A Historical Commentary On Diodorus Siculus Book - 15

A HISTORICAL COMMENTARY ON DIODORUS SICULUS BOOK 15 P. J. Stylianou CLARENDON PRESS OXFORD 1998 -iii-

Oxford University Press, Great Clarendon Street, Oxford OX2 6DP Oxford New York Athens Auckland Bangkok Bogota Bombay Buenos Aires Calcutta Cape Town Dar es Salaam Delhi Florence Hong Kong Istanbul Karachi Kuala Lumpur Madras Madrid Melbourne Mexico City Nairobi Paris Singapore Taipei Tokyo Toronto Warsaw and associated companies in Berlin Ibadan Oxford is a trade mark of Oxford University Press Published in the United States by Oxford University Press Inc., New York © J. Stylianou 1998 The moral rights of the author have been asserted First published 1998 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. Within the UK, exceptions are allowed in respect of any fair dealing for the purpose of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act, 1988, or in the case of reprographic reproduction in accordance with the terms of the licences issued by the Copyright Licensing Agency. Enquiries concerning reproduction outside these terms and in other countries should be sent to the Rights Department, Oxford University Press, at the address above British Library Cataloguing in Publication Data Data available Library of Congress Cataloging in Publication Data A historical commentary on Diodorus Siculus, Book 15 P.J. Stylianou. (Oxford classical monographs) Revision of author's thesis (Ph. D.) -- Oxford. Includes bibliographical references and indexes. 1. Diodorus, Siculus. Bibliotheca historica. Book 15. 2. History, Ancient -- Historiography. 3. Greece -- History -- To 146 B.C. I. Title. II. Series. DF231.S79 1998 938 -- dC21 98-12863 ISBN 0-19-815239-6 1 3 5 7 9 10 8 6 4 2 Typeset by Regent Typesetting, London Printed in Great Britain on acid-free paper by Bookcraft (Bath) Ltd., Midsomer Norton -iv

PREFACE THE present book has its origins in an Oxford doctoral thesis which was examined as long ago as December 1981. By the beginning of 1982 fate (as Diodorus would have put it) had taken me away from Oxford, and indeed away from England, and into other fields of academic endeavour, and many years were to pass before I could turn serious attention to the revision of the thesis, above all by the inclusion of the Sicilian portions (excluded in the original thesis), and its preparation for publication. I am consoled by the fact that this lengthy gap enabled my thinking on Diodorus to mature and that the book may be less imperfect as a consequence. Little need be said about the importance for Greek history of the Bibliotheke Historike (or Library of History) of Diodorus Siculus, in particular of those books of it which cover the fourth century (it emerges plainly enough, I hope, in both the Introduction and the Commentary). The absence of detailed historical commentaries on any of these books has long been felt. If the present study goes a little way towards filling this gap it will have served its main purpose. In so far as the finished product is aimed at a specific category of reader, it is the undergraduate and graduate student with Greek, as well as, of course, the professional Classical scholar and ancient historian. But I trust it may also be of use to Greekless students armed with a translation of Book 15. For a work of this kind naturally much needed to be consulted, in terms of both primary material (the ancient evidence) and secondary literature. With regard to the latter, I should say that with very few exceptions I have not made use of anything published after 1993. This is regrettable, especially in the case of major new contributions such as M. Dreher Hegemon und Symmachoi ( 1995), but a line had to be drawn somewhere if I were ever to finish. To have made only a hasty and partial use of the most recent books and articles would have been unfair, above all to them.

As far as the text employed in the lemmata and in quotations -v- from Book 15 elsewhere in the Commentary and in the Introduction is concerned, I have used both of the best present-day texts, that of Vogel in the Teubner edition, and that of Vial in the Budé. Professor David Lewis recommended that Vial should be used for Book 15 (CAH2 vi. 147 n. 123) and there are good reasons for so doing. Vial's work in arranging the various manuscripts which contain Book 15 into two families, and her thorough collation of the four principal manuscripts (P and M and their respective descendants X and F) marks a definite advance on the limited work of Vogel in this direction (on which cf. Chamoux, p. clxi), and in many places her text is to be preferred to that of Vogel (cf. 13. 1n.; 13. 4-5n.; 22. 2n.; 57. 1n.; 64. 1n.). In other instances, however, her judgement is less sure and her text is less satisfactory as a result than that of Vogel (cf. 19. 4n.; 30. 3n.; 52. 3n.). Where there is disagreement between them, therefore, arising out of variant readings or lacunae in the manuscripts, I have followed the edition I considered more correct and given my reasons for so doing. Occasionally I have preferred other readings (see, e.g., 52. 1; 75. 3 and nn. there). A useful outcome of Vial's work is the observation that MS F (of the fifteenth century) represents what might aptly be described as an 'edited' text. This is seen not only in the case of the lists of Roman magistrates (Vial, p. xxiii and n. 2; Chamoux, pp. ciiif. and civ n. 74), but elsewhere also (cf. 21. 2n.; 30. 3n.; 57. 1n.). It might be apposite to add in this connection that recourse to the great edition of Wesseling of 1746, truly a monument to eighteenth-century scholarship, repays the effort every time. But the definitive text of Book 15 does not yet exist. In favour of Vogel it can also be said that the Teubner text is that used in the Loeb edition of Books 1-20, the version more generally available in the English-speaking world. It is for this reason that I draw attention in the Commentary to instances of inaccurate or misleading translation of the Greek in this edition of Book 15. For books other than 15, and especially for Books 16-20, I have relied on the Teubner edition, though often comparing this with the Budé text where possible. For the fragmentary Books 21-40 I have used F. R. Walton's invaluable edition in the Loeb. It should not be necessary to stress that the present work is not a textual commentary. My comments on the text, therefore, are confined to instances of historical significance. In the course of research carried on intermittently over such a long time I have incurred many debts. The first and greatest debt I owe to the two eminent scholars who supervised the writing of my doctoral thesis, Professor Antony Andrewes and Mr George Cawkwell. The contrasting style of the two men, both acknowledged masters in their respective spheres, was bracing, and to move from New College (later 13 Manor Place) to University College and back was exciting and a constant stimulus. Until his death in 1990, Professor Andrewes continued to take a keen interest in my fortunes and to urge, whenever we met, and in a succession of notes and letters, the speedy revision and publication of the thesis. Indeed, in 1985 he very courageously even announced its not too distant publication (in Essays Starr, 196 n. 6). All this helped to keep 'Diodorus Book 15' alive at a period when I could spare little time for such matters. Professor George Forrest first acted as my internal examiner, but fast became a friend and mentor, and proved a sustaining force over the years. The sad news of his death reached me while engaged in composing this preface. Similarly I owe no less a debt of gratitude to Mr Robin Lane Fox who arrived at New College shortly after I began my research. It was at Professor Andrewes's request, not long before his death, that the late Professor David Lewis read the whole unrevised thesis and commented on it fully, with the sagacity, exacting scholarship, and attention to detail he was justly famed for. Later, and much to its profit, he also read the first draft of the section on Diodorus' sources for the West and reciprocated by very kindly allowing me to read his chapter on Sicily between 413 and 368 BC before its publication in CAH2 vi. Finally, Dr Simon Hornblower, with characteristic generosity and enthusiasm, boldly volunteered to read the revised version and, a punishingly busy teaching and research schedule notwithstanding, he did so with customary speed and efficiency, offering expert advice and saving me from many a pitfall. I am deeply in his debt. But for him the book would have been a great deal poorer and even longer in appearing. Other friends or colleagues who have helped in various ways, whether in discussions or in correspondence, include the following: Professors Michael Osborne, Brian Shefton, John Lazenby, and Antony Woodman, Mr Jeremy Paterson (who first guided me towards Diodorus as a research topic), Dr Robin Seager (my -viiexternal examiner), Dr George Georghallides, Dr Constantine Hadjistephanou, Professor Vassos Karageorghis, Dr Ian Carradice, Drs David Davison and Rajka Makjanić, Dr Rowland Smith, and Professor David Potter. I am especially grateful to the Chairman and Members of the Editorial Committee of the Oxford Classical and Philosophical Monographs for accepting the book for publication and to the Craven Committee of Oxford University for subsidizing a lengthy visit to Greece in 1977. I must also express my gratitude to the editorial staff of the Oxford University Press for their adroit handling of the book through the various stages of production. For his kindness, patience, and fatherly encouragement, His Eminence Archbishop Gregorios of Thyateira and Great Britain has my heartfelt and respectful thanks. Revision of the thesis was made possible by a generous grant awarded by the A. G. Leventis Foundation. I am most grateful to the Foundation and its Chairman, His Excellency Constantine Leventis, for this as well as their long-standing interest in my research. I owe a great debt of gratitude to Sir Reo Stakis for a research grant readily awarded for another purpose, but which also proved helpful in the final stages of publication of the present book. On a personal level, I wish to thank my wife for her steady encouragement and never-failing support, both moral and academic, without which, I have not the slightest doubt, this book would not exist, and to apologize to our children for devoting so much of my time to Diodorus rather than to them. October 1997 P. J. S

-viii-

CONTENTSAbbreviations

Maps INTRODUCTION 1. The Character of the Bibliotheke (a) The setting (b) Chance and utility (c) Style and diction 2. The General Plan and Date of Writing of the Bibliotheke 3. The Sources (a) The chronographer (i) The framework (ii) The other chronographic material: the obvious kind (iii) The other chronographic material: the less obvious kind (iv) The chronographer's errors (b) The narrative sources (i) Ephorus: the oikonomia (ii) Ephorus: the sources (iii) Ephorus: attitude to states and individuals (iv) Ephorus: merits and defects 4. Diodorus' Methods COMMENTARY Appendix Bibliography Index of Authors and Passages Discussed Greek Index General Index

xi xviii 1 1 1 3 15 17 25 25 25 31 43 45 49 84 104 110 121 132 141 552 556 581 586 587

-ix-

INTRODUCTION

"The Hellenica is for connoisseurs", writes Cawkwell in the introduction to the Penguin translation of Xenophon's Hellenica. For different reasons Diodorus' Bibliotheke Historike is not for the uninitiated. For the cardinal fact about Diodorus is that he was a second-rate epitomator who generally used first-rate sources. Had these survived the Bibliotheke would have been of interest only as an example of a first-century BC work of compilation. Unfortunately, however, the abridged historians are mostly lost and we must perforce do what we can with Diodorus. The problem is to discover the principles that should be applied. It is imperative to know in each instance if the information is Diodorus' or his source's; and the extent to which he has mishandled it if the latter. This introduction, therefore, aims at an understanding, on the one hand of Diodorus himself, and on the other of the sources he used. Only then can the evidence of the Bibliotheke be properly evaluated.

The first section is concerned with the character of the Bibliotheke: what and how typical of its time was it? What is one to make of one of its most conspicuous characteristics, the constant moralizing? What can we conclude from the language? The second section looks at how and when Diodorus produced the Bibliotheke. In the third section the sources which lie behind Book 15 are examined in detail. The fourth and final section is concerned with Diodorus' methods.

INTRODUCTION "The Hellenica is for connoisseurs", writes Cawkwell in the introduction to the Penguin translation of Xenophon's Hellenica. For different reasons Diodorus' Bibliotheke Historike is not for the uninitiated. For the cardinal fact about Diodorus is that he was a second-rate epitomator who generally used first-rate sources. Had these survived the Bibliotheke would have been of interest only as an example of a first-century BC work of compilation. Unfortunately, however, the abridged historians are mostly lost and we must perforce do what we can with Diodorus. The problem is to discover the principles that should be applied. It is imperative to know in each instance if the information is Diodorus' or his source's; and the extent to which he has mishandled it if the latter. This introduction, therefore, aims at an understanding, on the one hand of Diodorus himself, and on the other of the sources he used. Only then can the evidence of the Bibliotheke be properly evaluated. The first section is concerned with the character of the Bibliotheke: what and how typical of its time was it? What is one to make of one of its most conspicuous characteristics, the constant moralizing? What can we conclude from the language? The second section looks at how and when Diodorus produced the Bibliotheke. In the third section the sources which lie behind Book 15 are examined in detail. The fourth and final section is concerned with Diodorus' methods.

1. THE CHARACTER OF THE BIBLIOTHEKE (a) The setting The second and first centuries BC witnessed the growth of a numerous and wealthy middle class in the towns of Italy and Sicily; what we would today term a bourgeoisie. The conditions of peace established by Augustus were especially favourable to -1it. 1 There was an accompanying thirst for a smattering at least of higher learning and the booksellers of the late first century were not slow in meeting it. 2 This is the sort of readership that Diodorus of Agyrium in Sicily aimed to reach. 3 It is possible that he was commissioned by a bookseller; or he may have been his own employer. At any rate, the lively trade in books like the Bibliotheke may be indicated by the speed with which Diodorus worked. There was no final revision and indeed he did not even, while writing, take the trouble to correct glaring errors of which he was aware. Perhaps time was short. It is also perfectly possible, of course, that he simply could not be bothered (see pp. 24, 137 ). This is further proof that Diodorus' prospective buyers would have cared more for the appearance of the rolls and their general 'educational' and entertainment value than for the soundness of their historical detail. An impression of quality had nevertheless to be maintained, for no buyer, however modest his learning, would have been interested in a work of obviously inferior scholarship. Diodorus, therefore, gave his work the form of a serious universal history such as that of Polybius and a strong didactic tone calculated to appeal to a first-century reader. And the paraphrased extracts which make up the Bibliotheke derive from some of the best ancient historians, now largely or totally lost to us, Ephorus, for example, Timaeus, Hieronymus, Polybius, and Posidonius. Otherwise the Bibliotheke was what its title implies -- a concise history of the world for Everyman. Hence its arrangement: the contents of each book are defined at the beginning as well as at the end, and for the historical period the narrative is arranged annalistically. If readers desired to know what had occurred in a particular year all they had to do was to take up the appropriate book and read under the magistrate in question. The nature of the ____________________ 1 Rostovtzeff, 1957: i. 9ff., 54ff.; Rawson, 1985: 215ff. on D. 2 Cf. Kenyon, 1951: 81ff.; Reynolds and Wilson, 1969: 22f. Though of the 1st century AD we might profitably bring to mind Petronius' repulsive creation Trimalchio, a member of the Italian bourgeoisie, who prides himself on being cultured and on possessing two libraries, one Greek and one Latin ( Satyricon48). Also the essay by Lucian ( 2nd century AD), Πρòζ τòν ἀπαíΔεΥτον καì Πολλά ßιßλíα, ὠνούμενονa wealthy and vulgar man who collects masses of books for show (I: καὶ. Θζδαυρòδ ἔτοιμοδ (εἰ) τοÎδσ καΠήοιγ αὑτὡν(τὡ ßιßλíΩν)).That there were booksellers as a profession in Rome at the time of D, and that money could be made from the sale of books, cf. Strabo 13 C 609. 3 Cf. Sordi, p. vii. For the few biographical facts we possess about him see Schwartz, RE v. 663; Oldfather, pp. vii f.;

Sordi, pp. v f.; Sacks, 1990: 161 ff. -2Bibliotheke as a historical compilation is betrayed by its title and Pliny, two generations later, had no hesitation in classing Diodorus with the compilers of handbooks ( NH Praef. 24ff.), a thriving profession in the late first century BC . 4

(b) Chance and utility One of the more polished parts of the Bibliotheke is its general preface (I. 1-5); naturally, for the work stood to be bought or rejected on the strength of what was said there. The beginning is therefore, purported to be a history of the world and its professed aim was to benefit mankind. Diodorus proceeds to explain in what way such a work could be of use to the reader. By the examples it provides of good men earning the praise of their fellows while wicked ones meet with public disgrace or worse, by itself always praising the first and censuring the second, history encourages men to improve themselves and society; or at least it restrains their impulses to do evil. It offers experience and understanding, qualities which can admittedly be gained by association with wise and knowledgeable men, but history is far superior to them because of the great number of facts and paradeigmata at its disposal (1. 1. 4). The more of these the better, hence the superiority of the general history over the local. History, we are told, possesses no value in itself unless it is to instruct. We are not concerned here with the validity of such a 'philosophy' of history. But it is important to note that it is found throughout the Bibliotheke. The moral and didactic purpose of

____________________ Cf. Hornblower, 1981: 22ff. We may compare Pompeius Trogus, a contemporary, who also produced a historical compilation in the guise of a unisersal history entitled Historiae Philippicae. D and Trogus appear to have made the same claims in their general prefaces; cf. D I. 3. 2ff. and Pomp. Trog. praef. 2f. -3-

repeated elsewhere (15. 88.1; cf. 23. 15; 30. 15; 31. 15. 1) and exemplified on numerous occasions. Divine justice (τò ΘεîΟν, τò Δαιμóνιον) in the Bibliotheke always overtakes the arrogant, lawless, and sacrilegious (cf. 14. 63. 1f.; 14. 70. 4; 15. 48. 4; 16. 61. 1-64. 3; 21. 16. 5; 22. 11. 2). Such a conception of historical causation could not be simpler. God or Fortune (ἡ τύχζδ) lay at the back of most things and indeed nothing delights Diodorus more than to relate τὰ παλογα τ δ τύχη , those sudden and inexplicable changes in the circumstances of men which intrigued and fascinated antiquity. Diodorus is in effect writing a history of surprises and reversals (cf. 15. 33. 1; 15. 54. 5; 17. 46. 6-47; 17. 66. 2, 4; 17. 108. 6; 18. 53; 19. 11. 7; 20. 13. 3; 31. 10. 2). Also very much in evidence, though Diodorus does not spell this out as one of his aims, is what might be called the 'entertainment' element. He makes a point of relating when he can strange customs and events (τὰ παράΔΟζα) for apart from their possible educational value they served as καì for his readers (cf. 1. 83. 1; 2. 44. 3; 2. 47. 1; 3. 4. 1; 3. 30. 4; 17. 46. 6; 17. 63. 4; 18. 26. 2; 19. 98). None of the above characteristics is original or unique to the Bibliotheke. On the contrary, some of these elements went back a long way. The utilitarian view of history for instance was a commonplac, in antiquity and reached back at least to Thucydides. 5 The idea that god or fortune intervened in human affairs had much older roots -- it is in fact a necessary concomitant of religious belief. Xenophon in the fourth century was certainly under its sway (cf. H. 5. 4 1; 7.5 26). Still, it is a question of degree, and, thesecharacteristics became far more pronounced in late Classical And Hellenistic historiography. A more recent element, which in fact permeates the Bibliotheke, was the sensationalist, highly rhetorical way of writing history which Polybius roundly condemns, but cannot quite free himself from. 6 ____________________ 5 Thuc. I 22. 4; 2. 48. 3; cf. Romilly, in Histoire et historiens, 41 ff.; Polyb. 1. 1. 2 with Roveri, 1964: 106; Walbank, 1972: 28. 6 Walbank, 1972: 34ff. On the 'dramatic' approach to writing history see further, id., 1955: 4ff.; HCP i. 8f.; Wiseman, 1979: 143ff.; Gabba, 1981: 52f. See also the remarks of Murray, 1972: 211. Though its roots were much deeper, paradoxography flourished in the Hellenistic age. Leading practitioners of the genre such as Euhemeruis, Iambulus, and Dionysius Skytobrachion are known to us because they were extensively used by D in the early books of the Bibliotheke. Alongside genuine historical detail in the works of these Writers there stood the fictional and the fantastic. D's readers would not in general have been able to -4Difficulties arise when one seeks, as one must, the origins of the various moralizing and other trends in the Bibliotheke, especially in relation to source criticism. For though Diodorus clearly subscribed to this sort of historiography, this does not necessarily mean that he was responsible for its various manifestations in the Bibliotheke. As far as the citation of good and bad examples for the purpose of providing ethical instruction is concerned, the fact that it is found in different parts of the Bibliotheke induced Kunz to argue against Laqueur that its origins cannot be pin-pointed. 7 It is a fact nevertheless that the ethical viewpoint is not propounded in precisely the same way or with

the same intensity in all the books. There is an important difference, for example, between Books 11-15 and 17-20. (Book 16 is difficult; Ephorus ran out in the course of it: see pp. 95 ff.) The moralizing judgements on individuals in the form of set epainoi or psogoi are characteristic of the first group and do not occur in the second; cf. 11. 11 (praise of Leonidas and the Three Hundred); II. 38. 6 (praise of Gelon); 11. 46f. (censure of Pausanias, and praise of Aristides); 11. 58. 4ff. (praise of Themistocles); 11. 82. 1-4 (praise of Myronides) 15. 81. 1-4 (praise of Pelopidas); 15. 88 (praise of Epaminondas). Book 17 and the narratives based on Hieronymus in Books 18-20 are quite different in this respect. History's duty to provide moral instruction is less in evidence here. Alexander is commended for his treatment of the Persian royal captives (17. 38. 4-7) and his great achievement is remarked on twice (17. 1. 3-5; 117. 5), but he receives no eulogy as such. Demosthenes' death is not even noticed, but what is really surprising is that Eumenes, whose treatment is detailed and highly favourable, is not accorded the epainos of history. The praiseworthy actions of Eumelus of Panticapaeum are enumerated (20. 24. 4-25. 3). Ptolemy, who is no less favourably treated than Eumenes, is said to have been delivered from the greatest dangers by the gods because of his arete, that is, his reverential treatment of Alexander's corpse (18. 28. 3ff.). Olympias, on the other hand, got precisely what she deserved in the end because of her cruelty and inability to bear her good fortune with moderation and humanity (19. 11. 4ff.). The ____________________ distinguish between the two and D in fact calls such narratives 'history' (5. 41. 4; 42. 4); cf. Gabba, 1981: esp. 58f. 7 Kunz, 1935: 30ff. -5above is a substantial part of the clearly expressed moralizing in this part of the Bibliotheke. Brief remarks at the end of a person's life seem to be preferred here; cf. 17. 46. 5 (on the Tyrians); 17. 48. 5 (on Amyntas); 19. 51.5 (on Olympias); 20. 37. 6 (on Cleopatra); 20. 42. 5 (on Ophellas). Books 21-40 survive only in fragments and a proper evaluation of them is thus impossible. But it is evident that stereotype moral appraisements similar to those in Books 11-15 occurred in some at least of the books of the second half of the Bibliotheke; cf. 23. 15 (censure of M. Atillus Regulus); 29. 18ff. (eulogies of Philopoemen, Hannibal, and Scipio Africanus); 34/35. 33 (eulogy of P. Cornelius Scipio Nasica Serapio and his family). The praise/blame duty of history is reiterated; Cf. 23. 15. 1; 30. 15; 31. 15. 1; 32. 26. 1; 34/35. 9; 38/39. 18. 1. 8 The most logical explanation for this lack of uniformity in the Bibliotheke is that Diodorus' sources did not all concern themselves with the purely ethical function of history to the same degree or in the same way. Ephorus was Diodorus' main authority for Books 11-16 (below pp. 49 f.). We shall therefore first look at Ephorus before turning to Diodorus' other and later sources. How much of the moralizing in Diodorus is due to Ephorus? The case for the praise/blame task of history seems clear enough. The Evagoras of Isocrates was, as the author himself says, the first ever encomium in prose on a contemporary. 9 And it set a pattern for imitation. It was meant to praise Evagoras as a good and valiant prince, and Isocrates puts forward the view that such men ought to be eulogized because, being contemporaries, the truth could be employed in composing their encomia, and also because the younger generation would thus be encouraged to emulate and even surpass them (9. 5ff., 77). The genre proved popular. We possess Xenophon Agesilaus and we hear of encomia on Gryllus, Mausolus, Philip, Alexander, and others. 10 The tradition that Ephorus and Theopornpus were the pupils of Isocrates is a firm one 11 and all we know about these two, the most influential ____________________ 10 Diog. Laert. 2.55; FGH 115 TT6, 8, 48; FF 255 -7; cf. Isocr. 5. 17; Blass, ii. 284. 11 FGH 70 TT 1ffff., 8, 27f. 8 The fragments are enumerated according to the Loeb edn. vols. xi and xii. 9 9. 5ff. Before this time prose encomia were on mythical persons; cf. Gorgias' as well as Isocrates' On Helen. Arist. Rhet. 1368a, however, claims priority for an encomium on a Thessalian called Hippolochus. -6historians of the fourth century, indicates that they incorporated Isocratean features into their historical works, in particular the kind of moralizing which laid emphasis on praise and censure. 12 Whether Isocrates himself advocated a certain form of historical writing remains uncertain and does not in any case concern us here. 13 For the character of Theopompus' history we have the testimony of an ancient admirer, none other than Dionysius of Halicarnassus. It is in basic agreement with the surviving fragments. Theopompus wrote about thingsΘαὰρμαδτά or ΠαρU=1F71ΣΟ⌣α not just because they were entertaining, but also because they were beneficial to his readers. 14 He wrote about the foundations of cities, the lives of kings, the customs of nations, Greek and barbarian, νóμΟυδ . . . .Πρά⌣ειδ καì τέλ καì And he did so voluminously, for the benefit of man. He looked into the souls of men to divine causes which earned him the reputation of being ßάδ2κανοζ, but really he was performing the duty of a surgeon. He was always ready to reproach cities and generals for their villainous designs and unjust deeds, Πολὰδ γάρ ἐν τοάτοιζ. It is thus clear that for Theopompus history had a definite ethical role to play. The tradition that Isocrates once remarked that Theopompus needed the bridle, Ephorus the spur (FGH 70 T28) is probably not historical, but its implication seems to be that Ephorus was more restrained than Theopompus, and it indeed appears that he did not exhibit the bitterness and propensity to cheap gossip of Theopompus. Otherwise there cannot have been much difference between the two, and the fragments as well as the assessment of Polybius show that the ethical character of the Bibliotheke was very much the character of Ephorus' universal history too. Polybius states that Ephorus was Δεινóτατοζ . . . ἐν ταîζ παρεκßάδι καì. . . γνωμολογíαιζ, καì . . . ὅταν που τον ἐπιμετρούντα λóγον ΔιατíΘζται (12. 28. 10 = FGH 70 T23); and in fact he seems to be following ____________________

Cf. Avenarius, 1956: 157ff. Contra Sacks, 1990: 24ff., but see the present author's review, 1991: 391f. Cf. Beck, 1964: 277ff.; Bradford Welles in Studies Caplan, 3ff.; Schepens, 1977: 100f.; Hamilton, 1979: 290ff. 14 Cf. Isocr. 2. 48f. Homer and the tragic poets successfully combined καìκαì with ὠΦέλεια. 15 Dion. Hal. Ad Pomp. 6. For Theopompus' improper writing see Polybius' criticism at 8. 10 (12). 1-2. On Theopompus in general, Connor, 1968: chs. 1 and 6. 13

-7the example of Ephorus in his practice of assigning praise and blame whenever and wherever he judges it appropriate to do so. 16 FGH 70 F42 shows the didactic, through paradeigmata, nature of Ephorus' history, while F 191, frr. 2-6, a good instance of the sort of parekbasis praised by Polybius, proves that Ephorus, transferring Isocratean usage to the realm of history, appraised public men morally at the end of their lives. Also that the eulogies and censures in Diodorus Books 11-15 have their roots in Ephorus: F 191 frr. 2-6=11. 58. 4-59. 17 What of the second half of the Bibliotheke? Diodorus based Books 28-32 on Polybius, and Polybius as we have seen was a fervent practitioner of the kind of ethical historiography which gave public men their due. At 4. 20. 5 (=FGH 70 F8) he quotes with approval a distinction between history and music from the general preface of Ephorus. Ephorus had probably argued that the duty of history in contrast to music was not to thrill, but to instruct. Diodorus' eulogies of Philopoemen, Hannibal, and Scipio Africanus are derived from Polybius and are exactly in the Ephoran form (D 29. 18ff. = Polyb. 23. 12-14). Nor did Polybius ever fall to draw the moral lesson of history, and Diodorus must have found him fertile ground indeed; cf. D 30. 17 = Polyb. 28. 21; 18 30. 18 = Polyb. 28. 18; 32. 26. 1 = Polyb. 38. 4. 8. Occasionally Diodorus would tamper with Polybius' text in order to render the moral lesson more prominent; cf. 31. 15 = Polyb. 30. 18 -- Diodorus repeats the praise/blame duty of history and alters the reaction of the senate to highlight Prusias' worthlessness. After Polybius, Posidonius formed Diodorus' main authority. It is thought that Posidonius did not bestow epainoi and psogoi though his history did not lack an ethical side. The moralizing, it has been said, was implicit and subject to the pragmatic concep____________________ 16 Cf. 2. 56. 11f.; 6. 11. 10; 10. 21. 8; 18. 41. 1; HCP ii. 223; Avenarius, 1956: 157 ff. 17 Cf. FGH 70 T23 n. The attribution to Ephorus of P. Oxy. xiii. 1610 (F191) by its first editors, Grenfell and Hunt ( P. Oxy. xiii. 98 ff.), has found wide support. It has been questioned by Africa, 1962: 86ff. (cf. also Milns, Vindex Humanitatis, 56f.) and though his strictures on the editors' over-enthusiastic use of D in restoring the fragments are not entirely unjust, the arguments of Grenfell and Hunt for Ephorus remain sufficient. 18 I cannot see that D has garbled Polybius' censure of Ptolemy Physcon as Drews, 1962: 384 n. 7 , claims. -8tion of historiography. 19 It is noticeable, however, that Athenaeus found Posidonius very much to his taste. The impression one gets from the fragments is of a highly sensationalist historian. 20 Whatever the truth, it would be safe to assume that Diodorus was by this time (he had already written thirty-two books) well accustomed to applying epainoi and psogoi, and that he did so whenever a suitable opportunity presented itself in Posidonius' narrative; cf. 34/35. 9; 34/35. 33; 38/39. 18. 1. For the similar instances in Books 21-7 (Cf. 21. 1; 23. 15; 24. 5; 25. 10. 5; 26. 1. 3; 26. 4; 27. 6. 2. Chapters 23. 15 and 26. 24 were inspired by Polybius; see n. 21), one must again assume that Diodorus was either following the example of his authorities or reverting to the didactic habits of Ephorus after the interval of Books 17-20. The problem is that the sources for Books 21-7 are uncertain, with the exception of Philinus and those sections deriving from Polybius. 21 Philinus probably provided Diodorus with his narrative of the First Punic War. 22 Unfortunately we know very little about the character of Philinus' history. 23 Two important points ought to be stressed. First, it should be noted that the elaborate moral appraisements of Books 11-15 are really confined to Books 11 and 15. The deaths of such personages as e.g. Cimon, Pericles, and Dionysius are passed over almost in ____________________ 19 Cf. FGH iic. 160. For Posidonius the historian in general see Strasburger, 1965: 40ff. 20 This is not to ignore the warnings of Brunt, 1980: 477ff., that mere fragments can be seriously misleading as to the character of a lost work. FGH 87 F36, however, is long enough to provide us with a genuine flavour of Posidonius' Histories. Brunt cautions that a single long excerpt, like this one, 'might in principle be uncharacteristic. We should get a very false impression of Thucydides, if the only substantial part of his work that survived were the stories of Cylon, Pausanias, and Themistocles in 1. 126-38' (1980: 483). And so we should, though the analogy to draw for F36 is not with the stories of Cylon, Pausanias, and Themistocles, but rather with the account of Cleon which is not uncharacteristic of Thucydides. F36, the story of Athenion and the establishment of his 'tyranny' at Athens c.87, was no digression in Posidonius, but central to his main narrative. 21 It seems best to accept that D based Books 23 and 24 on Philinus (cf. Schwartz, RE v. 688), and that he did not begin to rely almost entirely on Polybius till Book 28 ( Schwartz, RE v. 689ff.). But he had already made at least a partial use of Polybius. The Carthaginian Mercenary War in Book 25 is usually thought to derive from Polybius; see Walbank, HCP i. 130f. 22 Both D 23. 15 and Polyb. 1. 35 are thought to have based their similar remarks on Atilius Regulus on Philinus; see HCP i. 92 f. 23 In spite of La Bua, 1966: 262ff.; cf. Badian review, Riv. di filol. 96 ( 1968), 203ff. In fact Philinus may have had not dissimilar historiographical conceptions to those of Polybius; see Walbank, 1945a: 1 ff. and esp. 10f. -9

silence. It is difficult to believe that Ephorus did not comment formally on the passing of these men. 24 The conclusion must, therefore, be that Diodorus, in spite of his promise at 15. 88. 1 and elsewhere, was inconsistent in this respect as in all others. The brief remarks on Pericles (12. 46. 1) and Philistus (16. 36. 3f. may be all that is left of lengthier Ephoran appraisements. (I am assuming here what I shall attempt to show below, i.e. that the substance of the western narratives in Diodorus is based on Ephorus.) Second, Diodorus was obviously free to mould the epainoi and psogoi in his own image, though the extent to which he did so is not always possible to gauge and it should certainly not be exaggerated. Drews is unconvincing when he argues that nearly all of the moralizing in Diodorus Books 11-15 is Diodorus' own and should not be ascribed to Ephorus for, as we have seen, all the available evidence indicates that Ephorus was not as devoid of the moralizing element as Drewsimagines. 25 It is an unproven assumption that 15. 88, the encomium on Epaminondas, is 'a patchwork paragraph which Diodorus may have assembled from his reading of Ephorus' and from elsewhere. 26 In particular, to claim that the moral qualities of the great man mentioned by Diodorus cannot be due to Ephorus, for Ephorus can only have concerned himself with the military ability of Epaminondas, is to misunderstand the character of Ephorus work. The moral qualities and education of public men were very much the concern of Ephorus as of his teacher Isocrates. 27 Indeed, paideia may have been one of the unifying themes of Ephorus' universal history. It would appear that he saw the possession of paideia or agoge by the citizens of a state, but especially by their leaders, as central to its well-being. 28 I am not, therefore, convinced that the paideia of Epaminondas which intrudes repeatedly into the Bibliotheke (10. 11. 2; 15. 39. 2; 52. 7; 16. 2. 2f.) is not an ____________________ 24 Alcibiades and Agesilaus were both 'grey areas', good as well as bad, and Ephorus perhaps did not moralize on their passing, preferring clear instances of good or bad; cf. pp. 119f. 25 1962: 386 and n. 15. Drews' view is now argued further by Sacks, 1990: 25 ff. 26 1962: 389. 27 Isocrates considered himself an educator if anything. For his proud concept that Athens was the very School of Hellas see 4. 50; 15. 295f. Cf. too Dion. Hal. Isocr. 1. For the high occurrence of the words παιδεία, παίδευσις, and παιδεύω in Isocrates see Preuss, Index Isocrateus, s.vv. 28 See pp. 112 f. below and cf. Schepens, 1977: 116f.; Burde, 1974: 19ff. -10Ephoran contribution, and there is in fact firm evidence that it is. Strabo 9 C 400f. ( FGH 70 F119) tells us that while Ephorus praised Boeotia for its natural advantages, he had nothing good to say about the Boeotian leaders, for these men generally lacked agoge and paideia and thus limited the success their country could have enjoyed. The short-lived exception was Epaminondas and in

same opinion is expressed in Diodorus 15. 79. 2 and 88. 4.

Nothing could be clearer regarding the Ephoran viewpoint. When of course Diodorus makes Epaminondas and Conon contemporaries, while Agesilaus is said to have lived a little earlier (15. 88. 2; but see n. there), we should see this as one of Diodorus' frequent blunders, as when Plato and Aristotle are assigned to the pentecontaetia (12. 1. 5). But these typically Diodoran confusions should not be seen as proof that the substance of what is being said does not derive from Ephorus. We should rather see 12. 1-2. 1 and 15. 88 as botched summaries of Ephoran encomia on Athens and Epaminondas respectively. Greek (primarily Attic) learning praised at 12. 1. 4f. was a theme to which Ephorus must have returned again and again, and at 15. 76. 4 we appear to have the remnants of such an instance (see n. there). It was seen above that throughout the Bibliotheke divine justice is active in punishing wrongdoing and it is not unreasonable to assume that this reflects Diodorus' own personal beliefs. This should not, however, be taken to mean that instances of this in the Bibliotheke did not originate in some form in the sources. Whenever checks can be applied, the unwisdom of drawing too sharp a distinction between Diodorus and his sources is seen again and again. While it is true that in Polybius, for example, C ____________________ 29 Thus also Arist. Rhet. 1398b; Justin 6. 8. 1-3 and Nepos, Epam. 10. 4. And see Polyb. 6. 43. 4ff.; Plut. Pel. 4; Mor. 864d. The Boeotians were proverbially dullwitted and it may well have occurred to many in the educated circles in Athens and elsewhere to associate the 'flash in the pan' Theban hegemony with the meteoric careers of Epaminondas (who had received philosophical training) and Pelopidas. It is not unlikely that Callisthenes was one of these, though the Aristotle passage does not prove it; nor does it prove that Ephorus borrowed his interpretation from Callisthenes (as was suggested to me by Simon Hornblower in a letter dated 21 Oct. 1995). -11δαιμόνιον is very rarely mentioned as dealing out retribution, yet Polybius did not really distinguish between τὸ δαιμόνιον and τύχη (nor did Diodorus, it would appear), and many instances are to be found in his history of Tyche punishing wrongdoing and sacrilege, instances adopted to a greater or lesser extent by Diodorus. 30 In the case of Ephorus too, it would seem, the gods did not often miss their chance to intervene benevolently in human affairs, punishing the wicked and profane and rewarding the good. Strabo 9 C 422. provides a particularly good example not only of this aspect of Ephorus, but also of his tendency to rationalize myths and of the high regard in which he held the Delphic oracle. It was in the course of his discussion of the origins and nature of the Spartan constitution (commonly ascribed to an oracular response) that Ephorus stated the view (one with which Xenophon would have agreed wholeheartedly) that a person lacking in proper respect of the gods (εὐσέβεια) was most unlikely to be just in his dealings with men (7. 12. 7).

The account of the Sacred War in Book 30 of the Histories was of course written by Demophilus, but there is every reason to suppose that the minds of father and son worked in the same way in this respect. A major theme of that account was the sacrilegious conduct of the Phocian leaders, their families, followers, and allies; and the divine punishment which befell them all: 16. 14. 3; 16. 61-4; FGH 70 F96 (Athenaeus). Athenaeus' quotation from Demophilus leaves the fate of the wives of the Phocian leaders incomplete and should not be taken to represent all that could have been found in Book 30 on the subject. The wives' ultimate fate is given by Diodorus (16. 64. 2). There is no need, therefore, to suppose 31 that Diodorus, in this instance at least, having decided that his source did not go far enough, turned to an alternative source which drove the moral lesson home fully. In contrast to this, Ephorus and Demophilus stressed the good fortune of Philip who had championed the Delphic oracle (16. 1. 4; 16. 64. 3). Widely held at the time, this point of view is well and eloquently put by Aeschines in his speech Against Ctesiphon (132ff.), delivered in 330. The attitude of Ephorus to the gods and the supernatural is further illustrated by his treatment of the earthquakes and tidal waves which struck the coast of Achaea in 373. ____________________ 30 Cf. HCP i. e.g. 20f., 20 n. 6, 93, 147. 31 As does Drews, 1962: 390f. -12While not neglecting to cite the natural explanations of the catastrophe of the physicists, it is clear from Diodorus' account that he was himself inclined to accept a religious interpretation (15. 48f.). Ephorus' piety nevertheless, as befitted a pupil of Isocrates, was guided by paideia. Mere superstition he seems to have abhorred, as shown by Diodorus' comments on Epaminondas' handling of the omens prior to the battle of Leuctra (15. 52ff.). A particular problem is the assessment of the incidence of Tyche in Ephorus and its relation to Diodorus. The Ephoran fragments are of no help. Only once does the word occur (F63) and then its meaning is that of an accidental happening. But Tyche, viewed as an agent or cause outside human control, whether personal (providence, fate) or impersonal, 32 can hardly have been absent from Ephorus. It figures very significantly in Isocrates 33 and Demosthenes 34 and it was certainly a characteristic of Theopompus (see p. 7 ). It must, therefore, have been present in Ephorus too. The difficulty is to distinguish between those Tyche references which are Ephoran and those which are Diodoran. The term, it is true, occurs throughout the Bibliotheke and may thus reflect Diodorus' own thinking. But it is clear that in many, if not indeed most, instances Diodorus did not initiate, but merely adopted the opinion of his sources. The unaccountability of Tyche at 12. 62. 6, for example, obviously derives from Thucydides (4. 12. 3) through Ephorus. As far as the books which are not based on Ephorus are concerned, it is evident that Tyche is very much an integral part of the narrative of Book 17, 35 and it should logically be attributed to the source or sources 'used for that book. Similarly, it is frequently the case that the action of Tyche in Books 18-20 can be traced back to Diodorus' sources with some confidence. For example, there is no reason why Hieronymus should not be credited with the references to Tyche at 18. 8. 7; 18. ____________________ 32 The definitions are those in the Liddell and Scott Lexicon. Both meanings are well attested in Classical writers and in Hellenistic writers they are of course ubiquitous. 33 Cf. 6. 47, the instability of human fortunes because of Tyche. Cf. too the many other instances of Tyche in Isocrates in Preuss, Index Isocrateus, s.v. 34 Cf. Preuss, Index Demosthenicus, s.v. 35 At 6. 3; 20. 1; 29. 4; 31. 6; 35. 7; 38. 5; 46. 2; 46. 6-47; 59. 7; 66. 2, 4; 69. 6; 101. 2; 107. 2; 108. 6; 116. 1. There are a number of 'ordinary' uses of Tyche too in the sense of the verb τυγχάνω, e.g. at 13. 5;94. 3; 115. 6. Cf. Sinclair, 1963: 41f. on Tyche in Book 17. -13-

13. 4; 18. 20. 1; 41. 6; 42. 1; 53. 7; and 67. 4. Sections 59. 4-60. 1 seem to be an instance where Diodorus elaborated Hieronymus. 36 Let us look specifically at Book 15. There are seven instances of Tyche, at 33. 1; 33. 3; 54. 5; 63. 1; 63. 2; 82. 6 and 84. 2. Also one of πεπρωμένη at 74. 4 and one of τὸ χρεών at 80. 3. The πεπρωμένη is mentioned in connection with the death of Dionysius. Ch. 74 may not, however, derive from Ephorus, but from Timaeus (see p. 84 ). What of 33. 1ff.? It is quite possible that Ephorus, in putting into the mouth of Agesilaus an excuse as to why the king had not attacked the strong allied positions, included a reference to the workings of Tyche. But the sententious statement at 33. 3, ἀγαθὴ γὰρ ἡ τύχη τοὺς μέγα φρονου+̑ντας παραδόξως σφη+̑λαι καὶ διδάξαι μηδὲν ἄγαν κατελπίζειν, is likely to be an addition by Diodorus to highlight the moral point. At 54. 5 we are told that Jason persuaded the Spartans and the Thebans to respect τὰ παράλογα τη+̑ς τύχης and conclude a truce. Something similar is stated by Xenophon (H. 6. 4. 23) so that we may assume that the mutability of the human condition was mentioned in this context by Diodorus' source. The tone of 63. 1f. is eulogistic and quite consistent with what Ephorus would have written on the Spartan appeal and the Athenian decision to help. Athens enjoyed a reputation for always helping the weak and needy which her panegyrists never failed to mention (63. 2n.), and certainly Ephorus would not have failed to do so when opportunity arose. The references to Tyche are simple and straightforward (cf. Isocr. 5. 44; X. H. 6. 5. 45), and they may well have occurred in the Ephoran text. The two references at 82. 6 and 84. 2 are likewise simple and unadorned and are paralleled in Polyb. 9. 8. 13 where, however, Polybius cannot resist adding his own comment on the unpredictability of Tyche. διόπερ εἰκότως οἱ συγγραφει+̑ς, he says, meaning Callisthenes and Ephorus. Callisthenes was probably the one who blamed Tyche for the failure of Epaminondas. Ephorus adopted this explanation and thus it found its way into the Bibliotheke (see also 82. 6n.). As a general rule, therefore, we should suspect wordy expositions of the power of Tyche. But there is no reason why we should reject as non-Ephoran simple and to-the-point references to Tyche such as those found in Isocrates (see n. 33.

____________________ 36 Hornblower, 1981: 106 and passim, agrees that Tyche played a part in the Histories of Hieronymus though she thinks that he 'perhaps gave less prominence' to its role than Polybius. -14-

c) Style and diction In a fundamental study J. Palm has shown that Diodorus' late Hellenistic Greek has stamped itself on the entire work independently of the sources used. 37 This does not nevertheless mean that the language of the sources has been totally obliterated. Far from it. Diodorus did not abridge uniformly and systematically. Sometimes he copied from a source verbatim; more frequently he paraphrased (see p. 132 ). But even when paraphrasing Diodorus was not beyond being influenced by the diction of his sources. At times, when struck by a word or phrase, he would retain it. Κοινὴ ἀναρχία at 45. 1 is almost certainly Ephoran, as is ἡμερατηγοί at 82. 6 (cf.n. ad loc.) and σατράπαι καὶ στρατηγοί at 90. 1, 3. Similarly, the description of Sparta as ἀπόρθητος at 15. 65. 1 and 81. 2. 38 Much of the Ephoran terminology is retained too; see 19. 2 n.; 34. 4n. One might further point to the proximity in language sometimes exhibited between Diodorus, Polybius, and Plutarch, certainly the result of the use of a common source (see p. 105 ). On the other hand he would sometimes replace obscure terms with more up-to-date ones intelligible to his readers. Section 62. 2 may be one such case (but see n. ad loc) and cf. 22. 2n. and 54. 5-56n. for instances of Diodorus both retaining Ephoran terminology and substituting his own. In spite of Photius' praise 39 Diodorus' is not an attractive style. The empty and inept rhetoric and the poverty of vocabulary are its most irritating characteristics. It is an almost formulaic way of writing. Embassies, military preparations, campaigns, battles, sieges are related repeatedly, but with little variation in their description. Diodorus is not interested in an event per se, but only in what that event can offer by way of moral instruction or entertainment. Hence it does not really matter to him if that event is narrated in very much the same way as other events. One Diodoran battle, for example, is very much like another. They are frequently described as ἰσχυραί or καρτεραί -certainly not Ephoran words 40 for they are found throughout the Bibliotheke; cf. 11. 7. 1 11. 12. 6; 11. 32. 2; 12. 6. 2; 15. 3. 6; 16. 86. 2; 18. 14. 3; 18. 44. 4; ____________________ 37 Palm, 1955; see also Hornblower, 1981: 263ff. and cf. Chamoux, pp. lxix ff. 38 Cf. ML no. 95; Lys. 33. 7. 39 Bibliotheke cod. 70, Budé edn., R. Henry, i. 103. 40 As Jacoby thought: FGH iic. 12. -1519. 83. 4; 19. 89. 2; 20. 87. 3; 20. 89. 2. 41 They often begin with trumpets sounding the charge; cf. 15. 55. 3; 85. 3; 17. 33. 4; 19-30. 1; 41. 3. There is almost always a high point when victory hangs in the balance and can go either way. Then something occurs, the death of a general or the intensified efforts of his opponents, and the issue is decided; cf. at random 15. 3. 6; 15. 17. 1; 16. 4. 5ff.; 16. 12. 3ff.; 16. 86. 2f.; 17. 11. 4ff.; 17. 33. 6ff.; 18. 32. 1; 19. 41. 3-42; 19. 83f. 42 The height of absurdity is reached when the Athenian fiasco before the walls of Amphipolis in 422 is given this treatment, both sides being said to have fought λαμπρω+̑ς, the battle being ἰσόρροπος to begin with, and both leaders exhibiting ἀνυπέρβλητον φιλοτιμίαν (12. 74. 1f.)! This surely cannot be Ephorus. Many of the words and phrases, often mere clichés, making up Diodorus' descriptions are monotonously similar if not precisely the same. Successful statesmen and generals, for instance, are invariably described as διάφοροι in ἀνδρεία or ἀρετή and στρατηγικὴ σύνεσις; cf., again at random, 15. 16. 2; 15. 21. 1; 15. 29. 2; 15. 56. 3; 16. 18. 1; 16. 48. 2; 16. 65. 2; 17. 7. 2; 18. 13. 6. When they fall in battle they do so μαχόμενοι ἡρωικω+̑ς or ἀγωνισάμενοι λαμπρω+̑ς; cf. 15. 17. 1; 15. 21. 2; 15. 55. 3; 16. 48. 5; 16. 63. 1; 17. 63. 4; 18. 15. 3. These and countless other such expressions are so stereotyped that they are meaningless in themselves. Such repetitive language of course facilitated the task of an epitomator, though historical accuracy could suffer in the process. To give some examples, Diodorus is inaccurate when he says that the Spartans ἐξέπεμφαν Cleombrotus in 371 and Polytropus in 370 (51. 4; 62. 1; see nn.). But in this way he saves himself the trouble of explaining the precise circumstances of the events. Of course, we should not treat every instance of ἐξέπεμφαν as hiding further detail. ἐξέπεμφαν at 45. 4 and 46. 3 (cf. 46. 2 ἔπεμφαν) means just that. Also, Diodorus is very fond of the adverb πανδημεί which he uses frequently, even when inappropriate in the context; cf. 13. 19. 4; 47. 4; 52. 1; 61. 2; 63. 4; 15. 21. 2; 25. 4; 26. 2; 36. 1f.; 52. 2; 20. 31. 5. Some of these instances are factual; others are not. (For two more examples of Diodoran usage see p. 133.) It might be worth noticing one or two characteristics of Diodorus' poor, epitomator's style. The heavy reliance on adverbs ____________________ 41 FGH 70 F191 proves that the expression is not Ephoran; see pp. 132 f. 42 Barber, 1935: 143f. and Meister, 1975: 74, ascribe these standardized descriptions to Ephorus. F191 (see previous n.) proves otherwise. -16-

is one such. There are, for example, no less than ten in 15. 1, while the short chapter 26 contains eight. Another trait of a poor stylist is the inept way of introducing persons and places into a narrative, though this could also be the result of careless abridgement of his sources; cf. 11. 64. 4 -- Ithome is now named when earlier, in 64. 1, it was simply referred to as χωρίον ὀχυρόν; 12. 39. 1 -- Pericles is now called formally 'son of Xanthippus' after he has been mentioned several times in the previous sections; 15. 91. 5 -Mithrobarzanes is given his name when in 91. 2 he was introduced as the nameless κηδεστής of Datames; cf. 15. 34. 3f. -- Pollis' ethnikon and title are repeated. 43 In the preface to Book 20 Diodorus inveighs against those historians who weighed down their works with long speeches. And in point of fact the Bibliotheke is largely free from this ancient historiographical device. But not entirely so for Diodorus decided to include speeches when the situation demanded it (20. 2. 1), though his criteria when he does do so defy comprehension. At 13. 52. 2 he describes the speech of the Spartan envoy Endius as brief and laconic; διόπερ ἔκρινα μὴ παραλιπει+̑ν τοὺς ῥηθέντας λόγους, he says. On the other hand the speeches at Syracuse three years earlier are considerably longer (13. 20ff.). A small number of speeches are found elsewhere too, for example, at 13. 102 and 14. 65ff. There are none in Book 15.

2. THE GENERAL PLAN AND DATE OF WRITING OF THE BIBLIOTHEKE At. 1. 4. 6-5. 1 Diodorus gives an outline of the whole work as it stood at its completion. It consisted of forty books. The first six dealt with the period before the Trojan War. The following eleven were a universal history from the Trojan War to the death of Alexander, and the remaining twenty-three gave an account of all subsequent events down to the beginning of Caesar's Gallic War which fell in the first year of the 180th Olympiad when Herodes ____________________ 43 We may compare the new fragment of the Hellenica Oxyrhynchia (P), Koenen, 1976: 70ff. (=McKechnie and Kern, 30, 32) lines 10, 17f., 23f., 47. P was not noted for his style (cf. Bruce, 18ff.;McKechnie and Kern, 22ff.). Of course such instances might also be explained as the result of D's careless abridgement of his source; cf. e.g. 14. 32. 4 -- the Three Thousand suddenly drop into the narrative, D having previously neglected to mention them. -17-

was archon at Athens (60/59 BC). In order to make his historical compilation readily serviceable to the readership he was aiming for, it was Diodorus' express intention to provide it with a firm chronological framework (1. 3. 2, 8). He therefore informs the reader that whereas he made no attempt to fix accurately events before the Trojan War, διὰ τὸ μηδὲν παράπηγμα παρειληφέναι περὶ τούτων πιστευόμενον (1. 5. 1), he followed Apollodorus in assigning 80 years to the period from the Trojan War to the Return of the Heraclids and 328 years to the 1st Olympiad, reckoning the dates by the reigns of the Spartan kings. From the 1st Olympiad to the start of the Gallic War, the limit of his work, he calculated 730 years. Thus the entire Bibliotheke, he tells us, embraced 1,138 years, χωρὶς τω+̑ν χρόνων τω+̑ν περιεχόντων τὰς τω+̑ν Τρωικω+̑ν πράξεις (1. 5. 1). Something is seriously amiss here, as 730 years from the 1st Olympiad give us 46/5 and not 60/59 BC; nor can the figure 730 alone be wrong because of the sum 80+328+730=1138. And yet twice Diodorus insists that his work closed with the beginning of the Gallic War (1. 4. 7; 1. 5. 1), and in fact the surviving fragments run out with the late 60s. We may, therefore, accept his statement that, as it stood completed, the Bibliotheke reached down to the year 60/59 BC, and reject his figures as mistaken in some way. And the mistake is certainly Diodorus'. But this bears directly on the date of writing and the original (and unfulfilled) scope of the Bibliotheke. There are a number of significant passages: 1. Sections 3. 38. 2f.; 5. 21. 2; 5. 22. 1: Diodorus accompanies references to Caesar's achievements in Britain with the undertaking to relate these when he comes to deal with Caesar's deeds. Taken together with the figures in 1. 5. 1, this all but proves that Diodorus' original intention had been to bring his narrative down to the year 46/5 when Caesar can be said to have rounded off his deeds with the final battle of the civil war. And if that was the original intention then clearly Diodorus cannot have begun writing before that date. 44 It has been suggested nevertheless that Diodorus was at work on the Bibliotheke at least as early as 56 BC. At 1. 44. 1 (cf. 1. 46. 7) Diodorus says that he visited Egypt in the 180th Olympiad (60/59-57/6 BC) when Ptolemy the New Dionysus (Ptolemy XI, 80-51 BC) was king. And a little further on, in computing the various periods of foreign as well as native rule in Egypt, he states that the Macedonians, the last of the foreign ____________________ 44 Rubincam, 1987: 325, 327, argues similarly. -18-

rulers, ruled for 276 years (1. 44. 4). The 276 years are consequently subtracted from 331/0 BC, the year Alexander captured Egypt according to Diodorus, and the resulting date, 56/5 BC (counting inclusively), is taken

as proof that Book I was already being written in that year. 45 But this cannot be. How could Diodorus start writing in the 50s and intend to bring his history down to 46/5 BC? 46 The problem of 1. 44. 4 disappears if we ignore the precise figure of 276 years and appreciate that the computations in 1. 44 are not done down to a specific year of any significance, but to the 180th Olympiad when Diodorus visited Egypt. That is all. There is no hint that he was writing at that time. As he puts it, 'they say that the country has been ruled by men for almost 5,000 years down to the 180th Olympiad during which we visited Egypt when Ptolemy the so-called New Dionysus was king' (1. 44. 1). Sixty-nine Olympiads intervened (reckoning inclusively) between the 112th Olympiad when Alexander conquered Egypt and the 180th, i.e. 276 years. It was, of course, misleading of Diodorus, as well as unnecessary, to translate the sixty-nine Olympiads into years. As it happens, the figure of 5,000 years itself is that of the original chronology of his source (almost certainly Hecataeus) which Diodorus has neglected to bring up to date, 47 in the same way that he neglected to correct the figures in 1. 5. 1. 48 2. Sections 1. 4. 7; 4. 19. 2; 5. 21. 2; 5. 25. 4; 32. 27. 1, 3: The deification of Julius Caesar is mentioned. The wording seems to point to the official recognition of Caesar as a god by the senate on 1 January 42 BC. (He gained the eponymia of god for his arete: 32. 27. 3.) The first passage comes from the general preface which Diodorus wrote and appended to Book 1 of the Bibliotheke at the work's completion. The rest of the passages, on the other hand, are part of the narratives of Books 4, 5, and 32. At 3. 38. 2 Diodorus mentions Caesar, but not his deification. This has therefore been ____________________ 45 Oldfather, p. ix; cf. Burton, 40 n. 3. 46 Burton indeed sees this, but because of her insistence that Book 1 was being written as early as 56 she creates considerable difficulties for herself: 40 f. 47 Murray, 1970: 145 n. 3: cf. n. 139. 48 Oldfather, p. viii, seems correct that D was in Egypt in the first year of the 180th Ol. On the other hand there is no reason to assume that the Roman citizen killed by a mob was connected with a Roman embassy, or indeed that there was a Roman embassy at all in Egypt at the time: τοὺς παρεπιδημου+̑ντας τω+̑ν ἀπὸ τη+̑ς Ι+̔ταλίας (1. 83. 8) need not mean an embassy. And, pace Sacks, 1990: 165, ἐπιδημία (1. 83. 9) could be of a short duration. -19-

taken by some as an indication that Books 1, 2, and 3 were written before Caesar's deification and the rest of the Bibliotheke after that date. 49 It would be nice if we could rely on this observation because it would provide us with a terminus ante quem as well as a terminus post quem for the writing of the Bibliotheke. Unfortunately for the view Caesar is also mentioned at 5. 22. 1 without being called a god. Diodorus is inconsistent if he is anything and we should not make too much of 3. 38. 2. Those who would have Diodorus writing the early books at least in the 50s (see above) need to prove that the passages in question are all later additions, added perhaps at the time the Bibliotheke was being revised. 50 Not only is there no such proof, but the indications are that Diodorus never did revise his work (witness e.g. the inappropriate figures discussed above and see pp. 137 ff.). 3. Section 12. 26. 1: Diodorus refers to the rostra which had 'at that time' (τοι+̑ς . . . τότε) stood in front of the Curia. As these were removed by Caesar c. 45 BC, Book 12 must have been written some years after that date. 51 4. Section 16. 7. 1: the Roman colony at Tauromenium mentioned here was almost certainly sent out in 36 BC.

5. Sections 13. 35. 3 and 16. 70. 6: in the first passage Diodorus says that Diocles' laws remained in force in many Sicilian cities until the time all the Siceliots were granted Roman citizenship. In the second, that the amphipoly (an eponymous magistracy) retained its importance at Syracuse for more than 300 years, from 343/2 when it was instituted by Timoleon, until the Romansn shared their citizenship with the Siceliots, which was the time the Bibliotheke was being written. It would appear then that Diodorus is not here referring to the grant of ius Latii to the communities of Sicily by Caesar ( Cicero, Ad Atticum 14. 12. 1), but to a full enfranchisement by Sextus Pompeius after 43 BC. 53 The references thus provide us with a terminus post quem for Books 13 and 16 ( 43 BC), though not with a terminus ante quem, 54 even if we accept ____________________ 49 Cf. Sartori, 1983: 547; accepted by Sacks, 1990: 171, who, however, points out that Caesar was being called a god from the day of his funeral. If so, this would push the terminus back to spring 44. 50 Thus e.g. Burton, 41 n. 1, though referring only to 5. 21. 2 and 5. 25. 4. 51 Cf. Casevitz, p. xi n. 1. 52 Cf. Rubincam, 1985: 521 f. 53 Sacks, 1990: 207ff. 54 As Sacks, 1990: 162 n. 6, 208, thinks. -20that Augustus in 36 Bc deprived all or most Sicilians of their recently gained citizenship: in 13. 35. 3 and 16. 70. 6 Diodorus is concerned not with the fortunes of the Roman franchise in Sicily, but with those of the laws of Diocles and the Syracusan amphipoly. The above evidence leads to the conclusion that Diodorus began writing some time in the late 40s at the earliest, with the initial intention of bringing his work down to 46/5. How long did the project last? The claims at 1. 4. 1, often treated with more respect than they deserve, are not to be countenanced, even if a large part of the thirty-year period were assigned to preparation. 55 The Bibliotheke is entirely derivative and Diodorus' methods slipshod, so much so that the work could have been dashed off in a very few years. A superior writer like Dio Cassius wrote eighty books in twelve years: 72. 23. 5. 56

A closely related question concerns the point at which Diodorus decided to stop with the year 60/59, contrary to his original intention, and the reason for so doing. Sacks develops rather an interesting theory. 57 The abandonment of the initial terminal date of 46/5, he thinks, was not fortuitous, but forced on Diodorus, a 'Greek émigré' living in Rome who 'venerated Caesar', but 'had reason to abhor Caesar's nephew', 58 by a deteriorating political climate. Long before he reached it, Diodorus had judged it unwise to treat the contentious period of 60-46 BC. Argued at length the thesis appears unnecessarily contrived and oversubtle for a writer of the level of Diodorus, even if we agreed with Sacks that the decision to stop with 60/59 was taken well in advance of it. And this does not quite follow from 32. 27. 3 where Diodorus attached to his description of the destruction of Corinth in 146 a eulogy of Caesar apropos of his later rebuilding of the city. Sacks ____________________ 55 Cf., most recently, Chamoux, p. xxxiii. Similarly Sacks, 1990: 161 ff. , following Rubincam, 1987: 324 ff., envisages D as engaged in 'research' from 60/59 down to 46/5 and then as settling in Rome and writing until c.30 when the Bibliotheke was published. And see p. 77 : D 'spent years of research in Alexandria, possibly working in the Great Library'! What, in any case, is the evidence that he began writing in 46/5? 56 Though ten years are also said to have been spent on gathering the necessary material. Nicolaus of Damascus, a contemporary of D, may have taken no more than ten years to write his massive universal history of 144 books: Wachsmuth, 1895: 105. But Nicolaus' dates are rather uncertain: Jacoby, FGH iic. 231. 57 1990: 160ff., esp. 172ff. Cf. also Rubincam, 1987: 327ff. 58 1990: 183. -21-

insists 59 that the account of the actual refoundation of Corinth was the correct place for such a eulogy and concludes that Diodorus must already have decided to go no further than 60/59. Perhaps, but as Diodorus never intended to take his narrative down to 44 when Caesar was assassinated (and epainoi in Diodorus commonly come at the end of a man's life: see pp. 5 ff.) any suitable point might do. Similar eulogistic comments which precede a person's death and serve to herald a fuller treatment of the subject further on are not unknown in the Bibliotheke, though generally Diodorus concludes his remarks with a promise to return to the matter in due course (cf. 15. 39; 16. 1. 3-6; 65. 9). Such a statement would not of course have been relevant to the purpose of the Byzantine excerptor who may therefore have omitted it. Why, in any case, should a fervent admirer of Caesar hesitate to describe his achievements, as he had promised (almost certainly several years after Caesar's death), in a Rome dominated by the dictator's nephew? The real reason why Diodorus decided to stop with 60/59 may be a great deal simpler. 60 It may well be that his chronographer ran out and at the same time he could not procure an account of contemporary events suited to his methods and purposes. It may even be that he had decided on forty for his number of books (see below), and having miscalculated (not for the first time: see p. 24 ) the length of one or more of those dealing with events in the first century, he decided to bring the work to an end with 60/59, as convenient a closing point as any. Such mundane reasons are far from impossible in the case of Diodorus. 61 The annalistic history of Books 11-40 presumably began when Diodorus had enough dated events at his disposal to make such a system practicable. That, however, is not likely to have been the case before the beginning of the series of Athenian archons. For the interval between the return of the Heraclids and the 1st Olympiad Diodorus claims to have reckoned by the reigns of the Spartan kings (cf. FGH244 F63b, 64). From the 1st Olympiad to the beginning of the archon series, the list of Olympiads would have sufficed for his purposes (cf. 2. 32. 3). When necessary, with the aid of the chronographer, he would doubtless have resorted to ____________________ 59 1990: 175 f. 60 Cf. also Stylianou, 1991: 394f. 61 For various other suggestions see Schwartz, RE v. 665; Oldfather, i, pp. xviii f.; Perl, 1957: 6f.; Laqueur, 1958: 285ff.; Burton, 39ff. -22-

reckoning by intervals from an epochal event, the Trojan War in his case. Indeed, he is not averse to using this method in later books too (cf. 14. 2. 4; 19. 1. 10; 20. 2. 3). Certainly Diodorus' annalistic system began before Book 10 (see 10. 3. 1 -- Pythagoras' floruit is dated by archon Thericles and the 61st Olympiad) and Book 9 may have been in that form in its entirety, to judge from the contents (Athenian sixth-century history, Croesus, Cyrus the Great, etc.). Book 8, however, can only have been so in part. It covered the Messenian Wars, some early tyrants, foundations, etc. Diodorus may have commented on his system as it was gradually introduced -- first the Olympiads, then the archons, and finally the consuls. Having decided on the overall plan Diodorus would then have turned his attention to the subject matter of the individual books; not necessarily all forty at once, and one might even question that Diodorus made up his mind on forty books from the beginning. But that is not at all unlikely. He may have adopted the number forty from Polybius whose work he tried to imitate in other respects. 62 To facilitate his task he must have divided the great length of time he had to cover into periods. That was only sensible. At 1. 4. 6f. he tells us that the Bibliotheke was divided into three major parts and this must be what he planned in broad outline at the outset. Certainly the hexad of pre-Trojan deeds, three Greek and three barbarian, represents his original plan. He must also have divided the period from the Trojan War to the death of Alexander, and that from the latter event to the end of the work, into smaller, more manageable units, though we can only guess what these were and how many books he assigned to each. The relevant statements he makes at various points are only of limited help as they refer to what has already been written and not to his original intentions. Nevertheless, it would be very surprising if Diodorus departed significantly from the original plan. From the anticipatory remarks found scattered throughout the Bibliotheke it is evident that Diodorus must have read, if only superficially, most of his sources, both the narrative and the chronographic, before he began writing. 63 In addition he had his

____________________ 62 Cf. Walbank, 1945b: 41; Hornblower, 1981: 25. His attempt to present his accounts in a year-by-year, region-byregion, form was probably borrowed from Polybius. 63 There are a great many of these remarks scattered throughout the Bibliotheke; see now Rubincam, 1989: 39 ff. To her tabulation on p. 56 add 16. 50. 8. -23- own knowledge, however deficient, of the past. It would not, therefore, have been particularly difficult to decide in advance, almost certainly with the aid of the chronographer, on the number of books for each period and even on their contents. For some of the books these would have been immediately obvious, for example, αἱ του+̑ Φιλίππου πράξεις (360-336) naturally belonged in one book, as did αἱ του+̑ Α+̔λεξάνδρου (336-323). The period from the Peace of Antalcidas (certainly a landmark and the beginning of Callisthenes' Hellenica) to the reign of Philip would readily suggest another. The less obvious books would in the end be defined by the more obvious ones. This method of working would have made for some inflexibility, but that would not have worried Diodorus; his framework constructed, the narrative material would simply have to fit it. Certainly on a number of occasions Diodorus explains that he cannot deal even with important events at length as that would destroy the symmetry of his work (cf. 1. 8. 10; 1. 29. 6; 2. 31. 10; 8. 12. 15). This does not constitute evidence for the above hypothesis and unfortunately there is no such evidence to be had. 64 However, it seems most probable that Diodorus decided on the number of books from the beginning and even on the specific contents of many of these books though obviously he was able to introduce modifications as he went along if necessary. For example, it appears that the fact that the pentecontaetia is split between two books, 11 and 12, is due to Diodorus' having run into difficulties with the size of Book 10 (see p. 104 ). The same is true of Book 13. Diodorus had intended to conclude with the beginning of Dionysius' Second Carthaginian War, that is, with the year 398, and he said so at 13. 1. 3. As, however, Book 13 grew to a length beyond that of the average book, he decided to finish instead with the end of the Peloponnesian War in Greece and the end of the First Carthaginian War in Sicily. Yet he did not hesitate to lie that the original intention had been fulfilled (13. 114. 3). The fact that this concluding paragraph is worded differently from all the others (almost as if by way of an apology) indicates that Diodorus was aware of the inconsistency. 65 ____________________ 64 D occasionally refers back to books, twice by number too (at 16. 46. 5; 20. 57. 6), but almost never ahead (the one exception is 11. 90. 4); but then neither does Polybius who certainly revised his work. See Rubincam, 1989. 65 Cf. ἐπιτετελεσμένης τη+̑ς προθέσεως at 13. 114. 3 with κατὰ τὴν ἐν ἀρχῃ+̑ πρόθεσιν at 11. 92. 5; 12. 84. 4; 14. 117. 9; etc.; see Laqueur, 1958: 282. I cannot, however, -24-

It should be noted that Diodorus is always anxious to define his books at the beginning and usually at the end of each too. For the historical period he was able to do so exactly: Book 10 ended with the year previous to Xerxes' crossing from Asia to Europe (11. 1. 1). Book 11 began with the year of Xerxes' invasion of Greece when Calliades was archon at Athens and the 75th Olympiad was celebrated (11. 1. 1-2) and ended in the year before Cimon's campaign against Cyprus (11. 1. 1; 92. 5; 12. 2. 3), and so on. These dated events were of course derived from the chronographer. 66

3. THE SOURCES (i) The framework

(a) The chronographer

The most obvious indication that Diodorus used a chronographic source is the series of Athenian archons, Roman consuls, and Olympiads which provides a chronological framework for the Bibliotheke. The archons and Olympiads quite evidently derive from a Greek chronographic source based ultimately on the system of Eratosthenes and Apollodorus. 67 What though is the origin of the colleges of Roman magistrates and indeed of the rest of the Roman material? The amount of this in Book 15 is so insignificant that it might as well be dealt with here as a whole. It is not impossible that Diodorus found the colleges in a Greek chronographer, perhaps even in the very same Greek chronographer he used for everything else. The Chronica of Castor of Rhodes is the obvious candidate, 68 especially as the work very conveniently reached down to 61/60 BC. Schwartz objects that ____________________ agree with his complicated explanation, 281 ff., based on D's misuse of different sources. I see the contradiction as evidence that the Bibliotheke was not revised. See also the preface to Book 17 (another large book) where D says that at the same time as the deeds of Alexander he intends to relate καὶ τὰ ἅμα τούτοις συντελεσθέντα ἐν τοι+̑ς γνωριζομένοις μέρεσι τη+̑ς οἰκουμένης, something he did not do, as he must have been aware; see further p. 137. 66 Cf. Perl, 1957: 8 ff. 67 Though far from being infallible (see pp. 45 ff.), the general reliability of D's chronographer is recognized by scholars; cf. Schwartz, RE v. 669; Hornblower, 1982a: 39 and n. 18. 68 As many have thought; cf. Perl, 1957: 141 n. 4. -25-

Diodorus cannot have been aware of the Chronica because in 40. 8 he states that he did not fix the dates of events prior to the Trojan War, since there existed no chronological record of them. 69 Castor, of course, had worked out such a chronological system. However, the text of 40. 8 (an excerpt from a Constantinian collection) breaks off in midsentence. At 1. 5. 1, on the other hand, where Diodorus makes the very same point, he goes on to complete the

sentence by adding that for the period before the Trojan War there existed no trustworthy (πιστευόμενον) chronology. What does seem to exclude the possibility that Diodorus used the Chronica is something else. In Diodorus Athenian and Roman magistrates are correlated in such a way that consuls enter into office during the year of the archon to whom they correspond and not, as it has sometimes been thought, six months before it began. 70 Thus Euthydemus the archon of 431/0 corresponds to the consuls of 430. 71 Castor, by contrast, appears to have equated the archon year 61/0 with the consul year 61. 72 In fact, the peculiarities of the fasti in the Bibliotheke are best explained on the supposition that Diodorus extracted this information from a list of chief Roman magistrates and then himself proceeded to introduce it into the framework of archons and Olympiads of his Greek chronographer on the basis of well-known synchronisms. 73 One might further accept the view that the fasti derive from a different source from the passages on Roman history. 74 The latter vary from brief notices which give every appearance of deriving from a chronographer -- they are almost always at the end of a year's narrative (cf. 11. 37. 7; 11. 40. 5; 53. 6; 68. 8; 12. 30. 6; 12. 34. 5; 12. 37. 1; 13. 6. 8; 42. 6; 14. 11. 6; 14. 34. 7; 43. 5; 15. 27. 4; 15. 35. 3; 47. 8) -- to fairly lengthy sections which clearly originated in a narrative source (cf. 12. 24. 2-25; 64. 1-3; 14. 113-117. 7). In three instances Roman notices are sandwiched between what appear to be Greek chronographic entries (16. 31. 7; 16. 36. 4; 45. 8). There are none at all in Books 17 and ____________________ 69 RE v. 665. 70 Perl, 1957: 4 and n. 2. 71 And not 431. This invalidates Meyer's solution of D's muddle over the beginning of the Peloponnesian War, Forschungen, ii. 504, followed by Gomme, HCT i. 4. 72 FGH 250 F5; Perl, 1957: 4f., 141f.; Cassola, ANRW ii. 30. 1. 743f. Contra Beloch, RG107ff. 73 Schwartz, RE v. 665; Perl, 1957: 83 f., 151 ff. 74 Perl, 1957: esp. 123 ff., 137f.; Cassola, ANRW ii. 30. 1. 739ff. -2618. In Books 19 and 20 Roman affairs are related regularly, almost every year, and in relative detail. It is thus possible to surmise that (a) Diodorus used consular fasti; (b) his Greek chronographer included brief notices on Roman history and, of course, synchronisms (with e.g. the Sack of Rome); and (c) Diodorus in addition employed a Romann historical handbook of some description, a narrative history perhaps, from which he extracted detail when he so chose. 75 The main distinguishing features of the fash in Diodorus are as follows. As a general comment it is a fact that the colleges are far less well preserved in Diodorus than, say, in Livy or the Fasti Capitolini (FC). The colleges are frequently incomplete and the names corrupt 76 as a result of careless handling, and Diodorus himself, though the main, is not the only culprit as shown by the serious differences in this respect between MS P and MS M. 77 More specifically, for the years 386/5-382/1 Diodorus repeats the colleges of 391/90-387/6 (394-390 V (Varronian). In 15. 75. 1 (367/6) he mentions a year of anarchy which is in stark contrast to the five-year anarchy (solitudo magistratuum) spoken of by Livy (6. 35. 10) and noticed in the FC (375-371 V). The college of 367 V is passed over and the four 'dictator-years' which feature in the FC (333, 324, 309, and 301 V) are unknown to him, as also to Livy. 78 As it is fairly evident that all existing fasti have a common origin, 79 the various differences between them, when not due to negligence or plain error, should be seen as the outcome of attempts to overcome certain difficulties of which the chief was the need to accommodate the lists to synchronisms known from Greek sources. 80 It is then possible to hypothesize that the repetition of the five colleges at the beginning of Book 15 was merely a device employed (possibly by Diodorus himself) for the purpose of raising the college of magistrates in whose year Rome was known to have been sacked by the Gauls to meet the year (387/6) in which this event had been dated by Greek historians. 81 The additional years of ____________________ 75 These conclusions are drawn primarily from and for Books 11-20. 76 This is readily shown by comparison with Broughton, MRR i. 89 ff. and CAH2, vii. 2. 629ff.; cf. Beloch, RG253 ff.; Perl, 1957: 31 ff., 90ff. 77 Cf. Vial, p. xi n. 1 and see also nn. in the commentary. 78 Perl, 1957: 107ff. 79 Beloch, RG4ff.; and for the consular tribunes in D and Livy, see Drummond, 1978: 80ff. 80 Cf. Stuart Jones, CAH1 vii. 321 f. 81 Pinsent, 975: 10; Cornell, CAH2vii. 2. 348 f.; Cassola, ANRW ii. 30. 1. 735 ff. -27-

anarchy and the 'dictator-years' were alternative means of achieving the same objective, and the Varronian system appears to have resulted from the adoption of both devices. 82 It is not impossible that the fasti used by Diodorus (like those used by Livy) did indicate a five-year period of anarchy, but that Diodorus erroneously reduced this to one year, in the same way that he mistakenly overlooked the college of 367 V. His repetition of the colleges of 394-390 V, therefore, may have been by way of compensating for this. 83 From 444 V until 391 V intermittently, and from 391 V until 367 V continuously, the consulship was replaced by colleges of consular tribunes. 84 The latter, unbroken sequence of consular tribunes is dated 383/2-364/3 in Diodorus, whose list is often in serious disagreement (mainly regarding the number of these annual magistrates) with that of Livy and the FC. For a detailed treatment of the matter the reader is referred to the work of Drummond. 85 His findings are cited in the commentary whenever applicable. His general conclusion, however, is worth stating here. Wherever fewer names are given by Diodorus than Livy and the FC the fault seems to lie with Diodorus and his copyists, and the additional names in Livy and the FC should not, therefore, be suspected as interpolations. 86 It is also interesting to observe that down to c.395 V the figure Diodorus gives, as is his habit, for the total number of annual magistrates before he names them is generally in agreement with the number of

names cited by Livy and the FC, though Diodorus himself may go on to name fewer tribunes. After c.395 V, however, it is usually the case that the figure and the number of names subsequently cited are in agreement, though often lower than the number of magistrates given by Livy and the FC. In addition, after c. 390 V Diodorus seems to have taken greater care in reproducing the names on his Jist. 87 No explanation suggests itself for these oddities except the obvious one, that Diodorus was inconsistent here as everywhere. ____________________ 82 Cf. Cornell, CAH2 vii. 2. 349; and see Perl, 1957: 106ff. and Werner, 1963: 173 ff. for the time-gap between the Varronian chronology and D's Roman dates. 83 Cf. Perl, 1957: 113f. 84 Designated military tribunes with consular powers in the sources. For the nature and functions of the consular tribunate cf. Drummond, CAH2 vii. 2. 193 ff. 85 1980; cf. 1978: 87ff. 86 A tendency so to view them goes back to Mommsen, Forschungen, ii. 224ff., and is still to be seen in studies such as Pinsent, 1975. 87 Drummond, 1980: 58f. -28-

The commonest and simplest way in which the archon is given is ἐπ+̕ ἄρχοντος γὰρ Α+̔θήνησι. This accounts for 118 of the 175 instances in Books 11-20 and it may well represent the formula of the chronographic source itself. 88 Frequently, however, this simple formula is rendered more elaborate and the elaboration, one assumes, is Diodorus' own. When the simple formula is used a year's narrative is usually sealed with the phrase ταυ+̑τα μὲν οὐ+̑ν ἐπράχθη κατὰ του+̑τον τὸν ἐνιαυτόν or some such phrase (for a more sophisticated example cf. 11. 26. 8). The more elaborate archon formulae on the other hand combine both functions of closing one year and introducing the next one. There does not appear to be any logical pattern in these variations. They are undoubtedly conscious and due to Diodorus' desire to vary his style, but on the whole he seems to have acted at random. An interesting break in his practice occurs at 13. 38.1 when for the rest of Book 13 and the whole of Book 14 the simple formula is shunned and the less simple one used. The simple formula reappears with a vengeance at the beginning of Book 15. What is important about the Roman formulae is the verb and consequently they exhibit a little more variety than the archon formulae. The simplest and most popular formula is Ρ+̔ωμαι+̑οι κατέστησαν ὑπάτους. This occurs 76 times. κατέστησαν also dominates the formulae introducing the tribunes. There are 21 such cases in all which thus gives us 97 instances of κατέστησαν. Two other forms of the same verb are found with some frequency -κατεστάθησαν and καθειστήκεσαν. The total is therefore 130, well over half the number of surviving year formulae. Other verbs occurring are (in descending order of frequency): διεδέξαντο, μετέλαβον, παρέλαβον, διέλαβε, παρειλήφεισαν, ὑπη+̑ρχον, διώκουν, εἰ+̑χον, ἐγένοντο, ἠ+̑σαν, παρέδοσαν, εἵλοντο. What was said about the archon formulae applies here too: no pattern is to be observed in the formulae variations with one possible exception which broadly corresponds with that noticed in the archon formulae. The simple form does not occur at all in Book 14. Instead we have in this book an unparalleled number, eight in all, of verb variations often in succession. ____________________ 88 On the analogy of the Atthides which began each year very simply with the archon; cf. Jacoby, 1949: 90, 94ff. Admittedly D does not supply the archon's demotic, with the single exception of the famous Demetrius of Phalerum (20. 27. 1). In this he resembles the Parian Chronicle. For the testimony available for the Athenian archons see Hicks and Hill, pp. 314 ff.; Develin. -29Little need be said about the formulae introducing the Olympiads except to note that they exhibit less variation than the archon and consul formulae, and that the formula of the three successive Olympiads of 408, 404, and 400 is varied. As we have said, the formulae variations were certainly the result of a conscious effort by Diodorus to enhance his style. It is interesting nevertheless that within the section covering the period 411/10-387/6 he consistently takes more pains than usual with the year formulae. Even more interesting is that, with the exception of the historiographical notices, the few literary entries in the surviving Bibliotheke are confined almost entirely to the years 406/5-397/6. This brief burst of literary chronology, therefore, seems to correspond to the rest of the pattern. It may all be the result of a simple coincidence helped along by Diodorus' haphazard method of working; or an external stimulus may have been at work. 89 Whatever the reason, one is led to assume that Diodorus occasionally felt the urge to be less monotonous with his formulae. Sometimes, as at 13. 38. 1-14. 1 10. 1, he sustained the urge better than at other times. It is necessary to look briefly at the names in the year formulae. How accurate are they? Diodorus' archon list compares quite favourably with other such lists, with, for example, the Parian Chronicle (see the references in n. 88). Any mistakes are not in the order of names, but in their form or spelling. As the list was intended to be of practical use, that is, the name of each archon was meant to 'fix' a particular year and it did not really matter whether it was historically correct or not, the great majority, if not all, of the mistakes, one suspects, are not due to Diodorus; he either inherited them or they were committed after him by copyists. Even so the number of serious mistakes is small, about twelve altogether. For example, the archon for 469/8 is Apsephion, not Phaion or Phaidon; for 458/7 Habron, not Bion; for 436/5 Lysimachus, not Nausimachus. Another twenty-one or so are less serious. For instance, Diodorus has Callias for Calleas ( 377/6) and Chariandrus for Charisandrus ( 376/5). Some fourteen names we ____________________ 89 The switch from Ephorus to Timaeus at the conclusion of the Sicilian expedition? Pinsent, 1975: 10 n. 5, is not quite right: the change in the formula introducing the Roman magistrates at the beginning of Book 15 (the simple form is resumed) is rather the tail-end of something which began earlier and does not, therefore, indicate a change of source, whether chronographic or narrative, at this point.

-30know from Diodorus alone. Of these only 'Akatestorides' ( 474/3) is certainly wrong; it should be 'Akestorides'. The position with the Olympic stadion victors appears similar to that of the archons though the absence of ample, reliable, and independent (in particular epigraphic) evidence makes it difficult to check Diodorus' accuracy, for usually he is only competing against writers such as Africanus and Pausanias, and against papyrus fragments of later chronographic works. But the divergences are on the whole insignificant. Diodorus, for example, gives Scamandrios for the victor of 476 while P. Oxy. 222 and Dion. Hal. 9. 18 give Scamandros. (ii) The other chronographic material: the obvious kind In addition to the year formulae Diodorus borrowed other information from the chronographer. The provenance of much of this is unmistakable and it falls into four classes. These are, in order of importance: (1) notices on the reigns of kings and tyrants (the kinglists), (2) literary chronology, (3) notices on the beginnings and ends of wars, (4) foundations of cities. Most of this information, and especially the kinglists and the literary notices, is given in small groups, often at the beginning (immediately after the year formula) or, more frequently, at the end of a year's narrative; occasionally in the middle (cf. 15. 9. 2; 93. 1; 16. 63. 2). There is usually no mistaking the nature of this material for it stands out by its formulae, arrangement, and subject matter. In most instances, therefore, there is a clear break with the narrative. When, however, the formulae do not occur and the subject matter is the same, it is difficult to separate chronographic material from narrative. This is infrequently the case with the kinglists and the literary chronology, more frequently so with the notices on wars, while the foundations are given mostly in the narrative. We shall therefore concern ourselves mainly with the kinglists; the literary notices and those on wars will receive less attention and the foundations none. We may begin by looking at the formulae of these entries. First the kinglists. This is the biggest group and it provides us with information on the reigns of rulers of a number of states, Greek as well as barbarian. It is more complete for some states, less so for others. A typical entry has the form: 'A, ruler of B, died -31- after X years of rule. C succeeded him and ruled for Y years' (cf. 13. 75. 1; 15. 60. 4). Often, though, more detail is volunteered, such as the cause of a ruler's death (14. 89. 2; 15. 23. 2); the relationship between successor and the person succeeded (cf. 14. 89. 1); other circumstances (cf. 15. 93. 1; 16. 36. 3). Thus the formulae themselves are uniform throughout, differing only in the amount of detail they offer. What is interesting about the literary chronology is that the great majority of notices concern historians. The number of entries on poets and philosophers is small, just nine, and their quality poor. In contrast, for the period from 477 to 354 the Parian Chronicle contains no fewer than eighteen notices, almost all on poets; and this is a very brief chronicle on stone. Clearly, though, the composer of the chronicle had a predilection for verse; he has nothing to say about historians. Diodorus' interests were reversed. As a 'historian' himself he paid special attention to historiographical chronology; the rest of the literary notices he largely ignored or lumped together indifferently. Witness, for example, 14. 46. 6. Ctesias receives separate and accurate treatment while a number of dithyrambists are thrown together, the last one even losing his ethnikon (cf. the Parian Chronicle 65, 68, 69). But with that Diodorus felt he had discharged his duty to verse chronology with which he could not generally be bothered. As with the kinglists, the formulae of the notices on historians are fairly uniform. The fullest entry will supply the name of the historian, his provenance, the subject of his work, how many years he covered, and in how many books. The entries, it may be noted, correspond to the beginnings and ends of the histories. The kinglists For the specific purposes of Book 15 our main concern is with the Macedonian, Spartan, and Persian kinglists. 90 But first some general remarks are necessary. For the principles governing the kinglists Ed. Meyer's masterly study 91 is still fundamental. The kinglists in Diodorus have gone through several stages and at each of these mistakes are likely to have crept in. As a simple list of local ____________________ 90 The kinglists are conveniently assembled by Schwartz, RE v. 666 ff., though he has mysteriously overlooked the Macedonian kings. For the Pheraean tyrants, Dionysius, and Clearchus of Heraclea see 60. 5 n.; 73. 5 n.; 81. 5 n. 91 Forschungen, ii. 440-53. -32-

rulers, related to the local calendar, the individual list was subject to the pressures of local politics though these would only have made it untrue to history; it would still have been continuous and self-consistent. Different Greek states laboured under different calendars and when the various lists came to be incorporated into date-tables based on a universally accepted system of chronology, the Olympiads and Athenian archons for the Classical and Hellenistic periods, discrepancies arose. The errors arising from the different year-systems apart, mistakes were also made when distributing the dates into the sort of handbook Diodorus used because of the misreading or the corruption of names, faulty calculation, etc. And then of course Diodorus was not above blundering when using the handbook, and neither were the copyists to whose efforts we owe our text of Diodorus. Our first task should be to reconstruct the original, selfconsistent lists. Only then can their historical evaluation, that is, their relation to real events, begin. As the lists are continuous we must reckon exclusively (by simple subtraction). We should also bear in mind that the lengths given for the reigns are likely to have departed less from those in the original lists than the dates for their beginnings and ends.

The Macedonian list

400/399 (14. 37. 6)

Archelaus II 7/Orestes/Aeropus II 6.

395/4 (14. 84. 6)

Aeropus II 6/Pausanias 1.

394/3 (14. 89. 2)

Pausanias 1/Amyntas III 24.

393/2 (14. 92. 3f.)

Amyntas loses kingdom, but regains it shortly afterwards and rules for 24 years. Some say that Argaeus ruled for 2 years and then Amyntas returned.

370/69 (15. 60. 3)

Amyntas III 24/Alexander II 1.

368/7 (15. 71. 1)

Alexander I I/Ptolemy Alorites 3 (see 71. 1n.).

365/4 (15. 77. 5)

Ptolemy Alorites 3/Perdiccas III 5.

(360/59 (16. 1. 3)

Philip II 24.)

(336/5 (16. 95. 1)

Philip II 24.)

(335/4 (17. 1. 4)

Alexander III 12.)

(324/3 (17. 117. 5)

Alexander III 12. 7.)

The bracketed entries may not derive from the chronographer; they are in the narrative. The seven years for Archelaus is a mis-

-33-

take by a copyist; it should be seventeen. 92 Archelaus is said to have been killed accidentally while his son Orestes was murdered by Aeropus. Aeropus himself died from natural causes, but his son Pausanias was murdered by Amyntas, the father of Philip. I shall ignore the two entries on Alexander's successors; figures are given at 18. 36. 7 and 19. 11. 5, but they do not appear to be chronographic entries and they make no contribution to the solution of the problems presented by the list. It was said above that in reconstructing the original lists the figures for the reigns should be treated with less suspicion than the dates for their beginnings and ends. At any rate this is certainly the case for Diodorus' Macedonian list. The 6, 1, 24, 1, 3, 5, and 24 years given for Aeropus, Pausanias, Amyntas, Alexander, Ptolemy, Perdiccas, and Philip add up to 64 and this is indeed the period from 400/ 399, Archelaus' death, to 336/5, Philip's death. The difficulties in Diodorus, therefore, arise from the fact that some of the transition years have been incorrectly dated. Specifically, the transition years of Aeropus/ Pausanias, Pausanias/ Amyntas, and Amyntas/ Alexander have been displaced, moved back, by a year. Aeropus did not die in 395/4 under archon Diophantus, but in the following year under Eubulides, for he was still alive when Agesilaus marched through Macedonia at the end of July or the beginning of August 394 (see below). This has caused the following entry, the murder of Pausanias after a year's rule and accession of Amyntas, to be moved back too; these events belong in 393/2 and not 394/3. Twenty-four years, the reign of Amyntas, from 393/2 bring us down to 369/8 and not 370/69 as Diodorus says. Alexander reigned for one year and was murdered and succeeded by Ptolemy in 368/7. This date is given correctly by Diodorus. The displacement, therefore, stops with the Amyntas/ Alexander transition year. What caused the displacement? The answer probably lies in the method the chronographer employed in distributing the available information in his date-table. The Macedonian list he had was presumably one of names and dates in sequence (cf. the list at D 7. 15), the latter perhaps given according to the Macedonian calendar. But he would also have been in possession of certain synchronisms with well-known Greek dates; for example, that ____________________ 92 All the MSS give seven. Eusebius 1. 227 (Schoene) gives the 'correct' 17; see Beloch, GG iii. 2. 50. -34-

Archelaus was killed in the year of Socrates' death or that Aeropus died in the year of the battle of Cnidus. 93 He fitted the kings into his handbook sometimes by using the synchronisms and/or the converted Macedonian dates, 94 and sometimes by reckoning by their reigns as indeed he did in the case of the Spartan kings (see below). Now the battle of Cnidus was fought under Eubulides, 394/3 (Lysias 19. 28). Diodorus, however, places it under Diophantus, 395/4, and 14. 84. 7, a chronographic entry, perhaps explains why he did so. Theopompus, he tells us, brought his Hellenica to an end with this year and the battle of Cnidus, having covered seventeen years; he began where Thueydides left off, that is, at the battle of Cynossema and 411/10 (13. 42. 5). Obviously the chronographer reckoned inclusively and seventeen years from 41/10 (archons Mnasilochus and Theopompus) gave him 395/4 (archon Diophantus). He had likewise reckoned inclusively the years covered by Thucydides' history: twenty-two, from 432/1 to 411/10 (12. 37.2; 71.2). The chronographer, therefore, wrongly dated the end of Theopompus' Hellenica and the battle of Cnidus to 395/4. The death of Aeropus may have been associated with this battle (14. 84. 6f.) and thus it too, as well as the accession of Pausanias, was placed under 395/4. He dated the following four events, the murder of Pausanias, accession of Amyntas, death of Amyntas, and accession of Alexander, by reckoning by the reigns of Pausanias and Amyntas, one and twenty-four years respectively, and all of these, therefore, are one year too high. 95 It would seem, however, that he dated the murder of Alexander not by his reign of one year which would have given him 369/8, but by a definite date he possessed and perhaps a synchronism too -- Alexander was murdered in the same year that Dionysius died, the year of archon Nausigenes, 368/7. 96 The displacement thus stood rectified. That Aeropus' death was displaced by a year was suggested by Beloch 97 and accepted by Geyer. 98 Both, however, blamed Diodorus for it and neither realized its full implications. Beloch, preoccupied with the necessity, as he saw it, to have Pelopidas ____________________ 93 Cf. Swoboda, 1883:17 ff., on the synchronistic character of D's chronographer. 94 'Equated' is perhaps the correct word to use. The equation would be with the archon or Olympiad year system. 95 The expulsion of Amyntas too (one year later?); see Syncellus 500 and 19.2n. 96 Cf. the Parlan Chronicle 73 which puts the two events in the same year; so does D of course. 97 GG iii. 2. 55f. 98 1930: 108. -35-

visit Macedonia and treat with Alexander in the summer of 369, insisted on 370 as the year of Amyntas' death. He therefore shortened Amyntas' overall reign to twenty-two years, from 392 to 370. 99 Geyer followed suit, but allowed Amyntas to reign from 393 to 370. 100 Hammond 101 is critical of Beloch: 'The error of Beloch was to reckon exclusively in year-units . . . but if we count inclusively and make 400/399 common to Archelaus and Aeropus, then six years is correct.' This nevertheless is to miss the whole point about the lists. While historiographical notices and notices on wars are sometimes reckoned inclusively, sometimes exclusively, the kinglists, as was said above, were designed to be continuous and inclusive reckoning is therefore out of the question. Doubtless the process of balancing the lists had involved some pushing and shoving. The difficulties facing the chronographer (for example, the different and not always reconcilable calendars) made this inevitable, but the end-result must have been a consistent and continuous series of reigns. The problem with Diodorus' list is simply one of 'mechanical' displacement.

By reckoning inclusively Hammond accepts 395/4 as the year of Aeropus' death and sees no contradiction with Polyaen.2. 1. 17 where a confrontation between Agesilaus and Aeropus in the summer of 394 is related. 'The point which is best attested is that Aeropus died of disease in 395/4 (D.S. 14. 84. 6), and right at the end of that year because he was outwitted by Agesilaus a month or two before the eclipse of the sun on 14 August 394 (Polyaen. 2. 1. 17 and 4. 4. 3; X. HG. 4. 3. 10).' But X. Ag. 2. 1 says that Agesilaus covered the distance from the Hellespont to Boeotia in under a month (cf. Nepos, Ages. 4. 4: 'in thirty days'). I see no reason to disbelieve this as do Beloch 102 and Hammond. 103 He stood poised on the Boeotian frontier on 14 August. He must, therefore, have been traversing Macedonia two weeks earlier at the most, at the very end of July and beginning of August when Eubulides had already assumed the archonship for 394/3. 104 Another line of argument is ____________________ 100 1930: 110. Likewise Cloché, 1960: 104 ff. 101 1979 : 168 and n. 2. 102 iii. 2. 217. 103 1979: 168 and n. 2. 104 Travelling by land the Peloponnesian reinforcements to Potidaea in 432 clearly reached Thrace from Corinth in some two weeks, three at the most: Thuc. 1. 59-60. 99 GG iii. 2. 56 ff. -36-

even more conclusive. When Agesilaus reached Amphipolis and consequently before he entered Macedonia and confronted Aeropus, he was met by Dercylidas fresh from the Peloponnese and bearing news of the Spartan victory near Corinth (X. H. 4. 3. 1). The battle was fought under Eubulides (Tod 105 -- the Dexileos monument ). Aeropus, therefore, survived the beginning of Eubulides' archonship by several weeks at the very least.

105

We may conclude that Diodorus' Macedonian list should be corrected by moving down the death of Aeropus/accession of Pausanias, as well as the next three entries, by one year. Thus reconstructed the list satisfies the needs of history too. Its data appear basically sound. Those of the other chronographic lists, which are substantially in agreement, 106 being less tidy and thus perhaps closer in some respects to the historical realities, should be used to elucidate and supplement Diodorus. 107 To date the death of Amyntas in 369/8 cannot unfortunately help us decide between the chronologies of Niese and Beloch for the 360s (see p. 449 ). For the question whether Amyntas was expelled once or twice and when, see 19. 2n. The Spartan list: the Agiads

408/7 (13. 75. 1)

Pleistoanax 50/Pausanias 14.

394/3 (14. 89. 1)

Pausanias 14/Agesipolis I 14.

380/79 (15. 23. 2.)

Agesipolis I 14/Cleombrotus I 9.

370/69 (15. 60. 4)

Agesipolis II 1/Cleomenes II 34.

309/8 (20. 29. 1)

Cleomenes II 60. 10/ὁ Α+̔ρέτα υἱός 44.

____________________ 105 I am not convinced by Beloch, GG iii. 2. 217 f. (followed by Jacoby, FGH iiiB Suppl. p. 156 ad 324 F47) that 'Nemea' was fought at the close of the year of Diophantus ( 395/4). Dexileos was killed C under archon Eubulides like the rest of the cavalrymen shown on Tod 104. ἐν Κορίνθῳ signifies the great battle fought between the rivers Nemea and Rachiani in the coastal plain much nearer to Corinth than Nemea ( Pritchett, 1965-85: ii. 73 if.). The 'battle of Corinth' is therefore more correct than the 'battle of Nemea'. It was known

in Athens as the 'battle of Corinth' (Andoc. 3. 18, 22; X. Ag. 7. 5; Demosth. 20. 52; Plato, Menex. 245e; Lys. 16. 15). I doubt if even as much as a month separated Corinth from Coronea; see Lys. 16. 16 καὶ οὐ πολλαι+̑ς ἡμέραις ὕστερον μετὰ ταυ+̑τα. 106 Once we know how to read them. For example, the two years of Argaeus (or one according to the Canon and three according to the Exc. Barb.) should be included in the twenty-four of Amyntas. Amyntas ὁ μικρός or Amyntas II, the pretender mentioned by Arist. Pol. 1311b, and given one year by the lists, 'reigned' concurrently with Pausanias. 107 Orestes, for example, ruled for a few years before he was murdered by Aeropus, the total of both reigns amounting to six years, and this is what we find in D. Aeropus presumably had reckoned his reign from the death of Archelaus. -37-

As we can see the Agiad list is virtually complete for the fourth century. The deaths in battle of Leonidas and Cleombrotus are mentioned in the narrative at 11. 10.4 (480/79) and 15. 55. 5 (371/0). The accession and death of Pleistarchus and the accession of Pleistoanax are not recorded. The figure 34 for Cleomenes at 15. 60. 4 is corrupt and should be ignored. Also corrupt is the name of Cleomenes' successor; it should read ὁ Α+̔ρεὺς υἱὸς Α+̔κροτάτου. Acrotatus was a son of Cleomenes (19. 70. 4; Plut. Agis 3. 6f.). But problems do remain, one in particular. It is impossible that Pausanias was exiled as late as 394/3. The battle of Corinth was fought very early in that year and Agesipolis was already king (p. 37 ; X. H. 4. 2. 9). Most probably Pausanias was tried immediately after his return from Boeotia in the summer of 395 (X. H. 3. 5. 25) and thus 395/4 is the best date for his exile. Meyer, therefore, 108 puts forward the suggestion that at Sparta, as in Babylon, 109 the less common system was followed of reckoning the year during which a king died as his last and the following year as the first of his successor. When the chronographers later utilized the local Spartan lists, unaware of the Spartan system, they assumed that a king's first year was also the year of his predecessor's death. Meyer consequently suggests that all Diodorus' dates for the Spartan kings ought to be pushed back by a year. Pausanias, therefore, fled and was condemned in 395/4 -- which is correct. One point, not spelled out by Meyer, requires elucidation. Between the beginning of the Attic official year on the new moon following the summer solstice, and that of Sparta on the new moon after. the autumn equinox, there existed a gap of some three months. During this time events would belong to different Attic and Spartan years. Chronographers would on the whole equate the two systems without concerning themselves about the gap. This three-month gap is the reason why the death of Agesipolis I and accession of Cleombrotus, the death of Cleombrotus and accession of Agesipolis II, the death of Agesipolis II and accession of Cleomenes are all dated correctly in Diodorus' list under 380/79, ( 371/0), 370/69; they occurred during it. Reckoned by Spartan years they happened in 381/0, 372/1, 371/0. The exile of Pausanias on the other hand took place in 395/4 by either system, for if it belonged in the Spartan year 396/5 then the year in Diodorus ____________________ 108 Forschungen, ii. 503, 511. 109 Ibid. 448. -38would have been 395/4 (the correct year), while in fact it is 394/3. The presence of both correct and incorrect (they are out by a year) dates in Diodorus serves as an indication of the soundness of Meyer's rule. The Agiad dates should, therefore, be reconstructed in accordance with his rule; see his table. 110 One correction: Leonidas was killed in the year of Calliades, 480/79, but in the Spartan year 481/0, so that Pleistarchus reigned for twenty-two and not twentyone years. The Spartan list: the Eurypontids

476/5 (11. 48. 2)

Leotychidas 22/Archidamus II 42.

434/3 (12. 35. 4)

Archidamus II 42/Agis II 27.

346/5 (16. 63. 2)

Archidamus III 23/Agis III 15.

338/7 (16. 88. 4)

Archidamus III 23/Agis III 9.

330/29 (17. 63. 4)

Agis III 9.

As we can see the Eurypontid list is in rather a sorry state. The last of the entries appears to be the remnant only of a chronographic formula; it has been assimilated into the narrative. The third comes from a digression and its proper place in the list is not therefore known. The end of the reign of Agis I I is not noticed, nor the accession of Agesilaus. Agesilaus' death is mentioned in the narrative at 15. 93. 6 (362/1) with no figures given and as a result the accession of Archidamus III is missing too. While the lengths of the reigns given are correct (with the exception of the figures for Agis III), the reigns themselves have been seriously displaced; those of the fifth century at any rate, for those of the fourth cannot be evaluated as there are hardly any. Demaratus, the predecessor of Leotychidas, appears to have been deposed in 491/0 and the twenty-two years of Leotychidas bring his exile down to 469/8. 111 There is thus a dislocation of seven years which has been carried forward to Archidamus II and Agis II. Archidamus II died sometime between the summer of 428 and early 426 and not in 434/3 (as the evidence of Thuc. 3. 1. 1; ____________________ 110 Ibid. 510f. 111 Rather than his death ( ibid. 507f. ) for his 22 years are unlikely to include his years in exile. The 14 years of Pausanias (13. 75. 1; 14. 89. 1) certainly do not include his long years in exile. On the other hand, Pleistoanax's 50 years certainly do include the exile years, but since these were in the middle his case cannot form a parallel. The year 469/8 is not necessarily too late for the Thessalian expedition and the king's exile. For a discussion see Gomme, HCT i.406f. -39-

26. 2; 89. 1, as well as of Diodorus himself, 12. 42. 6; 47. 1; 89. 1, proves). And Agis II cannot of course have died in 407/6 (434/3 minus 27) for he survived the Peloponnesian War by several years. What caused the dislocation? Perhaps a confusion between the respective archons for 476/5 (Φαίδωνος) and 469/8 (ΦαίδνοςAHLP/Φαίδωνος -- FIKM). This attractive suggestion was first made by Krüger and restated by Andrewes: 112 the exile/death of Leotychidas really belongs in 469/8, but Diodorus, misled by the Phaidon of 476/5, placed the event under the earlier year. The real name of the archon for 469/8 appears to have been Apsephion (on the evidence of the "Parian Chronicle"56; Plut. Cim. 8. 8; Diog. Laert. 2. 44), but this makes no difference for it must have been corrupted to Phaion/Phaidon by the time Diodorus' chronographic source was composed, or rather it was so corrupted in one branch of the transmission of the Athenian archon list; the branch which reached Diodorus. If this be not the case then the Krüger/ Andrewes suggestion collapses. Where I would disagree with Andrewes 113 is in the assumption that Diodorus is responsible for the confusion of the date of Leotychidas' death. If we look at 11. 48 we see that everything else in that chapter is in order, beginning with the year formula. And the entry on Leotychidas is sandwiched between the year formula and the entry on Anaxilas of Rhegium. The mistake, therefore, must have already stood in Diodorus' handbook. One can say more. If the error is Diodorus', why did he repeat it at 12. 35. 4 (434/3)? Neither of the archons for 428/6 could possibly be confused with the archon for 434/3, not even by Diodorus. No, but perhaps he reckoned by Archidamus' forty-two years from 476/5 which of course gave him 434/3. Yes, but if he placed Leotychidas' death/ Archidamus' accession by the archon why did he not place Archidamus' death/Agis' accession by the archon too? And vice versa: if he entered the latter by Archidamus' reign why did he not enter the former by Leotychidas' reign also? In any case, how could he place Archidamus' death/Agis' accession by Archidamus' length of reign? Diodorus presumably worked with his chronographic handbook before him. He would complete a year's account before proceeding to the next. He could not possibly, therefore, have moved forward by forty-two years to ____________________ 112 In White, 1964: 145 n. 21. 113 And Smart, 1967: 136; Gomme, HCT i. 53, and many others. -40place Archidamus' death/Agis' accession. I conclude that the mistake was not made by Diodorus, but was already present in the chronographic source. This provides us with an interesting glimpse into the working method of the chronographer. It would appear that he distributed the Spartan kings sometimes by the archons and sometimes by their reigns. 114 We have seen that he followed a similar method in fitting the Macedonian kings into his handbook. It is possible that the muddle arose long before Leotychidas, perhaps in matching up the archon and ephor lists, and that we are only witnessing its tail-end here. At the other end of the list the main problem is the two figures for the reign of Agis III; 15 years at 16. 63. 2 and 9 years at 16. 88. 4 and 17. 63. 4. The latter figure, twice repeated, is probably the 'correct' one, that is, it is the figure which stood in the original list. 115 It is not easy to account for the mistaken figure 15. Meyer 116 thought that the mistake was caused because Diodorus' source for 16. 63. 2 assumed that Archidamus III died in 346/5 straight after the Sacred War. While his son's reign was extended from nine to fifteen years to cover this eventuality, Archidamus' reign was for some reason, which Meyer did not explain, left at twenty-three years. Meyer further believed that this discrepancy of 8 years (346/5 minus 338/7) carries on right through all of Diodorus' Eurypontid dates and that it therefore explains the fifth-century displacement too. He reconstructed the

list accordingly. 117 Fortunately for Meyer the dates of Agesilaus are missing and consequently his hypothesis cannot be tested. In any case he failed to take into account the fact that 16. 63. 2 is part of a digression and was not necessarily under 346/5 originally. That rather invalidates his theory which depends on the year 346/5. A suggestion made earlier by Volquardsen, 118 which sees the tangle as Diodorus' own handiwork, seems more plausible and may well be correct. After concluding the story of the Sacred War (under 346/5) Diodorus ____________________ 114 This is essential if Andrewes' suggestion is to stand, though the suggestion cannot invalidate the working method of the chronographer here proposed if it is proved incorrect. 115 To obtain nine, however, we have to reckon inclusively which we said we must not do with the kinglists. Thus either 338/7 or 330/29 or both are wrong. Megalopolis certainly was not fought as late as 330/29; cf. de Ste Croix, 1972: 376f.; Bosworth, 1975: 27ff. 116 Forschungen, ii. 506. 117 Ibid. 507. 118 1868: 22. -41-

turned aside to relate the fate of the various groups of people who had profaned Delphi. The death in battle of Archidamus was part of that story and it is possible that Diodorus, mistakenly assuming that this occurred in 346/5, attempted to 'correct' the chronographer, with unfortunate results. This being the position it would be pointless to attempt a reconstruction of the Eurypontid list on the Meyer analogy. The real dates of the kings of course are another matter. The Persian list

465/4 (11. 69. 6)

Xerxes more than 20/Artaxerxes I 40.

425/4 (12. 64. 1)

Artaxerxes I 40/Xerxes 1.

424/3 (12. 71. 1)

Xerxes 1 (ἔνιοι γράφουςι 2 months)/ Sogdianos 7 months/Darius II 19.

405/4 (13. 108. 1)

Darius II 19/Artaxerxes II 43.

362/1 (15. 93. 1)

Artaxerxes II 43/Artaxerxes III Ochus 23.

(335/4 (17. 5. 3-6. 2)

A summary from Artaxerxes III to Darius III.)

The death of Darius III is mentioned in the narrative; no figures are given. The summary at 17. 5. 3-6. 2 derives from the narrative and not the chronographic source. 119

Little need be said about the list as a whole except to remark that it compares very favourably with the various other lists 120 and the evidence of the Babylonian tablets. 121 The death of Artaxerxes I and the accession of Darius II involve certain complications. At 12. 64. 1 (425/4) we are told that Artaxerxes died having ruled for forty years; Xerxes succeeded and ruled for one year. At 12. 71. 1 (424/3) the chronographer says that Xerxes died after a rule of one year, but adds a variant: ὠς δ+̕ ἔνιοι γράφουσι, μη+̑νας δύο τὴν δὲ βασιλείαν διαδεξάμενος ὁ ἀδελφὸς Σογδιανὸς, μη+̑νας ἑπτά. του+̑τον δ+̕ ἀνελὼν Δαρει+̑ος κτλ. The variant was true to history, whereas in a conventional kinglist which was meant to be of practical use Sogdianos could be omitted and the whole year given to Xerxes -perhaps because the latter was in office on a new year's day? 122 Of greater relevance here are the accession and death of Artaxerxes II, especially the latter which is a problem in Diodorus. ____________________ 119 Cf. Meyer, Forschungen, ii. 488. 120 See the tables ibid. 457 and 493. 121 Parker and Dubberstein, 1956: 17 ff. 122 On all this see Stolper, 1983: 223 ff. -42

According to Diodorus Darius II died in the year of archon Alexias (405/4) a little after the surrender of Athens. It is impossible to ascertain whether this information is a deduction by Diodorus from the fact that the death of Darius II was entered immediately after the entry on the fall of Athens in his handbook, or whether it is the considered opinion of the chronographer himself. If the former, it is next to worthless as an indication of the approximate time of the year of the king's death; if the latter, it deserves a little more respect. A Babylonian document of Darius II dated 17 September 405 records a receipt of rent due for the period covering the nineteenth year of Darius and the first of Artaxerxes II. 123 And we know that the Babylonian year 404/3 is Artaxerxes' first year. Consequently his accession year ran from April 405 to April 404. As the Athenian surrender may have taken place as early as March 404, 124 Diodorus' comment may in fact be correct. The year 405/4 as that of transition is therefore correct. Diodorus gives Artaxerxes 43 years and places his death in 362/1. He gives his successor Ochus 23 years. Thus the latter's murder falls in 339/8 and this is about right; cf. the summary at 17. 5. 3-6. 2. Also correct is the sum 43 + 23 = 66; and 66 is the number of years from 405/4 to 339/8. As the figure 43 is repeated twice, and as Diodorus' statements are internally consistent, we may accept that these data are the chronographer's own. Yet Artaxerxe II died not in 362/1, but between late November 359 and April 358. 125 He reigned for 46, not 43 years. The other lists are equally wrong though they give different figures. It is no longer possible to discover the source of error, but it is certainly older than Diodorus. 126 (iii)The other chronographic material: the less obvious kind There is a tendency to view Diodorus' chronographer as little more than a list of annual magistrates and Olympiads. All the evidence, however, points to a fairly detailed handbook which, on the Greek side at any rate, traced its descent back to Eratosthenes' Chronographiai and Apollodorus' Chronica. On the form and considerable contents of the latter we have the testimony of ps.____________________ 123 See Joannès, Textes économiques babyloniens récents, 103 ( 1982), no. 34 (2/ VI/19). 124 HCT iv. 12. 125 Parker and Dubberstein, 1956: 19. 126 Meyer, Forschungen, ii. 489, 496f. as well as his table p. 493. -43Scymnus, Orb. descr. 16 (= FGH 244 (Apollod.) T2). It is borne out by the fragments; cf. especially FF3, 13, 26, 44, 45. For the kind of information chronographers included in their Chronica cf. also FGH 245 F1 and 246 F1. Diodorus' chronographer would not have been radically different from the Chronicon of Oxyrhynchus (P. Oxy.12 = FGH 255) or the Olympiades of Phlegon of Tralles, FGH 257 (see in particular F12 from the Library of Photius) or P. Oxy. 2082 = FGH 257a. The information these last two supplied was not negligible and Diodorus' chronographer compares favourably with them. We have seen that the obvious chronographic entries have to do with kinglists, literary chronology, the beginnings and ends of wars, and the foundations of cities (p. 31 ). A great many of these are accompanied by circumstantial detail. Less obvious chronographic notices concern such events as campaigns, battles, the capture of cities, treaties, etc. These are usually not to be distinguished from the narrative, they are part of it, though occasionally their nature is not to be mistaken; cf. 11. 86. 1 (the five-year truce between Athens and the Peloponnesians 127 ); 14. 11. 1 (the murder of Alcibiades). Book 16 is particularly fruitful perhaps because Diodorus, deserted by Ephorus in the course of it, had to rely more heavily on the chronographer; cf. 16. 31. 6-7; 34. 3-4; 36. 2-5; 45. 7-9; 52. 9-10; 74. 1-2. 128 The presence of such material in the chronographer is in any case demanded by the fact that Diodorus did manage to fit the narrative sources he summarized into his annalistic scheme with a modicum of success. He could not have done so if his handbook had not the range and detail of, say, Phlegon Olymplades. 129 In searching for the hidden chronographic material there is perhaps one means by which some of them can be distinguished. At 12. 69. 1 we read: ἅμα δὲ τούτοις0 +03C0ραττομένοις περὶ τὸ Δήλιον ἐγένετο παράταξις Α+̔θηναίων πρὸς Βοιωτοὺς διὰ τοιαύτας τινὰς αἰτίας. One wonders whether a sub____________________ 127 Cf. Meiggs, 1972: 453f. 128 Most of these are acknowledged as chronographic entries by Schwartz, RE v. 668 (though cf. Hammond, 1937b: 90 f.). Many of them are of course kinglists, but I refer to the brief notices of campaigns and treaties which accompany them. 129 The significance of the chronographer in the Bibliotheke is seriously underestimated in some recent attempts to redeem D the historian. For instance, he is all but unmentioned in Sanders, 1987: 110 ff.; Sacks, 1990 ;

and Chamoux. Much of the credit given to D (e.g. for erudition and organizational ability) really belongs to the chronographer. Such an approach to the subject marks a retrograde step in Diodoran studies. -44stantial proportion of such instances, and there are a great many throughout the Bibliotheke, 130 are not in fact chronographic entries which have been made to introduce narratives. Thucydides certainly did not begin his account of the Delium campaign thus (4. 76), and neither did Ephorus, we may assume. One may then hazard a guess that the chronographic entry ran: καὶ περὶ τὸ Δήλιον ἐγένετο παράταξις Α+̔θηναίων πρὸς Βοιωτούς, and Diodorus has linked it to his abridged narrative by the ubiquitous expression διὰ τοιαύτας τινὰς αἰτίας. There are other very likely cases; cf. 12. 8. 1; 12. 9. 1; 12. 30. 2; 12. 38. 1; 15. 5. 1; 15. 20. 1; 15. 25. 1, to mention but a few. Of course not every such instance need represent a chronographic notice. 131 They should all be treated individually on their merits. (iv) The chronographer's errors Though fairly reliable the chronographer is by no means always right. 132 A number of chronological blunders in the Bibliotheke were certainly committed by him and not by Diodorus, a fact not always recognized. We have already seen that the chronographer is responsible for the dislocation of the Macedonian and Spartan kinglists (pp. 34 f.). And at 11. 86. 1 the five-year truce between Athens and the Peloponnesians is noted, by the chronographer, under 454/3 which must be wrong. 133 We may now look at the vexed question of the misdating of the beginnings and ends of wars. At 12. 37. 2 under 432/1 a chronographic entry correctly informs us that Thucydides began his history of the Peloponnesian War with this year. 134 Yet the outbreak of the war, the attack on Plataea, comes under Euthydemus, 431/0, headed by the statement: ἐπὶ δὲ τούτων. . . ἐνέστη πόλεμος ὁ κληθεὶς Πελοποννησιακὸς. . . ἀναγκαι+̑ον δ+̕ ἐστὶ. . . προεκθέσθαι τὰς αἰτίας αὐτου+̑. We should see this as a chronographic entry which Diodorus has utilized to head the narrative of this year (see above). ____________________ 130 Cf. McDougall, s.vv. 131 Cf. 12. 59.3=Thuc. 3. 92. 1. 132 Cf. e.g. Hornblower, 1990a: 74. 133 Cf. Meiggs, 1972: 454. 134 Correctly, pace Gomme, HCT i. 52 n. 4. Thucydides did begin with 432/1. His famous κατὰ θέρη καὶ χειμω+̑νας system was not applied before this date -Thuc. 2. 1. -45-

The incorrect dating of the beginning of the war is thus due to the chronographer. How did it come about? We see that both the Peace of Nicias (12. 74. 6) and the conclusion of the war (13. 107. 5) are dated correctly, and equally correct are the lengths given for the war as a whole, twenty-seven years, and for the Archidamian war, ten years. The error, therefore, may have arisen by the chronographer reckoning backwards from the end of the war or from the Peace of Nicias. In trying to make sense of such confusions we should distinguish between historiographical entries and notices on wars. The two were probably worked out and introduced into the handbook separately. Similarly, at 14. 86. 6 (394/3) we are told, by the chronographer presumably, that the war now begun, called the Corinthian, lasted for eight years. In fact the war began in the previous year and Diodorus himself relates the formation of the anti-Spartan alliance and the battle of Corinth under 395/4 (14. 82. The war called the 'Boeotian' at 14. 81. 3 (under 396/5) is not a mistake for the Corinthian War, but must refer to the fighting in Boeotia). The Corinthian War was ended by the King's Peace in 387/6, and the chronographer may have hit on 394/3 by inclusive reckoning of the eight years from the Peace. At 15. 25. 1, a probable chronographic notice (see above), the Boeotian War is dated as beginning in 378/7. No figure for its duration is given. In reality the war broke out in 379/8. As in the case of the previous two examples, we may surmise that 378/7 is the result of the chronographer reckoning backwards from the conclusion of the war. The three years of the Social War are likewise correct if we reckon inclusively; it began in 358/7 and ended in 356/5 (16. 7. 3; 22. 2. The four years given at 16. 22. 2 is probably a mistake). Diodorus' dates for the Sacred War are a more formidable problem and they may or may not be amenable to explanation along similar lines. This is because of the difficulty, one ancient scholars, no less than modern ones, may have experienced too, of deciding what events precisely constituted the beginning and what the end of the war. At 16. 14. 3-5 (under archon Agathocles, 357/ 6) three historiographical notices inform us that: (a) Demophilus, the son of Ephorus, narrated the Sacred War which his father omitted, beginning with this year, which saw the capture of Delphi and the pillage of the oracle by Philomelus. The war lasted the sacred treasure. (b) Callisthenes concluded his Hellenica with the capture of the sanctuary and the lawless action of Philomelus (cf. 14. 117. 8), and (c) Diyllus began his history with the pillage of the temple. The Sacred War itself (ὁ κληθεὶς ἱερὸς πόλεμος), on the other hand, is clearly marked by two chronographic notices, its beginning at 16. 23. 1 (under archon Callistratus, 355/4) where it is given nine years (note that the phrase διὰ τοιαύτας τινὰς αἰτίας links the entry to the narrative), and its end at 16. 59. 1 (under archon Archias, 346/5); here it is said to have lasted for ten years (again, we may note the link-phrase τύνδε τὸν τρόπον). Duris too thought the war had lasted for ten years, adding that it was brought to an end in the tenth (FGH 76 F2), and so did Aeschines, a contemporary (2. 131; 3. 148). There is no serious problem with the figures nine and ten, the result it would seem, of both inclusive and exclusive reckoning. Now it is probable, as often thought, that Diodorus' information is in basic agreement with historical reality. If that is so, there is nothing more to say except to remark that the chronographer has done his job as well as we know he can: the Phocians captured Delphi in the course of the Attic year 357/6 (spring/early summer 356), but the political circumstances were such (the continuing Social War and the attitude of Thessaly) that the Amphictyons did not officially declare war against them until 355/4; and the war was officially ended by the meeting of the Amphictyons

which passed sentence on the Phocians and appointed Philip to preside over the celebration of the Pythia held shortly afterwards (Attic year 346/5). The eleven years of war at 16. 14. 3, it would seem, are meant to include the whole sequence, from the capture of Delphi until the punishment of the Phocians. Diodorus, of course, as is his way (see p. 135 ), does not only include the Amphictyonic meeting and the Pythian festival under 346/5 (and headed by the chronographic entry about the ending of the Sacred War: 16. 59. 1), which is correct, but also the last actions of the war which led to the agreement between Philip and Phalaecus and the surrender of the Phocians, which is wrong. These events took place in the first half of 346 under archon Themistocles (347/6). Immediately after mentioning the surrender of the Phocians Diodorus adds (from Demophilus) that Philip had thereby concluded the Sacred War (16. 59. 4). There was, therefore, something of a conflict about this point between his historiographer and his chrono-47-

grapher. Not so, however, about the time the Sacred War was declared. If this event is told under 354/3 (16. 28. 4), whereas the chronographic entry about the beginning of the war is under 355/4, this is because of the way Diodorus has distributed the summarized narrative of Demophilus. The main scholarly controversy concerns the date the Amphictyons declared war. According to Diodorus' chronographic source, as we have seen, this happened in 355/4 and many scholars concur. Others disagree. In their view the epigraphical evidence from Delphi shows that the Sacred War was declared a year earlier, in autumn/early winter 356 (archon Elpines, 356/5). 135 If this is right, the likely explanation for the chronographer's error is that, as in the cases cited earlier, he counted back from 346/5 and therefore gave 355/4 as the year the war was declared. The historiographical entries at 16. 14. 3-5, on the other hand, as commonly agreed, correctly mark the date of the seizure of Delphi (thus also Paus. 10. 2. 3) with which Callisthenes concluded his Hellenica and Demophilus and Diyllus commenced their coverage of the war. Finally, a blunder very probably committed by the chronographer, 136 though in this case no likely explanations are forthcoming. In Diodorus the Kερκυραι+̇κά are recounted under the four successive years 439/8-436/5. These dates are wrong; they are too high by some three years. 137 However, there can be little doubt that the responsibility for the original mistake, the date of the outbreak of hostilities, which caused Diodorus to distribute the Ephoran narrative under the wrong years, lies with the chronographer: ἅμα δὲ τούτοις πραττομένοις κατὰ τὴν Ε+̔λλάδα συνέβη τὸν Κορινθιακὸν κληθέντα πὸλεμον λαβει+̑ν διὰ τοιαύτας τινὰς αἰτίας, we read at 12. 30. 2 (439/8). This may be accepted as a chronographic source with some confidence. The cause of the error cannot be known, but it is interesting that the five-year truce between Athens and ____________________ 135 See, most recently, Buckler, 1989: 148ff., a refinement of an older view (see e.g. Beloch, GG iii. 2. 264ff.). Buckler's analysis of the evidence of D is unsatisfactory, 1989: 155f. Not only does he not distinguish between different kinds of chronographic entry (he lists them together as though they were all of the same sort), but he seems unaware of the role and nature of the chronographer in D. For the chronology which puts the emphasis on D's testimony, by no means undermined by the evidence of epigraphy, see Hammond, 1937a: 54ff.; Griffith, 1979: 227 ; Hornblower, 1983: 246f. 136 "Pace" Gomme, HCT i. 198, who sets the blame squarely on D's shoulders. 137 Gomme, HCT i. 196 ff. -48Sparta is likewise dated too high by three years (11. 86. 1). Perhaps both events were wrongly reckoned from the same point -- the end of the Peloponnesian War? (At 13. 8. 8 (414/3) the Decelean War is given twelve years instead of nine, another three-year discrepancy.)

