THE TANGLED WAYS OF ZEUS
This page intentionally left blank
The Tangled Ways of Zeus and other studies in and around Greek Tragedy
ALAN H. SOMMERSTEIN
1
3
Great Clarendon Street, Oxford ox2 6dp Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide in Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Taipei Toronto With offices in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan Poland Portugal Singapore South Korea Switzerland Thailand Turkey Ukraine Vietnam Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries Published in the United States by Oxford University Press Inc., New York # Alan H. Sommerstein 2010 The moral rights of the author have been asserted Database right Oxford University Press (maker) First published 2010 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this book in any other binding or cover and you must impose the same condition on any acquirer British Library Cataloguing in Publication Data Data available Library of Congress Cataloging in Publication Data Data available Typeset by SPI Publisher Services, Pondicherry, India Printed in Great Britain on acid-free paper by the MPG Books, Bodmin and King’s Lynn ISBN 978–0–19–956831–4 1 3 5 7 9 10 8 6 4 2
Preface This book is the ‘further volume mainly consisting of articles on tragedy’ that was foreshadowed in the preface to my previous collection Talking about Laughter and other studies in Greek Comedy (Oxford, 2009). The ‘mainly’, and the ‘in and around’ in the title of the present volume, refer to Chapters 1 and 2, which are concerned with issues relevant to all varieties of Greek drama. Most of the essays in this volume, as in the previous one, have been published before—all of them, like all but one of those in the previous volume, within the last twenty years. With one exception, as before, I have excluded papers which first appeared in books or periodicals likely to be widely accessible in academic libraries in the English-speaking world, or in e-journals accessible without subscription; the exception, included at the special request of OUP’s editor and referee, is Chapter 7. Five of the chapters have not appeared previously: Chapters 3 and 17 are papers presented at Classical Association conferences, Chapter 4 is the Gaisford Lecture which I had the honour of delivering at Oxford in 2007, Chapter 15 is a semi-popular lecture, and Chapter 18 was to have been delivered at a London meeting of the Society for the Promotion of Hellenic Studies (see the final note to that chapter). All papers appear here essentially as they were originally published or presented, with the original pagination (if any) inserted (in square brackets and in boldface) at the point where each new page of the original began (in Ch. 8 the same has been done in the footnotes, since these were originally endnotes). In general, the only changes in the text have been those of style, referencing, brief explanatory additions, and the correction of errors that should have been spotted in proof, specified in the preface to Talking about Laughter; but those chapters which have not previously appeared in print (3, 4, 15, 17, and 18) or which appeared in publications where the usual scholarly apparatus would have been inappropriate (9, 12, and 19) have been provided with the footnotes which they would have had originally if that had been possible. Most of the chapters are followed by an
vi
Preface
addenda section drawing attention to significant subsequent developments in the relevant scholarship, or to points where my own thinking has changed since the time of the original paper. For a second time I express my gratitude to Hilary O’Shea and her colleagues at Oxford University Press; to the British Academy, the Society for the Promotion of Hellenic Studies, and especially the University of Nottingham, for the support they have given me in attending many conferences; and to those who invited me to write or present so many of these papers—Guillermo De Santis, H. D. Jocelyn, Ben Jourdain, Manfred Lossau, Benedetto Marzullo, Pantelis Michelakis, Duncan Murdoch, Christopher Pelling, David Stuttard, Isabelle Torrance, and (once again) Bernhard Zimmermann. alan h. sommerstein Nottingham
Contents Abbreviations
ix
Introduction
1
1. The titles of Greek dramas
11
2. Violence in Greek drama
30
3. Adolescence, ephebeia, and Athenian drama
47
4. Sherlockismus and the study of fragmentary tragedies
61
5. The seniority of Polyneikes in Aeschylus’ Seven
82
6. The beginning and the end of Aeschylus’ Danaid trilogy
89
7. The theatre audience, the Demos, and the Suppliants of Aeschylus
118
8. Sleeping safe in our beds: stasis, assassination, and the Oresteia
143
9. The tangled ways of Zeus
164
10. The omen of Aulis or the omen of Argos?
171
11. Pathos and mathos before Zeus
178
12. Oresteia Act II: two misconceptions
189
13. Aeschylus’ epitaph
195
14. Dearest Haimon
202
15. ‘They all knew how it was going to end’: tragedy, myth, and the spectator
209
16. Alternative scenarios in Sophocles’ Electra
224
17. Sophocles’ Palamedes and Nauplius plays: no trilogy here
250
viii
Contents
18. ‘The rugged Pyrrhus’: the son of Achilles in tragedy
259
19. What ought the Thebans to have done?
277
References Index locorum General index
288 307 331
Abbreviations This list includes only abbreviations not listed in the Liddell-Scott-Jones Lexicon (Oxford, 1940), in its Revised Supplement (Oxford, 1996), or in the revised third edition of the Oxford Classical Dictionary (Oxford, 2003). In some cases abbreviations of ancient author names or text titles may be less drastic than those used in these works, or may differ from them owing to transliteration conventions (e.g. by using k rather than c). Where the number of a fragment, hypothesis, etc., is followed by the name of a scholar, the reference is to that scholar’s edition of the author or text in question; where there might be ambiguity about the meaning of such a reference, the editor’s name is listed below. AJP CA CAG Davies Fowler h.Herm. HSCP Hypoth. IGUR
YCS
American Journal of Philology Classical Antiquity Commentaria in Aristotelem Graeca (Berlin, 1882–1909) M. Davies, Epicorum Graecorum Fragmenta (Go¨ttingen, 1988) R. L. Fowler, Early Greek Mythography, i. Texts (Oxford, 2000) Homeric Hymn to Hermes Harvard Studies in Classical Philology Hypothesis (headnote to, or synopsis of, a literary text) Inscriptiones Graecae Urbis Romae, ed. L. Moretti (Rome, 1968–90) C. Mu¨ller, Fragmenta Historicorum Graecorum (Paris, 1841–84) Nachrichten der Akademie der Wissenschaften in Go¨ttingen, Philologisch-historische Klasse Parola del Passato Prolegomena Quaderni di Storia Rheinisches Museum fu¨r Philologie Seminari Romani di Cultura Greca University of California Publications in Classical Philology (for epic) M. L. West, Greek Epic Fragments (Cambridge, Mass., 2003) (for iambus and elegy) M. L. West, Iambi et Elegi Graeci (Oxford, 1971–2) Yale Classical Studies
#
scholium number
Mu¨ller NAWG PdP Proleg. QSt RhM SemRom UCPCP West
This page intentionally left blank
Introduction As in the introduction to my previous volume of selected papers Talking about Laughter (Sommerstein 2009), I will first briefly explain the organization of the book, and then attempt to define a few themes that surface frequently both in the chapters that follow and in my other work on tragedy. Like Talking about Laughter, this book is arranged so as to proceed from the general to the particular. Chapters 1 and 2 deal with issues that are relevant to all varieties of Greek drama. Chapter 1 is concerned with the titles which appear at the beginning of all surviving ancient and medieval copies of Greek dramatic texts (if the beginning of the copy is preserved at all) and which ancient authors, from Aristophanes onwards, regularly use to refer to them: are they authentic? And if so, were they designed to give the first audience information (or maybe sometimes misinformation) about the subject or themes of the play? Chapter 2 is about onstage violence; it seeks to establish exactly what forms of it were and were not considered acceptable in the various genres of Athenian drama, and what were the reasons—religious, aesthetic, practical, or other—which underlay these rules or conventions. In both cases the answers turn out to be complex, varying with time, genre, and, so far as violence is concerned, with type of violence. Chapter 3 attempts not so much to debunk as to amend a commonplace of much recent study of Athenian drama—the association often drawn between certain kinds of tragic1 plot and rituals 1 Initiatory and ephebic themes have also sometimes been detected in Old Comedy (e.g. Thiercy 1986: 305–27; Bowie 1993; Lada-Richards 1999); on this, see ch. 9 of Talking about Laughter.
2
Introduction
(religious and/or civic) associated with the transition of the adolescent male to the status of adult citizen and soldier; it argues that no recognized set of such rituals existed in fifth-century Athens, and that tragedy’s undoubted interest in the problems of the adolescent male is a reflection not of ritual but of life itself.2 Chapter 4, though as a methodological study it is relevant to fragmentary tragedies by any author, takes most of its specific examples from Aeschylus. This is in part due to the fact that I was working on the fragments volume of the Loeb Aeschylus3 at the time when I was preparing the lecture on which this chapter is based; but in part it is simply because Aeschylus has always been my favourite tragic dramatist.4 I do not find it much easier than did the Aristophanic Dionysus5 to provide an explanation of my preference, unless it is that Aeschylus gives me the impression6 of having strong and clear convictions and communicating them effectively while never forgetting that he is a dramatist and that his characters must speak for themselves, not for him. This chapter leads on to a series of nine further Aeschylean chapters (Chs. 5 to 13)—one on Seven against Thebes, two on the Danaid trilogy, five on the Oresteia, and one on an epigram which, so we are told, Aeschylus himself composed to be inscribed over his own tomb. Sophocles is represented by five chapters (Chs. 14 to 18), the first three of which deal in different ways with perhaps the three most famous of his surviving plays, Antigone, Oedipus the King, and Electra, while Chapter 17 is concerned with a group of fragmentary plays having to do with two great crimes committed by the Greek army at Troy—the judicial murder of Palamedes and the sacrilegious seizure of Cassandra. Chapter 18 begins and ends with Philoctetes, 2 I shall be returning to this theme, with a particular focus on the late plays of Sophocles and Euripides, in Sommerstein (forthcoming e). 3 Sommerstein (2008) vol. iii. 4 Counting both solely and jointly authored publications, and including forthcoming items, I have produced altogether, on Aeschylus, six books and twenty-two articles; on Sophocles, three books and nine articles (including Chs. 15, 17, and 18 of this volume, published here for the first time); on Euripides, five articles. 5 Who had, or at least gave, no reason for choosing Aeschylus rather than Euripides to bring back with him from Hades to earth except that he is the one ‘whom my soul wishes to choose’ (Ar. Frogs 1468). 6 Others find otherwise (Goldhill 1984, for one).
Introduction
3
but its actual subject is Sophocles’ portrayal, or rather portrayals, of Neoptolemus, the character whom he has introduced—almost certainly for the first time ever—into the story of how Philoctetes was made to come from Lemnos to Troy, and a character for whom it seems likely that neither Aeschylus nor Euripides, in a combined output of more than 150 dramas, ever created a role for an actor to play. The book ends with a short piece on Euripides’ Bacchae (Ch. 19), which asks a question that Athenians must often have asked themselves towards the end of a tragic drama: how, if at all, could the characters have averted the catastrophe that has befallen them—and could they reasonably have been expected to do so? Throughout the book the reader may notice a tendency to focus on ‘hard-edged’ issues and on questions to which there is in principle a definite answer, or would be if we had sufficient evidence. This reflects my belief that the study of ancient, or any other, literature is a branch of history.7 A literary work8 is a human artefact, designed and created by a particular person or persons, over a particular period of time9 in the past, in a particular cultural milieu or series of milieus; and the main aim of literary scholarship as I pursue it10 is, for each such product, to establish, so far as it is possible to do so, 1. its form and content, both as regards verbal and any non-verbal elements; 2. by whom, over what period of time, and against what cultural, intellectual, etc., background(s), it was created; 7
‘All classicists, whether they know it or not, are historians: of linguistic, graphic and literary behavior . . . or of political, social and economic behavior . . . or of art, religion, science or philosophy’ (Dover 1983: 55 [1988: 54]). 8 I was about to use the fashionable term ‘text’, but this risks being seriously misleading, especially in the case of drama, which always—except for radio plays— includes the visual (Aristotle’s opsis) as an essential component, and for ancient Greeks always included non-verbal auditory elements (Aristotle’s melopoiia) as well. It is not for nothing that dramatists are called playwrights (or in Greek ØÅÆ ‘makers’), despite the ever-increasing efforts of the semi-literate to limit them to the role of playwrites (of which the Greek equivalents, æƪø fi (Ø)ªæç and Œø ø fi (Ø)ªæç, are first found in the Hellenistic period). 9 ‘Over a particular period of time’ rather than ‘at a particular time’: no nontrivial work, literary or other, comes into being instantaneously, and some literary works have taken centuries to attain their definitive form. 10 I do not wish to be taken as making here any comment at all on the validity of literary scholarship whose main aims are different from these.
4
Introduction 3. why it was given the form it had rather than any other—which necessarily includes, among other things, consideration of 4. the effects which it was designed to produce upon the audience, readers, etc., for whom it was composed; and 5. the effects which it actually produced both upon those originally envisaged recipients and on other, usually subsequent recipients.
These are exactly the same kinds of question that a historian would be seeking to answer about any other human action in the past: what was done, by whom, when, and under what circumstances; the causes which brought it about that that was done, rather than something else (or nothing); the intended effects of the action, and the effects which it actually produced. The historian of literature, like other historians, will often find that the evidence is not sufficient to ground a firm answer to all—or sometimes to any—of these questions; and this is particularly likely to be the case in antiquity, where our evidence is frequently very scanty. But even when certainty cannot be attained, it is often possible to reduce considerably the area and the extent of uncertainty, and this is what I have been trying to do in most of the chapters that follow. More than half these chapters (Chs. 1, 4, 5, 6, 9, 10, 12, 15, 16, and 18) are to some extent concerned, in one way or another, with the interconnected topics of the dramatists’ exploitation and modification of myth11 and of how much, and what, their audiences would know, or be able to guess, in advance about the subject-matter of a play and the course of its action. Almost every Greek tragedy was based on, or at least took its start from, some story about the distant past which had often been told before and would be familiar (at least in outline) to most spectators.12 But, so far as we can tell, no 11
I have attempted a general discussion of this issue in Sommerstein (2005). Aristotle, to be sure, asserts that ‘even the known stories are known to few’ (Poet. 1451b25–6); but he cannot have really believed that only a minority knew anything about myths like those of Oedipus, Thyestes, and Alcmeon (cf. Poet. 1453a20–1, Antiphanes fr. 189), otherwise he would be entirely undermining his own statement a few lines earlier (1451b15–19) that tragic dramatists tended to use for their characters the names of persons who (were believed to have) actually lived because that added credibility to their plots (‘what has actually happened is obviously possible’). The passage is best explained by Baldry (1954: 154 n. 2: ‘the names [cf. O ÆÆ 1451a15, 22] of the figures of legend were familiar and generally accepted, but the incidents connected with them (except, presumably, those bare essentials 12
Introduction
5
tragedian was ever content simply to put the story on stage in its inherited form (or one of its inherited forms); there was always some novel feature, an alteration or a complication, which made the story to some extent a new one—in some cases an almost completely new one, as possibly in Sophocles’ Antigone13 and certainly in Euripides’ Orestes. Consider then the position of an audience at the beginning of a play, once they understood at what point of what story-cycle the action was commencing: how much can we take it that they ‘knew’ (or thought they knew) about how the action would proceed and how it would end? And in what ways might a dramatist play on, and massage, such audience expectations? These questions are considered, respectively, in Chapter 15 (with special reference to Sophocles’ Oedipus the King) and in Chapter 16 (with special reference to Sophocles’ Electra and, more briefly, to Euripides’ Hippolytus). Chapter 1 considers what information, if any, fifth-century spectators would be likely to have, before they entered the theatre, about the plays they were about to see, and in particular (1) whether the titles now attached to tragic dramas were already attached to them at the time of their first production and (2) if so, whether they were designed, always or sometimes, to inform, to enlighten, to mystify, or to mislead. Chapters 9, 10, and 12 are all about the Oresteia, and they are all in different ways concerned with a dangerous temptation to which interpreters of tragedy, scholarly and other, have repeatedly proved vulnerable: that of reading back into a scene information which was not yet available to its first audience.14 In the parodos of
which P ØÐ º Ø [a slight misquotation of 1453b23]) were known only to a few. Hence the playwright was at liberty to invent incidents (outside the fixed essentials) and to attach them to the familiar O ÆÆ.’ 13 If Aesch. Seven 1005–78 is to be regarded as spurious (Taplin 1977: 169–91; Sommerstein 2008: i. 147–9), there remains no evidence whatever that Sophocles’ basic story—an edict prohibiting the burial of Polyneices, defied by Antigone at the cost of her life—was known at all before he composed his play (cf. Baldry 1956: 33–4; Sommerstein 2005: 165); indeed, according to the only earlier sources we possess (Ion of Chios PMG 740 may well be later than Sophocles’ play, since Ion lived to the 420s), two of the play’s three other main characters, Haemon (Oidipodeia fr. 3 West) and Ismene (Mimnermus fr. 21 West; Pherecydes fr. 95 Fowler), were dead—in Haemon’s case, long dead—before the defeat of the Seven and the mutual killing of Eteocles and Polyneices. See Ch. 15, pp. 209–10. 14 Bain (1979) exposed the danger, but critics continue to succumb to it.
6
Introduction
Agamemnon, when we hear of the omen of the eagles, there is nothing to indicate that it appeared at Aulis—in fact such indications as there are suggest that it appeared outside the palace at Argos; in seeking an answer to the question ‘why is Agamemnon placed in the position of having to choose between sacrificing his daughter and abandoning the expedition against Troy?’ we have no right to refer to the curse of Thyestes, since we have no way yet of knowing that Thyestes, let alone his curse, will ever be so much as mentioned in the drama; when Orestes is contemplating his matricidal revenge, even when he is actually committing the deed, we do not yet know (though, as we later discover, he does) that Apollo has promised to protect him afterwards; and contrariwise, earlier in the same play, when Electra prays for the return of Orestes and the punishment of Agamemnon’s murderers, she has no idea (though we do) that it will be Orestes who inflicts that punishment. Various other aspects of the dramatists’ treatment of myth are explored in Chapters 4, 5, 6, and 18. The first and last of these it will be more convenient to discuss later in this introduction. Chapter 5 looks at an issue—the relative ages of Eteocles and Polyneices—on which Sophocles and Euripides directly contradict each other, and shows which option Aeschylus chose and also how (and when) he presented it, half verbally and half visually, to his audience. Chapter 6,15 inspired by the studies of Sicherl (1986) and Ro¨sler (1993), attempts to establish how Aeschylus presented the myth of the Danaids, of which literally dozens of variant versions are known,16 and provides support for Ro¨sler’s view that The Suppliants was the second, not the first, play of its tetralogy. With this we have reached the borderline of my second topic, that of trilogies and tetralogies. While Sophocles and Euripides did occasionally present a connected suite of plays as a City Dionysia production,17 the form is primarily associated with Aeschylus and his
15 The Addenda for this chapter are particularly important, since they correct a major error in the original paper (but also show that it was not vital to the argument). 16 Cf. Garvie (1969: 163–81); Friis Johansen and Whittle (1980: i. 44–50); Gantz (1993: 203–8). 17 Sophocles, on an unknown occasion, produced a Telepheia (IG ii2. 3091.7–8); but a corrupt and muddled passage of the Suda (815) seems to imply that this was
Introduction
7
contemporaries.18 The only example that survives complete (minus the satyr-play) is, of course, the Oresteia; chapters 9 and 12, which originated as pre-performance talks for the Actors of Dionysos, both plead for this trilogy to be viewed and interpreted as a whole, despite the considerable difficulty of presenting it as a whole within a timespan acceptable to present-day audiences.19 Chapter 17 considers a group of five plays by Sophocles from which repeated attempts have been made to extract a connected trilogy, and argues, not merely that none of the existing proposals is convincing, but that every conceivable trilogy composed of three plays out of these five would have such serious flaws that it is overwhelmingly likely that all five plays were produced on separate occasions. Between them, Aeschylus, Sophocles, and Euripides were believed by Hellenistic scholars to have composed nearly three hundred plays; of these, just thirty-three survive more or less complete.20 For the
unusual, saying that Sophocles ‘began the practice of competing with play against play instead of with a tetralogy’ ( æƺªÆ Scaliger: æƺªÆ( ) GB: æƺª ØÐ ŁÆØ [the ending comes from Iªø Ç ŁÆØ, three words before] AVM). Euripides produced the suite Alexandros, Palamedes, Trojan Women, Sisyphus in 415 (Aelian VH 2.8), though the plays are more loosely linked than those of most Aeschylean tetralogies. 18 Four connected tetralogies by Aeschylus are attested, and seven more can be reconstructed with fair to high confidence; see Gantz (1979, 1980), Sommerstein (1996a: 53–70; 2008: i.pp. xxix–xxx), and, for a complete survey of proposals made up to 1985, Radt (1985: 111–19). We also have a reference to a tetralogy called the Lycurgeia by Aeschylus’ contemporary Polyphrasmon (Hypoth. to Aesch. Seven) and, a generation later, to a tetralogy Pandionis by Aeschylus’ nephew Philocles ( Ar. Birds 281). In Sommerstein (forthcoming a) I argue that Aeschylus’ production of 472 bc—the only production of his which we know to have consisted of four prima facie unconnected plays—was in fact a connected suite, all the first three plays either foretelling or enacting major events from the wars of 480/79 against Persia and Carthage in correct sequence. 19 It is gratifying to find that increasing efforts are being made to overcome this. The 188 Aeschylean productions since the beginning of 2001 recorded on the database of the Archive of Performances of Greek and Roman Drama at http:// www.apgrd.ox.ac.uk (accessed 20 May 2009) included forty-seven productions of the Oresteia as a whole (or of adaptations based on all three plays of the Oresteia and on no other ancient play) compared with forty-six of Agamemnon and seven each of Choephoroi and Eumenides. 20 Including Prometheus Bound and Rhesus, which, whoever actually wrote them, were generally (the former universally) accepted in antiquity as the work of Aeschylus and Euripides respectively.
8
Introduction
remainder, we have two or three thousand ‘fragments’—papyrus remnants and ancient quotations; a fair number of plot-summaries or fragments thereof, and sometimes other ancient statements about the content of a play; a great deal of mythographic material, much of which shows signs of being wholly or partly of tragic origin; and a good many artistic images, some of which explicitly declare themselves to have been inspired by a theatrical performance (e.g. by the presence of a piper or of a stage-like environment) while for others this can be reasonably inferred. In recent years there has been a marked increase of interest in these ‘lost’ or ‘fragmentary’ plays,21 and since about 1998 much of my research has been devoted to them.22 They feature significantly in several chapters in this volume. I have already mentioned the discussions of lost or partially lost trilogies, or putative trilogies, in Chapters 6 and 17. Both in these papers and elsewhere I have sometimes been, or been seen as being, rather on the bold side in extracting from our evidence, in favourable cases, inferences about the structure and content of fragmentary plays; I attempt in Chapter 4 to explain and exemplify some aspects of my methodology. In Chapter 18, I use the evidence of fragmentary plays to fill out and diversify Sophocles’ portrayal of Neoptolemus: the Neoptolemus of Philoctetes proves to be almost as abnormal, against the background of the usual presentation of Neoptolemus in Sophocles and elsewhere, as the Odysseus of Ajax. The multifarious connections between the tragic drama of Athens and the society in which its composers and spectators lived have rightly been a major concern of scholarship over several decades. Four aspects of these connections are discussed in four chapters of
21 The last fifteen years have seen, among much else, the completion of TrGF by Kannicht (2004), the publication of a four-volume Bude´ edition of the fragments of Euripides (Jouan and van Looy 1998–2003), and the inclusion of the fragmentary plays of all the ‘big three’ in new Loeb editions, two of them for the first time (LloydJones 1996; Collard and Cropp 2008; Sommerstein 2008: vol. iii). Harvey (2005) presents a comprehensive review of the current state of the field. 22 In addition to the Loeb Aeschylus (see previous note) and the papers reprinted in the present volume, this has included two collaborative editions of fragmentary plays of Sophocles (Sommerstein, Fitzpatrick, and Talboy 2006; Sommerstein, Clark, and Talboy forthcoming), a conference volume on Sophocles (Sommerstein 2003), and several more articles (Sommerstein 2000b, forthcoming a, c, d).
Introduction
9
this volume.23 Chapter 3, on tragedy’s presentation of problems and crises in the transition to male adulthood, has been described above. Chapter 7 discusses the socioeconomic composition of the Athenian theatre audience at different periods, and the extent to which it may have influenced tragic (as well as comic) dramatists in search of competitive success. Chapter 8 suggests that certain prominent themes in the Oresteia are designed to allude specifically to the assassination of Ephialtes in 462/1, and that this can throw additional light on the trilogy’s well-known and thinly (if at all) disguised interest in issues crucial to Athenian internal and external politics at the time of production. Lastly, Chapter 13 aims to reclaim for Aeschylus (or at least for a member of his family) the famous epitaph which Athenaeus and Pausanias assert he composed for himself, against the arguments of Page (1981: 131–2) seeking to prove that it was a Hellenistic composition; my counter-arguments are drawn partly from its language and partly from its content. If I am correct, the epitaph itself is a notable document testifying to the connection that was perceived to exist between tragic drama and Athenian citizenship. Aeschylus had won fame as a dramatist throughout the Greek world, and particularly in Sicily where he was residing at the time of his death. Yet in this epitaph there is not one word that even remotely alludes to his profession or to the triumphs he had achieved in it; he wanted to be memorialized (or his family were sure that he wanted to be memorialized) as one who had fought in the Athenian ranks among ‘the first of all Greeks . . . who stood firm when they saw Persian dress and the men who wore it’.24 I end this introduction, as I begin Chapter 13, with my translation of that epitaph: In Gela, rich in wheat, he died, and lies beneath this stone: Aeschylus the Athenian, son of Euphorion. His valour, tried and proved, the mead of Marathon can tell, The long-haired Persian also, who knows it all too well. 23 See also Sommerstein (2006), which uses the evidence of tragedy to refute the claim (Omitowoju 2002) that for classical Athenian males the question whether a (free) woman had or had not consented to an act of illicit sex was of little or no importance, and Sommerstein (forthcoming b), which analyses in detail the relationship between Orestes’ trial in Eumenides and procedures in actual Athenian homicide trials. 24 Hdt. 6.112.3.
This page intentionally left blank
1 The titles of Greek dramas Every ancient Greek play that we know of has a title attached to it; some have two alternative titles, or even three. What is the status of these titles? How did they originate? Who was responsible for attaching them to plays? Were they known to the plays’ first audiences, and did those audiences expect them to provide a clue to the understanding of the plays? I shall begin by stating my conclusion, and afterwards discuss the evidence. I believe that, from the 470s at any rate, dramatic compositions were given titles by their authors which were recorded in the state archives; that from 446 or thereabouts, these titles were known to the public in advance of performance; that the titles by which plays were known in later antiquity and today are authentic in the overwhelming majority of cases; and that once the advance publication of titles began, dramatists started using them to mystify, tantalize, and sometimes mislead their audiences as to what their play was going to be about. I shall now present the evidence in favour of this conclusion, before considering other evidence that might prima facie seem to tell against it. The strongest evidence for the basic authenticity of play-titles is the fact that there existed records of Athenian productions—the Didaskaliai, edited by Aristotle or under his supervision,1 transcribed
1 The work is ascribed to Aristotle in Ar. Birds 281, 1379, (Arethas) Pl. Apol. 18b, Harpokration 54, and Photius s.v. Z ı ŒØ, and listed in the catalogue of his works by D.L. 5.26.
12
The titles of Greek dramas
on to stone (in an updated form) early in the third century bc2 (IG ii2.2319–23) and used by Hellenistic literary historians—which listed the results of dramatic competitions at the City Dionysia and Lenaia, with precise dates and titles,3 and [2] which ancient scholars regarded as a document so complete and reliable that it could be used with confidence to prove a negative, as, for instance, that Euripides’ Andromache, Nikophon’s Sirens, or Metagenes’ Thouriopersai, were never produced at Athens.4 We know that this record was kept sequentially, for those who used it had no difficulty distinguishing between different eponymous archons of the same name, such as the two named Kallias who held office in 412/11 and 406/5.5 We know too that it went back at least as far as 472, when Aeschylus won first prize with Phineus, The Persians, Glaukos Potnieus, and Prometheus [Pyrkaeus]:6 further than that we cannot trace 2
Probably in 279/8 (Snell 1966: 27). One text, IG xiv. 1097 ¼ IGUR 216 ¼ Kallias test. 4 KA, might seem to indicate that some titles were missing from the Didaskaliai. This inscription seems to be a reworking of the Didaskaliai arranged not, as Aristotle’s book (and presumably the official records) were, by festival, date, and rank-order, but by dramatist, rank-order, festival, and date, giving successively, for each dramatist, his Dionysia victories, Lenaia victories, Dionysia second places, Lenaia second places, and so on; the last item in each entry is normally the title of the play in the dative case (eight titles survive, in whole or in part), but in lines 2 and 3 we find merely Œø ø fi Æ fi . The fact that both untitled plays are apparently by the same author suggests that the reason for the absence of titles is something special to that author. If so, A. M. Wilson (1973) may have been right to suggest that the subject of IG xiv. 1097.1–6 was Ekphantides rather than (as usually supposed) Kallias, for it would fit in with other features in our evidence about Ekphantides. Ekphantides was a contemporary of Kratinos, and a highly successful dramatist with four victories at the City Dionysia (IG ii2. 2325.49; of contemporaries, Kratinos had six, Krates and Telekleides three each), but his style came to be regarded as either obscure or old-fashioned (he was nicknamed ˚Æ Æ, either Øa e Åb ºÆ æe ªæç Ø [Hsch. Œ716] or [Øa e] IŁ ŒÆd ƺÆØ
[ Ar. Wasps 151]) and, remarkably for a leading dramatist active from the 450s (as is evidenced by his position in the Dionysia victor-list) to the 430s (cf. Ekphantides fr. 5), only two titles and six fragments of his work have survived. It is as though Ekphantides, late in his career, was unwilling to move with the times: he clung to a dated style, he was reluctant to let copies of his works go into circulation (only one play of his is ever quoted by later writers), and sometimes, in defiance of what had become established convention, he even insisted on leaving them nameless. 4 Cf. Eur. Andr. 445 and Ath. 6.270a. 5 Ar. Lys. Hypoth. I Henderson Kd ˚ƺºı ¼æå ıÐ a ˚º ŒæØ , Ar. Frogs Hypoth. I(c) Dover Kd ˚ƺºı ¼æå ıÐ a ت Å. 6 Aesch. Pers. Hypoth. 3
The titles of Greek dramas
13
it. I do not see how such a catalogue could possibly have been created in retrospect if the information had not been recorded contemporaneously, presumably by the magistrates in charge of the festivals or their staff. And the existence of such contemporaneous records at once creates a presumption in favour of the authenticity of the information they transmitted. Of course, once the data were roused from their sleep in the archives and got into the hands of scholars, they were liable to be edited and adjusted, and in some cases this has almost certainly happened,7 but the onus of proof must always be on whoever alleges that it has. Two secondary indications also point towards the general authenticity of the traditional titles. One is the evidence of fifth- and fourthcentury liter-[3]ary texts: in Aristophanes, as Taplin pointed out (1975: 185 n. 5), ‘all the play titles found . . . seem to be those transmitted to us’, and the same is true, so far as we can tell, for references in contemporary and later comedy, in Aristotle, and for the few found in Plato and the orators. The other is the very eloquent evidence of some of the dramatic titles themselves. When Brown (1984: 268) wrote that ‘the title of a Greek play is never whimsical or paradoxical, or designed to tell us anything interesting’, he must have been thinking exclusively of tragedy: plenty of comic titles are ‘whimsical or paradoxical’ in a high degree,8 and no bookseller, or Hellenistic scholar, could ever have invented them as, in Brown’s words, ‘simply a label to identify the play being talked about’. At first sight there is considerable evidence that tells against the position I have so far taken—and before now I have drawn attention to some of it myself.9 This evidence can be treated under three headings: the use of supplementary designations to distinguish
7 As in the cases of Aeschylus’ Seven against Thebes and Eumenides (below, p. 15); probably also in those cases where a didascalic record gives a play a distinguishing epithet or ‘sur-title’ (below, pp. 14–15), e.g. Glaukos Potnieus (Aesch. Pers. Hypoth.), Iphigeneia at Aulis ( Ar. Frogs 67). 8 A small selection: Aristophanes’ Wasps and Frogs (discussed below, p. 24) and also Aiolosikon; Krates’ Tolmai (‘Acts of Audacity’); Kratinos’ Dionysalexandros and Pytine (‘The Wicker Flask’); Metagenes’ Thouriopersai; Pherekrates’ Doulodidaskalos (‘The Slave-Teacher’), Hipnos (‘The Kitchen’), and Leroi (‘Stuff and Nonsense’); Platon’s Nyx Makra (‘The Long Night’); Telekleides’ Sterroi (‘The Hard Men’). 9 Cf. Sommerstein (1989a: 435–6, b: 12).
14
The titles of Greek dramas
between homonymous plays; the mismatches between some early tragic titles and the content of their plays; and the frequent attestation of alternative titles for the same play. Plays of the same name by the same author, usually tragedies and satyr-plays, often have distinguishing epithets or phrases appended to their titles; thus we have Oedipus Tyrannos and Oedipus at Kolonos, or again Iphigeneia among the Taurians and Iphigeneia at Aulis, while Sophocles’ Aias and Euripides’ Hippolytos were more fully known as Ajax Mastigophoros and Hippolytos Stephanephoros to distinguish them, respectively, from Aias Lokros and Hippolytos Kalyptomenos.10 Such sur-titles (if I may so call them) [4] are never mentioned before the Hellenistic period (Aristotle, for instance, speaks only of Oedipus and Iphigeneia), though as West (1979a: 131) has acutely observed, some of them at least cannot have originated so late, since a sur-title like Mastigophoros or Kalyptomenos alludes to a memorable visual feature of its play and therefore must have been created in an environment where a play was thought of as primarily a theatrical event rather than a literary text—that is, somewhere more like the Agora of Athens than the Mouseion of Alexandria. West is very likely
10 Presumably Sophocles’ surviving Philoktetes must once have had a sur-title to distinguish it from Philoktetes at Troy. Comedies of the same title are often given distinguishing numbers rather than sur-titles (as in the case of Aristophanes’ Clouds, Thesmophoriazusae, Wealth and Aiolosikon, Eupolis’ Autolykos, and Menander’s Adelphoi and Epikleros). In some such numbered pairs the ‘second’ play is a mere revision of the ‘first’ (as in the case of Clouds), but this is by no means always the case—if, with Galen (xv. 424.4–9 Ku¨hn; cf. Ar. Clouds Hypoth. I Dover ¼ I Guidorizzi), we distinguish a revised script from a new one by whether it (1) has the same basic structure and content (Ł Ø) and (2) retains a substantial portion of the text in unchanged form: by this test Clouds and (according to Galen) Autolykos are clear cases of revision, but Thesmophoriazusae and Wealth (on which see Sommerstein 2001: 28–33) equally clearly are not. (Where homonymous plays are distinguished by sur-titles, so far as we can tell, one is never a revision of the other.) In tragedy numerical designations are less common, except for some reason in the case of Sophocles, for whom we have four numbered pairs (Athamas, Lemniai, Tyro, Phineus) and apparently one numbered triple (Thyestes). Butrica (2001: 51–62), while his almost total rejection of the evidential value of these numerical designations is excessive, has done well to remind us never to take it for granted that they tell us reliably which of two homonymous plays was the earlier: I would be more confident than Butrica that in general the numerical labels originate with Hellenistic scholars, but their conclusions must often enough, especially in the case of tragedy, have been based on little or no solid evidence.
The titles of Greek dramas
15
right to see the origin of sur-titles in the book trade. The very fact that it was necessary to devise them itself shows, paradoxically, the stability of the basic titles themselves: it was easier for the booksellers to assign distinguishing epithets to the two Hippolytos plays than to redesignate one of them as Phaidra,11 even though we know that in at least one of these plays Phaidra was the character who made the deepest impression on contemporaries.12 To be sure, it also shows that nobody other than booksellers was concerned to ensure that different plays by the same author had different titles, and that includes the authors themselves; but it does not follow from this that the dramatists thought titles were of no importance, only that in the fifth century they did not regard titles as of permanent importance and were not concerned about the convenience of future readers. I shall in fact be arguing that the time period when a play’s title did matter, to author and audience, was quite a short one, beginning a few days or weeks before the performance, and ending, normally, some time during the first half of the performance itself. Another kind of apparent counter-evidence I can be brief about, since I have discussed it more than once elsewhere.13 Two plays of Aeschylus, Seven against Thebes and Eumenides, have titles, universally attested in antiquity (and listed in surviving didascalic notices),14 which fail to match their scripts: in Seven the city and people of Thebes are never once mentioned under that name, and in Eumenides the Erinyes, on being received into Athens, seemingly acquire the name not of Eumenides but of Semnai Theai (Eum. 1041). I have argued elsewhere that these anomalies are to be explained by supposing that in Aeschylus’ time, when a connected tetralogy [5] was produced, only one title was officially recorded for the whole production (such as Oresteia or Lykourgeia); individual plays will have acquired titles of their own later on, either via the book trade or when they were produced separately in revivals, and these will have been inserted into the published version of the
11
Though the manuscript L does in fact title the play as Phaidra at the beginning (but as Hippolytos Stephanephoros at the end). 12 As witness Ar. Thesm. 547, 550, Frogs 1043, 1052. 13 See n. 9 above. 14 Aesch. Seven Hypoth.; Ar. Frogs 1124.
16
The titles of Greek dramas
Didaskaliai when there was sufficient evidence to establish what plays a tetralogy had consisted of.15 A large number of comedies, and a smaller but still significant number of tragedies and satyr-plays, are sometimes referred to, both in testimonial sources and at the beginning or end of actual play-texts, under double (and occasionally triple) titles such as Semele or Hydrophoroi (Aeschylus) and Dramata or Kentauros (Aristophanes).16 In most cases the same play will also be mentioned elsewhere under one, sometimes both, of the separate titles alone; thus a comedy by Strattis, Makedones or Pausanias, is mentioned twice under the double title, four times as Makedones, and once as Pausanias—and a single author, Athenaios, uses all these three
15 For Aeschylus such evidence appears usually to have been available, but it may not have been so, say, for Polyphrasmon, whose Lykourgeia appears in a didascalic notice for 467 without any indication of its component plays; similarly for Philokles’ Pandionis ( Ar. Birds 281) and Meletos’ Oidipodeia ( [Arethas] Pl. Apol. 18b) the Didaskaliai apparently only contained the tetralogy title. The Prometheus plays present a related but slightly different problem. When Prometheus Desmotes and Lyomenos were produced together, as they surely were (West 1979a: 130), they can hardly both have been officially registered as Prometheus; and yet there is no trace of either of them ever having had any other name. This could be explained, if the production was a connected tetralogy, by supposing that it was originally given a collective name, say Prometheia. However, there is good reason to believe that this production was not a connected tetralogy or even trilogy (for Prometheus Pyrphoros, often thought to have been its third or maybe first play, is probably nothing but a variant—and a better—name for the satyr-play of 472, Prometheus Pyrkaeus; see Sommerstein 1996a: 319–21). If the two plays were written and produced by Euphorion in the late 430s (so Bees 1993), one is tempted to wonder if they may have been staged as a connected pair at the Lenaia,* where tragic contests were just beginning (or so it is usually assumed, e.g. by Snell 1971: 29) and each competitor (as the layout of IG ii2. 2319–23.66–83 indicates) apparently put on only two plays. 16 For particulars see Appendix I (comedies) and II (tragedies and satyr-dramas), where for each play are shown the number of times it is cited or referred to (1) under each of its alternative titles separately, (2) under both or all of them together, and (3) where there are three alternative titles, under two of them. The data are mainly based on TrGF and PCG (except that for lost plays of Euripides they are based on Nauck 1889). The listings do not include the many cases in which two titles have been conjectured by modern scholars to be alternative names for the same play, except in a handful of cases where the indirect evidence is generally regarded as overwhelming; they also exclude three comedies (Anaxandrides, Zographoi or Geographoi; Alexis, Phaidon or Phaidrias; Timotheos, Metaballomenos or Metapheromenos) whose double titles look like textual variants rather than real alternatives.
The titles of Greek dramas
17
designations of the play on different occasions.17 It seems prima facie obvious that no dramatist would have given his work a double title; if he did, moreover, we would expect that when the title was abbreviated for convenience it would always be abbreviated in the same way, just as no one speaks of the discomfiture of Malvolio [6] in Shakespeare’s What You Will.18 More likely the phenomenon normally has its source in the book trade: there were texts in circulation labelled (say) Makedones alongside others labelled Pausanias, and the double title arose when it was observed (presumably in Alexandria) that these two were different designations for the same play. This would imply that the title on a fifth- or fourth-century book-copy of a play was not necessarily the same as the title under which it was entered in the didascalic records; and since the extant evidence for double titles is very patchy indeed,19 we can never be certain that the didascalic title of a play has actually survived. However, the scope for uncertainty here is not as great as it may seem. In regard to comedy, where a play is cited under both double and single titles, we usually find that only one of the single titles is attested, and only in a small minority of cases is the evidence for the two single titles even approximately equal.20 We are thus usually dealing not so much with two competing titles, but with a standard title and an occasional variant. We have three test cases of comedies which have been preserved, in whole or in large part, and for which alternative titles are known but one of the two is overwhelmingly preponderant in the sources: one by Aristophanes 17 The double title at 9.396a, 13.589a; ÆŒ at 7.302e, 323b; —ÆıÆ Æ at 14.654f. 18 In the days when the humorous magazine Punch, then immensely popular, also bore (but did not give prominence to) the alternative title The London Charivari, it once carried the story (told in the first person) of a man who asked at a news-stand for The London Charivari; the newsvendor had never heard of it, asked the customer more than once if he was sure of the title, searched frantically through his stock and failed to find the magazine. After several minutes the customer, tired of waiting, gave up and said, ‘Oh, hang it, give me Punch.’ 19 Thus of fifty-five comedies (down to the mid third century) for which double titles are known, in thirty-six cases the double title as such is attested only once each, and in two more cases it is not directly attested at all. 20 Of forty-three comedies that are referred to under both double and single titles, in thirty cases only one of the two single titles is attested, and in seven more one of the single titles has at least a four-fifths majority of citations over the other.
18
The titles of Greek dramas
(Lysistrate or Diallagai)21 and two by Menander (Dyskolos or Misanthropos; Thrasonides or Misoumenos). In each case there is good reason to believe that the better-attested title (Lysistrate, Dyskolos, Misoumenos) is the original one.22 [7] The only (more or less) surviving comedy whose title is in real doubt is Menander’s Samia, which is referred to three times under that name and twice apparently as Kedeia;23 here Samia is probably to be preferred because it follows a common Menandrian titling pattern (naming a play after a woman character of non-Athenian origin)24 and fits the linguistic preferences of the text.25 In tragedy and satyr-drama double titles are much rarer, and they are chronologically bunched. Of the fifteen plays explicitly attested as being known by double titles, nine are by Aeschylus, his predecessors or contemporaries; two by Sophocles; two by other fifthcentury dramatists; none from the fourth century, but two from the third. Very significantly no play of Euripides—the most popular with posterity, and the most often cited, of all the tragedians—is ever 21
I leave aside the absurd variant Adoniazousai, reported by the scholia on 389. The title Diallagai is derived, as the scholia note, from a single scene late in the play (1114–87), and specifically from the mute character Diallage who appears in it; and the noun ØƺºÆª is not particularly prominent in the play (it appears first at line 932, and four times in all, against nine for its synonym Næ Å), whereas the eloquent name of Lysistrate appears as early as line 6 and eleven times altogether, far more often than that of any other woman in Aristophanes and several times, contrary to normal convention, in the mouths of strange men (cf. Sommerstein 1980a: 395–6). The central character of Dyskolos, Knemon, is spoken of as Œº right at the start (7) and more than once again (184, 242, 747, 893; cf. also 296), whereas the word Ø Łæø never appears in the play at all. Thrasonides, as Arnott (1996: 251) has pointed out, became an alternative title for Misoumenos because the name of the character, like that of Falstaff or Micawber, ended up being better known than that of the work in which he figured. 23 Probably to be understood as The Arranged Marriage, in reference to a distinctive feature of the plot (the two old men, Demeas and Nikeratos, returning from abroad, arrange a marriage between their children, unaware that the latter have already made them grandparents); a possible alternative interpretation is The Carer, referring to Chrysis who has taken charge of Moschion’s and Plangon’s baby as if it were her own. 24 Cf. Andria, Boiotia, Thettale, Leukadia, Messenia, Olynthia, Perinthia, also Melia (known only from Men. test. 42.24 KA). 25 Chrysis, though always addressed by name, is usually referred to in the first half of the play as ‘the Samian woman’ (so in 21, 36, 265, 354; ‘Chrysis’ in 56, 70, 314, and invariably from 406 on). The root ŒÅ- is nowhere found in the surviving text of the play. 22
The titles of Greek dramas
19
designated by a double title. This pattern, combined with the evidence we considered earlier, strongly suggests that from about 450 bc onwards, tragic titles were highly stable,* the didascalic title being so well known to the literary public that the possibility of using a different one to label a book-text normally did not arise. There are one or two exceptions, the best-known being Bakchai, which is called Pentheus two or three times by Stobaios and also in the manuscript L (but Bakchai in a surviving quotation from the Didaskaliai);26 the general pattern, though, is very clear, and makes it likely that whatever had been the case earlier, from the mid fifth century the titles of tragedies and satyr-plays were not only officially recorded but publicly announced at the time of production. Comedy shows a similar chronological gap, but a much shorter one. From Aristophanes and his contemporaries such as Pherekrates, Eupolis, and Platon, to Menander and his contemporaries such as Diphilos, Philemon, and the two Apollodoroi, double titles are fairly evenly scattered through the record, and one is also recorded from the very earliest of Athenian comic dramatists, Chionides. The gap comes in the generation before Aristophanes, in the work of Kratinos and his contemporaries such as Krates, Telekleides, and Kallias, that is, the dramatists who peaked in the [8] 440s and 430s: between Chionides and the outbreak of the Peloponnesian War, not one comedy is attested, directly or indirectly, as having had a double title.27 Looking only at them we would be almost bound to assume that the explanation was the same as for tragedy: that at this time comic titles too were being publicly announced and thus becoming so firmly associated with their plays that the official titles were difficult to displace. How then can we explain the fact that a few years later we do find double titles for comedies beginning to appear with fair regularity? It is hardly likely that the public announcement of titles was abandoned for comedy and retained for tragedy; and we 26
Ar. Frogs 67. Although just after the end of this period, Luppe (1966: 184–91) makes a good circumstantial case for identifying Kratinos’ Idaioi with his Dionysalexandros of (probably) 430—to which, indeed, one could add that Kratinos fr. 90, which seems to indicate that in Idaioi someone was shaved, fits well with the references to hairremoval in two citations from Dionysalexandros, fr. 39 (referring to shears ‘with which we shear the sheep and the shepherds’) and fr. 48 where someone, probably Dionysos, says ‘I looked like a plucked fleece’.* 27
20
The titles of Greek dramas
have already seen reason to be sceptical about the supposition that authors might have given double titles to their own plays (though it would be risky to rule it out in all cases). Perhaps the explanation is that, on the one hand, more book-texts of comedy were now going into circulation than previously (so that there were more opportunities for deviant titling), while on the other hand there was still nothing like such intense public interest in comedy as there was in tragedy (so that the titles of comedies were much less deeply burned into potential book-buyers’ minds than those of tragedies were): certainly explicit references to particular comedies are rare in classical prose texts, and comedy itself refers to tragic dramatists much more often than it does to comic.28 From all this it follows that the traditional titles of tragedies and satyr-plays are broadly reliable, with the important exception of plays from the first half of the fifth century that formed part of connected tetralogies. In the case of comedies there is sometimes more room for doubt, especially where our sources only mention a particular play two or three times, or where there is a fairly even split of attestations between two alternative titles; but among those comedies of which we possess substantial stretches of continuous text there is only one, Samia, about whose original title there can be any [9] serious uncertainty.29 We have also concluded that from some time near the middle of the fifth century, the titles of plays were publicly announced before they were performed. Can we say how long before? Probably not on the actual day of production; in the 420s, we know, the herald still proclaimed, addressing the didaskalos, ‘X, bring on your chorus’ (Ar. Ach. 11), not even deigning to recognize the 28 Between 432/1 and 405/4 there are 102 known explicit comic references to tragic dramatists who were alive during that period (counting the extended appearances of Euripides and Agathon on stage in three Aristophanic plays as just one reference each); the corresponding figure for comic dramatists (excluding references within a play to its own author) is thirty-seven. The passages concerned are listed and discussed in Sommerstein (1996b, esp. 329–30, 348–9). The disproportion would be even greater if we included references to poets of earlier generations (above all Aeschylus) and/or passages in which plays are quoted or parodied without the author being named. 29 I am not now speaking of grammatical or orthographic variants such as Sikyonios/Sikyonioi; these are not problems of unstable titling, but of ordinary textual criticism.
The titles of Greek dramas
21
existence of actors, let alone such new-fangled things as titles. The obvious opportunity for an announcement would be the proagon held in the Odeion a few days before the festival, at which producers and performers were presented to the public30—which, significantly, is likely to have been instituted at the time when the Odeion was built, probably between 450 and 446.31 If a dramatist was aware that his audience would come to see his play knowing what it was going to be called, might he, in one way or another, exploit that knowledge? In the fifth century the use of names for publicity was well understood. Politicians could even use the names of their children to advertise their policies,32 and there is no 30 Aischines 3.67; cf. Pl. Symp. 194b (referring to the Lenaia of 416), Life of Euripides 20 (referring to the Dionysia of 406). 31 The building of the Odeion was associated with the relaunching or reorganization of the musical contests at the Panathenaia, of which it became the venue (Plut. Per. 13.11). We are told in Ar. Clouds 971 that Phrynis was the first to win a Panathenaic victory with the kithara (æøÐ Ææ’ ŁÅ ÆØ ŒØŁÆæø fi ØŒfiÅÐ ØŒÅÐÆØ —Æ ÆŁ ÆØÆ: sic fere codd. EMA) ‘in the archonship of Kallias’; Davison (1958: 40–1), noting that of the three fifth-century archons of this name none held office in a Great Panathenaic year during Perikles’ lifetime, argued that ˚ƺºı should be emended (with M. H. E. Meier (in Ersch and Gruber 1818–89: iii. 10.284)) to ˚ƺºØ åı (archon 446/5), that the Panathenaic musical victor-lists available to later scholars began with the Periclean relaunch, and that therefore the Panathenaia of summer 446 was the first at which musical contests were held in the Odeion, which must therefore have been completed in the interval since the previous Great Panathenaia in 450. Hose (1993), starting from an inferior (as it stands, unintelligible) text of the Clouds scholion (æøÐ ŒØŁÆæÆØ Ææ’ ` ’ ŁÅ ÆØ ŒÆd ØŒÅÐÆØ —Æ ÆŁ ÆØÆ: sic fere codd. RV, which in the Clouds scholia have suffered much more than E both from casual corruption and from erroneous ‘correction’, cf. Holwerda 1977: pp. x–xxvi) argues for the Kallias archonship of 406 as the date of Phrynis’ victory, but this is surely ruled out by the implication of Arist. Met. 993b15 that Phrynis was famous well before Timotheos; Timotheos died aged 90 between 366/5 and 357/6 (FGrH 239 A 76) and therefore was already middle-aged in 406/5, and anecdotes associate him with Archelaos of Macedon (reigned 413–399) and with Euripides. Hose’s argument that if Phrynis had won a Panathenaic victory in 446 his style would not have seemed disturbingly novel in 423, when Clouds was produced, is fallacious: Euripides won his first Dionysian victory in 441 yet could still be seen as a controversial and subversive innovator in Frogs, produced in 405 after his death. 32 Examples are Italia, Asia, and Sybaris, daughters of Themistokles (Plut. Them. 32.3); Lakedaimonios and Thettalos, sons of Kimon (Plut. Kim. 16.1); Demos, son of Pyrilampes (Pl. Gorg. 481d, etc.); and probably Idiotes, son of Theogenes (PA 7445; see Sommerstein 1977a:273–4). Harvey (2000: 95–7) argues that the rare name Eunomides, borne by the father of the comic dramatist Phrynichos, is another example.
22
The titles of Greek dramas
a priori reason why dramatists in [10] competition with each other should not have used the names of their plays to advertise too. The title of a serious drama, if made known to the potential audience in advance of the performance, can have one of three main functions. First, it can inform them, in general terms, what the drama is going to be about: examples of this are Antony and Cleopatra, Saint Joan, and Titanic. Secondly, it can give them an idea of the play’s mood, atmosphere, or theme, as do A Man for All Seasons, Waiting for Godot, or Love Story (in serious drama, as in the novel, this is largely a nineteenth- and twentieth-century phenomenon). And thirdly, it may be designed to keep them in suspense or even to mislead them: Julius Caesar may be such a case (how many people would have expected the title character to have so small a role and to die midway through the play?) and so too are Die Zauberflo¨te (which sounds like a fairy-tale but proves to be something very different) and To Kill a Mockingbird (which could, from the title, be about practically anything). Many Sophoclean and Euripidean tragic titles are, to be sure, of the first class: Antigone, Aias, Elektra (both), Alkestis are obvious examples. Sometimes, too, the designation of a chorus virtually reveals the mythical episode being handled: the odds are high, for instance, that a play called Trachiniai will be about Herakles and Deianeira, and one can be fairly sure that a play called Bakchai will portray the death of an enemy of Dionysos even if one cannot tell whether it will be Pentheus, Lykourgos, or Orpheus. But there remain, especially in the Euripidean corpus, many titles that will mystify, and some that will mislead. One pattern that Euripides favours, for instance, is to name a play after a major figure of myth who appears in it, when the play deals with a story which would not be the first one a potential spectator would associate with that character’s name. Such plays include Andromache, Iphigeneia (in Tauris), Helen, Melanippe (Desmotis), and Hypsipyle. No spectator would be likely to guess that Andromache would be set in Phthia, Iphigeneia in the Crimea, Helen in Egypt, or Melanippe in Italy, and most would certainly expect Hypsipyle to be about the story of the Lemnian women rather than about an episode in the saga of the [11] Seven against Thebes. Nor would many have expected a play called Herakles to be about the hero’s slaying of his wife and children; if Euripides had wanted to signal that theme
The titles of Greek dramas
23
clearly, he had the option of naming his play Megara—and he did not do so. Another practice of Euripides is to recycle a plural title that had been used by an early dramatist (Phrynichos or Aeschylus) for a play which, while it might recall its precursor in one way or another, dealt with a completely different myth. He does this three times for certain: Kressai (Aeschylus on Polyidos and Glaukos, Euripides on Katreus and Ae¨rope); Hiketides (Aeschylus on the Danaids, Euripides on the recovery of the bodies of the Seven against Thebes); and Phoinissai (Phrynichos on the Persian War, Euripides on Eteokles and Polyneikes). He may have done it a fourth time in another play with a title that Aeschylus had used, Herakleidai. Euripides has certainly been misleading here in one way, for in his play the children of Herakles do not form the chorus (that consists of Marathonians), and it is more than possible that, as in the other cases cited, he has changed the subject-matter as well, but in the absence of better evidence about Aeschylus’ play we are in the dark here.33 Sophocles may have done the same kind of thing at least once. His Oedipus at Kolonos was presumably, like other plays from homonymous pairs, called just Oedipus when it was produced. Of course it would be expected to differ from Sophocles’ earlier Oedipus, but that did not necessarily mean it would present an entirely different episode in Oedipus’ life (consider here Euripides’ two Hippolytos plays), and certainly no one would think of the Kolonos episode, from a time when the main tragic events of Oedipus’ life were long over (for the curse on his sons, which features in this play (1370–96), is elsewhere always located in Thebes).34 So far as comedy is concerned, there are very few lost plays whose action we can even attempt to reconstruct with enough confidence to venture an opinion on how well their titles describe them, but with the eleven surviving plays of Aristophanes it is otherwise. Eight of the 33 Aeschylus might, for instance, have dramatized the conflict between Keyx and Eurystheus which ended in the expulsion of the Herakleidai from Trachis (D.S. 4.57): this story seems to have been well enough known that Sophocles in Trach. 1147–56 has in effect to apologize for the absence from Trachis of most of the children and of Alkmene). 34 So strong is this tradition that Sophocles feels compelled to represent Oedipus as merely repeating, at Kolonos, curses that he has uttered on one or more earlier occasions (OK 1375).
24
The titles of Greek dramas
eleven titles refer to the (or a) chorus; but two of these are positively misleading (the chorus of Wasps are not wasps, and in Frogs the frogs are only a subsidiary chorus who appear in a single scene), and four more (Acharnians, Clouds, Birds, Thesmophoriazousai) while correctly identifying the chorus provide little or no information about the subject of the play (could anyone have guessed that Clouds would be about Socrates or Thesmophoriazousai about Euripides?). Knights may have been a different matter, in view of the known hostility between Kleon and the cavalry (Ach. 5–8) and between Kleon and Aristophanes;* and Ekklesiazousai accurately defines the first half of its play—except that the Assembly scene it seems to foreshadow takes place offstage, the chorus leaving the scene to attend the meeting! Of the three non-choral titles, Peace and Wealth both refer to deified abstractions who confer blessings (though both turn out to be rather passive beings, and one is not even animate); Lysistrate signals that a woman, and probably a woman of citizen sta-[12]tus, will be the chief character, but gives no clue to the plot (we cannot tell in advance that her perfectly ordinary name is itself the best clue to the plot).35 In short, the main function of an Old Comic title is to arouse interest and curiosity, to mystify and tantalize; which, significantly perhaps, is also often the function of the opening scene of an Old Comic prologue. This function would be best served if normal practice was to publicize, in advance of performance, the title and nothing else. Indeed, if anything beyond the title was publicized, the prologue would often be in danger of becoming a rather damp squib.36 On the other hand, Aristophanic prologues are usually so constructed that knowledge of the title itself does not let the cat prematurely out of the bag; the only exception is the late play Wealth, where many spectators, if guided by the title, are likely to have realized, long before they are told at line 78, that the blind beggar whom Chremylos is following, in obedience to Apollo’s response when he had enquired whether to encourage his son to enrich 35 Some wiseacres may even have suspected that the central character might turn out not to be a woman at all, but the effeminate Lysistratos alluded to in Lys. 1105; compare, in reverse, Aristophanes’ The Dramas or Niobos, a (per)version of the Niobe story with a male character in the role of Niobe.* 36 It is perhaps significant that at the proagon the performers appeared without their masks ( Aischines 3.67) and presumably in plain clothes, thus giving no clue as to what character(s), or even what kind(s) of character(s), they would be playing.
The titles of Greek dramas
25
himself by crime, is likely to be the blind god of Wealth37—but in this case the usual function of the comic title is being reversed: rather than mystifying and misleading the audience, it gives them what the audience of tragedy so often have, namely privileged information denied to the characters. Greek play titles, then, were primarily designed for advance publicity—sometimes to inform the audience about the content of the play, and sometimes to keep them guessing, often wrongly, not only before the performance but for some time after it had begun. It is much more doubtful whether we are ever meant to think of a title as providing a clue to the understanding of its play. I know of no case (though I may have overlooked one) where a Greek dramatist makes an important point about his play by dwelling on the significance of its title, as Shakespeare does at the end of Measure for Measure38—or even humorously demonstrates its insignificance as Oscar Wilde does at the end of The Importance of Being [13] Earnest.39 Once the audience have been let into the secret (if any) of what the play is really going to be about, the title has fulfilled its purpose; the public are not going to feel cheated if Antigone has been forgotten by the end of Antigone, or Chrysis reduced to an insignificant role by the end of Samia. After the first performance of a play, a title really was what Brown said it was, just a convenient label. Before that performance, and in its early scenes, the title was a crucial part of the author’s campaign for competitive success.40 37
On the blindness of Wealth (or rather of one of the two or maybe three differently imagined divine personalities/personifications of this name) cf. Hipponax fr. 36, Timokreon PMG 731, Eur. fr. 776 ¼ Phaethon 164–7 Diggle, Amphis fr. 23, Pl. Laws 631c, Men. fr. 74, Theokr. 10.19; see Sommerstein 2001: 5–8. 38 ‘ . . .An Angelo for a Claudio, death for death; | Haste still pays haste, and leisure answers leisure; | Like doth quit like, and Measure still for Measure.’ (Shakespeare, Measure for Measure V.i.410–12; subsequently Claudio turns out to be alive, and Angelo is spared.) 39 Gwendolen: Ernest! My own Ernest! I felt from the first that you could have no other name! Jack: Gwendolen, it is a terrible thing for a man to find out suddenly that all his life he has been speaking nothing but the truth. . . . Lady Bracknell: My nephew, you seem to be displaying signs of triviality. Jack: On the contrary, Aunt Augusta, I’ve now realised for the first time in my life the vital Importance of Being Earnest. (Oscar Wilde, The Importance of Being Earnest, sub finem.) 40 This chapter was originally published in Seminari Romani di Cultura Greca 5 (2002) 1–16 (p. 16 contained the bibliography). I am most grateful to Edizioni Quasar for giving permission for this republication. It was based on a paper presented at the Annual Meeting of the Classical Association at the University of Wales, Lampeter, in April 1998. I am grateful to those who participated in the discussion on that occasion, and to the anonymous referees for Seminari Romani.*
26
The titles of Greek dramas
APPENDIX I
Attic comedies known to have had alternative titles, down to 262 bc
Author
Titles
Citations: Title 1
Chionides Pherekrates Eupolis Phrynichos Aristophanes
Platon Philyllios Strattis Metagenes Nicochares [14] Euthykles Araros Ephippos Epikrates Euboulos
41
Persai / Assyrioi Epilesmon / Thalatta Ipnos / Pannychis Astrateutoi / Androgynoi Tragoidoi / Apeleutheroi Dramata / Kentauros Dramata Dramata / Niobos Lysistrate / Diallagai Hellas / Nesoi Lakones / Poietai Xantai / Kerkopes Plyntriai / Nausikaa Makedones / Pausanias Aurai / Mammakythos Homeros / Asketai / Sophistai Amymone / Pelops Asotoi / Epistole Parthenis / Kampalios Homoioi / Obeliaphoroi Triodous / Rhopopoles Lakones / Leda Odysseus / Panoptai Oinomaos / Pelops Prosousia / Kyklops Semele / Dionysos
0 6 8 14 9
Title 2 Double title 0 1 1 0 1 2
1 1 2 1 1 11
many 9 7 5 1 4 3 2
4 0 1 0 0 0 1 0 0
1 0 1 0 0 0 0 0 0 1
0 0 0 1 0 1 0 0 0 0
7 241 1 1 1 1 2 3 1 (H/A) 1 (H/S) 1 1 1 1 1 3 1 1 1 1
9
The double title appears in the list of Ar.’s plays called the Index Ambrosianus (Ar. test. 2a.19 KA) and in the scholia to Lys. 1114.
The titles of Greek dramas Sophilos Antiphanes
Aristophon Alexis
Timokles Diphilos Menander
Philemon Apollodoros G. Apollodoros C. [15]Theognetos Anon.
42
Tyndareos / Leda Knoithideus / Gastron Oinomaos / Pelops Skythai / Tauroi Stratiotes / Tychon Thorikioi / Diorytton Didymai / Pyraunos Agonis / Hippiskos Demetrios / Philetairos Dorkis / Poppyzousa Krateia / Pharmakopoles Leukadia / Drapetai Pannychis / Erithoi Philokalos / Nymphai Tokistes / Katapseudomenos Kentauros / Dexamenos Eunuchus / Stratiotes Achaioi / Peloponnesioi Androgynos / Kres Arrhephoros / Auletris/-trides Dyskolos / Misanthropos Hypobolimaios / Agroikos Misoumenos / Thrasonides Samia / Kedeia Metion / Zomion Ptoche / Rhodia Philadelphoi / Apokarteron Proikizomene / Himatiopolis Phasma / Philargyros Oneiros / Progamon
0 1 0 3 2 1 1 0 5 0 1 2 1 0 0 1 4 1 5 5 19+ 16 11+ 3 1 2 0 0 1 0
27 0 0 0 0 0 0 0 4 0 0 0 0 1 0 1 0 1 0 1 5 0 0 143 244 0 1 1 0 0 0
1 1 1 1 3 1 1 2 1 3 6 1 4 1 1 1 1 1 1 3 142 1 0 0 1 2 2 1 2 1
The Hypothesis in the Bodmer codex ends I تæç (ÆØ) Ø Łæø. Subscriptio to POxy 2656. 44 Stobaios 1.6.9, 4.29.10. 43
28
The titles of Greek dramas
APPENDIX II
Attic tragedies and satyr-dramas known to have had alternative titles, down to c. 250 bc
Author
Titles
Citations: Title 1 Title 2 Double title
Thespis Pratinas Phrynichos Aeschylus
Sophocles
Euripides*
Ion Philokles Chairemon Sositheos Nikomachos
Athla Peliou / Phorbas Dymainai / Karyatides Antaios / Libyes Dikaioi / Persai / Synthokoi Argo / Kopastes Kares / Europe Phryges / Hektoros Lytra Semele / Hydrophoroi Theoroi / Isthmiastai Atreus / Mykenaiai Manteis / Polyidos Nausikaa / Plyntriai Pandora / Sphyrokopoi Alope / Kerkyon Bakchai / Pentheus Hippolytos Steph. / Phaidra Polyidos / Glaukos Phoinix / Kaineus Tereus / Epops Thersites / Achilleus (Thersitoktonos) Daphnis / Lityerses Tyndareos / Alkmaion
0 0 1 0 1 1 6 3 2 0 6 4 5 8 many many 12 6 0 1 3 0
0 0 0 0 0 0 1 0 1 1 3 2 0 545 446 147 3 0 0 2 0 0
1 1 1 1 1 2 3 2 2 1 0 0 1 0 0 0 0 4 1 0 1 1
45 The title Kerkyon is used only (but invariably) by Eustathios in his Homeric commentaries. 46 For particulars see Diggle (1994: 291). 47 See n. 11 above.
The titles of Greek dramas
29
ADDENDA p. 16 n. 15 my speculation here that Prometheus Desmotes and Lyomenos were produced ‘as a connected pair at the Lenaia’ is of course incompatible with my speculation elsewhere (Sommerstein 1996a: 326 n. 14) that it was with these plays (and others) that Euphorion won first prize at the City Dionysia of 431. p. 19 if I am right to suggest that ‘from about 450 bc onwards, tragic titles were highly stable’, that would tell somewhat against the suggestion by West (2000: 347–50) that the Kares or Europe traditionally ascribed to Aeschylus was actually a somewhat later work by his son Euphorion. p. 19 n. 27 on the relationship between Cratinus’ Idaioi and Dionysalexandros, see now Bakola (2009: 84–6, 96). p. 24 I should also have mentioned the threat made by the chorus in Ach. 300–2 that they will ‘cut up [Cleon] to make soles for the Knights’ shoes’: this is actually out of character for the bellicose Acharnians, and would probably be perceived as coming from the poet (cf. Ach. 377–82, 502–3, 659–64). See Olson (2002) ad loc. p. 24 n. 35 also Strattis’ Atalantos (that this is the correct title is shown by Miles 2009). p. 25 n. 40 I had submitted this paper to SemRom before I became aware of Kaimio (2000). I was glad to see that Kaimio and I agreed on essentials, and on such specifics as that play titles were announced at the proagon (ibid. 55, cf. p. 21 above). p. 28 the century and more between Nauck (1889) and Kannicht (2004) have yielded a few more citations for Euripides’ Alope and Polyidus under their usual titles, but none for the titles Cercyon or Glaucus. I should have noted that the title Glaucus, like Cercyon, is used by only one ancient author (in this case Stobaeus).
2 Violence in Greek drama It is a cliche´ that violence, especially violent death, was not presented visually on the Greek tragic stage. Yet violence could be presented verbally, through ‘messenger-speeches’ or other forms of narrative, with virtually no limits on the intensity of the horrors described; nor were the dramatists in the least squeamish about presenting on stage the results of violence in the most appalling form—the blinded Oedipus or Polymestor, the dismembered corpse of Pentheus whose head adorns his mother’s thyrsos, Aias amidst the animals he has tortured and slaughtered. The convention is evidently a very powerful one; but what precisely are the constraints, and how are they to be explained? Any conventional practice or constraint in Greek drama might in principle be ascribed to causes of one (or perhaps more than one) of several kinds. 1. Religion. On the one hand, the production of plays was part of a festival of Dionysos, and the god himself was considered to be watching the performances; hence it might in principle be the case that certain kinds [42] of spectacle were avoided as being potentially offensive to him. On the other hand, certain acts and utterances were forbidden absolutely by religious laws, and these would also be avoided in drama;1 for example, though two of Aristophanes’ surviving plays have choruses I am most grateful to Eleanor OKell, and to an anonymous referee for Ordia Prima, for their comments on earlier version of this chapter. 1 Aeschylus, to be sure, is said to have been accused of divulging secrets of the Eleusinian Mysteries in his plays, and this story was well enough established in Aristotle’s time for him apparently to take its truth for granted (EN 1111a8–10). But even if true, it of course proves only that, as we would expect, such divulgation was punishable—not that Aeschylus had actually done what he was accused of. At any rate, if charged, he must certainly have been acquitted, since if convicted he could hardly have escaped the death penalty. See Sommerstein (1996a: 25–6).
Violence in Greek drama
31
consisting, respectively, of women celebrating the Thesmophoria and of deceased initiates of the Eleusinian Mysteries, neither contains anything that might divulge any of the secret elements of these rituals to those who did not already know them.2 2. Competition rules. This is certainly the appropriate explanation for the requirement that the chorus consist of fifteen (earlier twelve) members, and almost certainly also for the restriction of tragedy to a maximum of three (earlier two) speaking actors, with all that flows from that.3 3. Theatrical practicality. There are many things that simply could not be done on stage, given the resources of the Athenian theatre. There could not be abrupt changes of scene, except in the rare cases where the chorus made a temporary exit. Before the invention of the mechane, it had been impossible for characters, even gods, to be seen flying; contrariwise, after the creation of the skene, it became virtually impossible4 for the entire action of a play to be set indoors. At all times, because of the use of [43] masks, it was impossible for characters to eat,5 though they 2
Aistophanes sails near the wind of this prohibition in Thesm. 626–33, when the disguised male who has infiltrated the Thesmophoria is tested by being asked questions about ‘last year’s rituals’—but he then quickly snatches away the forbidden fruit: the old man’s answers, though (at first) apparently accurate, reveal only what all male spectators thought they knew anyway—that if women assemble together in the absence of men, their main activity will be drinking. We shall soon be seeing many other examples of this ploy of threatening to violate a convention and then avoiding the violation, often at the last moment. 3 For example, the sending away of Tekmessa at Soph. Ai. 985–9, and her silence from the time she returns (1168) till the end of the play: evidently the returning Tekmessa is played by a non-speaking performer, because from 1316 to 1373 three speaking actors will be on stage playing other parts (Teukros, Agamemnon, Odysseus). On the parallel rule in comedy (where the limit in Aristophanes’ time was four), see MacDowell (1994). 4 ‘Virtually’, because there is at least one close approach; there is good reason to believe that the great contest in Aristophanes’ Frogs (830–1478) is imagined as taking place in the hall of Plouton’s palace, with his throne and the ‘Chair of Tragedy’ visible to the audience (see Sommerstein 1996c: 229). This would be done by means of the ekkyklema. 5 Olson (1998: 313) was more than justified in rejecting the suggestion of Sommerstein (1984: 152 n. 36 that the chorus of Aristophanes’ Peace begin at 1315 (his 1317) to eat the food that Trygaios has placed before them; as he points out, the very last words of the play (Œi ı ÅŁ Ø, ºÆŒıÐ Æ Ł ) show that the chorus have then still not begun to eat the ºÆŒıÐ they saw ‘roaming about unattended’ at 1314. They are presumably just about to do so (perhaps they already have food in their hands) when they are interrupted by the return of Trygaios.
32
Violence in Greek drama
could (and in satyr-drama and comedy, sometimes did) mime drinking.6 4. Aesthetic/artistic considerations. The language, metre, music, costumes, masks, etc., of tragedy were subject to many conventions that made them, in general, much more restrained and stylized than those of comedy (satyr-drama being in most respects intermediate between the two), and most of these can in origin have had no other rationale than the dramatists’ belief that they were artistically appropriate to the genre. 5. Inertia. Once a convention has become established within a genre it may become, as it were, a rule of the craft, and be maintained by its practitioners for no reason that can be articulated except that it is what they have always done. It is generally difficult or impossible to distinguish a convention maintained for this reason from one maintained for aesthetic/artistic reasons, particularly since those who live within a cultural tradition are very prone to invent (and then hand on, as part of the tradition) ostensibly rational explanations for irrational features of the tradition; but sometimes it can be discerned that a traditional practice which is irrational or arbitrary under the circumstances obtaining at the time under consideration would have been rationally explicable, or even inevitable, under the circumstances of an earlier time. In the case of drama, one can well envisage that the circumstances of its early history (one actor, no back-scene, etc.) might have given rise to conventions that came to be regarded as part of the essence of the genre, and continued to be observed even when the circumstances from which they had arisen no longer existed. Whether any such ‘fossil’ conventions can in fact be identified is another matter. I began by saying it was a cliche´ that violence was not presented visually on the Greek tragic stage. But ‘violence’ (and likewise its Greek equiv-[44]alents, Æ and its derivatives) does not have a single, universally agreed meaning; and some acts that most people would regard as violent unquestionably can be presented visually on the tragic stage. Most of us, I am sure, would regard it as an act of violence if a person was seized and dragged against her will from a 6
Cf. Eur. Kykl. 551–78; Ar. Knights 105–24.
Violence in Greek drama
33
holy place in which she had taken sanctuary; and in tragedy we find no less a personage than Poseidon agreeing with us.7 Yet actions such as that are certainly not kept off the tragic stage: forcible seizure, detention, and restraint are in fact relatively common, and manhandling and dragging are not unknown.8 Only two kinds of onstage violence are completely, or almost completely, avoided in tragedy: killing (including suicide) and striking (whether with a body part or with a weapon or other implement). These two forms of violence will now be examined in more detail. Killing is never represented on stage, in comedy9 or in tragedy (two apparent exceptions will be considered presently). This applies alike to killing by sword or poison or any other means; it applies to homicide, to suicide, to death by accident or act of god. But it is only true under a very precise definition of killing. What may not be shown is the act that causes death—the blow of the weapon, the tying of the noose10 (not that that could have been staged anyway, regardless of conventions), the drinking [45] of the poison (which, contrariwise, would have been very easy to simulate, and would have made a most convenient device for presenting some of tragedy’s many
7 Eur. Tro. 70 Œ’ `YÆ ØкŒ ˚Æ æÆ Æ fi : Athena has just described this seizure as an act of hybris against her and her temple, and Poseidon is in full agreement. 8 Cf. [Aesch.] Prom. passim; Soph. Ant. 939 (Antigone), OT 1522 (Antigone and Ismene), Phil. 814–18, 1003–54, 1300–2 (Philoktetes), OC 831–47 (Antigone), 872–86 (Oedipus), possibly also (as a referee for Ordia Prima suggests) OT 1152–66 where the Theban shepherd may have been subjected to arm-twisting or other lowgrade physical pressure during his interrogation; Eur. Herakleidai 61–76 (Iolaos), Andr. 425–726 (Andromache), Hek. 1282–6 (Polymestor), Tro. 786–9 (Astyanax), IT 456–69 (Orestes and Pylades), Hel. 1629 (Theoklymenos), Or. 1567–1672 (Hermione), Ba. 434–52 (Dionysos), and above all Andromeda (frr. 122, 127, 128). See also Fitzpatrick (2003: 255) on Sophocles’ Aias Lokros, where the audience may have seen Aias dragging Kassandra away after seizing her from Athena’s temple. There appear, however, to be no instances in the six undisputed plays of Aeschylus. 9 It is almost, but perhaps not quite, true in our evidence that death cannot even occur during the action of a comedy. The exception, if there is one, appears in Eupolis’ Demes (fr. 99.<112>), where (if Ko¨rte’s brilliant restoration Ææ’ ˇ’Ø[ ØÐ Æå ] is correct) Aristeides orders a sykophant to be taken away for execution; cf. Sommerstein (2000a: 442, 448–9 n. 33). 10 More strictly, perhaps, the kicking away of the stool (or equivalent act); it would presumably not be impossible for a captive, say, to appear on stage with a halter around her neck.
34
Violence in Greek drama
female suicides). There is no bar on the stage presentation of death itself, and in fact it occurs twice in surviving tragedy, in Euripides’ Alkestis and Hippolytos, in both of which the title character is brought on stage dying from the effect of an act of god that we have not seen actually take place. Still less, of course, is there any bar on the exhibition on stage of the corpses of those who have died by violence; in at least one case (Neoptolemos in Euripides’ Andromache) we see a character dead whom we had never, during the play, seen alive. Just as we may see the events that follow a fatal act, we may also see the events leading up to it. In two cases, both of them suicides, our view is cut off only at the very last moment. The more famous and more controversial case is that of the suicide of Aias. At the beginning of his last speech (Soph. Aias 815–22) he announces that he has just planted his sword in the ground, ready to fall upon it; at its end he makes it plain (864) that he has spoken his last words on earth. If we did not know the conventions, and if we did not have the rest of the play, we might well be tempted to suppose that the suicide itself had occurred in sight of the audience. But in fact we know from 891–2 that when Tekmessa discovers his body, she is in a grove ( ) and, crucially, out of sight of the chorus11 (who ask each other whose voice they are hearing). It follows that the sword must have been planted where the audience could not see it, and that the fatal stroke it (passively) delivers occurs ‘offstage’.12 Less problematic, from our point of view,13 is the suicide of Euadne in Euripides’ Suppliants: she leaps to her death (on the burning funeral pyre of her husband [46] Kapaneus) from some elevated point, no doubt on the skene—and no doubt lands behind it, so that we see the leap but not the fatal 11 Garvie (1998: 211), who is for other reasons committed to the view that ‘the suicide [took] place in full view of the audience’ (ibid. 203), is forced to suppose that at 891 ‘the chorus . . . must be looking in the wrong direction’. Since the chorus have been on the scene, searching for Aias, since 866, they could hardly have failed to see the corpse by now if it had been visible to the audience, and moreover, on hearing Tekmessa’s sudden cry, they would instinctively have turned towards its source (for they perceive where the voice is coming from, though not whose it is). 12 Where exactly this was in the theatrical space, and how and when the body was brought into the prominent position which it certainly occupies by 1003 at latest, are difficult questions which fortunately are not germane to the present discussion. 13 Though in some respects no less difficult in detail (e.g. the chorus can apparently see Kapaneus’ pyre: Supp. 980–3, 1009–11).*
Violence in Greek drama
35
landing.14 Similar, in a sense, is what happens in Sophocles’ Niobe (fr. 441a): we apparently see Apollo and Artemis shoot their arrows at Niobe’s children, but they are shooting into the house (cf. line 4 ø)15 and therefore we will not see the arrows hit their mark. Another way whereby the audience can be taken to the brink of a killing on stage, but not over it, is for the killing itself to be prevented at the last moment by some intervening cause; the closest shave of this kind is in Sophocles’ Philoktetes (1299–302) where Neoptolemos forcibly prevents Philoktetes from shooting the arrow which he has aimed at Odysseus and which, if loosed, would certainly have been fatal.16 It is not only humans who may not be killed on stage, in any form of drama; the same prohibition seems to apply to animals.17 In two plays of Aristophanes, Peace (937–1022) and Birds (848–1057), elaborate preparations are made for a sacrifice, and some of the preliminary rituals (in Peace, all of them) are performed onstage; in both they are abruptly cut short, and the actual slaughter takes place out of sight of the audience. Each time, to be sure, an excuse is given: in Peace it is said that a bloody sacrifice ought not to take place in the presence of the goddess Peace (1019–20); in Birds the sacrifice has been interrupted five times by annoying visitors, and Peisetairos decides he must get out of their way if he ever wants to finish it (‘let’s get away from here as quickly as we can’, Birds 1056). In Peace, additionally, we are given a joky extra-dramatic reason for keeping the sacrifice offstage—to save a sheep for the choregos (Peace 1022). In fact, however, the cost of a sheep or goat would be trivial in
14 If, when Euadne made her leap, she had found herself falling into a deep pool instead of onto a pyre, she would probably have survived; it is thus not the leap that kills her, but the impact and/or the fire. 15 Alternatively, as suggested by Ewans in Ewans, Ley, and McCart (2000: 169), Apollo (onstage) may be giving instructions to Artemis (within) to shoot the girl cowering in a storeroom. 16 For Philoktetes’ arrows, inherited from Herakles, never miss their target (Phil. 105, 198). 17 Herakles in Prometheus Unbound shot the eagle that was tormenting Prometheus (Aesch. fr. 200); but this bird will not have been visible to the audience (cf. Winnington-Ingram 1983: 191), any more than were the birds that Ion threatens to shoot in Eur. Ion 154–83.
36
Violence in Greek drama
comparison with the choregos’ overall expenses—particularly since he could (and seemingly as a rule did, at least for comedy) choose an infe-[47]rior animal18—and the avoidance of onstage sacrifice must have a different, and probably a religious, explanation. One might think of suggesting that it would be thought inappropriate, perhaps dangerous, to have performers go through the motions of making an offering to a god when the whole procedure was only a pretence; but this cannot be so, since other kinds of offering not involving bloodshed (e.g. libations and the burning of incense) are common in drama. More plausibly it might be argued that the problem was precisely that in one important respect the procedure would not be a pretence. Slaughter could not be simulated;19 a real animal would have to be really killed, not in order to please the god to whom it was ostensibly being offered but merely for the purpose of making a show. The complex, and partly obfuscatory, rituals that surrounded Greek animal sacrifice and especially its central act20 show how serious a matter the taking of a domestic beast’s life was perceived to be, even though it was a necessity both for humans and for gods; it would have been unthinkable to do such a thing outside its true ritual context. If that explanation of why animal slaughter was avoided in drama is correct, it evidently cannot be applied directly in explaining why drama also avoided direct presentation of the killing of humans; for simulating that would have presented no difficulty to producers. But a quite different parallel may be suggestive. Tragedy talks incessantly about killing but never shows us (a simulation of) the act itself, though it can come very close to doing so. In precisely the same way, comedy, especially Old Comedy, talks incessantly about sex— but never shows us (a simulation of) the act itself, though, again, it can come very close to doing so.21 Sex and killing (including suicide) 18
Cf. Ar. Birds 901–2 (‘nothing but beard and horns’); Men. Dysk. 438 (‘it’s nearly dead already’), Sam. 399–404. 19 It would not have been possible to represent the animal by a dummy, since a sacrificial beast had to be led, not carried, to the altar. A four-legged ‘pantomime animal’ seems to have been used for the donkey in Aristophanes’ Wasps (for it brays on cue at 179 and 180), but would hardly have been workable for any beast smaller than that. 20 See Burkert (1972/1983), esp. 3–8. 21 Notably in Ar. Lys. 904–53 (where Myrrhine says ‘yes, in a minute’, or the equivalent, at least six times to her desperately frustrated husband Kinesias, and then, when she has no excuses left for procrastination, runs away) and Thesm. 1201–10
Violence in Greek drama
37
could both cause ritual pollution (Parker [48] 1983: 74–143), and both were prohibited in sanctuaries (as was childbirth);22 when we find that (simulations of) both are also avoided in drama—which was performed in a sacred place, during a festival, and in the presence of the god’s image, altar, and priest—the likeliest explanation is that we are dealing with an extension of this taboo. How deeply entrenched it was is well exemplified by a passage in Aristophanes’ Frogs (1079–82) where, in a catalogue of the shocking immoralities of Euripidean drama, ‘giving birth in temples’ is given precedence over brother–sister incest.23 Thus the avoidance of killing in tragedy is probably to be ascribed to religious considerations. What of the other form of violence that tragedy avoids, striking? There can be no question here of a religious prohibition: blows are frequently struck in comedy, and in a famous surviving scene (Ar. Frogs 615–73) Dionysos himself is subjected to a flogging. Yet in tragedy, full though it is of quarrels and angry words, no one (that we know of) ever inflicts a blow on another person24— with one remarkable exception, to which we will come. As in the other cases we have considered, there are some close approaches, but the line is never crossed,25 as may be seen by an examination of the passages concerned.
(where the Scythian archer takes the dancing-girl Elaphion somewhere just offstage, and returns, highly satisfied, nine lines later). 22 Whose onstage presentation, even had there not been religious considerations telling against it, would almost certainly have been judged impracticable. 23 Herodotos (2.64), noting that apart from the Greeks and Egyptians almost no nation known to him prohibits sex in sanctuaries, and referring to the plausible argument that such acts cannot be offensive to the gods since animals and birds freely mate on sacred sites, feels impelled to add ‘These people [i.e. barbarians other than Egyptians], citing arguments like this, do what I for my part find distasteful [emphasis mine].’ 24 Characters and choreutai may, however, strike themselves; breast- and headbeating are sometimes explicitly attested in tragic texts (e.g. Aesch. Pers. 1054; Cho. 30–1, 423–8) and may have been common in other passages of lamentation (significantly called Œ at least as early as Aristotle, Poet. 1452b24); see Foley (1993). 25 It has often been thought that the line is in fact crossed in the prologue of Euripides’ Herakleidai. The confrontation of the Herald and Iolaos ends with the latter lying sprawled on the ground (75–6), and recent editors (Wilkins 1993; Kovacs 1995; Allan 2001) speak of his being ‘knocked’ down; but ¼ ææ’ ‘begone’ (67) is an odd word to accompany a blow, and it is more likely that the Herald merely drags or pushes Iolaos out of his way and the old man stumbles and falls.
38
Violence in Greek drama
In Aeschylus’ Suppliants (825ff.) the Egyptians who attempt to seize the Danaids, with a view to forcing them into marriage to their cousins, make some extremely colourful threats, ranging from dragging by the hair (884, 909) to decapitation (840) and including the promise to ‘get you sit-[49]ting in the boat, streaming blood’ (847); yet when the Argive king comes to their rescue he can threaten the Egyptian herald with condign punishment ‘should you lay a finger on them’ ( N łÆ ØÆ, 925), implying that they have not at that moment been touched. Of menacing choreography there has doubtless been plenty; of violent contact, none. Towards the end of Aeschylus’ Agamemnon a fierce quarrel breaks out between Aigisthos and the chorus of Argive elders. Aigisthos orders his bodyguard to have their swords at the ready (1650–1); the chorus may prepare to mount a doomed defence with the nearest thing they have to a weapon, their walking-sticks (1652)—but at the moment when the storm is about to break, Klytaimestra intervenes, begs Aigisthos to desist (1654–6), and sends the elders home (1657ff.).26 When Oedipus learns that he is not, as he had thought, the son of Polybos of Corinth, but had been brought to Corinth as a baby by a shepherd who had, in turn, been given him on Mount Kithairon by another shepherd working for Laios, he orders the latter to be summoned and interrogates him. The Theban shepherd, however, not only proves reluctant to answer his questions, but is furious with the Corinthian who, without being asked to, answers them for him, and he at least threatens the Corinthian with violence (Soph. OT 1146), as is shown by Oedipus’ urgent response (‘Stop, don’t hit him’, 1147). It is not clear from this passage alone whether a blow is actually delivered, but the Corinthian does not cry out (indeed, he never speaks again), and since both men are aged (990, 1001, 1009, 1013, 1112–13, 1121, 1147, 1153) it is a reasonable supposition that 26
There is a somewhat similar confrontation between the chorus (old men again) and Kreon in Sophocles’ Oedipus at Kolonos (835–86), when Kreon’s men have seized Antigone and he himself is attempting to seize Oedipus. The chorus seem to crowd around Kreon (842–4, 856), and one can well imagine (though nothing in the text makes it certain) that they brandish their sticks; but they fail to prevent the seizure (Kreon’s last words in the scene, oæØ, Iºº’ I ŒÆ (883), mark his apparent triumph) and can only cry out for assistance, which rapidly arrives in the shape of Theseus.
Violence in Greek drama
39
the Theban does no more than raise his fist or staff. A little later, still proving recalcitrant, he is threatened by Oedipus with beating or torture27 (1152–4); the threat is enough to crack his resistance for the moment, though it has to be renewed each time another question is put (1158, 1166).28 [50] In Euripides’ Andromache, Andromache and her son are about to be led off to their deaths by Menelaos when Peleus arrives (Andr. 545) in response to a message she had sent him some time earlier (79–90). She supplicates him to save her, and he at once orders the attendants guarding the condemned pair to release them ‘before someone gets hurt’ (577–8). Menelaos tells them to disregard the order, and says to Peleus ‘you will never take her out of my hand’ (587); to which Peleus replies ‘Yes, I will, when I have bloodied your head with this staff ’ (588).29 On these words he no doubt brandishes the staff, but he is not within striking range, as Menelaos’ taunting reply shows: ‘Touch me, and you’ll know—come closer!’ (589). The confrontation thereupon turns into a verbal agon, which as usual30 leads to no decisive conclusion; but then Peleus simply orders the attendants out of the way (715–16) ‘that I may discover whether anyone is going to stop me untying this woman’s hands’—and no one, Menelaos included, makes any attempt to interfere. More than a decade later, in Helen, we find another instance, on a smaller scale, of the same pattern. Theoklymenos, rushing into the palace to take revenge on his sister Theonoe for helping Helen and Menelaos escape, is obstructed by a slave of Theonoe,* who seizes his robe (1629) and will not let go. ‘Get out of my way’, says Theoklymenos (1628); ‘I won’t’, says the servant—and then for nine lines nothing happens, as the two conduct a hemistichomythic verbal fencing bout. The action unfreezes when Theoklymenos says ‘You 27
To which, as a slave (1123), he is legitimately liable. It is possible that some form of physical pressure other than blows was applied to the old man (see n. 8 above). 29 ŒÅÐæ can denote either a ruler’s sceptre or an old man’s walking-stick; Peleus is in fact still a king (Andr. 22–3), but during the rescue-scene (545–765) his age is mentioned fourteen times and his status as a ruler only once (and that after Menelaos’ departure: 759–60). In IA 311 (see below) it is another matter, since Menelaos is not an old man. 30 ‘The agon in Euripides rarely achieves anything’ (Lloyd 1992: 15). 28
40
Violence in Greek drama
seem to long for death’ (1639), to which the slave gives a resolute reply, and at that point the Dioskoroi appear. There is no sign that any blow has been struck. The even briefer confrontation of Menelaos and the Old Man in Iphigeneia at Aulis (309–16) has features reminiscent of both the scenes just mentioned. The Old Man (a servant of Agamemnon), on his way from Aulis to Argos with a letter countermanding Agamemnon’s previous instructions to send Iphigeneia to Aulis, has been intercepted by Menelaos, who has seized, opened and read the letter. The Old Man tries to grab it back, and a tug of war ensues; Menelaos, in almost the same [51] words Peleus had used—to Menelaos!—in Andromache, threatens to ‘bloody your head with my sceptre’; the Old Man, like the servant in Helen, retorts that it is glorious for a slave to die for his master. A moment later, however, he is appealing to Agamemnon for assistance, and Menelaos is evidently in firm possession of the letter. Had Menelaos struck the Old Man, the latter would certainly have complained to Agamemnon of this assault by Menelaos on a slave not his; so we can safely assume that there was again no blow, and that Menelaos was able to tear the letter out of the Old Man’s hand (as he had already done once before) simply because he was the younger man and had the stronger arms and fingers. Even in the late Rhesos the convention continues to be observed. The author carefully keeps the Greeks and Trojans apart, for all but seventeen lines of the play. When the Greek raiders, Odysseus and Diomedes, arrive on the scene (565), even the chorus are absent, and Dolon has already been killed. Paris is seen approaching at 627, but Athena tells Odysseus and Diomedes to ignore him, and gets rid of him herself by providing some disinformation31 while they are slaying Rhesos offstage. But then, unexpectedly, they do meet the enemy onstage after all, as the chorus return with menacing words and actions; but menaces is all they are, because the chorus, not knowing for certain whether they have met friends or foes, cannot attack until they have asked for the password (684, 688)—and Odysseus knows it. I mentioned that there was one great exception to the rule that no person strikes another in Greek tragedy. This is the impaling of 31
Pretending to be Paris’ guardian goddess Aphrodite, she assures him that the rumour he has heard, that there are Greek spies in the Trojan camp, is false (661).
