 ÆƺØH Å (Massalia/Marseille, IACP no. 3): 5.196/5 ªÆºº Æ (Megalopolis, Arkadia, IACP no. 282): 1.1, 1.2.1, 1.8.9, 6.1.1, 6.1.5, 6.1.6, 6.1.14, 6.3.5 ªÆæ (Megara, IACP no. 225): 1.5.1, 6.3.1, 6.3.3, 6.3.6, 7.1, 16.4.2, 17? ºØ ÆØ (Melitaia, Achaia Phthiotis, IACP no. 438): 1.2.1 Æ (Messene, Messenia, IACP no. 318): 1.2.1, 1.8.1, 1.8.13, 6.1.13, 6.2.2, 6.3.3, 11.1  Æ E (Metapontion, Italy, IACP no. 61): 1.2.1 ÅŁıÆE (Methymna, Lesbos, IACP no. 797): 7.1, 7.2, 7.3 Å æº Å (Metropolis, Thessaly, IACP no. 403): 18 غ"Ø , غ Ø (Miletos, Ionia, IACP no. 854): 7.1, 16.2.1, 16.3? ıŒØ (Mykonos, Cycladic Islands, IACP no. 506): 2.1 ıºÆ (Mylasa, Karia, IACP no. 913): 14.12 ıæØÆE (either Myrina, Lemnos, IACP no. 502, or, less likely, Myrina, Aiolis, IACP no. 822): 14.6.1, 14.9 ı غÅÆ (Mytilene, Lesbos, IACP no. 798): 1.2.1, 1.4.1, 1.5.1, 4.2, 7.1, 7.2, 7.3, 14.5.1 ˝ Ø (Naxos, Cycladic Islands, IACP no. 597): 14.5.1, 16.4.3 ˝ÆæŁÆŒ , ˝ÆæŁÆŒØ (Narthakion, Achaia Phthiotis, IACP p. 687): 13 ˝Æı Œ Ø (Naupaktos, West Lokris, IACP no. 165): 5.188/7, 5.187/6 ˝ØŒÅ (Nikomedia, Propontis): 11.5 ˝ıÆ (Nysa, Karia): 4.3, 14.9  Oº Ø (Olynthos, Kalchidike, IACP no. 588): 2.2  O Ø (Opous, East Lokris, IACP no. 386): 17?, 18  O Ø (Opous, Elis, IACP no. 261): 11.1  OæŁøØ (Orthosia, Karia): 4.3  OæåØ (Orchomenos, Boiotia, IACP no. 213): 1.2.3, 1.5.1  OæåØ (Orchomenos, Arkadia, IACP no. 286): 6.1.7 —ƺ (Paleis, Akarnania, IACP no. 132): 1.8.9 —Æ ØŒÆÆØ , —Æ ØŒÆÆ Å (Pantikapeion, Black Sea, IACP no. 705): 4.2, 5.195/4 —ÆæØÆ (Parion, Propontis, IACP no. 756): 9, 14.5.2 — æØ (Paros, Cycladic Islands, IACP no. 509): 2.1 —Æ æ (Patrai, Achaia, IACP no. 239): 11.1

Appendix

359

— çØ (Paphos, Cyprus, IACP no. 1019): 5.176/5 —ººÆ , ÆŒ KŒ —ººÅ (Pella, Macedonia, IACP no. 543): 1.2.1, 5.190/89 —ººÆ (Pellene, Achaia, IACP no. 240): 1.2.1, 1.2.2, 6.1.10, 6.1.13, 11.1 —ººÆf KŒ B ¸ÆŒøØŒB (Pellana, Lakonia, IACP no. 341): 10.1.2 —Ææ"ŁØ (Peparethos, Western Aegean, IACP no. 511): 1.2.1, 1.2.3, 4.1 —æªÆ (Perge, Pamphylia, IACP no. 1003): 1.5.1 —æªÆÅ (Pergamon, Aeolis, IACP no. 828): 6.1.13 —ºÆ ÆØ (Plataia, Boiotia, IACP no. 216): 4.2, 6.3.6 —ºıæ"Ø , ˚æc —ºıææ"Ø (Polyrhenia, Crete, IACP no. 983): 1.8.9, 6.3.2 —æØÅ (Priene, Ionia, IACP no. 861): 14.3, 14.5.2, 14.8, 14.10 —æŒ"Ø (Prokonnesos, Propontis, IACP no. 759): 10.1.6 —æåØ (Proschion, Aitolia, IACP no. 154): 1.8.13 —æH (Pronnoi, Akarnania, IACP no. 135): 1.8.10 —ıææÆE (Pyrrha, Lesbos, IACP no. 799): 7.1  ŪE (Rhegion/Rhegium, Italy, IACP no. 68): 5.190/89, 15.1  Ø (Rhodes, IACP no. 1000): 1.2.1; 2.3, 4.1, 5.180/79, 7.3, 14.5.2  øÆE (Rome): 1.2.1, 5.195/4, 5.190/89, 5.189/8, 7.3, 16.4.3 ƺıæØÆe (Selymbria, Propontis, IACP no. 679): 7.2, 7.3 Æ (Samos, Ionia, IACP no. 864): 8, 14.3, 14.5.2, 14.6.2, 17 ÆŁæfi  (Samothrace, Northern Aegean, IACP no. 515): 4.2 ÆæØÆ (Sardis): 4.3 æçØ (Seriphos, Cycladic Islands, IACP no. 517): 10.1.5 Ø Æ (Side, Pamphylia, IACP no. 1004): 6.1.1 ØØ (Sidon, Levant): 1.4.2 ØŒıØ , ŒıØ (Sikyon, Peloponnese, IACP no. 228): 1.2.1, 1.4.1, 5.176/5, 11.1 Øø (Sinope, Black Sea, IACP no. 729): 1.5.1, 4.2, 11.5 ŒÆæç (Skarpheia, East Lokris, IACP no. 387): 18 Œ ıÆE (Skotoussa, Thessaly, IACP no. 415): 5.189/8, 18 º (Soloi, Kilikia Tracheia, IACP no. 1011): 1.6  çÆ Å (see comm. ad loc.): 17  æÆ ØŒ (Stratonikeia, Karia): 14.5.2, 14.7  ıç ºØ (Stymphalos, Arkadia, IACP no. 296): 6.3.3 ıÆ (Synnada, Phrygia): 4.3 ıæÆŒØ (Syracuse, Sicily, IACP no. 47): 1.2.1, 1.6, 8, 10.2.7, 15.1? æØ (Syros, Cycladic Islands, IACP no. 523): 10.1.5 ƺ ø (Talana, Ainis, IACP no. 424): 1.8.7 ÆƪæÆE (Tanagra, Boiotia, IACP no. 220): 6.3.6, 17 ÆæÆ E (Taras/Tarentum, Italy, IACP no. 71): 1.4.2, 1.8.9, 5.189/8, 8, 11.6, 15.1

360

Appendix

ÆıåæØ (Taucheira, Libya, IACP no. 1029): 6.1.3 ª Å (Tegea, Arkadia, IACP no. 297): 6.1.3, 6.1.13, 6.3.1, 6.3.2, 6.3.4, 11.1 ºÅ (Telmessos, Lykia, IACP no. 936): 2.1 Ø (Tenedos, East Aegean, IACP no. 793): 8, 10.1.6 "œ (Teos, Ionia, IACP no. 868): 4.2, 4.3, 14.6.2, 14.10 "Ø (Tenos, Cycladic Islands, IACP no. 525): 4.2, 10.1.5, 16.4.2 ?Ø Æ ?/?ı Æ ? (Otherwise unattested): 6.1.15 ºø (Tlos, Lykia): 5.188/7 æŁı (Torthyneion, Arkadia, IACP no. 302): 1.2.1 æØŒŒÆE , ¨

ƺ KŒ 挌Š(Trikka, Thessaly, IACP no. 417): 10.2.2, 18 æØå (Trichoneion, Aitolia, IACP no. 156): 5.188/7 æØÇ"Ø , æ Ø (Troizen, Argolid, IACP no. 357): 6.3.6, 18  ( (Hymisseis, Karia, IACP no. 890): 7.3 ƺÆÆE (Phalanna, Perrhaibia, IACP no. 468): 18 ƺÆ æØ (Phalasarna, Crete, IACP no. 981): 6.1.2 ƺøæØÆ

(Phaloria, Thessaly, IACP no. 411): 1.1 Ææ ºØ , ¨ƺ KŒ Ææ ºı (Pharsalos, Thessaly, IACP no. 413): 2.1, 18 Æź Å (Phaselis, Lykia, IACP no. 942): 8  Å , ØŒ (Pheneos, Arkadia, IACP no. 291): 6.1.9, 6.1.13, 6.3.2 Øƺ (Phi(g)aleia, Arkadia, IACP no. 292): 1.1 ºØ Ø (Phleious, Argolid, IACP no. 355): 6.1.13, 6.3.2, 6.4, 11.1? ºª æØ (Pholegandros, Cycladic Islands, IACP no. 513): 2.3 øŒÆ øŒÆØ (Phokaia, Ionia, IACP no. 859): 4.2, 14.9 ÆØæø (Chaironeia, Boiotia, IACP no. 201): 6.3.6 ƺŒØ (Chalkis, Euboia, IACP no. 365): 1.2.1, 1.8.4, 1.8.5, 5.179/8, 10.1.1, 10.2.1, 11.3 ƺåÅØ , see ˚ƺåÅØ , ˚ƺåÆØ (Kalchedon) æÆ Å KŒ F — ı (Chersonesos Pontica, Black Sea, IACP no. 695): 5.195/4, 5.192/1, 7.1 E (Chios, Ionia, IACP no. 840): 1.8.4, 2.1, 5.194/3, 5.186/5, 5.185/4, 5.184/3, 7.3, 14.9 ’#æØ (Oropos, Boiotia, IACP no. 214): 5.187/6

Index of Regional Ethnics (Including References to Ethnē, Koina, and Kings) `N øº (Aitolian koinon, IACP pp. 379–80): 4.1?, 6.2.1, 8 ŒÆæ  (Akarnanian koinon, IACP pp. 351–2): 5.194/3 Øæ Å (Epirote koinon, IACP pp. 338–9): 1.2.1

Appendix

361

Ææa Æغø

ºı (representatives from king Attalus): 14.5.2 åÆØ (Achaian koinon, IACP pp. 472–7): 15.2 BØ Ø (Boiotian koinon, IACP pp. 431–3): 4.2 Ie F ŒØF H æd e ˜Øı åØ H H Ie % øÆ ŒÆd  EººÅ ı (koinon of Dionysiac artists of Ionia and the Hellespont): 14.10, 14.11 ˜ºł (Dolopian ethnos, IACP p. 683): 4.1? ¨æø  (Thesprotia, Epirus, IACP pp. 340): 1.8.2 ¨ƺ (Thessalian koinon, IACP pp. 680–3): 1.6?, 4.1? ˚æ" (Crete, IACP pp. 1144–5): 6.1.5, 6.3.6, 11.4 ¸Œæ (Lokrian koinon, IACP pp. 391–2, 664–5): 8 ÆŒ (Macedon): 1.2.1, 4.1?, 6.1.14, 8, 10.2.2, 16.3 ƺØ , źØ (Malis, IACP pp. 685–6): 4.2, 10.1.2 º (Molossian koinon, IACP pp.338–9; cf. Øæ Å ): 1.6 øŒ (Phokian koinon, IACP pp. 399–400): 8

Bibliography Ager, S. (1996) Interstate Arbitrations in the Greek World, 337–90 B.C. Berkeley. Alcock, S.E. (1995) ‘Pausanias and the Polis: Use and Abuse’, in Sources for the Ancient Greek City-State, ed. M. H. Hansen. Copenhagen: 326–44. Ampolo, C. and Parra, M. C. (2001) Da un’antica città di Sicilia: i decreti di Entella e Nakone, catalogo della Mostra. Pisa. Anderson, G. (2009) ‘The Personality of the Greek State’, JHS 129: 1–22. Aneziri, S. (2003) Die Vereine der dionysischen Techniten im Kontext der hellenistischen Gesellschaft: Untersuchungen zur Geschichte, Organisation und Wirkung der hellenistischen Technitenvereine (Historia Einzelschriften 163). Stuttgart. Arafat, K. and Morgan, C. (1994) ‘Athens, Etruria and the Heuneburg: Mutual Misconceptions in the Study of Greek-Barbarian Relations’, in Classical Greece: Ancient Histories and Modern Archaeologies, ed. I. Morris. Cambridge: 108–34. Archibald, Z. (2001) ‘Making the Most of One’s Friends: Western Asia Minor in the Early Hellenistic Age’, in Hellenistic Economies, ed. Z. Archibald et al., London: 245–72. Arnush, M. (1992) ‘Ten-Day Armistices in Thucydides’, GRBS 33: 329–53. Arnush, M. (2000) ‘Argead and Aetolian Relations with the Delphic Polis in the Late Fourth Century bc’, in Alternatives to Athens, ed. R. Brock and S. Hodkinson. Oxford: 293–307. Ashton, R. (2000) ‘More Pseudo-Rhodian Drachms from Central Greece: Haliartos (Again), Chalkis, and Euboia Uncertain (?)’, NC 160: 93–116. Austin, J. L. (1975) How to do Things with Words, 2nd edn., ed. J.O. Urmson and M. Sbisà. Oxford. Badian, E. (1994) ‘Herodotus on Alexander I of Macedon: A Study in Some Subtle Silences’, in Greek Historiography, ed. S. Hornblower. Oxford: 107–30. Bagnall, R. S. (1976) The Administration of the Ptolemaic Possessions outside Egypt. Leiden. Bardani, V. (1987) ‘¯Œ BØøÆ’, Horos 5: 75–7. Barnes, C. H. L. (2006) ‘The Ferries of Tenedos’, Historia 55: 167–77. Baslez, M.-F. (1989) L’étranger dans la Grèce antique. Paris. Behrwald, R. (2000) Der Lykische Bund: Untersuchungen zu Geschichte und Verfassung. Bonn. Beloch, K. J. (1927) Griechische Geschichte, 2nd edn., vol. IV.2. Berlin and Leipzig.

364

Bibliography

Bickerman, E. J. (1958) ‘Autonomia. Sur un passage de Thucydide (I, 144, 2)’, RIDA 5: 313–43. Bielman, A. (1994) Retour à la liberté: libération et sauvetage des prisonniers en Grèce ancienne: recueil d’inscriptions honorant des sauveteurs et analyse critique. Paris. Billows, R. A. (1990) Antigonos the One-Eyed and the Creation of the Hellenistic State. Berkeley. Billows, R. A. (2008) ‘International Relations’, in The Cambridge History of Greek and Roman Warfare, vol. I: Greece, The Hellenistic World and the Rise of Rome, ed. P. Sabin, H. van Wees, and M. Whitby. Cambridge: 303–24. Bissa, E. M. A. (2009) Governmental Intervention in Foreign Trade in Archaic and Classical Greece. Leiden. Blondé, F., Mulliez, D., and Müller, A. (eds.) (2000) ‘Thasos, d’Archiloque aux guerres médiques: questions de topographie et d’urbanisme’, Topoi 10: 1–56. Blümel, W. (1994) ‘Two Inscriptions from the Cnidian Peninsula’, EA 23: 157–8. Blümel, W. (2000) ‘Ehrendekrete aus Olymos’, EA 32: 97–100. Boffo, L. (1995) ‘Ancora una volta sugli “Archivi” nel mondo greco: conservazione e “pubblicazione” epigrafica’, Athenaeum 83: 91–130. Bommelaer, J. F. (1969) ‘Une nouvelle proxénie de Delphes’, BCH 93: 92–6. Bonanno, A. (2005) Malta: Phoenician, Punic, and Roman. Malta. Bousquet, J. (1958) ‘Inscriptions de Delphes’, BCH 82: 64–91 Bouvier, H. (1978) ‘Honneurs et récompenses à Delphes’, ZPE 30: 101–18. Bowersock, G. W. (1965) Augustus and the Greek World. Oxford. Bowersock, G. W. (1985) ‘Hadrian and Metropolis’, in Bonner HistoriaAugusta-Colloquium 1982/1983, ed. J. Straubb. Bonn: 75–88. Braun, T. (2004) ‘Xenophon’s Dangerous Liaisons’, in The Long March: Xenophon and the Ten Thousand, ed. R. J. Lane Fox. New Haven: 97–130. Bravo, B. (1980) ‘Sulân: Représailles et justice privée contre les étrangers dans les cites grecques’, ASNP 10: 675–987. Brélaz, C. (2008) ‘L’adieu aux armes: la défense de la cité grecque dans l’empire romain pacifié’, in Sécurité collective et ordre public dans les sociétés anciennes, ed. C. Brélaz et P. Ducrey. Geneva: 155–204. van Bremen, R. (1996) The Limits of Participation: Women and Civic Life in the Greek East in the Hellenistic and Roman Periods. Amsterdam. Bresson, A. (1980) ‘Rhodes, l’Hellénion et le statut de Naucratis (VIe–IVe s. a. C.)’, DHA 6: 291–349. Bresson, A. (2000) La cité marchande. Paris. Bresson, A. (2003) ‘Merchants and Politics in Ancient Greece: Social and Economic Aspects’, in Mercanti e politica nel mondo antico, ed. C. Zaccagnini. Rome: 139–64.

Bibliography

365

Bresson, A. (2005) ‘Les cités grecques et leurs inscriptions’, in L’écriture publique du pouvoir, ed. A. Bresson, A. Cocula and C. Pébarthe. Paris: 153–68. Bresson, A. (2007–8) L’économie de la Grèce des cités ( fin VIe–Ier siècle a.C.) (2 vols.). Paris. Briant, P. (1998) ‘Droaphernes et la statue de Sardes’, in Studies in Persian History. Essays in Memory of Professor David M. Lewis (Achaemenid History 11), ed. M. Brosius and A. Kuhrt. Leiden: 205–26. Bringmann, K. (1993) ‘The King as Benefactor: Some Remarks on Ideal Kingship in the Age of Hellenism’, in Images and Ideologies: Self-definition in the Hellenistic World, ed. A.W. Bulloch et al. Berkeley: 7–25. Brock, R. and Hodkinson, S. (eds.) (2000) Alternatives to Athens. Oxford. Brousseau, E., and Glachant, J.-M. (eds.) (2008) New Institutional Economics: A Guidebook. Cambridge. Brulé, P. (1978) La piraterie crétoise hellénistique. Paris. Brun, P. (1989) ‘L’île de Kéos et ses cités au IVe siècle av J.C.’, ZPE 76: 128–38. Brun, P. (1996a) ‘Problèmes de la micro-insularité en Grèce Égéenne’, REA 98: 295–310. Brun, P. (1996b) Les archipels égéens dans l’antiquité grecque (Ve–IIe siècles av. notre ère). Paris. Brun, P. (2004) ‘La datation de IG II² 404, décret athénien concernant les cités de Kéos’, ZPE 147: 72–8. Buckler, J. (1998) ‘Epameinondas and the New Inscription from Knidos’, Mnemosyne 51.2: 192–204. Bull, H. (2002) The Anarchical Society: A Study of Order in World Politics, 3rd edn. New York. Buraselis K. (2004) ‘Some Remarks on the Koan Asylia (242 B.C.) Against its International Background’, in The Hellenistic Polis of Kos: State, Economy and Culture, ed. K. Höghammar. Uppsala: 15–20. Burrell, B. (2004) Neokoroi: Greek Cities and Roman Emperors. Leiden. Burton, G. P. (2000) ‘The Resolution of Territorial Disputes in the Provinces of the Roman Empire’, Chiron 30: 195–215. Buzan, B. and Little, R. (2000) International Systems in World History: Remaking the Study of International Relations. Oxford. Cabanes, P. (1976) L’Épire de la mort de Pyrrhos à la conquête romaine. Paris. Camia, F. (2009) Roma et le Poleis. Athens. Carey, C., et al. (2008) ‘Fragments of Hyperides’ Against Diondas from the Archimedes palimpsest’, ZPE 165: 1–19. Cargill, J. (1995) Athenian Settlements of the Fourth Century B.C. Leiden. Carlsson, S. (2010) Hellenistic Democracies: Freedom, Independence, and Political Procedure in Some East Greek City-States. Stuttgart.