(b) The narrative sources The 'traditional' position is that Diodorus generally relied on one main authority at a time. First established by Volquardsen in a pioneering study published in 1868, the principle is currently under attack though in the view of the present writer it remains unshaken in its essentials. The principle rests securely on the simple fact that whenever he can be checked, that is, by comparing his text against those of surviving sources, it can easily be demonstrated that Diodorus tended to follow closely one source for as long as possible. That being the case, it is but sensible to assume that that was his usual method -- unless it can be proved otherwise. 138 Those who argue for the use of a multiplicity of sources, always and as a matter of course, also tend to regard Diodorus as something more than a mere epitomizer and an incompetent one at that. 139 What follows presupposes the correctness of the 'traditional' view. As far as the Greek and Persian narratives of Books 11-15 (16) are concerned there is very wide agreement that Diodorus epitomized Ephorus. This was firmly established long ago and requires no detailed argumentation. 140 The evidence is impressive: ____________________ 138 Cf. Hornblower, 1981: 19ff. and see further below on D's methods. 139 The most recent and sustained attempt to 'rehabilitate' D (Sacks, 1990) cannot be deemed a success; cf. Stylianou, 1991. It should be stressed that the conception of D as a noteworthy writer capable of synthesizing accounts out of a variety of sources has not been proved for any section of the Bibliotheke, not even for Book 1, in spite of claims to the contrary, e.g. by Burton, 1 ff. Cf. Murray, 1975: 214f. Murray's point is well taken that if the composition of the account of Egypt in Book 1 'was the work of Diodorus himself, then he must be acclaimed one of the great historians of the Hellenistic period' ( 1975: 215). Schwartz, RE v. 670ff., Jacoby, RE vii. 2750ff.; id., FGH 264 F25 n., and Murray, 1970: 144ff.; id., 1972: 207; 1975: 214ff., appear quite correct that, as in other books, so in Book 1 D relied on one main source (Hecataeus of Abdera in this

instance). The addition of material from a second source (chapters 37-41 probably derive from Agatharchides who was himself employed as a main source elsewhere in the Bibliotheke), and D's own comments interspersed here and there, do not invalidate the thesis. 140 By Cauer, 1847; quardsen, 1868: 52ff.; Collman, 1869; Holzapfel, 1879; Schwartz, RE v. 663 ff. -49(1) the agreement in tone and structure of Books 11-15 (16), standing, as they do, in sharp contrast to Books 1820; 141 (2) the mention by Diodorus of Aeolis and/or Cyme, the home-town of Ephorus, more frequently than their importance seems to warrant. 142 (3) Above all, there is the striking agreement between the fragments of Ephorus and the corresponding passages in Diodorus. On the other hand, the extent to which Diodorus employed other, 'secondary', sources to supplement Ephorus is a thorny question. We have seen that a fair amount of the moralizing is Diodorus' own rather than Ephorus' (pp. 3 ff.). And it would appear that some of the factual information too (apart from the chronographic material) did not originate with Ephorus. For example, 12. 38 should not be assigned to Ephorus. When Diodorus says that he has given Ephorus' aitiai of the Peloponnesian War (12. 41. 1), we should understand him to mean chapters 39 and 40. 143 The anecdotal stories about Myronides and Tolmides at 11. 81. 4f. and 11. 84. 2-5 may derive from a biographical handbook of the kind he occasionally employed for the history of the West (cf. Polyaen. 1. 35; 3. 3 and see below), and the same may perhaps be true of 15. 44 (but see n. there). Otherwise my inclination is to minimize Diodorus' reliance on sources other than Ephorus. This is in any case a reasonable assumption to make for an epitomator like Diodorus. The quest for Diodorus' sources for the history of the Greek West, by contrast, is considerably more complicated and still highly controversial. As it is necessary to adopt some kind of position, an examination of the matter is unavoidable. Diodorus, of course, was himself a western Greek writing, one supposes, primarily with a western readership in mind, and he might consequently have been disposed to take greater pains here than with the history of other regions. We should also expect him to volunteer more of his own knowledge and that seems to be the ____________________ 141 On which see Hornblower, Cauer, 22ff. 142 On the propensity of Ephorus to push Cyme to the fore at every conceivable opportunity: FGH 70 F236. The nature of the references too is significant; see e.g. 11. 36. 5f; 11. 37. 1; and Burn, 1984: 550f. 143 I therefore agree with Jacoby, FGH 70 F196n. But cf. Meyer, Forschungen, ii. 329ff. and Schwartz, RE v. 680, who argue otherwise. Sections 38. 2-4 give a ludicrous tale, different both in tone and content from 39-40 (these two chapters are in line with Plut. Per. 30ff.), as Schwartz himself admits. Perhaps 38. 2-4 was borrowed from a Hellenistic handbook of political stratagems and awkwardly attached to Ephorus' summarized aitiai by means of the last sentence of 38. 4. -50-

case. 144 This, however, should be balanced against the fact just noted that Diodorus was an epitomator, one of limited abilities too, and as such he would always seek to simplify his task. His sources should not, therefore, be multiplied unnecessarily, especially within single narratives. His usual method, as we have seen, was to rely on one of the standard works at a time, and I see no reason at all for supposing that he abandoned this method for the western narratives. The obvious sources indicated are Ephorus and Timaeus. All the same, down to at least 357/6 he had Ephorus' western narratives ready to hand, sometimes (perhaps even always) in the same books as the eastern narratives and with the connections between the two highlighted (see pp. 89 ff.), and good evidence is required before suspecting an alternative source. 145 This may serve as a principle. It may be apposite to stress at this ____________________ 144 Cf. 13. 35; 13. 90. 5-7; 14. 16. 3-4; 14. 95. 4-5; 16. 70. 6; 16. 82. 4-5; 16. 83. 2-3. Also n. 163 below. Most of this material seems to be D's own, seeking either to 'improve' his sources or to highlight his native town of Agyrium. 145 For most scholars, Ephorus and Timaeus are the only serious candidates here. Volquardsen, 1868, held that Timaeus was responsible for almost all of D's western narratives in Books 11-16. Schwartz, RE v. 681, 685ff., endorsed Volquardsen's thesis in a modified form which admitted a greater use of Ephorus, for instance in Book 15. Others who consider Timaeus as the main source, supplemented to a greater or lesser degree by Ephorus, are Barber, 1935: 160ff., Stroheker, 1958: 11ff., Meister, 1967, and now Pearson, 1987. In fact Pearson has all but returned to Volquardsen's position. Laqueur, RE via. 1082ff., on the other hand, considered that the framework of D's western history was Ephoran, but with the very frequent interspersion of passages lifted from Timaeus. A more sensible version of Laqueur's view is that of Jacoby, FGH iiib (text) 529, 541f., which sees Ephorus as the main, and Timaeus as the secondary, source. See also Lewis, CAH2 vi. 121. Caven too, 1990: 2ff., sees Ephorus as basic though reinforced with information from Timaeus, Philistus (directly), and from many other sources as well. His description of D as a kind of Pliny, indefatigably gathering material from all directions, is not of course one I can share. Sanders, 1981: 394ff., 1987, who also believes that D consulted a wide variety of sources, proceeds on the principle that Philistus was the main source for the Dionysian narratives (though not for Book 15), but used directly by D and not through the means of Ephorus or Timaeus. This view seems to me misconceived and based on a series of mostly erroneous assumptions (see pp. 63, 70 and cf. nn. 165, 174, 178, 179, 194, 198). It is worth mentioning the work carried out in the Istituto di Storia Antica of Palermo by E. Manni and his pupils. The 'Palermo School', as it were, set itself the task of ascertaining the sources of D and, more generally, of re-evaluating the Greek historians of Sicily. Its findings can be followed in the local journal Kokalos. But one should first read Walbank's penetrating assessment, 1968-9: 476ff. Walbank all but destroys the 'School's' chief tenet that for the Sicilian narratives in Books 8-24 D followed an intermediary source, the Sicelica of Silenus of Caleacte ( FGH175), a most obscure figure who possibly never even wrote a history of Sicily. -51 point that because of Diodorus' erratic ways no worthwhile pattern is discernible from the incidence of named fragments. All the same, it does indeed appear to be the case that named fragments occur when Diodorus is taking greater pains (which he does do from time to time); and they are mostly the outcome of a clash between the sources he is using, whether between the chronographer and the narrative source (e.g. at 14. 11. 1; 15. 60. 5), or between narrative source and narrative source (e.g. at 13. 54. 5); or occasionally when Diodorus is impressed by his source making a show of something (e.g. at 13. 41. 3). But even here we must be careful: Hieronymus, for instance, is not cited once as a source, and yet clearly Diodorus made a better job of abbreviating him than Ephorus. I suspect that he did not encounter any contradictions in Books 18-20. Most scholars regard 11. 20-6 either as a continuous series of excerpts from Timaeus, 146 or as an abridgement of Ephorus to which nevertheless material from Timaeus was introduced, 147 or as a combination of Ephorus, Timaeus, and at least one other source. 148 In accordance with the principle we have set, the question to be asked is, is there anything in these chapters which could not come from Ephorus? A major argument of those who credit Timaeus with the chapters is the highly favourable, indeed eulogistic, treatment of Gelon, and the elevation of the victory over Carthage to the level of that over Persia. But why should that be surprising in Ephorus? The very concept of an alliance between Persia and Carthage, such as described by Ephorus, suggests the belief that what was accomplished by the western Greeks was no less significant than the repulse of Xerxes. This, after all, is how Pindar saw it, and the scholion ad Pythia I146 ( FGH 70 F186) shows that Ephorus concurred: by his defeat of the Carthaginians at Himera Gelon freed not only Sicily, 'but also the whole of Greece'. It is true that Polybius, in a well-known passage, censures Timaeus for wildly exaggerating the achievements of the Siceliots (12. 26b), but the strictures, justified one supposes, are not quite borne out by Diodorus' text. Diodorus' praise of Gelon is ____________________ 146 e.g. Jacoby, FGH 70 F186n.; Meister, 1967: 42f.; Pearson, 1987: 132ff.; Walbank, 1989/90: 41ff.; Ameling, 1993: 26ff. 147 Laqueur, RE via. 1083ff. 148 Cf. Brown, 1952: 346ff. -52-

admittedly extravagant, but not to an extent that a fourth-century panhellenist from Asia Minor like Ephorus would find objectionable. The claim at 11. 22. 6 that no one before Gelon, so far as memory served, had employed such a stratagem, or achieved such signal success, against the barbarians, is paralleled, for example, by what is claimed for

Cimon and the victory at the river Eurymedon in 11. 61. 149 In fact, far from pointing to Timaeus, 11. 22. 5-24. I seems to reveal the pen of Ephorus. It may be compared to 11. 11; 11. 46. 1-3; 11. 58. 4-59. These are the sort of comments one should expect from him. The comparison between Gelon and the other Greek leaders, much to the former's advantage, is not incompatible with Ephorus. On the contrary, he seems to have liked such contrasts which he used to enhance any moral points he tried to make (cf. e.g. 11. 59. 1-2). In any case, why should not someone like Ephorus, writing a century and a half after the events, declare, after looking at the facts available, that of all the wartime Greek leaders Gelon was the most successful (which is true), and his victory at Himera more decisive than Salamis and Plataea (which is equally true). It is noticeable nevertheless that 11. 23 begins with the information that many historians have compared the Siceliot victory with those in Greece, but they have not all arrived at the same conclusions. Ephorus, therefore, may have been less unequivocal than Diodorus. On the other hand, the opinion that virtue and benevolence in a ruler were the best means of ensuring the goodwill of his subjects, is an Ephoran commonplace (see pp. 112 ff.). We should not of course blame Ephorus for the statement that Gelon 'grew old' in the exercise of his rule. This is just Diodorus.

The synchronism between Himera and Thermopylae (11. 24. 1) is often thought to have been invented by Timaeus, but this is unwarranted. Herodotus of course reports a synchronism between Himera and Salamis (7. 166), and the earlier dating of Himera in Diodorus may indeed represent an attempt (surely Siceliot) to enhance the glory of Gelon and Sicily. But Timaeus was neither the earliest nor the only Siceliot patriot, and it is likely that Ephorus found the synchronism in the Siceliot sources he employed, in Antiochus, for example, or in Philistus. At any rate, ____________________ 149

See esp. 11. 61. 7. Pearson, 1987: 137, considers the story of Gelon's stratagem preposterous, and therefore Timaean in origin. But the stratagem said to have been used by Cimon at the Eurymedon may strike one as equally preposterous.

- admittedly extravagant, but not to an extent that a fourth-century panhellenist from Asia Minor like Ephorus would find objectionable. The claim at 11. 22. 6 that no one before Gelon, so far as memory served, had employed such a stratagem, or achieved such signal success, against the barbarians, is paralleled, for example, by what is claimed for Cimon and the victory at the river Eurymedon in 11. 61. 149 In fact, far from pointing to Timaeus, 11. 22. 5-24. I seems to reveal the pen of Ephorus. It may be compared to 11. 11; 11. 46. 1-3; 11. 58. 4-59. These are the sort of comments one should expect from him. The comparison between Gelon and the other Greek leaders, much to the former's advantage, is not incompatible with Ephorus. On the contrary, he seems to have liked such contrasts which he used to enhance any moral points he tried to make (cf. e.g. 11. 59. 1-2). In any case, why should not someone like Ephorus, writing a century and a half after the events, declare, after looking at the facts available, that of all the wartime Greek leaders Gelon was the most successful (which is true), and his victory at Himera more decisive than Salamis and Plataea (which is equally true). It is noticeable nevertheless that 11. 23 begins with the information that many historians have compared the Siceliot victory with those in Greece, but they have not all arrived at the same conclusions. Ephorus, therefore, may have been less unequivocal than Diodorus. On the other hand, the opinion that virtue and benevolence in a ruler were the best means of ensuring the goodwill of his subjects, is an Ephoran commonplace (see pp. 112 ff.). We should not of course blame Ephorus for the statement that Gelon 'grew old' in the exercise of his rule. This is just Diodorus. The synchronism between Himera and Thermopylae (11. 24. 1) is often thought to have been invented by Timaeus, but this is unwarranted. Herodotus of course reports a synchronism between Himera and Salamis (7. 166), and the earlier dating of Himera in Diodorus may indeed represent an attempt (surely Siceliot) to enhance the glory of Gelon and Sicily. But Timaeus was neither the earliest nor the only Siceliot patriot, and it is likely that Ephorus found the synchronism in the Siceliot sources he employed, in Antiochus, for example, or in Philistus. At any rate, ____________________ 149 See esp. 11. 61. 7. Pearson, 1987: 137, considers the story of Gelon's stratagem preposterous, and therefore Timaean in origin. But the stratagem said to have been used by Cimon at the Eurymedon may strike one as equally preposterous. -53 it is evident from the scholion ad Pindar, "Pythia" I 146 ( FGH 70 F186) that Ephorus, unlike Herodotus, accepted the story (the same story presumably which Diodorus might have related if he had so chosen; cf. 11. 26. 4) that Gelon had every intention of helping the Greeks, and was in fact on the point of doing so, but was prevented by the Carthaginian invasion. Furthermore, the statement that by his victory Gelon freed not only Sicily 'but also the whole of Greece' does perhaps carry the additional implication that Ephorus dated Himera before Salamis: not only was the battle of Himera of equal importance with Salamis and Plataea, but by being won first, as Diodorus explains, it inspired the mainland Greeks to win their own battles (11. 23. 2), thus ensuring the salvation of Greece. And there are two other points which seem to declare Ephorus as the source behind these chapters. First, and the synchronism between Himera and Thermopylae apart, there is the pairing of the two battles as τήν τε καλλίστην νίκην καὶ τὴν ἐνδοξοτάτην ἡ+̑τταν (11. 24. 1). This instantly brings to mind 11. 11. 2 from the encomium of Leonidas and his men (certainly Ephoran): τοιγαρου+̑ν οὑ+̑τοι μόνοι τω + ̑ν μνημονευομένων κρατηΘέντες ἐνδοξóτεροι γεγóνασι τω + ̑ν ἄλλων τω + ̑ν τὰς καλλίστας νίκας ἀπενηνεγμένων. Second, it is a great deal easier to believe that Ephorus rather than Timaeus is responsible for the statement that Themistocles, 'having been driven out from the whole of Greece, took refuge with Xerxes, his greatest enemy' (11. 23. 3). We know from Plutarch (Them. 27. 1-2) that Ephorus was one of those historians (the majority) who brought Themistocles to the court of Xerxes. Thucydides and Charon of Lampsacus, on the other hand, said that Xerxes was already dead and Artaxerxes on the throne when Themistocles went up to Susa. Plutarch himself was of the opinion that Thucydides' dating agreed best with the existing chronological records. It is rather unlikely, therefore, that Timaeus, no mean chronographer himself ( FGH566 T10, 11), would have followed

Ephorus, a historian he criticized for inaccuracies ( FGH 566 F110) and the popular tradition rather than Thucydides and Charon. Some other features of Diodorus' narrative of the campaign and battle of Himera which have been held to point to Timaeus, or at least away from Ephorus, are the following. 150 ____________________ 150 It is not possible here to deal with all the arguments which been advanced in this connection. The weakest ones are those based on language; e.g -541.

At 11. 20. 3 it is said that while Hamilcar marched against Himera, his fleet escorted him by sailing along the coast. This is seen as one of several imaginary elements which amount to a 'fantastic' account, which can only be Timaean. 151 But there is nothing 'fantastic' or specifically Timaean in this. There are plenty of instances in ancient history of such close co-operation between land and sea forces. See, from Diodorus for example, from indubitably non-Timaean contexts, 11. 77. 1; 14. 21. 2; 15. 41. 4; 18. 12. 2; 20. 73. 2; 20. 76. 6. 2. Brown, though rightly dismissive of Laqueur's 'scissors and paste' theory, nevertheless proceeds to identify at least two different sources in the account of Himera. 152 He considers a source, which he calls the 'Amilkon' source, as responsible for 20. 1-21. 3, and a second source, the 'Amilkas' source, as responsible for 21. 3-22. 3. At 22. 4 Diodorus, he thinks, returns to the first source. The second source is more 'theatrical' in character than the first. 'An improbable device like that of the intercepted messenger suggests an armchair historian', and therefore it 'may well be Timaeus'. 153 But, apart from the unlikelihood that a compiler like Diodorus would change source in mid-narrative, (a) the narrative forms a unity, the same details occurring in the sections he attributes to different sources; for instance, the ξύλινον τεîχος and the ὑπερκείμενοι λóφοι; (b) the variation between 'Amilkon' and 'Amilkas' is hardly significant. Even if we could presume that Ephorus and Timaeus were consistent in the form of name they gave, one perhaps preferring 'Amilkon' and the other 'Amilkas', the truth is that no pattern can be deduced from Diodorus. At 11. 20. 1, for example, where the Teubner text prints 'Amilkona', MS P has 'Imilkona'. And in Book 13, chapters 80-91, Diodorus sometimes writes 'Imilkon' and sometimes 'Imilkas'. As we shall see these chapters are based on Timaeus. 154 ____________________ Pearson, 1987: 137 n. 52, quotes the phrase τη + ̑ς κατὰ τὰς ναυ + ̑ς φλογòς ἀρΘείσης εἰς ὕφος, from 11. 22. 3, and opines that 'variations of this . . . occur in other Timaean passages of Diodorus, 14. 73. 3, 20. 67. 2'. I would question the Timaean origin of either passage, especially of the first (see below), but as it happens the very same phrase is employed by D at 20. 48. 7, most certainly not a Timaean passage; and cf. 18. 22. 5. 151 Meyer, GdA iv. 1 ( 1944 edn.), 376 f.; Pearson, 1987: 136. 152 1952: 347ff. 153 1952:353. 154 Brown, 1952:353 n. 83, advances the additional argument that 'the Α + ̔μίλκων source prefers not to repeat the proper name. Instead we find οὑ + ̑τος (20. 2) or αὐτóς -553.

It has been said that the claim at 11. 24. 4 that the Carthaginians feared an invasion of Africa by Gelon could only have been made in the light of Agathocles' campaigns, and that it therefore points to Timaeus. 155 This is not an inevitable conclusion. The attempt by Dorieus, foiled by Carthage, to found a colony on the Libyan coast c.514 was a well-known fact; and Herodotus (7. 158. 2) represents Gelon as having entertained a plan to free the region known as the 'Emporia' in the Syrtes from Carthaginian control. 156 4. The scenes of deep mourning at Carthage on receipt of the news of the catastrophe at Himera have been seen as an example of the 'tragic' school of writing history which characterizes thirdcentury historiography -- Timaeus, therefore, rather than Ephorus. We may agree that pathos must have figured heavily in the writings of Timaeus, but Ephorus' Histories cannot have been entirely devoid of the dramatic element (cf. 13. 102; 14. 5. 1-4); and Diodorus of course can be relied on to have enhanced whatever he found in his source, whenever the spirit so moved him. 5. Chapter 25 has been universally attributed to Timaeus, entirely on the basis of 13. 81. 4-84. 6 which is certainly from Timaeus. 157 The attribution should be queried. In 11. 25 Diodorus describes the public works constructed by the large number of prisoners captured by Acragas in 480. In 13. 81ff., on the other hand, we have a comprehensive description of the wealth and grandeur of Acragas in 406. Some of the detail is common to both, especially the dimensions of the artificial lake, but that is hardly surprising if Ephorus and Timaeus employed the same sources, Philistus in particular; and of course the Histories of Ephorus were available to Timaeus. 6. The epithets εὐεργέτης, σωτήρ, and βασιλεύς with which Gelon is said to have been acclaimed by the Syracusans (11. 26. 6) are ____________________ (20. 5). Usually his name is merely assumed as the subject of the verb. This suggests different authorship even if the form Α + ̔μίλκων crept in by scribal error.' But this is to carry source criticism to ludicrous lengths. In ch. 20 only Hamilcar is involved and hence the proper name does not have to be repeated; not so in 21. 4f. where Gelon and Hamilcar alternate. 155 Brown, 1952: 350. 156 Polyb. 1. 82. 6; Dunbabin, 1948: 412. Cf. Talbert, : 196. 157 Cf. e.g. Meister, 1967: 43. Jacoby, FGH566 F26 n., asks us to 'vergleichen' the two passages, 11. 25 and 13. 81. 484. 6, the latter of which he accepts as Timaean, and goes no further. -56 taken by some to betray Timaeus and the Hellenistic age. 158 It is true that such epithets were a commonplace after Alexander, but their use in the Classical period is well attested. Demosthenes, for example, a contemporary of Ephorus, reviles the Thebans and the Thessalians because φίλον, εὐεργέτην, σωτη+̑ρα τὸν Φίλιππον ἡγου+̑ντο (De corona 43); and cf. Plut. Pel. 12. 7: the Theban liberators in 379/8 were hailed as εὐεργέται and σωτη+̑ρες by the populace. As for βασιλεύς, there can be no doubt that Gelon and the other Deinomenids were often addressed by this title (cf. Pindar, Ol. 1. 23; Pyth. 1. 60; 3. 70; Herod. 7. 161. 1). Whether the title belonged to

them officially

159

is another question, one not crucial to our present purpose.

I see no good reason, therefore, for denying that in 11. 20-6 we have the remnants of an Ephoran narrative which has been subjected to the usual 'tender mercies' of Diodorus. On the other hand, there is some reason for believing that Timaeus was his source for the Sicelica in Book 10, at least for the events which led up to the Perso-Carthaginian treaty (the historicity of which Timaeus undoubtedly accepted) and the invasion of Sicily. He tells us at the beginning of Book 11 (11. 1. 1) that he concluded Book 10 with the year previous to that of Xerxes' invasion of Europe, and with the demegoriai delivered at the congress at Corinth about the proposed alliance between Gelon and the Greeks. Fragments only survive from these demegoriai (10. 34), but they do read very much like the irrelevant, rhetorical platitudes that Polybius censures Timaeus for in Book 12. 160 ____________________ 158 They sound 'emphatically and oddly hellenistic' according to Hornblower who nevertheless thinks that D 'himself probably added the titles . . . in the same way he regularly gives his early Egyptian pharaohs the hellenistic royal virtues' ( 1983: 48). But it has never really been doubted that D's sources for Book 1 were Hellenistic; cf. n. 139. Rutter, 1993: 176 ff., argues similarly with Hornblower. He futher suggests that the story of the crowning of Damarete by the Carthaginians related at 11. 26. 3 (and indeed D's portrait of Gelon and Damarete in general) is (1) D's own contribution and (2) unhistorical. Nevertheless, to concede the correctness of the second point (and I do not, at least not on the grounds advanced by Rutter) does not prove the first. Likewise Zahrnt, 1993: 353ff., also dismisses D's account of the campaign of Himera and of the events of the year 480 in the West as largely tendentious and worthless though he blames Philistus and Timaeus for this. 159 As Oost argues, 1976: 224ff. He is less convincing when he proposes the same for the Dionysii: see n. ad 15. 74. 5. 160 The speeches of Nicolaus, Gylippus, and Theodorus are very different in this respect (see below). Surprisingly, 10. 34 escapes the attention of Pearson both in his book, 1987, and in his article, 1986. -57All the same, though Timaeus represented not only that envoys were sent by the Greeks to Gelon (10. 33 -- the version we find in Herodotus), but also that Gelon sent envoys to Corinth offering his help (Polyb. 12. 26b. 1-2), a version of events clearly more favourable to the tyrant, we are not justified in concluding 161 that the latter was an invention to be found in Timaeus, but not in Ephorus. As we have seen, FGH 70 F186 is proof that Ephorus took a highly favourable view of Gelon; and what does τὸν τύραννον ἱκετεύοντας εἰς τὸν τω+̑ν Ε+̔λλήνων σύλλογον ἐλθει+̑ν in the fragment exactly mean? We turn our attention now to Diodorus' account of western history down to the end of the great Athenian expedition in 413. We may at once concede that 12. 12-21 (the lawgivers Charondas and Zaleucus) cannot possibly derive from either Ephorus or Timaeus. 162 It is likely that Diodorus lifted this from some sort of Hellenistic handbook. 163 There is also near-unanimous agreement that 11. 48. 3-8 (the rivalry between Hiero and Polyzelus) and the narratives of the Athenian expeditions against Sicily, with the exception of the speeches of Nicolaus and Gylippus, are based on Ephorus. 164 Otherwise, the preponderance of scholarly opinion is that for this period Diodorus relied mostly on Timaeus. However, none of the arguments advanced convinces. Obviously, I cannot here attempt to refute these arguments in detail. Nor is it in fact necessary. For while the major part of this material does not admittedly allow itself to be assigned to the pen of either Ephorus or Timaeus, there are sections which seem to indicate Ephorus, whereas nothing really suggests Timaeus. 165 ____________________ 161 With Jacoby, FGH566 F94 n. 162 Schwartz, RE v. 685. Both lived much earlier than the 5th century ( FGH 70 F139; Arist. Pol. 2. 1274a-b) while Timaeus actually denied the existence of Zaleucus ( FGH566 F130). And 12. 20, compared with Strabo 6 C 260, further shows that the chapters cannot derive from Ephorus. 163 11. 89, the digression on the deities known as the Palici, which Meister, 1967: 51, and Pearson, 1987: 141, confidently attribute to Timaeus, I would credit to D himself. For one thing, the many stoas and lodgings (11. 89. 8) cannot have existed in the first half of the 3rd century BC. 164 Schwartz, RE v. 686, following Volquardsen, 1868: 90, is in a minority in believing that FGH 566 F93 completes and makes intelligible 11. 48. 3-8, and that the latter, therefore, is Timaean in origin. 165 In attributing 11. 67-8 (the downfall of the tyranny at Syracuse) to Timaeus, Meister, 1967: 47, points to the statement that the Syracusan rebels occupied μέρος τη+ ς πὸλεως . . . τὴν ὀνομαζομένην Τύκην (11. 68. 1). This, he thinks, can only derive from Timaeus because Ephorus erroneously considered Τύκη, or Τύχη, to have -58 In addition to 11. 48. 3-8 and the accounts of the Athenian expeditions, I would see the hand of Ephorus in the following. The contrast drawn between good and bad rulers in 11. 53 and 11. 66-8, the bad rulers causing their own downfall by the harshness of their rule, seems to point to Ephorus. We may compare especially 11. 68. 7 with 15. 1. 3. Chapter 11. 68 leads on to 11. 72-3 and 76 -- the mercenary revolt at Syracuse which involved the rest of Greek Sicily, and the general settlement which followed. The analysis given of the staseis and tarachai is quite consistent with the aims and character of Ephorus' Histories (see pp. 127 f.). The same is even more true of 11. 86-7, where 11. 87. 1 bears strong resemblance to 11. 55. 2; and the praise of the πΛλαιἀ καὶ σπουΔαία ἀγωγή (11. 87. 5) is Ephoran mainstream philosophy. The trial of Ducetius (11. 92. 2-4) also argues strongly for Ephorus. It is quite likely that Diodorus' source provided a pair of speeches at this point, one in defence of Ducetius by a representative of the χαριἑστατοι τω + ̑ν πρεσßυτἑρΩν, 166 and one against him by a demagogue. But speeches or not, the passage corresponds closely to 13. 19. 4 ff. and the speeches of Nicolaus and Gylippus (13. 20 ff) which, in my view, are almost certainly from Ephorus (see below). Finally, there is 12. 26. This panoramic view of most of the ____________________ been an island, as FGH 70 F66 proves. Sanders too, 1987: 72, who is concerned to show that Ephorus did not use Philistus' Περὶ Διονυσίου, thinks that this 'is clearly indicated by his error in detail in referring to the Syracusan suburb of Tyche as an island . . . an error unlikely to have stemmed from . . . Philistus'. But F66, from Stephanus

Byzantius, cannot be right as it stands: s.v. Tὑχη πóλις Σικελίας πλησίον Συρακουσω + ̑ν. Ἔϕορς ἐν τῃ + ̑ ιß (12th) νήσον Tυχίαν ϕησίν. Apart from the improbability that Ephorus would consider Tyche an island, Stephanus generally quotes from one or more authors by way of proof of his claim; and his claim here is that Tyche was a polis. The quotation from Ephorus, therefore, must be garbled or corrupt (as in F76). It is possible of course that Ephorus also mentioned an island which bore the name Tychia. It is far from certain that Palice (11. 88. 6; 90. 1) and Trinacia (12. 29. 2-4) are one and the same city, as Meister, 1967: 51 f., and others believe, and that therefore two different sources, Timaeus and Ephorus respectively, are indicated. 166 Δει + ̑ν γἀρ σκοπεÍ + ̑ν οὐ τί παΘει + ̑ν ἄΞιóς ἐστι Δουκἑτιος, ἀλλἀ τί πρἐπει πρα + ̑Ξαι Συρακοσίοις, was the main argument used. Pearson, 1987: 141 f., who of course thinks that this is Timaeus writing, sees proof of it in Polybius 12. 14. In the process of chiding Timaeus, Polybius comments that the main consideration of sensible men when planning to avenge themselves on their enemies is not so much what their enemies deserve to suffer, but how they themselves ought to act. Polybius, Pearson believes, is reminding Timaeus of what he had written earlier, in his account of the career of Ducetius. But Polybius does not say that, and Pearson's argument is therefore circular. In fact the advice is a rhetorical commonplace; cf. Thuc. 3. 44. -59 Mediterranean world at peace is more likely perhaps to have been conceived by the mind of a universal historian like Ephorus who was concerned with war and peace and the circumstances which gave rise to them (cf. pp. 110 ff.). Diodorus nevertheless must have derived 12. 26. 1 from the handbook on Roman history he employed. To this he joined his none-too-competent abbreviation (12. 26. 2-4) of the relevant Ephoran chapter.Pearson discusses in extenso the speeches in Timaeus, 167 with special reference (naturally) to what Polybius says about them in Book 12 as well as to the speeches of Nicolaus, Gylippus, and Theodorus in Diodorus. He identifies the speeches of Theodorus (14. 65 ff.) and Nicolaus (13. 20 ff.) as Timaean, while that of Gylippus (13. 28 ff.) may be either Ephoran or Timaean. His main argument is that Polybius' criticisms are fully illustrated by the speeches of Nicolaus and Theodorus. Needless to say, there is here a strong element of subjectivity, not least on the part of Polybius himself -- some of his complaints against Timaeus would hold equally well against Thucydides (the fact, for example, that he crammed into his speeches more arguments than could possibly be used on an actual occasion: Polyb. 12. 25i). It may be conceded nevertheless that Timaeus' speeches were probably more trite and banal than those in other leading historians. However, by contrast with 10. 34 (see above), nothing is found in the speeches of Nicolaus and Theodorus to justify Polybius' criticism. We shall return to the speech of Theodorus later. Meanwhile the following further points argue for the Ephoran origin not only of the speech of Nicolaus, but of that of Gylippus too: 168 1. The speeches of Nicolaus and Gylippus clearly form a set; the second answers the first, point for point. So, if Diodorus adapted the first from Ephorus, he must have adapted the second from the same source also. 2. To judge from Plutarch, Nicias29, the role assigned to Gylippus by Timaeus was totally different from that given to him in Diodorus. Ephorus is therefore indicated as Diodorus' source. 169 3. At 13. 30. 3, in the speech of Gylippus, it is claimed that the ____________________ 167 1987: 145 ff.; 1986: 305 ff. 168 We should always bear in mind what Schwartz, RE v. 681, points out, that D transformed the speeches stylistically, as indeed he did, with greater or lesser success, most material culled from his sources. 169 Cf. Barber, 1935: 165. -60-

Athenians had taken a decision to destroy Syracuse and Selinus, selling their populations as slaves, and to impose a phoros on the rest of Sicily (cf. 13. 29. 4; 13. 30. 3). First stated at 13. 2. 6 in the narrative (which is of course Ephoran in origin) as a secret agreement between the generals and the boule, this almost certainly unhistorical element must have been borrowed by Ephorus from a Siceliot source, probably Philistus. 170 4. The occasion for the speech of the old man Nicolaus is admittedly a little theatrical, but that is not surprising in an author like Ephorus; cf. 11. 50. 6 and 13. 102. 1 ff. In fact the speeches of Nicolaus and Gylippus give every impression of being a deliberate imitation by Ephorus of the twin speeches of Cleon and Diodotus in Thucydides. 171 5. The strong Isocratean touches in the speech of Nicolaus were well noted by Schwartz who correctly surmised that they pointed to Ephorus. 172 With respect to the western accounts, therefore, in Books 11, 12, and 13 down to the end of the great Athenian expedition, none of the arguments which have been adduced for the use of Timaeus, whether as the main or even as a secondary source, convinces; whereas, as we have seen, there are many reasons, more or less sound, for detecting Ephorus. This does not of course exclude the possibility that an occasional detail from Timaeus was introduced (as happened later -- see below). But as neither Ephorus nor Timaeus survives, the extent to which this happened is impossible to estimate. We noted earlier (pp. 29 f.) that for the period between about 411 and 387 Diodorus took greater pains over the phrasing of the formulae with which he opens each year's narrative. And we suggested that a possible explanation for that might be sought in a change of source at this point, to wit, in his having taken up again Timaeus as his main source for western history. Whether that is so or not, there can be no doubt that Timaeus was indeed Diodorus' ____________________ 170 Meister, 1967: 62, who rightly assigns to Ephorus the narrative of the great expedition, nevertheless selects a couple of items, including this one, and suggests that they derive from Timaeus. But this is the kind of unwarranted assumption which makes nonsense of source criticism. If an account is admitted to be based on source A, good evidence is required before any part of it is attributed to source B; and Meister advances no such evidence.

171 172

Cf. Pesely, 1985: 320 f. RE v. 681. -61-

source for the Carthaginian invasions of 409 and 406. It is noticeable that the various narratives, though separated as usual by accounts of mainland Greek and eastern affairs, display unity and internal cohesion. For example, 13. 54. 1 clearly carries on from 13. 44. 6, 13. 75. 2 from 13. 63. 6, 13. 80. 1 from 13. 62. 6, 13. 96. 5 from 13. 91. 1, and 13. 108. 2 from 13. 96. 5. On the other hand, the recurrence of the sort of detail (for instance, the Campanian mercenaries mentioned at 13. 44. 1, 13. 62. 5, and 13. 80. 4) which in different circumstances might be explained by the supposition of a single source at work is not applicable here because the same sources (above all Philistus) provided the factual basis for both Ephorus and Timaeus. 173 The proof that Timaeus is in fact the source responsible for this lies in the figures for the Carthaginian forces we find embedded in the narrative of the campaigns. 174 Timaeus gives the size of the invading Carthaginian army in 409 as 100,000 men (13. 54. 5). These were joined by 20,000 Sicels and Sicans (13. 59. 6). And it is precisely this army of 120,000 that we meet operating against Himera (13. 59. 6; 13. 60. 3). The Carthaginian navy on this occasion amounted to 60 triremes and 1,500 merchantmen. Diodorus does not attribute these figures to either Ephorus or Timaeus, but if we concede that the account as a whole is Timaean there is a prima facie case for the naval figures being his too (and see below). In 406 the Carthaginians again had at their disposal 120,000 men, according to Timaeus (13. 80. 5). Their navy this time numbered some 1,000 merchantmen and 90 triremes. Though ____________________ 173 This needs to be qualified, especially in the case of Timaeus-- see text below. Both Ephorus and Timaeus must have used historical writings other than that of Philistus: Ephorus at D 14. 8. 4 ff. did not, for instance, follow Philistus (cf. Plut. Dion 35. 6 f.); and Polycritus of Mende reached D 13. 83. 3 through Timaeus. Other possibilities are Hermeias of Methymna ( FGH558), Athanas of Syracuse ( FGH562), and Timonides of Leucas ( FGH561). Whether Theopompus on Sicily could have been available to Ephorus (as he certainly was to Timaeus) is a moot point. Historical compositions apart, a great deal of material of all sorts must have existed in Athens in the second half of the 4th century, written as well as oral, and largely hostile to the memory of the tyrants. The role of the comic stage and the Academy in both generating and disseminating such material was probably significant ( Stroheker, 1958: 18 ff., esp. 23 f.; cf. Caven, 1990: 222 ff.; Sanders, 1987: 1 ff.). 174 The significance of these numbers was seen by Schwartz, RE v. 685, though not by Sanders, who insists that all that can be deduced from such fragments 'relates precisely to the immediate material to which they allude' (1987: 111). But fragments such as these are a great deal more important when (as here) they can be shown to underpin the narrative. -62 Diodorus does not attribute the latter information to Timaeus, the same comment applies as for the fleet of 409. Once more we see that the army of the account of the campaign of 406 is that of Timaeus (13. 85. 1; 13. 87. 1). By itself the evidence of the numbers suffices for the attribution of the account to Timaeus. But there is more. As generally recognized, 175 the lengthy description of the wealth of Acragas (13. 81. 4-84. 6) unquestionably derives from Timaeus; not only is Timaeus quoted twice in the course of it, but it partly tallies with another Timaean fragment from Aelian ( FGH566 F26). Timaeus is also referred to at 13. 85. 3, in the narrative. In the light of all this, it would be illogical to credit Diodorus with all of the pathos in 13. 89. 176 Two questions demand answers. The first concerns the origin of the discrepant Ephoran numbers for the Carthaginian army (at 13. 54. 5, 13. 60. 5, and 13. 80. 5) which accompany those from Timaeus. Did Diodorus find these in Timaeus, 177 or did he consult his copy of Ephorus? The second alternative is much to be preferred. For one thing, if the presence of the Ephoran numbers is really the result of Timaeus' criticism of them, why is there not a single reference to Philistus in this connection? As a matter of fact Philistus, unlike Ephorus and Timaeus, is never cited for historical detail of any kind by Diodorus, and this circumstance is a further argument that Diodorus did not use him directly. 178 It would thus appear that when Ephorus and Timaeus are cited together, whether in Book 13 or in Book 14 (see below), that is because they were both consulted on the point in question. A good ____________________ 175 cf. e.g. Jacoby, FGH566 F 26 n. 176 Meister, 1967: 76 ff., and Pearson, 1987: 152 ff., employ a great many more arguments in favour of Timaeus, but few of these are plausible. 177 As pro-Timaeus 'fundamentalists' like Pearson, 1987: 152 and n. 93, propose. 178 Philistus on the origin of the Sicans at 5. 6. 1 is clearly derived from Timaeus. Contra Sanders (see n. 145). The fact that Philistus is mentioned five times in the Bibliotheke as a historian (perhaps six: 23. 17. 1) is no evidence of course that D was acquainted at first hand with his work. Three of these references, as it happens, are chronographic notices and it was established long ago, what is in any case evident, that mere mention in the chronographer does not mean that the historian mentioned was used by D ( Volquardsen, 1868). It is surprising, therefore, to find Sanders claiming just that: 1987: 110, 117, 141. And on p. 121 he suggests that the fact that D at 16. 71. 3 'refers to a specific part of the Philippica (of Theopompus), the Sicilian excursus, certainly suggests Diodorus' direct acquaintance with the work'. Chamoux, pp. xxiii ff., seems to be similarly in error (while his comments on p. xxii on German scholars of the 19th century in general and on Schwartz in particular are inaccurate and unfair). That D did not use Philistus directly, but through Ephorus and Timaeus, see p. 70 and nn. 165, 174, 179, 194, 198. -63illustration of how Diodorus worked is provided by 13. 60. 5 and 13. 87. 1. In the first instance Ephorus was consulted and he is therefore cited alongside Timaeus, the latter's figure as usual being a great deal more restrained than that of

the former. In the second instance Diodorus it seems did not bother to check up and cite Ephorus' figure and only the low, Timaean figure is given, unassigned this time. The second question concerns the absence of discrepant Carthaginian naval figures, whether by Ephorus (in Bk. 13) or Timaeus (in Bk. 14). It should not be assumed from this that the two historians gave the same numbers. The truth must be that Diodorus, erratic as always, simply did not take the trouble to note the divergence. Ephorus' numbers, by the way, are exactly in line with those he gives for Persian armies and are therefore his own. 179 Timaeus' numbers on the other hand were probably taken unchanged from Philistus. (Xenophon's interpolator seems to have got his Sicilian facts from either Philistus or Timaeus.) If then Timaeus was Diodorus' main source for the western events of these years, did he continue so using him, or did he at some point forsake him for Ephorus? An important passage to note here is 13. 90. 5-7. Amongst the objects plundered from Acragas in 406 Diodorus includes the famous bull of Phalaris (13. 90. 4). This he does in direct contrast to Timaeus who seems to have denied that such a thing had ever existed at Acragas. In a manner strongly reminiscent of Polybius (12. 25) Diodorus goes on to censure Timaeus who, while making a great show of accuracy, was proved, in the case of the bull, to be merely improvising. Timaeus' bitter criticism of his predecessors is thus rendered even more inexcusable. Authors, Diodorus opines, ought to be forgiven for errors committed in good faith. If, on the other hand, the misrepresentations are deliberate then they deserve to be censured; ὅταν κολακεὐοντες τινἀς, he concludes, ἢ δι + ̔ ἔχϕραν πικρóτερον προσßἀλλοντες ἀποσϕἀλλωνται τη + ̑ς ἀληΘείας. The conclusion is obviously inappropriate to the case in point, that is, Timaeus' ____________________ 179 See pp. 130 ff. This is overlooked by Caven, 1990: 31 f., who therefore surmises that Ephorus' figures derive from Philistus. On the other hand, Timaeus' numbers, he thinks, being 'more modest . . . probably represent "corrections" of Ephorus rather than figures taken from a source other than Philistus'. Conversely, and rightly, Sanders, 1987: 72, infers that Timaeus' figures are those of Philistus. But he is wrong to see the differences between Ephorus' and Timaeus' figures as further evidence that Ephorus did not use Philistus (see n. 194). -64-

views on the bull of Phalaris. So how did it come to be here and what does it signify? At 21. 17. 1-3 Diodorus once again digresses for the purpose of rebuking Timaeus, this time for his grossly unfair treatment of Agathocles. Timaeus allowed personal enmity (διἀ τὴν πρòς αὐτòν ἔχϕραν) so to distort his vision as a historian that his account of the deeds of Agathocles was quite unacceptable. This does not necessarily mean that Diodorus shunned Timaeus altogether as a source for the life and times of Agathocles. 180 On the other hand, 21. 17. 1-3 must be allowed a great deal more weight than Pearson allows it, who merely takes it to indicate that for his narrative of Agathocles Diodorus employed Timaeus, but took the trouble to moderate the anti-Agathocles bias. On the whole I tend to agree with Schwartz that Duris was Diodorus' main authority for Agathocles though with the occasional reference to Timaeus. 181 Can 13. 90. 5-7 be taken to imply something similar? There can be little doubt that both passages, 13. 90. 5-7 and 21. 17. 1-3, were inspired by Diodorus' reading of Polybius Book 12. 182 We need not suppose that Timaeus mentioned the bull of Phalaris in connection with the sack of Acragas. But whether he did or not it seems evident that Diodorus turned to Polybius Book 12 again at this point and reread a number of chapters, including those which discussed Timaeus' treatment of the bull of Phalaris and Agathocles. What else can he have found in these chapters? Most of Book 12 of Polybius is unfortunately lost to us. We cannot be certain of the correct sequence of all the surviving fragments which, moreover, may not be fully representative of the original contents of the book. From 12. 24. 3 f. we learn that Timaeus had branded Dionysius as an effeminate, 183 and Polybius may well have had more to say about Timaeus' one-sided account of the tyranny of Dionysius. That Timaeus, in contrast to Ephorus, was bitterly hostile to the tyrant is sometimes questioned. 184 It seems to me, however, that there are sufficient ____________________ 180 Pearson, 1987: 227 ff. 181 RE v. 687 f. P. Oxy. 2399 (possibly Duris -- accepted as such in Lustrum, 21 ( 1978)) lends support to this view -- see Turner comments, P. Oxy. xxiv. 99 ff. 182 On D, Polybius, and the bull of Phalaris see Walbank, HCP ii. 380 ff.; Schepens, 1978: 117 ff. 183 On Dionysius' interest in textiles cf. 14. 109. 1; Jacobsthal, 1938: 205 f. 184 e. g. by Caven, 1990: 3, 231 f. (arguing against Stroheker, 1958: 16 ff.). Lewis adopts an agnostic position: 'there is no real reason for supposing that one was more -65indications for supposing just that. Timaeus was not called Epitimaios for no good reason (whereas nothing remotely similar was ever said of Ephorus) and Momigliano makes the additional point that many of the victims of Timaeus' vitriolic attacks were either pro-Macedonians or supporters of despotic governments. In his opinion, Timaeus' hostility towards Aristotle stemmed from a political motive; in other words, it was due to the philosopher's collaboration with Macedonia. 185 But why should that have annoyed Timaeus? The answer of course lies in Timaeus' Sicilian background and experiences. Timaeus' city of Tauromenium was destroyed by Agathocles and Timaeus was forced to flee to Athens; hence his complete misrepresentation of the reign of Agathocles against which Polybius (12. 15) and Diodorus (21. 17. 1-3) protest. The gross vilification of Demochares, a nephew of Demosthenes, which so shocked Polybius (12. 13 f.), may well have been provoked, in the view of Momigliano, 186 by some opinion of Agathocles in Demochares' historical work which Timaeus construed as favourable.

Of course, the destruction of Tauromenium. by Agathocles was only the tail-end of a story which began nearly a century earlier. In 403 Dionysius I had utterly destroyed the city of Naxos, whence Timaeus' family apparently hailed, selling its entire population as slaves (14. 15. 2). If Andromachus (Timaeus' father) did not suffer this fate, because he had not yet been born perhaps, his father (Timaeus' grandfather) must have done. When in 358/7 Andromachus gathered together the remnants of the Naxians and founded (or refounded -- see 14. 59. 1-2) Tauromenium (16. 7. 1), he almost certainly did so against the wishes of Dionysius 11. 187 Timaeus was not the man either to forgive such sins against his family or, conversely, to overlook favours done to them. What Marcellinus says cannot be the entire truth, but he may well have a point when he suggests that Timaeus praised Timoleon above measure because he allowed Andromachus to continue as tyrant at Tauromenium. The intense hatred Timaeus must have felt against ____________________ hostile to Dionysius than the other': CAH2 vi. 123. On the contrary, Sanders, 1987: 79 ff., seems perfectly correct in arguing that Timaeus was violently hostile to Dionysius. 185 Momigliano, Essays, 46. 186 Ibid., 47. 187 Timaeus' hostility to Dionysius is also indicated by the fragments: FGH566 FF29, 105. -66the Dionysii, and the contrast in this respect between him and Ephorus, are made quite clear by the comments of Plutarch in Dion 35. 6-36. 2, which passage recounts the death of Philistus and the way this event was treated by Ephorus, Timonides, and Timaeus. Taking as a pretext Philistus' zeal for, and loyalty to, the rule of the Dionysii, Timaeus, Plutarch says, filled his work with insults and defamatory 'comments against him. The violence of Timaeus' language against a leading supporter and the 'official' historian of the tyrants (incomprehensible, if Timaeus did not hate them) evokes Plutarch's condemnation: it is forgivable in those who lived and suffered under the tyranny to exhibit such bitterness, but later historians ought to moderate their language. Timaeus had not of course experienced the tyranny of the Dionysii, but, as we have seen, his parents and grandparents had. If Timaeus went too far, so did Ephorus, according to Plutarch ( Dion 36. 3-4), though in the opposite direction: he dared praise Philistus. Plutarch does not explain as well as he might, but Ephorus' praise seems to have been not just for Philistus' abilities as a historian, but also for the loyalty, disinterestedness, and perhaps even patriotic motives (as Philistus himself claimed?) with which he served both Dionysii. It is quite possible that Ephorus, as was his habit (pp. 6 ff.), accorded Philistus a set epainos, the gist of which survives in Diodorus' account of his death (16. 16. 3 f.) where he is described as the πιστóτατος and πρακτικῴτατος of all the philoi of the two tyrants. We are in no way justified in deducing from all this that Ephorus' account was favourable to the elder tyrant, but it must surely have been fairer than that of Timaeus. Timaeus would not have written as he did about Philistus had he not been actuated by hatred against the Dionysii. Nor would Ephorus have praised Philistus' abilities and character if his viewpoint was not more detached than that of Timaeus. 188 So what kind of history of the Dionysii did Timaeus write? Commenting on Timaeus' account of the reign of Agathocles Polybius accuses him of deliberately leaving out all the good things and including only the bad ones (12. 15. 10 f.; cf. D 21. 17. 2), obviously with the aim of showing the dynast in the worst ____________________ 188 Having made an Italiot of Ephorus, presumably because he supposes Cyme, his home-city, to be the city in Italy with the same name, Caven proceeds to surmise that Ephorus 'would have had his own reasons . . . for disliking Dionysius' (1990: 3). This mistaken conjecture intrudes several times in Caven's book: pp. 65, 142, 146, 187. -67 possible light. In all probability the elder Dionysius did not fare any better in Timaeus' hands. And this is where the qualification as to Timaeus' use of Philistus I mentioned earlier comes in. If Timaeus' hatred of Dionysius is granted, we must inevitably conclude that he is unlikely to have borrowed from Philistus material to the tyrant's credit. 189 When such material is therefore found in Diodorus it must have reached him through Ephorus rather than Timaeus. Ephorus himself was concerned with virtuous conduct, not least in public life (see pp. 112 f.). The various references to Dionysius in Isocrates may serve as an indication of Ephorus' own attitude. For Isocrates, Dionysius was an upstart who, having stopped at nothing in pursuit of power, succeeded in making himself the most powerful Greek ruler of his time ( Philip65; Nicocles23; Archid. 44 f.). Still, democrat of the old school though he was, Isocrates might not have disapproved, if Dionysius had employed his power, however unjustly gained and maintained, to serve the interests of Hellenism. This he could not be said to have done in any consistent manner. On the contrary, he had collaborated with Sparta to the detriment of those interests. 190 Not inconsistent with this view of Dionysius is that presented by Diodorus at 14. 2. 2: though the most fortunate of rulers Dionysius, like his close allies the Spartans whose rule was equally harsh and unjust, was hated by his subjects against whom he had constantly to be on his guard. Section 14. 2. 2 comes from the preface to Book 14 which appears to be an adaptation of an Ephoran preface (see p. 101 ). This less hostile, more objective estimate of Dionysius as a man and as a tyrant is found in other writers too, 191 and it may well derive from Ephorus. 192 Whether or not there was in Polybius Book 12 criticism of Timaeus' unfair treatment of Dionysius, the last sentence in Diodorus 13. 90. 7, about historians missing the truth when they choose to flatter or denigrate because of hostility, is a signal that ____________________ 189 Sanders, 1987: e.g. 81, 131; Cf. Caven, 1990: 2. 190 Paneg. 126; Peace99. Isocrates' short-lived interest in Dionysius as a panhellenic leader (Letter 1) does not

alter this opinion. E.g. the portraits of the tyrant found in Nepos ( Dion1 f.; De reg. 2. 2) and Cicero ( Tusc. disp. 5. 57 ff.), though much of the account in Cicero of the inordinate lengths to which Dionysius went to guard himself against assassination is likely to derive from elsewhere. Nepos, De reg. 2. 2, in particular is very close indeed to the impression of Dionysius one derives from D's account. 192 Cf. FGH556 F40n. 191

-68 Diodorus was aware of Timaeus' propensity to distort events when his personal feelings were involved. 193 He was about to embark on his account of Dionysius and, had he been a more competent writer, he might have tried combining Ephorus and Timaeus, using the narrative of the former to temper the more extreme features of that of the latter. Certainly, what we read in Diodorus, though by no means sympathetic to Dionysius, cannot, in my view, be a simple abridgement of Timaeus. We may be reasonably certain that Timaeus' account was unrelentingly hostile. In fact, Diodorus being Diodorus, what he did do, I think, was to lay aside Timaeus and take up Ephorus again as his main authority. In any case, that Book 14, which holds the major part of the story of Dionysius, is based on Ephorus, is indicated by quite a body of evidence. 1. The estimate of Dionysius, of his character and actions, is not, as we have just seen, what we should expect from Timaeus. It is the portrait of a dynamic and resourceful, albeit unscrupulous tyrant, objectively drawn, warts and all. His one and only concern was how to safeguard and expand his rule. To achieve this aim he hesitated at nothing. This was the opinion of Isocrates (see above), and Ephorus evidently shared it. All of Dionysius' virtues and vices stemmed from this one desire. But of vices, the lust for power and the cruelty and suspicion which this could engender apart, virtually the only one we come across in the pages of Diodorus is an unseemly (for a ruler) predilection for the composition of bad verse. There is no hint of effeminacy (with which Timaeus had charged the tyrant: Polyb. 12. 24. 3 f.), cowardice, or sexual promiscuity; quite the contrary. The energy, bravery, astuteness, and even, on occasion, a certain nobility of character of Dionysius are shown by many passages some of which almost amount to an aristeia of the tyrant. 194 In several instances the sources which lie ____________________ 193 The claim at 21. 17. 1 that Timaeus showed very high regard for historical truth in the rest of his work, though not when he came to relate the deeds of Agathocles, is an incautious overstatement. It is belied by 13. 90. 57, Plut. Dion 36. 1-2, and some at least of Polybius' criticism in Book 12. 194 We may compare, for instance, 14. 18. 4; 14. 18. 6f.; 14. 42. 1 f.; 14. 44.3; 14. 45.2; 14. 73. 2; 14.88. 2 ff.; 14. 100. 3 ff. In 14. 73.2 it is claimed that in 396 he was just as eager to destroy the enemy fleet as the Syracusans. It is in such passages that the hand of Philistus shows through most clearly. The surviving fragments are not very helpful, but cf. FGH 556 F28 with 14. 41-3. The general grounds on which Sanders, 1987: 71 f., denies use of Philistus to Ephorus are tenuous in the extreme (why should Ephorus have eschewed 'the -69behind the narrative chiefly Philistus through Ephorus) plainly seek to exonerate him from any responsibility for military disasters. 195 The general impression one derives is that Dionysius was not by nature a cruel man -unless circumstances demanded it, in which case he was second to none in this respect. With one notable exception, when he unwisely let his desire for revenge Let the better of him (14. 112), he appears as a pragmatist and a man well in control of himself, able to show either mercy or severity as his policy demanded (14. 9. 5 ff.; 14. 14. 8; 14. 41. 1; 14. 45. 1). Such an account of Dionysius, let it be said again, cannot be described as friendly and in places it is positively hostile. Philistus' account, on the other hand, must have been consistently well disposed to the tyrant, if not indeed eulogistic; see 16. 16. 3-4; Plut. Dion 36 (cf. Paus. 1. 13. 9); Pel. 34. 1; Dion. Hal. Ad Pomp.5 (ἦϑ+00F3ς τε κολοκικὀν καὶ Φιλοτύ ραννον ἐμφαί+03BDε ι καὶ μικρολóγον). Consideration of these texts excludes the possibility of Philistus' being Diodorus' direct source. 196 If Diodorus' record of Dionysius is both favourable and unfavourable the obvious source indicated surely is Ephorus, not Philistus and certainly not Timaeus. Naturally enough, the one part of Book 14 where there is unremitting and not entirely fair criticism of Dionysius and his policies is the speech of Theodorus (14. 65 ff.). There, Dionysius is accused by one of his political enemies of cruelty, treason, incompetence, impiety, and cowardice. This is the kind of hostile tradition that Timaeus would have inherited from his family and further elaborated. All the same, the speech is unlikely to have ____________________ æuvre of a Thucydidean historian like Philistus' on Dionysius (p. 71) when he did not object to using Thucydides in his entirety or the Oxyrhynchus historian?), while the positive evidence he adduces to prove it (p. 72) is unsustainable (see p. 63 and nn. 165, 174, 179, 198). 195 For the costly failure to capture Segesta in 396, Leptines, if anyone, was to blame (14.53.5; 54. 2 f.). For the serious defeat in the naval battle off Catane in the same year, Leptines was definitely to blame, we are told, having failed to follow his brother's instructions (14. 59. 7; 60. 2). The decision not to meet the Carthaginians in the field after the naval defeat, but to return to Syracuse, was taken by the tyrant following the advice of his philoi and it was the right decision it seems, in the circumstances, the implication being that the Sicilian Greeks were wrong to desert the tyrant because of it (14. 61. 1-3). The same Sicilian Greeks had been anxious to fight earlier near Segesta and again the text goes out of its way to point out that Dionysius was wiser in shunning a battle (14. 55. 5). Finally, in the campaign of 392, the Syracusans appear in a bad light while Dionysius is shown to be correct in his assessment of the strategy required (14. 96. 1-2). 196 As maintained by Sanders, 1987: esp. 11 off. -70-

originated with Timaeus, as many suppose. 197 It is very much of a piece with the narrative, 198 and that is based on Ephorus, not Timaeus. The difference between Ephorus and Timaeus is not that the former totally ignored the hostile tradition. Clearly he did not. But, being dispassionate about the subject, he was able to treat the tyrant more equitably. 199 2. The oikonomia of Book 16 of Ephorus (as well as of the following books perhaps) may have appealed to Diodorus. Con____________________ 197 e.g. Laqueur, RE via. 1142; Meister, 1967: 93; Pearson, 1987: 179 ('the speech . . . must be taken almost word for word from Timaeus'!); id., 1986: 350 ff. One of Pearson's arguments for Timaeus (1987: 180) is that the speech and its circumstances are entirely fictitious. Thus also Caven, 1990: 115, who nevertheless believes that the speech 'is much more likely to have come from a collection of rhetorical exercises . . . unless . . . it is a product of Diodorus' own pen'. According to Pearson, the delivery of such a speech would never have been allowed. But apart from the obvious retort that this does not necessarily exclude Ephorus, the circumstances of Syracuse and its tyrant in 396 were quite exceptional. The moment was one of deep crisis and the citizenry, as Theodorus points out, were armed and in the company of potentially sympathetic allies (14. 67. 2 f.). What better time for agitation against Dionysius? By good fortune we have the fragmentary description of a very similar episode at Syracuse in 310 (P. Oxy. 2399), possibly from the pen of Duris (see n. 181). It is obvious that his information is derived from partisans of Agathocles. All the same, it is conceded that Diognetus, the agitator, spoke at length and with some effect before being removed. 198 198 That the speech is an integral part of the surrounding narrative is obvious even from a cursory reading; cf. e.g. 14. 45. 5 with 64. 4; 67. 3; 70. 1; and 14. 41. 1 with 68. 4. Clearly there are untruths in the speech, but these are understandable either as the rhetorical exaggerations (e.g. 66. 1, 4) or the deliberate distortions of a political enemy. Still, the inconsistency between, say, 61. 4 and 68. 6 f. may well be due to the fact that what D gives at either, or indeed both, places may not represent all of the detail (and reasoning) available in his source. It is also worth noticing the connection between chs. 14. 10 and 14. 70 and the speech itself, all three pointing to Ephorus as the underlying source. Unconvincingly, Sanders (1987: 134 ff.) assigns the speech, along with the rest of the Dionysian narratives in D, directly to Philistus. The historian, he suggests, had thereby sought to expose the dishonesty and incompetence of the leader of Syracusan resistance to the tyrant and the unrealistic nature of the policies which he advocated (1987: 139), and this may be the reason why D, contrary to his custom, decided to include the speech (1987: 140). But the speech gives no such impression and Sanders' claims could only be upheld if we ignored 14. 70 (and 14. 10) wherein the true feelings of D's source about the tyranny of Dionysius are given. 199 Pearson, 1987: 183 (in line with Stroheker, 1958: 79), considers 14. 75. 1-4 (the agreement between Himilco and Dionysius, and the motive ascribed to the latter) a 'highly improbable story', and its source can therefore be identified 'fairly confidently' as Timaeus. But whatever the degree of historical truth ( Caven, 1990: 118, accepts the compact as authentic), there is no reason why such a story could not have been narrated by Ephorus. Pearson further queries whether 300 sacks of money could have been carried into the acropolis unnoticed ( 1987: 183 n. 86). By land that would of course have been impossible, though not by sea. -71 trary to what is almost a communis opinio, Ephorus did not as a matter of principle treat the western narratives in isolation, in separate books (pp. 89 ff.). Moreover, one of his aims as a universal historian was to highlight the connections between events and developments in different parts of the Mediterranean. The preface of Book 14 of Diodorus was fashioned out of the preface to Book 16 of Ephorus. 200 It stresses the moralizing viewpoint that all men, but especially those in authority, ought not to commit evil deeds if they want to escape the censure of history. ἐμθανη + ̑ δἐ τὰ κατά παραδείγματα, says Diodorus (as Ephorus must have done though probably more subtly), λαμßάνειν ἔΞεστι τοι + ̑ς α + ̔ναγνου + ̑σι τὰ κατὰ μἐρος τη + ̑σδε τη + ̑ς ßίßλου (14. 1. 3). The three examples chosen are the rule of the Thirty at Athens, the Spartan experiment with empire, and the successful, but unjust and cruel, tyranny of Dionysius. Diodorus, therefore, must have found Ephorus' account ideal almost in every respect, not only in what was said, but also in the arrangement of the material -- barring of course the difficulty arising out of the fact that the work of Ephorus, unlike that of Polybius, was not a year-by-year account. A good example of the means by which Ephorus linked appropriate narratives, Hellenica with Sicelica in this instance, is provided by 14. 10. The chapter is of undoubted Ephoran origin and it serves as proof that one source is responsible for both eastern and western accounts in Book 14. It bears close relationship with Isocrates, On the Peace 99 and Paneg. 126. The collaboration between Sparta, who was making a bid for empire at the time, and Dionysius is emphasized and condemned by both. Ephorus in fact provided information highly discreditable to Sparta. In the immediate aftermath of the revolt of 404 which all but toppled Dionysius, and while the tyrant must have felt very insecure, a Spartan officer called Aristos was dispatched to Syracuse, τῳ + ̑ μἐν λóγῳ προσποιοὑμειν καταλάειν τὴν δυναστείαν, τῃ + ̑ δ' ἀληθεία σπεὑδοντες αὐΞη + ̑σαι τὴν τυραννίδα. Acting as a sort of agent provocateur he betrayed those who had trusted in him (τοὐς δἐ λπεὑδοντες προδοὑς) and delivered them up to Dionysius. In this way the rule of the tyrant was strengthened though Sparta was disgraced thereby (14. 10. 2 f.). Very much the same thing is said to have happened in 396 when the then Spartan commander sent ____________________ 200 This incidentally shows that the Ephoran book began with the year 404. -72-

to aid Dionysius and Syracuse against Carthage betrayed the hopes of the Syracusans (τη + ̑ς ἐλευθερίας προδóτης), and earned curses for his country (14. 70. 1-3). 201 Feeling more secure as a result of Aristos' treachery, Dionysius proceeded to disarm the Syracusans. He then built a second wall round the acropolis (14. 10. 4). The building of the first wall is described at 14. 7. 2. Ephorus is therefore indicated as the source for all of these narratives. 3. The numbers for the Carthaginian forces in the account of the campaign of 397/6 seem to be those of Ephorus.

To begin with we may observe that the statement at 14. 54. 4, οἱ δἐ Kαρχηδóνιοι πυΘóμενοι τó μἐγεΘος τη + ̑ς Διονυσίου δυνἀμεως, ἔκριναν πολὐ ται + ̑ς πυθóμεναι + ̑ς αἐτòν ὑπερΘἑσΘαι, fits Ephorus' figures better than those of Timaeus, considering the size of Dionysius' force (14. 47. 7). As far as the numbers given for the land forces are concerned, there is no real problem: 14. 76. :2, in the narrative, is possible only if Ephorus' figures are followed. 202 And we may compare 14. 62. 3, also in the narrative, where it is said that the army consisted of 300,000 men, ώς μἑν τινες ἀνἑγραφαν, the latter remark surely Diodorus' who may have recollected that Timaeus gave a different figure. On the other hand, there is something very seriously wrong with the numbers for (as well as the description of) the Carthaginian navy, as these stand in the text of Diodorus: 203 14. 54. 5: the Carthaginian armada consisted of 400 long ships and more than 6oo other ships which transported the provisions, the engines of war, and the rest of the supplies. 14. 56. 1: Himilco was anxious to capture Messene because of its good harbour which could receive all of his ships (ναυ + ̑ς), οὔσας πλείω τω + ̑ν ἑΞακοσίων. 14. 59. 7: in the sea battle off Catane the Greeks were greatly outnumbered. εἰ + ̑χον γἀρ (i. e. the Carthaginians) σὐν ται + ̑ς ὁλκἀσι καὶ ται + ̑ς ἂλλαις ται + ̑ς ἐπικώποις, οὔσαις χαλκεμßóλΟις, ναν + ̑ς οὐκ ἐπἀττους πεντακοσίων. ____________________ 201 201 The difference in the form of the name, Aristos at 14. 10. 2, but Aretes at 14. 70. 3, is of no significance. 202 Cf. Schwartz, RE v. 685. 203 As is not generally appreciated. Caven, for example, 1990: 107ff., does not seem to be aware of the problem. -73-

14. 62. 2: the Carthaginian fleet which entered the Great Harbour of Syracuse consisted of 250 (the reading of text is corrupt, 204 almost certainly involving a lacuna between τρισχιλίων and θἑρουσαι. It is extremely suspicious that only 600 transports are said to have accompanied 400 warships and such a vast land force. We may compare 11. 20. 2: 200 warships, but more than 3,000 merchantmen; 13. 54. 1: 60 warships, but 1, 500 merchantmen; and 13. 80. 5-7: 90 warships, but more than 1,000 merchant vessels. While it is impossible to iron out all difficulties, it seems reasonably safe to accept the following: i. The figures τρισχίλιαι and δισχίιαι at 14. 62. 2 are probably sound though it appears that they have been transposed. The accompanying statement that the great harbour of Syracuse was completely filled by the enemy fleet certainly argues for thousands of ships. ii. The 400 long ships at 14. 54. 5 (at the beginning of the campaign) are reduced to 250 at 14. 62. 2, perhaps as a result of losses in the battle off Catane and the stationing of a strong squadron at Messene (cf. 14. 56. 1). iii. There is a group of ships which stands apart from the long ships and the merchantmen. It numbers 'more than 500' at 14. 62. 2. It is perfectly obvious that something was said about it at this point which has dropped out of the text: '(lacuna) θἑρουσαι Δἐ πλείους τω + ̑ν πεντακοσίων.' Or perhaps: '(lacuna) θερ (corrupt), οἐ + ̑σαι δἐ πλείονς τω + ̑ν πεντακοσίων.' What can we say about this group? Two things. First, when Dionysius and the Syracusans attacked the Carthaginian naval station near Dascon, they found, and set fire to, 40 penteconters in addition to merchantmen and some triremes (14. 73. 2). Penteconters were warships equipped with 50 oars and, of course, a bronze ram. Like triremes, they could also be used as transports. Second, at the battle off Catane the Carthaginians, as we have seen, had no less than 500 ships, 'including the merchantmen and the other oared ships which had bronze rams'. So says Diodorus, but without making sense. It ____________________ 204 This is universally acknowledged; see the app. crit. of the Teubner edn. -74

would appear that he badly telescoped his source. At the beginning of the campaign the Carthaginians had 400 long ships (triremes certainly), and the figure of 500 here must have corresponded to a different class of long ships in Diodorus' source, penteconters and, perhaps, triaconters, which had been serving as transports. We may now proceed to surmise that at 14. 54. 5 Diodorus gives the number of front-line long ships (400), as well as the number of the other oared ships which nevertheless served as transports (more than 600), but has neglected to mention the merchantmen. These, on the basis of 14. 62. 2, must have numbered about 2,000. And the total force, therefore, amounted to some 3,000 craft. At 14. 56. 1 he mentions only the second group. The discrepancy between the 600 at 14. 54. 5 and 14. 56. 1, and the 500 at 14. 59. 7 and 14. 62. 2 may be due to Himilco having left a part of his fleet at Messene. Doubtless Philistus' detailed account made all this clear. It is not very likely that Ephorus confused this information, but it is quite likely that, once again (see above), he exaggerated the figures. The incomplete and corrupt details we meet in Diodorus must be his own handiwork as well as that of the copyists. 205 Thus the Carthaginian naval numbers in Book 14 are those of Ephorus, while those at 13. 54. 1 and 13. 80. 5-7, being more modest, betray Timaeus -- to whom we have assigned the corresponding narrative. 4. One feature that is frequently assigned to Timaeus with some confidence is the description of the catastrophe which struck the Carthaginian army besieging Syracuse in 396, and in particular the claim that it was divine retribution for its sacrilegious acts (14. 63. 1-2; 14. 76-7). 206 But the reasons advanced lack cogency. For example, it must be the case that the account of Ephorus (much more than that of Timaeus) was based largely on that of Philistus; and not only for hard facts, one must suppose, but for much of the interpretation of less 'sensitive' issues which did not affect his particular view of Dionysius. At many points (see above) Diodorus' narrative bears the strong imprint of a contemporary witness who participated in the events. The vivid account

of the siege of ____________________ 205 205 I do not, therefore, agree with Schwartz, RE v. 685, that the numbers at 14. 56. 1 and 14. 59. 7 are incompatible with those at 14. 54. 5, and that this circumstance points away from Ephorus. 206 Cf. Meister, 1967: 103. -75 Syracuse in 396, including the detailed description of the plague, can have originated in no other source but Philistus. And it would be perverse to deny him the attribution of the pestilence to a divine agency, a punishment meted out to the Carthaginians of Thucydides. That Philistus believed in the intervention of the gods in human affairs, because of their sacrilegious conduct. and that he exemplified this in his Sicelica, is proved by the fragments (cf. FGH 556 FF57, 58). This was the sort of account Ephorus found in Philistus and which he shaped to suit his purposes. Like Philistus, Ephorus did not exclude the supernatural from historical causation (see pp. 12 f.). Chapter 14. 76 reads very much like a Diodoran rendering of an Ephoran ἐπιμετρὼν λóγος (Polyb. 12. 28. 10). We have seen that the figure of 150,000 Carthaginian dead from the plague (14. 76. 2) can only derive from Ephorus. Also, the explanation offered for the revolt of the Libyans against Carthage (14. 77. 1) and the figure given for the rebel army (14. 77. 3) are more likely to have originated with Ephorus than Timaeus. 5. Additional sundry grounds can be adduced in favour of Ephorus. For example, what is said about the origin of the name of Tauromenium at 14. 59. 2 (Cf. 14. 87. 4) cannot derive from' Timaeus. 207 And the opinion expressed at 14. 100. 1 seems to be that of Ephorus: in his campaign of 39o Dionysius decided to begin by capturing Rhegion because he considered the city to be the προπολεμητήριον of Italy. Relying on Diodorus' choice of vocabulary can be a dangerous occupation, but perhaps not in this instance. At 14. 98. 3, in an undoubted Ephoran context, we are told that the Great King decided to declare war on Evagoras because of the strategic position of Cyprus and her naval strength, whereby it was in a position to προπολεμει + ̑ν τη + ̑ς Ασίας. Such a use of προπολεμει + ̑ν and προπολεμητήριον is not found anywhere else in Diodorus. 208 6. Might it still not be possible, however, to maintain (with e.g. ____________________ 207 The chronographic entry at 16. 7. 1 probably does. Pearson, 1987: 177, nevertheless contrives to see Timaeus' "unmistakeable signature . . . in the etymology of Tauromenium" at 14. 5 9.2. 208 The meaning of προπολεμοὑσης at 20. 84. 1 is quite different. -76-

Laqueur) that though there is demonstrably a lot of Ephorus in Book 14 there may also be a fair amount of Timaeus? In my view this is excluded by the unity and consistency in tone and detail of the Dionysian account. Sordi 209 nevertheless claims that she can distinguish in Diodorus' account of the relations between Dionysius and the Itallots 'due racconti distinti e paralleli', 210 one deriving from Ephorus and the other from Timaeus. The result of a simultaneous and clumsy use of the two historians (one of them, Timaeus, a noted chronologist and the other not) was a whole series of duplicated or not entirely consistent narratives which seriously disfigure Diodorus' account of the reign of Dionysius and its chronology. From the many arguments she advances, she selects for special emphasis the following: (i) the 'undoubted existence' of the 'doublets' of the embassy to the Olympic Games and the failed epigamia of the tyrant with Rhegion; (ii) the 'clear contradictions' between 14. go and 14. 95 about the presence in Sicily of the Carthaginian commander Mago; and (iii) above all, the significant differences she perceives in tone and tendency in the various chapters on Dionysius which must point to different points of view and therefore to different sources. 211 This is all wholly unconvincing: i. The second mention of the Olympic embassy at 15. 7. 2 is not a 'doublet' and need not come from a different source from 14. 109 (see below). The same is true of the epigamia. At 14. 107. 3(under 388/7) Diodorus says that Dionysius wished to punish the Rhegians because of an insult. For at the time he had approached the city with the offer of a marriage alliance, it is said (θασί) that the demos of Rhegion replied they could only grant him the daughter of the public executioner. In 14. 44. 4 f. (under 398/7), on the other hand, where the incident is related in its proper context, it is merely said that the Rhegians, having discussed the matter in assembly, simply turned down Dionysius' request to be granted a Rhegian bride. Sordi accounts for this 'inconsistency' by assigning 14. 107 to Timaeus and 14. 44-6 to Ephorus. But the insult connected with the rejection of the offer of epigamia is also mentioned in 14. 106. I which clearly carries on from 14. 105, a chapter Sordi rightly agrees in attributing to Ephorus. Ephorus, it is to be presumed, doubtless reflecting contemporary opinion, saw Dionysius' refusal to be reconciled with Rhegion after his great ____________________ 209 1992: 51 ff. 210 Ibid. 61. 211 Ibid. 62. -77-

ii. iii.

victory at Eleporus when he came to terms with the Italiot League as stemming from a private grudge of this sort; and so he mentioned now (14. 106. 1; 107. 3 f.) what he had had no reason to mention earlier. The only thing that is perfectly clear about 14. 90 and 14. 95 is that they relate not one (as Sordi) but two quite distinct events. There is nothing even mildly surprising in what is said about Mago: he had initially been stationed in Sicily in a defensive capacity (14. 90. 2), but was later put in command of the expeditionary force there (14. 95. 1.). In my view Sordi seriously underestimates the amount of criticism of Dionysius that Ephorus incorporated in his work (see above). I quite agree, and have argued so, that Timaeus was as hostile to Dionysius as he was to Agathocles, 212 but such intense and unqualified hatred is nowhere to be met in Diodorus' account, with the exception of the speech of Theodorus (on which see above).

What then of Book 15? The main fact about the western material in Book 15 is its sketchy and generally unsatisfactory nature. And'yet Dionysius was as active in the period 386-367 as before. The usual explanation offered for this circumstance is that the account of Philistus was much briefer for these years as a result of his exile in the mid-380s. As he was the major source for Dionysius, the accounts of the later writers who relied on him could not but reflect his dearth of detail; and inevitably that of Diodorus too, whether he followed Timaeus 213 or Ephorus in the main 214 or some other source. 215 As far as the supposedly impoverished state of Philistus' narrative of the second half of the reign of Dionysius is concerned, the truth is that it is a hypothesis incapable of proof. Thucydides, the historian Philistus emulated, was only marginally affected by his ____________________ 212 Ibid. 64. 213 Volquardsen, 1868: 103ff. 214 Schwartz, RE v. 68 1 f.; Meister, 1967: 100 ff. 215 Barber, 1935: 168 f., is sceptical that either Timaeus or Ephorus were used and is inclined to leave the question open. Likewise Sanders, 1987: 119 ff., 143 ff.; id., 1988: 54 ff. Hammond, 1938: 144, opts for Theopompus. As far as the latter is concerned, compelling evidence is necessary, which Hammond fails to supply, before supposing that D suddenly, and for no apparent reason, abandoned both Ephorus and Timaeus for Theopompus. From this it follows that the sceptical position is out of place here. If we are reasonably certain of the sources for Books 11-14, why should 15 form an exception? -78 exile, and not in terms of the quantity he produced. Certainly Thurii, the city to which Philistus went as an exile according to Diodorus (15. 7. 4, but see n. ad loc.), was well placed for plenty of information to reach him. 216 In fact it is probably the case that Diodorus himself is to blame for the sorry nature of Sicelica in Book 15. Caven, on the other hand, who rightly exonerates Philistus, Ephorus and Timaeus, nevertheless blames not Diodorus himself, but 'the transmission of Diodorus' text'. 217 Specifically, he cannot believe that Diodorus is responsible for the very brief and unsatisfactory description of the important Third Carthaginian War in chapters 15-17. In his opinion Diodorus' account of this war was, to begin with, as adequate as those of the first and second wars against Carthage, but then 'at some remote point in the stemma of a sadly corrupt . . . archetype from which our extant manuscripts derive, 218 the passages dealing with the third Punic War were . . . collected together [his italics] in order to provide a single connected narrative of this war [for dramatic reasons, Caven suggests] . . . The section of Diodorus' book containing this connected account . . . was subsequently lost, and the resultant gap in the history of the decade was filled by a clumsy epitome . . . Certainly as regards their whole manner and style, chapters 15 to 17 read like an epitome, rather than a continuation of Diodorus' own narrative.' 219 ____________________ 216 Cf. also Caven, 1990: 186, who further points out that Ephorus and Timaeus could have obtained information from elsewhere if Philistus failed them. See further n. 173. Sanders, 1987: 45 f., draws attention to the fact that according to FGH 556 F28 the great preparations of the tyrant for war against Carthage in 398 were related in Book 8 of the Sicelica, and deduces from this that the war must have taken up the next two books also, so that of the four books (8-11) of περὶ Διονυσίου the first three dealt with the period 405-396 and the fourth, very briefly and unsatisfactorily, with the rest of Dionysius' career. Such a top-heavy view of the περὶ Διονυσίου is unwarranted. A less extreme and more acceptable position is that of Stroheker, 1958: 192., which assigns Book 9 alone to the great war. It should be noted, however, that FF30-2 from Book 8 mention Libyan nations and places and should therefore correspond either to D 14. 54. 4-6 or to 14. 77. It is thus not impossible that the whole of the first war against Carthage was narrated in Book 8. Besides, we know that Philistus accorded Dionysius' funeral detailed treatment (FF28, 40b), and this was at a time when, according to Plut. Dion 11. 4 ff. (which Sanders accepts), Philistus was still in exile. If he could obtain information about the tyrant's funeral why not about other aspects of his rule? 217 1990: 186. 218 He refers to Vogel's comments in the introduction to the Teubner edn. (vol. i, p. xxi) about the early corruption of the manuscript tradition. But Vogel is talking about simple textual corruption, not about what Caven proposes. 219 1990: 188. Caven thinks that 15. 24. 1-3 provides support for his thesis; but see pp. 82 f. -79

This is an extraordinary theory, and it is unconvincing. On the contrary, chapters 15-17 read very much like a Diodoran epitome. We have seen (pp. 9 f.) that one of the main characteristics of Diodorus the epitomator is his lack of consistency. Sometimes he lets us know that he has abbreviated his source more severely than usual: for instance, at 13. 104. 8. And at. 15. 15. 3 he informs us that he has deliberately omitted most of the detail of the warfare in Italy and Sicily. The Fourth (and last) Carthaginian War (ch. 73) is equally briefly told, but there is no suggestion there of an epitome of Diodorus' epitome. It is salutary to recall that Hellenica in Book 15 do not fare any better than Sicelica -- witness, for example, the single sentence on the Spartan treatment of Phlius (15. 19. 3). And yet we know that Ephorus dealt with Greek affairs of this period in no less than seven books. What might we not assume about the scale and scope of this part of Ephorus' work if Diodorus Book 15 was all the evidence we had! But whatever the truth of the matter, on its own this circumstance is hardly indicative of any particular source. Ephorus must have been briefer on Sicily than Timaeus, though very much more detailed than Diodorus. 220 Nor can anything be concluded vis-à-vis source criticism from the fact that there are no named fragments in Book 15 (see pp. 51 f.). Turning now to the material itself, the question to ask once again is whether there is anything which suggests that Diodorus had turned to a source other than Ephorus. Under the year 386/5, and sandwiched between mainland Greek and eastern affairs, we have two chapters on Dionysius. What these purport to give us is a picture of the tyrant at home, as it were, of how he spent his time when not on campaign. The chapters are largely taken up with Dionysius' literary concerns, namely his great fondness for writing poetry and with the problems this created. Such stuff lends itself to anecdote and the anecdotal nature of the chapters is obvious. Isocrates provides evidence ( Archid. 44 f.) that the truth about the tyrant became confused with legend while the tyrant still lived. And Ephorus. was just as likely as Timaeus to include such material in his Histories, if only in the

____________________ 220 I cannot, therefore, go along with the view of Lewis, CAH2 vi. 121, 123, that the poor state of Sicelica in Book 15 is due to the fact that D no longer used Timaeus, but relied entirely on Ephorus. This is seriously to underestimate D's ability to butcher his sources. -80form of a digression. Chapters 6 and 7 in fact give every impression of deriving from an Ephoran narrative 'chunk' of this kind. It is of crucial importance to note the statement at the beginning of chapter 6: relieved of the wars with Carthage, Dionysius enjoyed plenty of peace and leisure. And so he turned to the writing of poetry. Now, the peace treaty with the Carthaginians was concluded in 392, 221 and therefore the only safe chronological conclusion to draw for the events in chapters 6 and 7 is that they come after that date, and not necessarily under or after 386/5, as some suppose, though two of these events (Plato's visit and the exile of Philistus and Leptines) can be dated a little more closely. 222 This observation invalidates the argument of Schwartz that there is a stark contrast between 15. 7. 2 and 14. 109, pointing to two different sources. 223 In fact there is no contradiction. In a general discussion of Dionysius' literary pursuits, Ephorus briefly repeated the well-known tale of the discomfiture of Dionysius' embassy to the Olympic Games of 388. Only this time he connected the tale with the court intrigues which resulted in the exile of Philistus and Leptines. The latter was deprived of his naval command in 390/89 (14. 102. 3) and may have been exiled at the same time or, more likely (see 7. 3 n.), shortly afterwards. The favourable terms in which these two are described is a further argument that Ephorus rather than Timaeus is the source here. 224 Under the years 385/4 and 384/3 we have five consecutive chapters of western history, 13-17. Nothing in these chapters is inconsistent with Ephoran origin. The only possible objection is the statement at 15. 13. 5 that Dionysius surrounded Syracuse with a defensive wall. This has been seen as a restatement of the information given at 14. 18 (under 402/1), and therefore as indicating different sources. 225 The conclusion is unsound. First of all, ____________________ 221 Beloch, GG iii. 2. 370; Berve, Tyrannis, i. 233. 222 15. 7. 1 and 15. 7. 3 nn.; cf Stroheker, 1958: 233 n. 43. 223 RE v. 681 f. As he assigns 14. 109 to Timaeus, he is inclined to see Ephorus as responsible for 15. 7. 2. In this he is followed by Meister, 1967: 99 f. (a 'doublet' resulting from the use of two different sources). That the two passages are incompatible is a generally held view; cf. Barber, 1935: 168; Vial, p. x; Sanders, 1988: 56 f. For Lewis too, CAH2 vi. 139 n. 82, 15. 7. 2 is 'a partial doublet' of 14. 109. See further ad loc. 224 By contrast, the Plato incident is of no significance vis-à-vis the sources, pace Sanders, 1988: 56. 225 Meister, 1967: 101, considers Ephorus responsible for 15. 13. 5, and Timaeus for 14. 18. -81-

there would of course be nothing surprising in one and the same source, whether consciously or unconsciously, repeating the same thing. But in fact it is not obvious that what is given at 15. 13. 5. is the same as what is described at 14. 18 and 14. 42. 5. Some of the detail at 15. 13. 5 is clearly new (the gymnasia and the temples), and the safest interpretation of the section perhaps is that at this point Ephorus discussed Dionysius' building activity in general, both that which had preceded the war with Carthage and that which followed it (see n. ad loc.). The words which open the section, Διονυ+́σιΟς σχολὴν ἄγων(the lacuna is perhaps best placed before Διονυ+́σιΟς; see ad loc.), inescapably remind one of the beginning of chapter 6: enjoying a great deal of leisure after the treaty with Carthage in 392 Dionysius devoted himself to poetry. And not only to poetry, but, to judge from 15. 13, to architecture at home as well as to the foundation of colonies in the Adriatic. The summary of the war against Carthage of 383 betrays no animosity against Dionysius. Quite the contrary. Dionysius, but especially his brother Leptines, are credited with great bravery. It is true that this may be due to Diodorus' style, to some extent. But at the very least it suggests that there was no hostility in the account he followed. Also, Dionysius' demand that the Carthaginians should retire from the cities of Sicily, a demand which they considered 'grievous and haughty' (15. 15. 4-16. 1), does not quite square with Timaeus' known attitude. It harmonizes well, on the other hand, with Ephorus' policy of reporting what was to Dionysius' credit (as here) no less than what was to his discredit, as, for example, at 14. 75. 3. 226 Chapter 15. 24 (under 379/8) is the next to deal with western affairs. Meister considers the section 15. 24. 3 as certainly Timaean -- a ΘεóπεμΠτος ἀτΥχíα which struck Carthage and which was averted with difficulty after the deity had been propitiated by sacrifices. On the other hand he assigns the previous section, 15. ____________________ 226 Meister, 1967: 102, 104, is amongst those who see a contradiction between the two passages, such as would arise out of two different sources having been used. But a historian could claim that in 396 Dionysius did not consider it to be in his interest to destroy Carthaginian power totally; and the same historian could maintain that at a later date the tyrant demanded that Carthage should evacuate all Sicily. On the other hand, we have no reason to suppose that Ephorus was fairer to the Carthaginians than Timaeus. Meister, 1967: 105, is misled by 15. 17. 5 into assuming this. But he overlooks 15. 16. 1! On Ephorus' attitude to barbarians see pp. 114 f. -8224. 2, to Ephorus -- a straightforward mention of an outbreak of the plague in Carthage as a result of which the Libyans and the island of Sardinia revolted. These reversals, that is, the epidemic and the revolts, encouraged Dionysius to declare war on Carthage, according to 15. 73. 1 (under 368/7). Between 15. 24. 3 and 15. 73. 1 there are no sections on western history. Meister joins 15. 73. 1 directly on to 15. 24. 2 and considers 15. 24. 3 an insertion from Timaeus, a secondary source for Book 15. To this conclusion he is led not only by the contents of 15. 24. 3, but also by the contradiction he perceives between 15. 24. 2 and 15. 73. 1, on the one hand, and 15. 24. 3, on the other: according to the latter, the plague and the revolt of the Carthaginian allies had come to an end long before the last war

of Dionysius against Carthage in 368. By contrast, 15. 73. 1 claims that it was the pestilence and the rebellion which induced Dionysius to go to war. Yet 15. 24 is under 379/8 while 15. 73 is under 368/7. Such a chronological inexactitude highlights once again the difficulties Diodorus faced when using Ephorus. 227 The last point may readily be granted, to the extent at least that Diodorus failed, and not for the first time, to appreciate the timescale of the account he abbreviated: the plague and the rebellion may have begun in 379/8, but are likely to have continued for a number of years; and Dionysius may have begun preparing for war before 368. 228 We notice in any case that the epidemic and the rebellion are described as 'having taken place' (γεγενημε+́νην) when Dionysius decided to go to war (15. 73. 1. See n. ad 15. 1.) Chronological oddities are no reason for denying 15. 24. 3 to Ephorus. The alleged quick conclusion of the revolt is very likely due to compression by Diodorus of his source. Nor can the supernatural element be held to indicate Timaeus rather than Ephorus (see above and pp. 12 f.). What should be noted, on the other hand, is the fact that 15. 24. 2 and 15. 24. 3 clearly form part of the same narrative: the plague and the revolts break out in 15. 24. 2 and are brought under control in 15. 24. 3. After that we must wait until 15. 73-4 (under 368/7) for western history. Chapter 15. 73 briefly traces the course of the last war of Dionysius against Carthage and concludes with the bare report of the tyrant's death. The detail about the length of his reign and that ____________________ 227 Meister, 1967: 103. 228 This is not exactly Meister's view. But see further ad 15. 1. -83-

of his son who succeeded him was clearly taken from the chronographic source. Otherwise 15. 73 may be credited to Ephorus. And to Ephorus should also be assigned 15. 74. 5, the tone of which bears no malice towards the Dionysii; quite the opposite. Sections 15. 74. 1-4 are a different matter. The four sections are of course a digression, and in Diodorus the phrase introducing them (οἐκ ἀνπίκειον) sometimes marks a change of source too (cf. 12. 11. 4; 15. 44. 1; and 12. 38. 1 for a similar phrase). And this seems to be the case here. 15. 74. 1-4 can hardly derive from Ephorus. It would appear that, wanting to add some spice to the story of Dionysius at this point, Diodorus turned to his alternative source, Timaeus. The story is certainly slanderous enough to qualify for Timaean authorship. 229 In Book 16, chapters 5, 6, 9-13, and 16-20 could be fairly described as an aristeia of Dion. They clearly form a single narrative, evidently from the same source. In 16. 5 (under 359/8) we are taken back in time in a review of the first years of the tyranny of Dionysius II. There then follows the story of how Dion succeeded in freeing Syracuse from the tyrant. The story stops at the high point of Dion's career (16. 20. 6 -- under 356/5), after which nothing more is said about him, or about the West, until 16. 31. 7 (under 354/3), when a chronographic entry informs us of the murder of Dion. We can only guess why Diodorus did not expand on the entry though it does appear likely that Ephorus did not continue his western narrative beyond the year 356 (see p. 97 ). At any rate, the assumption that Ephorus is the source which lies behind these chapters seems confirmed by the manner in which Philistus' death is reported (16. 16. 3 f.): see pp. 66 f. (i) Ephorus: the Oikonomia. We saw above that Diodorus relied on Ephorus for his accounts of mainland Greek and Persian history. Also, that the same author very probably provided Diodorus with the chief part of his western narratives. We may feel reasonably confident in addition about the way Diodorus worked, employing one main narrative source at a time (p. 49 ). His text then can be used in conjunction with the testimonia and fragments assembled by Jacoby to obtain ____________________ 229 Cf. Laqueur, RE via. 1150. The arguments Meister, 1967: 104, Sanders, 1988: 56 f., and Caven, 1990: 3, advance against this are not insurmountable. -84a reasonably clear idea of the kind of historian Ephorus was, that is to say of the scope and quality of the Histories. 230 Sometime about the middle of the fourth century an important innovation was introduced (perhaps by Ephorus himself) to the writing of lengthy prose works, not least of histories; the division into books of the various logoi or narratives. As we see the new method practised by Ephorus and Theopompus, their histories were not merely divided into books, but each book carried a proem or preface (16. 76. 5=FGH 70 T 10 ). The primary task of the preface was to define the book (or books) following (see below), but both Ephorus and Theopompus also used it as a device for, inter alia, publicizing their historiographical aims and techniques and, inevitably, criticizing other writers. The procedure of dividing historical compositions into books will in turn have necessitated the numbering of these books and we should have no hesitation in accepting the known numeration of Ephorus' lost books as Ephorus' own. On the other hand, it is unlikely that Ephorus also named his books and the title ' Europe' which Strabo assigns to Book 4 (FGH 70 F 42 ; cf. FF 30 and 33) should probably be attributed to the work of Alexandrian scholars. 231 Crucial for the quest for Ephorus' oikonomia is Diodorus' statement at 5. 1. 4 that Ephorus τω + ̑ν . . . ßίßλων περιἐχειν κατἀ γἀνος τἀς πρἀΞεις. This is a little vague and it has been variously interpreted, but its natural meaning is that within each book the praxeis were distributed according to their genos. Chapter 5. 1 seems to

have been patterned on a preface by Ephorus, one which discussed and sought to justify his method. Diodorus, doubtless echoing Ephorus, stresses the importance to a historian of the correct arrangement (oikonomia) of his material. Style and experience, toil and diligence are fine and justly deserving of praise, but they should not exempt a historian from censure if he has failed to give to his work a proper oikonomia (5. 1. 1-2. The rather irrelevant comment at this point (5. 1. 3) about Timaeus having earned his nickname of Epitimaios because of his lengthy and out-of-place censures is Diodorus' own). Ephorus, on the other hand, in writing a general history, succeeded not only with his style, but also with his oikonomia. Diodorus himself ____________________ 230 The possibilities as well as the limitations in this sort of investigation are well delineated by Brunt, 1980. 231 Cf. FGH iic. 27, 48 ff. -85claims to have been persuaded that Ephorus' kata genos arrangement was the best and that he would therefore adopt it and hold to it as far as possible (5. 1. 4). But what defined a genos? Diodorus' immediate application of the principle was a reductio ad absurdum. He decided to call the book in hand (the fifth) nesiotike and to deal in it with all the islands. But to lump together such different places as Crete and Britain simply because they were islands made little sense in the circumstances, and Diodorus in fact quickly disregarded the original intention (5. 24. 1) by also including in Book 5 accounts of Gaul and Spain. 232 Earlier, at 1. 9. 5, in a section where Ephorus is mentioned, Diodorus, very probably expressing an Ephoran idea, defines Greek and barbarian history as of different gene and undertakes to treat them separately. This argues for the thematic unity of each book, or group of books, and from what Diodorus says and does elsewhere too, as well as from the fragments of Ephorus, the nature of arrangement kata genos emerges even more clearly. It was intended, it would appear, to be an improvement above all on the annalistic method of Thucydides and the author of the Hellenica Oxyrhynchia. This is stated plainly in 16. 1, another preface modelled on one by Ephorus (see below and p. 101 ). As a pioneer in this respect Ephorus would no doubt have felt the need to explain the merits of his method to the reader on more than one occasion. Historians, we are told, ought as far as possible to include in their books the praxeis of kings or cities autoteleis, that is, complete in themselves, from beginning to end. For half-finished (hemiteleis) praxeis interfere with the expectations of the reader and are not as easy to remember or as intelligible (16. 1. 1-2). Three centuries later Dionysius of Halicarnassus, without acknowledging Ephorus, criticized Thucydides' oikonomia along very similar lines ( Thuc. 9). It is important to appreciate that praxeis is the key word in the passage and Diodorus' (or rather Ephorus') exhortatiori clearly is that praxeis of cities or kings should be included in books complete. He is not suggesting that the entire careers of kings should be included in single books though that, naturally, would depend ____________________ 232 For the writing of works on islands in the Hellenistic period see Gabba, 1981: 55 ff. In 5. 1-2. 1, however, D is concerned with a technical aspect of historiography, and his chief inducement for the decision to devote a book to islands was the oikonomia of Ephorus' work. Gabba in any case is wrong that Book 5 is taken up entirely with the histories of islands (1981: 56). -86on the importance of the king concerned and the amount of material available to the historian as well as, of course, the scale of his work. But the emphasis, it must be repeated, is on praxeis: the individual praxeis should not be split between books. It was Diodorus' intention to narrate all of Philip's praxeis in one book (16. 1. 3). 233 Ephorus cannot have intended to do so. The sheer number and magnitude of them and Ephorus' detailed treatment of his own time excluded that if nothing else did. Of course, at every level geography was an important factor (though not the only one -- see below) in determining the grouping together of the praxeis, and certainly at the highest level the kata genos approach involved, as Jacoby has suggested, the apportionment of the various narratives among the major areas -- Greece, the East, the West, and (after 360) Macedonia. 234 In Thucydides' relatively restricted scheme of a monograph on a single war the management of the material was no less geographic, the narrator needing to move from one area of conflict to another. The wider canvas of a history of the world would have made the distribution of the material into Hellenica, Sicelica, Persica, and, later, Macedonica unavoidable. Such ἐπί μἐρους histories flourished in the fourth century and Ephorus' main task had been to bring them together within the framework of a single composition. We know, for example, that he used the Persica of Ctesias, the Sicelica of Philistus, and the Hellenica of Callisthenes and the Oxyrhynchus historian (see below). Traces of this sort of arrangement in Ephorus seem to survive in three passages in Diodorus, at 11. 20. 1, 16. 5. 1, and 16. 64. 3. All three are normal transition formulae which Diodorus regularly employs when switching from one area to another, but unusual because of the expression ἐτερογενει + ̑ς πρἀΞεις which occurs in them. In the first and third instances the switch is from Europe ( Greece) to Sicily, while in the second the narrative is transferred from Macedonia to Sicily. Greek and Macedonian praxeis are therefore deemed to be of a different genos from Siceliot ones. But we still need to know what Ephorus did ____________________ 233 If any part of the preface to Book 16 is to be assigned to Ephorus then 16. 1. 1-2 surely deserves first consideration. Section 16. 1. 3, especially the first sentence, is certainly D speaking. We must not use 16. 1. 3 to interpret 16. 1. 1-2. 234 FGH iic. 26. For arguments about Ephorus' oikonomia not dissimilar to those advanced here see Vannicelli, 1987: 165 ff. The present section was completed before I saw Vannicelli's paper, and

its core formed part of my Oxford doctoral thesis of 1981. -87within these broad gene or categories. Strabo (13 C 623 = FGH 70 F 236 ) mentions a criticism levelled at Ephorus that ἐν τῃ + ̑ διαριΘμήσει τω + ̑ν ἂλλων πρἀΞεων . . . οὓτως ἐπιθωνει + 'κατα' δἐ τ αὐτòν καιρòν Kυμαι + ̑Οι τἀς ἡσυχίας ἠ + ̑γοὐ. This rather points to a region by region arrangement of the praxeis on a broadly synchronistic approach (κατἀ δἀν αὐτρóν καιρóν). In what way then did his disposition of events differ from that of Thucydides? We get our answer by looking at what he made of the history of the fifth and fourth centuries. One tendency, as we see clearly in Diodorus, was to group events around some prominent individual or theme and to allow the resulting narrative to run on until it formed a praxis autoteles. The time-span thus covered varied and might be quite lengthy. The books on the fifth century covered fairly long periods and narratives would naturally tend to extend over quite a few years. Many books on the fourth century, on the other hand, were taken up with shorter periods, some covering no more than two years (see below). Inevitably the praxeis autoteleis would also be of a shorter time-span. For the Peloponnesian War down to 411 Ephorus of course relied heavily on Thucydides and the way he rearranged the latter's text sheds ample light on his method. For example, the siege and surrender of Potidaea are narrated in six consecutive sections in Diodorus (12. 46. 2-7). In other words they are made into one episode. In Thucydides on the other hand they form two (2. 58, 70). Similarly, the Athenian involvement in the West in the Archidamian War which is spread over three years and three books in Thucydides (3. 86, 88, 90, 115; 4. 24 f., 48. 6; 58 ff.; 5. 4. 2 ff.), is made into one story in Diodorus, 12. 53-54. 7 (under 427/6). We may further compare 12. 55 (Mytilene), 12. 56 (Plataea), 12. 58 (the plague). Also, we note how at 12. 60 ff. and 12. 65 the actions of, respectively, Demosthenes and Nicias have been taken out of their correct chronological setting in Thucydides and grouped together to form single narratives. 235 For the pentekontaetia Ephorus did not rely entirely or even primarily on Thucydides' brief account and the alternative sources he employed are lost in any case. All the same, Diodorus' text leaves us in no doubt that Ephorus' oikonomia here was the same as what we have just seen. For instance, 11. 54-9 (under 471/0) ____________________ 235 Drews, 1963: 245 ff., is thus certainly wrong in thinking that Ephorus adapted himself to Thucydides' annalistic pattern. -88-

gives us the whole of Themistocles' later career, from his ostracism until his death in Persia, complete with a typically Ephoran assessment of the man at the end. Chapters 11. 60-2 (under 470/69) are an aristeia of Cimon, from the capture of Eion until the battle of Eurymedon. In the same way, 11. 81-3 (under 457/6) is an aristeia of Myronides, and 11. 84 (under 456/5) of Tolmides. The unmistakable clue in the case of Myronides is the Athenian expedition to Thessaly of the mid-450s. According to Thucydides, who does not name the Athenian general in command, this took place a couple of years after the battle of Oenophyta and in his narrative it is separated from that campaign by other events (1. 111. 1-2). In Diodorus, on the other hand, the leadership of the expedition is assigned to Myronides and the expedition is tacked on to the campaign which led to the victory at Oenophyta and the subjection of Locris and Phocis; and the narrative is concluded with the return of Myronides to Athens in glory (11. 83. 3-4). For the fourth century, apart from the fragments of the Hellenica Oxyrhynchia, no source employed by Ephorus survives, but Ephoran narratives of the same sort as before are still detectable in Diodorus here and there. For example, we have the various Persian narratives (i. e. the Cypriot War (14. 98. 1-4; 15. 2-4; 15. 8-11), the campaigns against Egypt (15. 29. 1-4; 15. 41-4; 16. 40. 3-51), the Great Satrap Revolt (15. 90-3), etc.), the Spartan attack on Mantinea (15. 5; 15. 12), the Olynthian War and related topics (15. 19-23), the foundation and early years of the Second Athenian Confederacy (15. 28. 2-5; 15. 29. 5-30) (and see also the analysis of Book 15 below, pp. 98 ff.). Of course, it does not follow from the fact that geography was central to the kata genos arrangement, that the main gene, i.e. Hellenica, Persica, etc., were assigned separate books. It was Jacoby's contention, one widely shared, 236 that each of the Ephoran books was restricted to just one of the major areas. This is not quite right. Not only is it not supported by 5. 1. 4 (see above), but there is not the slightest evidence that Ephorus took pains to treat Persian affairs, for example, in separate books from Greek affairs. What evidence there is points in the other direction. FGH 70 F76, referring to the early stages of the Cypriot War, comes from Book ____________________ 236 Cf. Drews, 1963: 244ff.; id., 1976: 497; Meister, 1971,: 506ff.; Schepens, 1977: 115f. -89-

19 and Book 19 also dealt with Greek affairs. 237 Similarly FGH70 FF64, 65, and 191 (from Book 11) also show that Persica and Hellenica were narrated in the same books. Persian history in any case was not treated with the same thoroughness and consistency by Greek historians as Greek history. Even Ephorus' famous universal history was still primarily the history of the city-states of the Greek mainland. 238 One would guess, and this is supported by the contents of Diodorus Books 11-16, that most, if not all, of the Persian narrative in Ephorus concerned

events which were of immediate interest to Greece. The Persian narratives were few compared to the Greek ones. They occurred in sizeable 'chunks' here and there. But there was no connected history of Persia as such, only a series of events which bore directly on Greek history, and my contention is that these were juxtaposed with the Greek events to which they were related. What of western affairs? These were largely independent from those of mainland Greece and it would not be surprising if Ephorus, like Theopompus in the Philippica (16. 71. 3), assigned them their own books. This is in fact the common scholarly view and the numbered fragments are supposed to support it. As far as Ephorus' account of the fifth and fourth centuries is concerned we have only one fragment from Book 12, two from Book 16, three from Book 28, and one from Book 29. As all of these are thought to refer to the West, all four books are usually given over to western affairs. 239 The view is almost certainly wrong, in part at least. The first fragment from Book 16, F68, is undoubtedly from a western narrative. It corresponds to 14. 9. 9 (under 404 BC ). F69, on the other hand, is at first glance mysterious. 'bezieht sich vielleicht auf die Karthager oder ihre libyschen hilfstruppen' is Jacoby's only comment. 240 Jacoby, of course, hampered by his theory of separate books assigned to the various geographic regions, was at a serious disadvantage -- it is not possible to relate F69 to any known event of western history at this time. However, once we are prepared to overlook the theory, two notorious and ____________________ 237 It should be mentioned that the book number of the fragment is not beyond dispute. It is either ιή or ιθ+́, see Reid, 1974: 124 n. 3. But this changes nothing, for again Book 18 contained Greek history. 238 Cf. Mioni, 1949: 23. 239 FGH iic. 26; Barber, 1935: 41ff., 173f. 240 FGH iic. 57. -90interlinked episodes from Spartan history come readily to mind. The fragment could refer to either of them or indeed to both. F69 is a scholion on Hom. Od. 3. 214-15. In addressing Telemachus, Nestor, who is apprised of the problems the young prince is facing at home, enquires if the hostile attitude of the people of Ithaca is due to the utterance of a god. The scholiast comments that on many occasions kings were put aside by their subjects (μεθίστασαν -- the same verb as used by Plutarch in the quotation below) in obedience to oracles; and adds that in Book 16 Ephorus narrates about the gods. The text of the scholion is not free of problems, but there can be little doubt that Ephorus had spoken about oracular responses and the removal of kings. What springs to mind immediately is the foiling of the boy Leotychidas by Agesilaus and Lysander -- an oracle was involved in that instance and a prince was actually put aside. The second episode concerns Lysander's alleged attempt to undermine the constitutional position of the twin royal houses of Sparta. The best and most detailed account of the attempt is that of Plutarch in his Lysander, who attributes his information to Ephorus. 241 Briefly, the story is as follows. Not willing to resume the station of a private citizen after the Peloponnesian War, Lysander conceived a plan to reform the office of king at Sparta so that it would be open to all Heraclids, perhaps even to all Spartiates (διενοει+̑το τὴν ἀρχὴν ἐκ δυει+̑ν οἴκων μεταστήσας εἰς κοινὸν ἀποδου+̑ναι πα+̑σιν Η+̔ρακλείδαις -- 24. 5). The most effective way of achieving his aim, he realized, was by first bringing to bear on the Spartans the fear of the gods in the form of suitable oracular responses (25. 2). There then follows the unsuccessful attempt to suborn the oracles of Delphi, Dodona, and Zeus Ammon. 242 ____________________ 241 Impossibly Plutarch sets the conspiracy in 395, after Lysander's return from Asia (24ff.). This is obviously wrong as shown by: (1) Plutarch's own evidence: 20. 6-8 (Cf. 24. 2); (2) D 14. 13 (under 403/2),which is set in the immediate aftermath of the Peloponnesian War (cf. Nepos, Lys. 3. 1); (3) the fact that Lysander would probably not have had enough time in 395/4 to hatch such a complicated plot (if historical, of course), as Grote points out (ix. 334 n. 1). 242 The nameless 'historian and philosopher' at 25. 5 is probably not Ephorus, though not because of the appellation 'philosopher' which worries some commentators (e.g. Flaceliére and Chambry, Budé edn. of text, 161f. and ad loc.). For Strabo, 1C1, Ephorus was a philosopher no less than a historian. But Smits, in his edition of the Lysander ( Amsterdam, 1939), appears correct in arguing that Plutarch at 25. 5 is explicitly supplementing Ephorus. However, whether Ephorus is responsible for the story within a story or someone else, the argument put forward here is not affected. Book 16, then, included both Hellenica and Sicelica, just as Books 11 and 19 contained both Hellenica and Persica. Jacoby's thesis, therefore, of separate books for the different regions, always and as a matter of course, becomes unsustainable. Attention was drawn earlier to the expression ἑτερογενει+̑ς πράξεις which Diodorus three times uses when switching from Greek or Macedonian events to Siceliot ones. This is further evidence that praxeis of different gene could, and often did, occur in the same books, for otherwise there would have been no need for Ephorus (if the expression, as it seems probable, goes back to him) to stress that he was moving from one genos to another. 243 It is true, of course, that Books 4 and 5 of Ephorus, which formed a geographical preparation, concentrated the first on Europe and the second on Asia. However, to assign one book to Europe and another to Asia in this instance is one thing; assiduously to distribute the historical narratives of the various areas into different books as a matter of principle, quite another, and we have seen that the evidence tells against it. I would maintain on the contrary that the Greek, Persian, Siceliot, and (later) Macedonian narratives could, and often did, occur side by side or rather alternatively in the Ephoran books just as they do in those of Diodorus. What, however, determined the extent and nature of the juxtaposition of the various narratives in a book or group of books? Was it a random process or was there a purpose to it? Of course, the amount of information available to Ephorus was an important factor, as was the fact that he was more interested in some areas than others. It is not surprising, therefore, that as a rule Greek metropolitan affairs took the lion's share of space. Books 23-5, for example, dealing in great detail with the 360s, were probably pure Hellenica. By contrast it is to

be strongly doubted whether any books were reserved for Persia alone. But the deciding factor here surely was the theme of each book or group of books, for there cannot be any doubt that Ephorus aimed at the thematic unity of his books. We have seen that he had pioneered the division of historical works into books, each an identifiable unit with a preface. The remnants of some of these prefaces in Diodorus are illuminating. The preface to Diodorus Book 12, the full version of which must have headed Ephoran Book 11, shows that the latter ____________________ 243 Drews, on the other hand, 1976: 497f., thinks that the expression puts Jacoby's thesis beyond any doubt. -92concentrated on the pentecontaetia. The Persica narratives in it (see above) were pertinent to this theme. The preface to Diodorus Book 14 was based on the preface to Book 16 of Ephorus. Its central theme was the initial period of the Spartan hegemony and its evil consequences, one of which was the buttressing of the tyranny of Dionysius I. Hellenica in this instance, therefore, and the corresponding portions of Sicelica were narrated in the same books, one of which certainly was Book 16 (see above). Books 28 and 29, on the other hand, seem from the fragments 244 to have dealt exclusively with Sicilian matters, with the later career of Dionysius I from a point which cannot be determined, and with those of Dionysius II and Dion until the latter's defeat of Nypsius in 356/5. Thus Ephorus' method was flexible and made to serve the themes he wanted to emphasize. Whether Macedonica were also treated separately at some stage is a matter best considered in connection with the problem of the ending of Ephorus' work. It would be helpful to look first at the contents of the books of the Histories which treated the fourth century. 245 Persica apart, Books 18 and 19 certainly dealt with Greek affairs, with the Spartan adventure in Asia Minor and the Corinthian War ( FF71-8). So did Book 20 (F79, the dioikismos of Mantinea), Book 21 (F80, the naval battle off Naxos), and Books 23-5 (FF81-5, the Theban hegemony). We have no fragments from Book 22. The only fragment from Book 26 mentions a town in Egypt. The fragments from Book 27 refer to Philip's warfare in Paeonia and Thrace (FF87, 88). Books 28 and 29, as we have seen, dealt with Sicilian affairs from at least 385 to 358 (FF89-92), and Book 30, composed by Ephorus' son Demophilus, narrated the Sacred War (FF93-6; 16. 14. 3). We may proceed to look at the contents of these books in greater detail. The outbreak of the Corinthian War clearly delimited Books 18 and 19 as is indeed indicated by the fragments. Equally, the King's Peace stood between Books 19 and 20. At the end of 19 or at the beginning of 20? Dressler assigned it to the end of 19, 246 Jacoby to ____________________ 244 FF89-92. It is worth emphasizing that the number and nature of the fragments are such as to allow no certainty. 245 I am assuming that all book references are correct, something about which we cannot be certain in every single case; see, for instance, n. 237. 246 In his pioneering work now rendered seriously out of date by Jacoby collection of fragments, 1873: 22f., 25 (XXI printed wrongly for XIX). -93the beginning of 20, 247 while Barber, like Dressler, places it at the end of 109. 248 It may be of some significance that Diodorus himself deals with the Peace at the end of his Book 14 and makes a fresh start in 15 with the postPeace events, for it is probable that he was following Ephorus' arrangement here. (The preface to Book 15 could add some support, see below.) But the best argument for the Dressler suggestion is the fact that the Peace forms an excellent ending, but a poor beginning. F78 from Book 19 possibly refers to the peace negotiations, the final terms of which recognized Clazomenae as belonging to the Great King. Would Ephorus have ended Book 19 at this point and left the conclusion of the Peace to the beginning of Book 20? It does not seem likely. 249 One fragment survives from Book 21, F8o, 250 and none from Book 22 so that it is difficult to decide on the end of 20 and the beginning and end of 21. All we have to go on is Diodorus' arrangement of his material and the knowledge that important events would have been selected by Ephorus to stand between his books. There are two such here: the liberation of Thebes and the battle of Leuctra. Dressler's analysis of Diodorus Book 15 251 is useful in helping us to reach a decision. Dressler perceived that 15. 25. 1 smacks of a fresh beginning and we can take this a step further. At 15. 23. 2-3 (380/79) the surrender of Olynthus is related, and 15. 23. 3-5 is a review of Spartan strength at this juncture. This point, therefore, would make a natural finale for Ephorus Book 20. There then follows a gap as far as Greek affairs are concerned. Chapter 24 (379/8) deals with western affairs. And then a new chapter in Greek history is opened with 15. 25. 1. It is not unreasonable to accept that we have here the beginning of Ephorus Book 21. The gap at 15. 24 could be of some significance. Ephorus we may assume began with the outbreak of the Boeotian War, but naturally went back a little to cover its origins and deal with the ____________________ 247 FGH iie. 28. 248 1935: 34. 249 Callisthenes' Hellenica, a major source for Ephorus, cannot help. According to 14. 117. 8 Callisthenes began his Hellenica with the year of the King's Peace. But it appears that Book 1 was introductory, the detailed narrative beginning with Book 2 and the liberation of Thebes. In Book 1 he perhaps outlined the political situation of the early 4th century comparing it to that of the 5th, especially in relation to Persia, and the King's Peace would have been related in that context; cf. Jacoby, RE x. 1690, 1694ff.; FGH iid. 416f. 250 Not available to Dressler who based his investigation on Müller's collection of fragments. 251 1873: 23f. -94-

to have found the outbreak of the war under Nausinicus in his chronographer, and hence placed the whole episode under that year. He was left with a vacant year in his annalistic system between the end of Ephorus Book 20, 380/79, and the beginning of Book 21, 378/7 as the chronographer saw it, and this year he filled with western events. Section 24. 1 looks like a chronographic entry (see pp. 201 f.) to which Diodorus attached 24. 2-3, probably an abbreviated Ephoran chapter from Book 28. For the end of Book 21 Dressler's judgement again appears sound. The Common Peace of 372/1 could well have marked its end. Besides, if Book 21 began with 378 it could hardly have gone on to include Leuctra. At any rate it came down to at least 376 (F80). Book 22. then, from which no fragments survive, dealt with Leuctra and the events which followed the Theban victory. Book 23 certainly included the second invasion of the Peloponnese (FF81, 82), and it concluded with spring/summer 367, if indeed Book 24 (F84) covered the period from spring 367 till summer 365 (see 75-76 n.). Book 25 commenced with summer 365 and finished with the battle of Mantinea (F85). 252 F83 from Book 23 is interesting. The only likely context for it is Epaminondas' naval venture in 364 (79. 1 n.). Yet that can hardly have been related in Book 23. It is possible, however, that during his second invasion of the Peloponnese Epaminondas handed some territory over to one of the Boeotian allies (cf. 57. 1 n.; 75. 2 n.) and Ephorus in an aside commented on a similar action in 364. The contents of Books 26 and 27 are controversial and bear directly not only on the oikonomia of Ephorus, but also on the thorny question of the end of the Histories. At 16. 76. 5 (341/0) Diodorus states, and the statement is from the chronographer, that Ephorus τὴν ἱστορίαν κατέστροφεν εἰς τὴν Περίνθου πολιορκίαν. Before that, at 16. 14. 3 (357/6), another chronographic entry informs us that Demophilus, the son of Ephorus, wrote up in one book the history of the Sacred War which was omitted by his father, beginning with the capture of Delphi by the Phocians. It is clear from this, as well as from the fact that the continuing siege of Perinthus is a most inappropriate point at which to conclude a universal history, that Ephorus had been forced (by old age? Sickness?) to make a somewhat abrupt end of his work. The ____________________ 252 FGH iic. 28. -95original plan (pp. 110 ff.) remained unfulfilled. Still, the query remains: by what path was the year 341/0 reached? Book 30 of the published work was the account of the Sacred War by Demophilus ( "FGH"70 T9b; FF93 ff.). If we are right to assume (see above) that Books 28 and 29 were taken up entirely with the West then the year 341/0 must have been reached either by Book 26 or by Book 27. But what did these books contain? Jacoby assigns Book 27 to Philip, or rather Macedonica, and allows it to reach 341/0 and be incomplete at the time of its publication. 253 Jacoby keeps an open mind about the contents of Book 26: Greek history after Mantinea or the East. Barber is more definite; he fills Book 26 with Persian history from the revolt of Evagoras down to the death of Artaxerxes II or even later. 254 Hammond, on the other hand, thinks 255 that Book 26 treated Persian affairs down to the siege of Perinthus and was complete, while Book 27 included not just Macedonian affairs, but Greek affairs also, and broke off with 357/6, after which Diodorus' coverage of Philip becomes scanty, and was, therefore, incomplete. Hammond sees proof for his hypothesis in 16. 14. 5, a chronographic entry stating that Diyllus of Athens began his history with the year 357/6 and wrote twentysix books of Greek and Sicilian affairs; and in 16. 76. 6, another chronographic entry under 341/0, which states that Diyllus began his second syntaxis from where Ephorus left off and included the deeds of the Greeks and the barbarians till the death of Philip. By subtracting the second entry from the first Hammond deduces that Ephorus brought his Persian history down to 341/0, but his Greek and Sicilian narratives stopped with 358/7. 256 The immediate objection to Hammond is that the siege of Perinthus is related from the Macedonian point of view. 257 In any case, the event could hardly come under Persica. To the Persians the event was only of marginal importance; not so to Philip. Nor are the entries on Diyllus as straightforward as Hammond imagines. He assumes that the entry at 16. 14. 5 refers to the first ____________________ 253 FGH iic. 29f., followed by Barber, 1935: 39f. 254 1935: 38f. Barber is definitely wrong about the revolt of Evagoras: its beginning came in Book 19: F76 and above pp. 89 f. 255 1937b: 85ff. 256 In this he is followed by Sordi, pp. xiif., without reservations, and by Drews, 1963: 255 n. 33, with reservations. 257 So Drews, 1963: 254, too. It must be admitted, however, that this is not a conclusive objection; cf. e.g. 11. 1. 3ff., where the Greek and Persian viewpoints alternate within the same narrative. -96syntaxis. The unlikely result of such a hypothesis is that Diyllus covered the sixteen years from 357/6 to 341/0 (the first syntaxis) in no less than twenty-six books! But twenty-six clearly refers to the number of books of the whole work of Diyllus which seems to have reached down to the year 297/6. 258 The entry at 16. 14. 5 is confused. 259 On the other hand Hammond is surely correct about one point at least. Diyllus began with the year 357/6 presumably because he wished to cover what Ephorus had not. What was that? Sicelica and Hellenica in general, thinks Hammond, including Macedonica. As far as Sicelica is concerned there are indications that Ephorus did not proceed beyond 357/6 (see p. 84 ). On the other hand, Diodorus is explicit that of Hellenica only the Sacred War was omitted (16. 14. 3). 260 We have no reason for supposing that other Greek events and the achievements of Philip between 357 and 341 were not related. Indeed it is perfectly possible that aspects of the Sacred War were touched on too. But the narratives in Diodorus of, for example, the Social War (16. 7. 3-4; 16. 21-22. 2), the Olynthian War (16. 53-5), the rebellion of Artabazus (16. 34. 1-2), Philip's attack on Methone (16. 34. 4-5), and the war between Argos and Sparta (16. 39. 1-7) may well be based on Ephorus. 261 There were definitely Persica in Book 26 (the one surviving fragment refers to an Egyptian city: F86) and Macedonica in Book 27 (the two fragments from this book mention a Paeonian town and a Thracian tribe: FF87, 88). It is certainly

possible that, as Laqueur suggested, 262 Book 27 related the consolidation of Philip's power in the north till the renewed outbreak of hostilities with Athens during the sieges of Perinthus and Byzantium. That would indeed amount to a group of praxeis autoteleis on a theme and be entirely consistent with Ephorus' method. We cannot, however, be sure that there were no Persica in 27 and no Macedonica in 26; and Hellenica (e.g. Athenian activities in the Aegean) may well have figured in both books. That too would not ____________________ 258 21. 5; FGH iic. 113. The exact number in the MSS at 21. 5 is uncertain. It could be 25, 26, or 27. 259 As Jacoby appreciated, though he let the difficulty stand: FGH iic. 112. The book number in FGH73 (Diyllus) Fi (=Athen. 4. 41) seems to me to require emendation to ἐνάτῃ καί δεκάτῃ. 260 It might be objected that Demophilus had already written about that. Demophilus, however, dealt with the ten years of the war in just one book. Perhaps Diyllus wanted to do it greater justice. 261 Hammond's arguments to the contrary are unconvincing: 1937b: 83f. 262 1911: 336ff. -97-

have been inconsistent with Ephorus' method. At any rate, we can be reasonably confident that the siege of Perinthus, related from Philip's point of view, was reached in Book 27, which book was probably finished. 263 On the other hand, Ephorus' original plan (pp. 110 ff.) cannot have been thereby completed. The story Ephorus never told was only partly narrated in Book 30 by Demophiluss. The above conclusions on the nature of Ephorus' oikonomia receive some support from an analysis of the structure of the first half of Diodorus Book 15. Sections 15. 1. 1-5 probably reproduce the substance of a general Ephoran preface relating to the period 386-362 which was thus meant to serve as an introduction to six books of Ephorus, 20-5 (see further below). Book 20 contained both Greek and Persian narratives. Did Ephorus keep the two entirely separate, finishing one completely before beginning the other, or did he integrate the two to a certain extent thus fulfilling the demands of cause and effect? Either might be thought to be in accord with his oikonomia, though in fact the second happens to be true of Book 20. As we have seen, the early stages of the Cypriot War which broke out c.390 and drew to its close in the late 380s were narrated in Book 19 (on the Corinthian War), but its highpoint and conclusion reasonably enough were told in Book 20 which covered the years 386-379. The Ephoran narrative has left traces in Diodorus. One such is at 15. 5. 1-3 which so obviously marks a fresh beginning after the end of the narrative of the conclusion of the King's Peace. πρου+̇παρχούσης τοι+̑ς Ἕλλησι κοινη+ ς εἰρήνης τη+̑ς ἐπὶ Α+̔νταλκίδου . . . Λακεδαιμόνιοι κτλ. A condemnatory outline of the Spartan policy of the years 386-379 follows, possibly deriving from an Ephoran 'secondary' proem specific to Book 20 (see below), and then immediately upon this the reasons for the Spartan attack on Mantinea and the attack itself (15. 5. 3-5). It would appear, therefore, that Ephorus began Book 20 With Greek affairs. The gist of 15. 5. 1-3 is repeated at 15. 19. 1 and this passage, another beginning, which precedes the outbreak of the Olynthian War, also provides an indication that the Greek narrative was not continuous, but that something intervened between ____________________ 263 It is interesting to observe that Pompeius Trogus, surely under the influence of Ephorus, whether direct or indirect, also narrated the siege of Perinthus from the Macedonian viewpoint and made it stand between his Books 8 and 9: prol. lib. 8 and 9. On Trogus' sources cf. Schanz and Hosius, ii. 324. -98-