Violence in Greek drama
41
Prometheus. Not the clamping of his arms, chest, and legs ([Aesch.] Prom. 55–62, 71–81); that is merely a more drastic version of the common enough actions of seizure, arrest, or physical restraint. But the driving of a wedge ‘right through his chest’ (64–5) is unique: not only is it a blow,32 but if the victim were not immortal, it would be a fatal blow. [52] Commentators and critics do not appear to have taken on board the uniqueness of this action not only in Aeschylus but in all tragedy. They have, of course, tried to explain how it was managed theatrically that Prometheus should seem to be impaled, but not why it was acceptable for the audience to see him seemingly impaled, in breach of so strong a convention. Whether or not Prometheus Bound is by Aeschylus, it is certainly a fifth-century play, and was almost certainly performed in Athens.33 Either, then, the convention that prohibited the striking of blows in tragedy was for some reason not applicable in this case, or it was deliberately broken for effect. This question cannot be decided separately from that of how to explain the convention itself. If, for example, the avoidance of striking was based on religious considerations or on a competition rule, it would have been no more possible to break the convention than it would have been to have a fourth speaking actor in tragedy or to present simulated sex on stage in comedy; if it was a purely artistic norm, an occasional breach of it might well be risked if it was dramatically effective, as might (say) the use of lexemes from an undignified stratum of vocabulary (Sommerstein 2002), or a major, signalled change of imaginary location (such as occurs in Aeschylus’ Eumenides and probably34 Sophocles’ Aias). 32 Probably only one. Hephaistos, though reluctant to perform his terrible duty, finishes each task with remarkable speed once he has begun it (e.g. securing Prometheus’ legs, P ÆŒæø fi Ð ø fi , in the interval between lines 74 and 75); evidently he works with divine ease, and one hammer-stroke therefore ought to be enough for the wedging. The arm clamps need more than one (cf. 58), but that is because Hephaistos does not at first put forth his full strength (as Kratos warningly complains). 33 Aristophanes adapts Prom. 59 and 613 in Knights 759 and 836 respectively (425/4), and Kratinos’ Ploutoi, produced a few years earlier (probably in 429: West 1979a: 141; Kassel and Austin 1983: 204, 206), is based to a substantial extent on Prometheus Unbound (cf. especially Kratinos fr. 171). 34 Scullion (1994: 109–28) argues that there is no change of scene at Aias 814; against this, see Garvie (1998: 203–4).*
42
Violence in Greek drama
In fact, of the five explanatory strategies (1–5) above, we can quickly rule out a religious one (1); for the occasion and the environment of performance were no less sacred in the case of comedy than of tragedy, yet there is no restriction on the striking of blows in comedy. For the same reason, the convention can hardly be a matter of theatrical practicality (3).35 One might fleetingly think that it could be a historical hangover, due to inertia (5), from the days of single-actor drama, when it would not have been possible for one character to strike another. However, even in single-actor drama, violent confrontation between a character and the chorus would still have been a possibility; as we have seen, this is avoided by a [53] whisker in Agamemnon, and it probably occurred from time to time in comedy.36 And in any case, such an explanation would, as it were, explain too much: in single-actor tragedy it would have been equally impossible for one character to seize, hold, or drag another, yet as we have seen this becomes quite frequent at least from the 440s onwards. It is hardly plausible prima facie that there should have been a rule of the tragic competition (1) prohibiting the striking of blows; in any case such a rule, if it had existed, would have had to be observed without exception—and the violation of the convention in Prometheus is about as total and as spectacular as it could possibly have been. Thus the only kind of explanation available to us is an aesthetic/ artistic one (4): tragic dramatists felt (and/or expected their audiences to feel) that the infliction of blows by one dramatis persona on another would in general not be appropriate to the type of effect
35 See, however, n. 37 below on the problems of striking blows when wearing masks. 36 In the surviving comedies of Aristophanes there are many threats of violence by the chorus against individual characters or (less often) vice versa; there is no absolutely clear case of such a threat being carried out (though water is thrown over the chorus at Peace 969, and one half of the chorus do likewise to the other half at Lys. 381–5), but Olson (2002: 153) is willing to accept that in Ach. 280–3, when the Acharnians break up Dikaiopolis’ phallic procession and put his daughter and slaves to flight, they have ‘let loose a wild volley of stones’ (‘perhaps represented by bits of leather or the like’, he prudently adds), and it is hardly likely that these all missed their targets.
Violence in Greek drama
43
which, as tragic dramatists, they were trying to create,37 and therefore avoided presenting it—while implicitly reserving the right to do so in exceptional circumstances. If so, the author of Prometheus Bound evidently considered that the circumstances of his play were exceptional. He was already committed to showing Prometheus being subjected, on stage, to a bodily punishment equivalent to the Athenian penalty of apotympanismos (cf. Sommerstein 1996a: 309–10) though in one way even greater inasmuch as it could not be terminated by death and was indeed intended to last for ever (26–7); but he could have done this—and had he not written verses 64–9, he would have done it, very neatly—without violating the no-striking convention. But this was not enough for him. Prometheus’ agonies must be [54] shown to exceed all human bounds not only in their duration but also in their intensity: he must be shown suffering, over an extended period, the pain of a wound that no mortal can imagine because no mortal could survive it. Moreover, if this wound is to be inflicted at all, it must be inflicted onstage. Prometheus is to spend the entire play onstage clamped and pinned to a rock, and therefore we have to see the clamping and pinning being done. In principle, to be sure, the play could have begun (as Prometheus Unbound presumably did) with Prometheus already in his bonds, and he could have informed us in a prologue of what had just happened to him; but then we would lose the opening scene, and the play would seem even more static than it already is. We would also lose the vivid sense of the ruthlessness of Zeus’s tyranny conveyed by the implacability of Kratos and Bia,38 nor would we see how extremely reluctant Hephaistos is to carry out the punishment of Prometheus’ crime even though (as Kratos reminds him, 7–8) he was the victim of it if anyone was. Whether this artistic decision by the author was a correct one is
37
It may be relevant that, as Eleanor OKell points out to me from experience, it is difficult to place a blow with precision when wearing a mask. In tragedy, a blow that went wild might produce a ludicrous effect, and one that connected too well (and hurt) risked breaking the boundary between fiction and reality, which tragedy in general strictly respects. In comedy, one might argue, the latter would be no great problem, and the former might actually be an advantage. 38 Made manifest, we must suppose, by the appearance of both of them, as well as by the words of the former.
44
Violence in Greek drama
another matter—a matter, in the first instance, for the audience and the competition judges to decide; and we do not know what their decision was.39 What we can say is that the author’s decision is of a piece with what he does in several other scenes of the play in which he goes close to or beyond the normal limits of tragic practice for spectacular effect—for example the ‘flying’ entry of the chorus, the gratuitous use of the mechane and of a ‘four-legged bird’ in the Okeanos scene, and the swallowing up of Prometheus and his rock at the end of the play—and that whether or not this approach won him success in the competition, it did not win him imitators. We may conclude, then, that the cliche´ with which we began can be disassembled into two conventions. The first—based on religious considerations, and unbreakable—was that the audience must not, in any form of drama, witness any act or occurrence which impinged on a human or animal body so as to be the proximate cause of a death. The second convention applied only to tragedy; it was based on artistic considerations, [55] and we find it broken in one production which we know to have been exceptional in other ways. It was that the audience must not, in tragedy, witness any person inflicting a blow on any other person. We can safely assume that the tragic dramatists were fully conscious of this convention, at least by the time of Aeschylus’ maturity: Aeschylus, Sophocles, and Euripides can all be found teasing the audience, as it were, with the possibility that they might violate it, while being careful not to transgress the limits the convention set. We do not know who first made it part of the tragic dramatist’s style manual, or why. One obvious possibility to consider initially is that the object was to draw a clear distinction between tragedy and satyr-drama, especially once they had begun to be produced together by the same poets as part of the same competition; but this is unlikely in view of the fact that in satyr-drama too, so far as our evidence goes, the infliction of blows was normally
39 We know that Aeschylus’ son Euphorion won four victories with suites of hitherto unperformed plays which he presented as the work of his father (Suda 3800), and much the least implausible hypothesis about the authorship of the Prometheus plays is that they were written by Euphorion as part of such a production (West 1990b: 67–72). We cannot be sure, however, that Euphorion won first prize every time he used this ploy.
Violence in Greek drama
45
avoided.40 Perhaps at first it was simply taken for granted that performers in the early, simple tragic and satyric performances would not hit each other (after all, they certainly would not have done so in the pre-dramatic choral performances from which these genres originated), and this tacit assumption was only articulated as a conscious convention when there appeared on the scene the new genre of comedy, which, like the ŒøÐ Ø from which it took its name, had no such inhibitions.41
ADDENDA p. 34 n. 13 it has been plausibly suggested by Scott Scullion (see Morwood 2007: 219–20) that what the chorus can see is not the pyre itself but the smoke rising from it. p. 39 the speaker whom I, like Diggle (1994), Kovacs (2002), Burian (2007: 288–9), and others, have identified as ‘a slave of Theonoe’ (or ‘Second Servant’ or the like), is labelled . by the sole manuscript, and this attribution to the chorus-leader has been defended by several recent critics (e.g. Dale 1967: 165–6; Kannicht 1969: ii. 422–4; Taplin 1977: 90; Allan 2008: 338). This speaker, however, is a person who is in some sense a slave both of Theoclymenus and of Theonoe (using the word Å in reference to both: 1627, 1640) and who has at least as close a connection with the latter as with the former (being ready to die defying Theoclymenus for Theonoe’s sake: 1640–1); whereas the chorus have not at any time during the play given any indication that they consider themselves 40 In the satyric corpus of Krumeich, Pechstein, and Seidensticker (1999)—plus Euripides’ Kyklops, whose text they do not include—there is no passage or fragment indicating clearly or even probably that someone is or has been struck, on stage, by another person (statements about intentions for the future are not adequate evidence, in view of the frequency of unrealized threats in all the dramatic genres). The nearest approach is at Eur. Kykl. 683–4, where the blinded Polyphemos hits his head on a rock (part of the skene wall?) as a result of being misdirected by the chorus. 41 This chapter was originally published in Ordia Prima 3 (2004) 41–56 (the last page contained only bibliography). I am most grateful to Ediciones del Copista for giving permission for this republication.
46
Violence in Greek drama
Theonoe’s slaves in any sense at all, never speaking of her as their mistress but always as ‘the young woman’ or ‘the maiden’ or ‘the Nereid’s daughter’ or ‘the prophetess’ (318, 324, 328, 515, 996). He is perhaps an old family slave (his mask and hair will have made his age evident) who had once served the virtuous Proteus. p. 41 n. 34 Scullion’s thesis is no more satisfactory in the modified form offered by Heath and Okell (2007), who suggest (p. 364) that the grove in which Ajax’s body will be found, part of ‘the wild, unoccupied territory beyond the camp’, is represented by one of the flanking doors of the skene concealed by appropriately painted scenery. Ajax would not choose for his suicide a spot within sight of his own hut; that, indeed, would be inconsistent with his declared intention to go to an untrodden place where he would not be seen by anyone (Aj. 657–9), nor could he have known in advance that Tecmessa and the chorus, warned by a messenger that he had deceived them, would depart from the camp in search of him, thus conveniently leaving the area clear for him to return to. Whether we assume with Scullion that Ajax does not go to the seashore at all, or with Heath and OKell that he goes there and then comes back to the grove, we would have to put him down as a fool whose careful plan to deceive his friends and evade their efforts to save him depended for its success on a flukish coincidence. See also Finglass (2009: 273). Finglass (forthcoming) notes further, as Lobeck had done in 1835, that Teucer’s alarmed and urgent response to the information that Eurysaces is ‘alone by the hut’ (Aj. 985–9) would be absurd if the hut were only a few steps away; Scullion’s counter-argument (1994: 123–4) effectively elides the words Ææa ŒÅ ÆØÐ Ø as if they did not exist, while Heath and OKell seem to elide the preceding word when they say that Eurysaces ‘is brought to [Teucer] from where he is being kept safe [emphasis mine: AHS] in the camp’ (Heath and Okell 2007: 374 n. 36).
3 Adolescence, ephebeia, and Athenian drama For nearly three decades now, it has been increasingly difficult, in studies of Athenian drama, to avoid the words ‘initiation’ and ‘ephebeia’. Almost every play in which an adolescent male appears, from Aeschylus’ Choephoroi1 to Euripides’ Bacchae,2 has at one time or another been declared to reflect an initiation rite or an ephebic ritual. The late John Winkler has argued,3 bringing some highly persuasive artistic evidence, that the choruses of tragedy were themselves composed of ephebes, and that the very word ‘tragedy’ means etymologically ‘the ephebes’ song’; while in one of the best books on Aristophanes in the last twenty years—Angus Bowie’s Aristophanes: Myth, Ritual and Comedy—three of Aristophanes’ plays (Knights, Clouds, and Wasps) are analysed as comedies of ephebeia, but in two of them it turns out to be a ‘reversed ephebeia’, the initiation not of an adolescent but of an old man (Strepsiades and Philokleon).4 I cannot recall having seen it suggested—not yet—that some plays may present (what shall we call it?) the ‘transgressive ephebeia’ of a woman, but doubtless that is only a matter of time. I suggest that Oedipus at Kolonos might be a fruitful subject for this type of investigation.*
Vidal-Naquet (1969/1990: 154–6); Zeitlin (1978: 160–1 ¼ 1996: 98–100). Especially by Richard Seaford; see Seaford (1981) and (1996), where references to relevant commentary notes are collected on p. 42 n. 70. 3 Winkler (1990). 4 Bowie (1993: 45–112). 1 2
48
Adolescence, ephebeia, and drama
Those who have not yet been, shall we say, initiated into these discussions may be somewhat surprised at the apparent ubiquity of initiation and ephebeia in fifth-century drama. In one respect their bewilderment will soon disappear, as they discover that ‘initiation’ is being used in two different senses. In one it refers to the rites of mystery-cults like that of Eleusis; this is the sense in which the term is used, for example, in the admirable discussion of Aristophanes’ Frogs by Ismene Lada-Richards which <was presented at the 1996 Classical Association conference>.* There is also, however, another sense of ‘initiation’ which has not till recently been part of the vocabulary of classical scholars but is familiar to anthropologists, referring to rituals associated with the transition from childhood to adult membership of the community. And in this sense, in an ancient Greek context, ‘initiation’ and ‘ephebeia’ mean, at least for males, much the same thing. From now on, to avoid confusion, I will use only the terms ‘ephebeia’, ‘ephebe’, and ‘ephebic’ in relation to rituals of this kind applying to adolescent males in ancient Greece. I will use them whether or not there is evidence that these specific terms were applied to the rituals in question by contemporaries. But having cleared that mildly red herring out of the way, some simple folk may still find themselves wondering about other things. In particular, if sufficiently long in the tooth, they may recall that when they were students no one ever mentioned ephebes or ephebeia in connection with fifth-century Athens, and wonder whether someone has uncovered new evidence, inscriptional or artistic or both, that the system of ritual and training for 18–20-year-olds, described in the 320s, in elaborate detail, by the author of the Ath.Pol.,5 and attested in changing forms by massive epigraphic, artistic, and literary evidence from his time almost to the end of antiquity (to be precise, from 334/3 bc until ad 265/6),6 was already in existence a hundred years or so earlier. In case there are some who are worried that they may have missed this exciting new evidence, I should reassure them at once. There isn’t any. The nearest thing to it was an inscription published in 1965 (SEG xxiii. 78), recording two honorary decrees of the tribe 5 6
[Arist.] Ath.Pol. 42.2–5. See Reinmuth (1967).
Adolescence, ephebeia, and drama
49
Akamantis. The first decree (honorand and subject unknown) was passed in 361/0 (archon Nikophemos). The second was in honour of a man named Autolykos who had been superintendent (kosme¯te¯s) of the ephebes; the archon’s name is lost, and the first publisher restored it as Nikophemos also, but David Lewis showed in 1973 that this was epigraphically untenable.7 There is still a complete absence of inscriptional references to the ephebeia before the year 334/3; from that moment there is an absolute flood of them. A hundred years, and thousands of inscriptions, after Wilamowitz’s Aristoteles und Athen, his basic conclusion remains sound:8 the middle 330s saw a revolutionary innovation in the way Athens inducted its young men into the citizen body. There certainly were Greek communities in the classical period— in Sparta, Arcadia, Crete, to name a few—which did have elaborate, highly structured, wholly or partly ritual patterns of transition to adulthood for boys and sometimes for girls also; in Sparta it took not less than twenty-three years, from the age of 7 to the age of 30, for a male to complete this transition. For Athens, there has been much valuable recent study of transition rituals for girls and especially of the bear ritual (arkteia) of Artemis at Brauron.9 But if there was at Athens a ritual, or series of rituals, for boys at all comparable to these or to the later ephebeia, our sources, of all kinds, for the fifth century (and indeed most of the fourth as well) are completely silent about them.10 All they speak of is intermittent military training and guard and patrol duties, later supplemented by gymnastic training. There is no evidence whatever that young males in the fifth century (or indeed in the first half of the fourth) were perceived as passing through a ‘transitional’, ‘liminal’, ‘ephebic’ stage of life between the ages of 18 and 20; indeed there is positive evidence that they were not so perceived but were regarded as having the full rights and responsibilities of adult citizens as soon as they underwent dokimasia and were registered in their demes at the age of 18. Those who
7
Lewis (1973: 254). Wilamowitz-Moellendorff (1893: i. 192). 9 See especially Sourvinou-Inwood (1988).* 10 See Sommerstein (1996d: 53–9) <¼ Sommerstein (2009: 192–7)>, where the statements made in this and the next two sentences are documented. 8
50
Adolescence, ephebeia, and drama
doubt this may care to ask themselves a simple question: what is the fifth-century Attic Greek word for an ephebe? I ought at this point to give a brief sketch of what I do believe to be the history of the institution that was transformed, in the 330s, into the two-year ephebeia. From the time of Kleisthenes, if not earlier, an Athenian male attained adulthood at the age of 18 and became liable for military service. He must always have needed training before he became capable of serving as a hoplite in battle. It could well have been in Kleisthenes’ own time, or very soon after, that the practice was established whereby, after registration in their demes, the new citizens, in military dress and equipment, swore what was later called the ‘ephebic oath’11 and then, during some part of the next two years, underwent hoplite training and served in the frontier guard (or, in emergencies, anywhere). This system continued without substantial change throughout the fifth century. When not actually performing military duties the young men lived where they would and did what they wanted, with the right to speak in the Assembly, to bring prosecutions in the courts, to get married, and so on.12 Some time in the first half of the fourth century there was a change. Aischines clearly regarded his time in the youth corps, in the late 370s, as an important epoch in his life; he refers to it in two of his three surviving speeches, and uses it to demonstrate that one of his then companions, Misgolas, is not as young as he looks.13 Then from Xenophon’s Poroi (4.51–2) we discover that by 355 the time of training was divided into two periods, the first in gymnasia, the second on light military duties, and that when engaged in training activities the young men were supposed to receive a maintenance allowance from the state, but that recently (doubtless owing to the Social War) they had not been receiving the allowances regularly and as a result attendance at training had fallen off. Even in 355, however, Xenophon has no name available by which he can designate the trainees, and has to speak, clumsily, of ‘those who are designated 11 This oath survives in three versions, one inscriptional (Tod ii. 204,* from Acharnai) and two quasi-literary (Pollux 8.105, Stobaios 4.1.8). There is an excellent discussion of it by Siewert (1977), who inter alia demonstrates its archaic, predemocratic origins. 12 See Sommerstein (1996d). 13 Aischines 1.49 (Misgolas), 2.167.
Adolescence, ephebeia, and drama
51
for gymnastic training . . . those who are designated for fortress garrison duties’. The word ‘ephebe’ therefore became established in Attic between 355 and 346/5, when Aischines uses it, retrospectively, as though it was familiar to everyone.14 Finally in 335 came the major reorganization15 whereby ephebic training became a virtually full-time activity and the ephebes lived a semi-segregated life for two years, under specially chosen supervisors and trainers, wearing a distinctive uniform, taking their meals in common, taking part collectively in various festivals, exempt from all normal civic duties, and forbidden to engage in litigation except regarding inheritances. There is no evidence that any of these six features of the Aristotelian ephebeia existed before 335; and if we take them away we are left with something that Oscar Reinmuth16 had good reason to call ‘a purely military organization’, and one that, to judge from the extreme scantiness of references to it in sources such as comedy and Plato, was not considered by most Athenians to be of any great social or civic significance. This is not to deny that some of the plots of fifth-century drama, especially tragedy, may be ultimately based upon rituals that were ephebic in the broad sense I have indicated. Indeed some of them almost certainly were. We know that many Greek communities had elaborate ephebic rituals in the classical period; it is highly likely that many more did in early times; and it is virtually certain that most of these rituals had myths associated with them.17 Indeed I well remember listening with pleasure, thirteen years ago,
52
Adolescence, ephebeia, and drama
another Classical Association conference at Nottingham>, to Richard Buxton’s masterly demonstration of this, apropos of wolves and werewolves, principally in Arcadia.18 We can be confident, then, that the corpus of myths exploited by tragic poets in the fifth century included many that were of ephebic origin. That is not at all the same thing as saying that either the tragic dramatists or their audiences were aware of that fact. What they must have been aware of—and what one can hardly help being aware of today—is the acute interest that Athenian drama displays, especially in the last three decades of the fifth century, in the adolescent male, his education, and his socialization. I stress the chronological point. There survive complete thirty fifth-century Athenian tragedies (I omit Rhesos and Iphigeneia at Aulis). Of these, eleven were almost certainly produced before the outbreak of the Peloponnesian War, eighteen thereafter, and about one (Oedipus Tyrannos) the matter is doubtful. If we ask which of these thirty show a major concern with the adolescent male, his education, and his socialization, for the war period the list runs as follows: Sophocles, Electra (Orestes), Philoctetes (Neoptolemos) Euripides, Hippolytos, Suppliants (the Seven), Ion, Orestes (Orestes and Pylades), Bacchae (Pentheus) What would a corresponding list look like for the pre-war period? It is doubtful whether there are any surviving plays at all from that period in which the problem is addressed in anything like the way it is in the seven plays I have just mentioned. It is instructive to compare the role of Orestes in Aeschylus’ Oresteia, in Sophocles’ Electra, and in Euripides’ Orestes. In Aeschylus, Orestes is presumably more or less of ephebic age, but this is of virtually no dramatic consequence: if he had been twenty years older his dilemma and his sufferings would hardly have been different; they arise, not because he is new to adult life and ill-acquainted with its demands and difficulties (as is that greatest of all dramatic ephebes, Hamlet),19 18
Buxton (1987). Hamlet, as is well known, is 30 years old in the gravedigger scene (Shakespeare, Hamlet v. i. 138–57); but at the beginning of the play he is an undergraduate, wanting to resume his studies at Wittenberg. If Hamlet were an Athenian tragedy, scholars would probably identify Hamlet’s time at sea as a period of ephebic liminality. 19
Adolescence, ephebeia, and drama
53
but because he finds himself compelled, by Apollo and by the situation, to kill his mother. In Sophocles, we find ourselves forced to consider the issue of Orestes’ upbringing because the man who educated him and who still largely controls him, his slave paidagogos, accompanies and instructs him in two crucial scenes,20 and is always obeyed without question; while in Euripides we confront the terrifying picture of two young men and one young woman who are radically asocial—whose appealing loyalty to each other is accompanied by a total lack of scruples of any sort in relation to anyone else. Another possibly significant comparison, though more difficult because it is of a play that survives with one that does not, is that between Euripides’ two Hippolytos plays.* In the first play, if Phaidra was anything like the shameless wanton that contemporaries who had seen the play, and ancient scholars who had read it, agree in describing, Hippolytos must have been essentially a victim. If he is not experienced enough to fight back effectively in the way Bellerophon apparently did in Euripides’ contemporary and partly parallel play Stheneboia, that can hardly be put down to faulty education. In the surviving Hippolytos, on the other hand, Hippolytos’ upbringing in the traditional and much-praised pursuits of aristocratic adolescents, hunting and chariot-driving, has left him thoroughly ill-prepared for his future roles as husband and responsible citizen, with a contempt and loathing for those of different background (including all women) that do much to bring about his ruin. His father may be absurdly wrong to suppose him an Orphic and a vegetarian,21 but he certainly does see himself as one of a small circle of the elect set against, and superior to, the vulgar ‘mob’—in which he includes his father.22 The ephebic and educational features of the other tragedies I have mentioned will be sufficiently obvious:* the self-discovery of Neoptolemos as he wavers between the influence of two, or maybe rather three, rival preceptors—the tempting, corrupting Odysseus, the
20 21 22
Soph. El. 1–85, 1326–75. Eur. Hipp. 952–4. Ibid. 78–81, 986–9.
54
Adolescence, ephebeia, and drama
honest, pathetic Philoktetes, and the image of a father he has never met; Adrastos’ funeral oration on the Seven against Thebes,23 with its concluding claim that virtue depends vitally on education; the sudden exposure of the sheltered, naive Ion to shocking facts and actions that make him ready to violate the sanctity of the same temple whose pious servant he has been all his life; and a Pentheus bearing the responsibilities of kingship when too young to exercise them prudently, with catastrophic consequences for himself and Thebes. During the same period education is also a major concern in several of the comedies of Aristophanes. The education of the young is central in Clouds, as it was in the very first of Aristophanes’ plays, the lost Banqueters; and by comic inversion, as Bowie in particular has seen, the education of the old (Demos, Strepsiades, Philokleon) is a theme of Knights, Clouds (again), and Wasps.24 These plays all belong to the 420s, but a decade or so later, in Birds, two scenes late in the play25 introduce young men whose upbringing has gone wrong (leading one of them to contemplate murdering his father, and the other to make a living as a sykophant); the former (this being comedy) proves redeemable, the latter is whipped off the stage. Why were both tragedy and comedy in this period so much concerned with the socialization of the adolescent male? I wish to suggest that, rather than reflecting the pattern of an existing Athenian ephebic transition-scheme, what they reflect is contemporary anxieties arising precisely from the fact that such a scheme did not exist. If one tries to generalize the relevant themes of the plays we have described, most of them tend to fall into two groups. One is that of the youth whose education, often of a traditional type, has left him unable to cope satisfactorily with the realities of adult life: such are Hippolytos, Ion, Pentheus. The other is that of the youth whose education has led him to cast aside all normal social duties and loyalties: sometimes, like Orestes in the play of that name, he lives only for himself and a few of those like him; sometimes there is a 23
Eur. Supp. 860–917. See Bowie (1993) and Sommerstein (1996d: 59–64) <¼ Sommerstein (2009: 198–202)>, with the paper by Niall Slater (1996) to which it is a response. 25 Ar. Birds 1337–71, 1410–69. 24
Adolescence, ephebeia, and drama
55
corrupting teacher who for loyalty to gods, parents, and polis has substituted loyalty to him, the teacher, as in Clouds, Philoktetes, and Sophocles’ Electra. In at least two plays known to us, the two patterns were combined. One was Aristophanes’ Banqueters with its two contrasted youths, products of traditional and sophistic training, the ‘virtuous boy’ and the ‘buggered boy’, the latter (who absconded from the school to which his father sent him) running rings round the former in argument.26 In the other, Sophocles’ Philoktetes, the two schemata are both focused on the same character: Neoptolemos is taken from his sequestered home on Skyros (which in the end he wishes he had never left)27 to become a hero and a leader of men when hardly out of boyhood—and also to become, willy-nilly, the pupil of the amoral Odysseus. These patterns can be seen to correspond to well-marked and acute anxieties which are known from other sources to have been present in the Athens of this period, anxieties which have more than once been summed up in the phrase ‘generation gap’28 and which arose precisely from the fact that the inherited Athenian educational system had ceased to meet the needs of society. That system of paideusis was designed primarily for paides—those under 18. Its ‘musical’ component prepared the young to perform their religious functions as members of choruses and to play their part in the social rituals of the symposium, and taught them their ethical duties through the medium of archaic poetry; its ‘gymnastic’ component trained their bodies, directly for athletic competition and indirectly for war. It offered, however, no preparation at all (except basic literacy) that would train a boy for his future role as citizen of a democratic community, and none that would protect him against seductive creeds—new philosophies or alien religions—which might lead him away from the proper performance of that role. The result was that boys who had been very well educated for life in the Athens of the mid-sixth century were being pitchforked at 18 into the Athens of the late fifth, with the right to marry, to speak in the Assembly, to squander their property, and to do all other acts and things which free and independent Athenians might of right do. 26 27 28
Ar. Clouds 529; Ar. frr. 205, 206, 233. Soph. Phil. 969–70. See Forrest (1975); and Handley (1993).*
56
Adolescence, ephebeia, and drama
Meanwhile those who could afford it were looking for teachers who could prepare their sons for life in the real civic world; and those who claimed to provide such preparation were numerous. But there was little agreement among them on principles or methods, and in general they were deeply suspect, especially to the majority who could not afford their fees. We need not perhaps blame the teachers as much as contemporaries did. Fathers no doubt wanted their sons to be good at looking after the interests of the community, but it was even more imperative that they should be good, in a world full of rivals and of sykophants, at looking after themselves. Unfortunately the latter objective sometimes got in the way of the former; and that this is no mere prejudice of crusty contemporaries is proved by the events of 411 and 404/3. Twice the political system was overthrown by members of that section of the community which had gone in most enthusiastically for the new educational trends; and on both occasions, if the leaders of the revolutions tended to be middle-aged, there were plenty of the younger generation who were willing and eager to serve as its storm-troops. In 411 they provided the deathsquads which removed from circulation those who expressed strongly democratic views in public, and the ‘hundred and twenty young men whom [the Four Hundred] employed if there was any physical work needing to be done’;29 in 404/3 they supplied the guards who scared the council into silent acquiescence in the murder of Theramenes,30 as well as the cavalry who came to be so much reviled that, in the 390s, even to be suspected of having served among them31 was the kiss of death to a political career, while (as Socrates found) to be suspected of having taught some of them was the kiss of death, full stop. In the late fifth century both the old and the new education appeared to be deeply flawed. The products of the old system were ill prepared to resist predators and parasites; the products of the new were only too well prepared to be predators and parasites, if they were that way inclined; and if the former had an uncritical attitude towards traditional wisdom, the latter might have an equally 29 30 31
Thuc. 8.66.2, 8.69.4. Xen. Hell. 2.3.23, 50–1, 55. Like Mantitheos, the speaker of Lysias 16.
Adolescence, ephebeia, and drama
57
uncritical attitude towards the thoughts of their particular guru.32 The drama of the period duly reflects these uncertainties. It does not itself offer any escape from them. Some form of training for civic life is clearly a necessity if one is not to meet the fate of Hippolytos or Pentheus—or indeed of the Seven against Thebes, men of high virtue (as Adrastos describes them) who nevertheless, when among the leading figures in Argos, not only failed to prevent but actually incited the launching of an impious and doomed expedition. (They may have had a good upbringing, but none is described as having had any kind of intellectual education.) On the other hand, the wrong kind of training can be even more disastrous, as Orestes and Pheidippides demonstrate. Is there a right kind of training? In extant drama only Philoktetes seems to offer some degree of hope, as personal association with an older man of high character,33 without any formal teaching, leads Neoptolemos to (re)discover his true self; but it is by no means clear that this happy outcome would have ensued had not Neoptolemos had in addition both the inherited endowments and the inspiring example of Achilles to help him find the right road. Some of those who associated with Socrates notoriously failed to find it. What was needed was to create a new system. But what form could that system take? Sparta was, or seemed to be, the current model of a successful polis, and some eyes were turned in that direction; but its whole social system was so radically different from the Athenian as to make the transplantation of the Spartan agoge to Athens utterly impracticable, though some Athenians in the fourth century did send their sons to undergo it at Sparta,34 and some intellectuals sketched out model educational systems with a Spartan tinge. Xenophon’s Persians, in the Cyropaedia, have a 10-year ephebeia (from 17 to 27), during which they live as if in camp, practise archery, 32
This is beautifully skewered by Aristophanes in Clouds 1432; when Strepsiades points out an inconsistency in his son’s exploitation of the nomos/physis antithesis (he wants to behave like a cockerel when it comes to assaulting his father, but not when it comes to eating dung and sleeping on a wooden perch), Pheidippides replies ‘It’s not the same thing, man, and Socrates wouldn’t think it was [emphasis mine].’ 33 A figure reminiscent of the ideal archaic erastes/mentor, as exemplified, over a century before, by the relationship between Theognis and Kyrnos. 34 Such as Phokion (Plut. Phok. 20.4).
58
Adolescence, ephebeia, and drama
spearmanship, and (of course) hunting, and carry out light military and police duties when ordered.35 Plato in the Laws36 creates a sort of Young Pioneer Corps carrying out what would now be called ‘defence and civil infrastructure projects’ in the countryside for two years between the ages of 25 and 30; some features in the description of their activities and lifestyle hark back to Sparta (they are even at one point called by the Spartan term kryptoi) and/or foreshadow aspects of the Athenian ephebeia as it was remodelled twelve years after Plato’s death (this is especially true of the five agronomoi, who very much resemble the later so¯phronistai). But these remained paper schemes. Meanwhile teachers of rhetoric, and of elementary political science, were multiplying; but they still charged high fees, and their work, valuable though it was in training a political class, was no solution to the problem of educating a citizen body; nor indeed did even the most eloquent of these teachers, Isokrates, claim that it was. As late as the 350s, as we have seen, all that Athens could offer its 18-year-olds was a course of training extending intermittently over two years, of an entirely gymnastic and military character, ill funded and ill attended. It speaks eloquently for the inadequacy of this system that Phokion, one of the most prominent men in Athens, about this time sent his son Phokos to undergo the last two years or so of the Spartan agoge in the hope of curing his drunken and extravagant habits. It is possible that about 350 there was a financial and administrative reorganization of the training programme, and that it was in connection with this that the trainees were first recognized as a distinct group within the citizen body, given the name ‘ephebes’, and perhaps (though we cannot be certain of this) given exemption from certain civic duties, say from festival liturgies with their temptation to reckless expenditure. But we hear of no reform to the training itself until after the shock of defeat at Chaironeia and vassalage to Philip and Alexander. The pivot of the new system was the so¯phroniste¯s or ‘corrector’ who took charge of all the ephebes of his own tribe. He was a civilian, and not directly responsible for 35
Xen. Cyr. 1.2.4 and 9–12. Pl. Laws 760b–763c (infrastructure projects, 760e–761d; guard and patrol duties, 761d, 763a; common quarters and meals, 762b–c; strict obedience, 762e; austere lifestyle, 762e–763a; hunting, 763b; called kryptoi, 763b). 36
Adolescence, ephebeia, and drama
59
gymnastic or military training, which were in the hands of specialist trainers. He was elected, not chosen by lot, and by a most unusual method: chosen by the Assembly from a short list of three drawn up by a meeting of parents (or rather fathers) who had taken an oath to vote on the basis of the candidates’ ‘character and suitability for having charge of ephebes’. He took the boys on a tour of the city sanctuaries before the training proper began; he held the common purse into which all their pay was put; he ‘looked after them in all other respects’; and presumably it was expected that he would train them in so¯phrosyne¯, that quality in which Pentheus was so disastrously lacking37 and whose nature Hippolytos so disastrously misunderstood.38 If he blatantly failed, doubtless the usual procedures of magisterial audit could be applied against him. Here we see an attempt to translate into reality the idea of ‘personal association with an older man of high character’, and to teach by experience the virtues and benefits of cooperation. The educational gap perceived by tragedians and others nearly a century earlier had at last been filled. It was, of course, too late. The first new-style ephebes, 18 years old at the beginning of the Athenian year 335/4, became eligible to hold office at the beginning of the year following their thirtieth birthdays. That year was 323/2, the year when the classical Athenian democracy perished.39
ADDENDA p. 47 on ‘transgressive ephebeia’: this prophecy does not appear to have been fulfilled. p. 48 on Ismene Lada-Richards’ paper: the substance of this paper was incorporated in Lada-Richards (1999). p. 49 n. 9 see now also Giuman (1999); Gentili and Perusino (2002); and Parker (2005: 232–49). 37
Eur. Ba. 504, 1150, 1341. Eur. Hipp. 80, 731, 949, 995, 1007, 1100, 1365. This paper was delivered at the Annual Conference of the Classical Association at Nottingham in April 1996. It is published here for the first time. 38 39
60
Adolescence, ephebeia, and drama
p. 50 n. 11 this inscription is now best read in Rhodes and Osborne (2003) no. 88. p. 53 Thomas Talboy and I have discussed the Hippolytus plays, together with Sophocles’ Phaedra, in detail in Sommerstein, Fitzpatrick, and Talboy (2006: 248–89); our analysis of the lost Hippolytos Kalyptomenos (pp. 255–66) confirms that in this play Hippolytus was ‘essentially a victim’. We also argue (pp. 266–72) in support of the traditional view, taken for granted in the text above, that the lost Hippolytus was produced earlier than the extant play, against the proposal of Gibert (1997) that it was the later of the two, and we tentatively date it to the period 436–433 inclusive. p. 53 on Philoctetes see now especially Lada-Richards (1998). p. 55 n. 28 it appears that the vogue of the phrase ‘generation gap’ in classical studies itself lasted just one generation; Handley’s paper is the last one recorded by L’Anne´e philologique whose title contains this phrase (the first appeared in 1970).
4 Sherlockismus and the study of fragmentary tragedies There are three things I wish to say before I begin. The first is to express my gratitude for the honour of being chosen to deliver this lecture—an honour which I feel to be all the greater because it is taking place within the walls of a college that has ancient historic ties with the school, Merchant Taylors’,1 where I first learned Greek from—let me name them now, I will come back to one of them later—David Powell-Price, Paul Grundy, Robin Davis, Simon Cherrett, and Martin Lowry. The second is to express my sorrow at the cruel fate that has brought it about that this should be happening on the day of the funeral of a former Gaisford lecturer and an outstanding contributor to our understanding of Greek tragedy and many other things, Christiane Sourvinou-Inwood. I recall with pleasure working with her when she was a member of the editorial committee of the Journal of Hellenic Studies during my editorship. There are some things that are best expressed simply. She was a scholar. I will remember her. And the third is to apologize for my title.2 Some of you have doubtless had the experience of being required to submit an abstract for a conference paper many months in advance, only to find that as the paper develops it becomes something not at all like what 1 St John’s College, Oxford, was founded in 1555 by Sir Thomas White, a former Master of the Worshipful Company of Merchant Taylors of the City of London, which six years later established a ‘Grammerschoole’ in the City (now at Sandy Lodge, near Rickmansworth, Hertfordshire); soon afterwards Sir Thomas created a large number of scholarships at the college reserved for alumni of the school. 2 See note 69 below.
62
Sherlockismus and fragmentary tragedies
you expected it to be when you wrote the abstract. Well, I have, anyway, and it may have happened again. Let us just say that it is just as well that I began my title with the word ‘towards’, and ended it with a question mark. Later on I may find myself suggesting an alternative title that might be a little more appropriate. But now to business. There sometimes seems to be a slight aura of the not-quiterespectable about the enterprise of reconstructing fragmentary Greek tragedies; I have from time to time felt this in the comments of referees, and expect in the near future to feel it anew in those of reviewers.3 But this activity is only a particular case of what one is doing, much of the time, in almost every branch of our discipline: reconstructing past history from evidence which, by the standards of almost all later historical periods, is wretchedly scanty, in quantity and in quality, both absolutely and in proportion to the total amount of evidence that must once have existed. It only attracts particular attention, I suggest, because side by side with these fragmentary dramas there survive others for which, most exceptionally, the evidence is not scanty—indeed, so far as the words of the script are concerned, it is pretty well complete—and for which, therefore, at least in some respects, no reconstruction is necessary. That great good fortune, and our further good fortune in recent generations in having discovered very substantial portions of several other tragic texts (especially those of Euripides), should not lead us to suppose that there is nothing worth while to be gained by attempting to piece together other dramas for which we have, maybe, no more than a few quotations, or paraphrases, or summaries in ancient sources and, if we are lucky, twenty or thirty tattered lines of papyrus. ‘Piece together’ suggests a jigsaw puzzle, and the image is not a bad one. We have, perhaps, from 2 to 5 per cent of the puzzle pieces, mostly isolated but occasionally in clumps, a few of them perhaps with an indication that they belong, say, near the top left corner. We usually do not have the picture which toymakers always print on the puzzle lid, or at best we only have some torn bits of it in the form of a papyrus Hypothesis; but we virtually always have
3
Of Sommerstein, Fitzpatrick, and Talboy (2006); and Sommerstein (2008).*
Sherlockismus and fragmentary tragedies
63
other pictures, complete or partial, of closely related scenes—or, to speak without allegory, other accounts of the same mythical episode or parts thereof. The challenge is to recover as much information as possible about the picture that lies behind the assortment of pieces presented to us. We know, at any rate, that what we are trying to do is not inherently impossible. ‘Aeschylus’ Palamedes’, for example, to take a play that I have had some dealings with in the past,4 is the name of a real historical object, a drama which was produced, in the tragic competition at the City Dionysia, some time in the first half of the fifth century bc, and whose script was known to scholars of the Hellenistic period. And about every tragedy, even one of which not a single word is preserved, we know at least three things. I am assuming henceforth that we are dealing with a play of Aeschylus, Sophocles, or Euripides. First, we know that the play respected the conventions of the tragic genre, as they were at the time of production. We know that it had a chorus, and not more than three actors. We know that the action was structured so as to give at least the impression of continuity, and that while a choral song could cover the passage of time to some extent, it could not cover an interval of months or years. We know that if the action involved a violent death, the violence did not take place on stage (though the death might, as in Euripides’ Hippolytos).5 We know that in any play produced later than about 460 bc, there will be some kind of imaginary enclosed space behind the performing area—usually a building, but sometimes a cave or even, as in Oedipus at Kolonos, a sacred grove—out of and into which characters, and very occasionally the chorus, may come and go. And so on—I need not go through all the specifications. Every responsible scholar who proposes reconstructions does so in accordance with these conventions, though there may of course be disputes as to precisely what the conventions were at any given moment (such as the disputes there have been over exactly when it was that the mechane was introduced). Secondly, we know that the play respected the overall structure of the mythical complex of which its story formed part. No tragedy can 4 5
Sommerstein (2000b). See .
64
Sherlockismus and fragmentary tragedies
have an ending which results in Troy not being captured, or in Herakles ending his earthly life anywhere other than Mount Oita, or in Medea taking her children with her to Athens, or in anyone dying childless if he or she was traditionally credited with famous offspring. (That is the strongest reason for believing that Verrall and Dawe6 were right to regard lines 822–31 of Seven against Thebes as spurious, even though it is not a reason that Dawe himself would accept:7 Aeschylus could ignore the well-known mythological datum that Thebes was captured and sacked by the Epigonoi a generation after their fathers’ attempt to do so had failed—after all, his characters did not know that this was going to happen—but he could not explicitly contradict it by positively asserting that Eteokles and Polyneikes had both died childless, any more than the poet of the Odyssey could explicitly assert, or make a character assert, that Klytaimestra died otherwise than at the hand of her son.8) Towards the end of the fifth century, to be sure, we do find, in plays such as Philoktetes and Orestes, that the action is moving towards, or has even virtually reached, a conclusion radically incompatible with the traditional continuation of the myth when it is forcibly brought back on to the rails by a deus ex machina; but the fact that this occurs twice in the five plays that survive from the last three or four years of Euripides’ and Sophocles’ lives, and never in any of their eighteen or nineteen earlier surviving tragedies, strongly suggests that during most of their careers, let alone that of Aeschylus, it had not been a possible option. These two points are fairly obvious. A third is perhaps a little less so. We can be sure—this is the third point—that every lost tragedy told a story that differed, to a greater or lesser extent, from that of 6
Verrall (1887) ad loc. (lines 807–16 in his numbering); Dawe (1978a: 88–9). Cf. Dawe (1978a: 89): ‘I have myself a vested interest in retaining 822–31 as the genuine work of Aeschylus, because the adjective IŒ ı is the clearest of the many indications given to us that Eteocles and Polyneices died childless.’ The extinction of the house of Laios is certainly a presupposition of the latter part of Seven (once it is stripped of the non-Aeschylean dramatis personae Antigone and Ismene); but it is nowhere else made explicit. 8 What he actually makes a character (Nestor) assert is that Orestes killed Aigisthos, and then held for the Argives a feast for the funeral of Aigisthos and Klytaimestra—which would give an uninformed hearer (such as Telemachos?) the impression that Klytaimestra had committed suicide. See my discussion in Ch. 16 below, pp. 224–5. 7
Sherlockismus and fragmentary tragedies
65
every other tragedy based on the same slice of myth. Whenever we have evidence on the subject—as in the case of the three plays on Orestes’ revenge, or of Dion Chrysostom’s account of the three Philoktetes plays9—this is what we find. Partly in combination with my collaborator Tom Talboy, I have made quite extensive use of this principle in the volume of fragmentary plays of Sophocles that I recently co-edited in the Aris & Phillips series,10 and I don’t really wish to go over it again before the reviewers have been heard from. Of course, though we can always be sure that each dramatist will have introduced changes, we will not be able to tell a priori exactly what changes he introduced, since there will always be plentiful opportunities for innovation at various points in the story. Plentiful—but, bearing in mind that the overall structure of the mythical complex must be respected, not unlimited. And sometimes, at least, therefore, one can bring to bear the principle to which one of my early teachers—I think it must have been Martin Lowry—called Sherlockismus: when you have eliminated the impossible, whatever remains must be the truth. Sherlock Holmes himself, of course, added to that ‘however improbable’;11 in the kind of problems we are concerned with, it would be more appropriate to add ‘however little direct evidence there may be in its support’. Indeed, in at least four of the five cases I am going to consider, there is no direct evidence at all in support of the conclusion that I shall be arguing for, but the conclusion still, I would maintain, deserves to be accepted simply because every alternative to it can be ruled out. Perhaps that should have been my title today: ‘Sherlockismus and the study of fragmentary tragedies’? I used this principle in 200012 to establish (well, I think I established) that Oxyrhynchus Papyrus 2253 (Aesch. fr. 451k in TrGF) contained the opening lines of Aeschylus’ Palamedes.* It was already fairly generally accepted that they contained the opening lines of some play of Aeschylus,13 and it was obvious that this play had to 9
Dio Chrys. Or. 52. Sommerstein, Fitzpatrick, and Talboy (2006). 11 Conan Doyle ([1890] 1929) 161 and ([1892] 2001) 248. 12 Sommerstein (2000b). 13 The fragment had been attributed to Aeschylus by its first publisher, Edgar Lobel (in Lobel, Wegener, and Roberts 1952: 1), mainly on the ground that it was one 10
66
Sherlockismus and fragmentary tragedies
do with the Trojan War saga-cycle; or more precisely, given the apparent reference to ‘captains of Greece’ taking revenge on Paris, that it was set at a time when the combined Greek expeditionary force against Troy was in being. The ensuing reference to a ‘friendly reconciliation’—which could hardly mean a reconciliation with the Trojans, given the tone in which Paris’s crime is spoken of—strongly suggested, as Lloyd-Jones (1957) had seen, that there had been a serious quarrel among the Greek leaders. Furthermore, the speaker of these lines was someone who was not himself a partisan in that quarrel, since he is anxious for it to be resolved and speaks about the quarrelling parties collectively in the third person. At this point everybody’s thoughts would turn first, as those of Lloyd-Jones and Mette did,14 to the quarrel of Achilles and Agamemnon, and to Aeschylus’ Myrmidones. But we know how Myrmidones began, and it began, not with spoken iambics like these, but with chanted anapaests.15 There are, at most, ten other known plays of Aeschylus set in the relevant parts of the Trojan saga. Five of these can be ruled out right away because there is no plausible way to make them begin with a serious quarrel among the leaders of the Greek army. A sixth, Nereides, can also be ruled out, for the same reason, if West is right to suppose that it centred on the death of Achilles;16 on the more traditional view (which I believe to be correct) that it was the middle play of a trilogy based on the Iliad, it can be ruled out just the same, because in Achilles’ part of the camp, where the action must have been set, no neutral figure would have been available to speak this prologue.* Three further plays can be excluded too: of over a score of papyri written by the same hand (listed in TrGF iii. 10) some of which were provably, most of them probably, and all of them possibly Aeschylean; he also drew attention (p. 15) to a similarity between line 1 of the fragment and the opening of Eumenides. Lobel nevertheless found ‘nothing to show that this is the beginning of a play’, but Bruno Snell in his review of the POxy volume (Snell 1953: 437), and most since, have been fairly confident that it is. 14 Lloyd-Jones (1957: 584–5) (‘likelier than not to come from the Achillean trilogy’ and if so, from the first play rather than the second or third); Mette (1959: 77–8) (printing it among the fragments of Myrmidones, though with two warning question-marks). 15 Aesch. fr. 131: the scholia to Ar. Frogs 992 say Ø b Iæåc ÆoÅ ıæ Ø ø
`Nå ºı. 16 West (2000: 341–2).*
Sherlockismus and fragmentary tragedies
67
Hoplo¯n Krisis because there was no quarrel between Odysseus and Ajax until Thetis arrived, at some point during the play, to present the armour of Achilles as a prize for competition;17 Threissai, its sequel, because in the second play of a trilogy the audience would not need to be told that they were watching an episode of the Trojan War, and lines 5–618 would thus be superfluous; and Iphigeneia because it would be very surprising, at the start of a play that is going to end with Iphigeneia’s sacrifice, for a neutral character to begin with a prayer that did not have Artemis as one of its addressees and that made no mention of the detaining winds. That leaves Palamedes as the only play that could fit the bill; therefore this must be the opening of Palamedes. It follows, furthermore, that at the moment when the action of Palamedes opened, the dispute between Palamedes and his enemies had already become public—in other words, the false accusation against him had already been made—but it was still possible to hope that no harmful consequences would come of it (which would have to mean that Palamedes would be proved innocent). That was applying Sherlockismus to eliminate all but one of a limited set of plays. It can also be applied so as to eliminate all but one of a limited set of mythical episodes that are candidates for supplying the plot of a poorly attested play. In my collaborative edition of selected fragmentary plays of Sophocles, published , I did this with his play Phthiotides (The Women of Phthia)19 in a way that caused serious concern to the late Malcolm Willcock when he read the finished manuscript of the volume—so much so that he wanted me to bury the play in the middle of the volume instead of placing it at the beginning. The reason I had placed it at the beginning, by the way, was a purely mechanical one. Since there is no general criterion for dating Sophoclean plays in the way that Euripidean plays can be dated by their metrical practice, the only sensible way to arrange them in an edition is alphabetically; and if it seems surprising that a play beginning with the letter phi should come first in such an arrangement, the explanation is that I regarded, 17
Aesch. fr. 174. ] º ø fi c ÆØ ±æƪc | [ªı ÆØŒe KŒ] æ ıØ — æ[Ø]Æ []Å —æØ . See Sommerstein, Fitzpatrick, and Talboy (2006: 14–16), of which the following paragraphs are a summary. 18 19
68
Sherlockismus and fragmentary tragedies
and regard, Phthiotides as being just another name for Sophocles’ Hermione. We possess only three fragments ascribed to Phthiotides; in the ancient citations of two of these, by the way, the play-title is corrupt, and in one of them (fr. 694) it was corrected by none other than Thomas Gaisford—the patron hero, of course, of these lectures, and, as it happens, the subject of the one to be given by Chris Stray. Fortunately, and unusually, every one of the fragments of Phthiotides tells us something about the play. In the first (694) someone tells a young man, surely by way of rebuke, that he has a lot to learn; in the second (695) one old man says to another ‘I shall lead you like a child’; and in the third (696), there is mention of an actual or metaphorical lawsuit connected with the killing of a father (ÆæŒ ŒÅ). Thus we know that the characters of Phthiotides included, at least, two old men, who at some point went somewhere together, and one rash young man, and that the play was concerned, among other things, with repercussions following the killing of someone’s father. In addition, it is highly likely, though not certain, that a play with a chorus of women of Phthia would have Phthia as its setting. Now all this is compatible with what we know of Hermione. Of Hermione we have even fewer attributed fragments (two), but we do have the equivalent of a Hypothesis (if a rather brief one) in Eustathios’ commentary on the Odyssey,20 and one or two other mythographical passages21 that seem likely to derive from Sophocles’ play in whole or in part. We can be sure that Hermione was set at Phthia (not at Delphi, as Welcker supposed), since in fr. 202 someone is greeting his or her native soil, having evidently just returned to it, and even if we imagine, say, Machaireus (the Delphian who killed Neoptolemos)22 as a character in the play, there would be no reason to present him as having returned to Delphi from a journey abroad. We can also easily identify the rash young man in Phthiotides as Neoptolemos (who, we know, foolishly attempted to exact justice 20
Eustathios on Od. 4.3. Pherecydes frr. 64, 135A Fowler; [Apollod.] Epit. 6.13–14. According to Strabo 9.3.9; Eustathius loc.cit.; and [Apollod.] loc.cit. (who, however, says he was a Phokian). 21 22
Sherlockismus and fragmentary tragedies
69
from Apollo for the killing of Achilles)23 and the two old men as Peleus and maybe Phoinix. This evidence was enough for Pearson,24 following Vater,25 to conclude that it was ‘most likely’ that Phthiotides was the same play as Hermione; but we can, I think, go further. If it wasn’t the same play as Hermione, we must find something else with a Phthian setting for it to be about. What could this be? Matthiae and Welcker suggested that Phthiotides was the same play as Sophocles’ Peleus, but this will not work. There is no room in Peleus for a second old man befriending Peleus: we know that at the start of the play he was being cared for solely by one devoted woman (fr. 487), and we know he was eventually rescued by Neoptolemos,* who in this play is likely to have been a virtuous character and will hardly have been spoken to as he is in fr. 694; Radt’s verdict on Matthiae’s proposal, ‘vix probabiliter’,26 is entirely justified. Ernst Ahrens suggested27 that the young man was Peleus himself, who accidentally killed his father-in-law Eurytion during the Kalydonian boar-hunt. This won’t do either. Peleus would not have returned to Phthia directly after killing its king; indeed our main source for the story28 says that he fled to Iolkos. He did, of course, return in the end, but apparently only after the death of his first wife Antigone, Eurytion’s daughter (and only possible avenger).29 Antigone’s death—suicide, according to [Apollodoros] (whose source is probably Pherekydes) on receiving the false news that Peleus was to remarry in Iolcus—might make a good subject for a tragedy, but who then could the young man be (certainly not Peleus, who wouldn’t even be in Phthia at the time), or the two old men for that matter? I wasn’t quite right to say in my edition that no other suggestion had been made; Lewis Campbell30 did propose that the young man
23 Eustathius loc.cit. (explicitly giving this as the version found in Sophocles’ Hermione); [Apollod.] loc.cit. 24 Pearson (1917: ii. 306). 25 Vater (1851) 180–1, 189. 26 Radt (1999: 392). 27 Ahrens (1844: 287–8). 28 [Apollod.] Bibl. 3.13.2. 29 Ibid. 3.13.3; Tzetzes on Lykophron Alex. 175, citing Pherekydes fr. 1 Fowler. 30 L. Campbell (1879–81: ii. 541), in a note on Phthiotides (not specifically on this fragment): ‘Qy. On the education of Achilles?’
70
Sherlockismus and fragmentary tragedies
might be Achilles, but no one seems to have taken this seriously and Radt does not even mention it. Apart from anything else, Achilles did not grow up at Phthia but on Mount Pelion with Cheiron; he will, to be sure, have returned home at least briefly before joining the Trojan expedition (and the Iliad consistently implies as much), but nothing of significance happens during his stay there (except indeed whatever Sophocles contrived for the purposes of his satyr-drama Achilleo¯s Erastai). What we know of Phthiotides, then, fits what we know of Hermione, and it doesn’t fit any other Phthian story we can identify. Therefore, I contend, we can firmly conclude that Phthiotides and Hermione are one and the same play. I would now like to examine with you a few more plays in the Aeschylean corpus which I’ve found myself having to look at in the course of my work on a new Loeb edition of Aeschylus.31 Aeschylus, it has to be said right away, is even tougher reconstruction material than Sophocles. We thought things were pretty bad with Hermione and Phthiotides, for which we had five quoted fragments (plus some testimonial material); for almost half the plays of Aeschylus (I make the number thirty-six out of a total of about seventy-eight) we don’t even have that much. The first play I want to look at, Niobe, is rather better attested than that, but I wish to concentrate mainly on one speech, the speech in which Niobe’s silent grief was described—the speech from which Plato wrenched two lines out of context to ‘prove’ that Aeschylus disseminated falsehoods about the gods.32 Of this speech we now have twenty-one lines (fr. 154a) from a papyrus, but there is still no consensus on who the speaker is—the speaker, that is, either of the whole passage or of most of it (it is possible that lines 10–13 are spoken by the chorus-leader). The first editors33 suggested it might be Niobe herself, but this view has deservedly not attracted significant support for a long time now. Her father Tantalos, and her husband Amphion, are both referred to in the third person,34 and 31 32 33 34
Sommerstein (2008). Pl. Rep. 380a. Vitelli and Norsa (1933: 108–14). Tantalos in lines 2 and 10; Amphion in line 12.
Sherlockismus and fragmentary tragedies
71
can safely be excluded—indeed Amphion is probably dead (line 5 suggests that Niobe’s marriage is now at an end)35 and Tantalos, we know, is not yet on the scene (line 10). We can also exclude the chorus-leader, in the first place because chorus-leaders do not normally make set speeches,36 in the second place because the speaker is addressing a group of people (lines 5 and 14) who must certainly be the chorus. Reinhardt37 suggested that the speaker might be Leto, the goddess whom Niobe had offended, but not many will be able to see this passage, particularly its first nine lines, as the speech of a triumphant enemy. The serious candidates who remain standing are Niobe’s nurse (or the nurse of her children, though that is probably a distinction without a difference),38 Niobe’s mother (that’s all we can call her, since mythographers and scholiasts give her several different names),39 and Amphion’s mother Antiope. Of these, Niobe’s mother, an identification first proposed by Eva Keuls40 (on the basis of some late and flimsy artistic evidence), has perhaps been the most popular in recent years, having been advocated by Antonio Garzya and by Alain Moreau.41 Since all these candidates are female, I shall be using feminine pronouns to refer to the speaker, though the papyrus text itself does not reveal the speaker’s gender.
35 ›æA PØ[] æ Ø ª ı ‘you see the final outcome of this marriage’. In a fifth-century lyric poem by Telesilla (PMG 721) Amphion, like his children, was shot by Apollo and Artemis (cf. perhaps Hyginus Fab. 9, where he is killed trying to storm the temple of Apollo in revenge for their deaths);* in Ovid (Met. 6.271–2) he commits suicide following the slaying of his sons. 36 The rare exceptions include two in Aeschylus, Pers. 215–25 and Ag. 489–502. 37 Reinhardt (1934: 249–50). 38 In trag. adesp. 7 a nurse speaks of how she cared for Niobe’s children, in terms reminiscent of Aesch. Cho. 749–62, and it is tempting to see the opening three-word trimeter, with the sesquipedalian compound º ÆŁø , as Aeschylean, and the fragment as positive evidence that a nurse was a character in Aeschylus’ play. But three-word trimeters seem to have had a revival in the late fifth century and after (there are more in Eur. Phoin. than in any genuine play of Aeschylus bar one, and the short Rhesos has three: see Griffith (1977: 91–2)); and åºÆ Ø —a rare exception to the general avoidance of diminutives in tragedy—occurs otherwise only in Euripides (Supp. 110, Or. 42) and Chaeremon (fr. 14.9). 39 Euryanassa ( Eur. Or. 4, 11; Dositheus FGrH 54 F 1); Eurythemiste ( Eur. Or. 11); Klytia (ibid., citing Pherecydes fr. 40 Fowler); Dione (Hyginus Fab. 9). 40 Keuls (1978: 53, 61–4). 41 Garzya (1987: 193); Moreau (1995: 299–301).
72
Sherlockismus and fragmentary tragedies
The first thing we can say about the speaker is that she is well informed about recent events in Niobe’s family. At no point in her speech, so far as we can tell, does she express any uncertainty about anything, and she apparently knows, not only about the terrible fate of Niobe’s children, but also about its antecedents: Niobe was ‘elated by the surpassing beauty’ either of herself or of her children (line 21), and as a result was ‘rash in speech’ (line 18) and thereby aroused divine hostility. Moreover, the chorus are assumed not to know much of this—perhaps little beyond the fact of the deaths—otherwise it would be pointless for the speaker to inform them of it. Hence the speaker must be more closely connected with the house of Amphion, in which the disaster has occurred, than the chorus are. This at once rules out Niobe’s mother, even if we were to suppose that she had somehow reached Thebes, the scene of the action, before her husband Tantalos. Antiope, on the other hand, would be living with her son and daughter-in-law, and would know all about their affairs; and the fact that the blame for the disaster is placed on Niobe might at first suggest that the speaker is someone more closely connected to her husband than to her. Only at first, though. The speaker, while dilating on the sorrows of Niobe, seems to have no sympathy at all to spare for Amphion. She speaks of Niobe’s marriage as ‘a harbourless shore of a union’ (line 3) and calls the disaster that has now struck ‘the final outcome of this marriage’ (line 5): in effect she is saying ‘if Niobe hadn’t married Amphion, this would never have happened to her’—the very opposite of what we would have expected Amphion’s mother to say, especially when Niobe’s own rashness has been the cause of the catastrophe—and when it is Amphion, not Niobe, who lies dead! On the other hand the attitude shown in the speech is precisely what one would expect from an old and loyal servant of Niobe herself, and this is not inconsistent with the speaker’s acceptance that Niobe is to blame for what has happened. In tragedy, servants who have attended their master or mistress from childhood—nurses and paidagogoi—have considerable licence to criticize them when appropriate, in their own best interests: consider the paidagogos of Orestes (in Sophocles’ Elektra)42 and the nurses of
42
Soph. El. 82–5, 1326–38.
Sherlockismus and fragmentary tragedies
73
Hermione (Eur. Andr. 825–78) and above all of Phaidra.43 Nor is it incongruous that such a servant should have a sound understanding of how gods behave and how mortals should behave towards them: the servant of Hippolytos (Eur. Hipp. 88–120) is arguably the wisest and most pious mortal to appear in that play. So both the unsuitability of every other candidate, and the appropriateness of what is said on the lips of a loyal servant, point unmistakably to Niobe’s nurse as the speaker in this passage. And we can also get some purchase on the chorus: they must be residents of Thebes (we cannot suppose they are members of Tantalos’ retinue who have arrived before him, since they would not then need to be informed that he was coming) but not members of Amphion’s household. We cannot tell from this speech whether the chorus are male or female, but since they appear to have come to the palace to enquire about Niobe, out of curiosity or sympathy or a mixture of both, they are most likely to have been women of the city, given the well-known stereotypes of women as insatiably curious (think of Pandora) and as always likely to stick together. Another problematic papyrus is POxy 2251 (fr. 451h Radt). This is a lyric fragment in which a chorus of women grieve bitterly over the ruin of a man who had been noted for his hospitality and justice; as their grief extends to tearing out their hair (lines 6–7), it is virtually certain that someone has died, but we cannot tell whether this is the hospitable man himself or some member or members of his family. There has from time to time been support for the view early put forward by Margaret Cunningham44 that the fragment is from Aigyptioi and the victim Pelasgos, but there has never been any adequate response to the obvious objection that a play called Aigyptioi ought to have a male chorus. Another early suggestion, by Bindzus,45 was that the reference was to the murder of Eioneus (or Deioneus) by his son-in-law Ixion, which was the subject of Aeschylus’ play Perrhaibides; but in that story it was Ixion who
43
Eur. Hipp. 176–238, 304–14, 433–524; cf. also Soph. fr. 693a (Lloyd-Jones) [¼ El. 1050–4] with the discussion by Talboy and myself in Sommerstein, Fitzpatrick, and Talboy (2006: 276–7, 303–5). 44 Cunningham (1953). 45 Bindzus (1956: 228).
74
Sherlockismus and fragmentary tragedies
entertained his father-in-law, not the other way round. And there, it seems, the matter has rested, perhaps dampened by Lloyd-Jones’s remark46 that ‘for all we know many hospitable persons may have suffered destruction in lost plays of Aeschylus’. Maybe. But it is worth while examining the evidence, bearing in mind, in the first place, that we are looking for a play with a chorus of women—and women sympathetic to the hospitable hero—and secondly that our fragment strongly suggests that disaster has come to this man, at least in part, because of his hospitality—which is what made the Pelasgos option so tempting. He welcomed a guest or guests, and then something terrible happened which either was, or was thought to be, the result of his having done so. From the plays in the Aeschylean corpus, we can immediately rule out all those which are satyr-dramas and all those which had male choruses. That at once reduces the number of plays that come into consideration from seventy or so to little over forty. And of these only a very few give any positive reason to believe that they might come anywhere near meeting the criteria set out above. One might think of Argeiai—which was about the aftermath of the defeat of the Seven against Thebes—with Adrastos as the victim, ruined by his kindness to Tydeus and Polyneikes; but Adrastos is always the survivor of the expedition against Thebes, nor do we even hear of his losing his daughters (one of them, Argeia, finds herself at Thebes in various accounts,47 but she never perishes, though in Statius she has a narrow escape). Another conceivable option—though about the play itself, absolutely no information has survived except its title—might be Polydektes, if the righteous, hospitable victim was the title character’s half-brother, Diktys (who we know appeared in the presumably related satyr-drama Diktyoulkoi); but no source claims that Diktys was murdered, and an early one, Pherekydes (fr. 12 Fowler), says that Perseus, after returning with the Gorgon’s head and turning Polydektes to stone, made Diktys king of Seriphos in his place. In fact we are really left with only three possibilities. One is that the 46
Lloyd-Jones (1957: 571). Hes. fr. 192 M–W <¼ 135 Most>) brought her there for Oedipus’ funeral. In some later accounts (Statius Theb. 12.177–463, 677–804; Hyginus Fab. 72) she joins with Antigone in giving funeral rites to her husband Polyneikes. 47
Sherlockismus and fragmentary tragedies
75
fragment comes from one of those plays for which we have nothing at all, not even an informative title, to reveal what story it was that they presented—for example Hiereiai or Propompoi. The second is that it comes from a play of which we have no trace in any other source, like the Tennes-play from which fr. 451o (POxy 2256 frr. 51–3) is widely thought to come. And the third is that it comes from Kressai (The Cretan Women). We can be virtually certain that Kressai told the story of Glaukos, the young son of Minos of Crete, who disappeared and was eventually found dead by the visiting Corinthian seer Polyidos; Minos shut Polyidos up with the corpse, in effect burying him alive, and refusing to release him unless he restored the boy to life, which Polyidos managed to do.48 Minos had earlier been told that Glaukos would be found by the man who could suggest the best analogy to describe a marvellous cow in Minos’ herd which continually changed colours between black, white, and red; Polyidus compared it to a blackberry, and we have a couplet from Kressai (fr. 116) in which the colour changes of the blackberry are described in very much the same terms as they are in a fragment (Soph. fr. 395) of Sophocles’ play Manteis or Polyidos. Minos was, of course, famous for his justice (cf. Odyssey 11.568–71, Pl. Apol. 41a), and he was entertaining a guest, Polyidos, at the time when Glaukos disappeared. If our fragment does come from this play, it tends to suggest that Polyidos was suspected of being responsible for Glaukos’ death; this idea is not made explicit in any of our sources, but it would make Minos’ treatment of him (otherwise surprising, in such a paragon of justice) a great deal more intelligible. And the fact that Minos was a son of Zeus would make the chorus’s implicit rebuke to Zeus, in lines 2–3 of our fragment,49 all the sharper. We cannot formally exclude the other possibilities that I mentioned, but what we can say is that the papyrus fragment fits rather neatly in Kressai and fits no other Aeschylean play that we know anything at all about.
48
The story is told in full by [Apollod.] Bibl. 3.3.1–2; and Hyginus Fab. 136. They call on Z[eus] Xe[nios] to see what has happened to []e Œ . In the next two lines they appear to ask whether there is any åæØ K [ ]E [I ] æ[]Ø ØÐ ØŒÆØ. 49
76
Sherlockismus and fragmentary tragedies
I have examined so far—including Palamedes—three Aeschylean tragedies. Three tragedies ought to be followed by a satyr-drama, so let’s finish with a look at Proteus, the play that followed (or, if you prefer, concluded) the Oresteia. Proteus dealt with the adventures of Menelaos in Egypt. We know that as in Homer, Proteus’ daughter, Eidothea or Eido, was a character or at least was referred to.50 We do not know precisely what role the satyrs played, though it is tempting to suppose, in the first place, that (as so often) they were captives and the visiting hero liberated them,51 secondly that they took, shall we say, a considerable interest in Helen. In Homer, of course, Proteus informs Menelaos of the murder of his brother (among other things);52 there is no telling just what information, and/or what instructions, he gave in Aeschylus’ version. In the past I have suggested53 he might have given instructions for the marriage of Menelaos and Helen’s daughter Hermione to Orestes; and Griffith has proposed54 that he may have revealed that Iphigeneia had not in reality been sacrificed. I will make another suggestion presently. We do, of course, have another, complete play about Menelaos in Egypt: Euripides’ Helen. In general, students of that play have taken very little notice of its satyric precursor: I can find no mention of it whatever in the editions of Helen by Dale, Kannicht, or Burian,* or in Norman Austin’s book Helen of Troy and her Shameless Phantom (credit to Matthew Wright, at least, for referring to the play, very briefly, on two occasions).55 Students of Proteus very definitely have, however, been taking notice of Helen lately: first Margaret Cunningham, then Mark Griffith, have suggested that Aeschylus, like Euripides, may have used Stesichoros’ idea that the Trojan War was fought for a phantom divinely substituted for Helen while the real Helen was miraculously taken to Egypt.56
50
Aesch. fr. 212. Cf. Seaford (1984a: 33–6). 52 Odyssey 4.472–569, esp. 512–37. 53 Sommerstein (1996a: 190). 54 Griffith (2002: 241–6, esp. 243). 55 Dale (1967); Kannicht (1969); N. Austin (1994); Wright (2005: 128 n. 220, 149; Burian (2007)). 56 Cunningham (1994); Griffith (2002: 239–49). 51
Sherlockismus and fragmentary tragedies
77
The passages in Agamemnon to which Cunningham and Griffith refer (404–26, 737–43) certainly show that, as one might expect, Aeschylus was aware of Stesichoros’ unorthodox presentation of the story. They do not, however, in themselves show that he himself used it; and there is some evidence that he did not. In Homer, Proteus is ‘the Old Man of the Sea’; he comes up out of the sea and, like other mermen and mermaids, has to be pinned down before he will speak.57 In Euripides, he and his son Theoklymenos after him are ordinary terrestrial kings living in an ordinary terrestrial palace; his only marine connection is that he married the Nereid Psamathe.58 So of course it had to be, since Helen could not have stayed for seventeen years at the bottom of the sea. So too it is in Herodotus, where Proteus is ‘a man of Memphis’ and has his palace there;59 so it is in Euripides’ Elektra, where Helen is said to be returning to Argos ‘from the house of Proteus’;60 and so it must have been in Stesichoros (now that we know that in his account Helen did stay with Proteus61—even if not how she got there). So it must have been in Aeschylus too, if he used the phantom story. Can we tell whether it was or not? In Homer, Menelaos is stranded for three weeks on a small and apparently barren island, where the only way to get food is by fishing.62 In Euripides he is shipwrecked on the coast of mainland Egypt and goes straight to the palace to seek assistance;63 there is no question of his enduring a long period of semi-starvation. In Herodotus he sails safely up to Memphis, already knowing that Helen is there, and simply asks for her return.64 Again this is how it has to be: if Proteus is to be a suitable host for Helen, he cannot live on a barren island. It follows that if we find a hungry Menelaos reduced to a hunting and/or fishing existence, the phantom is excluded. And that seems to be what we do find. In fr. 211 there is a reference to garum fish-sauce; not that this will have been available to
57 59 61 62
58 Odyssey 4.363–461. Eur. Hel. 6–7. 60 Hdt. 2.112.1, 2.114.1. Eur. El. 1280–1. Stesichoros PMG 193 (¼ POxy 2506 fr. 26 col. 1). 63 64 Od. 4.354–69. Eur. Hel. 408–34. Hdt. 2.118.3–119.1.
78
Sherlockismus and fragmentary tragedies
Menelaos, but, as Germar and Krumeich have seen,65 the very mention of it strongly indicates that he (or the satyrs—it makes no difference) are living wholly or mainly on fish. There would be no need to do this if there was a city and palace within reach, as there would have to be if Proteus was to figure in the play as a human king. Therefore we are on the desert island (or, just possibly, a barren and isolated part of the mainland coast) and Proteus is the Old Man of the Sea. Therefore he cannot have the real Helen with him. Therefore there was no phantom in Aeschylus’ version. Unless, that is, one is prepared to suppose that in Aeschylus, uniquely, the real Helen had been concealed somewhere other than Egypt; and that must surely be a Remotely Conceivable Alternative66 if ever there was one—not, of course, because it would be unique, but because there is no reasonable other place for her to be (certainly not Sparta itself, as used to be supposed67—somebody would have noticed if she’d been there all the time!) Rather it seems that Aeschylus followed the Odyssey quite closely, subject always to the need to accommodate the indispensable satyrs. That may give us a hint as to a possible linkage between the tragic Oresteia and its afterpiece. Here I emphasize that I am no longer inferring; I am speculating. In the Odyssey (3.309–12) Menelaos arrives at Mycenae (not Sparta) on the day of the funeral of Aigisthos and Klytaimestra; and in due course we learn that Proteus had hinted to him (4.546–7) that he might find Aigisthos already dead when he got back. It follows—though this is never spelled out—that if it had not been for the three-week delay off Egypt (which was his own fault—he’d failed to make sacrifices to the gods), Menelaos would
65
Germar and Krumeich (1999: 181). For this term see Page (1959: 57). 67 On the basis of Dio Chrys. Or. 11.40–1, who says, evidently misinterpreting the opening of Stesichorus’ ‘palinode’ (PMG 192.2), that according to Stesichorus Helen ‘did not sail anywhere at all’, and contrasts him with ‘some others’ (such as Herodotus?) according to whom she was abducted by Paris and then came to Egypt. See for instance Bowra (1936: 122). The publication of POxy 2506 killed this view. Presumably in Stesichorus, as in Euripides, it was literally true that Helen did not sail away from Sparta, because she was transported thence to Egypt by miraculous means (so, with unnecessary hesitancy, Gantz (1993: 575)). 66
Sherlockismus and fragmentary tragedies
79
have got there before Orestes and would have had the glory of killing Aigisthos himself: I say Aigisthos because, as is well known, nothing is ever said in the Odyssey about how or at whose hand Klytaimestra died. What then if, in Aeschylus’ play, Proteus had told Menelaos (as in Homer) about Agamemnon’s murder and then told him (not quite as in Homer) that he should hurry home because Orestes was just about to take revenge on his father’s murderers (both of them, of course)—adding explicitly that if it hadn’t been for Menelaos’ neglect of the gods he’d already have been home by now? In the Aeschylean context, this would no longer be merely a matter of who got the credit for removing Aigisthos: instead it would undercut one of the fundamental assumptions on which the tragic Oresteia was based. Orestes himself, Apollo, Athena, and apparently Zeus all speak and act on the basis that Orestes has absolutely no alternative but to kill his mother. Now we would find that all that was necessary to avoid this horror was to wait a day or two, and Menelaos, who was no blood relation of Klytaimestra, would come home and do the job himself. This undercutting of a basic feature of the trilogy would itself, of course, be undercut by the fact that it was being propounded in a satyr-drama, where all the values of tragedy were typically stood on their heads. Euripides, in his own remake of part of the Oresteia, brought the undercutting remark right into the tragedy itself. When the Dioskoroi arrive on the scene in Electra after the killing of Klytaimestra—on which, and on Apollo’s role in it, they have some decidedly negative comments68—they casually inform Orestes and Electra that Menelaos and Helen have already arrived at Nauplia (1278–80), in other words that the whole filthy business they have just gone through was entirely unnecessary. And they then make it a double whammy by adding that Helen had been in Egypt and it was only a phantom that went to Troy (1280–3)—in other words that the whole Trojan War was unnecessary too. Perhaps the Aeschylean Proteus also had a second barrel: as we have seen, he couldn’t have used the idea of the phantom Helen, but Griffith’s suggestion that he revealed the truth about the sacrifice, or rather non-sacrifice, of
68
Eur. El. 1244–6, 1296–1302.
80
Sherlockismus and fragmentary tragedies
Iphigeneia would make a very good alternative, retrospectively depriving Klytaimestra of what she seems to feel as her most powerful motive for killing her husband (Ag. 1412–20, 1523–9, 1555–9). The art of tragic reconstruction must always be a matter of combining the logic of myth with the direct and indirect textual evidence, identifying the possibilities, and (in favourable cases) eliminating all but one of them—and the cases I have looked at, by Aeschylean or Sophoclean standards, are favourable ones. And it is ‘reconstruction’, often, in a very limited sense; I have not today claimed to be able to establish the plot structure of any lost Aeschylean play, though that is partly because I have chosen to illustrate the subject in breadth rather than in depth. It is more reducing the range of uncertainty. I hope I have persuaded you that this is not a futile endeavour.69
ADDENDA p. 62 n. 3 thus, gently, Gibert (2008: 261): ‘the feeling often lingers that we do not really know nearly as much as Sommerstein believes we do’. p. 65 I have therefore in my Loeb edition printed fr. 451k Radt in the Palamedes section (Sommerstein 2008: iii. 184–5) and numbered it 180a. p. 66 n. 16 against West, see Michelakis (2002: 31 n. 21, 53 n. 71), followed by Sommerstein (2008: iii. 156–7). p. 66 on Nereids: in addition, since Nereids was not the first play of its trilogy, the same argument that is employed below to rule out Threissai would apply to Nereids too. p. 69 Peleus rescued by Neoptolemus: perhaps ‘know’ is too strong, since there was at least one version of the story of Peleus’ old age in 69
This paper was delivered as the annual Gaisford Lecture at St John’s College, Oxford, on 31 May 2007, under the title (announced long in advance, and not appropriate to the lecture’s actual content) ‘Towards a grammar of tragic reconstruction?’ I have substituted the alternative title that I suggested during the lecture itself (see p. 65), but otherwise left the text in its original form.
Sherlockismus and fragmentary tragedies
81
which he and his grandson never met ( Eur. Tro. 1128); but it has long been the general consensus that Sophocles’ Peleus is most likely to have featured the rescue by Neoptolemus (see e.g. Lloyd-Jones 1996: 252–3). See further Chapter 18 below. p. 70 n. 35 Amphion seems to have been killed in Sophocles’ Niobe too; see POxy 3653 fr. 2 col. 1.21–4. p. 76 the more recent edition of Helen by William Allan (2008) does refer to Aeschylus’ play, though he still gets rather brusque treatment: one five-line footnote (21 n. 105) and part of another (29 n. 136), compared with four pages for Stesichorus and two for Herodotus.
5 The seniority of Polyneikes in Aeschylus’ Seven Which was Oedipus’ elder son, Eteokles or Polyneikes? There are surprisingly few explicit statements on the matter.1 The earliest come in tragedies of the last decade of the fifth century, and they contradict each other. In Euripides’ Phoinissai (71–2) Eteokles is the elder; in Sophocles’ Oedipus at Kolonos (374–5, 1293–5, 1422) Polyneikes is. We cannot tell from internal evidence whether one, both, or neither of these two versions is derived from earlier tradition, since each is integrated with the plot of its own play. In Phoinissai Eteokles’ seniority is used (69–76) to account for the fact that when the brothers agree to rule Thebes in alternate years (a story not heard of before Euripides), Eteokles is allowed to rule for the first year (at the end of which he refuses to surrender power). In Oedipus at Kolonos Polyneikes, who claimed the automatic right to succeed to Oedipus’ throne as his elder son (1292–4), is bitterly resentful at having been ousted and dishonoured by his younger brother, and this is one of the main reasons (1422–3) for his refusal to heed Antigone’s anguished plea to him not to destroy his family and city. Two later literary or quasi-literary sources2 name Eteokles as the elder brother; as we shall see, however, one artistic source tells a contrary tale.
1
I ignore witnesses (e.g. Pherekydes FGrH 3 F 95) who merely list Eteokles’ name before Polyneikes’ (or vice versa) without any explicit reference to relative ages, since in such cases there is no certainty that the brothers are being named in order of seniority. See, however, n. 15 below. 2 D.S. 4.65.1; A Iliad 4.376; the first-named is clearly dependent on Euripides.
Polyneikes in Aeschylus’ Seven
83
I wish to consider in this chapter what is implied in Aeschylus’ Seven against Thebes regarding the relative ages of the brothers. Nothing is said directly about the matter by anyone in the play, but certain indications point clearly, I would argue, to the conclusion that Aeschylus imagined Polyneikes as the older brother. In the first place, while both brothers are ¼ Œ Ø (828),3 so [106] that their deaths effectively extinguish the royal line (690–1, 813, 881–2, 954–60),* they appear to differ in their marital status. Eteokles is plainly an unmarried man. His wish never to be ØŒ (188, cf. ı
Æø 195) with women, parallel in form to the wish of Orestes never to have a woman like Klytaimestra as his ØŒ (Cho. 1005–6), implies that he is not ØŒ with any woman at present. The marriage of Polyneikes, on the other hand, is neither affirmed nor denied in the play; but the role of Tydeus as instigator of the war (571–5) strongly suggests that the traditional story of the combat between him and Polyneikes, resulting in Adrastos choosing them both as his sons-in-law,4 is presupposed (indeed no source explicitly leaves Polyneikes unmarried; and if Tydeus and Polyneikes were not both Adrastos’ sons-in-law, some explanation would be badly needed of why one of them should be urging Adrastos to launch an expedition to restore the other to his homeland).5 3 Outside tragedy, on the other hand, it is normally assumed that both brothers left sons behind them. Polyneikes’ son Thersandros certainly figured in the Cyclic Epigonoi, for the next poem in the Cycle, the Kypria, told of his death at the hands of Telephos in the course of the Achaians’ attack on Teuthrania (Proklos Epit.Kypr. 48 Davies <¼ Kypria Arg. }7 West>); in the fifth century cf. Pind. Ol. 2.43 and the Epigonoi monument set up by the Argives at Delphi (Paus. 10.10.4). Laodamas, the son of Eteokles, for his part, is regularly assumed to have been king of Thebes when the Epigonoi attacked the city (e.g. Hdt. 5.61; Paus. 9.5.13, 9.8.6, 9.9.5; [Apollod.] Bibl. 3.7.3; T Iliad 4.406), and Ion of Chios (PMG 740) portrayed him as having Antigone and Ismene burnt to death in the temple of Hera (when and under what circumstances we do not know). It is curious that no surviving source identifies Laodamas’ mother, but there is no reason to doubt that he was born in wedlock; this indeed is virtually presupposed by the tradition reflected in Paus. 1.39.2, 9.10.3, according to which Laodamas became king when a young child, immediately upon his father’s death, and Kreon acted as regent for him. 4 Eur. Supp. 131–50, Phoin. 408–25. 5 Similarly Gantz (1993: 509): ‘Aischylos’ Hepta presumably understands the same basic events [including the marriages of Polyneikes and Tydeus], since Adrastos . . . and Tydeus are . . . part of the expedition with Polyneikes . . . and no explanation of their presence is considered necessary.’
84
Polyneikes in Aeschylus’ Seven
Secondly, Eteokles must look young enough to be called Œ
(686) by a chorus of women who, though it is some time since they called themselves ÆæŁ ø . . . ºå (109–10, cf. ÆæŁ ø . . . ºØ 171–2, øºØŒøÐ . . . øºø 454–5), must still be wearing masks consistent with that designation. This strongly suggests that [107] he is beardless, as e.g. Orestes normally is in fifth-century art;6 Polyneikes, on the other hand, is specifically stated in the text to be bearded (666). Hence when the two brothers, one in death, are brought on stage (848ff.), there will be one conspicuous difference between them. They will be wearing similar armour, they have similar wounds (887), they are similarly mourned; but one has a beard and the other does not. Their quarrel has ended with the assignment of equal lots of Theban land to each (727–33, 816–19, 906–14); but one of the corpses that will be taken to be buried in those lots bears a highly visible sign to remind us that in one important respect they were unequal. This deduction from the text of Seven finds remarkable confirmation from a precious piece of artistic testimony. An engraved bronze Etruscan hand-mirror from Vulci, of the early fourth century,7 depicts two warriors, each grasping the right shoulder of the other and about to deliver a fatal sword-thrust to the left breast (cf. Seven 888–90 Ø’ Pø
ø ı Ø . . . › ºªå ø
º ıæø ø ).8 One of the warriors has a beard, the other has none. And both figures are labelled with names: the bearded man is phulnice, the beardless one evtucle. The only known earlier evidence for the brothers being distinguished in this particular way is that which we have found in Seven against Thebes, and it is hardly an extravagant suggestion that this iconographic pattern derives ulti6 For full discussion, and illustration of most relevant representations, see Prag (1985). 7 London, BM, Br. 621 (LIMC s.v. Eteokles #17), from Vulci. A similar bearded/ beardless pairing appears, without names, in two later Etruscan representations of the brothers dying, Oxford Ashm. 1965.359 (carnelian scarab, 3rd c., LIMC ibid. #25) and Siena Mus. Arch. 731 (alabaster urn, first half of 2nd c., LIMC ibid. #34). 8 Contrast the quite different account of the brothers’ deaths in Euripides (Phoin. 1407–22): Eteokles, surprising Polyneikes by ‘the Thessalian trick’ of a momentary feigned retreat, stabs him through the navel, and he falls; then, as Eteokles is bending over the supposed corpse to strip it, Polyneikes with his last strength thrusts his sword into his brother’s liver.
Polyneikes in Aeschylus’ Seven
85
mately from that famous play,9 [108] whether via lost Attic pictorial representations or via a production in Italy.10 From the moment when we first learn of the brothers’ deaths, the text of Seven is almost obsessive in its insistence on the symmetry between them. Whenever they are referred to, either the reference is to the pair collectively11 or else words relating to one brother are immediately followed by similar words relating to the other;12 the only distinction between them ever mentioned in this section of the play is that Polyneikes had been in exile.13 Why should Aeschylus, while preserving verbal symmetry so carefully, have made such a striking breach in visual symmetry, if not to keep it before our minds that Polyneikes is the elder brother? His seniority must thus be of significance for the understanding of the play—and doubtless for the same reason that makes seniority important in the Euripidean and Sophoclean plays, namely for the bearing it has on the brothers’ dispute. Polyneikes claims that right (and hence the goddess Dike) is on his side in the quarrel (644–8); this claim is contemptuously dismissed by Eteokles (662–71), but the upright prophet Amphiaraos by implication agrees with Polyneikes (584) though insisting that no claim of right entitles him to make war on his own city. The nature of the claim is not made explicit, but if it is (1) justified and (2) related to Polyneikes’ seniority, it can only be that, as in Sophocles, he holds that this seniority entitles him to the kingship of Thebes (which, unlike Oedipus’ material possessions, cannot be split between the brothers; the story of its being held on a time-share basis was
9 Famous among the western Greeks too, as witness the praise of it by Gorgias (fr. 24 DK). 10 For the latter possibility cf. Taplin (1993: 12–29),* though Etruscan material does not come within his purview. 11 805–19, 826–39, 845–6, 849–50, 875–960 passim, 1000–4. 12 So constantly in 961–92. At 998–9 the transmitted text has Ng Ng ı ø
ŒÆŒøÐ ¼ Æ ¯ Œº Ø IæåÅªÆ without any corresponding apostrophe to Polyneikes, but virtually all modern editors assume a lacuna, and it would be all too easy for a line or two to drop out in a passage in which almost every line begins with N.* 13 979, 991; also 981 if ıŁ d (V? K: øŁ d cett.) is the correct reading.
86
Polyneikes in Aeschylus’ Seven
probably unknown to Aeschylus, and in any case would require, if anything, that Eteokles should be the senior brother as in Euripides). And now we can observe one thing more. Into Eteokles’ savage [109] denunciation of his brother, in which he affirms with a crushing accumulation of negatives that at no time in his life has Polyneikes had anything to do with Dike (664–9), is slipped a phrase that undermines his whole argument: Dike never inclined her countenance to him when he was born, or when he was growing, or in his adolescence, or when his beard grew (666). It is precisely because his beard has grown (whereas Eteokles’, as we can see for ourselves, has not) that Polyneikes’ claim against his brother (though not against Thebes) is countenanced by Dike. This is of course not the only revealing slip in this speech of Eteokles’: later he will claim the sanction of Dike ( ¼ºº Æкº K ØŒ æ;) for facing Polyneikes in single combat ‘ruler against ruler, brother against brother, enemy against enemy’ (672–5), as if the second of these three phrases, like the first and third, were the most natural and proper thing in the world. It may well be asked: if the seniority of Polyneikes is so important to Seven, why is it not mentioned explicitly at an early stage, as in Euripides and Sophocles? The first indication of it, and that only an indirect one, comes when the play is already two-thirds over;14 and not till the bodies are brought on, at line 848, will it become apparent to every spectator that Polyneikes is the elder. This is not how a dramatist brings a new and vital fact to the notice of his audience. We may deduce, therefore, that the fact was not new to them. It is possible that the seniority of Polyneikes was firmly established in tradition, but it is unlikely, given the preponderance of literary evidence already noted on the other side.15 More probably there 14 To be sure, we know before then that Polyneikes has a claim of right which Amphiaraos considers valid, but we have no way of knowing what the claim is based on. 15 There are faint indications that the tradition before Aeschylus was divided on the question of seniority. On the one hand, the one narrative we have that is both explicit on the matter and independent of surviving tragedy, that in A Iliad 4.376, makes Eteokles the senior (and Hellanikos, in the late fifth century, may have done so too: see Jacoby on FGrH 4 F 98, noting that Hellanikos makes Eteokles propose a division of Oedipus’ inheritance and arguing that such an initiative would ‘naturally’
Polyneikes in Aeschylus’ Seven
87
had been mention of the brothers’ [110] relative ages in Oedipus, the play that preceded Seven; perhaps e.g. Eteokles in that play had expressed resentment at his brother’s claim that seniority gave him the right to be head of the oikos and ruler of Thebes when Oedipus should die.16 When the rights and wrongs of the brothers’ quarrel come back on to the agenda in Seven 576ff., the audience will have recalled that earlier scene, and thus will have understood the significance of the seemingly casual mention of Polyneikes’ beard at Seven 666. At any rate, for Aeschylus, Polyneikes was the elder brother and Eteokles a usurper—but for all that, Eteokles, who had the support of, and was defending, the polis of Thebes (cf. perhaps Soph. OC 1298 ºØ b Æ), deservedly won the victory for Thebes while just as deservedly perishing himself as Polyneikes did. And in the end they will lie side by side, equal in everything—except that Polyneikes will take to his grave the visible evidence of his maturer manhood.17
come from the elder brother). On the other hand, in the cyclic Thebais (fr. 2 Davies <¼ West>) Oedipus is said to have cursed both his sons because of an improper action committed by Polyneikes alone, as if Polyneikes had been acting on his brother’s behalf as well as his own, and Davies (1989: 25) has suggested that this may indicate that ‘our epic . . . regarded [Polyneikes] as the elder brother’. It has even been speculated (Mastronarde 1994: 27 n. 3) that ‘the brothers were originally twins’: ı Æ . . . ŒÆŒ (Seven 849, cf. 781–2) might conceivably be a distant echo of such a tradition. 16 In classical Athens, seniority among brothers was not normally of significance in inheritance law (they would either divide their father’s estate equally or else leave it undivided and own it jointly), but it could become so if there was a dispute over an ‘asset’ which could neither be divided nor shared, especially over marriage with an epikle¯ros. Thus in Menander’s Aspis, when the supposed death of Kleostratos leaves his sister as an epikle¯ros, her elderly and avaricious uncle Smikrines claims that he has a prior legal right to take her in marriage ahead of his younger brother Chairestratos (Aspis 141–3, 185–7, 254–6), and the rest of the family, though desperate to prevent the marriage, do not dispute this claim but instead first try to compromise, offering to renounce all claim to Kleostratos’ property in return for being allowed to find the girl a more suitable husband (261–9), and then, when this offer is refused, commence an intrigue designed to outwit Smikrines by dangling before him the bait of another epikle¯ros with a much larger estate. 17 This chapter was originally published in Museum Criticum 30/31 (1995/6) 105– 10. I am most grateful to Pacini Editore for giving permission for its republication.
88
Polyneikes in Aeschylus’ Seven ADDENDA
p. 83 while it is true that in Seven it is consistently assumed that the deaths of the brothers ‘effectively extinguish the royal line’, the passage that explicitly calls them IŒ ı is probably spurious; see Chapter 4 above. p. 85 n. 10 add Taplin (2007: esp. 5–15); but his material does not include anything that clearly relates to Seven. p. 85 n. 12 I should not have said that ‘virtually all modern editors assume a lacuna’ when the (then) two most recent, Hutchinson (1985) and West (1990a), deleted ’¯ Œº Ø IæåÅªÆ as a gloss— as indeed now does Sommerstein (2008). Several manuscripts (including the oldest, M) have these words only in the margin, and some others omit them entirely.