366

Bibliography

Carusi, C. (2003) Isole e peree in Asia Minore: contributi allo studio dei rapporti tra poleis insulari e territori continentali dipendenti. Pisa. Cartledge, P. (1979) Sparta and Lakonia. London. Cartledge, P. and Spawforth, A. (1989) Hellenistic and Roman Sparta: A Tale of Two Cities. London. Cassayre, A. (2010) La justice dans les cités grecques: de la formation des royaumes hellénistiques au legs d’Attale. Rennes. Chandezon, Chr. (2003) L’élevage en Grèce (fin Ve–fin Ier s. a.C.): l’apport des sources épigraphiques. Bordeaux. Chaniotis, A. (1993) ‘Ein diplomatischer Statthalter nimmt Rücksicht auf den verletzten Stolz zweier hellenistischer Kleinpoleis (Nagidos und Arsinoe)’, EA 21: 33–42. Chaniotis, A. (1996) Die Verträge zwischen kretischen Poleis in der hellenistichen Zeit. Stuttgart. Chaniotis, A. (1997) ‘Theatricality Beyond the Theater: Staging Public Life in the Hellenistic World’, Pallas 47: 219–59. Chaniotis, A. (2005) War in the Hellenistic World: A Social and Cultural History. Oxford. Chankowski, V. (2008) Athènes et Délos à l’Époque Classique: Recherches sur l’administration du sanctuaire d’Apollon délien (Bibliothèque des écoles françaises d’Athènes et de Rome 331). Athens. Charneux, P. (1953) ‘Inscriptions d’Argos’, BCH 77: 387–403. Charneux, P. (1956) ‘Inscriptions d’Argos’, BCH 80: 596–618. Charneux, P. (1958) ‘Inscriptions d’Argos’, BCH 82: 1–15. Cherry, J. F., Davis, J. L., and Mantzourani, E. (1991) ‘Introduction to the Historical and Epigraphical Evidence’, in Landscape Archaeology as LongTerm History: Northern Keos in the Cycladic Islands from Earliest Settlement until Modern Times, eds. Cherry, J. F., Davis, J. L., and Mantzourani, E. Los Angeles: 235–43. Chroust, A. H. (1973) ‘Athens Bestows the Decree of Proxenia on Aristotle’, Hermes 101: 187–94. Clinton, K. (1996) ‘Review of Lawton (1995)’, BMCR, accessed 25 July 2012:

Clerc, M. (1893) Les métèques athéniens: étude sur la condition légale, la situation morale et le rôle social et économique des étrangers domiciliés à Athènes. Paris. Cohen, E. E. (1973) Ancient Athenian Maritime Courts. Princeton. Cohen, E. E. (1992) Athenian Economy and Society: A Banking Perspective. Princeton. Coleiro, E. (1971) ‘Maltese Coins of the Roman Period’, NC 11: 67–91. Collar, A. (2013) Religious Networks in the Roman Empire: The Speed of New Ideas. Cambridge.

Bibliography

367

Combe, T. (1815) A Description of the Collection of Marbles in the British Museum, vol. II. London. Condoléon, N. M. (1949) ‘Inscriptions de Chios’, RPh 23: 9–13. Connor, W. (1967) ‘Two Notes on Cimon’, TAPA 98: 67–75. Constantakopoulou, C. (2007) The Dance of the Islands: Insularity, Networks, the Athenian Empire, and the Aegean World. Oxford. Cooper, F. A. (2009) ‘The Fortifications of Epaminondas and the Rise of the Monumental Greek City’, in City Walls: The Urban Enceinte in Global Perspective, ed. J. D. Tracy. Cambridge: 155–91. Craven, M. (2010) ‘Statehood, Self-determination and Recognition’, in International Law, 3rd edn., ed. M. D. Evans. Oxford: 203–51. Crowther, C. V. (1992) ‘The Decline of Greek Democracy’, JAC 7: 13–48. Crowther, C. V. (1993) ‘Foreign Judges in Seleucid Cities (GIBM 421)’, JAC 8: 40–77. Crowther, C. V. (1994) ‘Foreign Courts on Kalymna in the Third Century B.C.’, JAC 9: 33–55. Crowther, C. V. (1995) ‘Iasos in the Second Century bc III: Foreign Judges from Priene’, BICS 40: 91–138. Crowther, C. V. (1996) ‘I.Priene 8 and the History of Priene in the Early Hellenistic Period’, Chiron 26: 195–250. Crowther, C. V. (1999) ‘Aus der Arbeit der Inscriptiones Graecae 4: Koan Decrees for Foreign Judges’, Chiron 29: 251–319. Culasso-Gastaldi, E. (2004) Le prossenie ateniesi del IV secolo a.C: gli onorati asiatici. Alexandria. Culasso-Gastaldi, E. (2005) ‘Per un bilancio comparativo sulle prossenie ateniesi del IV secolo a.C.’, in Il cittadino, lo straniero, il barbaro, fra integrazione ed emarginazione nell’antichità. Atti del I Incontro Internazionale di Storia antica, Genova, 22–24 maggio 2003 (Serta Antiqua et Mediaevalia 7), ed. M. G. Angeli Bertinelli and A. Donati. Rome: 45–75. Currie, B. (2005) Pindar and the Cult of Heroes. Oxford. Curty, O. (1995) Les parentés légendaires entre cités grecques. Geneva. Daux, G. (1936) Delphes au IIe et au Ier siècle depuis l’abaissement de l’Étolie juisqu’ à la paix romaine, 191–31 av. J.-C. Paris. Daux, G. (1937) ‘Alcibiade, proxene de Lacedemone’, in Mélanges offerts à A. M. Desrousseaux. Paris: 117–22. Daux, G. (1943) Chronologie delphique. Paris. Davies, J. K. (1978) ‘Athenian Citizenship: The Descent Group and the Alternatives’, CJ 73: 105–21. Davies, J. K. (1984) ‘Cultural, Social and Economic Features of the Hellenistic World’, CAH2 7.1: 257–320. Davies, J. K. (1994) ‘On the Non-Usability of the Concept of “Sovereignty” in an Ancient Context’, in Federazioni e Federalismo nell’Europa antica, ed. L. A. Foresti, et al. Milan: 51–65.

368

Bibliography

Davies, J. K. (2002) ‘The Interpenetration of Hellenistic Sovereignties’, in The Hellenistic World: New Perspectives, ed. D. Ogden. London: 1–21. DiMaggio, P. J. and Powell, W. W. (1983) ‘The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields’, American Sociological Review 48:147–60. DiMaggio, P. J. and Powell, W. W. (1991) ‘Introduction’, in The New Institutionalism in Organizational Analysis, ed. W. W. Powell and P. J. DiMaggio. Chicago: 1–38. Dimitrova, N. M. (2008) Theoroi and Initiates in Samothrace: The Epigraphical Evidence (Hesperia Supplement 37). Princeton. Dmitriev, S. V. (2005) City Government in Hellenistic and Roman Asia Minor. Oxford. Domingo Gygax, M. (2009) ‘Proleptic Honours in Greek Euergetism’, Chiron 39: 163–91. Doukas, N. (1805) ¨ıŒıı ˇºæı  æ ı — º

Å ØÆŒ —º ı ØºÆ Œ:  ÆçæÆ Ł Æ, åºØÆ Ł Æ ŒÆØ ŒŁ Æ  Ø Ææ ˝ ç ı ˜ ŒÆ (10 vols.). Vienna. Dover, K. J. (1974) Greek Popular Morality in the Time of Plato and Aristotle. Oxford. Dreher, M. (1995) ‘Poleis und Nicht-Poleis im Zweiten Athenischen Seebund’, in Sources for the Ancient Greek City-State, ed. M. H. Hansen. Copenhagen: 171–200. Dubois, L. (1986) Recherches sur le dialecte arcadien, (3 vols.). Leuven. Dubois, L. (2000) ‘Hippolytos and Lysippos: Remarks on some Compounds in  -, Ø’, in Greek Personal Names: Their Value as Evidence (Proceedings of the British Academy 104), ed. S. Hornblower and E. Matthews. Oxford: 41–52. Durrbach, F. (1921–2) Choix d’inscriptions de Délos, (2 vols.). Paris. Eckstein, A. M. (2006) Mediterranean Anarchy, Interstate War and the Rise of Rome. Berkeley. Eckstein, A. M. (2008) Rome Enters the Greek East: From Anarchy to Hierarchy in the Hellenistic Mediterranean, 230–190 bc. Oxford. Ehrenberg, V. (1960) The Greek State. Oxford. Eilers, C. (2002) Roman Patrons of Greek Cities. Oxford. Elias, N. (1983) The Court Society, trans. E. Jephcott. Oxford. Engen, D. T. (2010) Honor and Profit: Athenian Trade Policy and the Economy and Society of Greece, 415–307 B.C.E. Ann Arbor. Erskine, A. (2010) ‘Between Philosophy and the Court: The Life of Persaios of Kition’, in Creating a Hellenistic World, ed. A. Erskine and L. LlewellynJones. Swansea: 177–94. Étienne, R. (1990) Ténos II: Ténos et les Cyclades du milieu du IVe siécle av J.-C. au milieu du IIIe siécle ap. J.-C. Paris.

Bibliography

369

Étienne, R. and E. Dourlot (1996) ‘Les Cyclades’, in Le Cicladi ed il Mondo Egeo, ed. E. Lanzillotta and D. Schilardi. Rome: 21–7. Fabiani, R. (2013) ‘Iasos between Mausolus and Athens’, in Euploia: La Lycie et la Carie antiques, ed. P. Brun, L. Cavalier, K. Konuk, and F. Prost. Bordeaux: 317–30. Fabiani, R. (forthcoming a) I decreti onorari di Iasos tra cronologia e storia (Vestigia series). Munich. Fabiani, R. (forthcoming b) Der Beitrag Kleinasiens zur Kultur- und Geistesgeschichte der griechisch-römischen Antike. Tagung der kleinasiatischen Kommission der Österreichischen Akademie der Wissenschaften. Fentress, J. and Fentress, E. (2001) ‘ “The Hole in the Doughnut”, Review of Horden and Purcell (2000)’, P&P 173: 203–19. Fernoux, H. (2011) Le Demos et la Cité: Communautés et assemblées populaires en Asie Mineure à l’époque impériale. Rennes. Ferrary, J.-L. (1997a) ‘De l’évergétisme hellénistique à l’évergétisme romain’, in Actes du Xe Congrès International d’Épigraphie Grecque et Latine, Nîmes, 4–9 octobre 1992, ed. M. Christol and O. Masson. Paris: 199–225. Ferrary, J.-L. (1997b) ‘The Hellenistic World and Roman Political Patronage’, in Hellenistic Constructs: Essays in Culture, History, and Historiography, ed. P. Cartledge, P. Garnsey, and E. S. Gruen. Berkeley: 105–19. Ferrary, J.-L. (2005) ‘Les mémoriaux de délégations du sanctuaire oraculaire de Claros et leur chronologie’, CRAI 149: 719–65. Feyel, M. (1936) ‘Nouvelles inscriptions d’Akraiphia’, BCH 60: 11–36. Figueira, T. J. (1993) ‘Aigina and the Naval Strategy of the Late Fifth and Fourth Centuries’, in his Excursions in Epichoric History: Aiginetan Essays. Maryland: 325–61. Flacelière, R. (1937) Les Aitoliens à Delphes: contribution à l’histoire de la Grèce centrale au IIIe siècle av. J.-C. Paris. Flensted-Jensen, P. and Hansen, M. H. (2007) ‘Pseudo-Skylax’, in The Return of the Polis. The Use and Meanings of the Word Polis in Archaic and Classical Sources. Papers from the Copenhagen Polis Centre 8, ed. M. H. Hansen. Stuttgart: 204–42. Forrest, W. G. (1987) ‘A Samian Proxeny Decree’, Horos 5: 91–3. Fossey, J. M. (1983) ‘The Proxenia Decrees of Tanagra’, Horos 2: 119–35. Fossey, J. M. (1985) ‘Aristoxenos the Boiotian Again’, LCM 10: 39–40. Fossey, J. M. (1996) ‘The Proxenia Decrees of the Aitolian Federation’, AncW 27: 158–67. Fournier, J. (2010) Entre tutelle romaine et autonomie civique: l’administration judiciaire dans les provinces hellénophones de l’empire romain 129 av. J.-C–235 ap. J. C. Athens. Francotte, H. (1910) Mélanges de droit public grec. Liège. Fraser, P. M. (1960) Samothrace: The Inscriptions on Stone, (Samothrace 2.1). London.

370

Bibliography

Fraser, P. M. (2003) ‘Agathon and Kassandra (IG IX 12 4 1750)’, JHS 123: 26–40. Fraser, P. M. (2009) Greek Ethnic Terminology. Oxford. Frei, P. and Marek, C. (1997) ‘Die karisch-griechische Bilingue von Kaunos: eine zweisprachige Staatsurkunde des 4. Jh.s v. Chr.’, Kadmos 36: 1–89. Frei, P. and Marek, C. (1998) ‘Die karisch-griechische Bilingue von Kaunos: ein neues Textfragment’, Kadmos 37: 1–18. Fröhlich, P. (2010) ‘L’inventaire du monde des cités grecques. Une somme, une méthode et une conception de l’histoire’, RH 655: 637–77. Gabrielsen, V. (1994) ‘Subdivisions of the State and their Decrees in Hellenistic Rhodes’, C&M 45: 117–35. Gabrielsen, V. (2001) ‘The Rhodian associations and economic activity’, in Hellenistic Economies, ed. Z. H. Archibald, et al. London: 215–44. Gabrielsen, V. (2007) ‘Brotherhoods of Faith and Provident Planning: The Non-public Associations of the Greek World’, MHR 22: 183–210. Gamauf, R. (1999) Ad statuam licet confugere. Untersuchungen zum Asylrecht im römischen Prinzipat. Frankfurt. Gauthier, P. (1972) Symbola: les étrangers et la justice dans les cités grecques. Nancy. Gauthier, P. (1982) ‘Les saisies licites aux dépens des étrangers dans les cités grecques’, Revue historique de droit français et étranger 60: 553–76. Gauthier, P. (1984) ‘Les cités hellénistiques. Épigraphie et histoire des institutions et de régimes politiques’, in —æƌ، ı ˙´ ˜Ø Ł   ı æı ¯ººÅ ØŒ ŒÆØ ¸ÆØ ØŒ ¯تæÆçØŒ. Athens: 82–107. Gauthier, P. (1985) Les cités grecques et leurs bienfaiteurs. Athens. Gauthier, P. (1989) ‘Grandes et petits cités: hégémonie et autarcie’, Opus 6–8: 187–202. Gauthier, P. (2001) ‘Les assemblées électorales et le calendrier de Samos à l’époque hellénistique’, Chiron 31: 211–27. Gavazzi, C. (1951) ‘Ricerche sulla prossenia nella Tessaglia’, Epigraphica 13: 50–86. Gawantka, W. (1975) Isopolitie: Ein Bertrag zur Geschichte der zwischenstaatlichen Beziehungen in der grieschischen Antike (Vestigia 22). Munich. Gehrke, H.-J. (1985) Stasis: Untersuchungen zu den inneren Kriegen in den grieschischen Staaten des 5. und 4. Jahrhunderts v. Chr. (Vestigia 35). Munich. Georgiou, Z. (1908) ‘Epigraphai Chiou kai Erython anekdotoi’, Athena 20: 119–354. Gerolymatos, A. (1985) ‘Fourth Century Boiotian Use of the Proxenia in International Relations’, in La Béotie antique, ed. G. Argoud and P. Roesch. Paris: 307–9. Gerolymatos, A. (1986) Espionage and Treason: A Study of the Proxenia in Political and Military Intelligence Gathering in Classical Greece. Amsterdam.

Bibliography

371

Gerolymatos, A. (1987a) ‘A Note on Apollonophanes of Kolophon and IG I3 65’, Euphrosyne 15: 227–33. Gerolymatos, A. (1987b) ‘Ekphantos of Thasos. An example of political use of the Athenian æ Æ’, AncW 15: 45–8. Giddens, A. (1984) The Constitution of Society: Outline of the Theory of Structuration. Cambridge. Giovannini, A. (1994) ‘Greek Cities and Greek Commonwealth’, in Images and Ideologies: Self-Definition in the Hellenistic World, ed. A. Bulloch et al. Berkeley: 265–95. Giovanni, A. (2007) Les relations entre États dans la Grèce antique du temps d’Homère à l’intervention romaine (ca. 700–200 av. J.C.). Stuttgart. Goldhill, S. (1996) ‘Review of M. H. Hansen and K. Raaflaub, eds., (1995) Studies in the Ancient Greek Polis’, BMCR, accessed 10 October 2012:

Graham, A. J. (1964) Colony and Mother City in Ancient Greece. Manchester. Graham, S. (2006) ‘Networks, Agent-Based Models and the Antonine Itineraries: Implications for Roman Archaeology’, JMA 19.1: 45–64. Graindor, P. (1906) ‘Fouilles de Karthaia (1)’, BCH 30: 92–102. Grainger, J. D. (1999) The League of the Aitolians. Leiden. Grandjean, Y. (2011) Le rempart de Thasos. Athens. Grandjean, Y. and Salviat, F. (2000) Guide de Thasos, 2nd edn. Athens. Graninger, D. (2011) ‘ “In as much land as the Pheraioi rule”: A Note on SEG 23 418’, Tyche 26: 87–90. Granovetter, M. S. (1973) ‘The Strength of Weak Ties’, American Journal of Sociology 107.5: 1143–78. Granovetter, M. S. (1983) ‘The Strength of Weak Ties: A Network Theory Revisited’, Sociological Theory 1: 201–33. Grant, T. D. (1999) The Recognition of States: Law and Practice in Debate and Evolution. Westport. Gray, B. (2011) ‘Exile and the Political Cultures of the Greek Polis, c.404–146 bc’ (unpublished DPhil thesis, University of Oxford). Grieb, V. (2008) Hellenistische Demokratie: politische Organisation und Struktur in freien griechischen Poleis nach Alexander dem Grossen. Stuttgart. Gruen, E. S. (1984) The Hellenistic World and the Coming of Rome. Berkeley. Gschnitzer, F. (1958) Abhängige Orte im griechischen Altertum. Munich. Gschnitzer, F. (1973) ‘Proxenos’, RE Suppl. XIII: 629–730. Guarducci, M. (1969) Epigrafia Greca, vol. II. Rome. Le Guen, B. (2001) Les associations de technites dionysiaques à l’époque hellénistique (2 vols.). Nancy and Paris. Günther, W. (1988) ‘Milesische Bürgerrechts- und Proxenieverleihungen der hellenistischen Zeit’, Chiron 18: 383–419.