Mantinea and Olynthus -- the Cypriot War it seems. Resumed now from the previous book, the war ran perhaps without interruption to its conclusion. It must have formed a long and complicated account with few chronological indications 264 and this was probably one of the reasons for Diodorus' confusion (2-4 n.). The war ended in 380 and after its conclusion Ephorus turned aside to relate the revolt of Glos and the trial and acquittal of Tiribazus. These events seem to belong to 384 and 383; at any rate they were over by early 382 at the latest (2-4 n.). What should be noted is that Ephorus saw a close connection between the revolts in Asia against the King and the actions of Sparta in Greece, a connection which he spotlighted by the arrangement of his material. It would appear that Sparta had not given up her dreams of conquest in Asia Minor even after the King's Peace. She had given her blessing to Glos' attempt to establish a power base in Ionia against the King, perhaps hoping for a chance to intervene if the attempt proved successful (9. 4 n.). At the collapse of the revolt, however, Sparta renounced her Asian designs and concentrated on Greece (15. 19. 1). Ephorus consequently seems to have begun Book 20 with Greek affairs (Mantinea), switched to Persian affairs (the Cypriot War), and then returned and stayed with Greek affairs till the end of the ____________________ 264 Unlike his oikonomia, Ephorus' chronology has received little attention; cf. Jacoby, FGH iic. 27; Burde, 1974: 102 f. It is usually dismissed with the comment that it cannot have been much in evidence in his work. This appears roughly correct though an examination of D shows that Ephorus cannot have been worse in this respect than, say, Xenophon. There are quite a few chronological indications in D Books 11-16, the great majority of them from Ephoran contexts. Many of them can be shown to derive from Ephorus' sources; cf. D 12. 47. 3 = Thuc. 2. 79. 2; 12. 51. 2=2. 101. 5; 12. 52. 1=3. 1. 1; 12. 81. 4=5. 83. 1; 13. 8. 7=7. 16. 2-17. 1. 12. 81. 5-82. 1 is an adaptation of Thucydides; cf. 5. 81. 2. But as in the case of the numbers (p. 129 ) Ephorus had access to information other than that of Thucydides; cf. 11. 78. 4; 12. 80. 3. The relation between Ephorus and the Oxyrhynchus historian (P) was no doubt similar. We know that P followed Thucydides' chronological system (cf. Bruce, 8 f.). We can safely assume, therefore, that the observation of the seasons, summers and winters, in D in connection with campaigns is derived from P; see 13. 49. 2; 14. 17. 12; 14. 35. 7; 14. 38. 3; 14. 79. 3. There is also some more straightforward chronological information which probably derives from P too; see 13. 64. 7; 14. 38. 3; 14. 38. 5; 14. 80. 8. Book 15 offers a number of chronological references: 5. 3; 12. 1; 41. 2; 41. 4; 43. 4; 65. 2; 67. 1; 70. 1; 94. 1; cf. 30. 3; 34. 1. The

fragments of Ephorus are disappointing. The nearest we get to chronology is with FF207, 218. Thus the conclusion must be that there was probably not enough chronological detail of the right kind in Ephorus to facilitate D's annalistic task. Ephorus' primary concern was with presenting 'complete in themselves' narratives in a broadly synchronistic order. -99book. But even this latter narrative cannot have flown uniformly for Ephorus had to deal with the capture of the Cadmea at some stage. This Spartan outrage did not bring up the rear because of 15. 23. 3-5 -- the surrender of Olynthus and review of Spartan power (i.e. the end of Book 20). Nor did it come at the head of the renewed Greek narrative after the Cypriot War because of 15. 19. 1-2 -- the beginning of the Olynthian War. That the Theban account was somewhat independent in Ephorus from the Olynthian War is shown by 15. 20. 1-2: a new beginning was made stressing the reasons for the Spartan coup de main. Diodorus' muddle over Phlius further illustrates Ephorus' method. The suppression of Phlius receives but one sentence in Diodorus and it is sandwiched between the beginning of the Olynthian War and a report of a confrontation between Agesilaus and Agesipolis over foreign policy. All of these are under 383/2. But Phlius was not attacked till 380. Ephorus presumably had dealt with this phase of Spartan aggression and aggrandizement en bloc, analysing the origins of the Olynthian War, commenting on internal Spartan politics, and digressing to include the subjection of Phlius. Diodorus abridged all this as best as he could and placed it under 383/2. No less informative is the next Persian narrative in Book 15: 29. 1-4 (under 377/6). Scholars generally attach no special significance to the position of these four sections, but assume that Diodorus must have misplaced them quite irrationally. Diodorus' poor oikonomia will not, however, do as an explanation here. The year 377/6 is a very full one, no less than eleven chapters long. Diodorus, therefore, must have had a reason for placing these sections under 377/6. Either one of the events mentioned (the last perhaps, Iphicrates' dispatch to Pharnabazus) belongs in that year and was so shown in the chronographer, or Ephorus inserted a Persian narrative at this point. The second possibility is surely the correct one. After relating the foundation of the Second Athenian Confederacy and the Spartan change of policy towards their allies, and just before the raid of Sphodrias which dramatically hastened the outbreak of hostilities, Ephorus it seems paused to bring his eastern affairs up to date. We may note that the theme of the newfound Spartan philanthropia towards their allies and their military preparations which concludes chapter 28 is resumed again later at 31. 1. What we therefore appear to have at 15. 29. 1-4 is a very much condensed Ephoran review of Persian affairs covering -100-

several years from before the death of Acoris till the dispatch of Iphicrates probably in 380/79 (41 1 n.). In carelessly abridging his source Diodorus has missed the point at which Acoris died and was succeeded by Nectanebos. All these events, Greek and Persian, were in Book 21 of the Histories. The story of the Egyptian campaign is resumed at 41. 1 and concluded in four chapters, 41-4 (374/3). Between 29. 4 and 41 there intervenes a series of Greek narratives. Section 34. 3 (τὰ μὲν οὐ+̑ν περὶ τὰς πεζὰς δυνάμεις . . . τω+̑ν δὲ ναυτικω+̑ν κτλ.) shows that Ephorus dealt first with the war on land before turning to the naval war. The King played a part in the Peace of 375, one of his motives being the wish to enlist mercenaries for the war against Egypt (38. 1). The narrative of the Peace concluded, therefore, Ephorus turned aside to relate the Persian campaign against Egypt. He then returned to Greek affairs. Again, all of these events were included in the same Ephoran Book, 21 or perhaps 22 (see pp. 94 f.). The prefaces: an original feature of Ephorus' Histories, as we have seen, was the preface which accompanied each of his thirty books. 265 Laqueur, in a detailed study of Diodorus' prefaces, 266 argued that those to Books 4-16 and 20 were adapted from Ephorus, while the rest were concocted by Diodorus himself after the manner of Ephorus. Kunz, 267 after subjecting the prefaces to a thorough analysis, concluded that the differences which the prefaces exhibit between themselves, and the fact that those to the later Books, 21 ff. (ignored by Laqueur), are not dissimilar to the others, but cannot be based on Ephorus, show that it is erroneous to consider all the prefaces as having a common source. All the same she fully endorsed Laqueur's view that Ephorus is the source for the prefaces to Books 11-15, and this is generally accepted. 268 The derivation of the prefaces to the following books, on the other hand, remains a matter of controversy and this is especially so for Book 16. However, as the comments on the kata genos arrangement of historical works and the praise of Philip are very probably Ephoran in origin, the substance of the preface can hardly be denied him (see also pp. 86 f.). The preface to Book 15 is a censure of the Spartans well in line ____________________ 265 16. 76. 5; Barber, 1935: 68 ff. 266 1911: 161 ff. 267 1935. 268 See e.g. Andrewes, Essays Starr, 190; Stylianou, 1991: 389 ff. -101with Diodorus' declared aim of praising the good and censuring the bad. After the restatement of this principle (1. 1), the epitimesis of Sparta begins in earnest (1. 2-5). The preface ends with the usual link-paragraph (1. 6). Diodorus' comments on the downfall of Sparta were no doubt commonplace by the first century BC. But that does not mean that they did not originate in the fourth century and with the school of Isocrates of which Ephorus was a famous product. 269 The censure of Sparta at 15. 1. 1-5 most appropriately applies to the period 386-362, that is, from the beginning of Spartan misdeeds following the King's Peace under which the Spartans had sworn to respect the autonomy of the cities, to the battle of Mantinea which confirmed the verdict of Leuctra and irrevocably deprived Sparta of her former leading position. The statement of Sparta's misbehaviour is repeated elsewhere, first and foremost at 15. 5. 1-3. By way of explaining the Spartan attack on Mantinea, and concerning the attitude and actions of the Spartans, Diodorus writes (I paraphrase): the Spartans soon found the terms of the Peace of Antalcidas unacceptable and returned to their

bad old ways of wishing to dominate Greece. They stirred up strife in the cities and this gave them the excuse to intervene. They first of all enslaved the weaker cities and then turned their attention to the more important ones. Now, it seems obvious that the above is not a description of Spartan policy and actions in any one year, but it applies rather to the whole period 386-379. The beginning of the Spartan drive was marked clearly enough by the attack on Mantinea. The gist of 15. 5. 1-3 is found also at 15. 9. 5 and 15. 19. 1 (cf. 15. 19. 4). What is one to make of these passages? The most acceptable explanation is that 15. 5. 1-3 is the remnant of a lengthy, rhetorical condemnation of Spartan misdeeds in the period 386-379 which Ephorus added to Book 20, probably in the form of a preface (cf. 5. 2-3 n.). No doubt he was glad to return to this theme every time an opportunity presented itself -- further infringements by Sparta of the autonomy clause -- and 15. 9. 5 and 15. 19. 1 may be all that is left of such occasions. The culmination of the Spartan policy of aggression was reached after the subjection of Olynthus and before the Theban coup d'état of winter 379/8 when Greece lay at Sparta's ____________________ 269 See n. 11. For the correspondence between D and Isocrates cf. Volquardsen, 1868: 49 ff.; Barber, 1935: 75 ff. and app. VIII. On Ephorus and the decline of Sparta see pp. 113 f. -102-

feet; and that was also the end of Book 20 of Ephorus, indicated, as has already been argued, by 15. 23. 3-5. Incorporated in the beginning of Diodorus Book 15, therefore, are the remnants of two Ephoran censures of Sparta; a general one applying to the period 386-362, and a particular one referring to the period 386-379. The second is very likely the preface to Ephorus Book 20. What of the first? There seems to be only one possible answer. Such a comprehensive censure of Sparta reasonably belongs at the beginning of Spartan misdeeds following the King's Peace, that is, in 386/5 and Ephorus Book 20. Thus Book 20 appears to have possessed a general preface which was relevant to the next five books also, 21-5, and pertinent to the years 386-362; and a particular one specific to Book 20 which castigated Spartan misbehaviour in the years 386-379 and which, following upon the general preface, introduced the narrative -- Mantinea. In Diodorus the two are separated by Persian narrative presumably because the chronographic entry on Mantinea came after that on the invasion of Cyprus. 270 There is an interesting analogy between the preface to Book 15 of Diodorus and that to Book 12. A comment on the unpredictability and mutability of the human condition, 12. 1. 1 is, like 15. 1. 1, what Diodorus exemplifies with the rest of the preface. The striking fact about this preface is that although Book 12 begins with the year 450/49, its preface would suit a book dealing with the pentecontaetia; cf. especially 12. 1. 4: ἀπὸ γὰρ τούτων τω+̑ν χρόνων (the Persian invasion) ἐπὶ ἔτη πεντήκοντα πολλὴν ἐπίδοσιν ἔλαβεν ἡ Ε+̔λλὰς πρὸς τὴν εὐδαιμονίαν. The artistic achievements of that time are referred to, Pheidias mentioned by name, as are also the great men of Athens, Miltiades, Themistocles, Aristides, Cimon, Myronides, 'and many others' for really 12. 1-2 is a eulogy of Athens, though not just the Athens of the pentecontaetia for at 12. 1. 5 the great Athenian philosophers are remarked, Socrates, Plato, and Aristotle, and of the orators, Pericles and Isocrates 'and his pupils'. It would appear, therefore, that 12. 1-2 is an incompetently abridged preface of Ephorus to the book ____________________ 270 Laqueur, 1911: 333 n. 2 would assign the original of D 15. 1 to Ephorus Book 25, but this cannot be; cf. 15. 1. 2, ἐπ+̕ ἐκείνους τοὺς χρόνους, ἐν οἱ+̑ς Λακεδαιμόνιοι περὶ Λευ+̑κτρα . . . ἡττηθέντες. Jacoby, FGH iic. 28, rather more plausibly, to Book 21 or Books 21-5. But Book 21, as we have seen, began with 378 and the Boeotian War, and I would consider it more probable, therefore, that the preface which D cannibalized introduced Books 20-5. -103-

which narrated the pentecontaetia. 271 But this preface clearly extolled Athenian achievements of both the fifth and fourth centuries down to the time of Ephorus, a pupil of Isocrates; naturally, as Athenian greatness was generally thought to have begun after the Persian Wars and as a direct consequence of that momentous event. 272 Like the preface to Book 20, the present preface was not meant to serve the book to which it was attached alone. 273 It would seem that Diodorus miscalculated with the arrangement of his material and was unable to fit the Persian Wars into one book as Ephorus presumably did. He was thus forced to break up the pentecontaetia. This is quite probably the reason why Book 11 lacks a preface: its beginning fell in the middle of an Ephoran book. 274 But Diodorus could not resist the appeal of the Ephoran preface to the book on the pentecontaetia; he abridged it and used it to head his own Book 12. All the same he was aware of its incongruity; hence the explanatory remark at 12. 2. 2: ἀλλὰ περὶ μὲν τούτων ἀκριβέστερον τὰ κατὲ μέρος ἀνεγράφαμεν ἐν δυσὶ βίβλοις, ταύτῃ τε καὶ τῃ+̑ πρὸ ταύτης. (ii) Ephorus: the sources For the history of the fourth century Ephorus used a multiplicity of sources, written (including the occasional inscription: FGH 70 FF122, 199) as well as oral. It should be noted at the outset that though certainly aware of the works of Xenophon he appears not to have used the Hellenica (to any significant degree at least; but see 23. 3-4 n.) and to have made only partial use of the Anabasis. 275 His use of Callisthenes' Hellenica on the other hand can hardly be in doubt. Completed by 335 the Hellenica (it covered the years 386-357) was thus available to Ephorus. 276 According to a quota____________________ 271 Probably Book 11. The lack of fragments makes the definition of Books 10, 11, and 13-15 highly problematical. 272 Thuc. 1. 118. 2 with HCT ad loc. and cf. Isocr. 6. 42 f. 273 Barber, 1935: 69 cannot be right.

274

Though Laqueur, 1911: 166, thinks that the preface to Book 11 has been lost. Book 13 carries no proper preface either, again it would seem because its beginning did not coincide with an Ephoran beginning. 275 Cf. Barber, 1935: 126 f. Westlake, Studies, 260 ff., thinks that the Anabasis was not used at all. 276 FGH iic. 24; Barber, 1935: 131 ff. For the probable dates of composition of the Histories see the text below. On Callisthenes' Hellenica in general cf. Pearson, 1960: 29 ff. -104tion from Porphyry in Eusebius ( FGH 70 T 17) Ephorus plagiarized Daimachus, Callisthenes, and Anaximenes. Ephorus' use, if not plagiarism, 277 of Callisthenes is corroborated by a comparison of a number of passages in Diodorus with fragments of Callisthenes or with passages in authors whose authority was Callisthenes. It can be shown that in addition to plain facts Ephorus was not averse to borrowing his source's diction, terminology and even, on occasion, historical interpretation (see 20. 2 n.; 43. 4 n.; 48. 1 n.; 50. 2-3 n.; 82. 5 n.; 82. 6 n.). The digression on Messenian history (66. 2-6 ) was probably inspired by a similar digression in Callisthenes. 278 But Ephorus used much else besides. In explaining the earthquake and comet of 373/2, for instance, he combined Callisthenes with Heraclides of Pontus and Democritus (48. 1 n.; 50. 2-3 n.). And the similarities and differences between Diodorus' account of the Mantinean campaign of 362 and those of our other sources (not Xenophon) are partly explained by the fact that Ephorus used material other than that derived from Callisthenes (82. 5-84. 2 n.). His use of Callisthenes is further indicated by a comparison of Diodorus' accounts of Pelopidas' actions in Thessaly and Macedon with the relevant chapters in Plutarch Pelopidas. There are substantial similarities between the two even in language (67. 3 n.; 67. 3-4 n.). Westlake 279 explains these as due to a common source employed by Ephorus and Plutarch; he cautiously suggests Callisthenes and the suggestion is plausible. It is at any rate difficult to offer a better candidate and Sordi, 280 therefore, accepts Callisthenes more confidently. The discrepancies between Diodorus and Plutarch on the other hand are easily explained by the fact that both Ephorus and Plutarch were able enough writers to supplement Callisthenes from other sources as well as introduce views of their own. Porphyry's statement that Ephorus drew material from Daimachus and Anaximenes is difficult to prove. We may accept that he used them for some part of the Histories, but further investigation is unfruitful. The first part of Anaximenes' Hellenica dealt with Greek and barbarian history from the very beginning down to 362 in only twelve books. Ephorus cannot, therefore, have ____________________ 277 Cf. Cavaignac, 1932: 156. 278 Cf. Barber, 1935: 132 f. 279 1939: 11 ff. 280 1958: 193 ff. -105extracted much detail from it (76. 4 n.; 89. 3 and n.). About Daimachus we know next to nothing. 281 We cannot even be certain that he dealt with fourth-century history at all (cf. 75. 2 n.). The assumption that Ephorus used the writings of the Boeotians Dionysodorus and Anaxis for his account of the Theban hegemony remains unfounded as the dates of these two historians escape us and virtually nothing survives of their works (95. 4 n.). The difficulty of pin-pointing such sources does not of course mean that they did not exist. We know that for the pentecontaetia and the Peloponnesian War he supplemented Thucydides in various ways, 282 and a great deal more information must have been available to him for the years after the King's Peace. The various local histories which grew in number as the fourth century wore on are an obvious candidate, especially those of Athens, the Atthides. The publication dates of the Atthides of Cleidemus, Androtion, and Phanodemus 283 make it possible for Ephorus to have used all three or at least that of Androtion, a fellow pupil of Isocrates, certainly for his later books. Little can be said about other local histories he must have consulted. 284 Ephorus himself wrote a local history of his home city, Cyme, entitled Ε+̔πιχώριος (70 FF1, 97-103), and it is evident that he possessed a considerable body of information about the history and traditions of Cyme (see 2. 2 n.; 18. 2-4 n.). Political pamphlets and essays would have been readily available to him, 285 and his own knowledge of the fourth century would have been augmented by the circle of Isocrates which included some of the most famous Athenian (and indeed Greek) generals and statesmen of the day. Whether he used Theopompus' Hellenica is a question which Jacoby leaves open, 286 but it is not totally impossible, especially if the Hellenica were indeed published before 335. 287 It is in any case certain that he used ____________________ 281 FGH65; J. Trevett, CQ 40 ( 1990), 416 f. 282 Cf. HCT iv. 68, 106. He clearly made much use of Hellanicus -- compare FGH 70 F117 with 4 F188 -- though not uncritically; cf. Strabo 8 C 366 (5. 5) = FGH 70 F118, where we see Ephorus using the pamphlet of King Pausanias of Sparta on the laws of Lycurgus to 'correct' Hellanicus. For writers before Thucydides see Dion. Hal. On Thue. 5. 283 FGH iiib. Suppl. 57 ff., 103, 173. 284 Jacoby has collected all the known local histories: FGH iii. 580 ff. Often no more than titles survive. Cf. his discussion of the genre, 1909: 109 ff. 285 FGH iic. 31. 286 Ibid. 287 Cf. Momigliano, 1935: 180 ff. -106the Hellenica Oxyrhynchia for the period 411-394 (or 387). Messenian history see 66. 2-6 n.

288

For the kind of information available to him on

For Ephorus' sources for western history see pp. 53 ff. The sources for eastern history present even more formidable problems of Quellenforschung. A line should be drawn between internal Persian history and events in the Mediterranean provinces of the Persian empire. Apart from occurring not far from Greece the latter invariably involved

large numbers of Greek mercenaries and it was consequently easier for a Greek historian to acquaint himself with their details. For example, thousands of Greek mercenaries participated in the struggle between Persia and Egypt; they were employed by both sides. 289 The campaign of 374/3 is related from Iphicrates' point of view. It reads in fact very much like an apologia (Iphicrates' own?) as well as a eulogy of the Athenian general (15. 41 ff.). Ephorus would not have had much trouble in collecting that kind of information in Athens. Hence the relative detail and reliability of the account. He drew some of his detail from Callisthenes (43. 4 n.), but he doubtless possessed other information too. Thousands of Greeks took part in Ochus' reconquest of Cyprus, Phoenicia, and Egypt in the 340s. Amongst the generals were Phocion of Athens, Lacrates of Thebes, and the brothers Mentor and Memnon of Rhodes. Diodorus' narrative is again detailed and it gives the impression of being accurate. It is very probable that Ephorus based his account not on any written sources (none perhaps was yet available), but on oral information. The Great Satrap Revolt is a little more difficult. While information on events in western Asia Minor could have been obtained without undue hardship, it is less easy to see how Ephorus could have acquainted himself with Datames' exploits in Cappadocia though we do hear that Datames employed mercenaries (91. 2 ff. But were they Greeks?) as did one of his adversaries, Autophradates, according to Nepos, Dat. 8. 2, who specifies them as Greek. 290 Even more difficult is the trial of Tiribazus (15. 10 f.). How could a Greek ____________________ 288 Depending on the date at which the Oxyrhynchus historian ended his work, which is not certain; cf. Bruce, 4; Breitenbach, RE Suppl. 12. 402. For a number of reasons (cf. p. 215 ) I have a slight preference for 394. 289 Many of these were Athenians; cf. Hicks and Hill, no. 122. For Greek mercenaries in the east in general cf. Parke 1933: 57 ff., 105 ff. 290 Cf. Parke, 1933: 107. -107historian learn, and learn accurately, of what went on at the Persiany court? We are told that Tiribazus appealed to 'all the Greeks present' (15. 10. 2). We are reminded of Clearchus' presence at the trial of Orontas before Cyrus in 401 ( X. An.1. 6. 5 ff). Should we then envisage a link, direct or indirect, between one of these Greeks and Ephorus? If Ephorus did not derive this kind of information orally then he must have used some other author. Jacoby 291 denies that he used the Persica of Dinon of Colophon. He followed Ctesias and not Dinon (or Xenophon) in giving Artaxerxes 400,000 troops at Cunaxa (14. 22. 2; Plut. Artox. 13. 3), and his account of the death of Cyrus seems to be a combination of Xenophon, Ctesias, and some other source which brought into focus the role of Tissaphernes (14. 23 f. A local Cymaean tradition about Tissaphernes?). Dinon's epic tale was not used. Dinon had said that Conon had not been put to death by the Persians, but had escaped, perhaps aided by Tiribazus ( FGH690 F18 = Nepos, Conon9. 5. 2). We do not know what Ephorus' version was, but Diodorus makes the comment that Conon was arrested and punished, whatever that implies (14. 85. 4; 15. 43. 5). 292 Dinon had said that Ochus sacrificed the sacred bull Apis after his reconquest of Egypt ( FGH690 F21). Diodorus does not mention this. Would he have left out such a glaring example of sacrilege if he had found it in Ephorus? Still, none of the above proves that Ephorus did not use Dinon elsewhere. It is a fact of some significance that the story of Tiribazus, specifically the Cadusian war and the trial, can only be followed by a combination of Diodorus and Plutarch's Artoxerxes. Now, Tiribazus plays an important part in Plutarch. He figures in chapter 5 (at a hunt), in chapter 7 (he urges the King to fight in 401), in chapter 10 (he provides the King with another horse at Cunaxa), in chapter 25 (he extricates the King from Cadusia), and in chapters 27 ff. (he conspires against the King and is killed). Chapter 10 is Dinon's version of the death of Cyrus, as Plutarch tells us. It is probable, therefore, that Tiribazus' prominent role in the Artoxerxes is due to Dinon whose Persica was used extensively by Plutarch for this life. He is cited no less than seven times. It is equally probable that Dinon is behind the romantic elements in Diodorus 15. 10 f. -- the trial of Tiribazus. Some of ____________________ 291 FGH iic. 31. 292 Cf. Swoboda, RE xi. 1332. -108-

these details are immediately suspicious: the skins of the corrupt judges stretched on the judicial benches (10. 1 - obviously an adaptation of Herod. 5.25); Tiribazus' slaying of the two lions and rescue of the King (10. 3); the nicely balanced verdicts of the three judges. (11. 1). This is precisely the sort of romantic fiction we meet in Plutarch too. This does not mean that the entire account of the trial should be dismissed. 293 We may accept the kernel of 15. 10f. as factual: that Tiribazus was tried after some delay occasioned by the Cadusian campaign, acquitted, and reinstated to his former honours while Orontes suffered an eclipse at the conclusion of the Cypriot War. The unsuccessful negotiations between Euagoras and Tiribazus, Orontes' charges against the latter, and the terms of the settlement would gradually have become known in Cyprus and Greece. It would not have been difficult for an imaginative mind to reconstruct the trial adding the romantic touches we have noted. Ephorus may also have used the Persica of his compatriot Heraclides though this was a very brief work (Diog. Laert. 5. 94) and it seems to have been based on Ctesias and Dinon. 294 The precise dates of Dinon and Heraclides are a problem. Jacoby raises a question over the respective floruit of c.360-330 and c.350 which he gives them, 295 but that cannot exclude the possibility that Ephorus used them (see below for his dates). In conclusion it is worth noticing that apart from the well-known Persica there was a considerable eastern 'tradition' available in antiquity ( FGH696 Anhang) written as well as oral. When Xenophon (An. 1. 8. 29) gives alternative versions of Artapates' death (καὶ οἱ μέν φασι . . . οἱ δ+̕) he is not referring to written, but to oral sources; that was what he had heard at the time or later; cf. An. 1. 2. 12,14; 1. 4. 5; 1. 6. 11; 1. 10. 2; 4. 3. 2. We may note especially 2. 1. 17: οἰ+̑σθα δὲ ὅτι ἀνάγκη λέγεσθαι ἐν τῃ+̑ Ε+̔λλάδι ἃ ἂν συμβουλεύσῃς. The Greeks who served in Persia disseminated much information, true and false. ____________________ 293 Stevenson, Achaemenid History, ii. 3 1 f., considers the trial in D (the whole of which she assigns to Dinon) to

be 'pure invention'. Drews, 1973: 202 n. 127. 295 FGH689 and 690. 294

-109(iii) Ephorus: attitude to states and individuals The didactic, ethical nature of Ephorus' work (pp. 6 ff.) inevitably conditioned his attitude to states and individuals and his interpretation of the wider political background of the fourth century. If we had some idea about when approximately he wrote we might be able to deduce something more not only about his purpose in writing a universal history, but also about the slant he gave to it. There are a number of indications which, when taken together with the little information we have about his life and circumstances, point to the late 330s and 320s as the time of composition: FGH 70 F223 (= Clem. Alex. Strom. 1. 139. 3): Ephorus reckoned 735 years from the return of the Heraclids (his starting point: 4. 1. 3; 16. 76. 5) to the crossing of Alexander into Asia, i.e. 335/4. Considering that no Ephoran narrative went beyond 341/0 and that some even stopped with 357/6, the most likely place for such a calculation was the general preface where Ephorus presumably announced his aim and explained his procedure. (16. 76. 5 seems inconsistent with F223, but see 15. 1. 3 n.) The very fact that Ephorus perceived the period from the return of the Heraclids to the crossing of Alexander into Asia as an era shows that he was writing at a time when Alexander's actions in Asia had become irreversible, i.e. after c. 330. Indeed, it would appear from this that his original intention had been to carry his history down to this epochal point. 296 FGH 70 F 119 (= Strabo 9 C. 401 (2. 2)): in the opinion of Ephorus the Thebans were not suited for the hegemony of Greece as was shown by the fact that they lost it, having only just gained it, immediately Epaminondas was dead. This part of the Histories, 297 therefore, was written not just after 362 and the battle of Mantinea, when the political and military situation was as yet unclear, but after 346, when Philip and not Thebes concluded the ____________________ 296 In F223 Clement is not saying that Ephorus 'brought his work down . . . to 334' as Barber, 1935: 171, wrongly supposes. That of course would be incorrect. Barber's scepticism about the fragment (1935: 8f.) seems arbitrary. Clement obviously did not invent the information and even if the numeral were suspected of corruption, the statement still stands, 'until the year of Euaenetus in which, they say, Alexander crossed to Asia'. But it is not in fact difficult to reconcile Clement with D: 1. 3 n. 297 Whether from Book 2, as Jacoby, FGH IIC. 24, or from one of Books 22-5, as Milns, Vindex Humanitatis, 48. -110-

Phocian War, and indeed after 338 and the battle of Chaeronea even.

298

FGH 70 F217 (=Tertullian, De an. 46): Philip's dream which signified the future greatness of Alexander. Book 27, therefore, in which presumably the dream was related (pp. 95 ff.), must have been written after c. 330. FGH 70 F221 (= Plut. Tim. 4 and Clem. Alex. Strom. 1. 135. 1): it would appear that Ephorus was aware of Timoleon's later achievement in Sicily when writing about Corinthian politics of the mid-360s, probably in Book 24 (cf. F84). That book, therefore, cannot have been written earlier that the early 330s. Several passages in Diodorus are also relevant: 15. 88. 4: not only did Thebes lose the hegemony of Greece following the death of Epaminondas, but because of the unwisdom of her leaders she was in the end destroyed. This bears close affinity to F119 (see above). It comes from the epainos on Epaminondas and it can be assigned to Ephorus Book 25 with a fair degree of confidence: post-335, therefore. 16. 1. 5: there is no reason at all why 16. 1. 3-5 should not be assigned to Ephorus, a eulogistic assessment of Philip which, as was Ephorus' way (see p. 123 ), anticipated the narration of the king's praxeis. Section 5 points to a post331/o date for Book 26 or 27 (pp. 95 ff.). 16. 14. 2: the favour Philip carried with the Thessalians stood Alexander in good stead -- if this was taken from Book 27 then that book was subsequent to 330. 16. 56. 7: even if ἔνιοι δὲ τω+̑ν συγγραφέων were taken to be an addition by Diodorus himself, it seems most likely that it was occasioned by Demophilus' comparison of the amount of treasure looted from Delphi by the Phocians with that captured by Alexander in Persia, that is, again a date post-330 is indicated for Book 30. 16. 64. 1: the fate of Athens at the hands of Antipater in 322 is clearly implied: Book 30 was, therefore, written after that date and it is reasonable to believe that Demophilus undertook the completion of the Histories very soon after Ephorus stopped writing. ____________________ 298 As Milns, Vindex Humanitatis, 48, suggests.

-111-

Finally, there is the fact that for Books 20 ff. Ephorus had Callisthenes' Hellenica available to him (see p. 104 ), a work written between c.343 and 335. 299 By contrast with all these termini post quos there is not a single terminus ante quem. Barber 300 proposes F121 (= Strabo 9 C 426 f. (4. 7)) as one. There Ephorus 'in speaking of Naupactus, puts it under Locrian control: after 338 It fell into the hands of the Aetolians'. Barber, therefore, concludes that the book from which the fragment derives (either 2 or 3) must have been written before that date. 301 But nothing of the sort can be deduced from F121 and Barber's construction, 302 whereby Ephorus began writing in the early 350s and, producing about one book a year, stopped writing c.330, thereby becomes impossible to sustain. Similarly Jacoby considers Ephorus to have been a slow worker who wrote his twenty-nine books (the thirtieth being that of his son) between c. 350 and c. 330. 303 The evidence looked at above seems rather to favour the view that the Histories were planned and composed in the late 330s and 320s. 304 If so, Ephorus' stimulus for undertaking the composition of a universal history should be sought in the unification of Greece under Macedon and the panhellenic war against Persia planned by Philip and launched by Alexander. Hence, as has already been suggested, the 735 years calculated by Ephorus from the return of the Heraclids until the crossing of Alexander, a descendant of Heracles on his father's side, into Asia (F223) may indeed represent the intended (and unfulfilled) timespan of the Histories. Neither Philip, nor Alexander, nor proMacedonian men of letters like Isocrates, ever tired of stressing the Heraclid ancestry of the Macedonian royal house. Writing late in the fourth century Ephorus could look back over the fifth and fourth centuries which had seen both the glory and the shame of the city-states of Greece. Unlike Xenophon who gave up in despair at the reigning chaos, Ephorus sought to explain and by explaining to instruct: virtuous and upright conduct in public life, whether within a state or in its relations with other states, ____________________ 299 Jacoby, FGH iic. 24. 300 Following Jacoby, ibid. 301 1935: 12. 302 1935: 12f. 303 FGH iie. 24 f. 304 304 Niese, 1909: 170 ff., is thus on the right lines though he goes too far in proposing that the whole work may not have been written till after Alexander's death. -112-

engendered by education (paideza or agoge) and respect for the gods (eusebeta); that was his entirely conventional solution, one shared by most of his contemporaries. We have already observed the central role that education and piety played in Ephorus (pp. 10 ff.). The case of the rise and fall of Sparta furnished him with his best example and it is evident that he gave it considerable space. The story, as recovered from Diodorus and the fragments, goes like this: the rigorous Lycurgan agoge, which put the emphasis on manliness (ἀνδρεία) and concord (ὁμόνοια), made Sparta the leading state in Greece. These twin virtues were achieved through a moderate and frugal way of living. So long as the agoge was maintained Sparta retained her hegemonic position. The hegemony was lost and Sparta herself almost destroyed when the Lycurgan laws were set aside, one by one, and the Spartans chose instead a life. of luxury (τρυφή) and laziness (ῥαθυμία), having allowed themselves to be corrupted by coined money and wealth, the direct result of the acquisition of the arche of the sea. The first chance of acquiring the hegemony of the sea in the 470s was wisely not taken up. Seventy years later, however, when Athens collapsed, her Aegean dominions passed on to Sparta. Not only did this corrupt the state internally, but it led to a grossly arrogant and tyrannous conduct where her allies were concerned. The widespread hatred thus generated proved Sparta's undoing. 305 Not surprisingly, Isocrates had earlier given the very same explanation (On the Peace 95 ff.): seduced by empire the Spartans abandoned their ancestral customs and became unjust, indolent, lawless, money-loving, contemptuous of their allies, greedy for the possessions of others, and unconcerned about oaths and the terms of treaties, that is, impious. Their crimes against their fellow ____________________ 305 Cf. 7. 12; 11. 50; 14. 2. 1; 14. 10; 15. 1. 2-4; 15. 5; 15. 19. 4; etc.; Plut. Lys. 17; FGH 70 F149 = Strabo 10 C 480 (4. 16); Polyb. 6. 46. 6; 48, 2-5 (almost certainly from Ephorus: Walbank, HCP i. 732, 734). I fall to see the basis for Sacks' assertion that Ephorus gave moral decadence alone as the cause of Sparta's downfall, and that the mistreatment of the allies, which we find in D as a reason, is D's own explanation: 1990: 49. In fact, Ephorus must have given both reasons and indeed they go together, τρυφή leading to tyrannical conduct, as D tells of Pausanias, the victor of Plataea (11. 44. 3-6; 46), comments which are undoubtedly derived from Ephorus. Pausanias' fate was seen as a harbinger of Sparta's own. And see the view of Isocrates in the text below. The fact that Ephorus may (we do not know) have given hedonism and luxurious living alone as the reason for the decline of Miletus ( FGH70 F183; cf. D 8. 20, of which Sacks, 1990: 48f., misses the point) is no argument that that is as far as he went in the case of Sparta also. -113-

Greeks were many and their misconduct did not cease until they had brought upon themselves the defeat at Leuctra. The opinion of some of his contemporaries notwithstanding, Isocrates was convinced that it was not the disaster at Leuctra that was responsible for the downfall of Sparta, but that both resulted from the hatred of allies who had been mistreated for many years ( ibid. 100 ; cf. also Panath. 56). Xenophon, in his old age at least (H. 5.

4. 1; 6. 4. 2f.; Lac. Pol. 14), and Theopompus (ap. Plut. Lys. 17) held views not dissimilar to these.

306

It is often said that Ephorus was strongly biased in favour of Athens and against Sparta while he viewed Thebes impartially or even with admiration. 307 An examination of the fragments and Diodorus Books 11-16 does not quite support this belief. What emerges is that Ephorus' main bias, as has been argued, was in favour of virtue whoever exhibited it, state or individual. This apart he was, as befitted a pupil of Isocrates, an ardent admirer of Athens and a panhellenist. Indeed, a high regard for the past glories of Athens and panhellenism were to some extent identified in Isocratean circles. 308 Ephorus' panhellenism is best shown by his treatment of nonGreeks when in contact with Greeks, especially on campaign. Witness, for example, 11. 77: the Egyptians appear in a bad light vis-O-vis their Greek allies (Athenians in this instance). Noteworthy also is the emphasis he laid on the importance of Greek soldiers to Persia and Egypt; cf. 15. 29; 15. 38. 1; 15. 41-4: Pharnabazus failed because he did not follow the advice of ____________________ 306 Tuplin's important study of Hellenica 2. 3. 11-7. 5. 27 ( 1993) is a valuable reminder that, lover of Sparta and ardent admirer of Agesilaus though he undoubtedly was, the mature Xenophon entertained grave misgivings about Spartan foreign policy, not least about the conclusion of the King's Peace and the way this was used by Sparta in pursuit of hegemony. Nevertheless, in my view, Tuplin goes too far in maintaining that Xenophon shows no bias whatever in favour of Sparta in Hellenica, that he does not, in other words, write as a philolacon. This seems to be rather an extreme position, not fully supported by his detailed analysis of the text. On the contrary, what Tuplin succeeds in doing, it seems to me, what is in any case obvious from the text itself, is that Xenophon was a disappointed philolacon, but a philolacon none the less. Instances of this are noted in the commentary wherever appropriate. At the same time attention is drawn to the difference in viewpoint of Xenophon and D (Ephorus) with regard to Athens, Sparta, and Thebes. See e.g. at 19. 3n.; 33. 56 n.; 34. 1-2 n.; 34. 2nn.; 38. 1n.; 47. 5-6n.; 55-56n.; 63. 1-2n. 307 Cf. Barber, 1935: 84ff.; Westlake, 1939: 16ff.; Shrimpton, 1971: 310ff. 308 Cf. Jaeger, Paideza, iii. 71 ff. -114Iphicrates (41. 2 n.; 43. 1-2 n.); 15. 92 f. (92. 3 n.): Tachos came to grief because he ignored the counsel of Agesilaus; 16. 44ff. Sections 16. 47. 7-48. 2 are particularly illuminating: King Nectanebos met with disaster because he allowed his earlier success against Persia to go to his head, not realizing that he owed it all to Diophantus the Athenian and Lamius the Spartan; and 16. 49. 2-6 and 49. 7-50. 6 are interesting comments on the affinity between Greek mercenaries serving different barbarian masters. 309 The ethical purpose of the Histories necessitated a degree of detachment and, his admiration for Athens apart, Ephorus could be termed apolitical. The struggle between democrats and oligarchs was as intense in the fourth century as in the fifth, but Ephorus remained uncommitted. Democrats were censured no less than oligarchs; cf. 15. 5. 2 and 40. 1-2: the behaviour of the democratic parties was far from praiseworthy. Ephorus' deep admiration for Athens is obvious; when praise was due it was invariably bestowed while censure, though by no means absent from the pages of Diodorus, was very much the exception. This was no political bias as such, but rather the esteem of a man of letters for the cultural and artistic centre of Greece. 310 Indeed, Ephorus' sentiments are voiced often enough by modern scholars too. Such a viewpoint inevitably involved a less than fair treatment of Sparta, Athens' great enemy for three generations. Ephorus genuinely considered Athens a virtuous state which had benefited Greece. Sparta, on the other hand, especially at the height of her power, he saw as a self-centred state which had, with one notable exception, harmed Greece. 311 His use of pro-Athenian sources (above p. 106 ) would have encouraged such an attitude. There are instances where his account was falsified in Athens' favour. The falsification was not his, but one may censure Ephorus for not being sufficiently astute, or not caring enough, to question his sources. One suspects that as such pro-Athenian accounts coincided with his own beliefs he was content to accept them. Many examples can be cited. The Athenian contribution to the fight against Persia is more than emphasized as is also Spartan selfishness and jealousy; cf. 11. 15. 3; 11. 27. 2-28. 1; 11. 30. 2ff. ____________________ 309 This was a point of view shared by Isocrates; cf. 4. 125, 134ff.; 5. 90ff. 310 Cf. Jacoby, 1049: 130. 311 Cf. Isocr, 4. 103 ff,; 12. 53 ff. Again, this is an opinion many a modern scholar would agree with; cf. e.g. Forrest, 1980: esp. 95, 152f. -115-

(11. 30. 6 οἱ περὶ τὸν Παυσανίαν καὶ Α+̔ριστείδην!); 11. 33. 1. The pentecontaetia was the Athenian golden age and Ephorus' narrative was suitably eulogistic. Overcoming Spartan suspicions and opposition the Athenians fortified their city (11. 39 f.) and under the leadership of their great men Themistocles and Aristides founded their naval hegemony (11. 41 ff.). Significantly Ephorus made the most of the Spartan share in the downfall of Themistocles and played down the Athenian part (11. 54ff.). The Spartans acted out of spite against Athens (11. 54. 2; 55. 5f.). Luckily we possess Thucydides' version (1. 135 ff.). We may compare especially 11. 56. 2-4 (the Spartans chased Themistocles from Greece) with Thuc. 1. 136-137. 2 (the Spartans and Athenians were anxious to capture Themistocles). In the 450s Athens was fighting a war on two fronts and Ephorus was not slow to praise; cf. 11. 77. 3 ff.; 11. 80. 6 (the battle of Tanagra was ἀμφίδοξος. This certainly reflects Athenian traditions; cf. Plato, Menex. 242a. Thuc. 1. 108. 1, on the other hand, says that the Spartans were victorious); 11. 82; 11. 85. 2; 12. 1-2. 1 (in effect a eulogy of Athens; see p. 103 ); 12. 2. 1 (the Persians are forced to leave the Greeks of Asia free κατὰ συνθήκας); 12. 3. 1 ( Athens fights Persia ὑπὲρ τω+̑ν κατὰ τὴν Α+̔σίαν Ε+̔λλήνων); 12. 4. 4-6 (the famous Peace of Callias of which fourth-century Athenians were so proud); 12. 26. 2 (the Peace of Callias contrasted with that made by Sparta in 387/6; cf. Isocr. 4. 115 ff.).

For the Peloponnesian War Ephorus relied heavily on Thucydides and perhaps as a result the corresponding narrative in Diodorus is less eulogistic. The Plataean incident which led to the outbreak of war seems to have been given a proPlataean slant. It is said that the Thebans killed and captured many people outside Plataea, ἅπασα δὲ ἡ χώρα ταραχη+̑ς καὶ διαρπαγη+̑ς ἔγεμεν (12. 41. 7). Thuc. 2. 5. 7, however, is explicit that the Thebans departed from Plataean territory οὐδὲν ἀδικήσαντες. Worse, according to Diodorus the Plataeans gave up their Theban prisoners as they had promised; but we know from Thuc. 2. 5. 7 that in fact they killed them. The assembly of the Syracusan demos to decide the fate of the Athenian captives in 413 presented Ephorus with an opportunity to eulogize Athens. The speech of Nicolaus 312 is pure Athenian ____________________ 312 D extracted the speech of Nicolaus from Ephorus (see p. 60 ). Of course he added his own elements, including the singularly inept use of the term Πελοποννησιακὸς πόλεμος at 13. 24. 2. His source perhaps had written κατὰ τὸν ὑποκείμενον πόλεμον. This is not to deny that Ephorus used Πελοποννησιακὸς πόλεμος. -116-

panegyric in places (cf. 13. 25. 1-3; 26. 2-27. 2) and in some of its statements it strongly agrees with Ephoran fragments and passages of Isocrates; cf. 26. 2-3 with FF31 and 42; 26. 2-3 with Isocr. 4. 28-30, 38 ff., 54ff.; 27. 1. with Isocr. 4.50 ( Athens the school of Hellas). The first half of the fourth century belonged first to Sparta and then to Thebes, and Athens received less attention, but Ephorus praised whenever he could. The Athenian demos is twice commended for its moral qualities (15. 46. 6; 63. 2). More striking was Ephorus' assertion, which probably originated in the circle of Isocrates, that Athens had in the 370s saved Greece from a conspiracy of Sparta, Dionysius, and Persia (47. 2-4, 7n.). Moreover, it is to be noted that many events are related from the Athenian point of view and in a way calculated to enhance Athens and Athenians; cf. 15. 32. 5-33. 4 ( Chabrias saves the Thebans in 378); 47. 5f. ( Ctesicles saves Corcyra in 373/2); 69. 2ff. ( Chabrias saves Corinth in 368). Explicit criticism of Athens there is, but very little. In 11.70. 1-4 it is explained that heavy-handed Athenian behaviour caused revolts. In 12. 75. 2, 4 (cf. 15. 38. 4) both Athens and Sparta are said to wish to enslave Greece and they are accused of pleonexia, though no comment is offered shortly afterwards (12. 76. 3) on the destruction of Scione. We do not unfortunately know what Ephorus said about the treatment meted out to Melos for Diodorus appears to have dealt with the relevant Ephoran narrative in one sentence (12. 80. 5). In 13. 102. 5 the Athenian demos is censured for unjustly executing the generals who won Arginusae. At 15. 23. 4 and 15. 29. 8 it is admitted in retrospect that Athens had been unpopular because of the cleruchies. It is very likely that Ephorus had commented on this earlier and that Diodorus overlooked it. And at 16. 57; 64. 1 Athens is criticized, admittedly by Demophilus (but see p. 12 ), for impiety. By contrast Sparta is severely and repeatedly criticized for her selfish policies which in the end brought about her downfall. In addition to the instances cited above, cf. 12. 67. 3-5; 14. 2. 1; 14. ____________________ The term occurs in at least nine other places in Books 12-15: 12. 37. 2; 12. 74. 6; 12. 75. 1; 12. 81. 5; 13. 107. 5; 14. 2. 4; 14. 10. 1; 14. 13. 1; 15. 35. 2. Of these, with the exception of 12. 81. 5, the first six cases seem to be from the chronographic source or to be inspired by it. Section 12. 81. 5 and the last three instances, on the other hand, may well be Ephoran. -1176; 14. 10. 1-3; 14. 44. 2; 14. 70. 2f. (the Spartans arranged Greece to their advantage and then strengthened the tyranny of Dionysius). It is to be noted that Lysander was acting under orders when he set up the harmosts, decarchies and other oligarchies (14. 13. 1). Ephorus, therefore, had probably argued against the view that Lysander alone was to blame; 14. 11. 2; 14. 19. 4; cf. 14. 21. 2 (the Spartans helped Cyrus for their own selfish interests. X. H. 3. 1. 1 ascribes the Spartan involvement to a sense of honour); 14. 17. 6, 1 1; 14. 70. 1-3; 15. 1; 15. 5; 15. 19. 1, 4; 15. 20. 2; 15. 23. 5; 15. 28. 2; 15. 29. 6. Sparta is not always censured, however, and this proves that Ephorus was not anti-Spartan as such. What he castigated was Spartan policy vis-à-vis Greece and, as he considered that Athens generally acted in the interest of Greece, Spartan opposition to Athens. When nevertheless Sparta behaved virtuously then he praised; cf. 11. 4. 4; 1 1. 9. 2; 1 1. 1 1; 11. 74. 5; 15. 30. 3-4; 15. 65. 1-5. Scathing enough prior to 387, criticism of Sparta, which naturally reaches a crescendo in the account of the years 387/6-379, subsides after this time and essentially stops with Leuctra. The one exception after that is 16. 57. 1, 4, the impiety of Sparta in connection with the Phocian War, as Demophilus saw it in Book 30. 313 As far as Thebes is concerned Ephorus appears uncommitted. The Thebans were men of brawn rather than brains (20. 1 n.). If we perceive a sympathy in the pages of Diodorus that is because Thebes was the victim of imperialist Sparta. He obviously admired Pelopidas and Epaminondas, especially the latter whose paideia enabled Thebes to gain the hegemony of Greece for a while (pp. 10 f. and below). The treatment of individuals followed the same lines. The decisive factor was not their nationality, but their character and conduct. Good men were praised, bad censured, the greatest amongst them with an epainos or Psogos at the end of their lives (pp. 5 ff). Amongst Spartans, for example, Brasidas is treated favourably (12. 43. 2 f.; 62. 1 ff.; 67 f.; 74) as is Callicratidas (13. 76. 2; 98. 1 f.; 99. 5). There is no condemnation of Lysander himself which may seem surprising, but then Ephorus chose to see him as the tool of the Spartan government (see above). Amongst ____________________ 313 I see no inconsistency in Ephorus' treatment of Sparta or a change of tone so severe as to require Westlake's assumption of the clumsy use of different sources by Ephorus: 1986: 263 ff.

-118-

Athenians the Thirty are condemned (14. 2. 1) as is Chares (15. 95. 3), while Theramenes and Timotheus are praised (14. 3. 6-5. 4; 15. 36. 6). Amongst Thessalians Jason is admired (15. 57. 2; 60. 1f., 5) while Alexander is censured (15. 61. 2f.). For the Siceliots see pp. 52 ff. A closer look at the treatment of some of the leading figures of the period covered by Diodorus Book 15 agrees with the above analysis of the Ephoran viewpoint. We may begin with Agesilaus. We have seen that Sparta was roundly condemned for her policies in Greece after the King's Peace. ϕύσεί

ϕίλαρχοῦυτες καì πολεμικοì ταîς αόρέσεσιυ ὄυτεσ (15. 5. 1), the Spartans exploited the Peace and enslaved Greece

anew. The motive force behind this policy is soon revealed: Agesilaus, being by nature δραστικòςb καì ϕιλοπóλεμος in contrast to his colleague Agesipolis, aimed at the δυύαστεíα Greece (15. 19. 4). Yet Agesilaus is not explicitly condemned. It would appear that Ephorus was in two minds about him. 314 On the one hand he abhorred his policies in Greece, while on the other he could not but admire his energy and considerable military ability. We may note especially 15. 31. 3f. where Agesilaus' renown as a soldier is acknowledged, and 88. 2 where in the epainos on Epaminondas, he is the only Spartan mentioned; cf. also 92. 2. The account of Agesilaus' campaigns in Asia Minor is instructive. The king was clearly striking a blow for Greece then. Yet Athens was ranged against Sparta at the time and an Athenian, Conon, the father of none other than Timotheus, employed Persian money and ships to smash Sparta's naval power. The defeated and killed Spartan admiral was Agesilaus' brotherin-law (X. H. 3. 4. 29). While Ephorus, therefore, followed the example of his source, the Oxyrhynchus historian, and gave a good account of the military achievements of Agesilaus, he appears to have been unwilling to see the campaigns as a Greek national effort. The fact that the Oxyrhynchus historian (P) repeatedly refers to Agesilaus' soldiers as 'Greeks' (11. 3ff.; 12. 1ff.; 13. 1; 14. ____________________ 314 As Plutarch was who both praises and censures Agesilaus. The praise is for actions in the interest of Greece; the censure, for his oppressive policy vis-à-vis his fellow Greeks, in particular the Thebans. It is unnecessary, therefore, to postulate, at any rate on this basis alone, two sources in the Agesilaus, one pro- and the other antiSpartan (as Westlake, 1939: 20). We may compare Isocrates' ambivalent attitude to Sparta; cf. Bradford Welles, Studies Caplan, 23. With even less justification Westlake perceives glaring inconsistencies in Ephorus' treatment of Agesilaus, as in his treatment of Sparta (see previous n.), which he explains in the same way. -119-

2; 21. 2; 22. 3) while Diodorus calls them 'Lacedaemonians' (14. 80. 1, 5) is surely of some significance. 315 It should be said in this context that Ephorus, reacting perhaps to popular Athenian opinion (re), did not consider Cnidus as either a Greek or an Athenian victory (15. 35. 2). The great men of Thebes were more straightforward. The build-up of Theban power in the 370s and 360s was a major theme in Ephorus, and he explained the short-lived Theban hegemony as due to the brilliance of Epaminondas. The way in which Epaminondas' arete and paideia overcame all difficulties on the road to hegemony is stressed several times in Diodorus. At 38. 3 the Thebans are encouraged by Epaminondas to stand fast against Spartan pressure to dissolve their Confederacy, διà τῆς ἰδíας ἀρετῆς ἐμποιοῦύτος ϕρóύύημα τοîς πολíταις ; cf. 39. 2f.; 50. 6; 52. 2-7 (his paideia overcomes the superstitious fears of the masses); 53. 4 and 54. 4 (he does the same with the soldiers); 55f. (he wins Leuctra for Thebes). The hegemony lasted for as long as Epaminondas was alive to guide it wisely and humanely; cf. 57. 1 (he saves Orchomenus from destruction); 66 (he resurrects Messenia); 69. 1 n. (he saves the Boeotian exiles); 72. 1 (he spares the Spartans); 78. 4-79. 2 (he persuades the Boeotians to challenge Athens at sea); and cf. his epainos, 88, 316 especially 88. 4. Pelopidas lacked Epaminondas' paideia, but he was brave and patriotic. He is seen as the imitator of Epaminondas who tried to do in the north what Epaminondas did in the Peloponnese (71. 2). It should be stressed at this point that Ephorus did not ascribe panhellenic aims to the two Thebans. To Ephorus Thebes was simply the successor of Sparta as hegemon in Greece (cf. 88. 4; F119), but with this important difference: she treated the Greeks rather better, at least when guided by Epaminondas and Pelopidas. Otherwise Ephorus made it very clear that what Pelopidas and Epaminondas did in Thessaly and the Peloponnese they did as patriotic Thebans; see 67. 3f.; 71. 2; 79. 1f.; 81. 1. The same is said by Nepos, Pel. 51, who used Ephorus directly or indirectly, while Isocrates, naturally, puts the matter much more strongly: ἠνώάλονύ ____________________ 315 Pace Walker, 1913: 55. P occasionally refers to them as 'Peloponnesians'; but 'Peloponnesians and their allies' (21. 2; 22. 1), whereas D uses just 'Lacedaemonians' (once 'Greeks':14. 80. 6). 316 A more elaborate affair than normally (as preserved in D at least), as Vial, 111 n. 1, rightly observes. -120-

μέν ταîς πóλεσι παîς πóλεσ2o παîς ἐπóλμων καπαδονλου̑σϑαι (5. 53). Diodorus (Ephorus) is very different in this respect from Plutarch. The latter (he was of course a Boeotian) emphasizes that Pelopidas' intervention in Thessaly was disinterested, his motive being love of freedom (Pel.31. 6; 33. 9; 34. 7). 317 The case of Magnesia and Achaea Phthiotis in 364 is a striking example of this difference between Ephorus and Plutarch. According to Diodorus 80. 6 these areas were handed over to Boeotia (see n. ad loc.). Plut. Pel. 35. 3, on the other hand, says that they were set free. (iv) Ephorus: merits and defects

Ephorus has not found much favour with modern scholars. Schwartz 318 and Laqueur 319 thought very little of him while Jacoby treated him as a mere compiler, albeit one on a large scale -- a writer of little originality except for his oikonomia. 320 More recently Schepens, in a notable study, has sought to defend him against some of the more serious accusations levelled against him. 321 On the whole his picture of Ephorus seems more balanced and more believable than the usual one. As far as flaws go there is first the obvious criticism that a didactic historian will not be an entirely objective one and there is considerable truth in this. We have already seen that the proAthenian sources he used led him astray sometimes, and Ephorus could be charged with not being critical enough of his sources (p. 115 and 12. 1n.). The ethical aim he set himself was largely to blame for this. If the version of an event he met in a source did not contradict the moral point he wanted to make, and it might even be that it enhanced it, Ephorus was inclined to accept it and search no further. This is reminiscent of Isocrates' use of history as a servant of paideia. 322 On the other hand, there is no evidence that ____________________ 317 Whether these sentiments in Plutarch should be traced back to Callisthenes is a thorny question. Sordi, 1958: 194ff., vigorously argues that they should: Callisthenes saw Epaminondas and Pelopidas as champions of freedom and panhellenism. Likewise, Momigliano, 1934a: 196. This is likely, though it cannot be proved. But there are no grounds for ascribing pro-Theban bias as such to Callisthenes. 318 RE vi.1ff. 319 1911: 161ff., 321ff. 320 FGH iic. 22ff. 321 1977: 97ff. 322 See the references cited in n.13. -121he falsified facts; he merely opted for the version which exemplified virtuous conduct or suited his particular viewpoint best. 323 The one definite, and understandable, exception concerns his undoubted inflation of numbers for barbarian armaments (see below). The famous Peace of Callias is a case in point. Unlike Callisthenes who interpreted it as a tacit understanding between Athens and Persia and not as a formal peace treaty (FGH 124 F16), and Theopompus who dismissed it as another instance of Athenian falsification of history (FGH 115 FF153, 154), Ephorus wholeheartedly accepted the full Athenian version of a formal treaty which, unlike the Spartan Peace of Antalcidas in 387/6, left Greek Asia free and rid the Aegean of Persian warships (12. 3. 1; 12. 4. 4-6; 12. 26. 2). Such a version satisfied both his love of Athens and his panhellenism. Luckily for him he was right about this and Callisthenes and Theopompus wrong. He appears equally correct, against Aristotle ( Poetics 1459a) and his modern adherents, 324 that Persia and Carthage were in close touch in 480. But whatever the truth of the matter there is not even a hint of a suspicion that he invented the story. It was part of the historical tradition he inherited and as a Greek patriot he was glad to pass it on. It is of course true that often his interpretation of the evidence and the theories he built on it, left much to be desired. It is quite possible, for instance, that he is the source behind Polyb. 13. 3. 2-4 and Strabo 10 C 448 (1. 12). In which case the only sensible conclusion to draw is that Ephorus, working like scholars in all ages, brought together different strands of evidence, including an archaic inscription of doubtful authenticity (very probably cited in a local history), and constructed the theory of an archaic ban on missile weapons. Such a ban fitted in well with Ephorus' ethical and panhellenic ideas. It is probably unhistorical, but that Ephorus invented the documentary evidence 325 is not to be credited. On the other hand, it is not to his credit that, though he must have been perfectly well aware of what Thucydides had said about the circumstances of Themistocles' flight to Persia, he chose to give the popular version (11. 56ff.; Plut. Them. 27. 1), doubtless because it was more dramatic and contrasted even more sharply the respective treatment of the great man by Athens and the Great ____________________ 323 Cf. Schepens, 1977: 114f. 324 e.g. Meister, 1970a: 607ff. 325 The inscription -- thus Wheeler, 1987: 157ff. -122-

King. Such shortcomings deprive Ephorus of a place amongst the very foremost Greek historians. Didacticism, however, was the rule rather than the exception with ancient historians and ancient critics did not hold this aspect of Ephorus against him. On the contrary, Polybius praised Ephorus for his γνωμολογíαι and ἐπιμετρου̑ντες λóγοι(12. 28. 10 = T23). See also Polyb. 5. 33. 2; 6. 45. 1 and Strabo1. 1. 1; 8 C 332 (1. 1); 9 C 422 (3. 11); 10 C 465 (3. 5) for praise of Ephorus as a historian. What did offend on the other hand was Ephorus' style. The ancient pronouncements are agreed that it was flat and repetitive. According to Dio Chrysostom, Ephorus wrote a great deal, but boringly (18. 10 = T25), and Polybius elsewhere states that he did not vary his vocabulary (6. 45. 10 = F148; cf. also TT22, 24, 28, and 29). The few surviving verbatim quotations seem to bear this out. 326 To what extent was Ephorus one of the first 'rhetorical' historians? Some ancient remarks provide a hint while Diodorus and F191 furnish due confirmation. Ephorus, as we have seen (p. 101 ), was one of the first if not the first historian to employ the preface. Theopompus, who also used prefaces, is censured by Dionysius of Halicarnassus for employing these to praise himself and attack other historians (Ant. Rom. 1. 1. 1 = FGH 115 F24); and Photius considered the prefaces of Ephorus and Theopompus very similar (FGH 70 F7: τῌ3+̑ τε διανοíᾳκαì τοîς ἂλλοις ἐστìν ὁμοιóτατα ). Many of Ephorus' personal comments, therefore, which Polybius admired, would have been made in the prefaces, not only those to books (FF8, 9, 110, 111), but also those he habitually appended to narratives within books. 327 Diodorus' own prefaces which he patterned upon Ephorus support this (pp. 101 ff.). But Ephorus' work was interspersed in addition with epainoi and psogoi, digressions of all sorts and speeches (cf. T21), and the remnants of many of these survive in the Bibliotheke. Such occasions would have provided Ephorus with the opportunity to indulge his γνωμολογíαι and ἐπιμετροῦντεςλóγοι in as rhetorically polished a manner as possible. His narrative on the other

hand seems to have been generally straightforward and unadorned. As Barber puts it: 'The encomia, prefaces and ____________________ 326 Cf. FF9, 20, 21, 31b, 47, 63, 90, 191 and Kalischek's detailed analysis, 1913: 64ff., as well as Barber, 1935:80f., 148ff. One of the less appealing characteristics was a predilection for the superlative degree; cf. FF9, 90, 191 frr. 4-5. 327 Andrewes in Essays Starr,189ff. -123-

speeches rose like peaks above the plain of a dull and uninspiring narrative.' 328 F191 corroborates this. Frr. 2-5, the epainos on Themistocles, are rhetorical in tone while frr. 6ff., the narrative, are not. The embellishments in the corresponding Diodorus narrative are Diodorus' own (pp. 132 f.). Closely related to the question of style is Schwartz's accusation that Ephorus regularly duplicated parts of his narrative as a literary artifice. 329 But none of the Diodoran passages he cites is decisive. On the contrary, while one may concede that Ephorus' language was repetitive there is not the slightest evidence that he manipulated the facts contrary to truth for the purpose of artistic effect. 330 Barber, who defends Ephorus against Schwartz's charge, nevertheless makes one exception -- the battle narratives were standardized rhetorical exercises. 331 Yet this is seriously to underestimate the extent to which Diodorus could impose his own language on his sources. Such expressions as καρτπερà μáηη,ἰσóρροποςἦν ἡ μáχη, ϕιλοτιμíς γενομένης etc. are Diodoran not Ephoran (pp. 15 ff.). The night attacks at 11. 10. 1; 11. 30. 2; 11. 61. 2-7; 11. 80. 3-6, a dramatic device according to Barber, may not be unhistorical, especially the last two instances, and Ephorus in any case derived this information from a source or sources he used in addition to Herodotus and Thucydides. With respect to the night fighting at Thermopylae (11. 10. 1) it is to be noted that the Spartan hoplites were apparently familiar with movement in darkness: X. Lac. Pol. 5. 7; Plut. Lye. 12. 14. We know nothing about Ephorus' personal circumstances and he may well have lacked military experience. All the same, he appears to have rendered the campaigns and battles of his sources fairly adequately. Blunders in Diodorus, where they can be checked, are usually shown to be Diodorus' and not Ephorus'. 332 Polybius specifically criticizes Ephorus' description of the battle of Mantinea in 362 as being untenable (12. 25f 3-5). It is clear, however, both from what Polybius says and from Diodorus, that Ephorus collected a great deal of information about the battle, and that he conscientiously tried to make sense of it. The result may ____________________ 328 1935: 152. 329 RE vi.15. Meister, 1982: 28ff. is in full agreement. 330 See Barber, 1935: 139f.; Stylianou, 1989a: 359f. 331 1935: 140ff. 332 Cf. pp. 386f., 395f.; Stylianou, 1989a: 353ff. -124- not have been as unsatisfactory as Polybius claims (85. 2n.). For Ephorus' interest in sieges and siege techniques see 26. 4n. and 34. 4n. On the other hand Polybius praises Ephorus' sea battles. The two he mentions, Cnidus in 394 and Salamis (in Cyprus) in 385, are too briefly related in Diodorus to enable us to evaluate Polybius'judgement. Just how much Polybius himself knew about naval matters is another question. 333 But there is other evidence which shows that his praise was not entirely unwarranted. What is certain is that Polybius was not praising the sort of rhetorical generalities we usually encounter in Diodorus. Ephorus' naval battles were more detailed, containing πολλà τχν χρησíμωνἀπενέγκασϑαι πρòς περισεις (Polyb. 12. 25f 2). The same state of affairs obtains elsewhere in Diodorus; for example, at 12. 33. 4 (battle of Sybota) and 12. 48 ( Phormio's victories), where only the bare bones of Thucydides' original accounts remain. Ephorus must have made better use of his sources as is shown whenever Diodorus chooses to retain more detail. A good example is 13. 10. 3. The explanation offered there seems nicely to supplement that of Thuc. 7. 36: the prows of the Syracusan ships were not merely stronger, but also lower and thus more effective against Athenian triremes with weaker and higher prows. The battles of the Decelean War are instances of the basic soundness of Diodorus as opposed to Xenophon which is now widely recognized. 334 These are the kind of descriptions Polybius admired. Admittedly Ephorus owed his facts to the Oxyrhynchus historian for the battles down to 394 (or perhaps 387: see n. 288); but not for the action off Cypriot Salamis which Polybius cites as an instance of an admirable account of a sea battle. The charge that he exhibited undue local patriotism was levelled against Ephorus in antiquity (FGH 70 F236) and it seems substantiated by the number of times Cyme is mentioned in Diodorus (p. 50 ). It is not the case that Ephorus fabricated evidence in order to glorify his native city. Everything in Diodorus about Cyme is historical in the sense that it was part of the local Cymaean tradition (2. 2n.; 18. 2-4n.). But the inclusion of a story such as that in ____________________ 333 Cf. HCP ii. 393. 334 For example, the actions off Cyzicus and Notium; cf. Bruce, 35ff.; Littman, 1968: 265ff.; Breitenbach, 1971: 152ff.; Lanzillotta, 1975: 135ff.; Andrewes, 1982: 15ff. But cf. Westlake, Studies, 224ff., for a more even-handed approach. -125-

15. 18 in a universal history was rather inappropriate. Ephorus nevertheless was not unique amongst universal historians in being inordinately proud of the city of his birth. 335 The merits of Ephorus are obvious enough even from the meagre fragments and Diodorus' text. It is true, as we have seen, that he was not political in the sense that, say, Thucydides was, and he certainly lacked the latter's acute political acumen. Ephorus' chief concern was with ethical conduct. But this should not be taken to mean indifference to political issues as such or to historical causation. We have also seen that he set himself the task of explaining why states rose and fell and specifically why Greece had failed in the end, and for this mere moral generalities (which of course he did not eschew) did not suffice. For one thing, as a universal historian of some ability (for which he elicits the praise of Polybius: 5. 33. 2) clearly he took care to bring out the connections between events and political developments in different parts of the Mediterranean. Some of these still survive in Diodorus; for example, the Perso-Carthaginian treaty of 480 (11. 1. 4f.), the symbiosis of Sparta and Dionysius (14. 10; 44. 2; 70. 1-3), and the fact that the inability of the Great King to bring the Phoenician navy into play in 411 was due to problems he was facing in the Levant (13. 46. 6). This factor was either unknown to Thucydides or he underestimated its importance. 336 Books 11-15 of the Bibliotheke are often compared with Books 18-20, and Ephorus is blamed for the much poorer state of the former group -- his unfortunate oikonomia, we are told, presented Diodorus with insurmountable problems. There is truth in this though the comparison is not entirely fair. Books 18-20 cover only twenty-two years. Hieronymus, like Thucydides, wrote contemporary history -- an account of the wars of the Successors in which he was himself a participant. His theme was fairly simple and straightforward compared with what Ephorus undertook to synthesize. Books 11-15, by contrast, cover 120 years. Book 15 alone covers twenty-six. This latter period was dealt with by Ephorus, let us recall, in no less than six books, without counting the western narratives. As a rule, the more severe the abbreviation, the greater the ability required and Diodorus had little of that as it was. We get a better idea, therefore, of the quality of Ephorus' narrative whenever Diodorus gives more detail. A good example is ____________________ 335 Cf. Burde, 1974: 115ff. 336 HCT v. 290. -126-

the account of Dionysius' preparations for declaring war against Carthage in 398 (14. 41ff.). The assignment and analysis of motives is particularly well done. At 14. 41. 1 we have Dionysius' true motives (according to Ephorus) for declaring war; they are those of an active and ambitious tyrant. When everything is ready Dionysius convenes and addresses the Syracusan assembly 14. 45. 2-4). Almost certainly Ephorus included a speech at this point and Diodorus has rendered the oratio recta into a brief oratio obliqua. But the essence of the speech, one trusts, is preserved and it gives us the public declarations of Dionysius which, as Ephorus had intended, do not quite tally with his true motives at 14. 41. 1. Again, in giving their assent to Dionysius' proposal for war, the Syracusan citizens had their own agenda (14. 45. 5) and this is taken up again later, at 14. 64. 4, and leads naturally on to the attempted revolution of 396 and the speech of Theodorus which Diodorus this time reproduces (14. 65ff.). The speech, a call to freedom by the Syracusan political opposition, is ably crafted and fits well into the rest of the narrative, as does the one by Nicolaus at 13. 20ff., and as the one by Hetoimaridas at 11. 50. 6, which Diodorus has not retained, must also have done. The contrast between 14. 41ff. and the bare bones (or less) given at 15. 15ff. and 15. 73 is striking and Diodorus, certainly, is responsible for this (see pp. 78 ff.). On a general level Ephorus offered a more complete and coherent picture of the contemporary Greek world than Xenophon and the loss of the books dealing with the fourth century is particularly unfortunate, as even Gomme conceded who had as low an opinion of Ephorus as Schwartz and Jacoby. 337 We could hardly begin to understand certain fourth-century aspects but for Ephorus. The Common Peace, for instance. Ephorus gave this central aspect of fourth-century politics a great deal of attention. 338 Unlike Xenophon who gives a very insufficient picture of the nature and significance of the Common Peace, Ephorus appears to have grasped it fully. Furthermore he tried to explain why it failed. A major cause of disruption was the ever present strife between democrats and oligarchs. The product of this strife, the political exile, and the envy and fear which the leading states felt against each other, guaranteed that no treaty could last for long; cf. ____________________ 337 HCT i. 45 and n. 3. 338 Cf. Martin 1944: 27ff.; Ryder, 1965: pp. xivff. -127-

15. 5. 1-3; 20. 1; 25. 1; 26. 1; 40; 45; 77. 1-2; 94. 1. The malady was so deep-seated that in one instance Ephorus aptly described the existing situation not as koine eirene but as koine anarchia (45. 1). All this does not suggest a politically naive historian. The Second Athenian Confederacy is another such instance. Xenophon ignored it totally. Ephorus, on the other hand, gave a full account of its origins and development. The accuracy of his information is striking even in Diodorus' severe and clumsy epitome (15. 28. 3; 29. 7f. and nn.). The rise of Thebes with the corresponding decline of Sparta and the Theban hegemony took up no less than six books (pp. 94 f.). This part of the Histories, therefore, must have been extremely detailed. Again and again Diodorus supplies vital information omitted by Xenophon (cf. 34. 3n.; 45. 4n.; 76. 3n.). We form an idea of the kind of detail Ephorus gave from the fragments and Diodorus' epitome; cf. F76 = 14. 98. 2; F81 and 69. 1n.; F82; F83 and 79. 1n.; F84 and 75. 2n.; F86; 7576n.; 81. 1-4n. One might add that Ephorus, in contrast to Xenophon, cared enough to make it clear that there was, after 378, a Boeotian Confederacy (28. 1nn.); and of course he gave what Xenophon should have given, but

did not, an account of the reorganization of the Peloponnesian League (15. 31. 1-2). For the factual basis of all this material Ephorus, of course, relied extensively, though by no means exclusively, on written sources and we have seen that occasionally he borrowed their opinions too (p. 105 ). The charge, however, that he was no more than a compiler is undeserved and indeed inappropriate. Given the task he set himself, heavy dependence on the work of others, predecessors as well as contemporaries, was inevitable. 339 It is perfectly clear nevertheless that he bent whatever he borrowed to his purpose and that in the absence of written sources he was well able to synthesize narratives from oral sources. The account of the reconquest of Phoenicia, Cyprus and Egypt in the 340s is impressive even in Diodorus (16. 40 ff.) An unusual feature in Ephorus appears to have been an interest in literary history. There is at any rate a passage in Diodorus (15. 76. 4) which seems to derive from an Ephoran literary digression (see n. ad loc.). As the matter has a direct bearing on the kind of historian he ____________________ 339 Cf. Schepens, 1977: 102ff. -128-

was it would be appropriate in this context to examine the numbers in Ephorus, that is, the numbers in Diodorus' narratives of Books 11-16 which are based on him. Ephorus derived this information from the various sources he employed, both written and oral, but it is clear that he felt free to manipulate it to suit his aims and the results of this manipulation are not in general to his credit. The numbers are usually to do with army and navy strengths and casualties. Greek and barbarian figures were arrived at differently and they should be considered separately. 340 For those of the Siceliots and Carthaginians see pp. 62 ff., 73 ff. The Greek numbers. Such an inquiry is bound to be untidy. And the starting point must inevitably be Thucydides as the only surviving original source for Diodorus (Ephorus) apart from the Oxyrhynchus historian. Meiggs has compared the numbers in Diodorus with those in Thucydides and has concluded that they are substantially those of Thucydides. 341 Most of the differences (minor ones at that) are due to Ephorus' wish to appear independent from his source. Two such cases may be mentioned, not commented on by Meiggs. First, the Boeotian infantry at Delium in 424 numbered 7,000 hoplites, more than 10,000 light-armed, and 500 peltasts, according to Thuc. 4. 93. 3, that is, more than 17,500. In Diodorus this is rendered as πεξοὺς μ ν οǵ0 πολὺ λείποντας τ̑ ν Δ σμυρίων (12. 69. 3). Second, in reviewing the Athenian resources at the beginning of the Peloponnesian War Ephorus gave 12,000 front-line and 17,000 reserve hoplites (12. 40. 4) while Thucydides gives 13,000 front-line and 16,000 reserve hoplites (2. 13. 6). The total, 29,000, is the same for both and it indicates that Thucydides was Ephorus' source. There are, however, a number of instances where Ephorus supplied information additional to that of Thucydides and some of these include figures; cf. 12. 54. 5;70. 1;73. 3;77. 4;80. 4;13. 38. 5. Such instances apart, it remains true that most of Diodorus' numbers for the pentecontaetia and the Peloponnesian War down to 411 are basically those of Thucydides and are reliable in so far as Thucydides' figures are reliable. Similar results would almost certainly have been obtained from ____________________ 340 Some allowance should be made for the vulnerability of numbers to corruption, observable in the MSS of D every now and then; cf. e.g. Vogel's app. crit., ii. 252. Such numerals as Λ', M', and N' are particularly liable to corruption. But the problem should not be exaggerated: Develin, 1990. 341 1972: 447ff. -129-

a comparison of Diodorus with the Hellenica Oxyrhynchia (P), had not the latter been so imperfectly preserved. The numbers are the same at 13. 65. 1f. = P 1. 1; 13. 71. 2, 4 = P 4; 14. 79. 8 = P 9. 2 (though after severe restoration). P 11.3(= Diodorus 14. 80. 1) is too fragmentary for any conclusions to be drawn. 342 The difference at 14. 80. 2 (1,400 soldiers) = P 11. 4 (? hoplites, . . . τακοσǷ7ους light-armed) is again perhaps due to Ephorus' wish to appear independent of his source. The 600/6,000 difference at 14. 80. 4 = P 11. 6 is the result of corruption rather than exaggeration. The numbers in Book 15 appear generally sound though they cannot always be checked. Obvious exaggerations there are: e.g. the claim that at Tegyra the Thebans faced and defeated twice as many Spartans (37. 1; a proBoeotian source is certainly responsible); and the Lacedaernonian losses at Leuctra which are given as 4,000 (a pro-Boeotian source may again be at work). The barbarian numbers. It is not surprising that Greek writers in general exaggerated the strength of non-Greek armaments; that of armies, wildly so. Figures for fleets are usually more manageable. Ephorus was no exception. A frequent figure given for Persian armies is 300,000 (cf. 11. 74. 1; 12. 3. 2; 15.2. 1; 16. 40. 6; X. An. 1. 4. 5; 7. 12). It would not have occurred to many ancient writers that such an incredible figure would have been impossible on logistical grounds alone. Not many ancient writers would have been aware that there is a difference between the number of men in a populous state capable of bearing arms, and putting all these men in the field at the same time. Thus, it often happened that fairly trustworthy figures for a state's military resources, the total numbers of cavalry and infantry, for example, were translated into mammoth expeditionary forces. Sections 2. 5. 3-7 and chapter 25. 13 are instructive. 343 Greeks would look at their own numbers, then consider the vast expanse of Asia and multiply accordingly. The circumstance that the same or similar figures are given for both Persian and Carthaginian armies (see pp. 62 f.) rather spoils the rationale that they have any basis in fact.

____________________ 342 But see Bruce, 78f. 343

25. 13 goes back to Fabius. For a proper use of the Roman καταγρϕαί see Polyb. 2. 24 with HCP i. 196ff. Three hundred thousand for the citizenry of Sybaris (12. 9. 2) is of course an absurdly inflated figure even for a reputedly fabulously wealthy Italiot city in the 6th century; but note its even more absurd conversion into an expeditionary force (12. 9. 5). If Ephorus is responsible for this information he must have had different standards for archaic times. -130-

On the other hand a case could be made for the factual basis of the number 300 for Persian navies which recurs in Diodorus in both fifth- and fourth-century contexts (at 12. 3. 2;13. 38. 4;13. 41. 4;13. 42. 4;13. 46. 6;15. 2. 1;15. 41. 3;16. 22. 2;16. 40. 6). As a rough estimate of the potential naval strength of the Great King 300 is a priori not impossible when compared with Greek fleets. After the loss of Ionia in 479 Persia relied almost entirely on Cyprus and Phoenicia for her fleets, and fairly reliable evidence shows that these two areas could, in an all-out effort, supply the King with a fleet not far short (if at all) of 300 triremes. Conon's fleet in 394 numbered go triremes, 80 Phoenician and 10 Cilician (14. 79. 8;83. 4). In the 380s Evagoras of Salamis eventually had 150 triremes, 130 Cypriot and 20 Tyrian (15. 2. 4;3. 4). Arrian tells us that in 333/2 the Phoenician cities of Aradus, Byblus, and Sidon had so triremes in operation with the Persian fleet while the Cypriot kings joined Alexander with 120 (An. 2. 20. 1, 3); Diodorus adds that the Tyrian fleet was so strong at the time (17. 41. 1). The sum of these last figures comes close to the reported strength of Memnon's fleet, 300 (17. 29. 2), though Arrian says it numbered 400 (1. 18. 5). At any rate Alexander did not consider it wise to risk battle with his 160 (An. I. 18. 4). As an estimate, therefore, of the full Persian (i.e. Cypriot and Phoenician) fleet, 300 seems reasonable. In practice, however, royal fleets usually fell far short of that figure. In 397/6 Herodas of Syracuse witnessed the preparation of a royal fleet in Phoenicia to be used against Sparta. He heard that the order was for 300 triremes to be manned (X. H. 3. 4. 1). It may thus be that 300 was the official Persian estimate of the potential of Cyprus and Phoenicia. In the event, however, the fleet which defeated the Spartans at Cnidus numbered only some ninety triremes (14. 79. 8;83. 4). It should be further noted that when Herodas reported at Sparta what he had heard and seen, Lysander was of the opinion that the Greek fleet would be superior to the Persian (X. H. 3. 4. 2). Hard-headed Greeks like Lysander were under no illusions as to the true naval strength of Persia. We may further compare Thucydides' figure of 147 triremes for the royal fleet which came as far as Aspendus in 411 (8. 87. 3). 344 Thucydides is emphatic about the number though he does make Tissaphernes ____________________ 344 Isocr. 16. 18 gives 90 ships, perhaps a confusion with Conon's Phoenician fleet in 394. -131-

say that fewer ships had been collected than the King had ordered (8. 87. 5). Ephorus nevertheless was no Thucydides and he should be criticized for repeatedly supplying the figure 300 as the actual operational strength of Persian fleets. Doubtless in many instances no reliable information could be obtained and Ephorus may have felt justified in giving the usual top estimate. On the other hand he must have come across Thucydides' figure for the fleet of 411; yet he opted for the usual estimate of 300. Perhaps he reasoned that Thucydides' figure was half the Persian fleet and as a panhellenist he felt no qualms about doubling it. And perhaps the fleet which defeated the Spartans at Cnidus was only given its true (and low) strength because of Conon's (and Athens') involvement.

4. DIODORUS' METHODS That the Bibliotheke is so obviously a work of compilation, one hastily and incompetently carried out, can be illustrated abundantly and it has in fact already been demonstrated. The present section is concerned with bringing this central aspect of Diodoran studies, that is, the use and misuse of the sources, into an even sharper focus. We may begin by comparing P. Oxy. xiii. 1610, an extensive Ephoran fragment (F191 -- see n. 17), with the relevant part of the Bibliotheke. The correspondence is impressive, the differences being inept rhetorical flourishes, omissions, misconceptions, and confusions on the part of Diodorus. He will sometimes stay very close to the language of his source, but usually he will paraphrase (p. 15 ). The epainos on Themistocles, for example, seems to have been repeated almost verbatim in places (frr. 3-5 = 11. 59. 3). Note in fr. 6 Ephorus' plain statement that he is returning to his narrative which Diodorus expands to the following pompous declaration: περὶ μ ν οὑν τηπς ΘεμǷ2στοκλ ους ἀρετη+̼0+ς εὶ καὶ παρλεονα+́καμεν παρεκßα+́ντες, ἀλλ' ου ν οǵ1κ α+́+ ξ ον κρίναμεν τὴν ἀρρετὴν νὐτου+ παρλιπειν ἀνεπισǷ5μαντον (11. 59. 4). This is very characteristic and easily spotted. If we look further at fr. 6 and compare it with 11. 60. 1-2 we notice that Diodorus is three times as long, but only with verbiage. Ǳ0χǷ9ντων in the fragment becomes κατεχǷ9ντων in Diodorus and εǳ1+ιλον becomes Ǳ0χεǷ2ρώσατο. A similar fate befalls frr. 9+10+53. The plain παραταχλεǷ7σας ΔǷ2 πολὺν χρǷ9ν of the fragment assumes the usual proportions of a Diodoran -132-

action: γενομένου Δ ǰ0γ νος ǳ0σχυυου+ καǷ6 τω+̑ν στόλων ἀμϕοτέρων λαμπρ ς ἀγωνιξομένων κτλ(11. 60. 6). The 'hundred' of the papyrus becomes 'more than a hundred' in Diodorus, surely a rhetorical exaggeration and not out of respect for Thucydides. 345 Two important observations to be made are that στρατηγου+ντος in fr. 6, which is correct, becomes στρατηγǷ9ν έλǷ9μενοǷ2 at 11. 60. 1, which is not. This is because Diodorus was ignorant of the Athenian constitution. He assumed that generals were elected for specific campaigns.

346

Further, Δὑναμιν ǰ0ξǷ2όλον in the same section is a favourite

expression of Diodorus. It would seem that whenever the strength of a force was not stated in his source (as it is not in fr. 6), Diodorus made up for it with this or a similar phrase (cf. 11. 65. 3;11. 71. 4;11. 78. 2, 4;12. 4. 1;12. 7. 1;12. 42. 3;12. 47. 1;15. 19. 3;15. 21. 1). The omission of whole episodes (e.g. the recovery of the bones of Theseus -- fr. 7) is highly instructive; Diodorus did not abridge uniformly -- he would extract and abridge. The rate at which he did so doubtless varied depending on the amount of narrative available and his own choice. The rules he followed, if any, in abbreviating his sources are impenetrable. It is undoubtedly true to say that if an account in Diodorus is full, the corresponding source must have been at least as detailed. The reverse, however, it should be emphasized, does not necessarily hold. No less instructive, and indeed this is a characteristic trait of Diodorus, is his tendency to muddle even moderately complicated accounts in his sources. In this instance he has contrived to remove the naval action of the battle of Eurymedon to Cyprus. 347 Too little of Ephorus survives to carry this investigation further, but if that were possible the results would certainly be the same. For instance, a verbatim quotation from Ephorus by Stephanus Byzantius (F76) is repeated word for word by Diodorus except that he drops the adverb ἔτι. 348 A comparison of Diodorus with other writers he employed such as Agatharchides, Polybius, and Posidonius produces similar results. Diodorus' own utterances, ____________________ 345 Grenfell and Hunt, P. Oxy. xiii. 122. Frr. 9+10+53 are restored somewhat differently by Rubincam, 976: 357ff. 346 Cf. 12. 7. 1;12. 27. 1;12. 81. 2;15. 29. 7;15. 47. 4. For 12. 7. 1 and 12. 27. 1 see Thuc. I. 114. 1; 116. 1 and cf. Jacoby, 1949: 306 n. 37. 347 Certainly Ephorus is not to be blamed for this: Stylianou, 1989a: 353ff. It is most unlikely that Κέπρον is the correct reading in frr. 9+10+53 of FGH 70 F191 (= P. Oxy. xiii. 1610): Stylianou, 1989a: 359ff. 348 I would not attach any historical significance to Ǳ4τὲ as does Reid, 1974: 134. -133-

which are often in evidence, and which are largely of a moralizing or a procedural nature, hardly amount to a 'philosophy of history' and they do not earn him promotion to the rank of an original writer; nor do the blunders which generally ensue whenever he attempts to innovate. It would be unnecessary to enter upon such an investigation here, especially as this has been recently done, and done well, by J. Horriblower. 349 A good example (taken at random), not cited by Hornblower, is Diodorus25. 2-5 which corresponds with Polybius 1. 65-88. It can be seen that Diodorus has followed Polybius very closely, even reproducing the latter's didactic statements and moralizing reflections, often word for word; cf. e.g. 25. 4. 2 = Polyb. 1. 84. 5-6;25. 5. 1 = Polyb. 1. 84. 10;25. 5. 2 = Polyb. 1. 86. 7;25. 5. 3 = Polyb. 1. 88. 3. On Diodorus and Polybius see also p. 8. It might be helpful if Diodorus' misuse of his sources were examined more closely under the following three headings of omissions, chronological peculiarities, and general vagaries. Omissions. Diodorus' unsystematic and injudicious method of extracting from his sources has resulted in many and often serious omissions. Accounts are frequently telescoped or left incomplete and. important information passed over. It would serve little purpose to list all such instances, even those in Book 15 (they are noted in the commentary), especially as we cannot be certain in every single case that the omission is Diodorus' and not the source's. Two major omissions in Book 15 concern the Peace at Athens in 371 (pp. 408 ff.); and the years 367/5 are sketchy in the extreme (75-76n.). For instances of telescoped narratives cf. 15. 29. 1-4 (the point at which Acoris died is missed); 32. In. (the description of the Spartan army); 34. 2n. (the second invasion of Boeotia in 377); 36. 5 n. (Corcyra omitted from Timotheus' successes); 77. 4n. (the Arcadian successes in 365/4). For incomplete accounts cf. 14. 84. 1f. (the best part of the battle of Coronea is not related); 15. 32. 3ff. (only one episode of the 378 invasion of Boeotia is given); 15. 34. 1f. (only the end of the second invasion is related); 15. 34. 3 (the preliminaries to the naval events of 376 are ignored). ____________________ 349 1981: 27 ff. Contrast Sacks, 1990: e.g. 83 ff. As a defence of D against extreme views of Quellenforschung (rarely held these days in any case) Sacks' case is fair enough, as far as it goes, even though most of the points he advances are small beer which cannot really be proved either way. Hornblower's analysis and conclusions remain essentially indisputable. -134-

The narrative sources were not alone to suffer at the hands of Diodorus; the chronographer suffered equally as the sorry state of the kinglists shows (pp. 32 ff.). Diodorus neglected much material supplied by the chronographic handbook (p. 32 ). It is worth pointing out that the years 452/1 (archon Chaerephanes) and 321/19 (archons Archippus I and Neaechmus) are missing along with all the usual chronographic material (the annual magistrates). There is no reason to suspect lacunae. In the first case it is possible that the year was simply overlooked by Diodorus; or even that he found nothing in it to expand and therefore ignored it. For the second case Goukowsky has suggested that Diodorus was confused by the fact that the archon for 318/7 was also called Archippus. 350 A more probable explanation, however, is that Diodorus, by consulting the chronographer, realized that his narrative had moved ahead, way past the point at which the year of archon Philocles (322/1) should have been concluded and that of Archippus I begun. His solution, not untypical of the man, was to skip the chronographic information mentioned. Errors in chronology. Diodorus' method of working, as well as the fact that the chronographer was not infallible, are responsible for the chronological muddles which characterize the Bibliotheke. We have already looked at the kind of errors for which the chronographer is to blame (pp. 45 ff.). We may now look at Diodorus' own mistakes. The inept combination of narrative histories and chronographic source is sufficient to explain many of them. Diodorus would attach an abridged portion of narrative which included an event he found in his chronographic source to that archonyear under which the event was placed in the chronographer. If the latter had done his job well then that event would be correctly dated; but the rest of the events in the narrative might not be. Depending on the position of the dated event and the timespan of the summarized narrative, they might be misdated by several years, either too low or too high or both. The following cases are typical.

Chapters 15. 28-35 are a long, condensed version of Ephorus' account of the three years which followed the liberation of Thebes. They deal with the restoration of the Boeotian Confederacy, the foundation and expansion of the Second Athenian Confederacy, ____________________ 350 Budé edn. of D 18 (1978), p. xxvi. -135the raid of Sphodrias, the outbreak of war, Agesilaus' invasions of Boeotia, and the war at sea. All these events are placed under 377/6. It is clear that the narrative straddles this date. It trails behind it to begin with, but by the end it has actually moved ahead -- the battle of Naxos was fought in September 376 (34. 3 n.). The same is true of 15. 41-7, which deals with the Persian expedition against Egypt, the renewed war in Greece, the destruction of Thespiae and Plataea, and the whole episode of the siege of Corcyra. The year 374/3 is correct for the first of these events, but not for the rest; 15. 47. 7 belongs in summer 372. Such runaway narratives are generally halted and the chronology rectified by the next event Diodorus dates from the chronographer. Thus Diodorus is continually in and out of trouble with his chronology. There are many examples of such muddles. Under 471/0, for example, Diodorus deals with the ostracism of Thernistocles, his sojourn in Argos, his flight to Persia, and eventual death; under the following year, with Cimon's operations in the Hellespont, the Aegean, and the eastern Mediterranean. Both sets of events required a number of years to run their course. Again in Book 15, under 383/2 there occur the origins and beginning of the Olynthian War, the subjection of Phlius, and internal Spartan politics. Phlius, nevertheless, was not attacked till 380 (19-23n.); and the whole of the Great Satrap Revolt is related under 362/1. Clearly the oikonomia of Ephorus must bear part at least of the blame for many of these confusions. Even a painstaking epitomator, and Diodorus took few pains, might have run into trouble in trying to give Ephorus' Histories an annalistic form. When, however, Diodorus followed a source which used a year-by-year system, his chronology was less unsatisfactory. 351 Other chronological curiosities are the result of Diodorus' own arrangement of his material, an arrangement often dictated by the necessity of fitting together the historical accounts of different regions while keeping to a strictly annalistic system. For example, the otherwise continuous narrative of the Sacred War in Book 16 is interrupted for three whole years, 351-348, by Persian affairs, and for a fourth, 348/7, by Philip's siege and capture of Olynthus. This does not of course mean that the Thebans and Phocians remained inactive meanwhile. Likewise, the Greek narrative is ____________________ 351 This is the case with Hieronymus; see Hornblower, 1981: 34. -136interrupted at 11. 48-9 by western affairs though 11. 50. 1 makes it clear that there was no such interruption in reality. Occasionally happy results ensued. It is not impossible that Mantinea figured twice in the chronographer, the beginning of the siege under 386/5, and its conclusion under 385/4. But it is not unlikely that Diodorus, using his initiative, and noticing perhaps in his source a statement to the effect that τὸ μ ν λέρος Δὲετέλεσαν οἱ Mαντὲνεις . . . ἀγωνὲζόμενοὲ ... το ... του+Δ χεὲμ νος ἐνστος(15. 12. 1), split the Ephoran narrative in two. That enabled him to cover all three major areas in succession for both 386/5 and 385/4. The single chronographic entry perhaps was under 386/5, and 15. 5. 1, ἅμα Δ τούτοὲς Πραττομένοις ... Δι° τοιαύτας αίτίας, may be all that is left of it (pp. 44 f.). At any rate Diodorus' arrangement coincided with the historical reality for the Mantinean War seems indeed to have begun in summer of 385 under archon Mystichides and to have finished in the following winter under Dexitheus (12. 1 n.). General vagaries. A mixture of incompetence, lack of care, and ignorance is responsible for the vagaries which abound in the Bibliotheke. Many of these we have already met. Laqueur 352 examines a large number of anacolutha and other defects. It is impossible, however, to agree with his explanation that these are caused by the regular use of two or more narrative sources for the same account. Of course Diodorus did occasionally supplement the narrative source he was following with information from an alternative source (p. 50 ); and he not infrequently included his own commonplace reflections (pp. 3 ff.). But many of the instances cited by Laqueur are more simply explained as the result of a desultory and even haphazard use of one source. Diodorus would often not take care to smooth out the grammar of the extracts he joined together; hence the anacolutha. We might add 15. 70. 2 to Laqueur's list -- the passive προσεΔέχλησαν makes little sense in the context (see n. ad loc.). The anacolutha in fact are a good indication not only of Diodorus' slipshod methods, but also of the fact that he did not revise his work. 353 ____________________ 352 1958: 257ff. 353 Other defects listed by Laqueur have an even simpler explanation. For example, 12. 3. 2, ἐστρατηγουνV Αρτάßαζος {καὶ Μεγάßαζος Αρτάßαζος} μ ν κτλ, is surely an instance of the copyist's eye jumping from the first mention of Artabazus to the second. And at 15. 41. 3 where the MSS read "Artabazos" instead of "Tharnabazos" we should see a manuscript corruption rather than an insertion from an additional source as Laqueur, 1958: 273, implausibly suggests. On the other -137-

A complete list of Diodorus' blunders would be of a very respectable length. I confine myself to giving some representative examples with special, though not exclusive, reference to Book 15. The most serious confusions in Book 15 concern the Peace of 375 from which the Thebans are said to have been excluded (ch. 38); the assertion that Timotheus was immediately reinstated to the generalship in 373 and that he sailed to Corcyra with Iphicrates (ch. 47); and the campaign and battle of Leuctra (chs. 51-6). All three are discussed in the commentary. A kind of error to beware of is the confusion of a pair of names; cf. 12. 78. 1-2 (Troezen for Epidaurus); 13. 36. 5;37.

4f.;38. 4f. etc. (Pharnabazus for Tissaphernes. In fact the latter is not mentioned at all in Book 13); 13. 64. 1 (Thrasybulus for Thrasyllus); 15. 19. 3 (Phoebidas for Eudamidas); 15. 59. 1 (Tegean for Mantinean. He gets it right at 15. 62. 2); 15. 82. 1 (Mantineans for Tegeans); 15. 93. 2 (Tachos for Nectanebos). For sheer careless blunders cf. 12. 35. 4 and 12. 42. 6 Archidamus is killed off in 434 and then resurrected in 431 -- see pp. 39 f.); 12. 63. 3 (the Spartans on Sphacteria are starved into surrender in 425); 12. 72. 3 (he places Antandrus on Lesbos, though this may be the result of abridgement); 15. 36. 4 (Chabrias is murdered in Thrace in 376/5; see n. there); 15. 77. 4 (the Hill of Cronus at Olympia becomes a polis). The well-known doublets result from an injudicious combination of chronographic source and narrative source; cf. 16. 31. 6=16. 34. 4-5; 16. 34. 3=16. 39. 4; or of narrative source and narrative source; cf. 12. 19. 1 f. = 13. 33. 2f.; or of the same section of a narrative source being summarized twice; cf. 13. 34. 1-3=13. 36. 1-5. I do not, however, consider either 15. 27. 4 and 15. 33. 5-6 or 15. 38. 3 and 15. 50. 4 to be doublets (see 27. 4 n.;38.339n.). A peculiar feature in Diodorus is that he is apt to deviate from other sources in the form of proper names, though to be fair it is not always clear whether the fault lies with him or with his source or with a copyist; cf. Κλέαρχον (13. 106. 9f.) = ΚλεανΔρίΔην (FGH 70 F193; ΚλεανΔρίΔαν Thuc. 6. 93. 2; cf. HCT iv ad loc.); ΕὐΔόκιμον (14. 97. 3) = ἝκΔικον (X. H. 4. 8. 20); ΣϕοΔριάΔης (15. 29. 5f.)= ΣϕοΔριαΔ (29. 5 n.); ΘηριπίΔην (15. 30. 3)='ΗριππίΔαν (30. 3n.); 'Ηγήλοχον (15. 84. 2) = 'Ηγνσίλεως (FGH 70 F85; X. Poroi 3. 7). ____________________ hand, an awareness of D's propensity to write nonsensical Greek does not dispose one to perceive lacunae and textual corruption wherever the text is defective in some way. Cf. e.g. 14. 16. 5 where the editors' suggestions may be misdirected. -138-

Most of the above, of course, were errors of omission rather than commission though the second category is well represented too and both together set the seal on Diodorus as a historical compiler of low stature. To obviate the charge of carelessness and incompetence, once the Bibliotheke was completed, Diodorus claimed that some of the books had been pirated and published πρó Γο+00FB διορϑωϑη̑ναι καì Γὴυ ἀκριßη̑ συνΓὴλειαν λαßεîν. These he disowned, but in order, he says, to safeguard the integrity of his work he proceeds to give a brief summary of the contents of the Bibliotheke (40. 8). He does the same in the general preface which again, as he emphasizes (1. 4. 6), was written after the completion of the work (1. 4. 6-5. 2); and he repeats his wish to protect both the reader and the Bibliotheke from book pirates (1. 5. 2). The apologia is transparent. Even if what he says is true, it is not obvious why the general plan of the work he gives should help the reader distinguish between the official, and supposedly revised, books and the pirated and unrevised ones -- unless the contents of the pirated books were different and this we cannot know. At any rate, what we have undoubtedly corresponds to what he says and, therefore, clearly our text of the Bibliotheke, in all its sorry state, must be the 'corrected' one. There is in fact no evidence either for pirated books or for a revision of the Bibliotheke. Quite the contrary. All the indications are that any revision was of the most perfunctory nature. 345 ____________________ 345 The variant drafts of the same material in Book 1 pointed out by Murray, whether in the same or in different MSS ( 1970: 170), constitute no such evidence though some are inclined so to see them (e.g. Sacks, 1990: 173 n. 60). Perhaps the same explanation applies there as for 13. 34. 1-3 = 13. 36. 1-5 (see text above on doublets). -139-

COMMENTARY 1 . The preface For D's prefaces in general and this preface in particular see pp. 101ff. For the scope and origins of the praise/blame function of history in the Bibliotheke see pp. 5ff. 1. 2. ἐν οἱ̑σ Λακεδαιμóνιοι κΓλ : cf. 11. 82. 3; 15. 33. 3; 15. 50. 2; 15. 56. 3. The rhetoric doubtless goes back to Ephorus, but D's inept handling has distorted the meaning. Both παραδóξνσ and ἀνελπíσΓνσ have approximately the same meaning and D's use of them was motivated by a desire for stylistic variation. The Spartans lost their hegemony at Leuctra, not Mantinea, though Mantinea ensured that Spartan power was broken for good, and this presumably was the point made by Ephorus; cf. Strabo 9 C 414 who probably reflects Ephorus (9 C 422). ἀνελπíσΓνσ is inappropriate in the context of 362 and Mantinea. The Spartan defeat at Leuctra was a great surprise (even to the Thebans) and it shocked Greece (cf. X. H. 6. 3. 20; 6. 4. 20; Plut. Ages. 29. 1). But the Spartans at Mantinea were just another contingent and their defeat hardly caused a stir -- Epaminondas' death did. 1. 3-5. For Ephorus and the decline of Sparta see pp. 113 f. 1. 3. ἔΓη πλεíω Γω̑ν πενΓακοσíων: cf. 50. 2 and 7. 12. 8 where Wesseling emends the ú to φ (cf. Jacoby, FGH iic. 86). For Ephorus the Spartan hegemony began with the reforms of Lycurgus and ended with Leuctra; see, in addition to D, Strabo 8 C 365f.=FGH 70 F118. Ephorus, therefore, dated Lycurgus to C.870. The same approximate date may be obtained by another consideration. Clement of Alexandria says that Ephorus calculated 735 years from the Return of the Heraclids to the year of archon Euaenetus (335/4) when Alexander crossed into Asia (F223). He thus dated the Return to 1070/69. D, on the other hand, says that Ephorus covered 750 years from the Return of the Heraclids to the siege of Perinthus in 341/0 when he concluded the Histories (16. 76. 5). This would place the Return in 1090. The difference is small and it can be eliminated if we accept Jacoby's emendation of -141-

D's number to 730 (N' is emended to Λ: FGH iic. 101f.). Now, Ephorus considered Lycurgus to have lived in the sixth generation after Procles (F149. 18), i.e. some 200 years after the Return, reckoning 30 years to a generation. Thus his floruit is c.870; cf. Barber, 1935: 171f; Burde, 1974: 102. According to Ephorus, therefore, some 700 years intervened between the Return of the Heraclids and Leuctra, and this is Isocrates' figure too (6. 12; 8. 95; 12. 204). 1. 6. The usual link-paragraph; see p. 25. Wesseling's emendation προ Γῃ+̑ γραφῃ+̑ χρóνουσ on the basis of 12. 2. 2 is probably correct. The meaning could hardly be in doubt in any case as D always uses some such statement in introducing his link-paragraph. εἱσ Γòν ῥηγíνων ἀνδραποδισμòν κΓλ: from the chronographer. Under archon Theodotus (387/6) D's chronographer noticed the conclusion of the King's Peace, the capture of Rhegion by Dionysius, and the Sack of Rome by the Gauls, and D arranged his narrative accordingly (14. 110ff.). Polybius dates the same three events seemingly in the same year which, however, he does not fix by citing a magistrate or even the appropriate Olympiad year. Instead he defines the year in question as being the nineteenth after the battle of Aegospotami and the sixteenth before the battle of Leuctra (1. 6. 1-2). Using Olympiad years and reckoning inclusively, 387/6 is obtained (Walbank, HCP i. 46f.). There can be little doubt that the synchronism of the three events (probably going back ultimately to Philistus; cf. the discussion of Werner, 1963: 69ff.) is correct. On the chronology of Dionysius' activities in the south of Italy see further 7. 2n.

Archon Mystichides, 386/5 2. 1. ῥωμαîοι κΓλ: repetition of the college (394 V) cited at 14. 97. 1 (391/90); see p. 27. In reality only two tribunes are given, M. Furius and C. Aemilius, the Kal here (not at 14. 97. 1) probably being someone's (presumably not D's) attempt to harmonize the figure of three with the actual number of names given; cf. Unger, Jahrb. f. class. Philol. ( 1981), 494. At 14. 97. 1 the 'edited' text of MS F (see Preface) adds a third name, KáΓλοσ Oὐη̑ροσ, not known to Livy who otherwise gives a full college of six tribunes (5. 26. 1-2). M. Furius and C. Aemilius are first and third in Livy's list, -142the second being L. Furius. It is easy to see how the second Furius was omitted, and omitted early in this particular tradition of the fasti: Drummond, 1980: 61 and n. 21; CAH2 vii. 2. 636; cf. also Perl, 1957: 79, 103 and, for the odd name in MS F, Pinsent, 1975: 30 n. 35.

