The Fragments of Sophocles, Volume 1 (Cambridge Library Collection - Classics)

Cambridge Library CoLLeCtion Books of enduring scholarly value

Classics From the Renaissance to the nineteenth century, Latin and Greek were compulsory subjects in almost all European universities, and most early modern scholars published their research and conducted international correspondence in Latin. Latin had continued in use in Western Europe long after the fall of the Roman empire as the lingua franca of the educated classes and of law, diplomacy, religion and university teaching. The flight of Greek scholars to the West after the fall of Constantinople in 1453 gave impetus to the study of ancient Greek literature and the Greek New Testament. Eventually, just as nineteenth-century reforms of university curricula were beginning to erode this ascendancy, developments in textual criticism and linguistic analysis, and new ways of studying ancient societies, especially archaeology, led to renewed enthusiasm for the Classics. This collection offers works of criticism, interpretation and synthesis by the outstanding scholars of the nineteenth century.

The Fragments of Sophocles Sir Richard Jebb (1841–1905) was the most distinguished classicist of his generation, a Fellow of Trinity College, Cambridge, and University Orator, subsequently Professor of Greek at Glasgow University and finally Regius Professor of Greek at Cambridge, and a Member of Parliament for the University. At his death, his planned volumes of the fragments of Sophocles, which would complete his edition of the complete plays and fragments, were not ready for publication, and the final editing of these three volumes was undertaken by W.G. Headlam and A.C. Pearson; the books were published in 1917. The first volume contains a general introduction; Volumes 1 and 2 present the text of the fragments and a commentary, and the final volume consists of addenda and corrigenda, spurious fragments and two indices. The plays are presented in Greek alphabetical order: Volume 1 contains fragments of plays from ‘Athamas’ to ‘Ichneutae’.

Cambridge University Press has long been a pioneer in the reissuing of out-of-print titles from its own backlist, producing digital reprints of books that are still sought after by scholars and students but could not be reprinted economically using traditional technology. The Cambridge Library Collection extends this activity to a wider range of books which are still of importance to researchers and professionals, either for the source material they contain, or as landmarks in the history of their academic discipline. Drawing from the world-renowned collections in the Cambridge University Library, and guided by the advice of experts in each subject area, Cambridge University Press is using state-of-the-art scanning machines in its own Printing House to capture the content of each book selected for inclusion. The files are processed to give a consistently clear, crisp image, and the books finished to the high quality standard for which the Press is recognised around the world. The latest print-on-demand technology ensures that the books will remain available indefinitely, and that orders for single or multiple copies can quickly be supplied. The Cambridge Library Collection will bring back to life books of enduring scholarly value (including out-of-copyright works originally issued by other publishers) across a wide range of disciplines in the humanities and social sciences and in science and technology.

The Fragments of Sophocles Volume 1 E di t e d by R ichard C l averhouse Jebb, W.G. Headl am and A.C. P earson

C A M b R I D G E U N I V E R SI T y P R E S S Cambridge, New york, Melbourne, Madrid, Cape Town, Singapore, São Paolo, Delhi, Dubai, Tokyo Published in the United States of America by Cambridge University Press, New york www.cambridge.org Information on this title: www.cambridge.org/9781108009867 © in this compilation Cambridge University Press 2009 This edition first published 1917 This digitally printed version 2009 ISbN 978-1-108-00986-7 Paperback This book reproduces the text of the original edition. The content and language reflect the beliefs, practices and terminology of their time, and have not been updated. Cambridge University Press wishes to make clear that the book, unless originally published by Cambridge, is not being republished by, in association or collaboration with, or with the endorsement or approval of, the original publisher or its successors in title.

THE

FRAGMENTS OF

SOPHOCLES

IN THREE VOLUMES VOLUME I

CAMBRIDGE

UNIVERSITY

PRESS

C. F. CLAY, MANAGEJJ HonSon: FETTER LANE, E.C. fEfitnfcurgfr:

ioo PRINCES STREET

u £ei» ffiotft: G. P. PUTNAM'S SONS JScmiag, (Mcutta anB lEatraa: MACMILLAN AND CO., LTD. Toronto: J. M. DENT AND SONS, LTD. ?TnftEo: THE MARUZEN-KABUSHIKI-KAISHA