6 The beginning and the end of Aeschylus’ Danaid trilogy Once upon a time, not very long ago, many of us thought we knew, in general terms, the structure and content of the trilogy by Aeschylus of which The Suppliants is the only part that survives. The production had consisted of The Suppliants, The Egyptians, and The Danaids, with the satyr-play Amymone. The surviving play portrayed the arrival at Argos of Danaos and his daughters, in flight from Egypt and the threat of a forced marriage with their cousins, the sons of Aigyptos; the acceptance of their supplication by the Argive king and people; the arrival in pursuit of the sons of Aigyptos, with an army; the attempt by the herald of this army to seize the Danaids, his confrontation with the Argive king, and his declaration of war; and ended with the Danaids and their father leaving for the city with an armed escort. It was known, from many other accounts of the legend, that in the end the Danaids were after all compelled to marry the sons of Aigyptos, and that on the wedding-night all but one of the bridegrooms were murdered by their brides. Of the fragments of The Danaids, one clearly referred, apparently in the present or future tense, to the ceremony of ‘awakening’ the bridal couples on the following morning, indicating that the action of the play was set on the day after the murders;1 another, comprising eight lines of a 1 Aesch. fr. 43. Seaford (1984b) demonstrates the close connection of this passage with nuptial rituals, with particular reference to the phrase f ŒæØ ŒÆd ŒæÆØ. He notes that if the speaker is Danaos (see below, p. 109) there may well also be ironical reference to rituals for the propitiation of the dead, among whom, as Danaos knows, the sons of Aigyptos are now numbered; it should be observed, however, that the presence of æ ı ØÐ (line 2) does not necessarily imply a reference to chthonic cults (cf. Supp. 140, 210, Ag. 840, 1647, Eum. 236) and that f ı çı (line 2) can as easily mean ‘the bridal couples’ (cf. Seven 757) as ‘the bridegrooms’.
90
Aeschylus’ Danaid trilogy
speech by Aphrodite emphasizing the universality of her power,2 showed that during the play, in one way or another, the action of the Danaids and their father, who set Aphrodite at naught, was condemned, and Hypermestra, the lone dissident who had spared her husband Lynkeus, was vindicated. How this was done, and what happened to the dramatis personae thereafter, remained obscure. As for the middle play, The Egyptians, it seemed likely that this was concerned with preparations for the marriages, perhaps in the [112] aftermath of a battle in which the Argive king, Pelasgos, had been killed, and Danaos had taken over power in Argos and undertaken to end the war.3 There remained some difficulties with this reconstruction which were not entirely due to lack of evidence. The title of the second play implied a chorus of Egyptians (presumably soldiers; not the prospective bridegrooms themselves, who would have been `NªıØ()ÆØ rather than `Nª ØØ).4 In that case it was at first sight hard to see how Danaos and his daughters could have contrived their murder-plot. Various possibilities, indeed, existed: perhaps the play, like Aristophanes’ Frogs, was named after a secondary chorus,5 the Danaids being the main chorus; perhaps the plot was so familiar a part of the story that the dramatist took the audience’s knowledge of it for granted, and exploited it to ironic effect in scenes of negotiation between Danaos and his enemies.6 But none of these expedients seemed fully satisfactory. In addition there was a fundamental problem in the interpretation of the surviving play. In many passages the Danaids express the 2
Aesch. fr. 44. Cf. Winnington-Ingram (1961: 144–7 ¼ 1983: 61–5); Garvie (1969: 197–9); Friis Johansen and Whittle (1980: 50–1). 4 When a tragedy is named after its chorus, and that chorus represents a group of brothers or sisters, the title takes a patronymic, not an ethnic form; in addition to ˜Æ Æ we have from Aeschylus HºØ , HæÆŒº ØÐ ÆØ and ˝Åæ , from Sophocles Å æÆØ, from Euripides HæÆŒº ØÐ ÆØ, —ÅºØ , and "Å ÆØ.* 5 So e.g. Cunningham (1953: 230–1, 1962: 190); Winnington-Ingram (1961: 146 ¼ 1983: 64); Lesky (1983: 70). It is perhaps worth observing here that there is no known or even probable instance of an Aeschylean tragic trilogy in which the three choruses all represent groups of the same gender, let alone in which they all represent the same group (Sommerstein <1996a:>, ch. 3). 6 Friis Johansen and Whittle (1980: 51). 3
Aeschylus’ Danaid trilogy
91
intense horror they feel in contemplating marriage with the sons of Aigyptos.7 Most if not all these passages can be understood in two ways. Is it these particular suitors to whom the Danaids object8—say because they are enemies of their father’s, or because they are physically or morally repulsive, or because they are trying to impose their will by force? Or are hostile to marriage in principle,9 as the cautionary advice of the secondary chorus, at the end of the play, seems to presuppose that they are?10 This is no trivial issue. For a woman to object, with her father’s [113] strong support, to a particular undesirable suitor was perfectly acceptable behaviour (though of course it became another matter if the objection was carried to the point of murder). For a woman to object to marriage as such was, in the view of virtually all archaic and classical Greek males of whose opinions we know anything, to fly in the face of the laws of nature, and certainly it was to subvert one of the fundamental principles of civilized Greek society. The choice between the two interpretations will thus largely determine the spectator’s entire view of the rights and wrongs of the dispute. And Aeschylus seemed to have made it extraordinarily difficult for the spectator to make that choice. Perhaps history should have warned us not to be so confident in what we thought was our knowledge. Until 1952, after all, it had almost universally been taken as certain that The Suppliants was by a long way the earliest surviving Greek tragedy; and when a papyrus published in that year11 showed that it was produced in competition with Sophocles and therefore no earlier than 466 bc, scholars were not lacking who roundly declared that the play could not possibly have been written so late and that the didaskalia in the papyrus 7 Supp. 8–10, 37–9, 80–1, 141–3 ¼ 151–3, 332–7, 392–5, 643–5, 788–807, 1031–3, 1052–3, 1062–3. 8 So e.g. Mackinnon (1978); Friis Johansen and Whittle (1980: 32–7). 9 So e.g. Winnington-Ingram (1961: 144 ¼ 1983: 60); Garvie (1969: 222). 10 Supp. 1034–42, 1056, 1059, 1061. The singers, whoever they are (see p. 101), are certainly on the side of the Danaids and not of their enemies, and can hardly therefore be suggesting that they ought to accept marriage with the men whose agents we have seen treating both them and the gods with the utmost hybris; nor could it reasonably be suggested that refusal of such a marriage as that constituted contempt for Aphrodite (1034 ff.). 11 POxy 2256 fr. 3.
92
Aeschylus’ Danaid trilogy
fragment must relate to a posthumous restaging.12 We are now all accustomed to the idea that the Danaid trilogy is a rather late work of Aeschylus,* and much has been done to show how its techniques and ideas relate to those of the Oresteia13—though some people took longer to adapt to the new evidence than others.14 In the last few years our understanding of the Danaid trilogy has been, or should have been, revolutionized. The makers of this revolution have been Martin Sicherl and Wolfgang Ro¨sler.15 Sicherl drew attention to a scholion on Suppliants 37 which finds no explanation within the surviving play. In this passage the chorus speak of the threatened marriage as one ‘which Right forbids’, ºŒæø
H ¨ Ø Yæª Ø. Since they immediately go on to claim that the Aigyptiads are hoping to ‘take possession of what belongs to their uncle, against his and our will’ (ç æØ Ø ÆæÆ ºç Æ ’ I Œ ø ), one would have thought that the natural interpretation was that the proposed marriages were criminal for that reason, namely because the brides would not be given in marriage, as was proper, but seized by force. This, however, is not the interpretation offered by the scholiast. His note ends with the words Ø e c ŁÆ ÆøŁB ÆØ e ÆæÆ, ‘because of their father not [114] being killed’, or in other words ‘for fear that their father may be killed’.16 Now the idea that Danaos may be in danger of being killed is one that does appear later in the play. In lines 980ff. he returns from Argos accompanied by armed men, and explains that the Argive people have voted him a bodyguard ‘lest I unexpectedly perish by a secret and violent death’ (987–8). This, however, is not presented as bearing any direct relation to the marriage of his daughters; the assassin would presumably be either an enemy agent who had entered the city unnoticed, or an Argive who resented the way his city 12
E.g. Earp (1953: 119); Murray (1955: 2); Wolff (1958–9). Special mention is due here to Garvie (1969); Herington (1970); and Taplin (1977). 14 When I came to Nottingham in 1974, I found that my predecessor’s last examination paper in Greek Tragedy had its first question ‘Why was the Supplices formerly considered the earliest extant play of Aeschylus?’ 15 Sicherl (1986); Ro¨sler (1993). 16 Sicherl (1986: 92) suggests, reasonably enough, that < º ŁÆØ> vel sim. may have dropped out of the text of the scholion. 13
Aeschylus’ Danaid trilogy
93
had been made to undertake a dangerous war for the sake of this foreigner. On the evidence of this play alone we would conclude that Danaos is running a greater risk to his life by refusing his daughters to the Aigyptiads than he would by agreeing to the marriages: the refusal, after all, has involved him in war in Egypt (cf. 741–2), a long and perilous sea voyage, and an appeal to the mercy of strangers. The scholiast, therefore, must have relied on evidence from elsewhere. Sicherl found that evidence in the form of a number of ancient references to an oracle which had been given to Danaos, to the effect that he would be killed by his son-in-law;17 and he proposed that this oracle should be seen as the background to, and the explanation of, the events of the surviving play. Danaos’ reaction to the oracle will have been to resolve that he would never have a son-in-law: that he would never allow his daughters to marry. Perhaps (though I shall be suggesting that this is unlikely) he confided the terms of the oracle to his daughters; perhaps he indoctrinated them to loathe the very thought of marriage and of sex; at any rate when we see them they seem as determined as he is that they shall remain in their virginity,18 and treat marriage as equivalent to [115] enslavement.19 There is no reason now, within what we know of the trilogy, to believe that Aigyptos’ initial suggestion of a marriage alliance between his sons 17 Sicherl (1986: 92–3). Such an oracle is referred to, in varying terms, by ancient commentators on Prometheus Bound (853), on Euripides’ Orestes (872), on the Iliad (1.42 and 4.171), and on Statius’ Thebaid (2.222 and 6.269). Only the two Statian scholia make it explicit whether the oracle told Danaos he would be killed by ‘his sonin-law’ or by ‘a son of Aigyptos’—and they contradict each other on the matter; but as Sicherl rightly argues, the passage Supp. 991 ff., where Danaos anxiously warns his daughters against the possible amorous attentions of Argives, shows that Aeschylus is assuming a version of the oracle which predicted Danaos’ death at the hands of ‘his son-in-law’ (or perhaps more accurately ‘the bedfellow of his daughter’). 18 Cf. especially Supp. 1015–17. Friis Johansen and Whittle (1980: 32) argue otherwise on the basis of Supp. 79–82, 337, 1031–3, 1052–3, 1062–7; but (1) 79–82 supports their case only with an emendation ( ØŒ ª Oberdick) which creates an anacoluthon (Ł d . . . oæØ . . . ıª ºØ’ i ØŒ ª ), (2) 337 is corrupt and in any case presupposes that to be a wife is to be a chattel (cf. Œ ŒÅ ı), (3) to say one loathes beef does not necessarily imply that one finds mutton acceptable, and likewise to say that one abhors a forced marriage with the Aigyptiads does not necessarily imply that one would accept some other marriage, (4) in 1052–3 the Danaids make no attempt at all to rebut the accusation that they are slighting Aphrodite. See Griffith (1986: 325). 19 Cf. especially Supp. 335–7, 392–3; see Griffith (1986: 325–6).
94
Aeschylus’ Danaid trilogy
and Danaos’ daughters was anything but a normal proposal of the kind that might have been made by anyone, and a proposal, moreover, that had distinct dynastic and political advantages (cf. Supp. 338) since it would avoid bringing outsiders into the circle of the royal house. Danaos’ refusal of the proposal—for which he can have given no adequate reason—was bound to be taken by Aigyptos as an insult, and it is not surprising that this should have led to war. Who is in the right? It is much harder to say than we once thought. It remains true that Aigyptos and his sons have no moral right to impose themselves on the Danaids by force; but it is also true that Danaos is committing what contemporaries regarded as a very great wrong against his daughters by keeping them unmarried. This is what Klytaimestra and Aigisthos did to Electra; it is what Akrisios (again because of an oracular prediction of his own death)20 tried to do to Danae. It is remarkable that when Lysias, in his speech against Eratosthenes, catalogues the crimes of the Thirty, he does so in the following manner: For these men expelled many of their fellow-citizens and forced them to live among the enemy; they put many to death unjustly and left them unburied; they disfranchised many citizens in good standing; and they prevented the intended marriages of many men’s daughters [emphasis mine]. (Lys. 12.21)
The last-mentioned is apparently a crime at least equal in gravity to the others; otherwise it would make a sad anticlimax.21 So far as we know it was not punishable, presumably because it was never considered necessary to make it so: a kyrios who kept his daughter or ward unmarried well beyond the normal age would inevitably be so stigmatized by the community around him that he would be unable to persist in his attitude. Even without going into issues of gender ideologies or biological theories, it is easy to see why such behaviour was considered so wrong. To keep a woman unmarried into adulthood was to condemn her to social death: the wife had a recognized place in society and in religion, so did the young parthenos, but the mature parthenos had none (as the married yet virgin Electra bitterly 20 This oracle formed part of the myth as known to Aeschylus’ (younger?) contemporary Pherekydes (FGrH 3 F 10). 21 On the heinousness of this crime cf. Schaps (1979: 41).
Aeschylus’ Danaid trilogy
95
complains in Euripides’ play of that name).22 Moreover, through no fault of her own and with no possibility of redress, [116] she was being deprived of the chance to bear children who would honour and support her in old age, and tend her tomb after her death. To be sure, the Danaids seem to agree with their father’s determination, and speak of it as their own; but this is hardly surprising considering that they have been under his exclusive control throughout their lives and have not been exposed to any alternative view of things. Aeschylus had never heard of the concept of ‘false consciousness’, but it is not unfair to interpret his portrayal of the Danaids in terms of that concept. They themselves speak of their father (Supp. 12) as a player manipulating pieces on a game-board, and they are indeed, more fully even than they know, pawns in his hands. We also have to re-evaluate the relations between Danaos, his daughters, and the Argives. The Argives grant asylum to the group as persecuted and suppliant foreigners, and are persuaded (rather deviously, as I have shown <elsewhere>)23 to protect them even at the risk of war. Would they ever have done this if they or their king had known that what they were really defending was Danaos’ right to keep his daughters in permanent spinsterhood? Such are some of the implications of that six-word scholion on Supp. 37. But Sicherl’s hypothesis raises, as Sicherl himself clearly saw, a fresh difficulty. If it is right, the oracle given to Danaos is a fact fundamental to our whole understanding of the action—and yet it is never mentioned. Nor can we assume that it was a fixed feature of the myth which could be taken for granted as being known to the audience in advance, since there certainly existed at least one version of the myth in which the oracle did not appear.24 Sicherl’s own 22
Eur. El. 310–11, cf. 178–80. <See Ch. 7 below.> 24 Pind. Pyth. 9.112–16 (performed in 474) has Danaos himself giving forty-eight of his daughters (i.e. all but Hypermestra and Amymone) to new husbands; the fact that he is alive to do so implies a version of the myth (such as we know from later evidence to have existed) in which his rejection of the Aigyptiads as sons-in-law, and their subsequent murder, was caused not by an oracle predicting his death but simply by enmity (probably a dispute with Aigyptos over the rulership of Egypt). This seems also to have been the version best known to Herodotos, who refers to the Achaian Archandros (2.98; cf. Paus. 7.16) as ˜Æ ÆF ªÆ æ; it may have included a reconciliation between Danaos and Lynkeus (cf. [Apollod.] Bibl. 2.1.5). 23
96
Aeschylus’ Danaid trilogy
solution25 was to suggest that the oracle was mentioned in the final play of the trilogy, The Danaids, and that when the audience were made aware of it they would be forced to revise in retrospect their assessment of what had been done and said previously. It is, however, hard to understand what advantage is gained by postponing the revelation so long. It is not as though the Danaids and their father were compelled to maintain a front of deception continuously during The Suppliants. For about half the length of that play (1–233 and 524–824) the father and daughters, or the daughters alone, are by themselves on stage and able to speak freely; yet still the author will not allow them to make any reference to the oracle, and the most acute [117] spectator will find it hard to perceive any indication that their motives in fleeing to Argos and seeking asylum are any other than those which they declare to Pelasgos. The spectator will, in fact, be in the same position as modern interpreters were before Sicherl wrote, and, not knowing what is to come, will experience, as Ro¨sler puts it, ‘nicht der Aufbau von Spannung, sondern Mißverstehen und Nichtverstehen’.26 It would be a very different matter if The Suppliants were not the first play of the trilogy: if, that is, the oracle, and Danaos’ response to it, had been mentioned in a preceding play. The audience would then have been fully aware of the deception being practised on the Argives, and it would be unnecessary to make Danaos and his daughters say things to each other in The Suppliants which were already well known both to them and to the spectators. It is the achievement of Ro¨sler to show that there is a hypothesis which enables precisely this to have been the case. this is, quite simply, that The Egyptians was not the second play of the trilogy; it was the first.27 There is no evidence against this hypothesis. The papyrus didaskalia gives the names of the third and fourth plays of Aeschylus’ production, ˜Æ [Æ]Ø ı[ ÅØ], in the expected place; but the names of the first two plays are lost, and we cannot tell whether 25
Sicherl (1986: 98). Ro¨sler (1993: 7). 27 This had been suggested, though with less cogent reasons, by several earlier scholars, beginning with Schlegel (for references see Garvie 1969: 185–6), and more recently defended as a possibility by Griffith (1986: 324). It is a curious feature of Sicherl’s article that in thirty pages The Egyptians is hardly mentioned. 26
Aeschylus’ Danaid trilogy
97
they were listed as IŒ Ø `NªıØ or vice versa. And the hypothesis explains so much. It accounts for the scholion on Supp. 37: the commentator was assuming that readers of The Suppliants were familiar with the preceding play.28 It solves the problems we noted about The Egyptians. That play did not deal with the making of the murder-plot, or with marriage negotiations at Argos: it was set in Egypt, with a chorus no doubt of Egyptian nobles or elders (as the title suggests), and dealt with the earlier relations between Aigyptos and Danaos. In addition to this, the hypothesis, as Ro¨sler emphasizes, makes intelligible to the spectator the seemingly obsessive and irrelevant anxiety of Danaos (Supp. 980–1013) about his daughters’ behaviour in Argos, because the spectator will be aware of Danaos’ real motive which he cannot avow in the presence of third parties like his Argive bodyguard;29 and it makes it possible to explain the curious assumption made by more than one character in The Suppliants, and never denied by the Danaids themselves, that under Egyptian law the Aigyptiads are legally entitled to possession of their cousins, regardless of the latter’s own wishes or of their father’s (cf. [118] Supp. 387ff., 916–18, 932–3)30—this law will have been spoken of more explicitly in the preceding play. There is, nevertheless, or so some may feel, a price to pay. The Suppliants ends with the entry of the Danaids into Argos, and with war about to break out between the Argives under Pelasgos and the Egyptians. The Argives are determined to defend the rights (as they see it) of the Danaids. Danaos’ status in Argos is that of a resident alien who has been assigned a bodyguard for his protection; he and his daughters are adamant that under no circumstances will they consent to marriage with the Aigyptiads. When the action resumes, on Ro¨sler’s theory, the situation has been utterly transformed. Pelasgos is dead; Danaos, presumably, is ruler of Argos; the war is over; the detested marriages have been celebrated; it is now the morning after, and all the bridegrooms but one have been slain. That may seem too much important action to be passed over in the interval 28 Cf. Eum. 64, where at the moment of Orestes’ first appearance in the play he is described as e b çfi ƪ Ø ŒÆåø , i.e. as at the end of Cho. 29 Ro¨sler (1993: 11–15). 30 See Friis Johansen and Whittle (1980: 34–7); Sicherl (1986: 99–101).
98
Aeschylus’ Danaid trilogy
between two plays of a trilogy.31 We must, however, be careful not to judge all Aeschylean trilogies by the standards of the Oresteia. The Theban trilogy must have passed over some vital events in the story: it almost certainly omitted Oedipus’ arrival at Thebes and his marriage (since the satyr-play, The Sphinx, was written around these events), and it may well also have omitted the quarrel between Eteokles and Polyneikes and the latter’s expulsion from Thebes. The trilogy based on the Iliad seems to have included in its three plays most of the main events of the last nine books of the epic (with the exception, probably, of book 23); but that based on the Odyssey seems to have passed straight from the Nekyia (in #ıåƪøª) to Odysseus’ arrival at his own palace (in Penelope), omitting all his intervening adventures. There would be no difficulty in putting the audience in the picture, at the beginning of The Danaids, by means of a prologue speech and/or of one or more retrospective choral songs. I believe, therefore, that Ro¨sler’s theory is correct and that The Egyptians was the first play of the trilogy. I should like now to carry the investigation further in various directions, both at the beginning of the trilogy and at the end. I have three questions to put to the evidence.
1. WHERE AND BY WHOM WAS THE ORACLE GIVEN? Since Danaos was in Egypt, the likeliest answer is surely the famous oracle of Ammon at Siwa. This oracle had long been well known to Greeks.32 Spartans may have consulted it as early as 514,33 and Herodotos was prepared to believe Kroisos had done so more than thirty [119] years earlier.34 The Athenian statesman Kimon consulted 31 Garvie (1969: 186), indeed, tacitly takes this for granted: ‘At the end of the Supplices we are still in the early stages of the action. It is impossible to see how the situation could be resolved in a single following play [emphasis mine].’ 32 On Greek relations with the Ammon oracle in the archaic and classical periods, see Classen (1959). 33 Plut. Lys. 25.3–4 (¼ Ephoros FGrH 70 F 260); cf. Malkin (1990: 542). 34 Hdt. 1.46.2–3.
Aeschylus’ Danaid trilogy
99
it in the 460s.35 Pindar composed at least one hymn to Ammon,36 and in the fourth Pythian ode, composed within a year or two of the Danaid trilogy, he mentioned the sanctuary of Ammon in connection with Libye, grandmother of Danaos.37 In Aristophanes’ Birds, half a century later, Ammon is treated as on a par with Delphi and Dodona.38 No other oracle in or near Egypt—indeed no other oracle in any non-Greek land—had anything like this degree of prestige with Greeks. Greeks, as is well known, identified Ammon with Zeus (so that Pindar can invoke ‘Ammon, lord of Olympos’).39 The oracle given to Danaos, then, was given by Zeus. This fact makes much that is said in The Suppliants even more ironic than, on Ro¨sler’s hypothesis, it would be in any case. For the Danaids place their absolute trust in Zeus.40 He is their ancestor; he was gracious to their Argive forebear Io; he will be gracious to them; he will protect them from hateful marriage and drown their pursuers at sea (both of which he fails to do). All this when it is Zeus himself who has predicted that at least one of them will marry! On the traditional view of The Suppliants, it is plausible to suppose that the audience, sympathizing with the Danaids’ apparent victimhood, will be to some extent perplexed and disturbed by Zeus’ failure to grant their prayers; they know, of course, from the myth, that the marriages will take place, but may well not fully understand why Zeus seems to be siding for the moment with the hybristic Aigyptiads. On the new interpretation the audience will be aware from the start that the prayers must be futile. Deluded as the Danaids may be, it would be surprising if they were to put such enormous faith in the support of Zeus if they knew of the oracle he had given their father. It is therefore likely that we are meant 35
Plut. Kim. 18.7. Cf. Pind. fr. 36, Paus. 9.16.1. 37 Pind. Pyth. 4.13–16 (performed 462) çÆ d ªaæ A’ K ±ºØºŒı b ªA ’¯çØ ŒæÆ Iø ÞÇÆ çı ŁÆØ ºÅ æ ˜Øe K @ ø Ł ŁºØ. 38 Ar. Birds 618–19, 716; cf. too Eur. Alk. 115–16, Eur. fr. 955h Nauck-Snell <¼ Kannicht>. 39 Pind. fr. 36; cf. Pind. Pyth. 4.16. 40 Supp. 1, 15–18, 26–39, 40–7, 168–75, 206, 209, 359–60, 385–6, 402–4, 437, 524–37, 571–99, 811–16, 892–902, 1052–3, 1062–9. 36
100
Aeschylus’ Danaid trilogy
to infer that they do not know. Danaos has not thought it safe to trust them with the secret: he has simply brought them up not to look forward to marriage as a natural and inevitable stage in life (as any normal parent would do) but to dread it as, literally, worse than death (cf. Supp. 154–61). I see nothing in Supp. 980–1017 that requires us to suppose that the Danaids know the truth.41 If they do not know, indeed, 1015 ( N ªæ Ø c Ł ØÐ º ıÆØ ) acquires a new and very piquant irony: [120] their determination, and their father’s, will be defeated not by any ‘new counsel’ of the gods, but by the counsel the gods have taken long since, of which Danaos is well aware and his daughters are not. If this is correct, we might expect that Danaos himself will never trust to, or appeal for, the support of Zeus; for he, unlike his daughters, knows he cannot hope to obtain it. This expectation is not entirely fulfilled. Supp. 206–11, despite textual confusion over the order and assignment of lines, is clearly a joint appeal to Zeus by Danaos and his daughters; it contains three prayers to Zeus (in 206, 210, 209) at least one of which must be uttered by Danaos.42 It must be remembered, though, that Danaos is ex hypothesi deceiving not only the Argives but also his daughters: he cannot let them become aware that Zeus is hostile to his and their aspirations, and he appears to confine himself (in 210, the only line of the three which recent editors give him)43 to echoing their words. The Danaids, at any rate, all through the play, regard Zeus as their patron and protector; in the name of Zeus Hikesios they force Pelasgos, and through him the Argives, to support them; and at the end, just as at the beginning, they are still praying to Zeus to release them from their troubles as he once released Io. But just before that final song, there has been a strong note of doubt sounded, by the secondary chorus who sing 1034–51 and parts of 1052–62. There is now something of a consensus that this 41 Pace Ro¨sler 1993: 11–15. The author of the scholion on 37 of course assumed that the Danaids knew of the oracle—but he may have assumed wrongly. 42 Danaos must be the speaker of 210, since YØ BÆ (210) is evidently a response by another speaker to Z f . . . YØ (206), and the first half of 206 continues a sentence (çıº ÆØ b BŁÆØ Ł Œ a Kç ) whose content shows that it is spoken by the chorus(-leader). 43 Page (1972); Friis Johansen and Whittle (1980); West (1990a).
Aeschylus’ Danaid trilogy
101
secondary chorus is to be identified with the Argive soldiers who escort Danaos and his daughters into the city;44 it has been pointed out by Seaford45 that the lyric exchanges have some of the features of responsive hymeneal songs which were typically sung by two choruses of opposite sexes. I have argued in the past that it is the maidservants of the Danaids who sing here, but I now believe this to have been an error, remarking, however, that if the maids do not sing they have no dramatic function and it would be best to remove them from the text altogether46 (which could be done by deleting 977–[121]9).47 But I am not at present so interested in who these singers are, as in what they say, especially about Zeus. They emphasize the close connection between him and Aphrodite (1035), foreshadowing perhaps her intervention in the following play. And then, in their antistrophe, they come closer than anyone in the play to expressing a true perception of the will of Zeus: Why, why did they have so fair a voyage in their swift-sped pursuit? What is fated, you know, that will happen:
44 They are so identified both by Friis Johansen and Whittle (1980) and by West (1990a), and this view had been accepted as ‘most likely’ by Taplin (1977: 232) and is endorsed by Ro¨sler (1993: 14–15) on the basis of Seaford’s arguments (see next note). 45 Seaford (1984b, 1987: 114–15). 46 For my earlier views see Sommerstein (1977b: 76–9), cf. Ireland (1986: 19–20). The sole direct textual evidence for the presence of maidservants on stage consists in the three lines 977–9—and anyone who accepts the basic approach to the trilogy adopted by Sicherl, Ro¨sler, and in this , must logically reject these lines, since Danaos could not give a ç æ (979) to daughters whom he wanted to keep unmarried. There is no other clear reference to the maids anywhere in the play (in 954 M reads çºØ OØ in the masculine, evidently referring to Argive soldiers, and the following words Łæ ºÆFÆØ show that this reading is correct) and no point at which they could plausibly be brought on stage (if they have been with the Danaids from the start of the play, it would make this ‘chorus of desperate fugitive exiles . . . the only chorus in all of Greek tragedy to trail around a crowd of permanent attendants’ (Taplin 1977: 233); if they enter at 977 or not long before, where on earth are they supposed to have come from?). It is likely that 977–9 were inserted by a later producer in order to validate the presence on stage of these supernumerary performers, just as Seven 861–74 and 996–7 were inserted to validate the presence of Antigone and Ismene (Taplin 1977: 176–80; Hutchinson 1985: 190–1, 207).* 47 Taplin (1977: 222–38) suggested, on these grounds and others, that the whole anapaestic interlude 966–79 was an interpolation, replacing a ‘strophic act-dividing song’; but it is not necessary to be as drastic as that.
102
Aeschylus’ Danaid trilogy the great, unfathomable mind of Zeus cannot be defeated. (1045–9)48
There is, of course, no need to suppose that the singers have knowledge of Danaos’ oracle. They can readily be assumed to be drawing their own conclusions from the failure of the Danaids’ prayers for the destruction of their enemies at sea. But the audience do have such knowledge: they know why the Aigyptiads were allowed to come safe to Argos; they know what is fated; they know what the unfathomable mind of Zeus intends, even if they do not necessarily know in detail how it will come about. The Danaids reject the warning. They leave the scene, as they entered it, determined to resist marriage to the end. And by the time they appear again, they will have carried out the last and most appalling action in their father’s struggle to escape the fate predicted for him.
(2) WHAT WAS THE PLOT OF ‘ THE EGYPTIANS’? There are three things about The Egyptians that we can be sure of. First, the action was set in Egypt; secondly, there was a chorus of male [122] Egyptians, almost certainly not the sons of Aigyptos; thirdly, the audience were told of the oracle given to Danaos. Very little is said in the surviving play about previous events in Egypt. We gather from Supp. 741–2 that there has been a war, in which the sons of Aigyptos fought;49 it has evidently gone badly for Danaos, since he was forced to flee from Egypt. Some accounts of the legend suggest that the war was for the kingship of Egypt;50 but if the ‘son-in-law’ oracle was fundamental to our trilogy, it is much more likely that the war resulted from the refusal by Danaos of a proposal 48
The Danaids themselves had sung earlier (87, 93–5), and will sing again (1057– 8), of the inscrutability of Zeus’ purposes; but their recognition of mysterious ways does not seem to make them in the slightest degree less certain that they deserve, and will receive, his unconditional support. 49 Cf. also perhaps Supp. 83–4 KŒ º ı Øæ Ø . . . çıªØ , 1006–7 ºf (sc. in Egypt) ºf b (on the voyage to Argos). 50 E.g. A Il. 1.42 (¼ [Apollod.] 2.1.4), Hyg. Fab. 168.1, Serv. on Verg. Aen. 10.497; see Friis Johansen and Whittle (1980: 47) (H).
Aeschylus’ Danaid trilogy
103
for a marriage alliance between the two families. It is not even necessary to suppose that the brothers were enemies or rivals before this refusal, though they may have been. The obvious person to inform the audience about the oracle is Danaos himself, and since we have seen he is going to keep it a secret even from his own daughters, he can speak about it only in a soliloquy, which must have formed the prologue of the play. A logical development of the action from then on would be something like the following—though I do not put this forward as the only possibility. Danaos is King of Egypt. He has, however, no son to succeed him, and the Egyptian people, as represented by the chorus of Egyptian notables, are anxious about the future of their kingdom. His brother Aigyptos comes to him with what seems like an admirable proposal: let Danaos’ fifty daughters be married to Aigyptos’ fifty sons, thus keeping both rulership and property entirely within the one family. Fearing for his life because of the oracle, Danaos has to refuse; but he cannot give Aigyptos the true reason for his refusal. He either gives no reason, or gives a false reason which Aigyptos does not believe. Aigyptos, slighted, seeks and secures the support of the people against Danaos, warning them of the danger to Egypt if the succession is not secured (this would correspond to Pelasgos’ persuasion of the Argive assembly reported in Supp. 615–20). There is a law in Egypt that a woman’s closest kinsman has the right to demand her in marriage. Under this law Aigyptos demands, on behalf of his sons, that Danaos surrender his daughters. On the demand being rejected, he declares war on Danaos—a civil war—with the enthusiastic support of the chorus (who perhaps acclaim him as king), thereby causing both great suffering Egypt and, as the audience will know, the death of all but one of his own sons. This, again, would correspond to a scene in the surviving play, the confrontation of Pelasgos and the Egyptian herald in Supp. 911–53, but with the roles reversed. The play perhaps ended with Aigyptos going forth to battle. It would be more impressive if he was accompanied (as Danaos is at the end of The Suppliants) by armed men; these are not likely to have been the members of the chorus (for even if the chorus were men of military age, they will [123] not have been in armour at the start of the play and can hardly have armed themselves on stage), and it is tempting to suggest that there was a secondary chorus composed of
104
Aeschylus’ Danaid trilogy
Aigyptos’ sons, who in that case will have made here their only appearance as living men in the trilogy.
3. HOW DID THE TRILOGY END? There is a substantial discussion in Ro¨sler’s article of what may have transpired in The Danaids.51 He argues, mainly on the basis of a passage of Pausanias (2.19.6), that the play centred on a trial and acquittal of Hypermestra before a popular Argive court, and he sees extensive parallels between the Danaid trilogy and the Oresteia. The suggestion (which has of course been made before in various forms) is in some ways attractive, but there are serious difficulties with it, several of which had been pointed out long before Ro¨sler wrote.52 The problem of how a trial-scene involving Hypermestra, Danaos, and Aphrodite could have been staged if, as The Suppliants seems to indicate, only two actors were available, is in itself not insuperable, since one could suppose (for instance) that Hypermestra did not speak, Aphrodite appearing to speak for her at the point when she was about to make her defence.53 Other problems are less easy to set aside. For example, what crime could Hypermestra have been charged with? What she has done, putting it at its worst, and even assuming (which, as already indicated, I do not believe) that she knew about the ‘son-in-law’ oracle, is to disobey an order by her father to murder her husband because there was a one-in-fifty chance of his becoming, subsequently, her father’s killer.54 If Danaos tried to 51
Ro¨sler (1993: 15–22). Notably by Garvie (1969: 205–11) and Friis Johansen and Whittle (1980: 53–4), who give references to earlier discussions. 53 Nor would I wish to exclude altogether the possibility that a third actor was already available at the time when the Danaid trilogy was composed; Aeschylus was under no obligation to use three actors in every play, and if our text of Agamemnon had ended at line 1071 we would not have known that three speaking actors were employed in Agamemnon. I am grateful to Prof. W. Luppe for stimulating me to spell this point out. 54 I leave for the delectation of ethical philosophers the problem raised by the fact that once Lynkeus’ brothers were dead, Hypermestra by sparing Lynkeus had in effect increased the probability of his becoming the killer of her father from 2 per cent to 100 per cent. 52
Aeschylus’ Danaid trilogy
105
demand that the Argives punish his daughter for an ‘offence’ like that, it would hardly have needed the intervention of Aphrodite to secure her acquittal. It would have made better sense, as Friis Johansen and Whittle note,55 if he had punished her on his own authority, perhaps by imprisonment as in several later sources.56 The hypothesis of a trial (whether of Hypermestra, or her sisters, or Danaos) also has a serious drawback of [124] quite a different kind, connected not with the Danaid trilogy but with the Oresteia. If the last play of the Danaid trilogy contained a trial, involving a popular jury and the intervention of a deity, this would have greatly reduced the impact of the similar scene in Eumenides only a few years later. From the moment when Apollo promises Orestes to find him ØŒÆÆ (Eum. 81) the audience would be expecting a repetition of the trial-scene of The Danaids, and that would be essentially what they were afterwards given. The only significant difference to which Ro¨sler is able to point57 that Orestes is acquitted on a tied vote, whereas in The Danaids he supposes (without evidence)58 that the verdict was unanimous. It is most unlikely that Aeschylus tamely repeated himself in this way. More probably the idea of a trial of Hypermestra derives from some later tragedy, composed under the influence of Eumenides or even of Euripides’ Orestes. And we happen to know of a play, and a famous one, that might be the source: the Lynkeus of Theodektes, a play so well known in the later fourth century that Aristotle can refer to it without troubling to give the author’s name,59 and one which may have contained a double trial (for at one point Lynkeus was led off to execution, and at another an ‘accusation of murder’ was made which was a turning-point in the plot and which can only have been against Danaos). I agree, therefore, with Garvie60 that the supposition that The Danaids centred on a trial of any sort is unnecessary and unlikely. 55
Friis Johansen and Whittle (1980: 53). [Apollod.] 2.1.5; Hor. Carm. 3.11.45; Ov. Her. 14.3 and 83–4. 57 Ro¨sler (1993: 21). 58 Paus. 2.19.6 says merely ŒæØŁ ØÐÆ . . . K ØÐ `æª Ø Iç ª Ø. 59 Poet. 1452a27–9; in a later reference (145<5>b29–32) the author’s name is given. The statements about the play made in this paragraph are derived from these two passages. 60 Garvie (1969: 205–11). 56
106
Aeschylus’ Danaid trilogy
Friis Johansen and Whittle, in their careful study of the evidence for the reconstruction of the trilogy, after reviewing the possible scenarios involving a trial, consider another possibility in these words.61 If trials are banished from Danaides, it is fairly easy to conjure up a plot in which (say) Hypermestra and Lynceus are initially separated [e.g. by the flight of the one and the imprisonment of the other] but eventually reunited, Aphrodite intervenes, and the other Danaids are purified and married a second time; but the only traditional element for it to turn on would be a feud between Danaus and Lynceus resulting in the former’s death, and it is doubtful if even the utmost ingenuity could plausibly reconcile this with the play’s assumed ending.
This suggestion, which is here dismissed so easily, is precisely the one which I am going to advocate.62 [125] The objection of Johansen and Whittle to what I shall call the Lynkeus theory appears to be that it entails an ending involving both the death of Danaos and the remarriage of his daughters, and that this would be incongruous: as they put it a few lines higher on the same page, Danaos’ death ‘could hardly have failed to blight the wedding celebrations’. Yet such a situation—father’s death followed after a short interval by daughter’s marriage—was a perfectly familiar one to fifth-century Athenians; it happened every time a man died leaving a marriageable daughter as an epikleros. Something similar could happen, too, where a man on his deathbed gave his own wife in marriage to another, as Pasion the banker and Demosthenes, father of the orator, both did in the 370s.63 In tragedy, one may compare the marriages of Hyllos and Iole in Sophocles’ Trachiniai or of Electra and Pylades in Euripides’ Electra, both following closely upon traumatic deaths with which the parties were intimately connected. Johansen and Whittle’s misgivings appear to be anachronistic. And there are powerful positive reasons for favouring the Lynkeus theory. If Sicherl and Ro¨sler are right in seeing the ‘son-in-law’ oracle, and Danaos’ attempts to evade its fulfilment, as the spring and origin of the action, the crucial feature of the trilogy’s conclusion 61
Friis Johansen and Whittle (1980: 54). Here I must give her due to my student Elizabeth Randle, who, without knowing anything of Sicherl’s or Ro¨sler’s work, reached very much the same conclusion independently. 63 Dem. 45.28 and 36.8 (Pasion); id. 27.5 (Demosthenes). 62
Aeschylus’ Danaid trilogy
107
can surely have been nothing other than the fulfilment of the oracle; that is, the death of Danaos at the hands of Lynkeus. Danaos’ death, indeed, is multiply determined: first, it is necessary to fulfil the oracle; secondly, Lynkeus has an obligation to avenge the murder of his brothers; thirdly, Aphrodite may well be eager to punish someone who has shown such contempt for her power; and fourthly, as we shall see in a moment, Danaos is a tyrannos, a person therefore (like Aigisthos, or Hipparchos at Athens) whose killing can be seen as a civic duty and a glorious achievement. I would like, therefore, to take the Lynkeus theory as a working hypothesis, and examine whether The Danaids can be plausibly reconstructed on the basis of it. We have the following bits of information or inference to work in: 1. That Danaos is granted a bodyguard by the Argive assembly in The Suppliants (980ff.) is seen by Friis Johansen and Whittle,64 surely rightly, as a pointer to his eventually becoming tyrannos of Argos after the presumed death of Pelasgos: the connection between bodyguards and tyrants was already almost as much a cliche´ as it was for Plato (Rep. 8.566b), was familiar to Athenians from the career of Peisistratos,65 and is exploited in the portrayal of Aigisthos in the Oresteia. This implies, in the first place, that in The Danaids he will be in a position to punish Hypermestra (and Lynkeus, if he can be caught) by his own authority, without the [126] verdict of any court or assembly at all; but also, secondly, that when Lynkeus kills him he will not only be avenging his brothers, but also (like Orestes in Choephoroi)66 liberating the Argive people. The contrast between democracy and autocracy is highlighted in The Suppliants not only in what is said but also in what is seen: Danaos will have been dressed in a manner befitting a king of Egypt;* Pelasgus is dressed so plainly (apart possibly from a sceptre) that the chorus cannot tell whether he is a king, a priest, or a private citizen (246–8).67 It will have been even more important in The Danaids. 64 Friis Johansen and Whittle (1980: 43). That Danaos became ruler of Argos seems to be a fixed part of the myth (ibid. 48). 65 Hdt. 1.59; Arist. Ach. Pol. 14.1–2; Plut. Sol. 30.2–3. 66 Cf. Cho. 302–4, 973, 1046–7. 67 I am indebted for this point to a postgraduate student, Stephania Matsouka. For the sceptre cf. 248 ƒ æææÆ .
108
Aeschylus’ Danaid trilogy
2. Of the two substantial fragments of the play, one (43) describes the morning after the weddings of the Danaids and Aigyptiads, when the couples were to be wakened, according to custom, with song. The text is corrupt, but it clearly describes something that is about to happen, not something that has already happened;68 that is, the action of the play opens early on the morning in question, before the wakening and therefore before the murders have been discovered. The Danaids, therefore, are still (supposed to be) asleep indoors, and so cannot yet be on stage as chorus; there must, therefore, have been a prologue. 3. Lynkeus, having been spared by his bride, must somehow be kept safe from the wrath of Danaos, whether by her contrivance or by his own, close enough to Argos to be able to return and take part in events later in the play. In Theodektes’ play (where he is apparently arrested by order of Danaos after the discovery that Hypermestra has given birth to a son)69 he must have remained in hiding in or near Argos; and concealment, possibly aided by disguise, though for a much shorter period, is a plausible device for Aeschylus to have used. If so, it might be desirable to put the audience in the picture by having Lynkeus appear early on, before the entry of the chorus, and explain what has happened and what he is going to do. This would imply a double prologue, an assumption which Johansen and Whittle find objectionable70 but which can easily be paralleled: Seven against Thebes has a somewhat similar structure, the prologue being divided midway (35/39) by the exit of the [127] citizen-soldiers and the entry of the Scout, while in the prologue of Eumenides (33/34) the Pythia exits and then herself re-enters.71 68 The passage contains no past-tense verb and nothing that can reasonably be emended into a past-tense verb; according to how the text is constituted, the main verb is either a future indicative ( rØ vel sim., line 1) or an imperative (Ł ø , line 3). 69 The first major event within the action of the play was F ÆØı ºBłØ (Arist. Poet. 1455b30); later, but still before the æØ ØÆ, Lynkeus is on the point of being put to death by Danaos (ibid. 1452a27–8). Lynkeus must thus have remained concealed for nine months or more, until the discovery of the child aroused the suspicion that Hypermestra had been a party to his escape; she may then have been forced to reveal his whereabouts. 70 Friis Johansen and Whittle (1980: 53). 71 See Taplin (1977: 362, 268–9) who compares the complex prologues of some late Euripidean plays (Iphigeneia in Tauris, Phoinissai).
Aeschylus’ Danaid trilogy
109
4. A context must be provided for the speech of Aphrodite of which fr. 44 is part. It is not possible, I think, to determine from the speech itself whether it comes from a defence of Hypermestra (as being excused and justified, in her defiance of her father, by her having acted under the influence of Aphrodite’s universal power) or from a denunciation of the other Danaids and their father (as having set that power at naught). Aphrodite may, of course, have done both these things in one speech. 5. Lynkeus must bring about the death of Danaos, to avenge his brothers, to liberate the Argive people from their tyrant, and to fulfil the original oracle. It is possible, though not necessary, to suppose that he came to know, and the audience were reminded, that he was fulfilling an oracle. This could have been done through a confession by Danaos before his death, or through a revelation by a divinity afterwards. In the latter case, since we do not want to multiply divine interventions praeter necessitatem, the divinity in question was no doubt Aphrodite. The beginning and end of the third play are starting to take shape. We start with the entry of a disguised Lynkeus. The audience will not have known to what point the action was going to leap after the end of The Suppliants, and may initially have been mystified by the appearance of this young man in, perhaps, slave or peasant costume. He explains that he has been spared by his wife when all his brothers were killed, that she has provided him with a disguise, and that he intends to lie low in or near the city and await an opportunity to take revenge on Danaos who, as his wife has told him, was the planner of the murders. His exit is followed by the entry of another character—the speaker of fr. 43. His identity is uncertain: he might be Danaos (in which case his language would be ironical),72 or he might be some innocent third party, say a palace servant responsible for the wedding celebrations. He possibly describes the festivities of the previous day, and then goes off to organize the awakening ceremony. But the ceremony never takes place: the only singing this morning is done by the Danaids themselves, who will now have entered. 72
Cf. Seaford (1984b: 227).
110
Aeschylus’ Danaid trilogy
Instead of the suppliant-branches of the previous play (significantly termed Kªå ØæØÆ in Supp. 21),73 they now bear (in their right hands instead of their left)74 the bloodied daggers with which they have slain their bridegrooms. That may have been the beginning. What of the end? Lynkeus’ revenge on Danaos, as we have seen, was a political as well as a personal act, and must have been accompanied or followed by his election as king [128] of Argos (a democratic king, as Pelasgos had been75 and Danaos had not). But what now of the Danaids, who as chorus are still present and as yet unpunished? Here, I suggest, is the occasion for Aphrodite’s intervention. She can explain what Lynkeus does not know. The Danaids are innocent. They acted on the orders of their father, and they did it because they were brainwashed by him (whatever the Aeschylean Greek is for ‘brainwashed’) into a pathological loathing of sex and marriage, a loathing that went against a fundamental principle of the universe (which fr. 44 describes). Danaos did this in order to avoid the fulfilment of the oracle about himself, whose terms are now revealed to Lynkeus and also, not less significantly, to the Danaids who thus learn for the first time how their father had deceived and exploited them. Now the oracle has been fulfilled. Danaos is dead. His daughters have no natural guardian and no close kin, except for Lynkeus. It is therefore his duty to purify them from their blood-pollution and to find them new husbands. We cannot tell how this was done, if indeed the method was specified at all. The foot-race of Pindar’s ninth Pythian ode76 is inappropriate to our trilogy, as Johansen and Whittle point out,77 since it involves the Danaids being treated as chattels, which is precisely how they were treated by the hybristic Aigyptiads: marriages arranged by Lynkeus under the auspices of Aphrodite ought to be handled with more regard to Desire, Persuasion, and Harmony, those companions of Aphrodite hymned by the secondary chorus in The Suppliants (1038–42). 73 74 75 76
Cf. Winnington-Ingram (1961: 148 ¼ 1983: 66). Cf. Supp. 192–3 ƒŒ ÅæÆ . . . åıÆØ Øa å æH Pø
ø . Supp. 365–9, 389–401, 517–18, 600–24, 739, 942–5; cf. Podlecki (1993: 72–6). Pind. Pyth. 9.112–16; favoured e.g. by Kitto (1961: 19–20) and Garvie (1969:
226). 77
Friis Johansen and Whittle (1980: 52).
Aeschylus’ Danaid trilogy
111
We are left with the middle of the play, the meat of the plot. How did Lynkeus secure his revenge? How much of the action took place on stage (in the presence of the Danaids) and how much was reported from elsewhere? Did an Argive popular assembly play a role, as Winnington-Ingram suggested?78 I do not think we can dogmatize, except that the action will certainly have been broken up by act-dividing choral performances79 of which there are unlikely to have been less than four (including the parodos).80 [129] I offer the following as a possible reconstruction: some portions are inevitably speculative. I have adopted Winnington-Ingram’s offstage Argive assembly—in which case it becomes necessary to contrive a means for Lynkeus to get to it undiscovered. In the Prologue the disguised Lynkeus describes his escape, and departs to a place of safety.81 A servant(?) describes the wedding festivities, and departs to arrange for the waking-song. There follows the Parodos: the Danaids enter with daggers and proclaim their victory over the hated males and the hated Aphrodite (cf. Supp. 1030–3). Possibly this song include reminiscences of the events whereby Danaos was forced to come to terms with the Aigyptiads, and of the hatching of the murder-plot. The First Episode begins with the entry of Danaos, with guards escorting Hypermestra. He announces his discovery of her ‘treachery’, consigns her to prison, and departs to hold a meeting of the Argive assembly at which he will demand, under dire threats, information on the whereabouts of Lynkeus. Hypermestra may or may Winnington-Ingram (1961: 148–50 ¼ 1986: 67–9). That is, strophic choral songs (or alternatively, especially in the last third of a play, choral utterances in anapaests) each preceded by an actor’s exit and followed by an actor’s entrance: cf. Taplin (1977: 49–60). 80 Of Aeschylus’ surviving plays (other than the doubtfully authentic Prometheus Bound), only Eumenides (two) and The Persians (three) have fewer than four such performances (inclusive of the parodos whether or not it is preceded by a prologue). In Eumenides the normal pattern is disrupted by the departure and re-entry of the chorus (231/244) and by their uniquely active role in the plot (see Taplin 1977: 384–7, 408–9) which has the effect that many of their utterances serve less to articulate the action than to carry it forward; in The Persians there is a fourth strophic choral song (623–80) which is followed by an entrance and, though not preceded by an exit, is preceded by a clear disengagement of actor from chorus as the Queen goes to Darius’ tomb to offer åÆ (the move of Danaos to the ‘look-out post’ at Supp. 624–5, cf. 713, serves a similar function). 81 Perhaps to Lyrkeia (about 60 stadia north-west of Argos), cf. Paus. 2.25.4–5. 78 79
112
Aeschylus’ Danaid trilogy
not speak in this scene; it may well have included an epirrhematic amoibaion between the Danaids and their father. After his departure, in the First Stasimon, the chorus express their hatred of Hypermestra and their hope that she will be put to death, together with her husband. The Second Episode may bring the chorus into contact with that husband himself. Enter (I suggest) Lynkeus, in the guise of (say) a shepherd who claims to know where Lynkeus is hiding. The chorus encourage him to go to the assembly and inform Danaos. In their Second Stasimon they will express their delight at the impending capture of Lynkeus, and their confidence that Danaos on his return will tell them the Argives have dealt with the man as he deserves. But instead of Danaos a messenger arrives (Third Episode) to report what happened in the assembly. Danaos had told the Argives that a plot against his life by the sons of Aigyptos had been foiled, and they were all dead except for Lynkeus, who must be caught. An unknown young man had said he knew where Lynkeus could be found and how to save Argos from its peril, and asked for authority to act and armed men to assist him. On being granted these with the consent of Danaos, he proclaimed ‘I am Lynkeus, and this tyrant, the murderer of my brothers, is the peril to Argos.’ Danaos was seized by the armed men and the people, and taken off to execution; Lynkeus was acclaimed as the new king (as his father was in the first play, and his father-in-law in the interval between the second and third). He will shortly be coming to exact similar justice on those who had actually carried out the murders. Alarmed and agitated, the Danaids in the Third Stasimon lament for their father and for themselves, and bitterly accuse Zeus of betraying them. [130] At the beginning of the Exodos Lynkeus enters as king of Argos, accompanied by Hypermestra as his queen.82 Lynkeus is about to order the arrest of the Danaids (an event heavy with reminiscences of the previous play) when Aphrodite appears. She explains, as outlined above, the innocence of the Danaids and the true guilt and motive of their father; she prophesies the glory of Lynkeus and his house, and the great heroes Perseus and Herakles (both sons of Zeus, 82
If only two actors were available (cf. n. 53 above), Hypermestra will have remained silent in this scene, whether or not she spoke on her previous appearance.
Aeschylus’ Danaid trilogy
113
like Epaphos long ago) who will spring from it.83 Her eloquence reconciles the Danaids to marriage, and persuades Lynkeus to pardon and purify them, subject to the agreement of the Argive people.84 The Danaids drop their daggers and declare their acceptance of their destined role; Lynkeus undertakes to give them in marriage. The play ends with a procession reminiscent of that at the end of The Suppliants, but this time led by Aphrodite. Here one further speculation may be in order. At the end of The Suppliants the Danaids were escorted into Argos by a party of Argive soldiers, some of whom had been detached by Pelasgos from his suite (the çºØ O of Supp. 954)85 while others had formed the bodyguard of Danaos (985ff.); and if recent critics are right,86 these men form the secondary chorus who warn the Danaids against disrespect for Aphrodite (1034ff.), speak of Desire, Persuasion, and Harmony as her divine companions (1038–42), and suggest that marriage is a woman’s destined lot (1047–51). What could be more appropriate than that these same men, already the protectors of the Danaids and their champions in battle, reappearing now as escorts to the new king and queen, should be accepted by the Danaids as their new husbands, and that the final procession should thus become a wedding procession marking the reintegration of the Danaids both as ‘normal’ women and as members of the Argive polis from which their ancestress was once driven out?87 83 Abas, the son of Hypermestra and Lynkeus, was the father (Hes. fr. 129, 135 M–W ¼ 77, 241 Most) of Akrisios, the father of Perseus’ mother Danae; in turn Perseus’ son Elektryon was the father of Alkmene, the mother of Herakles. The genealogy is presupposed by Prometheus Bound 773–4, 865–73. 84 In [Apollod.] Bibl. 2.1.5 the Danaids are purified by Athena and Hermes; but these deities seem to have little part to play in Aeschylus’ trilogy (cf. Friis Johansen and Whittle 1980: 47–8), and it is simpler to assume, in accordance with what Parker (1983: 375–92) has shown to be the normal mythical pattern, that the purification is carried out by the local ruler. 85 After an attempt had been made to carry off the Danaids by force, and after a virtual declaration of war by the Egyptians (950–1), Pelasgos cannot have left the Danaids on their own pending the return of Danaos from Argos. 86 See pp. 100–1 above. Seaford (1984b) lays particular emphasis on the quasinuptial aspects of Supp. 1018–73. 87 Hyginus (Fab. 170.11) does indeed say that the Danaids were married to Argives after Danaos’ death (though he puts this at a time when Hypermestra’s son Abas was already adult).
114
Aeschylus’ Danaid trilogy
[131] If anything like this reconstruction is correct, certain conclusions follow about the theme or themes of the trilogy as a whole. Its central character will not have been, as has generally been assumed, the collective persona of the Danaids, but the individual one of the father who manipulated them like the pieces on a gameboard (Supp. 12). It will not have been primarily about sexual violence, or about the inscrutable will of Zeus, though these things had their place in its structure. The dominant themes will have been three, all of which converge in the climactic event of the death of Danaos. The first, and probably the least important, is that familiar theme of Greek myth, the folly of attempting to evade the fulfilment of an oracle. Danaos, having fifty daughters, and being told that his son-inlaw would cause his death, resolved never to have a son-in-law (which was a crime against his daughters), and when forced to break that resolve, had the sons-in-law murdered (which was a greater crime): the result was that he died in disgrace and pollution, whereas if he had done nothing the oracle might well have been fulfilled in some quite innocent way. I say this was probably the least important theme, because the basic function of the oracle is to set the action going and to motivate the actions of Danaos: Hypermestra’s defiance is not the work of Zeus, who gave the oracle, but of Eros or Aphrodite,88 and Lynkeus’ revenge has massive motivation on the human level. The second theme is a political one. Danaos and Aigyptos are first shown contending for the rulership of a barbarian land; then Danaos comes to Greece, to democratic Argos, and creates a barbarian type of monarchy there until he is overthrown and democracy restored. The tyrannos disposes arbitrarily of the lives of others for his personal gain; but the democracy too can be short-sighted and foolish, for it first allows Pelasgos to hoodwink it into a dangerous war in a cause by no means as just as it seems,89 and then gives Danaos the means by
88
The love-motive implied by Prometheus Bound 865 (taking Æø with Æ , cf. Winnington-Ingram 1961: 147 ¼ 1983: 65–6), Friis Johansen and Whittle 1980: 46) fits in precisely with the central theme of the Danaid trilogy as identified below. 89 On this, and its possible relevance to contemporary Athenian politics, see .
Aeschylus’ Danaid trilogy
115
which he is able to seize absolute power. A sovereign people must be on the alert for such manoeuvres: in the words of John Philpot Curran, ‘the condition, upon which God hath given liberty to man, is eternal vigilance’.90 The third and predominating theme is that of the nature of marriage. This theme starts quite gently. Aigyptos, for good family reasons, wants a certain marriage for his sons; Danaos, for understandable personal reasons, wants no marriage at all for his daughters. In the next generation, however, the clash becomes radical, polar, ideological: for the Danaids, root-and-branch rejection of marriage—and therefore (though [132] they never understand this) of the organized polis society which at Argos wins such praise and blessing from them;91 for the Aigyptiads, marriage as seizure and possession by force,92 what can truly, and without any hyperbole, be called legalized rape.* Between these opposites there can only be conflict, and it is all too predictable that it issues in murder. A third way is first pointed to by the secondary chorus at the end of The Suppliants, who sing of Aphrodite’s companions Desire, Persuasion, Harmony, and Eros (1038–42); it is confirmed by Aphrodite herself, who speaks among other things of the mutual eros of the primeval couple Heaven and Earth, and embodied, surely, in the marriage of Lynkeus and Hypermestra, and perhaps, as I have suggested, in the marriages of the other Danaids to the same Argive warriors who had formed that secondary chorus. That is what marriage should be: based on mutual desire and affection, the foundation of the order of the universe, the order of the polis and
90 Quoted by Hale (1958: 78) (and as the epigraph on his title-page) from Curran’s speech on the right of election of the Lord Mayor of Dublin (10 July 1790). 91 Supp. 625–709, 1018–29. 92 This attitude is implicit in almost every word spoken or sung by the representatives of the Aigyptiads in Supp., and becomes explicit in 918 (‘I have found, and am taking away, my own lost property’), 920 (where the Herald invokes Hermes as ÆæØ, the god, one might say, of search-and-seizure) and 924 + 932 (KÆØæ ÆØ and IçÆØæ Ł , alluding to the legal procedure in situations where A seized the person B on the claim that B was his slave and C opposed the claim by ‘taking [B] away to freedom’, cf. A. R. W. Harrison (1968: 178–9, 221), MacDowell (1978: 80), Todd (1993: 187), and implying again that the Danaids are being claimed as their cousins’ property).
116
Aeschylus’ Danaid trilogy
the order of the oikos. That is the affirmation with which the Danaid trilogy ended.93
ADDENDA The arguments of this were presented again, more briefly, in Sommerstein (1996a:141–51). An important challenge to them has come from Garvie (2005) pp. xviii–xix, who quite rightly argues that I, and Sicherl and Ro¨sler before me, have misunderstood the scholium discussed on pp. 92–5; he shows that, given the facts of Greek usage in all relevant periods, Øa e c ŁÆ ÆøŁB ÆØ e ÆæÆ can only mean ‘because their father has not been killed’. The commentator is thinking in terms of Attic law: if the Danaids’ father were dead, they would be obliged (having no brother) to marry the nearest kinsmen who claimed them; since he is alive, he alone has the right to give them in marriage. This does not, however, refute the other evidence that Ro¨sler and I have cited of passages in Supp. which will be puzzling unless they can be understood in the light of a preceding play; in one of these (991–1013, in particular 1006–9; see p. 93 n. 17 and p. 97) Danaus appears to be saying that he and his daughters had fled from Egypt in order to preserve their virginity not only from their cousins but absolutely—which to any Greek would make no sense except in the light of the oracle story. See now Sommerstein (2008: i. 277–86, esp. 281–6). p. 90 n. 4 the opening sentence of this note should have been phrased more generally, e.g. ‘When a tragedy is named after a group of (human) brothers or sisters . . .’. Euripides’ Heracleidae certainly was not named after its chorus (which consists, not of the 93
This chapter was originally published in B. Zimmermann (ed)., Griechischro¨mische Komo¨die und Trago¨die (Stuttgart, 1995: 111–34) (the last two pages contained only bibliography). # 1995 J. B. Metzlersche Verlagsbuchhandlung und C. E. Poeschel Verlag GmbH in Stuttgart, Germany. Versions of it had been presented to seminars at the Universities of Du¨sseldorf and Trier. I am most grateful to Prof. Zimmermann and Prof. M. Lossau for inviting me to make these presentations, and to J. B. Metzlersche Verlagsbuchhandlung for giving permission for this republication.
Aeschylus’ Danaid trilogy
117
children of Heracles, but of old men of Marathon), and it is unlikely that Peliades or Temenidae was either. The daughters of Pelias must have been major players in the tragic action of the drama named after them, the deceived but willing agents through whom Medea contrived the death of their father, and the crucial event in that action, in which they were personally involved—the actual killing of Pelias and the boiling of his dismembered body—must have taken place offstage, whereas choruses normally remain onstage throughout. As to the sons of Temenus, they normally number only three or four (D.S. 7.13.1; Nicolaus of Damascus FGrH 90 F 30; [Apollod.] Bibl. 2.8.5; Paus. 2.28.3), and usually they either quarrel among themselves or plot a murder together—again, actions appropriate to tragic characters rather than tragic choruses. p. 92 Scullion (2002: 87–101) has argued that the evidence of the Hypothesis, and the internal evidence, are compatible with a date for Suppliants in the 470s. In fact the earliest date that fits our recorded information about Sophocles’ career is 470 (the date of Sophocles’ debut according to Eusebius’ chronicle [Soph. test. 32a/b Radt]— whereas Plut. Cim. 8.8 [Soph. test. 36 Radt] claims that his first victory, in 468, was achieved at the first attempt); but Scullion is right to remind us that the dating of Suppliants between Seven and the Oresteia, now enshrined in the sequence in which plays are presented in almost every edition and translation, should not be regarded as a certainty. p. 101 n. 46 in the Loeb (Sommerstein 2008: i. 416–17 and n. 200) I suggest (in essence following Taplin) that Supp. 977–9 were inserted ‘in place of a short choral song’ and, more speculatively, that the lines originally stood in Danaids. p. 107 I have my doubts now about my statement that Danaus ‘will have been dressed in a manner befitting a king of Egypt’. After all, he asks his daughters to present themselves as ‘needy foreign refugees’ (Supp. 202); he will not wish himself to give an impression of great power and wealth. p. 115 ‘legalized rape’: see Sommerstein (2006); also Zeitlin (1986: 137–44).
7 The theatre audience, the Demos, and the Suppliants of Aeschylus The tragedies and comedies composed for production at the Athenian dramatic festivals in the fifth and early fourth centuries bc are priceless historical documents. For much of the period they are, apart from inscriptions, the only contemporary documents we possess emanating from Athens itself. Their historical interpretation is problematic, of course, in many ways, some of which are discussed elsewhere in . The particular issue that I am going to take up may be approached by considering the question which, it has been said, historians should always put to themselves in respect of every document they use: who wrote it, for whom, and why? In the case of Greek tragedy and comedy, we usually know the answer to the first question, and we also know the answer to the third: the plays were composed with a view to being successful in a competition, before a small panel of judges whose identity was not known at the time of composition but was known at the time of performance, and who thus, even if their actual voting was secret, in practice (as many remarks in comedy make clear)1 were very liable to be influenced by the attitude of the mass of the audience. And as to The first half of this chapter is the same, with minor variations and revisions, as the first half of my paper ‘The theatre audience and the Demos’ (Sommerstein <1998>). I am grateful both to and to Juan Antonio Lo´pez Fe´rez for allowing and indeed encouraging me to publish both versions of the paper; and also to all those who commented on the two versions when they were presented, respectively, at Oxford in October 1992 and at Madrid in March 1993. 1 Cf. Ar. Knights 546–50; Birds 445–6; Eccl. 1141–2; and the criticism of Pl. Laws 659a.
The audience, the Demos, and Suppliants
119
the remaining question—‘for whom’—we know the answer to that too. Essentially the plays were written to be seen, to be heard, to be judged, to be appreciated by those who sat in the Theatre of Dionysos when they were first staged. There might be other audiences later. Plays seem often to have been reperformed at deme theatres in various parts of Attica. They had begun to be performed abroad, too, as early as Aeschylus’ time, and convincing arguments have recently been advanced2 that in the second half of the fifth century there were frequent productions of Athenian tragedy at least in southern Italy.* It was no longer necessary for the Athenian masters to go and produce their plays in person, as Aeschylus had done in Sicily; at any rate there [64] is no record of Sophocles doing so, nor Euripides until the last year or two of his life, and presumably there were by now people in various centres who possessed the necessary skills to produce a tragedy from its script—as was being done at Athens in the same period, in posthumous restagings of plays of Aeschylus.3 The copying of play-scripts may have begun because of the needs of such producers (and also to some extent of other poets), but by the 420s an embryo reading public probably existed for drama, and by 405 Aristophanes can take the existence of such a public for granted.4 But except in rare cases (such as Aeschylus’ Women of Aitna, first produced at Aitna itself)5 all these other publics were derivative from and in a sense parasitic on the original Athenian audience. A play by an Athenian author that did not get performed in the Theatre of Dionysos normally did not get performed anywhere else, nor did its script get copied.6 To understand fifth-century Athenian plays we need to understand the fifth-century Athenian theatre audience—and no other. 2
Taplin (1993). Aesch. test. 71–7 and 1.51–2 Radt. 4 Cf. Ar. Frogs 53, 943, 1114; Dover (1993: 34–5). 5 Aesch. test. 1.33–4 Radt. 6 Hence the great contrast in productivity between dramatists of the fifth century and those who lived later, Aristophanes for example writing an average of one play a year throughout his career (40 between 427 and c.386 bc), Menander, with plenty of opportunities for getting plays produced abroad, composing something between three and four plays a year (105 or more between 321 and c.292: Proleg. de Com. III 60 Koster, Apollod. ap. Aul. Gell. 17.4.5). ‘Just as Athens in [Menander’s] age imported dramatists, she also exported plays’ (Gomme and Sandbach 1973: 1–2). 3
120
The audience, the Demos, and Suppliants
The subject of this chapter, then, is: how was this audience composed, and did its composition change during the period to which our surviving tragedies belong? This question has not been discussed as much as it might have been in recent scholarship, where there has been a strong tendency, more often implicit than explicit, to equate the theatre audience with the citizen body of Athens.7 Such a tendency (from which I do not pretend that I have always been immune) is very understandable. It has often and rightly been emphasized that the dramatic festivals were civic as well as religious and artistic occasions. Comedy explicitly, and tragedy not infrequently by indirect means, can treat its audience as though it were identical with the body of adult male citizens which, perhaps within a few days, would be meeting again on the Pnyx to debate and decide major political issues. Obviously this was not literally true. Whichever particular dramatic performance we consider, we can be sure that many who were adult male citizens did not attend the performance, and that many who were not adult male citizens did. It may well be that the latter can be largely discounted. On the sorts of political and social issues that usually concern the historian, the children who certainly constituted a substantial fraction of the audience may safely be ignored; not that their reaction to a play would necessarily be without effect on the judges,8 but it would be likely to be influenced by [65] different kinds of factors, among which, in the case of comedy, costume, music, and slapstick might well be prominent. The question of whether there were women among the theatre spectators, and if so whether a significant proportion of them were of citizen status, remains controversial, but for our present purposes I do not think it matters very much. Jeffrey Henderson has argued persuasively for a revision of the accepted view on this question;9 but he does not dispute that comedy, from the fifth century to the third, normally treats the audience as if it were all male, and appeals for the appreciation of males of all ages but not of females. We have no comparable evidence for tragedy, because the conventions of tragedy forbid it to take explicit notice of the audience’s presence at all, but 7 Exemplified by many of the contributors to Winkler and Zeitlin (1990), including notably Goldhill (1990); Ober and Strauss (1990); and Henderson (1990). 8 Cf. Ar. Clouds 538–9; Men. Dysk. 967, Sam. 733. 9 Henderson (1991b).
The audience, the Demos, and Suppliants
121
there is no reason to doubt that the attitude of tragic dramatists was the same. In any case, I think there is good reason to be sceptical as to the likelihood of there having been any large number of citizen women present at the dramatic performances. In comedy it is more than once taken for granted as normal that a husband goes to the theatre and his wife does not: the adulterer in Ar. Birds 793–6, as soon as he sees his mistress’s husband in the councillors’ seats, knows that she will be alone at home—he does not need to cast his eyes up or round to the women’s section to make sure she is not there;10 and when it is comically suggested that the mothers of outstandingly valiant men should be given public honours, the proposal made is that just as a man might be awarded privileged seating (æ æÆ) at such festivals as the Dionysia, so these women should be awarded æ æÆ . . . at the Stenia and Skira, women-only festivals. The woman speaker’s statement in Ar. Thesm. 386 that she has ‘seen’ how women are vilified by Euripides is not such decisive evidence as Henderson claims,11 since the verb ‘see’ in such contexts (especially in the exordia of Assembly speeches, on which the passage in question is modelled) is an oratorical cliche´ often having little to do with literal vision.12 At any rate, whether or not their audience actually was allmale or even nearly so, the dramatists do seem to regard it as such. They waver rather more over whether to regard it as exclusively Athenian: occasionally we can definitely detect them taking an interest in non-Athenian spectators, as Euripides may be doing when he emphasizes the common ancestry of Athenians and Ionians in Ion.13 But on the whole it is true that the dramatists, tragic and comic alike, wrote as male Athenian citizens for male Athenian citizens. But which male Athenian citizens? Here we come up against a significant fact of which surprisingly little notice has been taken in recent discussion of [66] ‘drama and society’.14 In one important 10
Cf. also Ar. Thesm. 395–7 where a woman complains that husbands coming home from the theatre ‘eye us suspiciously and straight away make a search, in case we’ve got a lover hidden in the house’. 11 Henderson (1991b: 142–3). 12 See e.g. Ar. Eccl. 176; Thuc. 6.20.1; Dem. 5.1, 9.1, 14.3, 16.2, Prooem. 5.1, 6.1, 44.1. 13 Cf. also Sommerstein (1977c: 117–18) on Aristophanes’ City Dionysia plays; and Cassio (1985: 105–18 on Peace. 14 Wilson <(1997)> is a heartening exception.
122
The audience, the Demos, and Suppliants
respect the dramatic festivals arguably were not civic occasions of a normal kind. Every spectator, citizen or alien, had paid to attend.15 We are so used to the idea of paying to attend a performance that we do not immediately perceive how abnormal this was in the context of Athenian democracy and indeed of Greek society generally. Nowhere was it the case that a citizen had to make payment for exercising a civic function. Of course in many places he might have to have a property qualification in order to do so, and in most, perhaps all, he would sooner or later be debarred from civic activities if he did not pay his taxes and other debts to the state; but rules such as these are quite another thing from being required to pay a fee each time one entered an assembly or sat on a jury. At Athens in particular, moreover, at least from the time of Ephialtes or soon after, the expectation was the other way round: the exercise of a large and increasing number of civic functions—those of the courts, the council, many of the magistracies, and later on the ekklesia too—involved a payment, not by the citizen, but to him. Nor again was it at all normal to require a money payment from citizens before they could participate in one of the religious festivals of the community.16 Eventually, as is well known, the theatre did succumb to the logic of Athenian civic ideology, with the institution of the theorikon. This did not however occur until the middle of the fourth century;17 and when it did occur, the change took the form, not of abolishing the admission charge, but of providing from public funds the money with which it could be paid—money which the recipients could not be prevented from spending on other things if they chose to do so.18 Even when it was thus subsidized, attending the theatre remained an option, not a duty. In the fifth century it had been an option with a 15
For the ancient evidence see Pickard-Cambridge (1988: 265–8).* It was not for nothing that these festival were officially described not just as ‘of the demos’ (Å ØÆ) but as ‘paid for by the demos’ (Å ºB, cf. Thuc. 2.15.2; Dem. 21.53; Deinarchos fr. 18.3 Conomis). 17 Ruschenbusch (1979).* 18 These are two reasons why I find it hard to accept the suggestion of Wilson <(1997: 97–100)> that the admission charge and the theorikon were from the start closely connected and designed to signalize the distinction between members of the citizens’ club (who did not have to pay the charge) and others (who did). See, however, p. 128 below for a suggestion that an increase in the admission charge in the mid-fifth century may have been coupled with a compensating increase in state misthoi paid to citizens. 16
The audience, the Demos, and Suppliants
123
far from negligible cost. Two obols was a third of an oarsman’s or building worker’s daily pay,19 or two-thirds of a juror’s; alternatively, it was the price of four pieces of meat each large enough to fill the maw of a Herakles.20 And two obols was the price per head; and while the head of a household would normally not [67] be taking his wife along,21 the evidence of comedy shows clearly that his sons would expect to be taken once they were old enough to enjoy the performances, and there might often also be an elderly father (retired, like Aristophanes’ Philokleon, from the active management of the household) who would not much appreciate being left behind. The charge seems likely, too, to have been payable for each day of attendance; we hear nothing of the issue of tickets or tokens on the first day which could be used to gain free admission on subsequent days, and we know22 that when the theorikon was first introduced its amount was not two obols but a drachma, perhaps (as PickardCambridge suggested)23 to cover the three days of the tragic competition. Thus in the fifth century, to see the whole of the tragic and comic contests, lasting three or four days, might involve admission costs, for a family of four males, of up to thirty-two obols or more than five drachmas. Chicken-feed, of course, for most of those whom we meet in the pages of the orators, but a substantial sum for many a poor citizen, and perhaps also for some quite prosperous peasants if they did not participate, or participated only marginally, in the money economy.24 Metics, on the other hand, being mostly craftsmen or traders, and living overwhelmingly in the city and the Peiraeus, will in general have been much more cash-oriented, and it 19
Oarsman: Thuc. 6.31.3; building worker: e.g. IG i3. 475.253–6, 272–85. Ar. Frogs 553–4. Dover (1993: 264) thinks (comparing Eupolis fr. 156.3) that ‘half-obol’ portions of meat were rather small; but the Eupolis fragment does not necessarily have that implication, and Dover’s claim that ‘it is entirely in character that the Innkeeper should think of such a [small] portion as fair and proper’ finds no positive corroboration in anything else she or her companion says. If these innkeepers are being presented as stingy and grudging caterers, how comes it that they serve meat (normally a special-occasion food) at all? 21 Above, p. 121. 22 Philochoros, FGrH 328 F 33. 23 Pickard-Cambridge (1988: 267). 24 The contrast between urban cash-economy and rural barter-economy is interestingly discussed, with particular reference to Aristophanes’ Acharnians, by Olson (1991). The hero of Acharnians claims (35) that in his home village the word ‘buy’ was unknown. 20
124
The audience, the Demos, and Suppliants
is quite possible that the proportion of metics in the theatre audience was higher than in the free population of Attica as a whole. Is it an accident that already in the second quarter of the fifth century, two surviving plays of Aeschylus—Suppliants and Eumenides—are about a community which welcomes the arrival of a group of ‘metics’, explicitly so described?25 Thus even if we confine our attention henceforth to that part of the theatre audience which consisted of adult male citizens—and once we subtract children, women if any, visitors, and metics, we may well be left with less than half of the whole audience—we still cannot regard it as a representative sample of the citizen body. It consisted of those who had paid; in other words, of those who wanted and could afford to come. We cannot take it for granted, as Dover once did,26 that ‘the audience which had acclaimed Knights proceeded to elect Kleon general’. Neither the audience nor Kleon’s electorate was the whole citizen body. They were both parts of it, and overlapping, not identical parts. Nevertheless, they could both be addressed as if they were the whole, in the same way that a speaker addressing a jury—a much smaller portion of the community, and likewise unrepresentative in that it excluded men under 30—could use the second person plural indifferently to denote the instant jury, a [68] differently constituted jury in another case, the ekklesia, an Athenian military or naval force, and so on.27 Can we say anything with any approach to confidence about the social composition of that portion of the citizen body who could be expected to attend the theatre? One or two things, perhaps. On average they will certainly have been more affluent economically than the citizen population as a whole. On average, too, they will have been better educated; Aristophanes’ compliments in Frogs and elsewhere to the intellectual and literary sophistication of his audiences28 need not 25
Aesch. Supp. 609, 994; Eum. 1011, 1018. Dover (1968: p. lvi). 27 Cf. e.g. Aischines 1.173 (where ‘you’ denotes both the present jury and that which had tried Socrates fifty-four years earlier), 3.86 (an expeditionary force to Euboia about twenty years before the date of the speech), 3.125 (a meeting of the ekklesia). 28 Ar. Knights 233, 505–6; Clouds 521, 527, 535, 575; Wasps 1013–14; Frogs 676, 700, 1109–18. 26
The audience, the Demos, and Suppliants
125
be taken at face value, but he did devote half of that play to a contest in technical skill and ethical merit between two tragic poets—and what is more, it was the second half, the part that would be fresher in the judges’ memories—and the play did win him first prize. The political biases of Old Comedy* are fully in line with this view of the balance of its audience; as de Ste. Croix and Cartledge have shown in the case of Aristophanes,29 the dramatists regularly adopt a position of hostility to those politicians whom they present as relying strongly on the support of the poorer citizens—in the 430s to Pericles, in the 420s to Kleon, after his death to Hyperbolos, and so on—and of some degree of sympathy for those who were perceived as their enemies, such as Thucydides son of Melesias,30 Archeptolemos,31 or Nikias (in Eupolis’ Marikas).32 The ‘anti-war’ stance of Old Comedy, which is not confined to Aristophanes (consider Kratinos’ Dionysalexandros, in which, we are told, Pericles was ‘very effectively satirized by implication as having brought war upon the Athenians’),33 should be regarded as part of the same pattern; our evidence strongly suggests that the wars which Old Comedy regarded as ‘bad’ were those which subjected Attica to enemy invasion and devastation (and in which, therefore, those who owned land lost out by comparison with those who did not), while there is virtually no condemnation of external wars of conquest—at least not until after they have failed.34 All this points, certainly for the period of the Peloponnesian War, to an audience distinctly ‘right-wing’ by comparison with the population as a whole.35 [69] For earlier periods things are more difficult. 29
De Ste. Croix (1972: 355–76); Cartledge (1990). Ar. Ach. 703–12. Ar. Knights 327, 794–5. 32 Eupolis fr. 193 KA. 33 POxy 663 (¼ CGFP 70 ¼ Kratinos, Dionysalexandros test. I KA), ad fin. 34 In Lysistrata, produced in 411, the decision to launch the Sicilian expedition is regarded as an act of folly (Lys. 391–7 and perhaps 517–18, cf. 589–90); in Birds, produced three years earlier while the campaign was in progress, the only (implicit) criticism is of Nikias’ dilatory generalship (Birds 640). 35 I define a ‘right-winger’ as one who favours the active use of the power and institutions of the state to maintain or extend privilege and inequality among those under its jurisdiction, and a ‘left-winger’ as one who favours the active use of the power of the state to reduce or eliminate such privilege and inequality. Strictly, therefore, all Athenian politicians were right-wing, since they all supported legal discrimination against slaves, women, and aliens. But I will follow their own practice 30 31
126
The audience, the Demos, and Suppliants
The passing of a decree in 440/39 which restricted in some way the freedom of comic satire36 suggests that the gap between Assembly opinion and theatre opinion already existed then. Going back beyond 440, the evidence of comedy more or less dries up, since it now seems likely that almost all the plays of Kratinos of which substantial fragments survive belong to the 430s and 420s.37 A window of more or less reliable information in the 470s and 460s suggests a somewhat different picture. In that period the theatre seems to have been something of a political battleground. In 476 Themistokles was the victorious tragic choregos,38 in 472 Pericles39 (whose father, Xanthippos, had succeeded Themistokles as commander of the Athenian fleet in 479);40 probably on the first occasion,41 certainly on the second, the successful production was wholly or partly devoted to the Persian War, in a notable departure from the conventions of the genre, and both times with strong emphasis on the naval side of the conflict, so closely associated with Themistokles, who had created the navy and who had contrived the stratagem (recalled in Persians 353ff.) that had won the victory of Salamis.42 This does suggest that at that time
and confine the universe of discourse to adult male citizens, which is only what we always do when we speak of classical Athens as a democracy. 36 Ar. Ach. 67; the historicity of this decree is accepted by Halliwell (1984: 87; 1991: 57–9), despite his scepticism regarding the reliability of ancient notices of similar decrees said to have been enacted c.415 or later.* 37 There are twenty-five plays of Kratinos of which four or more fragments survive. Kassel and Austin (1983: 112–267) offer or report suggested dates for nineteen of these; only one of these dates (that for Drapetides, which may be associated with the foundation of Thourioi in 444/3) is earlier than 439. See, however, Mastromarco (1992: 368), who places three anti-Periclean plays of Kratinos in the late 440s and sees them as partly motivating the decree of 440. 38 Plut. Them. 5.5. 39 IG ii2. 2318.9–11. 40 Hdt. 8.131.3, 9.114–21. It was Xanthippos who, having captured Sestos, brought back to Athens the cables of Xerxes’ bridge over the Hellespont, of which so much is made in Persians (68–72, 109–13, 130–1, 722–6, 736, 744–51). 41 The victorious dramatist was Phrynichos, and it is generally supposed (though it cannot be proved) that his production on this occasion included his Persian War play(s)—the celebrated Phoinissai and possibly also ˜ŒÆØØ j —æÆØ j ŁøŒØ (cf. Lloyd-Jones 1990: 233–4 ¼ 1966: 22–3). 42 I discuss the political aspects of Persians more fully in Sommerstein (1996a: 410–13). For a sceptical view (and references to other discussions), see Pelling <(1997b)>.
The audience, the Demos, and Suppliants
127
the theatre audience was not felt to be prejudiced against ‘left-wing’ personalities. On the other hand there is the well-known anecdote in Plutarch’s life of Kimon43 about the competition in 468, when Aeschylus was defeated by Sophocles after the audience had allegedly demanded that Kimon and his fellow generals be appointed judges of the contest; this, however, is not only of dubious historical value,44 but even if true it [70] would not in itself show that the theatre audience had a ‘right-wing’ bias in 468, since Kimon was also at that time a dominant figure in Athenian politics generally. Nothing in the evidence we have is inconsistent with the hypothesis that in the 470s and 460s the theatre audience more or less faithfully mirrored the balance of political opinion in the population as a whole (or at least in that part of it which normally attended the Assembly), whatever that balance might at a given moment happen to be. Yet by 440—in a period when, despite occasional successes for their opponents in the lawcourts, the Periclean ‘left’ was in complete command of the Assembly—the theatre audience had become preponderantly ‘right-wing’ and anti-Periclean. It no longer contained an even cross-section of the civically active citizens: the poorer, more strongly democratic part of the population was less well represented in the theatre than it had been a generation before. Why should there have been this change? May it be that the theatre admission charge, which I have suggested acted in the later fifth century as a deterrent for some of these poorer classes, had been less of a deterrent in the 470s and 460s? Not that a charge of two obols is likely to have been significantly more affordable in the 470s than it was in the 430s. Rather the reverse, if anything: in the 470s Athens was still recovering from the Persian devastation, there was no public building programme and no state pay except on campaigns. It is more plausible to suppose that the admission charge itself was then lower. One can 43
Plut. Kimon 8.8–9. The suspicion inevitably attaching to anecdotes of this kind in the absence of early evidence in their support (on which see Lefkowitz 1981) is deepened by a chronological point raised by C. W. Mu¨ller (1984: 70ff.). It is known that Sophocles’ first victory was in 468 (Marm.Par. A 56), and Mu¨ller argues that Aeschylus is not likely to have been competing against him then, since Aeschylus produced the Theban tetralogy in 467 and tragic dramatists do not normally appear to have put on productions at the City Dionysia in successive years. 44
128
The audience, the Demos, and Suppliants
understand why it might have been raised at some stage. Between the 470s and the 430s there was a very substantial increase in the population of Attica, both citizens and, we may be sure, metics too.45 During the same period the City Dionysia in particular became more and more an international event, especially when the centre of the Delian League was transferred to Athens and the Dionysia became the occasion for the delivery of tribute. These developments must have put pressure on the capacity of the theatre. The first reaction, in the 460s, may have been to expand the capacity; the details are uncertain, but we know from the plays themselves that the theatre underwent some sort of physical remodelling in the late 460s,46 and this may well have involved an enlargement of the spectator space. The site, however, imposed physical limits to any further expansion, and meanwhile Athens went on becoming ever more populous, more prosperous, and more visited. In these circumstances there were only two things that could be done. One was to ration demand by regulation (for example by limiting the number of foreigners or of children who could attend); the other was to ration it by price. As any successful football club can witness, the former alternative would have been an administrative nightmare, [71] and it seems that the latter was chosen. Perhaps at the same time some of the misthoi paid (to citizens only) from public funds were increased in compensation, and this may be the explanation of the claim in some sources that the theorikon was instituted by Pericles; for the change we are talking about must have happened after the start of Pericles’ ascendancy, and probably after the move of the League treasury to Athens in 454/3. Clearly Pericles and his supporters cannot have intended to turn the theatre, with all its prestige, into a platform for their political opponents. Obviously they realized that fewer people would attend the theatre, but it did not occur to them that this would change the socio-political make-up of the audience—particularly if, as I have
45
The most useful attempt to establish population statistics for classical Athens is Hansen (1985). 46 The skene, the ekkyklema, and, in my view, the mechane all make their first appearance at this time; see respectively Taplin (1977: 452–9); Garvie (1986: pp. lii–liii); and Sommerstein (1989b: 153).
The audience, the Demos, and Suppliants
129
suggested, the increase was effectively subsidized for the poorer citizens. They may have forgotten two things: first, that the cost of attending had to be thought of in terms not of individuals but of families (or at least of all the males in a family); second, that people’s positive reasons for attending the theatre varied considerably, that a sharp rise in price would force them to re-examine those reasons, and that the slighter their education, and the tighter their financial circumstances, the more likely they were to decide that they had better things to spend their money on, subsidy or no subsidy. At any rate this was apparently the result. It will have taken a year or two for dramatists (mainly comic dramatists, no doubt) to notice the change in the balance of audience reactions and to adjust their political ‘angle’ accordingly. And by that time it was too late to do anything about it. Comic sniping became an unavoidable fact of life for ‘leftwing’ politicians, and it was protected both by the parrhesia typical of democratic Athens in general and by the special parrhesia which, as Halliwell has argued,47 was associated with festal occasions.* Occasionally a politician might try to bite back, as Pericles probably did in 44048 and as Kleon did when he tried to prosecute Aristophanes in the 420s,49 but on the whole they seem to have accepted the situation. In the end it proved, under normal circumstances, not to be as threatening as all that; to paraphrase Dover,50 the theatre audience acclaimed Knights, but that couldn’t prevent the Assembly from electing Kleon general. It could no more be anticipated that one day a restaging of Frogs would help procure the execution of a Kleophon51 than it could be anticipated that one day a lingering memory of Clouds would help procure the execution of a Socrates. From all this it follows that there are considerable, though not insuperable, difficulties in the way of using the texts of tragic dramas as an index of public opinion at the time when they were produced— quite apart from the difficulties [72] inherent in the nature of the 47
Halliwell (1991). See n. 36 above. Cf. Ar. Ach. 377–82 with scholia, 502–3 with scholia, 630–2; Wasps 1284–91; Ar. test. 1.19–21 KA. That at least one prosecution was actually attempted cannot be seriously doubted; see Mastromarco <(1998)>.* 50 See n. 26 above. 51 See Salviat (1989); Sommerstein (1993). 48 49
130
The audience, the Demos, and Suppliants
genre itself, whose conventions did not allow direct reference to contemporary persons and events. The public whose opinions mattered to the dramatists was not the same as the public whose opinions mattered to the politicians. A play which seems to us to have a distinct political ‘slant’, and which might have made the same impression on a representative sample of Athenians, may in reality simply have been reflecting, rather than seeking to influence, the views of the decidedly unrepresentative sample of Athenians who were actually likely to see it. In some cases, however, more positive conclusions may be possible. Taking first the later period when the composition of the audience was skewed ‘rightwards’, added significance automatically attaches to any play or plays that can be seen as having political implications which such an audience would not be likely to find congenial—bitter hostility to Sparta, say, or robust, principled defence of democracy, such as we find respectively in Euripides’ Andromache of about 42552 and in his Suppliants of 423 or 422.53 In such cases two main lines of explanation are open to us. One is that the dramatist held a particular view so strongly that he was determined to put it before his audience even at the risk of failure in the competition. The other is that the view in question had such overwhelming support among the population as a whole that it was in fact the majority view of the theatre audience as well. In the two cases I have mentioned the parallel evidence of comedy each time suggests that the second explanation is preferable. In Aristophanes’ Acharnians, which must be very close in date to Andromache, the hero, Dikaiopolis, while explicitly and repeatedly emphasizing that he is pleading ‘on behalf of the Spartans’,54 feels compelled nevertheless to insist that he hates them and desires their destruction.55 As a matter of fact hostility to Sparta seems to have been both deep and constant in Athens from the 440s, if not earlier, to the end of the century. The Spartan Menelaos, who is 52
Eur. Andr. 445–53, 594–600. Eur. Supp. 426–55; for the date of the play, see Collard (1975: 8–14), Cropp and Fick (1985: 23). 54 Ar. Ach. 309–14, 356, 369, 482, the last three times each marked by the striking anapaestic rhythm (in iambic trimeter verse) of the phrase bæ ¸ÆŒ ÆØ ø . 55 Ar. Ach. 509–12. 53
The audience, the Demos, and Suppliants
131
quite an attractive character in the Iliad and Odyssey, in surviving tragedy is invariably presented as a fool, a knave, or a coward.56 And even in Euripides’ Trojan Women, which has often been seen as an attack on the aggressive, expansionist spirit that was soon to launch the Sicilian expedition, and which was produced at a time when Athens and Sparta were at peace, the chorus go out of their way57 to express their special loathing of the thought that they may find themselves living as slaves at Sparta of all [73] places. As regards Theseus’ defence of democracy in Suppliants, it has often been noted that while Old Comedy, and Aristophanes in particular, frequently criticizes aspects of the Athenian democratic system which seemed dispensable but were really crucial to its effectiveness (such as the payment of jurors, or the activities of the ‘demagogue’ and the ‘sykophant’), there is no open criticism of the system itself:58 it may from time to time be insinuated that there is something absurd about poor citizens holding high office,59 it is never suggested that there is anything wrong with their having an equal vote in the Assembly. This sort of evidence about the degree of public support for particular attitudes and policies can be of real value. We know a fair bit (though, of course, not as much as we would like) about the decisions that were taken on various issues by ‘the Athenians’ in, say, the last third of the fifth century. We only rarely know by what margins the decisions were taken; in particular, the procedure of the Assembly made the recording of voting figures impossible, and it is exceptional for us to be told, as we are by Thucydides about the vote on the fate of Mytilene,60 that a decree was passed by a narrow margin. 56 The relevant plays are Sophocles’ Ajax and Euripides’ Andromache, Trojan Women, Helen, Orestes, and Iphigeneia at Aulis. 57 Eur. Tro. 210–13. In contrast with the tirades in Andromache, the tone of this passage cannot be attributed merely to ‘wartime’ feelings; when Trojan Women was being written and produced, Athens was not at war with Sparta but only with Melos (and that campaign had probably ended by the time the play was performed) and, in a desultory way, with Corinth (Thuc. 5.115.2). 58 A point made strongly by Dover (1972: 33–4). 59 Particularly, it seems, when poor men serve as ambassadors: cf. Ach. 614–17 (Lamachos and the aristocratic but allegedly insolvent Megakles) and Wasps 1267–74 (Amynias). 60 Thuc. 3.49.1.