2-4. The invasion of Cyprus The background and beginning of the war were given at 14. 98. 1-4 (under 391/o) and at 14. 110. 5 (under 387/6) it was noted that with the conclusion of the Peace of Antalcidas the King was free to attack Evagoras. The invasion of Cyprus is now related. The chronology of the Cypriot War is a well-known problem and it must be dealt with first. For an account of the origins of the war and of Evagoras in general see Stylianou, 1989b: 458ff. The chronology The difficulty lies chiefly in the contradictory nature of the evidence. At 15. 9. 2 (under 385/4) D states that the Cypriot War lasted for almost ten years, but that most of the time was spent on preparations. The actual fighting spread over no more than two years (but see ad loc.) and these were 'at the end' (Γòν ἐπì πâσι. Oldfather translates 'in all' and Vial en tout, but 'at the end' is supported by 15. 31. 2, Γὴν δ̕ ἐπì πâσι; cf. Philostr. VS 2. 11, 12, 13). He dates the invasion itself under 386/5. The decision of the King to suppress Evagoras on the other hand is set under 391/0 (14. 98. 3f.), a mere six years earlier. Then there is Isocrates. In the Evagoras he says that the war lasted for ten years (9. 64) while in the Panegyricus he represents the siege of Salamis as still in progress at the time of writing and adds that the King had already wasted six years in trying to subdue Evagoras (4. 135, 141). The problem is to decide when the Panegyricus was written and this can only be done from the internal evidence of the speech itself. Section 126 provides the key:καì Γὴν Θηßαíων Kαδμεíαν καΓὴλαßον, καì νûν 'Oλυνϑíους καì Φλειασíους πολιορκοûσιν. The siege of Phlius was the most recent of the three events mentioned. It began in spring 380 and it lasted for some twenty months (19-23 n.). The spring of 380, therefore, is a terminus post quem for the speech. The Olympic festival of high summer 380 provides an approximate terminus ante quem, for the speech must surely have -143been completed by then if, as generally (and probably correctly) thought, it was composed for that occasion. Mathieu, 1938: 5, would date its publication between July and September 380. It is true that no specific evidence links the speech with the festival, but Tuplin's rejection of the association, 1983: 181f., is due to the fact that it does not fit his calculations either for the length of the war or for its ending, which he dates in late winter 381/0. This may be a little too early, I think. Nor should we insist, as he does, 1983: 178ff., on fitting exactly between c.390 and 380 the ten years the war is said to have lasted. There is no need to suppose that the essay was first published in 384, and that it was inadequately revised and redistributed in 380, the passages on Cyprus belonging to the first edition, those on the Cadmea, Olynthus, and Phlius to the second (as Reid, 1974: 141f.). Whatever the value of the ancient claims that Isocrates worked on

the speech for many years ( Plut. Mor. 837f; Dion. Hal. περì συνϑἐσεως ὀνομáΓων 208; ps.-Longinus περì ὕΨους 4. 2f.; Quintilian 10. 4. 4) Beloch is probably correct that no matter how long it took Isocrates to complete the speech he would have taken the trouble to remove all anachronisms at the time of publication ( GG iii. 2. 227). At any rate, Isocrates' testimony that Salamis was being besieged at the same time as Olynthus and Phlius receives strong support from Theopompus and the Egyptian kinglists. The sequence of events in Book 12 of Theopompus' Philippica is not entirely clear from the outline of its contents by Photius, but it is plain that according to Theopompus the Cypriot War did not reach its end till after Nectanebos I had assumed the kingship in Egypt ( FGH 115 F103. 10f.). We know that civil strife which lasted for several months followed Acoris' death, and that by the end of that Nectanebos was in control of the whole of Egypt (Kienitz, 1953: 88f., 173ff., 178ff.). Kienitz places the transition period, Acoris-Nepherites II-Nectanebos I, in the summer months of 380 because he is under the spell of Beloch's view that Cyprus was not invaded till 381. In reality Acoris' death and Nectanebos' accession may belong as early as 382 or as late as 380 (thus rightly Tuplin, 1983: 185f.) or even 379, if the pre-Kienitz chronology is followed (cf. e.g. Bickermann, 1934: 77ff.), though this is less likely. They cannot certainly be as early as 385/4 as Reid's chronology would necessitate (see below). That would be totally irreconcilable with the Greek as well as the Egyptian evi-144-

dence. (For a discussion of the latter see Kienitz, 1953: 178ff.; Johnson, 1974: 1ff.) We see, therefore, that the testimonies of Isocrates and Theopompus combine to bring the final stage of the Cypriot War down to the close of the 380s. What then of the evidence of D? Before we deal with this it will be necessary to look at what modern scholars have made of the problem. Two broad groups can be distinguished, those who accept D's date of 386/5 for the invasion, and those who do not. Most notable amongst the first is Judeich, 1892: 119ff., who is followed by, e.g., Miller, RE xviii. 1. 1165, Lenschau, RE xix. 2. 1847, and more recently Ruzicka, 1983. (Ruzicka's suggestions are very close to some of the views put forward here about Persia and the Cypriot War. It is therefore necessary to state that my views are the same as those included in my Oxford doctoral thesis which was submitted in 1981.) For Shrimpton, 1991, see the end of this n. Judeich further accepts 391/0 as correct for the beginning of the war and 381/0 for its end. He does not, however, see the statement διεΓη̑ χρóνον Γòν ἐπì πâσι (9. 2) as indicating the last two years of the war (see above), but merely 386 and 385. He is then left with the task of accounting for the years 384-381. This he does by suggesting that the King repudiated the treaty concluded between Orontes and Evagoras, replaced the former with Tiribazus who had meanwhile been acquitted, and peace was eventually made by the latter in 381. Unfortunately for Judeich his solution is unsupported by any hard evidence. Reid proposal, 1974: 123ff. (a revival of Engel, 1841: 7ff.), that the war began perhaps as early as 394/3, and that, therefore, 386/4 are indeed its last two years and D is correct, is wrong because she rather arbitrarily dismisses Isocrates (as Ruzicka, 1983: 106 n. 6, also points out) and overlooks Theopompus and the Egyptian evidence. Beloch, GG iii. 2. 226 ff., leads the second and dominant group. He is followed by e.g. Meyer, GdA v. 312; Swoboda, RE vi. 1. 825f.; ibid. vii. 1. 1432; Cary, CAH1 vi. 58; Spyridakis, 1935: 63 n. 1; Hill, 1940: 140 n. 2; Meloni, 1950: 321 and n. 3; Kienitz, 1953: 82ff., 82 n. 2, 85 n. 2; Osborne, 1973: 522ff. (who, however, thinks that 'the actual operations against Evagoras, despite the apparent indications in Diodorus' account, probably began in about 383 BC); Burn, CHI 369; Sealey, 1993: 47. Beloch understands the statement διεΓη̑ χρóνον Γòν ἐπì πâσι συνεχω̑ς πολεμηϑεíς (9. 2) to mean that the war was only fought in -145-

earnest in its last two years. If then the war began in 391/0, and this is about right as we shall see, and ended c.380, then obviously the last two years were not 386/5 and 385/4, but 381 and 380. Isocr. 4. 140 is the only source for an unsuccessful expedition mounted against Egypt by Abrocomas, Pharnabazus, and Tithraustes which lasted no less than three years. This event, Beloch thinks, belongs immediately after the King's Peace because Isocrates says that it preceded the attack on Evagoras, and also because Chabrias, who is generally thought to have held the command on the Egyptian side, is said to have gone to Egypt after having served in Cyprus ( Demosth. 20. 76). We know from X. H. 5. 1. 10 that Chabrias was sent to Cyprus in the year of Antalcidas' nauarchy (388/7) and he presumably remained there till the conclusion of the Peace forced him to leave and take service in Egypt under Acoris. Beloch consequently assigns this particular Persian attack on Egypt to the years 385-383. There then occurred the invasion of Cyprus in 381 and the Cypriot War was finally concluded in the late summer of 380 after the Olympic festival or at least after the publication of Isocrates' Panegyricus. Having failed to mention the campaign against Egypt D erroneously moved up the two closing years of the Cypriot War to 386/5 and 385/4. Beloch's chronology appears satisfactory at first glance, but on closer examination it seems to create as many problems as it solves. It is important to realize that D's evidence derives partly from the chronographer and partly from Ephorus, and that he himself may have manipulated the information with perhaps infelicitous results. The chronographer is to be seen first and indubitably at 15. 1. 6 -- the statement that the capture of Rhegion by Dionysius (and of Rome by the Gauls) occurred in the year before the Persian campaign against Evagoras signifies only one thing: D saw the two events entered under successive years in the chronographer and decided to conclude Book 14 with the first and to begin Book 15 with the second; hence 15. 2. 1 (under 386/5): ἐπì ςἑ ΓοúΓων 'AρΓαξἐρξης. . . ἐσΓρáΓευσεν ἐπ̕ Eὐαγóραν. . . ßασιλέα, which may in substance be the chronographic entry. This observation is of considerable significance for the chronology and weighty evidence to the contrary is required before the date of 386/5 for the expedition is set aside. Secondly, 15. 9. 2, ὁ μὲν οὐ̑ν Kυπριακòς δεκαεΓὴς σχεδòν γεγενημένος καì Γò πλέον Γοû χρóνου περì περì παρασκευàς ἀσχοληϑεíς, διεΓη̑ χρóνον Γòν ἐπì

πâσι συνεχω̑ς πολεμηϑεìς ΓοûΓον Γòν -146-

Γρóπον καΓελúϑη, is probably from the chronographer, in part at least. Its position in the middle of the narrative is rather unusual, though by no means unique. D presumably, coming to the end of the war, consulted his handbook and entered the relevant chronographic information at this point. He did the same at 15. 93. 1 (see n. ad loc.). It is a fair guess that the chronographer placed the ending of the war under 381/0 and that he said nothing more than that it had lasted for almost ten years. D then, who had just abridged from Ephorus under the successive years 386/5 and 385/4 the entire course of the war from the invasion of the island until Evagoras' capitulation, may have added the comment about the war being fought continuously for no more than two years at its very end. For a similar instance of possible Diodoran improvisation see 16. 63. 2 and pp. 41 f. If we could be sure that this is what happened the difficulties surrounding the chronology of the war would disappear forthwith. The possibility remains, however, that the chronographer (who was not infallible: pp. 45 ff.) is responsible for all the detail in 15. 9. 2 as well as for setting the ending of the war under 385/4. It is worth noticing that both Hieronymus and Syncellus misdate Evagoras' rebellion (Helm edn. 119; Schoene edn. ii. 110). The narrative of the war derives from Ephorus. It is found in four places: 14. 98. 1-4 (under 391/0); 14. 110. 5 (under 387/6); 15. 2-4 (under 386/5); 15. 8-9. 2 (under 385/4). The war had certain consequences and these are related at 15. 9. 3-11 and 18. It is possible that Ephorus gave no precise dates either for the beginning or for the end or indeed for any event of the war. Nevertheless, the detail he supplied is invaluable for the chronology. The passage 14. 98. 1-4 deals with the background to the war and its declaration. It reaches back as far as c.415/4 when Evagoras regained Salamis (Stylianou, 1989b: 458ff.) and it comes down to the time when Hecatomnus of Caria was ordered to fight against Evagoras. The placing of this narrative under 391/0 is probably due to the chronographer rather than to any sequence of abridged Ephoran material (F76 owes its present position to Jacoby. The origins and outbreak of the Cypriot War could have been related in any number of places in Book 19). It seems to be correct for the beginning of the war, as this was marked, on the one hand, by the order to Hecatomnus and Autophradates of Lydia to campaign against Evagoras (and these two were only appointed to their -147-

satrapies following the failure of the proposed Peace of 392/1: FGH 115 F103. 4; 328 F149; 90. 3n.), and on the other, by the dispatch to Cyprus of the first of three Athenian naval squadrons (Stylianou, 1988). Section 14. 110. 5 (under 387/6), on the other hand, undoubtedly retains its original position in the sequence of Ephoran narrative-with the Peace of Antalcidas in effect, we are told, the King immediately turned his attention to the Cypriot War and made preparations to crush Evagoras. Unless, therefore, there is evidence to the contrary we should keep the invasion as close in date to 387/6 as possible. Not only is there no such evidence, but the chronographer, as we have seen, by setting the invasion under 386/5, confirms Ephorus. What is more, the following sections contain information which further supports a mid-380s date for the invasion. In deciding to revolt Glos, the Persian admiral, formed alliances with Acoris of Egypt and with Sparta (15. 9. 4 and nn.). After his murder (he must have moved to Ionia by then: 9. 4n.) his successor Tachos continued along the same lines of trying to elicit Egyptian and Spartan support against Persia. He did not long outlive Glos and the last sentence of ch. 18 runs: ΓοúΓων δὲ πραχϑένΓων αἱ μ ν καΓà Γὴν 'Aσíαν ἐπανασΓáσεις καΓελúϑησαν. And in the following chapter it is explained that after the deaths of Glos and Tachos the Spartans gave up their plans for new conquests in Asia and concentrated on Greece. They shortly afterwards moved against Olynthus -- summer 382. Both chs. 18 and 19 are under 383/2. It would be reckless not to pay attention to what D (i.e. Ephorus) says. The revolts of Glos and Tachos were over by early 382 at the latest and hence the whole sequence of events, from the invasion of Cyprus to the death of Tachos, should be placed earlier than that date. How long did this sequence take to run its course? It is highly probable that D has telescoped the later stages of the war. The events which he dates to 386/5 and 385/4, perhaps under the guidance of the chronographer (see above), may have taken considerably longer than two years. What is more, the narrative does not flow uniformly in terms of either time or geography. After concluding the Cypriot War at 15. 9. 2 D goes back in time to recount the revolt of Glos; and the scene gradually shifts from Cyprus to Ionia. The shift in chronology and geo-148satrapies following the failure of the proposed Peace of 392/1: FGH 115 F103. 4; 328 F149; 90. 3n.), and on the other, by the dispatch to Cyprus of the first of three Athenian naval squadrons (Stylianou, 1988). Section 14. 110. 5 (under 387/6), on the other hand, undoubtedly retains its original position in the sequence of Ephoran narrative-with the Peace of Antalcidas in effect, we are told, the King immediately turned his attention to the Cypriot War and made preparations to crush Evagoras. Unless, therefore, there is evidence to the contrary we should keep the invasion as close in date to 387/6 as possible. Not only is there no such evidence, but the chronographer, as we have seen, by setting the invasion under 386/5, confirms Ephorus. What is more, the following sections contain information which further supports a mid-380s date for the invasion. In deciding to revolt Glos, the Persian admiral, formed alliances with Acoris of Egypt and with Sparta (15. 9. 4 and nn.). After his murder (he must have moved to Ionia by then: 9. 4n.) his successor Tachos continued along the same lines of trying to elicit Egyptian and Spartan support against Persia. He did not long outlive Glos and the last sentence of ch. 18 runs: ΓοúΓων δὲ πραχϑένΓων αἱ μ ν καΓà Γὴν 'Aσíαν ἐπανασΓáσεις καΓελúϑησαν. And in the following chapter it is explained that after the deaths of Glos and Tachos the Spartans gave up their plans for new

conquests in Asia and concentrated on Greece. They shortly afterwards moved against Olynthus -- summer 382. Both chs. 18 and 19 are under 383/2. It would be reckless not to pay attention to what D (i.e. Ephorus) says. The revolts of Glos and Tachos were over by early 382 at the latest and hence the whole sequence of events, from the invasion of Cyprus to the death of Tachos, should be placed earlier than that date. How long did this sequence take to run its course? It is highly probable that D has telescoped the later stages of the war. The events which he dates to 386/5 and 385/4, perhaps under the guidance of the chronographer (see above), may have taken considerably longer than two years. What is more, the narrative does not flow uniformly in terms of either time or geography. After concluding the Cypriot War at 15. 9. 2 D goes back in time to recount the revolt of Glos; and the scene gradually shifts from Cyprus to Ionia. The shift in chronology and geo-148have gone up to the King already by 389 (the year Thrasybulus sailed out: Stylianou, 1988: 467ff., esp. 470 n. 33); which would have left him with plenty of time to marry and campaign against Egypt. Besides, the preparations may have been well under way by the time Pharnabazus joined the other two commanders. It would appear that Abrocomas had been put in charge of operations against Egypt as early as 401 (X. An. 1. 4. 3, 5; 2. 1. 14). In fact, what we know about Chabrias' activities in Egypt tends to exclude the second half of the 380s as the time of the attack on Egypt. Chabrias may have been in Egypt continuously from 386 till 380/79 when he was recalled (cf. 29. 2n.), but there is no evidence to connect him with the repulse of a Persian invasion. (Thus also Ruzicka, 1983: 106 n. 7.) Demosth. 20.76, καì ὅσ' ἐυΚύπρῳ τρόπαι' ἔστησευ καì μετà ταûτ' ἐν ΑìYύπU=1FF3, is imprecise and does not have to mean anything more than that Chabrias served well in both Cyprus and Egypt. In Cyprus we are told that he helped Evagoras to subdue the island ( Nepos, Chabr. 2. 2; cf. X. H. 5. 1. 10). In Egypt he may well have helped Nectanebos establish himself as king, and that had perhaps involved sufficient fighting to justify Demosthenes' expression ( Nepos, Chabr. 2. 1; Kienitz, 1953: 88f.). Again, though Nepos, Chabr. 2, is useless chronologically, it makes no mention of a Persian campaign against Egypt at the time of Chabrias' sojourn there. Consequently Kienitz, 1953: 85, is largely conjectural. Indeed, his dismissal of the view that this particular Persian attack belongs before 386, on the ground that such a placing is inconsistent with the known dates of Chabrias (1953: 85 n. 2), is simply begging the question for as we have said nothing in the sources connects Chabrias with the defeat of a Persian expedition. And we might also consider the following. When Pharnabazus began preparing for a second attempt on Egypt he lost no time in confronting Athens with the terms of the King's Peace and demanding the recall of Chabrias (29. 2n.). Now Chabrias presumably was in Egypt between 385 and 383 and if those are the years of the first attempt ( Beloch, GG iii. 2. 228f.; Shrimpton, 1991: 1ff..) why did Pharnabazus not remonstrate with Athens then and demand Chabrias' recall? Somewhat firmer evidence is provided by D 15. 29. 1-4 which passage, we have argued (p. 100 ), is a condensed Ephoran review of what had been happening in Egypt from before the death of -150-

Acoris, probably from the King's Peace even, down to the dis-

2). He was recalled and he did not therefore have to face a Persian attack. All he did in Egypt was to prepare. Arguments from silence are of course dangerous. All the same it is a little difficult to believe that had Chabrias achieved such a notable victory against Persia in Egypt, all the sources, Isocrates, D, Nepos, and especially Demosthenes at 20. 76, would have failed to mention it. The oikonomia of Book 12 of Theopompus as summarized by Photius ( FGH 115 F103) provides an additional argument in favour of D's date. After relating the beginning of the Cypriot War (103. 4), Theopornpus commented on the effect of the King's Peace on the situation; the King was now free to concentrate his forces against Evagoras. This is precisely what Ephorus said too (14. 110. 5). This particular narrative of Theopompus reached down to at least the naval defeat of Evagoras (103. 5-6). Then, and the sequence here is similar to that in D, Theopompus turned his attention to the Greek mainland where he noted that, while Athens adhered to the terms of the Peace, Sparta did not (103. 7). One cannot but relate this statement to D 15. 5 (cf. Jacoby, FGH iid. 373). One of the Spartan victims was Mantinea. The Mantineans had appealed to Athens, but their appeal was refused, the Athenians being unwilling to act contrary to the Common Peace (15. 5. 5 and n.). Theopompus went on to describe how the King's Peace had been concluded. He then returned to the Cypriot War and related it to its end. D's arrangement is broadly the same. Now Mantinea was attacked in 385 (5. 1n.) and we see that Theopompus placed the invasion of Cyprus before it. Last, but not least, it is surely of some significance that one of the two Persian commanders was Tiribazus, the satrap of Ionia and the man in charge of negotiations with Greece (90. 3n.). The result of these negotiations was the King's Peace in 387/6 from which Cyprus was specifically excluded. And Tiribazus' army and navy came from Ionia (see further 2. 2n.). The year 386/5 is therefore a much better date for the attack on Evagoras than 381. The war ended not earlier than spring/summer 380. The evidence of Isocrates (and Theopompus) is decisive on this point. While the chronographer therefore is very probably correct about -151-

the date of the invasion, either he or D (see above) is certainly wrong that the war was concluded in 385/4 after two years of continuous fighting. On the other hand Isocrates is partly wrong too -Tiribazus was not in Cyprus in 381/o; this is contradicted by Ephorus, as we have seen. The following chronology is therefore proposed. Phase 1 can easily have required several months. We do not know how long the armada spent in Cilicia, but conditions there may have enforced a fairly lengthy stay (2. 2n.). And then Evagoras' pirate campaign, the famine at the Persian camp, the mutiny and its suppression, and the improvement of the supply situation may have taken up the rest of the campaigning season of 385 so that the naval battle may well belong to the summer of 384. Then followed the Persian concentration at Citium and the beginning of the siege of Salamis which cannot have been very effective. Tiribazus had gone up to the King after the battle and he cannot have returned before some, say, three to four months had elapsed. Thus clearly a winter, that of 384/3, should be envisaged at stage 9. The money Tiribazus had brought with him and his own presence made a difference and Evagoras found the siege actively pursued on his return from Egypt. I would see stages 11, 12, 13, and 14 as belonging to 383 with early 382 as the absolute lowest limit. Thus phase 15, treated very briefly by D, must have endured for a considerable time. Phase 17 belongs last because of 10. 2 -- the war had already been concluded.In a recent paper (1991: 1ff.) Shrimpton dates both the invasion of Cyprus and that of Egypt in 387/6 immediately after the conclusion of the King's Peace and argues that strategic considerations demanded the simultaneity of the two campaigns. While I agree, and have argued so, that the attack on Evagoras followed on the conclusion of the King's Peace, I cannot go along with the rest of Shrimpton's suggestions: (1) The year 387/6 for the invasion of Cyprus flatly contradicts the chronographer whose dating, we have suggested, deserves -153

respect. Also, by setting both the sailing of the expedition and the naval action in the first half of the campaigning season of 386 Shrimpton leaves too little time for the sequence of events in 15. 2. 2-3-4. (ii) The sources are silent about a campaign against Egypt in the 38os. The only source which is not (Isocr. 4. 140f.) dates it before the invasion of Cyprus. Shrimpton adduces Justin 6. 6 as evidence for an attack on Egypt after the King's Peace. But Justin 6. 6 seems rather to correspond to D 15. 38 and to refer to the Peace of 375. (iii) Shrimpton's analysis of the strategy required for a successful campaign against Egypt cannot be faulted and one sees such a strategy being put to good use in the 340s (as Shrimpton explains). The obvious retort, however, is that the campaign of the 380s did after all fail. All the same, by dating the campaign before the King's Peace we do not necessarily impugn the strategic sense of the Persian high command. For Shrimpton has overlooked the attempt to deal with Evagoras earlier by Autophradates and Hecatomnus and which may have been intended to precede the assault on Egypt or even to be simultaneous with it. 2. 1. πολὺ δὲ ξρόνον. . . παρασκευáς : it is possible that the King had begun preparations of his own (as opposed to those of Autophradates and Hecatomnus in the west) right at the start of the war (14. 98. 4), but that these had progressed in a desultory fashion. Something is clearly amiss at 14. 98. 4. οûτος δὲ τàς ἐν ταîς ἄνω σατραπόλαις ἐπορευόμεγα+́λαι ςυνáμεσι ςιαßαíνει εἰς τὴ Κύπρον. As it stands with οτος as the subject of ἐπιπορευόμενος and διαßαíνει and taken to refer to Hecatomnus, the sentence makes no sense. It cannot be that 'Hecatomnus traversed the cities of the upper satrapies and crossed over to Cyprus in strong force' (thus Oldfather)! The King must be the subject of ἐπιπορευόμενος and indeed all the MSS read αὐτός (οûτος is due to Dindorf). It is possible that there is a lacuna either between ἐπι2πορευόμενος and μεγáλαις δυνα+́μεσι or, more likely, between the latter and ἐπιπορευόμεUι. As for the subject of διαßαíνει, whatever that is, it cannot be Hecatomnus, and perhaps the infinitive δ2ιαßαíνειν stood in the text originally, denoting intent. I would see μεγα+́λαις δυνα+́μεσι as a reference to those mentioned in this section, 2. 1, which were now about to cross into Cyprus from Cilicia. Frustratingly lengthy -154-

Persian preparations are not unknown in the 4th century (cf. 41. 2n.). Reid's suggestion (1974: 136) that Ephorus had looked ahead to the campaign which was eventually launched at this point in his narrative, and that D confused the reference, is perfectly possible of course. See also Hornblower, 1982a: 37 n. 10. τὸ μὲν γ°ρ π2εζν ... ριακοσι+́ων: there can be little doubt that the army is monstrously exaggerated. Not so the navy. Evagoras eventually possessed 200 triremes and he may still have been outnumbered. But see pp. 130 ff. 2. 2. Ο+̔ρόνταν κηδεστήν: the MSS disagree over the exact form of the name; cf. Vial, 5 and 121. OGIS391, 392 refer to him as Α+̔ροΛ+́νδης; see 391n. Orontes, the son of Artasouras the King's 'Eye' at Cunaxa ( Plut. "Artox". 12. 1 with OGIS264, 391, 392), had a long and eventful career stretching from the end of the 5th century to at least the Great Satrap Revolt. See Osborne, 1973: 515ff. and n. 34 for his family and early career. His high status and the favour he carried with Artaxerxes are attested by the fact that he was given a daughter of the King in marriage (X. An. 2. 4. 8; 3. 4. 13; OGIS391, 392). τῃς δὲ ναυτικῃς 2Τιρι+́ßαζον: what was Tiribazus' precise position? He is here said to have been in command of the fleet and there is no hint that he was in any way superior to Orontes. At 3. 2 however Glos is said to be the ἡγούμενος τῃς ναυτικῃς δυνα+́μεω2ς, a statement repeated at 3. 6, τον+̑ τε ναα+́ρξου τω+̑τ Περσω+̑ν Γλω+̑ (cf. 9. 3; 18. 1). This would leave Tiribazus redundant but for 8. 2 where he is reintroduced (he had been away to the King) as τὴ ν ἡγεμονι+́αν. In that position he began his negotiations with Evagoras. One's immediate impression is that there had been a command reshuffle at some time and Spyridakis, 1935: 66, suggests that Tiribazus' appointment to the supreme command was subsequent to the naval victory and the result of his visit to the King. Osborne, 1973: 526 n. 59, objects that 'it is unlikely that one of the two should not have held the overall command from the outset' and adds that 'of the coins minted in Cilicia for this campaign none have been found bearing Orontes' name, whereas a number have been found with Tiribazus' name'. Nevertheless, we hear of other Persian campaigns which were not conducted

under an overall commander. For example, Datis and Artaphernes in 490 appear to have been equal, and Isocrates (4. 140) mentions Abrocomas, Tithraustes, and Pharnabazus, but does not specify -155-

an overall commander though admittedly his authority is poor in this instance. But at the Granicus in 334 the Persians had no commander-in-chief ( Arrian, An. 1. 12. 8). Memnon was appointed so later ( ibid. 1. 20. 3 ; 2. 1. 1). Fear of revolts must have been the main reason for avoiding single commands and only when the military necessities demanded it, i.e. after the dual or multiple commands had failed, would a single or overall command be risked. Thus after Tithraustes, Pharnabazus, and Abrocomas had failed in Egypt we find Pharnabazus seemingly in single command (15. 29. 3; 41. 1-2), and Datames was made the chief commander in turn following Pharnabazus' lack of success (but see 29. 3n.). Memnon too was appointed commander-in-chief in the west at a time of crisis. Thus one could plausibly maintain that on meeting the King Tiribazus had explained the earlier reverses in the campaign as the result of ineffective leadership due to the dual command. (It would seem that he was just the man to speak his mind, cf. Plut. Artox. 24. 2.) Convinced, Artaxerxes sent him down with 2,000 talents and the supreme command. The evidence of the coins is not against this. Those bearing Tiribazus' name (perhaps all of them) may have been minted after Tiribazus' return. To argue entirely from D's terminology is dangerous for it is certainly defective. If Tiribazus was in overall command from the beginning why is he only referred to as ατρατηγὸς τῃς ναυτικῃς δυνα+́μεως at 2. 2? And is this not contradicted by 3. 2, 6 where Glos is termed a nauarch? A way out is to assume that Orontes and Tiribazus were originally appointed joint and equal στρατηωοι+́. As Tiribazus already had a fleet in the Aegean his command was specified as naval, while Orontes' command obviously had to be of the land forces; he was in his inland province of Armenia or at the court at the time of his appointment. (The circumstances were similar perhaps to those of Autophradates appointed στρατηωός and Hecatomnus appointed ναύαρξος in 391/o: FGH 115 (Theop.) F103. 4.) In fact, both of them recruited naval as well as land forces (see below) and while still in Ionia Tiribazus had probably appointed Glos nauarch of his fleet. The title stuck and it is in this light that we should see 3. 2, 6 perhaps. After Tiribazus' return as supreme commander, Glos was quite likely appointed to command the entire naval force and, again, his later title may be reflected in the terms at 3. 2, 6. It is possible that local tradition at -156-

Cyme knew him as 'Glos, the Persian admiral in the Cypriot War' (18. 1. For Ephorus' local Cymaean sources see p. 106 ). So Ephorus himself may not have been very clear about the exact line of command. Section 14. 35. 3 (under 400/399), where Glos is introduced as the man who μετὰ δέ τινας ξρóνουσ ἀΦΦηωησαμένου τω+̑ν ßασιλικω+̑ν δυν2lα+́μεωνserves as additional warning that we must be very careful with the terminology in D. Just how much responsibility should be laid at Ephorus' door is an issue which cannot be decided. Aeneas Tacticus, writing about the middle of the 4th century, refers to Glos as 'the King's nauarch' (31-35). For the MSS variants of Tiribazus' name see Vial, 122; Meloni, 1950: 292 n. 1. For the career of Tiribazus, Schaefer, RE via. 1. 1430ff.; Meloni, 1950: 292ff. Tiribazus had almost certainly been a subordinate of Orontes in Armenia in 401/0 ( Osborne, 1973: 518f.) and the combination of the two in this instance, whether Tiribazus was in overall command from the beginning or not, was potentially explosive. Tiribazus' appointment is hardly surprising in view of his key and successful role in the manuvres and negotiations which imposed the King's Peace on Greece (X. H. 5. 1. 25ff.). μεγα+́λης ἀποδοχῃσ 2τυγχα+́νοντα: a stock phrase; cf. 11. 40. 4; 12. 20. 1; 12. 42. 8; 12. 43. 3; 15. 7. 1; 15. 35. 2; 15. 36. 6; 15. 71 7; 16. 14. 1; 18. 36. 6. But Tiribazus must have been the man of the moment. For D's language see pp. 15 ff. ἐν Φωκαι+́ᾳ κα2io+́ὶ Κύμ:why in Ionia? For three possible reasons. First, the fleet used by Sparta and Persia in 388/7 to force Athens to consent to the King's Peace (X. H. 5. 1. 25ff.) probably formed the kernel of the fleet which sailed against Evagoras. Secondly, Greek mercenaries were by now a regular feature of Persian armies and Ionia was the most convenient area for their recruitment. The Peace would have facilitated this as did its first renewal in 375 (38. 1). Thirdly, Cilicia and Phoenicia, which under normal circumstances would have formed the natural centre for the gathering of a Persian naval force, are said to have been largely in revolt though the extent to which this was so is impossible to gauge: 2. 4; Isocr. 4. 161; 9. 62. But should Isocrates be exaggerating, the first two reasons are enough to explain why the expedition set out from Ionia. Similarly in 391/0 Caria and Lydia were assigned the task of crushing Evagoras ( FGH 115 (Theop.) F103. 4), and in the late 350s Idrieus of Caria was ordered to gather together an army and -157-

a navy and invade Cyprus (16. 42. 5ff.). In both instances Cyprus and a part of Phoenicia were in revolt. All the same, D has certainly misrepresented the situation. The entire Persian army and fleet cannot have collected at Cyme and Phocaea before setting out, but only the Greek mercenary corps and a substantial part of the navy. These sailed out under Tiribazus and reached Cilicia where they were met by Orontes with the Persian army proper and doubtless a Phoenician naval contingent. The size of the Persian fleet was considerable and could not all come from Ionia (2. 1n.). These latter forces should be seen as the result of the preparations begun in 391/0 (14. 98. 4. Armenia was one of the upper satrapies). The Greek element in the Persian armament is stressed by Isocr. 4. 135; cf. 124; Polyaen. 7. 20.

Why at Phocaea and Cyme? At first glance one suspects another instance of Ephorus allowing his local patriotism to pervert historical truth (see pp. 125 f.). Suspicion is heightened by the mention of Phocaea and Cyme at 11. 2. 3 -- Xerxes ordered his fleet to assemble at Phocaea and Cyme prior to the crossing of the Hellespont, and at 13. 100. 4 -- after the battle of Arginusae the coastline of Cyme and Phocaea was filled with corpses and wrecks. These statements are not corroborated by any other source and modern commentators have tended to disbelieve them. Unnecessarily though, as Samuel, 1968: 379ff., shows. Cyme and Phocaea were situated on the coast almost directly west of Sardis, Xerxes' base before he set out for Greece. They were at the head of the large bay of Smyrna and Clazomenae. The latter was one of the two islands (the other being of course Cyprus) demanded and obtained by the King in 387/6 (X. H. 5. 1. 31), and the reasons were undoubtedly military, as Ruzicka stresses (1983: 107f.). There is no difficulty therefore in accepting that both in 480 and in 386/5 the Persian fleet assembled at Cyme and Phocaea. The question whether Ephesus was in Persian hands at this time ( Beloch, GG iii. 1. 97 n. 3; cf. Meloni, 1950: 314f.) is irrelevant. Cyme and Phocaea and their immediate neighbourhood were more than sufficient for the purpose of assembling a fleet. Κιλικι+́αν: here they met up with Orontes and the other 'half' of the force (see previous n. This point is generally missed by scholars who assume that Orontes and Tiribazus set out together from Ionia). How long did they stay? If Cilicia was really mostly in -158-

revolt as Isocrates represents (4.161;9.62) then the Persians' primary task would have been the subjection of as much of the area as would enable it to be safely used as a base of operations. This task was almost certainly carried out at this time (cf. Hill, 1940: 137 and n.3), pace Osborne, 1973:527:'the failure to hold a secure base in Cilicia at the outset led to a crisis over provisions soon after the troops arrived in Cyprus.' The crisis in question had different causes, see 3. 1n. The Persian presence in Cilicia is illustrated by coins issued at mints at Issus, Mallus, Soli, and Tarsus, many with Tiribazus' name in Aramaic; see Head, 1911: 722, 724, 728, 730; Hill, 1940: 137 and n. 3; Osborne, 1973: 527 and n. 65. καὶ ΠεαιωΘὲντες. . . τòν πολεμον: this stage of the war is in total darkness. It is possible that Soli, Amathus, and Citium were still holding out against Evagoras (2. 4n.; 14. 98. 2=FGH 70 F76. The Ωτιεις of the fragment is certainly a corruption: Hill, 1940:132 n. 2; Reid, 1974: 124ff.) and that the Persians headed for one or all of them. (On Citium see further 3. 4 and n.) It is reasonable to assume that after the Persians had secured a base on the island, and as Evagoras avoided a battle, they began to dislodge from his hold city after city with the aim of isolating Salamis. Their progress was interrupted for a while by Evagoras' tactics (see below). We may compare Demetrius' campaign of 306. Having sailed from the Aegean he made for Cilicia where he augmented his force, as the Persians had done, with ships and men before crossing into Cyprus and landing on the coast of the Carpass peninsula: 20. 47. 1f.

2. 3-4. Evagoras' resources 2. 3. 'Γκοριν. . . συμμαχίαν ἐποιńσατο: Acoris of the 29th dynasty was king from c.394/3 till 382/0 (see 2-4n.). His alliance with Evagoras cannot be closely dated but it probably belongs before 386; cf. FGH 115 (Theop.) F103. 1, 13. All of these alliances were aimed against Persia: Svt 237. Acoris had also allied himself with Athens (Svt 236) as had Evagoras (Svt 234); cf. Stylianou, 1988: 469ff. δύναμιν ἀΞι: D usage, see p. 133. 'Eκατóμνου . . . δυνμεων: Hecatomnus of Caria and Autophradates of Lydia had been ordered at the outbreak of the war to begin hostilities against Evagoras (14. 98. 3; FGH 115 (Theop.) F 103. 4) -159F103. 4). Little if anything had been achieved, probably because Hecatomnus was already in secret league with Evagoras. The reason why Hecatomnus should want Evagoras to succeed is not hard to find. He was himself a native dynast, but ruling only because the King suffered him to do so. Ideally he would want to be free. The more thorns in the side of Persia, therefore, the better. For Hecatomnus see Judeich, 1892: 233ff.; Böckisch, 1969: 133ff., but especially Hornblower, 1982 a. Also now Ruzicka, 1992: 15ff. ὁμοιως δὲ καί πρòς τοὺς ἄλλοΥς κτλ: from our other sources we may mention the cities of Cilicia (Isocr. 4. 161; 9. 62) and perhaps Pisidia (FGH 115 (Theop.) F103. 13). Isocr. 4. 161f. is at pains to demonstrate that the Mediterranean provinces of Persia were festering with revolt. But if there was any specific information in Ephorus D's generalization has rendered it irrecoverable. 2. 4. σχεδóν τι πασων : cf. Isocr. 9. 62. Some Cypriot cities may still have been holding out; see Stylianou, 1989b: 471ff. TύροΥ: Isocr. 9. 62; Stylianou, 1989b: 473. και τινων ἑτέρων: two inscriptions of Acoris, Evagoras' ally, have been found at Sidon and Ace: Stern, 1982: 254, 255 fig. 380, 278 n. 20. ει+̥χε. . . δαψίλειαν: the trireme figures are uncorroborated, but they are quite credible; see p. 131. Isocr. 4. 141 is doubtless an exaggeration and in any case Evagoras had by then been defeated and was under siege. We do not know whether the troops which had gone out with Chabrias (X. H. 5. 5. 1. 10; cf. Demosth. 20. 76) were withdrawn at the conclusion of the King's Peace. There are hints that Athenian officers serving in Cyprus may have strongly disapproved of their city's acceptance of the King's Peace and consequent abandonment of Evagoras: Lysias19. 7; Harpocr. s.v. Xύτροι; Meyer, GdA v. 265; Beloch, GG iii. 1. 141; Hill, 1940: 134f. and 135 n. I; Stylianou, 1988: 463 ff.

ὁ ττ+̑ν 'A↔άßων ßασιλεó3ς: the MSS read ßαρßα+̠ρων which is clearly unacceptable and hence corrupt. 'Γρα+̠ων is Rhodoman's emendation (accepted by both Vogel and Vial). Sievers emended Bαρκαίων from FGH 115 (Theop.) F103. 1 (cf. 18. 20. 3). But Barca or Barce in Cyrene appears to have been a republic at this time (cf. Sethe, RE iii. I. 20; Hill, 1940: 136 n. 3). Two points are clear: Evagoras' ally was a king, and he sent troops. At 13. 46. 6 (under 410/9) D refers to a king, ττ+̑ν 'Γρα+̠ων who along with a king of Egypt was said to ἐπιßουλευειν τοις περὸ Φοινίκðν πράγμαιν. Sections 13. 46. 6 -160and 15. 2. 4 probably refer to the same person: Stylianou, 1989b: 462 n. 401, 473. καὶ ἄλÁοι τκνές: see 2. 3nn. 3. 1. λἤστρικάς ἒχων ναυ+ς κτλ: cf. 3. 43. 5; 15. 95. 14; 16. 5. 3; 82. 3. What type of warship was meant we cannot say. Burn, CHI369, conjectures that the piratical raids may have been carried out by Pisidians. Whatever the truth, Evagoras succeeded in severing the Persian supply lines. The ensuing famine caused a mutiny at the enemy camp which was only suppressed with difficulty. The entire Persian fleet now sailed back to Cilicia and brought over enough foodstuff to last them for a considerable while; cf. Hill, 1940: 137. The reference to the supplying of the Persian army is of some interest, but this use of the word ἀγορά is fairly common in D and in contexts from different sources; see McDougall s.v. 3. 2. ὁ τη+π+ς. . . Τλτ+̑: The MSS read τατ+̑ and τατ+̑ here and at 3. 6, but τατ+̑ and τατ+̑ at 9. 3. The emendation Glo/Glos is Wesselings' (cf. Vial's n. on p. 122 ). Glos was a son of Tamos, an officer of Cyrus, and had himself served Cyrus (14. 19. 5f.; 14. 35. 3ff.; X. An. 1. 4. 2; 1. 4. 16; 2. 1. 3). Glos' own name appears to be Carian (cf. Beloch, GG iii. I. 99 n. I) and was himself perhaps regarded as a Carian (Athen. 6. 256c -- if this is the same person), though his father, according to X. An. 1. 4. 2, was an Egyptian from Ephesus (cf. HCT v. 74). Polyaen. 7. 20 may refer to the present mutiny. Athen. 6. 256c is quoting Clearchus of Soli, a Cypriot: γεγονασι δὲ παρ ήμι+̂ν καὲ ἐπὲ Τλου+ του+ Καρος κτλ. It would appear that Glos made quite an impression locally to be remembered two generations later, but the statement is entirely mysterious. See 2. 2 n. 3. 4. ἑΞὲκοντα. . . διακοσίας: see 2. 4 and n. παραπλἑοντος: '(as the royal fleet) was sailing past (or along) on its way to Citium.' It seems that we should envisage the Persian fleet as sailing by Salamis in the direction of Citium when suddenly Evagoras sailed out and attacked. Where had the Persians been? Somewhere on the north coast of the island, we may guess, or the Carpass peninsula, or the neighbourhood of Salamis itself, for had they been at, say, Soli or Amathus, they would have reached Citium from the west. What was the status of Citium at this juncture? Were the Persians sailing to a friendly city or to a hostile one to capture it? -161-

If the latter were the case then Evagoras' naval attack was perhaps in defence of the city. But this is not a necessary conclusion. The theory of Babelon, 1893: pp. cxxxif., that Demonicus, son of Hipponicus, to whom Isocrates' first essay is addressed, ruled Citium at this time is now known to be false; the coins on which it was based belong to Lapethos (cf. Jenkins, 1972: 135). Nor can our passage by itself prove that Citium was on the Persian side at this time (as Beloch, GG iii. 2. 228), though this is the more likely assumption. After all, we do know that Evagoras did not quite succeed in conquering the whole of Cyprus (2. 4), and Citium was the strongest non-Greek city on the island. At 4. 1 it is said that οἱ δὲ Πέρσαι τέ+ῃ+̑ νιήσαντεσ εἰς Κίτιον πολιν ἀμΦοτέρας τὰς δυνάμεις ἣΘροισαÁ. πολιν is the reading of MSS P and X. M and F, on the other hand, read πολιν (Vial, app. crit. She prints πολιν like Vogel). If we could be sure that this was the correct reading we might venture the following reconstruction of the Persian moves. Both their army and fleet were stationed at Citium some time prior to the battle. The fleet then sailed out on some operation or other. On its return, as it sailed past Salamis, it was attacked by Evagoras. After the battle both Persian fleet and army were again together at Citium.

3. 5-6. The sea battle For the date see the discussion above. Isocr. 4. 141 and FGH 115 (Theop.) F103. 6 refer to the battle without adding any detail, and the detail which Ephorus undoubtedly gave has been omitted by D (see p. 125 ). All the same, brief though D is, his account is sensible enough. Evagoras gained an initial advantage because his own triremes were in line of battle whereas the enemy's were not, and also because while his own men were psychologically prepared for the fight, his opponents had been taken by surprise. Once, however, Evagoras' impetus had been spent, Glos counterattacked and carried the day. 3. 5. ἀΘροαις . . . ται+̠ς τριήρεσιν: 'with the triremes in close order.' Cf. 13. 9. 6; 13. 40. 1; 13. 71. 4; 14. 83. 6. 3. 6. ναυμαχία καρτερά: certainly a D expression, see p. 15. μετά του ßάρους: D frequently uses ßάρος in the descriptions of land battles; cf. 15. 55. 4; 15. 86. 5; 17. 11. 4; and McDougall s.v. He seems to mean chiefly the weight of a formation, be it infantry, -162-

cavalry, or elephants, in particular the pressure exerted by a dense array of men or horses. At Leuctra, for example, the sheer mass of the Theban phalanx pushed the thinner Spartan line right off the field and caused it to disintegrate (15. 55. 4 and Appendix). ßάρος cannot obviously mean the same in a naval engagement, for a ship could not add its weight to that of the friendly ship fighting before it. It must rather refer to the order and cohesion of a powerful battle formation which made its onslaught invincible. Evagoras' fleet must have become disorganized eventually and Glos, attacking in close order, gained a victory; cf. 20. 52. 3 (the sea battle off Salamis in 306), μετά ßάρους ἐπιΦεομένους. We should also see the word as conveying the notion of a steadfast and purposeful attacking movement, whether in the context of a land or a naval engagement. 4. 1. εἰς Κίτον[πάλιν]: see 3. 4n. ἐπορΘουν: 'they were besieging' or 'they were endeavouring to destroy' the city. Either is a possible (though not the usual) meaning of the word. 4. 2. δισχίΛιΛ τάΛαντα: cf. Isocr. 9. 60: εἰς τήν στρατείαν ταύτðν πλέον ἤ τάλαντα π ντακισχίλια καί μύρια κατðνάλωσεν (ßασιλεύς). He means the total expenses in the war. If the figure is exaggerated it cannot be so by much considering the length of the war as a whole and of the siege in particular. The 2,000 talents brought down by Tiribazus were probably spent very quickly in back pay and in the unsuccessful operations against the city in 383. πεζἧ+̨ σχίλια τßαΛὼν. . . προετέρðσε: the action, a minor one involving only a part of the Persian force, was perhaps fought on the coast somewhere between Salamis and Citium where the Persian army was possibly stationed already. See 3. 4n.; cf. Stylianou, 1989b: 477 n. 450. 4. 3. Πνυταγορν: Πυϑαγοραν omnes codices. The emendation is Wesseling's, confirmed by Isocr. 9. 62 and FGH 115 (Theop.) F103. 12. Cf. Tod 194: ΠνυτΛγορας, the later king of the same dynasty. νυκτος ἒλαΘε: on returning Evagoras found the siege energetically pursued (8. 1). But at this moment it seems, and with Tiribazus away (4. 2), the siege was not a close one. -163-

5. The Spartan attack on Mantinea 5. 1. ἅμα δὲ τούτοις πραττομένοις: the synchronism is correct. Sparta moved against Mantinea in the summer of 385 under Mystichides (386/5); see p. 137 and D's statement at 5. 3 that the Spartans οὐδε δὐο ἒτð ϕυλάΞαντες τάς κοινὰς σπονδάς, which likewise argues for 385. The King's Peace was concluded between late 387 and early 386; cf. Beloch, GG iii. 2. 225f. οὐδεν ϕρντίσανες τω+̑ν. . . σποδνδω+̑ν: the Spartans and their supporters of course saw things very differently; see below ad 5. 2-3. διὰ τοιαύτας αἰτίας: for this link-phrase see pp. 44 f. The chronographic entry on Mantinea would thus have come under Mystichides. κοÓνὴς εἰρήνὴς: the King's Peace of the previous year, 387/6 (14. 110. 2-4). On the term itself, which D doubtless derived from Ephorus, see Martin, 1944: 13ff.; Ryder, 1965: pp. xiff. Though sometimes so stated, Ephorus' use of the term in connection with the Peace of 387/6 was neither anachronistic (see Andoc. 3. 17) nor necessarily inappropriate. But this depends on the meaning attached to the word koine in the context of the Peace, a matter still occasioning debate. On the one hand the evidence clearly points to the universality of the agreement. Its failed precursor of 392 is described by Andocides (the conference at Sparta) as koine eirene and eleutheria for all the Greeks, in which they would all be able to share fully (3. 17); while the proposals earlier put to Tiribazus, the King's general at Sardis, by Antalcidas again encompassed the total of Greece and the Aegean (X. H. 4. 8. 14). There is no reason for supposing that the scope of the settlement was narrowed by 387. On the contrary, the unbiased reader of the King's edict in Xenophon (H. 5. 1.31) would assume nothing less than that the law the King was laying down concerned all states, be they islands or cities on the Asian and European mainlands; see H. 5. 4. 1; 6. 3. 9; 6. 3. 12. Cf. the freer rendering in D of the same edict which adds ἅπαντας to Ελλὴνας (14. 110. 3) and Svt 248 lines 6ff.; 257 lines 9ff. Also Isocr. 4. 176; 14. 5, 10, 12 (there was no difference in this respect between the Peace of 387/6 and its renewal in 375). It is perfectly true of course that ὁποτεροι in the edict (H. 5. 1. 31) refers to the two sides in the war (as Lewis stresses: 1977: 147). To conclude from this, however, that the Peace was imposed only on the combatants in the war ( Badian, Georgica, 39, 43f.) is not -164-

quite consistent with the evidence, including that of the edict itself. (Cf. Sealey, 1993: 13.) The reason surely why ὁποτεροι, the two sides in the conflict, was chosen for inclusion in the edict, is because clearly the Peace could only be seriously threatened by one or the other of the two great alliance systems at war at the moment (ὁποτεροι δὲ τύτὴν εἰρήνν μὴ δὴχοντΛι). On the other hand it is indubitably the case that many states did not participate in the Peace. The answer would seem to be that whereas the Peace was indeed koine in theory, open to all Greeks, in practice participation depended (a) on whether a state was willing to abide by the terms of the Peace and to take the oaths, and (b) on whether the King and the leading states participating (Sparta really in 387/6) did not object to that state's inclusion. (Cf. Clark, 1990: 61ff., 67.) We have examples of both categories. In fact, the contention that the King's Peace was confined to those actively involved in the Corinthian War (e.g. Badian, Georgica, 39, 43f.) seems to rest on the text of Xenophon alone. Badian is impressed by H. 5. 1. 35, the

statement that the peace the King sent down marked the first peace between the Spartans and the Athenians and their allies after the war which followed the demolition of the walls of Athens. And so it did; the Corinthian War was ended by the King's Peace. But is this any reason for curtailing the scope of the Peace? After all, Xenophon ('our best source' for the Peace in Badian's view) also chooses to represent the renewal of the Peace in 375 as a bilateral affair between Athens and Sparta and does not even bother to mention the involvement of the King (H. 6. 2. 1). Nor does the fact that Xenophon, a master at putting the emphasis where it suited him, never breathes a word about the Peace and the autonomy clause in connection with Spartan attacks on other states between the conclusion of the Peace and the capture of the Cadmea, prove that other 4th-century sources which do are wrong (see 5. 2-3 n.). For a correct view cf. Cartledge, 1987: e.g. at 242f., 259. καΘ ἤν αἱ πολεις . . .παρέΛαßον: turning now to the contents of the Peace, there was clearly more to it than the royal edict quoted by Xenophon. The question is how much more? The subject has been extensively debated, the evidence being so equivocal as to allow a minimalist view as well as one which sees the treaty as a more comprehensive document comprising a number of specific clauses. See e.g. the discussions by Momigliano, 1934b: 482ff.; -165Hampl, 1938: esp. 8ff.; Wilcken, 1941: 12ff.; Accame, 1941: 1ff.; Martin, 1944: 20ff.; id., 1949: 127ff.; Ryder, 1965: 34ff., 122f.; Cawkwell, 1973b: 52ff.; id., 1981 a: 69ff.; Sinclair, 1978: 29ff.; Cartledge, 1987: 196ff.; Clark, 1990: 47ff., esp. 56ff.; Badian, Georgica, esp. 35ff. Most of the relevant texts are conveniently assembled in Svt 242. Here I am concerned only with those aspects of the Peace which bear directly on D both here and at 15. 29-7. The affirmation of the autonomy of all Greek states, with the exception of those claimed by Persia, was at the very heart of the Peace. It was what the King had commanded and the question to settle is whether the autonomy clause was defined in any way. Was, for example, the withdrawal of the garrisons mentioned here by D stipulated in the Peace, or was it merely the natural result of the implementation of an unadorned autonomy clause (as Ryder, 1965: 123; Sinclair, 1978: 31)? D's language seems to point to the first, but there is better evidence. Writing in 355 Isocrates proposed that Athens should return to the treaty made with the King and the Spartans which ordered τοὺς 'Εγγὴνας αὐτονομους ει+̥ναι καὶ Ýὰς ϕρυρὰς ἐκ τω+̑ν ἀλλοτρίων ἐΞιένααι καὶ τὴν αὑτω+̑ν ἒχειν ἑκάστους (8. 16). It is illogical to see this as a reference to the Peace of 375 alone ( Ryder, 1965: 122). Isocrates must surely mean all three Peaces, the original King's Peace of 387/6 as well as its two renewals in 375 and 372/1. Though the renewals differed from the original Peace in some important respects, they cannot have differed in this respect and that is why Isocrates is not specific ( Cawkwell, 1973b: 52 n. 3; id., 1981a: 73). That the third clause, καὶ τὴν αὑτω+̑ν ἒχειν ἑκάστους, was part of the first Peace is shown by Isocr. 4. 177f. (rightly interpreted by Cawkwell, 1981a: 72) and by other evidence (Justin 6. 6. 1; Svt 257=Tod 123 lines 11f.; Martin 1944: 26; Ryder, 1965: 122; cf. Sinclair, 1978: 29f.). It might further be noted that in the opinion of Philochorus the Peace of 375 resembled (παραπαπλήσιος) that of 387/6 (FGH 328 F151). It would not be unreasonable therefore to assume that the provision for the removal of garrisons from the cities, which undoubtedly accompanied the Peaces of 375 and 372/1 (15. 38. 1f.; X. H. 6. 3. 18), was first made in 387/6. There was, we know, a demobilization clause in 372/1 (X. H. 6. 3. 18) and the similar language Xenophon employs at H. 5. 1. 35 could be interpreted as a reference to such a clause in 387/6 (as -166Cawkwell argues. Contra Sinclair, 1978: 31ff.). Cawkwell's further proposals, however, that the Peace also prohibited the later redeployment of forces, naval forces especially, and that Athens in particular could not launch or augment her navy, seem most unlikely: Clark, 1990. Cf. Plato, Menex. 245e: σὺν Θεω+̑+̂ ἄμεινον ἤ τοτε (i.e. 404) ἐΘέμε α τον πολεμον καὶ γὰρ τείχὴ ἒχοντες κτλ. Perhaps the most intractable aspects of the King's Peace concern a possible sanctions clause and Sparta's role in first enforcing the Peace and in upholding it subsequently. The following four inferences seem to me to conform well not only with the information of D, but also with the rest of the evidence: (a) in 387/6 the King was the guarantor of the Peace while Sparta saw to its implementation. (b) The King's role went further than the initial acceptance of the terms (otherwise why should Athens be at pains to make it very clear that, for example, her alliance with Chios (Svt 248=Tod 118) did not contravene the Peace?) as did that of Sparta. (c) Sparta certainly acted as the prostates of the King's Peace and its autonomy clause, as Xenophon makes perfectly plain (H. 5. 1. 32, 36). Whether she was so formally or merely appointed herself to that position (with the tacit approval of Persia) might be thought to make little practical difference. In fact, there is no reason why Sparta's prostasia, however expressed ( Cawkwell, 1981a: 77 n. 31, might not have been made explicit in the treaty document. (d) The Greek states in general were not obliged to aid the victims of aggression unless they wished to do so (μετὰ τω+̑ν ταυ+τα ßουλομένωÚ: X. H. 5. 1. 31; cf. D 14. 110. 3 and see Cawkwell, 1981a: 76ff.). There was no symmachia in other words. The members of the Peloponnesian League of course were obliged to follow Sparta wherever she led them (cf. de Ste Croix, 1972: 108ff.), a position in glaring contradiction to the autonomy clause as contemporaries saw (X. H. 6. 3. 7f.). There is support for the above view in the case of Mantinea in 385. It would appear that as the prostates of the Peace Sparta invoked the Peace and its autonomy clause against Mantinea. At any rate, Athens' response to an appeal for help from Mantinea, that she did not wish to παραßαίνειν τὰς κοινὰς σπονδάς, cannot be interpreted in any other way (15. 5. 5 and n.). The action Sparta took against Thebes in 382 is equally consistent with such an interpretation (5. 2-3 n.). For the use Sparta made of the autonomy clause in the years 386-379 see 5. 2-3 n. -167-

The position was similar in 375 and 372/1. Action against violators of the Peace was voluntary, not obligatory. The explicitness at X. H. 6. 3. 18 (372/1), τον μὲν ßουλομενον. . . τω+̑+̣ δὲ μὴ ßουλομένῳ, was directed against the Spartan habit of forcing unwilling allies to campaign against states accused of alleged breaches of the Peace (X. H. 6. 3. 7f.). But in spite of the clause the Spartans persisted in their habit as the rejection of Prothous' proposal shows (X.

H. 6. 4. 2f.). In 375 Athens joined Sparta as a co-prostates of the Peace with the sea as her special concern (38. 4n.). Along with Sparta she could now call for action against states contravening the Peace. That such action was voluntary is indicated by Isocrates' Plataicus. Unlike Cleiteles in 370/69 (X. H. 6. 5. 37) the Plataeans did not claim that Athens was under any obligation to help them. Plataicus 42ff. especially shows that it was up to the individual state whether it moved against an aggressor or not. The Peace of 371/0 at Athens was different. The participants swore to aid, as a matter of course, any wronged city which was party to the Peace. The Second Athenian Confederacy was now the pattern; see 57. 1n. It is not very likely that there was an exiles clause in the Peace though the possibility cannot be totally excluded ( Cawkwell, 1973b: 59 n. 1; id., 1981a: 80ff.). On the Peace see also 19. 1n. αὐτονομίαν: for the concept of autonomy see Bickerman, 1958: 313ff.; Ostwald, 1982; Cartledge, 1987: 249f.; Sealey, 1993: 241ff. Λακεδιμονιοι ϕύσει ϕιλαρχου+ντες κτλ: for Ephorus' censure of Sparta see pp. 113, 117 f. δυνααστείαν: though δυνααστεία is used here to describe the Spartan supremacy (as something which the Spartans had had before the King's Peace and the loss of which they regretted) and (probably) at 19. 4, ἡγεμονία is the more frequent term occurring; cf. 14. 10. 1; 15. 1. 2, 3, 5; 15. 19. 1; 15. 23. 3ff.; 15. 40. 1. δυνααστεία is used here as appropriate to the censure of Sparta. Ephorus probably did not consider hegemony as wrong in itself, but only its improper use; see pp. 112 ff. 5. 2-3. εὐΘὺς ου+ὐν τάς μὲν πολεις συνετάραττον κτλ: it is argued in the Introduction (pp. 102 f.) that what we have here is a description, and condemnation, of Spartan policy in the period 386-379. In lifting it out of Ephorus D seems to have garbled it somewhat. We are told that the way the Spartans went about regaining their former dominance was by inciting civil strife within the cities with -168-

the help of their supporters. A smaller group of cities is then distinguished which actually provided Sparta with plausible grounds for intervention. The cities, having recovered their autonomy, came under the control of 'democratic' regimes (τοὺς εήμους: see 40. 4n.) which proceeded to exact revenge from those who had been in charge at the time of the Spartan hegemony, banishing many of them. Their case was taken up by Sparta and in restoring them by force she enslaved again the cities concerned, first the weaker ones, but afterwards also the stronger, having kept the common peace for less than two years. And D proceeds to narrate the attack on Mantinea. But the last statement about the enslavement of the cities ought surely to apply to all the cities and not merely to those of the smaller group which provided Sparta with grounds for interference. What is more, it is likely that Ephorus illustrated his rhetoric with examples which D omitted. It is not easy, therefore, to find instances to fit this generalized castigation. It cannot apply to Olynthus, as far as we can see, while Thebes and Mantinea could only be said to fit the wider group of cities (though see below); and Phlius is difficult because the friends of Sparta there were exiled long before the Peace (X. H. 4. 4. 15). But there is more to the passage. An important element is the reference to the autonomy clause which, as D says, was used in some cities against the pro-Spartan factions. It is not impossible that, faced with the Spartan demands (X. H. 5. 2. 9. Later an ultimatum: H. 5. 3. 15) to receive back their exiles, the Phliasians appealed to the autonomy clause. Certainly the Spartan actions against Phlius and the other cities were seen as contraventions of the Peace by contemporaries (cf. Isocr. 4. 126, 176; 8. 100; Svt 257 lines 9 ff.), including Xenophon himself ( Tuplin, 1993: 87 ff. But Xenophon's view of Spartan policy was ambivalent: see p. 114 n. 306). Pace Badian, Georgica, 44 and n. 39, H. 5. 4. 1 is no evidence that Xenophon did not, in his heart at least, recognize any breaches of the Peace earlier than the capture of the Cadmea. But Cadmea was such a blatant instance that even he could not refrain from openly condemning it; and of course it was the Thebans, virtually single-handed, who 'punished' the Spartans for their misdeeds. D's brief mention of the subjection of Phlius (19. 3) comes immediately before his statement that Agesipolis, in direct opposition to Agesilaus, objected to current Spartan policies which went contrary to the -169-

Peace terms. Sparta, Agesipolis is said to have stressed, was engaged in enslaving the Greek cities in spite of the fact that in the general settlement she had taken an oath to preserve their autonomy (19. 4). Yet D speaks of a number of cities where the pattern he outlines was repeated. Can any examples be found? The substance of the present passage is repeated at 9. 5 and 19. 1, and almost identical opinions are expressed at 40. 1 f. (under 375/4; cf. n. ad loc.) and 45. 2 (under 374/3). Thus according to D (Ephorus) both the Peace of 387/6 and its renewal in 375 were followed by staseis in many cities which gave Sparta the chance to intervene, the difference being that on the second occasion Athens was no longer a mere observer, but actively opposed Sparta (45 ff.).

and οὑ+̑τοι (but see n. there), for when Timotheus sailed into the Ionian sea in 375 Zacynthus was in the hands of the pro-Spartan party and the exiles who flocked to him were democrats. One is therefore drawn to conclude that the sentence before the lacuna should be linked with 5. 2 (and 19. 1). It is precisely the same process which is recounted and in very much the same language. Under the aegis of the autonomy clause the pro-Athenian Zacynthian demos it seems expelled the pro-Spartan oligarchs who nevertheless returned, perhaps with Spartan help and under the same pretext of autonomy, and in their turn expelled the democrats. These events may therefore be placed in the period 386-379. Can any other candidates be found? Speaking of the months after Agesilaus' first invasion of Boeotia in 378 Xenophon made the comment that, just as at Thebes, proSpartan regimes had been installed (by Sparta) in all the cities of Boeotia so that in these cities too the

friends of Sparta needed help (H. 5. 4. 46). It would be unreasonable to suppose that Xenophon is referring only to the time following the liberation of Thebes. Oligarchies could not possibly have been set up in all the cities of Boeotia in a matter of months. A longer period should be envisaged. Now, the Boeotian cities had been detached from Thebes in 387/6 as a result of the Peace, and it would appear that Sparta exploited the internal conflicts which ensued, helping her own supporters, expelling the democrats, and even installing -170-

garrisons in a number of places. Plataea was restored and garrisoned at this time too (X. H. 5. 4. 10; Plut. Pel. 25. 8; Paus. 9. 1. 4; Glotz and Cohen, iii. 105; Sinclair, 1978: 38 f. and n. 36). We do not hear of actual conflict at Thebes, but there must have been latent strife at the very least, considering the intense enmity which existed between Ismenias and Leontiades and their respective parties (Plut. Pel.5; cf. Hack, 1978: 210 ff.). If the report that the capture of the Cadmea in 382 was the result of a plot on the part of Sparta is credited, then we must surmise that Leontiades and his associates had visited that city some time before the event, made their request (the idea had probably originated with them), and worked out the details of the undertaking. Xenophon (H. 5. 2. 25 ff.) represents the outrage as having occurred on the spur of the moment, but his own narrative hints to the contrary (20. 2n.). This is not to suggest that the matter was openly debated in the assembly and a vote taken to that effect (Xenophon denies thisH. 5. 2. 32 -- and he must be right). The plot would have been laid with the dominant faction at Sparta ( Agesilaus' surely). Did they have any legitimate grounds for acting thus against Thebes? It is possible that they had two such grounds. First, Thebes may have been accused of contravening the Peace. Sparta had declared war on the Chalcidian Confederacy, quite likely on the basis of the autonomy clause (19. 3 n.; Cawkwell, 1973b: 53; Sinclair, 1978: 37; Cartledge, 1987: 270. Contra Badian, Georgica, 430. Yet the faction of Ismenias at Thebes, which held the upper hand at the moment, was contemplating an alliance with Olynthus and had had it proclaimed that no Theban should join the Spartan expedition against Olynthus (X. H. 5. 2. 27, 34ss). As the prostates of the Peace (whether formally so or not: 5. 1 n.), therefore, Sparta may have had valid grounds for taking action though not for continuing to occupy the Cadmea ( Cawkwell, 1973b: 53 n. 3; id., 1981a: 78 f.). Second, the intervention may have been justified by reference to the duties of a member of the Peloponnesian League. The return of Thebes to the Spartan alliance after the King's Peace has been questioned (e.g. by Buckler, 1980b: 179 ff. and Cawkwell, 1981a: 78 ff. The case for Boeotian membership before 395 is less in dispute; cf. de Ste Croix, 1972: 335 ff.). Isocrates nevertheless is explicit that at the conclusion of the Corinthian War the Thebans όα+̔πολιπόντες ὑμα+̑ς ει+̔ς τὴν Λακεδαιμονίων συμμαχίαν ει+̔ση+̑λΘον (14. 27). And in the next section he says that the Thebans actually swore to -171follow Sparta against Athens (ὢμοσαν ἠ + ̑ μὴν α + ̔κολουΘńσειν μετ' ε + ̔κεíνων ε + ̔Φ ὑμα + ̑ς). One is tempted to see in this the oath members of the Spartan alliance took to follow the Spartans wherever they led them (cf. de Ste Croix, 1972: 108). The return of the proSpartan oligarchic exiles in 386 which had been demanded by Agesilaus (X. Ag. 2. 21) may have counterbalanced and even checked the anti-Spartan democratic faction of Ismenias and Androcleidas for a while; Thebes joined Sparta again. This perfectly explains the presence of Theban troops before Mantinea in 385 (Plut. Pel. 4. 5-8; Paus. 9. 13. 1). Plutarch in fact says that the Thebans and Spartans were still at this time 'friends and allies'. Cawkwell, 1981a: 78 f., prefers to see the Theban presence at Mantinea as the result of a 'volunteers' sanctions clause in the Peace. 'Perhaps these [Thebans] who were sent were "volunteers", including Epaminondas and Pelopidas eager for experience in war' (1981a: 79 n. 33). He may well have a point about the clause (5. 1 n.), but the rest of the suggestion is difficult to credit. It is easier to believe that the two men (and Pelopidas was already in all likelihood a prominent member of the anti-Spartan party of Ismenias and Androcleidas: Plut. Pel. 5. 1) were called up to serve in a contingent officially sent to the Peloponnese in fulfilment of Theban symmachic obligations. An indication of the change of policy at Thebes after the Peace is to be seen perhaps in the destruction of the stele bearing the treaty of alliance between Athens and Boeotia which was concluded in 395 ( Svt 223 =Tod 101). The accuser of Evander in summer 382 charges Thrasybulus of Collytus with having caused a revolution in Boeotia which cost Athens the Boeotian alliance (Lys. 26. 23). Schweigert, 1939: 1 ff., dates the abrogation of the alliance between 386 and winter 383. One might further note that Isocrates remains consistent. Even after the liberation of Thebes, he says, the Thebans were ready to be Sparta's slaves and an embassy of theirs to Sparta promised μηδέν κινει + ̑ν τω + ̑ν πρὸς αυ + ̔τοὺς ὡμολογημένων (14. 29). I take this to mean that they were willing to remain within the Spartan alliance (but see 28. 1 n.). Lastly, there is Buckler's assertion that there is not even a faint whisper in the philolacon Xenophon that the Thebans were breaking their oath to Sparta by opposing her war efforts against Olynthus and that Thebes consequently cannot have been the ally of Sparta at this time (1980b: 179 ff.). This is a fair point at first glance until one -172-

notices that in his speech to the Spartan assembly Leontiades does not distinguish between Theban behaviour before and after the Peace (X. H. 5. 2. 33 ff.). Thebes, he says, has been consistently hostile to Sparta since the end of the Peloponnesian War. Yet nobody would deny that Thebes was Sparta's ally then. In the eyes of Sparta, therefore, the Thebans may have been doubly guilty. They opposed Spartan efforts against Olynthus and they were thus accountable under the King's Peace; and their behaviour was not that of loyal members of the Peloponnesian League. The charge of warmongering levelled against Ismenias by Leontiades points to the King's Peace (ὡς πολεμοποιου + ̑ντα: X. H. 5. 2. 30); but the trial of Ismenias before judges from Sparta and the allies, and the charge of Medism and treason, seem to point to the Peloponnesian League (X. H. 5. 2. 35f.). A better case can be made for the use of the autonomy clause against Mantinea. There is no evidence for actual stasis at Mantinea before the Spartan attack. Xenophon tells us that the ßέλτιστοι could not refrain from attacking their democratic enemies at the city's surrender (H. 5. 2. 6) and so it is possible that these people had appealed to Sparta. Indeed it is even possible that some of them had been in exile and only returned when the city capitulated. More significantly, D says that Athens turned down the Mantinean appeal for help not wishing to παραßαíνειν τὰς κοινὰς συνΘήκας (5. 5). And Polybius adds that the Spartans claimed to have done no wrong in breaking up Mantinea (4. 27.

6). It would appear, therefore, that Sparta justified her action by reference to the autonomy clause. To Spartan minds (and Xenophon's too, though cf. Tuplin, 1993: 90 and also p. 114 n. 306) real autonomy was not achieved at Mantinea until after the α + ̑ργολíζοντες and other democratic 'riff-raff had been expelled, the connection with Argos thus severed, and the city undergone dioikismos (X. H. 5. 2. 6f.). This is consistent with, and sheds some light on, D's otherwise mysterious statement that the Spartans attacked Mantinea because they ὑπώπτευσαν αυ + ̔τη + ̑ς τήν αὔξησιν τη + U0311 γινομένην ἐκ τη + ̑ς ει + ̔ρήνης (5. 3). It is likely that Mantinea had taken the autonomy clause to mean that she could move even further away from Sparta (for her hostile attitude during the Corinthian War see X. H. 4. 5. 18; 5. 2. 2) and closer to Argos if she so chose. This is precisely what she did after Leuctra (cf. X. H. 6. 5. 3). The αὔξησις might also refer to territorial expansion. As in the 5th century Mantinea may have -173been subjecting the neighbouring towns. All the more justification therefore for Spartan action on the basis of the autonomy clause. (For Mantinean empire-building in the 5th century see Andrewes, HCT iv. 31 f. and for Sparta and Mantinea in general, Cartledge, 1987: 257 ff.) Nor is this all. An Attic decree ( IG ii2 33) names Ecphantus and other Thasian exiles as honoured with ateleia: εἰ + ̑ναι δὲ καὶ τοι + ̑ς ἄλλοις τοι + ̑ φεύγοσι Θασίων ἐπ' 'Aττικισμῳ + ̑ τὴν α + ̔τέλειαν καΘάπερ Mαντινευ + ̑σιν. The mention of the Mantineans dates the decree in or after 385/4. ( Pouilloux, 1954: 199 ff., is in a minority in arguing for a much earlier date.) In 390 Ecphantus and his pro-Athenian party had expelled a Spartan garrison and Thasos had joined Athens ( Dem. 20. 59). Cf. Osborne, 1981-3: ii. 49f., 52. It would not be surprising if Sparta had had a hand in the exile of these men. There is in any case other evidence for Spartan interference in the Aegean. Sometime between summer 382 and winter 379 the Spartan officer Herippidas ejected Jason's man Neogenes from Hestiaea in north Euboea and restored to the people their freedom, τὴν ἐλευΘερίαν (καί αυ + ̔τονομίαν? 30. 3 nn.). At the same time or a little later the neighbouring islands, including Sciathus and Peparethus, were won over for they are said to have been under Sparta in 378 (30. 5 and n.). A little earlier Athens, it appears, withdrew from Delos, perhaps after having been threatened by Sparta with the autonomy clause ( Sinclair, 1978: 43 f. D's broad condemnation, therefore, seems justified (cf. Rice, 1974: 165 f.). Most of the detail is lost to us of course, but there is enough to show that Sparta took full advantage of the provisions of the King's Peace to arrange Greece in her interest. The interventions in both Epirus and Macedon in 385/4 are a good indication of how far the arm of Sparta reached in these years (15. 13. 2f.; 19. 2n.). 5. 3. κατεδουλου + ̑ντο . . . ου + ̔δὲ δúο ἔτη φυλάξαντες τάς κοινὰς σπονδάς: compared with Thebes and Olynthus Mantinea could be described as α + ̔σΘενεστέρα, but D is writing loosely and the statement should not be pressed (see previous n.). There can be little doubt, however, that we are meant to associate the attack on Mantinea with the beginning of Sparta's misuse of the Peace less than two years after its conclusion. Hence the attack took place in the second year after the conclusion of the Peace. See 5. 1n. πλήΘουσαν α + ̔νδρω + ̑ν α + ̔λκíμων κτλ: there are textual problems here, -174-

but the meaning is not in doubt. Mantinea is thus joined with Thebes as an object for Spartan jealousy and suspicion because of her populousness and bravery (Cf. 20. 1). Xenophon explains that Sparta was anxious to deal with those of her allies whose loyalty during the war had not been wholehearted (H. 5. 2. 1). This is correct, as is Ephorus' explanation that Sparta was not prepared to allow democratic Mantinea to grow too strong and independent at her own doorstep. For a discussion of the population of Mantinea see S. and H. Hodkinson, 198 1: 27 1 ff. 5. 4. τὸ μὲν πρω + ̑τον . . . κώμας: according to Xenophon the Mantineans were only asked to demolish their walls as proof of their loyalty. Upon their refusal to comply, the Spartans moved against them (H. 5. 2. 1 ff.) εἰς τὰρχαíας πέντε κώμας . . . συνῴκησαν: this is Ephorus' number ( FGH 7o F79) and Strabo 8 C 337 (almost certainly deriving from Ephorus) comments Mαντινεíα μὲν ε + ̔κ πέντε δήμων ὑπ' 'Aργεíων συνΰκíσΘη. Xenophon on the other hand says that Mantinea was broken up into four separate villages, τετραχῃ + ̑ (H.5. 2. 7). According to Pausanias (8. 8. 9) some of the Mantineans were allowed to remain on the site of the city while the rest were dispatched to their former villages. It is possible, therefore, that the site of the city was one of the old κω + ̑μαι. Xenophon's τετραχῃ + ̑ might thus be reconciled with the rest of our authorities on the assumption that he was confused by the fact that four groups of Mantineans were expelled to the countryside, overlooking a fifth group which remained behind. But see the discussion of S. and H. Hodkinson , 1981: 261 ff. ; cf. Tuplin, 1993: 90 n. 10. δúναμιν ἐκπέμΨαντες: under the command of Agesipolis; X. H. 5. 2. 3; Plut. Pel. 4. 8; cf. Cartledge, 1987: 260. D does not mention this, the result perhaps of too severe an abridgement of his source. It is not inconceivable nevertheless that Ephorus, having portrayed Agesipolis as a champion of Greek autonomy against Agesilaus (19. 4), thought it better to omit the fact that the young king had commanded vigorously against Mantinea (cf. 12. 1 and 19. 4 nn.). 5. 5. οἱ δὲ Mαντινει + ̑ς . . . κοινὰς συνΘήκας: Mantinea may well have based her appeal on the autonomy clause (5. 2-3 n.). I agree with Seager, 1974: 40, and Sinclair, 1978: 38, that whatever the legal points involved, the Athenian decision not to help was based on the sober calculation that nothing could be done. -175ὑποστάντες . . . ἠμúνοντο: neither D nor Xenophon mentions what Plutarch and Pausanias do. Paus. 8. 8. 7 says that Agesipolis defeated the Mantineans in battle before laying siege to their city; and Plut. Pel. 4. 5-8 and Paus. 9. 13. 1 relate that during the battle Pelopidas was severely wounded and saved by Epaminondas. It emerges

therefore that a Theban contingent accompanied Agesipolis (5. 2-3 n.). The silence of Ephorus (if it is not in fact D who has skipped over the event) and Xenophon may be explained by different motives. While Xenophon never went out of his way to mention the two Thebans, Ephorus might have been reluctant to inform his readers that the later champions of Arcadian and Messenian autonomy had earlier distinguished themselves in the service of Sparta. According to Polyaen. 2. 25 the Spartan allies sympathized with and helped the Mantineans. καινω + ̑ν πολέμων α + ̔ρχήν: cf. 77. 1 (after the Peace Of 366/5) and 94. 1 (after the Peace of 362). Writing in the late 330s and 320S Ephorus could look back over the past two generations as a sequence of peace settlements which had failed to work (see pp. 110 ff.). The first Common Peace failed because the Spartans exploited it to serve their own interests. Thus the attack on Mantinea was the first in a series of Spartan outrages which eventually led to a fresh outbreak of general war in 378.

6-7. Dionysius at home On the nature of these chapters and on the relative dates of the events related see 7. 2 n. 6. 1. ποιήματα: Snell, TGF i. 240 ff.; Sanders, 1987: 2 and n. 7. In general on matters cultural to do with Dionysius cf. Loicq-Berger, 1967: 138ff., 226ff. And for the tyrant's interest in music, id., Revue belge de philologie et d'histoire, 44 ( 1966), 12 ff. 6. 2-5. Φιλόξενοσ κτλ: of Cythera, a leading composer of dithyrambs: Page, PMG nos. 814-35. He died in 380/79 according to the Marmor Parium. The anecdote of how the poet displeased the tyrant and ended up in the infamous Quarries of Syracuse is related (with variations) by many writers: Lucian, Adv. Indoct.15; Cicero, Ad. Att. 4. 6. 2; Athen. 1. 6e; Aelian, VH 12. 44; Plut. Mor. 334c, 471e; Schol. Aristoph. Plutus290; Suida s.v. ἄπαγέ με. On the tyrant and Philoxenus see Sanders, 1987: 15 ff., who rightly -176stresses the political, anti-Dionysius character of the anecdote. On the λατομíαι, Drögemüller, 1969: 18 and 110. 7. 1. Plato's first (and perhaps only; cf. e.g. Finley, 1979: 91 ff.) visit to Sicily, the historicity of which there is no reason to doubt, including the sequel, that is, the selling of Plato into slavery, whether at Syracuse (as D here; cf. Nepos, Dion 2. 2f.) or on Aegina (Plut. Dion4f.; Diog. Laert. 3. 18 ff.). Some scholars doubt what the sources unanimously affirm, that Dionysius had ordered this: Porter, 1943: 45 ff.; id., Dion, 52 ff.; Stroheker, 1952: 235f.; id., 1958: 104f. Caven, 1990: 168f., even doubts that philosopher and tyrant even met, let alone quarrelled. For the evidence see Berve, Tyrannis, ii. 653f. For a general discussion cf. LoicqBerger , 1967: 146ff., 292 ff.; Sordi, 1979: 2013ff. For the impact of Dionysius and Plato on each other cf. Sanders, 1987: esp. 5 ff., 13 ff., 21 ff., who further points to the way their relations were shaped by the politics of the day. Also Lewis, CAH2 vi. 154 f. As far as the date of the journey is concerned, the Seventh Platonic Epistle tells us that Plato was about 40 years old at the time ( 324a). The year 388/7 is therefore indicated. In any case, the Corinthian War was still being fought (Aegina was hostile: Plut. Dion 5. 7). For what it is worth, Olympiodorus, Vita Plat. 5, says that the person who bought Plato, one Anniceris from Libya, was on his way to Elis to compete in the Olympic Games, presumably those of summer 388. (Cf. Stroheker, 1958: 100 and 216 n. 85. Contra Caven, 1990: 260 n. 15.) παραπλὴσιον: i.e. the connection is not chronological. 7. 2. εἰς . . . τὴν ὋλυμΠιακὴν πανήγυριν: the theoria to the Olympic Games of 388. Following Ephorus, D related its fortunes in 14. 109 (under 388/7), in the correct chronological context (the siege of Rhegion), and he now briefly returns to it in a different context and to serve a different purpose. The two passages do not therefore present problems, either in chronology or source criticism (pp. 80 f.). Yet some are of the opinion that Dionysius' famous embassy was sent to the festival of 384 and not to that of 388, basing themselves principally on Lysias 33 (Olympiacus), which D associates with the embassy (at 14. 109 and not at 15. 7. 2% and which they prefer to date in 384. Grote argued the case at length, but none of his reasons is compelling (x. 103 n. 2.; xi. 48 n. 1; cf. Stroheker, 1958: 234 n. 43. Lewis, CAH2 vi. 149 n. 82, on -177the other hand, cautiously accepts Grote). Can we really envisage an Athenian metic delivering such a blast against the Great King at Olympia in the summer of 384? Would not the speech have amounted to incitement to break the King's Peace? The whole tenor of the speech seems to fit a time before the conclusion of the Peace and indeed before Antalcidas came down from Susa and the position in the Aegean changed dramatically. And cf. section 6, τὸν μὲν πρὸς α + ̔λλήλους πόλεμον καταΘέσΘαι. Section 7 may seem a little surprising for 388, but would Lysias have dared return to Athens if he had expressed such sentiments about Sparta in 384? Admittedly the speech cannot be dated decisively from internal evidence; nor, apart from D, is there any other evidence to help us. Section 15. 7. 2 does not provide a date. What we can reasonably deduce from 14. 107 ff., on the other hand, is that the summer in which the 98th Olympiad was celebrated, i.e. that of 388, was also the summer in which Dionysius launched his final assault against Rhegion. And this was the Olympic festival to which the notable embassy was sent and in the course of which the oration of Lysias was delivered (14. 108. 6-109). Pace Stroheker, 1958: 222 n. 52, this is the crucial synchronism to note. It is not impossible that the embassy merited mention in D's chronographer. But whether it did or not, it would appear that 14. 109 figured in his narrative source, embedded in the narrative of the siege of Rhegion (14. 108. 6-109. 1: χρονιζοúσης δὲ τη + ̑ς πολιορκíας . . .τω + ̑ν δ' Ὃλυμπíων ε + ̔γγὺς ὄντων κτλ): 388, therefore, rather than 384. ε + ̔ν τῳ + ̑ πρóτερον at 14. 107. 4 should be credited to the narrative source also (cf. 13. 82. 7; 15. 5. 3), rather than to D himself (as Lewis, CAH2 vi. 147 n. 119, suggests). It is part of a passage analysing Dionysius' motives. The battle of Eleporus

then and the treaty with Rhegion (14. 103-6) should come under 389, the beginning of the actual siege of Rhegion in summer 388, under archon Pyrgion ( 388/7), but perhaps a little before the Olympic festival, and the capitulation of the city almost a year later (14. 111. 1), in 387/6, under archon Theodotus (14. 111; 15. 1. 6 n.; Polyb. 1. 6. 1 f. with HCP i ad loc.). These dates (cf. Meyer, GdA v. 128 ff.) seem to me to be more consistent with the evidence than the 'low' chronology worked out by Beloch ( GG iii. 2. 366ff.) and accepted by Stroheker ( 1958: 114 ff.). Sordi's proposed chronology is eccentric (1992: 64 ff.). It dislocates the synchronism of the siege of Rhegion -178with the King's Peace and th e Sack of Rome, and it depends on unacceptable source criticism (see pp. 77 f.). 7. 3-4. The reason for the brief repetition of the story of the fiasco of the embassy to the 98th Olympic festival now becomes apparent: Dionysius' acute disappointment, D's source would have us believe, led to serious problems at court for which he was entirely to blame. 7. 3. τοùς ΦíΛοÚς úΦωΦτεÚεν OMEGAώς XIξΦιßουλεντας κτλ: see the discussion in Caven, 1990: 169 ff., convincing in its conclusions, except for the chronology. Following D's arrangement (and in line with most other scholars; cf. Jacoby, FGH556 (text) 497) Caven is inclined to date the conspiracy, or Dionysius' reaction to the belief that there was a conspiracy (the view in D), in 386/5. It has already been argued (pp. 80 f.) that the events in chs. 6 and 7 are set in the period after the peace treaty of 392, though a little more can be said about the date of the visit to Sicily of Plato (see above) and the exile of Philistus and Leptines. The latter cannot have fallen into disfavour before 390/89 (14. 102. 3) and the 'conspiracy' itself may also post-date the Olympic festival of 388, if D's source is correct in seeing a connection between the two. Cf. also Sanders, 1987: 54ff. Φíλιστος: FGH 556 TT with Jacoby's commentary; above p. 67. Cf. Zoepffel, 1965 ; Sanders, 1987: 43ff.; Pearson, 1987: 19ff. Λεπτíνńς ó Aàδελϕóς: on Dionysius' family and family connections see Beloch, GG iii. 2. 102ff.; Stroheker, 1958: passim; Sartori, 1966: 3ff.; Berve, i Tyrannis, i. 249 f.; ii. 651 f.; Sanders, 1987: i passim; Lewis, CAH2 vi. 151 ff. 7. 4. ουτοι . . . εìς Θουρíους κτλ: the exile of Philistus and Leptines involves intractable problems because of the contradictory nature of the evidence. D seems to be saying that Philistus and Leptines were exiled at the same time and on the same charges. And he is certainly saying that they went to Thurii, but that later (ὔστερον) they were reconciled to Dionysius, at the tyrant's own request, and returned to Syracuse, Leptines going on to marry a daughter of his brother. D is in fact the only evidence for the exile of Leptines though there are indications that all had not been well between the brothers (14. 102. 3; 17. in.; Aeneas Tact. 10. 21 f.; Plut. Dion 9. 5). At any rate, Leptines was certainly back at Syracuse and in favour again by the time the battle of Cronium was fought (15. 17. 1) though we do not of course know -179when exactly that was (15. 1 n.). The problem then is the length of Philistus' exile because Plutarch is categorical that he did not return to Syracuse until after the death of Dionysius. Also that he went somewhere in the Adriatic or Epirus and there composed most of his history (Dion 11. 6f.; Mor. 605c; cf. Nepos, Dion3. 2). There are various ways of getting round these difficulties (see Jacoby, FGH556 (text) 497 f.; Stroheker, 1958: 227 n. 116), but it is all too possible that D misread his source. Λεπτίνὴς . . . ΘυλατέρALPHAα: this must be Dicaeosyne, Dionysius' daughter by Doris the Locrian ( Beloch, GG iii. 2. 104, by a process of elimination).

Archon Dexitheus, 385/4 8. 1. Πωμαîοι κτλ: repetition of the college (393 V) cited at 14. 99. 1 (390/89) except that there the nomen of the second consul is missing, though MS F adds Kòσσων instead; see Livy 5. 29. 2; CAH2 vii. 2. 636; Perl, 1957: 79, 103; "Introduction" pp. 27f.

8-11. The conclusion of the Cypriot War 8. 1. ἐπὶ SIΣἐ τοúτων:for the chronology see 2-4n. ἐλάττονα τών προσοκίΘἑντων: Acoris had clearly decided that he would be backing a lost cause with resources he could hardly afford; cf. Kienitz, 1953: 87. Πολιοορκουμἑνίν ἐνερλῶσ: Tiribazus' return and the 2,000 talents had made a difference (4. 2 and n.). 8.2. τῶν ὃλων ἔχων τἠν ἡλεμονὶαν: see 2. 2 n. 8. 2-3. ώς Σοῶλος ΣεσΠότῃ κτΛ: the negotiations broke down because Tiribazus insisted that Evagoras should obey the Great King 'as a slave his master' while Evagoras would only obey 'as one king another'. Grote saw this as a mere point of honour and suspected its historical truth because of Isocrates' failure to mention it when it would have been grist to his mill (x. 31 f. and 32 n. 1). Hill, 1940: 138f., thinks that more than a mere point of honour may have been involved and suggests that Isocrates' silence indicates that the unacceptable condition was imposed by Tiribazus on his own authority with the object of wrecking the negotiations; it did not form part of the official terms. He concedes that its precise meaning escapes us. More relevantly Spyridakis, -1801935: 66, saw the matter as one to do not with empty words and honour, but with the future legal status of

Evagoras; 'seiner königlichen freien Entschliessung' was at issue. Spyridakis might have used the term 'autonomy', for that is what Evagoras was holding out for, as Meloni, 1950: 332, recognizes. Against Spyridakis, Martin, 1963: 230ff., concludes that the dispute was purely verbal and Osborne follows suit: 'As for the point itself . . . it hardly amounts to much in real terms, since it merely defined specifically and realistically the relationship that being a client king of the Persian Empire usually implied . . . The concession made by Orontes was hardly other than a purely verbal amelioration of Evagoras' position' ( 1973: 530). As he therefore sees the dispute as lacking in substance, Osborne surmises that something deeper and more sinister must have been afoot (see next n.). The correct explanation to my mind is that of Spyridakis and Meloni as further elucidated by Chaumont, 1972: 187f. D's term Σοῶλοσ should be seen as corresponding to the term bandaka used in Persian official language to designate satraps and other high functionaries ( 1972: 187 n. 4. On the evidence of Darius' Behistun inscription where Badaka or Bandaka occurs several times; see Kent, 1950: 121 col. ii line 30; 122 lines 49f.; 125 col. iii lines 13, 31; 126 lines 56, 85; etc.). As a bandaka Evagoras would have relinquished his autonomy and on his death his kingdom would have reverted to the Great King who might or might not have chosen to allow Evagoras' sons to succeed their father. Salamis would have become a Persian royal estate to be disposed of as the King pleased. Orontes' later concession, therefore, was a substantial one: Evagoras was to remain an autonomous vassal of the King, paying tribute, but otherwise his own master. We may compare Tithraustes' proposal to Agesilaus in 395 that the Greek cities of Asia remain autonomous but tributary (X. H. 3. 4. 25; cf. Lewis, 1977: 122ff.). Isocrates at 9. 63 seems to be making the same point and he is not, therefore, as silent on the point as Grote and Hill suppose him to be: 8. 3-4. οù συλχωροûντος Σἐ τοû ΤιριßάὠοΥ όρóντίς κτλ: according to D Orontes conspired against Tiribazus alone. Theopompus, -181-

on the other hand (as reported by Photius), represented a collusion between Orontes and Evagoras, though we should not suppose that Evagoras actually participated in the arrest of Tiribazus: καὶ ώσ ΓιρόßαὠοςἑΠολἐμίσεν, ὅΠΩσ τε Ηὐαλòρα ἐΠεßοὐλενσεν, ὅΠΩς τε αὐτòνòναλóρας πρòσ ßασιαßαλΩν σννἑλαßε μετ' 'Oρóντον (FGH 115 15 103. 9). Theopompus' assertion, in so far as he has not been misunderstood by Photius, is that Evagoras was the originator of the plot while its execution was left to Orontes. In what way could Tiribazus have conspired against Evagoras? We can only guess. But if it were in Tiribazus' power to allow Evagoras to remain an autonomous vassal, as Orontes later did, and he refused, insisting on the σοῶλος Σεσπóτί condition, then Theopompus may have interpreted this as a conspiracy, especially as Tiribazus was perhaps known to be ill disposed towards Evagoras, the friend of Conon and Athens (Tiribazus had favoured Sparta in the Corinthian War, even going so far as to arrest Conon: X. H. 4. 8. 15 ff.). What could Evagoras hope to gain by the removal of Tiribazus? Clearly better terms. He may even have thought that he stood a chance of restoring his position; Orontes had proved himself ineffective in Tiribazus' absence. What did Orontes stand to gain? Obviously the supreme command. He may well have resented the pre-eminence of Tiribazus (Cf. 2. 2 n.). It should be sufficient to have outlined the possible motives of the three men. Further speculation on the basis of our meagre evidence would be useless. Osborne, 1973: 528ff., exhaustively, and not always convincingly, considers all the different possibilities. This is partly because of his conviction that the unacceptable condition lacked real substance and was a mere pretext for not effecting an agreement. Once the importance of the Σοûλος ςεσΠóτȗ and ßασιλἐα ßασιλεî alternative is granted, however, it becomes perfectly clear why Evagoras should choose to hold out and even plot to have Tiribazus removed. The charges laid against Tiribazus could hardly have been bettered and were well calculated to 'suggest to Artaxerxes that he had a second Cyrus on his hands' (Osborne, 1973: 534, though surely not one aiming for the throne: 10. 2 n.). They were all the more potent as there was more than a grain of truth in them. The siege was indeed protracted and so were the negotiations. And he was known to be well disposed towards Sparta (10. 2; X. H. 4. 8. 15 ff.). And the troops favoured him more than Orontes. This last -182constituted the most serious charge. For Tiribazus' defence see 10. 2 nn. 8. 5. οῶ πράὠαντος τò ΠροσταχΘἐν: cf. Plut. Mor. 168e; Polyaen. 7. 14. 1. ΚαΣονσόοΥς: according to Strabo 11 C 508 their home was in northern Media; cf. Steph. Byz. s.v.: μεταὠὐ τἠς ΚασΠόας Θαλὰσσίσ καὶ τοῶ ĪĪUóντον. Most of what we know about them is connected with Persian attempts to subdue them. They were first conquered by Cyrus it seems (2. 33. 6), but subsequently rose in revolt. At the time of his death in 405/4 Darius II had been campaigning against them (X. H. 2. 1. 13) and on this occasion Artaxerxes II had to come to none too honourable terms with them in order to extricate his battered army from their territory if, as generally thought (cf. e. g. Syme, 1988: 139), Plut. Artox. 24f. should be connected with this war. Sekunda, however, 1988: 38f., makes a good case that Plutarch refers to another Cadusian campaign which took place in the late 370s, the same campaign mentioned in the prologue of Book 10 of Pompeius Trogus. Sekunda is also right of course in identifying the campaign in which Datames first distinguished himself ( Nepos, Dat. 1) with the first of these wars ( 1988: 38). On the other hand he is wrong to date this, as well as all the events narrated by D in 15. 8-11, in 385/4, simply because that is D's apparent date for them. D's text is rather more complicated than that (above pp. 143 ff.). The Cadusians were eventually suppressed by Ochus and they are found serving against Alexander (17. 6. 1; Justin 10. 3. 2 ff.; cf. Arrian, An. 3. 8. 4; 3. 11. 3). For their hardihood and warlike spirit cf. X. Cyr. 5. 2. 25; Strabo II C 523 f. See Pragek, 1906-10: 1. 47, 127 f.; Meier, RE Suppl. vii. 316 f.; Olmstead, 1948: 371, 401, 424, 490; Syme, 1988:

137ff. Σιείλκετο...Σικαστ̓7ρίου: 'the trial was postponed'. We cannot know how long the campaign against the Cadusians lasted. 9. 1. χαλεΠὣσ. . . ἐλκαταλειΠóντων: the revolt of Glos must have taken place at this time; see 9. 3-4 nn. Σ' ᾀρóντ̓7ς ΣιαΣεὠὰμενος κτλ: why did the fighting continue after the arrest of Tiribazus? If Evagoras had no hand in the plot, no explanations are necessary; on assuming the supreme command Orontes tried to succeed where Tiribazus had failed. But if Evagoras and Orontes had plotted together as Theopompus -183-

suggests (FGH 115 F103. 9), then it would seem that either their co-operation went no further than the downfall of Tiribazus, or that one of them (or perhaps both) went back on his word once Tiribazus was out of the way. This last stage of the war appears to have lasted for some three years (2-4n.) though it is treated summarily by D. Orontes' attempts to bring Evagoras to heel were severely embarrassed to begin with by the revolt of Glos who sailed off with what must have been a considerable part of the fleet. Then he may well have been hampered by a shortage of funds; and he does not appear to have been very popular with the troops, partly perhaps because their pay was not forthcoming (cf. Isocr. 4. 153), and partly because they felt aggrieved at the arrest of Tiribazus. At any rate Orontes was in the end forced to grant Evagoras the terms which Tiribazus had refused. And these were in fact the final terms, the settlement which the Great King ratified. Judeich's intricate reconstruction of the end of the war (2-4 n.) is unnecessary. 9. 2. Σιετἡ χρóνον: MS M (and its derivative and 'edited' F) reads Σιετἡ. Our other major MS P, on the other hand, reads Σιετι (and its derivative X ἔτι Σἑ).The number may therefore be corrupt. 9. 3. A jump back in time takes place here, from 380 to 383; see 2-4 n. and next n. 9. 4. εὐΘὑς οὓν: he must in fact have moved to Ionia (18. 1). Judeich's assertion that Glos was murdered before he rose in open revolt is unwarranted ( 1892: 130, accepted by Swoboda, RE vii. 1432; Hill, 1940: 139; Kienitz, 1953: 88). It is clear from 18. 1 that Glos did revolt (ἀΠοστὰτ̓7ς ὢν TAUοὓ ßασιλέως)and the sentence οὺ σννετέλεσε τἡν Προαίρεσιν refers to his plan to fight against the King in conjunction with Sparta and Egypt. His murder occurred in Ionia. Cf. Beloch, GG iii. 1. 98f. Ακοριν: Acoris was probably dead by the summer of 380 at the very latest (2-4 n.). Thus hardly enough time is left for all the events of the Cypriot War if the invasion of the island is dated in 381. Beloch, therefore, who furthermore dates Glos' revolt after the war in 379, assumes that D is mistaken. The alliance, he suggests, was with Nectanebos, not Acoris (GG iii. 1. 99. Likewise Hill, 1940: 139 n. 5; Kienitz, 1953: 89). This is both arbitrary and unnecessary and it amounts to tampering with the evidence in order to make it fit the theory. It is much better to accept an earlier date for both the invasion and the revolt of Glos. -184-

Πρòς Σἐ ΛακεΣαιμονίους κτλ: was such an alliance concluded and when? Those scholars who have accorded the alliance a few words (it is ignored by most, even by Bengtson in Svt) have on the whole accepted it as historical, in particular Beloch, GG iii. 1. 99; cf. Judeich, 1892: 190; Meyer, GdA v. 314f.; Hill, 1940: 139. There is obviously no way of deciding conclusively and one might choose to maintain (with e.g. Lewis, 1977: 58 n. 59) that there is no more substance to it than mere rumour. However, such an alliance was not at all unlikely at this time. There is first of all the question of chronology, though to contest that any one time between 385 and 379 was more appropriate for the alliance, as Ryder, 1963a: 105 ff., does, is to set off on the wrong foot. In fact Ryder's article is an example of how not to approach such problems. His greatest handicap is his acceptance of Beloch's date, 380/79, for the revolt of Glos. He then attempts to make a case that this date is more suitable for an alliance between Sparta and Glos than '384, when the first Spartan act of aggression since the King's Peace, the attack on Mantinea, had hardly begun' (1963a: 106). The revolt and murder of Glos nevertheless belong probably in 383 or at least not later than early 382 (previous n. and 2-4n.). D's narrative of the Cypriot War is likely to be telescoped so that the fact that he places the revolt and alliance under 385/4 is of no great significance. Ryder is aware of D's tendency to place a series of events into a single year to which not all may belong. Yet he accepts 385/4 as the date to be derived from D's text and proceeds to argue its unsuitability. In fact, enough appears to have been happening in Greece by 384 (5. 2-3 n.) to invalidate Ryder's objection. Equally, Ryder's main thesis fails to convince because his method of dealing with D's evidence is inconsistent. Beloch (GG iii. 1. 99) suggests that as Sparta was being fairly successful in Greece in 379, she was prepared to ally herself with Glos and try to regain her old power in Asia Minor. Ryder objects that on the evidence of D the Spartans were not interested in Asia Minor, but saw the alliance as a means of strengthening their position in Greece (1963a: 107). But this is to misrepresent 9. 4f. It is true that Glos promised to help Sparta regain her position of hegemony in Greece (which, by the way, she had on an even more secure basis in 379 than 384 -- an additional argument against 379 as the year of Glos' revolt and the present alliance), but that was only a -185side issue. D's (Ephorus') claim is that the alliance was aimed against the King. The Spartans were having second thoughts about the surrender of Greek Asia to the King and, sensitive to Greek criticism of their action, were eager to involve themselves again in Asia Minor. We may believe or disbelieve, but we cannot deny that that is what D says.

The probable date of the alliance between Sparta and Glos therefore is 383, early 382 at the very latest. It is not inconceivable that Sparta, or, more accurately, the dominant group in Sparta (that of Agesilaus), should contemplate new commitments in Asia Minor at this time. Greece lay at Sparta's feet once more as in 404, and many Spartans may have thought that the King had served his purpose by making this possible. Agesilaus' position vis-à-vis the King's Peace is not entirely clear (cf. Cawkwell, 1976a: 68 ff.). In 396 he had gone forth as a new Agamemnon to fight the barbarian and free Greek Asia. The task was interrupted by the outbreak of war in Greece. But there is no reason to assume that he discarded his dream. He had promised to the Greeks of Asia to return once Greece was quiet (X. H. 4. 2. 3). The proposals of Glos who possessed a strong fleet (an indispensable instrument of overseas empire as Agesilaus discovered to his own cost in 394), and who seemed to have ample funds (9. 3), may have seemed attractive and opportune and not to be rejected. Still, the question remains whether we should be speaking of an alliance, a firm commitment, or a mere understanding between the two sides. Whatever it was it was probably done secretly. It would not have been in the Spartans' nature or interests to advertise the fact until they were ready to act. They never did because both Glos and his successor, Tachos, were murdered (18. 1 f.). Similarly in 401 Sparta had supported Cyrus wholeheartedly, but not openly (14. 19. 4f.; 14. 21. if.; X. H. 3. 1. 1). 9. 5. ἐκΣòτους πεποἰ7κἑναι: Cf. 15. 10. 2 and 15. 19. 4 where ἒκΣοτοι is likewise used to describe the fate suffered by the Greeks of Asia under the King's Peace. All three instances recall Isocr. 4. 122, 141. -186-

10-11. The trial of Tiribazus It is impossible to discover exactly how much of this account is historical. In general the question concerns Ephorus' sources for 4th-century eastern history, a thorny problem; see pp. 107 ff. 10. 1. Σικαστὰς τρεἰς: Herodotus' evidence argues for a permanent body of royal judges (3. 14. 5; 3. 31. 3; 5. 25; 7. 194. 1; cf. Plut. Artox. 29. 8-12; Aelian, VH 1. 34; Ctesias, FGH688 F27; Amm. Marc. 23. 6. 82). Their number was perhaps indefinite; cf. Meyer, GdA iv. 1. 30f.; Lewis, 1977: 23. X. An. 1. 6. 4 probably represents an improvisation by the rebel Cyrus. For non-Greek sources mentioning royal judges and for a brief discussion of this obscure subject see Tuplin, Coinage and Administration, 119 f. For Persian concern with justice, see Ehtécham, 1946: 77 ff.; Frye, 1962: 104 ff. 10. 2. γεγεν̓7μἑν̓7ν. . . ὑπακούσειν: There is clearly a lacuna here (as first perceived by Estienne) which renders an exact emendation of the text impossible. But the sense of the passage is easily recoverable from the context: Tiribazus defended himself against the charge of collusion with Evagoras by pointing out that his own terms had been tougher than those granted Evagoras by Orontes. μἡ χῤ7ματίὠειν. . . Θανάτου: the King's death, the revolt were successful. But is it at all credible that Tiribazus (who was not of royal blood), if indeed traitorous, was hoping to supplant the King, and that such an unlikely charge could therefore have been brought against him? Many years later he did plot against the King's life, but under very different circumstances ( Plut. Artox. 27 ff.). Perhaps the later occasion conditioned the present one. If there was any truth in the charge then Tiribazus was probably aiming at an independent command in the west. He did not at any rate deny that he had sent to Delphi. While the evidence for enquiries about victory in war is plentiful, the evidence for enquiries about revolts and the like is not, and it is partly unhistorical; cf. Parke and Wormell, 1956: i, nos. 70, 139, 201, 204, 327 (a response was obtained though both the enquiry and the response are probably unhistorical), 395, 414. Thus, and in spite of Plut. Mor. 378d (Σιὸ τῳ+̑ μἐν εὶς τò χῤ7στἡριον ἐνταὓΘα κατιóντι Παρεγγνὣμεν ὄσια Φρονεἰν, εὔΦ̓7μα λἑγειν), there is no reason why Tiribazus could not have obtained a response if he enquired about his chance of success if he revolted; and of course he could have phrased his enquiry differently, but still with the same intention in -187

mind. Cf. Parke and Wormell, 1956: i. 212; Fontenrose, 1978: 266 (H73). περὶ Σἐ τῆς Φιλίας κτλ: Tiribazus had favoured Sparta in 392 and was later instrumental in enforcing the King's Peace which worked in the interests of both Sparta and Persia (X. H. 4. 8. 15 ff.; 5. 1. 25ff.). Such a charge against Tiribazus would have been especially effective because of Artaxerxes' deep suspicion of Sparta, the result of her support for Cyrus (cf. Osborne, 1973: 535). According to Dinon the King considered the Spartans to be the most shameless of all mankind (FGH 690 Fig). τοὺς κατἀ τἡν ̔Εσίαν ̔Ελλ̓7νας. . . ἐκΣòτους: see 9. 5 n. 10. 3-11. 1. The historicity of these sections is discussed in the "Introduction", pp. 107 ff. 10. 3. κατἀ γἁρ τινα κυν̓7γίαν κτλ: the King in this instance was in danger of being killed (as was the younger Cyrus at X. An. 1. 9. 6); hence the gratitude shown to Tiribazus for saving him (and to the anonymous would-be rescuer in the Anabasis). FGH 688 (Ctesias) F14. 43 is perhaps different in that the King on that occasion did not consider himself to be in danger. 11. 1. We may compare Herod. 1. 137. 1. 11. 2. Ορóντου. . . ἐσχάάταισ ὰ⊤ιμίαις: he was probably deprived of his province of Armenia. At any rate when he reappears in history he is said to be satrap of Mysia (90. 3 and n.).

12. The dioikismos of Mantinea

12. 1. τò μἐν Θἑρος. . . τοû Σἐ χειμῳνος ἐνστἁντος :the summer of 385 and the following winter; see 5. 1 n. and p. 137. Cf. Paus. 8. 8. 7: the siege was of short duration. πρóτερον εὶὠΘεισαν. . . πιστοτάτους τῳν συμμάχων: an obvious exaggeration. Mantinea was actively hostile to Sparta in the years 421-418 and passively so in the Corinthian War (X. H. 4. 5. 18; 5. 2. 2). What is said is far truer of Tegea (Herod. 9. 28. 3; Thuc. 5. 67. 1; X. H. 4. 2. 19). The object of the exaggeration was to heighten Sparta's dastardly deed in attacking a loyal ally. It is possible that Ephorus derived this information from an antiSpartan source. (For his sources see pp. 104 ff.) Or it may be that someone, Ephorus himself or a source, was guilty of an anachronism; the Spartans and Mantineans drew closer together later in the -188-

4th century. They fought side by side as allies at the second battle of Mantinea in 362 (85. 2), and were still allied in 352 and even as late as 331 against Alexander (16.39.3; Aesch. 3. 165; Beloch, GG iii. 1. 480f., 646). τÓû Παρὰ τὴν Μαντινειαν ΠÓταμÓû κτλ: X. H. 5. 2. 4 says that the river did not flow beside the city, but through it, and this seems correct. In any case at the rebuilding of the city after Leuctra the Mantineans, having learned their lesson (X. H. 5. 2. 7), caused the river to flow round the walls. Thus Ephorus (or his source) may have argued back from the topography of the city in his own day; cf. Underhill ad 5. 2. 4. Pausanias says very much the same as D: τóν τε ὄΦιν πÓταμòν ὰπÓΣτρἐϨαδ σΦίσιν ἐδ τò τεîχÓσ κτλ(8. 8. 7); cf. Frazer ad loc. and Fougères, 1898: 418ff., 433. Vial, 125, wrongly states that Pausanias agrees with Xenophon on this against D; and Fougères's assertion ( 1898: 420 n. 1) that D has confused the present action with that of Agis in 418 is unnecessary and almost certainly wrong. D at any rate omits the latter, and the two episodes are quite different. On this occasion the Spartans dammed the river flowing by the city (D and Pausanias), or through it (Xenophon), and turned it against the walls. For Agis' different stratagem in 418 see Thuc. 5. 65. 4 with Gomme's and Andrewes' nn. in HCT iv. 97f. and v. 457f. as well as Pritchett, 1965-85: ii. 41ff. The river no longer exists (the present-day Fidhias is not to be identified with it) and clearly conditions in the Mantinean plain must have been quite different in antiquity ( Pritchett, 1965-85: ii. 44; S. and H. Hodkinson, 1981: 266ff.). Pritchett's statement that the siege took place in the spring does not follow, however, for if there was more water in the plain in antiquity, autumn or winter would do as well; and D does say that the river was greatly increased from the rains (not melting snows). This rather indicates the autumn or winter seasons. And see n. above. D, who does not mention Agesipolis, ascribes the stratagem to the Spartans in general (cf. 5. 4n.). Both Xenophon and Pausanias on the other hand attribute the plan to the young king (cf. H. 5. 2. 4: (Agesipolis) ἐνἐßαλε . . . ἐδὴÓν . . . ὢρνττε . . . ῲκÓδóμησε . . . αìσΘóμενÓΣ . . . νÓμíσαΣ . . . . ἀΠἐχωσε). The view that the attack on Mantinea was planned by Agesilaus and that an unwilling Agesipolis was saddled with its execution ( Smith, 1954: 275ff.; Rice, 1974: 166ff.; cf. Cartledge, 1987: 259f.; Tuplin, 1993: 89, -189-

93) is unconvincing. Most, if not all, Spartans may have felt that democratic Mantinea needed to be taught a lesson, and Agesipolis could hardly have been more vigorous or acted with greater determination; cf. Polyaen. 2. 25 and see 12. 2n. Xenophon's treatment of Cleombrotus is quite different. He never fails to mention that king's alleged incapacity as a general. 12. 2. τῳν οìκιἳν ΠιΠτουσΘῃν : cf. X. H. 5. 2. 5. ἅΛΛο μἐν οùΘἐν κακòν . . . μετοισΘῃναι: this might at first glance belie the opinion that Ephorus had castigated the dioikismos of Mantinea. But it is possible that the original statement was ironic and directed at those defenders of the Spartan action (e.g. X. H. 5. 2. 7) who claimed that the dioikismos was no chastisement, but a benefit to the Mantineans; cf. Polyb. 4. 27. 6. Though not the severest of punishments, dioikismos was punishment enough for a Greek. We may note the concluding sentence of the chapter: ΣιóΠερ ήναγκἀσΘńσαν τὴν μέν ìΣíαν PIατρíΣα, κατασκάΠτειν, εìΣ δέ τἀΣ κώμαΣ μετοικῃσι Xenophon adds that sixty ἀργολὶΞοντεδ and democratic leaders were expelled at the city's surrender, and again Agesipolis' behaviour with regard to these men does not support the view (12. 1 n.) that the young king acted as the unwilling tool of Agesilaus. The exiled leaders had feared execution. But then Pausanias intervened and persuaded his son to allow the men a safe passage out of the city (X. H. 5. 2. 6). It is clear that Agesipolis had full powers to deal with the men as he pleased (the dioikismos was the decision of the 'Lacedaemonians' whatever that means, presumably the assembly), and the impression one forms from Xenophon is that the young king would have acted differently, either because he did not like the Mantinean democrats or because he could not have cared less, had not his father intervened. Or he might have done nothing and allowed the oligarchs to butcher their enemies. In saving the democrats, therefore, Agesipolis was probably doing his father a favour. In any case, we cannot be certain that Agesilaus would not have spared these men had he been in command. He is not known to have been a cruel man; on the contrary, cf. X. H. 5. 4. 55; Ag. 1. 21f. The case of the Theban Ismenias was rather different. No difference of opinion between Agesilaus and Agesipolis is to be seen in the instance of Mantinea at any rate. But see 19. 4n. for the tradition that Agesipolis espoused Greek autonomy. -190-

Some of the exiled democrats found refuge in Athens: 5. 2-3 n. and cf. Osborne, 1981-3: ii. 51f. εἰΣ τὰΣ U=1F00ρΞαίαΣ κώμαΣ: see 5. 4 n.

13. 1-4; 14. 1-2. Dionysius and the Adriatic

These passages are as important for an estimate of Dionysius' policy in this area as they are problematic. The evident lacuna at 13. 5, plainly of considerable length, and the difficulty about the identity of the place or places named in 13. 4 and 14. 2 do not help; and it may also be that, as so very often, D's use of his sources has been maladroit, though the degree to which that is so is lessened somewhat if the text is emended in the manner indicated by Stroheker (see below). For the context and chronology see pp. 81f.; 7. 2n.; 13. 4-5 n. 13. 1. τóν Àδρίαν ... τóν ≤óνιν: for the names see Beaumont, 1936: 203f.; Sanders, 1987: 44. ίδιοΠοιεῖσΘαι . . . κατασκευάσ̃, και ΠóλειΣ ἕΞειν κτλ: thus Vial, following Stroheker. Vogel and all other editors by contrast put the comma after ἰδιοΠοιεῖσΘαι and print the reading ἔΞἡ of MSS X and F. The building of the cities is thus entirely subordinated to the plan of gaining control of the Ionian poros (the Straits of Otranto) and the further alleged intention of sacking Delphi; and the initial statement about the decision to send out colonies to the Adriatic is in this way left isolated and mysterious. Stroheker's emendation (he prefers the reading ἒΞειν of MS A (Parisinus Coislinianus) which is nearest to the reading ἒΞει of the oldest and principal MSS P and M: 1958: 120f., 224 n. 78) makes better sense of the text (see Vial, 18 n. I): Dionysius had two aims and they were connected to some extent. He did not simply want to control the Straits of Otranto, but he also intended to plant colonies in the Adriatic Sea. Some attempt is then made in D (the miserable remains of what must have been an adequate account in Ephorus) to define the two aims, the first in 13. 1-3 and the second in 13. 4. Still, D's credit is not thereby saved for to his mind clearly (though surely not to that of his source), as the narrative goes on to show, Dionysius' actions in the Adriatic, Illyria, and Epirus were determined by an intention to plunder Delphi. ἔσΠευδε γἀρ . . . Ξρńμάτων: most commentators rightly dismiss this as nonsense. Nor does either the text or common sense -191warrant the view of many that Delphi is a mistake for Dodona, and that it is that sanctuary which Dionysius planned to pillage. Ephorus is likely to have presented the story as a rumour (Φασιν or λἐγεται), very much as Xenophon does that of Jason and Delphi (H. 6. 4. 30). Not surprisingly, there is a tradition (which reaches absurd lengths with Cic. De nat. deorum 3. 34) that Dionysius was not very respectful of gods and their temples, but there is no good evidence that he robbed Greek (as opposed to barbarian: 15. 14. 3f.) temples (cf. Caven, 1990164). On the contrary, it is perfectly clear that Dionysius went to considerable lengths to establish and maintain good relations with the major Greek sanctuaries (cf. 14. 109; 16. 57. 2f:, Tod 113 Lines 9ff.). συλῃσαι: Estienne's correction (printed by Vogel) of the MSS σνλήσειν (retained by Vial, 125 n., rightly perhaps) is demanded by the syntax. 13.2. διò καὶ πρóΣ ≤λλυριοùΣ . . . συμμαχὶαν: Svt 247. The alliance was probably connected with the foundation of the colony (or colonies) mentioned at 13. 4. If so, it must have antedated the attack on Epirus by several years (13. 4n.). What Illyrians were these? Several separate tribes or nations are mentioned by the ancient geographers in southern Illyria. Were they all independent kingdoms or did they already at this date form a single state, as they certainly did after the middle of the 3rd century? The latter is the view of Papazoglou, 1965: 143ff. Contra Hammond, 1966: 239ff. For a compromise (and perhaps correct) view see P. Carlier in L'Illyrie méridionale 1987, 39ff. On the Illyrians in general see Wilkes, 1992: esp. 91ff. (citing all the sources and extensive bibliography). 13. 2-3. Àλκἐτου τοû Μοττοû κτλ: though the Illyrians were in the end defeated and expelled with Spartan help, it is not necessary to assume (with e.g. Woodhead, 1970: 505) that Alcetas was expelled with them. It is perfectly possible that he made his peace with his people and as a consequence restored to the throne with the consent of Sparta. If so, he must have made a volte-face at some stage because by the mid-370s he is the ally of Athens (36. 5 and n.). As far as IG ii2 101, is concerned, an Attic decree of the year 373/2 in honour of a certain Alcetas, son of Leptines, of Syracuse, its circumstances remain quite obscure and it may well have nothing to do with the Epirot king ( Woodhead, 1970: 505 f.). -19213.3. κατγαγον: the reading of all the MSS, rightly retained by Vial; see her n. p. 126. ΛακεδαιμóνιÓι: see 5. 2-3 n. 13. 4. Πάριοι . . . Φάρον: FGH 70 F89. Foundations of cities were of particular interest to Ephorus. Pharos is modern Hvar, the island off the coast of Croatia. The actual settlement was probably at Stari Grad: Princeton Encyclopedia, s.v.; cf. Cibin, 1919: 125. We have no other date for the foundation, but D's date, 385/4, is generally accepted and though it may not derive from the chronographer it must be approximately right because of the wider context in which the foundation is set. συμπράΞαντοΔ αùτοῖΣ Διονυσίου: rather more is implied by way of help to the Parians than is suggested by Beaumont ( 1936: 203) and Woodhead ( 1970: 512). Indeed, it is not impossible that Dionysius contributed colonists to the new foundation. In any case, the connection with Syracuse was remembered many years later: Cibin, 1991: 125ff. 13. 4-5. ο↾τοΣ γὰρ ἀποικίαν κτλ: Dionysius, we thus learn, had founded a colony in the Adriatic 'not many years before' (i. e. the foundation of Pharos). Setting out from this colony he proceeded to do something else in the Adriatic which escapes us because of the lacuna. When the text resumes we are back in Syracuse and with

Dionysius' various projects there. So far there is general agreement. The problems start when we try to identify the colony founded, to provide it with a date, and to determine what else was done in the Adriatic. As far as the date is concerned, a date circa 401 ( Vanotti, 1991: 107ff.) is not impossible, but one after the treaty with Carthage in 392/1, when Dionysius became involved in Italy, is altogether more likely (cf. Stroheker, 1958: 223). What of the place? MS M gives its name (in the accusative) as Λισσóν, MSS P and X as Λίσον. When shortly afterwards the Parian colonists were attacked by the neighbouring Illyrians they were assisted by Dionysius' eparchos of a city the MSS call (in the dative) Λίσἣ (PX) or Λίσσἣ (M): 14.2. Are these one and the same place or two different places? And which? Many scholars are of the opinion that in both passages D is speaking of the same city though they disagree about its identity. Evans, for example, in Freeman, iv. 223ff., and Beloch, GG iii. 1. 118 n. 2, thought that the place in question is Issa, modern Vis, the island with the excellent harbour south-west of Pharos ( Fraser in L'Illyrie méridionale 1993, 169 n. -19317, is inclined to agree). Beaumont, on the other hand, 1936: 202 f., Woodhead, 1970: 507f., and Braccesi, 1977: 226ff., opt for Lissos, modern Lesh (Alessia) at the mouth of the river Drin in northern Albania. Other scholars by contrast believe that the reference at 13. 4 is to Lissos while that at 14. 2 is to Issa: Casson, 1926: 316; Stroheker, 1958: 122ff.; Hammond, Princeton Encyclopedia, s.v. Lissos. I agree with Vial that, palaeographically speaking, Lissos should be read at 13. 4 and Issa at 14. 2. One problem with this is that archaeology does not seem to support the first reading: the pottery at Lissos is Hellenistic or later and the earliest of the fortification walls seems to date to the late 4th century and to be of native Illyrian construction ( Prendi and Zheku, 1971: 35ff.; cf. Fraser in L'Illyrie méridionale 1993, 169 n. 17 and 174 (comment by N. Ceka). Contra Beaumont, 1952: 69, 72). This, however, is hardly decisive. Dionysius' foundation may have been on a small scale and of short duration and therefore difficult to trace on the ground. For the site and its importance see Caesar, BC 3. 26-7; Polyb. 8. 13-15; Fluss, RE xiii, s.v. Lissos (2) 731ff.; May, 1946: 54ff.; Walbank, HCP ii. 90f. Why would Dionysius want to plant a settlement at Lissos? Beaumont suggests that the chief aim of the colony was the control of the Straits of Otranto (1936: 203). But Lissos by itself could not have accomplished that (cf. Evans, in Freeman, iv. 223 f.). Woodhead's suggestion is even less compelling: 'Its most evident suitability was as a point of contact with Dionysius' Illyrian allies, through which the supply of arms could be channelled to them for the intended invasion of Epirus to the south' (1970: 507). But Dionysius did not need to found a colony for that, and one in such a locality too. Woodhead, of course, goes on to suggest that 'although Lissus was "occupied", and could thus be described as a "colony", it is likely that its usefulness came to an end with the failure of the Epirot scheme' ( ibid.). But D says nothing of the kind. D in fact does not connect Lissos itself with the Illyrian invasion of Epirus. Nor does he claim that this colony alone was meant to achieve control of the Straits or anything else. What he says, as we have seen (13. 1 n.), is that cities (in the plural) were to be founded in the Adriatic, both for the purpose of appropriating the Straits and so that the ships of the tyrant would be provided with -194harbours. The plan therefore envisaged a string or network of coastal settlements and it was launched it seems with the planting of Lissos. The unfortunate lacuna at this point prevents us from knowing what else Dionysius had gone on to do from Lissos, but it is very probable (as Stroheker, 1958: 123f., argues) that the missing text told of the planting of a Syracusan colony on the island of Issa and the installation there of an eparch; information that recurs at 14. 2. Whether anything else was mentioned is a great deal more doubtful -- which does not of course mean that Dionysius did not expand further in the Adriatic. General probability and certain other indications argue that he did. In the mid-320s Athens founded a colony and a naval station somewhere in the Adriatic. The purpose, we are told explicitly (Tod 200 lines 2.17ff.), was so that Athens would have her own permament commerce and supply of corn from the area, while by the establishment of a naval base the Etruscan pirates would be kept at bay. Athens had been having problems with pirates in the Adriatic, an area clearly of great importance to her, and had been taking measures to protect her shipping, since at least the mid330S (Syll. 962 lines 277ff. Cf. Dinarchus fr. XII Conomis; Hyperides fr. 8, Burtt). That Dionysius had similar concerns to these in the Adriatic fifty years earlier can, in my view, be taken as certain. When the younger Dionysius founded two cities in Apulia in the early 350s his aim, we are told, was to safeguard the Straits of Otranto against pirates whose activities were keeping merchant shipping out of the Adriatic (16. 5. 3; cf. 16. 10. 2). There can be no doubt that the younger Dionysius was following the policy of his father in this. A conception of the plans for, and actions in, the Ionian poros and the Adriatic Sea of the Dionysii such as that of Evans (Freeman, iv. 220ff.) and Stroheker ( 1958: 120ff.; cf. also Caven, 1990: 149ff.) is to my mind a great deal more credible and in accord with the evidence than Woodhead's 'nihilistic' view which reduces everything to 'a brief and unhappy meddling by proxy in the affairs of Molossia, a short-lived naval station at an unpromising coastal site on the fringes of Illyria, and a helping hand on two occasions for a group of colonists from Paros' (1970: 512.). Beaumont is less negative, but he limits Dionysius' interests to Lissos and the Straits of Otranto (1936: 202f.). The extent of Dionysius' achievement in the Adriatic is another matter. That Issa was a Syracusan foundation of the time of -195-

Dionysius is a common enough view and Fraser has recently advanced impressive new evidence in support of it by his study of the onomasticon of Syll. 141, the 'foundation' document of Kerkyra Melaina (Korcula), the island east of Issa ( L'Illyrie méridionale 1993, 170ff.). Inter alia, this interesting inscription lists the names of some 245 Issaeans who founded Kerkyra c.300 Bc. Contrary to what other scholars have thought Fraser, with the authority of his monumental

Lexicon of Greek Personal Names (LGPN) behind him, is now in a position to say that the names in the inscription are good Doric and West Greek names. Furthermore, the formal division of the large number of persons listed into the three Dorian tribes (which makes Syll. 141 a remarkable and indeed unique document) suggests to Fraser that Issa was an officially sanctioned and planned Syracusan foundation. This, he thinks, took place in the middle of the 4th century or earlier. Important though this is for future study of the ethnography of Syracuse in the first half of the 4th century (as Fraser points out), for our present purpose it seems to me that Fraser's findings constitute weighty evidence for the Dionysii having founded Issa. As Syracuse and Greek Sicily in general rapidly deteriorated after c.357, such a colony must have been sent out before that date, most likely by Dionysius I. Of other possible sites in the Adriatic Adria near the mouth of the Po is the most likely to have been refounded by Dionysius, or at least strengthened with the sending of more colonists (cf. Evans, in Freeman, iv. 226f.; Stroheker, 1958: 125; contra Beaumont, 1936: 202, but his objections are weak). Ancona in Picenum is said by Strabo (5 C 241) to have been founded by Syracusans fleeing the rule of Dionysius. There is no need to reject this (as Beloch, GG iii. 1. 119 n. 2, does; see also Stroheker, 1958: 125). On the other hand, Dionysius could have taken over the city at a later stage as, for example, he took over Aetna in Sicily (14. 7. 7; 9. 6-8; 14. 2). Finally, that Philistus participated to some extent in Dionysius' activities in the Adriatic is likely enough. To suppose, however, that his role was a major one and that consequently Dionysius' colonizing enterprises ceased after Philistus' exile (cf. Gitti, 1952a: 161ff.; id., 1952b: 225ff.; Sanders, 1987: 44) goes beyond the evidence. If Dionysius showed no further interest in the Adriatic after the late 38os (and we do not really know if that is the case) -196-

one might as well blame the outbreak of the Third Carthaginian War in 383/2 for that. 13. 5. It is generally acknowledged that there is an extensive lacuna at this point, either before (Vial) or after (Vogel) ΔιοννσιοΣ. σχολήν ἂγΩν κτλ: see pp. 81 f. The details and chronology of Dionysius' varied building activity at Syracuse largely escape us with the exception of the north wall of Epipolai, the building of which prior to the declaration of war against Carthage in 398 is well and accurately described in D 14. 18. 2ff. See Drögemüller, 1969: 98ff. Also Lawrence, 1946: 99ff.; id., 1979: 117.