Ail rights resetDed

THE

FRAGMENTS OF

SOPHOCLES EDITED WITH ADDITIONAL NOTES FROM THE PAPERS OF SIR R. C. JEBB AND DR W. G. HEADLAM

BY

A. C. PEARSON, M.A. FORMERLY SCHOLAR OF CHRIST'S COLLEGE, CAMBRIDGE

VOLUME I

Cambridge : at the University Press 1917

PREFACE

T

HE production of this book has been delayed by various causes, which require particular notice on the occasion of its appearance. It is well known that Sir Richard Jebb intended ultimately to include the Fragments in his edition of Sophocles; and in pursuance of this intention he delivered at Cambridge in the Michaelmas Term of 1895 a course of lectures on 132 selected fragments. The Ajax, the last to be published of the seven extant plays, appeared in the autumn of 1896; and it was then anticipated that the publication of the Fragments would be undertaken in due sequence. But the discovery of the Bacchylides papyrus drew the editor's attention in another direction, and, during the remainder of his life, the time which he could spare from public duties was mainly devoted to the preparation of a comprehensive edition of the Poems and Fragments of Bacchylides, which was published by the Cambridge University Press in 1905. Thus it fell out that, when after Sir Richard Jebb's death the task of completing the edition of Sophocles devolved upon Dr Walter Headlam, the material available for his use consisted solely of the notes prepared for the lectures already mentioned. Once again misfortune attended the prosecution of the scheme, in consequence of the premature death of Dr Headlam before he was able to put into shape the preliminary labour which for a number of months he had expended upon the text. Towards the end of 1908 I was entrusted by the Syndics of the University Press with the papers of both scholars, in order that the work so long deferred might be brought to a conclusion. I will frankly admit that, though conscious of having assumed

vi

PREFACE

a serious burden, I did not at first adequately realize either the magnitude or the difficulty of the task. I am afraid that, after these preliminary remarks, readers will be disappointed to find how small a share in the contents of these volumes has been contributed by my predecessors. Headlam, according to his wont, set to work thoroughly to explore the ground which he was preparing to develop, but he left very little evidence of the results at which he had arrived, and hardly anything in such a shape as could be adapted readily for publication. Yet even the adversaria of so eminent a scholar are of considerable interest, and not a few instances will be found where his insight has pointed out the way leading to the solution of a puzzling problem. Jebb's notes were of an entirely different character. Although well fitted to introduce to an undergraduate audience the salient features of some of the most interesting fragments, they were obviously unsuitable for reproduction as containing the matured judgement of their author upon the critical and exegetical questions which these fragments raise. They were chiefly the record of first impressions drawn up with the skill and taste which we have learnt to expect from such a source, but made without much exercise of independent research, or a full recognition of the departmental literature bearing upon the subject, so far as it was at that time accessible. To have printed any considerable portion of these notes would have been both misleading and unfair. Indeed, I am doubtful if I have not gone too far in including so much as will be found below; and it is with the greatest reluctance that I have in several cases quoted Jebb's notes, where I felt bound to argue in favour of a different conclusion. But my guiding principle has been this. The obscurity of the text of these fragments is so great, and so little has been done to dispel it, that we can only hope to arrive at the truth by a patient sifting of the clues suggested by competent authorities ; and an editor may often best recommend the solution which he considers probable by canvassing the views of other workers in the same field. Anyhow by this method the reader is the better enabled'to form his own judgement on the issues submitted to him : securus iudicat orbis terrarum. It will now be apparent that not only the responsibility for

PREFACE

vii

everything that appears in these volumes is entirely my own, but also the bulk of the commentary itself1. I must therefore explain the lines upon which I have worked. The general plan, modified only so far as was required by difference of subjectmatter, was prescribed by the character of the earlier volumes, and, although my predecessors had not advanced far in the appointed track, they had at least made it plain that the chief feature of the book should be a thorough and searching exegesis. Translation was less essential than in the complete plays and often impossible ; but in some of the longer fragments I am fortunate in being able to quote renderings made by Jebb and Headlam. In the elucidation of fragmentary and corrupt texts criticism and interpretation are complementary of each other. I have therefore endeavoured to present the critical data in as accurate a form as possible, taking Nauck's edition as my basis, and verifying, supplementing, and correcting its results so far as my opportunities permitted. It has not been possible for me to obtain unpublished information concerning the readings of the MSS of authors which have not been edited in accordance with the requirements of modern criticism ; but I have endeavoured to make myself acquainted with published results, although I cannot feel confident of having surveyed every part of so wide a field. In this respect not much has been done since the appearance of Nauck's second edition. It is true that editions of Stobaeus and Plutarch's Moralia, two of our most important sources, have been completed by Hense and Bernardakis. But Hense's results had been already communicated to Nauck, and the character of Bernardakis's edition is such that it is almost entirely useless for the present purpose. Our knowledge of the most important scholia is still imperfect, although progress has been made, especially in regard to Aristophanes and Pindar. Wendel's edition of the scholia to Theocritus appeared while this book was passing through the press. Much might be learnt from a critical edition of Eustathius, which is scarcely to be expected at present. But the lexicographers are the most 1 The letters J. and H. have been attached to the notes of Jebb and Headlam now first printed, and their full names are retained in references to their published writings.