132
The audience, the Demos, and Suppliants
When the theatre audience is a skewed sample, then, as it was in the 430s and later, a play which seems to be slanted against that bias may be of considerable significance to the ancient historian even while another play with an equal and opposite slant might tell us, politically speaking, little or nothing that we did not know already. In Aeschylus’ time, when the theatre audience may (as we saw earlier) have reflected fairly closely the current balance of political opinion in the population as a whole, there is likewise much to learn from plays whose own political standpoint is not, or not unambiguously, that which seems on our other evidence to have been currently most popular. To some extent we are hampered here by the notorious chronological uncertainties of this period. Was Aeschylus in Persians seeking in some measure to defend the record of Themistokles at a time when he was already under threat, or was he paying to a popular hero a tribute for which he could expect popular endorsement? We cannot be sure, because we do not know exactly when Themistokles was ostracized,61 nor for how long feeling had been building up against him. It is perhaps, however, significant that Aeschylus felt it necessary in 473/2 to return to a theme which had been handled by Phrynichos, with memorable success, probably only four years earlier, and which, if it was going to have an effect on people’s attitudes to current political leaders, would certainly help Themistokles, the architect of victory, as against Kimon whose personal role had been no greater than that of many an ordinary citizen. [74] There is of course a lot to say, and I have said a lot before now,62 about the political implications of the Oresteia.* Here I only wish to draw attention again to the remarkably strong backing given in Eumenides, in words put into the mouths of no less than four superhuman powers—Athena, Apollo, the Erinyes (alias Semnai 61 ‘The dates of Themistocles’ ostracism and condemnation [for treason] cannot be established’ (Rhodes 1970: 398); see more recently Podlecki (1975: 198) (ostracism in 472/1), Lenardon (1978: 106) (476–470), Frost (1980: 187–92) (condemnation 471/0, ostracism a year or two earlier), Hornblower (1991: 220) (‘some time in the late 470s–early 460s’). 62 Sommerstein (1989b: 25–32); cf. Bearzot (1992), who draws attention to the anti-democratic language which Aeschylus puts into the mouths of Klytaimestra (Ag. 883–5) and Aigisthos (Ag. 1617–18)—in contrast, we may note, with his Agamemnon, who, when his attention is drawn to possible political difficulties awaiting him at Argos, replies that they will be the subject of deliberation in public assemblies.
The audience, the Demos, and Suppliants
133
Theai), and Orestes in his capacity as a future cult-hero—to an adventurous, aggressive, pro-Argive and therefore anti-Spartan, foreign policy for Athens,63 even to the extent of wishing war upon the Athenians as a blessing (864) and naming Ares as one of their special patrons (918). Coming as all this does at the end of the trilogy, repeated as it is over and over again, it would have made so powerful an impression on the audience and the judges, and been so impossible for them to ignore, that one is almost bound to conclude, not only that Aeschylus himself was strongly in favour of this policy with all that it implied, but also that not just a majority of Athenians, but a very large majority, favoured it too, despite the heavy casualties it had already incurred.64 Otherwise we would have to suppose that Aeschylus was seriously jeopardizing his chances of success in the competition merely in order to rub home by repetition a message already delivered with perfect (and in terms of tragic drama, highly abnormal) clarity. I want to end, however, by considering another play of Aeschylus, The Suppliant Maidens. Here we are again in the grip of chronological uncertainty, since we do not know for sure the date of the play itself, but I think it may be possible to use two features of the play to help establish that date.* These features are both connected with the play’s most important character, Pelasgos, king of Argos: they are his insistence on the democratic character of the city he rules, and the means by which he persuades that city to grant asylum to the Danaids even at the risk of war with Egypt. George Forrest in a well-known article65 associated the portrayal of Argos in this play with the welcome Argos had given to the suppliant Themistokles some years before:66 the political message which he saw in the play was that Argos ‘had been prepared to fight the right wars and had had the right constitution’—and, by implication, that Athens ought to follow her example. How far I do and do not agree with this will become evident presently.
63 64 65 66
Aesch. Eum. 289–95, 397–402, 667–73, 762–77, 864–5, 913–15, 918–21, 1008–9. For these cf. IG i2. 929 <¼ i3. 1147> ¼ ML 33. Forrest (1960). Thuc. 1.135.3; Plut. Them. 23.1.
134
The audience, the Demos, and Suppliants
As is well known, the group of persecuted suppliants appealing to the ruler of a city is a frequent feature in tragedy.67 Usually the ruler (especially if Athenian) readily grants them asylum. Pelasgos has all the more reason to act [75] thus when he learns that, far from being aliens from another continent as their appearance suggests, they are themselves of Argive descent (Supp. 274–326). He is, however, reluctant to do so, and argues (365–9) that as the Danaids have taken refuge at an altar belonging to the city rather than to him personally, it is the city that must decide what is to be done. To the Danaids this is a distinction without a difference (370–5): is not Pelasgos the ruler of the city, and can he not therefore act on behalf of it, as Theseus (say) often enough does in tragedy on behalf of Athens? But he will not: ‘I will not make a definite promise before I have consulted all the citizens on this matter . . . I have already said that I will not do this without the people’s consent, even though I have the power (P æ ŒæÆH ), lest hereafter the people should say, if something untoward were to result, “You set store by a band of aliens, and ruined your city”’ (368–9, 398–401). Even when the Danaids attempt to blackmail him by threatening suicide (455–65) he agrees only to bring the matter before the people (517–19), which he had in effect already agreed to do before the threat was made. He is no coward desperately seeking a way to avoid war: that is shown by the way he is ready to face the even greater danger of the wrath of Zeus Hikesios (cf. 347, 359–64, 385–6, 413–16, 478–9) rather than commit his city without consulting its people. When the assembly is held, too, emphasis is put on aspects of its procedure which resemble that at Athens,68 particularly the vote by show of hands (604, 607, 621), the formal language of decrees (609–14), and even the typically Athenian punishment of ¼Ø Æ or loss of citizen rights (614 ¼Ø r ÆØ). The very word ‘democracy’ itself, as Victor Ehrenberg pointed out more than forty years ago,69 appears thinly disguised twice in the play (604 ı ŒæÆFÆ å æ, 699 e Ø e ºØ ŒæÆ Ø) for what may be the first time in any surviving Greek text.70 This is the proper 67
See Kopperschmidt (1971); Taplin (1977: 192–3).* Cf. Friis Johansen and Whittle (1980: 487–501). 69 Ehrenberg (1950: 522). 70 In remarkable contrast, B and its derivatives are entirely absent from the Athenian portion of Eumenides, where the Athenian people are called the º or æÆ. 68
The audience, the Demos, and Suppliants
135
way for a polis to take a major decision: by vote of the assembled people.* And yet Pelasgos is shown as obtaining that consent by blatant manipulation. Careful preparation is made for the assembly meeting: suppliant-branches are to be placed on the city altars for all to see (481ff.), but the men escorting Danaos into Argos are ordered not to divulge anything about him until the assembly meets (502–3). When it is held, Pelasgos persuades the people to grant asylum to the Danaids by warning them of the wrath of Zeus Hikesios if this is not done, without saying a word about the danger of war with Egypt if it is (615–20). Having the decision, as it were, in his pocket, he then defies the Egyptian herald and accepts the declaration of war in the name of his people (942ff.) without further consultation. And we know the result: this war, though fought apparently in the holy cause of protecting suppliants, ended unhappily both for the Danaids and for Argos: the Danaids were left with no alternative [76] but to accept marriage with their hated cousins, Pelasgos was almost certainly killed, and Danaos (who already has a bodyguard at 985ff.) probably became tyrannos of Argos.71 What we see is a democratic state, very like Athens itself, deceived with ‘tricks of oratory’ (623 ŠŪæı . . . æç) into voting for a war that was to prove disastrous—a war which, to judge by Pelasgos’ tactics, they never would have accepted if the issue had been put to them honestly. None of the ‘almost cheerful militarism’72 of Eumenides here! Let us be a little more precise. The political leader of a democratic state comes before its Assembly together with a lone foreigner, who has made himself a suppliant at the city’s altars and asked for the city’s help. The foreigner claims a personal connection with the city. He seeks the city’s assistance against enemies from his own country, in a cause whose justice is by no means clear-cut (note the question raised by Pelasgos in 387–91, which the Danaids do not answer; and note also the probability, raised in important recent articles by Sicherl and Ro¨sler, that Danaos has in reality quite different motives for seeking Argive protection against the Aigyptiads, never avowed in 71 Winnington-Ingram (1961: 142 ¼ 1983: 57); Garvie (1969: 198–9); Friis Johansen and Whittle (1980: 43–50). 72 Sommerstein (1989b: 253).
136
The audience, the Demos, and Suppliants
this play).73 The politician presents a carefully crafted case to the Assembly, in which some crucial facts are suppressed (among them the dubious justice of the suppliant’s cause), and secures the passage of a decree giving the city’s full support to the foreigner and those on whose behalf he had made his supplication. This involves the city in a dangerous and ultimately humiliating military campaign, and in the end the foreigner becomes its tyrannos. Now this synopsis, all but the last clause, is extraordinarily similar to a series of events that occurred at Athens, probably in 462.74 The political leader was Kimon. The lone foreign suppliant75 was Perikleidas the Spartan, who may well have been a proxenos of the Athenians at Sparta (as Kimon was of the Spartans [77] at Athens76) and had named his son Athenaios (as Kimon had named his Lakedaimonios).77 The threatening enemies from his own country are the ‘slaves’,78 or helots, or Messenians, at Ithome. The campaign
73
Namely, to keep his daughters unmarried because an oracle has warned him that he will be killed by his son-in-law (the crucial evidence for the relevance of this version of the legend is Supp. 37, Øa e c ŁÆ ÆøŁB ÆØ e ÆæÆ): see Sicherl (1986); and Ro¨sler (1993), who argues powerfully that for this and other reasons it is likely that Supp. was the second play of the Danaid trilogy and that its audience knew about the oracle because it had been mentioned in the preceding play, Aigyptioi.* 74 For the appeal cf. Plut. Kimon 16.9–10, Ar. Lys. 1138–44; for its consequences, Thuc. 1.102–5. Since the Areopagos council was stripped of its political power, on the initiative of Ephialtes, while Kimon’s expedition was in the Peloponnese (cf. Plut. Kimon 15.2–3) in the year 462/1 (Arist. Ath.Pol. 25.2), the expedition must have been sent in the campaigning season either of 462 or of 461; the former is more likely, given that the earthquake at Sparta, which triggered the helot revolt, occurred at about the same time as the Athenian disaster of 465/4 at Drabeskos (Thuc. 1.100.2–101.2; for the date, Thuc. 4.102.2–3 with Aischines 2.31 (p. 64.206–7 Dilts), cf. D.S. 12.32.3; Hornblower 1991: 154–7). See however n. 89 below. 75 Ar. Lys. 1144 actually refers, in connection with Kimon’s expedition, to a suppliant-branch (ƒŒ ÅæÆ), presumably placed on an altar by Perikleidas; this detail is not mentioned in Aristophanes’ text and probably derives from a fourth-century historian, perhaps Philochoros (so Henderson 1987: 201, comparing Lys. 1138 ¼ Philochoros FGrH 328 F 117; it is more than ‘merely a paraphrase of Ar.’ as Jacoby terms it in his note ad loc., FGrH iii. b (Suppl.) (1954) I 455 and ii. 365). 76 Theopompos, FGrH 115 F 88; the family’s proxenia is also mentioned by Andokides 3.3 in the course of his wildly inaccurate summary of Athenian–Spartan relations in the fifth century (trustingly copied out in Aischines 2.172). 77 Athenaios: Thuc. 4.119.1; Lakedaimonios: Thuc. 1.45.2, Plut. Kimon 16.1. 78 So termed in the Athenian–Spartan alliance treaty of 421 (Thuc. 5.23.3, cf. Thuc. 1.101.2).
The audience, the Demos, and Suppliants
137
that followed—the first which Athenians had ever fought in the Peloponnese—ended in the Athenian force being ignominiously sent home, as if they were hired labourers who were no longer needed. Soon afterwards Kimon was ostracized—and no doubt there were those who said that had he not been stopped, he would have turned the Athenians into subjects or slaves of the Spartans. If the similarity between the two episodes is not coincidental, it would imply that the Danaid trilogy was produced not long after the Perikleidas–Ithome episode—perhaps in the spring of 461, and quite possibly just before the vote of ostracism was due to be held79 (the holding of such a vote will have been decreed two or three months earlier, and had probably been anticipated for much longer than that). The date usually assigned to the trilogy is 463, the archonship of Archedemides; but this depends on the assumption that `$[ in POxy 2256 fr. 3.1 is the beginning of the archon’s name rather than of his title of office, an assumption which, as West notes, finds no support in the didascalia of Laios which occurs in another fragment of the same book.80 In my view a stronger objection to a date as late as 461 would be the absence in Suppliant Maidens of any sign of a skene; there is evidence for a visible skene not only in the Oresteia but also in the testimonia and fragments of two or three other late Aeschylean productions,81 and it is therefore prima facie surprising to find no such evidence in a play produced only three years before the Oresteia. We may note, however, that there are parts of the Oresteia itself in which the skene is ignored, namely the first half of Choephoroi (where the action is centred on the tomb of Agamemnon, and the palace is supposed to be somewhere offstage)* and the greater part of Eumenides (if Rush Rehm is right to suggest that the image of Athena, which Orestes clasps, is at the thymele in the centre of the orchestra, where he can be entirely surrounded by the
79 The decision whether to hold an ostracism was made in the sixth prytany (Arist. Ath.Pol. 43.5); the ostracism itself was probably held in the eighth prytany (see Jacoby, FGrH iii. b [Suppl.] [1954] I 316), on Philochoros 328 F 30). 80 West (1990a: 125), comparing POxy 2256 fr. 2 (Aesch. test. 58 Radt) Kd ¼æå () [¨ ƪ] ı. 81 See Taplin (1977: 455–6), who finds evidence for a skene in Hiereiai, the satyrplay Theoroi, and, less securely, in Edonoi (part of the Lykourgos tetralogy).
138
The audience, the Demos, and Suppliants
chorus as they weave their binding-spell about him).82 it may be that the Danaid trilogy resembled the Oresteia in this respect,83 and that the odd emphasis on housing provision [78] towards the end of the surviving play (Supp. 957–61, 1009–11) may look forward to later scenes in which a house became a crucial part of the action— no doubt the house within which forty-nine bridegrooms were murdered in a night (a record which makes the palace of Atreus, with a mere six murders in the space of a generation, almost virtuous by comparison). If Ro¨sler’s reconstruction of the Danaid trilogy84 is on the right lines, there is a positive argument in favour of bringing its date down as low as possible. He envisages the final play as containing a trial of Hypermestra, accused by her father and acquitted after Aphrodite intervenes. Such a scene would require three actors,85 which is not known to be true of any other genuine Aeschylean scene before the Oresteia.86 The introduction of the third actor might have coincided with the creation of the skene or might have been subsequent to it: it may have taken a little time for dramatists to become aware that the new configuration of the acting area, with the possibility of entries
82 Rehm (1988: 297–8) (which appeared too late to be taken into account in Sommerstein 1989b). 83 As in some other and more important respects: see Herington (1970), and now Ro¨sler (1993), who revives (mainly on the basis of Paus. 2.19.6) the idea that a trial of Hypermestra, the Danaid who did not kill her husband, was a central feature of the final play. 84 See nn. 73 and 83 above. 85 Even on the assumption that Lynkeus, Hypermestra’s husband, played no part in the trial. He must, on the Sicherl–Ro¨sler view, have been a crucial figure in the trilogy’s conclusion, since he is the only son-in-law of Danaos who is in a position to fulfil the oracle and bring about Danaos’ death. I am, however, sceptical about the whole hypothesis of an on-stage trial-scene in Danaids; for an alternative reconstruction of this play (using two actors only) see . 86 I take it that the present final scene of Seven is spurious, and that Prometheus Bound is either (and more probably) spurious (see Griffith 1977; West 1990b: 51–72; Bees 1993) or, if genuine, later than the Oresteia. On Psychostasia see Taplin (1977: 431–2), and note that Plut. Mor. 17a, `Nå º . . . ÆæÆÆ ÆØÐ ºØªØ F ˜Øe Ł b c ¨Ø , Ł b c ˙, does not necessarily imply that Plutarch believed that the ‘weighing-scene’ took place onstage (any more than Ar. Lys. 187–9 ’ ‹æŒ ›æŒ Ø Ł’ Æ;—‹ Ø Æ; N I’, u æ, çÆ , `Nå º , źçƪ Æ, implies that Aristophanes believed that in Seven against Thebes the oath of the Seven was taken, and a beast slaughtered, onstage).*
The audience, the Demos, and Suppliants
139
from, and exits in, three directions instead of two, and of keeping a character in a house just offstage from which (like Klytaimestra in Agamemnon) he or she could appear repeatedly at short notice, created a wealth of new dramatic options for which the old two-actor rule, designed for a different theatre, could not cater. At any rate there is distinctly less Aeschylean evidence for a third actor than there is for a skene, and correspondingly more reason to assume a late date for a trilogy in part of which a third actor has to be posited.87 I conclude, therefore, that there is no positive reason to reject the view that the Danaid trilogy was produced in 461, and that aspects of its action were designed to recall recent events involving Kimon and Athenian–Spartan relations and to strengthen feeling against him ahead of the ostracism vote.88 [79] On this occasion, as he had done in 472 and was to do again, I believe, in 458, and as Phrynichos had done before him, Aeschylus was using tragedy as a political weapon, and using it in the cause of that demokratia which may have been coined as a political catchword about this very time, a catchword whose echo is heard in the text of Suppliant Maidens.89 It was left for Aristophanes—or maybe for
87 That the surviving Suppliant Maidens does not require three actors is not in itself counter-evidence: in Agamemnon a third speaking actor is used only for the Kassandra scene, in Choephoroi only for the trifling part of the Servant (875ff.).* 88 The assertion by Ro¨sler (1993: 22) that ‘es ist gewiß kein Zufall, sondern hochsignifikant, daß wir auf solche [demokratische] Akzenten gerade im zeitlichen Umkreis der Reform des Ephialtes treffen’ becomes even more to the point if the trilogy was indeed produced in 461 rather than, as Ro¨sler assumes, in 464 or 463. 89 The essential features of the above argument would not be affected if one were to accept, with Badian (1993: 89–96), the evidence of Plut. Kimon 16.4–17.3, combined with Ar. Lys. 1144, that Kimon led two expeditions to aid Sparta against the helot-Messenian revolt, the first of them in 468/7, and that Perikleidas’ mission was on the earlier occasion (when Sparta was in real and considerable danger). In this case the Danaid plays could, from the point of view of their political subtext, have been produced at any time between 466 and 461 inclusive; there is of course no difficulty in supposing that, like Persians and like the Oresteia, they may have referred back to events of several years before which remained vitally relevant to current political debate. We can be fairly sure that enemies of Kimon had been endeavouring, though unsuccessfully, to bring about his ostracism (or otherwise remove him from the political scene) well before 462/1. An earlier date of production than I have assumed in the text would also make it unnecessary to raise any questions about a skene or a third actor. I am grateful to for drawing my attention to Badian’s discussion.
140
The audience, the Demos, and Suppliants
others who used him—to exploit the name and fame of Aeschylus, fifty years later, in Frogs for the purpose of undermining that same demokratia, before an audience who could still be addressed as if they were the demos of Athens but who in reality no longer properly represented it.90
ADDENDA The main thesis of this chapter—that the theatre audience cannot be regarded as a representative sample of the Athenian citizen body— is greatly strengthened if the now widely accepted view is correct that the capacity of the Theatre of Dionysus was much smaller than has usually been assumed hitherto, probably no more than 7,000 (see Csapo 2007: 97–100; and Goette 2007); after allowing for hundreds of officials, priests, etc., for the relatives and friends of chorus members, and for resident and visiting foreigners (including official delegations from up to 200 allied states), the number of seats left for ordinary Athenians and their families can only have sufficed, on any one day, for quite a small proportion of the citizen body. See, however, Mitchell-Boyask (2008) for the possibility that some potential spectator space has been disregarded in these calculations. p. 119 plays produced outside Athens: Euripides’ Andromache was one such play, according to Andr. 445, and his Melanippe the Captive, with its South Italian setting, may have been another. On tragic productions abroad see now also Taplin (2007). Two Attic comedies of the late fifth (or possibly very early fourth) century, Nicophon’s Sirens and Metagenes’ Thouriopersai, were not produced at Athens (Ath. 6.270a); the South Italian connections of the latter are obvious (and fr. 6 refers to the river Crathis whose mouth is near Thurii), and the Sirens too, like the other divinities and monsters that Odysseus meets on his wanderings, will have been thought of as located somewhere in the west (they were later placed on islands off Campania, e.g. Strabo 1.2.12–13). 90
This chapter was originally published in C. B. R. Pelling (ed.), Greek Tragedy and the Historian (Oxford: Clarendon Press, 1997) 63–79.
The audience, the Demos, and Suppliants
141
p. 122 n. 15 see now also P. J. Wilson (2007: 91–6). p. 122 n. 17 P. J. Wilson (2008: 94–5) is still on balance disposed to accept the evidence of some ancient sources which associate the origin and development of theoric distributions with the names of Pericles and of the early fourth-century politician Agyrrhius, though he inclines to the view that in the fifth century such distributions were made only occasionally, and that even in the first half of the fourth century they were not as regular as they became after the 350s. I remain more sceptical still. Wilson (95 n. 32)—and likewise Roselli (2009)—discount much too readily the silence of comedy, which took an intense interest in all matters connected with the theatre; this silence is broken, precisely when on Ruschenbusch’s view we would expect it to be, in the third quarter of the fourth century by Theophilus (fr. 12.8). The case for scepticism had already been excellently put by Cawkwell (1963: 55 n. 53). p. 125 the political biases of Old Comedy: I study these, via an analysis of the comedians’ choices of persons to satirize (and, in a very few cases, to praise), in Sommerstein (1996b). On the ‘antiwar’ (more accurately, pro-Spartan) stance of Aristophanes, see Sommerstein (2009: 209–11, 223–36). p. 126 n. 36 on the decree of 440/39, and subsequent attempts at political interference with comedy, see Sommerstein (2004a, b). p. 129 ‘the special parrhesia . . . associated with festal occasions’: I now believe this to be a myth, and hold that ‘neither in theory nor in practice was comedy above the law: its freedom of speech was no greater and no less than that of every Athenian’ (Sommerstein 2004a: 216), though there came to be ‘a “general cultural understanding” . . . that comedy’s privileged access to a vast (and, at the City Dionysia, an international) audience did not require it to be held to any higher or more restrictive standards than the law imposed on all alike’ (Sommerstein 2004b: 167). p. 129 n. 49 the reference to Mastromarco (1998)—which had not yet appeared when this chapter was originally published, and was then known to me only as a 1993 conference paper—was an error; that (excellent) study does not in fact discuss the evidence for the attempts to prosecute Aristophanes, but accepts them as factual without argument. See rather Sommerstein (2004b).
142
The audience, the Demos, and Suppliants
p. 132 the political implications of the Oresteia: see also Sommerstein (1996a: 288–95, 392–402, 416–21, 427–30). p. 133 establishing the date of Suppliants: I would now be very much more reserved about the possibility of establishing a specific political context for the Danaid trilogy (cf. Sommerstein 2008: i. 286). p. 134 n. 67 see now also Grethlein (2003). p. 135 Pelasgus’ refusal to commit Argos to war without securing the people’s consent by persuasion should probably be regarded as one aspect of a major theme of the trilogy, the opposition of consent and coercion; this theme is discussed, in relation to sex and marriage, in Chapter 6 above (especially the last paragraph) and in Sommerstein (2006). p. 136 n. 73 see Chapter 6 above and the Addenda thereto. p. 137 the palace in Choephoroi: Scullion (1994: 71–4) has convinced me that in the first half of Choephoroi, as in the second half, the palace is represented by the skene, and that the chorus (unusually) make their entrance (during lines 10–21) from the skene door. This is in full accordance with the pattern whereby the only persons to emerge from this door in the first half, and more, of the Oresteia are Clytaemestra and those executing her orders— though in Choephoroi several of them (the chorus, Electra, the nurse Cilissa) proceed to sabotage those orders. The pattern terminates abruptly with the death of Aegisthus (Cho. 869). p. 138 n. 86 even if (as is probable; see Ch. 1 above) Psychostasia was given its title only at a later date, it is unlikely to have been named after an event which the audience did not see. It is preferable to accept that the play did require a third actor, in which case it must either have been written shortly before (or even after) the Oresteia, or else, as proposed by West (2000: 345–7), be the work of someone other than Aeschylus. p. 139 n. 87 I should have added that in Choephoroi a third actor is also needed for the very small, but far from trivial, role of Pylades (in the climactic scene only; in earlier scenes, where Pylades is silent, the role can be taken by an ‘extra’).
8 Sleeping safe in our beds: stasis, assassination, and the Oresteia It has long been observed1 that among the innumerable thematic motifs woven into the texture of Aeschylus’ Oresteia, the idea of sleep is not the least prominent. It is introduced early—in the opening speech of the trilogy, where the Watchman complains about the difficulty of keeping awake through his long vigils2—appears in either a literal or a figurative context about a dozen times in the first 600 lines of Agamemnon,3 and remains important thereafter.4 Each of the three plays of the trilogy begins with someone being wakened from sleep, and each time the waking is more vividly presented and more emphasis is placed on it. In Agamemnon the Watchman, having seen the beacon-light that announces the fall of Troy, bids Klytaimestra rise from her bed and raise a cry of triumph (25–9); it is possible—I would say probable—that this cry was heard from backstage between the end of the prologue and the entry of the chorus (39/40).5 In Choephoroi we hear from the chorus—first in
[11] 1 See e.g. Petrounias (1976: 286–9); Macleod (1982: 129); Sommerstein (1989b: 220). 2 Ag. 12–17. 3 Ag. 179, 275, 290, 337, 346, 420–6, 491, 559–62, 565–6, 597. 4 Ag. 887–94, 980–3, 1218, 1247, 1357, 1448–51; Cho. 32–41, 285, 288, 315–19, 495, 523–52, 613–22, 670–1, 751, 882, 897, 1075–6; Eum. 46–142, 280, 705–6. 5 Cf. Thomson (1966: i. 20). The fact that no such cry is included in the poetic script of the play does not prove that none was uttered in performance: cf. Eum. 1043, 1047 (and see Sommerstein 1989b: 185–9 on the herald’s proclamation in Eum. 566ff.).
144
Stasis, assassination, and the Oresteia
the parodos (32–41), later and more fully when Orestes is about to take the first active steps towards his revenge (523–52)—of the terrible dream that made Klytaimestra wake screaming (34, 535) from her sleep and forthwith arrange for offerings to be sent to her husband’s tomb. And in Eumenides the same Klytaimestra who as a living woman was awakened in both these scenes (and probably again, as we shall see, at Cho. 882) appears as a ghost—or strictly, a dream of a ghost (116)—and awakens the sleeping Erinyes before our eyes. During most of the trilogy sleep is presented as unattainable, uncomfortable, or dangerous—sometimes as more than one of these simultaneously.6 The Watchman has to fight hard to keep sleep away, since if he succumbs, we gather, he will be punished.7 In the ‘Hymn to Zeus’, in connection with the idea of Ł Ø Ł ‘learning by suffering’, we hear of the ÅØ ø ‘toil reminiscent of grief ’ that drips before the heart (Ag. 179–80)—and we need not, for our present purposes, trouble ourselves as to whether it drips ‘in sleep’8 or ‘instead of sleep’,9 since both sleeplessness and frightening dreams are afflictions of which we shall hear much more. Soon we learn from Klytaimestra [2] that the Watchman is only one of an army of watchers at a whole string of beacon-sites stretching all the way to Lemnos, who have been awake every night (cf. 290) for, perhaps, several months waiting for the sight of a distant fire that will be the cue to light their own. At Troy, meanwhile, the Greek army can look forward to ‘a whole night’s sleep without sentries’ for the first time in ten years; the Herald will presently be giving a graphic account of their previous discomforts (559ff.). Earlier, back at Argos, Menelaos’ nights had been filled with vain and tantalizing dreams of Helen, who slips through his arms (424–5) as lightly as she had slipped through the palace gates when she departed with Paris (407–8); while another deserted spouse, Klytaimestra, claims to have lain awake crying, persistently dreamed of something terrible
6 7 8 9
See Sommerstein (1989b: 220). Ag. 14 ç ªaæ I Ł’ o ı ÆæÆÆ ØÐ . Ł’ o ø fi codd. ( ª’ o ø fi Page). I Ł’ o ı Emperius (adopted in M. L. West’s Teubner text (West 1990a).
Stasis, assassination, and the Oresteia
145
happening to her husband,10 and been awakened by any buzzing gnat (887–94). When Agamemnon has entered the palace, the Elders find it impossible to shake off an anxiety that refuses to be put behind them like a bad dream (980–3); if it is a dream, it is one from which they cannot wake. The only sleep in Agamemnon that is truly peaceful and unflawed is the sleep for which the Elders will finally pray: the ‘endless sleep’ of death (1448–51). So too in Choephoroi. Klytaimestra’s dream we have noted. Orestes, if he fails to avenge his father, is menaced (among other things) with ghastly nightmares sent by the Erinyes that will drive him mad (Cho. 288):11 accordingly he desperately endeavours to awaken his father’s spirit from its sleep below the earth (415, cf. 315–19)— and apparently fails.12 The Nurse recalls how Orestes as a baby woke her with his cries at all hours (751); his mother on the other hand remembers him sucking her breast in contented drowsiness.13 Choephoroi ends with the chorus hoping against hope that the curse on their master’s house may now ‘fall asleep and cease’ (1075– 6). At the beginning of Eumenides, however, we find that the curse, now embodied in the Erinyes who also call themselves Arai ‘Curses’ (Eum. 417), while it has indeed, literally and visibly, fallen asleep, has certainly not ceased; and though Orestes may claim that the blood-pollution on his hands is ‘dormant and fading’ (Eum. 280), his pursuers are far from agreeing. In the last 750 lines of the trilogy, the theme of sleep makes only one further appearance; and when it does, like many other themes in the latter part of Eumenides, it is transformed from a sinister theme 10
The audience, better informed than Agamemnon, are left free to speculate whether these dreams, like so much else in Klytaimestra’s speech of welcome, are entirely fictitious, or whether she did indeed dream about Agamemnon’s death, not, as she pretends, through fearful anxiety but through pleasurable anticipation! 11 The difficult line 285 may also describe a nightmare-vision; on the text and interpretation of this passage see Garvie (1986: 116).* [12] 12 I have argued in Sommerstein (1980b: 67–9) that in 479–96 Orestes and Electra are attempting without success to raise up their father’s ghost to give them advice, encouragement, and blessing. 13 There is, of course, no inconsistency here; Klytaimestra, like upper-class mothers in all ages, is assumed to have only had the baby with her when he was behaving nicely (Clark 1985: 79: ‘did anyone seriously expect the queen of Argos to tackle the night feeds and the nappy wash?’)
146
Stasis, assassination, and the Oresteia
to an auspicious one,14 when Athena declares that she is founding the new council of the Areopagos as ‘a guard-post of the land, watchful on behalf of sleepers’.15 It is from this phrase that I want to begin my main investigation. The words just quoted, which form the conclusion16 of Athena’s speech establishing the Areopagos council, strike, one might think, a somewhat surprising note. Whether one viewed the Areopagos as a council with a key role in the government of Athens, or as a homicide court and little else,17 there would seem at first sight to be no reason to single out the sleeping citizen as the object of its special protection. Moreover, the same idea was expressed in slightly different form, with ring-compositional symmetry, at the beginning of Athena’s account of the council’s [3] function: on the Areopagos, she says, the respect and fear of the citizens will prevent wrongdoing ‘by day and night alike’.18 Why all this stress on nocturnal crime? And what kind of crime is likely to be meant? Before its wings were clipped by Ephialtes in 462/1, the Areopagos council is likely to have been mainly concerned, through the procedures of eisangelia and euthynai,19 with political misdemeanours and malpractices of various kinds—treason, bribery, embezzlement no doubt prominent among them.20 Such crimes could of course be 14 See Lebeck (1971: 131–3); Sommerstein (1989b: 220 and 304 (index s.v. ‘themes of play and trilogy . . . ominous themes transformed into auspicious ones’)). 15 Eum. 705–6 ø o æ | KªæŪæe çæ æÅ Æ ªB. 16 The following four lines (707–10) are not part of the foundation-speech proper; they are a transition back to dealing with the case currently being tried, ending with an instruction to the judges to cast their votes. 17 On the question which of these views, if either, is implicitly promoted by Eumenides, see Sommerstein (1989b: 216–18). 18 Eum. 690–2 K b fiH Æ | IH ç ıªª c e c IØŒ ØÐ | å Ø ’ q Ææ ŒÆd ŒÆ’ Pçæ Å › H. On the possible ambiguity of this sentence see pp. 159–60. 19 Eisangelia to the Areopagos had been introduced by Solon (Arist. Ath.Pol. 8.4). On the possible responsibility of the Areopagos for euthynai (the audit of outgoing magistrates), not directly attested in ancient sources, see Rhodes (1981: 316–18); L. A. Jones (1987: 59); Carawan (1987); contra (very tersely), Wallace (1989: 80). 20 In the fourth century, at any rate, the main forms of eisangelia were (1) to the Assembly, for treason or corruption only (Hyp. Eux. 7–8) and (2) to the Council (of Five Hundred) for misconduct by a magistrate (Arist. Ath.Pol. 45.2) including embezzlement (Ant. 6.35); while in euthynai the two most serious charges that could be brought were embezzlement and bribery (punished with a fine of ten times the amount of money involved, as against a penalty of simple or double
Stasis, assassination, and the Oresteia
147
committed by night as well as by day, but there would be little reason, in general, to attach any specific significance to the time at which they were committed, and—what is perhaps more important—crimes of this kind are not a focus of interest in the Oresteia. We should think rather of the function which the Areopagos council performs in the play itself, and which, moreover, it performed in reality at the actual time when the play was produced, namely the trial of charges of homicide. In the case of homicide, it is easy to see why the distinction between day-time and night-time crime could be felt to be of considerable significance.21 At night, especially under ancient conditions of little or no artificial lighting, the killer will find it easier to lie in wait unnoticed, easier to avoid recognition by witnesses, easier to escape; often, too, he may find his victim asleep and totally helpless. The nocturnal killer is, in fact, both harder to guard against and harder to bring to book, and it is therefore intelligible if he inspired greater fear. I will presently be suggesting that nocturnal killing, and the role of the Areopagos council in connection with it, may have been matters of intense public concern in 459/8 bc when the Oresteia was produced. What I want to try and show now is that nocturnal killing is a major theme of the Oresteia itself. In Agamemnon there are three episodes of nocturnal killing, two actual and one hypothetical. In the order in which they are mentioned in the play, the first of the three is the battle and massacre that accompanied the capture of Troy, when ‘friendly Night’ cast her net over the city’s towers ‘so that neither the full-grown nor any of the young might overtop it’ (Ag. 355–60) and left the Trojans next
damages for other offences: Arist. Ath.Pol. 54.2). See Hansen (1991: 212–18, 221–4). There is some specific, though indirect, evidence that trials for these offences took place on the Areopagos before 462: Themistokles was prosecuted (and condemned in his absence) for treason by eisangelia (Krateros FGrH 342 F 11), and the charge of corruption against Kimon after the Thasos expedition arose out of his euthynai (Arist. Ath.Pol. 27.1 with Plut. Kim. 14.2–15.1). See Rhodes (1981: 319–20, 335–6); Bauman (1990: 22–31). 21 Just as it was in the case of theft, with which the Areopagos was not directly concerned: anyone caught stealing at night, no matter how small the theft, could be killed if necessary to stop him getting away, and his killer incurred no penalty (Dem. 24.113).
148
Stasis, assassination, and the Oresteia
morning lamenting the deaths of their dearest ones and their own enslavement (326–9). The second is an event that is only anticipated. In the song which began with the invocation of thanksgiving to Night (and Zeus) part of which has just been quoted, the chorus eventually find themselves reflecting on the great loss of Greek life in the tenyear war, and the resentment and anger against the Atreidai which is building up in Argos (445–51, 456–8); ‘and’, they continue, ‘I have an anxiety that waits to hear of something happening under cover of night’.22 These are guarded words, but when they are followed by reflections on the certainty of divine vengeance upon ‘those who have caused the deaths of many’, on the hounding by the Erinyes of ‘him who is fortunate without justice’, on the thunderbolt of Zeus which strikes those who enjoy ‘an excess of good renown’, and on the undesirability of being a sacker of cities (461–72), this confirms what the audience had probably suspected already, that the Elders’ apprehension is that an attempt [4] may be made to assassinate Agamemnon or Menelaos or both on their return.23 It is striking, and may be significant, that Aeschylus makes them think of such a deed as a nocturnal one. The third example of violent death by night in Agamemnon is the storm that wrecked the Greek fleet on its way home, with Agamemnon’s own ship the only one known to have survived. It seems likely that in the cyclic Nostoi the storm was mainly a daytime affair. What happened at night was something related but distinct: the fleet, having suffered considerable losses, was still at sea after dark (having presumably been prevented by the storm from putting in anywhere) and Nauplios, seeking revenge for the death of his son Palamedes, lured them to destruction on the Kaphereian rocks by lighting a beacon which they took to mark a harbour.24 Aeschylus, for whom 22 Ag. 459–60 Ø ’ IŒFÆ Ø æØ Æ ıŒÅæ ç. [13] 23 They are right: such an attempt will indeed be made. They are even right about the motive: anger over the death of a loved one in Agamemnon’s and Menelaos’ war. Where they are wrong—disastrously and, as Cassandra will find, incorrigibly wrong—is in believing that the danger comes from the citizens (cf. 809) when in fact it comes from within the palace. 24 Our evidence for the narrative of the Nostoi comes partly from Proklos’ synopsis of the poem (EGF p. 67 Davies < ¼ Nostoi Arg. }3 West>), and partly from later mythographic accounts which seem to derive from it ([Apollod.] Epit. 6.5–11;
Stasis, assassination, and the Oresteia
149
it is important that the storm should have been caused by divine anger (Ag. 636, 649), has no use for Nauplios and makes no mention of him; but rather than abandon the nocturnal destruction for which in the traditional account Nauplios was responsible, he extends the storm itself into the night—or rather, he contrives to give the impression that almost the whole of it took place at night. In the Herald’s twenty-one-line account of the storm (Ag. 650–70), nightfall comes between the third and fourth lines (cf. 653 K ıŒ ). Up to that point the account has included no descriptive material except for the one word Fæ ‘fire’ (651); in contrast, the eight lines that follow, narrating the events of the night and the following morning, include graphic description of the winds, waves, and rains and of the ships helplessly colliding, and end (658–60) with the sunrise revealing the whole Aegean dotted with spars and bodies. There follows an even longer section (661–70) narrating the escape of Agamemnon’s ship, also during the night,25 and adding further touches to the hearer’s picture of the disaster.26 In Aeschylus’ presentation of events, the nocturnal destruction of the fleet serves as fit retribution for the nocturnal havoc and sacrilege which the Greeks had wrought at Troy;27 but it also adds another weighty item to the growing tally Hyginus Fab. 116). Proklos’ summary seems to imply two separate storms, one ‘in the open sea’ in which Menelaos lost all but five of his ships, another ‘near the Kapherid rocks’ in which Aias son of Oileus perished. The later accounts have much more to say about the latter storm. According to Apollodoros it began ‘off Tenos’ (the next island but one south-east of Euboia in the Cyclades chain), and after the death of Aias ‘the others were blown towards Euboia during the night, and Nauplios lit a beacon on Mount Kaphereus’ which caused the wreck of the fleet. Hyginus places the whole of this disaster close to Euboia, but like Apollodoros he implicitly associates the storm itself with the daytime by making his first mention of night at the beginning of his account of Nauplios’ treachery: ‘tempestate et flatibus adversis ad saxa Capherea naufragium fecerunt, in qua tempestate Aiax Locrus fulmine est a Minerva ictus . . . ; ceteri noctu cum fidem deorum implorarent, Nauplius audivit . . . ’. Proklos’ summary of the Nostoi says nothing about Nauplios, but we know that he and Palamedes were mentioned in the poem (fr. 1 Davies < ¼ 11 West>), and pace Davies (1989: 83) there is no reason to doubt that his role in it was essentially the same as it was in later accounts of the same events.* 25 It was only with the coming of day that the men knew they were safe (667–8). 26 Notably the driving of ships against the rock-bound coast (666). 27 The connection was made by Klytaimestra in 338–44 (‘if they respect the city gods of the conquered land and the places where they dwell, then the conquerors will not be conquered in their turn . . . for they have yet to secure a safe return home’).
150
Stasis, assassination, and the Oresteia
of violent death by night. This list is further extended in Choephoroi in a way that brings the theme of nocturnal killing into close relation with that of sleep, in the story of Skylla and Nisos, told by the chorus as one of their exempla of female criminality just before Orestes begins to put his plan of revenge into action (Cho. 613–22). As they tell the tale, Skylla, bribed by Minos, cut off the lock of immortality on the head of her father Nisos while he slept,28 thus enabling Minos’ Cretans to kill him. All these nocturnal killings are, of course, outside the action of the drama properly so called. One of the four episodes occurred many years previously; two others occur far away, at Troy and at sea respectively; the fourth never actually occurs at all. The killings that have formed part of the action thus far—those of Agamemnon and Cassandra—are undoubtedly represented as taking place during the daytime:29 shortly before they happen, Cassandra prays to the Sun (Ag. 1323), and some time afterwards Aigisthos hails the ‘light of the day that brings revenge’ (1577). [5] But there remains one nocturnal killing, or rather two, that do form part of the action of the Oresteia, and can indeed be regarded as its central event, fully justifying, from the dramatic point of view, the fact that in due course the court set up to try Orestes will be given a special remit to prevent crime at night and ensure the safety of
The explicit reference to the destruction of sanctuaries and altars made later by the Herald (527) is under suspicion of being an interpolation (the latest editor, West 1990a, accepts the deletion first proposed by G. G. Salzmann), but it was in any case notorious that the sack of Troy involved gross desecrations of altars and temples, notably in the killing of Priam and the seizure of Cassandra, and already before Homer (cf. Od. 1.325–7, 3.132–5; see S. R. West in Heubeck, West, and Hainsworth 1988: 116) divine wrath at these was the accepted explanation of the storm that overwhelmed the returning fleet. 28 Cho. 621 Ł’ . . . o ø fi . To be sure, as Easterling (1993: 19) has noted, there is no explicit reference to night; but while the siesta was no doubt a regular feature of Greek life in antiquity as now (it is indeed referred to by the down-to-earth Herald in Ag. 565–6 Å æØ ÆØÐ ŒÆØ, along with crowded ships, damp bivouacs, and ubiquitous lice), I know of no case of a character in Greek heroic legend suffering the fate of Hamlet’s father, and we can safely assume that in the absence of explicit indication to the contrary, ‘killing in sleep’ would automatically be taken to imply ‘killing by night’. 29 As of course did the sacrifice of Iphigeneia, and also, it would appear, the killing and eating of Thyestes’ children (cf. Ag. 1592 Œæ ıæªe q Ææ).
Stasis, assassination, and the Oresteia
151
citizens while they sleep. I refer of course to the killing of Aigisthos and Klytaimestra. How can we know that we are meant to envisage these killings as taking place at night? The first indication is given a considerable time before the murders, when Orestes first approaches the palace. Asking the slave who opens the door to report to the masters of the house that a stranger has arrived bringing important news from abroad, he bids him make haste because ‘the dark chariot of Night is rapidly approaching’ and it is time for travellers to seek lodging (Cho. 660–2). The impression thus created at the beginning of the scene that it is not long before sunset is confirmed, in ring-compositional fashion, towards its end, when Klytaimestra assumes that the supposed Phokians have been on the road all day and it is now the proper time (› ŒÆØæ) for them to receive appropriate hospitality (710–11). When Orestes, with Pylades, has gone inside, the chorus, in praying for his success, invoke Hermes (according to the one manuscript) as ‘underworld’ Hermes and ‘nocturnal’ Hermes (727–8 åŁ Ø . . . ŒÆd . . .
åØ ). He has already been appealed to under the former title both by Orestes and by Electra (1, 124); but the latter title is not only new in the play, it is never applied to Hermes in any other surviving text of any sort, and has no relevance to the current situation unless it is taken to imply—consistently with what we will have gathered from the preceding scene—that the ‘contest’ in which Hermes is being asked to lend his assistance is going to be fought at night.30 Aigisthos arrives at the palace, walks into Orestes’ trap, and is killed. Directly thereafter we find a servant shouting furiously for the ‘women’s doors’, which are apparently bolted, to be opened (877–9), and when at first there is no answer, he complains that everyone is asleep (881–2). Garvie31 thinks this should not be taken literally, and Easterling (1993: 20–1) points out that Klytaimestra [14] 30 West (1990a), indeed, following a suggestion by Blaydes, makes this explicit by emending the suspect transmitted reading ŒÆd e
åØ to ŒÆØæe ıåØ (ØÐ ’ Kç æ FÆØ ØçźØØ IªHØ ). This may well be right (though West himself in his Studies in Aeschylus [West 1990b: 251] speaks of it only as ‘tempting’), but I neither wish nor need to rely on it. 31 Garvie (1986) ad loc.
152
Stasis, assassination, and the Oresteia
summoned Aigisthos in order to discuss the situation with him (cf. 672–3, 716–18) and we should therefore not expect her to have gone to sleep but to be awaiting his arrival. However, since the idea of summoning Aigisthos was first aired, Aeschylus has engineered a shift in our assumptions about what will happen when he comes. After 718 the notion of his being wanted to discuss matters with Klytaimestra does not appear again.32 Instead we are given to understand that he has been summoned to the palace to speak to the visitors and hear their story ‘man from man’ (736, 850). We hear this from the Nurse as soon as she appears (734–7, cf. 771), and when Aigisthos arrives he makes no mention of Klytaimestra and likewise thinks only of an interview with the visitors, as do the chorus (848–54). The ‘active’ Klytaimestra who, to Orestes’ surprise (cf. 664–7) and probably to the [6] audience’s, came to meet him at the outer door, has given place to a ‘passive’ Klytaimestra who remains within the women’s quarters. These quarters, moreover, are bolted up, which they certainly were not at 668 when Klytaimestra came to the outer door almost before Orestes had finished asking the doorkeeper to fetch someone in authority; the change suggests that it is now night-time, when the women’s doors might well be bolted on the inside for the inmates’ protection. The difficulty, too, which the servant experiences in getting any response to his frantic shouting and knocking does strongly suggest that we are to suppose that all the women of the house are indeed now asleep.33 I therefore see Klytaimestra as being here wakened from sleep for the second night in succession; last night it was by a terrifying dream
32 Indeed, given that one scene ends with Klytaimestra saying she will discuss the situation with Aigisthos (and giving the impression ex silentio that he is in the house), and that the next scene begins with the Nurse leaving the house to fetch Aigisthos for a quite different purpose, we may well be meant to deduce that what Klytaimestra said to the supposed Phokians was a mere blind, like some other aspects of her speech and demeanour in that scene (cf. 737–40). 33 One is tempted to refer here to a vase painting depicting the death of Aigisthos (see Prag 1985 pl. 12c–d, no. C19 [Aegisthus Painter, c.470]) in which Klytaimestra is shown running up to defend him with disordered hair such as no queen, surely, would ever have unless just roused from bed. In another painting of the same period (Prag 1985 pl. 16b þ17, no. C22 (Berlin Painter, 470–65)) Klytaimestra’s hair seems to have been done up in a hurry, for one lock has escaped and is hanging down.
Stasis, assassination, and the Oresteia
153
involving a snake which represented Orestes; this night it is, in effect, by Orestes himself. Against this interpretation of 877ff. attention has not unreasonably been drawn by Easterling (1993: 22–3) to the scene following the murder, in which Orestes displays to men and gods the bodies of the ‘twin tyrants’ and the bloodstained robe which Klytaimestra had used, years before, to trap and slay Agamemnon. Even if 986, where he appeals to the Sun to be his witness, is spurious, as has recently been argued on independent grounds,34 the emphasis of Orestes’ speech is so strongly visual that we are clearly not meant to suppose that the time is night. In reply to this one might simply argue that in ancient theatrical conditions imaginary darkness could easily be forgotten;35 but one can say more than this. It is dark no longer. The light, as the chorus sing, has now become visible (961); the house, long prostrate, can rise up (963–4). As at the beginning of Agamemnon and as at the end of Eumenides, ‘the people that walked in darkness have seen a great light’. And while the choral ode, sung by palace slaves, emphasizes the liberation of the house, Orestes, speaking to the people of Argos (cf. 1040 Æ æª ı), announces the death of Klytaimestra and Aigisthos as the liberation of the city and only in the second place as his revenge for his father and his house (974–5). The scene which commences with this declaration is, in fact, an assembly of the Argive people,36 which would begin, most appropriately, at sunrise.37 There is of course no difficulty in supposing that the play’s action is thus spread over parts of two days: Agamemnon—to take only the 34
West (1990b: 262–3), adopting a suggestion by W. S. Barrett. If 986 is deleted, the witness to whom Orestes appeals will be Zeus the Father (984–5). 35 Cf. Garvie (1986: 322); MacDowell (1971: 19). 36 It does not follow that there was a stage-crowd to represent them. Orestes may well have spoken straight at the audience, in effect conscripting them into the role of the Argive people (cf. Whallon 1980: 90; Hogan 1984: 142; and, for a parallel situation in Eum. 566ff., Sommerstein 1989b: 186). Garvie (1986: 316–17) thinks that Orestes ‘addresses the Chorus throughout, as the conventional representatives of Argos[; t]he fact that they are technically slave-women is forgotten’; but it would be impossible in production to ignore the gender of the chorus as manifested by their masks and costumes, which would make them an utterly inappropriate audience for a speech addressed to the people of the Argive polis. 37 As Athenian assemblies did (see Ar. Ach. 19–20, Thesm. 376, Eccl. 84–5).
154
Stasis, assassination, and the Oresteia
nearest example—contains first the arrival of the news of Troy’s fall, by beacon, on the very night when it happens, and then the arrival of the Herald, who has sailed from Troy with Agamemnon through a storm which, as he himself makes clear, occupied at least part of a day and the whole of a night. There is, then, nothing to prevent us from concluding, on the cumulative evidence of the indications discussed earlier, that the killing of Aigisthos and Klytaimestra is indeed to be imagined as happening at night. I could continue by tracing the association of violent death and night on into Eumenides, where it is embodied in the Erinyes, those children of Night38 and avengers of murder who hound their victims to death and beyond.39 I wish now, though, to consider the [7] possibility that there may be a connection between the two themes of the Oresteia that we have been looking at—the theme of sleep and the theme of nocturnal killing—and a recent event whose shock-waves were still reverberating through the Athenian body politic: the death of Ephialtes, the first political assassination that Athens had known for half a century, and the last it was to know for another half-century.40 For not long after he had introduced and carried, in 462/1, legislation which drastically reduced the powers of the Areopagos Council, Ephialtes had been murdered in his sleep. That is an assertion which may be thought to require documentation, especially since David Stockton has argued41 that there is no proof, and no strong reason to believe, that Ephialtes was murdered at all. I will now present the case for a verdict of unlawful killing. It will be helpful to compare the evidence we possess about the death of Ephialtes with that about the sudden death in 129 bc of Scipio Aemilianus, which Stockton cites as a parallel case. Scipio’s death is often casually spoken of by later writers (above all Cicero) as
38
Eum. 322–3, 416, 745, 793, 844, 1033. Ibid. 174–8, 264–75, 339–40. The most recent assassination before that of Ephialtes had been the killing of Hipparchos by Harmodios and Aristogeiton in 514; after Ephialtes there were no more until Androkles was murdered by oligarchs in the run-up to the coup d’e´tat of 411 (Thuc. 8.65.2). 41 Stockton (1982). 39 40
Stasis, assassination, and the Oresteia
155
a murder,42 but it is now widely believed that it is at least as likely to have been natural.43 In assessing the witnesses, we must distinguish between, on the one hand, mere passing references by writers who give no details and simply take the murder story (or one of the murder stories) for granted, and, on the other, statements evincing a serious interest in the facts of the death. Of such statements there survive, by my reckoning, seven.44 In not one of these is it stated as a fact that the death was murder. Six accounts start by saying Scipio was ‘found dead’, the seventh45 says he ‘died after dinner at home’; murder, suicide, and natural death are all canvassed as possibilities, in various combinations;46 those who mention murder differ as to the alleged method47 and the likely suspects;48 two49 report the diverse views of previous historians (though without naming them), and the earliest of our six authors, Velleius, specifically states that the 42 Cic. pro Mil. 16; de Amic. 41; de Fato 18; Rep. 6.12; Att. 10.8.7; cf. also e.g. Val. Max. 8.15.4 mors clandestinis inlata insidiis. Pompey in the Senate on 9 February 56 bc, when he declared se munitiorem ad custodiendam vitam suam fore quam Africanus fuisset, quem C. Carbo interemisset (Cic. ad QF 2.3.3), evidently took it for granted that most of his hearers would regard the murder as a well-known ‘fact’. [15] 43 The case for natural death was most strongly argued by Carcopino ([1928] 1967: 85–127); and it is now possible for a writer like Crawford (1981: 114) to assert without argument that Scipio ‘died of disease’. Scipio’s biographer, A. E. Astin (1967: 241), is non-committal, hinting, with Last (1932: 45), that Scipio’s friends may have thought it politically advisable at the time to treat his death as natural, but concluding, as Last had, that the facts can no longer be established. 44 Livy Per. 59; Velleius 2.4.6; Plut. Rom. 27.5 and C.Gr. 10.5–6 (which have inconsistent details and must derive from different sources); Appian BC 1.20; [Aur. Vict.] de Vir.Ill. 58.10; Orosius 5.10.9–10. 45 Plut. Rom. 27.5. 46 Murder is mentioned as a possibility in all seven accounts; suicide by Plutarch in Romulus and by Appian; natural death by Plutarch in Romulus (again) and by Velleius. 47 Plutarch in Gaius Gracchus, alone, claims there were marks of violence on the body; in Romulus, however, he reports that those who believe in a murder say it was by strangulation, and this also seems to be the implication of pseudo-Aurelius Victor’s statement that Scipio was found obvoluto capite, ne livor in ore appareret. 48 Livy’s epitomator, and Orosius (who may likewise depend on Livy), say that Sempronia was suspected; Plutarch in Gaius Gracchus, however, claims that suspicion fell on Gaius Gracchus and Fulvius Flaccus. Ten years after the event, C. Papirius Carbo was accused of complicity (P. Africani necis socius fuisti) when being prosecuted on other charges by L. Crassus (Cic. de Or. 2.170); two generations later, Cicero mentions Carbo’s responsibility for the death as a widespread belief (Fam. 9.21.3), while Pompey, as we have seen (n. 42), asserted it as a fact. 49 Viz. Velleius, and Plutarch in Romulus.
156
Stasis, assassination, and the Oresteia
majority of his predecessors had regarded the death as natural. In the light of all this, even if we had not had evidence50 that Scipio’s own closest friend, at the climax of the eulogy he composed for Scipio’s funeral, had spoken of his death as due to disease,51 we would still, I think, have had sufficient reason to believe that it had been so, and that the various and inconsistent tales of foul play were politically motivated rumour-mongering. What do we find in the case of Ephialtes? We have four sources of significance, which should probably be reduced to three, since Plutarch52 explicitly refers to Aristotle53 and offers no information independent of that source; in addition to Plutarch, the other sources are Antiphon54 and Diodoros.55 Every source states positively that Ephialtes was murdered. According to Diodoros he was ‘done away with’ (I ÆØæ Ł ) by night and ‘the ending of his life was obscure’ (¼Åº å c F ı º ı ).56 The Ath.Pol. has it that Ephialtes was done away with (the same verb, I fi ÅæŁÅ) by being killed in an underhand manner (ºç ÅŁ ) through the agency of 57 Aristodikos of Tanagra. [8] The testimony of Antiphon is of particular value. He stands nearer to the event than either of the other sources. He was probably old enough to remember it himself;58 some of the jurors trying his client, Euxitheos, for the alleged murder of Herodes certainly were. Laelius fr. 22 Malcovati4 eo morbo {rumtemovit (mortem obit Orelli, Stangl; see Malcovati 1975). Cf. also Lucilius 1093 Marx insperato abiit, quem una angina sustulit hora. 51 Cicero, however, is prepared to make Laelius implicitly accuse the family and friends of Tiberius Gracchus of responsibility for Scipio’s death (de Amic. 41), though earlier in the dialogue (}12) he speaks only of ‘what men suspect’. 52 Plut. Per. 10.8. 53 Meaning Ath.Pol. 25.4. 54 Ant. 5.68. 55 D.S. 11.77.6. 56 The handling of both these expressions by Stockton (1982) is unsatisfactory. He renders the first as ‘died’, without offering any evidence that I ÆØæ ØÐ ŁÆØ can ever be used of a natural death; the second he paraphrases as ‘nobody knew how his life ended’ when the Greek expression is equally capable of meaning that nobody knew who was responsible for his death. 57 The preposition used is Ø, and Plutarch rightly understood the author to have been asserting that Aristodikos was acting on behalf of others. [16] 58 According to [Plut.] Lives of the Ten Orators 832f he was born ‘at the time of the Persian wars’ and was slightly younger than Gorgias (whose accepted 50
Stasis, assassination, and the Oresteia
157
Of Herodes it was even truer than of Ephialtes that ‘the ending of his life was obscure’, for his body was never found. The defendant hints that he might well have drowned accidentally, drawing attention59 to the fact that he disembarked when drunk from a boat that was riding at anchor,60 and remarking that he ‘would probably have been in no condition to control his movements’.61 Antiphon decided to mention the case of Ephialtes as a parallel to that of Herodes: it would have been in his client’s interests to point out that nobody could be sure Ephialtes had been murdered at all—and we may add that Antiphon himself, as an opponent of the democratic system, would have had political motives of his own for propagating that view if there was any credibility in it. He does not. All he says is that ‘those who killed Ephialtes have never to this day been discovered’; by itself this might just possibly be understood as a way of hinting that no such persons ever existed, but such an interpretation is surely excluded by his further remark that ‘those who killed Ephialtes’ made no attempt to dispose of the body. Far from suggesting that Ephialtes might not have been murdered, Antiphon, contrary to his client’s forensic interests and his own political interests, takes it for granted that he was. Presumably the known facts left him no alternative. It thus appears that the circumstances of Ephialtes’ death were such as to rule out any explanation other than murder. We know, moreover, from Diodoros, that he was killed at night. Can anything more be extracted from the information we are given? Antiphon’s language at first sight implies that Ephialtes was attacked when in the
birth-date was 484–481 (cf. Apollodoros FGrH 244 F 33 with Jacoby’s note)). A birthdate not long after 480 is supported by the allegation later made against Antiphon (Ant. fr. 1 Thalheim) that his grandfather had been a spearman (æıçæ) in the Peisistratids’ bodyguard (and therefore presumably born no later than 535–530, since Hippias was expelled in 510). It is true that no contemporary or near-contemporary source refers to Antiphon in a context earlier than the 420s; but this may be explained by the general scantiness of literary evidence for the 440s and 430s (especially comedy), by Antiphon’s unwillingness to engage in public political activity such as might have put his name in the inscriptional record, and by the probability that it was only in the 420s that he began the novel practice of putting copies of speeches into public circulation. 59 Ant. 5.26. 60 Ibid. 5.21 uæ Ø. 61 Trans. M. Edwards (in Usher and Edwards 1985).
158
Stasis, assassination, and the Oresteia
company of others; it would have been unfair, he says,62 to demand that ‘those who were with him’ (f Æ) should be made to guess ( NŒÇ Ø ) the identity of his assailants on pain of being held guilty themselves if they could not. This might suggest a sudden attack out of doors (say when Ephialtes was on the way home from a symposium) with the attackers escaping under cover of the darkness.63 In those circumstances, however, it would in fact not have been at all unreasonable to suspect Ephialtes’ companions: it would have been the easiest thing in the world, if they had been the guilty parties, for them to invent a phantom gang of thugs who committed the murder and then conveniently vanished. And Antiphon would have wanted to stress the distinction between this case and his client’s; ‘and yet,’ he could have made Euxitheos say, ‘it would have been much more reasonable to suspect them than me, since they were in Ephialtes’ company when he was attacked, whereas Herodes, as I have told you, disembarked alone’. It is more likely, therefore, that f Æ is, as it were, an imperfect participle,64 meaning not ‘those who were with him (sc. at the time of the murder)’ but ‘those who had been with him [sc. shortly before, as I had been with Herodes]’—or, as we might put it today, ‘the last people known to have seen him alive’. If so, Ephialtes, like Herodes, parted from these companions alive and well—and he was dead by morning, with unmistakable evidence that the [9] death was murder: which points strongly to the conclusion that he was found dead in bed with stab wounds.65 The question that remains is, of course, whodunit. There appears at first to be a conflict of evidence here. Antiphon speaks of the killers in the plural, and says they were never discovered; Diodoros also, on the most natural interpretation of his words, speaks of the crime as never having been cleared up. The Ath.Pol., on the other hand, confidently asserts that the murder was done through the agency of 62
Ant. 5.68. As John Rich observed in discussion . See Ku¨hner and Gerth (1898–1904: i. 200 Anm. 9). 65 The contrary conclusion of Stockton (1982) that ‘his body bore no obvious marks of physical violence’ is based on what we have seen (p. 156 with n. 56) to be an overconfident interpretation of an ambiguous expression in Diodoros. I do not understand Stockton’s claim that the evidence of Antiphon supports his view. 63 64
Stasis, assassination, and the Oresteia
159
Aristodikos of Tanagra. There is, however, as Peter Rhodes and others have seen, an easy way to reconcile these accounts: Aristodikos ‘was known to be the actual killer but it was never discovered whose agent he was’.66 It is not clear how Aristodikos came to be suspected, though various scenarios are imaginable. It is likely, however, that he was never brought to trial: as soon as he got to know he was suspected he had every reason to flee the country, particularly since as a noncitizen, and in a case of such political importance, he ran the risk of being questioned under torture if he stayed.67 Such flight by a suspect would no doubt be taken by most Athenians as proof of guilt. Be that as it may, what is clear is that no one else, and in particular no Athenian or Athenians, ever went on trial for killing Ephialtes. In my commentary on Eumenides,68 I drew attention to what I saw as the ambiguity of one of the passages in which the Areopagos council is associated with protection against nocturnal crime: K b fiH Æ IH ç ıªª c e c IØŒ ØÐ
å Ø ’ q Ææ ŒÆd ŒÆ’ Pçæ Å › H
66 Rhodes (1981: 322)—though he mentions this solution only as one of two possibilities, the other being that Aristodikos’ responsibility was ‘a rumour that was not substantiated’ which the author or his source has reported as fact. In connection with Aristodikos, the ‘man from Boeotia . . . presumed to be the agent of an Athenian citizen’ (Rhodes loc. cit.), one is tempted to recall the foreign robbers who were supposed to have killed Laios (Soph. OT 122–3, cf. 715–16), who Oedipus instantly infers (ibid. 124–5) must have been in the pay of some Theban enemy or enemies of the king, as indeed had been suspected at the time (ibid. 126). 67 For the torture of free non-citizens, see Lys. 13.54, 59; Dem. 18.133 (cf. Dein. 1.63); Aischines 3.223–4 (cf. Dem. 18.137); Antiphanes fr. 274 KA (‘a man only has to ship in a consignment of pepper, and someone puts down a motion for him to be tortured as a spy’). It was carried out when ordered by the Assembly in cases where the security of the state was believed to be under threat. It may also (but this is disputed) have been possible for free non-citizens to be examined under torture in cases of homicide and wounding. See Lipsius (1905–15: 894–5); Thu¨r (1977: 18–19); Carey (1988). The ‘decree of the archonship of Skamandrios’ (And. 1.43) only applied to citizens (cf. Lys. 13.27). 68 Sommerstein (1989b) on Eum. 690–2.
160
Stasis, assassination, and the Oresteia And in this place reverence of the citizens and inborn fear will be a restraint from doing wrong, both by day and at night alike. (690–2)
This could mean, I argued, either that respect and fear of the Areopagos by the citizens would keep them from doing wrong by day or night, or alternatively that respect and fear of the citizens would keep the Areopagos from doing wrong by day or night. Could this ambiguity, and through it the whole nexus of ideas that we have examined relating to sleep and to nocturnal killing, have something to do with popular opinions and suspicions about the death of Ephialtes? It would not, at any rate, be all that surprising if many Athenian democrats had come to believe that the men behind Ephialtes’ death were members of the Areopagos itself, taking revenge for the way Ephialtes had dishonoured their council and ridden roughshod over its traditional rights and privileges.69 No wonder this same Areopagos—its powers clipped but its composition, thus far, essentially unchanged—had never convicted anyone of his murder! How far was it living up to the ideals Athena is made to set for it? [10] Œ æH ¼ŁØŒ F ıº ıæØ , ÆNØÐ , O Łı , ø o æ KªæŪæe çæ æÅ Æ ªB ŒÆŁÆ ÆØ. (704–6)
Untouched by thought of gain or of private interest? Demanding or deserving reverence? Sharp in anger against all wrongdoers? A wakeful sentinel over those who sleep? I suggest that words like these, though all will have taken them as a fine statement of what the Areopagos council as a homicide tribunal ought to be, were also designed to be taken by many spectators as a stinging definition of just what the current Areopagos council was not. I have, of course, been assuming that the controversies aroused by Ephialtes’ enactments, and by his murder, were still topical in 459/8. Even if we suppose that the murder occurred in 462/1 very shortly after the passage of the legislation,70 some three years before the 69 I am using this sort of language because it is scattered all over the Oresteia; see Sommerstein (1989b) on Eum. 95, 110, 150, 780. [17] 70 The only evidence for the date of Ephialtes’ death is the statement of the Ath.Pol. that it occurred ‘not long’ after his reforms were carried.
Stasis, assassination, and the Oresteia
161
Oresteia was produced, I see no reason to believe that it had ceased to be topical, and plenty of reason to believe that it had not. Quite apart from the well-documented tendency of Athenian politicians at all times to rake over the embers of old disputes, we know71 that in 458 or 457 there were Athenians who were prepared—as Isagoras, the rival of Kleisthenes, had been half a century before—to call in Sparta in order to ensure that the Athenian state did not fall too completely into the hands of the demos. We know too that Aeschylus was alarmed enough about the threat of violent civil strife to include in the final portion of Eumenides two passages containing explicit warnings against it. I argued in my commentary, following Dodds,72 that one of these (858–66), which has sometimes been suspected of being interpolated, was in fact a last-minute addition by Aeschylus; arguments of some force have since been brought forward by Christopher Carey73 for the view that it is not an insertion at all but was composed at the same time as its context. It remains true that the passage will draw attention to itself by its vivid language and the disproportionate length of the speech containing it (over half as long again as any of Athena’s other speeches in the scene). The later and undisputed passage on the same subject (976–87), sung by the chorus, while difficult of interpretation at some points, undoubtedly makes special reference to the danger of tit-for-tat killings (980–3, esp. 982 I Øç ı). So far as we know, no such killings took place. The projected coup d’e´tat was rendered abortive, and in the Tanagra campaign74 the Athenians and their allies inflicted just enough damage on the Spartan-led forces to deter them from further adventures. Two months later the Athenians conquered Boiotia—and almost the first thing they did was demolish the walls of Tanagra.75 There was a good strategic reason for this: Tanagra was the nearest Boiotian city, except Plataia, to the Attic border, and could make a dangerous base in hostile hands. Politically, though, the Athenians were guaranteeing themselves the enmity of
71 72 73 74 75
Thuc. 1.107.4–6. Dodds (1960: 23–4 ¼ 1973: 51–2). Carey (1990). Thuc. 1.107.5–108.2. Thuc. 1.108.3.
162
Stasis, assassination, and the Oresteia
virtually the entire Tanagraian people. Were they, one wonders, also motivated by the desire to avenge the act of Aristodikos upon the city to which he belonged? [11] About this same time there was what can be seen as a parallel, if gentler, settling of accounts on the home front. It was in 457 that the nine archonships were opened to members of the third of the four Solonian property classes.76 The two top classes, consisting essentially of horse-owners, must have been so small that while the archonships, and membership of the Areopagos, remained confined to them, almost a hereditary privilege.77 No longer. Of course in the short run the reform had little effect on the composition of the council; but it did mean that no member of it could henceforth expect his son to have more than a small chance of joining or following him in its ranks. The opponents of Ephialtes and Pericles now either withdrew from politics or sought to work, rather than subvert, the democratic system. The struggle which had begun, one might say, with the ostracism of Themistokles, and which had become a crisis in 462/1, was for the time being at an end: a struggle of whose later phases the Oresteia is a contemporary document, and to all intents and purposes the only one we have.78
76
The zeugitai, Arist. Ath.Pol. 26.2. When the Athenian cavalry was first established, some time in the mid fifth century, it was only 300 strong (And. 3.5). Not all citizens of military age and appropriate wealth can have actually served in the cavalry; some would have conflicting responsibilities (e.g. as generals, taxiarchs, or trierarchs) and some would be unfit for service. Making generous allowance for these we might assume that in Ephialtes’ time there were some 750 pentakosiomedimnoi and hippe¯s aged 20–49. On the population model employed by Hansen (1985: 12), this would imply that 25–30 men in these classes reached the age of 30 annually and so joined the pool of potential archons. Some of these would be disqualified for office by reason of crime, public debt, scandalous life-style, etc., so it may well be that any pentakosiomedimnos or hippeus who was qualified had close to a 50–50 chance of eventually gaining a seat on the Areopagos council. 78 This chapter was first published in J. H. Molyneux (ed.), Literary Responses to Civil Discord (¼ Nottingham Classical Literature Studies 1) (Nottingham: University of Nottingham, 1992) 1–17. I wish to thank all who contributed to the discussion of this paper at and after the Nottingham seminar on 20 May 1992 <when it was originally presented>, especially Paul Cartledge, John Rich, and my respondent, Pat Easterling. 77
Stasis, assassination, and the Oresteia
163
ADDENDA p. 145 n. 11 I would now, with H. L. Ahrens, regard Cho. 285 as an interpolation, probably from another Aeschylean play. p. 148–9 n. 24 on the storm and shipwreck see now Marshall (2003); Chapter 17 below; and my discussion of Sophocles’ Nauplios Pyrkaeus in Sommerstein, Clark, and Talboy (forthcoming).
9 The tangled ways of Zeus I chose this title before I was told that it was particularly appropriate to this production in ways that I will not now reveal. I got it from another play of Aeschylus, The Suppliant Maidens, which was produced five years before Agamemnon. In that play, right at the beginning, the chorus sing that ‘the desire of Zeus is not easy to hunt out, for the ways of his thought stretch dark and shady, difficult to make out’ (92–5). That is an idea which Aeschylus seems to like his choruses to express at the beginning of a tragedy. In another, earlier play, The Persians, the chorus of Persian elders say much the same thing (93–100): ‘The guileful deception of god—what mortal man can escape it? Who with light foot is master of a swift enough leap? For Ruin first fawns on mortals, with apparently kind intent, and then leads them astray into her net.’ In Agamemnon, the chorus, near the beginning of the play (160– 83), sing what has often been called a ‘Hymn to Zeus’, in which among other things they say, ‘Reckoning up everything, I cannot compare anything to Zeus if one wants truly to cast off the vain burden of worry’ (164–7). So, if one want to set one’s mind at rest, the only hope for doing so is Zeus; but one question that you may find yourself asking is: is even Zeus a sufficient hope? Let us look at the context. The chorus, the elders of Argos who have stayed behind while Agamemnon and the young men of the city have gone to fight the Trojan War, are recalling how the expedition first set out, and how an omen appeared to the two kings, Agamemnon and his brother Menelaus. The army’s professional prophet, Calchas, interpreted this omen to mean that the war would in the end be successful, but that there was a danger that the
The tangled ways of Zeus
165
goddess Artemis would demand a sacrifice that would bring disaster to someone (he doesn’t make it very clear to whom—but the audience know that it is Agamemnon who will suffer). Then comes the ‘Hymn to Zeus’; and then, directly afterwards, we hear how Artemis is indeed demanding a sacrifice—demanding that Agamemnon sacrifice his daughter, or else the expedition will not be allowed to set sail. We have also already heard quite a bit about Zeus from this same chorus. They have told us (40–67) that Agamemnon and his brother had been sent by Zeus, in his capacity as the god of hospitality, to punish the offence against hospitality committed by the Trojan Paris when he abducted Helen, the wife of Menelaus, from the palace he shares with Agamemnon. The punishment of Paris will also, as the chorus go on to say—and the audience know they are right— involve the death of many Greeks; but nevertheless we now know that in some sense Agamemnon is the agent of Zeus for punishing Paris. And as he sets out, Zeus sends an omen to encourage him; but it appears that this omen itself angers Artemis. The omen consisted in two eagles swooping down on a hare and killing and eating it. But the hare was pregnant, and the eagles ate the unborn leverets as well; and Artemis as protectress of young wild creatures is angry (134–43). And apparently, since she is unable to get back at Zeus, she decides to get back at Agamemnon, on the principle of ‘if I can’t kick you, I’ll kick your dog’. There was a strong tradition, centuries old when Aeschylus wrote this play (it goes back at least to the Cypria, a lost epic which told the story of the origins and early years of the Trojan War),1 that the reason why Artemis demanded the sacrifice of Agamemnon’s daughter was that Agamemnon had committed an offence against her, for example by killing a stag in a sacred place and/or boasting that he was a better hunter than Artemis—both of them, of course, acts that no sensible person would do. There is nothing like that in Aeschylus’ treatment, no mention of anything that Agamemnon has done wrong—and yet Artemis demands that he must either abandon the expedition, accepting the terrible loss of face involved, or else
1
Cypria Arg. }8 West; cf. Soph. El. 566–72.
166
The tangled ways of Zeus
sacrifice his child. Whichever he chooses, he is going to suffer. Whichever he chooses, also, Zeus is going to suffer, at least in prestige. If Agamemnon abandons the expedition, Paris will not be punished, and Zeus will have failed in his function of overseeing the laws of hospitality. If, on the other hand, Agamemnon performs the sacrifice—well, we know the outlines of the story, we know what is going to happen to him: the mother of that daughter will kill him in revenge—and that will harm Zeus too, because Zeus is not only the protector of hospitality, he is also, as king of the gods, the patron of human kings.2 That is where we are at the moment of the Hymn. Zeus has put Agamemnon into an impossible dilemma. Why has he done so? Is there a way out? I am raising these questions; I am not necessarily going to answer them, because I don’t know whether you’ll think that Aeschylus answers them, at least within the bounds of this play (which, as you know, is only the first of a three-part series). The Hymn itself has three stanzas. In the first, as I have mentioned, the chorus say in effect that there certainly cannot be a way out of this impasse unless Zeus can provide one—because (as they go on to say in the second stanza) he has eliminated all his rivals, and eliminated them so completely that the chorus don’t even mention their names (nor do they mention that one of them, whom Zeus has imprisoned, was Zeus’s own father). And then they sing a very difficult and ambiguous little stanza (176–83): Zeus, who made a road for mortals to acquire wisdom, who laid down to hold good the law ‘learning by suffering’. Pain, which reminds one of grief, drips before the heart instead of sleep, and good sense comes even to those who do not want it to come. There is, somehow, a violent favour done by the gods who sit on the august bench of command.
2 A point to which attention has already been drawn in Ag. 43 (‘twin thrones and twin sceptres given by Zeus’) and 114–15 ‘the king of birds [the eagle, the bird of Zeus] appearing to the kings of ships’).
The tangled ways of Zeus
167
What does all this mean? How is it relevant? To say that good sense comes ‘to those who do not want it to come’ has point only if someone has been, or will be, acting contrary to good sense: who, we must ask, may this be? From these reflections, we move directly, without a change of rhythm, into a long narrative showing Agamemnon coming, not without hesitation but eventually with firm decision, to the resolve to sacrifice his daughter. As he says (206–11), either alternative will be bad for him. But whose fault was it that he had the dilemma in the first place? You may think that I’m hinting that it was Zeus’s fault. Well, maybe; but let’s not jump to that conclusion yet. Why does Agamemnon sacrifice Iphigeneia? He knows it is a terrible thing. In fact, the chorus take the trouble to mention that, when they first hear from the prophet what Artemis is demanding, both he and his brother shed tears (202–4). Then, in a reported soliloquy, Agamemnon is heard considering various arguments on both sides, and the one that he mentions just before coming to his decision—presumably, then, the one that tips the balance—is this: ‘how can I become a deserter of my ships and lose my allies?’ (212–13). In other words, if he gives up the expedition, his loss of face with his allies (who comprise almost the whole of Greece) will be so great that they will not be willing to follow him in future: if you like, he will lose his hegemony. And rather than do that, he thinks it is better for him to sacrifice his daughter. Little does he know! Why is Agamemnon embarking on this war, which, even at such a cost, he is not prepared to give up? On the divine level, the purpose of the war is to punish Paris, and it is the will of Zeus that it should be waged; but there is no sign that Agamemnon is even aware of this. So far as he is concerned, the war is being fought in order to get back Helen and also to gain glory and booty. But if you ask ‘who has the greatest interest in getting back Helen?’, the answer must be her husband Menelaus, and we know what Menelaus’ reaction was to the proposal to sacrifice Iphigeneia: he shed tears, just like his brother. Even he, therefore, doesn’t want Helen back if this is the price that has to be paid. The chorus themselves, as they will make clear to Agamemnon at the moment of his triumphant return (799–804), thought, at least at the time, that to make war for such a cause was stupid anyway.
168
The tangled ways of Zeus
On the human level, therefore, this is a bad war, fought for no adequate reason, and certainly not worth a human sacrifice, let alone the sacrifice of a young girl by her father. What of the divine level? Zeus, we know, wants the war to be waged, in order to punish Paris; yet, as Calchas’ interpretation of the omen shows, he knows that the war will be long-drawn-out and therefore very costly in lives to both sides, and the omen itself, which Zeus caused to happen, has led Artemis to make a demand whose consequence must be either the abandonment of this expedition so important to Zeus, or else the eventual murder of its leader. What follows? One of three things. One possibility is that Zeus is to be thought of as being hostile or jealous or indifferent towards human beings. That is perfectly possible, from an ancient Greek point of view: plenty of Greeks, we know, thought just that, including the highly-regarded didactic poet Hesiod3 and, for that matter, a large proportion of the characters in Homer’s Iliad. But if we are really meant to think that, it is rather strange that the chorus describe Zeus as the god ‘who made a road for mortals to acquire wisdom’. Secondly, it is possible that Zeus is to be understood as having other reasons, independent of those we have already heard about, for wanting to destroy Agamemnon. This has been a very popular solution of the riddle,4 because later on we do hear, from Cassandra and later from Aegisthus, the story of how Agamemnon’s father Atreus, in the course of a quarrel with his brother Thyestes, treacherously invited Thyestes to a feast and there served him with the flesh of his own murdered children—and Aegisthus, the partner (both in bed and in murder) of Agamemnon’s wife Clytaemestra, is Thyestes’ surviving son. The trouble with this solution is that not only is this story of Thyestes’ feast never mentioned until nearly two-thirds of the way through the play (1090), but it is also never anywhere, in Agamemnon or in either of its sequels, linked in any way to the killing of Iphigeneia. If we are going to make this connection, we must make it without the least help from the poet. That leaves us with the third possibility, which may seem a surprising one: that Zeus does not know what is best for human beings, 3 4
Hes. Thg. 550–613, Works 42–105. See for example Lloyd-Jones (1962).
The tangled ways of Zeus
169
including his own prote´ge´s, and doesn’t even know what is best for himself. Dare we claim this? Maybe we can. After all, most ancient Greeks agreed most of the time that it’s a pretty rotten, stupid world; they also believed that this world was ruled by Zeus; from which it seems to follow that Zeus must be pretty rotten and pretty stupid. Certainly one can say that in this play, as far as it goes—and I remind you again that it isn’t the end of the story—the will of Zeus looks as though it is creating an utter shambles, both in the current colloquial sense and in the literal sense of what Cassandra calls a house of slaughter. The play ends not only with the death of Agamemnon, but also with that of another innocent young woman—a young woman, incidentally, who has been extremely badly treated by Apollo who, you may recall, is the twin brother of Artemis (there are some very tight coils and knots in the structure that Aeschylus has created); and for the city of Argos, where all this is taking place, it ends with the establishment of a usurping tyranny. And towards the end, when the chorus are reflecting on all this, they say it has all come about through Zeus, the cause of all, the doer of all; for what happens to mortals without Zeus? what of all this is not caused by god? (1485–8)
Nobody within the play answers this question. The only thing of which we can be certain is that these killings will in their turn be avenged: that is predicted by Cassandra (1279–84) and confidently anticipated by the chorus (1646–8, 1667). But avenged how? By Orestes, the son of Agamemnon, another innocent young person— innocent until he is forced to become the killer of his mother, partly by circumstances, partly by a command from Apollo backed by the threat of the most ghastly sufferings imaginable in case of disobedience.5 In short, as I have said, what we are shown in Agamemnon is a rotten, stupid world, and one almost finds oneself wondering, at the end of the play, if the ‘good sense’ to which Zeus has ‘made a road’ is nothing but the recognition that it is a rotten, stupid world.
5
Cho. 269–305.
170
The tangled ways of Zeus
But, of course, the end of this play is not the end of the story. I am reminded of an anecdote—not an ancient anecdote; at least H. D. F. Kitto, who recounted it in a lecture at Berkeley in 1960 and published it six years later in his book Poiesis,6 said that unluckily he had lost his reference to its ancient authority. He was actually talking about a different Aeschylean production—The Suppliant Maidens and its lost companion plays—but the point of the story applies so well to the Oresteia too that I shall exercise the privilege always claimed by ancient poets and alter the myth (and one bit of the dialogue) to suit my point. Kitto’s anecdote, then, as modified, concerns a man from Acharnae, in the countryside, who had brought his son to Athens to see the first production of the Oresteia. As they were trudging homeward through the gathering dusk, having set out before dawn, the boy suddenly asked, ‘Father, is Aeschylus a wicked man?’ The father smiled. ‘Why do you ask that?’ ‘Well,’ said the boy, ‘he showed us this appalling story, with innocent people being slaughtered all over the place and then that perfectly decent young fellow being given no choice but to kill his mother, and he made it quite clear that we were to understand that Zeus was behind it all. Is Aeschylus trying to tell us that Zeus is an evil being, and that it is no use to pray to him?’ ‘Ah,’ said the father, ‘that’s what comes of falling asleep half-way through a trilogy.’ On this occasion we shall have to fall asleep little more than a third of the way through, because the very good thing that you’re just about to see will then come to an end; but it will be continued on another occasion, and I, like the Actors of Dionysus, hope very much that you will see the continuation then, and that when you have seen it you will take steps to discover for yourselves how the whole story finally concludes. Like the audience in Athens in the spring of 458 bc, you may have a few surprises in store.7
6
Kitto (1966: 52). This chapter, which originated as a pre-performance talk given before a production of Agamemnon by the Actors of Dionysus, was first published in Dionysus (York) 13 (2000) 26–9. I am most grateful to the Actors of Dionysus for giving permission for this republication. As in the 2000 publication, I have left references to the original occasion of delivery unchanged, though the present version contains a few additional footnotes. 7
10 The omen of Aulis or the omen of Argos? The lyric section of the parodos of Aeschylus’ Agamemnon begins with a narrative (104–59) recounting an omen that appeared at the time when the expedition against Troy was being set in motion, and the interpretation of it by the prophet Kalchas, which includes a prediction of the capture of Troy (126–30; understood by everyone) and foreshadowings of the sacrifice of Iphigeneia (134–52; understood by the audience, and no doubt with hindsight by the chorus, but by no one at the time when it was uttered) and of the murder of Agamemnon (154–5; understood by the audience alone). Does the text enable us to say where the omen appeared? This question might seem hardly to be worth asking, but I shall show presently that our answer to it does make a substantial difference to our understanding and appreciation of the parodos. Meanwhile, here is a selection of statements on the matter from scholarly literature of the last forty years:1 The Chorus describe a portent, sent to the Atridae at Aulis, and Calchas’ interpretation thereof. (Denniston and Page 1957, on 104–257.) In Aulis the army saw two eagles eat a pregnant hare2 and its unborn <[sic]>. (Lesky (1972) 1983: 72.)
1 My (far from exhaustive) sampling of the literature over this period has revealed several authors who refrain from stating where the omen was seen, but only one who (like all the rest, without argument) explicitly adopts the view advocated here. This is Scott (1984: 34): ‘The old men recall the beginning of the Trojan expedition: the departure of the commanders, the omen that met them as they left the palace, and Calchas’ interpretation of the omen.’ 2 The English translation actually says ‘rabbit’, but Lesky wrote, correctly, ‘Ha¨sin’.
172
The omen of Aulis or the omen of Argos?
The juxtaposition of helplessness and brutality . . . is repeated, with greater force, in the omen which occurred while the Greek fleet [88] lay becalmed3 at Aulis. (Vickers 1973: 350.) The Iliad begins again, in effect, with the omen sent to Aulis. When the Argives marshalled under Agamemnon and Menelaus, two eagles swooped down on a pregnant hare . . . (Fagles and Stanford 1977: 24–5.) The Chorus . . . sing of how the fleet was mustered . . . at Aulis, and how a portent revealed to the prophet of the army that it was destined to triumph, but at grievous cost. (Lloyd-Jones 1979: p. xii; cf. his note on 116.) The imagery . . . recurs almost immediately with the chorus’ narration of the . . . appearance at Aulis of the eagles that devoured a pregnant hare. (Garner 1990: 29.) At Aulis, the expedition is sent on its way by an omen . . . (Goldhill 1992: 61.) We hear [in Ag. 825–8] echoes of the omen at Aulis, where eagles devoured the young of the pregnant hare. (Rehm 1992: 84.)
On what are these confident assertions based? The chorus actually make only one statement about the location of the events described in 104–59, namely that the eagles appeared YŒÆæ ºŁæø (116); and Fraenkel (1950: ad loc). correctly says that ‘of the geographical position of the ºÆŁæÆ we are here told nothing at all’. He clearly believes, however, that the location is Aulis, and points out that in Euripides’ Iphigeneia at Aulis the ‘lodging of the Atridae’ is frequently called ºÆŁæ , ºÆŁæÆ, or ÆÆ,4 so that the use of the word does not necessarily imply ‘a fixed royal palace’. [89] Granting this argument, it begs a more important question. If the location is meant to be Aulis, how are the audience supposed to know this? We, reading the text today, know that the story of the omen is part of a structure whose climax will be the sacrifice of Iphigeneia. The original audience did not know this. As many scholars have emphasized, the link between the omen and the sacrifice is almost certainly an innovation by Aeschylus—and a very 3 These six words are of course in direct contradiction to the text, in which the omen causes Kalchas to fear that the fleet may be becalmed. 4 Eur. IA 440, 612, 678, 685, 820, 854. It makes a difference there, however, that theatrically speaking the reference is to the skene which in other plays regularly represented the houses of kings and nobles. Similarly the skene can be called ÆÆ, Ø, or ºÆŁæ when it represents the hut of Ajax (Soph. Aj. 73, 80, 305) or even the cave of Philoktetes (Soph. Phil. 1453); it does not follow that these abodes could have been so called if they had not been embodied by that solid, visible structure.
The omen of Aulis or the omen of Argos?
173
puzzling innovation many critics have found it. An audience seeing and hearing the play in real time, for the first time, can have had no possible notion that such a link was going to be made until Kalchas mentions the resentment of Artemis at 134. Up to that point they cannot have known that the parodos would have anything whatever to do with events at Aulis. They will therefore have interpreted YŒÆæ ºŁæø in the light of its context and of the play so far. And interpreted in that light, it will certainly not be taken to refer to the C.-in-C.s’ billets at the port of embarkation. The only house of which we have heard so far has been the palace of the Atreidai at Argos. It has been visible to us throughout; it may have seemed to speak the first words of the play, if, as is likely, the Watchman was at first in a recumbent position;5 it is explicitly mentioned five times in his prologue (3, 18, 27, 35, 37), and just before the passage with which we are concerned the chorus have been apostrophizing Klytaimestra who is within the building (83–103).6 When the spectator hears of an omen which appeared Æغ FØ
H . . . YŒÆæ ºŁæø , how on earth can he be expected to put aside all this and think of a temporary residence at an Aulis to which no reference has yet been made?7 [90] If, on the other hand, the spectator assumes, in harmony with what has preceded, that the omen appeared to the two kings just as they were setting out8 from their palace on their journey (cf. ‹Ø
5
Cf. Sommerstein (1996a: 218 with n. 8). See Taplin (1977: 280–5). 7 It is curious that of the authors whom I have cited above for the view that the omen appeared at Aulis, the only two who have published translations of the play— Fagles and Lloyd-Jones—have both, in common with most other translators, rendered ºŁæø as ‘palace’. Fagles (and Stanford) make no comment on the apparent contradiction; Lloyd-Jones in a note on the passage explains, following Fraenkel, that ‘ “. . . the palace” simply refers to wherever the kings are staying’ (in that case why use the word ‘palace’, which in English means, in relation to royalty, only an edifice permanently dedicated to royal use, instead of some expression like Fraenkel’s ‘abode’ which would render the supposed sense accurately?). 8 A moment of exceptional augural significance; compare the eagle-omen seen by Telemachos just as he is about to drive away from the palace of Menelaos [!] in Odyssey 15.160–81, and that seen by Priam when about to depart on his mission to Achilles in Iliad 24.314–21. In less exalted contexts too, special ominous significance could be attached to persons or objects encountered immediately upon leaving one’s house; cf. Ar. Frogs 196 (fiH ı ıå KØ ), Luc. Pseudolog. 17. 6
174
The omen of Aulis or the omen of Argos?
104, Ø 111, Œº ıŁ 126, ›ø 157) to Aulis and the war, what follows will also make sense in the light of this. When Kalchas, interpreting the omen, perceives its possible sinister implications, he hopes that no divine resentment cast its dark shadow upon the great instrument forged to clamp Troy,9 smiting it in advance when it has been mustered (æÆøŁ ). (131–3).
The participle æÆøŁ is picked out by the metre, the last word of its sentence, enjambed after a pause;10 thus it is given a degree of emphasis, pinpointing the time of danger as the time when the whole army has been gathered together (sc. at Aulis),11 in contrast with the time at which Kalchas is speaking, when Agamemnon and Menelaos are leaving Argos with, presumably, only the [91] Argive contingent.12 If the united army is already in being, æÆøŁ is redundant and its emphatic placement pointless. 9
More literally ‘the great bit for Troy’s mouth’ (Lloyd-Jones). The sentence so far (r Ø ¼ªÆ Ł Ł Œ çfi æıb Ø ªÆ "æÆ) has been coextensive with a dactylic octameter catalectic, and æÆøŁ
begins a new colon which starts in a different rhythm (iambic) before reverting to dactyls. Wherever such iambo-dactylic cola appear elsewhere in this song (108, 116, 126) they are preceded by a clear pause in grammar and sense. Here, moreover, the sentence is grammatically complete without the addition of æÆøŁ . 11 How is it possible for Kalchas to deduce from the omen that this is the time of danger, and that Artemis’ retaliation is likely to take the form of adverse winds that will prevent the fleet from sailing (148–50)? The only answer suggested by the text is that the hare was prevented from running (ºÆ Æ ºØŁø æ ø 120 codd.) by the action of the eagles, which makes it appropriate, on the principle æÆ Æ ÆŁ ØÐ
which pervades the whole Oresteia, that any reprisal should also have the effect of preventing locomotion. 12 Of course the description of the Atreidai in 109–10 as the commanders of ‘the Achaians’ and of ‘the youth of Hellas’ does not entail that they actually had the whole Achaian/Hellene army with them at the time; it was not to the army but ‘to the kings of ships’ that the ‘king of birds’ appeared (114–15), and it is the commanders who are described as being ‘sped on their way’ by the omen (111). So too the whole episode is summed up at the outset (104) as ‹Ø Œæ ÆYØ , ‘the command(ers) who received a favourable omen on/for a journey’, and in 125 the eagles are called f IæåA, ‘[those who] gave a send-off to the commanders’ (Denniston-Page). It is more than possible that the scene we are meant to imagine is not that of the Atreidai marching away from the palace at the head of the Argive troops, but that of the two kings themselves, accompanied only by their personal staff (including Kalchas) and servants, coming out of the palace to go and join an (Argive) army that had mustered at a convenient parade-ground outside the city walls. 10
The omen of Aulis or the omen of Argos?