Archon Dieitrephes, 38413 14. 1. ΔιοτρέΦńΣ: thus all the MSS and Dion. Hal. Epist. ad Amm. 1. 5. Inscriptions, however, give ΔίειτρέΦńΣ and this must be the correct form of the name. The corruption clearly occurred early. Develin, 219. ↾πατοι: repetition of the college (392 V) given at 14. 103. 1 (389/8). Cf. Livy 5. 31. 2; CAH2 Vii. 2. 636; Perl, 1957: 79, 103. ΔίκΩν ΣυρακóσιοΣ : Dicon, a famous runner, was from Caulonia in Magna Graecia. That city had been razed to the ground by Dionysius in 389, its free inhabitants transferred to Syracuse (14. 106. 3). Dicon's change of ethnic, therefore, was due to this fact and not to bribery as Paus. 6. 3. 11 says. 14. 2. ≤σσῃ+̃: see 13. 4-5 n. On Issa see KP s.v.; Fraser in L'Illyrie mgridionale 1993, 168; Kirigin in Greek Colonists, 292ff.

14. 3-4. The raid against Etruria 14. 3. χρńμάτΩν ἀποúμενοΣ κτλ: our knowledge of the economic aspect of Dionysius' rule is little short of lamentable. The few stray remarks in the sources do not even allow us to ask the right questions: ps.-Arist. Econ. 2. 1349a 14-135Oa5; 1353b 20-6. See the discussions by Evans in Freeman, iv. 230ff.; Andreades, 1933: 105ff.; Bullock, 1930: 260ff.; Stroheker, 1958: 161ff.; and on the tyrant's coinage, Kraay and Hirmer, 1966: 280f., 287ff.; Kraay, 1976: 231ff.; Boehringer in Studies Thompson, 9ff. That Dionysius should be short of money is not surprising when one considers his many activities, some of which were definitely extravagant (cf. 14. -197-

109), as befitted a Sicilian tyrant. All the same, the major drain on his resources stemmed from his main preoccupation, the struggle against Carthage. In fact, by raiding Etruria Dionysius was probably trying to kill two birds with one stone -- he was punishing allies of Carthage whose piratical raids were notorious and at the same time he intended to use any loot he secured to finance a new war against Carthage (end of section 4 and 15. 1). But even this perhaps falls short of the true explanation of Dionysius' actions in the area (below). τριἡρειΣ ἐΞἢκοντα: according to Polyaen. 5. 2. 21 Dionysius had at his disposal not 60 but 100 triremes (thus also ps-Arist. Econ. 2. 1349b) and a number of horse-transports in addition. Dionysius' force, therefore, was significant, as is evident also in D from the fact that Dionysius went on after looting the temple to defeat the militia of Caere. See Caven, 1990: 191. πρóΦασιν . . . τῃ+̑ δ ἀλńΘείᾳ: the depredations of Etruscan pirates were real enough as the Athenians, for one, knew very well (13. 4-5 n.). The fact, however, that Dionysius plundered the sanctuary, coupled with a hostile, antiDionysius tradition which Ephorus did not entirely eschew (see pp. 68 ff.), suffice to explain the motive given here. But it is less than fair to Dionysius. Whether under tyrants or free, the struggle of Syracuse, the leading western Greek

city, with the Etruscans had gone on for at least a century. In 474 Hiero defeated the Etruscans in a naval battle off Cumae in Italy. Just over twenty years later the Syracusans, in two successive campaigns, aimed against the piratical activities of the Etruscans, or so we are told, ravaged the coast of Etruria and the islands of Corsica and Elba, both belonging to the Etruscans, even capturing the latter. Much loot and a great many prisoners were taken (cf. Pindar, Pyth. 1; D 11. 51= ML 29; D 11. 88. 4-5). In this instance, Strabo mentions the looting of the temple by Dionysius, but adds that the tyrant was sailing to Corsica at the time (5 C 226). This in fact was the time of the Gallic incursion into Italy and the aftermath of the sack of Rome, a fast-rising city on the periphery of Etruria and probably allied with Caere (Agylla -- see below). And since the 6th century Etruria (including Caere and Rome) had been friendly to (indeed allied with) Carthage and antagonistic to Greeks: Herod. 1. 166; Arist. Pol. 3. 1280a; Polyb. 3. 22 with HCP 1. 339ff. The Punic and Etruscan gold inscriptions found at Pyrgi, the harbour of Caere(below), in -198-

1964 (they date from c-500) further testify to this: Heurgon, 1966: 1 ff.; M. G. G. Amadasi, Le iscrizioni fenicie e puniche delle colonie in Occidente = Studi Semitici, 28 ( 1967), 159f.; Werner, 1974: 263ff.; Pallottino, 1975: 90. For the alliance between Rome and Caere and the close relations between the two see also Cornell in CAH2 vii. 2. 313ff. Justin (20. 5. 4-6) claims that envoys from the Gauls, 'qui ante menses Romam incederant', approached Dionysius who was engaged in war against the Italiots (the detail is hopelessly entangled) and concluded an alliance clearly aimed against their common enemies. When all this is considered together, and a few more hints in the ancient writers added (D 5. 13. 3f.; 14. 117. 7; Ptol. Geogr. 3. 2. 4; Strabo 5 C 220; Arist. Pol. 1. 1259a), it does not appear improbable that Dionysius had an 'agenda' in the Tyrrhenian Sea not so dissimilar to what he was aiming for in the Adriatic: Momigliano, Studia et documenta historiae et iuris, 5 (1939), 389f.; Stroheker, 1958: 128f.; Sordi, 1960: 62ff., 86; Toynbee, 1965: 24f.; Ogilvie, 1976: 159ff.; Braccesi, 1977: 203. ἱερὸν ἄγιον : ps.- Arist. Econ. 2. 1349b and Polyaen. 5. 2. 21 ascribe the sanctuary to Leucothea. Aelian, VH1. 20, joins Apollo with Leucothea. Strabo 5 C 226, on the other hand, speaks of Eilethyia. Α+̔γύλλης . . . Πύργοι: Agylla, according to Strabo, was the older name of Caere, the only Etruscan city to dedicate a treasury at Delphi (5 C 220). Pyrgi was Caere's port (Strabo 5 C 226). For both places see Princeton Encyclopedia, s.vv. οὐκ ἕαττον ταλἀντων χιλἱων : Polyaenus (5. 2. 21) says that 500 talents of coined metal were taken by Dionysius himself, but 1,000 talents of gold and many more of silver were removed by his soldiers.

Archon Phanostratus, 383/2 15. 1. ῥωμαÎοι κτλ : the college (391 V) cited at 14. 107. 1 (388/7), but with considerable variation; cf. Vial ad loc. and p. 127. The full list of six (not four) tribunes is given by Livy 5. 32. 1. For an explanation of what went wrong with D's list see Drummond, 1980: 60f.; cf. CAH2 vii. 2. 636; Perl, 1957: 79, 103; Pinsent, 1975: 46 n. 53. -199-

15-17. The Third Carthaginian War 15. 1. ἐπὶ δἓτούτων : as was his wont, D abbreviated an autoteles account of Ephorus (and the abbreviation here appears to be more severe than usual; see p. 80 ), which evidently covered a number of years, and set it under the archon in question, presumably because a particular episode in the account was noticed in the chronographic source under archon Phanostratus: the outbreak of the war and/or the invasion of Sicily by Mago. That the war cannot have been fought to its conclusion in a single year is apparent from some of the details given in 15. 15-17 as well as from the chronographic entry (cf. Beloch, GG iii. 2. 374) at 15. 24. 1: in 379/8 the Carthaginians sent an expedition to Italy and restored Hipponium. Clearly then the war was still being fought in this year (contra e.g. Hans, 1983: 69), but for how much longer? The usual chronology followed is that of Beloch who dates the two great battles which decided the war to 375 and the peace to 374 ( GG iii. 2. 376f.). He finds support for this arrangement in the fact that when Alcidas was dispatched to Corcyra the Spartans put it about that the squadron was heading for Sicily (15. 46. 2, under 374/3). In Beloch's view this can only have taken place in the aftermath of Dionysius' defeat at Cronium when the tyrant must have been in severe difficulties ( GG iii. 2. 376). Meyer, on the other hand ( GdA v. 169), prefers to date the end of the war in 376/5 so as to coincide with the termination of the Sicelica of Hermeias of Methymna (15. 37. 3). Beloch ( GG iii. 2. 377f.) rejects this, as does Stroheker (1958: 232 n. 30). The truth is we do not know when the war was concluded and 15. 46. 2 is no evidence that it continued until 374, even if Alcidas' squadron was ultimately aimed for Sicily. If we accept, as is generally done (cf. Beloch, GG iii. 2. 377; Huss, 1985: 140), D's date of 383/2 for its outbreak, there is no reason why Cabala and Cronium could not have been fought and peace made in, say, 378 (thus also Walbank, HCP ii. 495, but without argumentation), the year following the Carthaginian expedition to Italy (15. 24. 1), or soon thereafter. D dates to 379/8 the outbreak of a serious epidemic at Carthage and a revolt of her subjects (15. 24. 2-3), after (μετὰ δἓ ταῦτα) the refoundation of Hipponium. These two sections present problems which have been variously tackled by -200scholars. How do they relate to D's account of Dionysius' Third and Fourth Carthaginian Wars? At 15. 73. 1 (under 368/7) we are told that Dionysius was encouraged to start yet another war (the fourth and last) against the Carthaginians because he saw they were in a bad state, διά τε τὴν γεγενημἝνη παρ ατοîς λΟιμικ ὴν νóσον καì τὴν ἀπóσταιν των Λιß⇕Ων. Obviously the plague and the accompanying revolt are those recounted in

15.24.2-3 (under 379/8). It is equally obvious to Beloch that the plague cannot have lasted for ten years and that Dionysius would not have waited till the rebellion of the Carthaginian subjects was at an end before launching his attack. He finds support for this in Justin 20. 5. 10-11 where it is said that Dionysius was forced to return to Sicily from Italy by the arrival in the island of the Carthaginians, 'qui reparato exercitu bellum, quod lue' (Beloch is right to retain the reading of all the MSS against some editors, including Seel, who see 'lue' as an interpolation from 19. 2. 7) 'deseruerant, auctis viribus repetebant'. The Carthaginian general was Hanno and Beloch is doubtless correct that what we have here is the beginning of the Fourth Carthaginian War, in the course of which (as is clear from both D and Justin) Dionysius died. Beloch then points to the prologue to Book 20 from which we learn that the deeds in Africa of Hanno the Great took place after the death of Dionysius. These deeds, Beloch concludes, must be connected with the resubjugation of the Libyans. It follows that the epidemic broke out not in 379/8, but C.369/8, and that D was wrong to set it, as well as the revolt which it occasioned, under 379/8 ( GG iii. 2. 375f.). Stroheker, by contrast, prefers not to disturb D's date, but to use it to explain why the Third Carthaginian War dragged on till the mid-370s: the Carthaginian war effort had been paralysed as a result of the plague and the ensuing revolt and it was not until c.375 that they could intervene effectively in Sicily (1958: 132). The problem with Stroheker's solution (a fatal one in my view) is that both D and Justin connect the epidemic and the rebellion not with the Third but with the Fourth Carthaginian War. Beloch's arrangement is thus much to be preferred though perhaps not in its entirety. I see no reason for supposing that D attached 15. 24. 2-3 to 15. 24. 1 because of a relevant notice in the chronographer. This is not impossible, but it is more likely that (see pp. 940, finding himself somewhat embarrassed about what to -201-

include under archon Nicon, he abridged Ephorus' account of the plague and of the other problems which troubled Carthage in the period between her last two wars against Dionysius, and linked this to the chronographic entry on the restoration of Hipponium with the vague phrase μετὰ δἓ ταῦτα. Certainly at 15. 73. 1 D writes of the plague and the revolt of the Libyans as past events (γεγνημένης) when Dionysius declared war; of the revolt, but not of its suppression which, according to Justin (prol. 20) and certain hints in 15. 73 (see nn. there), was still in the future. All the same, it is still possible that the epidemic broke out much earlier than Beloch allows, as early even as 378 or 377, and this might explain why the Carthaginians did not follow up their victory at Cronium, but preferred to come to terms instead (15. 17. 5). ἐξήτει λαΒει+̑ πρóφασιν κτλ: Ephorus clearly considered Dionysius responsible not only for the present war, but also for the previous one, as well as for the last to be fought against Carthage by the tyrant: 14. 41. 1-2; 15. 73. 1. It is to be noted, however, that Carthage and the Italiots made common cause and that the first blow seems to have been struck by Carthage, which landed troops in Sicily and Italy. Dionysius reacted by taking the field himself; cf. Meyer, GdA v. 165f. By contrast with 14. 41. 1-2 and 45. 2-5, neither Dionysius' motives nor his public declarations are stated here. For Ephorus' view of the tyrant see pp. 68 ff. τὰςὑπò Κα⇁χηδονίους . . . πóλεις κρλ : the purpose and nature of Carthaginian involvement in western Sicily have attracted considerable attention. That it was some form of arche or imperium already at this time seems to be beyond question. All the ancient writers and most modern scholars are agreed on this, though the absence of detailed information allows divergent views. D reserves the term epikrateia (realm or dominion) for the respective Greek (Syracusan) and Carthaginian zones in Sicily (see McDougall s.v., though other terms are also used, e.g. chōra: 73. 2), a word which undoubtedly goes back to Philistus through Ephorus and Timaeus. This and the terms of the treaties concluded between Carthage and Dionysius (13. 114. 1; 14. 96. 3f.; 15. 17. 5; cf. 13. 59. 3; 14. 65. 2), which also go back to Philistus, leave little doubt that real control was exercised by Carthage in her own interest and that it was not merely a matter of bilateral alliances, more or less equal, between Carthage and the cities in the area. The word τεταγμένας used here also points to the fact that the cities in question were acknow-202ledged (by Dionysius) as belonging to Carthage. For discussions and further bibliography see Roussel, 1970: 54ff.; Whittaker in Imperialism, 59ff. (unduly sceptical in my view); Hans, 1983: 119ff.; Huss, 1985: 467ff., esp. 472f.; KOKALOS 28-9 ( 1982-3), 127ff. For the Carthaginians in Sicily, Tusa, 1988: 186ff. οἰκείως ἐχούσας πρòς ἀπóστασιν: presumably not the Punic communities; cf. Stroheker, 1958: 131; Hans, 1983: 70; Huss, 1985: 137f.; Caven, 1990: 192. 15. 2. Ι+̔ταλιώτας : Wesseling's emendation for the MSS παρóντας is necessitated by the context. It was a clever manœuvre to embroil Dionysius in a war on two fronts; cf. Stroheker, 1958: 132. The Italiots involved presumably were the Italiot League, founded (or refounded or perhaps simply extended: Polyb. 2. 39. 6 with HCP ad loc.) in 392/1 by way of a defence against Dionysius (14. 91. 1; cf. 14. 101. 1). If the League was disbanded in 389 by Dionysius following his victory at Eleporus as a prerequisite of the peace he granted to the Italiots (14. 105. 4), then what we find now must be a resurgence of the League achieved with Carthaginian support, the same support which saw the restoration of Hipponium shortly afterwards (15. 24. 1). In fact, it is not certain that Dionysius demanded the termination of the League in 389, and its ending by the tyrant and the neighbouring barbarians mentioned by Polybius (2. 39. 7) may well refer to the time of the capture of Croton (the leading city of the League: 14. 103. 4) by Dionysius a few years later (see below). An Italiot League, however, was in existence again soon afterwards it seems, centred this time on Taras and Heraclea. (Cf. Brauer, 1983: 55f.) Whether this League was a rump of the earlier one or a new creation (as Polyb. 2. 39. 7 would seem to demand) is impossible to say. τω+̑ν τεπολιτω+̑ν κτλ: the usual shorthand for Carthaginian military preparations involving citizen troops as well as large numbers of mercenaries. καταστήαντες . . . ßασιέα: i. e. Mago, one of two (or possibly more at this stage?) sufetes, was chosen to command the campaign; see Huss, 1985: 458ff.; but cf. Scullard, CAH2 vii. 2. 487, 490; Lewis, CAH2 vi. 128 and n. 40; Ameling, 1993: 80ff., 90ff. All three scholars incline to the view that sufetes and kings are not to be

identified. Institutions and offices at Carthage are a crux. See Ameling, 1993, especially 67ff., for a recent discussion (citing earlier contributions). It seems clear from the present passage (cf. -203-

13. 43. 5 and 13. 54. 1 with Ameling, 1993: go n. 116) that, at this stage at any rate, the king did not ex officio command the state's forces. πολλὰς μ⊎ριάδας στραιωτω+̑ν: the vagueness of the statement is an indication of the severity of D's abridgement of his source (p. 80 ). For Carthaginian land and naval armaments cf. Scullard, CAH2 vii. 2. 493ff.; Ameling, 1993: 190ff., 210ff. 15. 3. πολλαὶ μέν ου+̑ν κατὰ μέ : see previous n. We hear nothing about any of these 'many' battles and struggles in Sicily, but we know something about Dionysius' activities in Italy. While a seaborne attack on Thurii proved unsuccessful ( Aelian, VH12. 61), he managed to capture Croton ( Dion. Hal. Ant.20. 7. 3; Livy 24. 3. 8. Croton in Justin 20. 5. 1-3 is really Rhegion) and the pan-Italiot sanctuary of Hera at Lacinium near Croton ( ps.-Arist. Mirab. auscult. 838a; Athen. 12. 541a-b); cf. Stroheker, 1958: 132; Caven, 1990: 189, 193f. The capture of Croton perhaps took place in 379/8 if the figure given by Dion. Hal. Ant. 20. 7. 3 can be trusted. Whether the treaty between Dionysius and Carthage also provided for the future status of south Italy is a question which cannot be answered. τὰ καλούμενα Κάßαλα: the place is otherwise unknown; cf. Holm, Geschichte, i. 86, ii. 142; Ziegler, RE x. 1394. For intelligent guesswork see Manni, 1981: 39, 82. πλείους . . . τω+̑ν μ⊎ρίΩν . . . ἐξώγρησε: the figures are on a par with that given for the Siceliot dead in the following battle (15. 17. 4). None can be checked. They point to Ephorus as the source (pp. 62 ff.). ἐπί τινα λόφον . . . παντελω+̑ς: cf. Polyaen. 6. 16. 1, κατακλεισΘένασ ἐσ χω↔ίον ἄνυδρον. 15. 4. ὁ δέ Διούσιος ἀπεφήνατο κτλ : Polyaen. 6. 16. 1 says the same. Dionysius' demand of the Carthaginians is presumably factual and reported by Ephorus (from Philistus) rather than Timaeus; see p. 82. 16. 1. τῃ+̑ σ↔υνήΘει πανου↔γίᾳ : cf. FGH 70 F7;'; and for Ephorus' likely attitude to the Carthaginians and nonGreeks in general, pp. 114 f. 16. 1-3. π↔οσποιηΘέντες ου+̑ν εὐδοκει+̑ν ὁμολογίαις κτλ : similarly Polyaenus 6. 16. 1: οἱ πρόßουλοι (whatever they might be, though clearly they were with the army and acting now alone, -204Mago having been killed) τω+̑ν Καρχηδονίω ὡμολόγουν ταυ+̑τα ποιήσειν. But from this point on Polyaenus' information diverges from that of D. The latter represents that the Carthaginians asked for a few days' truce to consult their home authorities. This having been granted them, they buried Mago, elected his son to his position, and under his leadership they so revitalized their army that when the truce was over they inflicted a crushing defeat on Dionysius. The account is quite impossible. The Carthaginians could not have done any of these things while still ἐπί τινα λόφον ἐρυμνὸν καὶ ἄνυδ⇁ον παντελω6+̑ς. In any case, D's own narrative shows that when fighting resumed the two combatants were no longer at Cabala. In abbreviating his source D clearly missed at least one vital step. Polyaenus is less unsatisfactory in this respect in that he provides for the removal of the Carthaginian camp to the coast with Dionysius' consent. Less unsatisfactory, but no less problematic for, as Caven, 1990: 197, correctly observes, it is most unlikely that in the circumstances Dionysius would have allowed the Carthaginians to escape. What probably happened was somewhat along the lines indicated by Caven ( 1990: 198): the Carthaginians managed to escape during a truce (see below) and the battle of Cronium which they won came months later and was the outcome of another campaign. D (not some nameless epitomizer as Caven maintains: pp. 79 f.) telescoped the account of his source and produced one campaign out of two. 16. 1. τοι+̑ς ἄ⇁χουσι : these surely are the authorities back in Carthage and not those in the cities in the Carthaginian part of Sicily, as Whittaker in Imperialism, 297 n. 9, is inclined to believe; cf. Stroheker, 1958: 133; Huss, 1985: 139; Caven, 1990: 197. ἀνοχάς: this truce according to Caven (1990: 197) was only for the purpose of burying their dead. But both D and Polyaenus are insistent that terms for a settlement were discussed, that the Carthaginians pretended to go along with Dionysius' demands, and that the latter was deceived as a consequence. Some relaxation of vigilance on the part of Dionysius would better explain the fact that the battered Carthaginian army managed to escape. 16. 2. τὸν υἱὸν αὐτου+̑ : Himilco seems to have been his name: Polyaen. 5. 10. 5; cf. Beloch, GG iii. 1. 124 n. 1. 16. 3. ὡς δὁ τη+̑ς ὁμολγίας διη+̑λΘε χρόνος : for what may have taken place in the interval between the two battles (which must have been considerably longer than a few days' truce: above) -205-

Polyaen. 5. 10. 5 (perhaps also 6. 16. 3) gives us some idea; cf. Caven, 1990: 198f. τὸ καλούμενον KΚρόνιον: a town it would seem (Polyaen. 5. 10. 5), though possibly a mountain (D 3. 61. 3), or both. Whatever it was it is a fair guess that it was in western Sicily, in the vicinity of Panormus; cf. Stroheker, 1958: 232 n. 28; Manni, 1981: 84; Caven, 1990: 198f. 17. 1. Λεπτίνης κτλ : see 15. 7. 3f. and nn. Dionysius and Leptines had not always seen eye to eye and there existed a hostile tradition that the tyrant had either engineered his brother's death (Plut. Mor. 338b = De fort. Alex. 2. 5) or had not tried to prevent it ( Aelian, VH13. 45). D (Ephorus) will have none of it. 17. 4. πλείους . . . τετρακισχιλίων : see 15. 3 n. Πάνορμον: modern Palermo; cf. KP s.v. no. 3; Princeton Encyclopedia, s.v. 17. 5. ἐγένοντο διαλύσεις κτλ : SVt261. The third and last reported treaty between Dionysius and Carthage. The brief and vague way in which D describes these agreements presents problems. In the present agreement the two parties were to have ὡ+̑ν πρότερον ὑπη+̑ρχον κύριοι. The meaning of this seems to be that Dionysius and the Carthaginians would continue to hold what they held as of right (κύριοι) before the war broke out, i.e. according to the terms of the previous settlement of 392 ( Svt233). The Carthaginians would have been especially anxious to stop Dionysius from 'poaching' in their territory once peace was made (cf. 15. 15. 1-2). The one exception (ἐΞαίρετον) to the terms of 392 was that the Carthaginians (they were after all the victors in this war) would now receive Selinus (both the city and its land) and of the territory of Acragas that part which reached as far as the river called the Halycus. The clear implication of this is that in 392 Dionysius did not concede to Carthage either Silenus or any part of Acragantine territory. From which it equally clearly follows not only that the clause in the treaty of 405 about Selinus, Acragas, Himera, Gela, and Camarina remaining unwalled and under the obligation of paying photos to Carthage (13. 114. 1; Svt210) was revoked in 392, but also that these places were added to the territory of Dionysius. The difficulty with this is that nothing of the kind is said in 14. 96. 4 (the treaty of 392). What D in fact says there is that the Sicels and Tauromenium were recognized as -206-

belonging to Dionysius, but that otherwise that treaty resembled the one of 405. It has seemed highly unlikely to most scholars that the stipulation about the Greek cities of western Sicily in the treaty of 405 was retained in that of 392. In all likelihood what happened was that in summarizing his source D, typically enough, missed a statement to the effect that the five cities were handed over to Dionysius. The explicit statement here about Selinus and Acragas can have no other interpretation (cf. Meyer, GdA v. 120f.; Beloch, GG iii. 1. 581 and n. 4; Stroheker, 1958: 82ff.; Finley, 1979: 82f.). Some scholars disagree ( Hans, 1983: 121ff.; Caven, 1990: 200; Lewis, CAH2 vi. 145f.), but their arguments are not compelling, least of all the suggestion (Caven and Lewis) that the western Siceliot cities were no longer of any real importance to be a bone of contention. This is not what one concludes from D's text, from, e.g., 14. 41, 14. 46, and, of course, the terms of the present treaty. Sordi's ingenious suggestion (1980: 23ff.) that the reference to the 'previous peace' at 14. 96. 4 is not to that of 405, but to one concluded in 399/8 which Timaeus (assumed by Sordi to be D's source) concealed, is not supported by any solid evidence, least of all by that of D. For yet another attempt to save the credit of D and of his immediate source, be that Ephorus or Timaeus, and to blame Dionysius himself (his censorship) or his historian ( Philistus), see Zahrnt, 1988: 209ff. Another difficulty concerns the identity of the river Halycus, especially in view of the fact that both D (16. 82. 3) and Plutarch ( Tim. 34) when describing the treaty between Timoleon and Carthage refer to the river which was to mark the frontier between the Carthaginian and the Greek areas as Lycus. The majority view is that Lycus was probably a mistake for Halycus, the latter being the modern river Platani west of Agrigento (Acragas), at the mouth of which was situated Heraclea Minoa (cf. Svt261 with the bibliography cited). Navarra, however, 1964: 70ff., has argued that Lycus and Halycus were different rivers, the first to be identified with the river Platani and the second with the river Salso east of Acragas. This is accepted by Hans, 1982: 211ff.; 1983: 122. Now (and in this at least Hans may be correct), as the fact that only a part of the Acragantine chora was ceded to Carthage demands that the river mentioned in our treaty be located to the west of Acragas, she proposes that Λύκον be substituted for Α+̔λύκον in the text of D ( 1982: 216 ; 1983: 229 n. 36). -207D neglects to give the northern limit of the Carthaginian epikrateia. It is quite possible that Himera was surrendered to Carthage and its chora formed the eastern border of Carthaginian territory in the north ( Stroheker, 1958: 135). ἕτισε . . . τάλαντα χίλια: a heavy indemnity by any account and we may well believe, as Grote pointed out long ago (xi (1853), 59 and n. 2), that Dionysius paid it off by annual instalments, thus justifying Plato's statement that the tyrant became tributary to Carthage. And we might also associate ps.-Arist. Mirab. auscult. 838a and Athen. 12. 541a-b (the story of the precious mantle looted by Dionysius from the temple of Hera at Lacinium near Croton and 'sold' to the Carthaginians) with Dionysius' efforts to pay off this large sum.

18. The end of the revolt in Asia Minor and the contest for Leuce The year 383/2 is about correct for these events (2-4n.) though the contest may be somewhat later. D perhaps found the foundation of Leuce in his chronographer under Phanostratus and consequently placed this particular Ephoran narrative here.

18. 1. Γλω+̑ς ὁ ναυρχήσας . . . ßασιλέως : see 2. 2 and 9. 4nn. Ταχώς: a relative, perhaps a son, of Glos; see Beloch, GG iii. 1. 99 n. 1; Meyer, RE iva. 1996. The name survived in Ionia; see Meyer, ibid. Δεύκην: Leuce or Leucae west of Smyrna by the mouth of the river Hermus; Strabo 14 C 646; Burchner, RE xii. 2209; Treidler, KP 3. 594. Cf. Judeich, 1892: 191; Ramsay, 1890: 206; Jones, 1971: 33. It is possible that Leuce was a refoundation: IG i3 260 (453/2) viii line 8. 18. 2. μετ ὀλίγον δὲχρǵνον αὐτου+̑ τελευτήσαντος : Tachos appears not to have survived Glos by more than a few months. 18. 2-4. ἠμφισßήτησαν κτλ : having related the revolt of Glos and Tachos, Ephorus turned aside to tell of the dispute between Clazomenae and Cyme, his native city, over Leuce. The issue was unimportant and not particularly relevant, though Ephorus could hardly have invented the story. His source doubtless was local Cymaean tradition; see p. 106 ; Samuel, 1968: 375ff. The gist of the story is substantiated by the fact that the coinage of Leuce bears the swan type characteristic of the best Clazomenaean coins -208of this period ( Head, 1911: 568, 581. And see P. Kinns in L'Or perse, 184f.). Interestingly Clazomenae won, not Cyme! 18. 2. ἔΚρινε ἡ ΠυΘία : see Tod, 1913: 95; Fontenrose, 1978: 249 (H15), 16, 38, 212; Parker in Crux, 309f. (discussing the 'whimsical' procedure described here by D); Hornblower, 1991 ad Thuc. 1. 28. 2. 18. 4. ΠροΦΘασείαν : ΠροΦΘαΣίαν M and F. 'Anticipation'. See RE xxiii. 1. 817.

19-23. The Olynthian War

The chronology

No chronology of the Olynthian War can be entirely gentle with Xenophon's account and one must be prepared to perform some surgery. But the chronological indications in the Hellenica, D's evidence, and a morsel or two of information from elsewhere allow us to construct a system which is quite satisfactory as a whole though only approximate in some of its detail. I differ from Beloch, GG iii. 2. 231 ff., in several respects. The Cadmea was captured in the summer of 382 when the festival of the Thesmophoria was being celebrated at Thebes (X. H. 5. 2. 29) and the Pythian Games were being held (Aristid. 19. 258 Dind.). The Pythian festival was held in the local month of Boukatios which corresponded in part to the second Attic month of Metageitnion (Aug./Sept.). The Thesmophoria were celebrated in the autumn in most Greek states, but Delos and Thebes were exceptions. There they were celebrated in the summer (cf. Nilsson, 1906: 316f.). D placed the event under Evander (382/1), probably by using his chronographic source. Shortly before the capture of the Cadmea the Peloponnesian League voted to intervene in the Chalcidice and Eudamidas was dispatched north with the first levy of Peloponnesian troops (X. H. 5. 2. 20 ff.). D placed this under Phanostratus (383/2) (19. 3. The passage is confused; see ad loc.). Thus between Eudamidas' arrival in the Chalcidice and the capture of the Cadmea, both events belonging to summer 382, Evander succeeded Phanostratus as archon at Athens. The Cadmea therefore was captured most probably in the month of August 382 (cf. Beloch, GG iii. 2. 232f.). X. H. 5. 2. 29, Θερου⇌⇌ς δε ὄντος καὶ μεσημßρίας ΠλείΣτη ἦν ἐρημία ἐν ταîς ὁδοîς, confirms the season. -209-

The capture of the Cadmea doubtless overshadowed the Olynthian War for a while. The Spartans had to decide what to do. There followed the trial of Ismenias before a court in which all the allies, great and small, were represented (X. H. 5. 2. 35). We could hardly allow less than a month for these proceedings. Then Teleutias set out with his 10,000. He did not hurry (X. H. 5. 2. 38) and he probably took a month or so to reach his destination. After the first clash with the Olynthians, before which he had refrained from burning the crops and cutting down the trees (X. H. 5. 2. 39), Xenophon comments, καì τοῦτο μέν στρατευσáμενος τò Θερος κτλ (X. H. 5. 2. 43). There is no way therefore that this can be the summer of 382; it must be the summer of 381. The winter of 382/1 intervened between the trial of Ismenias and the departure of Teleutias for the north. At the beginning of the following spring (380: X. H. 5. 3. 1), Derdas achieved his notable victory over the Olynthian cavalry, and some time afterwards (προïóντος δε τοû χρóνου) Teleutias was defeated and killed (X. H. 5. 3. 3-6). Agesipolis was now sent out (X. H. 5. 3. 8 f.). On his arrival he ἔΦΘειρε τòνσîτον (late spring/ early summer 380: X. H. 5. 3. 18), captured Torone, and soon died κατà Θέρυς ἀκμήν ( July/Aug. 380: X. H. 5. 3. 19). Soon after Agesipolis' departure from Sparta, Agesilaus invested Phlius (X. H. 5. 3. 10ff.). The siege lasted twenty months (X. H. 5. 3. 25) and finished, as well as that of Olynthus, in the autumn before the liberation of Thebes in midwinter 379/8. Olynthus was invested by Polybiades, Agesipolis' successor (23. 2f.; X. H. 5. 3. 20, 26). The contemporaneity of the two sieges is shown also by Isocr. 4. 126. The chronographic entry on the death of Agesipolis is under Pytheas in D ( 380/79: 23. 2), which is almost certainly correct (pp. 37 ff.), while his march north and arrival in the Chalcidice are under Demophilus ( 381/0: 22. 2). The events may then be arranged thus:

382 Summer (archon Phanostratus)

Eudamidas to Olynthus.

382 (August) (archon Evander)

Capture of the Cadmea.

381 Spring

Teleutias to Olynthus.

380 Spring

(archon Demophilus) Teleutias killed.

380 Spring

Agesipolis to Olynthus.

380 Spring

Siege of Phlius begins.

380 Summer (archon Pytheas)

Death of Agesipolis.

-210-

19. 1. Λακεδαιμóνιοι . . . ἀπέγνωσαν : see p. 99 and 9. 4n. καì τοû Ταχῳ: Laqueur, 1958: 268f., proposes that D added this here from his additional source, Ephorus (!), from which he derived the story of the contest between Cyme and Clazomenae. The result was a falsification of the facts since 'hatte natürlich dieser Tod des Tachos auf die lakedaimonische Politik keinen Einfluss, wie es infolge des Einschubs erscheinen könnte' ( 1958: 269). I have already argued against the hypothesis that D habitually interwove two or more narrative sources (p. 49 ). Nor is there anything necessarily unhistorical in 19. 1. Tachos succeeded to the affairs of Glos (διαδεΞάμενος τὰς τούτου πράΞεις: 18. 1). If the Spartans were prepared to co-operate with Glos (9. 4n.), why not with Tachos? τὰ δὲ κατὰ τὴν Ἑλλάδα κτλ: see 5. 2-3 and n. there. συνεπιλαßομένου τοû . . . ßασιλέως: this could be seen as additional (though admittedly weak) evidence that the King was an actual participant in the King's Peace, something for which there is good epigraphic evidence (Svt 248 (=Tod 118) line 11), and not outside and above it (as Martin, 1944: 13ff.; Badian, 1987: 27; id., Georgica, 25ff.). See also 38. 2n. 19. 2. κατὰ δὲ τὴν Μακεδονíαν κτλ : Swoboda, 1883: 1ff., and Beloch, GG iii. 2. 57 f., thought that 14. 92. 3-4 (under 393/2) and the present passage represented a doublet. This view was modified by Geyer, 1930: 111ff., who, while insisting that there was but one Illyrian invasion and one land donation, those of 14. 92. 3-4, yet saw 15. 19. 2 not as a doublet, but as a conscious repetition by D of the earlier event as a way of introducing the Spartan intervention in the Chalcidice. Geyer's interpretation was taken up and a little further by Ellis, 1969: 1 ff., and De Salvo, 1972: 114 ff. Hammond, 1979: 172ff. , on the other hand, not only rejects the doublet theory, but also the view that any part of 19. 2 looks back to 14. 92. 3-4 and 393/2. He believes in two different invasions, each accompanied by a 'gift' of land (see n. below) to the Olynthians. I would agree with Hammond though not with absolute confidence. But the evidence, such as it is, seems to argue for two Illyrian invasions of Macedonia during the reign of Amyntas III. -211-

The first invasion occurred shortly after Amyntas' accession to the throne (14. 92. 3-4). He was forced to flee, though not before he presented Olynthus with τὴν σύνεγγυς χώραν. He was returned to Macedonia by the Thessalians after a while (μετ ὀλíγον δὲ χρόνον), though the chronographer gave an alternative view. 'Some say that after the expulsion of Amyntas the Macedonians were ruled by Argaeus for two years, and that it was after that time that Amyntas recovered the kingship.' We may then assume with Hammond ( 1979: 173 ) that on regaining the throne Amyntas received back his territory from the Olynthians. The surviving treaty between the king and the Chalcidians (Svt 231 = Tod 111) may date from this time. For the invading Illyrians and their king, Philip's later opponent Bardylis perhaps, see Hammond, 1979: 172 ; id., 1966: 243, 248 f., and the bibliography cited at 13. 2 n. It may well be that, as Hammond, 1979: 172, 175, says, Argaeus was a puppet king installed by the Illyrians. We know that in 385/4 the exiled Alcetas returned to Epirus through their agency (15. 13. 2-3 and n.). The second invasion took place some time prior to the Spartan involvement in the north. Section 19. 2 is introductory, but there is no need to assume that it looks back to 393/2. There is a possible context for it other than the late 390s. As we have just seen, sometime in 385/4 the Illyrians invaded Molossis in strength and had to be expelled by Sparta. One might then reasonably hypothesize that about this time the Illyrians broke into Macedonia also. As before, Amyntas had to hand over part of his land to Olynthus and flee, though this time he showed a little more spirit and initiative. Isocr. 6. 46 must refer to this occasion, for Sparta could not surely have sent help to Macedonia in the late 390s at the height of the Corinthian War when she had been driven off the seas sistently mistranslated, e.g. by Geyer, 1930: 115; Cloché, and bottled up in the Peloponnese. As Isocrates 1960: 110; Ellis, 1969: 4; Hammond, 1979: 174, sometimes with ludicrous results as attempts are made to locate the said χωρíον and the -212place of origin of the force Amyntas sent for. But ἐνΘένδε means 'from here' and Archidamus is saying that Amyntas sent to us, Spartans, for help, and he got it, and very effective it was too.) The question to ask then is whether the Spartan help is to be identified with any of the forces dispatched against Olynthus in 382 and afterwards. Xenophon says that when Sparta moved against Olynthus Amyntas was in a sorry state (X. H. 5. 2. 12 f., 38), and one might see this as corresponding to Isocr. 6. 46. Alternatively one might infer that a Spartan army had helped Amyntas shortly before this time against the Illyrians only, as Isocrates says, and at the same time perhaps as the Spartan intervention in Epirus, with the result that the king recovered most of his territory, but not of course the area he had given to Olynthus. Section 19. 2, τὴν ὅλńν ἀρχὴν ἀνακτńσαμένου, is an exaggeration, as would be Isocr. 6. 46, κατέσχεν ἅπασαν, if our second alternative were preferred. Section 19. 2 is an introductory outline and the statement ἀναλαßóντος ἑαυτóν is no evidence against the present interpretation. Amyntas did not recover his throne unaided (as Hammond suggests, 1979: 174, who has misunderstood Isocrates), but was helped by Sparta to do so. But once war broke out between Amyntas and Olynthus we can expect that the Olynthians won even more territory than originally ceded, including the Macedonian capital Pella (X. H. 5. 2. 13). It should be stressed that the evidence is of so equivocal a nature that Geyer's view remains possible. On the other hand there is no difficulty in believing in two Illyrian invasions. The Illyrian menace was more or less permanent in the first half of the 4th century, and a combination of D and Isocrates seems to argue for a second invasion about the middle of the 380s. τῳ+̑ δήμῳ τῳν ὈλυνΘíων: the precise relationship between Olynthus and the Chalcidians is an old problem. Was it a unitary state or a confederation of some kind based on Olynthus? And when did it originate? Cf. the discussions by Harrison, 1912: 93ff., 165ff.; West, 1918: 14ff.; Busolt, GS ii. 1501ff.; Hampl, 1935: 177ff.; Kahrstedt, 1936: 416ff.; Robinson, Olynthus, ix. 113ff.; Gomme, HCT i. 203 ff.; Larsen, 1955: 42ff.; id., 1968: 58ff.; Zahrnt, 1971: 49 ff. The problem has been rendered more acute by the contrast between the literary sources, mostly those of the 4th century, on the one hand, and the testimony of the inscriptions and coins, on -213the other. The question cannot be treated fully here, but I find Larsen's view the most sensible and closest to the evidence. He accepts that the Chalcidian state was a confederacy. Hampl's espousal of the unitary state theory (he is followed by Kahrstedt and Robinson) lacks weight, while an inscription discovered in 1934 near Olynthus which records the alliance between Philip II and the Chalcidians (concluded in 357/6: Svt 308 = Tod 158), explicitly refers to τὰς ἀρχὰς τὰς ΞΥνάς of the Confederacy (line 3). And we may compare the reference to the koinon of the Chalcidians in Svt 231 (= Tod 111) line 12, the alliance between Amyntas III and the Confederacy. Xenophon's evidence too rather argues for a confederacy. He contrasts συμπολιτεύειν (sharing of citizenship) with αὐτοπολîται εœναι, and νόμοις τοîς αὐτοîς χρῃσΘαι With τοîς πατρíοις νóμοις χρῃσΘαι (H. 5. 2. 12, 14). Zahrnt accommodates both views. At its inception in 432 the state was probably a unitary one, an 'Einheitsstaat' (1971: 49ff.). At the beginning of the 4th century, however, Olynthus took the initiative and the Chalcidian Confederacy was born (1971: 80 ff.). On the form of government of the Confederacy, cf. Larsen, 1968: 75ff. Can D contribute to the discussion? One's first impression is that D (Ephorus) uses 'Olynthians' and 'Chalcidians' indiscriminately, but on further reflection one sees that Ephorus' terminology follows that of his sources. In Book 12 D consistently refers to the 'Chalcidians' rather than to the 'Olynthians' and this is in fact Thucydides' own usage as a comparison of the relevant passages shows: 12. 42. 5 = Thuc. 2. 9. 4; 12. 46. 7 = Thuc. 2. 58. 1; 12. 50. 3, 7; 51. 2 = Thuc. 2. 95-101; 12. 67. 2 = Thuc. 4. 78 ff. There is one exception which, if anything, proves the rule. At 12. 47. 3 D speaks of the Olynthians helping the Bottiaeans against Athens while Thucydides says, correctly, that the troops from Olynthus were Chalcidian. The difference hardly requires explanation. In rewriting Thucydides, Ephorus opted for the popular, 4th century terminology here. Thucydides himself twice uses 'Olynthians' at 5. 3. 4 and 5. 39. 1 (as well as at 5. 18. 5. For the possible reasons cf. Gomme, HCT i. 204 and nn.

1 and 4); cf. D 12. 77. 5. From 14. 92. 3 ( 393/2) D invariably refers to 'Olynthians' and this is in line with almost all 4th-century writers. Theopompus, exceptionally, frequently uses 'Chalcidians'. No special significance should be attached to this usage. The situation is similar to -214-

that of Thebes and Boeotia. Xenophon regularly writes 'Thebans' when 'Boeotians' would have been more correct. D uses both terms (28. 1 n.; cf. further Kahrstedt, 1936: 424 n. 42). The reason for this is that although technically there were 'Boeotians' and not 'Thebans' when the Boeotian Confederacy was in existence, everyone knew that Thebes controlled the Confederacy. The 4thcentury nomenclature reflects reality not constitutional niceties. The break in terminology in D occurs at 14. 82. 3 ( 395/4) where for the last time the Xαλκιδεîς οἱ πρòς τρ̑ Θρέ+̑κῃ are mentioned in connection with the anti-Spartan alliance. Not long after that, at 14. 92. 3 ( 393/2), the 'Olynthians' replace the 'Chalcidians' for good (the exception is 16. 52. 9 where the reference is probably only geographical). Again, it would be wrong to attach any weight to the change in terminology at this point and seek a corresponding change in the political circumstances of the Chalcidians (as Zahrnt, 1971: 82ff, is inclined to do). The reference to the 'Chalcidians' at 14. 82. 3 is likely to derive from the Oxyrhynchus historian who perhaps ended his work with the year 394 and was one of Ephorus' sources (p. 107 ). Like Thucydides the Oxyrhynchus historian was accurate and painstaking. For the rest of the Histories Ephorus presumably employed the popular term 'Olynthians'. δωρńσαμένου: it is not necessary to view this as an indication of Ephorus' attitude to Macedonia. It should rather be seen perhaps within the context of his criticism of Spartan behaviour after the King's Peace. The Macedonian king had presented the land to the Chalcidians of his own volition and the Spartans had no business supporting him against the Confederacy. See following nn. ὁμóρου χώρας probably the same as the σύνεγγυς χώρα at 14. 92. 3 and to be identified perhaps with the area round Lete and Lake Bolbe given to the Chalcidians by Perdiccas in 432 for the duration of the war against Athens (Thuc. 1. 58. 2; Hammond, 1979: 172 ). Amyntas' 'gift' presumably was to be returned after the Illyrians had been expelled. 19. 3. Λακεδαιμονíους ποιńσάμενος συμμάχους: thus according to Ephorus the Spartans intervened in the Chalcidice to help Amyntas against Greek Olynthus; cf., similarly, Isocr. 4. 126. Xenophon paints a different picture: the Spartans marched north not to bring succour to any Macedonian, but in defence of the autonomy of the Greek cities of Acanthus and Apollonia, and in -215-

their own defence of course, before the Chalcidians grew too powerful and posed a threat to Spartan supremacy in northern Greece (X. H. 5. 2. 12 ff. This is seen by Tuplin, 1993: 94 ff., but understated somewhat in line with the main thrust of his argument: see p. 114 n. 306). The two accounts should be combined. All three, Amyntas, Acanthus, and Apollonia, appealed to Sparta and Sparta responded because she felt threatened by the Confederacy, especially if the Chalcidians were to ally themselves with Thebes and Athens (X. H. 5. 2. 15). But while the Spartans and their friends claimed they were defending the freedom and autonomy of the Greek cities (Xenophon), their critics said they were serving their own interests and those of the Macedonian king against Greeks (Ephorus, Isocrates). Amyntas may not have been a party to the King's Peace, though Aristid. Panath. 282 Dind. hardly proves that, pace Cawkwell, 1973b: 53 n. 3; id., 1976a: 77 n. 56. What is given there by Aristides is the line-up of non-Greek powers on the side of Sparta in the struggle to cut the Athenian corn supply at the Hellespont in 387. Though a Greek, Dionysius was essentially an outsider to mainland and Aegean politics and his help to Sparta (X. H. 5. 1. 28) was in return for services rendered by Sparta in the past (cf. 14. 10. 2f.; 44; 63. 4; 70. 2f.; Plut. Lys. 2. 7 f.). He cannot have been a party to the Peace, and neither can Seuthes. The latter had, it seems, been aiding Antalcidas and Tiribazus, while Hebryzelmis of the Odrysian Thracians, a rival of Seuthes, had been supporting Athens, to judge from the honorary decree praising Hebryzelmis passed under Mystichides ( 386/5: Tod 117). The Chalcidians had sided with Athens6 and Thebes against Sparta in 395 (14. 82. 3; Isaeus 5. 46) and their Confederacy had, in theory at least, been dissolved by the King's Peace. δúναμιν ἀΞιóλογον: a Diodoran expression, see p. 133. πολιτ̑ν: no Spartiates as such were sent out, but the Lacedaemonian contingent would have consisted of perioikoi, neodamodeis, and, perhaps, Sciritans; cf. X. H. 5. 2. 24. D uses πολîται and πολιτικοí denote troops from Lacedaemonia in general, no matter what their status; cf. 14. 36. 1; 15. 62. 1 (probably wrongly, see n. there); 63. 1. Xenophon does the same; cf. H. 4. 4. 19; 5. 3. 25; 6. 4. 26. ὑπὲρ μυρíους: ὑπέρ is probably a Diodoran rhetorical touch (p. 133). The decision was that each ally should send its contingent -216for a total force of 10,000, τò εἰς τοὺσ μυρíους σύνταγμα(X. H. 5. 2. 20), evidently an established division (onethird?) of the total field force of the League; cf. 11. 79. 5=Thuc. 1. 107. 2: 1,500 Lacedaemonians and 10,000 allies; X. H. 3. 4. 2=Ag. 1. 7: εἰς ἑΞακισχιλíους δὲ τò σύνταγμα τ̑ν συμμάχων (one-fifth?). Xenophon also says that any state which so wished could send money instead of men (H. 5. 2. 21). This appears to have been something new in the Peloponnesian League. But see 31. 1-2 n. Φοι íδᾳ: D is almost certainly confused here. The Spartans did not give Phoebidas the 10,000 troops they had voted for, and Phoebidas did not use these to capture the Cadmea before 'he went off on his own business' (20.

2). Nor did they then replace him with his brother Eudamidas giving the latter 3,000 hoplites and (sending him out to fight against Olynthus' (20. 3). The result of this confusion is that Teleutias is left with no troops apart from the vague δύναμιν ἀΞιóλογον (21. 1). The correct order of events is given by Xenophon (H. 5. 2. 20 ff.). The congress of the allies voted to dispatch τò εἰς τοὺς μνρíους σύνταγμα. As this required time to be collected (it marched out eventually under Teleutias), Eudamidas was sent north at once with 2,000 troops from Lacedaemonia. But even his force fell short of what had been decided and on departing Eudamidas requested of the ephors that the remainder of his troops be brought north by his brother Phoebidas. It was this body of troops which captured the Cadmea. But though D is confused, he supplies one important piece of information. He says that Eudamidas was given 3,000 troops. This must be the force voted for as a temporary measure before Teleutias could arrive with the 10,000. Xenophon says that Eudamidas departed with 2,000, but with more to follow under his brother. One may then deduce that Phoebidas occupied the Cadmea with 1,000 Peloponnesian troops. Cawkwell, 1976a: 77 n. 58, finds the splitting in two of the advance force surprising and suspects that perhaps Xenophon has got his details wrong. But the surprise diminishes when we consider that even 3,000 Laconian and other Peloponnesian troops would have required some time to muster. For Cawkwell wrongly assumes that it was the 2,000 Laconian troops that were divided in two. The Acanthian envoys had requested a temporary force ἐκ Λακεδαíμονóς τε, ὅσń ἂν ταχὺ ἐΞέλΘοι, καì ἐκ τ̑ν ἄλλων πóλεων ( H. 5. 2. 23). The Spartans and the allies consented and Eudamidas -217-

was sent off at once with 2,000 neodamodeis, perioikoi, and Sciritans ( H. 5. 2. 24). Phoebidas then must have led out primarily the contingents τ̑ν ἄλλων πóλεων, the allies, once these were ready. It is true that πλεις may be ambiguous in 5th- and 4th-century writers. It mostly refers to the allies, but it is sometimes used of the perioikic cities (cf. HCT iv. 73 f.; de Ste Croix, 1972: 345 f.). There can be little doubt, however, that the allies are meant here. The proceedings are in the congress of the allies, and the πλεις mentioned in the previous three sections ( H. 5. 2. 20-2) are obviously those of the allies. The surprise at the splitting in two of the advance force diminishes even further if we accept D's (Ephorus') perfectly credible claim that the Cadmea was captured as a result of a Spartan conspiracy. It may well be that Agesilaus and his group had seized this opportunity to kill two birds with one stone. Eudamidas of course had to proceed north without any delay. A Spartan presence in the Chalcidice was essential and urgent ( X. H. 5. 2. 23). But it is possible that Phoebidas was never meant to go beyond Thebes; that his declared destination, Olynthus, was simply meant to cover his real objective, Thebes (see further 20. 2 n.). It is interesting to observe that the troops which occupied the Cadmea were apparently non-Lacedaemonian. There is nothing unlikely in this. Sphodrias did after all march against the Piraeus with allied and mercenary troops ( X. H. 5. 4. 15 and 29. 6n.). συμμαχεîν τῳ+̑ ̔Αμúντᾳ : D is consistent in that the alliance against the Chalcidians was between Sparta and Amyntas; see 19. 3 and n.; 20. 3 and 21. 1. Though Xenophon, as we have seen (above), prefers to stress the appeal to Sparta of Acanthus and Apollonia, it emerges from his narrative that war was raging at this time between Amyntas and the Chalcidians and that the latter were winning ( H. 5. 2. 13). P. Oxy. i. 13, which purports to be a letter to a Macedonian king, might, if historical, constitute further evidence that the alliance sought between Olynthus and Thebes (X. H. 5. 2. 15, 34) was aimed against Amyntas in the first instance though doubtless its character would have been anti-Spartan; the Chalcidians might be seeking additional muscle against Macedonia, but the Thebans only cared for support against Sparta. -218-

19. 3. The attack on Phlius ἑτεραν δὲ δύναμιν . . . Λακεδαιμονίοις : and this is all D has to say about the attack on Phlius which in any case took place two years later (19-23 n.). For the events at Phlius see X. H. 5. 2. 8-10;5. 3. 10-17;5. 3. 21-5 with Tuplin, 1993: 90ff. Legon, 1967: 324 ff., argues that Phlius was democratic till its suppression by Agesilaus when an oligarchy was established, and this is the general view. Thompson, on the other hand, 1970: 224 ff., sounds a note of caution: Phlius may not have been democratic, and Agesilaus did not necessarily impose an oligarchy at the city's surrender. However, that constitutional questions were indeed the issue at Phlius, and that an oligarchic regime replaced a democracy, thus justifying D's claim that Phlius was forced to submit to Sparta, see Cartledge, 1987: 262 ff., 372. μα+́χέ νικήσαντες : is this just a Diodoran cliché? Xenophon certainly does not mention a battle; but then neither does he mention the fighting before Mantinea in 385 (5. 5 n.).

19. 4. Agesilaus and Agesipolis κατ° δὲ του+τον τὸν χρόνον κτλ : Meyer, GdA v. 294, considers this an Ephoran exaggeration and indeed one's first reaction is to distrust such antitheses, a penchant it seems of Ephorus (pp. 118 ff.). There was certainly a tradition which viewed Agesilaus as the villain responsible for all Spartan misdeeds, while his contemporary Agiad kings were invested with just and pacific natures; cf. Polyb. 9. 23. 7 (he contrasts Cleombrotus, the brother and successor of Agesipolis, with Agesilaus); Plut. Ages.35. 5; (Maximus of Tyre 35. 8c is obscure and probably corrupt; but see Tuplin, 1979: 356). And the exiled Pausanias may have been making the same point when he had it proclaimed of his son on a memorial he raised to him at Delphi that 'Eλλ°ς δ' ἀρετ°ν' ὁμοϕωνει (Tod 120). Ephorus appears to have been influenced by this tradition. It would be as erroneous to reject the tradition out of hand as to credit it fully. The facts indicate that each incident should be treated separately; and that while there

was probably enmity between Cleombrotus and Agesilaus, the latter and Agesipolis seem to have liked each other, their disagreement (not to be exaggerated) stemming only out of differences over foreign -219policy (X. H. 5. 3. 20; Plut. Ages. 20. 7-9. And cf. Xenophon's telling comment at H. 4. 7. 5 that the two kings competed with each other in prowess much as athletes do in the pentathlon). There was no disagreement as far as we can see between the two kings in the cases of Mantinea and Olynthus (12. 1n.;22. 2 n.). The capture of the Cadmea, we are told (X. H. 5. 2. 32; Plut. Ages. 23. 6), provoked a great deal of anger at Sparta, but only because Phoebidas had acted without orders. But there was general acquiescence once it was realized that the action was in the Spartan interest. Agesipolis may have been critical to begin with, and he may in fact have pressed for Phoebidas' trial and fine (cf. Plut. Ages. 23. 6). But there is no reason to suppose that he opposed the continued occupation of the Cadmea which was so obviously beneficial to Spartan interests in central Greece and to the successful prosecution of the war in the north. Phlius was different and it is in fact the only certain instance of a difference of opinion between the kings. Unlike Mantinea, Phlius had not been disloyal in the Corinthian War and Agesipolis, championing the existing government (probably a democracy; see above), may have pressed that reasonably loyal allies, whatever their constitution, ought not to be interfered with. ἐκδὄτους . . . τοὺς κατ° τὴν Γσίαί Εαν λλðα : see 9. 5 n. ἐν ται+̂ς κοιναι+̂ς συνΘήκαις . . . αὐτονόμους : the text has its problems here, but whether we follow Vogel or Vial (who misprints ὀμόσασαν twice and gives an inadequate app. crit.) the difference in meaning is nil. δραστεκός : 'active', 'efficacious'. Also said of Agesilaus at 31. 4 and 33. 1, and of others elsewhere (11. 81. 5;11. 88. 6;17. 16. 2; 19. 4. 3; 19. 71. 4). A vivid though unusual adjective, more properly a medical term; cf. Hornblower, 1981: 279 n. 21. Agesilaus is δραστεκός and Φιλοπόλεμος in contrast to the εἰρνικός and δίκαιος Agesipolis. Cartledge, 1987: 270, 418, not inappropriately translates 'interventionist'. For D's stereotyped language of personal description see pp. 15 ff. δΥναστείας : see 5. 1 n. -220-

Archon Evander, 382/I 20. 1. Εὐα+́νδρου : μὲν Εὐα+́νδρου (MSS PX ) rather than Mενα+́νδρον (MSS MF ): Demosth. 24. 138. About the man see Lysias 26; Sealey, 1993:17f.; Develin, 220. Πωμαι+̂οι κτλ : repetition, with some variations, of the college (390 V) cited at 14. 110. 1 (387/6), the college of the year of the Sack of Rome; see pp. 27 f. Though he correctly states that six tribunes were elected he names only four. For the complete college see Livy 5.36. 11; 5.35.5; CAH2 vii. 2. 636. And for an explanation for the omissions in D, Drummond, 1980: 58f. MS F, as often, adds names not to be found in the other MSS or in other fasti; Vial ad loc. and p. 129 ; Perl, 1957: 79, 103.

20. 1-3. The capture of the Cadmea 20. 1. ἑπὶ δὲ τούτων . . .αἰἱτίας : the date is correct (19-23 n.). This is perhaps the chronographic entry (pp. 44 f.). ὁρω+̑ντ≡ς τὴν Βοιωτίαν κτλ : Cf. 11. 82. 3;15. 26. 1;15. 39. 1;15. 50. 5;17. 11. 4; and 15. 5. 3 (and n. there). Another example of D's stereotyped and repetitious way of writing; which is not to deny that Ephorus had expressed a similar opinion. On the contrary, Boeotian prowess was very much an Ephoran theme ( Andrewes, Essays Starr, 191; see also p. 324 ). ὤσπερ ἀκρόπολιν . . . οὕσαν : a common enough metaphor. At 14. 82. 4 he says the same of Sparta vis-à-vis the Peloponnese; cf. ML 95(c) (Lacedaemon, the acropolis of Greece); Athen. 13. 573e ( Corinth, the acropolis of Greece); Eurip. Orest. 1094 (Delphi, the acropolis of Phocis). 20. 2. διότερ. . . τὴν καδμείαν : the charge that the Spartans conspired to seize the Cadmea is perfectly plausible (5. 2-3 n.). Since the Kina's Peace and the restoration of the Theban exiles when Thebes had for a while performed her duties as a loyal member of the Peloponnesian League (5. 2-3 n.), the pro-Spartan faction had been losing ground and now (summer 382) Thebes was poised once more, as in 395, to challenge Sparta. Certainly there were rumours of an impending alliance between Thebes and the Chalcidians (X. H. 5. 2. 15). No wonder the Spartans were worried. Agesilaus was suspected at the time of having been behind the plot (Plut. Ages. 23. 6-24. 2) and indeed it may be -221doubted whether Phoebidas would have dared to act thus against Thebes without the support of some powerful group at Sparta; and we know that he was Agesilaus' man (X. H. 5. 4. 41; Cawkwell, 1976a: 77f. and n. 59; Cartledge, 1987: 143f., 156, 373). Then there is, as we have seen, the splitting in two of the advance force against Olynthus and the possibility that Phoebidas' contingent was aimed at Thebes from the start (19. 3 n.). In addition, there is the fact that Phoebidas found himself outside Thebes on the day of the Thesmophoria when the Cadmea was in the charge of the women. A coincidence? Perhaps, but under the circumstances one may reasonably suspect not. Perhaps the Theban delegates who attended the congress of allies at Sparta (X. H. 5. 2.

11) were Leontiades and his friends, the anti-Spartan faction of Ismenias having boycotted the proceedings. The occasion would have been ideal for the hatching of a plot. And the timing could not have been better. It might be thought that Phoebidas' very presence at Thebes constitutes yet another suspicious circumstance. Thebes is not on the direct route north. But Ismenlas and his supporters did not regard his presence with suspicion and were caught off their guard. Phoebidas therefore must have had a valid reason for being at Thebes. I would see that reason as the collection of the Theban contingent against Olynthus-in vain, for a public proclamation forbade any Theban to follow him (X. H. 5. 2. 27). Not that that mattered; Phoebidas was after the Cadmea, not Theban troops. (Cawkwell's suggestion, 1981a: 78f., that the Theban contribution against Olynthus was on the basis of the sanctions clause in the King's Peace is not impossible.) ἐΠί τινος ἡγεμονόας : this is a little awkward in D as he has only just explained what Phoebidas' command was (19. 3). In Ephorus' lengthier accounts on the other hand it may not have been so. συνα+́ψας μα+́χψν : a rhetorical exaggeration or did the Thebans resist? Neither X. H. 5. 2. 30 ff. nor Plut. Pel. 5; Ages. 23, speak of any fighting. Cf. in general Grote, x. 78ff. τριακοσίους : X. H. 5. 2. 31 agrees and adds that they fled to Athens. The Aristides scholiast, quoting Androtion, speaks of 400 Theban exiles at Athens (iii. 278 Dind. = FGH 324 F 50 and see Jacoby's n. ad loc.); cf. Isocr. 14. 28; Nepos, Pel. 2. 1. Many of these are listed in IG ii2 37 (+addenda) and IG ii2 245, as having been awarded ateleia and isoteleia. See Walbank, 1982: 268 n. 19 and cf. Plut. Pel. 7. 1 and Aristides, Panath.283 Dind. -222ϕοιßίδαν ἐζὴμίωσαν . . . Θὴßῳν : cf. Polyb, 4. 27. 4; Plut. Mor. 576a; Nepos, Pel. 1. 3. Plut. Pel. 6. 1 says that Phoebidas was fined the very high sum of 100,000 drachmas, precisely the same fine that Agis was threatened with in 418 (Thuc. 5. 63. 2), and Pleistoanax actually fined in 446 according to Ephorus ( FGH 70 F 193. Ephorus in fact said 15 talents which may be a rounding of 162/3 Aeginetan talents, that is, 100,000 Aeginetan drachmas: Lewis, 1977: 39 n. 82). At 27. 3 D says that one of the three men who surrendered the Cadmea was fined so much money that he could not pay it, and it may well be that such large fines were often meant to ruin a man (or group) politically or force him into exile. The trouble is we know very little about this aspect of Spartan life. The famous restriction on owning coined silver or gold which was reaffirmed in 404 (Plut. Lys. 17) appears to have been circumvented by many Spartans, both before this date (cf. Michell, 1952: 298ff.; MacDowell, 1986a: 119), and after it. Andrewes' suggestion ( HCT iv. 90) that the kings, at any rate, were exempt from it may be correct; and of course the state itself had a treasury, certainly after 404. Plato, Alc. 1. 122e-123b, speaks of the mass of gold and silver in Sparta and the great wealth of the kings. And Posidonius had said ( FGH 87 F48c) that Spartans, forbidden by law from keeping either gold or silver, were in the habit of depositing them with the neighbouring Arcadians. The statement finds confirmation in a 5th-century inscription from Tegea (IG v. ii. 159=Buck, 1955: 267 no. 70). The document is in fact a deposit 'account' by a certain Xuthias, son of Philachaeus, almost certainly a well-connected Spartan. The place of deposit was the temple of Athena Alea. Xuthias' total deposit amounted to 400 minae of silver, just under 7 talents, a considerable sum. Phoebidas' fine, therefore, may well have been paid, perhaps with the help of Agesilaus (cf. Cartledge, 1987: 156). At any rate four years later he is still prominent at Sparta and still within Agesilaus' circle (X. H. 5. 4. 41). Dicaearchus' statement (ap. Athen. 4. 141c; cf. Plut. Lyc. 12. 3) that each Spartan contributed to his mess lo Aeginetan obols is difficult to reconcile with the provision against owning coined silver; unless the latter had become a dead letter by the 4th century; cf. X. Lac. Pol. 14. 32. See the comments of David, 1981: 5ff.; Cawkwell, 1983: 396; Cartledge, 1987: 88f. Xenophon says nothing about a trial or a fine though Agesilaus' -223-

cynically utilitarian comment (H. 5. 2. 32) would imply that, at the time of Leontiades' arrival at Sparta, Phoebidas was being tried or at least that the trial was imminent. The verbal proximity between D and Polyb. 4. 27. 4 suggests a common source; either Ephorus or, more likely, Callisthenes who was used by both Ephorus and Polybius. 20. 3. τὸν μὲν ϕοιßίδαν κτλ : as was seen (19. 3 n.) D has confused the order of events and the two brothers.

20. 3-21. The Olynthian War continued 21. 1. κοινῃ+̑ μετ Αμύντου : D will not deviate from the view that Sparta was helping the Macedonian king against Olynthus (19. 3 and n.; 20. 3). δύναμιν ἀΞιόλογον : for this Diodoran cliché, used twice in this section, see p. 133. For the military strength of Olynthus see X. H. 5. 2. 14 where the figure 800 (hoplites) should be emended to, perhaps, 4,000 or even 8,000 ( Cartledge, 1987: 269). Demosth. 19. 263 is confused and inaccurate rhetoric. ΘαυμαξU=1F79μυνος δ ἐπ ἀρετ Αρετῃ+̑' : he was certainly admired by Xenophon; cf. H. 4. 4. 19;4. 8. 23 f.;5. 1. 2. ff.;5. 2. 37 ff. The quick temper which caused his death is nevertheless criticized: 5. 3. 7. 21. 2. ἀναξεὸΞεύ≻σ οὓν κτλ : this is but the briefest outline of Teleutias' warfare in the Chalcidice, for the details of which see X. H. 5. 2. 37 ff. But sketchy though it is, the usual Diodoran insipid embellishments of battle scenes still occur.

Εὐδαμίδου : the reading of the 'edited' MS F. MSS PMX, on the other hand, read Φοιβίδον. It is likely that D wrote Φοιβίδον and F corrected. παραταξαμένων κτλ : the facts still emerge from D's severe abridgement of his source. There was first an indecisive battle (ταραταξαμένων . . . διεχωρίσθὴσαν) which is to be identified with that fought before Olynthus in early summer 381 (X. H. 5. 2. 39-43). Some time later (μετὰ δὲ ταυ+̑τα' ) there followed another battle and this is the second battle before Olynthus (spring 380) which cost Teleutias his life (X. H. 5. 3. 3-6). Πλείους . . . διακοσίων: Xenophon gives only the number of peltasts under Tlemonidas killed, 'more than a hundred', but he does say that in the final rout a great many fell (X. H. 5. 3. 4, 6). Still, the figure of 1, 200 may represent Teleutias' total losses. -224-

21. 3. It is clear from X. H. 5. 3. 8f. too that the Spartans now decided to make a major effort and win the war.

Archon Demophilus, 381/0 22. 1. Ρωμαι+̑οι κτλ : CAH2 vii. 2. 636 (389 V). Unusually D does not here give a figure for the total number of tribunes in the college. If the καὶ between Δεύκιον and Ποστούμιον is deleted (following Rhodoman), we have seven names in all (Vial, 28), which makes this college unique in D. It is possible nevertheless that the number of names originally given in the fasti employed by D was eight, Δεύκιον καὶ Ποστούμιον being a contraction of Δεύκιον Αἰ+ἱμίλιον καὶ Δεύκιον Ποστούμιον, both of which are given by Livy (6. 1. 8); see Drummond, 1980: 65 and n. 38. Livy himself gives the usual number of six and the difference between him and D is that the latter includes two more magistrates, L. Papirius and M. Furius. In this and other instances where D names more tribunes than Livy or gives different names, Drummond's opinion is that Livy's colleges are more trustworthy, D's discrepant information being the result not so much of interpolation as of confusion and contamination with other years ( 1980: 65ff.). It is noteworthy, as Drummond points out ( 1980: 71), that D's two additional names reappear in the college of 384 V. MS F adds Kόιντον after Ποστούμιον; but cf. Drummond, 1980: 65 n. 39.

22. 2. The Olynthian War continued Αλὴσίπολιν : as Agesilaus in 396, Agesipolis was given a staff of thirty Spartiates. In addition to the Lacedaemonian and allied troops ordered out, he was followed by large numbers of volunteers: perioikoi, trophimoi, nothoi, as well as contingents from the allies and cavalry from Thessaly, γνωσθὴ+̑ναι τᾠ Αγὴσιπόλιδι ßονλόμενοι (X. H. 5. 3. 8f.). There is not the slightest evidence to support Smith's assertion ( 1954: 279f. Also, though with greater circumspection, Cartledge, 1987: 226, 271) that the command was thrust by Agesilaus upon an unwilling Agesipolis who was then sent north 'suitably hedged about' by the thirty Spartiates. On the contrary, it is evident that the young king intended to imitate, and if possible surpass, his older colleague. Witness his invasion of the Argolid in 388 (X. H. 4. 7. 5). Xenophon's description of -225-

Agesipolis' Olynthian campaign is very brief, but the energy with which it was pursued is plain enough. There is nothing sinister about the thirty advisers. It may be noted that Agesilaus was accompanied for a second time by such a body when he went to Egypt in 361 (Plut. Ages. 36. 6). Nor is the composition of the volunteer force evidence that the young king 'enjoyed a great popularity among the depressed classes at Sparta and among the allies' because he stood up to Agesilaus who was responsible for the oppressive Spartan policies ( Smith, 1954: 279f.). Why should Agesipolis be popular with the depressed classes at Sparta? What these people would have wished for, political equality with the homoioi, Agesipolis could not, would not, give them. One suspects (and this is probably how Xenophon saw it too) that the enthusiasm of these people was due to the joys and benefits to be derived from a great campaign in the north under one of their kings. The motivation of the nonLaconian volunteers can be explained quite satisfactorily as due to personal considerations. They wanted to become known to the king, as Xenophon says. Many no doubt hoped to be rewarded for their services (as Xenophon had been). Cawkwell explains the volunteers otherwise ( 1981 ae: 78 f.). δύναΦιν ἱκανήν : see p. 133. Xenophon gives no figures either. ἐΨὴΦίσαντο . . . Πόλεμον : a Diodorism, see p. 16. Unger's emendation Πόλεμον (printed by Vogel though not by Vial) has no MSS support and is unnecessary. παρεμßολῃ+̑ : the surviving Peloponnesian troops were in fact quartered in the various allied cities (X. H. 5. 3. 6). The term, meaning fortified camp, is Hellenistic. Ephorus could not have used it. διεπολέμει κτλ : Cf. X. H. 5. 3. 18 f.

Archon Pytheas, 380/79 23. 1. ἐν Πὡμέ κτλ : in agreement with Livy 6. 4. 7; CAH2 vii. 2. 636 (388 V).

23. 2. ἐπὶ δὲ τούτων . . . ἐννέα : a chronographic entry. For the Spartan kinglists, see pp. 37 ff. -226-

23. 2-3. The Olynthian Way concluded 23. 2. Πολυßιάδον : he may well have been the son of the ephor Naucleidas (whose own father was called Polybiadas) who in 404/3 accompanied Pausanias to Athens and supported him against Lysander (X. H. 2. 4. 36; Athen. 12. 55 od; Aelian, VH 14. 7). But we should not read too much into this. Polybiadas' appointment was not necessarily political, and it is no proof that Agesipolis' circle was dominant in Sparta at this time. Nor is it safe to speculate (as Smith, 1954: 279 f., and Rice, 1974: 177 f., do; cf. also Cartledge, 1987: 270 that Agesilaus had either contested the Olynthian command or had it thrust on Agesipolis (see n. above). 23. 3. οὓτος δὲ παραλαχίαν : X. H. 5. 3. 26 is as brief as D. Polybiadas confined the Olynthians to their city and besieged them until they asked for terms. Ελλραϕίντων . . . συμμαχίαν : cf. X. H. 5. 3. 26: τὸν αὐτὸν μὲν ἐχθρὸν καὶ ϕίλον Δακεδαιμονίοις νομίζειν ἀκολονθει+̑ν δὶ ὃν ἡλω+̑νται καὶ σύμμαχοι ει+̑ναι. The standard oath sworn by members of the Peloponnesian League ( de Ste Croix, 1972: 108 f. with SEG xxvi. 461 = ML 67 addenda. For further discussion and partial modification of de Ste Croix's view see Pistorius, 1985: 78 ff.). The Spartans do not appear to have insisted on the complete dissolution of the Chalcidian Confederacy. Zahrnt, 1971: 91 ff., believes, perhaps rightly, that even after 379 there remained a Chalcidian state, shrunk to little more than Olynthus, the original 'Einheitsstaat' which had created the Confederacy in the 390S (19. 2 n.). Certainly, the continuing Chalcidian coinage is an argument that something remained, whether a nucleus Chalcidian state or a reduced Confederacy. For the same argues also the designation [χαλκι]δπὸ [Θρὰικὴς] on the stele of the decree of Aristoteles ( Svt257 (=Tod 123) B lines 5 f.) which marked their entry into the Athenian Confederacy in the mid-370s. In the same context too probably belongs Svt 250 (=Tod I 19), an alliance between Athens and the Chalcidians of Thrace ( West, 1918: 108 and n. 2, 125 n. 13; Accame, 1941: 87 ff.; Lewis, 1954: 33), which should therefore be the treaty which accompanied the entry of the Chalcidians into the Athenian Confederacy. The cities detached from the Chalcidian Confederacy will have entered the Spartan alliance individually; see next n. The surrender of Olynthus was followed it would seem by -227intense coining activity as the state tried to make good the deprivations inflicted on it by four years of war. Now for the first time the names of magistrates appear on the coins, with 'Asclepiodorus' leading the series. Robinson's explanation that this was a measure aimed at certifying the solvency of the state and continuing worth of the Chalcidian coinage is almost certainly correct ( Olynthus, ix. 156ff. Robinson's series requires some slight modifications; see Clement, 1970) πολλαὶ καὶ των ἄΛΛων . . . καταλεχθὴ+̑ναι : former members of the Confederacy and other cities in the area. See 31. 2: the tenth division of the reorganized Peloponnesian League consisted of the Ολύνθιοι καὶ τω+̑.

23. 3-5. Spartan power at its height 23. 3-4. διὸ καὶ κ≻τὰ του+̑τοΥς τοὺς καιροὺς κτλ : see X. H. 5. 3. 27 for very similar comments on Spartan power at this time, made at this very point of the surrender of Olynthus. 'The reason for this convergence of judgement is . . . that both historians (Ephorus and Xenophon) have in mind the forthcoming peripeteia of the Theban counter-revolution' ( Tuplin, 1993: 87). No doubt, but this would also appear to be an instance where Ephorus used Xenophon. The states mentioned are given in the same sequence ( Thebes, Corinth, Argos, Athens) and the same opinions are expressed about them. 23. 3. καὶ κατὰ θάλατταν : the Spartan hegemony lasted from 404 till 371 and in that time, it was commonly said, the Spartans had ruled both land and sea; cf. Demosth. 19. 264;Isocr. 5. 47;7. 7; Plut. Pel. 2. 10;6. 2; 13. 7. 23. 4. κλὴρουχίας : presumably an Ephoran comment on the Athenian habit of sending out settlers to the territory of their subjects, in particular of those that had been suppressed after a revolt (cf. Thuc. 1. 114. 3;2. 70. 4;3. 34. 4;50. 2). This was in the 5th century, but it had not been forgotten abroad especially as people in Athens still dreamed of regaining their former overseas possessions (cf. Andoc. 3. 15, 36), so much so that assurances had to be given in 377 (29. 8 and n.; Isocr. 14. 44; Svt 257 (=Tod 123) lines 25 ff.). ΔακεδαιμU=1F79νιοι . . . ἐπεποίὴντο : this is rather intriguing. Perhaps what Ephorus had meant was that the Spartans with their alliance, which was at its most πολνάνθρωπος at this moment, far out-228-

numbered any other state or group of states in Greece. And of course the Spartans were professional soldiers in constant training. It is possible, however, that there is a reference here to the reorganization of the Peloponnesian League which was completed by summer 378; see 31. 1-2n.

23. 5. Διóπεμ ρο ἐγιστρι ρU03C439Aλ : the close connection between Sparta and Dionysius is too well known to require extensive documentation (cf. pp. 68 ff., 216). After the King's Peace when Sparta ruled with the backing of Persia, the unholy triad of the King, the tyrant and Sparta was denounced by the critics of Spartan rule; cf. Isocr. 4. 126; and see 47. 2-4, 7n. συμμαν ἕσπεíιϑεσϑαι is careless Diodoran rhetoric; Ephorus will have phrased things better.

Archon Nicon, 379/8 24. 1. 'Pωμαîοι κτλ: This is 387 V; cf. CAH2 vii. 2. 637. Six tribunes promised and six named. Livy's college by contrast consists of only five (6. 5. 7) though 'almost certainly a sixth has been lost' ( Drummond, 1980: 65 and n. 37). If D's Γáιν Σευíλιον is the same as Cn. Sergius then D and Livy have three names in common ( Drummond, 1980: 65 and n. 37). See 22. 1n.

24. 1-3. Western affairs For these sections see pp. 82 f. and 15. 1 n. 24. 1 ἐπì δοÚτων ωαρχηδóνιοι κτλ: see 14. 107. 2 for the fate of Hipponium. 24. 2. τἱ. . . Λíßυες: for the African subjects of Carthage see Gsell, ii. 287ff.; Huss, 1985: 469ff.; Warmington, 1969: 55ff.; Ameling, 1993: 210ff. οἱ. . . τήν Σαρδóνα καΤοικοûντες: for Carthage and Sardinia see Barreca, 1968: 74ff.; id., 1971: 7ff.; id., 1974; Moscati, 1966: 215ff.; id., 1968; Whittaker in Imperialism, 68f., 72f.; Huss, 1985: 473. 24. 3. ϑυσíαις: doubtless of children: 20. 14. 4-7; Warmington, 1969: 147ff.; Stager, 1980: 1ff.; Scullard, CAH2 vii. 2. 514f.; Lipinski, 1988: 151ff. -229-

Archon Nausinicus, 378/7 25. 1. 'Pωμαîοι κΤλ: 386 V; cf. CAH2 vii. 2. 637. Four tribunes are announced and four given, two short of Livy's six (6. 6. 3); see Drummond, 1980: 61f. and pp. 27f.

25-27. The liberation of Thebes 25. 1. πì δ τοúτων. . . αἰτíας: D's date is 'out', too low, by a year. The mistake is probably due to the fact that the sentence represents the chronographic entry to which D attached the Ephoran narrative by means of the frequently occurring phrase διà τοιαυτας αἱτíας; see p. 135. It is evident from these chapters that Ephorus began Book 21 of the Histories with an account of the liberation of Thebes (pp. 94 f.). Thebes was freed in midwinter 379/8. The time of the year is obtained from X. H. 5. 4. 4 and Plut. Pel. 24. 2; 25. 2; cf. X. H. 5. 4. 14; Plut. Pel. 9. 2; Mor. 596c; Kallet-Marx, 1985: 135f. and nn. 35 and 36. The year itself is derived by working backwards from the summer of 375 when Timotheus operated in the Ionian Sea and won over Corcyra, Acarnania, and Cephallenia (X. H. 5. 4. 64). The decree introducing these states to the Athenian Confederacy is firmly dated to the autumn of 375 (Svt 262=Tod 126). In the previous year, 376, Cleombrotus had failed to invade Boeotia, and 377 and 378 are the respective years of Agesilaus' successive invasions of Boeotia (X. H. 5. 4. 35ff., 47ff., 59, 63). Therefore 379/8 is the winter of the coup. Cf. Beloch, GG iii. 2. 234. ἀδíκως: for Ephorus' attitude to Sparta and Thebes see pp. 117 f. πολλοùς. . . πεθυγαδευκóτων: see 20. 2n. καì συνεπιλαßομἐνων αϑηναíων: '[the exiles] secured the support of the Athenians' (Sherman); 'avec l'aide d'Athènes' (Vial). This appears correct, the absence of τω + ̑ν notwithstanding. A possible alternative 'and with the support of some Athenians' is unlikely because of the omission of τινω + ̑ν vel. sim. But even the first translation may represent in brief summary a version of the early stages of the liberation which is in broad agreement with that of Xenophon, H. 5. 4. 2f. (Grote, x. 124 n. 1. Burnett, 1962: 15 and Kallet-Marx, 1985: 141 n. 57 seem to have missed this information in D). D is so brief and vague that we would not be justified in assuming (with Rice, 1975: 100) that the timing envisaged for the -230-

Athenian help is different from that of Xenophon. Nor of course can we tell whether Ephorus considered this initial help to have been given officially or unofficially. It is generally thought that the inference to be drawn from Xenophon is that the help was unofficial, but, as Kallet-Marx argues ( 1985: 141 ff.), this is not at all certain. That some Athenians were privy to the plot and helped the returning exiles, whether officially or unofficially, is also stated by Dinarchus 1. 38 and implied by Plut. Pel. 14. 1, but only X. H. 5. 4. 9 gives any detail. For the close rapport between the Theban exiles and the anti-Spartan group in Athens cf. Aristid. Panath. 283 Dind. The precise involvement of Athens in the later stages of the liberation is even more problematical (see below). 25. 2. καì πρω + ̑τον δU=1F73ν κτλ: the fullest description of the coup is in Plut. De genio and Pel. 7ff., both of which differ in some respects from X. H. 5. 4. 2ff. (on which cf. Tuplin, 1993: 147f.), and indeed from each other. But considering the nature of the event, the secrecy required before it, and the boastful elaborations of the conspirators following its success, especially as they competed for power (cf. Plut. Pel. 25. 5), what is remarkable

is not the differences, but the fact that Xenophon's and Plutarch's accounts are materially the same. Nepos, Pel. 2-3, agrees closely with Plutarch and doubtless derives from the same original source. Polyaen. 2. 3. 1; 2. 4. 3; and the scholion ad Aristid. Panath. 279 Dind. give garbled and partly fictitious versions of the coup. Cf. Westlake, 1939: 21f. J. de Voto attempts a reconstruction in Daidalicon: Studies in Memory of R. V. Schoder, ed. R. F. Sutton ( Wauconda, Ill., 1989), 101ff. Borthwick, 1976: 148ff., suggests that the representation on the amphora of Panagjurischte ( -JHS94 ( 1974), pls. 4f.) refers to the Theban liberation, especially as described in Plutarch; but cf. Buckler, 1980a: 286f. 25. 3-27. 3. D's account of what followed is diametrically opposed to that of Xenophon. D represents a fairly lengthy siege of the Cadmea lasting for no less than a week and including vigorous fighting (even after the rhetoric has been stripped away) with the participation of a strong and official Athenian expeditionary force. The Athenian aim, we are explicitly told, in wanting to free Thebes, was to repay a debt of gratitude and at the same time to have the Boeotians on their side in the struggle against Sparta (0 5. 26. 1). Xenophon's account is far simpler. There was Athenian involvement, but small-scale though perhaps not unofficial (see n. -231-

above); two strategoi with a body of troops reached Thebes shortly after the coup and in time for the assault on the Cadmea. The garrison of the latter surrendered almost immediately after hardly any fighting. The expedition of Cleombrotus was not meant to relieve the Cadmea for its garrison had already surrendered when the king set out from Sparta (H. 5. 4. 9 ff.) Plutarch in the Pelopidas appears to have used Xenophon to a certain extent, but the narrative in both the Pelopidas and the De genio reflects mostly the version followed by Ephorus, a version which goes back to at least 373/2 to judge from Isocr. 14. 29. Though Plutarch does not explicitly mention an Athenian vote to succour Thebes, his statement, τήν τε συμμαχíÁν απεíπαντο τοîς Θηßαíοις ('αϑηναîοι), at Pel. 14. 1 (cf. 15. 1) might be taken to imply it. Like D too he describes a lengthy and arduous siege of the Cadmea with the Thebans battling against time to free the acropolis before the arrival of the Spartan army. Similarly he agrees with D that there were three Spartans in command on the Cadmea and that they were tried and condemned (but see 27. 3n.). The Ephoran version receives its strongest support from Dinarchus 1. 38f. (dated mid-320s). There the speaker in no uncertain terms relates that apart from the initial help to the returning exiles which resulted in the liberation of Thebes, the Athenian assembly voted on the proposal of Cephalus to assist the Thebans in force; καì

ξελϑóν↔ων

κεîσε τω

+ ̑ν ὗμετἐρων πατἐρων ὀλíγαις ἡμἑραις ἐξßλU=+1F75ϑη ὁ Λακεδαιμονíων θρούραρχος κΤλ . This is closely echoed by Aristid. Panath. 179 Dind.; 38. 486 Dind.; cf. schol. Aesch. 2. 117. Aristid. Panath. 283 f. Dind., is quite explicit: there was first the winter expedition which freed the Cadmea followed later by the campaign against Agesilaus. Scholarly opinion is divided over the issue with a majority leaning towards Xenophon. Some attempt a reconciliation of D and Xenophon: see Grote, x. 124 n. 1, who produced seemingly irrefutable arguments in favour of Xenophon; Schaefer, 1885-7: i. 16; Stern, 1884: 44ff.; Fabricius, 1893: 448ff.; Beloch, GG iii. i. 144ff.; Marshall, 1905: 12f.; Judeich, 1927: 171ff.; Accame, 1941: 18ff.; Burnett, 1962: 15f.; Cawkwell, 1973b: 56f.; Seager, 1974: 45f.; Rice, 1975: 95ff.; Sinclair, 1978: 42; Kallet-Marx, 1985. In fact the two accounts are irreconcilable. One must choose between them. It is felt by most scholars that as Xenophon is the more -232reliable of the two, his version should be accepted and that of D, Dinarchus, and Aristides (and Plutarch and Isocrates) rejected as being no more than an understandable confusion: D's 5,000 Athenians marched to the aid of Thebes in the summer of 378 at the time of Agesilaus' invasion, and not six months earlier (Grote's suggestion followed by many including Burnett and Rice). It is considered inconceivable that Xenophon could not have known of such a major expedition. However, it is not perhaps the case that Xenophon was not aware of the midwinter march to Thebes. It is simply that he chose to ignore an action which he would have ascribed to the influence of the ßοιωτιáζοντες at Athens (cf. H. 5. 4. 34). Does he not after all fall to inform us of the Athenian participation in the operations against Agesilaus in the summer when Chabrias employed his famous tactic? And only his accidental mention of Chabrias at H. 5. 4. 54 betrays the fact that the Athenians were again fighting alongside the Thebans in 377, for the second successive season. These are hardly less serious failings. There is in addition an apparent inconsistency in Xenophon's account which should put us on our guard. Xenophon represents Cleombrotus' expedition as aimed at Thebes and not at relieving the Cadmea for the latter fell, according to him, before Cleombrotus set out (H. 5. 4. 13). Yet he says ( ibid. ) that the Spartans put to death the harmost who had surrendered the Cadmea and had not waited for the relief force. What relief force is that? The ßοόɥϑεια sent for to Thespiae and Plataea (H. 5. 4. 10) which either had not materialized (Thespiae) or had failed to get through (Plataea)? One cannot help suspecting that the ßοόɥϑεια at H. 5. 4. 13 is none other than Cleombrotus' army which would have arrived in time had the Cadmea garrison held out for two or three days longer, as D and Plutarch tell us. Also, the indecision of Cleombrotus which draws Xenophon's criticism (H. 5. 4. 15f.) rather indicates that the king had set out with one specific order -to relieve the Cadmea. But as that had already surrendered when he entered Boeotia, he was left in effect without a definite objective. His soldiers cannot have been alone in wondering whether or not a state of war existed between Sparta and Thebes (H. 5. 4. 16). Better sense is made if Plutarch's information is brought into play at this stage. On reaching Corinth (or Megara; see 27. 3n.) Cleombrotus was met by the retreating Cadmea garrison. He -233thought it better (and safer?) to proceed, but he no longer had a definite goal and it is likely that the authorities in Sparta were equally uncertain for a while as to what action to take.

Thus D ought not to be dismissed on the strength of Xenophon's far from reliable testimony. As to whether an Athenian expedition to Thebes at this moment is credible or not, it must be admitted that none of the objections advanced against it is compelling. Grote's criticism that such an Athenian undertaking 'would have been the most flagrant and unequivocal commencement of hostilities against Sparta' (x. 124 n. 1; followed by Rice, 1975: 109 n. 34) is weak: the Athenian people would on the contrary have considered their action as helping to rectify an obvious breach of the King's Peace, the continued occupation of the Cadmea by Sparta. The expedition could hardly be construed as a declaration of war against Sparta, and Sparta did not construe it as such. Burnett's objections (1962: 15f.) are equally unsound; see Cawkwell refutation, 1973b: 57 n. 2. Nor can the argument stand ( Rice, 1975: 102) that the Athenians could not have voted to march to Boeotia on such short notice and risk encountering a Spartan army there. Is that not precisely what they did in 395 (X. H. 3. 5. 7ff.)? Not only is D's version not implausible, therefore, but it is also self-consistent. No less than a week would have been required for a message to reach Sparta from Thebes and for Cleombrotus to gather an army and reach the Isthmus; and we are told in fact that the Cadmea held out for as long as possible and was only starved into surrender (cf. Dinarch. 1. 39). The evidence of Dinarchus and Aristides may appear rather dubious at first sight. Dinarchus is clearly boasting and Schaefer for one felt justified in dismissing Aristides as high-flown rhetorical exaggerations in praise of Athens. There are inaccuracies too. It is not possible, for example, that the Theban exiles were granted Athenian citizenship, but only ateleia or isoteleia (IG ii2 37+ addenda; IG ii2 245; cf. Plut. Pel. 7. 1). These criticisms nevertheless fail to detract from the value of their explicit testimony, especially as the same testimony is found in D (Ephorus) and Plutarch and echoed in Isocrates. The charge that the Ephoran version is another instance of the fiction-making power of Athenian propaganda cannot be refuted. We have seen, however, that an Athenian expedition to free the Cadmea is not in itself improbable, and in the absence of a statement to the contrary (Xenophon does not -234-

contradict, he is merely silent), there is no logical reason why the Ephoran version should not be allowed to stand. Two points require comment. First, Xenophon's statement that Chabrias guarded the road to Boeotia via Eleutherae and Cleombrotus consequently was forced to take τὴν εἰς IIλαταιàς θ ρΥσαν (H. 5. 4. 14). The view that Chabrias' objective was to keep Cleombrotus out of Attica (cf. Grote, x. 130) should be dismissed without further ado for he could not have done so by holding Eleutherae, and it is not what Xenophon says he was there to do. Clearly he was guarding a main, perhaps the main, road from the Megarid into Boeotia. Hammond, 1954: 103ff., argues that the main road from the Peloponnese to Boeotia ran from the Isthmus to Khani Derveni, to Ayios Vasilios, to Kriekouki in Boeotia (nr. ancient Erythrae) -- see his fig. 1. Cf. Ober, 1985: 118 ff. After Ayios Vasilios it passed through the Attic territory of Oenoe and Eleutherae. According to Hammond, this is the shortest and easiest of all the routes into Boeotia and it must have been practicable for wheeled vehicles throughout its length (cf. Pritchett, 1957: 16ff.; id., 1965-85: i. 119f.). While over Cithaeron, and while still on the Megarian side, this road is joined by a track which climbs steeply from Plataea. This is the route Cleombrotus must have taken when he found the main road blocked by Chabrias at Eleutherae, Hammond suggests (1954: 113f.). Cawkwell, 1973b: 57f., believes that 'Chabrias was covering the direct route to Thebes to prevent the Spartans, as part of Athens' formal help to the Thebans' (i.e. Demophon's army), from invading Boeotia, and that consequently D's version is to be preferred to that of Xenophon. Kallet-Marx, 1985: 143f., who prefers Xenophon to D on this point, sees Chabrias' mission as part of a 'coordinated (Athenian-Theban) defense on the Kithairon' and as a further indication that the action of the two generals did not lack official support. Unfortunately, however, Chabrias' presence at Eleutherae cannot be used as proof for either version, for quite independently of any connection with what was or was not happening at Thebes at the moment, Chabrias served a purpose by being where he was: the main road passed through Eleutherae and Eleutherae was Attic territory. Why should the Athenians have allowed the Spartans to traverse their territory to attack the Thebans? Let them take another (and more difficult) -235route. Also, if the two generals had indeed acted unofficially in the Theban affair, Athens may have feared equally 'unofficial' reprisals by Cleombrotus' troops on their way to Boeotia. Hence Chabrias' presence at Eleutherae (cf. Sinclair, 1978: 42f.). Second, there is the condemnation of the two Athenian generals, seemingly a formidable obstacle in the path of Ephorus' version (cf. e.g. Burnett, 1962: 16). But again this incident can add weight to neither version. If the generals acted unofficially the legal reason for their condemnation will have been that they acted without due authorization by the state in a way thought (at least at the time of the trials) to have endangered it. All Greek states, and Athens especially, were ever sensitive about their generals acting independently and they always exercised a tight control over them (cf. Arist. Ath. Pol. 61. 2; Pritchett, 1971-91: ii. 29ff.). But as most such trials were political in character we may guess that they belong to that uneasy period ( Cawkwell, 1973b: 58ff.) when Athens was unsure about how to proceed. Their condemnation therefore does not prove that they had not taken the trouble to secure the consent of the boule and their fellow generals; cf. KalletMarx , 1985: 144f. 25. 3. οὐκ λáττους τω + ̑ν χαì πεντακοσíων: Plutarch gives the same figure, Pel. 12. 4; De genio 586e; 598f. Parke, 1927: 162ff., finds this excessive for a mere garrison. He points out that Callibius had only 700 men in Athens in 404/3 and proposes that the Cadmea garrison had been augmented by another force in winter quarters (see 27. 3n.). The Cadmea nevertheless is a far bigger area than the Athenian acropolis and less defensible, and in any case the Spartans may have learned from their earlier experience. With the help of the pro-Spartan faction, 1, 500 men ought to have been able to keep Thebes in subjection and had they intervened on the night of the coup the outcome would probably have been very different; cf. X. H. 5. 4. 10f.; Plut. Pel. 12. 4; De genio 598e-f. The three commanders may not have been condemned just for surrendering the Cadmea, but for their earlier timidity too. κ τóπερδεΞíων: a popular expression with D. Cf. 15. 32. 4; 32. 6; 69. 2; 69. 3; 18. 15. 6; 17. 4; 45. 1; 20. 29. 8; 44. 5; McDougall's.v. Literally it means 'from commanding places on the right'; cf. X. An. 4. 8. 2. As the right-hand side was the unshielded side, an attack from that direction was particularly dangerous. The

expression, however, came to mean simply -236-

(ground or place lying above one' (cf. X. An. 3. 4. 37; 5. 7. 31) which thus gave one the advantage, and this is how it is used by D. 25. 4. κ τńς Ἑλλáδος: a reference probably to Sparta's position as hegemon in Greece, or perhaps to her prostasia of the Peace (5. 1n.), or both (they coincided); see Plut. Ages. 28. 6, αὐτοîς μέν (Λακεδαιμονíοις) ὁδοΏ τι ὃλης ὑπαρχούσης κσπόλος 'Eδἑ τω + ̑ν Θηßαίωνγεγονότων (in 371 before Leuctra), and cf. Herod. 6. 72 with Plut. Them. 20. 1, τω + ̑ν Ἑλλήνων στόλος, for the force Sparta led to Thessaly in the 470s against the Aleuadae who had medized. Sparta had most of Greece on her side (or rather under her) and could further threaten Thebes with the King's Peace, perhaps in the name of Greece (cf. 38. 2: 'κατἑστησαν ΟἱEλληνες 50. 4: οὐ προσεδU=1F11χϑησαν U=1F51πὸ τω + ̑ν Ἑλληνων). U=1F51πομιμνήσκοντες: they may also have appealed to the autonomy clause in the King's Peace; cf. Isocr. 14. 24. καì αὐτοì συγκατήγαγον: not strictly correct. The Thebans protected and aided the Athenian exiles, but they did not join Thrasybulus' force in 404/3: 14. 32. 1; X. H. 2. 4. 2; Plut. Pel. 6. 5; Justin 5. 9. 4ff. 26. 1.ὁ δέ δU=0144μος τω + ̑ν αϑηναíων κτλ : see 25. 3-27. 3n. Ephorus' estimate of Athenian attitudes at this time is different from that of Xenophon. τὸρ γàρ ἔϑνος . . . : cf. 20. 1n. 26. 2. Δημοθω + ̑ν: Vater suggested that this could be the same man as the Demades (or Demeas) of the schol. Aristid. Panath. 284 Dind. (cf. Schaefer, 1885-7: i. 20 n. 1; PA 3693; Develin, 222). If that is so then Demophon (or Demades or Demeas) may have marched out to Boeotia twice in 378, on the first occasion to help free the Cadmea and on the second with Chabrias to resist Agesilaus' invasion. κατασΤαϑεὶς στρατηγὸς: he would have been a general already. For D's ignorance of the mechanics of the Athenian system and his misleading language when dealing with such matters, see p. 133. The annual election of generals in the 4th century was held after the sixth prytany ἐθ ὦν ἂν εὐσημίαγἑνηται ( Arist. Ath. Pol. 44. 4) though this does not necessarily indicate the seventh prytany; see Meritt, 1970: 277f.; Rhodes, CAAP ad loc. πεντακισχιλίους . . . ἱππεîς: in the summer of 378 the Athenians helped with 5,000 hoplites and 200 cavalry (32. 2), but this is no -237-

argument against D's account as Burnett, 1962: 15 supposes; cf. Cawkwell, 1973b: 57 n. 2. καταλἑξας: D uses καταλἑγω and καταγράθω in a non-technical sense, 'to enlist' or 'raise a levy', and for all kinds of states; cf. 11. 1. 5; 11. 48. 4; 11. 84. 4; 13. 2. 2; 13. 44. 6; 13. 97. 1; 13. 109. 1.; 14. 44. 2; 15. 15. 2; 15. 52. 2; 15. 61. 4; 16. 28. 1; 16. 73.3; 17. 39. 3. A discussion of the Athenian κατάλογοι is therefore unnecessary. (For which see HCT iv. 264, 295; Andrewes in Studies McGregor 1 ff.; Rhodes, CAAP 327, 566f.) 26. 4. ὁπλîται μἐν. . . δισχιλίων: if we subtract the Athenian force (as Bruce, 162, omits to do) we are left with 7,000 Boeotian hoplites and more than 1,500 cavalry. The number of hoplites is credible enough. It compares well with other figures for Boeotian contingents. At Delion in 424 there were 7,000 hoplites (Thuc. 4. 93. 3). At Mantinea in 418, 5,000 (Thuc. 5. 57. 2). At Nemea in 394, 5,000 (the Orchomenians were absent: X. H. 4. 2. 17). Seven thousand took part in the second invasion of the Peloponnese in 368 (15. 68. 1), and 8,000 were sent to rescue Pelopidas in the same year (15. 71. 3). We may believe D that many from the other Boeotian cities hurried to help free the Cadmea, and as Thebans of the older age-groups would have joined in too (whereas they would not have been included in an expedition), the 7,000 figure is not unreasonable, though one would not of course expect all Boeotians to have helped Thebes. The 1,500 cavalry, however, seems inflated. According to Hell. Oxy. 16. 4, each of the eleven districts of Boeotia provided 100 cavalry thus making a force of 1,100, and the historian is speaking of the older Confederacy (447387/6) which was wealthier and more populous perhaps than later Boeotia. The relevant cavalry figures for the campaigns mentioned above are 1,000, 500, 800, 600, 600. Thus even allowing for the fact that this was a home operation and that older 'knights' might have participated, D's figure seems exaggerated. It is possible that a different figure stood in the text originally and that διοχιλίων is a dittography. The cavalry was not meant to assault the Cadmea of course, but to operate against Cleombrotus if he arrived before the citadel fell. For the Boeotian army, see Salmon, 1953: 347ff. διεόμενοι . . . ἐπο ιοûντο: the expression and variants occur many times, almost always in the description of sieges. See e.g. 12. 61. 3 -238-

(425); 13. 55. 5; 56. 4; 56. 8; 59. 7 (409); 13. 64. 7 (409); 16. 74. 2, 5 (341); 17. 12. 2 (335); 17. 22. 1 (334); 17. 24. 4 (334); 17. 67. 5 (330); 17. 85. 6 (327); 18. 13. 1 (323); 18. 34. 4 (321); 20. 45. 7 (307); 20. 88. 8 (305). These narratives are based on no less than five different sources. Did all five independently use the expression or did D find it in one of them and make it his own? It has been suggested that sieges pursued in this fashion, by dividing one's force and attacking in relays, while not unknown to 5thcentury Greeks, were first used to advantage by the Carthaginians at the siege of Selinus in 409, to judge from D's information and the fact that his use of the cliché begins in earnest with that event. The Carthaginians possessed both the necessary manpower and disregard for life to make such a tactic worthwhile (see Sinclair, 1966: 249ff.; Garlan, 1974: 159ff.). The chapters dealing with the Carthaginian invasion of Sicily derive from Timaeus who in turn based himself on Philistus (pp. 61 ff.). It is possible, therefore, that the cliché had its origin in the latter author. The problem with this suggestion is the statement, presumably Ephorus' own, that Xerxes' troops were accustomed to fight ε + ̑κ διαδοχη + ̑ς(11. 8. 2). It would therefore appear that in Ephorus' opinion large barbarian armies fought in this fashion. But perhaps this is an idea he borrowed from Philistus who had, after all, observed the Carthaginians in action at first hand. The important point to be determined is whether D, once having enriched his vocabulary from whatever source or sources with this formula, proceeded to use it indiscriminately whether the facts of a particular case and the language of his sources demanded it or not. This is all too possible and the simpler expressions ροσπ≀beßολα + ̑ς ποιοΥ + Áμενοι and σΥνεχει + +0311ς ποιοΥ + ̑μενοι are used fairly indiscriminately in the description of sieges; cf. e.g. 12. 47. 2 with Thuc. 2. 75ff.; 12. 55. 5 with Thuc. 3. 18. 4f.; 12. 72. 3 with Thuc. 4. 75. 1; 12. 72. 9 with Thuc. 4. 131. 3. In none of these instances is σΥνεχει + ̑ς προσßολα + ̑ς justified by Thucydides' language. Whether Ephorus' language justified its use, or whether indeed Ephorus himself used it, or something similar, is impossible to say. All the same, the more elaborate formula seems to be different. At any rate at no instance can it be proved that the use of the phrase ε + ̑κ διαδοχη + ̑ς was not necessitated by the details of the siege, as far as these can be discovered. Specifically, and if we except the Carthaginian and Macedonian siege operations (where the use of -239-

the tactic requires no proof), 'in relays' does not seem inappropriate to the remaining examples which involve Greeks. In the case of Pylos (12. 61. 3) Thucydides himself says that the Peloponnesian warships attacked in relays. He does not use the term ε + ̑κ διαδοχη + ̑ς, but Ephorus, it would seem, did and clearly that was a correct rendition of Thucydides' description. For the capture of Pylos in 409 X. H. 1. 2. 18 gives no detail whatsoever. But the Spartans were in a hurry to capture the place before it was relieved from Athens (13. 64. 6) and it is probable that they were prepared to sustain heavier than usual casualties by attacking in continuous waves. Similarly, in the instance of the Cadmea, there is no good reason to deny that the expression τα + ̑ς προσßολα + ̑ς ε + ̑κ διαδοχη + ̑ς ε + ἐποιοΥ +̑ντο was justified by the facts and therefore that it was employed by Ephorus. We may compare Plut. Pel. 13. 1 ό Πελοπι + ίδας . . . προσßολα + ὰς ἐποιει + ̑το τανταχὁΘεν, being anxious to capture the citadel before the arrival of the Spartan army. Given that it was a race against time, what else could the Thebans have done but divide their forces and attack on a wide front (Plutarch) and in successive waves (D), hoping to exhaust the besieged and force them to surrender? It was a time when they could well forsake the normal Greek reluctance to sacrifice citizen troops in steady assaults on the walls in order to gain their objective quickly. There is in fact a good example of how D (Ephorus) might or might not substitute the formula for an expression which meant the same thing. written ἐκ διδοχη + ̑ς instead of κατὰ με + àρος, 'in turn' (for this At 15. 83. 4 in describing Epaminondas' sudden use of κατὰ με + àρος cf. 13. 108. 7f.;; Thuc. 4. 26. 3). Sinclair, therefore, is not at all persuasive when he argues ( 1966: 255) that this aspect of the siege, however expressed, was not in Ephorus 'since precise military details of this kind are hardly to be associated with Ephorus'. On the contrary, Ephorus appears to have been very much interested in matters of this sort; see pp. 124f. and 34. 4n. 27. 2. ἀπὸ τω + ̑ν σΥμμαχίδων . . . πολλαπλάσιοι: it is argued above (19. 3n.) that the Cadmea was captured by allied troops under -240-

Spartan command and we now see that the garrison continued to be composed mostly of allies. This would have added a touch of legality to the occupation. σιόπερ οὗτοι . . . ἀπηλλάγησαν: it may be significant of Ephorus' attitude that D, like Plutarch, does not mention the butchery which ensued (X. H. 5. 4. 12). Or the silence may be the result of D's abridgement. 27. 3. Λακεδαιμόνιοι . . . ὑστερήσαντες: they were under the command of Cleombrotus ( X. H. 5. 4. 14ff.; Plut. Pel. 13. 2). Plut. Pel. 13. 2 says that the retreating garrison was met by the advancing Cleombrotus at Megara (or Corinth acc. to De genio598f). Thus the siege of the Cadmea must have lasted for at least a week. Dinarch1. 39 13 on the other says ὀλίγαις ήμέραις, but from the arrival of the Athenians on the scene; see 25. 3-27. 3 n. hand mentions but one harmost at Thebes who was executed for surrendering the Cadmea. Parke, 1927: 159ff., accepts the testimony of Plutarch and D and proceeds to find roles for the three harmosts. Lysanoridas, he thinks, was in charge of the Cadmea while Herippidas held an independent command against the further expansion of Jason of Pherae. This is not warranted by the evidence. We may concur that there were three Spartan commanders, two of whom were executed, Herippidas and Arcesus, and the third, Lysanoridas, heavily fined (he had been away at the time: De genio578a). But Xenophon is more likely to be correct that there was but one harmost (cf. Stern, 1884: 58 n. 1). The other two officers would have been the second and third in command. We may compare Thuc. 4. 38. 1 (cf. 3. 100. 2; 109. 1. Arnold ad 3. 100. 2 and 4. 38. 1 explains correctly) and Plut.

De genio576a, ΛΥσανρίδαν δέ τὐτόν. A fragment of Theopompus (FGH 115 F240) speaks of unsavoury actions at Sparta in connection with the condemnation of Lysanoridas, with Agesilaus playing none too edifying a part, and this may be a reason for Xenophon's unsatisfactory brevity here. The harmost is unnamed, a sure sign that his memory was anathematized in Agesilaus' circle (cf. Cawkwell's introduction to the Penguin translation of the Hellenica and his nn. ad 4. 5. 11; 5. -2414. 13). Cf. 25. 3n. For the heavy fine which forced Lysanoridas to flee, see 20. 2n. 27.4 μετὰ δὲ ταΥ + ̑τ + ̕ . . . ἐπιßολήν: if this is the unsuccessful Theban attack on Thespiae in autumn or winter 378 which is related at 33. 5 f. then D is guilty of a doublet here. Cawkwell, 1973b: 55, considers this an instance of overlapping narrative due to D's method of epitomizing portions of Ephorus and using them to fill out his chronological scale. In fact he finds three overlapping narratives in this part of Book 15. 'Chapters 25 to 27 recount Theban affairs from the liberation in midwinter 379/8 down to the Theban failure to take Thespiae in winter 378/7. Chapters 28 to 30 recount the history of the Second Athenian Confederacy from its foundation in 378 to its expansion in the summer campaign of Chabrias in 377. Chapters 31 to 35 recount the Spartan military offensive against Thebes from 378 to 375. The first and third narratives both recount the Theban attack on Thespiae (27. 4 and 33. 5). The second and third both recount the Spartan reaction to the events of early 378.' The reference is to 28. 4 and 31. 1. This, though not totally incorrect, is too schematic and inflexible a way of treating D's text. Chs. 25 to 28 narrate the liberation of Thebes, the creation of the new Boeotian Confederacy and the Second Athenian Confederacy and (in a digression: 28. 2-3n.) the first steps of the latter, and finish with the Spartan reaction (28. 4-5). The time by now is the spring of 378. There then follows a review of Oriental history from the King's Peace to the present time (29. 1-4 and see p. 100 ), which splits the Greek narrative in two. At 29. 5 the Greek narrative is resumed with the raid of Sphodrias, the outbreak of war, and the further progress of the Athenian Confederacy down to summer 377 (30. 5). Section 31. 1 carries on where 28. 5 left off. The Spartan approaches to their allies are now beginning to bear some fruit and we are told what the Spartan ει + ̕ς τὸν πόλεμον παρασκεΥη + ̑ς . . . øροντίδα (28. 5) amounted to (31. 1-2). There is no question therefore of a doublet though clearly there is a chronological overlap of at least one year between 30. 5 and 31. 1. Nor is it easy to conceive of 27. 4 and 33. 5f. as an instance of overlapping narrative for how could Ephorus have reached the winter of 378/7 (the attack on Thespiae) from the winter of 379/8 (the liberation of Thebes)? Surely by way of the very events which Cawkwell assigns to the second and third Ephoran narratives he traces in D's text. And as -242in fact D does not omit these events, we cannot speak of a telescoped narrative at 27. 3-4 as we can at e.g. 77. 4. A different explanation is therefore required for 27. 4 and the simplest and most obvious is that Thespiae was attacked twice. Before withdrawing from Boeotia in midwinter 379/8 Cleombrotus installed Sphodrias as harmost at Thespiae. He left with him one-third of the relief force and money to hire mercenaries with (X. H. 5. 4. 15). His orders were to receive and help τΟὺς α + ̕øισταμ + ένοΥς τω + ̑ν Θηßαίων ( Plut. Pel. 14. 3). We may add that Thespiae was strategically placed for the campaigns against Thebes which the Thebans knew were bound to come. The city was made the centre of his operations by Agesilaus in the summer campaign of 378 (X. H. 5. 4. 38, 41). There is no difficulty therefore in envisaging two Theban attempts to capture Thespiae. The circumstances of the two are quite different. The first occurred immediately after the withdrawal of Cleombrotus and the departure of the Athenians as D says (27. 4). The Thebans in fact may have been anxious to launch an attack before Sphodrias had used his money and considerably increased his force. He is said to have had some 10,000 men when he marched against the Piraeus in the spring (29. 6). We hear too that Agesilaus rebuilt or strengthened the fortifications of Thespiae in the summer of 378 (X. H. 5. 4. 41. We should not perhaps allow ἐτείχισε its full force; see 46. 6n.). Was that a need made obvious by the winter attack? The second attempt occurred in autumn or winter 378 when Phoebidas was harmost (33. 5f.; X. H. 5. 4. 42ff.). Buckler too, 1979: 54, 56, accepts D's two attacks, as does Tuplin, 1993: 158 n. 35, but without argumentation. And see also KalletMarx, 1985: 133 n. 29.

27. 4. A chronographic notice ἅμα δὲ τούτοις Πραττομένοις Π + ̔ωμαι + ̑οι κτλ: perhaps from the Greek chronographer; see pp. 25 ff. Wesseling, not believing that Sardinia could possibly have formed a Roman objective at this time, proposed replacing it in the text with Satricum, a place in Latium some 30 miles south of Rome which did as a matter of fact receive a Roman colony in the 4th century (385 V: Livy 6. 16. 6; cf. CAH2 vii. 2. 280, 316). Some are inclined to agree (cf. Perl, 1957: 127; Didu, 1972: 310ff.). But whatever the precise motive -243and timing might have been, a Roman colony to Sardinia at approximately this time is far from impossible: see 14. 3n.; CAH2 vii. 2. 315.

rchon Callias, 377/6 28. 1. 'Π + ̔ωμαι + ̑οι κτλ: 385 V; cf. CAH2, vii. 2. 637. Four tribunes are announced and four seem to have been

named though the MSS are even more unsatisfactory here than elsewhere. If Kόιντον Λ is a corruption of L. Quinctius (Vial, 131) then three of D's names correspond with names in Livy 6. 11. 1 where five tribunes are cited -- six if Broughton is correct in restoring Cn. Sergius to the text: MRR i. 101f. See pp. 27 f. and Drummond, 1980: 61f.

28. 1. The restoration of the Boeotian Confederacy ἐπὶ δὲ τΟύτων. . . τωιησάμενοι: D therefore places the refoundation of the Boeotian Confederacy early in 378, a date supported by Plut. Pel. 13. 1 where it is said that Pelopidas was elected Boeotarch with Melon and Charon the day after the coup, while at 14. 2 Plutarch says that, at the time of the raid of Sphodrias (spring 378), Pelopidas was Boeotarch with Gorgidas, and at Ages. 24. 6 he names Pelopidas and Melon as Boeotarchs at this time. The reinstitution of the Boeotarchy implies the refoundation of the Confederacy, or at least the intention to refound it. Such an early date for the restoration of the Confederacy has not found unanimous acceptance though none of the arguments against D and Plutarch is compelling. Beloch, for instance ( GG iii. 1. 145 n. 2), rejects Plutarch's claim that Boetarchs were elected on the morrow of the coup because of Isocrates' statement (14. 29) that soon after the liberation the Thebans tried to come to an agreement with Sparta: α + ̕λλ + ̕' εΥ + ̕Θὺς ει + ̕ς Λακεδαίμονα πρέσßεις α + ̕πέστελλον, ἕτοιμοι δοΥλεύειν ὄντες καὶ μηδὲν κινει + ̑ν τω + ̑ν πρότερον πρός πρὸς αΥ + ̕τοὺς ω + ̔μολογημένων. This Beloch takes to mean that the new regime was promising to abide by the terms of the King's Peace which would presumably have excluded the restoration of the Confederacy. It was not Boeotarchs, therefore, but polemarchs that were elected. Beloch's argument will hardly bear examination. First, what does Isocrates mean? Were the Thebans really promising to confine their actions within the terms -244-

of the Peace? And how would Thebans in general view those terms anyhow? In 382 Ismenias and his faction did not hesitate to challenge Sparta's call for action against the Olynthians who were flouting the autonomy clause by any account. In 373 they captured and destroyed Plataea ει + ̕ρήνης οὔσης (the renewal of 375: Isocr. 14. 1 and passim). Nor did they dissolve the Boeotian Confederacy as they had promised to do at the conclusion of peace in 375. The Confederacy was clearly in existence before the outbreak of war in 373 (38. 3n.). For a telling comment on how the Thebans regarded or rather disregarded the King's Peace in the 370s we may compare Paus. 9. 1. 4f.: while the Plataeans declared that the Peace still held, the Thebans maintained that τήν τε ι + щрνην ΛακεδαιμονίοΥς ει + +0315 + ̑ναι τοὺς τρξαντας καὶ ὕστερον παραßάντων ε + ̕κείνων λελύσΘαι καὶ ἅπασιν η + ̕ξίοΥν τὰς σπονδάς (cf. Isocr. 14. 24). I therefore take Isocr. 14. 29 merely to mean that the Thebans were offering to remain within the Peloponnesian League (5. 2-3n.). Second, to argue as Beloch does is to oversimplify what must have been a complicated political situation after the liberation. While many Thebans may have counselled patience, can it be excluded as a possibility that men like Pelopidas urged a vigorous policy of opposition to Sparta and took the first step towards the restoration of the Confederacy by the election of Boeotarchs? Third, it is possible that a reason why Sparta rejected the Theban offer was because the Boeotarchy had been resurrected with all that that office implied (cf. Accame, 1941: 26). Cawkwell, 1972: 275, accepts Beloch and adds that 'the morning after the murder of the polemarchs was no moment to engage in constitutional discussions. Boeotarchy meant Boeotia, and the Boeotia of the 370s and the 360s was not the Boeotia described in the Hellenica Oxyrhynchia. . . All that would take time.' This is true, but no one would propose that the constitution and machinery of the new Confederacy were worked out in a day. The election of Boeotarchs in the exhilaration of the moment on the other hand would have taken little time and even less planning. Certainly 28. 1 can easily accommodate two to three months (it is meant to coincide with the beginnings of the Second Athenian Confederacy: 28. 2ff.), time enough for constitutional deliberations. Cawkwell further objects ( 1972: 276) that the restoration of the Confederacy would mean that there would no longer be a state -245-

called Thebes; it would be replaced by 'Boiotoi'. 'The whole point was in the name.' (He does not of course mean that Thebes ceased to exist as a state, but only that the Thebans styled themselves 'Boiotoi' in future; Buckler, 1979: 57, misunderstands.) Yet it was the 'Thebans' who joined the Athenian Confederacy (29. 7) and they are shown as such on Svt 257 = Tod 123 (date spring 377). Hence, he concludes, the refoundation of the Confederacy must be after this date. The objection is unsound. The Confederacy was certainly in existence by the late 370s yet the 'Thebans' alone were members of the Athenian Confederacy. Their name was not erased from Svt 257=Tod 123 and replaced by 'Boiotoi', and the reasons for this are obvious: the Second Athenian Confederacy stood for the autonomy and freedom of its members. Athens could hardly openly concede Theban claims to Boeotia. But it was politic to condone and condone she did, though sometimes with a guilty conscience it would seem (the Thebans justified their Confederacy as being in the interests of Athens and her allies: Isocr. 14. 21, 24ff., 33, 38). The embassy to Thebes voted for in the spring of 377 (Svt 257=Tod 123 lines 72ff.), which was to persuade the Thebans ὅτι ἂν δύνωνται α + ̕γαΘόν, should probably be seen as an Athenian attempt to moderate Theban actions in Boeotia ( pace Burnett, 1962: 13; cf. Cawkwell, 1973b: 48f.). The fact therefore that the Thebans alone were introduced to the Athenian Confederacy is no comment whatever on what Thebes was or was not doing in Boeotia. The sequence of events in D which Cawkwell notices but dismisses (1972: 276) is instructive of how Ephorus viewed the events of winter and spring 378. The close causal links are still there in D's account: the revolution at Thebes (25-7), the failure of Sparta to reimpose her will (28. 1), the birth of both the Boeotian and Athenian Confederacies (28. 1ff.; cf. Busolt, 1873-5: 679ff.). It might be worth noting the μὲν/δέ construction at 28. 1f. The first introduces the revival of the Boeotian Confederacy, the second the foundation of the Athenian

Confederacy. The matter of Pelopidas' Boeotarchies can in no way be used as evidence here, pace Bersanetti, 1949: 89ff., who argues in favour of the restoration of the Boeotarchy in 378, and Cawkwell who opts for 377. Against the latter it may be mentioned that the tradition in Plutarch that Pelopidas was a member of the first college of Boeotarchs may be false ( Fuscagni, 1972: 415ff.). -246There is no reason, therefore, to reject D's testimony that early in 378 the Boeotians set the foundations of a new Confederacy. This differed significantly from the old. The heart of the new Confederacy was a popular assembly meeting at Thebes and open to the citizens of all Boeotian cities (see n. below). The Boeotians and Thebans who together besieged the Cadmea were in effect the embryo assembly of the new Confederacy. D is explicit (28. 1): the Boeotians, encouraged by the Spartan failure to restore the situation, collected together and formed a 'common alliance'. These were the Boeotian people in arms, fresh from their success over the Cadmea garrison and Cleombrotus. Sordi, 1973: 82ff., rightly stresses the military character of the Boeotian federal assembly, a character to be traced back to the events of winter 379/8. Xenophon himself says (H. 5. 4. 46) that the demoi of the neighbouring cities kept departing for Thebes at the time of the second (27. 4n.) attack on Thespiae in autumn 378. While some of these demol will have fled to Thebes before 382 (and been expelled at the capture of the Cadmea: 5. 2-3n.), others should be identified with the Boeotians who hastened to help free the Cadmea and who remained to form the nucleus of the federal assembly. Xenophon's "dark but pregnant words" ( Cawkwell, 1972: 276), ε + ̕κ δὲ τούτον πάλιν αὐ + ̑ τὰ τω + ̑ν Θηßαίων ἀνεζωτΥρει + ̑το (H. 5. 4. 46), should not be pressed too far. Xenophon sees this 'rekindling' in military terms: with Phoebidas killed, the Thebans carried their attacks further afield. In any case why should the defeat of Phoebidas have originated the Confederacy? It was hardly more important than the check the great Agesilaus and his powerful army had received in the summer of 378, or the liberation of Thebes and the failure of Cleombrotus. (Cf. Buckler, 1979: 50ff., who also argues in favour of D and Plutarch.) The new Confederacy was marked by a new federal coinage; cf. Head, 1911: 351f.; Hill, 1906: 69ff.; Kraay, 1976: 113. Βοιωτοί: as in the case of Olynthus and the Chalcidic Confederacy (19. 2n.), D is not always consistent and correct in his use of the terms 'Thebans' and 'Boeotians' and for the same reasons. He is correct in the use of 'Boeotians' here and at 67. 3f; 80. 2; 82. 3; 16. 25. 1; 28. 3; 84. 5; 85. 3 -- all important passages for our understanding of the new Confederacy. He is wrong at 46. 6 (the decisions were 'Boeotian' not 'Theban') and 94. 2. Often the two terms are interchangeable and for stylistic reasons it would seem; -247cf. 32. 3f.; 62. 3f.; 71. 2. We should see Ephorus behind most of this usage. κοινήν σΥμμαχίαν: the precise nature of the new Confederacy is a crux. Was it a unitary state, a synoikismos of Boeotia, or a federal state? κοινὴ σΥμμαχία is too vague a term, but it does not argue for a unitary state or a synoikismos but rather for a federation. The same term is used at 57. 1 where the Thebans are said to have forced the Orchomenians to join τὴν τω + ̑ν σΥμμάχων τωλιτείαν(Vial's emendation). Elsewhere D refers to the Confederacy as σΥντέλεια, a union or community of states (38. 3f.; 50. 4; 70. 2), which does not necessarily imply the subjection of Boeotia to Thebes ( as Larsen thinks, 1968: 177 n. 1), for D uses the same term to describe not only the 5th-century Boeotian Confederacy (12. 41. 3) and the Delian League (11. 78. 4), but also the 4th-century Arcadian Confederacy (59. 1; cf. X. H. 7. 4. 12). Busolt, GS ii. 1426ff., argues for a 'Bundesstaat' (for earlier views see n. 2 on 1426). So does Larsen, 1968: 176ff., and Buckler, 1980a: 18ff. Sordi, on the other hand, 1973: 79ff., esp. 89-91 (following in the footsteps of Thiel, 1926: 19ff.), argues that the Confederacy cannot be defined as a sympoliteia, that is, a federal state. We are dealing rather with a synoikismos. After all, was it not dioikismos that the Thebans were threatened with in 371 ( Plut. Pel. 20. 1)? Her reasoning seems unsound. The threat of dioikismos was directed against Thebes herself. The order to dissolve the Confederacy was phrased otherwise; see 51. 3; X. H. 6. 4. 2f.; Plut. Ages. 28. 1f. Proof that the Boeotian cities enjoyed local selfgovernment is furnished by X. H. 7. 3. 5 1f.; Thebes had a boule and so, presumably, had the other cities. We may then accept that Boeotia under Theban leadership was a federal state where the deciding power lay with the assembly of the citizens of all Boeotian cities sitting at Thebes (ἡ κωινὴ σύνοδος τω + ̑ν Βοιωτω + ̑ν : 80. 2, or Simply, τὸ κοινὸν τω + ν Βοιωτω + ̑ν: 16. 25. 1; 85. 3). That it was democratic is shown by 79. 3 (cf. Buckler, 1980a: 30, 34ff.). The votes it would seem were counted by heads and not by cities which of course gave Thebes a definite advantage. The executive body of the state was the college of Boeotarchs of whom there appear to have been seven at this time (cf. IG vii. 2407, 2408 and now a third inscription: Roesch, 1984: 45ff. = SEG 34 no. 355; 15. 52. 1; 53. 3; Paus. 9. 13. 6f. The seven are perhaps the original eleven minus Orchomenus and Thespiae). How many of these -248were Thebans? We do not know, but as the assembly met at Thebes and the people attending voted as individuals, we may suspect that the majority were invariably Thebans, and indeed it probably often happened that all were from Thebes; unless of course there was a rule which allocated Thebes a certain number of Boeotarchs and the rest had to come from elsewhere. But we are totally in the dark here. Sordi, 1973: 82, thinks that Thebes reserved for herself all the Boeotarchies. (Does the fact that the lists from the 370s and 360s -- the inscriptions and Pausanias cited above -- give no ethnika signify that the named were all from Thebes?) For a detailed discussion of the little we know about the workings of the Confederacy, see Buckler, 1980a: 23ff. See also Roesch, 1982: 259ff., 287ff., 297f., whose concern nevertheless is with Hellenistic Boeotia.