viii

PREFACE

promising field of all, and, though a good deal of work has been done in sifting their records, very little of it has seen the light. Bethe's Pollux and de Stefani's Etymologicum Gudianum are both incomplete. Here too the recovery of fresh material from unedited sources which may be still preserved in the libraries of Europe has been shown to be more than a possibility by the labours of Reitzenstein, Rabe, and others. The actual increase of material that has accrued in the last twenty-five years is not completely measured by the fact that this edition contains almost exactly ioo more fragments than were published by Nauck. A considerable proportion of the accession comes from the recently discovered commencement of Photius, published by Reitzenstein in 1907. But the most important addition of all was of course the fragments of the Ichneutae and Eurypylus contained in the ninth volume of the Oxyrhynchus Papyri. The discovery was made at a time when the greater part of the present commentary was written, and I am glad to be able now to express my thanks to Prof. A. S. Hunt, who was kind enough to allow me to inspect the sheets of the new fragments before publication, and has more than once replied to my queries concerning the actual readings of the MS in doubtful cases. I must also acknowledge my indebtedness to the Committee of the Egypt Exploration Fund and the Delegates of the Clarendon Press for permission to include the Oxyrhynchus fragments in the pages of this edition. In the General Introduction I have endeavoured to describe the literary history of Sophoclean tragedy, to estimate the extent and variety of its activity, to discover the vestiges of the material with which it worked, and to show how its monuments were transmitted to posterity until they passed into oblivion and how finally its scanty relics were preserved for the instruction of our own times. In this way I have tried to answer the questions, why the majority of the plays were lost, and by what means their fragments survived. It will be evident that the third section follows in the main the lines which have been sketched in various writings by Prof, von Wilamowitz-Moellendorff. In dealing with the sources of our existing fragments I have entered at some length into the history of Greek philological literature

PREFA CE

ix

during the Roman and Byzantine ages, confining myself particularly to its connexion with the study of Sophocles. This is an arid region ; yet I am convinced that those who make themselves acquainted with its chief features will return not only with a clearer understanding of the limits within which the criticism of the Greek poets must proceed, but also with a deeper respect for the honest labours of generations of workers who struggled against the forces of barbarism to keep alive the purity of the classic speech. So much misconception prevails as to the significance of quotations made by these writers that no apology is needed for the space which has been devoted to them. The only work of reference in English which touches this branch of literature is Sir J. E. Sandys's History of Classical Scholarship. Shortly after the printing had commenced, it was decided to take advantage of the occasion by the preparation of a comprehensive index to the whole of the ten volumes. For this purpose Prof. Jebb's seven volumes have been carefully re-read, the old indexes have been consolidated, corrected, and considerably enlarged, and the entries so collected have been incorporated with those relating to the three volumes of Fragments. The work was at first undertaken by Mr G. V. Carey of Gonville and Caius College, who re-indexed the Ajax and Antigone; but, when he obtained a commission in the Army on the outbreak of the European War, the responsibility for the remaining portions passed into my hands. -It is hoped that the new indexes will be of service to students not only as a better means of access to the information which the volumes contain, but also as a register of Sophoclean usage for anyone who may attempt further researches in the sphere of tragic vocabulary and grammar. At the same time their users should be warned that they do not pretend to be anything more than a record of the material comprised in the commentaries ; for an attempt to provide by this means a complete digest of the language would have involved an enormous addition to a labour which was already sufficiently arduous. I have elsewhere discussed and tabulated the researches of those modern scholars who since the close of the eighteenth century have laboured directly on the fragments of Sophocles, «5