175
Then at the end of Kalchas’ speech our attention is drawn back to the royal house: ‘ . . . Ø ªaæ ç æa ƺ æ NŒ ºÆ, ø B Ø Œ Ø .’ Ø ˚ºåÆ f ªºØ IªÆŁØÐ KŒºÆª
æØ ’ I’ Oæ Łø ›ø YŒØ Æغ Ø. (154–7)
The omen appeared ‘near the house’, and its prime significance is for the house—in which, when Agamemnon and Menelaos departed from it,13 Klytaimestra remained ( Ø), and for which the consequences of the omen will issue in numerous violent deaths (cf. æØ ’). Between the omen and the beginning of its fulfilment is placed [92] the famous ‘Hymn to Zeus’;14 and after that ends we at last (184ff.) begin to hear of events at Aulis—though even now it is not till the fifteenth word of the sentence (IºÆ: 188) that it becomes clear that this is the episode being narrated, and not till 190–1 that the place is actually specified. On the interpretation here advanced, the specification of place in 190–1 can be taken straightforwardly as indicating that the action has now moved to Aulis from Argos.15 13 The account of their departure will find an echo in the narrative, in the next choral song (404ff.), of the departure of Helen (and Paris) from the same house; that too was accompanied by words of foreboding from prophetic lips (410–26, spoken by the ø æçBÆØ). It is striking that whereas the retrospective choral songs in Agamemnon thus dwell on past departures from the palace of the Atreidai, within the play itself no one ever comes out of the palace alive except (five times) Klytaimestra; the ‘watchdog of the house’ (607), like some Kerberos, keeps guard over what Kassandra calls ‘the gates of Hades’ (1291), letting others in only as she chooses (cf. Taplin 1977: 300, 307) and letting no one out at all. 14 Which thus in effect accompanies the journey of the Atreidai from Argos to Aulis—a journey which we can imagine, after hearing Kalchas, as being beset by anticipations both of certain ultimate triumph and of possible, mysterious disaster. The former kind of thoughts are echoed as the chorus sing of the triumphs (cf. æØÆŒBæ 171–2, KØ ŒØÆ 174) of Z f Æغ (cf. 355); the latter in the third stanza of the Hymn (176–83), in phrases such as Ł Ø Ł, ÅØ ø
, and åæØ ÆØ. 15 Fraenkel (1950, on 190–1), committed to the view that the scene had been Aulis ever since 104, attempts to explain the fact that Aulis is mentioned only here by asserting that ‘it is characteristic of Aeschylus’ way of telling a story, and of early narrative technique in general . . . that no definite geographical place is indicated until the story has reached the point at which the counter-currents in the Euripus and the north wind . . . have become of immediate importance’. But all we get by way of
176
The omen of Aulis or the omen of Argos?
Then follow Kalchas’ announcement that Artemis requires the sacrifice of Iphigeneia; Agamemnon’s soliloquy and decision; the sacrifice scene;16 and some concluding reflections, after which, as the parodos at last ends, the ç æa ƺ æ NŒ ºÆ, ø B Ø Œ Ø appears at the door of the rŒØ ƺ ØØ. One may well wonder why it has so often been taken for granted, in defiance of the plain implication of the text, that the [93] omen appeared at Aulis. The reason is not far to seek. The narrative has been unconsciously contaminated by recollections of another famous narrative about an omen that did appear at Aulis, to the whole army, and was interpreted by Kalchas. In Iliad 2.299–330 Odysseus calms the rebellious Achaians by telling them the tale. They are all witnesses to what happened when the Achaian ships were gathered at Aulis. They were sacrificing hecatombs under a plane-tree by a spring, when a snake leaped on to the tree and devoured eight nestling sparrows; the mother-bird fluttered about in distress, but the snake caught her by the wing and devoured her also. Kalchas interpreted this as an omen from Zeus: as the snake had consumed nine birds, so the army would fight against Troy for nine years, and take the city in the tenth.17 It is, of course, clear that the Aeschylean omen is designed to put its audience in mind of the Homeric one. It corresponds in three crucial features: the omen comes from Zeus; it involves the destruction of wild creatures, mother and young together; and it is parallels are the late placement of ÆæÆAØ in Ag. 59 and (in Fraenkel’s Appendix A (1950: iii. 805)) Hdt. 1.110–12 where the reader is kept in the dark as long as possible about the fact that the baby born to Cyrus’ foster-mother was a stillbirth; no parallel is offered for what Fraenkel assumes here, a specification of locality coming so late in the story that it gives the impression that the locality has changed when it has not. It clearly never occurred to Fraenkel for a moment that the location of the omen could be anywhere other than Aulis; he revealingly adds that ‘of course [emphasis mine] the audience, being familiar with the events, must have identified the locality long before’. 16 Here too, references to the royal house continue: Agamemnon speaks of his daughter as ø ¼ªÆº Æ (208) and has her gagged to suppress any çŁªª IæÆØÐ
YŒØ (237), and her last thoughts are of how she had often sung at banquets in her father’s I æH (243–7). 17 The omen and its interpretation also figured in the Kypria (cf. Proklos Kypria 44–6 Davies <¼ Kypria Arg. }6 West >), where, however, it was placed before the first sailing of the fleet and its mistaken attack on Teuthrania (of which the Iliad knows, or professes to know, nothing).
The omen of Aulis or the omen of Argos?
177
interpreted to portend that Troy will fall, though only after a long time (åæ ø fi Ag. 126).18 But there are also obvious and substantial differences, and in view of these there would be no reason for a hearer to adopt a recherche´ interpretation of a phrase like YŒÆæ ºŁæø , in preference to a straightforward one, merely because the recherche´ interpretation would bring the story closer to that told in the Iliad. Rather, here as in so many other respects,19 Aeschylus has taken an episode well known from earlier poetry and remodelled it to suit his own purposes. The omen appears, not to the army at Aulis, but to the kings at Argos, and accordingly the snake is replaced by Zeus’ royal eagle (or rather eagles); the episode is ingeniously brought into relation with the originally quite separate [94] story of the anger of Artemis and the sacrifice of Iphigeneia; and whereas in Homer the interpretation of the omen was simple, favourable, and addressed to the army, in Aeschylus it is double, the second limb being decidedly sinister, and concerns not the army (which is indeed going to suffer, but for reasons that have nothing to do with the omen) but the Atreidai and their kin. The parodos, like the prologue, puts us clearly on notice that this is going to be the tragedy of an oikos.20 18 In the Aeschylean version it is presumably to be understood that Kalchas deduces this from the fact that the hare was almost ready to give birth (ÆPŒ
æe ºåı 136). 19 See Prag (1985), esp. 79–84; March (1987: 79–98); Sommerstein (1996a: 190–204). 20 This chapter was originally published, under the title ‘Aesch. Ag. 104–59 (The omen of Aulis or the omen of Argos?)’, in Museum Criticum 30/31 (1995/6) 87–94. I am most grateful to Pacini Editore for giving permission for this republication.
11 Pathos and mathos before Zeus It was Zeus—so sing the chorus of Aeschylus’ Agamemnon (177–8)— who laid down (Ł Æ) the law Ł Ø Ł. This is not the place to investigate the exact meaning of that brief phrase or its significance for the Oresteia as a whole;1 though it cannot be too often emphasized, firstly that Ł Ø Ł is not the same as ÆŁ Æ ÆŁ E and does not imply that the person who learns is necessarily the same as the person who suffers,2 and secondly that despite the efforts of a long succession of translators to confine the application of the law to mortals3 to the exclusion of gods, nothing in its actual expression 1 On these matters see Kitto (1956: 6–8, 69–86); Conacher (1987: 11–12, 83–5); and Sommerstein (1989b: 19–25). 2 Cf. Booth (1976: 228). 3 So Thomson (1938) ‘man shall learn by suffering’; Murray (1952) ‘Man by Suffering shall Learn’; Vellacott (1959) ‘man must suffer to be wise’; Fagles (1977) ‘we must suffer, suffer into truth’; cf. Fraenkel (1950) ‘by suffering they shall win understanding’ and Lloyd-Jones (1979) ‘by suffering they shall learn’, both of whom make it clear in their notes that they mean ‘they’ to refer back to ‘men’ (¼ æ 176); Grene and O’Flaherty (1989), while translating 177 accurately enough as ‘knowledge comes in suffering’ nevertheless impose the time-honoured interpretation by speaking of Zeus as ‘having set this law firmly in our hearts [emphasis mine: AHS]’. Tony Harrison’s Oresteia (in T. Harrison 1985/1986) omits the passage altogether, along with the whole of the ‘Hymn to Zeus’ (160–83). Of the English translations I have sampled, only those of Weir Smyth <([1925] 1957)> ‘wisdom cometh by suffering’ and Lattimore (1953) ‘wisdom comes alone through suffering’ avoids, as the Greek does, any attempt to identify the sufferers <with> the learners either directly or indirectly. The German renderings collected by Neitzel (1980: 283 n. 3) show a greater tendency to circumspection: only four out of eight interpolate a reference to humanity. On Neitzel’s own interpretation of the passage (retaining the manuscripts’ reading fiH Ł Ø, and taking the sense to be ‘bringing it about, by means of suffering, that learning is effective’) I agree with Conacher (1987: 94 n. 21).
Pathos and mathos before Zeus
179
implies any such restriction. The main purpose of the present note, however, is to raise anew the question: if the law Ł Ø Ł was established by Zeus, what then was the relationship between ‘suffering’ and ‘learning’ under his predecessors, Ouranos and Kronos, of whom we have been reminded in the preceding stanza (168–73)? I say ‘to raise anew’ because the question is one that has been asked before. It seems to have been first raised by Kitto:4 What Aeschylus says about [Zeus] is that he brought a new law, Learn by suffering. How was this new? We cannot imagine that under his predecessors men learned without suffering; Aeschylus did not believe in a past Golden Age. The only interpretation is that under the earlier gods men suffered but did not learn; nothing came of hard experience.
Some years later the same question was discussed by Maurice Pope,5 who, without referring to Kitto, considered the same two possibilities of interpretation only to reject both: Are we to suppose [the chorus] to mean that such useful experiences as the sacrifice of Iphigenia were denied to mankind in the bad old days? The thought is grotesque. Or are we to suppose that such experiences were taking place all the time but that nobody learned anything from them? The thought is less grotesque, but it is a very complicated one to have to think. On the other hand the thought-sequence in the pessimist interpretation is natural and easy. Things have always been like this.
[110] Whether Pope’s favoured interpretation is really ‘natural and easy’ may be a matter of opinion. What is not a matter of opinion is that it will not square with the text: if you say that X ŁÅŒ the principle Y to hold as valid law (Œıæø å Ø 178), you are saying that before X did this Y was not valid law. D. J. Conacher in his rebuttal of Pope’s paper6 found no difficulty in accepting the explanation—substantially the same as Kitto’s—which Pope had
4 The quotation is from Kitto (1961: 68); earlier editions have no corresponding passage, the chapter having been rewritten for the third edition. Kitto had already stated the same conclusion in Kitto (1956: 81), but without the supporting argument. 5 Pope (1974: 111). 6 Conacher (1976: 332): ‘it is Zeus alone who put man on the road to using his intelligence and so to learn through suffering’.
180
Pathos and mathos before Zeus
considered ‘very complicated’. And there the matter has rested.7 Let us now re-examine it. Under Zeus, according to the chorus, there is suffering, and it leads to learning; there is learning, and it is the result of suffering. What was the position before Zeus? There are four logical possibilities; all of them have been canvassed in one or other of the studies mentioned above, but never all four together. The possibilities are these: 1. Things were the same as they are now (Pope). 2. There was learning, but no suffering (rejected by Kitto). 3. There was suffering, but it did not lead to learning (Kitto and Conacher; rejected by Pope). 4. There was neither suffering nor learning (dismissed by Pope as ‘grotesque’; not considered by Kitto). We have already ruled (1) out of court as inconsistent with the Aeschylean text. We can also safely reject (2), not on Kitto’s question-begging ground that ‘Aeschylus did not believe in a past Golden Age’ (which presupposes, not only that Prometheus Bound is genuinely Aeschylean, but also, less excusably, that that play rejects the idea of a past Golden Age—more on this presently—and that Aeschylus was incapable of rejecting for artistic purposes in one play a tradition which he had tacitly accepted in another) but, once again, because the text excludes it: Zeus is described in 176–7 as e çæ ØÐ æf ›Æ Æ, and it must follow that whether or not ‘suffering’ is being presented as an innovation by him, ‘learning’ certainly is being so presented. We are thus left with the choice between (3) and (4). Both of these are wholly consistent with the text of the passage: Aeschylus has given his audience no guidance as to which they should assume to be correct. They will therefore have been guided, not by his views about the primeval world, but by their own. Do we know anything of what their views were? Yes, we do; a great deal. Before the reign of Zeus came the reign of Kronos; and throughout all antiquity › Kd ˚æ ı was a
7
Neitzel (1980) does not consider the question whether learning and/or suffering existed before Zeus.
Pathos and mathos before Zeus
181
proverbial expression for a paradisal or utopian existence,8 free from every imaginable kind of suffering. Typical is the Hesiodic account (Works and Days 109–19):9 First of all the immortals who dwell on Olympos made the golden race of mortal men. They were in the time of Kronos, when he ruled in heaven. They lived like gods, having a soul free from grief, utterly without toil and trouble, nor did wretched old age come over them at all, but they always kept the form of their feet and hands, and rejoiced in feasting, remote from all evils. They died as though overcome by sleep, and all good things were theirs; the bountiful earth of her own accord brought forth abundant fruits for them, and they gladly inhabited their lands in peace, with many blessings.
[111] Later sources abundantly confirm this picture. There was universal peace10 and freedom from fear11 and from all diseases.12 Even animals did not eat one another,13 and the very climate was perpetually equable14 so that men had no need for clothing, bedding, or housing. Private property,15 slavery,16 and government17 were unknown; men were directly ruled by the gods18 with absolute justice,19 and themselves ‘partook of immortal, divine descent’20 and were not 8 Cf. [Pl.] Hipparchos 229b; Arist. Ath.Pol. 16.7; Plut. Arist. 24.3; Philodemos, On Piety 51 Gomperz <¼ B6798 Obbink>, citing inter alia the archaic epic Alkmeonis (fr. 7 Davies <¼ West>). On the history of the concepts ‘life under Kronos’ and ‘Golden Age’, see Baldry (1952: esp. 84–6); Guthrie (1957: 69ff.); Gatz (1967: esp. 114–28); Blundell (1986: 135–64). 9 Cf. also ibid. 90–2: ‘For formerly the races of men had lived on earth remote from evils and without harsh toil and painful diseases that dealt death to men.’ 10 Telecleides fr. 1.2, Pl. Polit. 272e, Laws 713e. 11 Telecleides fr. 1.3. 12 Ibid. 13 Pl. Polit. 271e. 14 Ibid. 272a; cf. perhaps Hes. fr. 204.124ff. M–W <¼ fr. 155.124ff. Most> which seems to describe a deterioration of climate at the end of the heroic age. 15 Plut. Cim. 10.7 c Kd ˚æ ı ıŁºªı Å ŒØ ø Æ . In Pl. Polit. 271e– 272a PŒ qÆ . . . Œ Ø ªı ÆØŒH ŒÆd Æø · KŒ ªB ªaæ I ØŒ the logical connection as the text stands is obscure, and one wonders whether ªı ÆØŒH
ŒÆd Æø may not be a gloss (based on recollection of the fifth book of the Republic), in which case Œ Ø would originally have referred to property in land and goods. 16 Plut. Comparison of Lycurgus and Numa 1.12; Luc. Saturn. 7, 20. 17 Pl. Polit. 271e. 18 Pl. Laws 713c–e. 19 Ibid. 713e; Plut. Mor. 266f. 20 Soph. fr. 278 Radt PÆ ƒ ª
Æ IçŁı ºÆå Ł Æ: cf. Gatz (1967: 115–16).
182
Pathos and mathos before Zeus
only better but physically bigger than the men of today.21 Most persistent of all is the idea that in the age of Kronos the earth spontaneously (ÆP Å) bestowed her produce on man in bounteous abundance without the need for toil, an idea that in fifth-century comedy repeatedly found expression in such delectable fantasies as this: Every gully ran with wine, and barley-cakes fought with loaves of bread at the portals of men’s mouths, begging to be gobbled up ‘if you like the whitest’; fish would cook themselves on the way home, and lay themselves on the table; beside the diners’ couches flowed a stream of soup rolling down hot chunks of meat, and there were channels of sauces there for whoever wanted them to give ample facilities for dunking each mouthful to be soft to swallow; and there were ground-cakes in little trays, sprinkled with seasonings; and roast thrushes with milk-cakes would fly down your throat, and there was great din of flat-cakes jostling one another to get between your jaws; and boys would play at knucklebones with savoury titbits and slices of sow’s womb.22
We can therefore take it as certain that when Aeschylus’ audience heard that Zeus, the overthrower of Kronos, had introduced the principle Ł Ø Ł, they understood this to mean that Zeus had done three things: brought suffering into the world, brought learning into the world, and established a causal relation between them. That Zeus had brought suffering into the world was, as we have seen, thoroughly traditional. That he had brought learning into the world may seem on our evidence to have been less so; at any rate descriptions of the age of Kronos are not as a rule explicit on this point, and Plato in the Politikos (272b–d) can make his protagonist float the idea that the men of those days used their ample leisure to discuss philosophy and gather wisdom. He admits, however, that this is not ‘the sort of account that is currently given concerning them’ (272c), though tantalizingly he does not tell us what were generally 21
Telecleides fr. 1.15. Telecleides fr. 1.4–14; cf. Cratinus fr. 176. Numerous other fragments, many of them cited by Ath. 6.267eff., take us to the same fantasy world without linking it explicitly (in the lines that have survived) with the age of Kronos; on this comic topos see Gatz (1967: 116–21); Heberlein (1980); Sutton (1980: 58–65); Zimmermann (1983: 59–61); Reckford (1987: 323–8).* 22
Pathos and mathos before Zeus
183
supposed to have been the topics of the conversations men used to have with each other and with animals. But the best elucidation of the degree of men’s intellectual ability and achievement in the age of Kronos probably comes from [112] Prometheus Bound. For our present purpose it matters little whether this play is or is not by Aeschylus;23 it is at any rate a fifth-century Attic drama, written for an audience whose background beliefs were the same as those of the audience of the Oresteia or of Telecleides’ Amphiktyones. It is customary to regard the ‘anthropology’ of Prometheus Bound as being based on the idea of progress from primitive and miserable beginnings, in sharp contrast with the degenerative anthropology first fully articulated by Hesiod. It should, however, be remembered that the sorry state from which Prometheus claims to have rescued humanity (Prom. 442–57) was not their state during the reign of Kronos, but early in the reign of Zeus. We are told nothing directly of how men fared under Kronos, but we do hear at some length (406– 24) of how the whole world is grieving for the fate of Prometheus and lamenting for the ancient honours held by him and his brethren (ı ÆØ ø 410)—the Titans under Kronos, defeated and imprisoned by Zeus; which strongly suggests that in this play, as in all other relevant Greek literature, it is assumed that the reign of Kronos was a time of felicity for mortals. Certainly the play nowhere denies this. And yet in the early days of Zeus’s reign Prometheus finds men lacking not only the necessities of decent life but the intelligence to devise means for procuring these necessities. They were ‘childlike’ (443); they ‘had eyes but saw not, ears but heard not, and . . . lived out their days haphazard and confused’ (447–9); they ‘did all things without understanding’ (456–7); it needed the intervention of Prometheus to make them ‘intelligent and possessed of mind’ (444). What has brought about the change? Did men formerly, under Kronos, possess intelligence, and has Zeus deprived them of it? Or was it rather that under Kronos they had no need of intelligence because, in the words of Hesiod, ‘the bountiful earth of her own 23 See (against authenticity) Griffith (1977); (for authenticity) Pattoni (1987). Pattoni does not in my judgement come near refuting Griffith’s case; but her comparison between the prologues of Prom. and of Sophocles’ Ajax (see especially Pattoni 1987: 181) makes it highly likely that Prom. is the earlier of these two plays.*
184
Pathos and mathos before Zeus
accord brought forth abundant fruits for them’, and that Zeus, by cutting off earth’s bounty, plunged men into a state of wretchedness from which they could escape only with the help of that intelligence which they did not yet possess? For the first explanation there is no evidence whatever; the second, on the other hand, might well be regarded as implicit in the traditional picture of the age of Kronos. It may even have been made explicit in some lost treatment of the myth earlier than Prometheus Bound: this would account not only for the way in which the author of that play seems able to take it for granted, but also for the use of the name of Kronos (and words derived from it) in colloquial Attic precisely to connote stupidity.24 In Prometheus Bound human ‘learning’ is almost exclusively (in the broad sense) technical. In the Oresteia it is rather moral and political wisdom that is to the fore. But the principle holds just as strongly that where there is no suffering, as under Kronos, there will be no learning. Man will be happy, no doubt, but his happiness will be that of a child or a pampered slave, utterly dependent on the favour of his divine providers. The harsh regime of Zeus seems a curse on man—a curse which Old Comedy repeatedly dreams of shaking off;25 and yet only under that regime can man be fully human, with the mind of an adult—and one meaning of Aeschylus’ çæ ØÐ , as Pope rightly emphasized,26 is precisely ‘to have the mind of an adult’. Before Zeus, the law Ł Ø Ł, even if it was theoretically true, was ineffective (¼Œıæ ), vacuous, [113] trivial: it is of no interest that X is the cause of Y, if X is something that can never occur. Zeus, by introducing suffering, for the first time made the law actually operative (Œıæø å Ø ). And thus, paradoxically, if human intelligence is a good thing (as Greeks on the whole do seem to have
24 Cf. ˚æ Ar. Clouds 929, Wasps 1480, Pl. Euthyd. 287b, Hyp. fr. 252; ˚æ Ø Ar. Clouds 1070; ˚æ ºÅæ, ˚æ Æ ø , ˚æ ŁŒÅ com. adesp. 1052–4 . In most of these passages the meaning is ‘old fool’; but Hyp. loc. cit. used ˚æ in the sense of Hæ (Bekk. Anecd. i. 104.7), and in Ar. Wasps 1480 the term is applied to tragic actors challenged to a dancing contest, and the three who take up the challenge (the sons of Karkinos) are in fact young men. 25 Thus Zeus is successfully defied in Aristophanes’ Peace, and actually deposed in his Birds and Wealth. Cf. Sutton (1980: 42–3); Reckford (1987: 326, 339–40, 359–63). 26 Pope (1974: 108–9).
Pathos and mathos before Zeus
185
believed it to be),27 the regime of suffering inaugurated by Zeus can be seen as a great blessing. Whether Zeus had that in view from the start is quite another matter. I see no reason to believe that the Zeus of the Oresteia did, any more than did the Zeus of the Prometheus plays.28 Such then seems to have been, in its simplest form, the common Greek understanding of how human life came to be as it now is. Man was once possessed of all happiness and devoid of all wisdom; then came Zeus and deprived him of happiness, so that he could survive only by acquiring wisdom. But it is not difficult to see how the same transition from bliss and unintelligence to suffering and intelligence could be interpreted in a sense almost diametrically opposite. What if Ł Ø Ł were replaced by Ł Ø Ł? What if, instead of suffering being seen as the stimulus to intellectual advance, the gaining of wisdom were seen as the cause of suffering? That may well seem to be what happens to the individual human being as (s)he passes from infancy to adulthood; but it is not an entirely straightforward matter to turn this idea into an origin myth. What, after all, is the sorrow-bringing wisdom that the growing individual learns? As the Sophoclean Ajax knew,29 it is the knowledge of the griefs and hardships that the world contains. But that kind of knowledge cannot have been the cause of the existence of those griefs and hardships. The connection has to be made in a slightly less direct way. Children are typically eager to know everything that adults know—including things that adults prefer to keep concealed from them; only when they themselves become adults do they discover how great a spiritual burden some of that knowledge is. That idea can readily enough be transposed into mythical terms. In the childhood of humanity, someone reached out for, and got, knowledge of that kind which humanity often feels it would have been better off without. That is the basis of the story of Pandora’s jar (Hes. Works 94ff.); for though we are not told why she removed the jar’s lid, Zeus had instructed Hermes to give her an KŒº qŁ (Works 67–8, 77–9; missing, together with
27
It may be sufficient here to cite two relevant texts: one is Soph. Ant. 332–64; the other is the Odyssey. 28 See Sommerstein (1989b: 23 n. 78). 29 Soph. Aj. 554–5 K fiH çæ ØÐ ªaæ Åb mØ , | ø e åÆæ Ø ŒÆd e ºı ØÐ ŁÆØ ŁfiÅ.*
186
Pathos and mathos before Zeus
the jar, from the parallel passage in the Theogony) and one may reasonably suppose that she removed the lid (while poor Epimetheus’ back was turned) for the same reason for which Greek women (or so their husbands alleged) were for ever removing the lids of storage jars or secretly opening cupboards: she wanted to know what was inside, in case it might be something worth stealing.30 And of course the same notion of reaching out for knowledge that proves a curse is also the basis of the story of Adam and Eve in Genesis. But with later Greeks the idea of Ł Ø Ł does not seem to have been popular. There is, as we have seen, plenty of nostalgia for the age of Kronos, but it does not, even in comedy, include the wish for a return to a state of blissful imbecility: it is one thing to wish for a life in which one will never need to put a strain on one’s mental apparatus, and quite another to wish that one did not have that apparatus. Most Greeks considered human wisdom a good thing, not just in the wretched world we now inhabit, but absolutely: no ancient Greek could ever, as Christians have often done, regard a mental defective or ‘holy fool’ as a person deserving admiration in his lifetime and worship after his death. It is better to be adult—çæ ØÐ . And what has caused [114] humanity to become mentally adult, in the Oresteia’s way of looking at things, is the harsh, ungiving, unforgiving world in which Zeus has caused them to live. And yet in one sense the Oresteia too can be said to join hands with Genesis. When Adam and Eve have eaten of the fruit of the tree of knowledge, God himself expresses the view that ‘this man has become like one of us’:31 they who were created in the physical image of God have now become, in a sense, mentally his peers as well.32 In the Oresteia the acquisition of wisdom is no matter of eating an attractive 30 Cf. Ar. Thesm. 418–27, 556–9, 812–13; Ekkl. 14–15; Alkimos, FGrH 560 F 2. Semonides fr. 7 West gives us one woman who ‘wants to hear everything, know everything and have her eyes everywhere’ (13–14), another who is constantly stealing (55–6—though she steals from neighbours’ houses or from offerings at altars rather than from her husband), and a third who ‘knows all the wiles and tricks of a monkey’ (78–9). 31 Genesis 3: 22. 32 Their only remaining inferiority is their mortality—for, as Genesis 3: 22 clearly indicates, they were created mortal and could have become immortal only through
Pathos and mathos before Zeus
187
piece of fruit: it comes by suffering—by blood and toil and tears over a period of generations. But when it has been attained, when humanity, and in particular the Athenians, have become ‘wise at last’ (øçæ ıÐ K åæ ø fi Eum. 1000),* the Erinyes who have been agents of so much of that suffering have this to say of relationship to the divine world: —ƺº ’ e æØÐ Z Æ –Ç ÆØ Ææ. You are under the wings of Pallas, and her father reveres you. (Eum. 1001–2)
That is the ultimate and paradoxical outcome of what the popular mind saw—and the Oresteia nowhere rejects this view—as the spiteful decision of Zeus to cut off the spontaneous bounty of Earth and condemn man to a life of toil and pain: the emergence of a human population who merit the awe and reverence33 of Zeus himself.34
eating the fruit of the tree of life. In Hesiod too man is mortal even in the age of primeval bliss (Works 116, 121). 33 Eumenides 1001–2, like Agamemnon 177 (see n. 3), has been mangled by a long succession of translators who have read it through the lenses of their (or perhaps, in some cases, their predecessors’) preconceptions about Greek religion. This has led to such renderings as the following: ‘yours at last the grace of Zeus’ (Thomson 1938); ‘therefore shall Zeus the Father love you’ (Murray 1952); ‘heaven’s . . . king . . . guards and governs well those favoured ones’ (Vellacott 1959); ‘blessed with Father’s love’ (Fagles 1977); ‘on you . . . her father looks with kindness’ (LloydJones 1979); ‘Zeus will favour you’ (Raphael and McLeish 1979); ‘with her father looking on you with favour’ (T. Harrison 1985/1986). Even Lattimore (1953) (‘grand even in her father’s sight’) tones down the Greek somewhat, as now does Podlecki (1989) (‘Father Zeus has respect for you’); of my sample only Weir Smyth (<[1925] 1957>) (‘the Father holdeth ye in reverence’) translates the words that were before him. Recently there has been a welcome tendency to recognize the true force of –Ç ÆØ (‘stand in awe of, especially gods and one’s parents’ LSJ; cf. Eum. 389, Supp. 652, Iliad 1.21) and hence the audacity of what Aeschylus makes the chorus say here; see Goldhill (1984: 278) and Chiasson (1988: 20). The translation of Grene and O’Flaherty (1989) omits the passage, together with almost the whole of the trilogy’s conclusion. 34 This chapter was originally published in H. D. Jocelyn (ed.), Tria Lustra: Essays and Notes Presented to John Pinsent, Founder and Editor of Liverpool Classical Monthly (Liverpool, 1993) 109–14. I am most grateful to Mrs Helena Hurt Pinsent for giving permission for this republication.
188
Pathos and mathos before Zeus ADDENDA
p. 182 n. 22 see now also Ruffell (2000); Farioli (2001: 27–137); Olson (2007: 75–7, 99–107) (text of, and commentary on, four of the fragments cited by Athenaeus). p. 183 n. 23 I would not now maintain the view that Prometheus Bound was earlier than Ajax; Bees (1993) seems to me to have demonstrated that it belongs to the 430s. p. 185 n. 29: the Sophoclean Deianeira (Trach. 142–54; note 143 ’ KŒ ŁØ ÆŁıÐÆ) and the Sophoclean Procne (Soph. fr. 583.3–5) knew it too. p. 187 this reading and interpretation of Eum. 1000 (where West 1990a had adopted van Herwerden’s emendation øçæ F K
Œæø fi ‘wise amid surfeit’) are brilliantly defended by Chiasson (1999/ 2000); I was probably wrong in the Loeb (Sommerstein 2008: ii. 477) to understand the phrase as ‘wise in due season’ (i.e. not needing to learn wisdom, too late, through painful experience), which gives K åæ ø fi a sense for which there is no clear parallel.
12 Oresteia Act II: two misconceptions The problem to which I want to draw your attention is one that we keep meeting, indeed one that we can’t really avoid, in the study of something that has been studied so long and so intensively as Greek tragedy. It is that because we usually know exactly how a play ends before we have begun to see it, we find it difficult to remember that the original audience did not have this precise knowledge. They were not, of course, completely ignorant, and certainly not in this particular case. The story of how Agamemnon was murdered they had seen already in Agamemnon. The story of how Orestes returned and killed his father’s murderers, one of whom was his mother, had been told many times before, and moreover this matricide had been predicted in the previous play, in the prophecies of Cassandra. So the original audience knew, as we do, that Orestes was going to return and kill his mother. Orestes himself, of course, also knows that he must kill his mother, because Apollo has told him so. But what about Electra? When Orestes arrives at the beginning of the play, he very soon sees a number of women coming along with the grave gifts which give the play its name. He recognizes one of them as his sister Electra, and he conceals himself. Thus Electra prays for divine aid against the usurpers at a time when, unknown to her, Orestes is actually present. Given that we know that Orestes has returned and intends to kill the two murderers, it is very easy for us to take it for granted that she is praying for Orestes to do just that. Let us look, though, at what is actually said. At first Electra is not sure what to pray for, and she asks the old women slaves of the chorus for advice. For whom, she asks (110), should I seek the favour of the gods? ‘For yourself, and for all who hate Aegisthus.’ ‘You mean,
190
Oresteia Act II: two misconceptions
therefore, for you and me? . . . Should I add anybody else?’ And they reply, ‘Remember Orestes’ (115). She thanks them for this good advice, and then, and only then, they go on (117–21) to recommend that she should ‘make mention of those responsible for the murder, and pray that some god or mortal man will come . . . to kill them in their turn’. Some god or mortal man: they don’t expect it to be Orestes, because they have already mentioned Orestes separately. Rather, they are suggesting that Electra should pray for two distinct blessings: first, the return of Orestes; secondly, the coming of an avenger. And that is what she does. Addressing the spirit of her father, she prays: ‘Pity me, and make our dear Orestes shine like a light for this house. . . . I beg you—and hear me, father—that by some turn of fate Orestes may come here’ (130–9). She adds a short prayer for herself (140–1), and then continues: ‘Such are my prayers for ourselves; for our enemies, I ask for someone to appear to avenge you, and that the murderers die in just requital for what they did’ (142–4). Just as the chorus-leader recommended, two separate prayers, one for the return of Orestes, the other for the coming of an avenger. After her, the chorus in song also pray for the coming of an avenger, whom they describe (160–3) as ‘a strong man with a spear, a war-god brandishing his bow in combat, ready for the work of fighting hand to hand and sword to sword’. That’s never Orestes, who is a very young man (and they are perfectly aware of his age) with no experience of combat. We, the audience, know—and the original audience knew too— that Electra’s two prayers, though distinct in intention, were in effect one and the same. But Electra doesn’t know this, and that is what is called dramatic irony. And the irony continues. When Electra comes face to face with the unknown person who is in fact Orestes, she still doesn’t know that he is the one who will have to kill the murderers— one of them his mother and hers. When he proves his identity to her, she still doesn’t know. When she joyfully embraces him and calls him her father, her mother, her sister, and her brother in one (238–43), she still doesn’t know. When he then prays to Zeus for aid (246–63), she still doesn’t know—because in that prayer he makes no mention of the revenge killing, asking only to be restored
Oresteia Act II: two misconceptions
191
to his rightful place in the house and city, with no mention of how this may be brought about. When does Electra know the truth? Only when Orestes, in his longest (or at least his longest surviving) speech in the trilogy, tells her and the chorus and us of what Apollo said to him at Delphi: ‘he ordered me to pass through all these dangers . . . telling of violent torments that would chill my heart, if I did not pursue those who contrived my father’s death “in the same manner”—meaning I must kill them in return’ (270–4). We knew Apollo had made that demand (though we probably didn’t know till now just how ghastly were the menaces that accompanied it); to Electra it is a bolt from the blue. At least, however, there seems to be some reassurance: Apollo has promised Orestes his support. Well, has he, actually? Or is that just another piece of hindsight? We know, or think we know, that Apollo will protect Orestes from the avenging spirits of his mother (her Erinyes or Furies) after he has killed her. How do we know this? In the first place, because that is what actually happens later in the trilogy. Fine, but we haven’t seen the end of the trilogy yet. Secondly, because it also happened (though not in quite the same way) in an important earlier version of the story (by the lyric poet Stesichorus).1 Fine again, but how can we be sure that this dramatization is going to follow the same track? Even from the scanty information we possess about Stesichorus’ Oresteia, we can see that Aeschylus has already made one quite significant change (having Clytaemestra kill Agamemnon with a sword instead of an axe),2 and if we had better information we would doubtless find he had made several more. At this point, therefore, our only evidence for any promise by Apollo consists in the opening words of this speech by Orestes: ‘Apollo’s great and mighty oracle will not betray me’ (269–70). That certainly does sound reassuring. But it sounds a good deal more reassuring to us than it can have done to the original audience, because we can look ahead to the next play, when the same words will be repeated by
1
Stesichorus PMG 217. In Stesichorus PMG 219, Clytaemestra dreams of seeing a snake ‘with the top of its head all bloodied’, which implies an axe-blow from above. That her weapon in Aeschylus is a sword was shown by Fraenkel (1950: iii. 806–9); the counter-arguments of Davies (1987) were answered by Sommerstein (1989c) and Prag (1991). 2
192
Oresteia Act II: two misconceptions
Apollo in person (Eumenides 64–6), and then they really will be an assurance and a promise: Never will I betray you: I will be your guardian to the end, and whether I stand near to you or far from you, I will not be kindly to your enemies.
But once again, we haven’t heard that yet. When Orestes now, before the event, talks about Apollo’s oracle, he says a great deal about dark and terrible menaces of what will befall him if he does not carry out his revenge; he says not one word about any protection that will be given him if he does. How long is it before we know that Orestes can indeed count on Apollo’s protection? When the plotters immediately proceed to make their long lament and prayer at Agamemnon’s tomb (306–509), they make appeals to a large number of gods—Zeus, the Earth, the underworld gods, the Erinyes, Persephone the queen of Hades—as well as to the mighty spirit of Agamemnon himself. Of Apollo they make no mention—nor, indeed, of any of the other gods of the upper world, of Olympus (for when they speak of Zeus,3 they make it clear that they are talking about ‘the other Zeus’, ‘the Zeus below’, a euphemistic name for Hades).* There is nothing here to suggest that Orestes can look to any of the Olympians for protection. When Orestes thereafter explains his plot to Electra and the chorus, we still don’t know he has a promise from Apollo. When he gains admission to the house, we still don’t know. When Aegisthus is lured to the house alone, we still don’t know. When son and mother actually confront each other, we still don’t know. Near the climax of that confrontation (924) she warns him ‘Beware your mother’s wrathful hounds!’—a threat of the tormenting fury of the Erinyes. Orestes’ obvious reply would be ‘Why should I be afraid of your wrathful hounds? I know Apollo will keep me safe from them.’ He says nothing of the sort. What he does say is: ‘If I fail to do this, how can I escape my father’s Erinyes?’ In other words: I know I’m damned if I do; but I’m also damned if I don’t. A little earlier, Orestes had hesitated for a moment, when his mother had appealed to the most basic ties of nurture and shown him the breast that he had once sucked. He turns to the friend who 3
Cho. 382–5: ‘Zeus, Zeus, who sends up from below avenging ruin, soon or late, against audacious, reckless human violence.’
Oresteia Act II: two misconceptions
193
has accompanied him silently all through the play, and asks ‘Pylades, what shall I do? Should I spare my mother’s life?’ (899). Here, again, was the moment for a reassuring reply; but he does not get one. Instead, speaking for the only time in the play, Pylades says: ‘Where would that leave Apollo’s oracle delivered at Delphi? Where would it leave your faithful, binding oath? Make anyone your enemy, rather than the gods’ (900–2). In other words: if you don’t do it, you will have to face, not only the avenging spirits of your father, but also the wrath of Apollo. Even after the killing, when Orestes appears standing over the bodies of the two usurpers and murderers, and displays for all to see the robe in which his father was trapped and slain, even then we still have heard nothing of any promise of protection by Apollo. But there is one strange thing about Orestes’ appearance now. He still has his sword in his right hand, but in his left hand he has an olive branch wreathed in wool (mentioned at 1035), the symbol of the suppliant. What does this mean? We are about to hear. The Erinyes, he tells us, are already attacking him mentally, he feels himself on the brink of madness (1021–5), and then he continues: While I still have my sanity, I proclaim to my friends that I killed my mother not without justice—the polluted murderer of my father, loathed by the gods—and the power that induced me to dare this deed was, more than anything else, Apollo the seer of Delphi, whose oracle told me that if I did this I would be beyond the charge of wickedness, but if I failed—I shall not tell the punishment of that; no one could shoot an arrow high enough to hit it. (1026–34)
So now at last, more than 96 per cent of the way through this play, more than two-thirds of the way through the whole Oresteia, we at last know that Apollo is not going to let this endless cycle of revenge continue indefinitely. He has made a promise, and Orestes, as he goes on to tell us, is going to Delphi to beg him to fulfil it. Till now, we had thought, Electra had thought, the chorus had thought, that there was no way out for Orestes, any more than there had been for all those who had suffered in the previous play. And now, at last, it seems that Apollo has in fact promised his protection, and (if we think about it—but we almost certainly won’t) that Orestes knew this all along.
194
Oresteia Act II: two misconceptions
But the horrendous Erinyes are already on his trail. He can feel them warping his mind. And then suddenly (1048) he sees them (which, at this stage, no one else can) and flees in headlong terror. So what is going to happen now? Can we tell? Do we know? It is significant that the last words of this play are an unanswered question: ‘where will it end, where will the power of Ruin go to sleep?’ They have no idea. It will take another play to provide an answer— and probably no one in that Athenian audience could have remotely guessed what the answer was going to be: ‘here and now, among yourselves, in Athens’. I urge you, when you have seen this play, to get acquainted with that sequel, Eumenides. Perhaps you may then perceive why I found it so utterly appropriate that when I published an edition of Eumenides, its official publication date (named in advance—no hindsight here!) was 9 November 1989: the day of the breaking of the Berlin Wall.4
ADDENDUM p. 192 while the Zeus of Cho. 382–5 certainly is ‘the Zeus below’, there is no particular reason to doubt that the mentions of Zeus in Cho. 395 and 409 refer to the supreme god of Olympus. It remains true, however, that in both these passages there is also, in close proximity, prominent mention of chthonic powers (399 ‘hear me, Earth, and you honoured gods below’; 405–7 ‘see us, you mighty rulers of the underworld, and you Curses of the dead’). 4 This chapter, which originated as a pre-performance talk given before a production of Choephoroi (under the title Grave Gifts) by the Actors of Dionysus, was first published in Dionysus (York) 14 (2000) 20–2. I am most grateful to the Actors of Dionysus for giving permission for this republication. As in the 2000 publication, I have left references to the original occasion of delivery unchanged, though a few footnotes have been added in the present version.
13 Aeschylus’ epitaph
`Nå º ¯Pçæø ŁÅ ÆØÐ Œ Ł Ø ÅÐ Æ ŒÆÆçŁ ıæçæØ ˆºÆ· IºŒc ’ PŒØ ÆæÆŁ Ø ¼º i YØ ŒÆd ÆŁıåÆØ Ø B KØ .1 In Gela, rich in wheat, he died, and lies beneath this stone: Aeschylus the Athenian, son of Euphorion. His valour, tried and proved, the mead of Marathon can tell, The long-haired Persian also, who knows it all too well.2
Athenaios (14.627c) and Pausanias (1.14.5) say that Aeschylus composed this epitaph himself; the Life transmitted in the Aeschylean manuscripts says merely that the people of Gela ‘gave him a costly funeral in their public cemetery and honoured him magnificently, inscribing these verses’, without specifying a composer. One is fully entitled to approach such traditions as these with scepticism, knowing what we know about the habits of ancient biographers, especially biographers of poets,3 and knowing also that this was not the only epitaph of Aeschylus circulating in antiquity;4 and Aeschylus is 1 Life of Aeschylus 11; the first half is quoted by Plut. Mor. 604f and by an anonymous commentator on Aristotle’s Nicomachean Ethics (CAG xx. 146.2–3 Heylbut), the second half by Ath. 14.627c. There are many variant readings, but the text is nowhere in serious doubt (except perhaps for the Aristotelian commentator’s reading B Æ j Œ Ł Ø IçŁØ{ } ). 2 Sommerstein (1996a: 24). 3 See Lefkowitz (1981) passim. 4 The others are AP 7.39, 40 (¼ Aeschylus test. 164, 165 Radt), ascribed to two epigrammatists of the Augustan age, Antipater of Thessalonike and Diodoros (of Sardis). A fragment (one pentameter) of a further epitaph, couched in the first person, is preserved in Life 17.
196
Aeschylus’ epitaph
certainly not likely to have written an epitaph that specified the place of his death.5 We may, [112] though, note that this was clearly much the most famous of the ‘Aeschylean epitaphs’ in existence,6 and that even if we can hardly suppose Aeschylus to have written it himself, a member of his family7 could have written it shortly after his death at the request of the Geloans. After all, there must have been some inscription on Aeschylus’ tomb, and given that he was a distinguished poet it would not be at all strange if the inscription was in verse. Denys Page, however, has attempted8 to demonstrate that the language and content of the epigram proves it to have been composed no earlier than the Hellenistic period. He brings four arguments. Three of these, I shall show, prove nothing; the fourth, far from proving the poem late, is actually evidence that it is a fifthcentury composition. 1. Page argues that an epitaph inscribed on a tomb at Gela would not have mentioned that the tomb was at Gela. In fact it is common for epitaphs to mention the place of death or burial if, but only if, the deceased was not a native or citizen of that place. Thus in CEG 104 (c.400) we read źРÆæ ]’ ŁÆ Œº ÆØÐ K Ł ÆØ E<æ>Åd ª øØÐ Ø AØ ºØÐÆ Ł ,
cf. also CEG 114 (possibly of the year 479), 528, 545. Page, to be sure, points out that in all these the deceased speaks in the first person, and asserts that ‘common sense’ precludes the possibility of a thirdperson epitaph informing its readers where it is located. Apparently then common sense deserted the writers of epitaphs in the fourth century, when we find in Athens an epitaph like this, in which third person alternates with second:
5 Though it would be foolish to insist that he could not have done so—so long as we reject the fantastic story about his death told in Life 10. 6 It is the only one which is quoted or referred to by any ancient author other than Aeschylus’ own biographer.* 7 Aeschylus’ son Euphorion, and his nephew Philokles, were both poets good enough to win first prizes for tragedy at the City Dionysia. 8 Page (1981: 131–2).
Aeschylus’ epitaph
197
. . . ı åø fi Kd Ææ, ƃ b Ł EØ ºØÆ çºÆØ Ł ÅE ’ ŁB ÆØ øÐ Æ e Kª ŒºØ Œæ łÆ IçŁØ ı (CEG 606.7–9)
[113] and in Cyprus: ØŒæa " å[ÅØ] ıæ, , ç ÆØ m I’ ¯ºÅ `Nº źıÐ ˚ æ Œ çøØ. (CEG 713)9
It is clearly a commonplace of the sepulchral epigram, in the case of one who died far from home, to name both the homeland and the place of death/burial—not, of course, in order to tell readers where they are, but to emphasize (1) the pathos of dying away from home and being buried far from one’s kin and (2) counterbalancing this, the honour of resting in the soil of a famous or pleasant land. The Aeschylus epitaph follows this same pattern—Aeschylus is an Athenian who has died at Gela, and Gela is praised for its fertility (‘rich in wheat’)—and in addition makes a further contrast between Gela as the site of Aeschylus’ physical tomb/memorial ( B Æ) and Marathon which enshrines and ‘can tell’ the memory of his martial valour. 2. Page claims that the use of the genitive in line 2 (ıæçæØ ˆºÆ) is unclassical. He unwarrantably takes it for granted, however, that the genitive is possessive and depends on B Æ;10 it could perfectly well depend on the word next to it, ŒÆÆçŁ , and be an example of the poetic genitive of ‘place within which’ (see Ku¨hner and Gerth 1898–1904: i. 384 and Moorhouse 1982: 59). 3. Page objects to the epithet ÆŁıåÆØ Ø as inappropriate to Persians; but it would refer very well to a feature of Iranian hair-styles to which, oddly enough, Page himself draws attention. In Iranian art nobles are generally represented as having ‘thick buns of hair on the nape’,11 and any fifth-century Greek who had seen a distinguished Persian (whether an enemy on the battlefield or an ambassador in the assembly) might have observed this (to Greeks) bizarre fashion and created this epithet to describe it. How likely [114] is it, on the other 9
Cf. also CEG 737. ‘No inscription of so early a date could possibly have called the tomb “the memorial of Gela”.’ 11 Cf. e.g. Culican (1965) pl. 18, 21, 22, 41, 42, 52. 10
198
Aeschylus’ epitaph
hand, that this epithet for the Persians could have originated after the fall of their empire? 4. The plain of Marathon, claims Page, is not appropriately described by the word ¼º. That is true provided, but only provided, that ¼º bears one of its normal meanings, which are (1) ‘sacred grove’; (2) ‘grove, glade’ whether or not sacred; (3) ‘sacred enclosure’ whether or not wooded.12 These three senses (the third only in poetry) account for all but ten of the 150-plus occurrences of ¼º and its derivatives in literary texts down to 100 bc.13 Of the ten exceptions, one is in our epitaph; the other nine are all found in poetry certainly or probably written between 475 and 410 bc—one by Bacchylides, one by Melanippides, two by Sophocles, and five by Aeschylus, in whose surviving work, moreover, there is no instance of ¼º in any of its regular senses. Three times ¼º is applied to the sea (Aesch. Pers. 111,14 Supp. 868; Bacchyl. 17.85).15 In another passage (Aesch. Supp. 508–9) it is used (twice) of an area which is specifically described as being both smooth (º ıæ )— hence presumably [115] treeless16—and non-sacred (ź ); in others again it denotes broad plains like those of Thebes (Soph.
12
On occasion ¼º in this sense may be applied to a whole city as being under the patronage of a particular divinity or hero: thus ‘sacred’ Onchestos is the ¼º of Poseidon in Iliad 2.506 (cf. h.Ap. 230, h.Herm. 186), Aigina of the Aiakidai in Pind. Olymp. 13.109, and Eleusis of Demeter in Pind. Isthm. 1.57. 13 And in the epigrams of the Anthology regardless of date. The statements I have made about the distribution of ¼º (with its derivatives) in its various senses are based on a TLG search of all authors dated from the eighth to the second century bc and also of Lyrica Adespota, Tragica Adespota, Comica Adespota, and the Anthologia Graeca. Typical instances of sense (1) are Odyssey 6.291, 20.278, Sappho fr. 2.2–3, Soph. Tr. 1167, Hdt. 6.78–80 passim; of (2), Iliad 20.8, Theogn. 1252, Arist. HA 618b19, 22, 34; of (3), Bacchyl. 3.19, Pind. Olymp. 7.49, Eur. Tro. 15, Ar. Thesm. 1149 (see also previous note). 14 Broadhead (1960) ad loc. takes Ø ¼º as ¼ ‘the demesne sacred to Poseidon’; but there is no mention of Poseidon in the context. Hall (1996) more cautiously suggests that the word ‘implies a sense of enclosure’ and so may refer to the Hellespont rather than the open sea; but an interpretation on these lines will not work for the other two passages cited in the text. 15 In contrast, when a millennium later Paul the Silentiary (AP 9.663) speaks of ŁÆºÆÆEÆ ¼º Æ, he means not the sea, but seaweed. 16 Particularly since the area being described was almost certainly, in theatrical fact, the floor of the orchestra.
Aeschylus’ epitaph
199
Ant. 844–5), of Argos (Soph. El. 5)17 or of Egypt (Aesch. Supp. 558,18 Melanippides PMG 757.3).19 Thus invariably (so far as our knowledge goes) in Aeschylus, and occasionally in the work of his contemporaries,20 but (unless in our epitaph) never before or after, ¼º denotes, not a (relatively small) area covered with trees, nor a (relatively small) area consecrated to a divinity, but a level area regardless of its size;21 it has become, in fact, a virtual synonym of , [116] which can likewise be applied to the sea (e.g. Aesch. fr. 150). This is the sense which ¼º bears in our epitaph, and it is therefore overwhelmingly probable that it was written by a contemporary of Aeschylus.
17 The reference cannot be to the city of Argos, since the epithet NæºBª applied to Io indicates that we are to think of her after her transformation, when in bovine form she grazed (as Aeschylus’ Danaids put it) K Ø Ø Ææe IæåÆÆ Ø (Aesch. Supp. 50–1, cf. 538–42, Prom. 673–82). 18 Strictly speaking this passage (˜E ¼º) might be regarded (as by Friis Johansen and Whittle 1980 ad loc.) as an instance of the sense ‘sacred enclosure’; but nowhere outside the ten passages now under consideration is ¼º unambiguously applied to any area larger than a city. 19 D. A. Campbell (1993: 23), under the spell of the conventional rendering of ¼º, translates ¼º Æ here as ‘glades’, not reflecting that glades are neither sunny ( PºØÆ) nor suitable places to practise chariot-driving (K ±æ Ø Øçæ åØ Kªı Ç ’; chariots figure likewise in Soph. Ant. 844–5!). 20 For the evidence for the chronology of Melanippides’ life see D. A. Campbell (1993: 15); add that he was dead by the time Pherekrates’ comedy Cheiron was produced (cf. Pherekr. fr. 155.6–7 KA). 21 How did this sense of ¼º originate? Most probably by a slipshod extension of sense (3). Where, as in the passages cited in n. 12, a city is described as the ¼º of a god or hero, it may not always be obvious whether the reference is to the urban centre or to the entire territory of the polis; if the latter alternative is taken, it then becomes legitimate to describe a whole country, even a very large one, as the ¼º of its patron god. The final stage comes when, in a passage like Aesch. Supp. 558 (though of course that passage cannot itself have been the model), a phrase like ˜E . . . ¼º is taken to be synonymous not, as sense (3) would require, with ˜Øe ƒ æe but with ˜Øe . We cannot tell by whom or when this final step was taken, except that it had been taken by 472 bc; but noting the remarkable fact that of the ten surviving instances of ¼º ¼ four are in Aeschylus’ Suppliants and two more (Melanippides PMG 757.3 and Soph. El. 5) relate to persons appearing, or prominently mentioned, in that play (the Danaids and their ancestress Io), I would like to suggest that the person responsible may have been the tragic dramatist Phrynichos who is known to have written at least two plays (`Nª ØØ and ˜Æ Æ ) based on the Danaid myth.
200
Aeschylus’ epitaph
There is another consideration too. It has often been observed that the epitaph quoted in the Life makes no reference to Aeschylus’ art. Indeed, anyone who knew of Aeschylus only from the epitaph would be entirely unaware that he was a man of any special individual distinction. To be sure, he had fought bravely at Marathon, but so had ten thousand others. Would anyone have written a fictitious epitaph for a man like Aeschylus in such terms? Certainly the two other ‘Aeschylean epitaphs’ we possess, written by known later authors and making no pretence to authenticity, are not of this kind: one dilates on his contribution to the development of tragedy, the other calls him ‘the great’. Other fictitious epitaphs for archaic and classical poets similarly praise them as individuals.22 Not so our poem: to its author, Aeschylus was first and last an Athenian who did his duty as a citizen and a soldier. Such an attitude bears the stamp of that poet who, more than any other, articulated the ideological principle of Athenian democracy according to which communities prospered through the loyalty, solidarity, and courage of their peoples, and came to grief through the selfishness and folly of their leaders.23 If Aeschylus did not write the epitaph himself, it was written by someone who knew what Aeschylus would have wanted to be remembered for: not for winning thirteen victories for himself at the City Dionysia, [117] but for winning, with his comrades, one victory for Athens on the plain—or, to use a favourite word of his own, the ¼º—of Marathon.24 22 Of nine known epigrams on Sophocles (Soph. test. 1.72–3 and testt. 177–84 Radt), of which four are ostensibly sepulchral, every one refers to his poetry or drama; of nine on Euripides (Eur. test. 1.18 and testt. 63, 64, 66, 89, 96–9 in Kovacs 1994), all except the disparaging 64 praise him as poet, dramatist, or winner of glory; of the numerous sepulchral epigrams on poets in AP 7 not one praises the poet exclusively or even mainly, as our Aeschylean epitaph does, for being a good citizen (I exclude of course the two epigrams on Solon, AP 7.86–7, since their authors, like every other educated Greek, will have pigeonholed Solon not among poets but among statesmen and lawgivers). 23 See Sommerstein (1996a: 413–30); Ober (1989: 166–70). 24 This chapter was originally published in Museum Criticum 30/31 (1995/6) 111–17. I am most grateful to Pacini Editore for giving permission for this republication.
Aeschylus’ epitaph
201
ADDENDUM p. 196 n. 6: this is true only if ‘ancient’ is taken to exclude ‘medieval’; the epigrams by Antipater and Diodorus are both quoted in the Suda (3203; 118, Ł366).
14 Dearest Haimon Malcolm Davies1 has recently restated and amplified the case for accepting the manuscripts’ attribution of Antigone 572 (t çºÆŁ’ `x , u ’ IØ Ç Ø Ææ) to Ismene rather than Antigone, and the increasingly strong scholarly consensus in favour of this attribution has since been crowned by its adoption in the Oxford edition of Lloyd-Jones and Wilson.2 The arguments in its favour are on the whole powerful: not so much the arguments deriving from the contribution this passage might make to the presentation of Antigone in the play as a whole (it is as easy—or as hard—to extract from the play a plausible and coherent Antigone who does speak 572 as to extract one who does not) as those based on formal and verbal features of the immediate context. For Antigone to speak here would be a breach of the regular conventions of stichomythia.3 Kreon’s reply (¼ªÆ ª ºı ØÐ ŒÆd f ŒÆd e e ºå) makes perfect sense if addressed to Ismene,4 who is pressing him on the touchy subject of Haimon’s betrothal in a manner he would be bound to find highly vexatious, but seems remarkably weak if spoken to Antigone who is guilty of something, in his eyes, vastly worse. And if Antigone is addressed, would we not expect her to reply?5 These 1
Davies (1986). Lloyd-Jones and Wilson (1990a); cf. Lloyd-Jones and Wilson (1990b: 127–8). 3 Cf. Mastronarde (1979: 95–6). 4 See (in particular on e e ºå ¼ ‘the marriage you speak of ’) Davies (1986: 21–2); Lloyd-Jones and Wilson (1990b: 26–7, 128). 5 Hence there was a certain logic in the proposal of Dawe (1978b: 106–8) to give 574 (and 576) as well as 572 to Antigone; unfortunately the lines are utterly unsuited to her (see West (1979b: 108), Lloyd-Jones and Wilson (1990b:128), who concur in giving 574 and 576 to Ismene). 2
Dearest Haimon
203
considerations, while not decisive, do set up a definite prima facie case in favour of the manuscript attribution, and put the burden of proof on those who would reject it. [72] There remains, however, one consideration of the same order which should give us pause. If the manuscripts are right, Ismene apostrophizes Haimon as çºÆ . Would it be proper for her to speak of him thus? The question I am posing should be carefully distinguished from the question whether ‘“dearest Haemon” is more appropriately spoken by a fiance´e than by a mere cousin’6 or whether Ismene can be allowed to use the expression in Antigone’s presence.7 The question that needs to be asked is whether it would be proper in any situation, given a woman and a man placed with respect to each other as Ismene and Haimon are, for the former to call the latter çºÆ—an issue in principle quite independent of what Antigone’s connections are with Haimon, or indeed of her very existence. Some have thought the answer to this question self-evident,8 but an investigation of the actual use of çºÆ by women speaking to or about men in tragedy suggests otherwise. It suggests, in fact, that if Ismene speaks Ant. 572, her use of çºÆ is unique in surviving Greek tragedy. The number of relevant passages in the extant plays, if we omit Ant. 572 itself, is fifty-two9—six in Aeschylus, fourteen in Sophocles, and thirty-two in Euripides. They fall into three classes:
6
Brown (1987: 169) (not his own view). G. Mu¨ller (1967: 111). 8 For example Knox (1968: 755), who thinks that the argument of Mu¨ller (1967) ‘can be dismissed in view of the frequent occurrence of this form of çº as a common salutation’ and then cites two examples, both spoken by males to males (one by the leader of a chorus of satyrs!); Hester (1971: 30 n. 1) (‘the epithet she [Ismene] applies to Haemon is in no way remarkable’); Paduano (1982: i. 292–3 n. 39) (‘e` risibile l’obiezione secondo la quale la cognata (e perche´ non anche cugina?) Ismene non puo` apostrofare Emone con çºÆŁ’’). 9 This does not include such passages as Aesch. Cho. 1051 where the speaker is calling the addressee çºÆ not to herself but to some third person. 7
204
Dearest Haimon
1. In the vast majority of cases the person addressed or described as çºÆ is a very close member of the speaker’s family—father,10 [73] brother,11 husband,12 or son.13 2. There are two examples (one Sophoclean, seven Euripidean) of the usage discussed by Gregor (1957), where a person is called çºÆ because he is a bringer of what may be best defined as welcome news. News may be welcome because it is good news (Soph. Tr. 232; Eur. Herakl. 788, Supp. 641, El. 229, 767; at Eur. Hec. 990 the speaker pretends to welcome good news which she actually knows to be false) or because the speaker hopes it will be good (Eur. Hec. 505) or simply because it brings release from suspense (Eur. Phoin. 1072). 3. It is also possible for a woman to address in this way a man who has been, or who she hopes will become, her saviour. Thus the chorus of Seven against Thebes so address Eteokles (677) on whom the city’s safety, and therefore their own, depends. Antigone speaks of Theseus and his çغø . . . OÆ ø (Soph. OC 1103), who have saved her and Ismene from their Theban kidnappers. Electra (Soph. El. 1354, 1357) ecstatically greets Orestes’ paidagogos, whom she describes as øcæ ø ªÆ . . . n Œ¼ ’ øÆ KŒ ººH ø . Andromache, supplicating Peleus to save her and her child from death, apologizes for being unable to make the full suppliant gesture of touching his çغŠª Ø (Eur. Andr. 574). With these we may also class Eur. IT 1065 where Iphigeneia brackets herself with Orestes and Pylades as æ ØÐ . . . f çغı:
10 Aesch. Cho. 496; Soph. El. 462; Eur. IA 652 (curiously, the father is Agamemnon in all three cases). 11 Aesch. Cho. 235; Soph. Ant. 81 (Antigone speaking of the dead Polyneikes; the only place, except 572, where çºÆ is used at all in this play), El. 808, 903, 1126, 1158, 1163, 1208, 1224, 1286; Eur. El. 1322, IT 815, 827, Phoin. 166, 1437, 1702, Or. 217, 1045, IA 1452. 12 Aesch. Cho. 893; Eur. Andr. 222, El. 345, HF 490, 514, 531, Hel. 595, 625, 899, 1299. We should also include here Aesch. Ag. 1654: it is not clear at what moment (if ever) Klytaimestra becomes Aigisthos’ ªÆ c ªı , but she has already publicly avowed their relationship (1434–6) and in a few moments, like a bridegroom of the wrong gender, she will be leading him into her house. 13 Aesch. Pers. 851; Eur. Suppl. 793, Tro. 757, Ion 1409, Ba. 1298.
Dearest Haimon
205
their coming has given her the chance to escape from her hated Taurian exile.14 [74] If our line is spoken by Antigone, it can reasonably be assigned to class (1); to be sure, Haimon is a prospective rather than an actual husband, and to that extent there is no exact parallel, but the betrothed maiden is not a very frequent character-type in tragedy and the absence of a parallel could not really be regarded as suspicious. If, however, the speaker is Ismene, things are much more difficult. There can evidently be no possible justification for the superlative under rubric (2) or (3); we are not informed at any stage in the play of any action at all by Haimon which affected, or was likely to affect, Ismene directly for good or ill. And Haimon falls, in relation to Ismene, well outside the category covered by (1), which excludes not only cousins but even, so far as the evidence goes, closer relations such as uncles, nephews, and grandsons,15 while Ismene’s connection with him as a prospective sister-in-law is of so little significance that classical Attic had no word to describe it.16 Abnormal as t çºÆŁ’ `x would thus be on Ismene’s lips,17 it would be hard to claim that this should outweigh the other evidence that favours her as the speaker of 572. Rather we should perhaps accept the abnormality, acknowledge that the expression is a startling
14 Orestes is of course her brother, but we can hardly suppose she calls Pylades çغ because he is her cousin (917–19) and Electra’s husband (913–15); she has this day met him for the first time, and he was not born when she disappeared at Aulis (920–1). 15 Note that in Euripides’ Orestes, though the cousins Orestes and Pylades can speak of each other as ıªª ØÐ (733, 1233), Orestes can also use language implying that they are not (804–6, contrasting Ææı with e ıªª and I cæ . . . ŁıæÆØÐ with › Æ ø ). 16 Classical Attic uses one term, ŒÅ , reciprocally for all male relations by marriage; there is no general term for a female relation by marriage, and we know (in prose) of only one specific term, Ł æ ‘mother-in-law’ (Dem. 45.70). See Thompson (1971). Indeed even the elaborate, Indo-European-descended set of affinity terms we find in Homer does not include, and seems never to have included, any term for ‘wife’s sister’ or for ‘sister’s husband’. 17 Even in comedy, Aristophanes and Menander together can show only one passage (Ar. Lys. 853) where a respectable wife, maiden, or widow addresses or speaks of a man not of her immediate family as çºÆ.
206
Dearest Haimon
one which would draw attention to itself, and ask whether there are special circumstances that would explain and justify its use. Mary Whitlock Blundell18 has seen a special circumstance in Ismene’s ‘affectionate’ nature. Certainly Ismene is as full of feel-[75]ings of çØºÆ towards Antigone as the latter is barren of them towards her,19 and she has just offered to lay down her life to show solidarity with and moral support of Antigone, to do honour to Polyneikes (545), and because for her life without Antigone would not be worth living (548, 566). But despite Antigone’s gibe that Ismene is Kreon’s ŒÅ (549), she has said nothing hitherto to make us suppose that her affectionate feelings extend beyond her immediate kin. Indeed her first words in the play (11ff.) suggest that for her çºØ means, essentially, immediate kin; if she has heard no word of her çºØ since her brothers killed each other, then Kreon and Haimon, in whose house she and Antigone have been living (cf. 491, 531–3), do not count for her as çºØ. No more than Antigone does she think of Kreon as her uncle or her guardian: she thinks of him as the ruler and representative of the state (cf. 44 IææÅ º Ø with 47 ˚æ I ØæÅŒ, also 60 ıæ
ø , 63 Iæå Ł’ KŒ Œæ Ø ø , 67 ØÐ K º Ø HØ). Thus Ismene’s outburst is not only unique in expression; it is also unprepared for. Nothing has led us to imagine that she regarded Haimon as çº at all; yet now she calls him çºÆ. If her personality, as revealed up to this point, cannot by itself explain this, can the immediate context do so? Ismene, we have seen, is unconditionally loyal to her living immediate kin,20 but all her interactions in the play are with persons who do not show such loyalty. In the opening scene, on discovering that she is not prepared to sacrifice her life for the dead, Antigone calls her an enemy (86, 93) and a contemnor of that which the gods hold in honour (77 a H Ł H Ø ’ IØ Æ’ å ). When they meet again, and Ismene gives the highest possible proof of her
18
Whitlock Blundell (1989: 108 n. 11). Cf. ibid. 111–15. 20 The qualification ‘living’ is crucial: she will defy the state and risk her life for Antigone, but not for the dead Polyneikes. Antigone’s priorities are the reverse: in spirit she herself has long been dead (559–60). 19
Dearest Haimon
207
devotion to Antigone, Antigone not only rejects her self-sacrifice but does so in as cutting and hostile a manner as she can, and Ismene feels that she has been treated with contempt unbecoming a sister (544 Ø, ŒÆت Å, ’ IØ fi Å). But at least Antigone’s attitude, however wrong, is itself based on feelings of çغƗfeelings that Ismene can understand because she too [76] shares them, though not to the same extent.21 In Kreon, on the other hand, she finds herself confronted with someone who seems, to judge by his words, to have no feeling of çØºÆ for his immediate kin; someone who, it might be said, h ıªª ÅÐ h çº r 22; someone who, when it is drawn to his attention that he is injuring his own wholly innocent son, can reply (569) in terms showing no more consideration for the son’s feelings than if Haimon were a ram or a bull.23 Such an attitude is the very negation of everything we have seen Ismene say and do; no wonder she is moved to speak of it in the same language that Antigone has used to her and she to Antigone about behaviour that each saw as contempt for the claims of immediate kin—u ’ IØ Ç Ø Ææ. It may be that she speaks of Haimon as çºÆ out of fellow-feeling for a fellow-victim of such unwarranted contempt; it may be that she is not calling him çºÆ to herself at all, but reminding Kreon that his son is, or ought to be, çºÆ to him; at any rate her extraordinary expression is to be accounted for by, and serves to draw our attention most arrestingly to, the extraordinary indifference of Kreon to his son. It is the climax, and almost the end, of her role in the play. Within four lines she has recognized that in pleading for Antigone to be spared, even for Haimon’s sake, she is beating her head against a stone wall; within nine she has been taken offstage, not to appear again; at the end of the following scene her condemnation is casually 21 She has begged the pardon of her dead kinsfolk for failing to act on their behalf (65–6), and when she offers to die with Antigone she does so partly in order to do honour to their dead brother (545). 22 Men. Asp. 117–18 (the prologue-goddess speaking of Smikrines whom she has just described as ‘surpassing all men in wickedness’). 23 This brutality may be compounded by the impersonal-sounding plurals of 571 (ŒÆŒa Kªg ªı ÆØÐ ŒÆ ıƒØ ıªH), especially if the audience knew or suspected, from previous dramatic or poetic versions of the story, that Kreon actually had only one surviving son.
208
Dearest Haimon
rescinded (769–71), and by 941, when Antigone speaks of herself as the only remaining member of the royal house, Ismene’s very existence has been forgotten. But she has had her moment, and she has had her function: to represent the principle of being ØÐ çºØ . . . çºÅ (99), the principle that both Antigone and Kreon in their different ways refuse to recognize.24 24 This chapter was originally published, under the title ‘Soph. Ant. 572 (“Dearest Haimon”)’, in Museum Criticum 25–8 (1990–3) 71–6. I am most grateful to Pacini Editore for giving permission for this republication.
15 ‘They all knew how it was going to end’: tragedy, myth, and the spectator I am going to begin with three quotations. One is from a book I have admired for very many years, H. W. Fowler’s Modern English Usage. It comes from the article on irony, when Fowler is trying to explain the meaning of ‘dramatic’ irony. The facts of most Greek plays were not a matter for invention, but were part of every Athenian child’s store of legend; the audience, that is, were in the secret beforehand as to what would happen. But the characters, Pentheus and Oedipus and the rest, were in the dark: one of them might utter words that to him and his companions on the stage were of trifling import, but to those who hearing could understand were pregnant with the coming doom.1
We all know, do we not, that this is a major difference between the plots of Greek tragedies and those of most modern dramas: in Greek tragedy the audience knew from the start what was going to happen by the end. And it presumably follows from that that the key events of the plot were fixed in advance and unalterable—a dogma which has long been invaluable to critics seeking to explain away apparent flaws and inconsistencies in the structure of plays. The trouble is, it can easily be shown that the dogma is false. To begin the process of loosening up ideas on this, may I draw your attention to the results of some research done about twenty-three centuries ago on the plots of two other Sophoclean plays, Antigone and Ajax. The surviving ancient commentaries on these plays begin with headnotes (known as Hypotheses) summarizing the plot and 1
H. W. Fowler ([1926] 1965: 305).
210
‘They all knew how it was going to end’
giving various bits of background information, including some information about alternative versions of the story. It turns out that there is nothing sacrosanct about the Antigone story as Sophocles tells it: Ion [a contemporary of Sophocles] in his dithyrambs says that both sisters were burned to death in the temple of Hera on the orders of Laodamas, son of Eteokles; while Mimnermos [an elegiac poet, about 200 years before] says that Ismene on the one hand was killed by Tydeus, on the instructions of Athena, because she was having an affair with Theoklymenos . . . 2
and unfortunately the rest of this sentence is lost and we don’t know how, according to Mimnermos, Antigone (on the other hand) perished; but it does look as though Mimnermos got rid of both sisters even before their brothers had died at each other’s hands, while Ion kept them alive for years thereafter. Oh yes, and neither must we forget Euripides’ lost Antigone, in which Antigone became the wife of Haimon and bore him a son.3 But Antigone, before Sophocles, was rather a peripheral figure in the Theban saga. The suicide of Ajax, in contrast, was one of the most famous events in heroic myth, implicit in the Odyssey,4 explicit in two of the cyclic epics,5 repeatedly presented in art, lyric, and drama—and all the more striking because suicide is something that men in Greek myth hardly ever do. And yet there were those who told the story otherwise, according to the Hypothesis to Sophocles’ play: ‘Some say that Ajax was wounded by Paris and returned to the ships streaming blood; others that an oracle was given to the Trojans to throw mud over him, because he was invulnerable to iron, and thus he died; others that he took his own life, among whom is Sophocles.’ Nevertheless, the dogma that poets cannot alter the crucial elements of their myths is no modern invention. Its explicit formulation, indeed, goes back to the Poetics of Aristotle, from which comes the second of my three quotations. It is not possible to abolish the stories that have been handed down—for example, that of Klytaimestra being killed by Orestes or Eriphyle by [her 2
Hypothesis II (Pearson) to Antigone. The citations are Ion PMG 740 and Mimnermos fr. 21 West. 3 Hypothesis I (Pearson) to Sophocles’ Antigone. 4 Od. 11.543–64. 5 Aethiopis fr. 6 West; Little Iliad Arg. }1 and fr. 3 West.
‘They all knew how it was going to end’
211
son] Alkmaion; rather, the poet must himself find a way to make good use of the traditional material.6
Well, er, actually it is possible, because one poet very well known to Aristotle did in fact abolish both. Namely Homer. Regarding his treatment (or rather non-treatment) of the death of Eriphyle, I have not time to go into detail;7 but this is what is said in the Odyssey about the death of Klytaimestra—at any rate the fullest of the various accounts that are given (none of the others adds any fresh data):* Seven years Aigisthos ruled over Mykenai, rich in gold, after killing the son of Atreus, and the people were subdued under him; but in the eighth year evil came upon him in the shape of glorious Orestes returning from Athens, and he killed his father’s murderer, the guileful Aigisthos, who had killed his famous father. Then after killing him, he was giving a funeral feast to the Argives for his hateful mother and for the cowardly Aigisthos; and on the same day there came to him Menelaos of the great war-cry bringing many possessions, all the burden that his ships bore.8
The ancient commentators recognize that Homer either does not know of, or is deliberately glossing over, the killing of Klytaimestra by her son.9 Evasion is more likely than ignorance, since Homer does mention Klytaimestra’s funeral and one would have expected him to mention her death unless he had a special reason not to; and we can add that Homer is not only evasive but also misleading, since (as Jebb
6
Arist. Poet. 1453b22–5. (1) Od. 11.326–7: Odysseus says that among the women he saw in the underworld was ‘hateful Eriphyle, who took precious gold for the life of her own husband’ (Amphiaraos)—and at that point he breaks off his whole narrative and suggests to his Phaiakian audience that it is time to go to bed. (2) Od. 15.244–9: the poet, speaking of Amphiaraos in the course of an elaborate genealogy, says he ‘did not reach the portal of old age, but perished at Thebes because of the bribes his wife took; and his sons were Alkmaion and Amphilochos’—and at that point he abruptly moves to another branch of the family, descended from a brother of Amphiaraos’ grandfather. 8 Od. 3.304–12. 9 They are cited at the beginning of Ch. 16 below. 7
212
‘They all knew how it was going to end’
pointed out a century ago)10 one can hardly imagine ‘the Argives’ attending a feast in the house of a matricide—rather the impression is being given, certainly to Telemachos and possibly also to Homer’s audience, that Klytaimestra died not by Orestes’ hand but by her own. I argue in Chapter 16 below that Sophocles in Electra has taken note of this bit of Homeric trickery and has done something similar himself: at first he encourages us to believe that in his version, Klytaimestra is not going to be killed by Orestes at all, only Aigisthos is; once Klytaimestra has appeared on stage and we have come to know her character and Electra’s hatred of her, that idea is soon buried, but a new red herring is created in its place—the possibility that Electra herself may take part in the killing of her mother. And in the end she does so in spirit, though not physically; so that while in literal terms Sophocles has not changed the myth at all, by encouraging his audience to expect that he may do so he has created a whole series of effects that would have been quite impossible if they had known for certain that nothing of the sort could possibly happen. The long and short of it is that within very wide limits, poets, and particularly dramatists, treat myth as they please: almost the only constraint is that the de´nouement of a play must not be radically incompatible with the well-known course of subsequent events—so that it would not, for example, have been possible for Euripides to have Achilles sacrifice his life at Aulis in an attempt to save Iphigeneia. Even this constraint is partly nullified by the deus ex machina convention. In Sophocles’ Philoktetes, Neoptolemos is about to take Philoktetes home to Greece, thus giving Philoktetes’ enemies their just deserts for their treatment of him—but also making the capture of Troy impossible—when they are stopped by the appearance of Herakles telling them they must go to Troy and win glory there (though even so, Sophocles doesn’t bother to tidy up on Odysseus, and so far as the play goes he seems to be left stranded on the island). Next year,11 in Euripides’ Orestes, Orestes and his companions have killed Helen, taken her daughter Hermione hostage, and are just about to burn down the palace of Argos when Apollo appears, 10
Jebb (1894: p. xi n. 3). Philoktetes was produced in 409 (Hypothesis I [Pearson] to Philoktetes), Orestes in 408 ( Eur. Or. 371). 11
‘They all knew how it was going to end’
213
announces that Helen is not dead at all, and gives a series of orders which cancel the entire action of the play as if it had never happened (I am not sure there was anything like this in the known history of Western drama until the scriptwriters of Dallas nullified an entire series by revealing that all the events in it had been part of a dream of Bobby Ewing’s*). And in fact the one thing about every tragedy of which the spectators could be certain, when they sat down to watch it, was that in at least some important respects it would be different from any treatment of the same story that they had seen or heard of before. Only, they did not know what the differences would be, and therefore, paradoxically, their presumption would be, until they were given positive reason for believing otherwise, that any particular event or episode would be treated in the manner most familiar from tradition. These principles clearly gave the dramatist wide scope for misleading and surprising his audience. And it is in the light of these principles that I’d like to take a look today at Oedipus the King. I should add that what I’m going to be saying about this play owes a lot to numerous discussions with my former research student, David Fitzpatrick. I will start (at last) with my third quotation, from a book which like Fowler’s has a special place in my heart, but for a different reason: it is the first book I ever reviewed. In 1971 Philip Vellacott, whose name some of you will know as the first Penguin translator of Aeschylus, Euripides, and Menander, published a book called Sophocles and Oedipus12 in which he attempted to show that Sophocles in Oedipus the King portrays Oedipus as having known for many years that he had killed Laios and that Laios was his father. (Twenty years later, contrariwise, Frederick Ahl13 tried to show— oddly enough by exactly the same method—that Sophocles’ Oedipus is an innocent man who perpetrates a terrible miscarriage of justice against himself.)14 Vellacott recognized that his interpretation 12 Vellacott (1971), which was reviewed by Sommerstein (1972) in Farrago, the journal of the Cambridge University Classical Society. 13 Ahl (1991). 14 Vellacott’s thesis has now been put forward again, independently it seems and in a slightly different form, by Maiullari (1999). In Maiullari’s version, the killing of Laios was the work of a conspiracy involving Kreon and Teiresias (who supplied a gang to assist Oedipus), and they have now turned against their former partner with
214
‘They all knew how it was going to end’
involved a number of difficulties. One of them was that by the beginning of the discovery scene (the confrontation of the two shepherds) there would be nothing left to discover, since on Vellacott’s reconstruction of events before and during the play, not only Iokaste but just about everyone else—Oedipus, Kreon, the Chorus— would by then be aware of the truth. He admitted that this made the discovery scene redundant, but explained that ‘the self-blinding as a result of the discovery was an unalterable part of the saga and Sophocles was bound to include it’.15 Unfortunately for Vellacott, we know that this was not the case. Euripides also wrote a play about Oedipus. It has not survived, but we know that in the story as told in this play, Oedipus did not blind himself at all: he was blinded forcibly by ‘the servants of Laios’16— probably, though not certainly, after he had been revealed as Laios’ killer but before he was known to be Laios’ son. What is more, there was yet another version of the story, reported by an ancient commentator on Euripides’ Phoenician Maidens, in which Oedipus was already blind before he ever came to Thebes—blinded, back home in Corinth, by his adoptive father Polybos, when Polybos learned of the prediction that Oedipus would kill his father.17 Euripides’ Oedipus, to be sure, was later than Sophocles’; and the version in which Oedipus was blinded in childhood or adolescence may for all we know have been later still. But like the other evidence we have seen, they do show that we have to be very careful indeed what we declare to be an ‘unalterable part of the saga’. Sophocles in this play has actually done a great deal, both in the way of making actual innovations in the story and of creating red herrings by playing with the possibility of innovation, but I want to concentrate on just two issues. The second, to which I will return, is whether we do all know how it will end—I mean, actually end. The first is a very simple question: who is Oedipus’ wife?
the aim of putting Kreon on the throne. Like Vellacott, he claims (in the manner of A. W. Verrall) that the action is designed to be understood in one way by the simple spectator and in quite another by the sophisticated. 15 Vellacott (1971: 234). 16 Eur. Phoen. 61, citing Eur. fr. 541. 17 Eur. Phoen. 26.
‘They all knew how it was going to end’
215
Now come off it, you’ll say, any fule knos that.18 Oedipus’ wife is Iokaste, his mother, the widow of Laios. He married her when he came to Thebes after Laios’ death, they had four children, and when the truth was discovered she hanged herself. That, indeed, is how it turns out in Sophocles. That may also have been, more or less, how it was in Aeschylus’ Theban trilogy, though we cannot be sure: at any rate there is no evidence inconsistent with such an assumption, and some that supports it. But that was not the only version that existed. The Odyssey, for example, is out on a limb: Oedipus married his mother (here called Epikaste), the gods ‘quickly’ made it known to men, Epikaste hanged herself and cursed her son, but he continued to rule over Thebes.19 The clear implication is that Oedipus’ children were born of some other mother, and according to Pausanias the cyclic epic, the Oidipodeia, said just that; the name of the second wife, according to him, was Euryganeia.20 Not everyone believes that Pausanias got it right, but it is independently known that a version of the story like this existed by the fifth century. In some accounts Euryganeia belongs to an entirely different family; in others, however, she is the sister of Iokaste.21 Remember this; it will prove important. And don’t worry about her being Oedipus’ aunt; Greeks never regarded unions like that as incestuous. Very well, then: in Sophocles’ play, how soon do we know that Oedipus is still married to his mother? Because if she is dead (or, as one version apparently had it, divorced) we may find ourselves in a quite different dramatic situation—indeed in any one of several situations, depending on the answer to supplementary questions such as (1) whether (and whom) Oedipus has remarried; (2) if he has, who is the mother of Eteokles and the rest; (3) if they are the children of Iokaste, whether the new wife has had other children . . . and so on and so on. In some of these situations, the potential excitement of the story may well be greater, not least because of its
18
The orthography is intentional; see Willans and Searle (1954: 35). Od. 11.271–80. 20 Paus. 9.5.10–11. 21 Eur. Phoen. 53 gives this account with both versions of Euryganeia’s parentage, attributing the first to the fifth-century mythographer Pherecydes (fr. 95 Fowler). 19
216
‘They all knew how it was going to end’
possible unfamiliarity; in all of them, the horror will be less, because the factor of ongoing incest will be missing. Let us proceed through the play, taking care not to make use of information we don’t yet possess.22 The first indirect reference to Oedipus’ marriage comes at lines 69–70 where Oedipus says he has sent ‘Kreon, son of Menoikeus, my brother-in-law’ to enquire of the Delphic oracle. This designation, however, does not necessarily prove that Oedipus is still married to Kreon’s sister—and even if he is, we do not know which sister (you remember that in one version his second wife was another sister of Kreon). And when we hear what the oracle says, it encourages us to concentrate on the killing of Laios. After the first song of the chorus, Oedipus makes the famous proclamation and curse that eventually turns out to have been made against himself—for the murder of Laios. In the course of this he refers again to his marriage, and his words (258–65) require close examination. I translate them literally, except for two which I do not translate at all. Now, since I happen to have the rulership which he formerly had, and to have a marriage and homosporos wife, and koina of common children would have been born if his offspring had not been unfortunate—but in fact fortune leaped upon his head; wherefore I shall fight on his behalf as if he were my father . . .
The irony of this is obvious. What is less obvious is that Sophocles has here taken great care not to say anything that makes it certain that Oedipus is currently married to Iokaste. The proof of this lies in the two words that I have not translated. Oedipus, in the fictive reality (if that is an intelligible expression), knows that he is married to the widow of Laios, and for him homosporos, literally ‘same-sown’, means that the same woman has been the metaphorical soil in which both Laios and he have sown their seed. Sophocles knows, furthermore, that the widow of Laios is also the mother of Oedipus, and he is going to use this word homosporos again, in the mouth of Teiresias (460), to refer to that incest: the woman who has received Oedipus’ seed also 22
For the importance of this principle—too often forgotten in discussion not only of OT but of many other texts—see Bain (1979).
‘They all knew how it was going to end’
217
received the seed that made Oedipus. But the regular meaning of homosporos in tragedy is something different from both of these, namely ‘sibling’, and therefore, for the audience, Oedipus may be saying no more than that his current wife (note the present tense) is the sister of Laios’ wife. Similar ambiguities lurk in the involved phrase ‘koina of common children’. Why not say just ‘common children’—meaning that if Laios had had surviving children, they would have been half-siblings of Oedipus’ children? That, after all, appears to be what Oedipus himself means by the phrase. But to let him express it like that would destroy the ambiguities, which are again triple. Sophocles, who knows everything, is hinting at the horrendous perversions of kinship that this family has seen, which have made Oedipus ‘brother to his children, husband to his mother, successor to his father’s bed’ (457–60, cf. 1403–8), one who has caused Iokaste to bear ‘a husband by her husband and children by her child’ (1250). But in a theatre spectator’s mind the phrase could also mean ‘common of common children’: in other words, if Laios had had surviving children, Oedipus’ children would have been their cousins—as would have been the case if Oedipus is now married to another daughter of Menoikeus. Teiresias has meanwhile been sent for. Asked to give information about the killers of Laios, he refuses; Oedipus can see only one explanation for so antisocial a response, and furiously accuses Teiresias of being behind the murder himself; whereat Teiresias tells him that, on the contrary, he is ‘the unholy polluter of this land’, that he is the killer of Laios and moreover that ‘you are living unawares in a most shameful relationship with those closest to you, and cannot see the plight in which you are’ (366–7). To those who know all (the author and the gods), the meaning of that is obvious. To Oedipus it may or may not remind him of the oracle he had received at Delphi many years ago, but if it does he knows it to be nonsense; if it doesn’t, it will be merely incomprehensible. The audience as yet know nothing of this oracle: there is no evidence whatever that such an oracle, given to Oedipus (as opposed to Laios), had been imagined by anyone before Sophocles, and when the audience do hear of it, which is not until line 791, it will force them to look at many aspects of the story in an entirely new light. Of course they will recognize that
218
‘They all knew how it was going to end’
Teiresias is referring here to Oedipus’ incest; but his words still do not reveal whether that incest is a continuing fact. For all they know, the vague plural ‘those closest to you’ (tois philtatois) might refer only to the children. That still eliminates quite a number of possible scenarios; it means for sure that Oedipus’ children, or at least some of them, were the fruit of his union with his mother. Hence there are now only two main possibilities—although one of them still has a number of sub-options: 1. Oedipus married Iokaste and had children by her; subsequently she died (perhaps by suicide, perhaps not) and he then married her sister; if her death resulted from her discovering that Oedipus was her son, she either did not inform Oedipus or he knows and has kept it secret. 2. Oedipus married Iokaste and had children by her; she is still alive and still his wife; neither has any inkling of the truth about Oedipus’ birth. Teiresias will have more to say on this subject, but it will not shed more light. His words in 413–14—‘You cannot see what trouble you are in, nor where you are living, nor with whom [plural] you share your home’—do no more than repeat what he has already said. The seas and mountains, he says, will resound with Oedipus’ laments when he becomes aware of ‘the marriage into which you sailed with a fair wind, a treacherous harbour to your house’ (422–3) and of a situation that ‘will put you on a level with your children as well as with yourself ’ (425); but again, none of that tells us anything new either. And the same applies even to his final words, which have worried many critics so much that they have wanted to remove Oedipus from the scene so that he doesn’t hear them: ‘He shall be revealed as being both brother and father to the children with whom he lives; both son and husband to the woman who bore him; both murderer and marital supplanter (homosporos) of his father’ (457–60). You notice how careful Sophocles still is: he speaks of the children with whom Oedipus lives, but not of the wife with whom he lives. By now, of course, if we knew about the oracle, we’d be certain that Oedipus had remembered it, particularly after he’d asked Teiresias who his father was (437); since we don’t, we are left to the pleasure and horror of a situation in which we know that Teiresias
‘They all knew how it was going to end’
219
is telling the truth and know also (as we think) that Oedipus has no chance of perceiving this. This is how things continue to stand during the confrontation between Oedipus and Kreon. Kreon points out that his sister is married to Oedipus and that she, and Kreon himself, are virtually equals in power with the king (577–81); but as before, he could as well be talking about Euryganeia as Iokaste. The vital information comes at last, quite casually, from the chorus-leader. It is not often in Greek tragedy that words spoken by this very functional personage change our whole perspective on a play, but this is one of those cases. Oedipus and Kreon are quarrelling furiously, when the palace door opens and a middle-aged woman of obviously high rank comes out. And the chorus-leader says: ‘Stop, my lords. I see here, coming to you from the house, at a timely moment, Iokaste, with whose help you should now come to a sensible settlement of this present quarrel’ (631–3). Not Euryganeia; Iokaste. And in case there were any doubt about the matter, Iokaste proceeds to give instructions to both men in a manner that makes it quite clear who is the most powerful person in this family, and Kreon a moment later refers to Oedipus explicitly as her husband (639). This is the first of a series of revelations that will reorient the whole play as we enter its second half; the others, which follow Kreon’s reprieve and departure, are better known—the story of Laios’ death at a spot where three roads met, of the exposure of his child, of Oedipus’ youth, the oracle he received, and his fatal encounter with a party of travellers. But it is before all these things are revealed that Sophocles, having toyed with us for so long, finally lays his first card on the table by making it clear to us at last who Oedipus’ wife actually is. He does something similar, we may observe, in Antigone, where he allows the plot to proceed a considerable distance, and Antigone to be condemned to death, before letting us know that Antigone is betrothed to the son of Kreon23—and leaving us to reflect on the fact that Kreon must have known this all along. So now we do all know how it is going to end. Or do we? We have been told by Teiresias (and Teiresias, we know, is always right) that by
23
We first learn this, from Ismene, at line 568.
220
‘They all knew how it was going to end’
the end of this day, Oedipus’ parricide and incest will be known by all, and he will be a blind man; we have not been told that Iokaste will be dead, but the suicide of disgraced or despairing women is so routine in Sophocles that we will probably not be tempted to suspect that he may keep Iokaste alive (though twenty years later, in The Phoenician Maidens, Euripides will do just that). But what will happen to Oedipus? This is a question that has left many scholars confused, and has led some, among them Jenny March and Roger Dawe, to conclude that the ending of Oedipus the King has been fundamentally remodelled by some later producer.24 I hope to show that no such supposition is needed. The consensus of earlier accounts had him remaining in Thebes. In Homer, as we have seen, he continued to rule there; in the cyclic epic, the Thebais, he lived in retirement in the palace, maintained (somewhat grudgingly) by his sons;25 more than one source speaks of his funeral at Thebes,26 and the last genuine words of Aeschylus’ Seven imply that his sons will be laid to rest beside him.27 In addition, he is regularly spoken of as cursing his sons: in the cyclic epic because of their offensive behaviour towards him, in Aeschylus possibly for this reason, possibly because of his discovery that they are the fruit of incest, most likely (I believe) from a combination of both these causes, one coming on top of the other.28 From an early stage in Oedipus the King we are primed to expect something different. Apollo demands the death or banishment of the murderer of Laios (100), and these two alternatives are frequently brought to our attention again. Oedipus mentions them when explaining the situation to Teiresias (309) and twice accuses Kreon of plotting to have him killed or exiled (659, 669–70) while himself demanding a similar punishment for Kreon (622–3, 641). Meanwhile, however, Teiresias has specifically predicted exile, not death, for Oedipus (418, 454–6); and Teiresias, we know, is always right, so that’s what’s going to happen. We wonder where he will go, for 24
March (1987: 148–54); Dawe (2001). Thebais frr. 2, 3 West. 26 Il. 23.679–80; Hes. fr. 192 M–W ¼ 135 Most. 27 Aesch. Seven 1004.* 28 Depending on the text and interpretation of Seven 785–6, which I discuss in Sommerstein (1989a: 440–5). 25
‘They all knew how it was going to end’
221
Sophocles has not yet written Oedipus at Kolonos; but that he will go, there can be no doubt. Oedipus also wonders where he will go, once he begins to suspect that he may be the killer of Laios after all; he knows, at any rate, that it cannot be Corinth, for that (he believes) is where his father and mother are (823–7). But these reflections are stilled for the time being when it occurs to him (836–47) that he may possibly not be the killer of Laios after all (because the eyewitness spoke explicitly of an attack by robbers in the plural), and he then never has a moment to revert to them until after he has discovered the whole truth. Once that has happened, and Iokaste is dead and Oedipus selfblinded, the two options remain. To be sure, Oedipus rejected immediate suicide because he could not have borne, once in Hades, to look on the faces of those he had so terribly wronged (1371–5), but now he is blind that consideration no longer applies, and he begs the Thebans to ‘hide me somewhere abroad, or kill me, or throw me into the sea’ (1411–12). Death or exile are still the alternatives, and we know which will be chosen. For one thing, Teiresias has told us (and Teiresias is always right). For another, Oedipus’ terrible crimes were committed unwittingly; indeed on his story, which we are given no reason to doubt, his killing of Laios would by Athenian law have been no crime at all29—he had nothing to reproach himself for except that the dead man was his father. But whose will the decision be? Oedipus’ appeal to be banished or slain is made to the Theban elders, but it is answered by the new
29 Because Laios (together with one of his attendants) had been the first to use violence (804–5), even before making an assault on him which could easily have been fatal (807–9). We do not have an explicit statement of this aspect of Athenian homicide law (Antiphon 4.2.1 and 4.3.2 cannot be relied on as evidence for the Athenian legal provisions), but we do have good evidence about the way such cases were treated in practice. A certain Boiotos, when drunk at a symposium, struck a fellow diner, Euaion; Euaion killed Boiotos there and then. He was in the end convicted of murder, but only by a majority of one, and Demosthenes (21.71–5), who tells the story, clearly expects his jury to sympathize with Euaion. Much more would they have sympathized with a man who had been subjected to a murderous assault by the leader of a party of five men on a lonely country road; indeed there was a specific legal provision permitting a traveller waylaid by highwaymen to kill them without penalty (Dem. 23.53). In Oedipus at Colonus, Oedipus repeatedly insists, on these grounds, that he had done nothing criminal (OC 270–2, 546–8, 991–9).