28. 2-4. The foundation of the Second Athenian

Confederacy

The date

D's setting for the foundation of the Second Athenian Confederacy requires no detailed defence. Busolt pointed out long ago that it is erroneous to separate the Boeotian and Athenian Confederacies in time and to treat each in isolation (1873-5: 679ff.). The anti-Spartan group in Athens, chiefly the imperialists, and those who befriended the Theban exiles, seem to have largely coincided, as literary and epigraphic evidence indicates (cf. Dinarch. 1. 38f.; Aesch. 3. 138f.; Plut. Mor. 575d-f; Svt 248 (=Tod 118) lines 39f.; Svt 256 (=Tod 121) line 20; Svt257 (=Tod 123) lines 76f.; Svt 259 (=Tod 124) line 7). While their Theban protégés freed Thebes and resuscitated the Boeotian Confederacy, these people (the ἀἑιολογώτατοι in Athens: 28. 2) put into effect plans for the foundation of the Athenian Confederacy. Certainly, the Confederacy is not sufficiently explained as the Athenian reaction to the aquittal of Sphodrias. D's sequence of events received further support from Burnett, 1962: 1ff., and from Cawkwell, 1973b: 47ff. Cawkwell sees the formation of the Confederacy as a defensive measure on the part of an Athens which felt very nervous in the first three months or so after the liberation of Thebes. It was founded well within the terms of the King's Peace as indeed appears from D's account: the -249Athenians appealed to the Greeks to defend their 'common freedom' (28. 2). While the defensive nature of the Confederacy is obvious ( Svt 257 (=Tod 123) lines 9f.), it should be added that Athens cannot have been motivated entirely by fear or altruism. She had never ceased to dream of restoring her 5th-century empire (cf. Isocr. 4. 20; X. H. 3. 5. 10; Andoc. 3. 15). It was not a question of if, but of when and how, and now, after the liberation of Thebes, with Spartan unpopularity at its zenith, and with the Thebans deeply grateful to, if not in fact already allied with, Athens (see 28. 5n.), many Athenians must have seen a chance of taking the first steps at least in that direction by championing the freedom and autonomy promised Greece by the King's Peace. The raid of Sphodrias is a good indication that something was afoot at Athens. If, as Xenophon represents (H. 5. 4. 34), the Athenian attitude towards Sparta was one of relative goodwill till the raid of Sphodrias and it was the ßοιωτιáζονιες who, taking advantage of Sphodrias' acquittal, turned the people against Sparta, then Sphodrias' action makes no sense (cf. Busolt, 1873-5: 683 n. 1). Sphodrias' very objective, the capture of the Piraeus, points to the foundation of the Confederacy having been set in motion. Without the Piraeus there could be no Athenian sea-alliance. The Second Athenian Confederacy, therefore, was founded in the early months of 378, certainly before the raid of Sphodrias (see also 29. 5-6n.). (The Attic honorific decree dating to 379/8 published by Pritchett, 1972: 164ff., however restored, must refer to an alliance of some sort. But, contrary to what Pritchett thinks, it does not support an early date for the foundation of the Confederacy: Kallet-Marx, 1985: 138 n. 48.) It goes without saying that, though founded at this time, the foundation must have been on ground already prepared. A new Athenian alliance was far from welcome to Sparta and though feelers must undoubtedly have been put out to friends in the Aegean before the liberation of Thebes, these will hardly have been publicized. But bilateral alliances such as that with Chios (made in summer 384, following the destruction of Mantinea) were intended for mutual defence against Sparta and were at pains to stress the two states' strict adherence to the terms of the King's Peace; and this in fact was also the case with the new Athenian League. Though the direct outcome, therefore, of the momentous events of winter 379/8, the League had deeper -250roots and its establishment cannot have come as a surprise to anyone. This is as far as the evidence allows one to go. The suggestion by Kallet-Marx ( 1985: 133ff.) that the Athenian invitation to the Greeks to join together in a new league (28. 2) dates to late summer or autumn 379 goes beyond the evidence. The suggestion is based on the assumption that there was simply not enough time between the campaign of Cleombrotus and the raid of Sphodrias for the diplomatic manœuvres mentioned by D in 28. 2. Therefore the process must have been initiated earlier than the winter of 379/8, a possibility strengthened, he thinks, by the likelihood of a temporal overlap at 27. 3-28. 2. But the two to three months between the winter campaign of Cleombrotus and the raid of Sphodrias (29. 5-6n.) are surely ample time for the actual launching of the alliance. Communication by sea in the relatively small area of the Aegean was reasonably fast, even in late winter/early spring, and the only thing that can be said with any degree of certainty about 28. 2-3 as far as Athenian affairs are concerned is that it looks forward in time (see next n.). As for the supposed overlap at 27. 328. 2, to suspect one is not the same as to prove it. Not all narrative sequences in D are overlapped and firm indications are needed before one is posited. 28. 2-3. The two sections should be seen as an indifferent abridgement of what may have been a narrative of some length in Ephorus dealing with the origins, creation, and successes of the first few years of the Confederacy; this last theme, as was the way with Ephorus (p. 123 ), in the form of a general and rhetorical synopsis which anticipated the detail of later narratives. Thus viewed, surprising elements in the passage become intelligible. Πῤῤςßεις τοὺς ἀξιλεγωτáτονς. . .ἐξΠμφαν probably indicates not just the diplomatic activity initiated early in 378 which resulted in the foundation of the Confederacy, but also that of the following years, in particular of spring 377 ( Svt 257=Tod 123; 30. 1f. and nn.). πολλί πό πἴοσέΘεντο τοῖσ 'AΘðναίοτς clearly points to the future for only six states joined Athens in 378 (cf. Accame, 1941: 28f.). The call to defend freedom and autonomy against Sparta will have reached Sparta's allies too, especially those outside the Peloponnese (έπί τàς ὐπò τοὺς Λακεδαιμονίονς τεταγμένας πóεις). We know that many of these joined Athens by 375 (διóπεῤ Πολλοὶ -251. . . ἀπέκλινον Πρò ιοὐς 'AΘðναὶονς. See nn. ad 30. 1, 5; 31. 2; 34. 4; 36. 5). We may compare Isocr. 14. 15; Plut. Pel. 15. 1: καὶ περιîóντες έδέχονιο καί Πρσἡγοντο τοὐ σ ἀποστατικὡσ τὡν 'Eλλήνων Ĕχοντασ. Strictly speaking, D is wrong of course when he says that the first to 'break away' (Πρòσ ταστ ἀπóσταστν) from Sparta

were Chios and Byzantium, but this is an understandable consequence of his careless abridgement of Ephorus. It is possible that Ephorus had spoken of the apostasis of Greek states in general, with Chios and Byzantium setting the trend, in the sense that they openly now repudiated Sparta's role as prostates of the Peace. 28. 2. 'AΘðναîοτ σέ πρέσßεισ κιλ : we have no reason to query the statement that the initiative for the foundation of the Confederacy came from Athens; cf. 28. 2-4n.; Hampl, 1938: 133ff. The appeal to the cities to bestir themselves in defence of the common freedom (ἀνιέχεσΘατ ιńσ κοινńσ ἐλενΘερíασ) echoes the King's Peace and this must have been deliberate. And not only the King's Peace, but it harks back to older struggles for freedom, to the Persian Wars and the Ionian Revolt. Significantly, not only did the decree of Aristoteles put the emphasis on freedom and autonomy (lines io and 20), but the stele bearing the decree was set up beside the statue of Zeus Eleutherios (lines 65f.). Nothing could be more evocative (cf. Cargill, 1981: 115, 132). 28. 3. Xîοι κτλ : the drawbacks notwithstanding, the worth of D's text is shown by the four states he mentions as being the first to join Athens: Chios first and Byzantium, then Rhodes and Mytilene. On the front of the stele of the decree of Aristoteles ( IG ii2 43= Svt 257=Tod 123, and now Cargill, 1981: 14ff.), which dates from Feb./Mar. 377, we find these same four cities along with Methymna inscribed by the same hand as cut the text, immediately below the heading 'AΘðναíων Πóλεισ αἴμμαχοι, and in the following vertical order: Xîοι, Mντιλðναîοι, MðΘνμναîοι 'Pόδιοι, Bνζáντιοι. To the right of the Chians, heading another column and cut by the same hand, is θðßαîοι. The Euboean cities below the Thebans were cut by a different hand ( Fabricius, 1891: 596f.; Cargill, 1981: 33). Thebes, we know (29. 7), was admitted to the Confederacy after the outbreak of war in spring 378. It was not therefore one of the founding cities and neither was Methymna as the epigraphic evidence shows (below). On the other hand, both of them were probably members and certainly allies of Athens before the close of the year of Nikon ( 379/8). The Confederacy -252was created by Athens in close consultation with the Chians, Rhodians, Mytileneans, and Byzantians. The decree of accession of Byzantium survives ( Svt 256=Tod121= IG ii2 41) and it seems to have been passed when the Confederacy was still at the planning stage; there was no synedrion as yet. All four cities therefore may be assumed to have formed the first and founding members of the Confederacy ( Accame, 1941: 36). It may further be assumed that an oath similar to that of the Byzantians was taken by the other three states ( Svt 256 lines 4-6: ἔναι Bνζα[ντíζσ 'AΘðνxαäων] σνμμáχοσ κ[αì τω+̑ ἂλλων σν]μμáχων). The Chians almost certainly went first and set the pattern (see next n.). The occurrence of the four names in D probably reflects the fact that these four states were known to have been the original members. Isocr. 14. 27f. comments that while Thebes deserted Athens at the conclusion of the King's Peace, Chios, Byzantium, and Mytilene remained loyal (σνμπαρέμειναν). He does not include Rhodes perhaps because that state had been pro-Spartan before the Peace. We may therefore accept (with Accame, 1941: 33) D's order, Chios, Byzantium, Rhodes, Mytilene, as the actual order of accession of those cities. Nor is such an acceptance contradicted by the fact that the order in the decree of Aristoteles is different and it includes Methymna in third position, for there appears to have been an earlier list of the original members to which Methymna was added in fifth place, with Thebes in all probability bringing up the rear ( Svt 258 (=Tod 122), the accession of Methymna, lines 8ff.: αναΛµα⊤ α⋮τÓν Λ¶κπ τνσ ολÐσ ωσΘερ κα∞ οÓ αλλοι σνμμαχωι αναγραμμενΟι σιν; cf. Cawkwell, 1973b: 50. The reference surely is to a stele set up for public viewing and not to an archival list, as Cargill, 1981: 107 n. 24, thinks). Thus the position stood for nearly a year till spring 377 when the Confederacy was placed on a broader footing and advertised to the Greek world in general. On the stele of the decree of Aristoteles the original six members were inscribed in a geographical order it would seem, beginning with Chios which is nearest to Athens and ending with Byzantium which is furthest away. Thebes naturally enough was inscribed by itself to the right of Chios. χιοι: it would appear that Chios was already allied with Athens before the King's Peace and even before Thrasybulus' expedition in 389 (14. 94. 4; cf. Nepos, Con. 5, and IGii2 23, an Attic decree in honour of a Chian voted in 388/7). In the summer of 384, in the -253wake of the dioikismos of Mantinea, the alliance was renewed within the framework of the King's Peace ( Svt 248=Tod 118), and when early in 378 Athens called upon the Greeks to defend their common freedom, the Chians must have been the first to respond and the first to be admitted to the newly conceived koinon synedrion. But here there is a difficulty. When the alliance between Byzantium and Athens and her allies is specified to be καΘαπεð XÍοισ ( SVt 256=Tod 121 line 7), and when prospective members of the Confederacy are promised in 377 to be allied ἐΦ οισσΠερ Xιοι ( Svt 257=Tod 123 line 24), which Chian alliance is meant? The common assumption is that Svt 248=Tod 118 is meant (cf. Tod, p. 66 ; Burnett, 1962: 10; Buckler, 1971a: 506f. Cargill, 1981, does not tackle such questions). Can that be so? Accame, 1941: 34f., thinks not. Svt 248=Tod 118 was a symmachia between Athens and Chios alone, whereas the symmachia of Svt 256 (=Tod 121) line 7, Accame thinks, must have been between Chios on the one hand and Athens and her allies on the other, and it must have been concluded shortly before that of Byzantium (see previous n.). It must have contained specific definitions of the duties of the two sides which were subsequently to apply in the case of Byzantium and indeed the other allies also. τðν Ωε σóóΩÉΠΠÉ Π XÍοις thus saves time and effort by referring to the Chian treaty. Accame further proposes that the new type of alliance between Athens and Chios which superseded the old, forming the basis of those alliances ( Byzantium with Athens and the other allies', ' Rhodes with Athens and the other allies', etc.) which were concluded within the embryo framework of the Confederacy and out of which the synedrion grew, survives in IG ii2 35. This is commonly taken to be a duplicate of IG ii2 34 ( Svt 248 = Tod 118) and in fact it reproduces it word for word from line 9 to line 24. But the horkoi section of IG ii2 35 is not preserved and that is where the two inscriptions differed, Accame suggests. While in IG ii2 34 the Athenians and the Chians pledged themselves to reciprocal aid in case of attack, in the second treaty of 378, apart from the reciprocal aid, the Athenians pledged themselves to abide by the decisions of the koinon of the allies which they were then creating, and the Chians by those of the Athenians and the koinon. We may compare

Svt 263=Tod 127, the alliance treaty between Athens and Corcyra of 375. -254Whether or not Accame is correct about IG ii2 35, his proposal that a series of alliances on a new footing heralded the creation of the koinon synedrion is eminently plausible and I accept it, with one qualification: these alliances, as I argue below (30. 2n.), were bilateral treaties with Athens, but formed now within the framework of the League. B↑ζαντιοι: see previous nn. The beginning of the surviving part of Svt 256=Tod 121 does not seem to refer to a previous alliance between Athens and Byzantium ( Accame, 1941: 16f.: see below). The sentence probably refers to mere friendship. We have already argued that the decree comes from the very origins of the Confederacy. και με↑τσ. . . M⇑τινναÍοι: see previous nn. For Rhodes before the King's Peace see Accame, 1941: 15f. Like Chios, Mytilene appears to have been allied with Athens already at the time of Thrasybulus' expedition in 389 (14. 94. 34; X. H. 4. 8. 28) and friendly relations continued after the King's Peace (Isocr. 14. 28). και των αλλων τινσιωτων: there is no need to assume that this is a reference to Methymna (cf. Cawkwell, 1973b: 51). The phrase should be seen rather within the context of the generalizing statements in sections 2-3 (see ad loc.) about the growth of the Confederacy. The decree of accession of Methymna ( Svt 258=Tod 122) shows that by that time the Confederacy had been organized; the synedroi of the allies were involved in the oath-taking. Methymna joined after Chios, Byzantium, Rhodes, and Mytilene, but perhaps before Thebes (28. 3n.) In seeking to establish that the Methymnaean alliance was later than the Byzantian ( Svt 256=Tod 121) Cawkwell, 1973b: 50f., points to the five Athenians listed at the end of the latter inscription who were elected to receive the oaths of the Byzantians. The Methymnaeans on the other hand simply swore to the synedroi of the allies as well as to the Athenian generals and hipparchs. 'Thus the Byzantian alliance belongs to an earlier phase of procedure more akin to the normal before the Confederacy was founded' (1973b: 51). The argument is untenable. The five-man commission occurs afterwards too in accession treaties; see Svt 304=Tod 153 lines 4ff.; IG ii2 230=Svt 340 lines 13ff. The real difference between Svt 256 and 258 in this respect is that the boule was not involved in the oath-taking in 258 because Methymna was already an ally (30. 2n.). -255ο δε δÐ⇁οσ μετÁωρισΘεεισ: the foundation of the Confederacy. It occurs in no other literary source and the value and basic reliability of D's account have already been demonstrated by the discussions above. The details he gives of the constitution of the Confederacy are confirmed by epigraphy and, to a certain extent, by the later 4thcentury evidence of Aeschines and Demosthenes. The initiative for the foundation must have come from Athens, as D says. But she would not have moved unless she felt certain of a favourable reception by the cities approached. Once that condition was realized, the next step could be taken. We may assume that the envoys sent out (28. 2) were directed to communicate to the cities visited, Chios, Byzantium, Mytilene, and Rhodesfirst and foremost, all of which were allied with Athens already or were to be so very soon, the Athenian plan for creating a union of likeminded states to defend their freedom against Sparta. They were also instructed, we may further assume, to request the sending of ambassadors to Athens empowered to deliberate with the Athenian demos about the formation of such a union. This is too simplistic perhaps, and there may very well have been a series of embassies to and fro between Athens and the cities concerned. At any rate the Athenian demos, elated by the goodwill of the cities, 'established a koinon synedrion of all the allies (thus the founding members were all allied to Athens by this stage) and appointed synedroi of each city'. The establishment of the synedrion therefore was the result of a vote of the ecclesia and this we may accept. The literal meaning of και με ↑τσ⇑τι νναÍοι, however, is hardly credible (though Cargill, 1981: 112, seems prepared to admit it as an early usage of the League). In my view, it should be taken to mean simply that the decree of the ecclesia which established the synedrion specified, naturally enough, that member states should each send envoys to Athens to act as their representatives in the synedrion. The koinon synedrion having been set up, it proceeded in common with the Athenian demos (this is the meaning of εταχΘνδ αΠò τÐς κνωΠò τÐς) to work out the constituting principles and basic procedures of the Confederacy: the synedrion was to meet in Athens, all cities large and small were to have one vote, they were to be autonomous, and they were to accept the Athenians as leaders. These and other joint decisions which D does not mention made Up Πò τÐς των AΘðναιων και ιωσνÁÁαχωνwhich -256governed the Confederacy (cf. Svt 262=Tod 126 line 24; Svt 263=Tod 127 lines 34f.). Whether or not we speak of a 'constitution' of the Confederacy (cf. Accame, 1941:38), the dogmata of the Athenians and their allies which defined and guided the organization should not be seen as not admitting of change or amendment. If conditions demanded it and a dogma of the allies allowed it, for example, a garrison and an archon could be installed in a member state in spite of Svt 257=Tod 123 lines 21ff. The correct interpretation of the promise not to do certain things ΠαÐα τα δóγματα των σνμμαχων ( Svt 259=Tod 124 lines 25 f.) is that if the dogmata allowed it then those things could be done. Cf. Cargill, 1981: 154f. The Athenian motives for founding the Confederacy are not hard to discern (28. 2-4n.). But what did its members stand to gain? Fear of Sparta did of course play its part with certain states, notably Thebes. But this fear should be seen in a wider context, that of the endemic struggle between democrats and oligarchs which had centred on Athens and Sparta since the middle of the 5th century. The Second Athenian Confederacy was largely an organization of democracies (cf. Cawkwell, 1981b: 48). This polarization is clearly attested by the literary evidence, including D (cf. 4on.; 45. 1; Isocr. 4. 16; X. H. 6. 3. 14). And in two instances that we know of, Athens

took measures to guarantee the democratic government of a member state ( Svt 263=Tod 127, Tod 156). Piracy was another reason why many members would want to place themselves under Athens as the Athenian navy could provide them with a degree of protection (cf. Cawkwell, 1981b: 48). κωινòν σ↑νεδριων:for the workings of the synedrion see Accame, 1941: 107ff.; Cawkwell, 1981b: 48ff.; Cargill, 1981: 115ff. Whatever Athens' position vis-à-vis the allies in the case of the synedrion of the Delian League ( Hammond, 1967b: 57ff., esp. 60f., and de Ste Croix, 1972: 303ff., believe she stood outside it), there is no doubt that now Athens had no vote in the allied synedrion (cf. Aeschines 2. 60f.; 3. 69f.; Demosth. 19. 144). Svt 257=Tod 123 lines 41ff., 57ff., shows that the synedrion enjoyed certain legal powers in cases concerning the alliance and its members, but we are very vague about this (cf. Cargill, 1981: 121ff.). It is not at all clear that there was a joint Athenian-allied court as lines 57ff. led Marshall, 1905: 35f., and Accame, 1941: 138ff., to believe. It is more likely that in this as in all other respects Athens and the allied -257synedrion acted separately: Larsen, 1955: 63f.; Lewis, ΘΟρΟΣ Meritt, 89 n. 39; Cargill, 1981: 12ff. σ↑νεδðο↑ς: the general view is that while each member state had one vote it could send to Athens as many synedroi as it wished. Some states sent only one synedros ( Svt 304=Tod 153 lines 8f.); others sent several (Tod 131 lines 26ff.). As this is all the evidence at our disposal, Accame, 1941: 109f., by 'correcting' Tod 131 is able to insist that all states had but one synedros. It may well be doubted nevertheless whether such an emendation is legitimate; cf. Larsen, 1955: 55; Cargill, 1981: 113f. 28. 4-5.οι δε ðο↑ςγóνιοι τνΜ óðμðν κτλ : the language is clearly D's; cf. 16. 67. 1; 18. 33. 5. But equally clearly it is a brief paraphrase of an Ephoran passage which outlined the measures taken by Sparta to counteract the success Athens was having with her appeal to the Greek world. A certain amount of distortion in such cases was inevitable (compare 12. 76. 1-2, D's version of Ephorus' rewriting of Thuc. 5. 30ff.). All the same, we would not be justified in taking the statement to represent such a complete volte-face on the part of Sparta as to be unhistorical (thus Westlake, 1986: 269, who is further tempted to 'account for Diodorus' new appraisal of the situation by adopting the reasonable hypothesis that Ephorus, turning from a source violently hostile to Sparta to one largely sympathetic, postulated a modification of Spartan policy in order to avoid glaring inconsistency'). The Spartan measures, understandable enough in the circumstances, consisted of (a) paying more and better attention to their allies, and (b) looking to their war machine. For the results of (a) and for what (b) actually involved we have to wait till 31. 1 when D, doubtless reflecting Ephorus' own arrangement, takes up the Spartan narrative again. Meanwhile he leaves Greek history altogether and turns to eastern affairs. It would therefore be mistaken to see 28. 4-5 and 31.1 as instances of overlapping narrative. 28. 5. σ↑μμαχονων. . . σ↑νεδðιο : Cawkwell, 1973b: 58, does not see this as evidence that a treaty with Thebes existed prior to the raid of Sphodrias: 'The clause can have a conditional force' (58 n. 3). It can, of course, but this is rather a desperate and indeed unnecessary solution. Cawkwell's denial, rather surprising in view of the fact that on the whole he prefers D's version of events to that of Xenophon, stems from his belief that the King's Peace excluded such an alliance, and that it was only after Athens put aside the -258-

Peace that an alliance was concluded. The Peace, however, was not put aside by Athens (29. 7n.), and it is certainly possible that, as D says, an alliance existed between Athens and Thebes before the raid of Sphodrias ( Burnett, 1962: 2, 15; Cargill, 1981: 58f., who, therefore, should have included this treaty (as he does not) among those of this type ('allies of Athens and the League') he discusses in ch. 4 of his book, designating it (rather than Svt 263 =Tod 127) the first of the kind). How long before the raid? The time of the liberation of the Cadmea when the Athenian army marched out in force to succour Thebes (26. 1-2) is perhaps the most appropriate time, and indeed Plutarch speaks of an alliance between the two states which, however, so he says, the Athenians renounced shortly afterwards when the Spartans invaded Boeotia (Pel. 14. 1). It was only in the aftermath of the raid of Sphodrias, according to Plutarch (Pel. 15. 1), that the Athenians renewed their alliance with Thebes. But Plutarch is probably speaking loosely ( Cawkwell, 1973b: 58. Contra Kallet-Marx, 1985: 144 and n. 75) and reading too much into the text of Xenophon (at H. 5. 4. 19 and 5. 4. 34). For the conversion of the Theban alliance into membership of the Confederacy, see 29. 7n.

29. 1-4. Eastern affairs: Persian and Egyptian preparations for war On these four sections, see pp. 100 f. 29. 1. 'Α+̔+́γρις : see 2-4 n.; 2. 3n. μεγάλοΥς γὰρ μισΘοὺς κτλ: for Greek mercenaries in Egypt see Parke, 1933: 57ff., 105ff.; Kienitz, 1953: 114ff. An Athenian law on silver coinage passed in 375/4 seems to indicate that a large quantity of Attic silver owls was being struck outside Attica at the time, and that this was causing problems in Athens. Egypt has been identified as a major centre for such activity. The object of the Egyptian kings obviously was to have available recognizably Athenian silver with which to pay their Greek mercenaries ( Stroud, 1974: 157ff., esp. 169f.; 92. 3n.). 29. 2. μετεπέμφατο Ξαßρίαν : Chabrias was a general in Athens in the winter of 379/8 (X. H. 5. 4. 14), and he must therefore have been back in Athens by early spring 379 at the latest. His present activity in Egypt consequently belongs between 386, when he presumably evacuated Cyprus (2-4n.), and 380; unless he was

-259ordered out of Egypt by Athens even before that year. But it seems better to assume that the Persians turned their attention to Egypt at the conclusion of the Cypriot War (summer 380), and that that was when Pharnabazus demanded Chabrias' recall. Cawkwell, 1973b: 47 n. 3, suggests that the events in 29. 1-4 could belong to 375. This cannot be. It was Acoris who hired Chabrias (29. 1), and Acoris was dead by summer 380 (2-4n.). Also, the ἔ2τη δέ πλείω wasted in preparations which caused Iphicrates' patience to wear thin (41. 2) require a longer span than the eighteen months or so from late 375 to early 373, when Pharnabazus and Iphicrates finally attacked Egypt (15. 41ff.). For Chabrias' career see PA 15086; APF p. 560f.; Pritchett, 1971-91: ii. 72ff.; Develin, 223. ἂνεΥ τῃ+̑ς τοΥ̑ δήμοΥ γνώμης: 'the insistence is unnecessary and should therefore arouse suspicion. Perhaps it echoes an Athenian disclaimer, issued when Pharnabazus complained' ( Sealey, 1993: 79). Pritchett, 1971-91: ii. 100, maintains that though Chabrias went to Egypt as a private individual, he was not acting contrary to Athenian wishes or interests for was not Egypt allied to Athens? Yet, whether or not Athens sanctioned Chabrias' activities in Egypt, the former Athenian-Egyptian alliance (2. 3n.) no longer existed -- it had lapsed with the King's Peace. In fact Pharnabazus' protest (see below) must have been based on the King's Peace. Cf. FGH115 F105; Nepos, Chabr. 3. 29. 3. φαρνάßαξος : the former satrap of Hellespontine Phrygia. He had been called up to Susa to marry one of the King's daughters and he had then been entrusted along with two other generals with the task of reconquering Egypt. The three-year effort had failed (2-4n.) and we now find him in command of a force preparing for another expedition against Egypt. The impression given by D certainly is that though Pharnabazus had colleagues (15. 42. 4; 43. 1f.) he held the chief command. According to Nepos, Dat. 3. 5, these colleagues were Tithraustes (who had also taken part in the previous campaign against Egypt: Isocr. 4. 140) and Datames, who succeeded to the chief command after the recall of Pharnabazus, presumably following the expedition's failure. Nepos' testimony nevertheless is contradictory for he states that when sent by the Great King to join the army being prepared against Egypt Datames was given equal imperium with Pharnabazus and Tithraustes. And in Chabr. 3. 1 he writes of the praefecti regis Persae as sending envoys to Athens to demand the -260recall of Chabrias from Egypt. Though in Chabr. 2f. Nepos confuses the earlier with the present expedition against Egypt, 3. 1 seems to be the same occasion as related here by D. Ps.Demosth. 49. 28, 35 refers to the King's strategoi.

ἐξὲπεμφε δὲστειλαν: by the King's Peace Greece had conceded Asia and the islands of Cyprus and Clazomenae to the King (X. H. 5. 1. 31). Pharnabazus had every right therefore to demand the recall of Chabrias. The dispatch of Iphicrates to aid the Persians on the other hand was a gesture of goodwill. According to Nepos, Chabr. 3. 1, Chabrias was ordered out of Egypt by Athens on pain of death. 29. 4. 'ΛĴφικράτην . . . ἐξαπέστειλαν: he went out now, in 380/79; see 29. 2n. and 41. 1n. For his career see PA 7737; APF pp. 248 ff.; Pritchett, 1971-91: ii. 62ff.; Develin, 245f.

29. 5-6. The raid of Sphodrias

The date

If it were the case that the tradition found in Aristid. 19. 419 Dind. and Plut. Pel. 14. 6; Ages. 24. 7, that Sphodrias' men had been thwarted by the Eleusinian deities, arose because the raid was made during the truce observed for the Lesser Mysteries (held during the month Anthesterion), we would have a definite date for the raid. But Aristides is adamant that the spondai of Eleusis were never violated. As it is, the raid can only be dated approximately by being placed in the correct sequence of events of spring 378. Why did the raid take place? Because, as has been argued, the Athenian Confederacy was being founded and the raid was an attempt to prevent it (28. 2-4n.). As the diplomatic manœuvres preliminary to the Confederacy had already taken place, the raid could hardly belong earlier than early spring. On the other hand, as Kallet-Marx rightly points out (1985: 136f.), several weeks should be allowed between the raid and the beginning of Agesilaus' Boeotian campaign of 378. End of May is the terminus ante quem for the latter (see p. 287 ) and therefore a date in April is indicated for the former. 29. 5. Σφοδριάδου: Σϕοδρίας according to X. H. 5. 4. 15f. and elsewhere, and Plut. Pel. 14. 3, 6; Ages. 24. 4ff. See p. 138 ἐφ' ἡγεμονίας: he had been left as harmost at Thespiae by -261-

leombrotus (X. H. 5. 4. 15). Typically D omits the relevant background. φυ+́σει . . . προπετου+̑ς: cf. FGH 124 (Callisthenes) F9, εὐήΘης λίαν καί κοϕος τὰς ἐλπίδας, and Plut. Pel. 14. 3; Ages. 24. 4. Plutarch clearly derives from Callisthenes here, but not necessarily D (Ephorus).

ἒπεισεν . . . Πειραιᾳ: it would seem that there were at least three versions current at the time: that King Cleombrotus was behind the raid (Ephorus); that the Thebans were responsible who wanted to embroil Athens in war with Sparta (X. H. 5. 4. 20 Plut. Pel. 14. 2ff.; Ages. 24. 6); that Sphodrias alone was responsible, his motive being the desire to emulate Phoebidas ( Ages. 24. 4f.). Xenophon's explanation should be rejected for if indeed Sphodrias had been bribed by the Thebans ἐπὶ κακῳ+̑ τῃ⌣ πóλεως (H. 5. 4. 30), even if, as Plutarch says, the money and advice were given by a person (or persons) professing Spartan sympathies, Agesilaus, we may be certain, would not have let him off so easily (see further n. below). And one may rightly doubt whether Sphodrias would have dared to take such an action without the instigation and promised support of some powerful group in Sparta (cf. Grote, x. 135 n. 1; Beloch, 1884: 139; id., GG iii. 1. 147; Meyer, GdA v. 379). I am inclined therefore to accept the Ephoran version that Sphodrias had been incited by Cleombrotus to capture the Piraeus. Why?

Because the king and his adherents felt that such an action was required if the Athenian Confederacy were to be nipped in the bud. Agesilaus was trying to achieve the same, but by peaceful means ( Cawkwell 1973b: 55). Macdonald, 1972: 38ff., argues for the complicity of Cleombrotus too. Kallet-Marx, on the other hand, 1985: 150f., sees no need to choose among the various ancient versions, all of which are possible, though none capable of proof. In fact, Kallet-Marx, 1985: 149f., believes that the capture of the Piraeus was not a feasible objective, that Sphodrias was aware of this, but that by raiding Attica he hoped to scare the Athenians into remaining neutral in the coming war against Thebes, at the same time discouraging the growth of the new league. But even if impossible (though see next n.), the capture of the Piraeus might still have been Sphodrias' objective. With Piraeus in their hands, Spartan pressure on Athens would have been infinitely more effective. Xenophon, of course, as Kallet-Marx points out, does not seem to -262have believed that Sphodrias really intended to capture the Piraeus; or so he says (προσποιησάμενος τòν Πειραιᾳ καταλφεσΘαι: H. 5. 4. 20). But Xenophon apparent disbelief in the expedition's stated objective (nowhere does he say that he considered the mission impossible) may well be conditioned by the fact that he held Sphodrias, a political opponent of Agesilaus, responsible for letting Thebes off the hook by unnecessarily embroiling Athens in war with Sparta. And as is usually the case with Xenophon, all the information he gives is chosen so as to establish his claim, in this instance, that Sphodrias was a corrupt, incompetent, and culpable officer. 29. 6. πλείους τῳν μυρίων . . . ἐπανῃλΘεν : Parke, 1933: 84 n. 5, finds the figure 'incredible'. But Cleombrotus had left behind a third of his army and money to hire mercenaries with (X. H. 5. 4. 15). Ten thousand, therefore, is not so incredible a figure for Sphodrias' total force at Thespiae though we may well doubt that all 10,000 were taken along. Xenophon, however, speaks of στράτενμα πί1μολν (H. 5. 4. 21) and we ought to remember that, once taken, the Piraeus would need to be defended against Athens. In fact the size of the force may have been one of the causes of failure. Otherwise, of Plutarch's two reasons for the failure, want of luck and want of courage ( Pel. 14. 6; Ages. 24. 6f.), the first seems to me the more appropriate. Given the lack of maps and clocks it seems hardly fair to accuse Sphodrias of a miscalculation though miscalculate he clearly did. But not by much. The opinion that the mission was impossible, and therefore a pretence (see previous n.), can be countered with the fact that Sphodrias did after all succeed in reaching Thria near Eleusis. One more hour of darkness and it might have been impossible for the Athenians to react quickly enough to save the Piraeus. κατηγορηΘεὶς . . . ἀπελύΘη: Xenophon's story (H. 5. 4. 25ff.) is an apologia of Agesilaus. Contra Tuplin, 1993: 126ff., who concludes that Xenophon included such 'inside' information here 'with the intention of revealing in no friendly spirit what sort of place Sparta really was' (1993: 128). But surely the story was meant to explain Agesilaus' conduct (H. 5. 4. 24) and the explanation is typically Xenophontic (cf. e.g. H. 4. 1. 29 ff.) and one consistent with the mores of Peloponnesian gentlemen. We cannot know whether he himself believed it or simply wrote down what he was told by Agesilaus or some friend of Agesilaus (Etymocles?). He seems to -263have considered the acquittal 'most unjust' (H. 5. 4. 24), and he may have been thankful to have a story which did not of course exonerate Agesilaus, but which nevertheless presented him in a humane light. The Thebans were the evil plotters and Cleombrotus and his circle were guilty of protecting Sphodrias. Agesilaus, on the other hand, had merely given way to the tearful pleas of young lovers. The whole story is nonsense of course and hardly to be credited. Yet de Ste Croix is 'inclined to say that this is the most valuable account we have in any source of the way in which great issues might be settled at Sparta' ( 1972: 134)! The very beginning of the story shows it to be nonsense. Sphodrias, Xenophon tells us (section 26), addressed his son Cleonymus and asked him to intercede with Archidamus. But Sphodrias was not in Sparta at the time. By Xenophon's own admission he had disobeyed the summons of the ephors and had fled (H. 5. 4. 24). The question to ask here is why Agesilaus acquitted Sphodrias, and the answer which readily presents itself is that the king had failed in the end to persuade the Athenians by peaceful means to desist from founding their Confederacy and from lending support to Thebes. It is only a guess, but a reasonable one, that the Spartan embassy in Athens at the time of the raid represents Agesilaus' attempt to dissuade Athens. Etymocles, one of the three ambassadors, belonged to the group of Agesilaus (X. H. 5. 4. 32; Cawkwell, 1973b: 55. Rice, 1975: 111, postulates that the ambassadors were in Athens in connection with the trial of the two Athenian generals. But nothing in the sources connects the embassy and the trial and there are in any case insuperable chronological problems ( Kallet-Marx, 1985). Cawkwell's proposal is far more satisfactory). Having failed, he may well have decided that war with Athens was now inevitable, and his comment that τήν γὰρ Σπάρτην τοιούτων δει+̑σΘαι στρατιωτῳν (X. H. 5. 4. 32) should perhaps be seen in this light. War was coming. Why, therefore, waste a good soldier? For Sphodrias cannot have been bribed by the Thebans (29. 5n.). He was Cleombrotus' man to the last (he fell at Leuctra as did his son Cleonymus: X. H. 6. 4. 14), and his motives had been patriotic (cf. X. H. 5. 4. 34). But Agesilaus will also have had domestic politics very much in mind and it is quite likely that his position at home was strengthened considerably by this decision ( Cartledge, 1987: 137f.).

ἐν τῳ+̑ συνεδρίῳ: the gerousia is meant. The gerousia formed the -264-

highest criminal court at Sparta ( de Ste Croix, 1972: 131ff.). Synedrion could denote a court (cf. Lys. 9. 6) or a council-chamber (X. H. 2. 4. 23). At 11. 50. 2, 6 D correctly calls the Spartan body 'gerousia', doubtless retaining the Ephoran term. But D, like other Greek writers of the Hellenistic and Roman periods, did not generally distinguish between the terms boule, synedrion, gerousia, or syngkletos when speaking either of the Greek boule or of the Roman or Carthaginian senates; cf. 4. 83. 7; 14. 113. 6f.; 16. 65. 6ff. This is true even of Polybius; cf. 1. 11. 1; 1. 20. 1; 1. 21. 6 (with HCP i ad loc.); 1. 31. 8; 10. 18. 1; etc. In fact D may be following Polybian usage here.

29. 7-8. The outbreak of war In the absence of any other evidence one might assume that all the acts listed here by D were the immediate consequence of the decision to go to war. But section 8, ἐφηφίσαντο . . . ἐκτòς τη+̑ς Ἁττικη+̑ς, is clearly one of the provisions of the decree of Aristoteles of spring 377 (Svt 257=Tod 123 lines 25ff.), and the introduction of Thebes to the allied synedrion may not be as early as the late spring/early summer 378 though of course she was a member by spring 377 (28. 3n.). The military preparations nevertheless must have followed immediately upon the decision to fight. What we seem to have here therefore is D's compressed and not entirely understood version of an Ephoran chapter which recorded the measures taken by Athens in the year following the outbreak of war to improve her military situation as well as her reputation in Greece; 30. 1 dates to spring/summer 377. 29. 7. λελυ+́σΘαι τάς σπονδάς ὐπò Λακεδαιμονίων : not many would agree with Cawkwell that the vote meant that the Peace was now dissolved and that the declaration that Sparta had broken the Peace meant in effect that Athens considered the Peace to be 'null and void' (1973b: 54; id., 1981a: 75 f.). In 369/8 Athens praised the Mytileneans for their services in 'the past war' ( 378-371 or 373-371) and stressed that the Athenians had fought for the -265-

earlier Isocrates made the suppliant Plataeans say that Athens had fought the last war ( 378-375) ὑπέρ τῳν παρέ τοὺς ὄρκονς καὶ τὰς συνΘήκας τῃς αὐτομίας ἀ ποστερουμένων (14. 17). Later in the same speech the Plataeans entreat that Athens should be ready to make war on all, Thebes not excepted, in defence of the peace treaty; otherwise her claim to defend the King's Peace against Sparta would ring hollow (sections 42-4). And cf. Aristid. Panath. 293

Athens and Chios it was made abundantly clear that both sides adhered strictly to the King's

Peace. They would not contravene the Peace, even if others did (i.e. Sparta who had recently broken version of the treaty with Chios was made the basis for the alliances which resulted in the Second Athenian Confederacy (28. 3n.). A year after the outbreak of war the decree of Aristoteles (Svt 257=Tod 123) promised prospective members of the League (though not those 'who belonged to the King': lines 17f.) that they would be allied to Athens on the same terms as the Chians (lines 23f.). What emerges is that far from denouncing the King's Peace Athens actually presented herself as its new (and sole) prostates (cf. Accame, 1941: 48ff.; Seager, 1974: 45ff.). Accame's restoration therefore of Svt 257 lines 12-14 (1941: 51), a promise to maintain the Peace for ever, though not universally accepted, may well be correct. It is in any case difficult to see what else could have been stated there. στρατηγοὺς τρει+̑ς τρει+̑ς εἴλοντο: not necessarily elected to the generalship (p. 133 ), but rather entrusted with the conduct of affairs visὰ-vis the Confederacy and the war. Cf. Sealey, 1993: 58. ΤιμóΘεον: the son of Conon and good friend of Isocrates (cf. Isocr. 15. 101 ff.). For his career see PA 13700; APF pp. 506ff.; Develin, 223. χαßρίαν: see 29. 2n. Καλλίστρατον: for Callistratus see Beloch, 1884: 145ff.; Sealey, 1956: 178ff. and 187 for the election of both Chabrias and Callistratus in this year; PA8157 and addenda; APF pp. 277ff.; Develin, 223. -266-

ἐφηϕίσαντο . . .διακοσίας: this clearly corresponds to X. H. 5. 4. 34καί ἐκ τούτου οι+̕ ÀΘηναι+̑οι ἐπύλωσάν τε τòν ̑ τε ἐναυππηγου+̑ντο, τοι+̑ς τε Βοιωτοι+̑ς πάσῃ προΘυμίᾳ ἐßοήΘουν.Polyb. 2. 62. 6 records that the Πειραιύ, ναΥ+ς Athenians ἐκατòπηγου+̑ντο, τοι+̑ς τε Βοιωτοι+̑ς πὰσᾳ ἐßοήΘουν ἐκατòν δ' ἐπλήρουν τριήρεις. D's and Polybius' figures

are thus in conflict. Grote, x. 151 n. 2, sought to resolve this by assuming that while D's reference is to the vote to man 200 triremes and levy 20,000 hoplites, Polybius' is to what was actually sent out. This is not very convincing. More to the point is his suggestion that we must not be misled by D into thinking that these resolutions were taken by the Athenian demos alone, but we must regard them rather as joint decisions -- the Athenian demos and the synedrion of the allies. Such an assumption would make D's high figures more palatable and explain perhaps why Polybius' figures are exactly half those of D. The Confederacy was a 50: 50 partnership and the allies agreed to provide as many men and ships as the Athenians.

D omits to mention that to facilitate her war effort Athens looked to her finances. The 5th-century property tax, the eisphora (see Griffith, 1977: 3ff.; Sealey, 1984: 77ff.), was revived, and a new method of taxation, the symmories, was introduced in this connection (Demosth. 22. 44; FGH 328 F41, 45, 46 with nn. The year of Nausinicus, 378/7, of the sources should be understood as the year when the system began to operate, not as the year when it was devised). Polyb. 2. 62. 6f. says that the valuation of all property in Attica amounted to 5,750 talents, a figure confirmed by Demosth. 14. 19 where 6,000 seems to be a rounding off of the figure given by Polybius. For the valuation cf. HCP i. 268f. The method of collecting the eisphora is a crux. Cf. de Ste Croix, 1953: 30ff. (contra MacDowell, 1986b: 438ff.); Thomsen, 1964 (but with de Ste Croix review in CR 16 ( 1966)); Rhodes, 1982: 1 ff. D also overlooks the syntaxeis which were probably paid by the allies from the beginning (cf. Busolt, 1873-5: 703 f.; Accame, 1941: 131ff.; Mitchel, 1984: 23ff.). Cawkwell, on the other hand, 1963: 91ff.; id., 1981b: 51 (cf. Sealey, 1993: 64f.), finds it difficult to believe that the Athenians could have been so cynical as to introduce regular contributions in money right at the start and argues that syntaxeis were not introduced till 373 when they are

mentioned for the first time (ps.-Demosth. 49. 49). However, the existence of a common fund of the allies in 377 (Svt 257=Tod 123 lines 45f.) is perhaps an argument for pushing back the payment -267of syntaxeis to the very beginnings of the Confederacy. Also, if the assessment of the syntaxeis was done by the synedrion itself (cf. Tod 175 lines 27f.), that would have marked a real difference from 5th-century practice. Wilson's investigation of Athenian military finances from 378 to 375 (1970: 302ff.) in no way supports Cawkwell's thesis. Many of Wilson's figures can be halved or doubled, which renders pointless the whole exercise. On syntaxeis see Cargill's discussion, 1981: 124ff., who nevertheless ignores Svt 257 lines 45f. and does not concern himself with the date of their introduction.

ὁπλίτας . . . δισμυρίους: it would be possible to make a case that Athens had 20,000 front-line and reserve hoplites at

this time (cf. Hammond, HG 663 f., who accepts D's figures). In 369 according to D (63. 2), Athens dispatched 12,000 men to help Sparta, and if the reserve hoplites are added to these, the 20,000 figure no longer seems exaggerated. Nevertheless, the normal Athenian expeditionary forces in the 4th century were much smaller, about half the size of that given by D for 369 -- 5,000 in 378 (26. 2; 32. 2) and 352 (16. 37. 3), 6,000 in 394 (X. H. 4. 2. 17) and 362 (84. 2). Did 369 represent an extraordinary effort and were the 12,000 the full levy of Athenian front-line hoplites? But better sense is made perhaps of the 20,000 figure if it is seen as including the field levies of the allies. Certainly the member states were obliged to render military service (Svt 257=Tod 123 lines 46ff.; SVt 263=Tod 127) though, with the exception of Thebes, they are not mentioned as taking part in any of the land operations.

ἰππει+̑ς δέ πεντακοσίους:this is in line with other figures we have for the Athenian cavalry in the 4th century (see the references above). None of the islanders could presumably contribute cavalry. Thebes possessed a sizeable force (26. 4n.), but the estimate here, made right at the outbreak of war, may well pre-date Theban entry to the Confederacy (cf. n. below). ναυ+̑ς . . . διακοσίας: cf. Plut. Pel. 15. 1. Again, the assumption that the contingents of the allies are included makes the figure easier to accept. At 16. 21. 2 D says that 100 triremes were manned by the Chians, Rhodians, Byzantians, and their allies in 356, and a projected figure of 100 triremes for these maritime states (and Mytilene and Methymna) in 378 is not unreasonable, while the Piraeus dockyards held approximately 100 triremes when war broke out ( IG ii2 1604 as now re-dated by Clark, 1990). -268-

καταλÉἐξαι: see 26. 2n. προσελἀßοντο . . . πὰσιν: Thebes may already have been the ally of Athens and the Alliance (28. 5n.), and her introduction to the Confederacy may or may not have followed immediately upon the outbreak of war (29. 7-8n.). Nothing more can be added except Cawkwell's suggestion (1973b: 49f. that the honouring of a trierarch in IG ii2 40 may indicate the sailing season of 378, but after the breakdown of the Peace. The significance which Burnett sees in IG ii2 40 ( 1962: 3ff.), the first four lines of which are probably the end of the treaty of accession of Thebes, is illusory; cf. Buckler, 1971a: 506f. The stone is too fragmentary for any firm conclusions to be drawn, and Accame's use of it (1941: 38ff.) to cast doubt on D's evidence that Thebes joined after the synedrion had been set up is unjustified (cf. Cawkwell, 1973b: 49f.). I agree with Buckler (1971a: 506f.) and Cawkwell ( 1973b: 48) that Svt 257=Tod 123 lines 23ff. ἐπὶ Δἐ τ[οΣ] αντοισἐΦ οῖσπερ χιοι και ∞+ ńßαῖοι, does not mean that the Theban treaty was significantly different from the Chios treaty and an improvement on it (as Burnett, 1962: 3ff.), though I disagree with both that the Chios treaty here is that of 384 (see 28. 3n.). ἐπὶ Δἐ τοΣ ἹαοιΣπ πὰσιν shows that Thebes was a full member like everyone else, with the same rights and duties, including the obligation to contribute ships (cf. ps.-Demosth. 49. 14ff., 21). 29. 8.Ǳ0ΨńΦὶσαντο ... ÁττικᾶΣ : a clear reference to Svt 257=Tod 123, the decree of Aristoteles, lines 25ff. and 35ff. The decree mentions ἐγκτἠματα (as does Isocr. 14. 44), both public and private, and further specifies that any unfavourable stelai in Athens concerning would-be allies were to be destroyed. The stelai, we may guess, mentioned such ἐγκτἠματα as well as the former cleruchies of Athens (in 355 many Athenians still thought of regaining these possessions: Isocr. 8. 6). D's (Ephorus') remarks, therefore, are not inappropriate and together with the decree of Aristoteles they testify to the degree of offence caused in the Aegean by Athenian land expropriations in the previous century. νóμον ἔεντο is technically wrong, but cf. Cawkwell, 1973b: 48 and in general Cargill, 1981: 146ff. For the importance of private overseas possessions for many Athenians in the 5th century cf. Lewis, Studies Ehrenberg, 182; Andrewes, HCT v. 111 f.; Davies, 1981: 55ff. But see also de Ste Croix, 1972: 43f.; 1981: 604 n. 27. -269-

Burnett's view (1962: 11, which Cargill, 1981: 60, finds 'cogent') that the 'yoking together' in D of the abolition of the cleruchy system and the entry of Thebes to the Confederacy means that the two were contemporary, and that the first was cleverly designed by Athens in order to persuade her island allies to accept continental Thebes in the Confederacy, should be rejected. The fact that the two events are next to each other in D does not mean that they are contemporary, and in fact the admission of Thebes probably belongs to 378 (29. 7-8n.) while the decree of Aristoteles dates to spring 377. The purpose of lines 25ff. of the decree is made clear by both D here and Isocrates (next n.). Σιἀ Σἐ ταúτńσ ...ἥγεμονίαν cf. Isocr.14. 44 καì τῳ̑ν μενἐγκτńμáτων τῳ+ ν úμετἐρων ἀπἐστńτε, ßονλóμενοι τUY+1F75ν σνμμαχίαμὡδ μεγίστńν ποιῃσαι

30. The expansion of the Second Athenian Confederacy See 28. 2-3 n. 30. 1. πολλαι μεν . . . 'Éστιαίασ : immediately below the 'Thebans' on the front face of Svt 257=Tod 123 are found inscribed χαλκιδῃσ, Éρετρῃσ, áρεΘονσιΟι , Καρνστιοι, , 'ÍκιΟι, all five by the same hand, but a different one from that which cut the main text and the first six members of the Confederacy ( Fabricius, 1891: 596f.; Cargill, 1981: 33). The logical assumption is that all five were not members of the Confederacy in Feb./Mar. 377 when the decree of Aristoteles was passed; they joined some time afterwards, but before the expedition of Chabrias to Euboea in the same year, for by then the Euboean cities were already allies (30. 2). This conclusion is inevitable unless one is prepared to assume that the stonecutter who inscribed the text of the decree and the first six members was suddenly taken ill or dismissed. The decree of alliance between Athens and Chalcis which accompanied the city's accession to the Confederacy has survived ( Svt 259=Tod 124). It offers no contradiction. It was passed in the year of Nausinicus ( 378/7) and Chalcis consequently joined between Feb./Mar. and midsummer 377. An apparent copy of Svt 259=Tod 124, IG ii2 155 ( Schweigert, 1938: 626), may in fact be the decree of alliance between Athens and another Euboean city, either Eretria or Carystus. The following three points must be stressed in connection with the accession of the Euboean cities. -270-

First, the epigraphic is in fact the only evidence that the accession of the Euboean cities post-dated the decree of Aristoteles. D's text constitutes no such evidence (pace e.g. Accame, 1941: 70; Tod, p. 71 ). It has already been argued that the contiguity in D of the accession of Thebes (29. 7) and the decree of Aristoteles (29. 8) does not mean that the two were contemporary (29. 7-8n.). Equally, the fact that the accession of the Euboean cities follows on the decree of Aristoteles is no proof that the two events are necessarily meant to be in chronological sequence in the text. They are not, as the first sentence of 30. 1 shows: the εἰρńμεμń αίτία a because of which many cities were led to fall away to Athens is that stated at 28. 2, i.e. Sparta's arrogant and oppressive misrule. διóΠερ Πολλοὶ . . . αΠεκλιον ΠρòΣ τονΣ 'ÁΘńναιονδ, he says at 28. 2. Section 30. I looks back to that statement. Second, the entry 'Áρε ονσιοι between the Eretrians and the Carystians probably does not refer to a city in Euboea, as Meritt, 1963: 1f., and Cargill, 1981: 33 and n. 17, think. Eurip. Iphigeneia at Aulis170, cited as evidence by Cargill (and accepted as settling the dispute by Sealey, 1993: 284 n. 40), is a reference to the wellknown spring and not to a town. The entry should rather be identified with Mygdonian Arethusa in south-east Macedonia ( Accame, 1941: 72f.; Cawkwell, 1981b: 42 and n. 16). Situated some 6 miles from the coast it may well have belonged to τὴν δ' 'εΠὶ Πᾳσιν tenth division of the reorganized Peloponnesian League (31. 2). If so, it will have come under Spartan control during the Olynthian War and its present early, and surely voluntary, accession to the Confederacy will have amounted to apostasis (28. 2-3 and n.). The Euboean cities which joined the Confederacy at this point therefore were Chalcis, Eretria and Carystus -- Icus is an island near Peparethus. It is possible that Icus, like Sciathus and Peparethus, was under Spartan control prior to 377 (30. 5). If so one more candidate is found for apostasis at this early date. Third, it would seem that all five cities acceded voluntarily as indeed D says (προ νμóτατα σννεμαχńσαν), and it is unnecessary to speculate on the basis of Plut. Mor. 350f whether Timotheus was or was not responsible for their accession. (In any case Plutarch is probably referring to 357, pace Sealey, 1956: 188; id., 1993: 59 and n. 40, and he appears to go back in time: 357, 376, 394.) Nor is there a conflict between D and Plutarch here. D is not saying that Chabrias, or anyone else for that matter, liberated Euboea, but -271-

that the Euboeans freed themselves. Arethusa certainly cannot have joined as a result of an Athenian expedition in western Thrace in early 377 (cf. Cawkwell, 1981b: 42f.). Cawkwell furthermore makes the welcome point that many if not most of the cities entered on both the front face and the left side of Svt 257= Tod 123 will in fact have acceded voluntarily. ΛακεδÁιμονŷΩν . . . Σπαρτιἀταδ: úΠò ÁΘńναŶΩν is often used in D instead of Λακεμńμἑνń for stylistic variation; cf. 15. 30. 4; 31. 3f; 32. 1; 33. 1; 46. 1; 50. 2; etc. Occasionally the two terms are used technically as at 11. 4. 5. For the terms see HCT v. ad 8. 22. 1. πεπολĒμńμǰνń . . . úΠò ÁΘńναίΩν: Hestiaea had been destroyed by Athens following the suppression of the Euboean revolt in 447/6 and the place was colonized with Athenian cleruchs (12. 7; 12. 22. 2; Thuc. 1. 114. 3; 7. 57. 2; 8. 95. 7; Plut. Per. 23. 4). Theopornpus ( FGH 115 F387) says that the Hestiaeans were allowed to depart to Macedonia. Cf. HCT i. 345 ff. On the collapse of Athens in 404 and the expulsion of the colonists the original inhabitants were presumably allowed by Sparta to return. The evidence for this is the present passage of D. The hatred felt towards Athens by the Hestiaeans proves that they were the descendants of the people who had been expelled. D's language, reflecting that of Ephorus, shows that the latter did not always praise Athens and censure Sparta; see pp. 115 ff. 30. 2. οú μ+1F74ν ἐλ1λὰ τΟιδ συμμαχιαν κτλ : a reference ahead to the mid-370s when the Athenian Confederacy reached its widest extent and achieved recognition by Sparta in the King's Peace of 375 (see next n. and 38. 4n.). εἰδ συμμαχιν . . . εßδομńκοντα πóλε ιδ: as the text stands the figure seventy seems to include the five states mentioned earlier at 28. 3 (Chios, Byzantium, Rhodes, and Mytilene) and 29. 7 (Thebes). But perhaps D missed Ephorus' explanation that the seventy were in addition to the five states already mentioned. Such an oversight is all too likely and if that is indeed the case D's information is brought into line with that of Aeschines 2. 70 (cf. Vial, 37 n. 3; Cawkwell, 1981b: 41 and n. 6). It would seem that the Confederacy was generally believed to have included

seventy-five members at its height. Whether this belief was correct (as I think) or not is another matter. Aeschines' statement (2. 70) that all seventy-five cities were won over and made members of the synedrion by Timotheus, is clearly false. Isocr. 15. 113 and -272-

Dinarch. 1. 14 (cf. 3. 17) say that Timotheus captured twenty-four cities in all for Athens. Most of these will be the places taken in the north Aegean in the later 360s which were not, apparently, admitted to the Confederacy ( Griffith in Imperialism, 138; Cawkwell, 1981b: 45). Is there then a contradiction between the literary and the epigraphic evidence and where does the truth lie? On the stele of the decree of Aristoteles, on which all members were meant to be inscribed (lines 69ff.), as this has been restored from a large number of fragments, some fiftyeight names can be said to have been recorded, certainly not more than sixty (Tod, pp. 67 f.; Cargill, 1981: 45). How can the discrepancy between the stele and the literary sources be explained? In his review of Cargill Hamilton suggests that Ephorus created the discrepancy by enumerating the cities of Boeotia, the Olynthian Confederacy, and Acarnania separately as members of the League. The figure of seventyfive then passed into the 4thcentury tradition and 'may well have been fixed by the time Aeschines wrote some 30 years later' (1982a: 312f.). I find this most unlikely. Aeschines and Ephorus were exact contemporaries and the speech On the Embassy was probably delivered before Ephorus wrote that part of his history which dealt with the Second Athenian League (see pp. 110 ff.). But apart from this, confederated cities such as those proposed by Hamilton could not have sent representatives to the synedrion in their own right and it is difficult to see how anyone could have reckoned them separately as members. The common view is that D and Aeschines are correct and that for some reason the Athenians stopped inscribing the names of new members on the stele after a certain date. There is no general agreement either about the date or about the reason. Cargill takes a different line. His starting principle is that states whose names were not recorded on the stele were not members, that, therefore, there were never more than some fifty-eight members in all and that consequently the information of D and Aeschines should be rejected (1981: 45ff.). The proposal is especially provocative because Cargill, contrary to the scholarly consensus, is firmly of the opinion that Corcyra was not inscribed on the stele and was not therefore a member of the League. For Cargill Svt 263 (=Tod 127) is not a League alliance, but one between Corcyra on the one hand and Athens and her allies on the other, the first of this kind -273-

of treaty. In summer 375 Corcyra, Acarnania, and the cities of Cephallenia applied to join the League and their applications were approved, but before the Athenian envoys elected for the purpose could reach the cities concerned and administer the requisite oath (and a state's name could not be recorded on the stele, Cargill thinks, on the basis of his misreading of Svt 262 (see below), unless this were first done) something happened and the agreement was partly aborted: Acarnania and Pronnoi of Cephallenia took the oath and their names were inscribed on the stele ( Svt 257 B 10-12), the rest of the Cephallenian cities and Corcyra did not and were not admitted to the League ( Cargill, 1981: 68ff., 103ff.). To challenge Cargill's thesis it is first necessary to look at the procedure for joining the League and on the kind of alliances that that involved. Whatever the nature of the Delian League, the present League was a bicameral organization (28. 3n.). Questions then arise about the treaty or treaties that prospective members were obliged to conclude, questions which Cargill's study does not effectively get down to answering. While the evidence does not allow complete certainty, the indications are that members of the League were (a) allied with Athens herself in what amounted to bilateral treaties, and (b) allied with Athens' allies. These were general, 'League' alliances, that is, individual members were not allied with each other, but with the body of the allies. (a) and (b) were not independent acts, but were bound to each other in their provisions and oaths. There can be no doubt that this was the case with some of the members and we have no reason for supposing that it did not apply generally. Methymna ( Svt 258=Tod 122) was already an ally of Athens. To become also the ally of the allies of Athens (ὃπΩδ ὄν και πρòσ τòσ σNUμμαχοδ τóδ ÁΘńναιΩν νιαντοιδ ń σνμμαχι1α: lines 6-8), that is, a member of the League, she has to take the oath the other allies took. The oath is sworn to the synedrion of the allies as well as to the Athenian generals and hipparchs. The boule is not involved here in the oath-taking, as it generally was in the 4th century in alliance treaties, because Methymna was already an Athenian ally. Byzantium, on the other hand, a city not allied with Athens which has applied successfully to join the League then in process of being founded ( Svt 256=Tod 121 lines 3-6), must first enter into a bilateral alliance with Athens ( Svt 256). And this alliance is -274-

to be like that with Chios (lines 6f. Dittenberger's restoration seems safe: Accame, 1941: 34 n. 1). The boule, the generals, and the hipparchs are involved in the oath-taking. Of course, Svt 256 comes from the very time of the League's foundation and Svt 258 (Methymna) from its infancy, but there is no reason to deny that by the time the latter was concluded a definite procedure had been worked out which, as we can see ( Svt 258 lines 8-11), included the inscribing of a new member's name on a common stele, an ancestor it would appear of the one which was soon to carry the decree of Aristoteles (28. 3n.). At any rate, by the time Chalcis in Euboea joined in summer 377, the steps for admission to the League must have been settled. In vt 259 (=Tod 124) we have the bilateral (Chalcis with Athens) treaty of alliance which preceded Chalcis' entry to the League. τὴν πóλιν (lines 13f.) surely indicates the usual bodies in Athens, including the boule, charged with the taking of such oaths (cf. Rhodes, 1972: 44). Lines 21ff. point to the treaty as the first and, probably, indispensable step to membership of the League.

SVt 262 (= Tod 126) and Svt 263 (= Tod 127) easily fit into this framework. Corcyra, Acarnania, and Cephallenia wished to join the League and their wish was granted ( Svt 262). They were not allied with Athens, hence we observe the double process of (a) the oath-taking between the states and the boule, the generals, and the hippeis of Athens which would make them the allies of Athens (lines 15-17). A number of separate, bilateral alliances would have resulted and Svt 263 appears to be the treaty of Corcyra with Athens (cf. Tod, pp. 87 f.; SVt p. 220). Svt 267 (= IG ii2 98. Add. p. 658. Re-edited with additional fr. by Schweigert, 1940: 321 ff.), a badly preserved inscription which mentions garrisons and Athenian superintendents, may well be the decree of alliance between Athens and the cities of Cephallenia. (Or it may date to 372 when Iphicrates had to subjugate the cities of Cephallenia anew: X. H. 6. 2. 33, 38.) (b) The allies, that is, representatives of the synedroi, were also to swear 'the oath' to the states concerned (lines 17f.). This oath is not the same as 'the oaths' in line 15 sworn by the Athenian officials, but we should see in it rather the oath of Svt 258 (Methymna) lines 16-19. Of course, the ambassadors of the cities had also to take an oath to the synedroi ( SVt 258 11-16), something which the shorthand of SVt 262 17f. seems to have omitted, as it has also omitted it seems to mention -275-

that the Athenian generals and hipparchs were to take the 'League' oath alongside the synedroi (cf. Svt 258 11 ff.). In fact Svt 262 cannot sustain Cargill's interpretation of it. The decision of the Athenian demos concerning Corcyra, Acarnania and Cephallenia was that the names of the cities were to be inscribed on the common stele of the allies and the oaths administered to them (lines 10ff.). 'These things having been done' (πραχΘἐντΩν δἐ τοντ[Ων: line 18) the demos was to proceed to the appointment of commissioners who would sail out to administer the oaths to the cities concerned (lines 18ff.). These oaths or the names of the commissioners or indeed both are also to be recorded on the common stele of the allies: lines 20f. The reference here is not to the cities themselves, as Cargill mistranslates (1981: 104ff.; cf. Tod, p. 84 ). His view that the commissioners sent out to receive the oaths from the cities were also responsible for the recording of the cities' names on the stele, but that this could only be done after the oaths had been received, is thus based on a misunderstanding. And his further view that 'this is the only interpretation that can reasonably account for the fact that an alliance made with Korkyra, the Akarnanian confederacy, and (apparently) all four of the cities of Kephallenia led to the listing of only the Akarnanians and one Kephallenian city (Pronnoi) on the stele of the allies' ( 1981: 107) is clearly untenable. Svt 258 (Methymna) is equally clear that the listing of new members on the stele took place fairly speedily after the Athenian demos gave its assent and not subsequent to the administration of the oaths to the cities' home authorities. The suggestion by some scholars that names were 'saved up' for a while and then all listed at once, and that this is the reason for the lack of a geographical pattern (cf. Sealey, 1957: 105), does not tally with the evidence. Svt 263, in any case, is radically different from, e.g., Svt 293 (=Tod 147) which clearly belongs to Cargill's second category of alliances, 'allies of Athens and the League', but to which, equally clearly (pace Cargill), Svt 263 does not. With the usual preliminaries to such a document deliberately excluded, Svt 263 is best seen as an adjunct of Svt 262. Why then do not Corcyra and the rest of the Cephallenian cities appear on the left side of Svt 257, the decree of Aristoteles, along with Acarnania and Cephallenian Pronnoi? ( Svt 257 B 1 is certainly not Κερκν]ραιΩν: Bradeen and Coleman, 1967: 102ff. who suggested Theraians. Mitchel, 1984a: 39 ff., prefers Pheraians.) As -276-

Cawkwell says, it cannot be that 'they were simply omitted. The instructions of the Athenian demos were not lightly to be disregarded' (1981b: 42). His suggestion that space was found for an important city like Corcyra on the front of the stone is by no means impossible. But what of the remaining three cities of Cephallenia which Cawkwell ignores? A way out of the epigraphic difficulty might be to suppose that when Svt 262 speaks of the Cephallenians no more than Pronnoi is meant. This is unlikely. The cities of Cephallenia did not form a single state. They were independent communities with separate coinages (cf. RE ix. 204f.). When therefore Svt262 and 267 refer to the Cephallenians it is because all four independent states were following the same course of action at the time. Was space then also found for them on the front of the stele? The four cities of Ceos, all of which may have joined the League at the same time ( IG ii2; cf. Cargill, 1981: 61), perhaps provide an analogy: three are on the left side of the stele, but the fourth, Poiessa, is on the front. But perhaps there is a more satisfactory solution to the riddle. As we have seen, the oaths and/or the names of the men chosen to administer them to Corcyra, Acarnania, and Cephallenia were to be recorded on the common stele of the allies ( Svt26220-2). Nothing of the kind is to be seen on the stone, but, as Tod notes (p. 84 ), the stele 'is incomplete at the foot'. These are puzzling observations and it is surprising that they have not attracted more scholarly attention. Assuming, as we should, that the oaths and/or the names of the commissioners were recorded on the stele, this doubtless was done below line 96, if indeed the whole thing, as is quite possible, was not part and parcel of Aristoteles' second proposal (lines 91ff. The decree lacks a prescript. This is a little surprising, but cf. ML89 lines 48ff.). It is very probable that the states listed on the left side (with the exception of the last and perhaps the first which seem to have been inscribed by different hands: Fabricius, 1891: 596f.; Bradeen and Coleman, 1967: 102ff.; 45. 3n.), together with a certain number recorded on the front and now missing, joined the League at the time of the conclusion of the Common Peace of 375 (cf. Cawkwell, 1981b: 42ff.); and it is quite likely that Aristoteles' second motion sprang from this sudden growth in the League's membership. The oaths and/or the commissioners may well have concerned more states than those specified in Svt262. At any rate, there is no -277reason why Corcyra and the missing Cephallenian cities cannot have been listed near the foot (now missing) of the stele, immediately below the oaths and/or the names of the commissioners. And not only these four states, but the others also required to bring the total membership of the League to seventy-five. The extra 12 inches or so needed to accommodate fifteen to seventeen names would not make the stele impossibly tall. We cannot now know why the right-hand side of the stele remained uninscribed. Perhaps it was too close to another monument on that side and hence difficult of access?

We need not assume, from the instructions of the demos to the secretary of the boule to have the names of the cities inscribed on the common stele forthwith ( Svt26212-15), that these should all have appeared in close sequence, one after the other, on the stone. The cities after all were all independent states and, though grouped together in the preliminary decision of the demos to admit them as allies ( Svt 262), in all subsequent steps they would have been treated separately. What is more, the circumstances of some required further consideration ( Svt26225ff.; Svt26713ff.) and this may have delayed matters a little. It should not therefore surprise that, at a time when many other cities may have been queuing to join the League, the names of Corcyra, Acarnania, and the cities of Cephallenia were not all recorded together. See further nn. ad 36. 5. It follows that it is not necessary to posit a discrepancy between the literary and the epigraphic evidence in connection with the League's membership, nor to theorize (as do Woodhead, 1957a: 371; 1962: 258ff.; Sealey, 1957: 104ff.; Cawkwell, 1981b: 42ff.) about a second 'wave' of adhesions to the League in the late 370s (or even later) which were somehow not recorded on the common stele. Though wrong in other respects, Cargill is surely correct to insist that there were no members of the League which were not recorded on the common stele and that the League did not outgrow the decree of Aristoteles after 375 (as Sealey, 1957: 104ff., suggests). For a possible reason why, in my view, the League ceased growing after 375 see n. ad 38. 4. ἐπ ἴσńσ: see 28. 4; 29. 7. A vague but clear reference to the equal voting power enjoyed by all members. ο δ ÁΘńναμιο22ι . . . δúναμιν ἐΞἐπεμΨαν: the force was meant to defend the Euboean cities which had already acceded to the -278Confederacy (30. 1n.), and to make war on the pro-Spartan elements in north Euboea. The expedition does not provide us with much of a terminus ante quem for the accession of Euboea since Chabrias, its leader (section 5), could have sailed to Euboea before he helped the Thebans against Agesilaus in Boeotia in early summer. 30. 3. ßραχù μųν πρò τοúτων τω+̑ν χρóνων: an approximate date for these events can be obtained from speculation on the identity of the Spartan dispatched to Euboea. D's Therippides (see n. below) is unknown otherwise. But it is very probable that he is identical with the better-known Herippidas, one of the three Spartan officers on the Cadmea at the time of the coup (27. 3n.; Plut. Pel. 13. 3 -- if Bryan's emendation of Plutarch's text is correct. Lenschau, RE viii. 1. 684ff., does not take this into consideration, but brings Herippidas' career to a close with X. H. 4. 8. 11). If this is so, then Herippidas could have moved from Thebes to north Euboea on receipt of orders from Sparta, and dealt with Neogenes. Hence the events of sections 3-4 belong sometime between 382 and 379. According to Plut. Mor. 578a, Herippidas and the other two condemned commanders had succeeded Phoebidas in 382. Cf. Parke, 1927: 159ff. Ι+̔άσονος του+u=0311 Φεραíου: the famous Jason who after Leuctra made himself tagos of Thessaly (for a discussion of the date see 45. 3n., part 3). On Jason's early years see Westlake, 1935: 67ff.; Sordi, 1958: 156ff. Jason had designs to expand both by land and sea (X. H. 6. 1. 10ff.). The geography of Hestiaea was obviously of great importance to Jason's naval plans, as Pagasae, the port of Pherae, lies at the top of the Pagasaean Gulf, and the north coast of Euboea extends along the narrow mouth of the Gulf at no great distance from it. The importance of the place to the sea routes into the Gulf is well illustrated by X. H. 5. 4. 56; cf. Polyaen. 2. 7. Ε+̔στιαιέων . . . Ω+̔ρειτΩ+̑ν: Ω+̔ρειτΩ+̑ν (Ω+̔ρεíταις in section 4) is Le Paulmier's emendation of the MSS consensus ὠρωΠíων(ὠρωΠíοις in section 4), rightly accepted by Vogel though not by Vial. Theopompus ( FGH 115 F387) says that after the destruction of Hestiaea in 446 the Athenians planted a colony at Oreos which place had previously been a demos of Hestiaea. At any rate Hestiaea came to be known as Oreos too (Thuc. 7. 57. 2; 8. 95. 7; Strabo 10 C 445; see Moggi, 1976: no. 18, pp. 115ff. Vial objects to Le Paulmierw's emendation which was prompted, she says, by the fact -279-

that Oreos was the other name of Hestiaea. Now, Vial thinks, D here only makes sense if two separate cities are understood, and she therefore suggests that the MSS Ω+̔ρωΠíων should be emended to Ο+̔ρΟßíων, Orobiae known to have been a city in the region of Hestiaea (Strabo 10 C 445; cf. 9 C 405; Thuc. 3. 89. 2). Vial is wrong for while D in section 3 may be thought to be speaking of two different places, in the next section he is clearly speaking of the same place. It might at first sight appear that Ephorus understood the city to be called Oreos and the area Hestiaea, but as at 30. 1 Hestiaea is referred to as a polis (cf. 11. 13. 5; 12. 7. 1) one may conclude that the city (and the surrounding country) was known to Ephorus too as either Oreos or Hestiaea. In that case we must assume that the occurrence of both names here is the result of Ephorus' desire to vary his style and display his erudition. That Vial's emendation is wrong is proved by X. H. 5. 4. 56 f.: Alcetas, the Spartan commander at Oreos, had clearly been installed by Therippides/ Herippidas after the expulsion of Neogenes. And we may compare Polyaen. 2. 7; cf. Front. 4. 7. 19. Θηριπíδ1ην: the reading of the best MSS and therefore so printed by Vogel and Vial. But surely (whatever D may have written -- we cannot now know) Θηριπíδ1ην (MS F) must be the correct form of the name. 30. 4. ἐξεπολιóρκησε: MSS ἐπολιóρκησε. Dindorf's emendation ( Vogl), rejected by Vial, seems justified. 30. 5. τήν Ε+̔στιαιω+̑τιν χώραν . . . ΜητρÓπολιν: Vial ad loc. makes the absurd suggestion that Hestiaeotis, the region in west Thessaly, is meant here, her proof being the reference to Metropolis which she identifies with the city of that name in western Thessaly! But at section 5 the narrative is resumed from section 2, and Chabrias' operations were clearly in north Euboea. Having failed to carry Oreos, he laid waste the land, chose some rocky site nearby called Metropolis, fortified it, and sailed on. The absurdity of Vial's assertion is shown especially by the geography of the Thessalian Metropolis. It was situated in the very depths of Thessaly, 30 miles west of Pharsalus and some 50 miles inland. αὐτòς δέ ται+̑ς Κυκλάσι κτλ: strictly speaking Peparethus and Sciathus do not belong to the Cyclades (cf. Strabo 2 C 124) though both here and at 95. 1 D (or rather Ephorus: Accame, 1941: 80) thinks that they do. Peparethus and Sciathus would have been 'persuaded' to join Sparta either by Therippides/

Herippidas or by -280Alcetas (see above). Their neighbour to the east, Icus, which had perhaps been under Sparta too, had already joined Athens (30. 1 n.). It is not easy to find candidates for καí ἄλλας τεταγμένας ὑπò Λακεδαíοις. The same hand which cut the Sciathians and Peparethians on the front of Svt 257=Tod 123 may also have cut ΠερíνΘιοι, Μαρωνι+̑ται and Δíη+̑ς (Tod, p. 66. Accame, 1941: 76f., following Fabricius, 1891: 597, agrees though he assigns the five names to three groups inscribed by the same hand, but at different times. Cargill, 1981: 34, is sceptical). Perinthus and Maroneia in Thrace acceded voluntarily, we may guess, but the accession of the small town of Dion west of Oreos on the north Euboean coast, as well as that of Athenai Diades immediately to the east of Dion (line 90 though apparently cut by a different hand), may have been due to the efforts of Chabrias. Both places had perhaps been under Sparta. Dion and Athenai are not islands admittedly, but then D's language can usually admit of a certain degree of flexibility and imprecision. Accame, 1941: 78f., further suggests the Palesciathians whom he restores in line 85. Now, the statement that Chabrias sailed in the Cyclades may allow a wider scope than the vicinity of Sciathus and Peparethus, and Accame, 1941: 81 ff., offers a number of Cycladic islands which he suggests originally stood below line 85 on the front face. He may be right, at least in some of the names suggested, though proof is lacking; see also 34. 4 n. and cf. Cargill, 1981: 34ff. 31. 1-4. Spartan countermeasures 31. 1. οἱ δέ ὁρΩ+̑ντες. . . κατεσκεúασαν: see 27. 4; 28. 4-5 and nn. The Spartans put an end to their former bad ways, and by embassies, friendly words, and the bestowal of benefits regained the goodwill of their allies. The embassies and the friendly words require no explanation. The benefits may allude to the reorganization of the Peloponnesian League which was, perhaps, fairer on the allies. 31. 1-2. The reorganization of the Peloponnesian League See 28. 5 and 28. 4-5n. We are now told what the Spartan preparations were. They were chiefly concerned with the reorganization of their League. The reorganization, which is, surprisingly, not -281mentioned by Xenophon (or any other source), was based on geography and numbers, and the ten divisions or districts are given in a geographical order in D: (1) Lacedaemonia; (2) Arcadia (centred on Tegea?); (3) Arcadia (centred on Mantinea?); (4) Elis; (5) Achaea; (6) Corinth and Megara; (7) Sicyon, Phlius, and the Acte; (8) Acarnania; (9) Phocis and Locris; (10) Olynthus and the other allies in Thrace. Each of the ten divisions had to contribute a body of troops in approximately equal or rather equivalent numbers, and, this being the Peloponnesian League, the basis of the calculation must have been hoplites. The evidence for the numerical strength of the states concerned is admittedly limited, but it appears that each of the ten divisions was able to provide some 3,000-4,000 hoplites, or the equivalent in light-armed infantry, cavalry, money, or a mixture of all of these (I cannot see how Bengtson, GG273, reaches his estimate of 1,000 men per district). Elis, the 4th division, sent 3,000 hoplites to the help of Argos in 418 (Thuc. 5. 58. 1; cf. 5. 75. 5), and nearly 3,000 Elean, Triphylian, Acrorian, and Lasionian hoplites took part in the battle of Corinth in 394 (X. H. 4. 2. 16). Mantinea, if we suppose Tegea and Mantinea to have been the respective centres of divisions 2 and 3, sent nearly 3,000 hoplites to Argos in 418 according to D (12. 78. 4). At this time democratic Mantinea had conquered and was ruling over an extensive part of Arcadia (Thuc. 5. 29. 1; 33. 1; cf. HCT iv. ad loc.), and D's figure is not implausible and it should not be rejected as a mere Ephoran guess simply because Thucydides is silent on the point (for Ephorus' use of non-Thucydidean material, see pp. 106, 129 ). Lysias (34. 7) claims that in 403 the Mantineans numbered less than 3,000 men altogether, but the reference is suspect. At any rate, the permanent federal army instituted in 370 numbered 5,000 so that about 8,000 is a reasonable figure for the total Arcadian field army (reserves excluded) of hoplites (cf. Larsen, 1968: 194). The Spartans therefore divided Arcadia in two. The Corinthians had in 418 campaigned with 2,000 λογάδες hoplites (Thuc. 5. 57. 2; 60. 3). But in 378, after the tribulations of the Corinthian War, Corinth is unlikely to have had more than 2,000 hoplites in all, and Megara, its partner in the 6th district, could perhaps have put some 1,000 hoplites in the field. Sicyon, Phlius, and the Acte made up the 7th district. Sicyon -282sent 1,500 hoplites to Corinth in 394 (X. H. 4. 2. 16), while Epidaurus, Troezen, Hermione, and Halieis, i.e. the Acte (Paus. 2. 8. 5; Strabo 9 C 390), sent 3,000 (X. H. 4. 2. 16). Phlius is said to to have had a population of more than 5,000 men in 380 (X. H. 5. 3. 16). There followed Agesilaus' settlement of the city which resulted in many exiles (40. 5), and we might reckon 1,000-1,500 for Phlius' hoplite field force (in the early 360s it had no more than 60 cavalry: X. H. 7. 2. 4). With some 5,000 front-line hoplites the 7th division was probably one of the strongest. The strength of the Lacedaemonian army is a matter of controversy, but the reorganization of the League appears to exclude the possibility that it numbered (as some scholars suggest) no less than some 7,680 hoplites for then it would not have been on a par with the other divisions (32. 1n.). And divisions of 8,000 instead of 4,000 would result in a field force of some 80,000 hoplites which seems unacceptable (the figure in FGH324 (Androtion) F39 for 431 is corrupt (cf. Jacoby's n.), and Plutarch's 60,000 Peloponnesian and Boeotian hoplites (Per. 33. 5), at two-thirds strength presumably (Thuc. 2. 10. 2), is hardly to be credited). No worthwhile results are obtained from an examination of the other divisions; but see further 32. 1n. Why was the League reorganized at this juncture? A major aim must have been greater efficiency; as D Puts it, καÌ τόν διάταξιν καì τόν διαíρεσιν τω+̑ν στρατιωτω+̑ν καÌ λεριττóτερον ἐξειργáσαντο (31. 1). It is possible that the reorganization was initiated even before the Theban revolt at a time when, according to both D (23. 3ff.) and Xenophon (H. 5. 3. 27), Spartan might was at its zenith. D's cryptic words at 23. 4 may well hint at a greater

attention given to her alliance by Sparta at this point following the incorporation of Olynthus and the other Thracian cities. πολυανΘρωπíα may refer to an additional aim of the reorganization: an increased number of troops. The reorganization cannot have been completed overnight. It would have required discussion as well as, of course, the assent of the allies, and it may have taken up several months. The revolt of Thebes and of part of Boeotia intervened and the final arrangements thus excluded Boeotia. The new system would have been simpler than the old one and more satisfactory from the Spartan point of view. Under the old system the many cities of the League sent their levies individually (X. H. 3. 4. 3) and scores of Spartan officers were dispatched on -283-

each occasion to collect and command all the various contingents (Thuc. 2. 75. 3; X. H. 3. 5. 7; 4. 2. 19; 5. 1. 33; 5. 2. 7; cf. 4. 5. 7). Now, it would seem, only one officer per district was needed -- one ξεναγóς conducted the Corinthians, Epidaurians, Troezinians, and the rest to Sparta in the admittedly unusual circumstances of winter 370/69 (X. H. 7. 2. 3). This was a definite advantage considering the steadily diminishing number of Spartiates (though perioikoi may have acted as ξεναγοí too). D concludes the list of the ten divisions with an equation: 1 cavalryman=4 hoplites=8 light-armed infantrymen. He does not explain this, but clearly the equation was necessary if the districts were to be fairly assessed for most states contributed an assortment of all three arms. And we may guess that there was also in Ephorus a statement about those states which wished to contribute money instead of men, and the relevant equation. For already in 382, Xenophon tells us (H. 5. 2. 21), permission was given to any city which so wished to provide money instead of men at the rate of 3 Aeginetan obols per hoplite and 12 per cavalryman, and in 373 many of the cities preferred to contribute money rather than men to Mnasippus' force (H. 6. 2. 16). Xenophon's information confirms D's equation, 1 cavalryman=4 hoplites. The alternative of providing money would have suited at least those allies who were increasingly unwilling to campaign abroad (cf. X. H. 5. 4. 60 and Smith, 1954: 285f.). It would also have suited Sparta; she could use the money to hire mercenaries. Another advantage of the reorganization may have been a more equitable contribution of men or money which surely took into account a state's population and wealth. Each division was responsible for providing so many men and no more. This must have meant that some divisions could provide more than they actually did, but this could not be helped if they were all to be equal. The cornerstone of the new system must have been the size of the Lacedaemonian army which formed the first division. Cartledge, 1979: 291 (cf. id., 1987: 272), may well have a point when he says that the reorganization was 'designed perhaps to emphasize the burden carried by the Spartans and so to restore their prestige'. We do not know if the reorganization went beyond the land forces. Certainly Sparta had a number of island allies (30. 5; X. H. 5. 4. 65; 6. 2. 3), but it is debatable whether these were members of the Peloponnesian League. But coastal Peloponnesus had -284-

always contributed to Spartan fleets, and there must have been rules governing these contributions. For the forces raised as a result of the reorganization, which in fact confirm the above analysis, see 32. 1nn. 31. 2. τήν δ+̕ ὀγδóην κτλ: there is no problem about the states in the first seven divisions; they were all indubitably members of the Peloponnesian League. Those in the last three groups did not traditionally belong to the League, but this is no good reason for denying that they did at this time. The case for Olynthus and the allies of the 'Thraceward' area seems safe enough. On the capitulation of Olynthus in 379 the Olynthians had agreed to 'have the same friends and enemies as the Lacedaemonians, to follow them wherever they lead, and to be their allies' (X. H. 5. 3. 26) which indicates membership of the League, and D in fact states this explicitly ( 23. 3). The Acarnanians had concluded an alliance with Sparta in 388 (X. H. 4. 7. 1) and they may have entered the League at the same time, while we know nothing about the circumstances and nature of the treaties of Phocis and the Locrians. But the presence of all these states here strongly suggests membership of the League. Certainly this is more likely than the view that the ten divisions were a conglomeration of members and non-members (cf. de Ste Croix, 1972: 338). The Chalcidians seceded to Athens in 375 (Svt 250=Tod 119 (see pp. 277 f., 328 f. for the date); Svt 257 (=Tod 123) B lines 5f.) and so did the Acarnanians (Svt 257 (= Tod 123) B line 10; Svt 262=Tod 126). Λοκροí: ἀμΦóτεροι? Perhaps, though we have no way of knowing. Section 57. 1 does not help. οἱ ἐπì Θρᾳ+́κης . . .:σúηηαχοι see 23. 3 and 30. 1n. for a possible candidate, Arethusa. 31. 3-4. This rhetorical estimate of Agesilaus, as was the manner of Ephorus (p. 123 ), serves as an introduction to the warfare of the next two years conducted in Boeotia by the king. Agesilaus' energy and aptitude for war are stressed, but this is no eulogy; see pp. 119 f. 31. 3. δúναμιν νικήσας: the battle of Sardis in 395; see 14. 80. καì πέρας . . . κατέστησεν: it was widely believed that Agesilaus had come close to overthrowing the rule of the Great King, and that he would have succeeded had he not been recalled by his home government (cf. X. Ag. 1. 36; Plut. Ages. 15; Nepos, Ages. 4. 1-3). The origins of this belief may be traced to Agesilaus' attempt -285to sacrifice at Aulis like a second Agamemnon before invading Asia (X. H. 3. 4. 3 f.), and in the fact that at the time of his recall he was preparing to march as far as possible into the interior; or so our sources tell us (X. H. 4. 1. 41; Hell. Oxy. 22. 4). Isocrates (4. 144) thought that Agesilaus had conquered nearly all the territory west of the river Halys. In reality his success was more modest and it was likely to remain so (see next n.). But Isocrates too considered that Agesilaus' failure was ascribable to internal Greek politics (5. 86f.), and so did Polybius (3. 6. 11) who nevertheless was of the opinion that Agesilaus returned to Greece ἄπρακτος having encountered little or no resistance in Asia. D (Ephorus), like Polybius, reflects a tradition critical of, if not hostile to, Agesilaus and Sparta because of their Greek policies. But Ephorus was a moralist, a panhellenist (and an admirer of Athens), and whenever Sparta acted in accord with virtue and the interests of Greece, he praised (see pp. 114 ff.). In this instance, as a panhellenist, Ephorus readily shared the common belief that Agesilaus had come close to overthrowing the Persian empire. κρατω+̑ν τω+̑ν ὑπαíΘρων: Agesilaus' inability to capture fortified places is stressed by the Hellenica Oxyrhynchia (21. 5 f; 22. 3), and D's statement here that Agesilaus 'traversed a large part of Asia mastering the open country' is one more link in the chain Hell. Oxy.-Ephorus-D. Unlike Alexander, Agesilaus lacked the means necessary for capturing cities and forts, and this meant that his conquest of any part of the King's country could not but be temporary. Confined mostly to the open country, he was a mere raider. διά τινας . . . χρεíας: the epitomator's shorthand? Ephorus may have explained that the reason was the outbreak of the Corinthian War. 31. 4. ἀνὴρ . . . δραστικòς κτλ: see 19. 4 and n.; also pp. 119 f. του+̑τον . . . ἡγεμóνα: not so. D was ignorant of the mechanics of the Athenian constitution (p. 133 ), and we cannot expect him to have been better informed in the case of Sparta. When the Lacedaemonians ordered mobilization against an enemy one of the two kings (or some other important Spartan acting in their stead) was appointed to command. His command was just for that particular campaign (Agis at Decelea and Agesilaus in Asia were not typical, but exceptional cases). Cf. X. H. 5. 4. 3 5: οἱ δ+̕ αὐ+̑ Λακεδαιμóνιοι Φρουράν τε ἔΦηναν ἐΠì τοùς Θηßαíους καì τòν Α+̔γησíλαον -286. . . ἐδέοντο . . . ἄγειν τὴν στρατιάν, which is how D himself puts it earlier in section 3: ἡγει+̑το τη+̑ς στρατια+̑ς Α+̔γησíλαος ὁ ßασιλεúς. 32-33. 4. The first invasion of Boeotia 32. 1. ἡ+̑κεν εἰς τὴν Βοιωτíαν: from X. H. 5. 4. 56 we learn that Agesilaus destroyed the Theban corn harvest in two successive years. He must therefore have entered Boeotia by the end of May 378 at the latest: Hesiod, Works

and Days383 f. and pp. 253 ff. in West's edn. for the time of harvest. For the year see 25. 1n. ἔχων . . . μóραι: I take the generic στρατιώτας to refer to hoplites. Later in the same section we are told that Agesilaus' cavalry numbered 1,500. As one cavalryman equalled four hoplites (31. 2), Agesilaus had with him the equivalent of 24,000 hoplites, i.e. fivesixths of the League field forces, considering that five of the six Lacedaemonian morai marched out. The total field forces would thus have been equivalent to about 30,000 hoplites, 3,000 per district, which is indeed the conclusion reached above in the examination of the League reorganization (31. 1-2n.). In fact rather more than 3,000, for Agesilaus was also accompanied by light-armed troops for which D gives no figure (32. 4). One hoplite was equal to two light-armed so that our estimate of 3,000-4,000 hoplites as the approximate assessment for each district appears quite plausible. In the following year, 377, Boeotia was again invaded by the same force (34. 1). We have no figures for the League army on other campaigns of the 370s with the exception of Leuctra. Four Lacedaemonian morai were present then (X. H. 6. 4. 17), and Cleombrotus therefore must have had two-thirds of the League levy with him, i.e. the equivalent of 20,000 hoplites. Now, the 8th division, Acarnania, and the 10th, Olynthus and the Thracian cities, had fallen away by then (31. 2n.), i.e. the equivalent of some 6,000 hoplites. If we ignore the figures given by Polyaenus (2. 3. 8) and Frontinus ( Strateg. 4. 2. 6; cf. Anderson, 1970: 321 n. 26), and work on the basis of the numbers supplied by Plutarch in Pelopidas (D gives no figures for the Peloponnesians and Xenophon only for the Lacedaemonians, incidentally), we see that the result is not inconsistent with our previous calculations. Cleombrotus invaded with 10,000 hoplites and 1,000 cavalry ( Pel. 20. 1), i.e. with the equivalent of about 14,000 hoplites. αἱ . . . πέντε μóραι: the text is wrong as it stands; the Lacedae-287-

monian army was organized in six not five morai. The difficulty is resolved if the article is omitted (as indeed it is by a small number of admittedly secondary MSS (presumably a correction by a copyist). αἱ is Vogel's emendation of the οἱ of the rest of the MSS). If that is done, then Agesilaus had five of the morai with him, the sixth having been left at home, a not uncommon measure (cf. 14. 97. 5; Thuc. 5. 64. 3). The following spring Agesilaus invaded Boeotia with the same force (34. 1). As one mora was already at Thespiae ( X. H. 5. 4. 46 f.), the king must have led out four morai and again left one behind. τὴν δU1F72 μóραν . . . πεντακóσιοι: the organization and strength of the Lacedaernonian army are problems of considerable intractability which cannot be treated here. For discussions and a plethora of views, see Toynbee, 1969: 365ff.; Andrewes, HCT iv. 110ff.; Anderson, 1970: 221ff.; Cartledge, 1979: esp. 254ff.; id., 1987: 37ff., 427ff.; Forrest, 1980: 131ff.; Cawkwell, 1983: 385ff.; Lazenby, 1985; Figueira, 1986: 165ff. For our period, I accept the evidence of Xenophon that the Lacedaemonian army (both Spartiates and perioikoi brigaded together, though not perhaps in the lowest units, the enomoties: cf. Cartledge, 1987: 42. Contra Lazenby, 1985: 16ff.) consisted of six morai of hoplites (I omit the cavalry morai), each commanded by a polemarch and subdivided into two lochoi (H. 7. 4. 20; 7. 5. 10. Four lochoi, according to Lac. Pol. 11. 4 where the figure should perhaps be emended to two: Toynbee, 1969: 396), eight pentecostyes, and sixteen enomoties. This is broadly consistent with the information of Thucydides about the Spartan army (from his account of the battle of Mantinea in 418: 5. 66-73), though Thucydides evidently became confused over the exact structure and nomenclature of the army, erroneously assuming that the highest multiple of the enomoty was the lochos whereas in fact it was the mora. This apart, it is important to appreciate that both Thucydides and Xenophon agree on the essentials, above all on the number of enomoties in an army of six regiments (96 in all), on the maximum strength of the enomoty (about 40 men, i.e., it would seem, one man from each of 40 year-classes: Toynbee, 1969: 369f.), and therefore on the strength of the army as a whole, about 4,000 men (3,840 is the exact calculation). Many scholars nevertheless consider 4,000 Lacedaemonian hoplites too few and propose doubling Thucydides' figures: the -288historian spoke wrongly of the highest multiple of the enomoty as the lochos when he should have called it a mora, each mora consisting of two lochoi. Thus there were not 96 but 192 enomoties, and the army numbered some 8,000 men (7,680 is the precise figure; cf. e.g. Toynbee, 1969: 369ff.; Wade-Gery, 1958: 71ff., 80ff.; Andrewes, HCT iv. 111 ff.). The evidence is admittedly difficult, but it seems better to let the figures stand. After all, Thucydides, who reckoned the Lacedaemonian numbers at Mantinea in 418 to have been c.4,184 in all (i.e. 448 files, 8 deep on average, plus the 600 Sciritans), did not consider such a sum unreasonable and neither did Xenophon, it would seem, who also reckoned 16 enornoties in a mora, not 32. This agreement of Thucydides and Xenophon seems particularly compelling. Also, and decisively perhaps, there is the fact that Lacedaemonia was only one of the ten divisions of the reorganized Peloponnesian League (31. 1-2n.). She was the equal, for example, of Corinth and Megara together, or of half of Arcadia. She must therefore have provided no more than 3,000-4,000 hoplites. There are, it is true, certain difficulties with the low figures, but none is insurmountable. There is, for instance, the often-voiced objection (cf. Toynbee, 1969: 378f.; Lazenby, 1985: 7) that if Sparta had had no more than some 4,000 hoplites, she could not have imposed her will on Greece. However, all of the 4,000 were well-trained professionals, which was unusual in 5th-century Greece. The advent of the equally well-trained (and not more numerous!) Theban hoplites in the 4th century made all the difference. What is more, Sparta had the Peloponnesian League behind her. X. H. 7. 1. 23f. makes the Arcadians complain that they, being the most populous section of the Greek nation, had made Sparta what she was; and we might compare Timolaus' picturesque analogies at X. H. 4. 2. 11f. One might add too that Sparta did possess additional troops, the perioikoi and neodamodeis, which, when need arose, she used for garrisons or distant expeditions or even in battles. πεντακóσιοι: this is certainly the Ephoran figure for the strength of a mora as Plut. Pel. 17. 4 ( FGH 70 F210) confirms. At 15. 37. 1 we are told that 500 Thebans defeated twice as many Spartans. As there were thought to have been two morai present at Tegyra (Plut. Pel. 17. 4), we obtain again the number 500. Ephorus may have derived this by reckoning a call-up of the first 30 year-

-289-

classes, i.e. 480 men, and adding the officers (cf. X. Lac. Pol. 11. 4). Plut. Pel. 17. 4 gives Callisthenes' figure too, 700, which may be the mora at full strength, 640 men, plus the officers, plus one of the cavalry morai (cf. X. Lac. Pol. 11. 4). Polybius' figure, 900, perhaps refers to the later time of Nabis. ὁ δέ Σκιρι+́τńς κτλ: this is palpable nonsense as it stands. The beginning of the passage, ὁ δέ Σκιρι+́τńς κτλ . . . ἰδίαν ἔχων σύστασιν, seems factual enough, but most of the rest seems to refer to the hippeis rather. Ephorus appears to have included a description of the Spartan army at this point in the Histories and in summarizing it D 'has telescoped a reference to the Skiritai with one to the 300 "hippeis" ' ( Andrewes, HCT iv. 104). For the Sciritans cf. Toynbee, 1969: 208f. (but they were hoplites not light-armed, as Thuc. 5. 67. 1; 68. 3 shows); Andrewes, HCT iv. 103 f.; Anderson, 1970: 249ff.; Lazenby, 1985: 10. The 300 hippeis are one of the more elusive aspects of the Spartan socio-military set up. The belief that a special (or crack or elite) corps of 300 picked young Spartans called hippeis formed a royal bodyguard and fought with the king in battle in a separate formation is fairly general (cf. e.g. Andrewes, HCT iv. 112; Cartledge, 1979: 204, 275; id., 1987: esp. 204f. Anderson, 1970: 231 ff., is in a minority in casting doubt on these views). It rests on a handful of remarks in ancient writers, above all in Herodotus and Thucydides. Ephorus, to judge from D here, gave the usual story about the hippeis. Can the belief be substantiated? The most explicit statement in Herodotus is at 8. 124. 3 where it is said that 300 picked Spartiates, oΤοι οἵ περ ἱππέες καλέονται, honoured Themistocles on his visit to Sparta by accompanying him as far as the Laconian border. Two other statements hardly help to elucidate this statement. At 1. 67. 5 it is said that the five oldest men who retired from the hippeis every year were called agathoergoi and served the state in various ways for a year. And at 6. 56 we are told that when on campaign a Spartan king was guarded by 100 picked men. These are not called hippeis and some scholars in fact (e.g. Michell, 1952: 249) refuse to see them as such. (On Herodotus see also end of n.) A much clearer statement is provided by Thucydides. At 5. 72. 4 it is said that King Agis at Mantinea was posted at the centre of the battle-line and around him stood the 300 hippeis so-called (. . . ἣπερ ὁ ßασιλεὺς 'Aγις ἣν καὶ περὶ αὐτόν οἱ τριακόσιοι ἱππῃσ καλούμενοι). The trouble with this -290information is that it does not quite tally with Xenophon whose witness on things Spartan deserves greater credence even than that of Thucydides. The first thing to note about Xenophon is that, with one possible exception (H. 6. 4. 14; see below), he does not anywhere refer to the hippeis by name. In Lac. Pol. 4. 1-6, however, he describes a system by which 300 young Spartans were selected on the basis of merit. The selection was made by three men called hippagretal appointed by the ephors. Each hippagretes selected 100 young men. The purpose of the institution was the encouragement of ἔρισ περὶ ἀρετῃς with ἀνδραγαΘι+́α as the final objective. It is impossible to be definite about Xenophon's meaning, but he seems to say that (a) the three hippagretai were over 30 years of age (ἐκ τω+̑ν ἀκμαζόντων τρεîς ἄνδρας: 4. 3) and (b) the 300 themselves were men between about 20 and 30 years of age (ἄνδρας: 4. 3; τοîς γε μὴν τὴν ἡßńτικὴν ἡλικίαν πεπερακόσιν, ἐξ ὡ+̑ν ὡδń καὶ αἱ μέγισται ἀρχαὶ καΘίσταντι: 4. 7). The age-group from which they derived would then appear to be the τὰ δέκα ἀΦ ἥßńς which we encounter as a feature of the Lacedaemonian army (H. 2. 4. 32; 4. 5. 14). It is rather disconcerting that Xenophon does not say either here, at Lac. Pol. 4. 1-6 or 13. 6f., or anywhere else that these 300 acted as a royal bodyguard; nor does he refer to them as hippeis, though of course the term hippagretai ('those who gather the hippeis') implies that of hippeis. The hippagretai are mentioned again in the Hellenica in connection with the conspiracy of Cinadon (3. 3. 4-11). To arrest Cinadon as quietly as possible, the ephors sent him along to the senior hippagretes (3. 3. 9) who was to give him some of the young (or younger, νέων/νεωτέρων: 3. 3. 8f.) men for his assumed mission to Aulon. The men concerned are not specified as hippeis though they must have been of military age and presumably in their twenties; Cinadon himself is described as τὸ εἵδος νεAανίσκος (3. 3. 5) and as serving in the army (3. 3. 7). What Xenophon says, or rather does not say, about the hippeis and their role is disconcerting enough. No less worrying is the fact that, with the single possible exception noted above, the hippeis are nowhere mentioned in connection with battles he describes (often in some detail) which involved the participation of Spartan kings. Not only that, but Xenophon's evidence strongly suggests that the king and his companions (whoever and however many they were) fought as part of one of the morai, presumably the first, or leading, -291mora ( Lac. Pol. 13. 6). When Pausanias attacked the democrats in the Piraeus in 403 he did so initially with two morai (H. 2. 4. 31). He was with those two morai himself as the narrative shows: when the democrats began giving him trouble, Pausanias ordered τὰ δέκα ἀΦ ἣßńς to charge out (2. 4. 32). He himself followed with the others (σὺν δὲ τοîς ἐπńκολούΘει). Very soon afterwards the king found himself in trouble. Both polemarchs fell as well as a number of other Lacedaemonians and Pausanias was forced to retreat for the moment (H. 2. 4. 33f.). In 394 Agesilaus was returning from Asia when he met the allies at Coronea. He had been reinforced by one and a half moral for the battle. Again, no mention is made of the hippeis (H. 4. 3. 15 ff.; Ag. 2. 6 ff.). Even more surprising is the absence of the hippeis from H. 4. 5. 6ff.: Agesilaus was in the Corinthia in 390 when the decimation of the mora by Iphicrates was announced to him. Calling his officers together Agesilaus ordered the army to follow as soon as possible while he himself set out at once σὺν τοî δαμοσίαν( Lac. Pol. 13. 1, 7 explains who these were). The δορνΦροι also followed in haste. If these are not Agesilaus' bodyguards then they must be the men who had been guarding the prisoners (H. 4. 5. 6). Where are the hippeis? Xenophon does, however, on two occasions refer to the men who fought before the king, but without calling them hippeis or giving them a number. The thirty or so Spartans who fell while defending Archidamus in 364 (οἱ μαχόμενοι πρὸ αὐτοω+: H. 7. 4. 23) are described as agathoi and as being from amongst the most prominent of the citizens (H. 7. 4. 24). The same emerges from Xenophon's account of the Spartan rout at Leuctra (H. 6. 4. 13f.). The men who fought before Cleombrotus (οἱ πρὸ αὐτοω+ μαχόμενοι : H. 6. 4. 13) were Deinon the polemarch, Sphodrias τω+̑ν περὶ δαμοσίαν, his son Cleonymus (cf. H. 5. 4. 33), καὶ οἱ μὲν ἵπποι καὶ οἱ σΥμΦορεις τοω+ πολεμάρχοΥ καλούμενοι (H. 6. 4. 14). ἵπποι in the text is perhaps a corruption of ἱππεîς, but even so we see that (a) they were only one group of men amongst others who fought with the king, and (b) they may have

belonged to the polemarch like the σΥμΦOορεîς rather than to the king. The men who fought and fell with Cleombrotus were the very best of the Spartans, as Plutarch puts it ( Ages. 28. 8). Isocrates too says that the state appointed the most distinguished of the citizens to guard the king in battle ( Letter 2. 6). Xenophon's evidence, therefore, does not -292-

port the claim that a corps of 300 hippeis acted as a royal bodyguard. What is more, it conclusively proves that the king fought within one of the morai and not separately with the hippeis. The number 300 is in any case untenable. The system was that 300 young men were selected from a (presumably) much larger number of men between the ages of 20 and 30 ( Lac. Pol. 4. 3). However, by the time of the Peloponnesian War Spartiate numbers had become too small to cope with such a procedure (in the 370s there were no more than some 1,200 Spartiates of military age: H. 6. 4. 12, 15, 17); unless of course what we find outlined in the Lac. Pol. is the 'Lycurgan' theory, and practice fell far short of it, at least as far as the strength of the hippeis was concerned. This is not unlikely. The Lac. Pol. belongs to a genre. Within that genre Xenophon, whose stated aim (1. 12) was to show how the Spartans attained pre-eminence in Greece by adhering to the laws of Lycurgus, might well not feel inclined to point out that the Sparta of his time no longer had, say, 500-600 young men in their twenties. What then can we say about the 300 hippeis and their function? To be dogmatic would be inapposite as well as dangerous. Nevertheless, all the indications seem to point to this conclusion: the 300 hippeis were an ancient 'Lycurgan' institution meant to encourage healthy rivalry among the young Spartans. Once they had indeed been hippeis, but they were so no longer; they merely kept the title (Strabo 10 C 481-2). Once they had numbered 300, but from the middle of the 5th century at the latest they could no longer so do. However, the name, the 300 hippeis, was retained, and thus they were known at home and abroad. It was not their function to act as a royal bodyguard by nature of their office. Yet in elitist Sparta (and this was an elitism based only partly on merit; birth and wealth almost certainly counted for more) the eminent men who fought with the king must largely have coincided with the hippeis, past and present. Thucydides' statement at 5. 72. 4 about Agis and the 300 hippeis so-called should be seen in this light. It might be objected that perhaps a change took place between the time of Thucydides and that of Xenophon. Institutions of this kind, however, in an ultra-conservative state like Sparta rarely admit of radical change. After all, the title, hippeis, persisted long after the knights had lost their mounts. ( Kelly, 1981: 31 ff., suggests that the lochos of Amompharetus at -293-

Plataea mentioned in HerodotuṠ was in fact the 300 hippeis. The suggestion is based on a series of unsubstantiated assumptions. To voice but three objections: (a) Herodotus does not say that the Pitanate lochos numbered 300 men called hippeis -- yet he was aware of both of these items of information as we have seen. (b) The impression one derives from Herodotus is that Pausanias and Euryanax were not stationed with the Pitanate lochos. (c) As Den Boer, 1954: 288ff., argues, the MSS readings ireas and irees at Herod. 9. 85. 1-2 are preferable to Valckenaer's emendation irenas and irenes preferred by Kelly. Kelly, however, is right to criticize some of Den Boer's conclusions: all men who fell in battle and women who died in childbirth were exempt from the ban on named tombstones as Plut. Lyc. 27. 2 says, and the text should not be emended (as it is by Den Boer) to mean that only priests killed in war and priestesses could have inscribed tombstones. Also, the four Spartiates Herodotus names as buried in the first Lacedaemonian grave at Plataea were not the only occupants of that grave; Herodotus was clearly being selective. Nevertheless, neither point can lend support to Kelly's view that the first grave contained irenes (i.e. hippeis as he sees it) rather than irees.) 32. 2. 'AÁΘńναîοι . . . διακοσίους: Xenophon's failure (H. 5. 4. 35 ff.) to mention the Athenian presence tallies well with his policy of ignoring as many as possible of the anti-Spartan actions of the Athenian ßοιωτιάζοντες (25. 3-27. 3n.). On D's evidence, this was in fact the second time in 378 that the Athenians had marched out to aid the Thebans. For the first time, in midwinter 379/8, see 25. 3-27. 3 n. 32. 3. τοὲτΩν δὲ τω+̑ν δυνάμεων κτλ:D gives only one episode, albeit the most famous, of the campaign of 378. For a less inadequate picture of Agesilaus' operations in Boeotia we must turn to Xenophon's account which, however, is largely an apologia for Agesilaus' lack of success. The Theban strategy, Xenophon informs us, was a defensive one. A large part of the arable land was fenced off with ditches and stockades. For the novelty of this see Anderson, 1970: 134; Hanson, 1983: 67ff. For the likely course of this field fortification see Munn, 1987: 114 ff. He suggests that it ran from Cynoscephalae in a south-easterly direction keeping north of the Asopos. Behind these lines the Thebans and their allies kept guard. Finally, Agesilaus succeeded in passing within the lines and laying waste the land as far as Thebes. It was at this -294time that Chabrias' exploit took place though Xenophon does not mention it. λόΦ2ον τινά κτLAλ: we therefore know these facts about the hill: (a) it was fairly low and not too steep as Agesilaus tried to attack the enemy. (b) It was fairly extensive for it accommodated a large number of troops. (c) It was about 20 stades from Thebes (cf. Aristid. Panath. 284 Dind. πρὸ τńς πὸλεως τω+̑ν őńßαίων) and (d) it was oblong in shape and, occupied by the enemy, it appears to have frustrated Agesilaus' plan to march where he had wanted to march. Now we know that Agesilaus penetrated behind the lines at Cynoscephalae (X. Ag. 2. 22), a place which cannot unfortunately be located independently or with any certainty. Pindar's biographers cite it as the place of his birth and further say that it was between Thespiae and Thebes, but belonging to the latter (cf. Steph. Byz. s.v.). Its name signifies perhaps a group of hills (cf. RE xx. 1608f.; va. 1447f.; Suppl. x. 355; Munn, 1987: 113 and n. 26). That it was on the way to Thebes from Thespiae and close to the frontier is shown by X. H.