372

Bibliography

Günther, W. (1995) ‘Ein Proxeniedekret aus Myus’, in Rom und der griechische Osten: Festschrift für Hatto H. Schmitt zum 65. Geburtstag, ed. C. Schubert and K. Broderson. Stuttgart: 87–92. Günther, W. (1999) ‘Funde aus Milet 9: Ein milesisches Proxenie-Dekret des 4. Jahrhunderts v. Chr.’, AA: 475–9. Habicht, Chr. (1961) ‘Falsche Urkunden zur Geschichte Athens im Zeitalter der Perserkriege’, Hermes 89: 1–35. Habicht, Chr. (1982) Studien zur Geschichte Athens in hellenistischer Zeit. Göttingen. Habicht, Chr. (1987) ‘Fremde Richter im ätolischen Delphi?’, Chiron 17: 87–95. Habicht, Chr. (1993) ‘Proxeniedekrete von Akraiphia’, GB 19: 39–43. Habicht, Chr. (1997) Athens from Alexander to Anthony. Cambridge, MA. Habicht, Chr. (1998a) ‘Kleine Beiträge zur altgrieschichen Personenkunde’, REA 100: 487–94. Habicht, Chr. (1998b) Pausanias’ Guide to Ancient Greece, 2nd edn. Berkeley. Habicht, Chr. (2000) ‘Beiträge zu koischen Inschriften’, Chiron 30: 291–301. Habicht, Chr. (2002) ‘Die Ehren der Proxenoi’, MH 59: 13–30. Haensch, R. (2005) ‘Rom und Chersonesus Taurica: Die Beziehung beider Staaten im Lichte der Ehrung des T. Aurelius Calpurnianus Apollonides’, in Ethnic Contacts and Cultural Exchanges North and West of the Black Sea from the Greek Colonization to the Ottoman Conquest, ed. V. Cojocaru. Iaşi: 255–68. Haensch, R. (2009) ‘Not Official, but Permanent: Roman Presence in Allied States—the Examples of Chersonesus Taurica, the Bosporan Kingdom and Sumatar Harabesi’, in Diplomats and Diplomacy in the Roman World, ed. C. Eilers. Leiden: 209–48. Hall, J. (2008) ‘International Relations’, in The Cambridge History of Greek and Roman Warfare. Volume I: Greece, the Hellenistic World and the Rise of Rome, ed. P. Sabin, H. van Wees, and M. Whitby. Cambridge: 85–107. Hamon, P. (2001) ‘À propos de l’institution du Conseil dans les cités grecques de l’époque hellénistique’, REG 114: xvi–xxi. Hamon, P. (2005) ‘Le Conseil et la participation des citoyens: les mutations de la basse époque hellénistique’, in Citoyenneté et participation à la basse époque hellénistique. Actes de la table ronde des 22 et 23 mai 2004, ed. P. Fröhlich and C. Müller. Geneva: 121–44. Hamon, P. (forthcoming) Corpus des inscriptions de Thasos. Documents publics du ive siècle et de l’époque hellénistique. Hanell, K. (1934) Megarische Studien. Lund. Hansen, M. H. (1974) The Sovereignty of the People’s Court in Athens in the Fourth Century B.C. and the Public Action against Unconstitutional Proposals. Odense.

Bibliography

373

Hansen, M. H. (1995a) ‘Boiotian Poleis—a Test Case’, in Sources for the Ancient Greek City-State, ed. M. H. Hansen. Copenhagen: 13–63. Hansen, M. H. (1995b) ‘The “Autonomous City-State”. Ancient Fact or Modern Fiction?’ in Studies in the Ancient Greek Polis, ed. M. H. Hansen and K. Raaflaub. Stuttgart: 21–43. Hansen, M. H. (1998) Polis and City-State: An Ancient Concept and its Modern Equivalent. Copenhagen. Hansen, M. H. (2000) A Comparative Study of Thirty City-State Cultures: An Investigation Conducted by the Copenhagen Polis Centre. Copenhagen. Hansen, M. H. (2006) Polis: An Introduction to the Ancient Greek City-State. Oxford. Hansen, M. H. (2008) ‘Response to J. Rzepka, Reviewing M.H. Hansen, The Return of the Polis’, BMCR, accessed 29 October 2012: Hansen, M. H. and Fischer-Hansen, T. (1994) ‘Monumental Political Architecture in Archaic and Classical Greek Poleis’, in From Political Architecture to Stephanus Byzantius: Sources for the Ancient Greek Polis, ed. D. Whitehead. Stuttgart: 23–90. Hansen, M. H. and Nielsen, T. H. (eds.) (2004) An Inventory of Archaic and Classical Poleis. Oxford. Harris, E. M. (1995) Aeschines and Athenian Politics. Oxford. Harrison, A. R. W. (1968–71) The Law of Athens (2 vols.). Oxford. Harvey, F. D. (1985) ‘Dona ferentes: Some Aspects of Bribery in Greek Politics’, in Crux. Essays in Greek History Presented to G. E. M. de Ste Croix on his 75th Birthday, ed. P. A. Cartledge and F. D. Harvey. London: 76–117. Hasebroek, J. (1933) Trade and Politics in Ancient Greece. London. Hasenohr, C., and Müller, C., (eds.) (2002) Les italiens dans le monde grec: IIe siècle av. J.-C.–Ier siècle ap. J.-C. Paris. Hatzfeld, J. (1919) Les trafiquants italiens dans l’Orient hellénique. Paris. Hatzopoulos, M. B. (1996) Macedonian Institutions Under the Kings (Meletēmata 22, 2 vols.). Athens. Heberdey, R. and Wilhelm, A. (1896) Reisen in Kilikien ausgeführt 1891 und 1892. Vienna. Heller, A. (2006) ‘Les bêtises des Grecs’: Conflits et rivalités entre cités d’Asie et de Bithynie à l’époque romaine (129 a.C.–235 p.C.). Bordeaux. Heller, A. (2009) ‘La cité grecque d’époque impériale: vers une société d’ordres?’, Annales (HSS) 64: 341–73. Hellmann, M.-C. (1992) Recherches sur le vocabulaire de l’architecture grecque, d’après les inscriptions de Délos. Athens. Hennig, D. (1994) ‘Immobilienerwerb durch Nichtbürger in der klassischen und hellenistischen Polis’, Chiron 24: 305–44.

374

Bibliography

Hennig, D. (1997) ‘Die Beherbergung von “Staatsgästen” in der hellenistischen Polis’, Chiron 27: 355–68. Henry, A. S. (1983) Honours and Privileges in Athenian Decrees: The Principal Formulae of Athenian Honorary Decrees. Hildesheim. Herman, G. (1980) ‘The Friends of the Early Hellenistic Rulers, Servants or Officials?’ Talanta 12–3: 103–49. Herman, G. (1987) Ritualised Friendship and the Greek City. Cambridge. Herman, G. (1990) ‘Patterns of Name Diffusion within the Greek World and Beyond’, CQ 40: 349–63. Herman, G. (1997) ‘The Court Society of the Hellenistic Age’, in Hellenistic Constructs: Essays in Culture, History, and Historiography, ed. P. Cartledge, P. Garnsey, and E. S. Gruen. Berkeley: 199–224. Herrmann, P. (1965) ‘Neue Urkunden zur Geschichte von Milet im 2. Jahrhundert v. Chr.’, MDAI(I) 15: 71–117. Herzog, R. (1942) ‘Symbolae Calymniae et Coae’, RFIC 20: 1–20. Hinsley, H. (1986) Sovereignty, 2nd edn. Cambridge. Hitchman, R. and Mack, W. (in preparation) ‘Restoring Lost polis Ethnics in a Fragment of a Proxeny Catalogue from Astypalaia (I. Dor. Ins. 97)’. Hodot, R. (1976) ‘Notes critiques sur le corpus épigraphique de Lesbos’, Etudes d’Archéologie Classique 5: 17–81. Höghammar, K. (1996) ‘Honours for a Koan Judge in the Second Century B.C.’, MDAI(A) 111: 337–61. Horden, P. and Purcell, N. (2000) The Corrupting Sea: A Study of Mediterranean History. Oxford. Hornblower, S. (1982) Mausolus. Oxford. Hornblower, S. (1991–2008) A Commentary on Thucydides (3 vols.). Oxford. Hornblower, S. (2004) ‘ “This was Decided” ( ÆFÆ): The Army as polis in Xenophon’s Anabasis—and Elsewhere’, in The Long March: Xenophon and the Ten Thousand, ed. R. J. Lane Fox. New Haven: 243–63. Huybrechs, F. (1959) ‘Over de proxenie in Lakonië’, RBPh 37: 5–30. Jackson, M. O. (2008) Social and Economic Networks. Princeton. Jacques, F. (1984) Le privilège de liberté. Politique impériale et autonomie municipale dans les cités de l’Occident romain (161–244). Paris. Johnstone, S. (1999) Disputes and Democracy: The Consequences of Litigation in Ancient Athens. Austin. Jones, A. H. M. (1940) The Greek City From Alexander To Justinian. Oxford. Jones, A. H. M. (1964) ‘The Hellenistic Age’, P&P 27: 3–22. Jones, C. P. (1971) Plutarch and Rome. Oxford. Jones, C. P. (1978) The Roman World of Dio Chrysostom. Cambridge, MA. Jones, C. P. (1992) ‘Foreigners in a Hellenistic Inscription of Rhodes’, Tyche 7: 123–32. Jones, C. P. (1999) Kinship Diplomacy in the Ancient World. Cambridge, MA.

Bibliography

375

Jones, C. P. (2001) ‘The Claudian Monument at Patara’, ZPE 137: 161–8. Jones, C. P. (2006) ‘A Letter of Hadrian to Naryka (Eastern Locris)’, JRA 19: 151–62. Kadushin, C. (2012) Understanding Social Networks. Oxford. Kallet-Marx, R. (1995) Hegemony to Empire: The Development of the Roman Imperium in the East from 148 to 62 B.C. Berkeley. Kalliontzis, Y. (2007) ‘Décrets de proxénie et catalogues militaires de Chéronée trouvés lors des fouilles de la basilique paléochrétienne d’Hagia Paraskévi’, BCH 131: 475–514. Kalogéropoulou A. (1974) ‘Un nouveau décret de proxénie de Mégare’, AAA 7: 138–48. Kantor, G. (2008) ‘Roman Law and Local Law in Asia Minor 133 B.C.– A.D. 212’ (unpublished DPhil thesis, University of Oxford). Kantor, G. (2010) ‘Siculus cum Siculo non eiusdem ciuitatis: Litigation between citizens of differentcommunities in the Verrines’, CCG 26: 187–204. Kantor, G. (2011) ‘Review of Fournier (2010)’, JRS 101: 248–9. Keil, B. (1899) ‘Zur Thessalischen Sotairosinschrift’, Hermes 34: 183–202. Keil, J. (1913) ‘Ephesische bürgerrechts- und proxeniedekrete aus dem 4. und 3. jahrhundert v. Chr’, JÖAI 16: 23l–44. Keohane, R. O. (1986) ‘Reciprocity in International Relations’, International Organization 40: 1–27. Keohane, R. O. (1988) ‘International Institutions: Two Approaches’, International Studies Quarterly 32: 379–96. Killen, S. (2008) ‘Die Doppelaxt als Parasemon von Tenedos’, in Vom Euphrat bis zum Bosporus: Kleinasien in der Antike: Festschrift für Elmar Schwertheim zum 65. Geburtstag (Asia Minor Studien 65), ed. E. Winter. Bonn: 367–72. Klaffenbach, G. (1966) Griechische Epigraphik, 2nd edn. Göttingen. Knoepfler, D. (1978) ‘Proxénies béotiennes du IVe siècle’, BCH 102: 375–93. Knoepfler, D. (1995) ‘Les relations des cités eubéennes avec Antigone Gonatas et la chronologie delphique au début de l’époque étolienne’, BCH 119: 137–59. Knoepfler, D. (1997) ‘Le territoire d’Erétrie et l’organisation politique de la cité (dêmoi, chôroi, phylai)’, in The Polis as an Urban Centre and as a Political Community, ed. M. H. Hansen. Copenhagen: 352–449. Knoepfler, D. (2001a) Eretria XI: Décrets érétriens de proxénie et de citoyenneté. Lausanne. Knoepfler, D. (2001b) ‘La cité d’Érétrie et ses bienfaiteurs: réflexions en marge d’un nouveau recueil épigraphique’, CRAI 145: 1355–90. Knoepfler, D. (2002) ‘Un proxène malmené (Xénophon, Hell. VII 2, 16)’, Ktema 27: 71–8.

376

Bibliography

Knoepfler, D. (2005) ‘La prétendue domination d’Antiochos III sur Kéôs: á propos de deux décrets récemment publiés (SEG 48, 1130)’, Chiron 35: 285–308. Koehler, U. (1884) ‘Proxenenliste von Keos’, MDAI(A) 9: 271–8. Koenigs, W. (1981) ‘Bauteile aus Myus im Theater von Milet’, MDAI(I) 31: 143–7. Kolb, F. (1981) Agora und Theater, Volks- und Festsammlung. Berlin. Kralli, I. (2000) ‘Athens and the Hellenistic Kings (338–261 B.C.): The Language of the Decrees’, CQ 50: 113–32. Kreiler, B. M. (2010) ‘Zur Datierung eines Volksbeschlusses von Thasos und zum makedonischen Statthaltersitz im 2. Jh. v. Chr’, ZPE 174: 109–12. Kroll, J. (1993) The Greek Coins (Agora 26). Princeton. Kubitschek, W. (1894) ‘Kemplens Reise von Wien nach Constantinopel (1740)’, Archäologisch-Epigraphische Mitteilungen aus Österreich-Ungarn 17: 47–64. Kuhn, C. (2008) ‘Public Political Discourse in Roman Asia Minor’ (unpublished DPhil thesis, University of Oxford). Kurke, L. (1991) The Traffic in Praise: Pindar and the Poetics of Social Economy. Ithaca. Kvist, K. (2003) ‘Cretan Grants of Asylia: Violence and Protection as Interstate Relations’, C&M 54: 185–222. Labarre, G. (1996) Les cités de Lesbos, aux époques hellénistique et impériale. Paris. Lambert, S. (2011) ‘What was the Point of Inscribed Honorific Decrees in Classical Athens?’, in Sociable Man: Essays on Ancient Greek Social Behaviour in Honour of Nick Fisher, ed. S. Lambert. Swansea: 193–214. Lambert, S. (2012) Inscribed Athenian Laws and Decrees 352/1–322/1 bc. Leiden. Lane Fox, R. J. (2010) ‘Thucydides and Documentary History’, CQ 60: 11–29. Larsen, J. A. O. (1948) ‘ “Foreign Judges” in Cicero Ad Atticum vi. 1. 15’, CPh 43: 187–90. Latycheff, V. L. (1882) ‘Inscriptions de Narthakion’, BCH 6: 580–90. Lauterpacht, H. (1947) Recognition in International Law. Cambridge. Lawton, C. L. (1995) Attic Document Reliefs: Art and Politics in Ancient Athens. Oxford. Leaf, W. (1916) ‘The Commerce of Sinope’, JHS 36: 1–15. Lehman, C. L. (1994) ‘Xenoi, Proxenoi, and Early Greek Traders’, Helios 21: 9–20. Lerat, L. (1934) ‘En-tête d’un décret delphique de proxénie’, BCH 58: 168–72. Lewis, D. M. (1959) ‘Attic Manumissions’, Hesperia 28: 208–38. Lewis, D. M. (1962) ‘The Federal Constitution of Keos’, ABSA 57: 1–4. Lewis, S. (1996) News and Society in the Greek Polis. London. Lhôte, E. (2006) Les lamelles oraculaires de Dodone. Geneva.

Bibliography

377

Lianos, N. (1999) ‘The Area of the Closed Port of Thasos: A Preliminary Report’, 5th International Symposium on Ship Construction in Antiquity (Tropis V). Athens: 261–72. Liddel, P. (2007) Civic Obligation and Individual Liberty in Ancient Athens. Oxford. Liddel, P. (2009) ‘The Decree Cultures of the Ancient Megarid’, CQ 59: 411–36. Liddel, P. (2010) ‘Epigraphy, Legislation and Power within the Athenian Empire’, BICS 53: 99–128. Lintott, A. (1982) Violence, Civil Strife and Revolution in the Classical City 750–330 bc. London. Lintott, A. (1993) Imperium Romanum: Politics and Administration. London. Lintott, A. (2004) ‘Sula: Reprisal by Seizure in Greek Inter-Community Relations’, CQ 54: 340–53. Loraux, N. (2001) The Divided City: On Memory and Forgetting in Ancient Athens, trans. C. Pache with J. Fort. New York. Loughran, C. P. and Raubitschek, A. E. (1947) ‘Three Attic Proxeny Decrees’, Hesperia 16: 78–81. Low, P. (2005) ‘Looking for the Language of Athenian Imperialism’, JHS 125: 93–111. Low, P. (2007) Interstate Relations in Classical Greece: Morality and Power. Cambridge. Luppino, E. (1981) ‘La laicizzazione della prossenia: Il caso di Alcibiade’, CISA 7: 73–9. Ma, J. (2000a) ‘Fighting Poleis of the Hellenistic World’, in War and Violence in Ancient Greece, ed. H. van Wees. Swansea: 337–76. Ma, J. (2000b) ‘Public Speech and Community in the Euboicus’, in Dio Chrysostom: Politics, Letters, and Philosophy, ed. S. Swain. Oxford: 108–24. Ma, J. (2002) Antiochos III and the Cities of Western Asia Minor, rev. edn. Oxford. Ma, J. (2003) ‘Peer Polity Interaction in the Hellenistic Age’, P&P 180: 9–39. Ma, J. (2005) ‘The Many Lives of Eugnotos of Akraiphia’, in Studi Ellenistici 16, ed. B. Virgilio. Pisa: 141–91. Ma, J. (2009) ‘Empires, Statuses and Realities’, in Interpreting the Athenian Empire, ed. J. Ma, N. Papazarkadas and R. Parker. London: 125–48. Ma, J. (2011) ‘Court, King, and Power in Antigonid Macedonia’, in Brill’s Companion to Ancient Macedon: Studies in the Archaeology and History of Macedon, 650 bc–300 ad, ed. R. J. Lane Fox. Leiden: 521–43. Ma, J. (2012a) ‘Honorific Statues and Hellenistic History’, in Imperialism, Cultural Politics, & Polybius, ed. C. Smith and L. M. Yarrow. Oxford: 230–51.