222
‘They all knew how it was going to end’
king—who is Kreon. Can he resist the opportunity to take revenge on the man who so recently ordered his execution? To our surprise he at once says he has come neither to mock Oedipus nor to reproach him with anything about the past; to our greater surprise, he immediately orders Oedipus to be taken indoors, explaining that it is not proper that earth, sun, or rain should be exposed to such a pollution as he represents. From this point the option of death is forgotten, but Oedipus still pleads to be cast out of Thebes (1436–7). Kreon, however, wishes to enquire of Apollo first. Oedipus at least tacitly agrees (1446), but a moment later he is again begging to be allowed to wander on Mount Kithairon (1449–54). And then we hear about his sons—very briefly indeed: ‘As to my children—about the sons, Kreon, I tell you you need not worry. They are men, so wherever they are, they will never be short of a living’ (1460–1). More irony, of course, but no sign of a curse, and even perhaps a tacit abolition of it. The reference to being ‘short of a living’ will remind us that Oedipus’ sons quarrelled over the inheritance of his property, and we might reflect that that would be quite a sufficient motive for a quarrel without any need for a paternal curse. After that, for fifty lines and more, we concentrate entirely on the daughters, whom Oedipus entrusts to Kreon’s care—a scene of almost unbearable poignancy for anyone who has seen or heard of Antigone. Then Oedipus is again told to go into the house. He agrees on condition that he will afterwards be allowed to leave Thebes (1517–18); Kreon in turn agrees to this, on condition that Apollo gives his approval (1519–20); and Oedipus is taken inside, though not before being forcibly parted from Antigone and Ismene. And that is all. So what is going to happen? We have no idea. I’ve said more than once that we know Teiresias is always right; but didn’t he say that Oedipus would be expelled from Thebes this very day? Well, er, actually, he didn’t; what he said was that this day would destroy him (438) and that ‘at some time’ he would be driven from the land (417–18) and would wander as a blind beggar in foreign parts (454–6). That seems to be what’s going to happen, eventually; but it’s equally important that it hasn’t actually happened. Sophocles is thus enabled to have his cake and eat it. He has prepared us for an innovative exile, but he has not actually provided it; he has hinted that this Oedipus is not the sort of person to curse his sons, but he
‘They all knew how it was going to end’
223
has not actually deprived him of the opportunity to do so at some future time. And we may be left feeling that not only did we not after all know how it was going to end—we don’t even know now how it has ended. And how very, very appropriate that is for a drama that is all about a man who thought he knew everything and who has proved to know nothing.*30
ADDENDA p. 211 I have left this as I wrote it, but it is in fact misleading: Od. 3.304–12 is the only passage in the poem that makes any allusion at all to Clytaemestra’s death. Other references to the revenge of Orestes (1.29–47, 298–300) mention only the killing of Aegisthus (and 4.546–7 anticipates it). p. 213 Bobby Ewing’s dream: see http://www.ultimatedallas.com/ episodeguide/dreamzone.htm (accessed 11 Sept. 2009). The nullified series was that of 1985/6. p. 220 n. 27 to be precise, Seven 1004 implies only that it would be possible to bury Oedipus’ sons beside him. The speaker is saying that that would be painful for Oedipus, and one might well expect this realization to be followed by a decision to bury the sons in a different place; such a passage, if it existed, must have been cut out when the present spurious ending (1005–78) was added. Be all that as it may, the possibility of burying the sons beside the father could not even have been raised had not Oedipus’ tomb been in or very near Thebes. p. 223 the ending of OT has now been defended along similar lines by Burian (2009): it ‘remind[s] the audience of possible futures, but staunchly refuse[s] to choose among them’. 30 This lecture has been given at various times, in various forms, to various audiences; this version was given to a sixth-form conference organized by the Sussex Association of Classical Teachers at Roedean School in March 2006, and is reproduced here with minor editing.
16 Alternative scenarios in Sophocles’ Electra
It is not possible to abolish the stories that have been handed down—for example, that of Clytaemestra being killed by Orestes or Eriphyle by Alcmeon; rather, the poet must himself find a way to make good use of the traditional material.
Thus Aristotle (Poetics 1453b22–5); and we can safely assume that in all the Orestes- and Alcmeon-tragedies known to him, these basic mythical data were respected. It is, nevertheless, curious that he should have chosen, to illustrate his point, two items of myth which one poet at least, Homer, had to all appearance abolished or at least elided. As is well known, it is nowhere stated in the Odyssey how Clytaemestra died, and even the fact of her death is mentioned only once, indirectly, in a reference to the feast given by Orestes to the Argives for the funeral of her and Aegisthus (3.309–10). Aristarchus was uncertain whether this passage did or did not imply that Orestes had killed Clytaemestra as well as Aegisthus,1 and at least one other ancient scholar felt justified in affirming positively that Homer did not know of Orestes’ matricide;2 to mention the funeral without mentioning the death, however, suggests evasion more than it suggests ignorance, and most modern scholars suppose, with some
This chapter was originally published in Prometheus 23 (1977) 193–214. I am most grateful to Prof. Angelo Casanova for giving permission for this republication. I owe the original idea for the paper to discussions with my research student, Stephen Dailly, though its conclusions are rather different from those we reached then! 1 Od. 3.309–10 › b æÆæå çÅØ ‹Ø Øa ø ÆæıçÆ ÆØ ‹Ø ı ƺ `NªŁø fi ˚ºıÆØ æÆ· e b N ŒÆd e Oæı ¼Åº r ÆØ. 2 Od. 1.300 PŒ r › ØÅc e ˚ºıÆØ æÆ e F ÆØe æ .
Alternative scenarios in Electra
225
ancient support,3 that the poet knew of the matricide and deliberately suppressed it,4 because it would disturb the parallel he wished to draw between the constellations Agamemnon–Clytaemestra– [194]Aegisthus–Orestes and Odysseus–Penelope–Suitors–Telemachus and/or because it is the general policy both of the Iliad and of the Odyssey to suppress or tone down the more grotesque or atrocious features of inherited myth.5 The latter motive is the one that seems to have been at work in Odyssey 11.326–7 and 15.244–8, in both of which Eriphyle’s betrayal of her husband is mentioned but her death is not (even though Alcmeon is actually named in 15.248).6 After Homer, so far as we know, it is never questioned that, as a matter of mythic fact, Orestes did kill his mother. But if the tragic dramatists apparently refrained from ever altering this datum, they were under no obligation to refrain from playing with the possibility that it might be altered. The device of foreshadowing a major mythical innovation, and then not delivering it, is one that tragedy frequently employs. It can take three basic forms. 1. The mildest, and commonest, variant is that in which the characters, or some of them, anticipate a development which the 3 Od. 3.310 ç ÆØ Øa ø F ˇæı. e b ªaæ PçÅ æ r ‹Ø ŁÆł c ÅæÆ, e b Ł Æ Ææ ØÅ . 4 So e.g. Garvie (1986: pp. xi–xiii); and S. R. West in Heubeck, West, and Hainsworth (1988: 60). A somewhat different view is that of Davidson (1988: 51), who claims that ‘the matricide is implied’ in the reference to the double funeral; but if anything the passage insinuates, without asserting, the contrary. The very sentence that mentions the funeral says that it was held after Orestes had killed Aegisthus (› e
Œ Æ); moreover, as was pointed out by Jebb (1894: p. xi n. 3), ‘the Argives’ would hardly have been willing to come under the roof of a matricide and partake of his food. A hearer (such as Telemachus) who did not know the details for certain would be almost bound to infer that Clytaemestra had committed suicide, in grief or shame or despair, after the death of her lover. It is just possible that such a version of the myth was actually imagined by an archaic poet on the basis of the Homeric passage, and that Aesch. Cho. 978–9 ı Æ . . . ı ŁÆ ØÐŁÆØ is a detail ultimately derived from it. 5 On this policy see e.g. Griffin (1977), esp. 44. 6 Is it accidental that Odysseus breaks off his catalogue of women (11.328) at the very moment when he might have been about to speak of Eriphyle’s death as he had spoken of the deaths of Epicaste (277–9) and, just before, of Ariadne (324–5)? In this case, however, we have no evidence that any ancient scholar thought Homer had suppressed the matricide; both Od. 3.309–10 (citing Aristarchus) and Od. 15.248 say Homer ‘does not know’ of it.
226
Alternative scenarios in Electra
audience know to be contrary to established myth (and sometimes to indications already given within the play itself), and the audience’s perception of the characters’ error or blindness is exploited for dramatic and tragic effect. This is of course one of the basic forms of dramatic irony, and is found, for instance, in six of the seven surviving plays of Sophocles;7 it depends for its effectiveness on the audience not being deceived. 2. At the other extreme, as in Sophocles’ Philoctetes and Euripides’ Orestes, an entire plot may be on the point of reaching a de´nouement radically [195] contrary to tradition and threatening the integrity of large sections of the accepted corpus of myth, until the arbitrary intervention of a deus ex machina puts it back on its ‘proper’ course. In such cases the dramatist in a way has his cake and eats it: the innovative plot is allowed to run its full course, yet without invalidating the traditional story. In this case the effectiveness of the device requires that the audience should be deceived. While the innovative plot is progressing, they must be sufficiently captivated by it to disregard their ‘knowledge’ that in mythical ‘reality’ things turned out differently: Neoptolemus’ sacrifice of glory and acceptance of peril, for the sake of being true to his philia with Philoctetes, will be meaningless if the spectator is busy wondering how these developments are going to be reconciled with the mythical datum that both men went to Troy. 3. Intermediate between these two varieties is a third, with which this chapter will be mainly concerned, in which, while the course of
7 In each of his four earlier plays, the chorus at some point anticipate a happy outcome only to be tragically disillusioned (usually quite soon), and each time one of the principals shares in the misapprehension: Trach. 200ff. (chorus and Deianeira: safe and triumphant return of Heracles); Aj. 693ff. (chorus and Tecmessa, cf. 787–8, 807: Ajax’s supposed renunciation of suicide); Ant. 1115–54 (chorus and Creon: release of Antigone—but Teiresias has already foretold Haemon’s death, 1066–7); OT 1076–109 (chorus and Oedipus: Oedipus as child of Fortune or of a god—but Iocaste’s final words and exit, 1056–75, have shown that she realizes whose child he actually is). In Electra the effect is reversed, as the false tale of Orestes’ death is believed by everyone who hears it. At the end of Oedipus at Colonus Antigone asks for herself and her sister to be helped to return to Thebes in the hope of making peace between their brothers (OC 1769–72); we know—having been encouraged to think of Antigone by Polyneices’ request to his sisters (1407–10) to see to his burial if necessary—that this mission will end not only in failure but in Antigone’s death.
Alternative scenarios in Electra
227
the action itself is broadly in line with tradition, clues are planted by the author to mislead the audience into believing that he means to effect a major innovation; or alternatively, after encouraging the audience to expect an innovation of a particular kind, he surprises them with a quite different one. A well-known example of the latter pattern is the prologue of Euripides’ Hippolytus. Our evidence suggests8 that, in most earlier versions of the story of Phaedra and Hippolytus, Phaedra had killed herself after Hippolytus’ death,* when in some manner or other9 it had become known to Theseus and the world that she had not only fallen in love with Hippolytus but had made or authorized an adulterous proposition to him. The order of events was approximately: (a) Phaedra, rebuffed by Hippolytus, accuses him to Theseus of actual or attempted rape; (b) Theseus curses Hippolytus and he is killed; (c) the truth about Phaedra’s passion is revealed to Theseus; (d) Phaedra takes her own life. At the beginning of Euripides’ second Hippolytus, as always, the audience will have been expecting a treatment of the story that will be to some degree novel,10 but the extent and nature of the innovations will have been [196] unknown to them. Aphrodite considerately tells them what will happen: Phaedra, though smitten with love for Hippolytus, is keeping silent and confiding in no one; however— that is not the destined outcome of this passion; I will reveal the matter to Theseus, and it will be brought into the open. And the young man who is my
8 See Barrett (1964: 6–45); Halleran (1995: 21–37). I do not wish to be taken as expressing any view on the suggestion by Gibert (1997) that the surviving Hippolytus of Euripides may for all we know be the earlier rather than the later of his two plays of that name.* I presuppose only that the audience of the surviving play came to it with some knowledge of previous treatments of the myth; it may have been an earlier play by Euripides that was uppermost in their minds, or it may have been Sophocles’ Phaedra, or it may even have been an epic Theseis. 9 In some versions possibly through her own confession, as in Seneca’s Phaedra (1159–200). 10 Thus in one sense the oft-repeated cliche´ that ‘the facts of most Greek plays were not a matter for invention, but were part of every Athenian child’s store of legend’ (H. W. Fowler [1926] 1965: 305, s.v. irony) is the exact opposite of the truth: the one thing of which the spectators could be certain (and of which the dramatist knew they would be certain) was that the story they were going to see enacted would not be identical with any version of ‘the same’ myth that they had seen or heard before.
228
Alternative scenarios in Electra
enemy will be slain by his father with the curses which the sea-lord Poseidon granted him . . . and Phaedra will perish with a good name, but will nevertheless perish. (Hipp. 41–8)
In terms of the older tale, Aphrodite mentions coming events in the order (c–b–d), indicating one other modification (that Phaedra will die ‘with a good name’), and omits (a) altogether. This may well bewilder the spectator. In its context, ‘the matter’ (æAª Æ) ought to refer to ‘this passion’; if Theseus knows about that before he has cursed his son, how comes it that he utters the curse at all, and how can Phaedra possibly die with a good name? Again, no mention has been made of the rape allegation, and Theseus’ early knowledge of Phaedra’s passion would seem to leave no place for it: what entirely new twist, then, is Euripides meaning to substitute for it? Only as the action develops will it be realized how Euripides has played fast and loose with his audience. Aphrodite has not told any lies, but she has not told the whole truth, and what she has told she has put in a misleading order. The actual order of events turns out to be close to (d–a–b–c). The rape allegation is there after all, though it is made posthumously and Phaedra’s motives for it are in part11 different from those portrayed in earlier treatments. The curse and Hippolytus’ fatal injury occur, as tradition and logic require, before Theseus knows the truth, though he—and Hippolytus himself—are undeceived before Hippolytus dies. Aphrodite has led the spectator to expect far-reaching plot innovations; only one such innovation actually occurs (the retiming of Phaedra’s suicide), and it occurs contrary to what Aphrodite’s words seemed clearly to imply. A more subtle use of the same device is found in Sophocles’ Trachiniae. There is reason to believe that before Sophocles, Deianeira had been thought of as an active, assertive woman, capable of engaging in war, who resented Heracles’ continual infidelities and, when he capped them all by bringing Iole home, killed him with full intent 11 Only in part, for she is still concerned, as earlier Phaedras must have been, to avenge an insult (728–31) and to protect her reputation by forestalling an accusation by Hippolytus (689–92, 720–1)—though in this version his accusation, if made, will be a false one, and the practical benefits of Phaedra’s preserved good name will accrue not to her but to her children (717, cf. 421–5), the Athenian heroes Acamas and Demophon.
Alternative scenarios in Electra
229
and premeditation.12 Through the first 530 [197] lines of Trachiniae Sophocles does all he can to create a Deianeira who is the exact opposite of this. She has never taken an initiative of any kind in her life. She would have submitted to marriage with the repulsive Achelous if Heracles had not ‘set [her] free’ by overcoming Achelous in fight (9–21) while she sat at a distance, too terrified even to watch (21–5, cf. 523–8). She has accepted the frequent and lengthy absences of Heracles from home (28–35), and though the latest of them has lasted fifteen months and caused her great distress (37, 41–2, 46–51), she does not even think of taking active steps to seek news of him until prompted by her nurse (52–7). Lichas’ lies impose on her completely, and when she questions him about the noble-looking woman captive (who is in fact Iole) and gets answers that are plainly evasive (310–19) she does not press the matter; then, when her other informant (the angelos) tells her who the girl really is, the much younger women of the chorus curse Lichas (383–4) but Deianeira, the party actually injured, does not, and when Lichas returns and repeats his lie (401) it is not Deianeira but the angelos who confutes him. When Deianeira speaks to him again, she says not one word of anger either against him or against Heracles; she has no right, she says, to condemn either Heracles or Iole for being unable to resist the power of Eros (439–49), any more than she has ever condemned any of Heracles’ countless other mistresses (459–63). To do so would be to engage in a hopeless fight against a god (491–2). Both before and after she learned who Iole was, her overwhelming feeling towards her has been one of pity (298, 312, 330–1, 463–7); Lichas’ request to her to treat the woman kindly (486) is unnecessary because that was what she meant to do anyway (490). All the signs13 are that, having long since accepted her husband’s incurable promiscuity, she will quietly 12 The evidence is discussed by March (1987: 47–60). Gantz (1993: 864–5 nn. 88, 90) is sceptical but does not take account of evidence that Deianeira was often portrayed as a warlike character (Bacch. 5.165–75, çıa IºØªŒÆ to her brother Meleager; Ap.Rh. 1.1212; [Apollod.] 1.8.1; Nonnus 35.89–91). 13 One small hint is dropped just before the choral song, when Deianeira says that Lichas must take some gifts back with him (494–6); but even those who catch that hint, and realize that one of the gifts will probably be the poisoned chiton, will be most unlikely to fathom what could have made Deianeira send it (if with March 1987: 62–3 we assume that Bacch. 16 is later than Trach.).
230
Alternative scenarios in Electra
accept this latest manifestation of it as she has passively accepted every other misfortune of her life—and at this point the chorus are made to remind us of this by singing of the contest between Heracles and Achelous, during which Deianeira ‘sat waiting’ (525, 528) and then followed the victor ‘like a lonely heifer’ (530). And yet it is one of the certainties of myth that the capture of Oechalia and of Iole was almost immediately followed by Heracles’ death—and at the hands of Deianeira. How on earth will this Deianeira be capable of doing it, and what motive could she have that she has not already disavowed? We learn the answer shortly: she [198] too is not strong enough to resist Eros; she has decided to use magical means to win Heracles’ sole devotion—and the charm is a chiton (580) smeared with relics of Nessus and of the Hydra. Told so much, we know the rest at once. Returning now to the death of Clytaemestra, I wish to explore the ways in which Sophocles in Electra conditions his audience’s expectations concerning this event. Modern studies have expressed the most diverse views on this matter. On one side we are told that ‘the matricide . . . is kept before us throughout’,14 on the other that ‘we are never allowed to dwell in advance on the matricide[; i]n fact Klytaimestra’s murder is first spoken of . . . just a few verses before Orestes and Pylades go into the palace to kill her’.15 It would seem desirable, then, to examine closely the actual references in the play to the event, before it happens, always bearing in mind, in the first place, that the established and expected scenario was that Orestes, with Pylades, would kill Aegisthus and Clytaemestra,16 secondly that this was in principle capable of being modified in two ways: either Orestes might kill Aegisthus alone, with Clytaemestra committing suicide (as half-suggested in the Odyssey), or Clytaemestra might be killed by 14
Segal (1966: 474); cf. Sheppard (1927b: 164) and Bowra (1944: 218). March (1987: 105); cf. Owen (1927) and Ronnet (1969: 222). 16 This was clearly the canonical order, reflected in many artistic representations of the death of Aegisthus (see Prag 1985: 10–34 and Knoepfler 1993: 42–9) and taken for granted by Aeschylus—whose Orestes, on discovering that Clytaemestra is at home and Aegisthus out, does not think of doing anything but gain entry to the palace and await Aegisthus’ arrival; it is also the logical order, since Aegisthus, being a man, is presumed to be the more dangerous enemy, whom a tiny band of conspirators can hope to kill only if they take him by surprise. Sophocles’ reversal of the order is not prepared for in any way before it is announced at 1368 (‘now is the time to act; now Clytaemestra is alone . . .’). 15
Alternative scenarios in Electra
231
some person other than Orestes, say by Pylades alone or even by Electra.17 Sophocles, as we shall see, plays with both these options. References to the revenge in the prologue neither call in question, nor [199] explicitly confirm, the expectations about it with which the audience will have come to the play. The paidagogos speaks of having brought up Orestes as ‘an avenger of your father’s murder’ (14); Orestes himself says he asked the Delphic oracle ‘how I might gain revenge for my father from his murderers’ (33–4) and later expects to ‘shine like a star upon my enemies’ (66). Listeners who know, or think they know, that Orestes means to kill both his mother and Aegisthus, will understand these plural expressions as referring to the two destined victims; but the expressions used are so vague that if later on it becomes clear that Orestes actually has different intentions, we will not be able to complain of being misled. Neither the paidagogos nor Orestes has made any reference to Clytaemestra, or even to Aegisthus, as an individual. Nor can we make any deductions about the scope of Orestes’ plans from the fact that he reports Apollo as having called the vengeance justified18 (37, 70); Aeschylus, after all, 17
In Euripides’ Electra, Electra is a full participant in the murder and arguably the most guilty of the three: she urges a reluctant Orestes to go through with the act (962–84) although she, unlike him, had received no injunction from Apollo to do it (1303–4). I do not wish in this chapter to enter into the endlessly debated question of priority between these plays, except to remark that Sophocles’ raising, discussed below, of the possibility that Electra might have a role in Clytaemestra’s murder would seem very second-hand, rather than intriguingly novel, to an audience that had previously seen her actual role in Euripides’ play (cf. Cropp 1988: pp. xlix–l, who develops a rather different argument for the priority of Sophocles’ play also based on ‘the centrality of Electra’).* 18 For this purpose it makes no difference whether we read in 37 K Œı (codd.) or K Œı (Lange). Kells (1973: 82) argues that we cannot in fact be sure whether K Œı() is part of what the oracle said, or whether Orestes has inserted an epithet representing his own opinion into a paraphrase of the oracle’s response; but 70 (ŒfiÅ ŒÆŁÆæc æe Ł H ‰æ Å ) shows clearly that Orestes believes Apollo has endorsed the justice of his mission (see Erbse 1978: 286–7). Sheppard (1927a: 5) ingeniously observed that Apollo’s words could in principle be taken as mandating Orestes to carry out the killing that was righteous (sc. but not the one that was unrighteous); but what help does the context give a hearer (not reader) in extracting such an interpretation? Certainly not the fact, which Sheppard and others have found sinister, that Orestes had made up his mind to carry out the vengeance even before going to Delphi; the Aeschylean Orestes too feels it his duty to do the deed with or without an oracular command (Cho. 298–305). See further Bowra (1944: 216–17); Heath (1987: 136 n. 30); and Lefe`vre (1993: 43–4).
232
Alternative scenarios in Electra
had made Apollo say exactly that with specific reference to the killing of Clytaemestra, and claim the authority of Zeus for doing so (Eum. 614–18). At the end of the prologue, by default, we will still be expecting the traditional outcome. From this point to the recognition in 1211–26 Orestes’ current revenge-plan remains completely unknown to the other characters, affecting the action only through the impact of the false tale of his death, and everything we hear about a possible revenge-killing arises entirely from the hopes of Electra and the chorus, and the fears of Clytaemestra. And the first two things we hear about it are both in different ways misleading. Electra prays to the underworld gods, especially the Erinyes, to ‘avenge the murder of our father, and send me my brother’ (115–17). By making the two petitions in this order, she seems to be implying that, like her Aeschylean [200] counterpart (Cho. 138–44, cf. Cho. 114–21, 160–3), she does not expect, or wish, that her brother should be the one who carries out the vengeance.19 As in Aeschylus, we know that if she hopes for an avenger other than Orestes, she is mistaken; what we do not know, yet, is that we will be mistaken if we make the obvious deduction from her phraseology—this will become apparent only at 303–4. A much more important false clue is planted, immediately afterwards, in the first strophe sung by the chorus. The greater part of this stanza focuses on Clytaemestra, as the chorus first address Electra as ‘child of a most wretched mother’ (121–2) and then recall the death of Agamemnon ‘most impiously caught by the deception of your guileful mother, and betrayed by her20 evil hand’ (124–6). Yet the prayer with which they end is: ‘May he that brought these things about (› æ ) perish, if it is proper for me to speak thus!’ In
19 That this is what the Cho. passages imply was seen by Denniston (1939: p. xx) (cf. Sommerstein 1980b: 65). Garvie (1986) seems to take one view on 119 and another on 144, and makes no comment on the incongruity of 160–3 (a æıŁ c I æ wielding both sword and bow) if meant as a description of Orestes. 20 The text does not explicitly specify the owner of the ŒÆŒA fi å Øæ, but æ is more appropriate in relation to a treacherous ‘friend’ than to a hereditary enemy like Aegisthus.
Alternative scenarios in Electra
233
speaking of the murder the chorus refer only to Clytaemestra21—but then they curse Aegisthus.22 We can if we wish dream up reasons why they should choose to do this,23 but it is more important to attempt to explain why [201] Sophocles chose to make them do so. The nearcontradiction between the beginning and the end of the strophe cannot have been meant to pass unnoticed; and the listener who notices it can hardly take it other than as a hint from the dramatist that in this play Aegisthus will be the main, perhaps even the sole, victim of the avengers. After all, if those who are not Clytaemestra’s children are not prepared even to pray for her death, how much less will those who are her children be willing actually to kill her! A listener who draws this inference will seek to reinterpret earlier passages in the light of it, and will not find this difficult. One can, after all, take vengeance on one’s enemies, or on the murderers of one’s father, without necessarily killing both of them, particularly if it was Aegisthus alone who had actually struck the fatal blow(s);24 and 21
Whereas Electra (97–9) had spoken of her and Aegisthus as joint perpetrators. Most interpreters pass over this anomaly in silence. It is noted by WinningtonIngram (1980: 335), but it is not clear what significance he sees in it; nor do I understand what view Lefe`vre (1993: 25) is trying to take. Jebb (1894) and Kamerbeek (1974) ad loc. both suggest that the masculine singular may be ‘general, including both the authors of the crime’ (Jebb); but if that was what Sophocles meant, why did he not use the plural as he does almost everywhere else in this part of the play? As an alternative, Jebb thinks of taking the phrase as referring to Clytaemestra alone, comparing Ant. 464 (and indirectly Eur. Med. 1018); but it is one thing, as in those passages, to use the masculine in reference to ‘any person in the position of X’ even when X happens to be a woman, and quite another to use the masculine singular in reference to a specific, known female person.* 23 So Burton (1980: 186) suggests that they curse Aegisthus because they are ‘the wives of citizens who form a party in opposition’; but they are cursing him as a murderer, not as a tyrant! One might equally well (or equally ill) speculate that the women are reluctant to curse another woman (cf. C. F. L. Austin 1987: 78, on Ar. Thesm. 367) or that they are uncertain whether Electra would take kindly to a curse upon her mother (so Paulsen 1989: 30 n. 13). 24 On this matter—regarding which earlier accounts, literary and artistic, varied greatly—we are never in the play told anything definite (so rightly Machin 1981: 213–14), but nothing that we are told is inconsistent with the possibility that Aegisthus was the sole ÆPå Øæ, ÆP Å, ÆºÆ ÆØÐ (cf. 272, 587, 955; these terms are never applied to Clytaemestra). Electra in 97–9 spoke of two murderers but of only one weapon, and the chorus’s reference to Clytaemestra’s ‘evil hand’ in 126 might allude e.g. to her entrapment of Agamemnon in the famous enveloping robe (cf. Aesch. Ag. 1107–28 [note å dæ KŒ å æ 1110–11], 1382–3, Cho. 980–4, 997–1004). Aeschylus makes Clytaemestra the sole killer, with Aegisthus not even 22
234
Alternative scenarios in Electra
çƪ (lit. ‘slaughters’) in 37 might be a poetic plural referring to a single killing.25 Perhaps then this is going to be a Homeric Orestes, who kills only Aegisthus? We must wait and see, remembering that for the time being any ambiguous or vague expressions are likely to be interpreted in the light of this apparent signal, and secondly that even if it does become clear that Electra wants her mother as well as Aegisthus killed, that does not tell us that such is Orestes’ actual intention. And the next few references to the vengeance, all from Electra, are indeed ambiguous or vague. In 205–12 Electra sings of the ‘twin hands’ that inflicted death on her father and ruined her own life, and prays that ‘the great god of Olympus’ may see that the murderers ‘suffer sufferings in requital’; in 245–50 she asserts that all human decency (ÆN) and piety would be at an end if after such a death ‘they are not afterwards to pay the penalty of retaliatory killing ( c . . . ı’ I Øç ı ŒÆ)’; in 303–4, after a lengthy account of the hybris of Clytaemestra and Aegisthus and their op-[202]pression of her, she bemoans the failure of Orestes to ‘come and stop these things’. In none of these passages is either murderer singled out as an individual, and when they are mentioned separately, in contexts not concerned with the vengeance, the hostility expressed towards each of them appears about equal.26 And when a little later Electra asks Chrysothemis to pray that Orestes should ‘live, get the upper hand, and plant his foot upon his enemies’ (453–6), we can hardly be
present; Euripides is inconsistent—at one point, as (on the above interpretation) in Sophocles, Agamemnon is said to have met his death by ‘the guile of Clytaemestra and the hand of Aegisthus’ (Eur. El. 9–10, cf. 162–6), at another Clytaemestra herself is portrayed as having killed him with an axe (Eur. El. 1159–60). 25 Cf. Aesch. Eum. 450, Eur. Hec. 522. 26 In Electra’s account of the usurpers’ behaviour (266–302) Clytaemestra is given more space (21.4 lines against 9.6 for Aegisthus), and Electra twice quotes what purport to be her actual words; on the other hand the account begins and ends with Aegisthus, and the ending (300–2), with its accumulation of pejorative expressions in an asyndetic series, makes a particularly vicious climax. In the next hundred lines references to Clytaemestra as the common parent of Electra and Chrysothemis (326), whom the former accuses the latter of supporting in preference to her father (342, 366–7), are balanced by references to Aegisthus as the effective ruler, who, when at home, keeps Electra strictly confined within the palace (310–16), and whose return will be the signal for her to be imprisoned in an underground dungeon (379–86).
Alternative scenarios in Electra
235
meant to suppose that she is asking her cautious, not to say timid,27 sister to pray that Orestes should kill his mother, and that the latter is agreeing to do so. By this time, however, the spotlight of interest is already falling more and more on Clytaemestra. The process begins with Chrysothemis’ report (417–23) of her mother’s ominous dream. Yet the dream itself, considered both in its own right and in relation to its literary precursors, still tends to point, if anything, away from the idea of matricide. Both the earlier versions of the dream that we know of, that of Stesichorus and that of Aeschylus, had been bloody. In Stesichorus (PMG 219) Clytaemestra had seen ‘a snake, the top of its head all bloodied, out of which appeared the king, the son of Pleisthenes’;28 this is unlikely to have been the whole of what she saw, but we have no clue to what followed. In Aeschylus (Cho. 527–35, 543–50) she gave birth to a snake and suckled it, and it caused blood to flow from her breast—and in case we are in any doubt as to the significance of the dream, Orestes at once interprets it to mean that ‘I am turning into a serpent and killing her’. In Sophocles there is no serpent and no blood. In the dream Agamemnon returned to the upper world, in his own person, and ‘took the [203] sceptre that was carried formerly by him and now by Aegisthus, and fixed it in the ground by the hearth; and up from it sprouted a burgeoning bough, by which the whole land of the Mycenaeans came to be shaded’. This dream clearly portends that Agamemnon’s son will regain his royal rights and become a beneficent ruler of his country;29 it carries no direct suggestion of any bloodshed or violence, and Clytaemestra herself does not figure in it at all.30 Of course it is significant that 27 Note her quite unnecessary request (468–9) that the sympathetic chorus should not divulge what she has done. 28 That is, in the dream the serpent was transformed into Agamemnon; alternatively one might understand Æغ f —º ØŁ Æ as ‘a king, a descendant of Pleisthenes’ and suppose that the serpent engendered an Orestes restored to his royal rights, but ‘it would be curious for even . . . Stesichorus . . . to refer to [Orestes] as a “Pleisthenid king”, when Agamemnon himself is always thought of first and foremost as a king, and the expression would most happily fit him’ (Prag 1985: 74, citing Ibycus PMG 282 (a) 21). 29 For the image of the tree or bough cf. Aesch. Ag. 966–7, Hdt. 1.108, 7.19.1. 30 Unless, as suggested by Kamerbeek (1974) ad loc., the ‘hearth’ where Agamemnon plants the sceptre is a metaphor for his wife’s body (cf. Aesch. Ag. 1435–6 and
236
Alternative scenarios in Electra
the dream has appeared to her and not to someone else, and the restoration of Agamemnon’s son does bode disaster for her, but there is nothing in the dream, as there was in the Aeschylean dream, to indicate that she will perish at that son’s hands. Thus the dream is likely to confirm our suspicions that the de´nouement will be a nontraditional one. The choral song that soon follows conforms to the now familiar pattern: the chorus are certain that the dream will find fulfilment against ‘those who did the deed and shared in doing it’ (ØÐ æHØ ŒÆd ı æHØ , 497), avenging the events that followed from a murderous sexual relationship (493–4) described in language so obscure31 that it is impossible to tell whether the relationship meant is that between Agamemnon and Clytaemestra or that between Clytaemestra and Aegisthus.* Clytaemestra now appears on stage for the first time. Her very presence, and the obvious mutual loathing between her and Electra, continue the trend whereby attention is being concentrated upon her as Electra’s and Orestes’ main adversary; and in her first speech (516–51) she consistently speaks of herself as having killed Agamemnon, without mentioning Aegisthus once in connection with the murder.32 It is in Electra’s reply that Clytaemestra’s death, in revenge for Agamemnon’s, is specifically envisaged for the first [204] time in the play.33 At first it is envisaged only hypothetically: Electra, granting for the sake of argument that Agamemnon had sacrificed Hdt. 1.108 KŒ H ÆNø B ŁıªÆæ, and see Henderson 1991a: 143 }164); but note that the bough is said to sprout not from the hearth but from the sceptre (421 KŒ . . . F’). Agamemnon is, to be sure, performing a masculine act of reproduction with a quasi-phallic sceptre; but his ıæÆ › ØºÆ is not with Clytaemestra, who only sees it (417 NØ ØÐ ), but directly with the earthen floor of his home which nurtures the offspring of that sceptre. Orestes is the child of the house; he is not, in the dream, the child of Clytaemestra. There is an excellent interpretation of the dream, with strong emphasis on ‘the restoration of the interrupted [male] descent line’, by Des Bouvrie (1990: 264). 31 ¼º Œæ’ ¼ ı çÆ ªaæ KÆ ØÆØç ø ª ø ± غº ÆŁ’ xØ P Ł Ø. 32 Cf. Friis Johansen (1964: 15–16). 33 I ignore (despite Sheppard 1927b: 164) the suggestion in 435–8 that Clytaemestra’s offerings, instead of being placed in Agamemnon’s tomb, should be buried in the ground to be ‘kept below as possessions for her when she dies’; the context has no connection with the vengeance, and Electra’s implicit point is presumably that when Clytaemestra dies (no matter how) it will be as well for her to have provided graveofferings for herself in advance, since no one else will want to do so.
Alternative scenarios in Electra
237
Iphigeneia without necessity in order to please Menelaus, asserts (580–3) that this would still not justify his murder: ‘Take care that in laying down this law for humanity you do not lay up pain and repentance for yourself. If we are going to kill one person in return for another, you would be the first to die, if you got your just deserts.’ Electra’s use here of the first person plural is the first hint34 of another possible alternative scenario: that she might have a hand in the death of Clytaemestra. It prepares for a very important moment in the play, later in the same speech: the first clear declaration that matricide is the intention of at least one of Clytaemestra’s children. Electra has by now moved from her original theme, of establishing that Clytaemestra had no shred of excuse in justice or honour for killing Agamemnon, and is recriminating with her over her unmotherly behaviour towards her children; and in the course of this she says: ‘You have often accused me of bringing him [Orestes] up to be an avenger ( ØæÆ, lit. ‘defiler’) against you; well, if I had the strength, I would be doing that (’ . . . æø ¼ ), I assure you’ (603–5). Our uncertainty about Electra’s wishes, engendered by the false clue of 126–7, is now ended. She desires the death of Clytaemestra,35 and has no compunction about the agent of that death being a child of Clytaemestra’s. But which child? There is an ambiguity in ’. . . æø ¼ . Most interpreters36 have taken the meaning to be ’I would 34
An unconscious one so far as Electra is concerned; by ‘we’ she merely means the human race (cf. æØÐ 580). 35 Owen (1927: 51) argued that her assertion is compatible with her desiring the death only of Aegisthus; but if Sophocles had meant us to entertain that as a possibility here, he would never have coupled the singular Ø with so bloodstained a word as ØæÆ (rightly Sheppard 1927b: 164). 36 The commentators are mostly silent, though Kaibel (1911: 168) clearly takes the conventional view (‘Diesen falschen Vorwurf greift El . . . . auf, nicht um ihn zu widerlegen . . . sondern um zu wu¨nschen, er mo¨chte wahr sein’), and similarly Hogan (1991: 291) paraphrases ‘yet would I nurture him for vengeance if I could’; but most translators’ renderings imply, by grammatical details (note the words italicized below) rather than by anything more explicit, that they understand the passage thus, e.g. ‘I kept him alive, you have often said, to be your executioner. Yes, if I could, I would have done just that’ (Watling 1953: 86); ‘You say I am sustaining him that he may come as an avenger: would to God I were!’ (Kitto 1994: 121); ‘I would have done so . . . had I had the power’ (Lloyd-Jones 1994: 219). The paraphrase of our sentence by Winnington-Ingram (1980: 220) (‘Electra . . . is hoping for the return of Orestes to kill Clytemnestra’) is typical of the view taken by most critics who have taken note of the passage at all. I have found one translation that apparently
238
Alternative scenarios in Electra
have brought him up to be an [205] avenger against you.’ That Electra, as this reading implies, (1) desires her mother’s death and (2) would be content, indeed happy, for Orestes to inflict it, is appalling enough, however understandable. One feature of the text, however, suggests another reading even more appalling. That Electra has not in fact reared Orestes is due not to any lack of strength (NB Ł 604) but to lack of opportunity: the rearing (æç Ø ) of children, or even of adolescents, is not an activity requiring physical strength.37 A task that is normally thought of as requiring physical strength38 is the taking of violent revenge; and Electra’s words are therefore perhaps better understood to mean ‘if I had the strength, I would <not merely, as you allege, be inciting another to take revenge on you, but> actually (ŒÆd) be doing the deed!’.39 On this reading40 Electra (1) as before, desires her mother’s death and (2) is willing in spirit (though, or so she feels at present, too weak in flesh) to inflict it herself. Electra never said anything like that in Aeschylus,41 and it is unlikely [206] that she had ever been made to say or think anything of the sort by any poet before Sophocles. perceives, and attempts to capture, the ambiguity of the sentence, that of Kenneth McLeish (1979: 22): ‘You say I saved him to punish you one day. If only I had! If only I could!’ 37 Orestes has in fact been reared to young manhood (13–14) by the elderly paidagogos. 38 As Chrysothemis will point out (997–8); cf. also 1415. The possibility that a female avenger might make use of a method not requiring physical strength, such as the stereotypically feminine device of poison (cf. Medea, Deianeira; Eur. Hec. 878, Ion 616–17, fr. 464.2; Ar. Thesm. 430; Antiphon 1; see Just 1989: 265–8), is never raised in this play or, so far as is known, in any version of the Agamemnon–Orestes myth. 39 Whereas on the other reading, envisaging Orestes as the avenger, the imperfect tenses of this unfulfilled conditional sentence could be taken to refer to a (hypothetical) continuing action in the past, on the reading now under consideration the reference must be to (imaginary) present time, since taking violent revenge, unlike bringing a person up, is a single act rather than a continuing activity. 40 This interpretation is presupposed, rather than asserted, by Woodard (1964: 185, 188), and adopted explicitly by Gellie (1972: 115); neither seems aware that it is controversial, and no one to my knowledge actually discusses the meaning of the passage. 41 Quite the contrary, she prayed to be ‘more virtuous (øçæ æÆ ) and more pious in action’ than her mother (Cho. 140–1); in our play she specifically repudiates this aspiration as impossible in the circumstances in which she is placed (El. 307–8; cf. Garner 1990: 126) and admits, or asserts, that in crucial respects she has come to resemble her mother (605–9).
Alternative scenarios in Electra
239
As previously observed, to say that Electra desires a matricidal revenge is not to say that Orestes desires or intends it, but we are subsequently given two further pieces of evidence which prove that they are in fact at one on the matter. One of them comes, paradoxically, at the very moment when Electra is facing what she believes to be a final sundering from her brother, having just heard the false news of his death; the second when she is holding the urn which she supposes to contain his ashes, and lamenting over it in, but unaware of, his presence. Clytaemestra’s famously ambivalent first reaction to the tale told by the paidagogos (766–8) includes the sentence ‘It is a grievous thing, if I save my life by having an injury done to myself.’ That she regarded the existence of Orestes as a danger to her life we already knew (603–4). But now we discover that she had good reason to do so: ‘After leaving this land . . . he taxed me with his father’s murder and threatened to do terrible things, so that neither by night nor by day could sweet sleep enfold me, but time . . . was continually as it were leading me to my death’ (778–82). The circumstances make it impossible to suppose that Clytaemestra is lying here. There is no reason at all why she should show, in public, such relief and satisfaction at the death of her only son42 unless she genuinely feels it. Orestes, then, from exile, either in a personal letter to his mother or in a public manifesto, has made declarations that have convinced Clytaemestra that he intends to kill her or to have her killed. We knew that Electra was willing, indeed eager, for such an outcome; now we know it of Orestes as well, and his utterances in the prologue can now be understood in their straightforward, traditional sense. In case any trace of doubt remains, Sophocles finally removes it by having Electra mention, during her lament over the urn, that Orestes
42 Aeschylus carefully avoids letting her do so; her reaction to the news is to display intense grief (Cho. 691–9), and when the disguised Orestes expresses regret at having had to bring bad tidings (Cho. 700–6) Clytaemestra, while remembering her duty as hostess, says nothing to show that she is anything but sorrowful at the news. Even to the servants within the palace she presents a sad face (737–8), though the Nurse believes that inwardly she is laughing (738–9). On the relationship between the Aeschylean and the Sophoclean scene see Jouanna (1993: 180–2).
240
Alternative scenarios in Electra
had frequently sent her secret messages43 saying that he would come to take revenge [207] on their mother (1154–6). Aegisthus is not mentioned here: Clytaemestra is the one on whom vengeance is to fall. Orestes has been determined from the beginning to kill her, and both Electra and Clytaemestra herself knew it all along. Meanwhile the chorus also—who at first, it will be remembered, were not willing even to pray for Clytaemestra’s death in Electra’s presence—have given evidence of a very different attitude. When they are alone with the despairing Electra, soon after she has received the news of Orestes’ supposed death, they try to comfort her by evoking the parallel case of Amphiaraus, who like Agamemnon had been killed through the treachery of his wife, but who has become a lord of great power in Hades (837–41); after which they and Electra together recall how the guilty Eriphyle was killed, and Amphiaraus avenged, by ‘one who cared’ (846 ºøæ)—namely (as everyone knew, though it is not explicitly stated here) their son Alcmeon. Neither the chorus nor Electra give the least sign of concern about the fact that the murderer on whom Alcmeon was taking revenge was his mother; Electra grieves only that where Alcmeon succeeded Orestes has apparently failed (846–8), and the current of the chorus’s sympathy for her is not deflected. Later, despite having advised Electra to accept the prudent counsel of Chrysothemis and refrain from attempting to take revenge herself for her father (1015–16), they praise her determination to sacrifice her life if necessary for the sake of destroying ‘the twin Eriny(e)s’ (1078–80)44—and later still, when the avengers have gone into the palace specifically for the purpose of killing Clytaemestra, and then at the time of, and after, 43 We had heard of these messages before (171–2, 319), but of their content we knew only that Orestes said he yearned (Ł ØÐ 171) to return to Argos (or to be reunited with Electra) and promised he would come (318–19). 44 Paulsen (1989: 55–60) attempts to explain this change of mind in terms of the thought-processes of the chorus, suggesting that Chrysothemis’ stress on çæ ØÐ
(1056) has led them to think of the çæ Ø ÆØ Nø , the storks, and their filial piety (1058–62; cf. Winnington-Ingram 1980: 242 n. 82), and thus to favour Electra in the sisters’ quarrel because she is displaying similar piety; the logic of this analysis is somewhat convoluted, and I would prefer to note that Electra must score heavily with the audience, and surely also with the chorus, when she in effect forces Chrysothemis to admit for the second time (1041–2, cf. 338–9) that her policy of passivity is ¼ØŒ .*
Alternative scenarios in Electra
241
the killing itself, the chorus, while recognizing that the deed is a terrible one (1385, 1394, 1407–8, 1422–3), also regard it as the punishment of a great crime (1387), the vindication of the dead (1391–2, 1420–1), the restoration of the legitimate heir (1393), and the end of the long tribulations of oikos and polis (1413–14), and find themselves unable to condemn it (1423).45 The progressive trans-[208]formation of their attitude, from inability even to contemplate matricide to clear approval of it,46 runs parallel with the progressive revelation of the true intentions of Orestes and Electra. I have suggested that a crucial catalyst is the appearance of Clytaemestra on stage, and this finds confirmation in a brief but important remark by the chorus. Shortly after Electra has sung of the ‘caring’, and matricidal, son of Amphiaraus, she speaks of her own life as ‘month after month a constant spate of many terrible and hateful experiences’ (851–2) and the chorus reply (853) Y L Łæ ØÐ ‘we have seen what you are talking about’. This can only refer to what the audience too have seen:47 Electra’s squalid condition and her mother’s degrading treatment of her. What the chorus have seen, and what we have seen, gives credibility to Electra’s account of things we have not seen; and it is not so much (as in Aeschylus) the fact and the nature of Clytaemestra’s killing of her husband, but the accumulated wickedness48 of her behaviour over seven years or 45
The last two points both rest on textual emendations (Ø for in 1413, łª Ø
for ºª Ø in 1423); but the case for these, and against the transmitted readings, is so compelling that Kells (1973), to whose understanding of the play both passages as emended are prima facie very damaging, accepts both emendations without question and tries to remove their (to him) embarrassing implications by interpretative ingenuity. 46 Though unlike Electra (1154–6) and Orestes (1427), they can never actually bring themselves to utter the word Åæ in connection with the vengeance. 47 Since Electra’s presence outside the walls of the palace is unusual (see p. 234 n. 26 above), so that the chorus of women of the city can be taken to know little of how she has been forced to live. 48 In addition to adultery and murder, the crimes of Clytaemestra mentioned in the play are as follows: (1) mutilating Agamemnon’s corpse and wiping off his blood on his own head (445–6); (2) keeping festival every month on the day he was killed (277–81); (3) producing children with Aegisthus to usurp the position of Agamemnon’s children (589–90, cf. 653–4); (4) attempting to murder Orestes as a child (601, 1133; cf. 294–8) and publicly expressing satisfaction at the news of his death; (5) together with Aegisthus, deliberately keeping both Electra and Chrysothemis unmarried (165, 187– 8, 961–6); (6) together with Aegisthus, keeping Electra under house arrest (312–13,
242
Alternative scenarios in Electra
more49 thereafter, that makes her killing by her son, in this play, the lesser of two evils.50 It is fitting that after carrying out the killing, Orestes says not, as [209] we might expect, ‘Now my father is avenged’, but, to Electra, ‘You need not fear that your mother’s arrogance will ever degrade you again’ (1426–7). Thus far the implications of the raising, and eventual rejection, of the possibility of a scenario in which Orestes kills Aegisthus but refrains from killing Clytaemestra. But even as that alternative scenario is fading away, another is conjured up, in which, probably for the first time,51 Electra participates in the killing of her mother, or even carries it out on her own. The first indications of this appear, as we have seen, in Electra’s speech in the agon with Clytaemestra, when she says that Clytaemestra would in justice deserve to die ‘if we are going to kill one person in return for another’ (582–3) and when she counters the charge of bringing up Orestes to be an avenger against 517–18, 911–12) and forcing her to live and work like a slave (189–92, 814, 1192–4); (7) personally subjecting Electra to physical violence (1195–6); (8) together with Aegisthus, threatening to immure her in an underground prison (379–86; the phrase ÇHÆ ’ K ŒÆÅæ ç ØÐ ªfiÅ, with its echoes of Ant. 774, 885, 888, may half suggest that, as in Antigone’s case, this would really be a sentence of death by starvation—note that the prison is to be åŁ e B’ KŒ as if to avoid any risk of polluting the soil of Aegisthus’ realm should Electra die there). See Machin (1981: 208–12). 49 There are some indications that Sophocles is assuming a rather longer interval than Homer’s seven years (Od. 3.304–6) between Agamemnon’s murder and Orestes’ revenge: Electra thinks Orestes has been procrastinating (171–2, 319), and feels that ‘most of my life has already passed away from me’ (185–6). 50 That Sophocles wants us to see it as an evil is in my view certain; otherwise he would not have shown us, at the beginning of the play, Orestes instructing his agent to commit perjury (47–48; the instruction is quietly ignored by the agent)* and asserting that self-advantage justifies any lie (61; cf. the despicable Odysseus in Phil. 108–11), and at the end Electra—that heroine in some ways so reminiscent of Antigone—insisting that Aegisthus be denied burial (1487–90). Nobody, it seems to be suggested, can set aside the inhibitions that normally prevent us from even contemplating such acts as matricide without also setting aside other inhibitions essential to civilized life (cf. Aesch. Eum. 494–5 Æ XÅ ’ æª På æ Æ fi ı Ææ Ø æ ). And yet it does not follow that Sophocles wants us to see the matricide as wrong in the given circumstances. If we condemn Orestes for doing it, Electra and the chorus for approving it, they are entitled to answer us with the challenge posed by Dicaeopolis (echoing the Euripidean Telephus) in Ar. Ach. 540: ‘if you say they shouldn’t have done that, then what should they have done?’ So far as I can discover, none of the scholars who have taken this view (e.g. Sheppard 1927a, b; Friis Johansen 1964; Kells 1973) has attempted to meet this challenge. 51 Cf. n. 17 above.
Alternative scenarios in Electra
243
Clytaemestra by asserting ambiguously that ‘if I had the strength, I would be doing that’ (603–5). For the time being the idea remains dormant, and when Electra hears the false news of Orestes’ death her reaction is, for quite a long time, one of utter despair and the loss of all wish to live (674, 677, 808–22), a mood not in the least relieved by the consolations of the chorus, and, so at first it seems, only confirmed by the true but uncredited news of Orestes’ return brought by Chrysothemis (920–5). But then it suddenly revives52 as Electra proposes to her sister that the two of them should them-[210]selves avenge their father’s murder. Now in making this proposal (954–7), Electra names only Aegisthus as the intended victim. Are we to understand that at this moment she is only interested in killing Aegisthus?53 Or does she deliberately refrain from mentioning Clytaemestra because she is sure Chrysothemis would balk at such a suggestion?54 Or is it 52 The text does not directly indicate any motivation for the sudden change in Electra; but what immediately precedes it is Chrysothemis’ expression of distress (930–7) at the disappointment of the hopes that she had come, with such joy, to report to Electra. Electra had previously written off Chrysothemis as a coward (cf. pp. 234–5 above) who would assist the cause she knew to be right (338–9) only if she ran no risks thereby; in the present scene, both in her initial joy and in her subsequent sorrow, Chrysothemis’ words and actions have been spontaneous, emotional, uncalculated, from the very moment when, ‘putting aside propriety’, she ran up to Electra with her news (871–2). She has shown her real self as surely as Clytaemestra did when she greeted so eagerly the message that Orestes was dead (675); and if that is the real Chrysothemis, then Electra is not as she thought (813–14, 819) alone and friendless—at least she can now say (950) ‘we two are left alone’, using the extremely rare form of the first-person dual (º º Ł ). If the words of 930–7 are strongly reinforced (as they surely would be) by body-language, then added to what Chrysothemis has said already they will provide a credible explanation for Electra’s new-found resolve. Cf. Machin (1981: 226– 7) and Saı¨d (1993: 322–3) (‘La tirade ou` E´lectre explicite son projet est tout entie`re sous le signe du duel, du nous et de la coope´ration’). 53 Thus Linforth (1963: 103; cf. Kitzinger 1991: 321) suggests that Electra does not expect to be able to kill more than one of her enemies, and Aegisthus as the holder of power is the one whom it is essential to remove; Gellie (1972: 119) thinks Electra ‘is proposing the more difficult task, in fact the only real task’. Juffras (1991) argues with some plausibility that Electra is represented as proposing an act of tyrannicide (with reminiscences of the Athenian cult of Harmodius and Aristogeiton), so that Clytaemestra is irrelevant; but why should Electra expect a plan for a political assassination to be attractive to Chrysothemis, of all people? 54 So e.g. Waldock (1951: 185) (‘She names only one name—Aegisthus; it would have been out of the question to risk more’); Ronnet (1969: 224 n. 1); Lloyd-Jones (1972: 224); Kamerbeek (1974) ad loc.; Segal (1981: 284); Whitlock Blundell (1989: 160 n. 46).
244
Alternative scenarios in Electra
misguided to assign a motive to her (rather than to the dramatist) at all?55 The chorus at any rate, without having subsequently been given any further information,56 take it for granted (1080) that Electra is determined to kill both Aegisthus and Clytaemestra; and if this is to make any sense to an audience who recently heard her passionately appeal to Chrysothemis to join her in killing Aegisthus, that audience must be assumed to understand that in making that appeal Electra was revealing only part of her [211] intentions.57 And they have had sufficient evidence to justify that assumption. They know (603–5) that Electra desires Clytaemestra’s death and that the thought of a matricide gives her no qualms; they know that for her perhaps the most lamentable thing about Orestes’ supposed death is that he has been unable to play the part of Alcmeon to his mother’s Eriphyle (842–8); and ever since Clytaemestra appeared on stage (most recently at 929) Electra’s verbal assaults have been directed almost exclusively against her.58 It is impossible to believe that her animus against her mother has suddenly vanished;59 if she mentions only Aegisthus to her sister, it is because she wishes above all to win her over and will say only what she thinks is likely to contribute to that aim. For the same reason, while fully aware, as the chorus are, that she will be putting her life in the utmost jeopardy (1320–1, 55 Thus Gardiner (1987: 165) suggests that Sophocles wishes to keep the debate between Electra and Chrysothemis entirely on the level of expediency, without raising issues of moral principle as the suggestion of killing Clytaemestra inevitably would; she argues that Electra’s ‘real’ intentions, being (in Gardiner’s view) unknown to the audience, are irrelevant. 56 The plural ØÐØ KåŁæØÐ at 979 is not in itself decisive evidence, any more than were the plurals at 66, 209, 247, and 456. 57 And therefore that when she said ‘I must no longer conceal anything from you’ (957), she was lying. This is not necessarily out of character for a Sophoclean hero; in Ajax, after all, it is Odysseus, not Ajax, who consistently tells the truth! Electra herself shows a remarkable facility in deception at 1323–5 when, on being warned that someone is coming out of the house, she instantly shifts from conversing happily with her brother to formally directing two ‘strangers’ into the palace. 58 Between 389 and 955, indeed, Electra makes mention of Aegisthus as an individual only in 587–600, and then primarily in order to indict Clytaemestra for living with him (cf. 587 ı Ø, 594 ªÆ ØÐŁÆØ, 600 ı
ı) in the course of a sustained attack on her. 59 Indeed Sophocles will shortly provide us with further evidence that it has not, when he makes an exasperated Electra say to Chrysothemis ‘Go to your mother ( Åæd . . . fi ÅÐ) and tell her all this’ (1033), implying ‘I don’t regard her as my mother’.
Alternative scenarios in Electra
245
cf. 1078–9), in her plea to Chrysothemis she makes no direct reference to any risks at all.60 When Chrysothemis refuses to cooperate, Electra declares that she will make the attempt alone (1019–20, 1045); and this is no spur-ofthe-moment folly, for later on (1319–21), long after solo action by her has ceased to be necessary, she reaffirms that had Orestes not come she would have done it and ‘either nobly saved myself or nobly perished’. We know, of course, that Orestes is close at hand, but we cannot be sure at what moment he will appear; in particular, we cannot be sure that Aegisthus will not return before Orestes arrives,61 in which case Electra will have to act at once or not at all (cf. 379–86). But Orestes does arrive, and this event diverts the action back into its [212] traditional channel. Only not quite. When the paidagogos urgently tells Orestes and Pylades that ‘now is the time to act’ (1368)—against Clytaemestra—they duly go inside, while Electra, remaining out of doors (where she has been, onstage, without a break ever since her first entrance), prays to Apollo (1376–83) for their success—or rather for their and her success62—and then follows them inside. There has been some bafflement63 as to why Sophocles should have taken Electra offstage like this, only to bring her on again very shortly afterwards (1398). She has an important role to perform in the next few minutes, both for the conspirators (keeping a lookout for Aegisthus) and for the dramatist (bringing vividly to life for the
60
Possible dangers are admitted only via an indirect allusion (980 łıåB Iç ØÆ , in a speech put in the mouths of members of the general public); rewards, on the other hand, including ‘a worthy marriage’ (971–2), are predicted with confident future indicatives as the consequence not of success but of willingness to make the attempt (967), as though making it would guarantee success. 61 Orestes himself is evidently unaware of Aegisthus’ absence from the palace (Electra’s assumption that he has heard of it, 1307–9, is not warranted by anything in the text theretofore) and therefore of the importance of forestalling his arrival. 62 Cf. 1376–7 ¥ º ø ÆPØÐ Œº K F æe ØØ , 1380–1 ª F æçæø
ØÐ Iæøª. 63 See e.g. Linforth (1963: 108–9) who suggests three motivations, none of them very convincing: (1) on Electra’s part, to show the men their way about the house (but then she would need to go in with them rather than after them); (2) on the dramatist’s part, to enable her to say what she says in 1400–1 (a pretty trifling dramatic gain, as Linforth admits); (3) to mark ‘the end of [her] dramatic dominance’ (yet from her reentry to the end she directs virtually everything that happens onstage!).*
246
Alternative scenarios in Electra
audience what is happening out of their sight); that explains her reappearance, but it does not explain why she was made to exit in the first place. Like Gellie,64 I can see only one adequate explanation for that: when Electra goes into the palace, the impression is created that she has gone inside to take part in the murder. This has been prepared for by her (recently recalled) willingness to carry it out herself if necessary, by two remarks (1297, 1335–7) which seem to indicate that she, not Pylades, will be entering the palace with Orestes to do the deed,65 and by her expressed readiness (1319) to obey any directions Orestes gives. Now, thanks to the delay caused by her prayer, it is her entry into the palace, not that of Orestes and Pylades, that is accompanied by the grim song of the chorus (1384–97) in which Ares breathes out blood as the hounds that none can escape pass beneath the roof of the house. At this moment it can truly be said, with Karelise Hartigan, that ‘Sophocles has so structured his plot that we become uncertain as to the actual perpetrator of the [213] matricide’.66 Directly after the song Electra comes outside again. The killing is after all to be carried out by Orestes, assisted by Pylades. But although the second of our alternative scenarios has now at last evaporated, it has left behind an important residue. Electra does not take a physical part in the killing of Clytaemestra (or of Aegisthus); she does more than that. Far from her obeying Orestes’ directions, he obeys hers. She, not Orestes, answers Clytaemestra’s desperate appeal for the child to pity the mother (1411–12). She
64
Gellie (1972: 126). At 1297 the phrase ø fi Ð K ºŁ Ø ı is embedded in Orestes’ warning to her not to give the game away to Clytaemestra by looking happy (1296–9). See Gellie (1972: 124). At 1335–7 it is even clearer that the paidagogos’ injunction IƺºÆåŁ H ÆŒæH ºªø ŒÆd B . . . f åÆæA fi B must be addressed to Orestes and Electra, since Pylades has not engaged in any ‘long talk’, indeed has not said a word all through the play, and Electra is the only person who has indulged in ‘cries of joy’; cf. Saı¨d (1993: 326). In 1367, on the other hand, the dual çø fi Ð ‘you two’ refers unambiguously to Orestes and Pylades, in contrast with Electra whom he had been addressing just before. 66 Hartigan (1996: 90). In context Hartigan is actually referring to the killing itself, in regard to which her statement is something of an exaggeration (cf. rather Seale 1982: 75: ‘Electra appears mentally, if not physically, to appropriate the act by the violence of the language with which she wills it’). 65
Alternative scenarios in Electra
247
urges Orestes to strike a second time,67 and he does: in the words of Bernd Seidensticker,68 ‘the second blow is her blow’—and it is the second blow that is the killing one.69 She warns Orestes and Pylades to get out of the way when Aegisthus is seen approaching (1430),70 says she will deal with him (1437), and entraps him with an exquisite series of ambiguities (1448–65), finally interrupting an exchange between him and Orestes to insist on his immediate execution and the exposure of his corpse (1483–90): this last she demands as the only possible satisfaction for her past sufferings (1489–90), just as Orestes had reported Clytaemestra’s death to her by assuring her that she would no longer be insulted by her mother’s arrogance (1426–7). Orestes may carry out the killings, but they are done at Electra’s direction and for Electra’s benefit.71 Thus the phantom possibility of a matricide by Electra has proved in the [214] end, in one sense, not such a phantom after all. How about the other phantom possibility—that of a revenge without any matricide? That, I fear we must say, has vanished completely. Clytaemestra’s enemies may assert that one who behaves towards her children as she has done is no true mother,72 but a mother she remains, and is so called by her children shortly before and after they kill her.73 Her monstrosities made her murder inevitable when it had not seemed so at the outset, and her murder by Electra conceivable when probably no teller of the tale had ever thought of it
67
Or to strike twice as hard (so Lloyd-Jones and Wilson 1990b: 74). In discussion on Winnington-Ingram (1982: 257). 69 Or should we go even further? Machin (1981: 425) notes that Orestes does not strike even the first blow until Electra has ‘engaged openly in the action’ by calling out to those within the house (at 1411–12; her words at 1406 and 1410 had been addressed only to the chorus). 70 Addressing them as t ÆØÐ , which is a legitimate form of address between siblings (at least if of opposite sex) regardless of their relative ages (cf. OC 1255, 1420, 1431) but which here, covering Pylades as well as Orestes, will sound as though Electra is assuming superiority over them: the vocative ÆØÐ( ) is never used in addressing a person of even approximately equal age other than a sibling or a slave (see Dickey 1996: 65–72, 266–7). 71 Saı¨d (1993: 327) hardly exaggerates in claiming that ‘Oreste n’est plus qu’une main guide´e par la volonte´ d’E´lectre.’ 72 273–4, 597–8, 790, 1154, 1194, 1411–13. 73 1289, 1296, 1309, 1427. 68
248
Alternative scenarios in Electra
before; that murder is nevertheless itself a monstrosity,74 even if this play is not concerned with its consequences.75
ADDENDA p. 227 (also n. 8) on ‘earlier versions of the story of Phaedra and Hippolytus’ see the discussion by Thomas Talboy and myself in Sommerstein, Fitzpatrick, and Talboy (2006: 248–89); the question which of Euripides’ two Hippolytus plays was the earlier is considered on pp. 266–71. p. 231 n. 17 March (2000: 21–2), on the other hand, brings some persuasive arguments for the priority of Euripides’ play. Finglass (2007: 1–4) regards the issue as undecidable and ‘in the end . . . perhaps less important than it first appears’. I would not today have put so much emphasis at various points (e.g. pp. 238, 242) on the innovative nature of the role Sophocles gives to Electra. p. 233 n. 22 Finglass (2007) ad loc. says simply that › æ is ‘a generalizing masculine’, leaving it unclear which of Jebb’s two options he prefers. 74
See n. 50 above. Aegisthus’ famous reference to ‘the future woes of the Pelopids’ (1497–8) is not evidence to the contrary: if he is speaking of anything beyond his own death, it is of future events destined to take place under ‘this very roof’ ( c ªÅ ), and that excludes any possibility of an allusion to the pursuit of Orestes by the Erinyes (see Erbse 1978: 298; also Szleza´k 1981: 18). The only future events to which Aegisthus’ words might conceivably apply—leaving aside, of course, those which Euripides was to invent some years later in Orestes—would be the killing of Aegisthus’ son Aletes, and the near-killing of his daughter Erigone, by Orestes, narrated by Hyginus Fab. 122 and apparently the subject of a play attributed by some in antiquity (including Stobaeus, who quotes seven uninformative fragments) to Sophocles though probably in fact of fourth-century composition (see Wilamowitz 1929: 465–6 ¼ 1962: 483–4). We know that Sophocles did write an Erigone; we know nothing about its plot (not even whether its eponym was the daughter of Aegisthus or of Icarius) except that it will have been significantly different from that of the play that was Hyginus’ source (cf. p. 227 and n. 10 above), but it is possible that it included the death of a son of Aegisthus (cf. El. 589–90), at Orestes’ hands, in the Atreid palace, that Sophocles is here alluding to one of his own earlier plays, and that Aegisthus, far from predicting the future woes of his enemy, is ironically being made to predict, without knowing it, the extinction of his own male descent line.* 75
Alternative scenarios in Electra
249
p. 236 I was wrong to see ambiguity in El. 493–4; ØÐØ P Ł Ø makes it clear that the reference is to an illicit couple, i.e. Clytaemestra and Aegisthus. p. 240 n. 44 Finglass (2007), on El. 1058–97, does not find 1041–2, or anything else, sufficient to provide a realistic motivation for the chorus’s change of attitude, and prefers to hold that ‘dramatic necessity has taken precedence over consistency of characterisation’— the ‘dramatic necessity’ being to build up sympathy for Electra in preparation for the following scene. p. 242 n. 50 both March (2000) and Finglass (2007), on El. 47–8, hold that (in the words of the former) ‘a false oath would have been acceptable in the light of Apollo’s instructions to use guile’, but this is a complete misunderstanding of the nature of an oath. To swear to a false statement was to bring punishment down on oneself, automatically, from the god named in the oath, and was generally seen as unthinkable under all circumstances; indeed there were stories about entire families being wiped out merely because a member had contemplated committing perjury (Hdt. 6.86). p. 245 n. 63 March (2000), on El. 1382–3, suggests a fourth explanation, which is at any rate a bit stronger than Linforth’s three: to leave the chorus alone onstage so that ‘all the audience’s cconcentration can be on their emphatic reminder of the justice of Clytemnestra’s killing and of the gods’ involvement and support’. p. 248 n. 75 Finglass (2007), on El. 1442–504, argues that Sophocles has deliberately left the ‘future woes’ indeterminate, except that pursuit by the Erinyes is excluded (not least because Orestes certainly, and Electra probably, end the play not by fleeing into exile—as Orestes does in Aeschylus, and as both do in Euripides—but by going into the palace). I do not set very great store by my suggestion about Aletes and Erigone (though March 2000, on 1498, finds it attractive); it remains true that Sophocles, unlike Aeschylus, has gone to the trouble of telling us that Aegisthus and Clytaemestra have had at least one child, and therefore that there is a potential avenger or avengers for them in being.
17 Sophocles’ Palamedes and Nauplius plays: no trilogy here The story of Palamedes, and his father Nauplius, was one of the most popular of what might be called the side-issues of the Trojan War saga—popular with everyone except Homer, who either knew nothing of it or (more likely in my view) chose to ignore it because it did not fit with the way he wanted to represent Odysseus, either in the Iliad or the Odyssey.1 To summarize the basic story briefly, abstracting away, as always, from a lot of variability. Palamedes was about the only person in Greece who was cleverer than Odysseus, and when Odysseus tried to avoid joining the Trojan expedition by pretending to be mad, it was Palamedes who saw through and exposed his deception.2 This made Odysseus his deadly enemy, and his hatred was intensified by jealousy when Palamedes conferred enormous benefits on the army (by his ingenious inventions and/or by securing their food supplies).3 1
The Homeric epics can often be shown to have suppressed significant story elements of which their authors certainly or probably knew. Two famous examples of this—the matricides committed by Orestes and Alcmaeon—are discussed briefly in Ch. 15 and 16 above; on the general issue, particularly the contrast between the Iliad and Odyssey on the one hand and the cyclic epics on the other, see Griffin (1977: 39–46). 2 Cypria Arg. }5 West. 3 There are many catalogues of Palamedes’ inventions: in tragedy alone, we have Aesch. frr. 181a, 182, 182a; Soph. frr. 429, 432, 479 (discussed below); and Eur. fr. 578. According to Servius on Verg. Aen. 2.81, after Odysseus had failed to procure grain from Thrace for the Greek army, Palamedes ‘set out and brought back a vast amount of grain’; according to Lycophron 581, he was sent on a successful mission to secure the services of Spermo, Oeno, and Elaeo, daughters of Anius of Delos, who had the
Sophocles’ Palamedes and Nauplius plays
251
Eventually Odysseus, with or without assistance, contrived Palamedes’ death; in the cyclic epic, the Cypria, he and Diomedes drowned Palamedes on a fishing trip,4 but in tragedy the method employed is always to forge or plant apparent evidence of treasonable collaboration with the Trojans, with the result that Palamedes is condemned (usually by Agamemnon) and stoned to death by the army. When Palamedes’ father, Nauplius, heard of this, he came to Troy to protest, but none of the leaders took any notice of him, and he resolved to take revenge. He is reported as having done this in two ways, which are sometimes mentioned separately and sometimes together. First, he sailed (or rather rowed—he was the first great single-handed oarsman)5 around the Greek cities, encouraging the wives of the army’s leaders to have adulterous affairs (with other men—he was out to make trouble for the husbands, not to take risks himself).6 Secondly, at the time when the Greeks were on their way home, he went to Euboea and lit beacons on Cape Caphereus, which much of the fleet mistook for harbour-markers, so that they were lured on to the rocks with great loss of life.7 The events leading to the death of Palamedes were dramatized by all three of the great tragedians.8 Both Aeschylus’ play9 and, as we now know,10 that of Euripides went on to include the visit of Nauplius to Troy, and both will have ended with his aggrieved departure—which in Euripides’ case leads very neatly into the power of producing unlimited quantities of wheat, wine, and oil respectively (cf. [Apollod.] Epit. 3.10). 4 Cypria fr. 27 West. 5 Eur. Hel. 1128; Lycophron Alex. 1217. 6 [Apollod.] Epit. 3.9; Lycophron 384. With Penelope in Ithaca he either failed in his endeavours or did not make the attempt; but according to some authors he did induce Odysseus’ mother to commit suicide, by persuading her that her son was dead ( Od. 11.197, 202). 7 Nostoi Arg. }3 West; Eur. Tro. 75–90, Hel. 766–7, 1126–31; Seneca Ag. 470–578; [Apollod.] Epit. 6.5–11; Hyginus Fab. 116; Q.S. 14.449–658. 8 On the three tragic versions of the story see Scodel (1980: 43–61); Sommerstein (2000b); Jouan and van Looy (1998–2003: ii. 487–507); Collard (2004); Collard and Cropp (2008: ii. 46–59); and my discussion of Sophocles’ Palamedes and Nauplius plays in Sommerstein, Clark, and Talboy (forthcoming). 9 Aesch. fr. 181. 10 From a synopsis of Euripides’ play in an as yet unpublished Michigan papyrus; see Huys (1997: 24) and Luppe (2000: 273).
252
Sophocles’ Palamedes and Nauplius plays
opening of his next play, Trojan Women,11 when for entirely independent reasons Athena and Poseidon combine to create a storm that will ensure that the fleet, or what is left of it, will still be at sea, off Euboea, after dark, all ready to fall into Nauplius’ trap. Sophocles, however, wrote not one play but three about Palamedes and Nauplius: Palamedes, Nauplios Katapleon (The Arrival of Nauplius), and Nauplios Pyrkaeus (Nauplius the Fire-Raiser). The first of these clearly presented the entrapment and death of Palamedes; we have no direct evidence as to whether it continued any further. The second, as we now know from a papyrus hypothesis,12 dealt not (as had sometimes been thought previously) with one of Nauplius’ trouble-stirring visits to Crete or elsewhere, but with his justiceseeking visit to Troy. And the third was obviously about the episode of the deceptive beacons which gave Nauplius his culminating revenge. It was popular enough, four or five centuries later, to serve as the basis for a remarkable mechanical toy which presented a drama in five scenes on a miniature stage, and the description we have of this device, by Heron of Alexandria,13 tells us, what we would not otherwise have known, that this play included (presumably by way of report) the drowning of the lesser Ajax through the anger of Athena, which may even have been its climax—and probably an unexpected one. Athena appeared in two scenes of the toy drama; that does not in itself prove that she appeared in person in even one scene of Sophocles’ play, but it is highly likely that she did, because only she (or another deity) would be in a position to explain that she had been the cause of Ajax’s death, and why she had caused it. Now here we have three plays presenting, it would appear, three successive phases of the same story, without overlap so far as their action is concerned and without any inconsistencies of which we know. It is therefore tempting to wonder if they may have been parts of a connected suite of plays produced on a single occasion. Sophocles did not in general write connected tetralogies, but there is an
11 Trojan Women was the third play of Euripides’ production at the Dionysia of 415, the first two being Alexandros and Palamedes (Aelian VH 2.8). 12 POxy 3653 fr. 1.1–6. 13 Heron, On Automaton-Making 22.3–6 (see Marshall 2003).
Sophocles’ Palamedes and Nauplius plays
253
inscriptional reference to a Telepheia of his,14 and Euripides too produced a (somewhat loosely) connected suite of plays in 415.15 It has also been suggested,16 on the basis of Heron’s evidence about the death of the lesser Ajax, that the suite may have included Aias Lokros which we know covered events connected with this Ajax’s sacrilegious seizure of Cassandra at Troy and the army’s failure to punish it; and there is in addition Odysseus Mainomenos (The Madness of Odysseus), dealing with the event that kindled the enmity which eventually led to Palamedes’ death, which Lloyd-Jones17 has suggested might possibly have formed part of a suite of plays on the Palamedes–Nauplius story. For simplicity I will leave Odysseus Mainomenos aside for the moment—we will come back to it later—and concentrate on the other four plays. The sequence of these four plays, in terms of mythical chronology, is Palamedes—Nauplios Katapleon—Aias Lokros—Nauplios Pyrkaeus. The death of Palamedes occurred at a point in the war not long before the events described in the Iliad;18 the visit of Nauplius some time after that—we do not know how long—but certainly while the war was still on; the crime of Ajax during, or immediately after, the actual capture of Troy; and the wreck of the fleet, of course, during the voyage home. Thus, on the reasonable assumption that any trilogy of this type would have ended with the actual double catastrophe (in Nauplios Pyrkaeus) rather than with the mere threat of it (in Aias Lokros), there are three possible sequences of plays that could have gone to make such a trilogy—and all three have been put forward by recent scholars as possibilities for consideration. However, we have good reason to know how risky it is to posit trilogic connections between Sophoclean plays without clear evidence. If we possessed only a few fragments of each of Sophocles’
14
IG ii2. 3091.7–8. See n. 11 above. 16 Marshall (2003: 279). 17 Lloyd-Jones (1996: 249). 18 In Proclus’ summary of the Cypria (Arg. }12 West), ‘the death of Palamedes’ is immediately preceded by a distribution of spoils from sacked towns, in which Briseis is given to Achilles and Chryseis to Agamemnon, and followed by ‘the plan of Zeus to relieve the Trojans by causing Achilles to withdraw from the Greek alliance’. 15
254
Sophocles’ Palamedes and Nauplius plays
’three Theban plays’19—Oedipus the King, Oedipus at Colonus and Antigone—and had no direct information about their production dates nor any evidence even moderately reliable from which to infer them, there would certainly be found scholars to claim that they probably constituted (in the order named) a connected trilogy; in fact we know almost for certain that they were produced over a period of about forty years, with Oedipus at Colonus last,20 and, having the full texts of all three, we can see that there are clear inconsistencies between them (think of the portrayal of Creon, or of Ismene, in Antigone and in Oedipus at Colonus). And when we examine the three available options for a Palamedes-Nauplius trilogy, we find, even on the limited evidence we have, that none of them is persuasive. The simplest proposal, that of Schmid-Sta¨hlin (more recently supported by Clu´a),21 links together Palamedes with the two Nauplius plays. The main objection to this is that both Palamedes and at least one of the Nauplius plays included a major speech containing a detailed account of Palamedes’ intellectual achievements and the benefits he conferred on the Greek army,22 and both speeches made particular mention of his invention of board games (pessoi and kuboi).23 Sophocles might well have written similar speeches for two plays produced on different occasions, especially since they appear to have been delivered by different persons (the speech in 19 That is to say, his three surviving Theban plays. Sophocles wrote two other plays set at Thebes (Niobe and the satyr-play Herakleiskos [Little Heracles]), and at least one more, Epigonoi, which, like Oedipus at Colonus, is set elsewhere (in this case Argos) but in which the affairs of Thebes are of central concern. 20 Oedipus at Colonus, according to its second Hypothesis, was produced posthumously in 401. The date of 442 usually given for Antigone rests only on the uncertain evidence of a statement in the play’s first Hypothesis that Sophocles owed his election to a generalship (for 441/0) to his success with Antigone (which cannot have occurred at the Dionysia of spring 441 since Euripides was victorious on that occasion: Marm.Par., FGrH 239 A 60); but it has nevertheless been generally accepted that Antigone is earlier than Oedipus the King (see Stinton (1977) and Finglass (forthcoming) for some interesting stylistic evidence tending to confirm this). 21 Schmid and Sta¨hlin (1934: 446); Clu´a Serena (1985: 88). 22 Soph. fr. 479 (ascribed to Palamedes); fr. 429 (ascribed to Nauplios Pyrkaeus); fr. 432 (ascribed merely to Sophocles, but Nauplius named as speaker). 23 Soph. frr. 429 and 479.4.
Sophocles’ Palamedes and Nauplius plays
255
Palamedes was made either by Palamedes himself or, more likely, by a treacherous Odysseus posing as his friend and defender;24 the other speech is known to have been made by Nauplius). He would not have repeated himself like this within a single production. Indeed, we can go further. There are two fragments (429 and 432) ascribed to a Nauplius play that seem to come from a speech or speeches about Palamedes’ intellectual achievements. For fr. 432 our source does not give us the play-title; but fr. 429—which contains the mention of board games—is specifically ascribed to Nauplios Pyrkaeus. Hence we can be sure that Nauplios Pyrkaeus contained a speech of this type—and indeed there is reason to believe that fr. 432 also comes from this play.25 But is it likely that in Nauplios Katapleon, when Nauplius was pleading before the Greek leaders and army for some kind of recompense for Palamedes’ death, he did not remind them at length of how much they all owed to his son? Thus any suggestion that there was a Sophoclean trilogy on the Palamedes-Nauplius story, comprising three of the four plays we are at the moment discussing, is going to run into this repetition problem. And any suggestion that such a trilogy included Aias Lokros will run into another repetition problem as well. Two such suggestions have been made: the proposal of the late Timothy Gantz for a sequence Nauplios Katapleon—Aias Lokros—Nauplios Pyrkaeus, and that of Toph Marshall for a sequence Palamedes—Aias Lokros— Nauplios Pyrkaeus.26 The problem is this. We know that Aias Lokros contained an epiphany of Athena;27 but we have already seen that it is highly probable that Athena appeared in Nauplios Pyrkaeus as well. 24
This deception by Odysseus is attested by Servius on Verg. Aen. 2.81. I will be arguing in Sommerstein, Clark, and Talboy (forthcoming), as many have done previously (e.g. Robert 1920–6: iii. 1133 n. 2; Jouan 1966: 340–5), that the versions of the Palamedes story told by Servius and by Hyginus (Fab. 105) both derive from Sophocles’ play. The speech is unlikely to have been made by Palamedes himself, because it speaks of him in the third person (fr. 479.1), and the pronoun it employs is not ‹ (which can be used by a speaker referring to himself) but F (which cannot). 25 Because it refers to the Greek army in the third person (fr. 432.1), whereas in Nauplios Katapleon Nauplius would have been addressing himself to the Greek army and/or its leaders. 26 Gantz (1993: 696); Marshall (2003: 79). 27 Soph. fr. 10c.
256
Sophocles’ Palamedes and Nauplius plays
Not only is it a priori unlikely that Sophocles would have repeated himself in this way, but it would also be without a known parallel: there is no provable instance of the same divinity appearing on stage in two plays of a connected trilogy. We can now, as promised, bring in again Odysseus Mainomenos, but I fear only briefly. Lloyd-Jones speculated28 about a possible tetralogy comprising this play, Palamedes, and the two Nauplius plays. That, however, quite apart from the difficulties already mentioned, would require one of the four to have been a satyr-drama29— which has never been suggested, and for which there is no evidence. And a trilogy involving Odysseus Mainomenos fares no better. Its logical sequel is Palamedes; but if we want to avoid the repetition problem, the third play would have to be Aias Lokros, which on its own has no direct connection to the Palamedes story and would leave the trilogy without adequate thematic coherence. Aias Lokros makes sense, as part of a Palamedes trilogy, only if we have one of the Nauplius plays to follow, and we have already seen why that will not work. It is therefore most likely that all three of Sophocles’ Palamedes/ Nauplius plays were produced on separate occasions, probably at long intervals, and that two of them—Palamedes and Nauplios Katapleon—ended under the shadow of future events well known to the audience but completely, or almost completely, concealed from the characters. This is very common in Sophocles, occurring, in fact, in five of his seven surviving plays. Trachiniae ends with Heracles, already in agony, going to the further agony of a living pyre in obedience to an oracle, and his son bitterly reproaching the
28
Lloyd-Jones (1996: 249). In theory, one of them could instead have been a ‘prosatyric’ play like Euripides’ Alcestis—or, as I have elsewhere argued (in Sommerstein, Fitzpatrick, and Talboy 2006: 100–3), like Sophocles’ Syndeipnoi (The Diners). In a connected tetralogy, it is not necessary, or even usual, for the fourth play to be in chronological sequence with the first three; and in a Palamedes–Nauplius tetralogy the only possible candidate for either satyric or prosatyric status (both of which require a happy ending) would be Odysseus Mainomenos, of which there survive only six uninformative fragments totalling sixteen words. Even Redondo (2003), who is ready to declare a play probably satyric on remarkably little evidence, does not make any such suggestion in the case of Odysseus Mainomenos. 29
Sophocles’ Palamedes and Nauplius plays
257
apparently uncaring Zeus;30 we know, but no one onstage does, that the pyre is Heracles’ route to immortality.31 In the final scenes both of Oedipus the King and of Oedipus at Colonus, much of our attention is directed to Oedipus’ daughters; in one play he entrusts them to the care of Creon, in the other they seek assistance to return to Thebes in the hope of making peace between their brothers32—and we know too well what will happen there, both to the brothers and to Antigone, and what role Creon will have in it. In Electra Aegisthus is made to make an enigmatic reference to ’the future woes of the Pelopids’;33 we may not know precisely what he means,34 but in a moment, when the play abruptly terminates before the death of Aegisthus, it will become obvious that Sophocles has deliberately ended it, so to speak, too soon, and—to say the least—we cannot be as sure as the chorus seem to be35 that a satisfying closure has been reached. And in Philoctetes Sophocles quite unnecessarily makes Heracles warn Philoctetes and Neoptolemus, when they capture Troy, to be respectful towards the gods36—thus reminding us that Neoptolemus, whose blossoming heroism and maturing moral strength we have so much admired as the play unfolded, will become Neoptolemus the sacrilegious murderer of the aged Priam. Euripides does something similar in Trojan Women,37 but elsewhere he nearly always makes coming events
30
Soph. Trach. 1266–78. That the story of Heracles’ apotheosis from the pyre was known to the audience of Trachiniae (as Holt (1989) and R. L. Fowler (1999: 167–74) had powerfully argued, contra Stinton (1986: 84–99) and (1987)) is proved by Aesch. fr. 75a; see Hahnemann (1999). 32 Soph. OT 1462–510; OC 1769–72. 33 Soph. El. 1497–8. 34 See Ch. 16 above, and Finglass (2007: 526–8). 35 Soph. El. 1508–10. These lines are deleted by, among others, Dawe (1984) and Finglass (2007); but the chorus have already made their attitude clear, saying that they ‘cannot censure’ the killing of Clytaemestra (1423) and that, as regards her, Orestes and Pylades have ‘put things in a good state’ (1434—possibly echoing Aesch. Ag. 1673 where Clytaemestra herself is the speaker). 36 Soph. Phil. 1440–4; see Ch. 18 below. 37 When at the end of Trojan Women the Greeks sail away from Troy with their human and other booty, they are heading for a series of disasters, engineered by the gods and by Nauplius, of which both they and their captives (except perhaps Cassandra) are utterly unaware. 31
258
Sophocles’ Palamedes and Nauplius plays
clear to the characters as well as the audience; and Aeschylus too, so far as we can tell, does not normally leave the future hanging loose. This widespread Sophoclean pattern seems to have been followed in Palamedes and Nauplios Katapleon. In Palamedes, it is those guilty of Palamedes’ judicial murder, and those complicit with them, who are in the dark: we know he will be avenged. In Nauplios Katapleon it may be a little different. We can be fairly sure now, from the papyrus Hypothesis,38 that Nauplius left Troy with threats on his lips. We do not know precisely what these threats were; as we have seen, the tradition offered more than one possibility. Indeed, we do not even know whether Nauplius himself defined his intentions clearly enough for the audience to understand precisely what it was that he was threatening. Were they to think of his wrecking of families, his wrecking of the fleet, or both? If the former (with or without the latter) he will at least cause the destruction of one of those against whom he is most aggrieved, Agamemnon, and possibly also harm Odysseus.39 If the threat was solely against the fleet, then Nauplius’ sole important victim will be the lesser Ajax—for in virtually all accounts, just as in the Odyssey,40 he is the only one of the Greek leaders who perishes at sea. And in that case we will note, in the first place, that he is as much, or more, the victim of Athena than of Nauplius, and secondly that we have no evidence that Sophocles made him as an individual do anything to harm Palamedes or his father. That fact might give Nauplios Pyrkaeus itself an ironic closure. It is the drama of Nauplius’ revenge, but that revenge, while terrible, is erratic in its aim and may give him only limited satisfaction; the simultaneous revenge of Athena, in contrast, is swift, surgical, and almost effortless.41 38 See n. 12 above. The last surviving words of the Hypothesis say that Nauplius departed ØÐ ‚ººÅØ Æ[ : the only plausible supplements that have been proposed _ _ are I[ غøÐ ] (Maehler) and I[º ØÆ I غøÐ ] (Parsons). to complete the sentence 39 See n. 6 above. 40 Od. 4.498–511. 41 This chapter is based on a paper presented, as part of a panel convened by Isabelle Torrance, at the Annual Conference of the Classical Association (together with the Classical Association of Scotland) in Glasgow on 5 April 2009. It corresponds in substance to the last section (‘A trilogy?’) of my introduction to Palamedes, Nauplios Katapleon, and Nauplios Pyrkaeus in Sommerstein, Talboy, and Clark (forthcoming); I am grateful to Oxbow Books for giving me permission to include it in this volume.
18 ‘The rugged Pyrrhus’: the son of Achilles in tragedy The rugged Pyrrhus, he whose sable arms, Black as his purpose, did the night resemble When he lay couched in the ominous horse, Hath now this dread and black complexion smear’d With heraldry more dismal; head to foot Now is he total gules; horridly trick’d With blood of fathers, mothers, daughters, sons, Baked and impasted with the parching streets, That lend a tyrannous and damned light To their lord’s murder: roasted in wrath and fire, And thus o’er-sized with coagulate gore, With eyes like carbuncles, the hellish Pyrrhus Old grandsire Priam seeks. Shakespeare, Hamlet II. ii
When we think of Neoptolemus, the son of Achilles, as a character in Greek tragedy, we think overwhelmingly of just one play, the only surviving tragedy, in fact, in which he appears on stage as a living man: Sophocles’ Philoctetes. He is one of the two leading characters in this play, hardly less prominent than Philoctetes himself. Indeed, in some ways he is more prominent: he is on stage significantly longer than Philoctetes (though he has much less to say),1 and it might 1
Neoptolemus is present for lines 1–675, 730–1080, 1222–471 (in all 1,276 lines or nearly seven-eighths of the play); Philoctetes for lines 219–675, 730–1217, 1263–1471 (in all 1,153 lines). In the Oxford text of Lloyd-Jones and Wilson (1990a), Neoptolemus speaks about 361 lines and Philoctetes speaks or sings about 616, more than two-thirds as many again.
260
‘The rugged Pyrrhus’
almost be said that the main action of the play takes place in his mind, as we have recently been reminded in a superb paper by Ismene Lada-Richards (2009). And taking our impression of him over the whole play, he is a thoroughly sympathetic character, one of the most sympathetic male characters in the whole tragic genre. Introduced by Sophocles into a story with which he had previously had nothing to do, he is represented as a youth hardly more than a boy. Initially he is an obedient assistant to Odysseus who needs a little persuasion to embark on his mission of deceit, but once signed up, he performs it skilfully and effectively. Then from what seems like a combination of many factors—the impression produced upon him by Philoctetes’ admiration, affection, and trust, which he is about to betray; the impact of witnessing Philoctetes’ agony and helplessness when attacked by the malady that has beset him for nine years; the returning awareness of the principles on which he was reared and the example of the father he never knew—he collapses and reveals the truth to his victim. Odysseus manages to shock him (rather than convince him) into renewed compliance, but this time he recovers quickly, rebels again, defies all Odysseus’ threats, and gives Philoctetes back his precious bow; and when his final, honest attempt to induce Philoctetes to come to Troy fails, he prepares to fulfil his original promise to the sick man to take him home, until the deified Heracles intervenes to give Philoctetes instructions which, coming from such a source, he cannot refuse, and to promise glory to both men alike. This, we may well feel, is a true son of Achilles—of the Achilles of the Iliad when he is at his best as a human being—though of course without his almost superhuman prowess as a warrior; and we are constantly invited throughout Philoctetes to compare Neoptolemus with his father, explicitly by himself and by Philoctetes,2 and implicitly every time he is addressed, as he almost always is, as ‘son of Achilles’.3 It comes as a pretty big shock if one turns directly from this to consider the presentation of Neoptolemus in most archaic art and poetry. Neoptolemus comes across in these sources as something of 2
Phil. 91, 357–8, 874, 904–5, 1310–13; cf. 468, 719, 933. Phil. 3–4, 50, 260, 364, 542, 582, 1066, 1237, 1284, 1433; cf. 57, 96, 240–1, 242, 940, 1298. 3
‘The rugged Pyrrhus’
261
an adulescentulus carnifex, a teenage butcher. In art he is most typically associated with acts of brutal violence, brutally depicted: the slaying of Priam, the execution of Astyanax for the crime of having been born, the sacrifice of Polyxena.4 All three of these already figured in the poems of the Epic Cycle;5 the last-mentioned is not there explicitly attributed to Neoptolemus, but it is firmly associated with him in texts and images of the sixth century.6 Of deeds that can be placed to his credit we hear only of a considerable number of scalps won in battle (mostly on the night of the sack of Troy, but including the earlier killing of Eurypylus, the last major ally who tried to come to the Trojans’ aid)7 and probably the later rescue of the aged Peleus from exile or hardship.8 I should mention, too, what appears to be the earliest version of the story of Neoptolemus’ death at Delphi (found in Pherecydes, and assumed to be known to the audience of Pindar’s Sixth Paean),9 according to which he went to Delphi for some legitimate reason, got into a dispute with the priests over what he regarded as their pilfering of sacrificial meat, and was killed by one of them with a sacrificial knife; that is a story without strong positive
4 Neils (1994) nos. 87–94, 115–26 (Priam—often Astyanax too); Touchefeu (1984) nos. 7–20 (Astyanax); Touchefeu-Meynier (1994) nos. 24, 26 (Polyxena). 5 Little Iliad frr. 18, 25, 29 West (Astyanax and Priam, both killed by Neoptolemus); Sack of Troy Arg. }2 (Priam killed by Neoptolemus) and }4 (Astyanax killed by Odysseus, Polyxena by the Greeks with no individual specified) West. Glaucus of Rhegium (fr. 6a Mu¨ller) cites the cyclic epic Cypria for the story that Polyxena died of wounds inflicted by Odysseus and Diomedes the night Troy was captured, and that Neoptolemus buried her (Cypria fr. 34 Bernabe´ ¼ 27 Davies (not in West)); if the citation is right, the story must have been told in a flash-forward (perhaps in connection with the death of Troilus, an episode in which, in archaic art, Polyxena is regularly involved). 6 Touchefeu-Meynier (1994) no. 26; Ibycus PMG 307. 7 Little Iliad Arg. }3 (Eurypylus) and frr. 21, 23, 24, 27 West (three or four victims on the night of the sack, the most notable being Agenor, son of Antenor, who had defied Achilles when the latter was carrying all before him and then had been rescued by Apollo, Il. 21.525–605). 8 Nostoi Arg. }4 West ‘Neoptolemus . . . reaches Molossia and is recognized by Peleus’—which implies that Peleus was in exile from Phthia. Both the Iliad (24.486–92) and the Odyssey (11.494–503) seem to allude to an already existing tradition that Peleus suffered misfortunes after Achilles went to Troy; Priam and the ghost of Achilles both assume this to be a fact, though neither is in a position to know whether it is true! 9 Pherecydes fr. 64 Fowler; Pind. Paean 6.100–20.