x

PREFACE

and it is unnecessary to repeat here the nature of my obligations to them. Most of this literature is scattered in various periodicals or contained in dissertations which are even more difficult of access. In this connexion my thanks are due to Prof. R. Reitzenstein of Freiburg for supplying me with information respecting the contents of one of his dissertations which I had been unable to procure. Nor must I forget to mention the singular kindness of the late Dr Siegfried Mekler of Vienna, the editor of Dindorf's Sophocles in the Teubner series, who, hearing that I was engaged on this work, sent me a number of notes bearing on various points of difficulty. This will explain the occasional references to Mekler's unpublished views. Dr J. B. Pearson and Mr R. D. Hicks have kindly permitted me to print extracts from certain notes formerly communicated to Prof. Jebb in reference to frs. Jj6 and 1128. Notwithstanding the considerable output of labour directed to the criticism of the fragments, the attention which they have received is scanty in comparison with the mass of comment which has accumulated upon the extant plays. Hence I have been often compelled to rely largely on my own resources. This is, in fact, the first systematic effort that has been made to put together a continuous commentary, though I have the best of reasons for knowing that its imperfections are not due to that cause alone. I must warn readers that the printing of the book was seriously delayed by the stress of recent events and that it went to the press at the beginning of 1913. Everyone knows the difficulties and inconsistencies that are apt to occur in such cases, and that they cannot be satisfactorily cured by the list of corrigenda. I desire to acknowledge the generous support which I have throughout received from the Syndics of the Cambridge University Press, although the work has grown to a size which neither they nor I contemplated at the time of its inception. A. C. P. February, 1916.

CONTENTS OF VOLUME I PAGES

PREFACE

.

.

.

.

.

GENERAL INTRODUCTION

.

.

.

.

.

v—x .

§ i.

The number of the plays

.

.

xiii—xxii

§ 2.

The subjects of the plays

.

.

xxii—xxxii

§ 3.

The tradition of the text

.

.

xxxii—xlvi

§ 4.

The sources of the fragments

.

xlvi—xci

§ 5.

Bibliography .

.

.

.

.

xci—c

FRAGMENTS OF NAMED PLAYS :

Introductions, text and notes

.

.

1—270

GENERAL INTRODUCTION § I.

The number of the plays.

T H E anonymous Life of Sophocles1 records on the authority External of Aristophanes of Byzantium that 1302 plays were attributed3 evldenceto Sophocles, but that 17 of these were spurious. The statement is entitled to credit, as coming from Aristophanes ; and it has been referred with high probability to his work entitled Trpo? Toi>? KaWi/^d^ov -nivaica^. Not much is known of the book in question, but it may be taken to have contained corrections and enlargements of the well-known TrLvaice.*; of Callimachus, which was not merely a catalogue of the books contained in the Alexandrian library, but included biographical details concerning the various authors, and in the case of the Attic drama the dates of the production of the several plays, as well as other points of interest drawn from the StSaa-KaXiai of Aristotle5. Suidas, however, reports that Sophocles produced 123 plays, and according to some authorities considerably more. This information may be reconciled with the Life in two ways, i.e. by the adoption either of Boeckh's6 correction of Suidas, which makes the total 113 (piy' in place of picy'), or of Bergk's7 1

xi p . liv Bl. ^X€i $£ dpafmTa, ci's (prjaip 'Apii.yevelq. rrj iv A6\l6i (i.e. Eur. I.A. 993). Hence frs. 583, 769 and 941 have been assigned by some to Euripides. 2 Schol. Pind. Pyth. 5. 35, where however Schroeder suspects that a reference to the KoXx'Scs (fr. 340) has fallen out. 3 See 1 p. 213, 11 p. 185.

THE NUMBER OF THE PLA YS

xxi

to Aristophanes of Byzantium we are still able to identify about 112. Of all these there is, so far as I can see, only one, the Iberes1, of which it might be thought that it no longer existed in the Alexandrian epoch; and even of it we can only say that there is no positive indication of its survival. It has already been remarked that we have no record of the number of Sophoclean plays which were preserved in the Alexandrian library. Now, if Boeckh's hypothesis2 were correct, it would follow that we are still able to trace practically all the genuine plays as having passed into the keeping of the Alexandrians. But it is in the highest degree improbable that copies of every one of them survived throughout the interval between the fifth and third centuries. On the other hand, if we accept 123 as the actual total of the genuine titles, we are now in a position to say that some n o of the plays to which they belonged were known to the students of Alexandria. It is reasonable to infer that there are very few indeed3 of which Alexandria has left us no trace, and the result is a very remarkable testimony to the accuracy and comprehensiveness of our sources. The information available respecting the satyr-plays is not Satyrsuch as to disturb the previous calculation. There are sixteen p ays' plays universally admitted or strictly proved to be satyric4. To these we need not hesitate to add Aa/SaXo? and 'Hpaic\et
"Tfipis,

'TSpotfiopoi..