378

Bibliography

Ma, J. (2012b) ‘Epigraphy and the Display of Authority’, in Epigraphy and the Historical Sciences, ed. J. Davies and J. Wilkes. Oxford: 133–58. Ma, J. (2013) Statues and Cities: Honorific Portraits and Civic Identity in the Hellenistic World. Oxford. MacDowell, D. M. (2004) ‘Epikerdes of Kyrene and the Athenian Privilege of Ateleia’, ZPE 150: 127–33. Mack, W. (2011) ‘The Proxeny-lists of Karthaia’, REA 113: 319–44. Mack, W. (2012) ‘The Eresian Catalogue of Proxenoi (IG XII suppl. 127)’, ZPE 180: 217–29. Mack, W. (2014) ‘Consulting the Oracle at Dodona about a Female Proxenos? Lhôte no. 15 Reconsidered (SEG 56 663)’ ZPE 188: 155–8. Mack, W. (forthcoming) ‘Shepherds Beating the Bounds? Territorial Identity at a Dependent Community (IPriene 361–3)’, JHS 135. Mackil, E. (2004) ‘Wandering Cities: Alternatives to Catastrophe in the Greek Polis’, AJA 108: 493–516. Mackil, E. (2008) ‘A Boiotian Proxeny Decree and Relief in the Museum of Fine Arts, Boston and Boiotian-Lakonian Relations in the 360s’, Chiron 38: 157–94. Mackil, E. (2013) Creating a Common Polity: Religion, Economy, and Politics in the Making of the Greek Koinon. Berkeley. MacMullen, R. (1982) ‘The Epigraphic Habit in the Roman Empire’, AJPh 103: 233–46. Maddoli, G. (2007) ‘Epigrafi di Iasos: Nuovi Supplementi’, PP 62: 193–372. Magie, D. (1950) Roman Rule in Asia Minor: To the End of the Third Century after Christ. Princeton. Magnetto, A. (1997) Gli arbitrati interstatali greci, vol. I: Dalle origine al 338 a.C.. Florence. Magnetto, A. (2008) L’arbitrato di Rodi fra Samo e Priene. Pisa. Maier, F. G. (1959) Griechische Mauerbauinschriften, (2 vols.). Heidelberg. Maiuri, A. (1925) Nuova silloge epigrafica di Rodi e Cos. Florence. Malkin, I. (2011) A Small Greek World: Networks in the Ancient Mediterranean. Oxford. Malkin, I., Constantakopoulou, C., and Panagopoulou, K. (2009) Greek and Roman Networks in the Mediterranean. London. Manganaro, G. (1963) ‘Tre tavole di bronzo con decreti di proxenia del Museo di Napoli e il problema dei proagori in Sicilia’, Kokalos 9: 205–20. Marasco, G. (1983) ‘Alessandro, i diadochi e il culto dell’eroe eponimo’, Prometheus 9: 57–63. March, J. G. and Olsen, J. P. (1989) Rediscovering Institutions: The Organizational Basis of Politics. New York. March, J. G. and Olsen, J. P. (2006) ‘The Logic of Appropriateness’, in The Oxford Handbook of Public Policy, ed. M. Moran, M. Rein, and R. E. Goodin. Oxford: 689–708.

Bibliography

379

Marek, C. (1977) ‘Der Geldumlauf der Stadt Histiaia und seine Bedeutung für die Verteilung ihrer Proxenoi’, Talanta 8–9: 74–9. Marek, C. (1984) Die Proxenie. Frankfurt. Marek, C. (1985) ‘Handel und Proxenie’, MBAH 4: 67–78. Marek, C. (1988) ‘Review of Gerolymatos (1986)’, Gnomon 60: 594–8.  and Markoulaki, S. (2000) ‘Decreto inédito de Martínez Fernández, A. proxenía de Kísamos, Creta’, ZPE 133: 103–8. Mataranga, K. (1993) ‘Un étrange “Proxène” chez Énée le Tacticien’, CCG 4: 223–7. Matthaei, A. and Zimmermann, M. (eds.) (2008) Stadtbilder im Hellenismus. Die hellenistische Polis als Lebensform Bd. 1. Berlin. Matthaiou, A. P. (1983) ‘A Naxian Honorary Decree’, Horos 1: 39–44. Matthaiou, A. P. (1998) ‘æ E KتæÆç  —æı’, Horos 10–12: 423–36. Mattingly (2007) ‘Two Fifth-Century Epigraphic Texts Revisited’, ZPE 162: 203–10. Mauss, M. (1954) The Gift: Forms and Functions of Exchange in Archaic Societies. London. McLean, B. H. (2002) An Introduction to Greek Epigraphy of the Hellenistic and Roman Periods from Alexander the Great down to the Reign of Constantine (323 B.C.–A.D. 337). Ann Arbor. Meiggs, R. (1949) ‘A Note on Athenian Imperialism’, CR 63: 9–12. Meiggs, R. (1972) The Athenian Empire. Oxford. Meier, L. (2012) Die Finanzierung öffentlicher Bauten in der hellenistischen Polis. Berlin. Meier, M. H. E. (1843) Commentatio de proxenia, sive, De publico Graecorum hospitio. Halle. Mendoni, L. (1990) ‘Addenda et Corrigenda as Inscriptiones Ceae’, in POIKILA (Meletēmata 10). Athens: 287–307. Mendoni, L. (2007a) ‘Proxenikos katalogos apo tin Karthaia’, in Politeuesthai tous Keious kata poleis: Katalogos ’Ekthesis 16 Iouliou–30 Septembriou 2007, ed. M. Lagogianni-Georgakarakou. Athens: 44–5. Mendoni, L. (2007b) ‘ıº Å —æ øªæÆçÆ Å ˚ Æ: ˚ Ø Œ ı æ ’, in   Æ æªÆ. ØÅØŒe  ªØÆ  ŒÆŁÅªÅ BÆ Øº ˚. ¸ÆæØ ıŒÅ (`æåÆت ø Æ Supplement 5), ed. E. Semantone-Bournia. Athens: 529–58. Mendoni, L. and Mourtzas, N. D. (1990) ‘An Archaeological Approach to Coastal Sites: The Example of the Ancient Harbour of Karthaia’, Parnassos 32: 387–403. Merriam, A. C. (1885) ‘Inscribed Sepulchral Vases from Alexandria’, AJA 1: 18–33. Merritt, B. D. (1977) ‘A Proxeny Decree Restudied (IG I2 85)’, ZPE 25: 289–95.

380

Bibliography

Meyer, E. (2010) Metics and the Athenian Phialai-inscriptions: A Study in Athenian Epigraphy and Law. Stuttgart. Migeotte, L. (1984) L’emprunt public dans les cités grecques: Recueil des documents et analyse critique. Quebec and Paris. Migeotte, L. (1992) Les souscriptions publiques dans les cités grecques. Quebec and Geneva. Migeotte, L. (2004) ‘La mobilité des étrangers en temps de paix en Grèce ancienne’ in La mobilité des personnes en Méditerranée de l’Antiquité à l’époque moderne: procédures de contrôle et documents d’identification, ed. C. Moatti. Rome: 615–48. Migeotte, L. (2009) The Economy of the Greek Cities: From the Archaic Period to the Early Roman Empire. Berkeley. Millar, F. (1977) The Emperor in the Roman World (31 bc–ad 337). London. Millar, F. (1993) ‘The Greek City in the Roman Period’, in The Ancient Greek City-State, ed. M. H. Hansen. Copenhagen: 232–60. Millar, F. (1996) ‘Emperors, Kings and Subjects: The Politics of Two-Level Sovereignty’, SCI 15: 159–73. Miller, S. G. (1974) ‘A Family of Halikarnassians in North-Central Greece’, AJA 78: 151–2. Minon, S. (2007) Les inscriptions éléennes dialectales, VIe–IIe siècle avant J.-C., vol. I. Geneva. Misaelidou-Despotidou, V. (1980) ‘Proxenic Decree from Ancient Pharsalos’, ad 35: 226–40. Mitchell, L. (1997a) Greeks Bearing Gifts: The Public Use of Private Relationships in the Greek World, 435–323 bc. Cambridge. Mitchell, L. (1997b) ‘غÆ, ¯h ØÆ and Greek Interstate Relations’, Antichthon 31: 28–44. Mitchell, L. (2009) ‘The Rules of the Game: Three Studies in Friendship, Equality and Politics’, in Greek History and Epigraphy: Essays in Honour of P.J. Rhodes, ed. L. Mitchell and L. Rubinstein. Swansea: 1–32. Moggi, M. (1995) ‘I Proxenoi e la Guerra nel V Secolo A.C.’, in Les relations internationales: Actes du Colloque de Strasbourg, 15–17 juin 1993, ed. E. Frézoulas et A. Jacquemin. Paris: 143–59. Monceaux, P. (1886) Les proxénies grecque. Paris. Moreno, A. (2007) Feeding the Democracy: The Athenian Grain Supply in the Fifth and Fourth Centuries bc. Oxford. Moretti, L. (1977) ‘Mileto, le sue colonie e l’istituto dell’ evergesia’, RFIC 105: 5–11. Mosley, D. J. (1971a) ‘Cimon and the Spartan Proxeny’, Athenaeum 49: 431–2. Mosley, D. J. (1971b) ‘Spartan Kings and Proxeny’, Athenaeum 49: 433–5. Müller, C. (2005) ‘La procédure d’adoption des décrets en Béotie de la fin du IIIe s. av. J.-C. au Ier s. apr. J.-C’, in Citoyenneté et participation à la basse

Bibliography

381

époque hellénistique. Actes de la table ronde des 22 et 23 mai 2004, ed. P. Fröhlich and C. Müller. Geneva: 95–119. Müller, C. (2010) D’Olbia à Tanaïs: territoires et réseaux d’échanges dans la mer Noire septentrionale aux époques classique et hellénistique. Paris. Nichols, J. (2011) ‘Hospitality among the Romans’, in The Oxford Handbook of Social Relations in the Roman World, ed. M. Peachin. Oxford: 422–37. Nielsen, T. H. (2002) Arkadia and its Poleis in the Archaic and Classical Periods (Hypomnemata 140). Göttingen. North, D. C. (1990) Institutions, Institutional Change and Economic Performance. Cambridge. North, D. C. (1991) ‘Institutions’, The Journal of Economic Perspectives 5: 97–112. North, D. C. (1995) ‘Five Propositions about Institutional Change’, in Explaining Social Institutions, ed. J. Knight and I. Sened. Michigan: 15–26. Ober, J. (1983) ‘Two Ancient Watchtowers above Aigosthena in the Northern Megarid’, AJA 87: 387–92. Ober, J. (1987) ‘Early Artillery Towers: Messenia, Boiotia, Attica, Megarid’, AJA 91: 569–604. Ober, J. (1989) Mass and Elite in Democratic Athens: Rhetoric, Ideology, and the Power of the People. Princeton. Ober, J. (1996) The Athenian Revolution: Essays on Ancient Greek Democracy and Political Theory. Princeton. Ober, J. (2008) Democracy and Knowledge: Innovation and Learning in Classical Athens. Princeton. Ober, J. (2010) ‘Wealthy Hellas’, TAPA 140: 241–86. Ogilvie, R. M. (1965) A Commentary on Livy Books 1–5. Oxford. Oliver, G. J. (2007) War, Food, and Politics in Early Hellenistic Athens. Oxford. Oliver, J. H. (1954) ‘The Roman Governor’s Permission for a Decree of the Polis’, Hesperia 23: 163–7. Osborne, M. (1981–3) Naturalization in Athens. Brussels. Osborne, M. (2012) ‘Secretaries, Psephismata and Stelae in Athens’, Ancient Society 42: 33–59. Osborne, R. (1985) Demos: The Discovery of Classical Attica. Cambridge. Osborne, R. (1999) ‘Inscribing Performance’, in Performance Culture and Athenian Democracy, ed. S. Goldhill and R. Osborne. Cambridge: 341–58. Ostwald, M. (1969) Nomos and the Beginnings of the Athenian Democracy. Oxford. Ostwald, M. (1982) Autonomia: Its Genesis and Early History. Chicago. Panessa, G. P (1999) Philiai: l’amicizia nelle relazioni interstatali dei Greci. Pisa. Papageorgiadou-Banis, C. (1997) The Coinage of Keos. Paris.

382

Bibliography

Papazarkadas, N. (2007) ‘An Honorific Decree from Classical Siphnos’, REA 109: 137–46. Papazarkadas, N. (2009) ‘Epigraphy and the Athenian Empire: Reshuffling the Chronological Cards’, in Interpreting the Athenian Empire, ed. J. Ma, N. Papzarkadas, and R. Parker. London: 67–88. Papazarkadas, N. (2012) Review of Meyer (2009), CR 62: 553–5. Papazarkadas, N. and Thonemann, P. (2008) ‘Athens and Kydonia: Agora I 7602’, Hesperia 77: 73–87. Paris, P. (1892) Élatée: la ville, le temple d’Athéna Cranaia. Paris. Parke, H. W. (1957) ‘A Note on the Spartan Embassy to Athens (408/7 B. C.)’, CR 7: 106–7. Parker, R. (1985) ‘Greek States and Greek Oracles’, in Crux. Essays in Greek History presented to G. E. M. de Ste Croix on his 75th Birthday, ed. P. A. Cartledge and F. D. Harvey. London: 298–326. Parker, R. (1996) Athenian Religion: A History. Oxford. Parker, R. (1998) Cleomenes on the Acropolis: An Inaugural Lecture Delivered before the University of Oxford on 12 May 1997. Oxford. Parker, R. (2011) On Greek Religion. Ithaca. Paschidis, P. (2008) Between City and King: Prosopographical Studies on the Intermediaries Between the Cities of the Greek Mainland and the Aegean and the Royal Courts in the Hellenistic Period, 322–190 bc (Meletēmata 59). Athens. Paton, W. R. and Hicks, E. L. (1891) The Inscriptions of Cos. Oxford. Payne, M. J. (1984) ‘` ¯` ¯˝¯˚¯˝ : Honors for Romans and Italians in Greece from 260 to 27 bc’ (unpublished PhD thesis, Michigan State University). Pečirka, J. (1966) The Formula for the Grant of Enktesis in Attic Inscriptions. Prague. Peek, W. (1969) Inschriften von den dorischen Inseln. Berlin. Perassi, C. and Novarese, M. (2006) ‘La monetazione di Melita e di Gaulos. Note per un riesame’, in L’Africa Romana 16, ed. A. Akerraz, et al. Rome, vol. IV: 2377–404. Perdrizet, P. (1896) ‘—Ææ ÅÆ de villes sur des stèles de proxénie’, BCH 20: 549–62. Perdrizet, P. (1899) ‘Inscriptions d’Acraephiae (Suite de BCH, XXII, p. 241–260.)’, BCH 23: 90–6. Perlman, P. (1995) ‘ “Theorodokountes en tais polesin”: Panhellenic Epangelia and Political Status’, in Sources for the Ancient Greek City-State, ed. M. H. Hansen. Copenhagen: 113–64. Perlman, P. (1996) ‘Polis hypekoos: The Dependent Polis and Crete’, in Introduction to an Inventory of Poleis, ed. M. H. Hansen. Copenhagen: 233–87.

Bibliography

383

Perlman, P. (2000) City and Sanctuary in Ancient Greece: The Theorodokia in the Peloponnese. Göttingen. Perlman, S. (1958) ‘A Note on the Political Implications of Proxenia in the Fourth Century B.C.’, CQ 8: 185–91. Perlman, S. (1976) ‘On Bribing Athenian Ambassadors’, GRBS 17: 223–33. Perrin-Saminadayar, E. (2007) Éducation, culture et société à Athènes: les acteurs de la vie culturelle athénienne (229–88). Paris. Phillipson, C. (1911) The International Law and Custom of Ancient Greece and Rome, vol. I. London. Piccirilli, L. (1973) Gli arbitrati interstatali greci. vol. I: Dalle origine al 338 a.C.. Florence. Plassart, A. (1921) ‘Inscriptions de Delphes: la liste des Théorodoques’, BCH 45: 1–85. Plassart, A. and Blum, G. (1914) ‘Orchomène d’Arcadie: Fouilles de 1913: Inscriptions’, BCH 38: 447–78. Pocock, J. G. A. (1973) ‘Verbalizing a Political Act: Toward a Politics of Speech’, Political Theory 1: 27–45. de Polignac, F. (1995) Cults, Territory, and the Origins of the Greek CityState, trans. J. Lloyd. Chicago. Pouilloux, J. (1948) ‘Inscriptions de Thasos’, BCH 71–2: 262–72. Pouilloux, J. (1974) ‘Les décrets delphiques pour Matrophanés de Sardes’, BCH 98: 159–69. Prag, J. R. W. (2002) ‘Epigraphy by Numbers: Latin and the Epigraphic Culture in Sicily’, in Becoming Roman, Writing Latin? Literacy and Epigraphy in the Roman West, ed. A. E. Cooley (JRA Supplement 48). Rhode Island: 15–31. Price, S. (1984) Rituals and Power: The Roman Imperial Cult in Asia Minor. Cambridge. Pridik, A. M. (1892) De Cei Insulae Rebus. Berlin. Pritchett, W. K (1991) The Greek State at War: Part V. Berkeley. Psoma, S. (2008) ‘An Honorary Decree from Thasos (IG XII 8, 267) and the Samothracian Peraia During the Hellenistic Period’, in Thrakika Zetemata 1 (Meletēmata 58), ed. L. D. Loukopoulo and S. Psoma. Paris: 121–38. Puech, B. (2012) ‘Derniers affichages de l’octroi du droit de cité à l’époque impériale’, in Patrie d’origine et patries électives: les citoyennetés multiples dans le monde grec d’époque romaine, ed. A. Heller et A.-V. Pont. Bordeaux: 195–212. Putnam, R. D., with R. Leonardi and R. Nanetti (1993) Making Democracy Work: Civic Traditions in Modern Italy. Princeton. Quass, F. (1993) Die Honoratiorenschicht in den Städten des griechischen Ostens: Untersuchungen zur politischen und sozialen Entwicklung in hellenistischer und römischer Zeit. Stuttgart.