262
‘The rugged Pyrrhus’
or negative implications—it shows Neoptolemus as somewhat hottempered, to be sure, but not without good reason for anger. I said ‘in most archaic art and poetry’ because, as quite often happens, Homer is the big exception. The references to Neoptolemus in the Odyssey, coming as they do mostly in Odysseus’ conversation with Achilles in the underworld (11.492–537), are almost entirely concerned with his role in the Trojan War (Odysseus knowing nothing of any subsequent events)—and they completely ignore all the negative aspects of it, despite Odysseus’ promise to tell ‘the whole truth’ (11.507). To judge by the Odyssey, Neoptolemus was nothing but a fine young warrior of whom his father could be truly proud. He was forward in planning, second only to Odysseus himself and Nestor (11.510–12); he was forward in battle, with his killing of Eurypylus particularly notable; of those in the Wooden Horse, he alone showed no fear; and during the sack of the city . . . he received no injury (11.533–7). I am not sure, but I think we may be meant to notice what Odysseus is leaving out—though recent commentators have been strikingly silent.10 (One observes, not for the first time, how difficult it often is to perceive an absence—to notice when dogs do nothing in the night-time.) Certainly there are hints in the Iliad that Homer does know about two at least of Neoptolemus’ three atrocities. Andromache foresees the possibility of her son being hurled from the walls of Troy ‘by some Achaean . . . whose brother, or father, or son Hector has slain’;11 we are probably meant to know that while she is right about how her son will die, she is wrong about the perpetrator, since none of the leading Greek warriors could have such a motive—so it is not at all unlikely that Homer’s audience already knew Neoptolemus as the perpetrator.12 And certainly the 10 The omissions attract no comment from Stanford (1959); Heubeck (in Heubeck and Hoekstra 1988); P. V. Jones (1988); or De Jong (2001). Heubeck (on 11.533– 7) does raise the question ‘to what extent the details related in the Iliou Persis . . . belong to pre-Homeric saga’, but he is thinking of something that Odysseus does mention (that Neoptolemus took ‘a share and a noble prize’ away with him) and wondering whether this alludes to Andromache (mentioned in Sack of Troy Arg. }4 West). 11 Iliad 24.734–8. 12 In which case they will know that Astyanax was killed, not ‘by some Achaean . . . whose brother, or father, or son Hector [had] slain’, but by the Achaean whose father had slain Hector and who had himself slain Hector’s father.
‘The rugged Pyrrhus’
263
poignancy of the moment when Priam kisses the hand of the man who killed his son13—which is also the hand of the man whom his son (Paris) will kill—will be intensified yet further if we know besides that this same man’s son will kill Priam himself, and in a particularly horrific fashion. And if so, Odysseus in the underworld is being economical with the truth; probably we are meant to ascribe this to his anxiety to comfort the depressed Achilles. But the poet himself appears also to be playing that game; in the same way, when reporting Demodocus’ song of the Iliou Persis earlier in the Odyssey (8.499– 520), he skates over all the details of the sack except for one episode involving Odysseus, and makes no mention of Neoptolemus at all. There were thus three different Neoptolemoi, so to speak, available for exploitation by the tragic poets. There was the Neoptolemus whom Odysseus describes, a warrior sans peur et sans reproche, the Neoptolemus also who rescued his grandfather in a tale which, as we have seen, the Homeric poems may hint at more than once. There was the Neoptolemus of the Delphi story, a young man whose heart was in the right place but who might get himself into trouble, as young men sometimes can, through hot-headedness or lack of tact. And there was the ruthless thug who in different ways, but with equal cruelty, slaughtered an old man, a maiden, and a young boy— Priam, Polyxena, and Astyanax. How did the tragedians exploit these options? Well, for a start, there is a remarkable contrast in the extent to which they did so. Aeschylus never to our knowledge makes any mention of Neoptolemus, and none of his plays, extant or fragmentary, dealt with any story in which Neoptolemus was involved. (He did not figure in Aeschylus’ Philoctetes—or, for that matter, in the play on the same subject by Euripides—because according to what appears to have been the canonical sequence of events, reported in Proclus’ summary of the Little Iliad,14 the mission to fetch Neoptolemus from Scyros to join the war was not sent until well after Philoctetes had arrived at Troy.) Neoptolemus plays a role of sorts in several surviving Euripidean plays—quite an important one in Andromache, while he figures more 13 14
Iliad 24.477–9, 504–6. Little Iliad Arg. }}2–3 West.
264
‘The rugged Pyrrhus’
briefly in Hecuba and The Trojan Women15—but in none of these does he ever come on stage, except (in Andromache) as a corpse. At the other end of his life, he may have appeared as a baby in The Scyrians, which dealt with the unmasking of the disguised Achilles on Scyros, and in which reference is known to have been made to the physical condition of Neoptolemus’ mother, Deidameia.16 It is possible that he did appear as an actual dramatis persona in another lost play, Peleus; but that depends on whether the play was in fact built around the sufferings of Peleus in old age (rather than one or another of the many vicissitudes of his earlier life), and on that there is nothing remotely like decisive evidence (either way).17 And except for a brief mention in Orestes (1654–7), that (so far as we know) is that. Euripides, in fact, almost seems to go out of his way to keep Neoptolemus offstage, even in plays that deal with some of the major episodes of his life. There are one or two other characters who have an important role in the plot of this or that play but conspicuously fail to appear onstage (notably Odysseus in Iphigeneia at Aulis),18 but none is kept out of sight so often or so consistently as this. Of the three major tragedians, so it appears, it is Sophocles alone who made Neoptolemus a dramatic character in the ordinary sense of the word, what is more, he probably composed plays centring on six of the eight recognized major events in Neoptolemus’ life—his decision to go to Troy (The Scyrians); the battle in which he killed Eurypylus (Eurypylus, in which Neoptolemus almost certainly remained offstage); the sacrifice of Polyxena (Polyxena); the rescue of Peleus (Peleus); his marriage to Hermione; and his death (the last two
15
Eur. Hec. 23–4, 223–4, 523–68; Tro. 274, 573–6, 657–68, 697–705, 1123–46. Eur. fr. 682 (cf. 683). There is, however, no actual evidence indicating whether the infant Neoptolemus was brought onstage, and the discussions by Jouan and van Looy (1998–2003: iii. 62–70) and by Collard and Cropp (2008: ii. 159–61) make no mention of the possibility. 17 It has been almost universally agreed that the play presented the story of Peleus and the wife of Acastus, and Jouan and van Looy (1998–2003: iii. 531–6) take this for granted in their discussion, but they succeed only in showing that the fragments are compatible with this assumption; so far as I can see, they are equally compatible with several other possibilities. Collard and Cropp (2008: ii. 72–3) say only that ‘it seems likely’ that the play centred on ‘the failed seduction’ of Peleus by Acastus’ wife. 18 Cf. Eur. IA 107, 203–4, 524–35, 1362–6. 16
‘The rugged Pyrrhus’
265
both in Hermione)—plus a further play, Philoctetes, which brought him for the first time into an established story about another hero. In this chapter, I will be focusing mainly on the two plays out of these six in which the presentation of Neoptolemus is highly negative, Polyxena and Hermione, and also considering how this presentation was reworked by Euripides in Hecuba and Andromache respectively—for there are good grounds for holding that in each of these two pairs of parallel plays, that of Sophocles was the earlier.19 And then I shall have a little bit more to say about Philoctetes. But first, a rapid survey of the other three Sophoclean plays. It is clear that The Scyrians was about a successful attempt to bring a Peleid hero from Scyros to Troy, but there is still some controversy over whether this hero was Neoptolemus or Achilles. Like Jouanna,20 I have little doubt that it was Neoptolemus. In one fragment (Soph. fr. 557) a young person tells an old man that lamenting will not bring the dead back to life, and gives as an example his own grief for his father; this fits Neoptolemus very well and does not at all fit Achilles, whose father is alive. Sophocles, it seems, showed Neoptolemus determined to go to the war, no doubt against the opposition of his mother and grandfather—exactly as his father will have been in Euripides’ play and exactly as a Homeric hero should be. And yet they are not entirely wrong to be apprehensive about his future: he will survive the war without a scratch—but all the same, his family on Scyros will never see him again. Sophocles’ Eurypylus is known to us only from a long but (for the most part) extremely tattered papyrus,21 and from one quotation by Plutarch which seems to match it (fr. 210.8–9). Neoptolemus is, of course, seen as the enemy, but in those references to him that survive he seems to be quite favourably presented: he and Eurypylus are said (in the Plutarch quotation) to have approached each other on the battlefield ‘without boasting or insults’, and Eurypylus’ corpse was not maltreated by the victors (fr. 210.47–9). This restraint, for which Neoptolemus would have to be held largely responsible, contrasts strikingly with the treatment of Hector’s 19 20 21
See my discussions in Sommerstein, Fitzpatrick, and Talboy (2006: 20–1, 65–6). Jouanna (2007: 660–1). POxy 1175 þ 2081(b) ¼ Soph. frr. 206–222b.
266
‘The rugged Pyrrhus’
body by Achilles,22 and it would be all the more impressive if it is true that Eurypylus had been, in Greek eyes, guilty of perfidy in violating a pledge given by Telephus that he and his family would not fight against the Greeks. This pledge is known to us only from a confused note by a scholiast on Juvenal,23 but it would handily explain the surprising conviction of Eurypylus’ mother Astyoche that she has committed a crime which all will rightly think deserving of death.24 So far as we can tell, neither of the two chief mourners, Astyoche and her brother King Priam, makes any hostile remark about Neoptolemus (contrast Hecuba in Iliad 24.206–14); he is simply a warrior doing his job. There is no direct, incontrovertible evidence that Sophocles’ Peleus dramatized the rescue of Peleus from banishment and distress by the returning Neoptolemus, but it is certain that in this play Peleus was an old man in very reduced circumstances (Soph. fr. 487), and there is no other known story about him that would fit this and the other surviving fragments; we do hear of accounts in which Peleus wanders in search of his grandson and dies on the island of Icos,25 but in this play Peleus is living in a dwelling which, however poor, is apparently his own abode (fr. 487.1), not that of a foreign host, and fr. 492, referring to a ‘king of the land of Dotium’, suggests that the action is set on the mainland of Thessaly. Whether or not the account of the rescue by ‘Dictys of Crete’ (6.7–9) derives from Sophocles’ play (it is certainly tragic, but it seems to me rather too elaborate for Sophocles—Neoptolemus adopts one disguise to trap the sons of Acastus, his grandfather’s persecutor, and then a different disguise to trap 22
Iliad 22.395–405, 24.14–54. Juv. 6.655. The scholiast is commenting on a passage referring to Eriphyle (the wife of Amphiaraus, who, after being bribed by her brother Adrastus, persuaded her husband to join the expedition against Thebes), and—doubtless led astray by the similarity of the names Eriphyle and Eurypylus—confuses her with Astyoche, who, after being bribed by her brother Priam (cf. Od. 11.519–21; Little Iliad fr. 6 West; Acusilaus fr. 40 Fowler), persuaded her son to come to Troy to aid its defenders. A promise by Telephus not to assist the Trojans—but one binding on himself only—is also mentioned in A Il. 1.59. 24 Soph. fr. 210.45. 25 Eur. Tro. 1128 (where ‘Icos’ has become ‘Cos’ in the mss.); Pind. Pyth. 3.167; AP 7.2.9–10 (Antipater of Sidon). There was a cult of Peleus on Icos (Call. fr. 178.23–4). 23
‘The rugged Pyrrhus’
267
Acastus himself—and I suspect it of being fourth-century), Sophocles must clearly have presented Neoptolemus here as a man of valour, resource, and pietas. Now to Polyxena, famous in antiquity for the appearance of Achilles’ ghost, at the moment when the Greek fleet was about to set sail, demanding that Polyxena be sacrificed.26 On the role of Neoptolemus in this play we have no direct evidence; our few fragments are mostly concerned with the appearance (or appearances) of the ghost, with a quarrel between Agamemnon and Menelaus (adapted from Odyssey 3) over whether to depart at once or to stay and make (animal) sacrifices (Soph. fr. 522), and with some prophecies about the fate of the Achaeans in general and Agamemnon in particular (Soph. fr. 526 and probably 525). But two features of two other plays suggest that Neoptolemus was an important figure in this one. One is his large role in Seneca’s Troades, where he quarrels with Agamemnon over whether the sacrifice of Polyxena should be carried out or not; the other is his very small role in Euripides’ Hecuba, which is likely to be at least in part due to a desire to vary from Sophocles’ treatment. And there is also one key pointer in what we know of Sophocles’ play itself. In one fragment (524) a king, certainly Agamemnon, and evidently about to have to make a major decision, complains that whatever he decides, he is certain to be criticized by someone, just as even Zeus incurs resentment whether he sends rain or whether he doesn’t. This cannot refer to the quarrel with Menelaus, since in that quarrel Agamemnon was not an arbiter but was one of the contending parties himself; and since it is hardly likely that the play contained three major quarrels in which Agamemnon was involved, the decision over which he is agonizing must be the decision whether to sacrifice Polyxena, and he must be about to give a verdict between a character arguing for the sacrifice and a character arguing against it. The latter is probably Polyxena herself, since she ought to have a major role in a play that bears her name; the former was identified by Calder27 as Odysseus, but in view of the other indications I have mentioned is more likely to be Neoptolemus. 26
‘Longinus’ 15.7. Calder (1966: 35–7); see my remarks in Sommerstein, Fitzpatrick, and Talboy (2006: 61 n. 41). 27
268
‘The rugged Pyrrhus’
It is not clear whether Agamemnon gave the decision, or whether in fr. 524 he is clearing his throat for a decision not to decide—which is what he does in Seneca,28 passing the buck to the prophet Calchas. The sacrifice itself is presented relatively soothingly (so far as a human sacrifice can be) both in Euripides and in Seneca;29 in both accounts Polyxena behaves with noble courage, and in both Neoptolemus is in the end a reluctant slaughterer. But there existed in the fifth century another tradition about this sacrifice, which we know of from a late epigram apparently based on a painting by Polygnotus:30 in the epigram Polyxena is beseeching to be spared, and (on the most obvious interpretation) her garments have been torn, not by her own hand (as in Euripides)31 but by those of her handlers.32 If so, the Sophoclean Polyxena, like the Aeschylean Iphigeneia33 (and like the Polyxena of some archaic vases), may have gone to her death resisting and having to be restrained by force. So, taking all things together, there is enough to suggest that in Sophocles’ treatment, Neoptolemus may well have argued strenuously for the sacrifice to be performed, and may have carried it out in a notably brutal way. It has long been generally, and I am sure rightly, held, that Euripides’ Hecuba is later than, and (as Wilamowitz put it) ‘presupposes’,34 the Polyxena of Sophocles. And it is striking how much Euripides cuts down and defangs Neoptolemus’ role. Early on there is a bit of a red herring, when the ghost of Polydorus mentions the killing of Priam at the altar by ‘the polluted murderer, Achilles’ son’,35 which we may well take, especially if we recall Sophocles’ play, as a 28
Sen. Tro. 349–59. Eur. Hec. 518–82; Sen. Tro. 1118–64. APl 150 (Pollianus); the poet confuses Polygnotus the painter (cf. Paus. 1.22.6) with Polycleitus the sculptor, his contemporary. 31 Eur. Hec. 558–61. 32 She is trying to conceal her nakedness with her hand ºØ Þƪ (lines 3–4); the combination of these two ideas in one sentence suggests that her robe has been completely, or almost completely, torn in two, uncovering her whole body. In Euripides, where she tears the robe herself, she tears it only down to the waist (Hec. 559). 33 Aesch. Ag. 231–8. 34 Wilamowitz-Moellendorff (1889: 355 n. 28): ‘Hekabe . . . setzt Sophokles Polyxene voraus’. I set out the evidence in Sommerstein, Fitzpatrick, and Talboy (2006: 65–6). 35 Eur. Hec. 23–4. 29 30
‘The rugged Pyrrhus’
269
trailer for a major and murderous role for Neoptolemus in this drama too. It doesn’t happen. We are given a detailed account of the debate in the Greek army about whether the demand of Achilles’ ghost should be met,36 and Neoptolemus plays no part: Agamemnon opposes the sacrifice, the sons of Theseus advocate it, and the decisive voice is that of a character always available when amoral Realpolitik is wanted, Odysseus; he it then is, too, who takes Polyxena away from her mother. Neoptolemus still performs the sacrifice itself; Euripides could hardly have avoided that, since the sacrifice was part of his father’s tomb-cult. But he does so reluctantly, pitying his victim, and the army admire her courage and immediately prepare for her a splendid if improvised funeral. It is as if Euripides wanted to draw our condemnation away from Neoptolemus, who is about the least guilty (except for Polyxena herself) of all those involved in the sacrifice—when one compares him with Odysseus and the sons of Theseus who argued for it, with Agamemnon who only argued against it because of his passion for Cassandra,37 even with Hecuba who will later take a most grisly revenge on the Thracian Polymestor (including the killing of his children before his eyes, prior to putting them out) and who will end up, so we are told, literally dying like a dog for it.38 As in The Children of Heracles, where the persecuting tyrant Eurystheus becomes the protector of Athens against the descendants of those whom Athens rescued from him, nothing is as simple as you might expect.
36
Ibid. 117–40. Ibid. 120–9. Ibid. 1257–73; it is not made absolutely explicit that this is to be seen as a punishment for her treatment of Polymestor, and Mossman (1995: 196–201) argues that it should not be so regarded; but 1257–9 (‘You take pleasure in committing hybris against me!’—‘Should I not enjoy taking revenge on you?’—‘Well, you won’t be enjoying it when . . . ’) recall the idiom h Ø åÆæø vel sim. used in making threats of punishment or revenge both in comedy (Ar. Ach. 563; Knights 235, 828; Wasps 186; Thesm. 719) and in tragedy (Eur. Med. 395–8, HF 258, Or. 1593), and Buxton (1982: 183) notes that Hecuba’s transformation into a dog would make appropriate requital for the way she forced Polymestor to walk on all fours (Hec. 1056–9). That Hecuba expresses unconcern about this prospect (1274) is not evidence that the fate in store for her is not to be regarded as an evil one, any more than the words of the would-be tyrant in Solon fr. 33.5–7 West are evidence that being flayed alive was not thought of as an evil fate. 37 38
270
‘The rugged Pyrrhus’
In Sophocles’ Hermione and in Euripides’ Andromache, the story, which goes back at least to the Odyssey, of Neoptolemus’ marriage to Hermione, daughter of Menelaus and Helen, was brought into relation with that of his death at Delphi. I have argued elsewhere, more than once,39 that the Sophoclean play which is three times cited as The Women of Phthia was identical with Hermione, basically because its known fragments fit the plot of Hermione well and do not fit any other story to which we could plausibly assign a chorus of Phthian women, and I will not do so again here. Several other Sophoclean plays have similar double titles, one referring to the chorus: Atreus or The Women of Mycenae, The Prophets or Polyidus, Nausicaa or The Laundresses, Pandora or The Hammerers, and we might add Aeschylus’ The Carians or Europa (if his it is)40 and Semele or The Water-carriers. If an argument that I developed some years ago41 is correct, we could conclude with considerable probability that Hermione therefore dates from a time before the titles of plays were publicly announced in advance of performance, that is, before the completion of the Odeion in the early 440s. As already indicated, there are other grounds for believing that it is earlier than Euripides’ Andromache, which belongs to the mid 420s, which I have discussed in my edition of the fragments of Sophocles’ play; I would draw attention in particular to Euripides’ illogicality in having the Delphians leave Neoptolemus unburied,42 for no adequate reason (but very conveniently for the dramatist, who is thereby enabled to bring his corpse on stage back at Phthia),43 which can easily be explained as having been taken over from Sophocles’ play where, as we shall see, they had the best of reasons for not wanting to bury him in their soil. The crucial evidence for the presentation of Neoptolemus in Hermione is provided by an ancient note on the passage in the 39
In Sommerstein, Fitzpatrick, and Talboy (2006: 14–17), and in Ch. 4 above. Against its authenticity see West (2000: 347–50). In the Addenda to Ch. 1, above, I suggested that the play’s double title tells somewhat (though not massively) against West. 41 See Ch. 1 above. 42 Eur. Andr. 1156–7; see my discussion in Sommerstein, Fitzpatrick, and Talboy (2006: 20–1). 43 Whence it is almost immediately taken to Delphi again (Andr. 1239–42, 1263–4). 40
‘The rugged Pyrrhus’
271
Odyssey (4.3) in which Telemachus, arriving at Sparta, finds Menelaus celebrating the marriage of his daughter Hermione to the son of Achilles; I quote it in the version of Eustathius, which is somewhat fuller than that found in the regular scholia. Sophocles, they say, in Hermione narrates that while Menelaus was still at Troy, Hermione had been given in marriage by Tyndareos to Orestes, then later taken away from him and given to Neoptolemus according to the promise [made by Menelaus] at Troy. But when he [Neoptolemus] had been killed at Pytho by Machaereus, when he was trying to avenge the slaying of his father by punishing Apollo, she was restored again to Orestes.
‘Trying to avenge the slaying of his father by punishing Apollo’ sounds headstrong and dangerous. But it is not necessarily, in itself an act of unpardonable impiety; if that were all, Neoptolemus would have been going only a little further than Achilles had done when, finding that a disguised Apollo had decoyed him away and allowed the Trojans to escape to their city, he railed at the god: ‘Now you have taken great glory away from me and saved them easily, because you had no fear of future retribution: I tell you I would punish you, if I had the power!’44 However, we have some evidence that fleshes out a little this vague phrase ‘punishing Apollo’. There is another summary of the story which repeats two of the distinctive points ascribed by the Odyssey commentator to Sophocles’ version, but adds a crucial further detail. It comes in the epitome of pseudo-Apollodorus (6.13–14), who gives two versions of Neoptolemus’ end. One of these, which may or may not be based on Euripides, has him killed at Delphi by Orestes because he had taken Hermione away from him. The other says that ‘when he came to Delphi he demanded satisfaction from Apollo for [the death of] his father, plundered the dedications and set fire to the temple, and because of this he was killed by Machaereus the Phocian’. If this was Sophocles’ version (and, as just noted, it does correspond to Sophocles’ version in two of its three major features), it presents Neoptolemus as a robber and burner of temples, no better than (say) King Xerxes, the sort of criminal whom the law of Athens
44
Iliad 22.18–20.
272
‘The rugged Pyrrhus’
would not allow to be buried in its soil;45 the same sort of criminal, in fact, that dragged Priam from the altar.46 It finds an echo in Euripides, when Orestes, stirring up the Delphians against Neoptolemus, tells them, quite falsely, that he ‘is here a second time for the same reason he came before [i.e. to demand satisfaction], and wants to plunder Phoebus’ temple’ (Andr. 1094–5). In Sophocles’ play Neoptolemus was certainly a character (he is the speaker of fr. 20247 and the addressee of fr. 694), and presumably Sophocles took steps to present him as the sort of person who might be hybristic enough to commit so terrible a sacrilege. If he declared, before leaving Phthia, his intention of doing so, that might well call forth from an older kinsman or friend, such as Peleus or Phoenix, the kind of rebuke we hear in fr. 694: ‘You are young; you need to learn a lot, be told a lot, and be taught a great deal.’ But this play wasn’t only about Neoptolemus’ death at Delphi. It was also about Hermione, and she was prominent enough in the play to give her name to it. As in Euripides, she was unhappy in her marriage to Neoptolemus, and it is not unlikely that she had been sending secret messages to Orestes (cf. Eur. Andr. 964, a reference to such messages48 which in its own play is completely isolated and may well be an allusion to Sophocles’ treatment). Indeed, the most straightforward, though not the only possible, understanding of Eustathius’ words is that Hermione was already living with Orestes when she was ‘taken away’ by Menelaus (rather as Smikrines, in Menander’s Epitrepontes, tries to take his daughter Pamphile away from Charisios) to be given to Neoptolemus; two of Sophocles’ contemporaries, Philocles and Theognis, went one step further and said she was pregnant at the time.49 This breaking-up of a marriage
45
Xen. Hell. 1.7.22. Little Iliad fr. 25 West. 47 I prove this in Sommerstein, Fitzpatrick, and Talboy (2006: 14): Neoptolemus is the only character in the play who could be shown just arriving in his father’s homeland (ÆæfiÆ ªÅÐ) from abroad. 48 See Zielin´ski (1925: 114–17). More recent interpreters have generally either suspected the passage of being corrupt (e.g. Stevens 1971: 207), or taken Orestes to mean that he has disregarded messages from Hermione telling him not to come (Lloyd 1994: 151; but why on earth should she send such messages?). 49 Philocles fr. 2; Theognis trag. fr. 2. 46
‘The rugged Pyrrhus’
273
by a third party is particularly bad in this case because Menelaus, unlike Smikrines, is not revoking his own gift but someone else’s, and that of a person to whom he owes some respect, his father-in-law Tyndareos (but for whom, by the way, because of the oath he imposed on his daughter’s suitors,50 he would not now himself be living with Helen once more); and Neoptolemus, too, should have had more sense than to accept Hermione under such circumstances. We may well, however, suspect that that was not the worst of it, so far as Neoptolemus was concerned. Hermione was at the centre of the action for a long time, and in view of the circumstances of her marriage to Neoptolemus, it is likely that she was presented much more sympathetically than she is in Euripides; while one would expect that Neoptolemus, the hybristes of Delphi, would also be shown behaving hybristically towards Hermione, though nothing survives either to confirm or refute this. In short, it seems highly likely that the Neoptolemus of Sophocles’ Hermione was a man who had no respect or consideration for his elders, for his wife, for the gods, or for his own best interests, and little to arouse any sympathy except that he was still very young; of the characters in surviving tragedy, he was perhaps more reminiscent of Pentheus than of any other—and it may be no coincidence that one line of the play (fr. 695, spoken by Phoenix to Peleus or possibly vice versa) was taken over verbatim by Euripides in Bacchae (line 193, spoken by Cadmus to Teiresias). We may not think all that much of the Neoptolemus of Euripides’ Andromache either. I didn’t, before I became acquainted with the evidence about Sophocles’ play. But Euripides’ original audience did know of Sophocles’ play, and will have evaluated the new one in the light of it; and in that light it appears, once again, that Euripides has defanged Neoptolemus to a considerable extent. For most of the play we are allowed to see him only through the eyes of others, some of whom are sympathetic to him (Andromache, Peleus), while those 50 This oath (first mentioned in Hes. fr. 204.78–85 M–W ¼ 155.78–85 Most; cf. Stesichorus PMG 190) is in general taken for granted in tragedy as part of the background to the Trojan War (Soph. Aj. 1111–14, Phil. 72–3; Eur. IA 49–71); similarly Thuc. 1.9.1 assumes that it is commonly supposed that Tyndareos’ oath was what enabled Agamemnon to muster so large an expedition. See Jouan (1966: 156–64).
274
‘The rugged Pyrrhus’
who are hostile (Hermione, Menelaus, Orestes) all prove to be treacherous, double-dealing murderers. He finally appears in person only as a corpse, to be mourned by Peleus, traditionally a man of outstanding virtue, in this play the heroic rescuer of Andromache and her child who humiliates Menelaus by the sheer force of his personality, and now, save for a bastard great-grandson, the sole survivor of his family. He was apparently not responsible for the killing of Astyanax, which is mentioned by Andromache (9–10) without any reference to him, while Hermione (who could have made good rhetorical use of it) is not allowed to mention it at all. The murder of Priam is ignored completely. Hermione makes no complaint, at any time in the play, about her marriage to Neoptolemus or his treatment of her—and when she finds herself facing the dire consequences of her failed attempt to murder his only child, she goes so far as to say—to Orestes—that if he puts her to death, it will be no more than she deserves (920). For the making of the marriage, Orestes, the injured party, originally blamed Menelaus (967), though later he ‘forgave’ Menelaus’ breach of faith (he does not say why: are we meant to infer that it was because Menelaus was too powerful politically to risk making an enemy of him?) and instead tried to get Neoptolemus to withdraw from the match. All we are left with is three actions by Neoptolemus, all of them unwise, but none of them morally heinous. He agreed to marry Hermione when he was already living with Andromache and had a child by her, thus planting a conflict at the heart of his household (cf. 465ff.). He rejected Orestes’ request to give up this marriage, in a manner that Orestes perceived as grossly insulting, though he apparently spoke only what was true (977–8).51 And he demanded satisfaction from Apollo for the death of his father; one passage (1094–5), to which I have already referred, could be read as meaning that, as in Sophocles, he had plundered the temple or threatened to do so—and indeed I formerly did read it so—but if Neoptolemus had done that, he would hardly have left Delphi alive. The text says only that he wanted to sack the temple, and even that is only an allegation by Orestes. And in any case Euripides completely rebalances the moral scales by having 51
Namely, that Orestes had killed his mother and had then been pursued by the Erinyes.
‘The rugged Pyrrhus’
275
Neoptolemus return to Delphi to apologize and seek forgiveness for that error (51–5, 1104–8), and making that the occasion when Orestes, the Delphians, and Apollo combine to do him to death— Orestes recruiting the Delphians by a campaign of lies, and Apollo, who of course knows the truth, acting as his accomplice, rallying the Delphians with a fearsome cry when Neoptolemus seems on the point of defeating them single-handed (1147–9, cf. 1161–5). Philoctetes is almost certainly the latest of all the plays we have discussed—if we ignore the passing mention of Neoptolemus in Euripides’ Orestes, in which his marriage to Hermione is, as it were, abolished, presumably because after witnessing the murder of her mother, being seized as a hostage, threatened with instant death, and then told she is going to be married to the man who did all this, all within about half an hour, even Apollo thinks she deserves a bit of peace in her life. And in Philoctetes it is Sophocles’ turn to confound expectations. He does so, of course, in the first place, by having Neoptolemus in the story at all; his audience probably have no idea who the young man accompanying Odysseus might be, until he is named. After that, for a long time, we are encouraged to expect that Neoptolemus, despite his upbringing and his admiration for his father, will prove corruptible. He resists Odysseus’ seductive persuasion only briefly (86–120), and for the next 800 lines—well over half the play—carries out his instructions pretty well perfectly, displaying from time to time a remarkable talent both for straight lying and for treacherous equivocation. The intervention of the so-called False Merchant52 was entirely unnecessary; indeed, had Odysseus been less impatient, Neoptolemus would have delivered Philoctetes to him within a few more minutes. Even when Philoctetes suffers an
52 I do not know why translators, commentators, and critics regularly give this character this clumsy designation—or, even worse, refer to him as ‘the merchant’, which he is not. He is, we know, the man who had acted as lookout to warn Odysseus and Neoptolemus if Philoctetes was seen returning to his cave (Phil. 45–8, 125–9). He is formally, it seems, under Neoptolemus’ command, since Odysseus at first does not himself give him orders but asks Neoptolemus to do so (45), so presumably he is one of the crew of Neoptolemus’ ship. He should therefore be labelled ‘Sailor’. Raeburn (2008) reasonably lists him in the dramatis personae as ‘sailor posing as a merchant’, but then, in the scene where he is an active character, labels his speeches with ‘merchant’!
276
‘The rugged Pyrrhus’
agonizing attack of his malady in Neoptolemus’ presence, and then falls asleep, it seems to make no difference; when he begins to stir, Neoptolemus at once hisses a warning to the chorus to stay silent (865). And then, at the moment when all he has to do is walk off to his ship with Philoctetes leaning on his arm, he breaks down (895)— or, if you prefer, his true nature breaks through. In earlier plays by Sophocles we had had Neoptolemus the True Hero, and we had had two varieties of Neoptolemus the Villain. Euripides had given us, twice over, Neoptolemus the Invisible-But-Apparently-Not-So-VeryVillainous-After-All. It looked as though Sophocles was going to give us, this time, Neoptolemus the Sophist’s Apprentice. And now he has suddenly become at least Neoptolemus the Decent Youth, and we are not too surprised when offstage, some time between lines 1080 and 1222, he grows up into Neoptolemus the True Hero after all. When, on his words ‘Let’s go, if you want to’ (1402), the metre changes to trochaic tetrameters, the play seems about to end—and some audiences have applauded at 1408 as though it was the end. La prise de Troie n’aura pas lieu. But, of course, it isn’t the end. If it were, it wouldn’t be only the structure of mythological history that would suffer severely; so would other things, for which at this instant we probably care a good deal more. Philoctetes would be cheated of his cure, and both he and Neoptolemus of their glory. Heracles puts that all right. But having done that, he says a bit more, and these are his last significant words: ‘But remember this: when you conquer the land, be reverent in all that concerns the gods; for my father Zeus holds everything else of less account than that. Men’s piety does not die with them; whether they are alive or dead, it never perishes’ (Phil. 1440–4). He is addressing both Philoctetes and Neoptolemus. And every spectator knew that while one of them obeyed this injunction, the other did not. Sophocles didn’t need to put in that last touch. But he did put it in, giving us, in a final twist, a fourth Neoptolemus in the same play: Neoptolemus the Couldn’t-Keep-It-Up. Truly it seems that Neoptolemus was a character with whom one could do almost anything.53 53 This chapter is based on a lecture prepared for delivery at a London meeting of the Society for the Promotion of Hellenic Studies in 2009, a commitment that I was unable to fulfil owing to family circumstances.
19 What ought the Thebans to have done? Something goes terribly wrong in Bacchae. The god Dionysus, son of the Theban princess Semele, has returned to Thebes to establish his cult there. Every spectator watching the play at the City Dionysia knows that this cult, like Dionysus’ gift of wine, can bring great pleasure and great release of mental tension. And yet its arrival in Thebes produces catastrophe. Semele’s nephew Pentheus, the king of Thebes, is torn in pieces by his mother Agaue, her sisters Ino and Autonoe, and the other women of Thebes; Agaue returns to Thebes in triumph, believing she has killed a lion, and displaying Pentheus’ head on the end of her ritual rod (thyrsos); and finally the entire family, including Agaue’s aged father Cadmus, are expelled from the city. Why has this happened? In one sense, Dionysus himself answers this question right at the beginning. Semele, made pregnant by Zeus, had perished through the guile of Zeus’s ever-jealous consort Hera. The story (told somewhat allusively in Bacchae) is that Hera, in disguise, persuaded Semele to ask Zeus to visit her in his full divine splendour; Zeus, having promised Semele to do whatever she wished, could not refuse her request, and the lightning-fire of his presence destroyed her. Zeus snatched her unborn infant from the flames and sewed him up in his own thigh, eventually to be ‘born’ a second time. Cadmus turned Semele’s house and tomb into a [33] shrine (6–12). Her sisters Agaue, Ino, and Autonoe took a very different view (26–31). They claimed that Semele had really been pregnant by a mortal lover, at her father’s suggestion <she> had covered her shame by pretending Zeus was the father of her baby, and that Zeus had destroyed her to punish this lie. Because of this slander, the
278
What ought the Thebans to have done?
three sisters, with all the other women of Thebes, have been ‘driven in madness from their homes’, and made to wear Dionysus’ sacred garb and perform his rituals (32–8). That, however, need not have had catastrophic results. If Thebes, as Dionysus puts it (39–40), ‘learn[s] . . . that it is uninitiated in my bacchic cult’, and adopts it officially, Dionysus will doubtless let the women go home (cf. 804–7). But will Thebes do that? Not if Pentheus has his way. He ‘fights against the gods,’ says Dionysus, ‘or at least against me, debars me from libations, and does not mention me in his prayers’ (45–6); he may try to bring the women home by force—in which case Dionysus will ‘join with the maenads as their general’ (52) with unstated, but evidently fatal, results for Pentheus. Dionysus’ indictment of Pentheus may at first sight seem a little unjust. Pentheus was abroad when Dionysus came to Thebes (215), and has had no way of knowing that he even exists, much less that he demands universal worship. What is more, it is perfectly reasonable that he should believe Semele’s infant had perished (since that is what normally happens when a pregnant woman suffers a violent death, and no one in Greece knows of any evidence to the contrary) and perfectly reasonable that he should believe the baby’s father was not Zeus (for Zeus would not have destroyed his own offspring). But whatever sympathy we may initially have for him [34] is dissipated with extraordinary rapidity as we get to know him better. Dionysus’ condemnation is quite unjustified in terms of what Pentheus had done previously; but what Pentheus does subsequently shows, at the very least, that he was riding for a fall of some sort. Pentheus starts putting himself in the wrong almost from the moment he appears. He has imprisoned numerous freeborn women in chains (226–32) on the basis of unsubstantiated rumour (‘I hear’ 216, ‘they say’ 233). If he catches their priest (who is really Dionysus in disguise) he will execute him by stoning (356) or by decapitation (241)—a punishment utterly abhorrent to Greeks—and there is no indication that there will be any trial. He tells his own grandfather that he is making a ridiculous fool of himself (250–2, 344–5). He accuses the prophet Teiresias of being complicit in the introduction of the new cult in the hope of increasing his professional income (255–7)—and we know that when someone accuses Teiresias of corruption it is always a bad sign (ask Oedipus, or Creon
What ought the Thebans to have done?
279
in Antigone); later he orders the physical destruction of Teiresias’ seat of augury (346–51), thus putting himself in the wrong with Apollo as well as Dionysus. No wonder Teiresias ends the scene (367–9) by hoping that Pentheus, whose name means ‘man of grief ’, may not bring grief to Cadmus and his family. So far, as we have seen, Pentheus has spoken and acted entirely on the basis of —some of which will prove completely false (notably the allegation that the women’s bacchic rites are a cover for sexual debauchery). Now he begins to receive authentic information which makes it obvious that a god is at work. The guard who has arrested the supposed [35] priest reports that the imprisoned women have been miraculously liberated, their feet being unchained, and the prison doors opened, by no mortal hand (443–8): Pentheus’ response, incredibly, is to order the guards to let go of the priest’s arms, ‘because now he is in the net, he is not speedy enough to escape me’ (451–2). It is as if he had not heard what had just been said to him—not the last instance of such selective deafness or blindness. And after an interview in which contempt of the new cult is strangely mingled with curiosity (cf. 471–80), he has the ‘priest’ locked up in the palace stables, and gratuitously adds that his Asian followers will be seized as slaves (511–14). There follows another miracle, or series of miracles, of which this time Pentheus is an eyewitness. A mighty voice is heard; the palace shakes; the sacred fire on Semele’s tomb suddenly flares up; and presently the ‘priest’ walks calmly out of the palace, his prison having been demolished. Pentheus meanwhile has been kept busy tying up a bull (618–21), fighting the fire, and trying to kill a phantom (629–31). Surely by now he must have some suspicion that he is up against something too powerful for him? Not the least: his next order is to close all the city’s gates (653) as if this were a routine jailbreak. The final proof that Pentheus’ whole approach is both false and dangerous comes immediately afterwards, as a herdsman arrives with news from Mount Cithaeron. He has seen the bacchants on the mountain, and they are as orderly as a military garrison: they are divided into three companies each under a commander (680–2), they rise promptly at reveille (689–94) and adjust their uniforms [36] carefully (695–8), they perform their rituals at fixed times (723–4), and sexuality and drunkenness are nowhere to be detected. On the
280
What ought the Thebans to have done?
other hand, the impossible and the miraculous seem to be matters of routine: the women use live snakes as belts (698), suckle fawns and wolf-cubs (699–702), and by a stroke or a touch make the ground flow with milk, honey, water, and wine (704–11). All utterly amazing, and utterly unthreatening—except of course that society will collapse if the women aren’t restored to sanity, and to Thebes, fairly soon. This, though, cannot be done by force, as the rest of the herdsman’s narrative makes clear. When he and his friends, encouraged by a know-all from the city (717–21), try to capture the women, they immediately run amok, tear cattle in pieces, raid two villages, and put the armed inhabitants to flight; and loads do not burden them, fire does not burn them, weapons do not wound them. How will Pentheus respond to this? He responds in his accustomed manner. He has by now witnessed, or been credibly informed of, fifteen or twenty manifest miracles. And as ever, he neither accepts nor denies their reality; he just ignores them. He orders an immediate military expedition against the bacchants (780–5), vowing to ‘stir up a great deal of women’s blood’ on Mount Cithaeron (796–7)—oblivious, it seems, to the fact, of which we have recently been repeatedly reminded (682, 690, 720, 728), that one of these women is his own mother. It is perhaps significant that he refers to them as his ‘slaves’ (803), as if he were the King of Persia1—doubly ironic, this, considering how he prides himself on his Greekness (483, 779) in contrast to barbarians like the Lydian ‘priest’ and his followers. We now expect the threatened military expedition (cf. 52) to be launched, and Dionysus to take command [37] of his maenad army as promised. What happens is rather different. For at this moment Dionysus shows that he is, after all, a god of justice. Pentheus has shown himself a tyrant with no respect for man or god, ready to insult his grandfather, imprison or kill his mother, chop off heads on mere suspicion, and treat free people like slaves, wilfully blind and deaf to the plainest evidence that a superhuman agency is at work—and yet Dionysus offers him a way out. ‘I will 1 On the Greek belief that barbarians were the ‘slaves’ of their rulers, cf. Eur. Hel. 276, and see Hall (1989: 193–7). Another tyrant, Lycus, speaks similarly in Eur. HF 251.
What ought the Thebans to have done?
281
bring the women here’, he says, ‘without the use of arms’ (804). Thebes can be restored to normality without any blood being shed. Of course, there is a price: the definitive establishment of Dionysiac cult at Thebes (807–8). But why on earth not? On all the available evidence, the cult brings pleasure and, unless provoked, no pain. But Pentheus will have none of it: ‘Bring me my armour out here. And you, stop talking’ (809). The disguised god doesn’t stop talking, and Pentheus never gets his armour; in fact, the next time we see him he will be dressed as a woman, and the time after that he will be a set of detached body parts. Pentheus’ approach to the Dionysiac phenomenon has been a disastrous failure, and this naturally leads us to ask what alternative approach, if any, could have been successful. The play actually offers us a considerable range of options, before it narrows its focus to show us the consequences of the one chosen by Pentheus. The first of these options is presented by the chorus of Lydian bacchants in their opening song (parodos). One part of this song narrates the birth(s) and infancy [38] of Dionysus; other parts give a detailed and evocative picture of his ecstatic worship, first in the streets (64–87), then in the mountains (135–69). The song is full of ‘barbarian’ elements: the women are Asian, they wear the weird garb of bacchic ritual, carry and doubtless beat drums; there are repeated references to Lydia, Phrygia, even Syria, all of which to the Greek mind were places that slaves came from. The mountain ritual consists of running, leaping, falling, dancing, singing, shouting, with drums, pipes, and torches, and its focus and object is to hunt down goats, tear them in pieces and eat their raw flesh (139). It may well be doubted how anyone could think it wise to introduce such practices as this into a society that wished to be sane and safe. In particular, a crucial psychological prerequisite, apparently, is what the chorus call ‘communalizing the soul’ (75–6): abandoning individuality, merging one’s personality in the swarm. Such an experience can be extremely uplifting and pleasurable. It can also, however, be extremely dangerous: ‘herd instinct’, ‘crowd hysteria’, ‘mob violence’, are all designations of what can happen if this process gets out of hand. The Greek polis is a place where law rules and where the individual citizen is responsible for his actions. At the very least, even if one accepts that a degree of uninhibited release may be beneficial, it must
282
What ought the Thebans to have done?
be kept within some kind of boundaries. There is no sign that this chorus is willing to accept anything of the kind. They claim authority over the streets and houses and their inhabitants (68–70); they become animalized, wearing fawnskins and galloping like fillies (137–8, 164–9); they tear animals apart as if they owned them, not caring that goats might have a goatherd. If this is what Thebes and Pentheus are being offered, one can well understand if they feel they have to reject it. [39] Next we see two Thebans, Teiresias and Cadmus, both aged men. They may not be able to gallop like fillies or tear goats in pieces, but they do believe it is their duty to join in the worship of Dionysus. They have, though, two very different approaches to the cult. The fate of Cadmus in the play may well seem unjust. Early in the prologue Dionysus praises Cadmus for creating a shrine to Semele (10–11), and we learn also that he, unlike his daughters, had always publicly maintained that Semele’s lover had been none other than Zeus. On learning that Semele’s son has returned to Thebes as a god, Cadmus is eager to do him honour. That Cadmus is in the end bereaved of his treasured grandson is of course not inconsistent with this: the punishment of the guilty inevitably has side-effects on their innocent kin. But Cadmus also receives from Dionysus a specific punishment of his own: exile from Thebes to a barbarian land, and transformation into a snake (1330–4, 1354–60). It is true that he will eventually go to the Isles of the Blest (1337–8), but overall his fate is undoubtedly meant as a punishment, as Dionysus makes clear in words that are addressed to Cadmus as well as his daughters: ‘If you had known to be sensible, when you refused to be, you would now be happy, with the son of Zeus for your ally. . . . You have understood us too late; you did not know us when you should have done. . . . I was a god, and you insulted me’ (1341–7). And Cadmus admits the charge (1344, 1346, 1377–8), even while pleading for mercy. What is he admitting being guilty of? If the play offers an explanation at all (and surely it ought to), it must [40] come in the scene in which Cadmus and Teiresias are confronted by Pentheus. It may not matter much that Cadmus tells Pentheus that ‘even if, as you claim, this god does not exist, you should pretend he does’ for the honour of Semele and the family (333–6); taken by itself, that might be just Cadmus desperately trying to get through to
What ought the Thebans to have done?
283
Pentheus with an argument that might possibly appeal to him. But Cadmus had spoken like that to Teiresias too: I have come prepared, wearing this sacred attire of the god; for since he is my daughter’s son, it is right that he should be magnified and glorified to the best of our ability.
(180–3)
Admittedly Cadmus, as the first mortal ever to have a god for a grandchild, is in a unique position with no precedent to guide him, but his language strongly suggests that he is magnifying and glorifying Dionysus only because Dionysus is his grandson—doing, in fact, precisely what he later recommends Pentheus to do, and ‘telling a fine lie’ in the family’s interest. His attitude also changes the significance of what we heard in the prologue: the insistence that Semele’s lover had been divine, the creation of a shrine to her, now look like the contrivances of a head of family determined to put a positive spin on what might have been a very shaming episode. To say the least, we cannot be sure he actually believes Dionysus is a god. Now it is true that Greek religion, generally speaking, was a matter of practice rather than of belief. The gods, on the whole, didn’t mind what mortals thought about them, so long as they received their dues in the form of prayer and especially sacrifice. But as the [41] trial of Socrates would soon show, belief couldn’t be ignored, because it could have an effect on practice. If the belief spreads that the gods do not really exist, sooner or later the community will decide that it can use its resources better than by sending them up in smoke to these non-existent gods—and if by any chance the gods do in fact exist, such a decision could have unfortunate consequences. And Dionysus, in particular, can only be truly worshipped by those who throw their whole being into the act of worship—and one can hardly be doing that if one is thinking of Dionysus as a family asset. It is therefore appropriate that whereas the chorus of Asian devotees condemns Pentheus (263–5) and praises Teiresias (328–9), they say nothing at all about Cadmus. Teiresias, in tragedy, is nearly always right, and the audience may well look to him for some words of pious wisdom in an attempt to set Pentheus on a more prudent path. They may perhaps be disappointed. Teiresias speaks less like a prophet than like a sophist. He rationalizes everything; all the mystery and the ecstasy disappear.
284
What ought the Thebans to have done?
Dionysus is the inventor of (or perhaps is) wine, which is good for drowning sorrows, inducing sleep, and pouring in libations (278–85). He was never sewn up in Zeus’s thigh; that story was created through the confusion of two similar-sounding words (286–97). He also has (in reality rather minor) connections with prophecy and war (298–305). And while Dionysus will not compel women to be chaste, a woman who is truly chaste by nature will not change her nature under his influence (which, for an overwhelmingly male audience that was not disposed to think particularly highly of women’s moral capacity, would beg a very big question). All this is all very well, but it has little to do with the kind of ecstatic worship [42] that we have heard partly described, partly enacted by the chorus in their opening lyric. It is an attempt to tame Dionysus and make him into a reasonable god that reasonable, educated late fifth-century Athenians can rationally worship. If that is the kind of god he wants to be, he would never have come to Thebes accompanied by these outlandish barbarian women. Or so we probably think, until the scene ends and the outlandish barbarian women sing their second ode. And to our surprise we find that all the wild ecstasies have vanished, to be replaced by the not exactly sober, but at any rate bounded and socialized pleasures of that civilized, urban(e) institution, the symposium (376–85). The chorus commend ‘a life of quietness and good sense’ (389–90) and warn against the danger of ‘not thinking mortal thoughts’ and of ‘chasing great things’ instead of making the best of what one can get (395–9). They pray to be taken to Cyprus or Pieria (402–16), the lands of Aphrodite and of the Muses. The connection between Dionysus and the Muses would be obvious to every Athenian spectator; and Aphrodite, as the Herdsman will remind us (773–4), is (in the right context) an essential part of the good life. In the final stanza Dionysus’ association with ‘the painless delight of wine’ is re-emphasized, as is his democratic nature: he offers his gifts ‘equally to the prosperous and the inferior’ (421–2). More than once the chorus express a rejection of what may be called intellectualism (to sophon, 395); true intelligence (sophia) is something different—it is accepting the human condition and the opinions and practices of ‘the masses of ordinary people’ (430). This distinction between two contrasting connotations of sophia—being clever enough to know what’s good
What ought the Thebans to have done?
285
for one, and being too clever for one’s own good—runs right through the play.2 [43] Is this the answer we have been seeking? It seems surprisingly banal and naive, and resolutely oblivious to the tragic aspects of human existence. Accept life’s pleasures—song, dance, food, drink, sex, sleep, and good company (the listing is Kenneth Dover’s3—in another, though still Dionysiac, context—and all of them figure in the choral ode we have been looking at); don’t try to rationalize them into conformity with some high-flown philosophical schema; respect popular traditions, especially religious ones; that is true wisdom, and will bring happiness. Naive or not, this view does seem to be endorsed by all the Theban commoners we see in the play. The guard who arrested the disguised god was reluctant to do so and told his prisoner as much (441–2), and he tells Pentheus ‘this man . . . is full of miracles’ (449–50). The herdsman, though fearful of the king’s anger (670–1), urges him to accept the new god, not least because ‘they say . . . he gave mortals the vine which puts a stop to grief; and if there is no more wine, there is no more Aphrodite, nor any other human pleasure’ (771–4). And the messenger who reports Pentheus’ death concludes that ‘to know one’s place (so¯phronein) and revere the divine is best, and I think it is also wisest’ (1150–1). There is, after all, no contradiction in believing that life has both its tragedies, many of them unavoidable, and its pleasures, and that since total renunciation of the pleasures will not help one escape the tragedies, it is foolish not to accept them (within reason, of course). If Pentheus has rejected normal pleasures (and one certainly doubts if he’d make a good companion for a symposium), it seems only to have led him to seek perverted ones: his way of showing affection for his grandfather had been to ask whom Cadmus wanted him to punish (1310–12, 1318–22), [44] and he is lured to his death by 2
Good sophia: 266, 427, 480, 641, 656, 839, 1151. Bad sophia: 30, 200, 203. The two are contrasted at 395. Pentheus condemns the sophistries, as he sees them, of the priest who is in fact Dionysus at 489 and 655; in his later delusion he praises the sophia of the disguised god as the latter schemes to destroy him (824), and Agaue, also deluded, praises the god’s sophia in urging on the maenads to hunt the ‘lion’ who was in fact her son (1190). 3 Dover (1968: p. liii), listing the good things of which, in the world of Aristophanic comedy, ‘the normal man . . . takes as much as he can’.
286
What ought the Thebans to have done?
being offered the chance to spy on the maenads’ alleged sexual orgies. And his evidence-proof hatred and contempt for Dionysus and all he stands for leads to an appalling catastrophe that was entirely avoidable. It remains true that though Pentheus and Cadmus, Agaue and Ino and Autonoe, have all deserved to suffer, they all suffer far more than they deserve, or than anyone deserves—and that is the note on which the play ends. It may well be, as Richard Seaford supposes,4 that Dionysus’ speech as deus ex machina (most of which has been lost from the only manuscript) included instructions for establishing an organized cult of Dionysus at Thebes; but the Thebans who will benefit from that cult are not on stage. Those who are present are Cadmus, the lonely old man facing a future he loathes; Agaue, the mother who has killed her son; and that son’s dismembered body, over whose torn limbs, one by one, Agaue, in another lost passage, had lovingly lamented. As often in Euripides, human love seems the only consolation in a cruel world: father and daughter part with an embrace (1363–7), and Agaue will share her exile with her sisters (1381–2). They themselves may have made the world crueller than it need have been, but we can pity them nevertheless—as even the chorus do (1327–8). And we can go a little beyond pity. In the play’s last genuine words, Agaue says: May I go where foul Cithaeron shall never see me again nor I set eyes on Cithaeron, and where no thyrsos is dedicated, to remind me! [45] Let them be the concern of other bacchants! (1383–7)
Just as the savage zeal of Pentheus was counter-productive, so too, at least in one respect, has been the zealous savagery of his cousin Dionysus. It has benefited nobody, at least in Thebes (since Dionysus could have destroyed Pentheus, or rendered him harmless, in many less atrocious ways), and it has irrevocably alienated this sister of Semele. It is entirely understandable that Agaue should thus loathe and shun that which caused her ruin. She cannot be expected to 4
Seaford (1996: 252–3).
What ought the Thebans to have done?
287
reflect—but we should—that so many things in this world have caused the ruin of someone at some time that we cannot possibly shun them all. We have seen in this play that Dionysus is a risky, destabilizing god; but deliberately, obsessively to avoid all risk and instability is itself risky and destabilizing. And deliberately, obsessively to avoid pleasure as such merely ensures that life will be all pain—as if there wasn’t enough of that anyway.5 5 This chapter was originally published in T. Shasha and D. Stuttard ed. Essays on Bacchae (Brighton: aod publications, 2006) 32–45; and again in A. Beale (ed.), Euripides Talks (London: Bristol Classical Press, 2008) 23–32. Inserted page numbers refer to the first publication. I am most grateful to the Actors of Dionysus for giving permission for this republication, to which a few footnotes have been added.
References Ahl, F. M. 1991. Sophocles’ Oedipus: Evidence and Self-Conviction (Ithaca, NY). Ahrens, E. A. J. 1844. Sophoclis Fragmenta (Paris). Allan, W. 2001. Euripides: The Children of Heracles (Warminster). —— 2008. Euripides: Helen (Cambridge). Arnott, W. G. 1996. Menander II (Cambridge, Mass.). Astin, A. E. 1967. Scipio Aemilianus (Oxford). Austin, C. F. L. 1987. ‘Textual problems in Ar. Thesm.’ Dodone 16: 61–92. Austin, N. 1994. Helen of Troy and her Shameless Phantom (Ithaca NY). Badian, E. 1993. From Plataea to Potidaea: Studies in the History and Historiography of the Pentecontaetia (Baltimore). Bain, D. M. 1979. ‘A misunderstood scene in Sophokles’. Oidipous (O.T. 300–462). G&R 26: 132–45. Bakola, E. 2010. Cratinus and the Art of Comedy (Oxford). Baldry, H. C. 1952. ‘Who invented the golden age?’ CQ 2: 83–92. —— 1954. ‘Aristotle and the dramatization of legend’. CQ 4: 151–7. —— 1956. ‘The dramatization of the Theban legend’. G&R 3: 24–37. Barrett, W. S. 1964. Euripides: Hippolytos (Oxford). Bauman, R. A. 1990. Political Trials in Ancient Greece (London). Bearzot, C. 1992. ‘Ancora sulle Eumenidi di Eschilo e la riforma di Efialte (in margine ad una pagina di Chr. Meier)’. Prometheus 18: 27–35. Bees, R. 1993. Zur Datierung des Prometheus Desmotes (Stuttgart). Bindzus, D. 1956. ‘Griechische Arbeitsgemeinschaft der OIg des Staatlichen Gymnasiums zu Flensburg 1955–1956’. Dioniso 19: 227–8. Blundell, S. 1986. The Origins of Civilisation in Greek and Roman Thought (London). Booth, N. B. 1976. ‘Zeus hypsistos megistos: an argument for enclitic ı in Aeschylus, Agamemnon 182’. CQ 26: 220–8. Bowie, A. M. 1993. Aristophanes: Myth, Ritual and Comedy (Cambridge). Bowra, C. M. 1936. Greek Lyric Poetry from Alcman to Simonides (Oxford). —— 1944. Sophoclean Tragedy (Oxford). Broadhead, H. D. 1960. The Persae of Aeschylus (Cambridge). Brown, A. L. 1984. ‘Eumenides in Greek tragedy’. CQ 34: 260–81. —— 1987. Sophocles: Antigone (Warminster). Burian, P. 2007. Euripides: Helen (Oxford).
References
289
—— 2009. ‘Inconclusive conclusion: the ending(s) of Oedipus Tyrannus’, in S. D. Goldhill and E. M. Hall (eds.), Sophocles and the Greek Tragic Tradition (Cambridge) 99–118. Burkert, W. 1972/1983. Homo necans: The Anthropology of Ancient Greek Sacrificial Ritual and Myth (trans. P. Bing) (Berkeley). Burton, R. W. B. 1980. The Chorus in Sophocles’ Tragedies (Oxford). Butrica, J. L. 2001. ‘The lost Thesmophoriazusae of Aristophanes’. Phoenix 55: 44–76. Buxton R. G. A. 1982. Persuasion in Greek Tragedy: A Study of Peitho (Cambridge). —— 1987. ‘Wolves and werewolves in Greek thought’, in J. Bremmer (ed.), Interpretations of Greek Mythology (London) 60–79. Calder, W. M. 1966. ‘A reconstruction of Sophocles’ Polyxena’. GRBS 7: 31–56. Campbell, D. A. 1993. Greek Lyric, v. The New School of Poetry and Anonymous Songs and Hymns (Cambridge, Mass.). Campbell, L. 1879–81. Sophocles (2nd edn.) (Oxford). Carawan, E. M. 1987. ‘Eisangelia and euthyna: the trials of Miltiades, Themistocles, and Cimon’. GRBS 28: 167–208. Carcopino, J. [1928] 1967. Autour des Gracques (2nd edn.) (Paris). Carey, C. 1988. ‘A note on torture in Athenian homicide cases’. Historia 37: 241–5. —— 1990. ‘Aischylos Eumenides 858–66’. ICS 15: 239–50. Cartledge, P. A. 1990. Aristophanes and his Theatre of the Absurd (Bristol). Cassio, A. C. 1985. Commedia e partecipazione: la Pace di Aristofane (Naples). Cawkwell, G. K. 1963. ‘Eubulus’. JHS 83: 47–67. Chiasson, C. C. 1988. ‘Lecythia and the justice of Zeus in Aeschylus’ Oresteia’. Phoenix 42: 1–21. —— 1999/2000. ‘øçæ ıÐ K åæ ø fi : the Athenians and time in Aeschylus’ Eumenides’. CJ 95: 139–61. Clark, E. G. 1985. Review of Tyrrell (1984). LCM 10: 78–9. Classen, C. J. 1959. ‘The Libyan god Ammon in Greece before 311 B.C.’ Historia 8: 349–55. Clu´a Serena, J. A. 1985. ‘El mite de Palamedes a la Gre`cia antiga: aspects canviants d’un interrogant cultural i historic’. Faventia 7: 69–93. Collard, C. 1975. Euripides: Supplices (2 vols.) (Groningen). —— 2004. ‘Appendix Two [to Alexandros]: Euripides, Palamedes’, in C. Collard, M. J. Cropp, and J. Gibert (eds.), Euripides: Selected Fragmentary Plays II (Oxford).
290
References
Collard, C. and Cropp, M. J. 2008. Euripides: Fragments (2 vols.) (Cambridge, Mass.). Conacher, D. J. 1976. ‘Comments on an interpretation of Aeschylus’, Agamemnon 182–183’. Phoenix 30: 328–36. —— 1987. Aeschylus’ Oresteia: A Literary Commentary (Toronto). Conan Doyle, A. 1890. The Sign of Four (London); cited from The Complete Sherlock Holmes Long Stories (London, 1929). —— 1892. The Adventures of Sherlock Holmes (London); cited from The Adventures of Sherlock Holmes and The Memoirs of Sherlock Holmes, ed. I. Pears and E. Glinert (London, 2001). Crawford, M. H. 1981. The Roman Republic (London). Cropp, M. J. 1988. Euripides: Electra (Warminster). —— and Fick, G. 1985. Resolutions and Chronology in Euripides: The Fragmentary Tragedies (BICS Suppl. 43) (London). Csapo, E. G. 2007. ‘The men who built the theatres: theatropolai, theatronai, and arkhitektones’, in P. J. Wilson (2007: 87–115). Culican, W. 1965. The Medes and Persians (London). Cunningham, M. L. 1953. ‘A fragment of Aeschylus’ Aigyptioi?’ RhM 96: 223–31. —— 1962. ‘Second thoughts on Oxyrhynchus Papyri, XX, 2251’. RhM 105: 189–90. —— 1994. ‘Thoughts on Aeschylus: the satyr play Proteus—the ending of the Oresteia’. LCM 19: 67–8. Dale, A. M. 1967. Euripides: Helen (Oxford). Davidson, J. F. 1988. ‘Homer and Sophocles’ Electra’. BICS 35: 45–72. Davies, M. 1986. ‘Who speaks at Sophocles Antigone 572?’ Prometheus 12: 19–24. —— 1987. ‘Aeschylus’ Clytemnestra: sword or axe?’ CQ 37: 65–75. —— 1989. The Epic Cycle (Bristol). Davison, J. A. 1958. ‘Notes on the Panathenaea’. JHS 78: 23–42. Dawe, R. D. 1978a. ‘The end of Seven against Thebes yet again’, in R. D. Dawe, J. Diggle, and P. E. Easterling (eds.), Dionysiaca: Nine Studies in Greek Poetry . . . Presented to Sir Denys Page (Cambridge) 87–103. —— 1978b. Studies on the Text of Sophocles, III (Leiden). —— 1984. Sophocles: Tragoediae I (Leipzig). —— 2001. ‘On interpolations in the two Oedipus plays of Sophocles’. RhM 144: 1–21. De Jong, I. J. F. 2001. A Narratological Commentary on the Odyssey (Cambridge). Denniston, J. D. 1939. Euripides: Electra (Oxford). —— and Page, D. L. 1957. Aeschylus: Agamemnon (Oxford).
References
291
Des Bouvrie, S. 1990. Women in Greek Tragedy: An Anthropological Approach (Oxford/Oslo). De Ste. Croix, G. E. M. 1972. The Origins of the Peloponnesian War (London). Dickey, E. 1996. Greek Forms of Address from Herodotus to Lucian (Oxford). Diggle, J. 1984. Euripidis fabulae: Tomus I (Oxford). —— 1994. Euripidis fabulae: Tomus III (Oxford). Dodds, E. R. 1960. ‘Morals and politics in the Oresteia’. PCPS 6: 19–31. Reprinted in Dodds (1973: 45–63). —— 1973. The Ancient Concept of Progress and Other Essays on Greek Literature and Belief (Oxford). Dover, K. J. 1968. Aristophanes: Clouds (Oxford). —— 1972. Aristophanic Comedy (London). —— 1983. ‘Thucydides “as history” and “as literature”’. History and Theory 22: 54–63. Reprinted in Dover (1988: 53–64). —— 1988. The Greeks and their Legacy. Collected Papers, ii. Prose Literature, History, Society, Transmission, Influence (Oxford). —— 1993. Aristophanes: Frogs (Oxford). Earp, F. R. 1953. ‘The date of the Supplices of Aeschylus’. G&R 22: 118–23. Easterling, P. E. 1993. ‘Response to Sommerstein’, in J. H. Molyneux (ed.), Literary Responses to Civil Discord (¼ Nottingham Classical Literature Studies 1) (Nottingham) 19–24. Ehrenberg, V. 1950. ‘Origins of democracy’. Historia 1: 515–48. Erbse, H. 1978. ‘Zur Elektra des Sophokles’. Hermes 106: 284–300. Ersch, J. S. and Gruber, J. G. (eds.) 1818–89. Allgemeine Encyclopa¨die der Wissenschaften und Ku¨nste in alphabetischer Folge (Leipzig). Ewans M., Ley, G. K. H., and Mc Cart, G. 2000. Sophokles: Three Dramas of Old Age (London). Fagles, R., and Stanford, W. B. 1977. Aeschylus: The Oresteia (Harmondsworth). Farioli, M. 2001. Mundus alter: utopia e distopie nella commedia greca antica (Milan). Finglass, P. J. 2007. Sophocles: Electra (Cambridge). —— 2009. ‘Unveiling Tecmessa’. Mnemosyne 62: 272–82. —— forthcoming. Sophocles: Ajax (Cambridge). Fitzpatrick, D. G. 2003. ‘Sophocles’ Aias Lokros’. In A. H. Sommerstein (ed.), Shards from Kolonos: Studies in Sophoclean Fragments (Bari) 243–59. Foley, H. P. 1993. ‘The politics of tragic lamentation’, in A. H. Sommerstein et al. (eds.), Tragedy, Comedy and the Polis (Bari) 101–43. Forrest, W. G. 1960. ‘Themistocles and Argos.’ CQ 10: 221–41. —— 1975. ‘An Athenian generation gap’. YCS 24: 37–52.
292
References
Fowler, H. W. [1926] 1965. A Dictionary of Modern English Usage (2nd edn., rev. Sir E. Gowers) (Oxford). Fowler, R. L. 1999. ‘Three places of the Trachiniae’. in J. Griffin (ed.), Sophocles Revisited: Essays Presented to Sir Hugh Lloyd-Jones (Oxford) 161–75. Fraenkel, E. 1950. Aeschylus: Agamemnon (3 vols.) (Oxford). Friis Johansen, H. 1964. ‘Die Elektra des Sophokles’. C&M 25: 8–32. —— and Whittle, E. W. 1980. Aeschylus: The Suppliants (2 vols.) (Copenhagen). Frost, F. J. 1980. Plutarch’s Themistocles: A Historical Commentary (Princeton). Gantz, T. R. 1979. ‘The Aischylean tetralogy: prolegomena’. CJ 74: 289–304. —— 1980. ‘The Aischylean tetralogy: attested and conjectured groups’. AJP 101: 133–64. —— 1993. Early Greek Myth: Literary and Artistic Sources (Baltimore). Gardiner, C. P. 1987. The Sophoclean Chorus: A Study of Character and Function (Iowa City). Garner, R. 1990. From Homer to Tragedy: The Art of Allusion in Greek Poetry (London). Garvie, A. F. 1969. Aeschylus’ Supplices: Play and Trilogy (Cambridge). —— 1986. Aeschylus: Choephori (Oxford). —— 1998. Sophocles: Ajax (Warminster). —— 2005. Aeschylus’ Supplices: Play and Trilogy (2nd edn.) (Exeter). Garzya, A. 1987. ‘Sur la Niobe´ d’Eschyle’. REG 100: 185–202. Gatz, B. 1967. Weltalter, goldene Zeit und sinnverwandte Vorstellungen (Hildesheim). Gellie, G. H. 1972. Sophocles: A Reading (Melbourne). Gentili, B., and Perusino, F. 2002. Le orse di Brauron (Pisa). Germar, R., and Krumeich R. 1999. Proteus. In Krumeich, Pechstein, and Seidensticker (1999: 179–81). Gibert, J. C. 1997. ‘Euripides’ Hippolytus plays: which came first?’ CQ 47: 85–97. —— 2008. Review of Sommerstein, Fitzpatrick, and Talboy (2006). Exemplaria Classica 12: 259–66. Giuman, M. 1999. La dea, la vergine, il sangue (Milan). Goette, H. 2007. ‘An archaeological appendix [to Csapo 2007]’, in P. J. Wilson (2007: 116–21). Goldhill S. D. 1984. Language, Sexuality, Narrative: The Oresteia (Cambridge). —— 1990. ‘The Great Dionysia and Civic Ideology’, in Winkler and Zeitlin (1990: 97–129).
References
293
—— 1992. Aeschylus: The Oresteia (Cambridge). Gomme, A. W., and Sandbach, F. H. 1973. Menander: A Commentary (Oxford). Gregor, D. B. 1957. ‘' çºÆ’. CR 7: 14–15. Grene, D., and O’Flaherty W. D. 1989. The Oresteia by Aeschylus (Chicago). Grethlein, J. 2003. Asyl und Athen: Die Konstitution kollektiver Identita¨t in der griechischen Trago¨die (Stuttgart). Griffin, J. 1977. ‘The Epic Cycle and the uniqueness of Homer’. JHS 97: 39–53. Griffith, M. 1977. The Authenticity of Prometheus Bound (Cambridge). —— 1986. Review of Friis Johansen and Whittle (1980). Phoenix 40: 323–40. —— 2002. ‘Slaves of Dionysos: satyrs, audience, and the ends of the Oresteia’. CA 21: 195–258. Guthrie, W. K. C. 1957. In the Beginning (London). Hahnemann, C. 1999. ‘Mount Oita revisited: Sophocles’ Trachiniai in light of the evidence of Aeschylus’ Herakleidai’. ZPE 126: 67–73. Hale, L. 1958. John Philpot Curran: His Life and Times (London). Hall, E. M. 1989. Inventing the Barbarian: Greek Self-Definition through Tragedy (Oxford). —— 1996. Aeschylus: Persians (Warminster). Halleran, M. R. 1995. Euripides: Hippolytus (Warminster). Halliwell, F. S. 1984. ‘Ancient interpretations of O Æd Œø ø fi ØÐ in Aristophanes’. CQ 34:83–8. —— 1991. ‘Comic satire and freedom of speech in classical Athens’. JHS 111: 48–70. Handley, E. W. 1993. ‘Aristophanes and the generation gap’, in A. H. Sommerstein et al. (eds.), Tragedy, Comedy and the Polis (Bari) 417–30. Hansen, M. H. 1985. Demography and Democracy (Herning). —— 1991. The Athenian Democracy in the Age of Demosthenes (Oxford). Harrison, A. R. W. 1968. The Law of Athens, i. Family and Property (Oxford). Harrison, T. 1985/1986. Dramatic Verse 1973–1985 (Newcastle/ Harmondsworth). Hartigan, K. V. 1996. ‘Resolution without victory/victory without resolution: the identification scene in Sophocles’ Electra’, in F. M. Dunn (ed.), Sophocles’ Electra in Performance (Stuttgart) 82–92. Harvey, F. D. 2000. ‘Phrynichos and his Muses’, in F. D. Harvey and J. Wilkins (eds.), The Rivals of Aristophanes (London/Swansea) 91–134.
294
References
Harvey, F. D. 2005. ‘Tragic thrausmatology: the study of the fragments of Greek tragedy in the nineteenth and twentieth centuries’, in F. McHardy, J. Robson, and F. D. Harvey (eds.), Lost Dramas of Classical Athens: Greek Tragic Fragments (Exeter) 21–48. Heath, M. 1987. The Poetics of Greek Tragedy (London). —— and OKell E. R. 2007. ‘Sophocles’ Ajax: expect the unexpected’. CQ 57: 363–80. Heberlein, F. 1980. Pluthygieia: Zur Gegenwelt bei Aristophanes (Frankfurt). Henderson, J. J. 1987. Aristophanes: Lysistrata (Oxford). —— 1990. ‘The de¯mos and the comic competition’, in Winkler and Zeitlin (1990) 271–313. —— 1991a. The Maculate Muse: Obscene Language in Attic Comedy (2nd edn.) (New York). —— 1991b. ‘Women and the Athenian dramatic festivals’. TAPA 121: 133–47. Herington, C. J. 1970. The Author of the Prometheus Bound (Austin). Hester, D. A. 1971. ‘Sophocles the unphilosophical: a study in the Antigone’. Mnemosyne 24: 11–59. Heubeck, A., and Hoekstra, A. 1988. A Commentary on Homer’s Odyssey, ii. Books IX–XVI (Oxford). —— West, S. R., and Hainsworth, J. B. 1988. A Commentary on Homer’s Odyssey, i. Introduction and Books I–VIII (Oxford). Hogan, J. C. 1984. A Commentary on the Complete Greek Tragedies: Aeschylus (Chicago). —— 1991. The Plays of Sophocles: A Companion to the University of Chicago Press Translations (Bristol). Holt, P. 1989. ‘The end of the Trachiniai and the fate of Herakles’. JHS 109: 69–80. Holwerda, D. 1977. Scholia in Aristophanem. Pars 1, Fasc. 3.1: Scholia vetera in Nubes (Groningen). Hornblower, S. 1991. A Commentary on Thucydides, i (Oxford). Hose, M. 1993. ‘Kratinos und der Bau des Perikleischen Odeions’. Philologus 137: 3–11. Hutchinson, G. O. 1985. Aeschylus: Septem contra Thebas (Oxford). Huys, M. 1997. ‘Euripides and the “Tales from Euripides” ’. APF 43: 11–30. Ireland, S. 1986. Aeschylus (Oxford). Jebb, R. C. 1894. Sophocles: The Plays and Fragments. Part VI: The Electra (Cambridge). Jones, L. A. 1987. ‘The role of Ephialtes in the rise of Athenian democracy’. CA 6: 53–76.
References
295
Jones, P. V. 1988. Homer’s Odyssey: A Companion to the English Translation of Richmond Lattimore (Bristol). Jouan, F. 1966. Euripide et les le´gendes des chants cypriens (Paris). —— and Van Looy, H. 1998–2003. Euripide. Tome VIII: Fragments (4 vols.) (Paris). Jouanna, J. 1993. ‘E´lectre, trage´die du retour’, in A. Machin and L. Perne´e (eds.), Sophocle: le texte, les personnages (Aix-en-Provence) 173–87. —— 2007. Sophocle (Paris). Juffras, D. M. 1991. ‘Sophocles’ Electra 973–85 and tyrannicide’. TAPA 121: 99–108. Just, R. 1989. Women in Athenian Law and Life (London). Kaibel, G. 1911. Sophokles Elektra (2nd edn.) (Leipzig). Kaimio, M. 2000. ‘Tragic titles in comic disguises’, in J. Styka (ed.), Studies in Ancient Literary Theory and Criticism (Cracow) 53–69. Kamerbeek, J. C. 1974. The Plays of Sophocles, Commentaries V: Electra (Leiden). Kannicht, R. 1969. Euripides: Helena (Heidelberg). —— 2004. Tragicorum Graecorum Fragmenta, v. Euripides (2 vols., continuously paginated) (Go¨ttingen). Kassel, R., and Austin, C. F. L. 1983. Poetae comici Graeci, iv. AristophonCrobylus (Berlin). Kells, J. H. 1973. Sophocles: Electra (Cambridge). Keuls, E. C. 1978. ‘Aeschylus’ Niobe and Apulian funerary symbolism’. ZPE 30: 41–67. Kitto, H. D. F. 1956. Form and Meaning in Drama (London). —— 1961. Greek Tragedy (3rd edn.) (London). —— 1966. Poiesis: Structure and Thought (Cambridge). —— 1994. Sophocles: Antigone, Oedipus the King, Electra (2nd edn., rev. E. M. Hall) (Oxford). Kitzinger, R. 1991. ‘Why mourning becomes Electra’. CA 10: 298–327. Knoepfler, D. 1993. Les Imagiers de l’Orestie: mille ans d’art antique autour d’un mythe grec (Kilchberg/Zu¨rich). Knox, B. M. W. 1968. Review of G. Mu¨ller (1967). Gnomon 40: 747–60. Kopperschmidt, J. 1971. ‘Hikesie als dramatische Form’, in W. Jens (ed.), Die Bauformen der griechischen Trago¨die (Munich) 321–46. Kovacs, P. D. 1994. Euripidea (Leiden). Kovacs, P. D. 1995. Euripides: Children of Heracles, Hippolytus, Andromache, Hecuba (Cambridge, Mass.). —— 2002. Euripides: Helen, Phoenician Women, Orestes (Cambridge, Mass.).
296
References
Krumeich, R., Pechstein, N., and Seidensticker, B. 1999. Das griechische Satyrspiel (Darmstadt). Ku¨hner, R., and Gerth, B. 1898–1904. Ausfu¨rhliche Grammatik der griechischen Sprache. Zweiter Teil: Satzlehre (2 vols.) (3rd edn.) (Hannover). Lada-Richards, I. 1998. ‘Staging the ephebeia: theatrical role-playing and spiritual transition in Sophocles’ Philoctetes’. Ramus 27: 1–26. —— 1999. Initiating Dionysus: Ritual and Theatre in Aristophanes’ Frogs (Oxford). —— 2009. ‘ “The players will tell all”: the dramatist, the actors and the art of acting in Sophocles’ Philoctetes’, in S. D. Goldhill and E. M. Hall (eds.), Sophocles and the Greek Tragic Tradition (Cambridge) 48–68. Last, H. M. 1932. ‘Gaius Gracchus’, in S. A. Cook, F. E. Adcock, and M. P. Charlesworth (eds.), The Cambridge Ancient History, ix. The Roman Republic, 133–44 B.C. (Cambridge) 40–101. Lattimore, R. 1953. Aeschylus I: Oresteia (Chicago). Lebeck, A. 1971. The Oresteia: A Study in Language and Structure (Washington). Lefe`vre, E. 1993. ‘Die Unfa¨higkeit, sich zu erkennen: Sophokles’ Elektra’, WJA 19: 19–46. Lefkowitz, M. R. 1981. The Lives of the Greek Poets (London). Lenardon, R. J. 1978. The Saga of Themistocles (London). Lesky, A. [1972] 1983. Greek Tragic Poetry (trans. M. Dillon) (New Haven). Lewis, D. M. 1973. Review of Reinmuth (1971). CR 23: 254–6. Linforth, I. M. 1963. ‘Electra’s day in the tragedy of Sophocles’. UCPCP 19: 89–125. Lipsius, J. 1905–15. Die attische Recht und Rechtsverfahren (Leipzig). Lloyd, M. 1992. The agon in Euripides (Oxford). —— 1994. Euripides: Andromache (Warminster). Lloyd-Jones, H. 1957. Appendix, in Weir Smyth ([1925] 1957) 523–603. —— 1962. ‘The guilt of Agamemnon’. CQ 12: 187–99. —— 1966. ‘Problems of early Greek tragedy: Pratinas, Phrynichus, the Gyges Fragment’, in Estudios sobre la tragedia griega (Cuadernos de la Fundacio´n Pastor 13) 11–33 (abridged in Lloyd-Jones 1990: 225–37). —— 1972. ‘Tycho von Wilamowitz-Moellendorff on the dramatic technique of Sophocles’. CQ 22: 214–28. —— 1979. Aeschylus, Oresteia: Agamemnon (London). —— 1990. Greek Epic, Lyric, and Tragedy: The Academic Papers of Sir Hugh Lloyd-Jones (Oxford). —— 1994. Sophocles, i. Ajax, Electra, Oedipus Tyrannus (Cambridge, Mass.). —— 1996. Sophocles, iii. Fragments (Cambridge, Mass.).
References
297
—— and Wilson, N. G. 1990a. Sophoclis Fabulae (Oxford). —— —— 1990b. Sophoclea: Studies on the Text of Sophocles (Oxford). Lobel, E., Wegener E. P., and Roberts, C. H. 1952. The Oxyrhynchus Papyri, xx (London). Luppe W. 1966. ‘Die Hypothesis zu Kratinos’ Dionysalexandros’. Philologus 110: 169–93. —— 2000. ‘Nuove acquisizioni da papiri di Euripides’. SemRom 3: 267–79. MacDowell, D. M. 1971. Aristophanes: Wasps (Oxford). —— 1978. The Law in Classical Athens (London). —— 1994. ‘The number of speaking actors in Old Comedy’. CQ 44: 325–35. Machin, A. 1981. Cohe´rence et continuite´ dans le the´aˆtre de Sophocle (HauteVille, Que´bec). Mackinnon, J. K. 1978. ‘The reason for the Danaids’ flight’. CQ 28: 74–82. McLeish, K. 1979. Sophocles: Electra, Antigone, Philoctetes (Cambridge). Macleod, C. W. 1982. ‘Politics and the Oresteia’. JHS 102: 124–44. Maiullari, F. 1999. L’interpretazione anamorfica dell’Edipo Re (Pisa/Rome). Malcovati, E. 1975. ‘Tre frammenti di oratori romani’. Athenaeum 53: 364–7. March, J. R. 1987. The Creative Poet (¼ BICS Suppl. 49) (London). —— 2000. Sophocles: Electra (Warminster). Marshall, C. W. 2003. ‘Sophocles’ Nauplius and Heron of Alexandria’s mechanical theatre’, in A. H. Sommerstein (ed.), Shards from Kolonos (Bari) 261–79. Mastromarco, G. 1992. ‘La commedia’, in G. Cambiano et al. (eds.), Lo spazio letterario della Grecia antica i.1 (Rome) 335–77. —— 1998. ‘La commedia attica antica fra tradizione e innovazione’, in J. A. Lo´pez Fe´rez (ed.), La comedia griega: aspectos literarios, sociales y educativos (Madrid) 23–42. Mastronarde, D. J. 1979. Contact and Discontinuity: Some Conventions of Speech and Action on the Greek Stage (Berkeley). —— 1994. Euripides: Phoenissae (Cambridge). Mette, H. J. 1959. Die Fragmente der Trago¨dien des Aischylos (Berlin). Michelakis, P. 2002. Achilles in Greek Tragedy (Cambridge). Miles, S. N. 2009. ‘Strattis, Comedy, and Tragedy’ (Diss. Nottingham). Mitchell-Boyask, R. 2008. Review of P. J. Wilson (2007). Moorhouse, A. C. 1982. The Syntax of Sophocles (Leiden). Moreau, A. 1995. ‘La Niobe´ d’Eschyle: quelques jalons’. REG 108: 288–307. Morwood, J. 2007. Euripides: Suppliant Women (Oxford). Mossman, J. M. 1995. Wild Justice: A Study of Euripides’ Hecuba (Oxford). Mu¨ller, G. 1967. Sophokles: Antigone (Heidelberg).
298
References
¨ dipus (Mainz/ Mu¨ller, C. W. 1984. Zur Datierung des sophokleischen O Wiesbaden). Murray, G. G. A. 1952. The Complete Plays of Aeschylus (London). —— 1955. Aeschyli septem quae supersunt tragoediae (Oxford). Nauck, A. 1889. Tragicorum Graecorum fragmenta (Leipzig). Neils, J. 1994. ‘Priamos’. LIMC vii.1: 507–22. Neitzel, H. 1980. ‘—Ł Ø Ł, Leitwort der aischyleischen Trago¨die?’ Gymnasium 87: 283–93. Nesselrath, H. G. 1990. Die attische Mittlere Komo¨die (Berlin). Ober, J. 1989. Mass and Elite in Democratic Athens (Princeton). —— and Strauss, B. S. 1990. ‘Drama, political rhetoric, and the discourse of Athenian democracy’, in Winkler and Zeitlin (1990) 237–70. Olson S. D. 1991. ‘Dicaeopolis’ motivations in Aristophanes’ Acharnians.’ JHS 111: 200–3. —— 1998. Aristophanes: Peace (Oxford). —— 2002. Aristophanes: Acharnians (Oxford). —— 2007. Broken Laughter: Select Fragments of Greek Comedy (Oxford). Omitowoju, R. 2002. Rape and the Politics of Consent in Classical Athens (Cambridge). Owen, A. S. 1927. ‘" ’ Z Æ ŒÆd ºº Æ: the end of Sophocles’ Electra’. CR 41: 50–2. Paduano, G. 1982. Tragedie e frammenti di Sofocle (Turin). Page, D. L. 1959. History and the Homeric Iliad (Berkeley). —— 1972. Aeschyli septem quae supersunt tragoedia[e] (Oxford). —— 1981. Further Greek Epigrams (Cambridge). Parker, R. C. T. 1983. Miasma: Pollution and Purification in Early Greek Religion (Oxford). —— 2005. Polytheism and Society at Athens (Oxford). Pattoni, M. P. 1987 L’autenticita` del Prometeo incatenato di Eschilo (Pisa). Paulsen, T. 1989. Die Rolle des Chors in den spa¨ten Sophokles-Trago¨dien (Bari). Pearson, A. C. 1917. The Fragments of Sophocles (3 vols.) (Cambridge). Pelling, C. B. R. ed. 1997a. Greek Tragedy and the Historian (Oxford). —— 1997b. ‘Aeschylus’ Persae and history’, in Pelling (1997a:1–19). Petrounias, E. 1976. Funktion und Thematik der Bilder bei Aischylos (Go¨ttingen). Pickard-Cambridge, A. W. 1988. The Dramatic Festivals of Athens (3rd edn., rev. J. P. A. Gould and D. M. Lewis) (Oxford). Podlecki, A. J. 1975. The Life of Themistocles (Montreal). —— 1989. Aeschylus: Eumenides (Warminster).
References
299
—— 1993. ‘˚Æ’ IæåÅÐ ªaæ çغÆØ º : the concept of leadership in Aeschylus’, in A. H. Sommerstein et al. (eds.) Tragedy, Comedy and the Polis (Bari) 55–79. Pope, M. W. M. 1974. ‘Merciful heavens? A question in Aeschylus’ Agamemnon’. JHS 94: 100–13. Prag, A. J. N. W. 1985. The Oresteia: Iconographic and Narrative Traditions (Warminster). —— 1991. ‘Clytemnestra’s weapon yet once more’. CQ 41: 242–6. Radt, S. L. 1985. Tragicorum Graecorum Fragmenta, iii. Aeschylus (Go¨ttingen). —— 1999. Tragicorum Graecorum Fragmenta, iv. Sophocles (2nd edn.) (Go¨ttingen). Raeburn, D. A. 2008. Sophocles: Electra and Other Plays (London). Raphael, F., and McLeish, K. 1979. The Serpent Son: Aeschylus, Oresteia (Cambridge). Reckford K. J. 1987. Aristophanes’ Old-and-New Comedy, i. Six Essays in Perspective (Chapel Hill, NC). Redondo, J. 2003. ‘Satyric diction in the extant Sophoclean fragments: a reconsideration’, in A. H. Sommerstein (ed.), Shards from Kolonos (Bari) 413–31. Rehm, R. 1988. ‘The staging of suppliant plays’. GRBS 29: 263–307. —— 1992. Greek Tragic Theatre (London). Reinhardt, K. 1934. ‘Zur Niobe des Aischylos’. Hermes 69: 233–61. Reinmuth, O. W. 1967. ‘Ephebia’. Der Kleine Pauly ii (Stuttgart) 287–91. —— 1971. The Ephebic Inscriptions of the Fourth Century B.C. (Leiden). Rhodes, P. J. 1970. ‘Thucydides on Pausanias and Themistocles’. Historia 19: 387–400. —— 1981. A Commentary on the Aristotelian Athenaion Politeia (Oxford). —— and Osborne, R. G. 2003. Greek Historical Inscriptions 404–323 BC (Oxford). Robert, C. 1920–6. Die griechische Heldensage (Berlin). Ronnet, G. 1969. Sophocle poe`te tragique (Paris). Roselli, D. K. 2009. ‘Theorika in fifth-century Athens’. GRBS 49: 5–30. Ro¨sler, W. 1993. ‘Der Schluß der ‘Hiketiden’ und die Danaiden-Trilogie des Aischylos’. RhM 136: 1–22. Ruffell, I. A. 2000. ‘The world turned upside down: utopia and utopianism in the fragments of Old Comedy’, in F. D. Harvey and J. Wilkins (eds.), The Rivals of Aristophanes (London/Swansea) 473–506. Ruschenbusch, E. 1979. ‘Die Einfu¨hrung des Theorikon’. ZPE 36: 303–8. Saı¨d, S. 1993. ‘Couples fraternels chez Sophocle’, in A. Machin and L. Perne´e (eds.), Sophocle: le texte, les personnages (Aix-en-Provence) 299–327.
300
References
Salviat, F. 1989. ‘La deuxie`me repre´sentation des Grenouilles: la faute d’Adeimantos, Cle´ophon et le deuil de l’hirondelle’, in R. E´tienne, M. T. Le Dinahet, and M. Yon (eds.), Architecture et poe´sie dans le monde grec: hommages a` Georges Roux (Lyon) 171–83. Schaps, D. M. 1979. Economic Rights of Women in Ancient Greece (Edinburgh). Schmid, W., and Sta¨hlin, O. 1934. Geschichte der griechischen Literatur, pt. 1, vol. ii. Die griechische Literatur in der Zeit der attischen Hegemonie vor dem Eingreifen der Sophistik (Munich). Scodel, R. S. 1980. The Trojan Trilogy of Euripides (Go¨ttingen). Scott, W. C. 1984. Musical Design in Aeschylean Theater (Hanover, NH). Scullion, S. 1994. Three Studies in Athenian Dramaturgy (Stuttgart). —— 2002. ‘Tragic dates’. CQ 52: 81–101. Seaford, R. A. S. 1981. ‘Dionysiac drama and Dionysiac mysteries’. CQ 31: 252–75. —— 1984a. Euripides: Cyclops (Oxford). —— 1984b. ‘L’ultima canzone corale delle Supplici di Eschilo’. Dioniso 55: 221–9. —— 1987. ‘The tragic wedding’. JHS 107: 106–30. —— 1996. Euripides: Bacchae (Warminster). Seale, D. 1982. Vision and Stagecraft in Sophocles (London). Segal, C. P. 1966. ‘The Electra of Sophocles’. TAPA 97: 473–545. —— 1981. Tragedy and Civilization: An Interpretation of Sophocles (Cambridge, Mass.). Sheppard, J. T. 1927a. ‘Electra: a defence of Sophocles’. CR 41: 2–9. —— 1927b. ‘Electra again’. CR 41: 163–5. Sicherl, M. 1986. ‘Die Tragik der Danaiden’. MH 43: 81–110. Siewert, P. 1977. ‘The ephebic oath in fifth-century Athens’. JHS 97: 102–11. Slater, N. W. 1996. ‘Bringing up Father: paideia and ephebeia in the Wasps’. In A. H. Sommerstein and C. Atherton (eds.), Education in Greek Fiction (Bari) 27–52. Snell, B. 1953. Review of Lobel, Wegener, and Roberts (1952). Gnomon 25: 433–40. —— 1966. ‘Zu den Urkunden dramatischer Auffu¨hrungen’. NAWG [1966]: 11–37. —— 1971. Tragicorum Graecorum fragmenta, i. Didascaliae tragicae, catalogi tragicorum et tragoediarum, testimonia et fragmenta tragicorum minorum (Go¨ttingen). Sommerstein, A. H. 1972. Review of Vellacott (1971). Farrago 8: 20–9. —— 1977a. ‘Notes on Aristophanes’ Wasps’. CQ 27: 261–77.