5. 4. 15: in the winter of 379/8 Cleombrotus advanced from Thespiae to Cynoscephalae and remained there for sixteen days watching the attitude of Thebes. The trouble with identifying D's hill, even if we could locate Cynoscephalae, is that we do not know what Agesilaus' movements were once over the trench and stockade. The most convenient (though not quite direct) way of reaching Thebes from Thespiae is that followed by the presentday road which proceeds down the long Kanavari valley in a north-easterly direction. Once in the northern (or Aonian) plain, the road proceeds east to Thebes. Immediately to the west (less than a mile away) of the mouth of this valley there was situated the Cabeireion in 'a small valley or hollow in the hills opening northward on the Teneric plain . . . The hills on either side are low, bare, and treeless' (Frazer, v. 136; cf. Buck, 1979: 13f. On the Cabeireion see Schachter, 1981-94: ii. 66 ff.; Fossey, 1988: 211f.). It is possible that Agesilaus tried to reach the northern (Aonian) and western (Teneric) Theban plains, at the same time threatening Thebes, by marching down the Kanavari valley. If he did, the low hills of Pouridi (or Gyftoaidema) or Ntarani south-west of Thebes on either side of the road to Thespiae are good candidates for the elevated ground occupied by the Thebans and Athenians. The allies had perhaps hurried back and had succeeded in overtaking -295Agesilaus because the king, busy with plundering and laying waste the land, had proceeded more slowly. But perhaps the allies did not succeed in intercepting the Spartans until the valley of the Cabeireion and it was there that Agesilaus received his famous check (see further 33. 4n.). If this is what happened we may assume that Agesilaus was prevented from reaching the plains west and especially north of Thebes, and that he had to confine his activities to the south and south-west of the city for the time being. All the above suggested locations satisfy D's information that the hill was about 20 stades from Thebes. It is possible nevertheless, as suggested by Munn, 1987: 112 ff., that with such a large force, both infantry and cavalry, Agesilaus decided not to risk marching through the relatively narrow Kanavari valley, but preferred instead to move through the fairly open countryside to the south which leads from Leuctra in the direction of Thebes, passing on the left a ridge of hills (Rakhi Kendani) some 4 miles south-west of Thebes which Munn identifies with Cynoscephalae. Munn then identifies the elongated crest known as Konizos, some 2-3 miles southwest of Thebes, with D's hill (1987: 119 f. and see his map p. 114 ). His description of how the campaign developed at this stage may well be correct. This n. (as well as the one at 34. 1) is based in part on the 1: 100,000 Greek General Staff map and on a personal visit to the area west of Thebes in August 1977. For this region of Boeotia see Philippson, i. 2. 500ff.; Wallace, 1979: 5f., 67; Fossey, 1988. 32. 4. τὸ μὲν πρω+̑τον τοὺς ΫιλοΥς κτλ: D and Nepos, Chabr. 1. 2, agree that there was a skirmish of light troops before Chabrias' exploit and we should see Polyaen. 2. 1. 2 in that light too. àποκροΥσαμεξνων: Nepos, Chabr. 1. 2, on the other hand says that Agesilaus was victorious in the skirmish. Polyaen. 2. 1. 2 seems to support D. εκ τóπων úπερδεξíων: see 25. 3 n. 32. 5. των μισζοΦóρων àΦńγοúμενος: the 5,000 infantry and 200 cavalry of section 2 were Athenian citizens. But there were mercenaries present too under the specific command of Chabrias, to judge from the present passage and from X. H. 5. 4. 14, 54. There was at least one other Athenian general, Demades or Demeas, who may be the same as the Demophon who led the midwinter expedition to free the Cadmea (26. 2 n.). Neither D nor the rest of the sources (Polyaen. 2. 1. 2; Nepos, Chabr.1; Demosth. 20. 76) -296-

exclude the presence of other generals. What is more, D's statement does not preclude the possibility that Chabrias was in command both of the mercenaries and of the Athenian troops, the latter along with Demophon ( Demades/ Demeas) and perhaps others. Indeed the other sources are explicit that he commanded Athenian troops. παρήγγειλε τοîς στρατιώταις κτλ: who performed the stratagem? Clearly hoplites as they were armed with shields and spears, standing in a phalanx formation, and about to receive the attack of Agesilaus' hoplites. Demosth. 20. 76 and Polyaen. 2. 1. 2 are explicit that Athenian citizens were involved, while the latter includes the Thebans too under Gorgidas. Likewise Nepos, Chabr. 1. 2, speaks of the phalanx as a whole as having carried out Chabrias' instructions. D's Xαßρíας δ ὁ 'AΘńναîς τω+̑ν μισΘοΦóρων àΘńγοúενος Παρńγγειλε τοîς στρατιωταις is ambiguous. It could mean that the order was given to the mercenaries alone whose commander he was, or it could mean that Chabrias, whose specific duty on this campaign was to command the mercenaries, gave the order to the soldiers in the phalanx in general. D's account as a whole supports the second interpretation: the Thebans and their allies were occupying a strong position. Agesilaus advanced against them, but following Chabrias' tactic he decided to retreat (sections 3-6). The evidence of Demosthenes and Polyaenus too, as we have seen, asserts that the citizen troops were involved. The question, therefore, whether the mercenaries present also performed the stratagem becomes of small importance. If they did, then they must have been hoplites and not peltasts (cf. Anderson, 1963: 412). But it appears from X. H. 5. 4. 54 and Nepos, Chabr. 1. 2, that the mercenaries were peltasts, not hoplites. They were almost certainly the same body of peltasts as that at H. 5. 4. 14, some six months earlier. What exactly was the tactic? There is no real problem here. As Agesilaus' hoplites charged uphill, the enemy hoplites, smartly and all together (ων ποιńσáντων τò προσταχΘεν ωσπερ àΦ' ενòς παρακελεúσματος: section 6), stood at ease, i.e., while still keeping their formation they downed shields, resting them against their knees. Their spears remained upright. Their stance was exactly that of the great bronze statue of Athena Promachos on the Acropolis. The 'stratagem' itself was hardly novel (pace Nepos, Chabr. 1. 2). On the contrary, it was the normal way by which -297-

hoplites stood at ease (cf. X. An. 1. 5. 13). It was the timing that was crucial. The allies had been ready to receive the Peloponnesians as the latter advanced uphill, when suddenly they stood at ease -- an almost contemptuous action in the face of the advancing enemy. If Agesilaus had hoped to intimidate his opponents by his daring charge uphill and drive them off the ridge, he now realized that he had miscalculated. Intimidated himself by their εùταξíα and καταΦρóνńσις (section 6) he ordered a retreat. D's description of the tactic is the best we possess and it requires no elucidation from the other sources which have in fact misunderstood the stratagem. Polyaenus' τà μεν δóρατα ορΘα προτειναμενους 'with their spears held out upright' is wrong and even more wrong is Nepos' 'proiecta hasta' 'with spear advanced'. (Nepos is likely to have mistranslated his Greek source; cf. Themist. 9. 4 'tuam petens amicitiam' and Thuc. 1. 137. 4 διá τńν σńν Φιλíαν.) Nepos and Polyaenus have been interpreted in a way that is totally unacceptable by Müller, 1908: 46, and Parke, 1933: 77: 'At a word of command from Chabrias his soldiers went down on one knee and rested their shields against the other, and with spears pointed forward awaited the enemy's charge.' But a hoplite phalanx could hardly receive another thus. Müller's and Parke's misinterpretation of the tactic has led to a mistaken reconstruction of the statue of Chabrias; see 33. 4 n. For Parke's erroneous view that a new type of peltast carried out the stratagem see 44. 1-4n. 33. 1. σúμßολοι: no other source mentions symbouloi as accompanying Agesilaus in 378. During the Peloponnesian War symbouloi had been attached to unsuccessful nauarchs (Thuc. 2. 85. 1; 3. 69. 1; 8. 39. 2), and in 418, when Agis had displeased the Spartans by withdrawing from Argos without a battle, ten symbouloi were appointed to accompany the king (Thuc. 5. 63. 4). The regulation was temporary and the powers of the ten must have been limited (see Andrewes, HCT iv. ad loc.). The thirty symbouloi who followed Agesilaus to Asia and Agesipolis to Thrace were clearly different. They were advisers and helpers of the kings properly speaking; they were not meant to circumscribe the kings' authority in any way ( Busolt, GS ii. 676; Westlake, Studies, 257 n. 33). But symbouloi here may be a vague reference to the two ephors who generally accompanied the king in a consultative capacity (cf. X. H. 2. 4. 36; Lac. Pol. 13. 5). Andrewes ( HCT iv. 125) remarks -298that 'the ξúμβουλοι may throughout have played a large part in Ephoros' account of the campaign' of Mantinea in 418, and it may be that here too Ephorus loosely referred to symbouloi by way of setting the scene for Agesilaus' answer to the criticism of his decision not to fight. If candidates for symbouloi had to be found one could name, apart from the two ephors, the men περì δαμοσíαν mentioned at X. H. 6. 4. 14, but really all the officers and officials who took part in a council, a βουλή such as that held on the morning before the battle of Leuctra for example (X. H. 6. 4. 8). But whatever we conceive the symbouloi in this context to be we should not draw too rigid a distinction between them and the οἱ τàς ἡγεμονíας Ἔχοντες, as does Muttelsee, RE iva. s.v. Σúμβουλοι, under the influence of Kahrstedt, GS i. On the basis of D, Muttelsee contrasts sharply between symbouloi as being mere ἰδιω̑ται, and ἄρχοντες as enjoying ἡγεμονíα. Under the Spartan system, however, the only sharp distinction was between the kings and the other Spartiates, the latter being allἰδιω̑ται (X. H. 3. 4. 7; 6. 1. 14). This distinction apart it would be true to say that in the small and personal world of Sparta a man wielded as much power as his birth, wealth, and connections allowed him to. Official titles provided but a flexible framework of reference. The symbouloi of Agesilaus in Asia were ἄρχοντες/ἡγεμóνες too for they undertook whatever tasks or commands the king saw fit to assign to them (cf. X. H. 3. 4. 6, 10, 20; 4. 1. 15, 21. Kahrstedt, GS i. 162f., 200ff., 220 f., sees too much in the word ἡγεμόνíα and thinks that only D uses it correctly (!); cf. the review by Wade-Gery, JHS 46 ( 1926), 293 ff.). Α+̔γησíλαος ἀπεκρíϑη κτλ: Agesilaus' policy of invading Boeotia was criticized both by his fellow Spartans and by the Spartan allies (X. H. 5. 4. 13; 5. 4. 60; Plut. Ages. 26; Pel. 15. 3). Obviously he defended himself and we may have here a distant echo of that defence. τò παρáλογον τη̑ς τúχης: for the reference to Tyche here and in section 3, and whether it is Ephoran in origin or a Diodoran elaboration, see p. 14. 33. 2-3. This does not represent a favourable treatment of Agesilaus; see p. 119 ; and for Ephoran anticipatory rhetoric, p. 123. 33. 3. Μαντíνειαν. . . ἀπέβαλον: see I. 2n. 33.4. εἰκόνας. . . σχη̑μα: in fact only one statue was put up by the demos as the other sources say ( Nepos, Chabr. I. 3; Arist. Rhet.3. -2991411b; Aesch. 3. 243). The occasion was his sea victory over Sparta near Naxos in 376 (Aesch. 3. 243). A number of marble fragments discovered in the Agora seem to have belonged to the inscribed base of Chabrias' statue; see Schweigert, 1940: 314 ff.; Burnett and Edmonson, 1961: 74ff.; Anderson, 1963: 411ff.; Buckler, 1972: 466ff. A headless herm which has vanished from the Vatican is reported to have been inscribed with the name, patronymic, and demotic of Chabrias ( Richter, The Portraits of the Greeks ( 1965), 159). Burnett and Edmonson asserted on the basis of Müller's and Parke's mistaken interpretation of Polyaenus and Nepos (32. 5n.) that the statue represented Chabrias kneeling on one knee. This is indefensible; see the criticisms of Anderson and Buckler. As the latter points out (1972: 474), the kneeling position is unlikely in the context of Greek sculpture, while standing hoplites at rest abound and many examples date to the 4th century. Nor is the

base unusually long and narrow for a standing hoplite even if we assume a one-wreath front. The restoration of some of the dedications, especially that on fragments D and E on the front face, is not beyond dispute. Schweigert, who reconstructs a three-wreath front face, restores fragments D and E thus: [οἱ στρατÁ]ω+̑ται Ệ [. . . ] ANTEΩITΩI || [7-8]ντωι συνμαχεσáμενοι. This is the third citation under the wreath on the right. He can find no satisfactory restoration of the letters given in capitals, but he does suggest (1940: 319 n. 5) that τ[ω+̑ι μ]αντε[í]ωι τω+̑ι. . . is epigraphically possible. Burnett and Edmonson, following Wilhelm ( Anzeiger Oester. Akad. 84 ( 1947), 190 ff.), restore [οἱ στρατι]ω+̑ται οἱ ἐν τ[ω+̑ι Αἰ]αντε[í]ωι τω+̑ι II [ἐν Ἑλλησπόν]τωÁ σ[υμμαχ]εσá[με]νοι. As they prefer to reconstruct the monument with a one-wreath front (1961: 77, though in n. 7 they concede the possibility of a second wreath), the single and most important citation on this face, according to them, referred to some obscure and otherwise unrecorded event in the Hellespont. This seems rather unlikely. As D says, Chabrias was particularly proud of the tactic he employed against Agesilaus and when he was subsequently awarded a statue by the demos he asked that it should exhibit the posture assumed by his soldiers on that occasion (cf. Nepos, Chabr. I. 3). It would be a little surprising, therefore, if that famous event was not mentioned on the monument. If the position taken up by the Thebans and Athenians was in the neigh-300-

bourhood of the Cabeireion (32. 3 n.), a partial restoration of fragments D and E could be [οἱ στρατι]ω+̑ται οἱ ἐν τ[ω+̑ι ΚορΥβ]αντε[í]ωι τω+̑ι. . . The substitution of Corybanteion for Cabeireion is a little difficult, but it appears that at least from the early 5th century Cabeiroi, Corybantes, and Curetes were often identified in literature (cf. Strabo 10 C 466; 472f.; Waites, 1923: 25ff.; Nock, 1941: 580; Hemberg, 1950: 189. In late literature the confusion was widespread; cf. Clem. Alex. Protrept. 2. 19; Arnobius, Adv. nat. 5. 19). It is not impossible, therefore, that the Athenian(s) who composed the citation referred to the Cabeireion as the Corybanteion. There is no difficulty about fitting this in for fragments D and E could be further apart than shown by Burnett and Edmonson in their fig. 3. The name of the sculptor could take up a longer space, e.g. . . . ΗΣΣOΓΝ[ΙΑΔΟΓΕΠ]ΟΙΗΣΕ or . . . ΗΣΣΟΓΝ[ΙΑΔΟΓ (+demotic) ΕΠ]ΟΙΗΣΕ. The possibility remains of course that the monument had a twoor even a three-wreath front and that the success in Boeotia was mentioned further to the left above one of the other wreaths. It was not inscribed on the left or the right faces. The citations there are better preserved and they refer to other events.

33. 5-6. Second attack on Thespiae. Death of Phoebidas See 27. 4n. where it is suggested that this was the second attack on Thespiae. Sections 5-6 exhibit well the different points of view of Xenophon and Ephorus. As in the case of the Spartan defeat at Leuctra, Xenophon (H. 5. 4. 42ff.) makes the best of Phoebidas' defeat and seeks to explain it. He belittles the Theban victory, ascribing it to good luck and the cowardice of the Thespians. D's account is admittedly but a poor summary of Ephorus', but even so certain elements stand out in strong contrast to Xenophon. He concedes that the Thebans (he does not specify their strength whereas Xenophon says they were in full force), who actually assaulted the city wall, achieved nothing worthwhile, though the 200-man-strong Thespian advance guard was destroyed. Phoebidas then rushed out of the city and hastily attacked the retreating Thebans. He lost 500 men and was himself killed. 33. 5. προσβολáς: see 26. 4n. 33. 6. ἀξιόλογον: Ephorus therefore probably gave no figure; see p. 133. -301ἐπιθέμενος: cf. Plut. Pel. 15. 6; Polyaen. 2. 5. 2. λαμπρω+̑ς ἀγωνισáμενος: Diodoran rhetoric though Phoebidas probably did fight well and die bravely. Otherwise Xenophon would not have mentioned him, and he would not have added that they μαχόμενοι (H. 5. 4. 45).

34. 1-2. The second invasion of Boeotia As in the case of the first invasion, D concerns himself with but the climax of the campaign. For further information we must turn to X. H. 5. 4. 47-55 whose narrative again is one-sided and largely a Spartan apology. What principally mattered to Ephorus on the other hand (and this comes through in D) was the fact that for the second year in succession the mighty Spartans, led by the great Agesilaus, were checked in the open field. 34. 1. οὐ πολλῳ+̑ δ+̕ ὕσρε1ρον χρόνῳ: in fact early next spring 377; cf. X. H. 5.4. 47. This vague chronological indication may conceivably be Ephoran; see Introd. n. 264. τῃ+̑ αὐτῃ+̑ δυνáμει: see 3 2. In. δυσχωíας τινáς ἑτέρας: ἑτέρας because they had occupied a similar position in the previous campaign (32. 3). The present δυσχωρíαι are called Γραòς στη+̑θος by Xenophon (H. 5. 4. 50) which seems to suggest a gently sloping hill. He confirms that the place was στενòν. . . ἐπιεικω+̑ς καí δúσβατον; see also Polyaen. 2. 1. 12 κατελáβοντο λόφον ὑπρ τη+̑ς ὁδοΥ+̑ δύσβατον,καλούμενον Γραίας ἕδος κτλ, and cf. 2. 1. 24, the same incident, but wrongly assigned to 394. The exact site of Graos Stethos escapes us. It must have been south-east of Thebes, and at not too great a distance from the city for otherwise the action described by Xenophon becomes impossible: the Thebans, fearing for their city,

δρόμἳ ἔθεον εἰς τὴν πόλιν and reached there not very long afterwards. Agesilaus withdrew from Thebes on the same day and encamped at Graos Stethos (H. 5. 4. 54). The place therefore must have been not too far from Thebes and D's narrative agrees on this point. The south-east direction of Graos Stethos is indicated by Xenophon's comment on the route taken by the allies when hastening back to Thebes -- they took τὴν ἐπὶ Ποτνιὰς ὁδόν (H. 5. 4. 51). Potniae was about 10 stades south of Thebes (Paus. 9. 8. 1; Buck, 1979: 14; Fossey, 1988: 208ff.). Two further statements of Xenophon help to some extent, but -302-

also add difficulties to the identification of the site. First, he says that on returning from the Tanagran frontier Agesilaus marched, initially at least, ἐν ἀριστερᾳ+̑ ἔχων τὸ τεîχος (H. 5. 4. 49). What teichos is that? There are two possibilities: (a) Tanagra is meant (the usual interpretation). (b) Teichos is an inept reference to the trench and stockade ( Munn, 1987: 124: ff.). In either case Agesilaus was proceeding in a westerly, or rather southwesterly, direction away from Tanagra. Second, he says that while on Graos Stethos the allies had the trench and stockade behind them. With this information in mind, Bölte, reasonably enough, identified Graos Stethos with either of the hills known as Psilorakhi or Mikri Psilorakhi some 2-3 miles south-east of Thebes ( RE vii. 2. 1827f.). The road from Tanagra skirts these hills on the south side. Even more reasonably, Munn, 1987: 126ff., identifies Graos Stethos with Golemi, a hill not far from Mikri Psilorakhi, but on the south side of the road from Tanagra. Golemi is a great deal closer both to the road running north to Thebes through Potniae and to the Theban fieldworks along the Asopos river and, therefore, fits Xenophon's account even better. The trouble with both identifications, of course, is that Xenophon insists that the ditch and stockade in the rear of Graos Stethos also ran by Scolus (H. 5. 4. 49 f.); and the location of Scolus is controversial. Paus. 9. 4. 4 provides the best clue as to its whereabouts: on the road from Plataiver to Thebes the main road presumably which the Thebans did not follow in 373: Paus.9.1 6), before crossing the river Asopus, 40 stades down the river, are the ruins of Scolus (and cf. Strabo 9 C 408 Σκω+̑λος δ+̕ ἐστὶ κώμη τη+̑ς Παρασωπίας ὑπὸ τῳ+̑ Κιθαιρω+̑νι, δυσοίκητος καὶ τόπος καὶ τραχύς. It seems fairly clear from this that Scolus was on the south side of Asopos (as generally agreed: see e.g. Pritchett, 1957: 13; Burn, 1984: 510; Wallace, 1979: 87ff. Contra (now) Pritchett, 1965-85: i. 107ff.; ii. 178ff.; iii. 289ff.; iv. 97ff.; Fossey, 1988: 119ff.; Munn, 1987: 121ff., who does not nevertheless tackle the evidence of Pausanias and Strabo). But if the Theban stockade was on the south side of the river, the movements described by Xenophon and D become impossible because of the distances involved. One way out of the difficulty perhaps is to assume that the land of Scolus stretched north of the river and was consequently traversed by the trench and stockade, and that this is what Xenophon meant by τὸ κατὰ Σκω+̑λον σταύρωμα (H. 5. 4. 49); another, that the -303stockade, which encircled and protected the plains and other good agricultural land, as Xenophon says (H. 5. 4. 38), was extended south of the river at this point to take in Scolus, a place claimed by both Thebes and Plataea it would seem (Herod. 9. 15. 2; Strabo 9 C 409; Paus. 9. 4. 4). (Some scholars follow Meyer, 1909: 97 n. 2, in identifying Xenophon's Graos Stethos with the Graia of Strabo 9 C 404 (cf. Paus. 9. 20. 2; Steph. Byz. s.v. Τάναγρα and Ω+̔ρωπός) and placing it on the Theban-Tanagran frontier. This cannot be correct for it pushes Graos Stethos too far to the east. Meyer in fact is doubly wrong: Strabo's Graia was a place in Oropus near the sanctuary of Amphiaraos; cf. Buck, 1979: 19.) 34. 2. του+̑ δ+̕ Α+̔γησιλάου προςμαχἐνου κτλ: 2pe' aim was to show that the Thebans and their allies (we do not know whether the Athenians were present with as strong an army as in 378) had had the better of it, but D has butchered the action by telescoping it. The fighting began several miles south-east of Thebes at Graos Stethos (see previous n.), and the final act was played outside the walls of the city some hours later (X. H. 5. 4. 53). D has fused the two ends together leaving out the middle (for such shortcomings see p. 134 ). He suddenly faces us with the statement μετὰ δὲ ταυ+̑τα τω+̑ν ἐκ τη+̑ς Θηβαίων πανδημεί, and we are left to guess, with the help of Xenophon, that the Thebans, having abandoned the strong position they had been occupying, retreated to their city and part of them, the greater part perhaps, made their way inside the walls. Xenophon represents that the entire Theban army caused the Sciritans alone to fall back θâττον ἢ βάδην (H. 5. 4. 53). D, on the other hand, says that Agesilaus himself was involved in the fight against the Theban rearguard, and that he retreated when the rest of the Thebans came out to face him. This certainly was a better justification for the Theban trophy than the retirement of the Sciritans alone. Xenophon further claims that Agesilaus suffered no casualties. D nevertheless at 52. 5 mentions Lacedaemonian graves outside Thebes which dated to the time of Agesilaus' invasions. D's description exhibits his normal rhetorical elements; see pp. 15 f. οἱ δὲ Θηβαîοι... τότε πρώτως κτλ: cf. Plut. Ages. 26; Pel.15. Unlike Xenophon, Ephorus took great interest in the Theban leaders and the growth of Theban power; see p. 128. Related in -304-

some detail in Xenophon, Agesilaus' Boeotian campaigns achieved little and proved counter-productive in the long run; cf. Munn's good analysis: 1987: 133 ff.

34. 3-36. The war at sea

34. 3. πεζὰς δυνάμεις. . . τάλος: D passes over Cleombrotus' attempted invasion of Boeotia in early 376 for which see X. H. 5. 4. 59. πεζάς. . . ναυτικω+̑ν: Ephorus, therefore, treated the land and sea operations separately, as does Xenophon (H. 5. 4. 60 ff.). No parallel chronology is involved, however, as the sea war followed the land war. κατὰ τοὺς καιρούς: not strictly so. For the reason why all these events from 28. 1 to 35. 2 are placed under 377/6, see pp. 135f. The naval action near Naxos was fought on 16 Boedromion in the year of Charisandrus, 376/5, about Sept. 376 (Plut. Phoc. 6. 7; Cam. 19. 6; Mor. 349e-f; Polyaen. 3. 11. 2). The archon year is not given by either Plutarch or Polyaenus, but is obtained from the sequence of events in Xenophon (25. In.). διὰ τοαύτας αἰτίας: D ignores the background to the battle and gives only the immediate events leading up to it. See X. H. 5. 4. 60f.: the Peloponnesians decided to switch their attention from Thebes to Athens. They therefore manned a fleet and sent it out to starve Athens to submission. The Athenians soon found themselves under an effective naval blockade, the Peloponnesian ships having taken their stations at Aegina, Ceos, and Andros. The cornships could only reach as far as Geraestus, a small harbour at the southernmost point of Euboea. On the shortage of corn which ensued in Athens see Demosth. 22. 15. Πόλλις: a Pollis was nauarch in 396/5 ( Hell. Oxy. 9. 2; 19. I), but1 there is no proof that this is the same man. It is not necessary therefore to question our sources' information (X. H. 2. 1. 7; Plut. Lys. 7. 3) that a law at Sparta forbade a man to hold the nauarchy twice. On the Spartan nauarchy see Pareti, 1909: 71 ff.; Sealey, 1976: 335ff.; Andrewes, HCT v. 454f.; Stylianou, 1988: 467f. πυθόμενους σίτου πλη+̑θος κτλ: in contrast to Xenophon, D speaks of just one large convoy of cornships which Pollis decided to capture, and to safeguard which the Athenians sent out a fleet (not their entire fleet it would seem, and perhaps it was not even under the -305-

command of Chabrias; see below). But the difference between D and Xenophon is superficial: a large number of cornships must have gathered at Geraestus, not being able to proceed further as Xenophon says (H. 5. 4. 61), and it was these that the Athenians brought in in a convoy. But there is an important difference between D and Xenophon, which once again demonstrates the superiority of Ephorus' historical account over that of Xenophon, and it is this: Xenophon represents that the battle was fought in order to lift the naval blockade and, the Athenians having won, the blockade was lifted. If Xenophon had bothered to enquire where the battle was fought, the absurdity of what he says might have become apparent to him. For the battle was fought between Naxos and Paros and it cannot therefore have been related to the blockade. (But cf. Tuplin, 1993: 159ff. who thinks that Xenophon, in addition to underplaying the Athenian naval successes of the 370S on purpose, deliberately misrepresented the present event.) D's information is surely correct and a testimony to the worth of his source: Athens first dispatched a fleet to lift the blockade (the decision had not been easy according to Demosth. 22. 15). That was soon done, Pollis having apparently decided not to risk a battle. After that success Chabrias was sent out to the Cyclades with the entire fleet (section 4) where he began to lay siege to Naxos. Pollis now decided to intervene and the engagement was fought. ( D. Thomas, "Aristotle's Treatment of Historical Material in the Politics" (unpublished Oxford thesis, 1978), 104f., speculates from Arist. Pol. 1270b 12-14 that Pollis had been distracted for a while by an Athenian-inspired intrigue on Andros.) 34. 4. μετὰ δὲ ταυ+̑τα χαβρίας κτλ: though he omits the background to the naval war as we have seen, D gives valuable information on the events leading up to the battle (previous n.) and full information on the battle itself. In fact D and the schol. Aristid. Panath. 173. 16 Dind., are the only sources to note Chabrias' attack on Naxos and Pollis' attempt to relieve the city which resulted in the battle, though Demosth. 23. 198; 24. 180; Aesch. 3. 222, 243; Dinarch. 1. 75; Plut. Phoc. 6. 5; Cam. 19. 6; Polyaen. 3. 11. 2, 11, confirm D's locality by labelling the battle as that of Naxos. Naxos was clearly hostile to Athens before the battle. But had it already been admitted to the Athenian Confederacy and had it perhaps defected when Pollis appeared in the Aegean? Some -306scholars have thought so, though no evidence can be adduced to support this (cf. Beloch, GG iii. 1. 152). Certainly Tod 125 (=IG ii2 1635 -- new text with nn.: Inscriptions de Délos, no. 98), the accounts of the Athenian Amphictyons of Delos of the years 377-3, constitutes no such evidence ( pace Accame, 1941: 81ff.; Hornblower, 1982a: 190) for the states mentioned thereon of which one is Naxos, because the Myconians, Ceians, Siphnians, and Andrians joined the Confederacy not earlier than the late summer or autumn 375; they are entered on the left side of Svt 257 =Tod 123, but below the Acarnanians and Cephallenians (who joined in the second prytany of 375/4: Svt 262=Tod 126). Naxos is not to be found on Svt 257=Tod 123. On the other hand her accession to the Confederacy after the battle is perhaps indicated by Svt 321=IG ii2 179 which speaks of legal agreements between Athens and Naxos. Some of the literary sources refer vaguely to a large number of the Cyclades as having acceded to the Confederacy after the battle (Demosth. 20. 77, 80; Plut. Phoc. 7. 2). This may be an exaggeration, but it is in line with what has been suggested above (30. 2n.) about the possible number of names missing from the front of Svt 257. It is possible that one of the missing names was that of Naxos. Cargill, 1981: 37, leaves the question of Naxian membership open. If already an ally, Chabrias' attack on Naxos need not mean that the island was in revolt from Athens. An alternative explanation for the attack is that a pro-Spartan group on the island had been co-operating with Pollis.

Similarly, Aegina, Ceos, and Andros were used as bases by the Peloponnesian fleet (X. H. 5. 4. 61). An inkling of anti-Athenian activity in the Cyclades, occasioned apparently by Pollis' presence, is provided by Tod 125 lines 134ff., where it is said that some Delians in 376/5 (or shortly before) violently ejected the Athenian Amphictyons from the temple of Apollo. Pro-Athenians in the area included the Delian Pythodorus honoured with proxenia at Athens in 369/8, no doubt for 'services rendered', at the suggestion of Phoxias and Epicrates, the latter probably the same as the Amphictyon of that name for the period 377-373 ( Syll. i. 158; Tod 125 lines 10f., 62f.); see Osborne, 1974: 171f., who points to the very unusual provision that the money to pay for the inscription of a second copy of the decree to be set up in the temple of Apollo on Delos was to come from the sum exacted παρὰ τω+̑ τὰς ὀΆλόντων. He may very -307-

well be correct that these are the same as the people mentioned in Tod 125 lines 134ff. Pythodorus' proxenia was extended to his nephew in 363/2 and Osborne may also be correct that this was part of a policy to offset the results of the Theban naval venture in the Aegean (1974: 173f.). χαβρίας. . . ναύαρχος: in fact a strategos. D is using the term in its non-technical sense here as at 36. 5 and 18. 15. 9. Xenophon too occasionally uses nauarch in its general sense (cf. H. 5. 1. 13). Yet D's terminology in this respect is usually correct. He consistently calls the Athenian generals strategoi and the Spartan admirals nauarchs. With the exception of the three instances above no other case of an Athenian general termed nauarch is to be found, and there is but one instance of a Spartan nauarch called a strategos (47. I). The plural nauarchoi at 13. 99. 1 is understandable; and cf. 13. 76. if. and 13. 97. 5f. where Athenian strategoi and Spartan nauarchs are nicely contrasted. Occasionally D is very accurate as at 43.6 and 45.2 where the exact title of appointment seems to be given. In these cases D has departed from his source least. μηχανάς: of what kind? As the walls were shaken battering-rams must be meant (κριοί), perhaps protected by 'tortoises' (χελω+̑ναι), though stone-throwing catapults cannot be excluded. Aeneas Tacticus (32f.) writing just before the middle of the 4th century mentions all three as employed in sieges. For Greek siegecraft see Garlan, 1974. Ephorus evidently gave sieges some attention. He added interesting information to Thucydides' description of the siege of Samos in 441/0: 12. 28. 2f.; Plut. Per. 27. 3, who cites Ephorus as his source; see Garlan, 1974: 132ff.; Lawrence, 1979: 42. And the siege of Perinthus in 341/0, with which Ephorus brought his Histories to an end, is recognized by Marsden as technically accurate: 1977: 217f. 34.5. ἑξήκοντα καὶ πέντε: X. H. 5. 4. 61 says 60. ὀγδοήκοντα καὶ τρεîς: though the other sources are silent on the Athenian numbers (Aesch. 3. 222 should be dismissed) D's figure is perfectly reasonable and quite consistent with the epigraphic evidence. Approximately 100 hulls are recorded on IG ii2 1604 which dates to 379/8 (29. 7n.). Many of these are described as old and, when we also take into consideration that Athens was none too wealthy at this time, we may safely conclude that the number of triremes which could be manned and sent out would have fallen considerably short of the total number of hulls recorded by -308-

the dock superintendents. It may further be assumed that some of the 83 triremes would have belonged to the allies. For the obligatory allied contribution of both men and ships see 29. 7 nn. Κήδων: Plut. Phoc. 6. 5 says that Phocion commanded the left wing and that he was victorious after some sharp fighting. D's Cedon on the other hand was killed and the left wing had to be rescued by Chabrias who proceeded to win the battle. This does not seem to be a case of D inverting a pair of names (for which see p. 138 ) for the details are different. It is better to suppose that Plutarch's version of the battle was different from that of Ephorus. Different and incorrect; Phocion was too young in 376 to have commanded the left wing. As he was 80 in 322/1 ( Plut. Phoc. 24. 5) he must have been about 26 in 376 (cf. Plut. Mor.791f; 819a; Polyaen. 3. 12; Nepos, Phoc. 2. 1). One might conjecture that he had served under Cedon and that perhaps he had played some prominent part in the battle after Cedon's death; or perhaps he commanded the triremes dispatched by Chabrias from the right to aid the left. D is our only source here and only one other Athenian Cedon is known from the 4th century ( IG ii2 2308a). A third one in Archaic times is mentioned by Arist. A th. Pol. 20. 5; Rhodes, CAAP ad loc.; cf. PA 8281; Develin, 239. ἀναρρήττων: 'breaking open'. An unusual present tense of ἀναρρήγνυμι. For D's use of it see McDougall s.v. ἀναρρήττειν. ἃ δὴ κατιδὼν ὁ Ξαβρίας κτλ: Polyaen. 3. 11. 11 relates a stratagem by which Chabrias won the battle. This is, however, wholly inconsistent with D and Plut. Phoc. 6. 5 and rather unlikely in itself. 35. 1. This reads somewhat like an apologia: Chabrias could have done more; he could have destroyed the entire enemy fleet, had he not desisted in order to pick up the citizens who were in the sea. We know that the honours bestowed on Chabrias after the battle were unsuccessfully attacked by Leodamas as unconstitutional (Demosth. 20. 146). An attempt may have been made by the general's enemies to belittle his victory by pointing to the fact that the Athenians had lost almost as many ships as the Spartans. This section thus may be an echo of Chabrias' defence. Leodamas is named by Aesch. 3. 139 as a leading pro-Theban and the attack on Chabrias may indicate a struggle in Athens between pro- and antiThebans. Chabrias had allowed the Spartans to escape for political considerations, his enemies may have claimed. -309αἰτιασάμενος . . . ἔθαφαν: X. H. 1. 6. 35ff. is almost certainly correct that the anger of the people against the generals in 406 was due mainly to their failure to rescue the thousands of Athenian shipwrecked who were thus left to drown. Chabrias was anxious to prevent the same disaster from happening again and hence, as D himself goes on to say, 'refraining from pursuit, he gathered up the bodies of his fellow citizens which were afloat, saved

those who still lived, and buried the dead' (Sherman). It is interesting to observe that D has repeated the same mistaken, or at least incomplete, view first expressed at 13. 101. 1 f. Ephorus should therefore be held responsible? Cf. Andrewes, 1974: 115. 35. 2. Α+̔θηναίων . . . ὀκτωκαίδεκα: the other sources are silent on Athenian losses. Λακεδαιμονίων . . . ὀκτώ: Demosth. 20. 77 speaks of 49 enemy triremes captured which is certainly an exaggeration; unless the figure represents the total number of ships captured by Chabrias during his entire career. The schol. Aristid. Panath. 173. 16 (iii. 282 Dind.) agrees with D on the number of enemy vessels sunk, but mentions only two as captured; cf. IG ii2 1606 (374/3) and IG ii2 1607b (373/2) where five triremes are listed as captured by Chabrias. μεγάλης ἀποδοχη+̑ς: he was rewarded with ateleia, the privilege to be enjoyed by his descendants (Demosth. 20. 75 ff.). A statue was set up in his honour in the agora (33. 4n.) and he was awarded a crown (Demosth. 24. 180). Athen. 4. 165e claims that when he died the Athenians spent 1,000 drachmas on his funerary monument. Πελοποννησιακὸν κτλ: Ephorus' comment, see p. 116 n. 312.

35. 3. A chronographic notice ἅμα δὲ τούτοις πραττομένοις κατὰ τὴν Ι+̔ταλίαν κτλ: again, the entry on Roman affairs is very brief and at the end of a year's account and it is possible that it derives from the Greek chronographer; see pp. 26 f. M. Manlius Capitolinus was the hero of the defence of the Capitol against the Gauls (14. 116. 6). His attempted coup (dated 385/4 V) was a consequence of the social and economic problems which seem to have beset Rome at the time. For detail, most of it probably legendary, see Livy 6. 15f., 18-20. Cf. Perl, 1957: 44, 123 f., 127, 154; Cornell, CAH2 vii. 2. 231 f. -310-

Archon Charisandrus, 376/5 36. 1. Ρ+̔ωμαι+̑οι κτλ: 384 V; CAH2 vii. 2. 637. Once again the MSS indicate that the fasti employed by D were a great deal more sound to begin with than they are at present in D's text, and of a common origin with those used by Livy. MS M, by contrast with P, reads Σερούιον Κπρνήλιον Σουλπίκιον. In this way the number of names is brought into line with the number announced though we are still two names short of the six given by Livy 6. 18. 1. These names probably dropped out by mistake: pp. 27 f.; Drummond, 1980: 60, 64f.

36. 1-4. Chabrias in the north Aegean 36. 1. Τριβαλλοί: the Triballians, who were perhaps distinct from both Illyrians and Thracians, lived south of the Danube, between the Isker and the Morava; see Papazoglu, 1978: 58ff., 67ff., and map facing p. 664. σιτοδείᾳ: Ephorus' reason for the Triballian raid has been generally discounted in favour of other explanations such as pressure exerted on the Triballians by the Celts or the Autariatae. These explanations cannot, however, stand ( Papazoglu, 1978: 13 f. and n. 10). The Triballians were certainly not migrating. They were on a plundering raid and it is not impossible that famine had set the tribe moving south in search of food as D says. The great distance between the Triballian homeland and Abdera should not deter us from accepting this explanation ( Papazoglu, 1978: 14). For the Triballians' hostile actions near and far see Isocr. 12. 227. πανδημεί: a Diodoran word, see p. 16. 36. 2. πλείους τω+̑ν τριμυρίων: cf. Aeneas Tact. 15. 8 ἐνίκων πλη+̑θος πολύ καὶ μάχιμον. οἱ Α+̔βδηρι+̑ται κτλ: Ephorus knew that the tribesmen were not the neighbours of the Abderitans. They first had to traverse a part of Thrace before reaching Abdera. Their route to the Aegean was probably the valley of the Nestus. οἱ δ+̕ Α+̔βδηρι+̑ται κτλ: Aeneas Tact. 15. 8 speaks of a pitched battle. But the battle is more likely to have been as D says: the Abderitans suddenly fell on the retreating and disorganized Triballians. πλείους τω+̑ν δισχιλίων: cf. Aeneas Tact. 15. 8 κστέβαλον πολλούς. 36. 3. οἱ δὲ βάρβαροι . . . πάλιν ἐνέβαλον κτλ: there is a disagree-311-

ment here between D and Aeneas Tact. 15. 9. The latter, mainly concerned with stressing the need for caution, speaks of an ambush laid by Triballians into which the Abderitans fell. D on the other hand envisages a pitched battle during which the Thracian allies of Abdera went over to the enemy with the result that the Abderitans were defeated. Both D and Aeneas agree that the Abderitans' first easy victory had caused them to regard the Triballians with undue confidence and that their losses were very heavy indeed. Perhaps the two accounts should be combined: the Abderitans, two successive levies, fell into an ambush and abandoned by their Thracian allies

were annihilated. According to a garbled scholion on Aristid. Panath. 172. 7 (iii. 275 Dind.), the people of the neighbouring Greek city of Maroneia aided and abetted the Triballians against Abdera. The motives of the Maroneitans are not difficult to discern, if the report is true. The two cities had long competed for trade and influence in the area and while for most of the 5th century, as both coins and the Athenian tribute lists show, Maroneia lagged far behind Abdera, during the first half of the 4th century she seems to have drawn level with her rival ( Schönert-Geiss, 1979: 437 ff.). To employ the Triballians was a masterstroke. The magnitude of the Abderitan disaster is reflected in the coinage. It would seem that the city closed its mint for several years and when its coinage was resumed it was of a different type. Its trade and supply of silver must have been dislocated; cf. May, 1966: 242; Schönert-Geiss, 1979: 450. 36. 4. Ξαβρίας ὁ Α+̔ θηναι+̑ος κτλ: Chabrias' expedition to Thrace dates either to autumn 376, after the battle of Naxos, or, better, to spring or summer 375. Abdera is entered on the left side of Svt 257=Tod 123. All the entries on the side probably date to the same time (almost all were cut by the same hand; see 30. 2n.), i.e. summer/autumn 375, and perhaps we should not allow too much time to elapse between Chabrias' intervention and the city's accession to the Confederacy; and see section 5n. The only other source to refer to Chabrias' intervention is the schol. Aristid. Panath. 172. 7; 173. 17 (iii. 275, 282f. Dind.). But the scholiast has hopelessly confused it with earlier events in Thrace which likewise involved Chabrias. Aristides' own chronology in sections 172f. is sound on the whole. In section 172, in enumerating the Athenian achievements during the Corinthian War, he mentions the conciliation of the Thracian kings in 389 -312-

Stylianou, 1988 for the date) which, according to our main sources, was due to Thrasybulus of Steiria ( X. H. 4. 8. 26; D 14. 94. 2; cf. Lys. 28. 5; Isocr. 5. 6; Arist. Pol. 5. 1312a). In section 173, in praising Athenian energy following the liberation of Thebes, he touches on Chabrias' successes in the north Aegean after the battle of Naxos. In commenting Rehdantz, 1845: 63, thinks that the events mentioned by the scholiast date to the on 172. 7 scholiast A writes: time of Thrasybulus' expedition to the Hellespont. Schaefer, 1885-7: i. 43 f., on the other hand, ascribes them to 375 and Chabrias' operations in Thrace in that year. Both are partly right and partly wrong. When Thrasybulus sailed to Thrace in 389, Ergocles and, it seems, Chabrias too served with him as generals (Lys. 28. 4 f., 12; IG ii2 21=Svt 238A; PA 7310, 15086; Beloch, 1884: 314). What is more Chabrias appears to have been instrumental in the treaty between Athens and Seuthes. IG ii2 21 is badly preserved, but Chabrias' name appears three times in it. It is therefore reasonable to assume that he played a (leading) part in the conciliation between Seuthes and Amedocus too (about the latter cf. IG ii2 22=Svt 238B), though as he was junior to Thrasybulus, only the latter is mentioned by the literary sources. The scholiast however seems to have had additional information. But unfortunately Chabrias' involvement in Thrasybulus' diplomacy in the Hellespont has led the scholiast to confuse these earlier events with Chabrias' intervention in Abdera in 375: the first part of his comment on 172. 7 (as well as that on 173. 17), Αὐδηρίταις . . . Ξάλης, belongs to 375. The second, καὶ διαλλάξας . . . ἐποίησε, to 389, the time of Thrasybulus' expedition. (We may note the closeness in language between the scholiast here and X. H. 4. 8. 26 ἀλλήλοις μὲν διήλλαξεν . . . ἐποίησε For the scholiast and Maroneia see previous n. φρουρὰν . . . καταλλιπών: Abdera had been depleted of troops and the introduction of the garrison must have had the consent of the synedrion of the allies and the Abderitans themselves. The promise in Svt 257=Tod 123 lines 21ff. concerned unilateral Athenian action; see 28. 3 n. -313αὐτὸς . . . ἐδολοφονήθη: not so of course and D himself records Chabrias' death in battle at 16. 7. 4 under 358/7. Chabrias in fact is mentioned later in Book 15 at 68. 1 f.; 69. 1 ff. Meyer, GdA v. 396, thinks that this is a mistake on the part of D due to the severe abridgement of his source, and he may be right. In a lengthier narrative Ephorus perhaps spoke of the assassination of the garrison commander installed by Chabrias. It is even conceivable that the man murdered was that Chales, king of the Triballians, mentioned by the scholiast (see above), and that D misread Chabrias for Chales. But however that might be D's abridgement must still be made to bear most of the blame.

36. 5-6. Timotheus in the Ionian Sea 36. 5. Τιμόθεος δὲ κτλ: as this stands in the text it seems to lead straight on from, and to be in some way connected with, the events at Abdera. Perhaps on Chabrias' return to Athens Timotheus was put in charge of the fleet and sent west; hence D's expression παραλαβὼν τὴν ναυαρχίαν. ναυαλαβχίαν: see 34. 4n. Κεφαληνίαν, τὰς . . . πόλεις προσηγάγετο κτλ: for the admission to the League of these states, including Corcyra which D does not mention (see below), see n. ad 30. 2. Cargill does not believe that Corcyra and three of the four Cephallenian cities joined the League, but, as we have seen (30. 2n.), his use of the epigraphic evidence is unsatisfactory. Equally unsatisfactory is his reading of the literary evidence. 'The sources', he says, 'use phraseology the precision of which is significant' (1981: 69). When Xenophon tells us that Timotheus Κέρκυραν . . . ὑφ+̕ ἑαυτῳ+̑ ἐποιήσατο though he did not enslave or exile anyone or change the laws ( H. 5. 4. 64) he is 'hardly describing the recruitment of a League ally' for did not the decree of Aristoteles in any case and as a matter of course forbid such treatment? 'Xenophon's point can only be that Timotheos was showing unusual generosity to a conquered state' (1981: 70). Isocrates 'supports this interpretation' by grouping Corcyra with a number of states which were captured by Timotheus, but clearly not admitted to the League (15. 107 f.). Later sources too 'share the careful language of the

contemporary authors'. Nepos writes of Corcyra being brought back 'sub imperium Atheniensium' ( Tim. 2. 1) while D omits any mention of Corcyra from his account -314-

(1981: 71 f.). But this should not make us doubt the account's reliability, for this is 'the proverbial exception that proves the rule: Diodorus is discussing new League members recruited by Timotheos, and Korkyra was not one of them' (1981: 71). Of course, the assumption that D is discussing the recruitment of new League members is not justified by 15. 36. 5, but is based on what D says elsewhere and on other evidence. In fact, none of the sources cited by Cargill is explicit in the passages quoted that the states mentioned did or did not join the League. When Xenophon referred to Timotheus' humane treatment of the Corcyraeans the last thing on his mind it would seem was the Athenian League. What was on his mind surely was Chares' misbehaviour on Corcyra fifteen years later (for which see D 15. 95. 3) and he wished to contrast the latter with the former (see also 46. 1 n.). This is not to deny that Corcyra was in all probability taken by force (see below). But this does not support Cargill's case. After all, on his view too (and as Svt 262 makes perfectly plain), Corcyra was accepted for membership after its capture by Timotheus! As for D, his account is less straightforward than Cargill imagines (see below). Nor is Cargill's treatment of it consistent. Why see such great significance in the absence from the passage of Corcyra, 'the proverbial exception that proves the rule', but disregard D's clear statement that Timotheus won over (i.e. for the League) the cities (all four presumably) of Cephallenia? τὰς κατὰ τὴν Α+̔καρνανίαν: see Svt 262=Tod 126 and Svt 257 (=Tod, 123) B line 10. The Acarnanian decision to join the Confederacy would in fact have been a federal one. For the Acarnanian Confederacy see Larsen, 1968: 89 ff. Α+̔λκέταν: Alcetas, the king of the Molossians, and his son Neoptolemus are shown on Svt 257 (=Tod 123) B lines 13-14. The inclusion of Neoptolemus may represent an attempt by Alcetas to strengthen his son's chances of succession. Even so on his death the succession was disputed and Neoptolemus had to accept his younger brother Arybbas as co-ruler; cf. Tod, ii. pp. 216f. Nepos, Tim. 2. 1, says that 'Timotheus sociosque idem adiunxit Epirotas, Athamanas, Chaonas omnesque eas gentes, quae mare illud adiacent.' One might surmise from this that, apart from the Molossian tribes, Alcetas held sway over a number of neighbouring tribal groups too for no Athamanes or Chaones are shown on Svt 257=Tod 123; but this is hardly conclusive (30. -3152n.). The term 'Epirotas' does not seem to have been used earlier than the second half of the 4th century to describe the peoples of the area which was known as Epirus later. Ephorus and Theopompus spoke of the Η+̔πειρωτικὰ ἔθνη in which they included Chaonia in north-west Epirus (on the coast) and Athamania in south-east Epirus (inland); cf. Strabo 7 C 321, 323 f., 326; 8 C 334. The Athamanes though seem to have generally acted separately from the other Epirotan tribes (cf. 14. 82. 7; 16. 29. 1). But Nepos cannot be trusted to have reproduced his source faithfully enough to allow fruitful speculation here. For the Epirotan tribes see Hammond, 1967a: esp. 443 ff. and chs. 11 and 12 with maps on 464 and 614. καὶ καθὸλΟυ . . . ἐξιδιοποιησάμενος: under this vague generality D unwittingly hides Athens' most important gain in the area, Corcyra. Xenophon on the other hand mentions only Corcyra though he does take cognizance of Timotheus' further successes ( H. 5. 4. 64; cf. Isocr. 15. 109; Nepos, Tim. 2. 1). D's failure to list Corcyra is due as usual to his unintelligent abridgement of Ephorus: he has included the beginning of τὰ περὶ Τιμόθεον narrative, i.e. the general's arrival at Cephallenia, as well as the end, the decisive naval battle. But the important middle part of the story has been replaced by the vague sentence in question. See nn. above and ad 30. 2. The accession of Corcyra may not have been accomplished peacefully, the pro-Spartan oligarchs (for whom see 46. 1) having to be overcome first. Isocr. 15. 109 (cf. 107f.) speaks of Corcyra being captured (Κόρκυραν εἱ+̑λε). Aristid. Panath. 173. 17 (284 Dind.) points to ἀγωνίσματα on the island. Most explicitly though the scholiast ad It may well be therefore that the Spartan fleet under Nicolochus (not Nicolaus; Aristid. Panath. 173. 17 says: cf. X. H. 5. 4. 65; Polyaen. 3. 10. 12) was meant to succour a Corcyra threatened by Timotheus. Xenophon's language (οὐ μέντοι ἠνδραποδίσατο κτλ) seems to imply that some force was used, but that Timotheus was as moderate as possible. It is likely that the oligarchs had held the upper hand before Timotheus' arrival (which need not imply an oligarchic constitution: 46. 1 and n.) and it is -316even possible that they were allied with Sparta. For power politics on the Ionian islands see 45. 1 ff. and 46. 1 nn. ἐνίκησε . . . Λευκάδα: actually off Alyzeia on the

west coast of Acarnania ( X. H. 5. 4. 65; schol. Aristid. Panath. 282 Dind.), but as this city was opposite Leucas and as the battle was fought between the island and the mainland, it was also known as the περὶ Λευκάδα ναυμαχία ( FGH 70 (Ephorus) F2II; Polyaen. 3. 10. 4, 12, 17; Aristid. Panath.313). Dinarchus (1. 14; 1. 75; 3. 17) refers to it as the ἐν Κερκύρᾳ ναυμαχίαν. Isocr. 15. 109 says that Timotheus was sent out on his periplous of the Peloponnese with 50 triremes and 13 talents. X. H. 5. 4. 63 mentions 60 triremes, the same number he fought the battle with against Nicolochus' 55 ships ( ibid. 65 ). Polyaenus on the other hand gives the figure of 40 (3. 10. 6, 11; but see 3. 10. 12, τω+̑ω μὲν πλειόνων νεω+̑ν κτλ, which necessarily implies that he had more than 40 triremes, while at 3. 10. 16 he says that Timotheus' fleet was augmented by squadrons from the Corcyraeans and the other allies; X. H. 5. 4. 66 says that this was after the battle). For the battle itself see X. H.

5. 4. 65 f.; Front. Strat. 2. 5. 47; Polyaen. 3. 10. 4, 6, 11, 12, 13, 16, 17 (6, 12, 16 and 13, 17 respectively refer to the same two stratagems allegedly employed by Timotheus). Cf. Schaefer, 1885-7: i. 47 ff. and 49 n. 1. According to Polyaen. 3. 10. 4 the battle fell on the day of the feast of Scira which should be identified with the feast of the Scirophoria celebrated on 12 Scirophorion (June/July. On the Scirophoria see Jacoby n. on FGH 328 FF 14-16 ). 36.6. ταυ+̑τα δὲ πάντα κτλ: Timotheus, a close friend of Isocrates, could not fail to win praise from a wellknown pupil of the master (p. 119 ). Cf. Isocrates' own eulogy of Timotheus, 15. 101 ff. Xenophon too has praise for Timotheus' good conduct ( H. 5. 4. 64. As was said above, he is very probably contrasting Timotheus' behaviour with that of Chares). But more important than their connection with Isocrates, Timotheus' moderation towards his fellow Greeks could not but be praised by a moralist and a panhellenist. διόπερ οὐ μόνον παρὰ

τοι+̑ς πολίταις, ἀλλὰ καρὶ τοι+̑ς ἄλλοις Ἕλλησι μεγάλησ ἐτύγχανεν ἀποδοχη+̑ς, D says; cf. Isocr. 15. 116, 124, 127, 138. On the Ephoran viewpoint see pp. 110 ff. His fellow citizens, highly pleased both with his victories and with the Common Peace which followed (38. 4n.), heaped honours on him, including statues (Aeschin. 3. 243; Paus. 1. 3. 2; 1. 24. 3; Tod -317128; Nepos, Tim. 2. 3) and a monument at Delphi ( Vatin, 1983: 26 ff.).

37. 1-2. Tegyra This is but the briefest outline of what occurred. Plut. Pel. 16 fills in the background. Orchomenus was proSpartan and defended by a strong Spartan garrison; two morai it is claimed. D simply says πολλοι+̑ς στρατιώταις. When this garrison was away on an expedition to Locris, Pelopidas saw his chance of capturing the city and marched against it. Though he must have heard of the action Xenophon chose to ignore it (unless H. 6. 4. 10 is a reference to it). See Pritchett, 1965-85: iv. 103 ff. For the date see the chronological excursus below at 45. 3, part 1. 37. 1. πεντακοσίοις: cf. Plut. Pel. 16. 2 ἔχων μεθ+̕ ἑαυτου+̑ τὸν ἱερὸν λόχον καὶ τω+̑ν ἱππέων οὐ πολλούς. The Sacred Band consisted of 300 men ( ibid. 18. 1 ). One might therefore venture to suggest that D's figure embraces both horse and foot: there were 300 hoplites and 200 cavalry. μάχη καρτερά: cf. Plut. Pel. 17. 6. The action was fought at Tegyra some 5 miles north-east of Orchomenus ( Plut. Pel. 16. 3; 17. 1). Fossey, 1988: 367ff. D fails to name the place here, but does so at 81. 2 in the epainos on Pelopidas. Plutarch tells us that Pelopidas and his Thebans were returning home having failed to capture Orchomenus; a fresh garrison had arrived from Sparta and was guarding the city. At Tegyra the Thebans fell in with the Spartans who were on their way back from Locris ( ibid. ). πρὸς διπλασίους: as there were 500 Thebans present, the Spartan force is reckoned to have been 1,000 strong. Elsewhere D says that the mora was made up of 500 men (32. 1) and hence the two morai mentioned by Plut. Pel. 17. 3 are obtained. These are Ephorus' figures as Plut. Pel. 17. 4 confirms (=FGH 70 F210). Plutarch feels duty-bound to supply alternative figures for the strength of the mora which enhance even more the Theban victory: τὴν δ+̕ μόραν Ἔφορος μὲν ἄνδρας εἰ+̑ναι πεντακοσίους φησί, Καλλισθένης δ+̕ ἑπτακοσίους, ἄλλοι δέ τινες ἐνακοσίους, ὡ+̑ν Πολύβιός ἐστι ( Pel. 17. 4. For the strength of the mora and the Spartan army in general see 32. 1 n.). Thus a maximum of 500 Thebans defeated 1,000 or perhaps 1,400 or even 1,800 Spartans. As Jacoby puts it, 'die -318boeotische panegyrik ist deutlich' ( FGH ad 124 F18). And see following n. οὐΔéποε γàρ τΟŭτο αυνἐßν κτλ: similarly Plut. Pel. 17. 11 ff. though Plutarch used other sources apart from Ephorus as well as his own knowledge of Boeotian history and topography for his account of Tegyra. For the cult and sanctuary of Apollo Tegyraeus in the vicinity of Tegyra mentioned by Plutarch in connection with the battle see Schachter, 1981-94: i. 75; Fossey, 1988: 367ff. It has been suggested that the two moral and ἠ δἡαδοΞἠ τὠν ΛαηδαιυΟνí≀ν ( Plut. Pel. 16. 2f.; 17. 3, 9) correspond to Cleombrotus' army of four morai sent out in response to an appeal for help by Phocis ( X. H. 6. 1. 1), and that Tegyra therefore should be dated to 375 ( Cawkwell, 1963: 89 n. 49, though he has now changed his mind; see id., 1981b: 44). But this is very unlikely. Xenophon at H. 6. 1. 1 is hardly

reconcilable with Plutarch, and Cleombrotus probably did not go to Phocis in 375 (see the chronological excursus below). What is more it seems incredible that Cleombrotus' name should have escaped our sources if indeed it was half his army that Pelopidas had severely defeated. In fact the tradition that two Spartan morai were defeated is questionable. We may accept that two morai were involved, but not both in the battle. Only one mora clashed with the Thebans. The other, the replacement of the first, fresh out from Sparta, was at Orchomenus at the time (it prevented Pelopidas from capturing the city), and presumably it moved out and against the Thebans but after the battle. The Thebans did not in any case await it; they hastily withdrew as Plutarch says ( Pel. 17. 9f.). The oversight of the sources may of course have been deliberate as it redounded to the glory of Thebes and Pelopidas. 37. 2. One of a number of anticipatory comments which herald the Theban victory at Leuctra; see 34. 2; 39. 1; 50. 5 f. Schwartz, correctly, would assign them to Ephorus ( RE v. 1. 681). For the Ephoran attitude to Sparta and Thebes see pp. 117 f.

37. 3. A chronographic notice See Jacoby, FGH iilb. 558; cf. Sanders, 1987: 41 f. -319

Archon Hippodamus, 375/4 38. 1. 'Ι̑πποδáμου Ιπποδáμαντος is the correct form of the genitive as we can see in inscriptions. Develin, 240. 'Π≀μα οι κτγ: 383 V; CAH2 vii. 2. 637. Four tribunes are named, as announced, two short of Livy's usual college of six (6. 21. 1). See pp. 27 f. and Drummond, 1980: 61 f.

38-39. The Peace of 375/4 Svt 265. For the date see the chronological excursus below at 45. 3, part 1. 38. 1. 'ΑπταĒéπĒνς κτλ: the Great King was chiefly concerned with maintaining his hold on western Asia Minor, which meant preventing any mainland Greek state from becoming too strong in the Aegean. The King's Peace with its autonomy clause had been devised for that very purpose and hence the King's anxiety to have it reaffirmed. All the same, the reason which D (Ephorus) gives may well have been another of the King's main aims. At this time there was hardly a Persian campaign in the western satrapies which did not involve large numbers of Greek mercenaries and 20,000 of them took part in the invasion of Egypt shortly afterwards according to D (41. 3). D's statement that peace was now made at the initiative of the King, and the motive he ascribes to his intervention, are important as proving that not all of his account of the Peace of 375 is confused. (38. 3-39 n.). The participation of the King is placed beyond doubt by a fragment of Philochorus ( FGH 328 F151 with n.). Xenophon does not mention the King, but gives the initiative for the Peace to Athens. Further, he represents that the treaty was a bilateral one between Athens and Sparta ( H. 6. 2. 1), though he must of course mean that their respective allies were a party to it too; cf. Roos, 1949: 267f. Roos ( 1949: 277f.) is too kind to Xenophon when he suggests that the historian omitted the part of the King 'in order to give to Athens . . . and to Sparta . . . the full credit of it'. A more likely explanation would be that the pro-p Spartan and panhellenist Athenian chose to overlook the fact that in the 370s Sparta and Athens used the big stick of the King to intimidate Thebes. When in the 360s the situation was reversed Xenophon was not so reticent (cf. H. 7. 1. 33 ff.). Isocr. 14. 41 does -320not mean that the King was not involved in the peace negotiations, but only that he took no part in the preceding warfare. 38. 2. τ≀ν Δ' 'Ελλν≀ν . . . Πολéμων: this further stresses the fact that the Peace was a general one, a Common Peace, and not a bilateral affair as Xenophon represents. Of course the term 'the Greeks' does not mean that all the Greek states were present and took the oaths. The non-belligerent states (few, considering the extent of the Athenian and Spartan alliances) may not have taken part. But this matters little. The Greek League against Persia did not comprise all the Greek cities. It was 'Greek' in the sense that it was open to all the Greeks and claimed to represent Greece. The Common Peace likewise encompassed all the Greek states, belligerent or not. This was in fact its main characteristic. There is a general consensus amongst the sources that the Greek cities almost universally accepted the Peace readily (the Thebans appear to have been the only exception; see below); X. H. 6. 2. 1; FGH 328 ( Philoch.) F151; Isocr. 14. 41; cf. 15. 109 f.; Nepos, Tim. 2. 2. αυνéΘεντο πáντες τἦν ε∞ἠν♮ν κτλ: the conference was held at Sparta ( X. H. 6. 2. 1). The oaths sworn amounted to a renewal of the Peace of 387/6. ὔτ⌈ Πα ΛΠλ≉αλΟν α⇑τ≉ν τ♮ τΟ⇕ Λáκ≀νος 'Ανταγκíδομ Πποα≉κατο was Philochorus' comment on this Peace (F151). For its terms, which were in fact essentially those of the original Peace, see 5. In. A major difference concerned the sanctions clause; Athens now appears to have joined Sparta as a coprostates of the Peace (section 4n.). The appointment by 'the Greeks' of ἐÈαγΩγε ς who went from city to city and evacuated the garrisons was, of course, aimed at ensuring that all cities were left àω ↕ ντοι. It is possible that the same was done in 387/6 in spite of the silence of the sources.

The strictures of Badian notwithstanding, 1987: 27 ff., there can be little doubt that, in whatever manner, in 387/6, 375, 372/1, and 366/5 the King firmly gave his pledge to abide by the terms of the Peace 'he sent down', whether himself in person to Greek ambassadors (cf. 16. 43. 30. or, as on this occasion, through his emissaries (cf. X. H. 3. 4. 6); so that vt 248 (= Tod 118) lines 11-12 and Dion. Hal. Lys.12 (overlooked by Badian) are perfectly justified in referring to the King as 'swearing to the Peace'. Cf. 19. 1 and n. 38. 3-39. μóν≀ν Δé Ωνßαí≀ν κτγ: this is wrong of course. The Thebans were certainly not excluded in 375 for we know that they -321-

continued as members of the Athenian Confederacy (ps.Demosth. 49. 14, 21, 48ff.; IG ii2 1607 lines 49, 155; Accame, 1941: 230 lines 15 f.), and Isocr. 14. 1, 5, 14 tells us that Plataea was seized by Thebes (in 373) Áíρ≉νς . So what has gone wrong with D's account? The commonest view is that D has confused the two Peaces of 375 and 372/1; the Thebans were excluded from the latter Peace, but not from the former. For some of the scholars who hold this opinion see the references given by Lauffer, 1959: 315 n. 1. Add Ryder, 1965: 124. But D has his supporters too (cf. Lauffer, 1959: 315 n. 2), most notably Judeich, 1927: 182 ff., and Sealey, 1956: 189 ff. Roos, 1949, agrees that the Thebans were not excluded in 375, though he considers this not an error on the part of D, but a deliberate misrepresentation by the pro-Theban sources used by Ephorus. Lauffer too, 1959, considers chs. 38 f. and 50 a doublet, but he blames Ephorus and Ephorus alone for it (see below for both). The simplest and to my mind correct way of dealing with. 38. 3 f. is to regard it not as a doublet, but as a badly excerpted Ephoran narrative: while D has related some of the circumstantial detail bearably well he has, as is sometimes his wont, missed the all-p important upshot of it all, i.e. the fact that in the end Thebes was intimidated into accepting the Peace on the Spartans' and the Athenians' terms. Later, at 50. 4, when relating the Peace of 372/1, noticing the similarity between his two accounts, D added éαπÉρ καì Ππρον to his statement that the Thebans were éÁαππονδοι. This is a more satisfactory explanation than the assumption that D read the exclusion of Thebes back to this Peace from that of 372/1. In any event, the theory of a Diodoran doublet here appears untenable for it goes contrary to our understanding of how D worked. D followed down his chronographic handbook and his narrative source so that, naturally, he dealt with the events of 375 before those of 371. While, therefore, elements of the Peace of 375 could have entered that of 372/1, the reverse process is less easy to envisage. In other words we would have a doublet if the Thebans were excluded in 375 but not in 371 and D had said they were, misled by what he had previously read and written. It is better therefore to suppose that D botched the culmination of the Ephoran account here; he misunderstood Ephorus to mean that the Thebans were excluded from the Peace when in truth they had only been threatened with exclusion. In the end they had given in -322and taken the oaths as 'Thebans' and allowed the other Boeotians to do so too Κατà πóγιν. Roos' hypothesis that Ephorus' sources are responsible for the error solves nothing; it merely shifts the blame. In any case it is naive to reason that D cannot be responsible because the words παγÊρ and ≀γÊρ Καì ππóτηρν at 50. 4 prove that he was aware of the close similarity between his two accounts of the Peaces of 375 and 372/1 and 'it seems unthinkable that the same compiler should notice that he is telling the same story twice, make this clear by additions, and even then fail to see that he is condensing the same passage which he had already condensed once before!' First, D may not have seen anything wrong in having the Thebans excluded twice. Secondly, a compiler like D was perfectly capable of letting an error stand so as to avoid wasting time and effort in correcting it (see pp. 24, 137 ). Thirdly, the Ephoran accounts of the two Peaces were not dissimilar, but with this important difference: in 375 the Thebans had eventually yielded to Spartan pressure. In 372/1 they had not, and D was just the man to miss this vital point and further disfigure the narrative by having the Thebans excluded on both occasions. Besides, a potent reason why Ephorus could not have said that the Thebans were excluded twice is that this would have made nonsense of his Greek narrative of the years 375-371. There is nothing in sections 3-4 which is inconsistent with a 375 date and with the Peace of that year. The Thebans had vigorously objected to the Spartan interpretation of the autonomy clause in 387/6, and they had only backed down and conceded the independence of Boeotia after Agesilaus had taken the field against them. In 371 they answered the Spartan challenge and won at Leuctra. It would be very surprising therefore if they had made no objection at all to letting Boeotia free in 375. The issue must surely have been hotly debated at Thebes itself, in the Athenian assembly, in the allied synedrion, and at the peace congress at Sparta. Section 3 is an echo of that debate. Callistratus and the Athenians in general are very likely to have opposed Theban expansionism, and Epaminondas, who did not necessarily leap into fame at the peace congress of 372/1 for the first time, is certain to have defended the Boeotian Confederacy. Whether the altercation took place in Athens or Sparta (see n. ad τ≀ ΚοΙν≀ αμνηδρí≀below) it makes little difference; a 375 context is far from improbable. -323It was said above that the motive which D ascribes to the King for wishing to have the Common Peace reaffirmed shows that, at the very least, not all of D's account is in error. And there is something else. Section 4 concerns the co-prostasia of the Peace by Athens and Sparta which was certainly initiated in 375 (section 4n.). If therefore sections 1, 2, and 4 are not lifted from the Peace of 372/1, but correctly belong to 375, why should it be supposed that the details* of section 3, out of which section 4 proceeds, are drawn from the later Peace? It is simpler to assume that D has misrepresented the crowning detail of section 3, i.e. that the Thebans were not excluded from the Peace in the end. There is in fact in the second half of section 4 a statement which strongly indicates that in Ephorus the combined pressure of Athens and Sparta eventually succeeded in overcoming Theban objections on this occasion. The Athenians and Spartans, D says, resented the Theban claim to hegemony and they àπéαπ≀ν the

Boeotian cities from Thebes. What of ch. 39? It is remarked elsewhere that the growth of Theban power (and the corresponding decline of Sparta) was one of Ephorus' major themes, and it has left its traces in D (pp. 112 ff.). I would view ch. 39 as D's version of an outburst of Ephoran rhetoric of the usual sort (p. 123 ) which anticipated the Theban triumph at Leuctra. It anticipated, but it did not immediately presage. On commenting on the Theban attitude to the Peace of 375 Ephorus looked ahead to 371 when the Theban aspirations were realized. The various elements of ch. 39 are found elsewhere than 50. 5f., as Lauffer, 1959, has not noticed, who consequently proceeds to draw quite unacceptable conclusions (see below). We may note especially the prowess of the Thebans stressed at 11. 82. 3; 15. 20. 1; 26. 1; 37. 2; 50. 5. The recurrence in D therefore of the Theban valour and of the Spartan defeat at Leuctra surely indicates a theme which ran through Ephorus' history of this period. The close similarity in language to which Lauffer attaches such significance is, in reality, of no significance at all. D's language is monotonously uniform and we know that Ephorus too did not always take care to vary his description of the same or similar things (p. 124 ). A further argument against the doublet hypothesis can be adduced from the sequence of D's narrative in chs. 3840. Ch. 40 is generally (and probably erroneously; see below) considered to -324-

represent the reaction of some Peloponnesian states after Leuctra. But whether the events of ch. 40 belong after the Peace of 375 or after Leuctra (these being the only two possibilities), the conclusion is inescapable that ch. 39 cannot derive from the Ephoran narrative which introduced the campaign of Leuctra because if that were so and ch. 40 were given a post-Leuctra date, what has happened to the campaign itself? áγγá περì μèν τ ↑ατερον éν το ς κατá μéρος éκτεá ÅκτεΘ√ι ωαωéωτερΟν Δνγ≀αομεν νÁν Δ'éπí τò 2ALô ↑νεΞèς τ√ρΗűóμηΘα D declares at 39.3, so that what-p ever it was that intervened in Ephorus, it was not Leuctra. If, on the other hand, ch. 40 is assigned to the aftermath of the Peace of 375 (as it should be) then the true nature of ch. 39 is immediately apparent: it is the remnants of Ephorus' anticipatory comments on the Theban victory at Leuctra four years later. The transition sentence just quoted (39. 3), a formula in fact, is common in D. There are many such both in parts of the Bibliotheke based on Ephorus, and in those based on other authors (p. 23 and n. 63). They show that D was on every occasion aware that what he was saying, whether his own or his source's, belonged to the future. None of these transitional formulae (which in part served to whet the readers' appetite with things to come) are apologies for errors of anticipation, and the one at 39. 3 is not (as Lauffer, 1959: 317 f., thinks) D's way of admitting that he has jumped the gun, but that he is now, after consulting his date-book, aware of it and is taking measures to rectify the situation. I am in fact totally resistant to the view of Lauffer who, accepting the doublet theory, blames not D for it, or Ephorus' sources, but Ephorus himself (1959: 345 ff.). 'Daraus folgt aber, dass die Dublette selbst nicht von Diodor verschuldet ist, sondern auf zwei würtlich öbereinstimmende Stellen in seiner Quelle zurückgeht' (1959: 346). Chs. 38 and 50 are a good example of Ephorus' doubling technique, Lauffer thinks. Ephorus knew that the Thebans were not excluded in 375, but he deliberately falsified history for rhetorical and artistic reasons. This is too serious an allegation to be left unanswered. To substantiate his theory (and I fail in any case to see what possible artistic effect could result from such a misrepresentation of the facts) Lauffer points to the transitional formula at 39. 3, to the words πÁγ ν and≀απερ καì πρóτεον at 50.4, and to the fact that 50. 5f. is briefer than ch. 39. In Lauffer's opinion all these prove -325represent the reaction of some Peloponnesian states after Leuctra. But whether the events of ch. 40 belong after the Peace of 375 or after Leuctra (these being the only two possibilities), the conclusion is inescapable that ch. 39 cannot derive from the Ephoran narrative which introduced the campaign of Leuctra because if that were so and ch. 40 were given a post-Leuctra date, what has happened to the campaign itself? áγγá περì μèν τ ↑ατερον éν το ς κατá μéρος éκτεá ÅκτεΘ√ι ωαωéωτερΟν Δνγ≀αομεν νÁν Δ'éπí τò 2ALô ↑νεΞèς τ√ρΗűóμηΘα D declares at 39.3, so that what-p ever it was that intervened in Ephorus, it was not Leuctra. If, on the other hand, ch. 40 is assigned to the aftermath of the Peace of 375 (as it should be) then the true nature of ch. 39 is immediately apparent: it is the remnants of Ephorus' anticipatory comments on the Theban victory at Leuctra four years later. The transition sentence just quoted (39. 3), a formula in fact, is common in D. There are many such both in parts of the Bibliotheke based on Ephorus, and in those based on other authors (p. 23 and n. 63). They show that D was on every occasion aware that what he was saying, whether his own or his source's, belonged to the future. None of these transitional formulae (which in part served to whet the readers' appetite with things to come) are apologies for errors of anticipation, and the one at 39. 3 is not (as Lauffer, 1959: 317 f., thinks) D's way of admitting that he has jumped the gun, but that he is now, after consulting his date-book, aware of it and is taking measures to rectify the situation. I am in fact totally resistant to the view of Lauffer who, accepting the doublet theory, blames not D for it, or Ephorus' sources, but Ephorus himself (1959: 345 ff.). 'Daraus folgt aber, dass die Dublette selbst nicht von Diodor verschuldet ist, sondern auf zwei würtlich öbereinstimmende Stellen in seiner Quelle zurückgeht' (1959: 346). Chs. 38 and 50 are a good example of Ephorus' doubling technique, Lauffer thinks. Ephorus knew that the Thebans were not excluded in 375, but he deliberately falsified history for rhetorical and artistic reasons. This is too serious an allegation to be left unanswered. To substantiate his theory (and I fail in any case to see what possible artistic effect could result from such a misrepresentation of the facts) Lauffer points to the transitional formula at 39. 3, to the words πÁγ ν and≀απερ καì πρóτεον at 50.4, and to the fact that 50. 5f. is briefer than ch. 39. In Lauffer's opinion all these prove -325-

that D had knowingly come across a doublet in his source. The first two points have already been dealt with. The third, while it may indicate that D did not wish to repeat fully comments he had made in ch. 39, hardly proves Lauffer's contention for the doublet lies not in the repetition of the praise of Thebes and her leaders here (and elsewhere), but in the fact that the Thebans are said to have been excluded twice, in 375 and in 372/1. It is all very well to draw one's attention to the ancient criticism of Ephorus that he described similar things in very much the same language, but this does not mean that where the details were different Ephorus ironed these out thus falsifying history out of some perverse desire to repeat his accounts (p. 124 ). The verbal similarities apart, a comparison of ch. 39 and 50. 5 f. reveals significant differences which further shake Lauffer's thesis. Two statements made in 50. derives from, led up to the campaign of Leuctra; the original of ch. 39 did not.

narrative it

To conclude: while the theory of a doublet and, further, that others than D were responsible cannot be definitely disproved, it seems an altogether more satisfactory explanation that the details of ch. 38 are factual and belong to 375, with the exception of the statement that Thebes was excluded from the Peace. When we consider the mess made of other accounts (e.g. Leuctra and the Corcyra campaign of 373/2), it does not seem at all unlikely that D is responsible for the blunder. 38. 3. συντéγελαν: see 28. 1n. κολ≀ συνεδπí≀: the peace congress or the synedrion of the Athenian Confederacy? Judeich, 1927: 184 n. 1, and Sealey, 1956: 189, think the latter. Lauffer, 1959: 320, objects that the term koinon synedrion occurs elsewhere in D and indeed it is used of the assemblies or councils of the Greek League of Corinth (11. 3. 4; 11. 55. 4, 5; 17. 14. 1; 17. 73. 5), of the Delian League (11. 70. 4), of the Peloponnesian League (12. 39. 4), of the anti-Spartan alliance of 395 (14. 82. 2), as well as of the Second Athenian Confederacy (15. 28. 3; 29. 7; 30. 2). It is possible that the peace congress is indicated here though all the eleven instances of the -326-

term in D just cited refer to leagues and similar bodies and not to peace congresses. Either way it does not affect what was said above about 38. 3-39. If the peace congress at Sparta is meant, then Epaminondas and Callistratus clashed there in 375. If the syne-p drion of the Athenian Confederacy is meant, then Callistratus and Epaminondas opposed each other there in 375 in their respective addresses of the Athenian assembly and the synedrion of the allies. Of course it is all rather badly stated in D, but there is nothing surprising in that. Nor is there anything surprising in the assumption that the proposed renewal of the Peace was discussed beforehand in the synedrion of the Confederacy. It was one of the principles of the Confederacy that a decision to go to war or to make peace required the assent of Athens and the majority of the allies (Svt 257 = Tod 123; cf. Aesch. 2. 60; 3. 69). So the peace proposals must have been discussed in the synedrion. At the meeting Epaminondas, as a Theban representative, spoke against the dissolution of the Boeotian Confederacy which was required by the proposed terms of the Peace, but he could not carry a Majority of the synedrion. Later at Sparta the Thebans and the rest of the allies took the oaths κατá πóγελá πóγελςἔκőτολ ( X. H. 6. 3. 19 for 372/1, but doubtless that was the case in 375 too). The Thebans obviously swore as 'Thebans' (cf. Svt 257=Tod 123 line 79), and the Boeotian Confederacy was dissolved, at least in theory. There is nothing improbable in this. Whether the Boeotian Confederacy was dissolved in practice is another question. Almost certainly it was not. At any rate, in the summer of 373 when Plataea was captured and destroyed, and before the resumption of war, the Confederacy was very much in existence (46. 4; Paus. 9. 1. 6: both mention Boeotarchs; Isocr. 14. 8f.; cf. 33). The number of Pelopidas' Boeotarchies too argues against the interruption of that office in 375. When he fell in summer 364 he was Boeotarch for the thirteenth time ( Plut. Pel. 34. 7). He was not a Boeotarch in 371 ( ibid. 20. 3 ) and, probably, 378 (cf. Fuscagni, 1972: 415 ff.); but he must have been in 375 and 374 if the thirteen Boeotarchies are to be accounted for. What happened therefore was that the Thebans promised to dissolve the Confederacy (perhaps not in so many words), but did not. On the contrary, once Sparta had evacuated her garrisons, the Thebans began to take over the cities which had previously resisted them with Spartan help (46. 4 ff.; Isocr. 14. 8f.). The 'cold war' between -327Athens and Sparta which followed immediately upon the conclusion of the Peace (40 n.; 45. 1 and n.; chronological excursus at 45. 3, part 2) facilitated the Theban task. Athens needed Thebes and was therefore willing to tolerate Theban actions in Boeotia as she had done before the Peace (and as indeed Isocr. 14. 21, 24f., 26, 33, 38, says), while Sparta soon had her hands full in the Ionian Sea. Much of the Peloponnese too was dangerously restless (ch. 40). Circumstances in 371 were quite different. Sparta then had an army poised in Phocis and, more important, Athens was prepared to leave Thebes to her fate. Isocr. 14. 37, the Athenian threat to have the Thebans declared ἔκσπονδοι because of their conduct over Oropus, does not necessarily refer to this occasion (as Schaefer, 1885-7: i. 53f., and Sealey, 1956: 190f., suggest). It may even be that the Athenian threat was to exclude the Thebans from their Confederacy (Glotz and Cohen, iii. 137). If, however, the threatened exclusion was from the Peace, and as Isocrates does not mention the Boeotian cities, then the time perhaps was shortly after the conclusion of peace. Finally, against those who for the sake of simplicity would assign 38. 3, the clash between Callistratus and Epaminondas, to 371, merely because that is when Epaminondas stood up to Agesilaus ( Plut. Ages. 27. 5-28. 3) and they are averse to complicating matters by having Eparninondas confront opponents on more than one occasion, it should be said that the sources mention another confrontation between Callistratus and Epaminondas; see Nepos, Epam. 6; Plut. Mor. 193c-d; 810f. The place was Arcadia and the occasion either 366 ( Roy, 1971: 596 n. 145 ) or 362 ( Cawkwell, 1961: 84 n. 4).

38. 4. παπεξ≀πουν àγγ≉γοις κτγ: i.e. the prostasia of the Peace was now to be shared by Athens and Sparta (see 5. 1 n.; Hampl, 1938: 14 ff.). Nepos, Tim. 2. 2f. (whose ultimate source was probably Ephorus), makes the same point and so does Isocrates (15. 109 f.; cf. 7. 12; also Demosth. 9. 23 which goes a great deal further) in his own way. Both add what D does not, i.e. that the Athenians were so delighted with their position vis-α-vis the Peace that they dedicated an altar to Eirene and voted annual sacrifices; see FGH 328 F151 with n.; Robert, 1977: 211 ff. It does not appear that the co-prostasia was worked out in detail; see 67. 1 n. It is suggested above (30. 2n.) that some seventy-five states -328became members of the League, that all of them were recorded on Svt 257= Tod 123 and that the process was completed by autumn 375, in the aftermath of the conclusion of the Common Peace in that year. (For the date of the Peace see chronological excursus at 45. 3, part 1.) There is no evidence that any cities were added to the League after 375. The evidence usually cited for acquisitions in 373 hardly amounts to that: 47. 2f. is a very confused passage (see n. there). Why then did the League grow no further after 375? The answer should perhaps be sought in the re-establishment of the Common Peace which set Athens and her League in a new light. Not only was the latter recognized by Sparta and the King, but Athens was elevated to the position of co-prostates of the Peace. The joy with which the Peace was received indicates that many patriotic Athenians felt they were on the road to empire again. Certainly Athens is more self-assured and even aggressive after the Peace, and there are signs of renascent imperialism already in the later 370s (cf. Sealey, 1957: 95ff.; Hornblower, 1982a: 190ff.). It is not the case that Athens gradually went back on the principles enunciated in the decree of Aristoteles. These appear (the evidence is less than plentiful) to have been broadly honoured where the members of the synedrion were concerned. (To this extent Cargill 1981 is right.) It is rather that she saw her chance for empire in keeping future allies for herself and not enrolling them as members of the synedrion where they would be protected (cf. Griffith in Imperialism, 137ff.). συντελεὶασ: see 28. 1 n.

39. The growth in Theban power 39. 1. Πολλασ μάχαισ Προνενικóτεσ: see 34. 2; 37. 1-2; 37. 2n.; Plut. Pel. 15. 6ff.; 16f. 39. 2. ΠελοΠίδασ: the famous Theban mentioned here for the first time in D. For the most detailed modern treatment of Pelopidas see Bersanetti, 1949: 43ff. Γοργίασ: 'Gorgidas' according to Plut. Pel. 18. 1; 19. 3; Polyaen. 2. 1. 2, though MS F of Polyaenus consistently reads 'Gorgias'. Gorgidas is credited with the creation of the Sacred Band by Plut. Pel. 18. 1 and Polyaen. 2. 5. 1; Athen. 13. 602a ascribes this to Epaminondas, naturally. The creation (or rather reorganization; cf. 12. 70. 1) of this renowned body apart, his defeat of Phoebidas -329at Thespiae in 378 would have contributed to his reputation as a soldier (Polyaen. 2. 5. 2). ἘΠαμεινώνδας: for Epaminondas see Fortina, 1958; Cawkwell, 1972: 254ff., as well as the works cited by the latter on 254 n. 1. μετέσχε φιλοσοφίας: cf. Plut. Ages. 27. 6 (ἀνὴρ ἔνδοΞος ἐΠὶ Παιδείᾳ καὶ Άιλοσοφίᾳ); Pel. 3. 6; 4. 1; 5. 4. As noticed in the Introduction (pp. 10 f.), Ephorus ascribed the short-lived Theban hegemony to the paideia and general moral qualities of Epaminondas. With that man's death at Mantinea, Theban greatness was at an end: 79. 2; 88. 3f.; Polyb. 6. 43; cf. Plut. Mor. 864d on the Thebans' lack of education and polish. The tradition of Epaminondas' training in Pythagorean philosophy is a strong one; cf. 10. 11. 2; 15. 52. 7; 16. 2. 3 (certainly confused); Aristox. F38, 96 (Wehrli, ii); Plut. De genio579d-f; 583b-c; 585d-586a; Nepos, Epam. 2. 2; Paus. 9. 13. 1; Diog. Laert. 8. 1. 5; Cic. De orat. 3. 139; De off. 1. 155; cf. Arist. Rhet. 2. 1398b. As Lysis was a historical personality and the dates of the two men are not inconsistent, there is no reason to reject the tradition; cf. Fortina, 1958: 5f.; Lévêque and Vidal-Naquet, 1960: 294ff. ὀλίγοις Πολιτικοι+̑ς κατακóΨαι: for the facts of Leuctra see chs. 51ff. nn.

40. Strife in the Peloponnese D's picture of a deeply troubled Peloponnese in the wake of the Peace of 375 is usually viewed with incredulity by scholars who prefer to give ch. 40 a post-Leuctra date; cf. Grote, x. 271 n. 1; Stern, 1884: 93ff., 155; Meyer, GdA v. 420; Beloch, GG iii. 1. 174 nn. 2, 4;Glotz and Cohen, iii. 151 n. 22; Hammond, HG 495; Cartledge, 1979: 296; id., 1987: 266; Buckler, 1980a: 292 n. 1. Buckler's objection that D at 40. 1 'speaks of the Spartan hegemony as though it were a thing of the past' and that therefore the disturbances are better dated after Leuctra, is unsound for D speaks in the same way at 5. 2 and 45. 2, yet neither passage could possibly be assigned a post-Leuctra date. The general view is questioned by Lauffer, 1959: 318 n. 5, but without argumentation, and in any case the grounds on which he does so are very shaky as we have seen (38. 3-39n.). Equally Roy, 1973: 135ff., who also aims to question the date commonly given to ch. 40, spoils his case -330by rejecting Ephorus' albeit incompetently transmitted explanation of the troubles. Yet surely our only hope of placing these events in their correct context lies precisely in accepting what Ephorus had to say about them. Thus

viewed, not only does ch.40 cease to be an awkward problem, but it provides valuable insight into the condition of the Peloponnese in the mid-370s. We would seek in vain for such illumination in Xenophon. D explains the events of ch. 40 as the result of the conclusion of peace, μετὰ γὰρ Τὴν ΣνγξωρνΘει+̑αν (not of any Spartan defeat be it noticed), and at the end of the chapter, after commenting, αἱ μέν οὐ+̑ν κατὰ Πελοπóννησον Πóλεις ἐν τοIιαύταις σνμφοραι+̑ς ἦσαν, he passes to Persian affairs, the invasion of Egypt of 373 (chs. 41ff.). With that narrative concluded he returns to Greek affairs: κατὰ Δέ τὴν Ἑλλάδα κτλ (45. 1). The situation he describes at 45. 1 is basically that of 40. 1-2: we are still in the aftermath of the Peace of 375, with democrats vying with oligarchs for control. But while in ch. 40 he was concerned with the Peloponnese, he now moves to the Ionian Sea where the involvement of Athens soon led to the resumption of hostilities (45. 1-2). The first area of conflict in the west was Zacynthus and as this certainly does not belong to a post-Leuctra, but to a post-Peace of 375 context, it is but logical to conclude that the events of ch. 40 also belong to the same context. Firm evidence to clinch this is lacking, but as evidence to the contrary is not forthcoming, the mere silence of Xenophon and a vague feeling that such disturbances could not have occurred in the Peloponnese in 375/4 are not good enough for rejecting D's sequence here. And if on examining the events of ch. 40 themselves we discover nothing inconsistent with a pre-Leuctra date, then that would effectively place the ball in the court of those who would date ch. 40 after Leuctra. As so very often we must suspect that D has been maladroit in comprehending and abridging Ephorus. The five attempted coups occurred now after the conclusion of peace -- Ephorus would have commented on the aftermath of the Peace, and the last sentence of ch. 39, ἀλλὰ Περὶ μέν Τούτων ΤρεΨóμεΘα, should be understood as pointing to that. But sections 1-2, μετὰ Γὰρ Τὴν ΣυγχωρηΘει+̑σαν κτλ, outline a process which stretched back to 386 and the Peace of Antalcidas though the renewal of the King's Peace appears to have shifted it into a higher gear. Sections 1-2 should be seen in the -331-

same light as 5. 2 and 45. 2. We must not therefore allow the apparent contradiction between the explanation D provides for the coups (the revengefulness of the democrats once in control), and the coups themselves (in at least one instance it was the oligarchs who attempted a revolution), to mislead us, as Roy, 1973, is misled. Roy in fact is totally off the scent when he suggests that the Ephoran discussions at 5. 1-3 and here are anachronistic, that the democratic reactions to Sparta and her supporters properly belong after Leuctra, and that there was no connection between the autonomy clause of the King's Peace and the political troubles in the years 386-371, in spite of Ephorus' claim to the contrary. The contest between democrats and oligarchs with the respective backing of Athens and Sparta was just as much a feature of this period as of the 5th century (cf. e.g. Isocr. 4. 16; X. H. 6. 3. 14; de Ste Croix, 1981: 295ff.). D's (Ephorus') comments at 5. 1-3 and here are perfectly to the point as the details of the Zacynthian and Corcyraean affairs make plain (45-46. 1 -- overlooked by Roy). We should consequently be tolerant of the inconsistency between the general remarks at 40. 1-2 and the examples given at sections 2-4; and we should view sections 1-2 as referring to a lengthy strife which by 375 had resulted in some cities being under democratic and others under oligarchic regimes -- hence D's mixed bag of examples. Zacynthus and Corcyra parallel the examples of ch. 40 in this respect. In the first case the oligarchs are in control at the conclusion of peace, in the second, the democrats. The renewal of the Peace will have fed the hopes of all those, be they oligarchs or democrats, not in power in their cities at the time or even in exile. They could appeal to the autonomy clause in their effort to return and/or regain power. The presence of a clause about exiles in the Peace would provide an even better explanation, but such a clause is very doubtful (5. 1n.). For Ephorus and the Common Peace see pp. 127 f. 40. 2. Πρω+̑τον Δέ Τω+̑ν Άιαλέων κτλ: hardly anything definite is known about Phigalia in south-west Arcadia (present-day Pavlitsa) in the 4th century. In the 360s it appears that it followed the lead of Arcadia (cf. Ernst Meyer, RE xix. 2082) and it is thus assumed that before Leuctra it had been a loyal ally of Sparta. But in absence of any evidence to the contrary nothing prevents us from assigning a pre-Leuctra date to the switch from oligarchy to democracy at Phigalia and the banishment of the pro-Spartan -332-

party. If it could happen at Phlius at a time when Sparta ruled supreme why could it not happen at Phigalia in the mid-370s? The details of the bloodbath which followed in no way contradict this hypothesis. Άιάλειαν: thus MF, but Φιάλιαν PX. Phigaleia (or Phigalia) appears to have been the earlier version of the name (cf. Herod. 6. 83. 2; Paus. 8. 3. 1f; 8. 5. 7; 8. 39. 2ff.). It is found in 4th-century inscriptions (e.g. Syll. i. 239c) though by the beginning of the 3rd century at the latest the gamma seems to have been dropped (IG ii2 687; IG V. 2. 419ff.). The city's coinage on the other hand has both forms in the Hellenistic period ( Head, 1911: 418), and in literature too Phigalia and Phialia are found together, with Phigalia predominating. See Ernst Meyer, RE xix. 2065ff. Ephorus perhaps gave the earlier version, but we cannot be sure. Χραίαν: the Heraea occupied by the exiles prior to the raid on Phigalia is often identified with the city of that name in western Arcadia on the north bank of the Alpheus some 15 miles east of Olympia (cf. e.g. Bölte, RE viii. 1. 414). Sievers, Gesch. Griechenlands, 251 n. 5, objected to this because of the distance involved. And indeed Heraea is some 20 miles to the north, hardly a suitable base for a surprise operation against Phigalia. Sievers suggested that some fort near Phigalia was meant, but he offered no candidates. D describes the Heraea of the exiles as a χωρὶον ὀχυρóν which anything but fits Heraea on the Alpheus, a city located on a gentle slope by the river (Paus. 8. 26. 1). It is therefore proposed that D's Heraea is in fact the Heira, the mountain in northern Messenia which Aristomenes supposedly defended for ten years during the Second Messenian War (Paus. 4. 17. 10; 4. 20. 1, 5. The reading of the name in Pausanias is due to Schwartz). Leake, i. 469f., 501, placed Heira some 5 miles west of Phigalia on the south side of the river Neda near modern Sidherokastro. But there are no ancient remains there. Most scholars now would identify Heira 'with the hill of St Athanasius, which rises to the south of

the village of Kakaletri, on the upper waters of the river Neda' (Frazer, iii. 415). Subsequent studies have supported Frazer; cf. Gaertringen and Lattermann, 1911: 13ff.; Valmin, 1930: 118f.; Roebuck, 1941: 11f. The fortifications on the hill appear to belong to the Archaic period and to the 4th century BC. For a photograph of the hill see MME pl. 9. 2. The hill of St Athanasius is some 4 miles southsouth-east of Phigalia on the south side of the river Neda in -333northern Messenia. For Phigalia itself, little investigated to date, see Frazer, iv. 390f.; Orlandos, AD 11 ( 1927-8), 1ff.; Ernst Meyer , RE xix. 2067ff.; Cooper, 1972: 359ff.; Leekley and Noyes, 1976: 100. Διονυσίων . . . ὄντων: urban Dionysia with dramatic festivals were held in the spring (in Elaphebolion at Athens) and if the massacre occurred in spring 374 then that is a terminus ante quem for the first renewal of the King's Peace. But the incident at Phigalia could be as late as spring 373. For the chronology of this period see the excursus below. 40. 3. Οἱ Δέ τη+̑ς ΚορίνΘου κτλ: we know a little more about Corinth, but not enough to put the matter beyond dispute. At the conclusion of the King's Peace in 387/6 large numbers of Corinthians were driven into exile (X. H. 5. 1. 34). Many of them sought and found refuge in Athens (Demosth. 20. 52ff.). Others it seems retreated to Argos, a natural enough choice. It was these exiles, of democratic and anti-Spartan sympathies, who now attempted to return with disastrous consequences both to themselves and their supporters. The pro-Spartan regime established in 387/6 was a strong one and it survived the Leuctra disaster. There can be no question that it was oligarchic in character, though our evidence seems to suggest that it was far from being a narrow oligarchy. In dealing with the circumstances of the appointment of Timoleon to the Sicilian command in the mid340s, D speaks of a Corinthian gerousia which met in a bouleuterion, though decisions ultimately lay with a larger body, the synedrion (16. 65. 6-8). Plutarch in fact goes so far as to interpret as a democracy whatever it was his sources said about the Corinthian polity of the 360s and later (Tim. 5. 2; cf. 3. 1-2; 7. 2). Perhaps they spoke of a Corinthian demos (cf. end of following n.). 40. 4. ἐν Δέ τῃ+̑ Πóλει τω+̑ν Μεγαρέων κτλ: Megarian politics in the 4th century are highly problematic. We know that a narrow oligarchy was set up with Spartan help in 424 (Thuc. 4. 74. 3). We do not know how long that lasted, though it was still in power in 421 ( ibid. 5. 31. 6 ). Thucydides remarks that for a regime set up by so few men after a revolution it lasted a very long time (4. 74. 4). The implication seems to be that at the time Thucydides was writing (not later than c.400), the government in question no longer existed (contra Beloch, GG iii. 1. 70 n. 3; Accame, 1951: 61f.; Ernst Meyer, RE xv. 192; cf. Gomme, HCT iii. 536; Legon, 1981: 247, 263). -334-

1981: 247, 263). If Demosth. 18. 96 is to be given any credence, Megara was held by a harmost and garrison at some time between the end of the Peloponnesian War and 395. Perhaps Lysander had stepped in at the collapse of the said regime and installed a harmost, or perhaps the narrow oligarchy was still in power, but required the backing of Spartan spears as had been the case in Athens in 404/3. (For Lysander's Megarian connections cf. Meyer, RE xv. 192; Legon, 1981: 260f.) At any rate Plato in the Crito (dramatic date 399) speaks of Megara as a well-governed city (53b). It would thus seem that by that date Megara was under a moderate oligarchy, as indeed was Thebes. We are ignorant of the stance Megara adopted in the Corinthian War. Beloch (GG iii. 1. 70 n. 3, followed by Accame, 1951: 61f.) believes that it joined the anti-Spartan alliance of Athens, Thebes, Argos, and Corinth, and underwent a similar democratic revolution to Corinth, the democratic government being still in power at the time of Leuctra. As evidence he cites Plato, Theaet. 142c and the present section of D -- the reference to the Megarian demos. Theaet. 142C, however, probably refers to the warfare in the Corinthia in the 360s (cf. A. E. Taylor, Plato ( 1929 3rd edn.) 320), and in any case all it shows is that, at the time the injured Theaetetus was being transported home, Megara was not hostile to Athens. A state of neutrality could account for that. Megara was possibly neutral in the Corinthian War (cf. Sinclair, 1978: 33 n. 17; Legon, 1981: 263ff.). There are three pieces of evidence which prove conclusively that in the 370s Megara, whatever its constitution, was a loyal member of the Peloponnesian League. ( Legon, 1981: 274ff., on the other hand, sees Megara as neutral after 375.) First, there is D's statement that Megara and Corinth together constituted the 6th district of the reorganized Peloponnesian League in 378 (31. 2). Secondly, on his way back from Boeotia in 377 Agesilaus stopped at Megara (X. H. 5. 4. 58; Plut. Ages. 27. 1). Thirdly, it is clear from ps.-Demosth. 59. 35f. that when hostilities were resumed in 373 Megara was on the Spartan side and hostile to Athens. One might add that at the time of Epaminondas' second invasion of the Peloponnese Megara was still the ally of Sparta (68. 2 and n.). It is of course quite possible that Megara was both democratic and allied to Sparta at this juncture, as were Phlius in the Corinthian War (but see 19. 3 n.) and Syracuse in the -335Peloponnesian War. Phlius needed Spartan protection against Argos, and Syracuse and Megara against Athens. But it is also possible that Megara was still under an oligarchic regime of some description, in spite of D's κρατηΘέντες ὑΠò του+̑ Δήμον. The mere use of the word demos, especially in a 4th-century context, does not necessarily imply a democracy; cf. Larsen, CP 49 ( 1954) 14 n. 2; id., 1968: 175, 178; Tod 144 (362/1) line 30: Δη+̑μον [Φλειαρίων; Roy, 1971: 587 n. 95. In any case, what matters here is that there is nothing inherently improbable in an attempted coup at this time to change the existing political order. ὁμοίως Σικυωνίοις κτλ: the Sicyonian case appears plain enough on the surface. A pro-Spartan oligarchy was in control there till 368 though in this same year the city was forced to join Thebes (X. H. 7. 1. 18, 44ff.; D 15. 69. 1; 70. 3; see nn. ad loc.). The attempted revolution therefore was in all probability due to democratic elements. Cf. Griffin, 1982: 67f. and n. 19.

A tantalizing possibility is introduced by a badly fragmented citation from the Chabrias monument mentioned above (33. 4n.), restored as ο[ἱ στρτριω+ ται Οἱ . . . ] | [ . . . Σι]κυωνίων by Schweigert, 1940: 318f. Given that the monument dates from the mid- or later 370s an explanation could be found for the mention of Sicyonians within the context of the war against Sparta whose ally Sicyon was at the time (thus Schweigert, 1940: 318f.; Burnett and Edmonson, 1961: 87). Alternatively one might venture to suggest an Athenian involvement in northern Peloponnese in support of democratic movements in the area in the way that the Zacynthian democrats were supported (45. 3f.). A commercial treaty between Athens and Troezen (IG ii2 46) dated to the mid-370s by Woodhead (1957b: 225ff.; but cf. Lewis, 1959: 248ff.; Gauthier, 1972: 166f.) may point to Athenian diplomatic activity being used to exploit antiSpartan feeling in the Peloponnese. The possibility that the second agreement recorded on the reverse of the stone of IG ii2 46 was with Sicyon (only the last letter of the name survives -- N) is remote, but still a possibility ( Burnett and Edmonson, 1961: 86 n. 36). A restoration of the citation on the Chabrias monument such as o[ἱ στρατιΩ+̑ται Οἱ ßοηΘἡσαντες] | [τῳ+̑ Δήμῳ τω+̑ν Σι]κυωνίων is as likely as that of Burnett and Edmonson, 1961: 87. 40. 5. Παρὰ Άλιαίοις κτλ: a pro-Spartan government was installed at Phlius by Agesilaus in autumn 379 (19. 3n.) and this was strong enough to survive Leuctra and the subsequent attacks -336on it by its many enemies (X. H. 7. 2. 1 ff.). Xenophon does not mention any Phliasians exiled at the time of the settlement, but there must have been many for 600 were killed on this occasion alone according to D (and cf. X. H. 7. 2. 5 where numbers of exiles are spoken of in 368). Άρούριον ὀχΥρóν: this may be the same as τò ὑπέρ τοῦ Χραίον Τρικάρανον: held against Phlius by Argos later (X. H. 7. 2. 1). For the probable location of the latter see Pritchett, 1965-85: ii. 103 ff. I cannot agree with Legon, 1967: 335 n. 75, that the present incident is in fact to be identified with X. H. 7. 2. 5-9. The circumstances are totally different. The Phliasian exiles played but a small part in Xenophon's incident of 368; cf. Pritchett, 1965-85: ii. 109 n. 83.

Archon Socratides, 374/3 41. 1. Ρ+̔ωμαῖοι κτλ: 382 V; CAH2 vii. 2. 637. Only three tribunes are named, not four as announced, but L. Papirius (placed second in Livy's college of six: 6. 22. 1) must have been omitted in the way explained by Drummond, 1978: 87f.; id., 1980: 61f. MS F, as often, offers additional names (cf. Vial, 139). See also pp. 27 f.

41-44. The invasion of Egypt The Persian narrative is resumed from 29. 4. See pp. 100 f. 41. 1. ἐΠὶ Δέ τούτων Περσω+̑ν: this may well be the chronographer. The invasion of Egypt in fact spanned two archon years as it began ἀρχομένον τοῦ Θέρονς (May/June: section 4), and ended after the Etesians and the annual inundation of the Nile. The Nile begins to rise about the time of the summer solstice (1. 36. 7; Herod. 2. 19. 2) and it reaches its highest level by September (cf. Burton, 141). The invasion took place either in 374 or 373 and D's date thus marks either its end or its beginning. It was certainly over by autumn 373 for Iphicrates was back in Athens in the autumn of that year where he replaced Timotheus as general (X. H. 6. 2. 13; ps.-Demosth. 49. 9, 22). On the other hand it cannot belong earlier than 374 for surely it cannot antedate the Peace of 375 (probably concluded summer/autumn: see the excursus at 45. 3), and several months ought to be allowed for the journey of the Greek mercenaries to Ace. The expedition made but slow progress -337(section 5) and Kienitz, therefore (1953: 90 n. 3), queries D's statement that it set out ἀρχομένον τοῦ Θέρους. That does not leave enough time, he thinks, to accommodate the events of the campaign. He tentatively suggests that the campaign began either a year earlier in spring 374 or, more likely, early in 373 and D has got the season wrong. The first suggestion goes beyond the evidence, for the narrative of chs. 41-3 is of a single summer's campaign; and the second suggestion is unnecessary. The four to five months (May/June -- Sept./Oct.) envisaged by D's source seem sufficient enough time for the events related. ἀΆεστηκóτας: shortly after Darius' death in 405/4 Egypt was free of Persia and ruled by Amyrtaeus, though parts of it, notably Elephantine, continued under Persia for a little while longer; cf. Kraeling, 1953: 111ff. But even before this date Persia was not fully in control in Egypt to judge from 13. 46. 6 and Thuc. 8. 35. 2; cf. Kienitz, 1953: 73ff.; CHI511f. The Persians had made two previous attempts that we know of to reconquer Egypt. The first, under Abrocomas, had come to nothing because of Cyrus' revolt (X. An. 1. 4. 3, 5). For the second see 2-4n. ἡγοῦντο μισΘοU0386óρων: not so. Pharnabazus was clearly the supreme commander as is shown by 29. 3; 41. 5; and 43. 2. Iphicrates only commanded the Greek corps. Pharnabazus of course is the well-known former satrap of Hellespontine Phrygia (29. 3n.). He must have been in his seventies by now for he is said by Xenophon to have been older than Agesilaus (H. 4. 1. 31). For his career see Lenschau, RE xix. 1842ff μετάπεμΠτος: if D's statement that Iphicrates was sent to Pharnabazus at the same time that Chabrias was recalled from Egypt is accepted (29. 3 f.), then Iphicrates went out in 380/79 (29. 2n.). It may seem a little surprising that he spent some six years with Pharnabazus before they took the field against Egypt in 374/3, but that is not impossible. Certainly the sources are silent about Iphicrates in these years as far as Greece is concerned. He does not, for example, appear to have been involved in the warfare of the early years of the Confederacy (unlike Timotheus and Chabrias), so he may well have been in Persian service. We are told that the preparations against Egypt took up a number of years, much to Iphicrates' chagrin (41. 2; see next n.), though he was not idle all the time (41. 3n.).

41. 2. ἔτη δέ Πλείω: 'more years (than necessary)'; cf. 44. 1, τὴν ἐν -338-

τῳ+̑ Περσικῳ+̑ πολέμω πολνχρóνιον έμπειρí. The preparations began in 380/79 (29. 2n.; cf. Parke, 1933: 62; Kienitz, 1953: 89) and as the invasion took place in 374/3, some six to seven years were taken up by them. This is not unlikely. Persian preparations were notoriously slow and sometimes they came to nothing (see n. below). Appointed in 380/79, Pharnabazus may have found it difficult to procure enough Greek mercenaries for war broke out in Greece shortly afterwards; hence the King's desire to end that war (38. 1 and n.). ὁέν μέν←ϕικρáτης ὁρΩ+̑ν κπλ the account is consistently favourable to Iphicrates. He is raised to the same level with Pharnabazus (41.1, 3; 42.4), his military adroitness is stressed (42.5; 43.1), the blame for the expedition's failure is laid squarely on the timidity of the Persian generals as well. as on their jealousy and even fear of Iphicrates (43), and the narrative is concluded with a eulogy of Iphicrates' military genius (44). All this agrees well with Ephorus' attitude to non-Greeks. For the emphasis he laid on the importance of Greek generals and soldiers to Persia (and Egypt) and his panhellenism, see pp. 114 f. ὁ δ 0 ϕαρνá ßααος ἀπεκρíϑη κτλ: cf. section 5, ώς ἐπí αν γàρ οí πω+̑νΠερσω+̑ν στρατηγοì κτλ , and the comment at Hell. Oxy. 19. 2 regarding the financial difficulties Conon faced in 395; cf. also Isocr. 4. 142 which refers to the same (395) occasion. The shortage of money related in Polyaen. 3. 9. 59 probably refers to this period and the conspiracy at Ace mentioned at 3. 9. 56 may have been caused by the long delays and the money difficulties. 41.3. ἄκην: Ace or Acco (later Ptolemais and even later St Jean d'Acre or simply Acre) in north Palestine; see Strabo 16 C 758: εîϑ ἡ Πτολεμαïς ἐστι μεγáλη πóλιςἢν ἂκν ὠνóμαον πρóτε√ονέ+̑ ἐχðέîχρω̑ντο,ὁρμηττητηρíὡ πρòς τἠν Áíγ . For Persians at Ace see Stern, 1982: 254, 278. Polyaen. 3. 9. 63 speaks of a landing manœuvre by Iphicrates in Phoenicia against hostile Phoenicians. As Iphicrates did not to our knowledge operate in this area at any other time we must ascribe this episode to the years 380/79-374/3 when he was in Persian service. Kienitz, 1953: 90 n. 4, doubts the historical basis of the episode because D's detailed account says nothing about any fighting in Phoenicia, and because the Persians would not have chosen Ace as their base if Phoenicia was in revolt. D's silence is not of course an argument against the historicity of the episode; and we know that parts of Phoenicia and Cilicia had -339risen in revolt in the 380s at the instigation of Evagoras and Acoris (2. 2-4 nn.). Some of these places may still have been in revolt and Pharnabazus may have taken the opportunity of using Iphicrates and his mercenaries to subject them. Ace is mentioned in Demosth. 52. 20, Which reference dates to this time. The people mentioned there may have been hoping to trade with the Persian camp. FGH 115 (Theop.) F263 probably refers to the successful campaign of 344/3; contra Burn, CHI 372. ßαρßáρων . . . μυριáδες: the usual impossible information about barbarian numbers; see p. 130. 'Éλλфνων. . . δισμυρíω: repeated from 41. 1. In the longer narrative of his source the repetition would have been less awkward. Twelve thousand according to Nepos, Iphicr. 2. 4. Both figures are quite credible. Cyrus, a generation earlier, had recruited no less than thirteen thousand Greeks (X. An. 1. 2. 9). τριήρεις . . . : see pp. 131 f. τριήρεις . . . διακóσιαι: a hundred of these were used according to Polyaen. 3. 9. 63 to defeat Phoenician rebels (see above). 41. 4. ἀρχομενου δè τοû ϑéρους: see 41.1 n. For such chronological indications in Ephorus see Introd. n. 264. 42. 1. Νεκτáνεßις: Nectanebos I of Sebennytus had been on the throne since c.380 (2-4n.). ςυσπροσíτου . . . τές Áἰγúπτο⊎: see I. 30f. and cf. Polyaen. 3. 9. 38, Iphicrates' stratagem undertaken τῃς χω+̑ρας Ó↕σης Ó↕σ ἀλιμενον. 42. 2 ἑπτà στóμπτο⊎: cf.1. 33. 7f.; Burton, ad loc.; Burn, CHI 373 for a good map. 42. 4. Μενδ

σιον: the third mouth from the east.

43. 1. μετà δè τα↑τα . . . εσϕáλησαν: the only other reference to the quarrel between Pharnabazus and Iphicrates which appears to have contributed to the collapse of the campaign is in Plut. Artox. 24. 1: πóλεμον δè ταû το↕ς Áιγνπτíονς διà ϕαρναßáζον καì ¡ϕικρáτονς εζενéγκων 22a2a Áραζéρζης àπèτνχε, σπασιασàντων èκεíων. 43. 1-2. ó μèν ←ϕκρáπτης . . . π↑ϑóμενος κΠλ: according to Ephorus therefore Iphicrates and Pharnabazus differed on strategy. It would seem that the two had not been on the best of terms from the beginning (41. 2) and now, while the mercenary leader was in favour of moving against Memphis immediately, Pharnabazus

preferred to wait till all of his forces had concentrated; and he also turned down Iphicrates' second suggestion -340that he allowed to move against Memphis alone with his mercenaries. τó τε ϑρáσος αντοσ αντον καì την αρετην νπωπτενσε,μ≉κατ Íϕϑóνο⊎ν αντω καì διαßολàς αδíκονς προσῃπτον(section 2)! And Pharnabazus' Persians εϕϑóνον αντω καì διαßολàς αδíκολàς προσῃπτον(section 2). After Iphicrates had fled to Athens and Pharnabazus' envoys had demanded that he be punished as responsible for the failure of the campaign, the Athenians promised that if there was any substance in the charge Iphicrates would be punished as he deserved. Shortly afterwards he was elected general. It is very likely that we have in D a reflection of Iphicrates' apologia offered in Athens on his return. The ill success of the expedition which had involved so many Greeks would doubtless have been discussed extensively in Greece, and Iphicrates' explanation that he was not to blame for it, that Persian procrastination and jealousy of him were, would have found ready favour with the many who viewed the Persian military machine with contempt. It would have been grist indeed to Ephorus' panhellenist mill to accept such an explanation (see pp. 114 f.). It may at first sight seem surprising that Iphicrates proposed sailing against Memphis which was situated some 120 miles up the Nile from the Mendesion. But the capture of Memphis might have settled the issue in the Persians' favour. For the importance of Memphis' position see 1. 50. 3. To control effectively both upper and lower Egypt one had to hold Memphis. Possession of Memphis had been crucial in the 450s when the Athenians and Egyptians battled against Persia. On being called in to help, the Athenians had then sailed up from the Mendesion mouth and captured two-thirds of Memphis. Their failure to capture the rest had proved disastrous in the end (Thuc. I. 104, 109f.). One cannot of course expect Iphicrates to have been familiar with the details of the Athenian expedition to Egypt, but the importance of Memphis must have been common knowledge; cf. 16. 48. 7. 43. 2. την πολην στρατραιáν . . . εϕϑóνο⊎ν: Rhodoman rightly suspected a lacuna here, but the loss is not serious and the meaning of the passage is easily recoverable. Something like οι δèπερì ϕαρναßαζον(see above) is clearly indicated, as also suggestδη ed by Rhodoman. 43. 4. καì των ετησíν ἤ γενομενωω: this is one of several instances where Ephorus' use of Callisthenes' Hellenica is strongly indicated. Earlier in the Histories Ephorus had offered a different -341explanation (perhaps one of several?) for the annual inundation of the Nile (1. 39. 7). Here he associated it with the Etesians, which was Callisthenes' explanation ( FGH 124 F12 with n. and RE X. 2. 1706). For Ephorus' sources see p. 104. The Etesians, i.e. the summer winds which blow from the Mediterranean, have nothing to do with the flooding of the Nile of course; cf. Burton, 138 ff. 43. 5. καϑáπερ κóνων επαϑεν: the sources disagree about what actually befell Conon after his arrest by Tiribazus; cf. 14. 85. 4; Isocr. 4. 154; X. H. 4. 8. 16; Nepos, Con. 5. 3f.; schol. Aristoph. Eccles. 196; Swoboda, RE xi. 1332. 43. 6. στρατνγòν . . . επì τò να⊎τικóν: see 34. 4n. He replaced Timotheus as general and took command of the expedition to relieve Corcyra (47. 3 n.; X. H. 6. 2. 13). 44. Digressions by way of praise or censure are usually to be found at the end of a public man's life (pp. 5 if.). However, a chapter on Iphicrates' military arete at this point is understandable. Iphicrates owed his fame to his successful career as a general (44. 1; cf. X. H. 6. 2. 27 ff.) and if D (Ephorus) is correct in ascribing Iphicrates' military innovations to his long experience (πολνχρóνιον εμπειρíαν) as a general in Persian service, then ch. 44 is a good epilogue to the Egyptian campaign. And in fact the last sentence of the chapter shows that this eulogistic assessment of Iphicrates concluded the Egyptian expedition. Besides, Ephorus may have felt that such a favourable appraisal was the best way to end an account which took pains to prove that Iphicrates was not responsible for the failure of the campaign. 44. 1-4. προσλαßóμενον ο⊎ν αÚτòν κτλ: only D and Nepos, Iphicr. 1. 3 f., refer to the military innovations of Iphicrates. According to D the Athenian general replaced the large shield of the hoplite, the aspis, with the smaller and lighter pelte; hence the name of the new troops, peltastai. The result was greater mobility for the soldier. The spear was made half as long again, and the sword was almost doubled in length. The footgear Iphicrates made light and easy to untie and these boots became known as ιϕικρατíδς.(section 4 -- see n. there). Nepos' information is essentially the same, but with these differences. The length of the spear was doubled while that of the sword was simply increased and Nepos further adds that the body armour was changed: bronze breastplates and chain mail were replaced by linen corslets. -342First, both D and Nepos are obviously mistaken in supposing that the Greek heavy infantry in general, the hoplites, was converted to a medium-heavy force of peltasts (cf. Arrian's definition of the three basic Greek infantry categories, Tact. 3. 1-4), and that henceforth there were no longer hoplites, but only peltasts! And of course the peltasts were not new to the Greek world, but had been known since at least the early 5th century. It is surely inconceivable that Ephorus was responsible for these absurdities. At the same time one is puzzled by the fact that contemporaries D and Nepos made the same mistake, apparently independently of each other. There are two possibilities: (a) both D and Nepos (or perhaps the source used by Nepos which ultimately went back to Ephorus) misunderstood Ephorus, especially if (see below) Iphicrates armed some at least of his hoplites in this way while serving in Egypt in order to suit the conditions there, or (b) D and Nepos used the same biographical or

military handbook which had got its facts wrong. It is certain that D did occasionally use secondary sources to supplement his main authorities (see p. 50 ). One cannot decide between the two possibilities though (a) seems to me the more likely. The blunder apart, what is one to make of their information? This is a thorny problem and a detailed discussion here is impossible. The reader is referred to Lippelt, 1910; Kromayer and Veith, 1928: 89; Parke, 1933: 79ff.; Best, 1969: esp. 102ff.; Anderson, 1970: esp. 129 ff.; Griffith, Studies Edson, 161 ff. No two scholars agree on all points and I will confine myself mainly to expressing my own opinion. What emerges from D and Nepos is that the new equipment was meant to combine the best features of light and heavy troops. The long, thrusting spear and the longer sword were evidently intended for hand-to-hand fighting, while the light but adequate pelte and absence of metal armour rendered the new troops more mobile than hoplites. Unfortunately, there is not a scrap of evidence for peltasts of this type in action in the 4th century (pace Parke, 1933:81, who believes that Chabrias' stratagem in Boeotia in 378 was executed by this new type of peltast; see 32. 5n. Griffith, 1935: 5, 17, 196, 239, 317 -- he has now revised his opinion: Studies Edson, 163 n. 8 -- is of the opinion that the 'Iphicratid' peltast as explained by D and Nepos became the standard mercenary of the 4th century, playing an -343First, both D and Nepos are obviously mistaken in supposing that the Greek heavy infantry in general, the hoplites, was converted to a medium-heavy force of peltasts (cf. Arrian's definition of the three basic Greek infantry categories, Tact. 3. 1-4), and that henceforth there were no longer hoplites, but only peltasts! And of course the peltasts were not new to the Greek world, but had been known since at least the early 5th century. It is surely inconceivable that Ephorus was responsible for these absurdities. At the same time one is puzzled by the fact that contemporaries D and Nepos made the same mistake, apparently independently of each other. There are two possibilities: (a) both D and Nepos (or perhaps the source used by Nepos which ultimately went back to Ephorus) misunderstood Ephorus, especially if (see below) Iphicrates armed some at least of his hoplites in this way while serving in Egypt in order to suit the conditions there, or (b) D and Nepos used the same biographical or military handbook which had got its facts wrong. It is certain that D did occasionally use secondary sources to supplement his main authorities (see p. 50 ). One cannot decide between the two possibilities though (a) seems to me the more likely. The blunder apart, what is one to make of their information? This is a thorny problem and a detailed discussion here is impossible. The reader is referred to Lippelt, 1910; Kromayer and Veith, 1928: 89; Parke, 1933: 79ff.; Best, 1969: esp. 102ff.; Anderson, 1970: esp. 129 ff.; Griffith, Studies Edson, 161 ff. No two scholars agree on all points and I will confine myself mainly to expressing my own opinion. What emerges from D and Nepos is that the new equipment was meant to combine the best features of light and heavy troops. The long, thrusting spear and the longer sword were evidently intended for hand-to-hand fighting, while the light but adequate pelte and absence of metal armour rendered the new troops more mobile than hoplites. Unfortunately, there is not a scrap of evidence for peltasts of this type in action in the 4th century (pace Parke, 1933:81, who believes that Chabrias' stratagem in Boeotia in 378 was executed by this new type of peltast; see 32. 5n. Griffith, 1935: 5, 17, 196, 239, 317 -- he has now revised his opinion: Studies Edson, 163 n. 8 -- is of the opinion that the 'Iphicratid' peltast as explained by D and Nepos became the standard mercenary of the 4th century, playing an -343important part in the warfare of that century. But where is the evidence?). The peltast/hoplite hybrid is not to be found either in Classical or Hellenistic antiquity, a decisive fact which has not been stressed sufficiently. On the contrary, battles in Greece are still fought with hoplites, and the peltasts involved are, as far as we can see, the old hit-and-run javelin-men. ( Best, 1969: 41ff., is wrong that by 'peltasts' Xenophon means spearmen equipped with a shield of some description. A single passage, H. 2. 4. 12, which Best discusses at length, is enough to disprove this. πελτοϕóροι τε καì Ψιλοì ακοντισταíis obviously translated 'peltecarrying as well as light (i.e. without shields) javelin-men'. Both the word ακοντισταí (which Best persists in mistranslating 'spearmen') and the position of these troops (behind the hoplites on a slope) make it clear that they were javelin-men. Thrasybulus had intended them to throw over the heads of his hoplites, while the πετροßóλοιwho could naturally reach further, were posted behind the javelin-men.) All the same, the story of the new equipment cannot be a fabrication of Ephorus. There must be some substance to it especially as D provides a definite context for the reforms -- the years Iphicrates spent in Persian service (44. 1). One might therefore hazard a hypothesis (cf. Anderson, 1970: 130f.). Like Cyrus' Ten Thousand, the mercenaries Iphicrates commanded for the King against Egypt would have comprised both hoplites and peltasts (armed with javelins), with the former predominating. Xenophon in the Cyropaedia gives a good idea of the kind of opposition the

(7. 1. 3. 3); cf. An. 1. 8. 9.

Now, Iphicrates was distinguished from most of his fellow mercenary leaders because he combined brains with brawn. The list of his stratagems in Polyaenus is exceptionally long -- 63 in all. According to 3. 9. 38, on being informed that the Egyptian coast lacked harbours Iphicrates forthwith thought of an answer to the problem. It is possible, therefore, that Iphicrates re-equipped a part at least of his Greek force in such a way as to make it more effective against the heavy and unwieldy Egyptian infantry (cf. X. -344-

Cyr. 6. 4. 17). He gave them longer spears and swords in order to outreach the Egyptians, and light shields and no metal body armour so as to exploit Egyptian immobility. Interestingly both D and Nepos agree that a longer spear was substituted for a shorter one, not that javelins were replaced by spears, so that the troops thus reequipped seem to have been hoplites and indeed a hoplite would have been more adept at wielding what amounted to a pike than a peltast. We hear too that while at Ace Iphicrates undertook amphibious operations against Phoenicia (Polyaen. 3. 9. 63) and the new type of soldier described by D and Nepos would have been better suited to that kind of operation. That Iphicrates' innovations occurred in Phoenicia and Egypt seems to be indicated by Nepos' statement that metal armour was replaced by linen. Greeks generally wore woollen clothes, for linen was expensive. It had to be imported largely from Egypt (cf. Michell, 1940: 188, 290f.). But in Egypt linen was common and it was worn by the lower classes (cf. Herod. 2. 81. 1). Linen corslets had originated in Egypt and though not unknown in Greece (cf. Page, 1959: 215f.; Anderson, 1970: 22f.) they were characteristic of the Persians and other non-Greeks (Herod. 1. 135; X. An. 4. 7. 15; 5. 4. 13; Cyr. 6. 4. 2). In Greece corslets were made of leather rather than linen. In a hot, humid climate such as that of Egypt a linen corslet, apart from its lighter weight, was far more comfortable to wear than leather or metal; and linen was plentiful too. Hence Iphicrates' preference. One might suggest, therefore, that Iphicrates' reforms were temporary. They were meant for Egypt where they suited the military situation (and climate) but, as was said above, there is no evidence subsequently for this novel infantry. With Iphicrates' flight from Asia in 373, the new-style soldier disappeared from history. There was one exception. The Iphicratean boots (see below) were Iphicrates' one lasting reform (though cf. Griffith, Studies Edson, 163 f. and n. 11); that and the professionalism he applied to soldiering. To conclude: D 15. 44 and Nepos, Iphicr. 1. 3f. derive from an Ephoran chapter on the military ability and innovations of Iphicrates. But the Ephoran information was misunderstood by the 1st-century BC writers to whom hoplites and peltasts were mere names. D certainly had little idea of what a peltast was; cf. 16. 24. 2. ( Best, 1969: 7, 102ff., does not believe in the reforms of -345Iphicrates though he does believe ( ibid. 5ff.; Griffith, Studies Edson, 162f., agrees) that the approximate type of spear-carrying 'peltast' described by D and Nepos had existed in Thrace since at least the time of Peisistratus; Iphicrates merely made use of it. He bases this belief on a small number of Attic Late Archaic vase representations of Thracian peltasts seemingly armed with single spears rather than twin javelins. None of the vases, however, is later than c. 490. One might seriously question, therefore, whether these representations constitute evidence for the peltasts of the 4th century. Best also fails to allow for the artistic licence of the vase decorators which was considerable. As he himself notices the figures of spearmen on 5th-century vases later than c. 490 which he identifies as peltasts carry hoplite shields, not peltai. And Lucian Necr. dial. 27. 3 is hardly admissible as evidence. For Griffith, Studies Edson, 166, the only innovation connected with the Egyptian campaign was the rearming of the mercenary hoplites with longer spears.) 44. 1. εν τῳ+̑ Περσικῳ+̑ πολεμΰ: the war for, not against, Persia. The latter is the natural meaning, but cf. Thuc. 1. 128. 3 τòν Éλληνι2κòν πóλεμον, 'the Greeks' war', 'the war for Greece' (cf. HCT i ad loc.). 44. 2. σ⊎μμèτροÚς: 'symmetrical' or 'of suitable or moderate proportions'. The word 'pelte' generally signified a small, light shield. Such a shield was considered specifically Thracian. See Best, 1969: 3f. 44. 4. μéχρι τοû νûν: This is probably D rather than Ephorus. The type of boot invented by Iphicrates survived throughout antiquity so that D could well have been familiar with it in the ist century BC. Alciphron, Letters 3. 21 (Schepers and Loeb edns.), refers to Íϕικρατíδας. He is of the 2nd or 3rd century AD though the letters purport to be of 4th-century Bc Athenians. Sallustius, a 5thcentury AD cynic philosopher, is said to have worn Iphicratid boots (Photius from Damast. Vit. Isid. p. 1045); cf. Suidas s.v. Íϕικρατíδας; Steph. Byz. S.V. íϕικρατíς.