If the number 112 is accepted as a probable total of the plays comprised in the preceding list, it will be noticed that 43 of them, or over 38 per cent., belong to the Trojan Cycle, A similar calculation applied to the plays of Aeschylus and Euripides yields percentages of 23 and 21 respectively. These remarkable figures entirely confirm the evidence relating to Sophocles' Homeric proclivities ; and if the limits were enlarged so as to include the plays whose subjects lie on the borders of 1 This play should in strictness have been associated with class V: cf. Apollod. 3. 150. But it would be inconvenient to separate it from the other Trojan plays. "- The subject of the play is extremely doubtful: see II p. 325.

xxxii

GENERAL INTRODUCTION

the Homeric domain, the result would be even more striking. Want of information concerning the character and extent of the KV/CKO? prevents a closer enquiry. § 3. The tradition of the text. Tradition The seven plays which still survive have been handed down extant to u s m a number of MSS ranging from the eleventh to the plays. sixteenth century, of which the oldest and best is the well-known Laurentian, written in the first half of the eleventh century. But, whereas twelve MSS contain all the seven plays, either complete or with lacunae, and fifteen others four or more but less than seven, no less than seventy are restricted to the Ajax, Electra, and Oedipus Tyrannus alone, or to one or two of them1. The preponderance of the three plays is readily explained by their exclusive use for educational purposes during the Byzantine period. The existence of this selection may perhaps be traced as far back as to 500 A.D., if we may judge from the title of a treatise written by Eugenius, head of the imperial school at Constantinople under Anastasius I (491—518) and predecessor of Stephen of Byzantium in the tenure of that office2. The work was entitled KoKofieTpia rSiv fieXiicwp A.io-%v\ov ^otpoKXeovf icai RvpnrlSov a-jrb Bpafidroav ie'3. That is to say, Eugenius, no

doubt following earlier scholars, published analyses of the lyrical parts of the three tragedians similar to those which Heliodorus constructed for Aristophanes, but limited his activity to fifteen plays, three of Aeschylus, three of Sophocles, and nine of Euripides4. The number chosen corresponds to that of the plays selected from Aeschylus {Prometheus, Seven, and Persae), as well as of those taken later from Euripides {Hecuba, Orestes, and Phoenissae); and each of these groups consists of the first three plays according to the order of an earlier and larger collection. The history of the Euripidean tradition is more complicated, and does not concern us here, but the earlier collections made from 1

Thefiguresare taken from Jebb's text-edition of Sophocles, p. x m f. 3 Steph. Byz. p. 93, 1. Suid. s.v. Ei^vios. 4 So Colin in Pauly-Wissowa vi 987; Christ-Schmid, op. cit. 115 p. 879; Sandys, Hist. Cl. Schol. I 402; and C. H. Moore in C.R. xix 12. Wilamowitz, Einleitung, p- I97i62> thinks the statement obscure. 2

THE TRADITION OF THE TEXT

xxxiii

the works of Aeschylus and Sophocles undoubtedly comprised the fourteen plays which still exist. It must be recognized therefore that the survival of particular plays is due not to the accidental preservation of this or that MS from which all other copies were derived, but rather to the educational needs which prompted scholars to prepare annotated editions of select plays for the use of schools. Of the circumstances attending the publication of the earlier collection hardly a trace remains. An Argument is attached to each of the two plays Oedipus Coloneus and Antigone bearing the name of a certain Sallustius, and in the former he is described as Sallustius Pythagoras. In all probability therefore there was a tradition identifying him with Sallustius the Pythagorean1, the follower of Iamblichus and author of the treatise Trepl deSiv ical /coo-fiov2, who belonged to the latter part of the fourth century. This writer is probably the same man as the friend of the emperor Julian, and has also been identified with the sophist Sallustius to whom Suidas ascribes commentaries on Demosthenes and Herodotus3. Now, if Sallustius, the editor of the select plays, lived at so late a date as the second half of the fourth century, it is unlikely that he was the first compiler of the selection4. It will be shown later that the direct quotation of tragedies other than those contained in the select edition died out at the end of the second century. Further, it may be inferred, from the precise correspondence in subject between the plays chosen from each of the three great tragedians which contain the stories of Oedipus and Orestes, that the whole selection was made by a single person. Beyond this it is impossible to determine either author or date. WilamoTr

1

The title of the Argument is given in L as (TaKovcrrtov v irvBa/ybpov. Dindorf conjectured that irvdaySpov was the blunder of someone who did not perceive that the abbreviation represented inr66e<ns, but the view of Wilamowitz as stated above is preferable. F . Cumont, Rev. de Philol. XVI 53, rejecting the identification, points out that Sallustius was a Neo-Platonist. 2 For this work and its author see Gilbert Murray, Four Stages of Greek Religion, p. 163 ff. 3 A grammarian Sallust is quoted in schol. Ar. Plut. 725. Cumont assigns him to the sixth century. 4 Wilamowitz, op. cit. p. 199. Pius, a commentator on Sophocles, who is assigned to the second century (Christ-Schmid, op. cit. p. 345), is mentioned in schol. Ai. 408.