384

Bibliography

Rauh, N. K. (1993) The Sacred Bonds of Commerce: Religion, Economy, and Trade Society at Hellenistic Roman Delos, 166–87 bc. Amsterdam. Reger, G. (1991) ‘Apollodorus of Cyzicus and his Delian Garden’, GRBS 32: 229–37. Reger, G. (1992) ‘Athens and Tenos in the Early Hellenistic Age’, CQ 42: 365–83. Reger, G. (1994) Regionalism and Change in the Economy of Independent Delos, 314–167 B.C. Berkeley. Reger, G. (2003) ‘Aspects of the Role of Merchants in the Political Life of the Hellenistic World’, in Mercanti e politica nel mondo antico, ed. C. Zaccagnini. Rome: 165–97. Reger, G. (2004) ‘Sympoliteiai in Hellenistic Asia Minor’, in The GrecoRoman East: Politics, Culture, Society (Yale Classical Studies 31), ed. S. Colvin. Cambridge: 145–80. Reger, G. (2011) ‘Inter-Regional Economies in the Aegean Basin’, in The Economies of Hellenistic Societies, First to Third Centuries bc, ed. Z. H. Archibald, J. K. Davies, and V. Gabrielsen. Oxford: 368–89. Reger, G. and Risser, M. (1991) ‘Coinage and Federation in Hellenistic Keos’, in Landscape Archaeology as Long-Term History, ed. J. F. Cherry, J. L. Davis, and E. Mantzourani. Los Angeles: 305–17. Reiter, H. A. (1991) Athen und die Poleis des Delisch-Attischen Seebundes: die Proxenoi und Euergetai des attischen Demos in den Poleis des DelischAttischen Seebundes im Licht der attischen Proxenie-und Euergesiebeschlüsse des 5. Jahrhunderts v. Chr. Regensburg. Renfew, C. (1986) ‘Introduction’, in Peer Polity Interaction and Socio-Political Change, ed. C. Renfew and J. F. Cherry. Cambridge: 1–18. Reus-Smit, Chr. (2012) ‘Constructivism and the English School’, in Theorising International Society: English School Methods, ed. C. Navari. Basingstoke: 487–509. Rhodes, P. J. (1995) ‘Epigraphical Evidence: Laws and Decrees’, in Sources for the Ancient Greek City-State, ed. M. H. Hansen. Copenhagen: 91–112. Rhodes, P. J., with Lewis, D. M. (1997) The Decrees of the Greek States. Oxford. Rigsby, K. (1996) Asylia: Territorial Inviolability in the Hellenistic World. Berkeley. Ritti, T. (1969) ‘Sigle ed emblemi sui decreti onorari greci’, MAL 14: 259–360. Robert, L. (1951) ‘La circulation des monnaies d’Histiée’, in Études de numismatique grecque. Paris: 179–216. Robert, L. (1963) Noms indigènes dans l’Asie-Mineure gréco-romaine. Paris. Robert, L. (1966) Monnaies antiques en Troade. Geneva. Robert, L. (1969) ‘Laodicée du Lycos: Les inscriptions’, in Laodicée du Lycos: Le nymphée: Campagnes 1961–1963, J. des Gagniers. Paris: 247–389. Robert, J. and Robert, L. (1983). Fouilles d’Amyzon en Carie. Paris.

Bibliography

385

Robert, J. and Robert, L. (1989) Claros: Décrets hellénistiques I. Paris. Robinson, D. M. (1905) ‘Greek and Latin Inscriptions from Sinope and Environs’, AJA 9: 294–333. Robinson, D. M. (1906) ‘Ancient Sinope’, AJP 27: 125–53; 245–79. Roebuck, D. (2001) Ancient Greek Arbitration. London. Roesch, P. (1966) ‘Inscriptions du Musée de Thébes’, REA 68: 61–85. Roesch, P. (1970) ‘Inscriptions béotiennes du Musée de Thèbes’, BCH 94: 139–60. Roesch, P. (1984) ‘Un décret inédit de la ligue Thébaine et la flotte d’Épaminondas’, REG 97: 45–60. Roesch, P. (1989) ‘Décrets de proxénie d’Orchomène en Béotie’, in Architecture et poésie dans le monde grec: Hommages à Georges Roux, ed. R. Étienne et al. Paris: 219–24. Rood, T. (1998) Thucydides: Narrative and Explanation. Oxford. Rostovzeff, M. I. (1941) Social and Economic History of the Hellenistic World (3 vols.). Oxford. Roussel, P. (1916) Délos: colonie athénienne. Paris. Rousset, D. (2002) ‘Inscriptions hellénistiques d’Amphissa’, BCH 126: 83–95. Rousset, D. (2006) ‘Les inscriptions de Kallipolis d’Étolie’, BCH 130: 381–434. Ruggeri, C. (2004) ‘Senofonte a Scillunte’, Athenaeum 92: 451–66. Ruschenbusch, E. (1982) ‘IG XII 5 609: Eine Bürgerliste von Iulis und Koresia auf Keos’, ZPE 48: 175–88. Russell, F. S. (1999) Information Gathering in Classical Greece. Ann Arbor. Rutherford, I. (2004) ‘The Keian Theoria to Delphi: Neglected Data from the Accounts of the Delphic Naopoioi (CID 2.1–28)’, ZPE 147: 107–14. Rutherford, I. (2007) ‘Network Theory and Theoric Networks’, Mediterranean Historical Review 22: 23–37. Rutherford, I. (2013) State-Pilgrimage in Ancient Greece: Theōriā and Theōroi. Cambridge. Saba, S. (2010) ‘Hellenistic Institutions and Law: A Survey (2000–2010)’, Dike 12–13: 271–302. Salomon, N. (1997) Le Cleruchie di Atene: Caratteri e funzione. Pisa. Samama, E. (2003) Les médecins dans le monde grec: sources épigraphiques sur la naissance d’un corps medical. Geneva. Santangelo, F. (2009) ‘What Did the Cloatii Do for Gytheum?’, Historia 58: 361–6. Saprykin, S. (1999) ‘Proxenic Decrees of Tauric Chersonesus and the SeaRoutes in Pontus Euxeinus’, OTerr 5: 31–41. Savalli-Lestrade, I. (1998) Les philoi royaux dans l’Asie hellénistique. Geneva. Savalli-Lestrade, I. (2008) ‘Devenir une cité: Poleis nouvelles et aspirations civiques en Asie Mineure à la basse époque hellénistique’, in Citoyenneté et

386

Bibliography

participation à la basse époque hellénistique. Actes de la table ronde des 22 et 23 mai 2004, ed. P. Fröhlich and C. Müller. Geneva: 9–37. Schachter, A. and Slater, W. J. (2007) ‘A Proxeny Decree from Koroneia, Boiotia, in Honour of Zotion son of Zotion, of Ephesos’, ZPE 163: 81–95. Schaefer, H. (1932) Staatsform und Politik: Untersuchungen zur griechischen Geschichte des 6. und 5. Jahrhunderts. Leipzig. Schlaifer, R. (1944) ‘A Fragment of a Proxeny Decree from Ios’, Hesperia 13: 22. Schneider, A. M. (1938) Die Stadtmauer von Iznik (Nicea). Berlin. Scholten, J. B. (2000) The Politics of Plunder: Aitolians and their Koinon in the Early Hellenistic Era, 279–217 bc. Berkeley. Schuler, C. (2010) ‘Sympolitien in Lykien und Karien’, in Hellenistic Karia: Proceedings of the First International Conference on Hellenistic Karia, Oxford, 29 June–2 July 2006, ed. R. van Bremen and J.-M. Carbon. Bordeaux: 393–414. Schwahn, W. (1931) ‘Das Bügerrecht der sympolitischen Bundesstaaten’, Hermes 66: 97–118. Scibona, G. (2009) ‘Decreto sacerdotale per il conferimento della euerghesia a Nemenios in Halaesa’, in Alaisa-Halaesa. Scavi e ricerche (1970–2007), ed. G. Scibona and G. Tigano. Messina: 97–112. Scott, J. (2000) Social Network Analysis: A Handbook, 2nd edn. London. Seibert, J. (1979) Die politischen Flüchtlinge und Verbannten in der griechischen Geschichte: von den Anfängen bis zur Unterwerfung durch die Römer. Darmstadt. Shear, J. (2011) Polis and Revolution: Responding to Oligarchy in Classical Athens. Cambridge. Shear, T. L. (1978) Kallias of Sphettos and the Revolt of Athens in 286 B. C. (Hesperia Supplement 17). Princeton. Sheedy, K. A. (2006) The Archaic and Early Classical Coinages of the Cyclades. London. Sherwin-White, A. N. (1973) The Roman Citizenship, 2nd edn. Oxford. Sherwin-White, S. (1978) Ancient Cos: An Historical Study from the Dorian Settlement to the Imperial Period. Göttingen. Shipley, G. (2011) Pseudo-Skylax’s Periplous: The Circumnavigation of the Inhabited World: Text, Translation and Commentary. Exeter. Slater, W. J. (ed.) (1986) Aristophanis Byzantii fragmenta. Berlin. Slater, W. J. and Summa, D. (2006) ‘Crowns at Magnesia’, GRBS 46: 275–99. Souris, G. A. (1982) ‘The Size of the Provincial Embassies to the Emperor under the Principate’, ZPE 48: 235–44. Sourvinou-Inwood, C. (1990) ‘What is Polis Religion?’, in The Greek City from Homer to Alexander, ed. O. Murray and S. Price. Oxford: 295–322. de Souza, P. (1999) Piracy in the Graeco-Roman World. Cambridge.

Bibliography

387

Spawforth, A. (ed.) (2007) The Court and Court Society in Ancient Monarchies. Cambridge. de Ste Croix, G. E. M. (1961) ‘Notes on Jurisdiction in the Athenian Empire’, CQ 11: 94–112. de Ste Croix, G. E. M. (1981) The Class Struggle in the Ancient Greek World: From the Archaic Age to the Arab Conquests. London. Strassler, R. B. (ed. and trans.) (2009) The Landmark Xenophon’s Hellenika. New York. Stroud, R. (1963) ‘A Fragment of an Inscribed Bronze Stele from Athens’, Hesperia 32: 138–43. Swinnen, W. (1970) ‘Herakleitos of Halikarnassos, an Alexandrian Poet and Diplomat?’, AnSoc 1: 39–52. Swoboda, H. (1890) Die griechische Volksbeschlüsse: epigraphische Untersuchungen. Leipzig. Szanto, E. (1892) Das griechische Bürgerrecht. Freiburg. Talmon, S. (1998) Recognition of Governments in International Law: With Particular Reference to Governments in Exile. Oxford. Tataki, A. B. (1998) Macedonians Abroad: A Contribution to the Prosopography of Ancient Macedonia. Athens. Thériault, G. (1996) Le Culte d’Homonoia dans les cités grecques. Lyon. Thomas, R. (1992) Literacy and Orality in Ancient Greece. Cambridge. Thonemann, P. J. (2006) ‘Neilomandros’, Chiron 36: 11–43. Thonemann, P. J. (2007) ‘Magnesia and the Greeks of Asia (I.Magnesia 16.16)’, GBRS 47: 151–60. Thonemann, P. J. (2011) The Maeander Valley: A Historical Geography from Antiquity to Byzantium. Cambridge. Tissot, Ch. J. (1863) Des proxénies grecques et de leur analogie avec les institutions consulaires modernes. Dijon. Tod, M. N. (1913) International Arbitration Amongst the Greeks. Oxford. Todd, S. C. (1993) The Shape of Athenian Law. Oxford. Tracey, S. V. (2003) ‘A Major Athenian Letter-Cutter of ca. 410 to ca. 380: The Cutter of IG II2 17’, in Gestures: Essays in Ancient History, Literature, and Philosophy Presented to Alan L. Boegehold on the Occasion of His Retirement and his Seventy-fifth Birthday, ed. G. W. Bakewell and J. P. Sickinger. Oxford: 351–63. Traill, J. S. (1968) ‘The Bouleutic List of 303/2 B.C.’, Hesperia 37: 1–24. Tréheux, J. (1953) ‘Décret de Lampsaque trouvé à Thasos’, BCH 77: 426–43. Trevett, J. (1992) Apollodorus, the Son of Pasion. Oxford. Trevett, J. (1999) ‘Demosthenes and Thebes’, Historia 48: 184–202. Tzifopoulos, Y. (2010) ‘Proxeny and Citizenship Awards by Sybaritos, Crete’, in Studies in Greek Epigraphy and History in Honor of Stephen V. Tracey, ed. G. Reger, F. X. Ryan, and F. Winters. Paris: 355–68.

388

Bibliography

Ullrich, F. W. (1822) Disquisitionis de proxenia, sive, publico apud Graecos hospitio. Berlin. Vanseveren, J. (1937) ‘Inscriptions d’Amorgos et de Chios’, RPh 63: 313–47. Veligianni-Terzi, C. (1997) Wertbegriffe in den attischen Ehrendekreten der Klassischen Zeit. Stuttgart. Veyne, P. (1990) Bread and Circuses: Historical Sociology and Political Pluralism, trans. B. Pierce. London. Virgilio, B. (1969) ‘Rassegna di studi sulle prossenie greche’, RFIC 97: 494–501. Virgilio, B. (2011) Le roi écrit: le correspondance du soverain hellénistique, suivie de deux lettres d’Antiochos III à partir de Louis Robert et d’Adolf Wilhelm (Studi ellenistici 25). Pisa. Vincent, A. (1987) Theories of the State. Oxford. Vlassopoulos, K. (2007) Unthinking the Greek Polis: Ancient Greek History Beyond Eurocentrism. Cambridge. Vlassopoulos, K. (2011) ‘Review of Meyer (2010)’, BMCR, accessed on 12 September 2012: Vujčić, N. (2009) ‘Greek Popular Assemblies and the Discourses of Dio of Prusa’, EA 42: 157–69. Wacker, C. (1996) ‘Der akarnanischen Hafenplatz Panormos’, in Akarnanien: Eine Landschaft im antiken Griechenland, ed. P. Berktold, J. Schmid, and C. Wacker. Würzburg: 209–13. Walbank, F. W. (1981) The Hellenistic World. London. Walbank, M. B. (1978) Athenian Proxenies of the Fifth Century B.C. Toronto. Walbank, M. B. (1982) ‘An Athenian Decree Re-considered. Honours for Aristoxenos and Another Boiotian’, EMC 26: 259–74. Wallace, W. P. (1956) The Euboian League and its Coinage. New York. Wallace, W. P. (1962) ‘The Meeting-point of Histiaian and Macedonian Tetrobols’, NC 2: 17–22. Wallace, M. B. (1970) ‘Early Greek Proxenoi’, Phoenix 24: 189–208. Walser, A. V. (2008) Bauern und Zinsnehmer: Politik, Recht und Wirtschaft im frühhellenistischen Ephesos. Munich. Walsh, J. J. (1993) ‘Bones of Contention: Pharsalus, Phthiotic Thebes, Larisa Cremaste, Echinus’, CP 88: 35–46. Waltz, K. (1979) Theory of International Politics. New York. Wasserman, S. and Faust, K. (1994) Social Network Analysis: Methods and Applications. Cambridge. Welles, C. B. (1970) ‘Review of Bogaert, Banques et banquiers dans les cités grecques’, Gnomon 42: 804–10. Wendt, A. (1992) ‘Anarchy is What States Make of It: The Social Construction of Power Politics’, International Organization 41: 335–70. Wendt, A. (1999) Social Theory of International Politics. Cambridge.

Bibliography

389

Whitehead, D. (1977) The Ideology of the Athenian Metic (Cambridge Philological Society Supplement 4). Cambridge. Whitehead, D. (1983) ‘Competitive Outlay and Community Profit: çغØÆ in Democratic Athens’, C&M 34: 55–74. Whitehead, D. (1993) ‘Cardinal Virtues: The Language of Public Approbation in Democratic Athens’, C&M 44: 37–75. Whitehead, D. (2000) Hypereides: The Forensic Speeches. Oxford. Wiemer, H.-U. (2010) ‘Review of Carlsson (2010)’, Sehepunkte 10, accessed on 23 October 2012: Wilamowitz-Moellendorf, U. von (1909) Nordionische Steine. Berlin. Wilhelm, A. (1909) Beiträge zur griechischen Inschriftenkunde mit einem Anhange über die öffentliche Aufzeichung von Urkunden. Vienna. Wilhelm, A. (1939) Attische Urkunden IV. Vienna. Wilhelm, A. (1942) ‘Proxenie und Euergesie’, in his Attische Urkunden V. Vienna: 11–86. Will, E. (1979–82) Histoire politique du monde hellénistique, 2nd edn. (2 vols.). Nancy. Wilson, A. (2009) ‘Approaches to Quantifying Roman Trade’, in Quantifying the Roman Economy: Methods and Approaches, ed. A. Bowman and A. Wilson. Oxford: 213–49. Wilson, A. J. N. (1966) Emigration from Italy in the Republican Age of Rome. Manchester. Wörrle, M. (1988) ‘Inschriften von Herakleia am Latmos I. Antiochus III, Zeuxis und Herakleia’, Chiron 18: 421–76. Zelnick-Abramovitz, R. (2000) ‘The Xenodokoi of Thessaly’, ZPE 130: 109–20. Zelnick-Abramovitz, R. (2004) ‘The Proxenoi of Western Greece’, ZPE 147: 93–106. Zgusta, L. (1984) Kleinasiatische Ortsnamen. Heidelberg. Ziebarth, E. (1933) ‘Neue Beiträge zum griechischen Seehandel’, Klio 26: 231–47. Zimmermann, M. (1992) Untersuchungen zur historischen Landeskunde Zentral-lykiens. Bonn. Zuiderhoek, A. (2008) ‘On the Political Sociology of the Imperial Greek City’, GRBS 48: 417–45. Zuiderhoek, A. (2009) The Politics of Munificence in the Roman Empire: Citizens, Elites and Benefactors in Asia Minor. Cambridge.