45-46. The 'cold war' period, 375-373 See below for the chronology. 45. 1. κατà δé τὴν Éλλáδα κτλ:with the Egyptian campaign completed D, certainly following Ephorus' own arrangement, takes up again Greek affairs which were interrupted at 40. 5. -346What is said here rather confirms that disturbances occurred in 375/4 after the Peace of that year. The troubles which now plagued Greece are those spoken of at 40. 1. Both Athens (now a coprostates of the Peace) and Sparta exploited the unrest by backing their respective supporters in the cities. D goes on to exemplify his statement with the instance of Zacynthus. This is hardly a postLeuctra situation. After Leuctra Athens no longer vied with Sparta for support in the cities (see further 40 n.). διá: 'because of the unaccustomed government (or constitution)'. Both the original King's Peace and its renewal in 375 were followed by disturbances as many cities broke away from Sparta, or as they attempted to do so while switching from oligarchy to democracy. That was the picture painted by Ephorus and it is in sharp contrast to that of Xenophon (see 5. 1-3; 40. 1-2 and nn.). κοινὴν αναρχíαν: as opposed to κοινὴν ειρὴνην. This is almost certainly Ephoran. For Ephorus and the Common

Peace see p. 127. 45. 2. χρóνον ολíγον . . . ε ρὴνης: for chronological indications in Ephorus see Introd. n. 264. Just how brief was the period of peace? About two years it would seem as peace was made probably in summer 375 and fighting was resumed in autumn 373 (see the chronological excursus below). It should be noted that there was not now as in 378 a formal proclamation (by Athens) that the Peace had been broken by Sparta, and a declaration of war (29. 7). There was rather a gradual disintegration of the conditions of Incidents during it peace until Sparta and Athens found themselves fighting each other again in the Ionian eventually led to a renewal of hostilities. It is not surprising that peace did not last. Athens, with her enlarged Confederacy and her new position as co-prostates of the Peace, was no longer as timid as she had been, and it may even be that she was more to blame for the fresh outbreak of war than Sparta. In 378 on the other hand Sparta was clearly the guilty party (see 45. 4 n.). Ephorus at any rate did not on this occasion point the finger at Sparta, but seems to have considered both cities responsible for the renewal of war. -347διóπερ ενΖακűνϑΰ . . . σűνεμáχοűν: the text is obviously wrong as it stands. The exiles whom Timotheus received were democrats as Zacynthus was under a pro-Spartan oligarchy at the time (see 45. 3 n.; X. H. 6. 2. 3). Because of this Le Paulmier suggested a lacuna between πáντας and οτοι. If that is the case then a context must be found for the statement that the demos, once in power, exiled all those who had ruled the island under the Spartan epistasia. There is no real difficulty. The present passage should be seen in relation to 5. 1-3, i.e. to the strifes of the years after the Peace of Antalcidas. Though these troubles were of the same nature and had the same cause as those which followed the Peace of 375 (40 and n.), our passage belongs with the former because of its date the exiled oligarchs were again in control and their democratic opponents themselves in exile when Timotheus arrived on the scene in 375 before the Peace of that year. What did the missing text relate? Presumably it told of the return of the oligarchs, perhaps with Spartan help (5. 2-3 n.), and of the subsequent expulsion of the democrats. It is not impossible of course that the missing information is not the result of a lacuna, but of D's telescoping technique. εγο⊎μενον τοű ναűτικοû: see 34. 4n. 45. 3. ω ποσηγóρεűον Áρκαδíαν: there is no justification for the identification of Arcadia with the Nellus of Tod 123 (=Svt 257) lines 131ff. That entry is perhaps earlier than the conclusion of peace (it appears to be a spillover from the front face of the inscription), while the occupation of Arcadia post-dates the Peace ( Cawkwell, 1963: 88). Still, the democrats at Nellus were almost certainly the same as those planted at Arcadia by Timotheus. While still at Nellus they joined the Athenian Confederacy and were entered on the stele of the decree of Aristoteles. Cf. Mitchel, 1981: 73 ff., who argues the same. It is not clear whether the stronghold occupied was already called Arcadia or whether the exiles named it so. If the latter then Dugani's thesis is strengthened (1979: 325ff.). He sees the name as an anti-Spartan slogan, 'Arcadia' alluding to a future free and democratic Arcadia. We have seen (40. 4n.) that there is a hint of Athenian support for anti-Spartan movements in the Peloponnese. ßοηϕòν . . . Τιμóϕεον: cf. X. H. 6. 2. 2. D neglects to mention that Timotheus now withdrew to Athens. -348Excursus: the chronology of the years 375-371 It is necessary at this point to deal with the chronology of the years 375-371. This is done under three broad headings: (1) the Peace of 375/4; (2) the period of 'cold war' and the warfare of 373-371; (3) the date of Cleombrotus' dispatch to Phocis and the election of Jason as tagos of Thessaly (this, it is argued, belongs after Leuctra, but it is discussed here because of its association in Xenophon with Cleombrotus' expedition, and because it is, as a result, usually dated in 375). 1. The Peace of 375/4 The precise date of the Peace of 375/4 is in dispute. Dates given to it vary from summer 375 to autumn 374 (cf. Buckler, 1971b: 353 n. 4). The most likely date is summer 375 though the nature of the evidence cannot allow of certainty. Two considerations have been advanced in favour of a 375 date. (a) Timotheus' naval expedition to the Ionian Sea belongs to the summer of 375 (25. 1n.). He won the battle of Alyzeia on the day of the feast of Scira, i.e. some time in June/July of the same summer (36. 5n.). It is possible that the Peace was concluded on the day of the festival of the Synoecia (16 Hecatombaeon), i.e., sometime in July/Aug. ( Cawkwell, 1963: 90 n. 56 in conjunction with Jacoby, FGH iiib. 1. 523ff. Cawkwell's suggestion is convincing). But July/Aug. 375 or 374? The latter year seems to be excluded by the fact that Timotheus would have found it very difficult to winter in the west because of his severe shortage of money ( X. H. 5. 4. 66; cf. ps.-Arist. Econ. 1350a30; Beloch, GG iii. 2. 234; Cawkwell, 1963: 89; Tuplin, 1984a: 564 n. 77. Some disagree; cf. e. g. Sealey, 1957: 102; id., 1993: 285 n. 59). Hence peace was concluded about a month after Alyzeia, in middle or late summer 375. (b) The second consideration adduced in support of a 375 date is unacceptable. It is based on Xenophon's arrangement of his narrative of these events ( H. 5. 4. 63-6. 2. 1). Its best exponent is Cawkwell, 1963: 89 (and see Buckler, 1971b: 358. Cawkwell, 1981b: 44 and n. 23 presents a modified view). As this part of the Hellenica is crucial to this excursus, it is necessary to refute Cawkwell's (and Buckler's) arguments in detail.

Book 5 of the Hellenica appears to finish with the summer of 375 -349-

and with Timotheus in the Ionian Sea. Book 6 begins: οἱ μὲν ουν 'ΑΘηναιοι καί Λακεδαιμὸνιοι πεί ταυ+τα ἠ+̑σαν. οἱ δὲ Οηßαιοι ἐπεὶ κατεστρέψαντο τὰς ἐντῃ+ń Βοιωτίᾳ πόλεις, ἐστράτευον καί εἰς τὴν Φωκίδα. There follows the Phocian appeal to Sparta and the sending out of Cleombrotus. Σχεδὸν δὲ περὶ τουτον τὸν χρόνον, Xenophon continues, Polydamas of Pharsalus arrived at Sparta and requested help against Jason, unsuccessfully. Οἱ μἐν γάρ Λακεδαιμὸνιοι καί οἱ σύμμαχοι συνελέγοντο εἰς τοὺς Φωκέας, Xenophon goes on (6. 2. 1) . . . οἰ δ ΑΘηναιοι . . . πέμψαντες πρέσßεις εις ΛακεδαIOμονα εἰρήνην ἐποιήσανατι. Cawkwell thinks that 'if the Peace were to fall in 374, it would be necessary to stretch Xenophon's account uncomfortably'. The battle of Alyzeia, he believes, is synchronized in Xenophon with the Theban successes in central Greece, Cleombrotus' dispatch to Phocis, the appeal of Polydamas, and the Peace itself. 'Thus in view of his synchronisms there is no question of his account spilling over into a new campaigning year. All this fits satisfactorily into 376 and 375.' Yet Xenophon's synchronisms do not on the whole inspire one with confidence. Compare, for instance, 1. 1. 27 where he synchronizes the exile of the Syracusan generals with the battle of Cyzicus which belongs in 410 while the exile of the generals dates to 411, on the evidence of Thuc. 8. 85. 3 (see HCT v. ad loc.). Compare also 3. 2. 21 where he synchronizes the Spartan war against Elis with the operations of Thibron and Dercylidas in Asia Minor, and 3. 3. 1 where Agis' death and Agesilaus' succession to the throne are misdated ( Cawkwell, 1976a: 63 n. 8, is himself rightly sceptical of Xenophon's chronology). Buckler argues from the evidence of Xenophon too. First he accepts the synchronism of the conclusion of peace with the confrontation between Sparta and Thebes in Phocis ( 1971b: 353). Then he accepts the synchronism of Timotheus' sojourn in the Ionian Sea with Cleombrotus' presence in Phocis (p. 354 ). Otherwise, he argues, 'Xen. Hell. 6. 1. 17 makes no sense when he speaks of morai stationed near Sparta to be employed against Athenian triereis' (p. 354 n. 10). But this is a weak argument surely for Leucas, Acarnania, and Cephallenia are, if anything, further away from Laconia than Attica, and Corcyra is much further so. Thus, he postulates, 'if the Theban invasion of Phocis can be dated with a fair degree of certainty . . . it would put us one step closer to ascertaining the date of the peace treaty itself'. This he proposes to -350-

do by dating the battle of Tegyra to which, he assumes, the Theban move against Phocis was a strategic follow-up. Thus he passes from one fallacious argument to another for, granted that the battle of Tegyra was fought in the spring (because of the high water level of lake Copais -- pp. 356 f.), but the spring of which year? Buckler opts for 375, but in truth 376 would do just as well (Tegyra pre-dates the Peace, and 378 and 377, the years of Agesilaus' invasions of Boeotia, should surely be excluded). As we cannot decide between 376 and 375, the rest of Buckler's argument collapses: the operations in Phocis and western Boeotia cannot help with the dating of the Peace. (The analogy he draws (p. 358 and n. 28) between Xenophon's phrase οἱ μὲν ουν . . . περὶ ταυ+τα ἦσαν (6. 1. 1) and D's ἅμα δὲ τούτοις π⇁Áττομένοις Οηßαιοι μέν ἐστράτευσαν ἐπ+̕ Ο=̕ρχομενὸν (37. 1) is counter-productive for this very common phrase in D is useless for precise and often even for approximate dating. D's narrative here is broadly synchronistic and the ἅμα δὲ τούτοις π⇁Áττομένοις at 37. 1 roughly corresponds to the naval warfare of the two years 376 and 375.) Buckler concludes by stressing 'two attractive features' of peace in 375. First, Timotheus did not have to stay in the west over the winter. This is reasonable, as we have seen. But the second 'attractive' feature, that peace in 375 would explain why the Spartans did not invade Boeotia then (they had only just arrived in Phocis and the Peace intervened), is simply begging the question for we do not in fact know if Cleombrotus went to Phocis prior to the Peace of 375/4 (probably he did not; see section 3 of this excursus). It should be added further that even if the Theban victory at Tegyra (which was exaggerated in the Thebans' favour; see 37. 1n.) were to be dated to spring 375, it would not necessarily mean that Phocis was threatened in that year and before the Peace, for the evidence suggests that the Spartan garrisons were not evacuated from Orchomenus and elsewhere in Boeotia till after the Peace. To recapitulate: a (summer) 375 date for the Peace is probable, but only because of the necessity to have Timotheus back home before the winter and not because of Xenophon's arrangement of his material. -3512. The period of the 'cold way' and the warfare of 373-371 Cawkwell, 1963: 84 ff., appears correct in arguing on the basis of Isocrates' Plataicus and against Xenophon Hellenica that hostilities were not resumed till autumn 373. X. H. 6. 2. 2ff. represents a very short-lived peace broken by Timotheus' interference in Zacynthus: two of the Athenian envoys who had concluded the Peace had sailed directly (εὐΘύς) from Sparta to order Timotheus home; and directly (a second εὐΘύς) after the Zacynthians 'from the city' had complained to Sparta, the Spartans dispatched Mnasippus against Corcyra. The Athenians responded by sending out Timotheus, and the war was on. We know from ps.-Demosth. 49. 6 that Timotheus sailed from Athens in Apr./May 373. Hence, according to Xenophon, Mnasippus must have attacked Corcyra some time before that date and war broken out before the destruction of Plataea, destroyed εἰρήνης οὔσης (Isocr. 14. 1, 5, 14). As this contradicts Isocrates and Paus. 9. 1. 8 (where the destruction of Plataea is dated to the archonship of Asteius, 373/2), we had better assume that Xenophon is once more confused and turn to D for possible illumination. Ps.-Demosth. 49. 22 provides a terminus ante quem for the Spartan attack on Corcyra: as Timotheus' trial took place in Nov./Dec. 373, Mnasippus must already have been at Corcyra for some time; cf. section 13. The period of peace cannot have been as brief as Xenophon represents for D provides a series of what can aptly be termed 'cold war' events leading up to Mnasippus' attack on Corcyra, and these events must be allowed a certain amount of time to run their course. Xenophon's chronology is not saved even if the Peace is dated to 374 for he maintains that Mnasippus sailed before Timotheus and this, as was just seen, contradicts ps.-Demosthenes 49, Isocrates, Pausanias, and of course D. (For a defence of Xenophon's account see Gray, 1980: 315ff., but cf. Tuplin effective criticism of her case, 1984a: 537ff.)It would help the discussion if D's events in 45. 2-47. 7 were tabulated: 1. Timotheus lands exiles on Zacynthus (45. 3). 2. Zacynthian oligarchs appeal to Sparta (45. 4). 3. Spartan embassy to Athens (45. 4).

4. 5.

Aristocrates sent to Zacynthus (45. 4). Oligarchic revolt on Corcyra (46. 1). -352-

6. 7. 8. 9.

10. 11. 12. 13. 14. 15.

Appeal to Sparta of Corcyraean oligarchs now in exile (46. 1). Alcidas dispatched (46. 2). Appeal to Athens of Corcyraean democrats (46. 3). Athens votes to help Corcyra and Zacynthus. (a) Ctesicles is sent to Zacynthus (46. 3). (b) Naval preparations to aid Corcyra (46. 3). (c) (Timotheus sails out but remains in the Aegean; cf. 47. 2 f.). Mnasippus attacks Corcyra (47. 1). Timotheus is relieved of his generalship (47. 3). Great fleet prepared for Corcyra (47. 4). Meanwhile, Ctesicles is sent to Corcyra (47. 4). Ctesicles and the Corcyraeans defeat Mnasippus (47. 5f). Athenian fleet arrives in the west (47. 7).

The most important observation to be made from D is that the sending out of Timotheus (9b+c) was not in response to the expedition of Mnasippus (10) for the latter was not to sail for some months yet. This is correct as we have seen above. The dispatch of Timotheus appears to have been a precautionary measure following the appeal of the Corcyraeans (8) who believed themselves threatened by Alcidas (7). Both these naval expeditions (Alcidas and Timotheus) belong to the 'cold war' period. A reasonable and probably correct assumption is that as Alcidas' expedition proved a false alarm, Timotheus decided against hastening west and remained in the Aegean instead, acquiring more ships and crews and no doubt money (47. 2 n.; X. H. 6. 2. 12). Events 1-9 belong to the 'cold war' period. Events 3-7 and 9a are omitted by Xenophon, and 8 and 9b and c are mistakenly set after 10, as we have seen. (If Athens indeed interfered in northern Peloponnese at this time then the 'cold war' incidents should be extended; see 40. 4n.) Events 9a and 13 do not represent a doublet in spite of the near unanimity on the point amongst scholars, simply because the details given at 46. 3 and 47. 4 are different. We are dealing with two separate events at two different times. The appeal of the Corcyraean demos to Athens for help against the threat of Alcidas, real or supposed, seems to have been accompanied by an appeal by the Zacynthian exiles. At any rate, the Athenians decided to succour both (46. 3). Ctesicles was sent to Zacynthus, presumably with no troops as he was meant to lead the exiles, while a fleet was being prepared for Corcyra. -353This fleet eventually sailed under Timotheus in spring 373 (ps.Demosth. 49. 6) and cruised in the Aegean over the summer months finishing up at Calauria. By then, autumn 373, Mnasippus had arrived at and was besieging Corcyra and the city seemed lost. In their chagrin the Athenians relieved Timotheus of his command and elected Iphicrates in his place. While Iphicrates prepared over the winter of 373/2 the situation became desperate at Corcyra. Iphicrates could not sail before spring 372 and until his fleet was ready; a temporary measure was therefore taken: a small force was sent out to Corcyra to serve under Ctesicles who either went there directly from Zacynthus or, perhaps, from Athens, having returned home by then (9a probably belongs to autumn, winter, or spring 375/4; 13 to autumn, winter, or even spring 373/2). There is no need therefore to suppose that we are faced with a doublet here. Sections 46. 3 and 47. 4 refer to different occasions. What has led to the belief of a doublet is the fact that Xenophon first of all omits 9a (and really the entire series of 'cold war' events), and secondly associates 13, which properly belongs after the investment of Corcyra, with the fitting-out and dispatch of Timotheus' fleet in spring 373 which consequently belongs considerably earlier than the expedition of Mnasippus. D's order concerning the expeditions of Timotheus and Mnasippus is undoubtedly correct, and equally correct are the two tasks of Ctesicles: one came before the sailing of Mnasippus and concerned Zacynthus, the other after the investment of Corcyra and was aimed at saving that city ( Beloch, GG iii. 1. 157, does not see 46. 3 and 47. 4 as a doublet either, but I cannot agree with his explanation). The appeal to Sparta of the Zacynthian oligarchs (2) and the Spartan embassy to Athens (3) presumably followed not long after Timotheus' action on Zacynthus (1). No dates are available for the attempted oligarchic coup at Corcyra (5) and the expeditions of Aristocrates (4) and Alcidas (7). The fact that the Athenians apparently voted in favour of the Corcyraean and Zacynthian democrats at the same time (9) argues for the near contemporaneity of the expeditions of Aristocrates and Alcidas. Timotheus sailed in the spring of 373 with the considerable fleet of 60 triremes he had spent some time preparing (46. 3), and consequently the Athenian psephisma may belong to autumn or winter 374/3. -354-

Doubtless Ctesicles departed for Zacynthus at once as he had no preparations to make. We may therefore venture to date the expedition of Aristocrates to the sailing season of 374 though it could belong as early as autumn 375 (the Zacynthian exiles may have defended themselves for some considerable while before Athens dispatched Ctesicles). The expedition of Alcidas on the other hand was probably being prepared, or even just rumoured, when the Corcyraeans appealed to Athens (cf. the events of 434/3: Thuc. 1. 31 ff.). Though Alcidas may have sailed in the autumn of 374, he may belong as late as spring 373. The latter date is in fact preferable; Timotheus' time-wasting in the Aegean is better explained perhaps on the assumption that he had heard that Alcidas had avoided Corcyra and headed west (it is even possible that Alcidas' expedition never materialized). The above is in perfect agreement with the suggestion that Aristocrates and Alcidas were the Spartan nauarchs for the respective years 375/4 and 374/3: Beloch, GG ii. 2. 281, 289. Sealey, 1976: 355ff., on the other hand, ignores the evidence of ps.-Demosth. 49. 6, Isocr. Plataicus, and Paus. 9. 1. 8, and persists in supporting

Xenophon's sequence against that of D. His thesis is that Spartan nauarchs assumed their command in the spring (335 ff.), and that Mnasippus, therefore, being the nauarch for 373, attacked Corcyra in the spring of that year and before Timotheus left the Piraeus. The contemporary Xenophon, he thinks, is correct when he maintains that Mnasippus was already at Corcyra when Timotheus sailed. Furthermore, to set Mnasippus' expedition in spring 373 argues for 374 as the date of the Peace, for does not Xenophon explicitly say that very little time elapsed between the Peace and the attack on Corcyra? Sealey is wrong and should be rejected for even if the nauarchs entered upon their office in the spring (but cf. Stylianou, 1988: 467 n. 16), the case still remains that Corcyra was attacked in the autumn of 373. That can only be changed if Isocr. Plataicus and Paus. 9. 1. 8 are dismissed, and the authority of Xenophon is not weighty enough to justify such an action. So even if Mnasippus took up his command in the spring, he did not proceed against Corcyra till the autumn, many months after Timotheus sailed. The siege of Corcyra was probably pursued over the winter months (47. 1 n.). Sealey's thesis concerning the time of the year the nauarchs assumed their command does not negate D's sequence, even if granted. For then we have only to -355assume that Mnasippus was nauarch from spring 373 to spring 372. From spring 375 to spring 374 Nicolochus was nauarch (X. H. 5. 4. 65), and from spring 374 till spring 373 either Aristocrates or Alcidas. The one left out could of course have commanded without the official title of nauarch; cf. my remarks on the Spartan nauarchy, 1988: 467f. The Peace of 372/1 was concluded on 14 Scirophorion in the Attic year previous to that of the battle of Leuctra ( Plut. Ages. 28. 7), the year of the archon Alcisthenes (50. 1; Dion. Hal. Lys. 12), i.e. in June/July 371. The battle of Leuctra was fought shortly afterwards at the beginning of the year of Phrasicleides, 371/0 (51. 1; Marm. Par., FGH239 A72; Plut. "Ages". 28.7; Cam. 19.4; Paus. 8. 27. 8). Xenophon says that the defeat was announced at Sparta on the last day of the Gymnopaidiai (H. 6. 4. 16). The date of the battle is given as 5 Hecatombaeon, the first month in the Attic calendar, by Plutarch at Ages. 28. 7, but as 5 Hippodromius, the eighth month in the Boeotian calendar, at Cam. 19. 4. Pédech, 1972: 1 ff., seems correct therefore in reasoning that the date according to the Attic calendar is nothing but a careless conversion by Plutarch himself of what presumably was the traditional Boeotian date for the battle (5 Hippodromius was apparently a lucky day for Boeotians for two great victories were won by them on that day: Cam. 19. 4). Greek calendars were in notorious disagreement with each other ( Bickerman, 1968: 32f.) and Plutarch himself was aware of this ( Aristid. 19). Pédech is therefore justified in taking scholars to task for reckoning exclusively according to the Attic calendar (e.g. Beister detailed study, 1970: 59ff.). Pédech's own calculations, however, are unsatisfactory for he first supposes that the Gymnopaidiai at Sparta coincided with (were a part of) the Karneia, and secondly he fails to suspect that the date 5 Hippodromius may not be a 'true' (and accurate) lunar date, but a modified one. And of course it is a mere assumption that the Boeotian year began on the full and not on the new moon ( Plut. Pel. 24. 2 rather argues for the latter). Thus his too precise Julian date for the battle, 18 Aug. 371, is highly suspect. We know very little about the Gymnopaidiai and the Karneia, but it would appear that they were totally separate from each other in spite of Bekker, Anecd. Gr. i. 234; see Nilsson, 1906: 140ff. (cf. RE vii. 2. 2087ff. Gomme, HCT iv. ad 5. 82. 2αὐτὰς τὰς γυμνοπαιδίας:' the very day of the festival', is wrong. 'The very time of the festival' is -356a correct translation for the Gymnopaidiai lasted for several days as shown by Thucydides' context (5. 82. 2f.) and X. H. 6. 4. 16). Suspect though Pédech's Julian date for the battle of Leuctra is, his enquiry has unwittingly cast doubt on another aspect of the campaign -- the twenty-day period which Plutarch says intervened between the Peace and the battle ( Ages. 28. 7). It appears that Plutarch made his own calculation here by counting the days from 14 Scirophorion (the date of the Peace) to 5 Hecatombaeon (the date of the battle). While the Peace was probably enough concluded on 14 Scirophorion, the date of the battle as we have seen was 5 Hippodromius (Boeotian calendar), not 5 Hecatombaeon, and so the generally accepted twenty-day period should be rejected (and Beister's careful calculations on the basis of this interval, 1970: 13ff., reconsidered). About all this, and more, see also now Tuplin, 1987: 77 ff. The capture of Plataea belongs to the second half of the summer of 373. Paus. 9. 1. 8 dates it to the year of Asteius, 373/2, and says that it was in the third year before that of the battle of Leuctra. From Isocr. Plataicus we know that there was still peace at the time, and Mnasippus' attack on Corcyra therefore must have occurred not earlier than autumn 373 (see beginning of part 2). The fact that D places the destruction of Plataea under 374/3 is of no importance for under that year is found the entire narrative from the incident on Zacynthus down to the successes of Iphicrates in the Ionian Sea in 372. What is of significance, however, is D's (Ephorus') sequence. The destruction of Plataea and the subjection of Thespiae (46. 4-6) are inserted between the sailing of Timotheus which we know belongs to spring 373 (46. 3; 47. 2), and the attack on Corcyra (47. 1). We may as well allow D's statement that Thespiae was subjected after the capture of Plataea. Isocr. 14. 9 is no evidence that the reverse is true for Thespiae could have been coerced into joining the Boeotian Confederacy between the destruction of Plataea and the time of the speech. According to Isocrates Tanagra too was subjected at this time.

3. The date of Cleombrotus' dispatch to Phocis, and the election of Jason as tagos of Thessaly At the time of the peace negotiations of 372/1 at Sparta, Cleombrotus was in Phocis with an army. This much is clear (X. H. 6. 4. 2; cf. Plut. Ages. 28. 5). The date of Cleombrotus' dispatch -357to the north, however, is a matter which has not been satisfactorily settled. At H. 5. 4. 63 Xenophon states that

for two consecutive years, 376 and 375, the Theban territory remained free from invasion. As a result the Thebans marched Θρασέως. . . ἐπὶ τὰς περιοικίδας πόλεις καὶ πάλιν ἀνελὰμßανον. The obvious meaning of this is that the Thebans were busy capturing the towns in the proximity of Thebes (cf. Beloch, GG iii. 2. 237; Thespiae is described as one of the περιοικίδας at H. 5. 4. 46). Yet shortly afterwards Xenophon asserts that the Thebans ἐπεὶ κατεστρέψαντο τὰς ἐν τῃ+̑ Βοιωτίᾳ πὴλεις, ἐστράτευον καὶ εἰς τήν Φ ωκίδα (H. 6. 1. 1). So there can be little doubt that to Xenophon's mind at least the Boeotian cities had at this time, i.e. shortly before the Peace of 375, come under the sway of Thebes. Threatened, the Phocians appealed to Sparta and Cleombrotus was sent forth with four Lacedaemonian morai and the corresponding contingents of the allies, i.e. two-thirds of the total field forces of the Peloponnesian League. This happened just before the conclusion of the Peace of 375 according to Xenophon (H. 6. 2. 1). He does not tell us what became of Cleombrotus after the Peace, and in fact the king is not heard of again till after the conclusion of the Peace of 372/1 when we suddenly hear that the king is still in Phocis. 'Still' because certainly that is the impression one derives from Xenophon. Scholars have questioned his account on the following grounds. First, Cleombrotus must have been recalled after the Peace of 375 along with all the Spartan garrisons in Boeotian towns. Secondly, the king could not in any case have stayed in Phocis with two-thirds of the Peloponnesian levy for some four years. Thirdly, if he was still in Phocis in 373 when Plataea was destroyed and Thespiae subjected (see above), why did he not intervene? One way out of this difficulty is to assume (with e.g. Meyer, GdA v. 397f., and Tuplin, 1993: 185) that Cleombrotus was indeed withdrawn after the Peace of 375, but sent back some time before Leuctra, and Xenophon has neglected to tell us so. The objections to this view appear formidable, while Beloch's alternative suggestion is attractive. He suggests ( GG iii. 2. 236ff.) that it is not possible that Phocis was threatened before the Peace of 375 when Thespiae, Plataea, and evidently Tanagra too, not to mention Orchomenus, were held by Spartan garrisons (Isocr. 14. 13f.; Plut. Pel. 15f.). The garrisons were withdrawn as a result of the Peace of 375, and it was then that Thebes saw and seized her -358chance. Beloch is therefore unable to accept that Phocis was threatened twice, both times before a Peace, and that it was twice saved by Cleombrotus. He concludes that there was but one Spartan expedition, in 371 when Phocis was seriously threatened, and that Xenophon has made a mistake in placing these events before the Peace of 375; they belong before the Peace of 372/1. I am inclined to agree. From a close look at Xenophon it emerges that the Spartan army defeated at Leuctra had not been out for more than a year, and that there was no expedition to Phocis prior to the Peace of 375. H. 6. 4. 17 leaves little doubt that not more than months separate it from the πρóσΘεÚ and τóτε expedition referred to in the passage, certainly not more than a year in view of those men who had been left behind on ἐπ + ̔ ἀρχαι + ̑ς (annual magistracies) and who were now (a few days after Leuctra) ordered out. Of course, this still leaves open the possibility that there was a previous expedition in 375, but Xenophon's own information at H. 5. 4. 59-63 seems to exclude this. After Cleombrotus' failure to invade Boeotia in spring 376 there was discontent amongst the Spartan allies and at their meeting at Sparta the feeling was that they 'could man far more ships than were available to Athens and could starve her into submission; the same ships could be used to transport an army across to Thebes by whichever route they preferred, either through Phocis or through Creusis' (section 60. Warner's translation). And indeed a strong fleet was manned and Athens was blockaded. The battle of Naxos followed (section 61). 'Next the Spartans made preparations for transporting an army across the gulf to invade Boeotia, but the Thebans asked the Athenians to send a force round the Peloponnese. They calculated that, if this were done, it would be impossible for the Spartans ἅμα μἐν τὴν ἑαυτω + ̑ν χῴραν ϕυλἀττειν, ἅμαδἐ τἀς περὶ ἐκει + ̑να συμμαχίδας πóλεις, ἅμα δἐ στρἀττειν διαßιßἀζειν ἱκανòς ἐαυτούς(section 62). The Athenians obliged. Timotheus was sent to the Ionian Sea with sixty triremes. The result was, Xenophon emphasizes, that 'there was no enemy invasion of Theban territory either in the year when Cleombrotus was in command of the army ( 376) or in the year when Timotheus sailed round the Peloponnese' ( 375 -- section 63). It might be objected that a Spartan army nevertheless did reach Phocis though it failed to enter Boeotia. But this would be an illogical objection to raise for, if a major Spartan army could have -359reached Phocis, it could have reached Boeotia too and, what is more, it would have wanted to do so (cf. section 60). The whole point, as Xenophon says, was to get together a sufficient number of troops and transport them across the gulf, and the Spartans did not succeed in doing so in 375 because of Timotheus' threatening presence in the Ionian Sea. There was no Spartan expedition to Phocis in 375, therefore, and something is very seriously wrong with H. 6. 1. 1-6. 2. 1, as indeed something is seriously wrong with H. 6. 2. 2ff., as we have seen. (Tuplin, who considers such a view to be 'a gross misestimation of Hellenica' ( 1993: 185), does not explain how Cleombrotus could have reached Phocis in 375.) When was Jason elected tagos of Thessaly? In Xenophon, the appeal of Polydamas to Sparta is certainly chronologically linked to the expedition of Cleombrotus to Phocis. How else should σχεδòν δἐ περὶ του + ̑τον τòν χρóνον at H. 6. 1. 2 be understood? Xenophon therefore has misplaced both. But even if Polydamas' appeal is kept distinct from Cleombrotus' expedition, that does not mean that it is correctly dated by Xenophon (as Sordi, 1958: 170ff., and Cawkwell, 1981b: 44 and n. 23, think). The speech of Polydamas itself (H. 6. 1. 4ff.) is not particularly illuminating, chronologically speaking, though section 10 would appear to point to 375 rather than 371: early in 375 Jason may not yet have become the ally of Athens, but by autumn 373 the two were certainly allied (ps.-Demosth. 49. 10) so that section 10 makes no sense in a 371 context even if Jason had ceased to be an ally by then. However, it is usually forgotten (even by Cawkwell, 1981b: 44 and n. 23) that the speech is Xenophon's, not Polydamas', and composing it some twenty years after the events Xenophon was all too likely to throw together disparate elements of different dates. Nor can section 17 help (pace Sordi, 1958: 171) for the statement τὰς τε ἔΞω μóρας ὅσαι αἐτοι + 0311ς εἰ + 0311εν καὶ τὰς περὶ Λακεδαίμονα πρòς τἀς τω + 0311ν ΑΘηναίων τριήρεις καὶ τòν πρòς τοὐς ὁμóρους πóλεμον, can apply equally to 375 as to 372/1. If to 372/1 then the morai abroad would be those with Cleombrotus. If to 375,

the one or two morai on guard duty in Boeotian cities are sufficient to explain the expression τἀς τε ἔΞω μóρας. One mora was certainly on garrison duty at Orchomenus before the Peace of 375, and Polydamas' appeal might have coincided with the relief of that garrison by a fresh mora (37. 1 n.). In any case Xenophon is vague enough for even one mora abroad to justify his expression. -360Lastly, the date of Polydamas' appeal is not necessarily the date of Jason's election, and the inconsistency between D and Xenophon may be less severe. Xenophon apparent date of 375 (or shortly thereafter) for Jason tageia is generally accepted (cf. most recently Tuplin, 1993: 207 ff.). A notable exception is Beloch who prefers to bring down to 372/1 both Cleombrotus' expedition to Phocis and Jason's election (GG iii. 2. 237). Xenophon's date is undermined by an examination of his own text. First, the synchronisms at H. 6. 1. 1 f. are very questionable as we have seen (part 1). Second, H. 6. 1. 2-19 is a digression involving probably a considerable jump forward in time. ἐγὼ δἐ πἀλιν ἐπἀνειμι, ὅΘεν εἰς τἀς περὶ Ιἀσονος πρἀΞεις ἐΞἐßην, he concludes at H. 6. 1. 19. The next time he takes up Thessalian affairs he makes a considerable jump forward from 371 to 358 (H. 6. 4. 27ff.), and he concludes at H. 6. 5. 1 with almost the same words as at H. 6. 1. 19. It may be that the leap forward at H. 6. 1. 2-19 is represented only by sections 18-19, or it may be that the whole belongs later and it has been misplaced by Xenophon. If we look at H. 6. 4. 20-37 we see that sections 20-6, where others than Thessalians are involved (it is a Hellenica narrative as it were), are part of the normal narrative, whereas sections 27-37, being a Thessalica (see H. 6. 5. 1-Θετταλικἐ), leave the rest of the narrative behind and go forward in time till we reach the year 358. Hence the remark at H. 6. 5. 1 to warn the reader that he is returning to the proper narrative-- the immediate aftermath of Leuctra. Similarly, it is perfectly possible to argue that H. 6. 1. 2-18 belongs to 375 while sections 18-19 stretch over the years 375-371. Thus, even if we accept that Polydamas' appeal has not been misplaced by Xenophon, the recognition of Jason as tagos does not have to pre-date the Peace of 375 and it could belong as late as 371. This being the case it becomes legitimate to ask, why does Jason's election to the tageia have to be before Leuctra? ( Tuplin various suggestions, 1993: 209, do not amount to anything very much.) We have no other evidence but that of Xenophon and D, and with Xenophon's largely discredited D's comes into its own. The fact that D sets Jason's recognition under 370/69 is of no importance (cf. Beloch, GG iii. 2. 238). But what he has to say about it is, and it cannot be lightly dismissed. At 57. 2 Jason is seen increasing his power. The time I take to be the immediate aftermath of Leuctra, i.e. 371. (X. H. 6. 27, where the capture of -361Heracleia in Trachis is also mentioned, provides additional confirmation.) Then 57. 3-59. 4 deals with Peloponnesian affairs concluding with Agesilaus' campaign in Arcadia in winter 370. The Thessalian narrative is then resumed (60. 1). Jason is now offering to the Thessalians the hegemony of Greece. (It is possible that 60. 12 echoes an actual propaganda manifesto of Jason; cf. Westlake, 1935: 96.) He stresses that 'the Lacedaemonians had sustained a great disaster at Leuctra; that the Athenians laid claim to the mastery of the sea only; that the Thebans were unworthy of first rank; and that the Argives had been brought low by civil wars and internecine slaughter' (Sherman's translation). It is important to observe that the context is a post- and not a pre-Leuctra one -Sparta has been toppled and the agon for a new hegemon is on (the slaughter at Argos is the skytalismos related at 57. 3-58). διóπερ οἱ Θετταλοί, D continues, προσησἀμενοι τΩ + ̑ν ὅλων ἡγεμóνα Ιἀσονα, τούτῳ τἀ κατἀ τòν πóλεμον ἐπἐτρεΨαν. For hegemon we should read tagos, the latter term in fact not being used by any other author except Xenophon. Pace Tuplin, 1993: 208f., there can be no doubt whatsoever about this. τω + ̑ν ὅΘετταλοί in the Thessalian context (cf. 60. 5, τούτῳ τἀ κτω + ̑ν ὅ) points indubitably to the tageia of Thessaly. Hegemon in any case was the correct nonThessalian term to use for what the office of tagos amounted to, for traditionally tagoi were appointed by common consent to pursue some national enterprise, or to deal with some danger ( Robertson, 1976: 104ff.), and there could be no greater enterprise than the hegemony of Greece. This is highlighted by D's words, Θετταλοίτω + ̑ν ὅ Θετταλοίτούτῳ τἀ κ, which should be seen in connection with X. H. 6. 1. 19-- the military preparations of Jason immediately after his election. There is no reason, therefore, to reject the evidence of D. On the contrary it makes good sense whether on its own or even in combination with the details supplied by Xenophon. On his return from Leuctra the powerful ruler of Pherae was indeed τω + ̑ν ὅτούτῳ τἀ κ (X. H. 6. 4. 28; cf. Westlake, 1935: 95ff., who, however, preferring the usual date for the tageia, interprets D 60. 1-2 differently), and he became even greater when he appealed to the patriotism of his fellow Thessalians and was elected tagos. It was in the immediate afterglow of this success that he set in motion the great preparations for holding the Pythian festival of 370. The following dates of the period 375-371 may therefore be -362considered as certain or at least as reasonable and in their proper order:

Mid-/late summer 375

Peace of 375/4.

Timotheus lands exiles on Zacynthus.

Zacynthian oligarchs appeal to Sparta.

Spartan embassy to Athens.

Autumn/spring/summer 375/374

Aristocrates to Zacynthus.

Failed oligarchic coup at Corcyra.

Summer/winter 374

Appeal to Sparta of Corcyraean oligarchs now in exile.

Autumn 374/spring 373

Alcidas sent out.

Autumn/winter 374/373

Appeal to Athens of Corcyraean demos.

Ctesicles sent to Zacynthus.

Naval preparations to aid Corcyra.

April/May 373

Timotheus sails out.

Late summer/early autumn 373

Plataea destroyed.

Autumn 373

Minasippus attacks Corcyra.

Timotheus relieved of his command.

Autumn/winter 373/372

Athenian fleet prepared for Corcyra.

Autumn 373

Ctesicles sent to Corcyra.

Spring 372

Ctesicles and the Corcyraeans defeat Mnasippus.

Iphicrates in the Ionian Sea.

Autumn/spring 372/371

Cleombrotus in Phocis.

June/JulY 371

Peace of 372/1.

July/August 371

Battle of Leuctra.

Jason tagos of Thessaly.

371/370 (after Leuctra)

45. 4. Θετταλοί. . . τούτῳ τἀ κ: X. H. 6. 2. 3 mentions no such embassy, but says that the Spartans sent out Mnasippus against Corcyra as soon as the Zacynthians 'from the city' complained to them. As was seen, the whole series of events which led to the fresh outbreak of war is missing from Xenophon (Excursus at 45. 3, part 2). The result is that the reader is left puzzled. The troublespot, Xenophon assures us, was Zacynthus. But Zacynthus did not warrant the great armament of Mnasippus that we are suddenly faced with in his account, and Xenophon in fact is clear that the Spartan objective was Corcyra. Θετταλοίτούτῳ τἀ κ -363Mτω + ̑ν ὅτούτῳ τἀ κ' τούτῳ τἀ κ Θετταλοί (Zacynthus perhaps) τούτῳ τἀ κ (H. 6. 2. 4). Far better sense is made of the undertaking if D's information is taken into consideration. Cawkwell, 1963: 88, rightly cautions against Xenophon here, though to deny that Zacynthus was a casus belli is to oversimplify things: Zacynthus was one of two (or more; see 40. 4n.) incidents which led again to war. Θετταλοίτω + ̑ν ὅ: Timotheus' landing of the exiles on Zacynthus has always baffled scholars. The Athenian general had had severe financial difficulties and at one point it seems his men had come close to mutiny (X. H. 5. 4. 66; ps. -Arist. Econ. 1350a30). One would have thought that he would have been glad to receive news of the Peace and his orders to return home. So what was he doing acting as irresponsibly as a Sphodrias and providing Athens and Sparta with a new source of friction? This is indeed a puzzle and Beloch cannot be blamed for considering Timotheus guilty of the resumption of hostilities ( GG iii. 1. 156 n. 3). Elsewhere he even calls Timotheus' action a breach of the Peace ( GG iii. 2. 236). Sealey ( 1957: 106, 111) argues much the same. But the matter is less simple. It is just possible that there was a clause about exiles in the Peace (5. 1n.). If that be the case then Timotheus far from breaking the Peace was actually implementing it by assisting the exiles to return. The fact that Sparta opposed their return does not necessarily argue against such a clause, for the democratic exiles may have been branded as τω + ̑ν ὅτούτῳ τἀ κ, Θετταλοί, and/or Θετταλοί by their enemies, and as such they would have been outside the Peace. The Athenians on the other hand may have considered them as simple political exiles and thus supported them. But, an exiles clause or not, the Athenians would have based their support on other grounds too. They would have helped the exiles in the first instance because these democrats, a substantial section of the Zacynthian population, were their allies and in fact belonged to their Confederacy (45. 3 n.). A Common Peace had just been concluded which promised freedom and autonomy to everyone. Why should not the exiles (who had perhaps sworn to the Peace too; cf. X. H. 6. 3. 19) be allowed to return? Timotheus would have felt a personal obligation to help too. The democrats had not fought with him for the glory of it, but doubtless because they counted on him to aid their return. He and Athens may even have promised -364them that. Also, the exiles may have based a demand for help on the fact of their membership of the Athenian Confederacy (cf. Svt 257=Tod 123 lines 46ff.; Svt 263 =Tod 127). All of this taken together with the new, elevated position of Athens vis-ἀ-vis Sparta and the Peace, may have encouraged the Athenians to ignore the Spartan protests and to stand by Timotheus' action. Athens' coprostasia of the Peace appears to have been broadly divided on the basis of land and sea hegemony, the Spartans predominating in the former, the Athenians in the latter, and Athens therefore may have felt that the enforcement of the Peace in the Ionian Sea was her concern and not Sparta's (for the co-prostasia of the Peace see 38. 4 and n.). Θετταλοίτω + ̑ν ὅ: D often uses this word in the loose sense of 'help' though frequently help within the framework of an alliance; cf. 12. 81. 5; 13. 34. 4; 13. 38. 6; 13. 61. 1; 15. 47. 2. The Spartans may have felt that they had both a moral and a legal obligation to act in defence of the 'autonomy' of one of their allies for Sparta was still a prostates of the Peace albeit jointly with Athens now. We do not hear whether they based their action on the Peace; probably they did. 46. 1. τω + ̑ν ὅΘετταλοί: the oligarchic revolt, not mentioned by any other source, failed before the appeal to Sparta of the oligarchs for the Spartans hoped to capture the city Θετταλοί (46. 2; 47. 1; chronological excursus at 45. 3, part 2). Hence Beloch, GG iii. 1. 157, is incorrect. Corcyra appears to have joined the Athenian Confederacy against the opposition of the people who now attempted a coup (36. 5n.). I see no reason for deducing from X. H. 5. 4. 64 (as Tuplin, 1984a: 562, seems inclined to do) that democracy at Corcyra post-dated the admission of the island to the Athenian League. For Xenophon's comment see 36. 5n. Θετταλοί: cf. Thuc. 1. 36. 1f. with HCT i ad 1. 36. 2; X. H. 6. 2. 9. 46. 2. Θετταλοίτούτῳ τἀ κ: perhaps Alcidaswas sailing to Sicily (15. 1n.) and the Spartans thought of killing two birds with one stone. Or possibly Alcidas was only sailing to Sicily and the Corcyraeans misunderstood the Spartan intentions. If he did sail to Sicily, he did not remain there indefinitely for his twenty-two ships probably formed part of Mnasippus' fleet. See also 47. 1n. 46. 3. Θετταλοί: Athens had pledged herself to render -365the utmost possible assistance to the Corcyraean demos ( Svt 263=Tod 127). Her obligation would have been

heightened by her role as co-prostates of the Peace. τω + ̑ν ὅτούτῳ τἀ κ: D is certainly wrong at 51. 4 and 62. 1 and perhaps at 47. 4 where he uses Θετταλοί. But he is not wrong here; see the chronological excursus at 45. 3, part 2. Θετταλοί: τούτῳ τἀ κ according to X. H. 6. 2. 10 though one Hellenica MS reads Nἀ. Ctesicles is a much more common name in Athens than Stesicles, and he could thus be identical with the general of Lysias 9. 6 (see PA s.v.; Develin, 243f.).

46. 4-6. The destruction of Plataea 46. 4. Θετταλοίτούτῳ τἀ κ . . . Θετταλοί: whether Plataea (re-established by Sparta after the King's Peace: 5. 2-3 n.) petitioned Athens for an alliance requesting military protection (which she sorely needed after the departure of the Spartan garrison) and further promising to hand herself over to Athens, as D says, is not known from any other source though it is quite credible. But would Athens in her current need of Thebes have been willing to take such a course of action? It is possible, and the Thebans at least feared so, if D is to be believed. We hear at any rate that shortly before this date Athens had stood up to Thebes in the case of Oropus (38. 3n.). Isocr. 14. 26 makes the Plataeans say that they were destroyed twice because of their friendship for Athens, and X. H. 6. 3. 1 too speaks of friendship between Athens and Plataea at this time. It is possible of course that D misunderstood an Ephoran comment on Plataea's long record of loyalty to Athens; but see 46. 6n. ad Θετταλοί. The appeal for help must, to some extent, have been based on the Peace (cf. 5. 1n.). For Plataean suspicions of Thebes which were soon realized see Paus. 9. 1. 5. Θετταλοίτούτῳ τἀ κ: the Boeotian Confederacy was now, before the outbreak of war, (still?) in existence; see. 38. 3n. 46. 5. ΘετταλοίΘετταλοί: Paus. 9. 1. 6f. supplies a little more information about what happened. The attack was sudden like the more famous one which introduced the Peloponnesian War, and it caught the Plataeans off their guard. Θετταλοί: the Thebans were less generous according to Paus. 9. 1. 7 who says that the Plataeans were allowed to depart, the men with one garment each, the women with two. -366-

46. 6. Θετταλοίτω + ̑ν ὅ: cf. Isocr. 14. 1, 7, etc.; X. H. 6. 3. 1, 5; Paus. 9. 1. 8. τούτῳ τἀ κτω + ̑ν ὅ: 'they pillaged'. Thespiae, another Boeotian city invariably hostile to Thebes, was forced at this time to join the Boeotian Confederacy (chronological excursus at 45. 3, part 2). In 423 its walls had been pulled down by Thebes (Thuc. 4. 133. 1) and this is what may have happened now too; Isocr. 14. 19, 35, τούτῳ τἀ κ Θετταλοί/τούτῳ τἀ κ, possibly refers to Thespiae. These last walls had been rebuilt or strengthened by Agesilaus in summer 378 so that the city could serve as a base against Thebes (27. 4n.). X. H. 5. 4. 41 writes Θετταλοί, but Thespiae must have possessed some makeshift fortification already early in 378 when Sphodrias was left there by Cleombrotus. But the city wall had probably not been properly rebuilt since 423. Paus. 9. 13. 8 says that the Thespians were present at Leuctra on the Theban side albeit unwillingly; they were allowed to depart before the battle (cf. Polyaen. 2. 3. 3). After the battle, fearing reprisals, the Thespians occupied a stronghold near their city from which they were ejected by Epaminondas (Paus. 9. 14. 1-4). This is perhaps an indication that their city had indeed had its walls dismantled in 373. (It was dioecized, Tuplin suggests: Athenaeum, 64 ( 1986), 321 ff.) And it was probably at this time, after Leuctra, that Thespiae was razed to the ground and its people driven away (Demosth. 16. 4). Some at least were received at Athens ( Aristid. Panath. 59 (180 Dind.)); cf. X. H. 6. 3. 1, τω + ̑ν ὅ τούτῳ τἀ κ, but anachronistically, for Xenophon cannot be right when he speaks of the destruction of Thespiae before Leuctra (Θετταλοί may indeed be interpreted: 'leur ville n'est plus une cité, mais une simple τω + ̑ν ὅ du territoire thἐbain; ils n'ont donc plus aucun droit civique ni par conséquent aucun róle politique' -- Roesch, 1965: 45. But τω + ̑ν ὅ at H. 6. 3. 5 and the association of Thespiae and Plataea leave little doubt that Xenophon erroneously considered Thespiae to have been destroyed before Leuctra). Θετταλοίτούτῳ τἀ κ: cf. Aristid. Panath. 59 τω + ̑ν ὅ. Θετταλοί: known chiefly from the Hellenistic period, this institution had deeper roots. The term itself was in use in both literature and inscriptions by the late 4th century at the latest. It occurs in Arist. F 575 Rose (from Photius). It is unnecessary to regard this as an anachronism on the part of the excerptor, for -367-

similar terms such as τω + ̑ν ὅ were used in inscriptions already early in the 4th century ( Gawantka, 1975: 166 n. 7), while isopoliteia itself is found in two Milesian decrees of the time of Alexander ( Gawantka, 1975: nos. 19, 21). An early 4th-century inscription from Cyrene speaks of Θετταλοί (ML 5 line 12). Strabo 9 C 404 mentions isopoliteia in a quotation from Philochorus ( FGH 328 F113), a younger contemporary of Aristotle. The passage is confused it is true ( Gawantka, 1975: 93 n. 7), but there is no reason to deny that the term derives from Philochorus. D's use of the word here may be the earliest known instance of it in literature if its Ephoran origin is conceded (as it is not by Kahrstedt, 1934: 78 n. 5, 86 n. 2, and Gawantka, 1975: 178 n. 31a. Kahrstedt suggests that D incorrectly wrote isopoliteia instead of isoteleia). The institution of isopoliteia anticipated the term. For its history in the 6th and 5th centuries see Gawantka, 1975: 165 ff. A possible case of isopoliteia early in the 4th century (not dealt with by Gawantka) is the 'union' of Argos and Corinth ( Griffith, 1950: 247f.). Of relevance here is the isopoliteia granted to the Plataeans by Athens in the 5th century. According to Thuc. 3. 55. 3 the Plataeans defending themselves before the Spartan judges in the

summer of 427 claimed to be allies as well as citizens of Athens. In their speech the Thebans repeated the Plataean claim (Thuc. 3. 63. 2). Thus, while Plataea still stood, its people had been admitted, potentially at least, to the Athenian citizenship. The case of isopoliteia seems plain (see Gomme, HCT ii. 339 f. Contra Gawantka, 1975: 174ff., who rejects Thucydides as incompatible with ps.Demosth. 59. 104-6 and Isocr. 12. 92-4. But he agrees that Thucydides was aware of the institution of isopoliteia though not the term). At the destruction of Plataea the next step was taken of admitting the Plataeans in fact to the citizenship. The psephisma quoted at ps.-Demosth. 59. 104 therefore is the further enactment which in 427 fulfilled the original offer, and provisional acceptance, of potential citizenship referred to by Thucydides. We may compare the similar case of the Samians (ML 94 lines 12ff., 32ff.; Tod 97). Yet the Plataeans appear to have remained a distinct body in Athens, no doubt in the hope that they would eventually be restored to their country ( Gomme, HCT ii. 339f.). When in 420 they were settled in the territory of Scione (Thuc. 5. 32. 1) they -368probably reverted to their previous (pre-427) condition of σύμμαχοι κλὶ ἰσοπολι+̑ται. (For the frequent combination of σνμμαχία or ϕιλα and ἰσοπολιτεία in Hellenistic times see Larsen, 1966b: 53; Gawantka, 1975: 146.) This, one may venture to suggest, is what happened after the King's Peace too, when Plataea was reconstituted by Sparta, though the symmachia may have lapsed then and been replaced by simple friendship (Isocr. 14. 26;X. H. 6. 3. 1). And as Athens did not perhaps approve of the Spartan refoundation of Plataea, those Plataeans who returned may no longer have been isopolitai; but they were once more granted this privilege when they again fled to Athens following the second destruction of their city. So D may not be so wide of the mark (and anachronistic) when he says (46. 4) that in 373 the Plataeans were ἀντεχόμενοι τὴ+̑ς Αθὴναίων σνμμαχίας and that they were granted isopoliteia when they fled to Athens. διὰ τὴν χρὴστότὴτα του+̑ δήμου: for Ephorus' attitude to Athens see pp. 115ff.

47. The outbreak of war. The siege of Corcyra 47.1. στρατὴγόν : in fact a nauarch: see 34. 4n. ἑξὴκοντα καὶ πέντε: 60 according to X. H. 6. 2. 3. χιλίονς καὶ πεντακοσὶουσ: X. H. 6. 2. 5 agrees though he says that they were mercenaries and that they were in addition to Mnasippus' Lacedaemonian troops; cf. X. H. 6. 2. 18. The mercenaries were hoplites as Anderson, 1970: 307 n. 67, correctly observes. τοὺς ϕυγάδας: see 46. 1n. If Sparta based her action on the Peace, she did so with far less justification than in the case of Zacynthus. She was supporting exiles (guilty of an attempted coup) against the recognized government of Corcyra which had taken the oaths to the Peace. εἰσέπλευσεν εἰς τὸν λιμένα κτλ: for the topography of Corcyra see Dontas, 1965: 139ff. (map facing 144), and id., Praktika ( 1965), 66ff., and (1966), 85ff., for the archaeological reports on the excavations of 1965 and 1966. Also Kalligas, 1980: 81ff. The ancient city occupied the peninsula of Palaiopolis (now Analepsis) as far south as Cape Canoni ( Gomme topography, HCT ii. 370ff., which is based on that of Schmidt, 1890, must be rejected. -369probably reverted to their previous (pre-427) condition of σύμμαχοι κλὶ ἰσοπολι+̑ται. (For the frequent combination of σνμμαχία or ϕιλα and ἰσοπολιτεία in Hellenistic times see Larsen, 1966b: 53; Gawantka, 1975: 146.) This, one may venture to suggest, is what happened after the King's Peace too, when Plataea was reconstituted by Sparta, though the symmachia may have lapsed then and been replaced by simple friendship (Isocr. 14. 26;X. H. 6. 3. 1). And as Athens did not perhaps approve of the Spartan refoundation of Plataea, those Plataeans who returned may no longer have been isopolitai; but they were once more granted this privilege when they again fled to Athens following the second destruction of their city. So D may not be so wide of the mark (and anachronistic) when he says (46. 4) that in 373 the Plataeans were ἀντεχόμενοι τὴ+̑ς Αθὴναίων σνμμαχίας and that they were granted isopoliteia when they fled to Athens. διὰ τὴν χρὴστότὴτα του+̑ δήμου: for Ephorus' attitude to Athens see pp. 115ff.

47. The outbreak of war. The siege of Corcyra 47.1. στρατὴγόν : in fact a nauarch: see 34. 4n. ἑξὴκοντα καὶ πέντε: 60 according to X. H. 6. 2. 3.

χιλίονς καὶ πεντακοσὶουσ: X. H. 6. 2. 5 agrees though he says that they were mercenaries and that they were in addition to Mnasippus' Lacedaemonian troops; cf. X. H. 6. 2. 18. The mercenaries were hoplites as Anderson, 1970: 307 n. 67, correctly observes. τοὺς ϕυγάδας: see 46. 1n. If Sparta based her action on the Peace, she did so with far less justification than in the case of Zacynthus. She was supporting exiles (guilty of an attempted coup) against the recognized government of Corcyra which had taken the oaths to the Peace. εἰσέπλευσεν εἰς τὸν λιμένα κτλ: for the topography of Corcyra see Dontas, 1965: 139ff. (map facing 144), and id., Praktika ( 1965), 66ff., and (1966), 85ff., for the archaeological reports on the excavations of 1965 and 1966. Also Kalligas, 1980: 81ff. The ancient city occupied the peninsula of Palaiopolis (now Analepsis) as far south as Cape Canoni ( Gomme topography, HCT ii. 370ff., which is based on that of Schmidt, 1890, must be rejected. -369victory at Alyzeia, having had refitted his own ships and manned others from Corcyra, he had no more than just over 70 (X. H. 5. 4. 66). In 372 Iphicrates brought 70 triremes west. He captured 10 Sicilian, and, the Corcyraeans having doubtless made their contribution, his fleet numbered about go (X. H. 6. 2. 14, 33, 35, 38). It would appear that Corcyra, like the rest of Greece, was not what it had been in the 5th century. If the normal Athenian fleet of the 4th century was less than half the strength of the normal 5th-century fleet, then the Corcyraean navy cannot have been in a healthier state and the above figures tell us that it was not. Whereas at the beginning of the Peloponnesian War Corcyra had 110 or 120 triremes (Thuc. 1. 29. 4; 47. 1), it is doubtful if it had more than 20 or 30 at this time. No wonder therefore the Corcyraeans did not venture out to face Mnasippus; they were not up to it. Xenophon says as much: κατὰ θάλατταν δὲ εἰσήγετο αὐτοι+̑ς διὰ τὸ ναυκρατει+̑σθαι (H. 6. 2. 8). As for his statement that ἐΞ Οὐδεμια+̑ς γὰρ πόλεως πλήν γε Αθὴνω+̑ν οὔτε χρήματα Πλείω ἂν γενέσθαι (H. 6. 2. 9), it must be remembered that these are the Corcyraean envoys speaking in Athens, and in any case the statement may be correct, but in the context of the 4th century. Isocr. 15. 109, Kέρκυραν . . . πόιλν ὀγδοήκοντα τριήριες κεκτὴμένὴν, is hardly credible. Isocrates was either thinking in 5th-century terms, or he was reflecting Corcyraean boasting at Athens. Also, it served the eulogy of Timotheus to exaggerate. 47. 2. πάλαι . . . ἀπεστάλκεισαν: the preparations mentioned at 46. 3 completed, Timotheus sailed out in the spring of 373. We are now in the autumn of the same (Julian) year and the sentence looks back to Timotheus' spring sailing; see chronological excursus at 45. 3, part 2. ἑΞήκοντα: X. H. 6. 2. 11 gives the same figure, as does ps.Demosth. 49. 11. Allied ships were included amongst the 60: ibid. , sections 14ff. πρὸ τὴ+̑ς σνμμαχίας ταύτὴς: πρὸ denotes time here, 'before'. The urgency or non-urgency of the Corcyraean situation apart, Timotheus had other reasons for remaining in the Aegean for the time being (ps.-Demosth. 49. 6ff.). I do not see that D's phraseology supports Cargill thesis ( 1981: 72) about the nature of the relationship between Athens and Corcyra. 'Before sailing out in aid of the Corcyraean allies of Athens' is the meaning. 47. 2-4, 7. πλεύσας ἀπὶ Θρκὴς κτλ: though errors and confusions -371are not rare in D, this is one of two grotesque muddles in Book 15. The other is the campaign and battle of Leuctra. Ephorus is certainly not to blame for either (on Leuctra see 51. 3 and 54. 5-56nn.). The simplest and probably correct explanation for both muddles is that D, after a hasty and careless reading of the lengthy and complicated narratives of Ephorus which included digressions (see p. 123 ), put aside his source and wrote his own brief account; ludicrous confusion ensued. Ephorus treated Timotheus favourably (p. 119). We would expect him to have stressed at some length the man's services to Athens (as Isocr. 15. 107ff. does), and to have said as little as possible, and that in an apologetic vein, about his removal from office and trial. D mentions the first, but not the second. Isocrates also does not mention the 373 trial though he does, briefly, that of 357 ( ibid. 129 ). Ephorus may have digressed to relate that Timotheus was again, some years later, re-elected general, and this might explain D's statement that μετενόὴσεν ὁ δὴ+̑μος καὶ πάλιν αὐτω+̑ τὴρατὴγία ἀποκατέοτὴσεν (47. 3). That Timotheus was not restored to his command, but was tried, acquitted (though disgraced), and then sailed to serve the King, is clear from ps.Demosth. 49; cf. X. H. 6. 2. 13. What then of the sailing to Thrace and the new allies and ships? It was seen above (30. 2n.) that eulogists of Timotheus tended to speak collectively, and not necessarily truthfully, of his services to Athens. I am inclined to see the present information as a confused rendering of an Ephoran passage on T imotheus' successes in Thrace in the later 360s. (Others explain differently; cf. Woodhead, 1962: 258ff. Dušanič, 1979: 333f., is too fanciful.) The absurd statement in 47. 7, that Timotheus and Iphicrates arrived in the Ionian Sea together, is more satisfactorily explained. Aristid. Panath. 313 (291 Dind.) asserts that the crowning achievement of Athens in the 4th century was the stop she put to Dionysius' attempt to attack Greece with the aid of the barbarians, specifically Persia (cf. Lysias 33. 5ff.). Two Athenian generals foiled this plan, the one by capturing Dionysius' ships which had sailed to Greece, the other by defeating his allies, the Spartans, at Leucas, and assuming command of the sea. The schollast on this passage supplies Aristides' source -- Ephorus (iii. 294 Dind. = FGH 70 F211). The scholiast is of course mistaken in attributing the plan to Dionysius II. And Dionysius did not move against -372Greece with all his ships. Nor were the battle of Leucas and the capture of Dionysius' ships contemporary. But the errors of the scholiast provide a key to D's confusion: Ephorus would seem to have remarked that Iphicrates'

success in capturing the Sicilian ships complemented Timotheus' victory at Leucas three years earlier. Both generals put an end to the 'plans' of Dionysius to subject Greece with the help of Sparta (cf. 23. 5n.). D, somewhat like the scholiast, misunderstood this, assuming that Timotheus and Iphicrates were in command together. That Ephorus lies behind all three, Aristides, the scholiast, and D, is shown by a comparison of D 47. 7, τριήρασι δὲ Σικελικαι+̑ς . . . αὐτάνδρονς ελον, οὔσας λαßὠν , and Aristid. Panath. 313, τὰς ἀΠὸ Σικελίας ει+̑λον . . . ἀΠάσσας λαßὠν αὐτοι+̑ς ἀνδράσιν. X. H. 6. 2. 35 gives different details (see 47. 7n.). The scholiast makes nonsense of ἀπάσας: συνΓγαγὠν (Διονύσιος) ἁπάσας μικρου+̑ τὰς ἐν Σικελί ναυ+̑ς ἐχὡρει . . . ει+̑λον μὲν Πα+̑ν τὸ Διονμσίον ναυτικόν. Section 16. 57. 2 is proof (if such were needed) that Epharus is not responsible for the confusion in 47. 7 (though the Sacred War was recounted by Ephorus' son Demophilus: 16. 14. 3). There D correctly mentions Iphicrates alone in command in the Ionian Sea. 47.4. τετταράκοντα . . . τριάκοντα: in view of the previous note these numbers should be treated with care. X. H. 6. 2. 14 says that after energetic preparations Iphicrates set out with 70 triremes. He had to scour Attica to get together this many. One is tempted to see the 30 triremes which, according to D, Timotheus acquired from new allies in addition to his original 60 as what, in fact, remained of the original fleet by autumn 373. As for the 40 which supposedly resulted in a grand fleet of 130 triremes, these may be the ones assembled by Iphicrates. Xenophon's number is thus obtained. ἐποιου+̑ντΟ . . . παρασκευάς: a stereotype Diodoran description of military preparations (pp. 15f.). Bέλὴ is usually a generic term in D meaning 'missile weapons' whether thrown by hand or shot with bows or from catapults; cf. 12. 61. 6; 13. 55. 4; 17. 11. 3; 17. 44. 4; 20. 12. 3; 20. 47. 7; 20. 73. 2f. Often though belos in D signifies a catapult or, more generally, an engine of war; cf. 14. 42. 2; 14. 43. 3; 14. 50. 4 (cf. Philo, Bel. 82. 8; 97. 10). Similarly Polybius (cf. 5. 4. 6; 8. 5. 2) though, like D, by belos he can mean simply a missile weapon. στρατὴγὸν ἑλὸμενοιsee p. 133. -373Πεντακοσίων: 600 according to X. H. 6. 2. 10 who says they were peltasts. ἐΞπενΨαν: this was Ctesicles' second mission in the west; see chronological excursus at 45. 3, part 2; 46. 3 n. 47. 5. λάθρα . . . εἰς Κόρκυραν: cf. X. H. 6. 2. 10f. who says that Alcetas was asked to help get the men across to the island. As it was now late autumn or winter, the peltasts probably made their way to Epirus overland by way of Boeotia and Thessaly. At any rate it is difficult to see why Alcetas' help was needed if the troops sailed round the Peloponnese in their own ships. 47. 5-6. καὶ καταΛαßὠν τοὺς ἐν τέ+̑ πόκει κτλ: cf. X. H. 6. 2. 15ff. The different standpoints of D (Ephorus) and Xenophon should again be noted. Xenophon ascribes the Spartan failure (he is concerned to explain it) to a combination of incompetence and even corruption on the part of Mnasippus, and luck on the part of the Corcyraeans. D on the other hand represents that the able leadership of the Athenian Ctesicles was the decisive factor in the Corcyraean victory (see Introd. n. 306). Why did Mnasippus fail? Xenophon's aspersions on Mnasippus (generally accepted; e.g. most recently by Tuplin, 1993: 132) are suspect. The charge of incompetence cannot really be answered. But the implied charge of corruption can. By spring 372 and shortly before he was killed, Mnasippus had discharged some of his Mnasippus' expenditure indicates that his mercenaries and he owed the rest as much as two months' pay. finances cannot have been in a very healthy state. A trireme required at least two-thirds of a talent per month for its upkeep (cf. Wilson, 1970: 304f.). A fleet of 60 triremes therefore would have needed 40 talents a month. If the siege continued for five to six months, from autumn 373 till spring 372, Mnasippus would have had to spend no less than some 300 talents on his fleet alone-a very considerable sum and surely beyond the means of a Peloponnesian League unsupported by Persia. And then there were the mercenaries. A mercenary in the 4th century required about 7 or 8 obols a day for his pay and food (μισθός and σιτὴρέσιον; cf. Perlman, 1976-7: 258f.). Fifteen hundred mercenaries would have required some 10 talents a month. -374It would seem therefore that, as the siege dragged on over the winter months, Mnasippus ran out of money. In similar circumstances in 375 Timotheus had to lie to his men to avoid a mutiny (ps.-Arist. Econ. 1350a30). He had been saved by being recalled home after a relatively short stay in the west whereas Mnasippus had to persevere till the city was taken. It is reasonable therefore to suppose that the indiscipline of the mercenaries was due to a genuine shortage of money. It may be relevant here to point to a statement of Aelian (NA 11. 19) that when the tidal wave struck Helice in the winter of 373/2 (48. 1 n.) there were ten Lacedaemonian ships moored off the coast. Diog. Laert. 3. 20 (quoting Favorinus, Memorabilia) adds that the Spartan Pollis was in command and that he was drowned. What was he doing there? It cannot be proved, but it is likely that he was involved in ἀργυρολογία -- either levying additional money for the needy Mnasippus or even collecting the original contribution which some cities may have been slow in dispatching ( Anderson explanation, 1954: 88f., seems farfetched). 47. 6. τὸμὲν μὲν τρω+̑τον κτλ: the sudden charge of the besieged took Mnasippus by surprise: X. H. 6. 2. 17. Xenophon mentions captured and killed on the Peloponnesian side, but gives no figures. ὕστερον δὲ κτλ: cf. X. H. 6. 2. 18ff. Mnasippus chased the Corcyraeans back to the city, but near the wall they rallied. In the ensuing fight Mnasippus was killed. The graves from which the Corcyraeans fought back were between the hill (Soteros; 47. 1 n.) and the city wall. From these perhaps come the 4th-century fragments

discovered in the last war near the south-west end of the wall and now in the museum ( Dontas, 1965: 144). 47. 7. Tιμόθεον καὶ Λϕικ⇁άτὴν: see 47. 2-4, 7 n. Kισσίδὴς καὶ Κρίνιππος: X. H. 6. 2. 36 gives just Crinippus (likewise Polyaen. 3. 9. 55) though he later mentions Cissides as the commander of the second force sent by Dionysius to aid Sparta (H. 7. 1. 28). According to Xenophon Crinippus committed suicide unable to bear his misfortune. αὐτάνδρονς . . . ἐννέα: nine was the number of triremes captured, but X. H. 6. 2. 35 says that a tenth escaped. Polyaen. 3. 9. 55 gives the number captured as ten out of an original force of eleven (cf. 47. 2-4, 7 n.). λαθυροπωλήσαντες: X. H. 6. 2. 35 says rather that the captured -375Syracusans were ransomed. But a sale of booty did take place. On the evidence of 16. 57. 2 f. there were gold and ivory statues aboard the captured ships meant for Olympia and Delphi. These Iphicrates sold on orders from home and used the money to pay his troops. Schaefer, 1885-7: i. 66 n. 2, may be right that Dionysius' dedications were intended for the Olympic festival of summer 372. πλείω . . . ταλάντων: if we reckon 200 crew per trireme then each man fetched about 2 minae which is not unreasonable; cf. X. Mem. 2. 5. 2; Finley, 1960: 5.

47. 8. Chronographic notices NικοΛλὴ+̑ς ὁ εὐνου+̑χος κτλ: Theopompus (FGH 115 F103. 12) gives an outline of the events which led to the murder of Evagoras and his son Pnytagoras. Grote, x. 33 n. 1, prefers Theopompus to the versions of D and Arist. Pol. 5. 1311b. He is followed by Judeich, 1892: 131f., 132 n. 1. Hill, 1940: 140f., is not so sure. See also Beloch, GG iii. 2. 100 (99 ff. for the genealogy of the dynasty). The confusion in D (the chronographer's perhaps) may be due to the closeness of the names Nicocreon, the original conspirator, and Nicocles, one of Evagoras' sons who succeeded him to the throne. κατὰ δὲ τὴν Λταλίαν κτν: for the war see Livy 6. 27. 6-29; cf. Perl, 1957: 127; Cornell, CAH2 vii. 2. 318.

Archon Astelus, 373/2 48. 1. 'PΠωμαι+̑οι κτλ: 381 V; CAH2 vii. 2. 637; Livy 6. 22. 5. Six tribunes named, as promised, and in complete agreement with Livy, apart from the order in which they are named. See pp. 27 f.

48-49. The earthquake of 373/2 48. 1. ἐπὶ δὶ τούτων κτ': D's date is confirmed by Arist. Meteor. 343b; Paus. 7. 25. 4; Strabo 8 C 384; cf. Polyb. 2. 41. 7. The time of the year, winter, is given by Arist. Meteor. 343b; 344b; Paus. 7. 24. 12; Strabo 8 C 385. The disaster made a strong impression on contemporary Greece. Aristotle, Callisthenes, Heraclides Ponticus, and Ephorus all gave it space in their works. Xenophon of course would not -376have been interested in natural disasters unless personally involved. A host of later writers refer to it ( Bölte, RE vii. 2855ff.), but their details are second-hand unless, like Eratosthenes and Pausanias, they added personal observations to their accounts. In fact Ephorus' own information was derivative. We know that he used Callisthenes' Hellenica elsewhere (pp. 104f.). Did he use it here? The various accounts of the disaster can be divided in two on the basis of whether it is said that both Helice and Bura were submerged by the sea (which is wrong) or that Helice alone was submerged, but Bura simply flattened by the earthquake (which is correct: below 48. 3 n.). The mistake may have arisen from the fact that Bura probably possessed a little harbour-town, perhaps of the same name, which was of course submerged, and that event was wrongly transferred to Bura itself. We know from Pausanias that neighbouring Aegeira possessed a harbour-town of the same name. For Bura he mentions only the road which descended to the shore (7. 25f.). The submersion of the harbour-towns of Aegeira and Bura may be the reason why Philo, De aeternitate mundi 140, lists all three, Helice, Bura, and Aegeira, as having been swallowed up ( Marinatos, 1960: 190). The best statement of the correct view is in Strabo 1 C 59: Boν+̑ρα δὲ καὶ Ελίκὴ ὴ+̑ μὲν ν+̑πI+1F78 χάσματος ὴ+̑ δ ν+̑πὸ κύματος ὴ+̑ϕανίσθὴ '; cf. 8 C 384f. Strabo refers his details to Heraclides Ponticus and adds that Eratosthenes had, by his own testimony, visited the place where Helice had vanished. Pausanias also visited the area and saw the mountainous site of Bura (7. 24; 7. 25. 8). Bura cannot therefore have been submerged. Helice on the other hand had been in the plain near the sea. Furthermore Pausanias says that Bura was rebuilt by those of its people who survived (7. 25. 9: ἀποδὴμον+̑ντες ἢ στρατείας ἓνεκα κτλ. -- with Mnasippus at Corcyra?). Finally, there is Strabo's statement (8 C 386) that Bura was 40 stades distant from the sea. Polyb. 2. 41. 7 and Aelian, NA 11. 19, also get the facts correctly (though not Pliny, NH 4. 12, as Jacoby, FGH iid. 424, says). See Baladié discussion, 1980: 145 ff.

According to D both Helice and Bura were submerged. This would have mattered little had he been alone in saying so. But others say the same; ps.- Arist. De mundo396a; Ovid, Met. 15. 293; Pliny, NH 2. 206. Of greater importance is the fact that Callisthenes, as quoted by Seneca, stated the same incorrect view: NQ 6. 23. 4; (6. 32. 8); 7. 5. 3 f. It seems, therefore, that Ephorus' -377use of Callisthenes here is not just 'mindestens nicht unmöglich' ( Jacoby, FGH iid. 423), but very probable. It would appear, however, that Ephorus combined Callisthenes with Heraclides Ponticus (cf. nn. ad FGH 70 F212; 124 FF19-21). This is indicated by a comparison of D with Strabo 8 C 385 (cf. Wehrli, viii. 18f., 73; Gottschalk, 1980: 94f.). The ascription of the disaster to the anger of Poseidon in literature would seem to derive originally from Heraclides' περὶ εν+̑σεßείας (cf. Wehrli, viii. 18f., 73, and Gottschalk, 1980: 94f.). The story of the Ionian appeal in fact as we have it in D and Strabo may have been first related by Heraclides. Callisthenes was not of course a stranger to divine explanations (cf. Strabo 17 C 814), but we do not know if he combined here a natural with a supernatural explanation like Ephorus ( Jacoby, FGH iid. 423f., does not argue this pace Barber, 1935: 132, who thus sees this as more evidence for Ephorus' use of Callisthenes). In the same winter of 373/2 the temple of Apollo at Delphi was destroyed by fire according to the Marm. Par., Tod 205 s. 71, and Macrobius, Sat. 3. 6. 7 (quoting Epaphus). This is not mentioned in any other source, and probably the fire had not been caused by the earthquake. Cf. Pouilloux, 1962: 300ff., and Tod 140 (which, though, dates to spring 360: Pouilloux, 1949: 192ff.). ου+̑δέποτε γὰρ ἐν τοι+̑ς ἐπάνω χρόνοιςκτΛ: we cannot know whether this is D or Ephorus speaking. 48. 3. ἐπιλαßούσὴς δ ἡμέρας: wrong. The tidal wave accomanied the earthquake ( Arist. Meteor. 368b; cf. Marinatos, 1960: 189). D himself in fact implies that the wave was caused by the earthquake: τὴ+̑Ξαρθέν γὰρ ν+̑πὸ σαλάσσὴς μετεωρισθείσὴς κτλ. Strabo states this explicitly: ἐΞαρθὲν γὰρ ν+̑πὸ σειμον+̑ τὸ πέλαγος κτλ (8 C 384). 'Eλίκὴν: Helice had been the chief and probably oldest city in Achaea; cf. Bölte, RE vii. 2855ff.; Anderson, 1954: 72ff. It was famous especially for its temple and image of Poseidon (Paus. 7. 24. 5ff.; Strabo 8 C 384). For the possible location of the submerged city and for a theory of what happened see Marinatos, 1960: 186ff.; id., Archaeol. Reports ( 1966/7), 12. Bου+̑ραν: Buris in Latin, Bura in Greek authors (cf. Bölte, RE vii. 2855ff.). The site of the city has been identified by Ernst Meyer, 1939: 133ff.; id., 1957: 81ff. Bura lay on a precipitous spur to the east of the river Cerynites in eastern Achaea. It is mentioned again later by D (20. 103. 4 under 303/2) in connection with Demetrius' -378-

operations in the Peloponnese in that year. He granted the city its autonomy. 48. 4. περὶ Δ τ ν σμπτωμάτων κτλ : it would be reasonable to suppose that Ephorus, as well as pointing to the divine anger, also gave details of the opinions of the natural philosophers. These D omitted, choosing to concentrate instead on the anger of the gods. The story was grist to his moralizing mill and his interest is shown by the promise to deal with the subject of divine retribution on the sacrilegious later in his work. This he does in 16. 61-4 (from Book 30 of the Histories: see pp. 95 f.), and he may have done so in even greater detail in connection with the great earthquake at Rhodes in 227/6 mentioned in Book 26 fr. 8. For Ephorus and the supernatural, in which he was far from uninterested, see pp. 12 f. 49. 1 :ἐννέα πόλεις twelve according to Herod. 1. 145; Strabo 8 C 383. But the number will have varied. D's number is accepted for this time by Caspari, 1915: 182f. (Cf. 177f.), and Kleiner in Kleiner, Hommel, and Müller-Wiener, 1967: 14f. Magie, 1950: ii. 867 n. 48, suggests that D perhaps reflects a tradition that originally the Ionian League consisted of nine cities. Lenschau's solution, Klio, 33 ( 1940), 220f., is ingenious: of the twelve cities listed as making up the League (cf. Magie, 1950: ii. 867 n. 48, for references in addition to Herodotus and Strabo cited above), two were the islands of Samos and Chios which were given their freedom by the King's Peace, while Priene was not a city properly speaking at this time; hence the transfer of the Panionion to Ephesus. There remain therefore D's nine cities. This would place the Ionian embassy to Achaea after 386 which indeed agrees with the evidence of Strabo (see n. below). We might then conclude that D in his usual bungling way misunderstood the nine cities which sent the embassy to be the cities which originally celebrated the Panionia. εἰώΘεισαν κοννὴν. . . ἐν ἐρὴμῳ τόπῳ: the old Ionian League and its festival at Mycale; cf. Herod. I. 148; Strabo 8 C 384; 14 C 639; Nilsson, 1906: 74ff.; Wilamowitz-Moellendorff, Kleine Schriften, v. 1. 129ff. (art. published 1906); Caspari, 1915: 173ff.; Roebuck, 1955: 26ff.; Magie, 1950: ii. 866 n. 47; Kleiner in Kleiner, Hommel, and M llerWiener, 1967: 12ff. στερον Δ πολέμων. . . πλησίον τη̑ς Ἐϕέσου : for the problems of this passage, almost certainly due to D's incompetent abbreviation -379-

of Ephorus, see the discussion in Stylianou, 1983: 245 ff. It is also argued there that the Ephesia of Thuc. 3. 104. 3 should not be identified with the Panionia; contra Hornblower, 1982b: 241ff., and 1991: 527ff.

∏υΘώΔε: cf. Parke and Wormell, 1956: i. 212ff. ἀϕιΔύματα. . . ßωμω̑ν: 'to take copies of their ancient and ancestral altars'; cf. Strabo 8 C 385: a copy of the temple. On the meaning of ἀϕίΔRHρυμα see Ziehen, Leges Graecorum Sacrae ( 1896), 115 no. 40 on lines 13 f. 49. 2. τὸ κοιν τω+̑ν Αχαιω+̑ν : Strabo too, 8 C 385, mentions the koinon of the Achaeans (the information of both ultimately derives from Heraclides; see 48. 1 n.). The evidence seems conclusive that Achaea was organized in a Confederacy at this time. In fact the Confederacy had existed already in the 5th century; see Larsen, Studies Robinson, 797ff.; id., 1968: 80ff; Anderson, 1954: 72ff. Strabo, 8 C 385, says that the Ionians first approached Helice and on failing applied to the koinon. Anderson, 1954: 88, considers D's sequence more likely. 49. 3. συνήρπασαν : perhaps they were even killed: Paus. 7. 24. 6; Aelian, NA 11. 19. 49. 4. Διὰ τὸ Δοκειν τὸ παλαιὸν κτλ : this appears correct; see Nilsson, 1967: 447f., and cf. the large number of references in Farnell's geographical register of Poseidon cults, iv. 81ff. 49. 5. πρὸς Δ τὴν ∏ελοπόννησον κτλ : it is not the case that Ephorus in this instance inserted part of a 'scientific' explanation into his 'religious' account (as Gottschalk, 1980: 95 n. 23). These underground rivers would have been associated with Poseidon and his worship; cf. Paus. 8. 7. 2. Ephorus' purpose in adding this information was to prove further that the Peloponnese was especially sacred to Poseidon. Δύο . . . ϕανερούς: there were others, but these two visibly ran underground, or so Ephorus thought. For other wholly or partly underground rivers see FGH 566 (Timaeus) F41; Polyb. 16. 17. 5ff.; Paus. 8. 54. 1ff.; Strabo 6 C 270f., 275 -- the Alpheus: Pritchett, 1965-85: i. 123ff.; Walbank, HCP ii. 329. And Paus. 8. 7. 2 with Pritchett, 1965-85: i. 131ff., for a remarkable underground stream which runs from near the village of Nestani in Arcadia and rises in the sea off the coast of Argolis. Baladié, 1980: 93ff. ὅ τε γὰρ περὶ ϕένεον . . . ὑποΔεχομένων: this is the river Ladon, a -380tributary of the Alpheus. It originates near Pheneus in north-east Arcadia; see Paus. 8. 20 with Frazer, iv. 262f., and 231ff.; Strabo 8 C 389; Baladié, 1980: 98ff. ὅ τε περὶ Στύμϕηλον. . . ἐξίησιν: see Paus. 8. 222 with Frazer, iv. 268f.; Strabo 6 C 275; 8 C 371, 389. This river reappeared as the Erasinus in the Argolid plain. The ancient claims have been confirmed by modern science: Pritchett, 1965-85: i. 122f., 132. Στύμϕηλον: Vogel's emendation. The MSS read either τὸ Δτύμϕην or τὸ Δτύμϕιον. Pausanias calls it 'Stymphelos' except once (2. 24. 6) when he calls it 'Stymphalos'. That the people of the city called themselves 'Stymphalioi' is proved by their coinage (Head, 1911: 454) as well as by the usage of the ancient authors in general (Frazer, iv. 270f.). Vial's emendation 'Stymphalon' is not necessarily correct for we cannot know which form D (Ephorus) used here. Elsewhere D uses the form 'Stymphalos' (3. 30. 4; 4. 13. 2; 4. 33. 7; 19. 63. 1), but for none of these passages was Ephorus the source. 49. 6. οὐΔεὶς ἄλλος : Philo, De aeternitate mundi 140, adds Aegeira (48. 1 n.). Pollis and his squadron (47. 5-6 n.) were not guiltless to religious minds (Diog. Laert. 3. 20). Strabo 8 C 385 says that after the disaster the Achaeans gave to the Ionians the model of the temple they had requested. One wonders how, when city and temple had disappeared!

Archon Alcisthenes, 372/1 50. 1. Πωμαιοι κτλ : 380 V; CAH2 vii. 2. 638. Eight tribunes announced, but only seven given of whom five are also given by Livy (6. 27. 2), if Γάιον Τερένττον in D is indeed a corruption of Cn. Sergius ( Drummond, 1980: 65 n. 39). The names given together by D and Livy are also to be found in the FC. See 22. 1n. As often, MS F adds names not confirmed by other sources.

50. 2-3. The comet of 373/2 50. 2. ἔτη σχεΔὸν πεντακὸσια : see 1. 3n. τὸ Θειον προήμαινΕν: for other omens which foretold the Spartan downfall see 54. 1-3 and nn.; X. H. 6. 4. 7; Plut. Ages. 28. 6; Lys. 18. 1; Mor. 397e-f; Cic. De div. 1. 74-6; 2. 54. On Ephorus and the supernatural see pp. 12 f. -38150. 2-3. ὤϕΘη μὲν γὰρ κτλ : only two other sources mention this phenomenon -- Arist. Meteor. 343b; 344b; and Seneca, NQ 6. 26. 3; 7. 5. 3f; 7. 16. 2. Seneca derives his information from Aristotle, Callisthenes, and

Ephorus, to all of whom he refers by name. There was evidently controversy at the time concerning the nature of the object and its fate, as is shown by Arist. Meteor. 343b 15ff. Aristotle pronounced it a comet; and it dispersed without leaving anything behind. Others thought differently. Callisthenes described it as an extended flame or torch ('columna ignis immensi': NQ 6. 26. 3; 'talem effigiem ignis longi': NQ 7. 5. 3. Cf. Aristotle's list of meteoric phenomena at Meteor. 341b). Otherwise Aristotle's description is similar to that of Callisthenes and D: τὸ Δὲ ϕέγγος ἀπέτεινε μέχρι του+̑ τρίτου μέρου μέρου⌣ του+̑ οὐρανου+̑ οἱ+ιον ἄλμα (or ἄμμα, the alternative MS reading). Διὸ καὶ ἐκλ Θη Ƿ1ΔǷ9ς (Δοκ ς is perhaps the correct reading on the evidence of D and Seneca). Some said that on disappearing the object had left behind one or more stars. Amongst these, it would appear, was Democritus (his dates are somewhat uncertain, but it is possible that he was alive in 373/2: Diog. Laert. 9. 34, 39, 41, 43). Aristotle would have none of this (343b), but Ephorus accepted it; for which he is severely and unjustly criticized by Seneca, NQ 7.15.if. (Seneca is thus wrong that no one except Ephorus said that the comet split up). But Ephorus here, as elsewhere, used more than one source, and D's description makes it clear that Callisthenes was yet again the ultimate source: λαμπὰς μεγάλη καομένη ἀπὸ του+̑σχήματος ὀνομασΘεισα, πΥρίννη ΔοκǷ7ς (pp. 104f.). D does not mention the splitting up of the comet, but that is hardly disconcerting. 50. 2. μικρὸν Δ ὕστερον : in fact some eighteen months later. 50. 3. ἔνιοι Δὲ τω+̑ν κτλ : more is made of the explanation of the natural philosophers here; cf. 48. 1, 4nn.

50. 4-6. The Peace of 372/1 For the date see the chronological excursus at 45. 3, part 2. 50. 4. ΑρταÝὲρξης κτλ : the participation of the King in this Peace is often denied on the basis of Xenophon's silence; cf. especially Accame, 1941: 152ff. As 15. 38 and 15. 50. 4 are generally considered to represent a doublet, the mention of the King here is dismissed as a detail erroneously introduced from the Peace of 375. That 38 and 50. 4 are not in fact a doublet, see 38. 3-39n. -382Accame's contention (1941: 154) that, whereas in 375 the King is given a motive for his intervention, in 371 he is not, and that for this reason alone D's statement is dubious, is weak because the King had other reasons for intervening in 375, as in 371 (38. 1n.), and these are not supplied by D either. Luckily we are not entirely dependent on D here. There is the explicit testimony of Dion. Hal. Lys. 12 that the King was indeed a party to the Peace of 372/1: μετὰ γὰρ ΑλκισΘένην ΑΘηναιοί τε Καὶ ΑακεΔαιμόνιοοι καὶ ßασιλεὺς ὤμοσαν (Dionysius almost certainly derived this information from Philochorus; FGH 328 F151n.). Accame's objection that IG ii2 103 (=Tod 133 -- an Attic decree of 368 in honour of Dionysius I of Syracuse): τἠ+ῃ ßα]σιλέως εἰρήνῃ, ἣν ἐποιήσα[ντο ΑΘηναιοι] καὶ ΑακεΔαιμόνιο[ι] καὶ ο[ἱ ἄλλ1οι Ελληνες], proves that Dionysius was in error (1941: 154f). cannot be sustained for the epigraphic reference is probably to the Peace concluded at Athens after Leuctra (cf. Buckler, 1980a: 237f.), and no one would argue that the King was a party to that. The epigraphic reference cannot surely be to the original King's Peace (as Lauffer, 1959: 323) for no Athenian could ever have spoken of Dionysius and his sons as 'supporting the King's Peace which the Athenians. . . concluded', meaning the Peace of 387/6, when in fact Dionysius had, in the period 386-371, supported the Spartans in exploiting the Peace against Athens and the rest of Greece (very different is the language used when referring to the 370s in the decree commending the Mytileneans (Tod 131 lines 46ff.) passed in the very same year as Tod 133). That leaves us with only the present Peace and that of 371/0 concluded at Athens. But the first was overtaken by Leuctra, and the present tense, ßοηΘου+̑σιν (line 24 in Tod 133), points to the more recent of the Peaces ( Hampl, 1938: 21ff.). Tod 133 is thus additional evidence that the Spartans participated in the Peace at Athens; see below under 57. 1. It might be objected that such references as in Tod 133 are not to any particular peace: there was one peace in 387/6 which was renewed from time to time and that is what is being referred to. But neither the terms nor the circumstances of the various peaces were precisely the same so that contemporaries could, and in fact did, distinguish between them; cf. e.g. FGH 328 (Philoch.) FF149, 151. Accame's attempt to exploit Xenophon's silence and vagueness is unconvincing, for two reasons. First, it is Xenophon after all -383-

who is speaking at H. 6. 3. 10ff., and not Callistratus, and what he says is not necessarily factual. Second, Xenophon could have deliberately excluded the King from the Peace. (On X. H. 6. 3. 1ff. see Tuplin, 1913: 101ff.) Even if we accept that at the time of the congress Antalcidas was at Susa, this does not exclude the possibility that the King had already dispatched his edict, παρακαλω+̑ν . . . συνΘέσσαι κοινήν εἰρήÚην κτλ, as D puts it (50. 4), and that the Greeks had acted as a result of it. Personally I am inclined to see H. 6. 3. 12 as referring to this edict: ßασιλεὴς μὲν γὰρ Δὴπου ἔγραψε πάσας τὰς ἐν τῃ+̑ Ελλάδι πὸλεις αὐτΟνὸμους εἴναι. This is often seen as a reference to the original King's Peace (cf. e.g. Tuplin, 1993: 102), but unnecessarily so, and the last sentence of H. 6. 3. 12, given in the present tense, shows that the King was taking an active interest in Greek affairs at the time. It should be added further that, though not mentioned again, the motive Ephorus gave for the King's involvement in 375 (38. 1), may still have counted with the King, as preparations were being made about this time for yet another attempt to subdue Egypt ( Nepos, Dat. 3. 5; 4. 1; ps.-Demosth. 49. 28, 32, 35). For the King's involvement argues also Xenophon's phraseology at H. 6. 3. 18: ἐφηϕίσαντο καὶ οἱ ΑακεΔαιμόνιοι ΔέχεσΘαι τήνην. That rather points to a royal edict.

However, there is nothing objectionable in meeting half-way critics like Accame who assign the initiative for the Peace to Athens, by suggesting that both the King and Athens promoted the peace congress. Sparta too may well have had a hand; why else would Antalcidas have been sent to Persia (cf. Buckler, 1980a: 48 f.)? On the peculiarities of the Athenian embassy to Sparta see Mosley, 1962: 41ff.; Tuplin, 1977: 51ff. On Athenian policy at this time, Ryder, 1963b: 237ff. On Theban policy, Mosley, 1972: 312ff.; Cawkwell, 1972: 258ff.; Buckler, 1980a: 52ff. συνέΘεντο. . . εἰρήνην: Svt 269. For the terms see 5. 1n.; 38. 4n. συντέλειαν: see 28. 1n. Θηßαιοι. . . οὐ προσεΔέχΘησαν κτλ: this is hardly the way a writer partial to Thebes and the Boeotian Confederacy would have put it; cf. e.g. Plut. Ages. 27f.: the 'Greeks' there side with Epaminondas and Thebes. Not so in D (see p. 118 for the Ephoran viewpoint). X. H. 6. 3. 18ff. relates what occurred at Sparta though he omits the clash between Agesilaus and Epaminondas (for which see Plut. -384-

Ages. 27. 6-28. 4; Nepos, Epam. 6. 4). As members of the Athenian Confederacy the Thebans first swore to the Peace as 'Thebans'. Indeed they had no choice but to do so for the Confederacy did not, officially at any rate, recognize Thebes' control of Boeotia. On the following day, however, they demanded that their signature be changed to 'Boeotians'. On Agesilaus' retort that they could only be a party to the Peace as 'Thebans', they departed and their name was struck from the treaty. They were ἔκσπονΔοι. What does this incident signify? There was probably a combination of factors at work. In 375 the Thebans had, after some initial objections (38. 339n.), taken the oaths as members of the Athenian Confederacy and had thus agreed to the dissolution of the Boeotian Confederacy. That dissolution had been nominal rather than actual for we know that in 373 the Confederacy was in existence before the outbreak of war (38. 3n.). It may well be, therefore, that in 371 too the Thebans reasoned that they could get by if they went along with the Peace and pretended to acquiesce in the independence of Boeotia. But conditions were different now. The rapprochement between Athens and Sparta in 375 was suspect and, as the Peace disintegrated, her need of Thebes caused Athens to turn a blind eye to the latter's activities in Boeotia. Further, Sparta was not in a position to act in 375. Now, however, Cleombrotus was in Phocis with an army and Athens was determined to remain neutral. It would have dawned on the Theban leadership, therefore, that this time Sparta ( Agesilaus) was resolved to dislocate the Boeotian Confederacy, and for good. But there must also have been a great deal of fear and indecision on the Theban side ( Cawkwell, 1972: 258ff., 263ff.). Many Thebans would have preferred peace and safety to the prestige of the Boeotian hegemony (see 52. 1n.). 50. 5. ὡς κοινοὺς : i.e. Sparta was implementing the Peace in her role as prostates. σϕόΔρα γὰρ ὑϕεωρϊ̑ντο κτλ: see 5. 3n.: 20. 1n. 50. 6. κατὰ μ ν οὑ+̑ν του+̑τον τὸν ἐνιαυτὸν κτλ : D's chronographer would have mentioned the two events, the Peace and the battle of Leuctra, under the successive years of Alcisthenes and Phrasicleides, and D, in the absence perhaps of any chronological hints in Ephorus, assumed that some time elapsed between the Peace and the battle; hence the stock statement about great, yearlong preparations. In reality, a very few weeks separated the battle -385-

from the Peace (though not twenty days). Also, Cleombrotus was sent to Phocis in 372/1, perhaps shortly before the Peace (chronological excursus at 45. 3, parts 2 and 3). It is possible that Ephorus mentioned this in his account of the Peace, but explaining how he had come to be already in the field, and D wrongly assumed the dispatch of Cleombrotus to be after the Peace and directly aimed at Thebes, rather than before it and intended for the protection of Phocis. But D's account of Leuctra is a very confused affair (see below). In any case it is not necessary to suppose that a proTheban source was at work here. κατέλεγον: see 26. 2n. πολιτικάς: see 19. 3n. συμμάχων: for the contribution of the allies see 31. 1-2n.

Archon Phrasicleides, 371/0 51. 1. Πωμαιοι κτλ : 379 V; CAH2 vii. 2. 638. Eight names are announced and eight are given, six of which correspond to the six given by Livy (6. 30. 2), though three are evidently corrupt ( Drummond, 1980: 65 n. 39). See 22. 1n.

51-56. The campaign of Leuctra For the date see the chronological excursus at 45. 3, part 2. 51. 1. ἀναΔέξασΘαι μόνοι : see Plut. Pel. 20. 1; i.e. they were devoid of non-Boeotian allies. But the Boeotians were with them: 52. 2; X. H. 6. 4. 4; Paus. 9. 13. 8; Polyaen. 2. 3. 3. Jason was an ally (54. 5; X. H. 6. 4. 20), but he was not present at the battle. οὐΔεμι ̑ γὰρ . . . εἰρ νην: Sparta, as a prostates, had called for action against the Thebans, the co-prostates, Athens, acquiesced, and no city therefore could 'legally' aid Thebes; cf. 5. 1n.; 5. 2-3n.; 38. 4n. 51. 2. μονωΘέντων . . . ἐξανΔραποΔίσασΘαι : the opportunity was judged ideal to crush Thebes; cf. Plut. Ages. 28. 6; Buckler, 1980a: 54f. But ἐξανΔρποΔίσασΘαι cannot represent the Spartan intentions; cf. X. H. 6. 3. 20 ΔεκατευΘη+̑ναι. ῥκΔίως. . .καταπολεμηΘη̑ναι: Cf. X. H. 6. 3. 20. 51. 3. παρέΔωκαν. . . Κλεομßρ τῳ : as seen above (50. 6n.) D has got the timing and purpose of Cleombrotus' dispatch to Phocis wrong. That Ephorus was not to blame for the confusion is -386

obvious from D's own account: after the rejection of their ultimatum the Spartans immediately sent out Cleombrotus with an army. οὑ+̑τοι μὲν οὑ+̑ν Προάγοντες ὡς ἠ+̑κον εἰς Ξαιρώνειαν (not Κορώνειαν; see n. there), κατεστρατοπέΔευσαν (52. 1). On discovering that the passes before him were occupied by the enemy, Cleombrotus decided to invade Boeotia by a different route. πορευΘεὶς Δὲ Δι 0 τηςΦωκίΔος κτλ (53. 1). All this is nonsense of course if Cleombrotus set out from Laconia, and it only makes sense in the light of what we know from the other sources, i.e. that Cleombrotus was already in Phocis when ordered to invade Boeotia. προστάττοντες. . . ἀποκαστη+̑σαν: Aristid. Leuctr. pro Theb. 1. 7, 34 p. 644 Dind., says that the ultimatum was delivered by Cleombrotus. Grote, basing himself on Xenophon (X. 239 n. 1), doubts that there was communication between the two sides after the congress dispersed. But Xenophon's text rather argues in favour of it: ἐπέστειλαν Δὲ τῳ+̑ ΚλεομßρǷ9τῳ κτλ (H. 6. 4. 3). D's ultimatum from Sparta to Thebes may well have gone by way of Cleombotus. The Spartan demands cannot have been explicitly stated in the peace terms for then the Thebans would not have added their name to the treaty document on the first day. The demands represented rather the Spartan interpretation of the autonomy clause. The Plataeans in fact had declared their treatment at the hands of Thebes a violation of the King's Peace (Isocr. 14. 5, 10, 12, 17, 23f., 39), and the Thespians had doubtless added their voice too (cf. X. H. 6. 3. 1). The independence of Boeotia apart, Sparta could do no less than demand the resettlement of Plataea especially as she had been instrumental in the rebuilding of the town after the King's Peace of 387/6. All Spartan actions in Boeotia between 386 and 379 were probably based on the autonomy clause and the prostasia of the Peace (5. 2-3 n.). The association of Thespiae with Plataea here is misleading for probably Thespiae had only had its walls demolished before Leuctra (46. 6n.). Perhaps some Thespians (the anti-Theban faction; there was also a democratic pro-Theban group: X. H. 5. 4. 46, 55) had already been expelled by the time of the peace congress and their land confiscated (cf. Isocr. 14. 35, 44) so that the demand τὴν χώραν τοις πρότερον γεγονόσι κυρίοις ἀποκαταστη+̑σαι could apply to both Plataeans and Thespians. In any case it is hard to -387find another candidate for Isocrates' τὰ τείχη κατασκάπτοντες/τά τείχη κατεσκΘασι (14. 19, 35). The walls of Orchomenus were demolished after Leuctra, while those of Tanagra stood intact throughout the 4th century ( Roller, 1974: 260ff.). And Plataea, Thespiae, Orchomenus, and Tanagra are the only cities mentioned in the sources as having opposed Thebes at this time. I agree with Buckler, 1980a: 20ff.; id., 1977: 76ff., that Paus. 9. 13. 8-14. 4 is suspect in part. But I see no reason for rejecting the statement that before Leuctra the Thespians had a polis. It is Xenophon who should be rejected (46. 6 n.). Buckler, 1980a: 20ff., is correct to stress that with the exception of Plataea, Orchomenus, and Thespiae there is no evidence that the Thebans adopted harsh measures, such as the demolition of walls and the confiscation of land, against any other Boeotian cities; and they never felt it necessary to garrison the federated cities. 51. 4. ἐξέπεμΨαν : see 51. 3n.; 46. 3n. οἱ . . . σύμμαχοι προΘύμως εἰ+̑χον: not so the other sources and one suspects that this is D writing carelessly rather than Ephorus; cf. X. H. 6. 4. 15; Plut. Ages. 28. 5; Nepos, Epam. 6. 4; Paus. 9. 13. 9. 52. 1. χαιρώνειαν : thus the MSS. The emendation Κορώνειαν is Wesseling's and it generally stands unchallenged (it is printed by both Vogel and Vial), students of the campaign assuming that Cleombrotus reached as far as Coronea. Tuplin, 1979: 351 ff., is perfectly correct, however, to revert to the MSS reading. As he argues, not only is there no need to emend the text, but Chaeronea makes better sense in the reconstruction of Cleombrotus' route.

τω+̑ν συμμάχων ἀνέμEενον: these would have been the allies from central Greece such as the Phocian peltasts and the cavalry from Heraclea; X. H. 6. 4. 9; cf. Beister, 1970: 24f.; Tuplin, 1979: 355. Cleombrotus' orders presumably were for everyone to gather at Chaeronea. οἱ Δὲ Θηßαιοι Διὰ τὴν παρουσίαν κτλ: in view of the fact that the Boeotians were in position at the passes near Coronea first (52. 7; 53. 1), a shift back in time should be understood here: Cleombrotus had not yet reached Chaeronea when the Thebans marched out. Alternatively, Cleombrotus may have delayed at Chaeronea, either because he was waiting for his allies or deliberately ( Tuplin, 1979: 354f.), thus giving the Thebans the opportunity to march out and occupy the passes. -388ἐφηφίσαντο: D's poor ability to deal with what must have been a detailed and involved narrative in his source deserts him altogether in places. It should surely be dismissed as absurd that the Thebans voted (and as far as D is concerned acted upon the vote) to remove their women and children to Athens. On the other hand, Paus. 9. 13. 6 says that at Leuctra the three Boeotarchs who opposed Epaminondas suggested that they should send their women and children to Athens while they themselves made ready to undergo a siege rather than fight a battle. One may guess that Ephorus had said that many Thebans opposed Epaminondas' bold plan of meeting the Spartans head-on, and were in favour instead of remaining on the defensive as in 378 and 377. If a proposal was made in the assembly to send all non-combatants to Athens it certainly did not get very far. The story itself is suspect because of its Herodotean echo (Herod. 8. 41-1). But cf. next n. Ε+̔παμεινώνδαν . . . βοιωταρχω+̑ν ἕξ : each Boeotarch possessed equal authority in the field though it did, it seems, sometimes happen that the Boeotarchs voluntarily surrendered their authority to one or more of their colleagues because they judged them better suited to carry out a particular task ( Buckler, 1980a: 26f.). This is how D should be understood at 62. 4 and 68. 1. That D is wrong here (as is Nepos, Pel. 4. 2) is shown by his own information at 53. 3: Epaminondas had not more authority than his colleagues. Yet something may lie behind D's assertion. There must have been intense discussion at Thebes with Epaminondas playing a leading part, and Ephorus perhaps noted a proposal to evacuate the noncombatants to Athens, which was rejected (previous n.), as well as the eventual acceptance of Epaminondas' motion to march out against Cleombrotus. στρατηγὸν ἑλόμενοι : a Diodoran cliché (p. 133). The Boeotarchs were elected probably in late autumn perhaps a month or so before they assumed their office ( Buckler, 1980a: 28f.). 52. 2. πανδημεί : a favourite word with D, not always to be believed though here certainly true (p. 16). καταλέξας: see 26. 2n. εὐθτους : 'able and willing' or 'fit and qualified'. A Diodoran word often used in connection with men being enlisted; cf. 13. 97. 1; 14. 18. 4; 14. 44. 2; 16. 28. 1; 17. 31. 1. All Boeotians, of course, were liable to serve, but they might not all have been willing to do so. τοὺς σύμπαντας . . . ἑξακισχιλίων : Front. Strat. 4. 2. 6 gives 4,000 -389-

Thebans including 400 cavalry (but see Anderson, 1970: 321 n. 26). Otherwise we have no figures for the Boeotians (Polyaen. 2. 3. 8 does not give 6,000 pace Anderson, 1970: 321 n. 26, while the 6,000 at 2. 3. 12 is an addition from D 52. 2). D's 6,000, whether hoplites alone or hoplites and cavalry together, seems reasonable when compared to Boeotian numbers in the 5th and 4th centuries (26. 4n.). Some Boeotians doubtless stayed away (cf. Paus. 9. 13. 8). About half of the 6,000 would have been Thebans; cf. Busolt, 1905: 444f.; Wolter, AS iv. 299ff.; Judeich, 1927: 193f.; Anderson, 1970: 197f.; Buckler, 1980a: 55. For a study of the Boeotian federal army see Salmon, 1953: 347 ff. 52. 3. δυσχερη+̑ σημει+̑α : Ephorus' treatment of the omens is of interest. Though not unimportant in themselves, they are used here to show the superiority of Epaminondas' intellect and paideia over the mere superstition of the πολλοί. For Ephorus, paideia, and religion see pp. 10 ff. Cf. Sordi, 1974: 45 ff. κη+̑ρυξ . . . ἀνασώζειν : the text is corrupt beyond recovery. Perhaps it is best to emend the MSS τυφλόν to τυφλός (with Madvig) and ἄνδρα ἀποδεδρακότα to ἀνδράποδα ἀποδεδρακότα (again with Madvig): see Vogel's text. Vial's restoration τυφλὸν ἀνδράποδα ἀποδεδρακότα is ungrammatical. Thus, a blind herald was proclaiming the escape of slaves and, as was customary, warned against anyone exporting or concealing them; they were to be kept safe for their owners instead. For a suggestion why this was considered a bad omen see Sherman ad loc. 52. 5. γραμματεὺς . . . ἡγεμόνων : the grammateus of the polemarchs (or Boeotarchs) at Thebes and at other Boeotian cities was an annually elected official who assisted his superiors in their duties; cf. X. H. 5. 4. 2, 4: Plut. Pel. 7. 4; Mor. 577b. The office was evidently important though most of the evidence derives from Hellenistic inscriptions (see Roesch, 1965: 157, 161, 163; Busolt, GS ii. 1440). The evidence of D here indicates that one of the duties of the grammateus was that of aide-de-camp when the Boeotarchs were on campaign. δόρυ καὶ ταινίαν : Plut. Mor. 578c says that Theban hipparchs carried a spear and a seal as badges of office. Elsewhere he says that the magistrates chosen by lot at Thebes always carried with them a sacred spear (Mor. 597bc). At Plataea and Chaeronea on the other hand the archon was not allowed to carry or even touch metal (Mor. 274b; Aristid. 21. 4; cf. Roesch, 1965: 158). Evidence -390-

is lacking, but one wonders whether the spear and band carried by their secretary might not have been the Boeotarchs' symbols of office. It could be, however, that the grammateus had a band tied to his own spear so as to be readily discernible to the soldiers to whom he passed on the commands of the Boeotarchs. οἶ μετ+̕ Α+̔γησιλάου στρατεύσαντες: see 34. 2n. 52. 6. οὐδὲν . . . ἀποκριθείς : according to Front. Strat. 1. 12. 5 he made a reply. 52. 7. προκαταλαβόμενος . . . στενά : cf. Paus. 9. 13. 3 Ε+̔παμεινώνδας . . . ἀντεκάθητο ὑπὲρ τη+̑ς Κηφισίδος λίμνης. The Boeotian position was between Coronea and Haliartus at Mt. Tilphusius which in antiquity was only a few hundred feet from lake Copais; cf. Wolter, AS iv. 291. For the extent of the lake see Pritchett, 1965-85: ii. 89ff. For Mt. Tilphusius, which should probably be identified with both the ridges known as Petra and Palaiothivai, see Fontenrose, 1969: 125f.; Buck, 1979: 9; Wallace, 1979: 144 f. D here speaks of στενά, which could mean a single pass (cf. X. H. 6. 4. 3 ἐπὶ στενῳ+̑ τινι ἐφύλαττον), but at 53. I he writes παρόδους. Fossey, 1972: 1 ff., identifies two passes, one between Petra and the lake and the other between Palaiothivai and the village of Koutoumoula (see his map p. 3 ). In fact Fossey, 1972: 13 f., suggests that the forts on Petra and Palaiothivai together constitute ancient Tilphusium referred to at Demosth. 19. 141, 148 (cf. D 4. 66. 5; Strabo 9 C 410, 413). The forts, however, cannot be dated and Fossey's suggestion that they date to 371 is a mere guess. But their purpose is clear: they were meant to control the natural route from western to eastern Boeotia.

. The most likely route followed by Cleombrotus into Boeotia is that suggested by Burn, 1949: 321f. (accepted by Buckler, 1980a: 57 ff.; contra Wolter, AS iv. 291ff.; Pritchett, 1965-85: i. 56f.) except that the starting point was Chaeronea and not Coronea (52. 1n.). Cleombrotus first fell back to Ambrossus in Phocis and then, setting out from there, he marched over Mt. -391Helicon by way of Stiris and the modern villages of Kyriaki and Koukoura (see Buckler's map p. 56 ) and descended to Thisbe in Boeotia. Pritchett's objection to this route is based on D's statement, διεξελθὼν τὴν παραθαλαττίαν ὁδόν, but this, as Pritchett himself argues on p. 53 (against Wolter and Heurtley, BSA 26 ( 1923/5), 40), probably refers to Cleombrotus' march from Siphai to Creusis along the shore route round Mt. Koromboli. It is possible, however, that Cleombrotus traversed Mt. Koromboli; see Roesch, 1967: 292, who describes a route from Siphai to Creusis which passes over Mt. Koromboli (cf. Pritchett, 1965-85: ii. 177). See also Beister discussion, 1970: 36ff. ἐν παρόδῳ . . . ἐγάνετο : from Thisbe Cleombrotus proceeded south-east to Siphai on the coast and from there to the port of Creusis (see previous n.), thus safeguarding his communications with the Peloponnese, before heading inland towards Thebes. As for the route he followed to Leuctra from here, the one traced by Pritchett, 1965-85: i. 52ff., is possible, but see Tuplin, 1987: 74ff., for other possibilities. The townships taken were those of Thisbe, Siphai, and Creusis. At the last, ten triremes were captured (δάκα is Vogel's emendation. X. H. 6. 4. 3 says twelve). For all these places see Fossey, 1988: 157ff. (with maps). 53. 2. καταντήσας: Burn, 1949: 322, assumes that the Thebans reached Leuctra first. But this is contrary to D; and cf. X. H. 6. 4. 4; Plut. Pel. 20. 4; Anderson, 1970: 195, 320 n. 17. τὰ καλούμενα Λευ+̑κτρα : it is not certain that there was an ancient village called Leuctra, and if so, where it was located; Pritchett, 1965-85: ii. 49ff.; Buck, 1979: 10; Wallace, 1979: 152; Tuplin, 1987: 73 n. 7; Fossey, 1988: 154ff. λόφους τινὰς . . . Λευκτρικὸν πεδίον : on the battlefield and the respective camps of the combatants, not easy to fix, see Pritchett, 1965-85: i. 57; Beister, 1970: 44ff.; Wallace, 1979: 111; Buckler, 1980a: 60f.; Tuplin, 1987: 73f. The Theban victory monument discovered by Ulrichs in 1839 ( Reisen und Forschungen in Griechenland, ii. ( 1863), 103f.) was excavated and restored by Orlandos, Ergon ( 1958), 48ff. (= Praktika ( 1958), 43f.); ( 1959), 173; ( 1960), 222ff.; ( 1961), 229ff. τὸ μέγεθος τη+̑ς δυνάμεως: cf. 53. 3 πολλαπλασίονα 56. 3 πολλαπλασίων. Otherwise D gives no figures for Cleombrotus' army. In fact the king had some 10,000 hoplites and 1,000 cavalry as Plut. Pel. 20. 1 says; see 32. 1n. However, the battle was fought -392-

between the Lacedaemonians and the Thebans; the rest of the Boeotians and the Spartan allies did little if anything. Four morai up to thirty-five years from the minimum amounted to some 2,200-2,300 men of whom only 700 were Spartiates ( X. H.6. 4. 15; 32. 1n.). The Thebans probably numbered some 3,000 hoplites (52. 2n.). 53. 3. The story of the evenly split board of Boeotarchs is also given by Pausanias (9. 13. 6-7) though with some differences and additions, the most important being the names of Epaminondas' fellow Boeotarchs. Xenocrates, one of the three who supported Epaminondas, is mentioned elsewhere by Pausanias (4. 32. 5-6), an episode confirmed by IG vii. 2462 = Tod 130 (see n. below). Cf. Plut. Pel. 20. 3 and X. H. 6. 4. 6 for the dilemma which faced the Theban leaders. 53. 4. It was seen above (52. 3n.; 52. 3-7) that Epaminondas' paideia proved superior to the unfavourable omens. But the soldiers remained superstitious and we are now told how Epaminondas employed his ingenuity to overcome their fears by inventing favourable omens.

διόπερ τινω+̑ν . . . ἔπεισεν εἰπει+̑ν : see X. H. 6. 4. 7 f. Xenophon does comment that some people said that the various reports of supernatural happenings had been fabricated by the Theban leaders, but characteristically he is inclined to believe in the divine signs. Like D, Polyaenus (2. 3. 8, 12) and Frontinus ( Strat.1. 11. 16) say that Epaminondas was responsible for these portents. τὰ κατὰ τὸν νεὼν του+̑ Η+̔ρακλέοΥς ὅπλα κτλ : thus also X. H. 6. 4. 7 though he mentions only Heracles as thereby armed and gone to take part in the battle. In D, on the other hand, the rumour claimed that the ancient heroes in general picked up the weapons and set out to aid the Boeotians. Callisthenes (apud Cic. De div. 1. 74= FGH124 F22a), by contrast, said that the weapons were found lying on the floor; and cf. Polyaen. 2. 3. 8 for yet another version. For Heracles and the Heracleion at Thebes see Schachter, 1981-94: ii. 14ff. Τροφωνίου: for the oracle of Trophonius at Lebadea see Paus. 9. 39-40. 2 with Frazer's nn.; Philippson, 1939: 11ff.; Nilsson, 1967: 450; Schachter, 1981-94: iii. 66ff. προστέταχεν U=1F41 Θεός: a different reply is given by Paus. 4. 32. 5-6; or perhaps a different section of the same reply. See also Callisthenes, apud Cic. De div. 1. 74 = FGH 124 F22a; Polyaen. 2. -393-

3. 8. IG vii. 2462 = Tod 130 lines 4-5 has been seen as referring to the carrying out of the god's response; see Beister, 1973: 65ff., but cf. Tuplin, 1987: 94ff. ἀφ+̕ οὑ+̑ δὴ Βοιωτοὶ . . . Λεβαδείᾳ : Schachter, 1981-94: iii. 111ff. Inscriptions testify to the celebration of this festival; cf. IG vii. 552; 1711; 2487; 2532; 3091; RE iii. 82. For the priesthood of Zeus Basileus which presided over the festival see Roesch, 1965: 200 ff.; Schachter, 1981-94: iii. 114ff. 54. 1. Λεανδρίας: this episode is known only from D. Dindorf emended 'Leandridas', C. Müller 'Cleandridas'. On the last emendation is based the suggestion that he may have been a son of Gylippus, the victor of the Athenians in Sicily whose father was named Cleandridas. Both these men had been exiled for corruption; Thuc. 6. 93. 2; 7. 2. 1; D 13. 106. 9f. (the MSS read 'Clearchon' wrongly for 'Cleandridan'); Plut. Per. 22. 3; Nic.28. 4; Lys. 16; Poralla, P. 72; RE xi. 557. Whatever the correct name of this Spartan, he was not a deserter ( Anderson, 1970: 206), but an exile; and he did not tell the Thebans the story of the outraged maidens of Leuctra. This mistaken assumption was first made by Underhill, ad H. 6. 4. 7. According to D it was χρησμολόγοι ἐγχώριοι who told Eparninondas about the maidens. ἐν Λεύκτροις: the original oracle perhaps concerned Laconian Leuctra and was adapted to Boeotian Leuctra after the event. 54. 2-3. The story of Scedasus and his outraged daughters is the central portent of the battle and it is given, with variations, by all the major sources: X. H.6. 4. 7; Plut. Pel. 20. 47-21. 1; Mor. 856f; Paus. 9. 13. 5f.; cf. ps.-Plut. Mor. 773b-774d; Aelian fr. 77; Westlake, 1939: 13. Burkert, 1979: 74f., argues the commonplace character of the theme: 'Girls falling into the hands of enemies, to be raped and killed, guarantee the victory, the "dispersion" of the hostile army'. The name Scedasus means 'the disperser'. Schachter, 1981-94: ii. 122. 54. 2. τω+̑ν Λεύκτρου καὶ Σκεδάσου θυγατέρων κτλ : this is contrary to the other sources. Two explanations present themselves. Either Ephorus knew of two versions of the story in the first of which the maidens were the daughters of Scedasus while in the second their father was called Leuctrus, and these he combined; he is known to have done this elsewhere; cf. his combination of the separate traditions (Herod. 1. 65) of Lycurgus' inspiration from Crete and -394-

Delphi ( FGH 70 F149). Or, D confused his source: if Ephorus spoke of the daughters of Scedasus as Plutarch does at Pel. 20. 5, ἃς Λευκτρίδας καλου+̑σι διὰ τὸν τόπον, then it may be

A Historical Commentary On Diodorus Siculus Book - 15 - PDFCOFFEE.COM (2024)