xxxiv

GENERAL INTRODUCTION

witz, who has once for all laid down the conditions of the problem, arrived at the conclusion that the selection was first made in the age of Plutarch, and after the lapse of a century secured universal acceptance1. We shall now proceed to give some account of the Sophoclean tradition in antiquity in order to prepare the way for an examination of the sources from which our knowledge of the lost plays Athens to is derived. The existence of written copies of Attic tragedies andria. c a n be traced back to the period of their production, that is to say, to a date not later than the close of the fifth century B.C., as is proved by the well-known passage in Ar. Ran. 52 where Dionysus speaks of reading to himself the Andromeda of Euripides ; and the force of v. 1114 of the same play is very much impaired if we do not understand it as implying that the text of the tragedians was studied in literary circles. The learning by heart of tragic pifuet? is mentioned by Plato2, and Alexis includes tragedies in a list of books which are recommended for the improvement of Heracles, who however chooses a cookerybook in preference to all of them3. The earliest recognition of the necessity for maintaining the integrity of the tragic texts is to be found in the law of Lycurgus the orator, which required that an official copy of the plays of Aeschylus, Sophocles, and Euripides should be preserved in the archives, and that in future performances the actors should adhere to the text of this copy4. It has been suggested that the official text only contained such plays as still kept the stage in the fourth century, but, as the plays of Aeschylus were seldom reproduced at that time6, it was probably more comprehensive. In fact, if such an official copy was ever made, although its primary purpose was not so much directly critical as to check the licence of the actors, it may be presumed that the net was cast as wide as possible, and that the most authoritative sources were consulted6. It is 1

Wilamowitz, op. cit. p. 202. 3 legg- 811 A : cf. Herond. 3. 30. fr. t 3 5 , 11 345 K. 4 Plut. vit. X orat. p. 841 F. The documents were no doubt preserved in the Metroum (Frazer's Pausanias, 11 p. 68). 5 Haigh, Attic Theatre*, p.. 76. 6 Wilamowitz, op. cit. p. 131, hardly allows so much. But his view that the 2

THE TRADITION OF THE TEXT

xxxv

generally supposed that this was the copy subsequently borrowed by Ptolemy Euergetes, when he left a deposit of fifteen talents as security for its return. His professed object was to make a transcript for his own use; but, when this had been done, he sent back the transcript to Athens in place of the original, and the Athenians were obliged to content themselves with a forfeit of the deposit1. It may be open to doubt whether the reference is to Euergetes I (247—221 B.C.) or to Euergetes II Physcon (146—117 B.C.)'2, although the former is generally preferred ; but the historical truth of the story is of less importance than the inferences to which it leads. Thus the acquisition of the official copy was certainly not regarded as having settled the text of the tragedians, for otherwise the conjectures of Aristophanes would riot have been recorded in our scholia. A still more important fact to which the story testifies is the migration of tragedy in the third century from Athens to Alexandria3, both as the home of the Alexandrian Pleiad, and as the place where the study of the old tragedians was pursued with the greatest zeal. It is to Alexandria that we owe our existing texts, and almost the whole of the information that can be recovered concerning the lost plays. Aesthetic and historical criticism of the tragedians had been Peripapursued almost exclusively by the Peripatetics in pre-Alexandrian tetlcstimes. With the former we are not immediately concerned, but on the historical side the publication of Aristotle's StSaa/caXiai was of considerable importance. This work was a collection of extracts from the archives giving the dates and circumstances of production of all the tragedies and comedies recorded in the official lists. The particulars which the archon registered were the names of the competing poets and their plays, of the choregi, and of the leading actors, and the order in which the competitors were placed by the judges. The concluding sentences of the ordinance was probably ineffective is to be preferred to Rutherford's contention that the story is inconsistent with the references to actors' readings in the scholia to Euripides {Annotation, pp. 57-60). 1 Galen in Hippocr. epidem. Ill 1 (xvn 1. 607 K.). 2 Sandys, History of Classical Scholarship, I p. 58. The later date is advocated by Usener in Susemihl, op. cit. II 667, but see ibid. p. 682. 3 Haigh, Tragic Drama, p. 439 ff.