Index of Sources Italics have been used to indicate more significant discussions of particular texts. References in bold type indicate an entry number in the Appendix. Appendix (Extant Proxeny Lists) 1: 155 1.1: 97 n. 29, 1.2: 167 1.8: 53 n. 104 2: 155 3: 113, 155, 165, 175–6 4: 113, 174, 246 4.1: 54 n. 109, 155 4.2: 161 n. 32 4.3: 155 n. 17 5: 153 n. 14, 155, 166, 167, 168, 169, 178–9, 246 6: 153 n. 14, 155, 165–6, 169 n. 66, 171 6.3: 167, 175–6 7: 54 n. 109, 113, 123 n. 132, 152–3, 154, 155, 170, 174, 177, 188 n. 103, 209, 252 n. 26 8: 53 n. 104, 55 n. 115, 153 n. 14, 155, 156–64, 168–9, 170, 178–9 9: 123, 155 10.1: 14, 54 n. 109, 97–8, 113, 123, 155, 158 n. 21, 162 n. 36, 165, 171–2, 172–3, 174, 176–7, 182–8, 206, 209, 210, 226 10.2: 155, 186–7 11: 155, 165, 175 12: 53 n. 104, 82 n. 203, 155 13: 10, 114, 181 14: 153 n. 13–14, 155, 242 n. 8, 256 14.5: 53 n. 104, 167 14.6: 168 14.9: 167 15.1: 54 n. 109, 75 n. 183, 155 15.2: 75–6, 155, 167, 169 16: 155, 242 n. 8 16.2: 42 16.4: 42, 57 n. 123, 167, 168 17: 153 n. 13, 155, 167, 168, 242 n. 8 18: 155, 167

Other Inscriptions and Papyri Agora 15 71: 39 n. 54 16 37: 94 n. 12 16 39: 94 n. 12 16 45: 94 n. 12 16 59: 164 n. 40 18 384–5: 57 n. 123 BCH 22 (1899) p. 90–4, no. III: 286–7 22 (1899) p. 94, no IV.1: 209, 227 45 (1921) p. 1–85: 166 n. 50, 256 n. 43, 329 57 (1933) p. 492: 53 no. 104 68/69 (1944/5) p. 102 no. 11: 306 n. 26 82 (1958) p. 85–90, no. 1: 164 n. 40 Cabanes (1976) 43: 25 n. 8 CEG 887 C: 25 n. 8 CIG 1936: see no. 18 3655: 63 n. 138 CIL VI.viii.3 40890: 250 n. 21 CIRB 1–5: 229 Clara Rhodos 2 (1932) 219 no. 61: 57 n. 123 Dimitrova (2008) 1–13: see no. 14.1–13 FD III 1 3–11: 200 n. 46 1 12–46: 112 n. 85 1 18: 33 n. 26 1 79–88: 112 n. 84

392

Index of Sources

FD III (cont.) 1 86: 164 n. 40 1 142–54: 112 n. 86 1 152: 53 n. 106, 67 n. 150 1 153: 35 n. 34 1 155–96: 112 n. 83 1 356: 227 1 359: 164 n. 40 1 451: 221 n. 107 1 453: 306 n. 26 1 480: 40 n. 55 2 89: 306 n. 26 2 48: 202 n. 49 2 65–6: 257 n. 48 2 71–97: 112 n. 89 2 89: 36 n. 40, 306 n. 26 2 91: 51 n. 96 2 160: 244 n. 9, 244 n. 10 2 193: 164 n. 40 3 79–116: 112 n. 87 3 105: 244 n. 10 3 119: 61 n. 134 3 145: 25 n. 8 3 158–78: 112 n. 88 3 249–50: 25 n. 8 3 383: 70 n. 158, 207 n. 64, 306 n. 26 4 20: 164 n. 40 4 34: 244 n. 10 4 35: 244 n. 10 4 59: 244 n. 9 4 74: 244 n. 10 4 132–5: 25 n. 9 4 175: 221 n. 107 4 225: 164 n. 40 4 377: 98 n. 31 4 403: 164 n. 40 4 406: 164 n. 40 4 414: 164 n. 40 4 427: 306 n. 26 4 430: 306 n. 26 4 442: 244–5 GIBM 1154a: see no. 18 IAphr2007 8.27: 250 n. 21 15.330: 257 n. 45, 278 n. 113 IC II v 35: 97 n. 27 II x 1: 43 n. 66, 53 n. 104

IV 184: 194–5 IV 206 a-m: 286 n. 3 ID 74: 221–2 75: 221–2 76: 221 77: 39 n. 52 79: 221 n. 106 88: 108 n. 70 98: 185 n. 95, 221 n. 105 1517: 119, 295 2535: 257 n. 47 2536: 257 n. 47 2538: 257 n. 47 2539: 257 n. 47 IG I3 10: 74 n. 174 I3 27: 81–2 I3 32: 215 n. 87 I3 80: 82 I3 91: 2, 110 n. 77 I3 95: 111 n. 80 I3 98: 96 I3 102: 124 I3 107: 111 n. 80 I3 110: 37 n. 43 I3 155: 111 n. 80, 155 n. 18 I3 156: 108 n. 70, 111 n. 80 I3 181: 110 n. 77 I3 227: 94 n. 12 I3 229: 94 n. 12 I3 262: 165 n. 41 I3 263: 165 n. 41 I3 364: 110 n. 79 I3 426: 126 n. 147 I3 430: 126 n. 147 I3 1032: 185 I3 1154: 57 n. 123, 129 n. 158 I3 1454: 43 n. 66 II2 2: 226–7 II2 6: 94 n. 12, 116 n. 105 II2 13: 164 n. 40 II2 17: 96 II2 19: 45 n. 76, 128 n. 154 II2 33: 116 II2 41: 69 n. 158 II2 52: 94 n. 12 II2 63: 53 n. 105 II2 66 fr. c: see SEG 14 40 II2 76: 69 n. 158

Index of Sources II2 81: 43 n. 67 II2 83: 57 n. 121 II2 86: 164 n. 40 II2 103: 102 n. 49 II2 105: 102 n. 49 II2 132: 34 n. 30 II2 176: 54 n. 110 II2 287: 57 n. 121, n. 122 II2 374: 48, 56 n. 118 II2 467: 139 n. 188 II2 487: 39 n. 54 II2 495: 45 n. 73, 54 n. 111 II2 654: 45 n. 73 II2 657: 40 n. 55 II2 786: 56 n. 119 II2 835: 56 n. 119, 64 n. 142 II2 845: 43 n. 67 II2 1012: 228 II2 1136: 25 n. 8, 201–2 II2 1186: 215 II2 1570: 10 n. 23 II2 1745: 322 n. 44 II2 2331: 135 n. 174 II2 2455: see no. 10.2 II2 3882: 10 n. 23 II2 7862–81: 57 n. 123 II2 9304: 57 n. 123 II3 294 (=II2 206): 32–3, 164 n. 40 II3 316 (=II2 237): 126 n. 148 II3 324 (=II2 373): 38 n. 48, 47–8, 56 n. 118, 56 n. 119 II3 352 (=II2 351 + 624): 43 n. 67 3 II 367 (=II2 360): 47, 56 n. 119, 57 n. 121, 57 n. 122, 64 n. 142 II3 375 (=II2 365): 164 n. 40 II3 378 (=II2 448): 95 n. 15 II3 379 (=II2 343): 53 n. 106, 63 n. 140 II3 390 (=II2 284): 73 n. 172 II3 430 (=II2 283): 73 n. 172 II3 454 (=II2 416 fr.b): 63 n. 140 II3 493 (=II2 288): 57 n. 121 II3 495 (=II2 267): 53 n. 105 II3 516 (=II2 579): 198 n. 37 II3 550 (=II2 417): 323 n. 44 II3 1137 (=II2 844): 29–32, 67, 76 n. 186, 130 n. 162, 164 n. 40 II3 1291 (=II2 785): 53 n. 106 II3 1336 (=II2 844): 67 n. 149 IV 756: 73 n. 172 IV 853: 86, 252 V 1 4: 126 n. 148 V 1 39: 269 n. 85, 269 n. 86

393

V 1 819: 269 n. 85, 269 n. 86 V 1 932: 155 n. 18 V 1 1146: 10 n. 24, 56 n. 119 V 2 106: 324 n. 51 V 2 367: 212 n. 80 V 2 368: see no. 11 V 2 387: see no. 12 VII 1: 216–18 VII 5: 155 n. 18 VII 9: 51 n. 97 VII 10: 36 n. 38 VII 207: 217 VII 208: 217 n. 93 VII 213: 217 n. 93 VII 219: 217 n. 93 VII 223: 50 n. 92, 217 n. 93 VII 2418: 64 n. 142 VII 4131: 297 IX 12 1 12a: 211 n. 89 IX 12 1 13: see no. 1.1 IX 12 1 17: see no. 1.2 IX 12 1 21: see no. 1.3 IX 12 1 24: see no. 1.4 IX 12 1 25: see no. 1.5 IX 12 1 29: see no. 1.6 IX 12 1 30: see no. 1.7 IX 12 1 31: see no 1.8 IX 12 1 71: 164 n. 40 IX 12 3 717: 68 n. 154 IX 12 5 2018: 279–81 IX 1 1: 127 n. 151 IX 1 100: 101 IX 1 103: 51 n. 96 IX 1 222: 38 n. 49 IX 2 5b: 103 n. 52 IX 2 11: 108 n. 70 IX 2 62: 25 n. 8 IX 2 69: 52 n. 102 IX 2 90: see no. 13 IX 2 91: 328, 329 IX 2 132: 329 IX 2 219: 85–6 IX 2 458: 25 n. 8 IX 2 1103: 40 n. 55 IX 4 882: 57 n. 123, 83 n. 204, 129 n. 158 IX 5 1912: 103 n. 52 XI 4 517: 50 n. 95 XI 4 542: 44 XI 4 543: 126 n. 146, XI 4 559: 44 XI 4 588: 55 n. 114

394

Index of Sources

IG (cont.) XI 4 593: 164 n. 40 XI 4 600: 44 XI 4 613: 53 n. 104, 128 n. 154 XI 4 639: 50 n. 91 XI 4 646: 44 XI 4 666: 61 n. 134 XI 4 693: 226 XI 4 705: 44 XI 4 765: 68 n. 152 XI 4 772: 199 n. 42 XI 4 776: 51 n. 97 XI 4 777: 12 n. 29, 98 n. 31 XI 4 784: 44 XI 4 789: 56 n. 119 XI 4 887: 164 n. 40 XI 4 1049: 10 n. 24, 164 n. 40 XI 4 1054: 72 n. 168, 108 n. 70 XI 4 1055: 108 n. 70 XII 1 25: 10 n. 23 XII 1 32: 10 n. 23 XII 1 1032: 215–6 XII 3 250: see no. 2.2 XII 3 251: see no. 2.1 XII 3 252: see no. 2.3 XII 3 254: 76 n. 185 XII 3 333/1298: see no. 16.2 XII 3 497: 57 n. 123 XII 3 168: see no. 3 XII 3 1290: 127 n. 149 XII 3 1299/1300: no. 16.4, 42, 57 n. 123 XII 3 1625: 57 n. 123, 336 n. 61 XII 4 32: 121 n. 126 XII 4 57: 107 n. 69 XII 4 59: 107 n. 69 XII 4 75: 43 n. 67, 295 XII 4 129–30: 9 n. 17, 107, 119, 121, 155 n. 18 XII 4 138: 61 n. 135, 108 n. 70, 295 XII 4 139–48: 107 n. 64, 108 n. 70 XII 4 148: 127 n. 149, 330 XII 4 1039B: 121 n. 128 XII 4 1040: 121 n. 128 XII 5 111: 37 n. 43, 94 n. 13 XII 5 112: 10 n. 24, 94 n. 13 XII 5 305: 269 n. 85 XII 5 528: 68, 254 n. 35 XII 5 538: 68 XII 5 542: see no. 10.1 XII 5 599: 108 n. 72 XII 5 653: 198 n. 37 XII 5 715: 127 n. 151

XII 5 798: 49, 50 n. 90 XII 5 812: 25 n. 8 XII 5 816: 49 n. 88, 333 XII 5 819: 49 n. 88 XII 5 820: 226 XII 5 821: 227 XII 5 835: 34 n. 30 XII 5 843: 333 XII 5 917: 10 n. 23 XII 5 946: 257 n. 48 XII 5 1000: 54 XII 5 1004: 53 n. 104 XII 5 1072: see no. 10.2 XII 5 1073: see no. 10.2 XII 6 11: 40 n. 55 XII 6 17–40: 200 XII 6 19: 55 n. 115 XII 6 20: 58 n. 124 XII 6 23: 50 n. 93 XII 6 24: 44 XII 6 28: 68 XII 6 30: 51 n. 97, 54 n. 112 XII 6 31: 35 n. 33 XII 6 43: 44 XII 6 46: 44 n. 69 XII 6 56: 64 n. 142 XII 6 95: 266 n. 80 XII 6 118: 55 n. 113, 62 n. 137 XII 6 150: 198 n. 37 XII 6 471: 25 n. 8 XII 7 5: 58 n. 125, 72 n. 172 XII 7 36: 121 n. 127 XII 7 67: 126 n. 146 XII 7 221: 60 n. 133 XII 7 226: 223 XII 7 231: 223 XII 7 388: 40 n. 55, 54 n. 107 XII 8 2: 214 n. 86, 224 n. 121 XII 8 3: 73 n. 172, 223 n. 117 XII 8 151: 330 XII 8 152: 330 XII 8 154: 330 XII 8 155: 43 n. 66, 330 XII 8 157: 330 XII 8 159: 330 XII 8 267: 97–8, 137 n. 180 XII 8 391: 135 n. 174 XII 8 420: 57 n. 123 XII 8 640: 55 n. 97 XII 9 11: 272 n. 94 XII 9 187A: 224 n. 121 XII 9 196: 16 n. 37

Index of Sources XII 9 198: 127 n. 149 XII 9 211: 199 n. 39 XII 9 218: 35 n. 32 XII 9 221: 53 n. 104, 66 n. 143 XII 9 900A: 10 n. 24 XII 9 1186: 162 XII 9 1187: see no. 8 XII suppl. 114: 315 XII suppl. 127: see no. 7 XII suppl. 137: 199 n. 42f XII suppl. 139: 155 n. 18 XII suppl. 304: see no. 15.2 XII suppl. 313: see no. 15.1 XII suppl. 355: 137 n. 180 XII suppl. 358: 96–7 XII suppl. 549A: 224 n. 121 XIV 612: 16 n. 39, 108 n. 72 IGBulg I2 37bis: 127 n. 151 I2 41: 58 n. 124 I2 42: 127 n. 149 I2 43: 86 IGR III 638: 278 n. 112 III 982: 278 n. 112 IGUR 2: 16 n. 39, 17 n. 41, 37 n. 45, 55 n. 114, 108 n. 72, 109, 199 n. 38 3: 16 n. 39, 17 n. 41, 37 n. 45, 108 n. 72, 109, 229 IosPE I2 4: 271 n. 90 I2 21: 98 n. 31 I2 32: 40 n. 55 I2 79: 246–7, 252–3 I2 164: 186 n. 98 I2 179: 135 n. 174 I2 357–9: 245 n. 13 I2 364: 245 n. 13 I2 365: 245 n. 13 I2 697: 245 n. 13 IPArk 9: 220 n. 103 IvO 10: 69 n. 157 13: 69 n. 157 39: 199 n. 40

I.Beroia 117: 277 n. 110 I.Byzantion 1: 30 n. 20, 53 n. 106 3: 246–7 I.Erythrai 6: 53 n. 104 122: 203 n. 56 201: 301 I.Eph 232–42: 258 891–6: 257 n. 48 1389: 46 n. 81 1411: 46 1415: 46 1420: 46 1422: 46 n. 82 1442: 46 n. 82 1447: 55 n. 114 1449: 46 1453: 127 n. 149 1459: 49 1498: 258 2048: 258 I.Gonnoi 3: 38 n. 48 40: 108 n. 72 I.Histriae 32: 30 n. 20 I.Iasos 47: 44 51: 124 n. 137 53: 34 n. 28 608: 155 n. 18 612: 155 n. 18 I.Ilion 24: 97–8 53: 123 n. 132 68: 139 n. 187 I.Kalch 4: see no. 9 I.Kaunos 1: 229 28: 242 n. 8, 258, 331 I.Kios 1: 22, 127 n. 149

395

396

Index of Sources

I.Knidos 603: 54 n. 108 I.Labraunda 11–12: 230 40: 25 n. 9, 129 n. 160, 202 n. 49, 230 42: 109 n. 73 I.Lampsakos 6: 73 n. 172 7 (=I.Thasos 171): 54 n. 107, 84, 108 n. 70 I.Lindos 15: 224 n. 121 16: 55 n. 114, 224 n. 121 130: 10 n. 23 165: 10 n. 23 I.Magnesia 6: 102 n. 49, 127 n. 149 34: 155 n. 18 36: 204 n. 58 48: 128 n. 154 65: 264 n. 69 75: 264 n. 69 76: 264 n. 69 I.Mylasa 643: 202 n. 49 866: 223 n. 114 942: 129 n. 160 I.Oropos 1–2: 214–5 3: 42–3, 214 n. 85 10: 22–3 22: 50 n. 94 35: 110 n. 77 51: 51 n. 97 53: 101 63: 52 n. 102 68: 101 69: 100 77: 101 97–8: 100–1 103–5: 100–1 111: 100–1 112: 100 113: 100 114: 100, 101 115: 100 116: 101 117: 100, 101 118: 101

135: 100 140–3: 101 147: 50 n. 95 151: 100 152: 100 160: 101 174: 100 175: 55 n. 113, 67 n. 151 176: 100 186–7: 101 192: 101 208: 101 303: 10 n. 24, 53 n. 104 307: 44 415: 161 n. 32 418–9: 100 I.Perge 323: 269 n. 83 I.Priene 2: 16, 127 n. 149 3: 16, 126 n. 148, 199 n. 42 6: 16, 37 n. 45, 53 n. 105 8: 16 12: 16, 95, 125 n. 143 47: 124 n. 137, 155 n. 18 111: 265 120: 265 121: 265 I.Stratonikeia 229: 269 n. 85 I.Thasos 169–72: 54 n. 107, 84–5, 108 n. 70, 155 n. 18, 330 I.Thespiai 44: 75 n. 180 47: 75 n. 180 48: 75 n. 180 50–5: 75 n. 180 56: 75 n. 180 57: 75 n. 180 I.Thess.Enipeus 13: 265 n. 74 I.Thrac.Aeg. 8: 127 n. 149 400: 53 n. 104 I.Tralleis 33: 287 n. 5

Index of Sources

397

Knoepfler (2001a) 15: 53 n. 104

Minon (2007) 24: 82 n. 203

Le Guen (2001) 13: 228 n. 133 30: 228 n. 133 37: 228 n. 133 61: 228 n. 133 72: 17 n. 41, 228 n. 133 75: 17 n. 41, 228 n. 133 76: 17 n. 41, 228 n. 133 77: 17 n. 41, 228 n. 133

ML 4 (=IG IX 4 882): 57 n. 123, 83 n. 204, 129 n. 158 80 (= IG I3 98): 96 Nomima 37–8: 83 n. 204, 125 n. 144 OGIS 494: 228 n. 133, 278 n. 112

LSAG 199.15: 83 n. 204, 125 n. 144

Peek (1969) 62: 164 n. 40 97–8: see no. 3

Maddoli (2007) 20B: 44 n. 70

Perlman (2000) E.3–6: see no. 6.1–4

Maiuri (1925) 8: 10 n. 23 18: 107

Praktika (1930) 35: 10–11

Michel 534: 63 n. 138

P.Cair.Zen I 59037: 120 III 59341a–b: 120

Migeotte (1984) 9: 10 n. 24 10: 10 n. 24, 209 24: 10 n. 24, 56 n. 119 25: 10 n. 24 48: 10 n. 24 56: 10 n. 24 61: 10 n. 24, 94 n. 13 117: 10 n. 24 Migeotte (1992) 158: 64 n. 142 Milet I 3 96: 287 n. 5 I 3 99: 287 n. 5 I 3 104: 38 n. 48 I 3 115: 287 n. 5 I 3 124: 222 I 3 139C: 35 n. 36 I 3 140A–B: 123 n. 132 I 3 154: 98 VI 3 1029: 222–3 VI 3 1060: 221 n. 112 VI 3 1067: 287 n. 5 VI 3 1069: 287 n. 5

P.Col.Zen. I 11: 121 P.Mich. I 23: 120 RC 3: 95 n. 19, 205–6, 255 RDGE 9: 328, 329 Rigsby (1996) 103: 228 111: 17 155: 210 n. 71 156: 210 n. 71 159: 210 n. 71 160: 210 n. 71 RO 14: 218–21 22: 172 n. 74, 194 n. 16, 195 n. 22 29: 94 n. 13 39: 129 n. 157, 172 n. 74, 173 n. 76, 320 40: 172 n. 74, 185 55: 25 n. 9, 129 n. 160, 202 n. 49, 230