xxxvi

GENERAL INTRODUCTION

Argument to the Agamemnon of Aeschylus had been recognized as a fragment of Aristotle's treatise1, and a remarkable confirmation of the accuracy of our tradition came to light in 1886, when an inscription was discovered on the Acropolis, recording the production of the Orestea, and agreeing exactly in the facts which it mentions with the text of the Medicean MS2. But the stoneinscription was not the original record; for, so far as the contests listed were earlier than Aristotle's time, the details to be inscribed were taken from his book or were otherwise due to his researches3. Aristotle also wrote a treatise in one book entitled irepl TparytpSiwv, and another also in one book entitled viicai, k.iowcriaical4'. Of the former nothing whatever is known, and there is no probability in Mueller's view6 that it was related to the Bi&ava>v, contained various items of information relating to the history of the Attic stage. Of less account is Hieronymus of Rhodes (c. 290—230 B.C.), another Peripatetic, who is quoted once or twice by Athenaeus as the source of certain anecdotes relating to Sophocles1, and may be compared with Dicaearchus as having been responsible for the statement that the plot of Euripides' Phoenix was drawn from the annals of a village community2. Passing to Alexandria, we find that at an early date (c. 285 B.C.) AlexZenodotus, the first librarian, shared with Lycophron and Alex- studies. ander Aetolus the task of putting in order the books in the library, and that to Alexander was assigned the special duty of superintending the arrangement of the tragedies and satyrplays3. Callimachus, who succeeded Zenodotus, completed the catalogue which his predecessor had begun and published it in

I2O books under the title vlvaices T&V ev irday iratZeia BiaXafiyfrdvTcov ical WV gweypayp-av. This celebrated work was more than a catalogue, since it contained biographical and other details of literary history, and, in the case of the dramatic writers, notices drawn from Aristotle's SiSaaKaXiai11 relating to the production of their plays at the Dionysia. Eratosthenes {c. 245 B.C.) confined his studies in the Attic drama to the production of a work on comedy (irepl ap^ala*; Kwfiwhias), but Aristophanes of Byzantium, who became chief librarian on the death of Eratosthenes (195 B.C.), was the first critic who laboured continuously on the text of the tragedians, and by his investigations laid a secure foundation for the benefit of later generations. We have already had occasion to refer to his work 71730? -roti? KaWifidxov irivaicas, which seems to have contained corrections of and additions to the treatise of Callimachus5. But his influence in the sphere of textual criticism was of much greater importance. There is no doubt whatever that he edited Euripides, for the allusions to him in the scholia do not admit of any other ex1

FHG II 450 n. See also vit. Soph. vi. His book was entitled irepl TroiijTav. 3 - TGFp. 621. Knaack in Pauly-Wissowa I 1447. 4 Schol. Ar. Nub. 552. For the whole subject see O. Schneider, Callimachea, 5 n 297 ff. Athen. 408 F. P. S.

C

xxxviii

GENERAL INTRODUCTION

planation. There is no similar evidence to prove that he was also responsible for editions of Aeschylus and Sophocles ; but, since no other assumption accounts equally well for the existence of the vnroO&aei'i attributed to him which are attached to plays of all three tragedians1, recent scholars have been unanimous in so concluding2. It may be added that the discovery of the Ichnentae papyrus, with marginal variants attributed to Aristophanes3, makes strongly in the same direction. Wilamowitz argued that the virodeaeL? were not accompanied by a commentary on the text, and that the edition of Aristophanes was intended rather for the general reader than for scholars4. However this may be, there is evidence that he wrote a vTr6/j,vrj/u.a— or what we should call ' lecture-notes'—on the Orestes, as well as on other plays 6 ; and much of the aesthetic criticism which is found in the scholia to Sophocles and Euripides has been attributed to him. We must not forget his lexicographical studies, plentiful remains of which are to be found in Eustathius, and less patently in Hesychius, Pollux, and Athenaeus, not to mention the excerpts still existing in medieval MSS which have been published by Boissonade and E. Miller6. These studies were entitled Xe^ei?7 or yXwaaai, and were divided into two main classes: (i) collections of dialectical variants, (2) varieties of subject-matter (e.g. irepl 6rofiaa[ai)/M quoted by schol. L on 451, 488. 2 Sdiol. cod. Barocc. At. 283. R. Schmidt, de Callistralo, p. 324. 3 Etym. M. p. 177, 55. 4 frs. 449, 624, 728. 5 Hecker introduced his name by emendation in schol. 0. C. 100.

xl

GENERAL

INTRODUCTION

His tragedies were for the most part adaptations of Euripides, and he was followed by Pacuvius (219—129) and Accius (170— 105), who constructed their plays with greater skill and included Sophocles among their models. Unfortunately the fragments of these writers are so scanty that they are very seldom of service in the reconstruction of the Greek originals1. The Romans were thus familiarized through the stage with the form and contents of Greek tragedy before they were trained to study them as literature. But after 146 B.C. the assimilation of Greek culture spread rapidly. That which was at first the exclusive possession of the Scipionic circle became in the next generation the common heritage of every educated Roman. The everincreasing demand for instruction brought the learning of Pergamum and Alexandria into contact with the ruling class of the imperial city. Educational requirements not only gave an enormous stimulus to the multiplication of copies of the most famous Greek authors, but grammar and criticism themselves were internationalized. Alexandria ceased to be the home of the most learned professors, and gradually lost its supremacy in the world of letters, although, as the birth-place of Didymus, Herodian, Harpocration, and many others, it continued for three centuries to preserve its reputation as the ultimate source of philological erudition. It is not surprising, in view of these circumstances, that the name of Didymus, who, after the lapse of more than a century, comes next on the list of Sophoclean editors, should be associated with a change in method calculated to adapt his lucubrations to the requirements of the Roman world. Didymus. Didymus is the most important name in our survey, not so much in consequence of his individual merit, although this has perhaps been undeservedly belittled, as because we owe to him more than to any other single person the preservation of such fragmentary knowledge as we possess respecting the lost plays of the Greek tragedians. The extraordinary industry of Didymus, which earned for him the epithet ^aX/cevTepo?, may be estimated by his performance in the field of literary criticism alone, in 1 The leading authority on the plots of the Roman tragedians is O. Ribbeck, die Romische Tragodie, Leipzig, 1875.

THE TRADITION

OF THE TEXT

xli

which he undertook the interpretation of Homer, Hesiod, Pindar and Bacchylides, the tragedians, Aristophanes and other comic poets, and the Attic orators. The importance of his collections to modern research becomes apparent when we learn that large portions of the existing scholia to Pindar, Euripides, and Aristophanes are drawn from the commentaries of Didymus, and that his writings are the ultimate source of the scholia to Sophocles. It is significant that in these, while the names of the older Alexandrian grammarians are scarcely mentioned or else are replaced by such general descriptions as ' the commentators1,' that of Didymus occurs at least nine times2. These commentaries were not simply virofj.vij/iaTa in the sense previously indicated, but were accompanied by a text3. The older grammarians had lectured to their pupils from a plain text, but conditions were now very different, and the wider public for whom the editions of Didymus were intended could not dispense with explanatory notes. The function of Didymus should not be misconceived. He was neither an original thinker nor an independent investigator : his province was to collect the results garnered by earlier scholars, and to make them serviceable to the needs of his contemporaries and his successors. It has been inferred that these books were the prototypes of the class afterwards represented by the medieval MSS, in which the margins surrounding the text are occupied with exegetical comment and critical variants4. Recent discoveries of papyri, which have been sufficiently numerous to familiarize us with the form and aspect of the papyrus roll, have only partially confirmed this conclusion. It is true that critical and explanatory notes are found in them, 1 See Cohn in Pauly-Wissowa v 452. There is a good instance in schol. Ant. 45, which shows that they were anterior to Didymus. 2 For Didymus as an interpreter of Sophocles see also fr. 718. Etytn. Gud. p. 81, 37 (Track. 1054) is to be added to the passages in which Didymus is referred to by name. 3 This is proved by schol. 0. C. 237, Ai. 1-225. The account given above follows Wilamovvitz, p. 166. He finds in the scholia to the O. C, besides the work of Didymus, traces of a i-irtuwqiia devoted chiefly to the explanation of antiquarian and mythological details. Cohn, however, is inclined to refer this part also to Didymus. 4 Wilamowitz, I.e. The same opinion is maintained in 'Die griech. Literatur des Altertums' in Kultur d. Gegenwarl, I viii2 (1907) p. 96. See also Susemihl, II 201,

xlii

GENERAL INTRODUCTION

and particularly in the Paris fragments of the Partheneion of Alcman, which are ascribed to the first century A.D., and the Oxyrhynchus papyrus of the Paeans of Pindar, which belongs to the early part of the second. But they are very scantily represented in the Hypsipyle and Ichneutae papyri, and there are obvious reasons why a continuous commentary was less suitable for inclusion in the papyrus roll than in the parchment book of a later age. It follows that the existing scholia, although in substance based upon the results of Alexandrian learning, are not formally and directly the completion of an original Alexandrian nucleus1. The growth of the various collections cannot be traced in detail, since for the most part the secure support of names and dates is wanting2. A glance at Nauck's Index of Sources will show that many of the fragments are quoted in the scholia to the writers enumerated above, but a still greater number is derived from the lexicographical labours of Didymus. The prefatory letter to Eulogius, which Hesychius placed in the forefront of his lexicon, refers to the separate vocabularies of comic and tragic diction (X