398

Index of Sources

RO (cont.) 76: 136 n. 179 95: 47, 56 n. 119, 57 n. 121, 57 n. 122, 64 n. 142 SEG 1 132: 169 n. 63 2 332: 55 n. 114 3 468: 40 n. 40 11 491: 269 n. 85 11 493: 269 n. 85, 269 n. 86 11 496: 269 n. 85 11 974: 86, 87, 125 n. 144, 164 n. 40 12 390: 127 n. 149, 155 n. 18, 301 13 239: 82 n. 203, 226 13 327: 218 13 415–16: 82 n. 203 14 40: 94 n. 14 15 411: 25 n. 9 16 373: 110 n. 77 18 36: 10 n. 23 18 112–13: 57 n. 123 18 143: 252 19 595: 10 n. 23, 134–8 21 940: 57 n. 123 23 418: 129 n. 160 23 420: 127 n. 151 23 424: 97 25 445: 200 25 539: 72 n. 168 26 311–12: 57 n. 123 26 351: 57 n. 123 26 352: 57 n. 123 26 476: 83 n. 204, 125 n. 144 28 48: 227 28 657: 10 n. 23, 186 n. 98 29 272: 57 n. 123 29 1130bis: 68 30 533: 51 n. 97 30 1117–8: 200 30 1120–1: 16 n. 39 31 536: 184 n. 95 32 38: 22 38 32 613: 35 n. 31 32 825: 40 n. 55 33 682: 295 36 982: 224 n. 121 36 983: 53 n. 104 37 340: see RO 14 37 388: see no. 17 38 464: 25 n. 8 38 662: 30 n. 20

38 39 41 42 43 44 44 44 47 47 47 48 49 50 50 54 56 56 56 56 56 56 56 56 57 59

1059: 53 n. 104 1243: 40 n. 55, 250 n. 21 663: 314 837: 16 n. 39 448: 10 n. 24 901: 322 n. 42 1146: 25 n. 9 1148: 25 n. 9 1196: 271 n. 90 1659: 53 n. 105 1745: 194 n. 18, 194 n. 20 999: 245 n. 13 500: 217 n. 93 543: 10 n. 23 1121–2: 222 1103: 269 n. 83 75: 57 n. 123 581–98: 210–11 663: 11, 94 n. 12 874: 245 n. 13 876: 245 n. 13 1083: 16 n. 39 1359: 278 n. 113 1710: 110 n. 77 228: 57 n. 123 1100: 109 n. 74

SGDI 2600: 164 n. 40 StV II2 232: 184 n. 90 II2 287: 184 n. 90, 210 III 508: 212, 260 Syll.3 110 n. 4: 108 n. 70 412: 66–7 492: see no. 8 534: 221 n. 107 534B: 221 n. 107 548: 25 n. 9, 69 n. 157 585: see no. 5 604: 204 n. 57, 306 n. 26 612: 179 n. 82 644–5: 107, 121, 155 n. 18, 307 n. 35 653A–B: 107, 306 n. 26 689: 25 n. 8 722: see no. 3 849: 276 n. 108 942: 25 n. 9, 202 n. 49 TAM II 420: 269 n. 85

Index of Sources II 583: 269 n. 85 II 175: 275 II 915: 269 n. 85 Tit.Cal. 1A: 53 n. 104 1B: 53 n. 104 85: 295 Tit.Cam. 41: 314

Literary Sources Aelius Dionysius iota 2.2: 69 n. 156, 70 n. 161 Aeneas Tacticus 22.29: 83 n. 204 Aeschines 2.89: 105–6 2.141–3: 114–15, 117 n. 109 2.172: 117 n. 106, 117 n. 109 3.42: 106 3.47: 108 3.258: 94 n. 12 Andocides 3.3: 117 n. 106, 117 n. 109 Androtion FGrHist 324 fr. 44: 132 n. 166 Anthologia Palatina 13.12: 73 n. 172 Antigonus Carystius Mir. 15a 1.1 Appian Hist. Mac. 11.7: 213 n. 144 Aristides Or. 23: 277 n. 109 Aristophanes Av. 1021: 71 Nu. 60–74: 110 Aristophanes Byzantinus fr. 301: 75 n. 181 Aristotle Pol. 1275a–b: 74 n. 178, 74 n. 179, 124–5, 266 n. 79 Pol. 1276a6–1276b13: 96 n. 20

Pol. 1309a9: 141 n. 195 Pol. 1329b37: 125 n. 145 Arrian Anab. 1.9.9: 116 n. 104, 129 n. 156 Athenaion Politeia 36.2: 95 n. 16 37.2: 95 n. 16 40.2: 97 n. 27 40.3: 96 n. 21 Callimachus Ep. 36 Cicero Att. VI.1: 251, 269 Verr. 2.2.32: 250–1 Verr. 2.4.145: 17 n. 31, 231 Deinarchus 1.45: 92–3 2.25: 94 n. 12 Demosthenes 7.38: 129 n. 156, 129 n. 158 9.41–3: 94 n. 12 15.15: 115 18.82: 71 n. 163, 106 n. 58 18.197: 136 n. 179 19.271: 94 n. 12 20.11–12: 96 n. 21 20.36: 95 n. 19 20.40: 126 n. 147 20.52: 116 n. 105 20.57: 99 n. 35 20.59: 323 n. 43 20.81: 98 20.132: 93 40.36: 64 n. 142 52: 77–81, 109–10, 131 57.64: 98 Dio of Prusa 38: 261 n. 57 38.28: 277 41: 261 n. 57 54.7: 271 n. 90 Diodorus Siculus 14.12: 70 n. 158 14.19.4–5: 133 n. 168 14.59: 213 n. 84 14.63.4: 69 n. 155

399

400

Index of Sources

Diodorus Siculus (cont.) 14.79.4–5: 69 n. 155 14.93: 231 n. 145 15.49: 213 n. 84 Diogenes Laertius 2.51: 55 n. 114 2.143: 159 n. 28

Pausanias 3.8.3–5: 140 8.4.5–6: 200 n. 46 10.32.4: 200 n. 46 10.9.6: 200 n. 46 Pindar Isth. 4.7–9: 106–7, 224 n. 123

Dionysius of Halikarnassos Ant. Rom. 15.5.2: 17 n. 41, 70 n. 158

Plato Grg. 505c: 38 n. 50 Laws 642b–c: 114, 132–4, 139

Hellenica Oxyrhynchia 2.1: 69 n. 155

Pliny ep. 10.114–15: 261 n. 57, 271, 273

Heraclides Ponticus fr. 46a: 213 n. 84

Plutarch Cim. 14: 114 n. 94, 115–16 Cim. 16.1: 110 Praec. Ger. Reip. 805a: 260 n. 55 Praec. Ger. Reip. 814e–815a: 275–6 X Or. 835f–836a: 97 n. 27

Herodas 2.10–15: 74 n. 178 Herodotus 3.55.2: 133 n. 168 6.57: 69 n. 157 6.86: 136 n. 176 8.136: 131 n. 164 8.140â: 131 8.144: 131 n. 164 8.85: 38 n. 50 9.85: 73 n. 172 Hyperides 5.25: 92–3, 105 fr. 76: 28 n. 15, 73, 93 Ion of Chios FGrHist 392 fr. 6: 71 n. 163

Pollux 3.59: 69 n. 156, 70 n. 161 Polybius 4.76.2: 58 n. 125 4.77: 118 n. 111 5.81: 118 n. 111 5.95: 129 n. 156 6.5–9: 285 12.11: 12 n. 30 20.5: 120 n. 121 21.26: 72–3 21.29: 263–4 23.9.12: 213 n. 83 28.14.4–5: 200

Isocrates 7.67–9: 96 8.53: 74 n. 177

[Skylax] 39.2: 217 n. 95

Lexicon Rhetoricum 201 p. 189, n. 4: 74 n. 176

Sophocles fr. 88.8: 58 n. 125

Livy 5.28.4–5: 231 21.12–13: 17 n. 42, 131 n. 164, 231 45.29: 261

Theophrastus Lap. 52–3: 185 n. 97

Lysias 28.1: 128 n. 155 Nepos Alc. 7.4: 342

Thucydides 1.129: 38 n. 50 1.145: 263 2.22: 133 n. 169 2.29: 5 n. 6, 66 n. 144 2.91: 194 n. 16 3.2: 140–1

Index of Sources 3.52: 117 n. 109 3.70: 72, 76, 140, 141 n. 197 5.22: 117 n. 109 5.43–5: 71, 72 n. 167, 117, 142–5 5.47: 220 5.76: 117 n. 109 6.89–92: 131 n. 163, 142 n. 199, 144–6 8.8: 70 n. 158 8.80: 70 n. 158 8.92: 133–4 Xenophon Anab. 5.4.2: 129 n. 156 Anab. 5.6.11: 229 Ath. Pol. 1.18: 225 Hell. 1.1.35: 70 n. 158 Hell. 1.3.15–19: 70 n. 158 Hell. 1.4.2: 133 n. 168

Hell. 2.3.18–20: 95 Hell. 3.1.1: 133 n. 168 Hell. 3.18: 213 n. 83 Hell. 3.2.26–31: 140, 141 n. 197 Hell. 4.5.6: 68–9 Hell. 4.18.15: 213 n. 83 Hell. 5.1.1–29: 184 Hell. 5.1.32: 213 n. 83 Hell. 5.2.19: 211 Hell. 5.4.61: 184 Hell. 5.4.22: 70–1 Hell. 6.1.4: 131 n. 163 Hell. 6.3.2–4: 117 n. 109, 131 n. 163 Hell. 6.4.24: 131 Hell. 6.5.33: 69 n. 155, 133 n. 168 Hell. 7.2.16: 129 n. 156 Symp. 8.39: 71 n. 163

401

General Index For the lists of proxenoi studied in the Appendix, a full index of polis-ethnics can be found at the end of the Appendix. References in bold type indicate an entry number in the Appendix. Achaian koinon 213 n. 83, 213 n. 84, 217, 218 Aeschines 105–6, 107–8, 114–15, 117 agathos, 34; see also proxenos-paradigm Agesilaus 68–9 Aglaos of Kos 119, 295 Aigai 158–9 Aigina 177 Aigosthena 216–8 Aitolian koinon no. 1 and Delphi 67, 160–1, 221 enguoi 74–5 and member cities 210, 211–12 proxeny network 166–7 Alexander I of Macedon 115–16, 131 Alexander the Great 68, 227 Alcibiades 71–2, 142–6 grandfather of 117 Anaphe no. 2, 155, 298 anarchy, interstate 196–200, 213, 231–2, 233–4, 270 Androsthenes of Gyrton 85 Antigonos Monophthalmos 16, 205 Antigonos II Gonatas 159–61 Antiocheia in Persis 17 Aphrodite 2 Apollodoros of Kyzikos 62–3 Apollodoros, son of Metrophanes 222 Apollodoros, son of Pasion, of Athens 77–80, 109–10, 131 Apollonia (Crete) 200 arbitrators, interstate, see interstate arbitrators Archedemos of Thasos 135–8 Argos 80–1, 142–5, 226 Aristoboulos, son of Persaios, Antigonid courtier (?) 159 Arsinoeis 195 Arthmios of Zeleia 94 n. 12 associations 216 n. 90, 228, 253 Assos 84

Astypalaia no. 3 proxenoi at 16 proxenoi of 57 n. 123 proxeny network 155, 165, 170–2, 175–6, 180, 256, 292 n. 16 asylia: granted to communities (inviolability) 17, 41–2, 76, 179 n. 82, 197, 206, 208, 212, 228, 259–62, 276, 277 granted to individuals (freedom from seizure) 128, 203, 245–6 ateleia 123–4, 127, 129 n. 160, 214 n. 86; see also isoteleia Athena 2, 202 Athens: and Augustus 271, 273 Athenian courts 68, 77–81, 127, 225, 253 boule 68, 97 n. 27, 102 n. 49, 228 and Delos 28–9 n. 15, 108 n. 70, 215, 221–2, 257 and Delphi 36 n. 40, 82, 112, 201–2, 257 embassies to 160 Athenian Empire 70–1, 139–42, 224–5 epigraphic practices 13–14, 81–2, 224–5, 239, 240–1 Four Hundred 96 graphē paranomōn 92–4 and Histiaia 159–60 and Keos 172–3, 184–5, 257 and Kydonia 30–2 metics 10 n. 23, 43 n. 67, 56–7, 74, 94, n. 11 127; see also, metics military campaigns 98 and Myrina 214 n. 86 and Mytilene 140–2 and Oropos 111–12, 208 particularities of honorific practice at 45

General Index and Plataia 226–7 Athenian proxenoi 30–2, 32–3, 82, 132–3, 133–4, 139–42, 164 n. 40 proxenoi at Athens 70–1, 71–2, 112–13, 142–6, 172–3, 180 sequences of honorific decrees at 47–8 Second Athenian League 172–3, 184–5, 194 n. 16, 195 n. 22 and Sparta 116–7 and Thasos 116 Thirty Tyrants 94–5, 96 Attic Stelae 126 L. Aurelius 84–5 Augustus 21, 87, 260 n. 56, 271, 273 autonomy: granted 41, 179 n. 82 and polis status, 191–2, 213, 219–20, 275 and koina 213 see also freedom bankers 10, 59, 64, 76, 77–8, 136, 138 benefactions, virtual 28, 60 n. 133 Bithynia 261, 271, 273 Black Sea: grain trade 161–3, 186 proxeny decrees 245–6, 249 Boiotian koinon 82 n. 203, 209, 211–12, 214–15, 227, 238–40, 279–80 Bosporan kings 126 n. 147, 229 Boulagoras of Samos 39–40 boulē 123, 271–3, 275, 280; see also under Athens bribery 105, 115–6 Brykos, deme of Rhodes 215–16 Byzantium 70 n. 158, 180, 246–7 Calabria 42 catalogues of proxenoi 14, 153–4 Chaironeia, battle of 6, 73 Chalkis 160, 161, 187, 208 Charmion of Kydonia (son of Eumaridas) 29–32 Chios no. 4, 98, 155, 174 Chremonidean war 159–61 chrēsimos 34, 37, 50, 59; see also proxenos-paradigm Chrysis, Athenian priestess 201–2 Cicero 17 n. 41, 251, 269 Cimon 110, 115, 115–7 civic titles 262, 276–7, 278, 279 citizenship, see politeia

403

Claros 257–8, 278 cleruchies 215, 223, 227 Cloatii brothers 10 n. 24 Cloudcuckooland 71 Corcyra 72, 140, 142 earliest attestation of proxeny 83 n. 204 Corinth 72, 86, 252–3 courts, legal access to 73–5, 250–1, 253 backlog of cases 266 and Roman Empire 250–1, 269 see also foreign judges; interstate arbitration; prodikia courts, royal 118–22: members of, see royal philoi as political community 41–2, 66 to spend time at, see diatribō Crete: proxenoi at 29–31, 42 n. 64, 75–6, 175, 298, 309 Cretan koinon 180, 295 and piracy 75–6, 123 n. 133 crowns: granted to proxenoi 47, 123, 216, 333 private display 108 proclamation of 106 C. Crowther 266–9 and passim Cyrene 156 Damocharis of Sparta 86–7 Deinarchus 92–3 delegations: between poleis 30, 68–73, 109, 120–1, 203–4, 205, 254 proxenoi sent on 69–70 proxeny granted to 158–61, 190, 201 to Rome 179 n. 82, 278 see also foreign judges; interstate arbitration; theoroi Delian League 172 n. 74 Delos amphictyons 185 n. 95, 221 n. 105 and Athens 28–9 n. 15, 108 n. 70, 215, 221–2, 257 and enktēsis 126 n. 146 epigraphic practices 12 n. 29, 13–4, 215, 235 and Keos 184–5, 257 honorific practices 39 n. 52, 44, 50–1 Delian proxenoi 10 n. 23, 10 n. 24, 53 n. 104, 55 n. 114, 56 n. 119, 60, 62, 68 n. 152, 72 n. 168, 98, 164 n. 40

404

General Index

Delos (cont.) proxenoi at Delos 108 n. 70 place for publication by other states 108 n. 70, 111 Delphi no. 5 and Aitolian koinon 160, 221 and Alexandria 169 and Athens 36 n. 40, 82, 112, 201–2, 257 competition at 111–12 dispute with Amphissa 207 n. 64 epigraphic practices 12 n. 29, 13–4, 215, 235 honorific practices 33 n. 26, 35, 151 and Keos 184–5, 257 and Kydonia 29–30 late evidence from 236–7, 243–6, 249 Delphic proxenoi 10 n. 23, 25 n. 8, 36 n. 40, 55 n. 114, 60, 61 n. 134, 69, 164 n. 40, 154–5, 169, 202 n. 49, 227, 228 n. 133 proxenoi at Delphi 25 n. 9, 53 n. 104, 70 n. 158 proxeny network 165–71, 178–9 rates of granting 167, 171 and Rhodes 169, 207 n. 64 and Rome 179 n. 82, 231 theōroi to 29–30, 168–9, 204 n. 57, 255–7 Delphic Amphictyony 40 n. 55, 158 n. 22, 160–1, 195 n. 22, 279–80 Demades 28 n. 14, 93–4, 105 demes 215–17, 218, 219–20, 221–2, 274 democracy 96, 113 n. 93, 271–3, 274 Demosthenes 92, 105 proxenos of the Thebans 114–15, 117 deō 50; see also proxenos-paradigm dependent communities 192, 213, 216–223, 274; see also, demes; cleruchies diatelēo 34, 40 n. 55, 49, 145; see also proxenos-paradigm diatribō 54–5, 119 dikaiosunē 221 P. DiMaggio 87 Dio of Prusa 245, 277, 279 Diodorus of Syracuse (resident of Rome) 17 n. 41, 55 n. 114, 109 Dionysiac artists, association of 202 n. 49, 228–9

Dionysius I of Syracuse 102 n. 49 Dionysodoros of Thasos 84–5 doctors: at court 121 as proxenoi 47–8, 51–52, 60 n. 131, 61, 226 n. 124 receiving other honours 215, 223 Metrodoros, a horse doctor 52 n. 102 Echinos 158 Ehrentafeln 107, 108 Elateia 101, 200 emporikos nomos 123 n. 132 Endios of Sparta 144 enguoi 74–5, 102–3, 288 enktēsis 56 n. 119, 95, 123–7, 203 Entella 200 entrenchment clauses 96–7 entynchanō 58 epeidē clause, see proxenos-paradigm, motivation clauses Ephesos 45, 46, 258 ephodos, see prosodos Epidamnos, battle at 72 Epidauros no. 6, 167–8, 171, 175, 256 Epidauros Limera 185 epigraphic habit; see epigraphic practices epigraphic practices: antigrapha 108 inscription on bronze 16–17, 82 n. 203, 94 n. 12, 109, 226, 239 cessation of inscription 242–3 destruction of stelae 94–6 early inscriptions 82–3, 223–4, 239 erasure 97–8, 277 functions of inscription 91, 99–101, 109–10, 111–12, 120 interpreting gaps in epigraphic record 214–15 publication at granting polis, function of 31–2, 111–12, 198–9 publication at polis of proxenos 62–3, 108–9, 121 regional dynamics 16–17, 238–40 reuse of media 98 n. 31, 99, 112–13, 238–9, 289–90, 338–40 inscription as selective 13–17, 236, 242–3 trends in inscription 234, 236–43, 256–8, 259–60, 262, 263, 267–8 Epimeinondas 184, 319, 322

General Index epimeleia (care) of polis for proxenos 128–30 of proxenos for citizens of polis 78, 131, 133–4 epinomia 105, 122 ephodia 76 Eresos no. 7: and Lesbos 153, 209, 256 proxeny catalogue 154, 155 proxeny network 165, 170, 174, 177 Eretria 98, 159–61 passim, 183–5 passim, 201, 210, 211 Erythrai 66–7, 203 n. 56, 265, 300 etheloproxenos 140–2 ethnics and communal identity 194–5, 198–9, 205, 225–7 external use of sub-polis ethnic 221–2 federal/koinon 158 n. 22 island 226 omission of in proxeny decrees 53 n. 104, 222 personal names derived from 110, 133, 215 n. 87 polis 51–2, 119, 194–5, 226–7 regional 42, 55–6 Euandros of Larisa 95, 96 Eudemos of Seleukeia 107, 121 Euenor, doctor from Akarnania 47–8 euergetēs, title 38–43, 53 n. 104, 335 n. 59 inappropriate for citizens 39–40 and kings 41–2 euergetism 37–8; see also kings and cities; services Eumaridas of Kydonia (father of Charmion) 30, 31, 32, 66–7 eunoia 34–6, 60, 84, 132, 202, 203, 246–7; see also proxenos-paradigm Euthykrates the anti-proxenos 28 n. 15, 73, 93 Euxitheos 98 exile 116–7, 142; see also under poleis; proxenoi federations, see koina female proxenoi 6, 25 n. 8, 201–2 female theōroi 257 n. 48 foreign judges 61, 71, 171, 262–3, 266–9 decrees for 98, 107 as proxenoi 64–5 formulaic language, functions of 12–13, 24–9, 47–8, 103–4 Four Hundred 96

405

freedom 41, 119 n. 116, 179 n. 82, 221 n. 105, 271; see also autonomy P. Gauthier 3, 7, 12, 40, 83 n. 204 Gennaios of Elateia 101 Gortyn 67 and Kaudos 194–5 M. Granovetter 149–50 F. Gschnitzer 7 hairesis 31, 35 M. H. Hansen 191–2, 194, 212–3, 219 hegemonic poleis, see under poleis Helisson 218–21 Herakleia (Pontika?) 77–80, 109, 131 Herakleides of Salamis 47 Herakleitos of Halikarnassos 160–1 Herakleodoros of Olynthos 134–8 Hestiaios of Thasos 84–5 hieromnēmones, see Delphic Amphictyony Hieron of Priene 95 Histiaia no. 8, 156–64, 178–9, 183–4 homonoia 279 honorific language, see proxenosparadigm honours: civic control of 106–8, 213–17 exceptional honours 123; see also crowns, statues function of 209? inscription as additional honour 15, 96 standard package granted to proxenoi 122–4, 243–4; see also asylia, ateleia, enktēsis, epimeleia, epinomia, isoteleia, politeia, proedria, prosodos hospitium publicum 230–1 hum and buzz 149 Hyperides 28 n. 15, 92–4 hypomnēma 108–9 Iasos 44, 45, 224 ideology 130–4 Ilion 96–7, 139 n. 187 imperialism 129–30, 139–42, 224–5 institutions: conservatism of 88, 249 definition 23 isomorphism 87–9, 199, 284 New Institutionalism 23–4 see also inter-polis institutions

406

General Index

International Relations Studies 190–1, 194, 196–8 inter-polis institutions, Greek system of 201–4, 204–6, 228–31, 254–5, 270–7 interstate arbitration 203, 207, 262–6, 274, 328 isomorphism 87–9, 199, 284 isopoliteia, see under politeia isoteleia (tax status of citizen) 56–7, 123–4, 127, 162 Kalchedon no. 9, 123, 124, 155, 156 Kallias, Spartan proxenos at Athens 70–1, 117 n. 109, 131 n. 163 Kallippos of Athens (proxenos of the Herakleians) 77–81, 109, 131 kalos 34; see also proxenos-paradigm kalos kagathos 34, 37 n. 42 Kalymna 53 n. 104, 274 Karthaia no. 10, 182–8 and Athens 68, 113 and Propontis 162 n. 36, 176–7 proxeny catalogue, inscribing of 97–8, 206 proxeny network, size of 14–15, 165 proxeny network compared 170–3, 180–2, 209–10 and Sparta 185, 226 Karystos 125 n. 157, 177, 300 Kassandros of Alexandria Troas 107, 305 n. 26 Kaudos, inhabitants of 194–5 Kaunos 120–1, 229, 258 Keos 183–8 passim and Athens 97–8, 172–3, 319 and Eretria 183–5 passim, 210 koinon 183–4, 206, 209–12 passim, 226 see also Ioulis; Karthaia; miltos Kephisiades, resident of Scyros 78 Kilikia 121, 251, 269 kings and cities 41–2, 230; see also courts, royal Kios 172 Konon 53 n. 104, 98 Kleitor no. 11 proxeny network 165, 175, 298 proxeny catalogue 155 Kleomachos of Oropos 100, 101 Knidos 2, 180, 184, 202 n. 49, 298 Knossos 67

koina 209–13, 278; origins of 179–80, 260 n. 56 proxenoi of 10–11, 25 n. 9, 103, 209 n. 67 see also Achaian koinon; Aitolian koinon; Boiotian koinon; Crete; Keos; politeia kōmai, see demes Korris, priest at Labraunda 229–30 Kos 61, 74, 108 n. 70, 274 Kydonia 29, 67, 200 Kyphanta 185 Kyzikos 62–3, 155 Labraunda 230 Lamspakos 62, 84 leagues, see koina Lebedos 96 n. 19, 205–6, 255 L. Licinius Anteros, of Corinth 86, 252 Lindos 108 n. 70, 224 Lipara 231 lists of proxenoi 4–5, 152–4 local region of primary interaction 151–2, 174–9, 179–80, 207–8, 279 logic of appropriateness 24 Lousoi no. 12, 53 n. 104, 82 n. 203, 155, 224 Lykon of Herakleia 77–81 Ma, J. 190–1 and passim Macedonia 115–6, 261 Macedonian cities: proxenoi of 17 n. 42; 116 n. 114 Macedonians receiving proxeny 55–6, 158–61, 187, 214 ‘magistrate proxenoi’ 69 n. 157 Magnesia on the Maiander: arbitrators from 264 n. 69, 328 asylia campaign 17, 197, 206, 228 proxenoi at 102, 208 proxenoi of 114 Malta 109, 229 Mantinea 218–20, 324 Mardonius 131 Marek, C. 7–8, 15, 41, 59–60, 63 Mausolus 129 n. 160, 229–30 Megara 112, 177, 216–18 Megillos of Sparta 114, 132–3, 139 mercenaries, 25, 83 n. 204, 287 n. 5, 303 merchants and ship-owners, association of 228 Methana 86

General Index Methymna 140–1 metics, not at Athens, 40 n. 80, 55, 56–7, 74, 109, 322 n. 42; for metics at Athens, see under Athens mētropolis 276–7 Miletos: asylia campaign 206 dispute with Priene 265 erasure at 98 Milesian proxenoi 114, 287 and Myus 222–3 proxenoi at Miletos 46, 313 miltos 185, 186 Minoa 60 n. 133, 223, 227 mobility 51–2, 62–3 money-lenders, see bankers monopoly on expression of honours in civic space 106 Montevideo Convention (1933) 192–3; 195 motivation clauses, see under proxenosparadigm Mylasa 222, 268 Myrina 214 n. 86, 223, 224 Mytilene 140–2 Myus 222–3 Nagideis 195 Narthakion no. 13, 114, 181 Naryx 279–81 Naukratis 32–3, 55 n. 114, 108 n. 70 negotiatores 251–2 neokoreia 258, 276 Neoptolemos of Aitolia 66 Neo-realism 196–7 networks 149–152, 173–82, 277–8; see also peer-polity interaction; private networks; proxeny networks; Social Network Analysis; and under individual poleis Nicias 143, 144 Nikaia 245, 277 Nikomedes of Kos 107, 119, 121 Nikomedia 245, 277 Odessos 86 oikēsis 124 n. 136 oligarchy 95–6, 124–5, 271–2 Olymos 222–3 Olympia 69 Olynthos 134–8 Orontas of Olbia 246–7

407

Oropos: and Athens 111–12, 208 epigraphic practices 13–4, 44, 53 n. 104, 238–40 Figs. 5.3–4 and Eretria 208, 211 independent 42–3, 214–15 proposers of decrees 99–101 and Thebes 214–15 Pagai 218 Panhellenion, Hadrianic 279–81 Panhellenion, at Naukratis 108 n. 70 parakaleō 50–1, 80 n. 198; see also proxenos-paradigm parasēmon, see relief sculpture on proxeny decrees Paros 62–3, 94 n. 13, 268–9 Pasion of Athens, banker 77–80, 109 patroneia 164 n. 40, 253 peer-polity interaction 190, 198, 231–2, 277–8 Peithias the etheloproxenos, of Corcyra 140–2 periodization 233–4, 270 of proxeny 5–7, 281 perioikoi 82 n. 203, 185, 226 Persaios of Kition, Antigonid courtier 159 Pherai 97, 131 philia (friendship) 34, 84–5, 200, 202, 261; see also proxenos-paradigm Philip II of Macedon 93, 114 n. 104, 129 n. 157, 135–6 Philistos of Kos 61, 295 philotimia 34, 37 n. 42, 106 n. 61, 173; see also proxenos-paradigm philosophers Phormion, slave of the banker Pasion 77–8, 109 phthonos 106 Pindar 11, 106–7, 225 n. 123 piracy 72, 80, 261; see also sylē; asylia Piraeus, men of the 133–4 Plataia 43 n. 67, 73 n. 172, 117 n. 109, 226 poleis: dependent poleis, see dependent communities non-polis actors 228–9, 237, 285; see also associations; dependent communities; koina destroyed 135–6, 195, 226–7

408

General Index

poleis: (cont.) poleis in exile 68, 195, 200 factions 91–2, 97–9, 140–1, 142, 147; see also stasis hegemonic poleis 129–30, 139–42, 192, 213–5, 224–5, 226 reciprocity in inter-polis relations 162, 198–9 polis identity: performance of 194, 198–9, 204–6, 213–24 and Rome 280–1 politeia (citizenship) 139 Aristotelian definition 124–5 granted to individuals 45–6, 122–4, 222–3, 224, 288–9 granted to communities (isopoliteia) 183–4, 202–3, 208, 210–11, 212, 259–60 incapacity of sub-polis communities to grant 215, 216–7 proxenias politeia 245 ‘religious citizenship’ 125 sympoliteia 140–1, 200, 205–6, 218–20, 222–3, 260 n. 56 synoicism, see sympoliteia see also ateleia; asylia; enktēsis; epigamia Pompey 271 Pomponius Bassos 275 W. Powell 87 Priene 94–5, 126 n. 148, 199, 265 epigraphic practices at 15–6 private networks 65, 144 n. 204 proxeny networks dependent on 91–2, 101–3 performing similar functions to proxeny networks 71, 80–1, 253 prodikia 123 n. 132, 127, 253 proedria 70 n. 161, 125 prohairesis, see hairesis prosodos 70, 122–3, 223 n. 118 prostatēs 74–5, 86, 246, 253 prōteia 276–7 prothumia 34–7, 47, 131, 134, 145; see also proxenos-paradigm Protogenes of Olbia 40 proxeneō 83 n. 204 proxenias politeia 245 Proxenides, son of Proxenos, of Knidos 2, 11 proxenikos nomos 123 n. 132

‘Proxenos’ as a personal name 110, 247–9 proxenoi: as ambassadors 69–70, 117, 130–1 catalogues of 14, 153–4 dedications by 10 n. 23, 134–8 education of 132–3, 146–7 in exile 116, 134–8 as foreign agents 115, 130, 139, 147 funerary monuments for 57 n. 123 ideology 130–4 killing of 129 n. 157, 142 lists of 14, 152–6 public performance of role 109–10 renouncing proxeny 117 women, see female proxenoi proxenos-paradigm 8, 23–7, 81–7, 87–9 benefactions, see euergetēs; services generalizing description, use of 29–31, 39–40, 46, 145 early attestations of 6 n. 10, 81–3 activity in external political contexts 51–6, 57–8, 65 late attestations of 84–7, 246–7, 252–3 expressed in motivation clauses, 27–8, 31, 32–33, 34, 35, 38, 45, 61 personal disposition 34–6, 60, 131, 203, 246–7 and polis identity 198–9, 201–4, 204–5, 207–13, 213–16, 221–4, 245–6 public friend of whole polis 2, 77–8, 103–4 honorific language, functions of 12–13, 24–9, 47–8, 103–4 intermediary role 48–9; see also services reciprocity 31, 37–8, 104, 131, 143–4, 164 specificity of 43–8 suppression of concrete details 103–4 utility 8, 32, 36–7, 48–9, 82, 147, 253, 282 xenos 39–40 proxeny: adoption of 87–8 atypical recipients 8, 24–5, 47–8, 55–6, 60 dark side 110 disappearance 236–43 early evidence 27, 224 and epigraphic practices, see epigraphic practices

General Index group grants 157–61, 168–70, 171–3, 181 hereditary relationship 31–2, 33, 52, 110, 132–3, 164, 247, 292 n. 16 inappropriate for major Hellenistic kings 41 mass grants 25, 202 narratives of post-Classical decline 5–8, 11–12, 83–4, 123–4, 172 ‘la proxénie liturgique’ 69 n. 157 opposing a grant 92–4, 97–8, 101 permanence of 36–8, 41 prestige value of 105–7 proposing a grant 28, 90, 91, 93, 99–103, 170 rates of granting 165–6, 166–73 scholarship on 4–8 seeking a grant 29–31, 104–105, 111–12, 118–19, 143–4, 169–70 selective inscription 13–17 rescinding a grant 94–5 use by non-polis actors 228–31 use and non-use within koina 208–13 proxeny networks: diachronic trends 170–3 patterns of distribution 174–82 proxenos as hub 79 renewal 164 size 165–70 Ptolemaic kings 160–1, 169, 295 Pylos 143, 144, 145 Python of Oropos 100, 101 Pythophanes of Phaistos 96 publication of proxeny decrees, see epigraphic practices quantitative probability distribution 234–7 T. Quinctius Flamininus 179 n. 82 reciprocity 131, 150, 162–3, 164 breakdown in 143–4 between cities 150, 162–3, 179–80, 198–9, 212, 260 between cities and proxenoi 104, 131, 150, 164 and euergetism 37–8 regions creation of 151–2 distribution of proxenoi 156–64, 172–83, 182–9

409

regional dynamics 8, 122, 169, 245–6 organization of proxeny catalogues by 52, 153 see also local region of primary interaction; epigraphic practices; provinces, under Rome relief sculpture on proxeny decrees 2, 53 n. 105 restriction on expression of external honours, 106 Rhodes 84–5, 115, 215–6, 224, 227, 264 L. Robert 4–7, 161, 218 Rome and Delphi 179 n. 82, 230–1 domination of 196, 233–81 passim embassies to 179 n. 82, 278 governors 250, 265, 269, 274–5 provinces 250–1, 269, 276, 279 Roman Emperor 232, 253, 262, 265, 277–81 granting proxenia (?) 230–1 Romans named proxenoi 42 n. 65, 179 n. 82, 253 non-Roman proxenoi at Rome 17 n. 41, 109 see also Augustus; patroneia royal philoi (friends) 28 n. 14, 62, 115–6, 118–22, 147, 195 n. 194, 230 continued links with native poleis 61–3, 120 Samos 44, 45, 102 n. 49, 223 Samian cleruchy at Minoa 223 Samothrace no. 14, 84, 167, 168, 242 n. 8, 258 Sardalos, king of Thrace 86 scholarship on proxeny 4–8 Senate 179 n. 82, 262, 264–5, 271, 276 services typically performed by proxenoi 49–51 advocacy 66–8, 250 burial 73, 251 care for prisoners 143–4 hospitality 70–1, 252–3 identification 68–9, 73–6, 252 information 71–2, 252 ransoming 72–3, 76, 251 role in facilitating trade 53, 63–4, 252 standing surety 76 continued utility under Roman Empire 249–54

410

General Index

services performed by proxenoi, atypical 64, 134–8 gifts of money 64 n. 142 loans 10 n. 24 Sicily 16–17, 250–1 Sicyon 218 Sidon 156 Sidyma 275 Sinope 162–3 Social Network Analysis 149–51, 179, 180–1 Sparta appointment of proxenoi by kings 69 n. 157 and Athens 70–1, 97, 110, 116–17, 180, 184 brachylogia 132 foreign judges 269 n. 85 perioikoi 185, 226 proxenoi of the Spartans 55 n. 114, 70–2, 86–7, 110, 116–17, 130–1, 140, 142–6 Spartans as proxenoi 57 n. 123, 69, 83 n. 204, 114, 132–3 Sphodrias 70–1 stasis 94–6, 141; see also under polis state identity 192–5; see also polis statues for proxenoi 16, 31–2, 62, 84, 121, 123, 134–8, 161 n. 32 stereotypical proxenos descriptions, see proxenos-paradigm Sthorys of Thasos 96–7 Stymphalos 200 sylē 97 n. 27, 128, 129–30; see also asylia; piracy symbola 253 sympoliteia, see under politeia syngeneia (kinship) 202, 229 n. 135, 261 synoicism, see under politeia Syracuse 17 n. 41, 42, 102 n. 49, 155, 162 Tarentum 70 n. 158, 156, 162 Tenedos 96, 141–0, 162–3, 185 Tenos no. 15, 49, 75–6, 155 Teos 96 n. 19, 205–6, 255

Thasos 84–5, 96–7, 134–8 Thaumakoi 85–6 Theangela 42 Thebes 106–7, 116 n. 104, 117, 214–5, 226–7 theōria 3 n. 3, 197, 200, 219–220, 255–9 and polis identity 203–6, 255–6, 261–2 theōroi appointed proxenoi 29–30, 59, 63, 153–4, 168–9, 173, 208, 242 n. 8 theōrodokia 255–7 Thera no. 16, 42, 57 n. 123, 167, 168 n. 62 Thessalonike 158–9 Thirty Tyrants 94–5, 96 Thouria 86–7 Thucydides of Pharsalos 133–4 trade: grain 63–4, 163 n. 39, 163, 186 miltos 185, 186 proxenoi as evidence for 51, 63–4, 161–3, 176–7, 185–6, 252 reports by traders 53 traders 9, 51, 53, 228 as proxenoi 63–4 treaties 258–72 unipolarity 270 K. Waltz 196 A. Wendt 197–8 A. Wilhelm 7 women, see female proxenoi; female theoroi xenia (guest-friendship) 2, 31, 35, 70–1, 89, 115–16, 202 private xenoi, see private networks renewal 164 royal xenoi, see royal philoi xenoi (foreigners) 127–8 Xenias of Elis 140–1 Zenon of Kaunos 120–1 Zoilos of Boiotia 216–17 Zoilos of Heraia 86