References
301
—— 1977b. ‘Notes on Aeschylus’ Suppliants’. BICS 24: 67–82. —— 1977c. ‘Aristophanes and the events of 411’. JHS 97: 112–26. —— 1980a. ‘The naming of women in Greek and Roman comedy’. QSt 11: 393–418. Reprinted with updates in Sommerstein (2009: 43–69). —— 1980b. ‘Notes on the Oresteia’. BICS 27: 63–75. —— 1984. ‘Act division in Old Comedy’. BICS 31: 139–52. —— 1989a. ‘Notes on Aeschylus’ Seven against Thebes’. Hermes 117: 432–45. —— 1989b. Aeschylus: Eumenides (Cambridge). —— 1989c. ‘Again Klytaimestra’s weapon’. CQ 39: 296–301. —— 1993. ‘Kleophon and the restaging of Frogs’, in A. H. Sommerstein et al. (eds.), Tragedy, Comedy and the Polis (Bari) 461–76. Reprinted with updates in Sommerstein (2009: 254–71). —— 1996a. Aeschylean Tragedy (Bari). —— 1996b. ‘How to avoid being a komodoumenos’. CQ 46: 327–56. —— 1996c. The Comedies of Aristophanes, ix. Frogs (Warminster). —— 1996d. ‘Response to Slater’, in A. H. Sommerstein and C. Atherton (eds.), Education in Greek Fiction (Bari) 53–64. Reprinted with updates in Sommerstein (2009: 192–203). —— 1998. ‘The theatre audience and the Demos’, in J. A. Lo´pez Fe´rez (ed.), La comedia griega y su influencia en la literatura espan˜ola (Madrid) 43–62. —— 2000a. ‘Platon, Eupolis and the “demagogue-comedy” ’, in F. D. Harvey and J. Wilkins (eds.), The Rivals of Aristophanes: Studies in Athenian Old Comedy (London / Swansea) 437–51. —— 2000b. ‘The prologue of Aeschylus’ Palamedes’. RhM 143: 118–27. —— 2001. Aristophanes: Wealth (Warminster). —— 2002. ‘Comic elements in tragic language: the case of Aeschylus’ Oresteia’, in A. Willi (ed.), The Language of Greek Comedy (Oxford) 151–68. —— (ed.) 2003. Shards from Kolonos: Studies in Sophoclean Fragments (Bari). —— 2004a. ‘Comedy and the unspeakable’, in D. L. Cairns and R. A. Knox (eds.), Law, Rhetoric and Comedy in Classical Athens: Essays in Honour of Douglas M. MacDowell (Swansea) 205–22. (Previously published in German in A. Ercolani (ed.), Spoudaiogeloion: Form und Funktion der Verspottung in der aristophanischen Komo¨die (Stuttgart, 2002) 225–54.) —— 2004b. ‘Harassing the satirist: the alleged attempts to prosecute Aristophanes’, in I. Sluiter and R. M. Rosen (eds.), Free Speech in Classical Antiquity (Leiden) 145–74. —— 2005. ‘Tragedy and myth’, in R. Bushnell (ed.), A Companion to Tragedy (Oxford) 163–80.
302
References
Sommerstein, A. H. 2006. ‘Rape and consent in Athenian tragedy’, in D. L. Cairns and V. J. Liapis (eds.), Dionysalexandros: Essays . . . in Honour of Alexander F. Garvie (Swansea) 233–51. —— 2008. Aeschylus (3 vols.) (Cambridge, Mass.). —— 2009. Talking about Laughter and Other Studies in Greek Comedy (Oxford). —— forthcoming a. ‘The Persian War tetralogy of Aeschylus’, in J. Davidson and D. Rosenbloom (eds.), Greek Drama IV (London). —— forthcoming b. ‘Orestes’ trial and Athenian homicide procedure’, in P. J. Rhodes, E. M. Harris, and D. Lea˜o (eds.), Law and Drama in Ancient Athens (London / Coimbra). —— forthcoming c. ‘An overlooked tragic fragment? PMG 960’. In F. Corte´s Gabaudan and J. Me´ndez Dosuna (eds.) In Memoriam A. Lo´pez Eire (Salamanca). —— forthcoming d. ‘Fragments and lost tragedies’, in A. Markantonatos (ed.), Brill’s Companion to Sophocles (Leiden). —— forthcoming e. ‘Problem kids: young males and society from Electra to Bacchae’. In F. Montanari and A. Rengakos (eds.), Crisis on Stage: Tragedy and Comedy in Late Fifth-Century Athens (Berlin). —— Clark, A. C., and Talboy, T. H. forthcoming. Sophocles: Selected Fragmentary Plays II (Oxford). —— Fitzpatrick, D. G., and Talboy, T. H. 2006. Sophocles: Selected Fragmentary Plays I (Oxford). Sourvinou-Inwood, C. 1988. Studies in Girls’ Transitions: Aspects of the Arkteia and Age Representation in Attic Iconography (Athens). Stanford, W. B. 1959. The Odyssey of Homer (2 vols.) (London). Stevens, P. T. 1971. Euripides: Andromache (Oxford). Stinton, T. C. W. 1977. ‘Interlinear hiatus in trimeters’. CQ 27: 67–72. Reprinted in Stinton (1990: 362–8). —— 1986. ‘The scope and limits of allusion in Greek tragedy’, in M. J. Cropp, E. Fantham, and S. E. Scully (eds.), Greek Tragedy and its Legacy: Essays Presented to D.J. Conacher (Calgary) 67–102. Reprinted in Stinton (1990: 454–92). —— 1987. ‘The apotheosis of Heracles from the pyre’, in L. Rodley (ed.), Papers given at a Colloquium on Greek Drama in Honour of R. P. Winnington-Ingram (London) 1–16. Reprinted in Stinton (1990: 493–507). —— 1990. Collected Papers on Greek Tragedy (Oxford). Stockton D. L. 1982. ‘The death of Ephialtes’. CQ 32: 227–8. Sutton, D. F. 1980. Self and Society in Aristophanes (Washington).
References
303
Szleza´k, T. A. 1981. ‘Sophokles’ Elektra und das Problem des ironischen Dramas’. MH 38: 1–21. Taplin O. P. 1975. ‘The title of Prometheus Desmotes’. JHS 95: 184–6. —— 1977. The Stagecraft of Aeschylus: The Dramatic Use of Exits and Entrances in Greek Tragedy (Oxford). —— 1993. Comic Angels and Other Approaches to Greek Drama through Vase-Painting (Oxford). —— 2007. Pots and Plays: Interactions between Greek Tragedy and Greek Vase-Painting of the Fourth Century B.C. (Los Angeles). Thiercy, P. 1986. Aristophane: fiction et dramaturgie (Paris). Thomson, G. 1938. The Oresteia of Aeschylus (Cambridge). —— 1966. The Oresteia of Aeschylus (2 vols.) (Amsterdam / Prague). Thompson, W. E. 1971. ‘Attic kinship terminology’. JHS 91: 110–13. Thu¨r, G. 1977. Beweisfu¨hrung vur den Schwurgerichtsho¨fen Athens (Vienna). Todd, S. C. 1993. The Shape of Athenian Law (Oxford). Touchefeu, O. 1984. ‘Astyanax I’. LIMC ii.1: 929–37. Touchefeu-Meynier, O. 1994. ‘Polyxene’. LIMC vii.1: 431–5. Tyrrell, W. B. 1984. Amazons: A Study in Athenian Mythmaking (Baltimore). Usher, S., and Edwards, M. 1985. Greek Orators, i. Antiphon, Lysias (Warminster). Vater, F. 1851. ‘Miscellaneorum criticorum fasciculus tertius’. Archiv fu¨r Philologie und Pa¨dagogik 17: 165–209. Vellacott, P. H. 1959. Aeschylus: The Oresteian Trilogy (2nd edn.) (Harmondsworth). —— 1971. Sophocles and Oedipus: A Study of Oedipus Tyrannus with a New Translation (London). Verrall, A. W. 1887. The Seven against Thebes of Aeschylus (London). Vickers, B. 1973. Towards Greek Tragedy (London). Vidal-Naquet, P. 1969. ‘Chasse et sacrifice dans l’Orestie d’Eschyle’. PdP 129: 101–25. Translated by J. Lloyd in J. P. Vernant and P. Vidal-Naquet, Myth and Tragedy in Ancient Greece (New York, 1990) 141–59. —— 1986. The Black Hunter: Forms of Thought and Forms of Society in the Greek World (trans. A. Szegedy-Maszak) (Baltimore). Vitelli, G., and Norsa, M. 1933. ‘Frammenti eschilei in papyri della Societa` Italiana’. BSAA 28: 107–21. Waldock, A. J. A. 1951. Sophocles the Dramatist (Cambridge). Wallace, R. W. 1989. The Areopagus Council to 307 BC (Baltimore). Watling, E. F. 1953. Sophocles: Electra and Other Plays (Harmondsworth). Weir Smyth, H. [1925] 1957. Aeschylus II (2nd edn.) (London / Cambridge, Mass.).
304
References
West, M. L. 1979a. ‘The Prometheus trilogy’. JHS 99: 130–48. —— 1979b. ‘Tragica III’. BICS 26: 104–17. —— 1990a. Aeschyli tragoediae cum incerti poetae Prometheo (Stuttgart). —— 1990b. Studies in Aeschylus (Stuttgart). —— 2000. ‘Iliad and Aethiopis on the stage: Aeschylus and son’. CQ 50: 338–52. Whallon, D. 1980. Problem and Spectacle: Studies in the Oresteia (Heidelberg). Whitlock Blundell, M. 1989. Helping Friends and Harming Enemies: A Study in Sophocles and Greek Ethics (Cambridge). Wilamowitz-Moellendorff, U. von 1889. Euripides: Herakles, i (Berlin). —— 1893. Aristoteles und Athen (Berlin). —— 1929. ‘Lesefru¨chte’. Hermes 64: 458–90. —— 1962. Kleine Schriften, iv. Lesefru¨chte und Verwandtes (Berlin). Wilkins, J. 1993. Euripides: Heraclidae (Oxford). Willans, G., and Searle, R. 1954. How to be Topp: a guide to Sukcess for tiny pupils, containing all there is to kno about SPACE (London). Wilson, A. M. 1973. ‘Not Callias, but Ecphantides? an alternative interpretation of I.G. xiv. 1097’. CR 23: 126–7. Wilson, P. J. 1997. ‘Leading the tragic khoros: tragic prestige in the democratic city’, in Pelling 1997: 81–108. —— (ed.) 2007. The Greek Theatre and Festivals: Documentary Studies (Oxford). —— 2008. ‘Costing the Dionysia’, in M. Revermann and P. J. Wilson (eds.), Performance, Iconography, Reception: Studies in Honour of Oliver Taplin (Oxford) 88–127. Winkler J. J. 1990. ‘The ephebes’ song: trago¯idia and polis’, in Winkler and Zeitlin (1990) 20–62. —— and Zeitlin, F. I. (eds.), 1990. Nothing to Do with Dionysus? Athenian Drama in its Social Context (Princeton). Winnington-Ingram, R. P. 1961. ‘The Danaid trilogy of Aeschylus’. JHS 81: 141–52. Updated version in Winnington-Ingram (1983: 55–72). —— 1980. Sophocles: An Interpretation (Cambridge). —— 1982. ‘Sophocles and women’. In J. de Romilly (ed.), Entretiens sur l’antiquite´ classique XXIX: Sophocle (Vanduvres/Geneva) 233–57 (including discussion). —— 1983. Studies in Aeschylus (Cambridge). Wolff, E. A. 1958/9. ‘The date of Aeschylus’ Danaid tetralogy’. Eranos 56: 119–39 and 57: 6–34. Woodard, T. 1964. ‘Electra by Sophocles: the dialectical design (Part I)’. HSCP 68: 163–205.
References
305
Wright, M. 2005. Euripides’ Escape-Tragedies: A Study of Helen, Andromeda and Iphigenia among the Taurians (Oxford). Zeitlin, F. I. 1978. ‘The dynamics of misogyny: myth and mythmaking in the Oresteia’. Arethusa 11: 149–84. Reprinted in Zeitlin (1996: 87–119). —— 1986. ‘Configurations of rape in Greek myth’, in S. Tomaselli and R. Porter (eds.), Rape (Oxford) 122–51. —— 1996. Playing the Other (Chicago). Zielin´ski, T. 1925. Tragodumenon libri tres (Cracow). Zimmermann, B. 1983. ‘Utopisches und Utopie in den Komo¨dien des Aristophanes’. WJA 9: 57–77.
This page intentionally left blank
Index Locorum In this index, works and passages are entered under the name of the author to whom they are traditionally ascribed, without regard to questions of authenticity. Unless otherwise stated, all tragic fragments are cited from TrGF, and all comic fragments from Kassel & Austin. Acusilaus fr. 40 Fowler: 266 n. 23 Aelian Varia Historia 2.8: 7 n. 17, 252 n. 11 Aeschines 1.49: 50 n. 13 1.173: 124 n. 27 2.167: 50 n. 13 2.172: 136 n. 76 3.86: 124 n. 27 3.125: 124 n. 27 3.223–4: 159 n. 67 Aeschylus Agamemnon 3: 173 12–17: 143 n. 2 14: 144 n. 7 18: 173 25–9: 143 27: 173 35: 173 37: 173 39/40: 143 40–67: 165 43: 166 n. 2 59: 175–6 n. 15 83–103: 173 104–59: 171–5 104: 173–4 108: 174 n. 10 109–10: 174 n. 12 111: 174 114–15: 166 n. 2, 174 n. 12 116: 172–4 120: 174 n. 11
125: 174 n. 12 126–30: 171 126: 174, 177 131–3: 174 134–52: 171 134–43: 165 134: 173 136: 177 n. 18 148–50: 174 n. 11 154–7: 175 154–5: 171 157: 174 160–83: 164, 178 n. 3 164–7: 164 168–73: 179 171–2: 175 n. 14 174: 175 n. 14 176–83: 166–7, 175 n. 14 176–7: 180 177: 187 n. 33 177–8: 178–88 179–80: 144 179: 143 n. 3 184ff: 175 188: 175 190–1: 175 202–4: 167 206–11: 167 208: 176 n. 16 212–13: 167 231–8: 268 n. 33 237: 176 n. 16 243–7: 176 n. 16 275: 143 n. 3 290: 143 n. 3, 144 326–9: 148 337: 143 n. 3 338–44: 149 n. 25
308 Agamemnon (cont.) 346: 143 n. 3 355–60: 147 355: 175 n. 14 404–26: 77, 175 n. 13 407–8: 144 420–6: 143 n. 3 424–5: 144 445–51: 148 456–8: 148 459–60: 148 n. 22 461–72: 148 489–502: 71 n. 36 491: 143 n. 3 527: 150 n. 27 559ff: 144 559–62: 143 n. 3 565–6: 143 n. 3, 150 n. 28 597: 143 n. 3 607: 175 n. 13 636: 149 649: 149 650–70: 149 651: 149 653: 149 658–60: 149 661–70: 149 666: 149 n. 25 667–8: 149 n. 25 737–43: 77 799–804: 167 809: 148 n. 23 840: 89 n. 1 883–5: 132 n. 62 887–94: 143 n. 4, 145 966–7: 235 n. 29 980–3: 143 n. 4, 145 1090: 168 1107–28: 233 n. 24 1110–11: 233 n. 24 1218: 143 n. 4 1247: 143 n. 4 1279–84: 169 1291: 175 n. 13 1323: 150 1357: 143 n. 4 1382–3: 233 n. 24 1412–20: 80 1435–6: 235 n. 30 1448–51: 143 n. 4, 145
Index Locorum 1485–8: 169 1523–9: 80 1555–9: 80 1577: 150 1592: 150 n. 29 1617–18: 132 n. 62 1646–8: 169 1647: 89 n. 1 1650–1: 38 1652: 38 1654–6: 38 1654: 204 n. 12 1657ff: 38 1667: 169 1673: 257 n. 35 Choephoroi 1: 151 10–21: 142 30–1: 37 n. 24 32–41: 143 n. 4, 144 34: 144 110: 189 114–21: 232 115: 190 117–21: 190 124: 151 130–9: 190 138–44: 232 140–1: 190, 238 n. 41 142–4: 190 160–3: 190, 232 235: 204 n. 11 238–43: 190 246–63: 190 269–305: 169 n. 5 269–70: 191 270–4: 191 285: 143 n. 4, 145 n. 11, 163 288: 143 n. 4, 145 298–305: 231 n. 18 302–4: 107 n. 66 306–509: 192 315–19: 143 n. 4, 145 382–5: 192 n. 3, 194 395: 194 399: 194 405–7: 194 409: 194 415: 145
Index Locorum 423–8: 37 n. 24 479–96: 145 n. 12 495: 143 n. 4 496: 204 n. 10 523–52: 143 n. 4, 144 527–35: 235 535: 144 543–50: 235 613–22: 143 n. 4, 150 621: 150 n. 28 660–2: 151 664–7: 152 668: 152 670–1: 143 n. 4 672–3: 152 691–9: 239 n. 42 700–6: 239 n. 42 710–11: 151 716–18: 152 727–8: 151 734–7: 152 736: 152 737–40: 152 n. 32 737–8: 239 n. 42 738–9: 239 n. 42 749–62: 71 n. 38 751: 143 n. 4, 145 771: 152 848–54: 152 850: 152 875ff: 139 n. 87 877ff: 153 877–9: 151 881–2: 151 882: 143 n. 4, 144 893: 204 n. 12 897: 143 n. 4 899: 193 900–2: 193 924: 192 961: 153 963–4: 153 973: 107 n. 66 974–5: 153 978–9: 225 n. 4 980–4: 233 n. 24 984–5: 153 n. 34 986: 153 997–1004: 233 n. 24 1005–6: 83
1021–5: 193 1026–34: 193 1035: 193 1040: 153 1046–7: 107 n. 66 1048: 194 1051: 203 n. 9 1075–6: 143 n. 4, 145 Eumenides 33/34: 108 46–142: 143 n. 4 64–6: 192 81: 105 95: 160 n. 69 110: 160 n. 69 116: 144 150: 160 n. 69 174–8: 154 n. 39 231/244: 111 n. 80 236: 89 n. 1 264–75: 154 n. 39 280: 143 n. 4, 145 289–95: 133 n. 63 322–3: 154 n. 38 339–40: 154 n. 39 389: 187 n. 33 397–402: 133 n. 63 416: 154 n. 38 417: 145 450: 234 n. 25 494–5: 242 n. 50 566ff: 143 n. 5, 153 n. 36 614–18: 232 667–73: 133 n. 63 690–2: 146 n. 18, 159–60 704–6: 160 705–6: 143 n. 4, 146 n. 15 707–10: 146 n. 16 745: 154 n. 38 762–77: 133 n. 63 780: 160 n. 69 793: 154 n. 38 844: 154 n. 38 858–66: 161 864(-5): 133 913–15: 133 n. 63 918–21: 133 n. 63: 133 n. 63 918: 133 976–87: 161
309
310 Eumenides (cont.) 980–3: 161 1000: 187, 188 1001–2: 187 1008–9: 133 n. 63 1011: 124 n. 25 1018: 124 n. 25 1033: 154 n. 38 1043: 143 n. 5 1047: 143 n. 5 Persians Hypothesis: 12 n. 6, 13 n. 7 68–72: 126 n. 40 93–100: 164 109–13: 126 n. 40 111: 198 130–1: 126 n. 40 215–25: 71 n. 36 353ff: 126 623–80: 111 n. 80 722–6: 126 n. 40 736: 126 n. 40 744–51: 126 n. 40 851: 204 n. 13 1054: 37 n. 24 Prometheus Bound 7–8: 43 26–7: 43 55–62: 41 58: 41 n. 32 59: 41 n. 33 64–9: 43 64–5: 41 74–5: 41 n. 32 71–81: 41 406–24: 183 410: 183 442–57: 183 443: 183 444: 183 447–9: 183 456–7: 183 613: 41 n. 33 673–82: 199 n. 17 773–4: 113 n. 83 865–73: 113 n. 83 865: 114 n. 88
Index Locorum Seven against Thebes Hypothesis: 7 n. 18, 15 n. 14 35/39: 108 109–10: 84 171–2: 84 188: 83 195: 83 454–5: 84 571–5: 83 576ff: 87 584: 85 644–8: 85 662–71: 85 664–9: 86 666: 84, 86, 87 672–5: 86 677: 204 686: 84 690–1: 83 727–33: 84 757: 89 n. 1 781–2: 87 n. 15 785–6: 220 n. 28 805–19: 85 n. 11 813: 83 816–19: 84 822–31: 64 826–39: 95 n. 11 828: 83, 88 845–6: 85 n. 11 848ff: 84 848: 86 849–50: 85 n. 11 849: 87 n. 15 861–74: 101 n. 46 875–960: 85 n. 11 881–2: 83 887: 84 888–90: 84 906–14: 84 954–60: 83 961–92: 85 n. 12 979: 85 n. 13 981: 85 n. 13 991: 85 n. 13 996–7: 101 n. 46 998–9: 85 n. 12, 88 1000–4: 85 n. 11
Index Locorum 1004: 220 n. 27, 223 1005–78: 5 n. 13, 223 Suppliants 1–233: 96 1: 99 n. 40 8–10: 91 n. 7 12: 114 15–18: 99 n. 40 21: 110 26–39: 99 n. 40 37–9: 91 n. 7 40–7: 99 n. 40 50–1: 199 n. 17 79–82: 93 n. 18 80–1: 91 n. 7 83–4: 102 n. 49 87: 102 n. 48 92–5: 164 93–5: 102 n. 48 140: 89 n. 1 141–3: 91 n. 7 151–3: 91 n. 7 154–61: 100 168–75: 99 n. 40 192–3: 110 n. 74 202: 117 206–11: 100 206: 99 n. 40, 100 209: 99 n. 40, 100 210: 89 n. 1, 100 246–8: 107 274–326: 134 332–7: 91 n. 7 335–7: 93 n. 19 337: 93 n. 18 338: 94 347: 134 359–64: 134 359–60: 99 n. 40 365–9: 110 n. 75, 134 368–9: 134 370–5: 134 385–6: 99 n. 40, 134 387ff: 97 387–91: 135 389–401: 110 n. 75 398–401: 134 392–5: 91 n. 7
392–3: 93 n. 19 402–4: 99 n. 40 413–16: 134 437: 99 n. 40 455–65: 134 478–9: 134 481ff: 135 502–3: 135 508–9: 198 517–19: 134 517–18: 110 n. 75 524–824: 96 524–37: 99 n. 40 538–42: 199 n. 17 558: 199 571–99: 99 n. 40 600–24: 110 n. 75 604: 134 607: 134 609–14: 134 609: 124 n. 25 614: 134 615–20: 103, 135 621: 134 623: 135 624–5: 111 n. 80 625–709: 115 n. 91 643–5: 91 n. 7 652: 187 n. 33 699: 134 713: 111 n. 80 739: 110 n. 75 741–2: 93, 102 788–807: 91 n. 7 811–16: 99 n. 40 825ff: 38 840: 38 847: 38 868: 198 884: 38 892: 99 n. 40 902: 99 n. 40 909: 38 911–53: 103 916–18: 97 918: 115 n. 92 920: 115 n. 92 924: 115 n. 92 925: 38
311
312 Suppliants (cont.) 932–3: 97 932: 115 n. 92 942ff: 135 942–5: 110 n. 75 950–1: 113 n. 85 954: 113 957–61: 138 966–79: 101 n. 47 977–9: 101, 117 980ff: 92, 107 980–1017: 100 980–1013: 97 985ff: 113, 135 987–8: 91–2 991ff: 93 n. 17 991–1013: 116 994: 124 n. 25 1006–9: 116 1006–7: 102 n. 48 1009–11: 138 1015–17: 93 n. 18 1015: 100 1018–73: 113 n. 86 1018–29: 115 n. 91 1031–3: 91 n. 7, 93 n. 18 1034ff: 113 1034–51: 100–2 1034–42: 91 n. 10 1038–42: 110, 113, 115 1045–9: 101–2 1047–51: 113 1052–62: 100–1 1052–3: 91 n. 7, 93 n. 18, 99 n. 40 1056: 91 n. 10 1057–8: 102 n. 48 1059: 91 n. 10 1061: 91 n. 10 1062–9: 99 n. 40 1062–7: 93 n. 18 1062–3: 91 n. 7 fr. 43: 89 n. 1, 108, 109 fr. 44: 90 n. 2, 109, 110 fr. 75a: 257 n. 31 fr. 116: 75 fr. 131: 66 n. 15 fr. 150: 199 fr. 154a: 70–3 fr. 174: 67 n. 17
Index Locorum fr. 181: 251 n. 9 fr. 181a: 250 n. 3 fr. 182: 250 n. 3 fr. 182a: 250 n. 3 fr. 200: 35 n. 17 fr. 211: 77–8 fr. 212: 76 n. 50 fr. 451h: 73–5 fr. 451k (¼ 180a Sommerstein): 65–7, 80 fr. 451o: 75 test. 1 (Life).10 Radt: 196 n. 5 test. 1.11 Radt: 195 n. 1 test. 1.17 Radt: 195 n. 4 test. 1.33–4 Radt: 119 n. 5 test. 1.51–2 Radt: 119 n. 3 test. 58 Radt: 137 n. 80 test. 71–7 Radt: 119 n. 3 test. 164–5 Radt: 195 n. 4 Alcimus FGrH 560 F 2: 186 n. 30 Amphis fr. 23: 25 n. 37 Andocides 1.43: 159 n. 67 3.3: 136 n. 76 3.5: 162 n. 77 Anthologia Palatina 7.2: 266 n. 25 7.39: 195 n. 4 7.40: 195 n. 4 7.86–7: 200 n. 22 9.663: 198 n. 15 Anthologia Planudea 150: 268 nn. 30, 32 Antiphanes fr. 189: 4 n. 12 fr. 274: 159 n. 67 Antiphon 4.2.1: 221 n. 29 4.3.2: 221 n. 29 5.21: 157 n. 60 5.26: 157 n. 59 5.68: 156 n. 54, 158 n. 62 6.35: 146 n. 20 fr. 1 Thalheim: 157 n. 58
Index Locorum Apollodorus (hist.) FGrH 244 F 33: 156–7 n. 58 Apollodorus (mythogr.) Bibliotheca 1.8.1: 229 n. 12 2.1.4: 102 n. 50 2.1.5: 95 n. 24, 105 n. 56, 113 n. 84 2.8.5: 117 3.3.1–2: 75 n. 48 3.7.3: 83 n. 3 3.13.2: 69 n. 28 3.13.3: 69 n. 29 Epitome 3.9: 251 n. 6 3.10: 251 n. 3 6.5–11: 148 n. 24, 251 n. 7 6.13–14: 271 6.13: 68 n. 21, 68 n. 22, 69 n. 23 Appian Bellum Civile 1.20: 155 n. 44 Aristophanes Acharnians 5–8: 24 11: 20 19–20: 153 n. 37 35: 123 n. 24 280–3: 42 n. 36 300–2: 29 309–14: 130 n. 54 356: 130 n. 54 369: 130 n. 54 377–82: 29, 129 n. 49 482: 130 n. 54 502–3: 29, 129 n. 49 509–12: 130 n. 55 540: 242 n. 50 563: 269 n. 38 614–17: 131 n. 59 630–2: 129 n. 49 659–64: 29 703–12: 125 n. 30 Birds 445–6: 118 n. 1 618–19: 99 n. 38 640: 125 n. 34 716: 99 n. 38
793–6: 121 848–1057: 35 901–2: 36 n. 18 1056: 35 1337–71: 54 n. 25 1410–69: 54 n. 25 Clouds Hypoth. I Dover: 14 n. 10 521: 124 n. 28 527: 124 n. 28 529: 55 n. 26 535: 124 n. 28 538–9: 120 n. 8 575: 124 n. 28 929: 184 n. 24 1070: 184 n. 24 1432: 57 n. 32 Ecclesiazusae 14–15: 186 n. 30 84–5: 153 n. 37 176: 121 n. 12 1141–2: 118 n. 1 Frogs Hypoth. I(c) Dover: 12 n.5 53: 119 n. 4 196: 173 n. 8 418: 51 n. 17 553–4: 123 n. 20 615–73: 37 676: 124 n. 28 700: 124 n. 28 830–1478: 31 n. 4 943: 119 n. 4 1043: 15 n. 12 1052: 15 n. 12 1109–18: 124 n. 28 1079–82: 37 1114: 119 n. 4 1468: 2 n. 5 Knights 105–24: 32 n. 6 233: 124 n. 28 235: 269 n. 38 327: 125 n. 31 505–6: 124 n. 28 546–50: 118 n. 1 759: 41 n. 33 794–5: 125 n. 31
313
314 Knights (cont.) 828: 269 n. 38 836: 41 n. 33 Lysistrata Hypoth. I Henderson: 12 n. 5 187–9: 138 n. 86 381–5: 42 n. 36 391–7: 125 n. 34 517–18: 125 n. 34 589–90: 125 n. 34 853: 205 n. 17 904–53: 37 n. 21 932: 18 n. 22 1105: 24 n. 35 1114–87: 18 n. 22 1138–44: 136 n. 74 Peace 937–1022: 35 969: 42 n. 36 1019–20: 35 1022: 35 1314: 31 n. 5 1315 (1317 Olson): 31 n. 5 Thesmophoriazusae 367: 233 n. 23 376: 153 n. 37 395–7: 121 n. 10 418–27: 186 n. 30 430: 238 n. 38 547: 15 n. 12 550: 15 n. 12 556–9: 186 n. 30 626–33: 31 n. 2 719: 269 n. 38 812–13: 186 n. 30 1149: 198 n. 13 1201–10: 37 n. 21 Wasps 179–80: 36 n. 19 186: 269 n. 38 1013–14: 124 n. 28 1267–74: 131 n. 59 1284–91: 129 n. 49 1480: 184 n. 24 fr. 205: 55 n. 26 fr. 206: 55 n. 26 fr. 233: 55 n. 26 test. 1.19–21 KA: 129 n. 49
Index Locorum test. 2a.19 KA: 26 n. 41 Aristotle Athenaion Politeia 8.4: 146 n. 19 14.1–2: 107 n. 65 16.7: 181 n. 8 25.2: 136 n. 74 25.4: 156 n. 53, 160 n. 70 26.2: 162 n. 76 27.1: 147 n. 20 42.2–5: 48 n. 5 43.5: 137 n. 79 45.2: 146 n. 20 54.2: 146–7 n. 20 Historia Animalium 618b19: 198 n. 13 618b22: 198 n. 13 618b34: 198 n. 13 Metaphysics 993b15: 21 n. 31 Nicomachean Ethics 1111a8–10: 30 n. 1 Poetics 1451a15: 4 n. 12 1451a22: 4 n. 12 1451b15–19: 4 n. 12 1451b25–6: 4 n. 12 1452a27–9: 105 n. 59 1452a27–8: 108 n. 69 1452b24: 37 n. 24 1453a20–21: 4 n. 12 1453b22–5: 210–11, 224 1453b23: 4–5 n. 12 1455b29–32: 105 n. 59 1455b30: 108 n. 69 Athenaeus 6.267e ff.: 182 n. 22 6.270a: 12 n. 4, 140 7.302e: 17 n. 17 7.323b: 17 n. 17 9.386a: 17 n. 17 13.589a: 17 n. 17 14.627c: 195 14.654f: 17 n. 17 Aurelius Victor de Viris Illustribus 58.10: 155 n. 44
Index Locorum Bacchylides 3.19: 198 n. 13 5.165–75: 229 n. 12 16: 229 n. 13 17.85: 198 Callias (com.) test. 4 KA: 12 n. 3 Callimachus fr. 178.23–4: 266 n. 25 Cicero de Amicitia 41: 155 n. 42, 156 n. 51 de Fato 18: 155 n. 42 Epistulae ad Atticum 10.8.7: 155 n. 42 Epistulae ad Familiares 9.21.3: 155 n. 48 Epistulae ad Quintum Fratrem 2.3.3: 155 n. 42 pro Milone 16: 155 n. 42 de Oratore 2.70: 155 n. 48 de Re Publica 6.12: 155 n. 42 Comica Adespota fr. 607: 184 n. 24 fr. 610: 184 n. 24 fr. 751: 184 n. 24
1.63: 159 n. 67 fr. 18.3 Conomis: 122 n. 16 Demosthenes 5.1: 121 n. 12 9.1: 121 n. 12 14.3: 121 n. 12 16.2: 121 n. 12 18.133: 159 n. 67 18.137: 159 n. 67 21.53: 122 n. 16 21.71–5: 221 n. 29 23.53: 221 n. 29 24.113: 147 n. 21 27.5: 106 n. 63 36.8: 106 n. 63 45.28: 106 n. 63 45.70: 205 n. 16 Prooemia 5.1: 121 n. 12 6.1: 121 n. 12 44.1: 121 n. 12 Dictys of Crete 6.7–9: 266–7 Diodorus Siculus 4.57: 23 n. 33 4.65.1: 82 n. 2 7.13.1: 117 11.77.6: 156 n. 55, 158 n. 65 12.32.3: 136 n. 74 Diogenes Laertius 5.26: 11 n. 1
Commentaria in Aristotelem CAG xx. 146.2–3: 195 n. 1
Dion Chrysostom Oration 11.40–1: 78 n. 67 Oration 52: 65 n. 9
Craterus FGrH 342 F 11: 147 n. 20
Dositheus FGrH 54 F 1: 71 n. 39
Cratinus fr. 39: 19 n. 27 fr. 48: 19 n. 27 fr. 90: 19 n. 27 fr. 171: 41 n. 33 fr. 176: 182 n. 22 Dionysalexandros test. 1 KA: 125 n. 33
Ecphantides fr. 5: 12 n. 3
Deinarchus
Ephorus FGrH 70 F 260: 98 n. 33 Epic Cycle Aethiopis fr. 6 West: 210 n. 5 Cypria
315
316
Index Locorum
Cypria (cont.) Arg. }5 West: 250 n. 2 Arg. }6 West: 176 n. 17 Arg. }7 West: 83 n. 3 Arg. }8 West: 165 n. 1 Arg. }12 West: 253 n. 18 fr. 27 West: 251 n. 4 fr. 34 Bernabe´ ¼ 27 Davies: 261 n. 5 Little Iliad Arg. }1 West: 210 n. 5 Arg. }}2–3 West: 263 n. 14 Arg. }3 West: 261 n. 7 fr. 3 West: 210 n. 5 fr. 6 West: 266 n. 23 fr. 18 West: 261 n. 5 fr. 21 West: 261 n. 7 fr. 23 West: 261 n. 7 fr. 24 West: 261 n. 7 fr. 25 West: 261 n. 5, 272 n. 46 fr. 27 West: 261 n. 7 fr. 29 West: 261 n. 5 Nostoi Arg. }3 West: 148 n. 24, 251 n. 7 Arg. }4 West: 261 n. 8 fr. 11 West: 149 n. 24 Oidipodeia fr. 3 West: 5 n. 13 Sack of Troy Arg. }2 West: 261 n. 5 Arg. }4 West: 261 n. 5, 262 n. 10 Thebais fr. 2 West: 87 n. 15, 220 n. 25 fr. 3 West: 220 n. 25 Epic Fragments Alcmeonis fr. 7 West: 181 n. 8 Eupolis fr. 99.24: 51 n. 17 fr. 99.112: 33 n. 9 fr. 156.3: 123 n. 56 fr. 193: 125 n. 32 Euripides Alcestis 115–16: 99 n. 38
Andromache 9–10: 274 22–3: 39 n. 29 51–5: 275 79–90: 39 222: 204 n. 12 425–726: 33 n. 8 445–53: 130 n. 52 465ff: 274 545–765: 39 n. 29 545: 39 574: 204 577–8: 39 587: 39 588: 39 589: 39 594–600: 130 n. 52 715–16: 39 759–60: 39 n. 29 825–78: 73 920: 274 967: 274 977–8: 274 1094–5: 272, 274 1104–8: 275 1147–9: 275 1156–7: 270 n. 42 1161–5: 275 1239–42: 270 n. 43 1263–4: 270 n. 43 Bacchae 6–12: 277 10–11: 282 26–31: 277 30: 285 n. 2 32–8: 278 39–40: 278 45–6: 278 52: 278, 280 64–87: 281 68–70: 282 75–6: 281 135–69: 281 137–8: 282 139: 281 164–9: 282 180–3: 283 193: 273 200: 285 n. 2
Index Locorum 203: 285 n. 2 215: 278 216: 278 226–32: 278 233: 278 241: 278 250–2: 278 255–7: 278 263–5: 283 266: 285 n. 2 278–85: 284 286–97: 284 298–305: 284 328–9: 283 333–6: 282 344–5: 278 346–51: 279 356: 278 367–9: 279 376–85: 284 389–90: 284 395–9: 284 395: 285 n. 2 402–16: 284 421–2: 284 427: 285 n. 2 430: 284 434–52: 33 n. 8 441–2: 285 443–8: 279 449–50: 285 451–2: 279 471–80: 279 480: 285 n. 2 483: 280 489: 285 n. 2 504: 59 n. 37 511–14: 279 618–21: 279 629–31: 279 641: 285 n. 2 653: 279 655: 285 n. 2 656: 285 n. 2 670–1: 285 680–2: 279 682: 280 689–94: 279 690: 280
695–8: 279 698: 280 699–702: 280 704–11: 280 717–21: 280 720: 280 723–4: 279 728: 280 771–4: 285 773–4: 284 779: 280 780–5: 280 796–7: 280 803: 280 804–7: 278 804: 280–1 807–8: 281 809: 281 824: 285 n. 2 839: 285 n. 2 1150–1: 285 1150: 59 n. 37 1151: 285 n. 2 1190: 285 n. 2 1298: 204 n. 13 1327–8: 286 1330–4: 282 1337–8: 282 1341–7: 282 1341: 59 n. 37 1344: 282 1346: 282 1354–60: 282 1363–7: 286 1377–8: 282 1381–2: 286 1383–7: 286 Cyclops 551–78: 32 n. 6 683–4: 45 n. 40 Electra 9–10: 234 n. 24 162–6: 234 n. 24 178–80: 95 n. 22 229: 204 310–11: 95 n. 22 345: 204 n. 12 767: 204
317
318 Electra (cont.) 962–84: 231 n. 17 1159–60: 234 n. 24 1278–80: 79 1280–3: 79 1280–1: 77 n. 60 1303–4: 231 n. 17 1322: 204 n. 11 Hecuba 23–4: 264 n. 15, 268 n. 35 117–40: 269 n. 36 120–9: 269 n. 37 223–4: 264 n. 15 505: 204 518–82: 268 n. 29 522: 234 n. 25 523–68: 264 n. 15 558–61: 268 n. 31 559: 268 n. 32 878: 238 n. 38 990: 204 1056–9: 269 n. 38 1257–73: 269 n. 38 1274: 269 n. 38 1282–6: 33 n. 8 Helen 6–7: 77 n. 58 276: 280 n. 1 318: 46 324: 46 328: 46 408–34: 77 n. 63 515: 46 595: 204 n. 12 625: 204 n. 12 766–7: 251 n. 7 899: 204 n. 12 996: 46 1126–31: 271 n. 7 1128: 251 n. 5 1299: 204 n. 12 1627: 45 1628: 39 1629: 33 n. 8, 39 1639: 39–40 1640–1: 45 1640: 45
Index Locorum Heracleidae 61–76: 33 n. 8 67: 37 n. 25 75–6: 37 n. 25 788: 204 Heracles 251: 280 n. 1 258: 269 n. 38 490: 204 n. 12 514: 204 n. 12 531: 204 n. 12 Hippolytus (Stephanephoros) 41–8: 227–8 78–81: 53 n. 22 80: 59 n. 38 88–120: 73 176–238: 73 n. 43 304–14: 73 n. 43 421–5: 228 n. 11 433–524: 73 n. 43 689–92: 228 n. 11 717: 228 n. 11 720–1: 228 n. 11 728–31: 228 n. 11 731: 59 n. 38 949: 59 n. 38 952–4: 53 n. 21 986–9: 53 n. 22 995: 59 n. 38 1007: 59 n. 38 1100: 59 n. 38 1365: 59 n. 38 Ion 154–83: 35 n. 17 616–17: 238 n. 8 1409: 204 n. 13 Iphigeneia at Aulis 49–71: 273 n. 50 107: 264 n. 18 203–4: 264 n. 18 309–16: 40 311: 39 n. 29 440: 172 n. 4 524–35: 264 n. 18 612: 172 n. 4 652: 204 n. 10 678: 172 n. 4
Index Locorum 685: 172 n. 4 820: 172 n. 4 854: 172 n. 4 1362–6: 264 n. 18 Iphigeneia in Tauris 456–69: 33 n. 8 815: 204 n. 11 827: 204 n. 11 913–15: 205 n. 14 917–19: 205 n. 14 920–1: 205 n. 14 1065: 204–5 Medea 395–8: 269 n. 38 1018: 233 n. 22 Orestes 42: 71 n. 38 217: 204 n. 11 733: 205 n. 15 804–6: 205 n. 15 1045: 204 n. 11 1233: 205 n. 15 1567–1672: 33 n. 8 1593: 269 n. 38 1654–7: 264 Phaethon 164–7 Diggle: 25 n. 37 Phoenissae 71–2: 82 166: 204 n. 11 408–25: 83 n. 4 1072: 204 1407–22: 84 n. 8 1437: 204 n. 11 1702: 204 n. 11 Rhesus 565: 627: 661: 684: 688:
40 40 40 40 40
Suppliants 110: 71 n. 38 131–50: 83 n. 4 426–55: 130 n. 53 641: 204 793: 204 n. 13
319
860–917: 54 n. 23 980–3: 34 n. 12 1009–11: 34 n. 12 Troades 15: 198 n. 13 70: 33 n. 7 75–90: 251 n. 7 210–13: 131 n. 57 274: 264 n. 15 573–6: 264 n. 15 657–68: 264 n. 15 697–705: 264 n. 15 757: 204 n. 13 786–9: 33 n. 8 1123–46: 264 n. 15 fr. 122: 33 n. 8 fr. 127: 33 n. 8 fr. 128: 33 n. 8 fr. 464: 238 n. 38 fr. 541: 214 n. 16 fr. 578: 250 n. 3 fr. 682: 264 n. 16 fr. 683: 264 n. 16 fr. 776: 25 n. 37 fr. 955h: 99 n. 38 test. 1.18 Kovacs: 200 n. 22 test. 63 Kovacs: 200 n. 22 test. 64 Kovacs: 200 n. 22 test. 66 Kovacs: 200 n. 22 test. 89 Kovacs: 200 n. 22 test. 96–9 Kovacs: 200 n. 22 Eustathius on Odyssey 4.3: 68 n. 20, 68 n. 22, 69 n. 23, 270–1 Galen xv. 424.4–9 Ku¨hn: 14 n. 10 Gellius, Aulus 17.4.5: 119 n. 6 Gorgias fr. 24 DK: 85 n. 9 Harpocration d54: 11 n. 1 Hellanicus FGrH 4 F 98: 86–7 n. 15 Herodotus 1.46.2–3: 98 n. 34
320
Index Locorum
1.59: 107 n. 65 1.108: 235 n. 29, 235–6 n. 30 1.110–12: 176 n. 15 2.64: 37 n. 23 2.98: 95 n. 24 2.112.1: 77 n. 59 2.114.1: 77 n. 59 2.118.3–119.1: 77 n. 64 5.61: 83 n. 3 6.78–80: 198 n. 13 6.86: 249 6.112.3: 9 n. 23 7.19.1: 235 n. 29 8.131.3: 126 n. 40 9.114–21: 126 n. 40 Heron of Alexandria On Automaton-Making 22.3–6: 252 n. 13 Hesiod Theogony 550–613: 168 n. 3 Works and Days 42–105: 168 n. 3 67–8: 185 77–9: 185 90–2: 181 n. 9 94ff: 185 109–19: 181 fr. 129 M-W ¼ 77 Most: 113 n. 83 fr. 135 M-W ¼ 241 Most: 113 n. 83 fr. 192 M-W ¼ 135 Most: 74 n. 47, 220 n. 26 fr. 204.78–85 M-W ¼ 155.78–85 Most: 273 n. 50 fr. 204.124ff M-W ¼ 155.124ff Most: 181 n. 14 Hesychius Œ716: 12 n. 3 Hipponax fr. 36: 24 n. 37 Homer Iliad 1.21: 187 n. 33 2.299–330: 176–7 2.506: 198 n. 12 20.8: 198 n. 13
21.525–605: 261 n. 7 22.18–20: 271 n. 44 22.395–405: 266 n. 22 23.679–80: 220 n. 26 24.14–54: 266 n. 22 24.206–14: 266 24.314–21: 173 n. 8 24.477–9: 263 n. 13 24.486–92: 261 n. 8 24.504–6: 263 n. 13 Odyssey 1.29–47: 223 1.298–300: 223 1.325–7: 150 n. 27 3.132–5: 150 n. 27 3.304–12: 211 n. 8, 223 3.304–6: 242 n. 49 3.309–12: 78 3.309–10: 224 4.354–69: 77 n. 62 4.363–461: 77 n. 57 4.472–569: 76 n. 52 4.498–511: 258 n. 40 4.546–7: 78, 223 6.91: 198 n. 13 8.499–520: 263 11.271–80: 215 n. 19 11.277–9: 225 n.6 11.324–5: 225 n. 6 11.326–7: 211 n. 7, 225 11.328: 225 n. 6 11.492–537: 262 11.494–503: 261 n. 8 11.507: 262 11.510–12: 262 11.519–21: 266 n. 23 11.533–7: 262 11.543–64: 210 n. 4 11.568–71: 75 15.160–81: 173 n. 8 15.244–9: 211 n. 7 15.244–8: 225 20.278: 198 n. 13 Homeric Hymns To Apollo 230: 198 n. 12 To Hermes 186: 198 n. 12
Index Locorum Horace Carmina 3.11.45: 105 n. 56 Hyginus Fabulae 9: 71 n. 35, 71 n. 39 72: 74 n. 47 105: 255 n. 24 116: 148–9 n. 24, 251 n. 7 122: 248 n. 75 136: 75 n. 48 168.1: 102 n. 50 170.11: 113 n. 87 Hypereides For Euxenippus 7–8: 146 n. 20 fr. 252: 184 n. 24 Ibycus PMG 282 (a) 21: 235 n. 28 PMG 307: 261 n. 6 Inscriptions CEG 104: 196 CEG 114: 196 CEG 528: 196 CEG 545: 196 CEG 606: 197 CEG 713: 197 CEG 737: 197 n. 9 FGrH 239 A 56: 127 n. 44 FGrH 239 A 60: 254 n. 20 FGrH 239 A 76: 21 n. 31 IG i3. 475.253–6: 123 n. 19 IG i3. 475.272–85: 123 n. 19 IG i3. 1147: 133 n. 64 IG ii2. 2318.9–11: 126 n. 39 IG ii2. 2319–23: 12 IG ii2. 2319–23.66–83: 16 n. 15 IG ii2. 2325.49: 12 n. 3 IG ii2. 3091.7–8: 6 n. 17, 253 n. 14 IG xiv. 1097: 12 n. 3 IGUR 216: 12 n. 3 Rhodes & Osborne 88: 60 SEG xxiii. 78: 48–9 Tod ii.204: 50 n. 11 Ion of Chios PMG 740: 5 n. 13, 83 n. 3, 210 n. 2
Laelius fr. 22 Malcovati4: 156 n. 50 Livy Periocha 59: 155 n. 44 ‘Longinus’ On the Sublime 15.7: 267 n. 26 Lucian Pseudologistes 17: 173 n. 8 Saturnalia 7: 181 n. 16 20: 181 n. 16 Lucilius 1093 Marx: 156 n. 50 Lycophron Alexandra 1217: 251 n. 5 Lycurgus fr. V 3 Conomis: 51 n. 15 Lysias 12.21: 94 13.27: 159 n. 67 13.54: 159 n. 67 13.59: 159 n. 67 Marmor Parium, see Inscriptions (FGrH 239) Melanippides PMG 757: 199 Menander Aspis 117–18: 207 n. 23 141–3: 87 n. 16 185–7: 87 n. 16 254–6: 87 n. 16 261–9: 87 n. 16 Dyskolos Hypothesis: 27 n. 42 7: 18 n. 22 184: 18 n. 22 242: 18 n. 22 296: 18 n. 22 438: 36 n. 18 747: 18 n. 22
321
322
Index Locorum
Dyskolos (cont.) 893: 18 n. 22 967: 120 n. 8 Samia 21: 18 n. 25 36: 18 n. 25 56: 18 n. 25 70: 18 n. 25 265: 18 n. 25 314: 18 n. 25 354: 18 n. 25 399–404: 36 n. 18 406: 18 n. 25 733: 120 n. 8 fr. 74: 25 n. 37 test. 42.24 KA: 18 n. 24
POxy 2656: 27 n. 43 POxy 3653 fr. 1.1–6: 252 n. 12, 258 n. 36 POxy 3653 fr. 2 col. 1.21–4: 81
Metagenes fr. 6: 140
Pausanias 1.14.5: 195 1.22.6: 268 n. 30 1.39.2: 83 n. 3 2.19.6: 104, 105 n. 58, 138 n. 83 2.25.4–5: 111 n. 81 2.28.3: 117 7.16: 95 n. 24 9.5.10–11: 215 n. 20 9.5.13: 83 n. 3 9.8.6: 83 n. 3 9.9.5: 83 n. 3 9.10.3: 83 n. 3 9.16.1: 99 n. 36 10.10.4: 83 n. 3
Mimnermus fr. 21 West: 5 n. 13, 210 n. 40
Pherecrates fr. 155.6–7: 199 n. 20
Nicolaus of Damascus FGrH 90 F 30: 117 Nonnus 35.89–91: 229 n. 12 Old Testament Genesis 3.22: 186 nn. 31–2 Orosius 5.10.9–10: 155 n. 44 Ovid Heroides 14.3: 105 n. 56 14.83–4: 105 n. 56 Metamorphoses 6.271–2: 71 n. 35 Papyri PMich ined.: 251 n. 10 POxy 663: 125 n. 33 POxy 1175: 265 n. 21 POxy 2081(b): 265 n. 21 POxy 2251: 73–5 POxy 2256 fr. 2: 137 n. 80 POxy 2256 fr. 3: 91 n. 11, 96–7, 137 POxy 2256 frr. 51–3: 75 POxy 2506 fr. 26 col. 1: 77 n. 61, 78 n. 67
Pherecydes of Athens fr. 1 Fowler: 69 n. 29 fr. 10 Fowler: 94 n. 20 fr. 12 Fowler: 74 fr. 40 Fowler: 71 n. 39 fr. 64 Fowler: 68 n. 21, 261 n. 9 fr. 95 Fowler: 5 n. 13, 82 n. 1, 215 n. 21 fr. 135A Fowler: 68 n. 21 Philochorus FGrH 328 F 30: 137 n. 79 FGrH 328 F 33: 123 n. 22 FGrH 328 F 117: 136 n. 75 Philocles fr. 2: 272 n. 49 Philodemus On Piety 51 Gomperz ¼ B6798 Obbink: 181 n. 8 Photius Lexicon s.v. Z ı ŒØ: 11 n.1 Pindar Isthmians 1.57: 198 n. 12
Index Locorum Olympians 2.43: 83 n. 3 7.49: 198 n. 13 13.109: 198 n. 12 Paeans 6.100–20: 261 n. 9 Pythians 4.13–16: 99 n. 37 9.112–16: 95 n. 24, 110 n. 76 fr. 36: 99 n. 36 Plato Apology 41a: 75 Euthydemus 287b: 184 n. 24 Gorgias 481d: 21 n. 32 Hipparchus 229b: 181 n. 8 Laws 631c: 25 n. 37 659a: 118 n. 1 713c-e: 181 n. 18 713e: 181 nn. 10, 19 760b-763c: 58 n. 36 760e-761d: 58 n. 36 761d: 58 n. 36 762b-c: 58 n. 36 762e-763a: 58 n. 36 762e: 58 n. 36 763a: 58 n. 36 763b: 58 n. 36 Politicus 271e-272a: 181 n. 15 271e: 181 nn. 13, 17 272a: 181 n. 14 272b-d: 182 272c: 182 272e: 181 n. 10 Republic 380a: 70 n. 32 566b: 107 Symposium 194b: 21 n. 30 Plutarch Aristeides 24.3: 181 n. 8
Cimon 8.8(-9): 117, 127 n. 43 10.7: 181 n. 15 14.2–15.1: 147 n. 20 15.2–3: 136 n. 74 16.1: 21 n. 32, 136 n. 77 16.4–17.3: 139 n. 89 16.9–10: 136 n. 74 18.7: 99 n. 35 Comparison of Lycurgus and Numa 1.12: 181 n. 16 Gaius Gracchus 10.5–6: 155 n. 44 Lives of the Ten Orators 832f: 156–7 n. 58 Lysander 25.3–4: 98 n. 33 Moralia 17a: 138 n. 86 266f: 181 n. 19 604f: 195 n. 1 Pericles 10.8: 156 n. 52 13.11: 21 n. 31 Phocion 20.4: 57 n. 34 Romulus 27.5: 155 nn. 44–6 Solon 30.2–3: 107 n. 65 Themistocles 5.5: 126 n. 38 23.1: 133 n. 66 32.3: 21 n. 32 Pollux 8.105: 50 n. 11 Prolegomena de Comoedia (Koster) III 60: 119 n. 6 Quintus of Smyrna 14.449–658: 251 n. 7 Sappho fr. 2.2–3: 198 n. 13 Scholia Aeschines 2.31: 136 n. 74
323
324
Index Locorum
Aeschines (cont.) 3.67: 21 n. 30, 24 n. 36 Aeschylus Eumenides 64: 97 n. 28 Prometheus Bound 853: 93 n. 17 Suppliants 37: 92, 95, 97, 116, 136 n. 73 Apollonius Rhodius 1.1212: 229 n. 12 Aristophanes Acharnians 67: 126 n. 36 Acharnians 377–82: 129 n. 49 Acharnians 502–3: 129 n. 49 Birds 281: 7 n. 18, 11 n. 1, 16 n. 15 Birds 1379: 11 n. 1 Clouds 971: 21 n. 31 Frogs 67: 13 n. 7, 19 n. 26 Frogs 992: 66 n. 15 Frogs 1124: 15 n. 14 Lysistrata 389: 18 n. 21 Lysistrata 1114: 26 n. 41 Lysistrata 1138: 136 n. 75 Lysistrata 1144: 136 n. 75, 139 n. 89 Wasps 151: 12 n. 3 Euripides Andromache 445: 12 n. 4, 140 Orestes 4: 71 n. 39 Orestes 11: 71 n. 39 Orestes 371: 212 n. 11 Orestes 872: 93 n. 17 Phoenissae 26: 214 n. 17 Phoenissae 53: 215 n. 21 Phoenissae 61: 214 n. 16 Troades 1128: 80–1, 266 n. 25 Homer Iliad 1.42: 93 n. 17, 102 n. 50 Iliad 1.59: 266 n. 23 Iliad 4.171: 93 n. 17 Iliad 4.376: 82 n. 2, 86 n. 15 Iliad 4.406: 83 n. 3 Odyssey 1.300: 224 n. 2 Odyssey 3.309–10: 224 n. 1, 225 n. 6 Odyssey 3.310: 225 n. 3 Odyssey 11.197: 251 n. 6 Odyssey 11.202: 251 n. 6 Odyssey 15.248: 225 n. 6
Juvenal 6.655: 266 n. 23 Lycophron 384: 251 n. 6 581: 250–1 n. 3 Pindar Pythian 3.167: 266 n. 25 Plato Apology 18b (Arethas): 11 n.1, 16 n. 15 Statius Thebaid 2.222: 93 n. 17 Thebaid 6.269: 93 n. 17 see also Eustathius, Servius, Tzetzes Semonides fr. 7.13–14: 186 n. 30 fr. 7.55–6: 186 n. 30 fr. 7.78–9: 186 n. 30 Seneca Agamemnon 470–578: 251 n. 7 Phaedra 1159–1200: 227 n. 9 Troades 349–59: 268 n. 28 1118–64: 268 n. 29 Servius on Aeneid 2.81: 250 n. 3, 255 n. 24 on Aeneid 10.497: 102 n. 50 Solon fr. 33.5–7: 269 n. 38 Sophocles Ajax 73: 172 n. 4 80: 172 n. 4 305: 172 n. 4 554–5: 185 n. 29 657–9: 46 693ff: 226 n. 7 787–8: 226 n. 7 807: 226 n. 7 814: 41 n. 34 815–22: 34 864: 34 866: 34 n. 11 891–2: 34
Index Locorum 891: 34 n. 11 985–9: 31 n. 3, 46 1003: 34 n. 12 1111–14: 273 n. 50 1168: 31 n. 3 1316–73: 31 n. 3 Antigone Hypoth. I (Pearson): 210 n.3 Hypoth. II (Pearson): 210 11ff: 206 44: 206 47: 206 60: 206 63: 206 65–6: 207 n. 21 67: 206 77: 206 81: 204 n. 11 86: 206 93: 206 99: 208 332–64: 185 n. 27 464: 233 n. 22 491: 206 531–3: 206 544: 207 545: 206, 207 n. 21 548: 206 549: 206 559–60: 206 n. 20 566: 206 568: 219 n. 23 569: 207 571: 207 n. 23 572: 202–8 574: 202 n. 5 576: 202 n. 5 769–71: 208 774: 242 n. 48 844–5: 198–9 885: 242 n. 48 888: 242 n. 48 939: 33 n. 8 941: 208 1066–7: 226 n. 7 1115–54: 226 n. 7 Electra 1–85: 53 n. 20
325
5: 199 13–14: 238 n. 37 14: 231 33–4: 231 37: 231, 234 47–8: 242 n. 50, 249 61: 242 n. 50 66: 231, 244 n. 56 70: 231 82–5: 72 n. 42 97–9: 233 nn. 21, 24 115–17: 232 121–2: 232 124–6: 232 126–7: 232–3, 237 165: 241 n. 48 171–2: 240 n. 43, 242 n. 49 185–6: 242 n. 49 187–8: 241 n. 48 189–92: 242 n. 48 205–12: 234 209: 244 n. 56 245–50: 234 247: 244 n. 56 266–302: 234 n. 26 272: 233 n. 24 273–4: 247 n. 72 277–81: 241 n. 48 294–8: 241 n. 48 300–2: 234 n. 26 303–4: 232, 234 307–8: 238 n. 41 310–16: 234 n. 26 312–13: 241 n. 48 318–19: 240 n. 43 319: 242 n. 49 326: 234 n. 26 338–9: 240 n. 42, 243 n. 52 342: 234 n. 26 366–7: 234 n. 26 379–86: 234 n. 26, 242 n. 48, 245 389: 244 n. 58 417–23: 235 417: 236 n. 30 421: 236 n. 30 435–8: 236 n. 33 445–6: 241 n. 48 453–6: 234 456: 244 n. 56
326 Electra (cont.) 462: 204 n. 10 468–9: 235 n. 27 493–4: 236, 249 497: 236 516–51: 236 517–18: 242 n. 48 566–72: 165 n. 1 580–3: 237 580: 237 n. 34 582–3: 242 587–600: 244 n. 58 587: 233 n. 24, 244 n. 58 589–90: 241 n. 48, 248 n. 75 594: 244 n. 58 597–8: 247 n. 72 600: 244 n. 58 601: 241 n. 48 603–5: 237–8, 243, 244 603–4: 239 604: 238 605–9: 238 n. 41 653–4: 241 n. 48 674: 243 675: 243 n. 52 677: 243 766–8: 239 778–82: 239 790: 247 n. 72 808–22: 243 813–14: 243 n. 52 814: 242 n. 48 819: 243 n. 52 837–41: 240 842–8: 244 846–8: 240 846: 240 851–2: 241 853: 241 871–2: 243 n. 52 903: 204 n. 11 911–12: 242 n. 48 920–5: 243 929: 244 930–7: 243 n. 52 950: 243 n. 52 954–7: 243–5 955: 233 n. 24, 244 n. 58 957: 244 n. 57
Index Locorum 961–6: 241 n. 48 967: 245 n. 60 971–2: 245 n. 60 979: 244 n. 56 980: 245 n. 60 997–8: 238 n. 38 1015–16: 240 1019–20: 245 1033: 244 n. 59 1041–2: 240 n. 44, 249 1045: 245 1050–4: 73 n. 43 1056: 240 n. 44 1058–62: 240 n. 44 1078–80: 240 1078–9: 245 1080: 244 1126: 204 n. 11 1133: 241 n. 48 1154–6: 240, 241 n. 46 1154: 247 n. 72 1158: 204 n. 11 1163: 204 n. 11 1194: 247 n. 72 1195–6: 242 n. 48 1208: 204 n. 11 1211–26: 232 1224: 204 n. 11 1286: 204 n. 11 1289: 247 n. 73 1296–9: 246 n. 65 1296: 247 n. 73 1297: 246 1307–9: 245 n. 61 1309: 247 n. 73 1319–21: 245 1319: 246 1320–1: 244 1323–5: 244 n. 57 1326–75: 53 n. 20 1326–38: 72 n. 42 1335–7: 246 1354: 204 1357: 204 1367: 246 n. 65 1368: 230 n. 16, 245 1376–83: 245 1376–7: 245 n. 62 1380–1: 245 n. 62
Index Locorum 1382–3: 249 1384–97: 246 1385: 241 1387: 241 1391–2: 241 1393: 241 1394: 241 1398: 245 1400–1: 245 n. 63 1406: 247 n. 69 1407–8: 241 1410: 245 n. 69 1411–13: 247 n. 72 1411–12: 246, 247 n. 69 1413–14: 241 1415: 238 n. 38 1420–1: 241 1422–3: 241 1423: 241, 257 n. 35 1426–7: 242, 247 1427: 241 n. 46, 247 n. 73 1430: 247 1434: 257 n. 35 1437: 247 1448–65: 247 1483–90: 247 1487–90: 242 n. 50 1489–90: 247 1497–8: 248 n. 75, 249, 257 n. 33 1508–10: 257 n. 35 Oedipus at Colonus 270–2: 221 n. 29 374–5: 82 546–8: 221 n. 29 831–47: 33 n. 8 835–86: 38 n. 26 842–4: 38 n. 26 856: 38 n. 26 872–86: 33 n. 8 883: 38 n. 26 991–9: 221 n. 29 1103: 204 1255: 247 n. 70 1292–4: 82 1293–5: 82 1298: 87 1375: 23 n. 34 1407–10: 226 n. 7 1420: 247 n. 70
1422–3: 82 1422: 82 1431: 247 n. 70 1769–72: 226 n. 7, 257 n. 32 Oedipus Tyrannus 69–70: 216 100: 220 122–3: 159 n. 66 124–5: 159 n. 66 126: 159 n. 66 258–65: 216 309: 220 366–7: 217 413–14: 218 417–18: 222 418: 220 422–3: 218 425: 218 437: 218 438: 222 454–6: 220, 222 457–60: 217, 218 460: 216 577–81: 219 622–3: 220 631–3: 219 639: 219 641: 220 659: 220 669–70: 220 715–16: 159 n. 66 791: 217 804–5: 221 n. 29 807–9: 221 n. 29 823–7: 221 836–47: 221 990: 38 1001: 38 1009: 38 1013: 38 1056–75: 226 n. 7 1076–1109: 226 n. 7 1112–13: 38 1121: 38 1123: 39 n. 27 1146: 38 1147: 38 1152–66: 33 n. 8 1152–4: 39
327
328
Index Locorum
Oedipus Tyrannus (cont.) 1153: 38 1158: 39 1166: 39 1250: 217 1371–5: 221 1403–8: 217 1411–12: 221 1436–7: 222 1446: 222 1449–54: 222 1460–1: 222 1462–1510: 257 n. 32 1517–18: 222 1519–20: 222 1522: 33 n. 8 Philoctetes Hypoth. I (Pearson): 212 n. 11 3–4: 260 n. 3 45–8: 275 n. 52 50: 260 n. 3 57: 260 n. 3 72–3: 273 n. 50 86–120: 275 91: 260 n. 2 96: 260 n. 3 108–11: 242 n. 50 125–9: 275 n. 52 240–1: 260 n. 3 242: 260 n. 3 260: 260 n. 3 357–8: 260 n. 2 364: 260 n. 3 468: 260 n. 2 542: 260 n. 3 582: 260 n. 3 719: 260 n. 2 814–18: 33 n. 8 865: 276 874: 260 n. 2 895: 276 904–5: 260 n. 2 933: 260 n. 2 940: 260 n. 3 969–70: 55 n. 27 1003–54: 33 n. 8 1066: 260 n. 3 1080–1222: 276 1237: 260 n. 3
1284: 260 n. 3 1298: 260 n. 3 1299–1302: 35 1300–2: 33 n. 8 1310–13: 260 n. 2 1402: 276 1408: 276 1440–4: 257 n. 36, 276 1453: 172 n. 4 Trachiniae 9–21: 229 21–5: 229 28–35: 229 37: 229 41–2: 229 46–51: 229 52–7: 229 142–54: 188 143: 188 200ff.: 226 n. 7 232: 204 298: 229 310–19: 229 312: 229 330–1: 229 383–4: 229 401: 229 439–49: 229 459–63: 229 463–7: 229 486: 229 490: 229 491–2: 229 494–6: 229 n. 13 523–8: 229 525: 230 528: 230 530: 230 580: 230 1147–56: 23 n. 33 1167: 198 n. 13 1266–78: 257 n. 30 fr. 10c: 255 n. 27 fr. 202: 68, 272 frr. 206–222b: 265 n. 21 fr. 210.8–9: 265 fr. 210.45: 266 n. 24 fr. 210.47–9: 265 fr. 278: 181 n. 20
Index Locorum fr. 395: 75 fr. 429: 250 n. 3, 254 nn. 22–3, 255 fr. 432: 250 n. 3, 254 n. 22, 255 fr. 441a: 35 fr. 479: 250 n. 3, 254 nn. 22–3, 255 n. 24 fr. 487: 69, 266 fr. 492: 266 fr. 522: 267 fr. 523: 267 fr. 524: 267, 268 fr. 526: 267 fr. 557: 265 fr. 583: 188 fr. 693a Lloyd-Jones: 73 n. 43 fr. 694: 68, 69, 272 fr. 695: 68, 273 fr. 696: 68 test. 1.72–3 Radt: 200 n. 22 test. 32a/b Radt: 117 test. 36 Radt: 117 test. 177–84 Radt: 200 n. 22 Statius Thebaid 12.177–463: 74 n. 47 12.677–804: 74 n. 47 Stesichorus PMG 190: 273 n. 50 PMG 192.2: 78 n. 67 PMG 193: 77 n. 61 PMG 217: 191 n. 1 PMG 219: 191 n. 2, 235 Stobaeus 1.6.9: 27 n. 44 4.1.8: 50 n. 11 4.29.10: 27 n. 44 Strabo 1.2.12–13: 140 9.3.9: 68 n. 22
Telesilla PMG 721: 71 n. 35 Theocritus 10.19: 25 n. 37 Theognis (eleg.) 1252: 198 n. 13 Theognis (trag.) fr. 2: 272 n. 49 Theophilus (com.) fr. 12.8: 141 Theopompus (hist.) FGrH 115 F 88: 136 n. 76 Thucydides 1.9.1: 273 n. 50 1.45.2: 136 n. 77 1.100.2–101.2: 136 n. 74 1.101.2: 136 n. 78 1.102–5: 136 n. 74 1.107.4–6: 161 n. 71 1.107.5–108.2: 161 n. 74 1.108.3: 161 n. 75 1.135.3: 133 n. 66 2.15.2: 122 n. 16 3.49.1: 131 n. 60 4.102.2–3: 136 n. 74 4.119.1: 136 n. 77 5.23.3: 136 n. 78 5.115.2: 131 n. 57 6.20.1: 121 n. 12 6.31.3: 123 n. 19 8.65.2: 154 n. 40 8.66.2: 56 n. 29 8.69.4: 56 n. 29 Timocreon PMG 731: 25 n. 37 Tragica Adespota fr. 7: 71 n. 38
Suda 118: 201 3800: 44 n. 39 Ł366: 201 3203: 201 s815: 6–7 n. 17
Valerius Maximus 8.15.4: 155 n. 42
Telecleides fr. 1: 181 nn. 10–12, 182 nn. 21–2
Velleius Paterculus 2.4.6: 155 n. 44
Tzetzes on Lycophron Alex. 175: 69 n. 29
329
330 Vitae Euripides 20: 21 n. 30 see also Aeschylus (test. 1), Aristophanes (test. 1), Sophocles (test. 1) Xenophon Cyropaedia 1.2.4: 58 n. 35
Index Locorum 1.2.9–12: 58 n. 35 Hellenica 1.7.22: 272 n. 45 2.3.23: 56 n. 30 2.3.50–1: 56 n. 30 2.3.55: 56 n. 30 Poroi 4.51–2: 50–1
General Index In this index, Greek names are shown in their most familiar form (usually latinized). The few untransliterated Greek words or phrases that are indexed are placed where they would have stood if transliterated. Titles of dramas and other works will be found under the names of the authors to whom they are traditionally ascribed. Major discussions are indicated by bold font. Abas (son of Lynceus) 113 nn. 83, 87 Acamantis (Athenian tribe) 48–9 Acamas (son of Theseus) 228 n. 11 Acastus 267 sons of 266 wife of 264 n. 17 Achelous 229 Achilles 57, 70, 253 n. 18, 260, 261 n. 7, 266 can’t perish at Aulis 212 ghost of, in Odyssey 261 n. 8, 262–3 ghost of, and Polyxena 267, 269 on Scyros 264, 265 Acrisius 94, 113 n. 83 actors limit on number of 31, 104 n. 53, 138–9, 142 non-speaking 31 n. 3, 112 n. 82 Actors of Dionysus 170 Adam and Eve 186 adolescents 47–60 see also Hippolytus, Neoptolemus, Orestes, Pentheus Adrastus 74, 83, 266 n. 23 Aegina 198 n. 12 Aegisthus 38, 78–9, 94, 107, 132 n. 62, 150–2, 168, 192, 204 n. 12, 211–12, 230, 233, 234, 235–6, 237 n. 35, 241 n. 48, 242, 243–4, 247, 248 n. 75, 249 Aegyptus 93–4, 95 n. 24, 103, 114, 115 sons of 89, 91, 92, 97, 102, 103–4, 108, 110, 115; see also Lynceus Aeschines (orator) 50 Aeschylus 82–201 Agamemnon 38, 139, 153–4, 164–88 see also Oresteia
Aigyptioi 73, 89–90, 96–8, 102–4, 116–17, 136 n. 75 Aitnaiai 119 Argeiai 74 Choephoroi 52–3, 107, 137, 139 n. 87, 142, 150–4, 189–94, 232, 235, 239 n. 42 see also Oresteia Cretan Women see Kressai Danaid trilogy 89–117 Danaids 89–90, 96, 104–13, 117 defeated by Sophocles (468) 127 Diktyoulkoi 74 Edonoi 137 n. 81 epitaph 9, 195–201 Eumenides 15, 105, 124, 134 n. 70, 137–8 see also Oresteia Heracleidae 23 Hiereiai 75, 137 n. 81 Hoplo¯n Krisis 67 Iliadic trilogy 98 Iphigeneia 67 Kares or Europe 29, 270 Kressai 75 Myrmidons 66 and the Mysteries 30 n. 1 Nereids 66, 80 Niobe 70–3 number of choral songs per play 111 n. 80 Odyssean trilogy 98 Oresteia 76, 78–9, 132–3, 137–8, 142, 143–94 see also Agamemnon, Choephoroi, Eumenides Palamedes 65–7, 251
332
General Index
Aeschylus (cont.) Perrhaebides 73–4 Persians 126 n. 40, 132, 164 on Persian and Punic wars 7 n. 18 Philoctetes 263 Polydectes 74 posthumous restagings 119 Prometheia (?) 16 n. 15 Prometheus Bound 40–4, 138 n. 86, 183–4, 188 Prometheus Pyrkaeus 16 n. 15 Prometheus Pyrphoros 16 n. 15 Prometheus Unbound 41 n. 33 Propompoi 75 Proteus 76–80 Psychostasia 138 n. 86, 142 Semele or Hydrophoroi 270 Seven against Thebes 15, 63, 82–8, 101 n. 46, 138 n. 86, 220, 223 in Sicily 119 Suppliants 38, 89–117, 124, 133–40, 164, 199 n. 21 Tennes-play 75 tetralogies and trilogies 7 n. 18, 15–16, 90 n. 5, 97–8 see also under names of particular tri/tetralogies Theban trilogy 98 Theoroi 137 n. 81 Threissai 67 using tragedy as political weapon 132–3, 139 Women of Aetna see Aitnaiai Aetna 119 affinity terms 205 n. 16 Agamemnon 148–50, 164–6, 167–9, 171, 174, 176, 235–7, 253 n. 18, 267–9 condemns Palamedes 251, 258 see also Atreidae Agaue 277, 285 n. 2, 286 Agenor (son of Antenor) 261 n. 7 agoge (Sparta) 49, 57, 58 agronomoi (in Plato’s Laws) 58 Agyrrhius (fl. 403–374/3) 141 Ajax (son of Oileus) 149 n. 24, 252–3, 258 Ajax (son of Telamon) 226 n. 7, 244 n. 57
alternative deaths for 210 suicide of 34 Alcm(a)eon (son of Amphiaraus) 211, 224–5, 240, 241 Alcmene 113 n. 83 Aletes (son of Aegisthus) 248 n. 75, 249 Alexandros (Trojan prince) see Paris ¼º 198–200 ambassadors 131 n. 59, 197 Ammon, oracle of 98–9 Amphiaraus 85, 86 n. 14, 211 n. 7, 240, 241, 266 n. 23 Amphilochus (son of Amphiaraus) 211 n. 7 Amphion 70–3, 81 Amynias (fl. 423–422) 131 n. 59 Androcles (d. 411) 154 n. 40 Andromache 262 see also Euripides (Andromache) animals did not eat one another under Cronus 181 never killed on stage 35–6 ‘pantomime’ 36 n. 19 talked with men under Cronus 183 Anius of Delos 250 n. 3 Anticleia (mother of Odysseus) 251 n. 6 Antigone (daughter of Eurytion) 69 Antigone (daughter of Oedipus) 83 n.3, 202–8, 219, 222, 226 n. 7, 242 n. 48, 257 alternative deaths for 210 see also Sophocles (Antigone; Oedipus at Colonus) Antiope 71, 72 Antipater of Thessalonice 195, 201 Antiphon 156–8 Apaturia 51 n. 17 Aphrodite and Cyprus 284 in Danaid trilogy 90, 91 n. 10, 93 n. 18, 101, 104, 107, 109, 110, 112–13, 114, 115 and Dionysus 285 in Euripides’ Hippolytus 227–8 Apollo 35, 212–13, 220, 232, 249, 261 n. 7, 275, 279 and Neoptolemus’ death 271–2, 275
General Index in the Oresteia 132, 169, 189, 191–2, 193 see also Delphi apotympanismos 43 Archandros (son-in-law of Danaus) 95 n. 24 Archedemides (archon 464/3) 137 Archeptolemus (d. 411) 125 archonships, eligibility extended (457) 162 Areopagus Council 136 n. 74, 146–7, 154, 159–60, 162 Ares 133, 246 Argeia (daughter of Adrastus) 74 Argos as ally of Athens 133 dedication at Delphi 83 n. 3 eagle/hare omen seen at? 171–7 plain of 199 Themistocles at 133 see also Aeschylus (Danaid trilogy; Oresteia) Ariadne 225 n. 6 Aristarchus 224, 225 n. 6 Aristodicus of Tanagra 156, 159, 162 Aristogeiton 154 n. 40, 243 n. 53 Aristophanes Acharnians 123 n. 24, 130 Birds 54, 125 n. 34 Clouds 47, 54, 55, 129 Daitaleˆs (Banqueters) 55 Frogs 31, 124, 129, 140 Knights 47, 54, 124, 129 Lysistrata 17–18, 24, 125 n. 34 mystifying or misleading titles 24 play titles cited in 13, 14 political attitudes 125, 131, 141 prologues 24–5 Thesmophoriazusae 31 violence in 42 n. 36 Wasps 47, 54 Wealth 24–5 Aristotle Didaskaliai 11–13 on myth 4 n. 12 Arkteia (bear ritual) 49 Artemis 35, 49, 165, 167, 169, 173, 174 n. 11 artistic images:
333
Bologna 230 (Aegisthus Painter, C19 Prag) 152 n. 33 Boston 91.227a/226b (Berlin Painter, C22 Prag) 152 n. 33 London BM Br. 621 (Etruscan mirror) 84 Neoptolemus in 260–1 Orestes in 84 Oxford Ashm. 1965.359 (Etruscan scarab) 84 n. 7 Siena Mus. Arch. 731 (Etruscan urn) 84 n. 7 Asia (daughter of Themistocles) 21 n. 32 assassination 148, 154–60, 243 n. 53 Astyanax 261, 262 n. 11, 263, 274 Astyoche (sister of Priam) 266 Athena 40, 113 n. 84, 132, 146 image of, in Eumenides 137–8 instructs Tydeus to kill Ismene 210 strikes lesser Ajax with thunderbolt, 149 n. 24, 252, 258 Athenaeus (son of Pericleidas) 136 Atreidae 148, 173, 174 n. 12, 175 n. 14; see also Agamemnon, Menelaus Atreus 168 audience 118–42 children in 120 conscripted into drama 153 n. 36 foreigners in 121 social composition of 124–9, 140 women in 120–1 Aulis 171–7 Autolycus (superintendent of ephebes) 49 Autonoe 277, 286 barbarians vs. Greeks 114, 280, 284 ÆŁıåÆØ Ø 197 Berlin Wall 194 board games 95, 114, 254–5 Boeotia 161 Boeotus (murder victim) 221 n. 29 book trade 13, 15, 20 Brauron 49 Briseis 253 n. 18 Cadmus 277, 282–3, 285–6
334
General Index
Calchas 164, 168, 171, 172 n. 3, 173, 174–5, 176, 268 Callias (archons of 456/5, 412/1, 406/5) 21 n. 31 Callias (comic dramatist) 12 n. 3 Callimachus (archon 446/5) 21 n. 31 Calydonian Boar Hunt 69 Campania 140 Caphereus, Cape (Euboea) 148, 149 n. 24, 251 Carcinus, sons of 184 n. 24 Cassandra 33 nn. 7–8, 150, 169, 257 n. 57, 269 sacrilegious seizure of 253 Ceyx 23 n. 33 choregoi 35–6, 126 chorus, secondary 90, 91, 100–1, 103–4, 113, 115 Chryseis 253 n. 18 Chrysis (in Menander’s Samia) 18 n. 25 Chrysothemis 234–5, 240, 241 n. 48, 243–5 Cicero 154–5, 156 n. 51 Cimon 98–9, 127, 132, 136–7, 139, 147 n. 20 sons of, 21 n. 32 Cithaeron, Mt 279–80 Cleisthenes (fl. 508/7) 50, 161 Cleon 24, 29, 124, 125, 129 Cleophon 129 Clytaemestra 63, 78–80, 94, 132 n. 62, 142, 143–5, 150–4, 175, 191, 204 n. 12, 212, 230–49 myth of, unalterable? 210–12, 223, 224–5 see also Aeschylus (Agamemnon, Choephoroi, Eumenides); Euripides (Electra, Iphigeneia at Aulis, Orestes); Sophocles (Electra) Clytia 71 n. 39 confirmation 51 n. 17 Corinth 221 costume 107, 117 Council of Five Hundred 146 n. 20 Crassus, L. Licinius (cos. 95) 155 n. 48 Crates (comic dramatist) 12 n. 3 Crathis, River 140 Cratinus 12 n. 3, 126 Dionysalexandros 19 n. 27, 29, 125
Drapetides 126 n. 37 Idaioi 19 n. 27, 29 Ploutoi 41 n. 33 Creon (son of Menoeceus) 83 n. 3, 206, 207, 208, 216, 219, 222, 254, 257 accomplice in Laius’ murder? 213 n. 14 Croesus 98 Cronus 179 felicity under reign of 180–4, 186 used to mean ‘fool’ 184 crowd hysteria 281 Curran, John Philpot 115 Cyprus 197, 284 Cyrus the Great 176 n. 15 Dallas 213, 223 Danae 94, 113 n. 83 Danaids see Aeschylus (Danaid trilogy; Suppliants) Danaus 89 n. 1, 90, 92–105, 106–10, 111–12, 114–15, 116, 117, 135 death humans subject to, even primevally 186 n. 32 on stage 34 decapitation 38, 278, 280, 281 Deianeira 228–30, 238 n. 38 Deidameia (mother of Neoptolemus) 264 Deioneus 73–4 Delian League 128 Delphi 99, 193, 217, 231 Neoptolemus killed at 68, 261–2, 263, 270–5 democracy 55, 59, 107, 110, 114, 126 n. 35, 127, 130–6, 139–40, 152, 157, 200, 284 Demodocus (Phaeacian singer) 263 Demophon (son of Theseus) 228 n. 11 Demos (son of Pyrilampes) 21 n. 32 Demosthenes (father of the orator) 106 Desire (Pothos), companion of Aphrodite 110, 113, 115 deus ex machina 64, 79, 212–13, 226, 286 Dictys 74 Dike 85, 86 Diodorus of Sardis 195, 201
General Index Diomedes 251, 281 n. 5 Dione (mother of Niobe) 71 n. 39 Dionysia audiences at 118–42 as international event 128, 141 tribute delivered at 128 Dionysus in Euripides’ Bacchae 277–87 flogged on stage 37 as theatre spectator 30 Dioskoroi 79 Dodona 99 Dotium (Thessaly) 266 Drabescus 136 n. 74 dreams 144–5, 152–3, 191 n. 2, 213, 223, 235–6 drinking on stage 31 Earth see Gaea eating on stage, impossible 31 Ecphantides 12 n. 3 education 53–9 Egypt customs of 37 n. 23 Menelaus and Helen in 76–7 plains of 199 supposed marriage law in 97, 103 see also Aeschylus (Danaid trilogy, Suppliants), Euripides (Helen) Eido/Eidothea 76 Eioneus 73–4 eisangelia 146 ekkyklema 31 n. 4, 128 n. 46 Elaeo (daughter of Anius) 250 n. 3 Electra 94–5, 106, 189–92, 204, 212, 232–49 Electryon (son of Perseus) 113 n. 83 Eleusis 198 n. 12 Epaphus 113 ephebeia 47–60 ‘ephebic’ oath 50 Ephialtes 122, 136 n. 74, 139 n. 88, 146, 154–62 Ephippus Epheboi 51 n. 14 Epicaste 215, 225 n. 6 Epicrates, law of (335) 51 n. 15 Epigoni 63, 83 n. 3 epikle¯ros 87 n. 16, 106
335
Epimetheus 186 epitaphs 195–201 Erigone (daughter of Aegisthus) 248 n. 75, 249 Erigone (daughter of Icarius) 248 n. 75 Erinyes 132, 145, 154, 187, 191, 192–4, 232, 248 n. 75, 249, 274 n. 51 Eriphyle 211, 224–5, 240, 266 n. 23 Eros 114, 115, 229 Eteocles 63, 82–8, 204 Etruscan art 84 Euadne 34–5 Euaeon (narrowly convicted of murder) 221 n. 29 Euboea 251–2 Eunomides (father of Phrynichus) 21 n. 32 Euphorion (son of Aeschylus) 29, 44 n. 39, 196 n. 7 Eupolis Autolycus 14 n. 10 Marikas 125 Euripides Alcestis 34 Alexandros 252 n. 11 Andromache 12, 22, 34, 39, 130, 131 n. 56, 140, 263–4, 265, 270, 273–5 Andromeda 33 n. 8 Antigone 210 Bacchae 54, 277–87 connected suites of plays 6, 253 Cretan Women, see Kressai Electra 53, 79, 106, 231 n. 17, 234 n. 24, 248 epigrams on 200 n. 22 Hecuba 264, 265, 267, 268–9 Helen 22, 39–40, 45–6, 76–7, 81, 131 n. 56 Heracleidae 23, 269 Heracles 22–3 Hippolytos Kalyptomenos 14, 15, 53, 60, 248 Hippolytos (Stephanephoros) 14, 15, 34, 53, 54, 227–8, 248 Hypsipyle 22 Ion 54, 121 Iphigeneia at Aulis 40, 131 n. 56, 172, 212, 264 Iphigeneia in Tauris 22, 204–5
336
General Index
Euripides (cont.) keeps Neoptolemus offstage 264 Kressai 23 Melanippe Desmotis 22, 140 mystifying or misleading titles 22–3 no double titles recorded 18–19 Oedipus 214 Orestes 5, 53, 54, 131 n. 56, 205 n. 15, 212–13, 226, 275 Palamedes 251–2 Peleus 264 Peliades 117 Philoctetes 263 Phoenissae 23, 82, 84 n. 8, 220 productions abroad 119 Rhesus 40 Scyrians 264, 265 Stheneboea 53 Suppliants 23, 34–5, 53–4, 130, 131 Telephus 242 n. 50 Temenidae 117 Troades 131, 252, 257, 264 Euryanassa 71 n. 39 Euryganeia (second wife of Oedipus) 215, 219 Eurypylus 261, 262, 265–6 Eurystheus 23 n. 33, 269 Eurythemiste 71 n. 39 Eurytion 69 Eustathius 28 n. 45 euthynai 146 Euxitheus (alleged killer of Herodes) 156–8 Ewing, Bobby 213, 223 ‘false consciousness’ 95 ‘False Merchant’ 275 n. 52 fishing 77–8, 251 Four Hundred (oligarchy of 411) 56 Fowler, H.W. 209, 227 n. 10 fragmentary plays 61–81, 89–117, 250–76 see also under titles of particular plays in their authors’ entries Fulvius Flaccus, M. (cos. 125) 155 n. 48 Furies see Erinyes Gaea (Earth) 115, 192 Gaisford, Thomas 68
Gela (Sicily) 195–7 ‘generation gap’ 55, 60 ghosts 144, 145 n. 12, 267, 268, 269 Glaucus (son of Minos) 75 Golden Age 179, 180, 181 Gorgias 156–7 n. 58 Gracchus, C. Sempronius 155 n. 42 Gracchus, Ti. Sempronius 156 n. 51 Hades 192 Haemon (son of Creon) 202–3, 205–7 marries Antigone 210 Harmodius 154 n. 40, 243 n. 53 Harmony, companion of Aphrodite 110, 113, 115 Heaven see Uranus Hector 262, 265 Hecuba 266, 269 see also Euripides (Hecuba, Troades) Helen 76–9, 144, 165, 167, 175 n. 13 killed (?) 212–13 phantom of 76–8, 79 Hellanicus 86 n. 15 Hellespont, Xerxes’ bridge over 126 n. 40 helots 136, 139 n. 89; see also Messenians Hera 277 Heracles 112–13 apotheosis 257 as deus ex machina 212, 257, 260 in Sophocles’ Trachiniae 228–30, 256–7 herd instinct 281 Hermes 113 n. 84, 151, 185 Nychios 151 Hermione marriage to Neoptolemus abolished 275 nurse of 73 taken hostage 212–13 see also Euripides (Andromache), Sophocles (Hermione) Herodes (Antiphon 5) 156–7 Heron of Alexandria 252, 253 Hesiod 181, 183 Hipparchus (son of Peisistratus) 107, 154 n. 40
General Index hippeˆs (Solonian property class) 162 n. 77 Hippias (son of Peisistratus) 157 n. 58 Hippolytus 53, 54, 57, 227–8 servant of 73 Holmes, Sherlock 65 Homer affinity terms in 205 n. 16 glossing over matricide 211–12, 223, 224–5 glossing over war atrocities 262–3 Odyssey 64, 211–12, 215 homicide justifiable 221 n. 29 nocturnal 147–60 see also assassination Hydra, the Lernaean 230 Hyllus 106 hymeneal songs 101 Hyperbolus 125 Hypermestra 90, 104–5, 106, 107–9, 111, 112, 114, 115, 138 Hypotheses 27 n. 42, 68, 117, 209–10, 254 n. 20, 258 Icos 266 Idiotes (son of Theogenes) 21 n. 32 ‘initiation’ 47–48, 61 n. 17 Ino 277, 286 intelligence, origin of 183–6 Io 99, 199 nn. 17, 21 Iocaste 213–21 Iolcus 69 Iole 106, 228–9 Ion of Chios 210 Ionians 121 Iphigeneia 67, 79–80, 150 n. 29, 167–8, 171, 172, 204, 205 n. 14, 268 Isagoras (archon 508/7) 161 Isles of the Blest 282 Ismene 83 n. 3, 202–8, 210, 222, 254 Italia (daughter of Themistocles) 21 n. 32 Italy, tragic productions in 119, 140 Ithome campaign 136–7 Ixion 73–4 killing, never seen on stage 33–7, 44 kryptoi 58
337
Lacedaemonius (son of Cimon) 21 n. 32, 136 Laelius Sapiens, C. 156 and nn. 50–1 Laius 159 n. 66, 213–14, 216–17, 219, 220–1 Lamachus 131 n. 59 Laodamas (son of Eteocles) 83 n. 3, 210 ‘left-wing’, defined 125 n. 35 Lenaea 16 n. 15, 29 Leto 71 Libya (daughter of Epaphus) 99 Lichas 229 literary scholarship, aims of 3–4 London Charivari, The 17 n. 18 Lowry, M.J.C. 65 Lycus 280 n. 1 Lydia(ns) 280, 281 Lynceus 90, 95 n. 24, 105, 106–9, 110, 111–13, 114, 115, 138 n. 85 Lyrceia 111 n. 81 Lysistratus (effeminate) 24 n. 35 Machaereus (killer of Neoptolemus) 68, 271 Marathon, battle of 195, 197, 200 marriage 38, 87 n. 16, 91, 93–5, 97, 100, 101–2, 106, 110, 113–14, 115, 229, 245 n. 60, 271–4 masks 24 n. 36, 31, 43 n. 37, 46, 84, 153 n. 36 ‘ Ł Ø Ł’ 185–6 mechane 31, 44, 128 n. 46 Medea 238 n. 38 Megacles (fl. 420s) 131 n. 59 Melanippides 199 n. 20 Meleager 229 n. 12 Meletus Oidipodeia 15 n. 16 Menander Dyskolos 18 Misoumenos 18 plays named after foreign women 18 productions abroad 119 n. 6 Samia 18 Menelaus 76–9, 144, 149 n. 24, 164–5, 167, 267, 271, 272–3, 274 as potential avenger of Agamemnon 78–9
338
General Index
Menelaus (cont.) negatively portrayed in tragedy 130–1 Merchant Taylors’ School 61 Messenians 136, 139 n. 88 Metagenes Thouriopersai 12, 140 metics 123–4, 128 military training at Athens 49–51, 58 Mimnermus 210 Minos 75, 150 miracles in Bacchae 279–80 Misgolas 50 mob violence 281 Muses 284 Mysteries, Eleusinian 30 n. 1, 31 myth, tragedy’s treatment of 4–6, 22–4, 63–5, 165, 191, 209–30, 263–76 Mytilene debate (427) 131 Nauplius 148–9, 250–8 first great single-handed oarsman 251 Neoptolemus 34, 53–4, 55, 57, 68–9, 80–1, 226, 257, 259–76 ignored by Aeschylus 263 kept offstage by Euripides 264 Nessus 230 Nicias 125 Nicophon Sirens 12, 140 Night, mother of the Erinyes 154 Niobe 70–3 mother of 71–2 nurse of 71–2 Nisus (king of Megara) 150 nurses 71–3, 145, 229 oaths 50, 59, 193, 242 n. 50, 249, 273 Odeion 21, 270 Odysseus 140, 212, 244 n. 57, 258, 260, 261 n. 5, 262–3, 264, 267, 269, 275 and Palamedes 250–1, 255 Odyssey see Homer Oechalia 230 Oedipus 213–23, 226 n. 7 blinded by Laius’ servants 214 blinded by Polybus 214
buried at Thebes 220, 223 children of 222; see also Antigone, Eteocles, Ismene, Polyneices cursing his sons 23, 87 n. 15 not cursing his sons? 222 and the shepherds 38–9 Oeno (daughter of Anius) 250 n. 3 Old Comedy (see also Aristophanes) political interference with 126, 141 political tendencies of 125, 141 Old Man of the Sea 77, 78 Olympus, gods of 192, 194 omens 164–5, 171–7 Onchestus 198 n. 12 oracles 93–6, 98–103, 106–7, 109, 110, 114, 116, 191–2, 193, 210, 216–19, 231, 256 Orestes 52–3, 54, 57, 79, 83, 84, 107, 133, 145, 150–4, 169, 189–94, 204, 205 nn. 14–15, 212–13, 224–5, 230–49, 271–2, 274–5 paidagogos of 72, 204, 231, 239, 246 n. 65 ostracism 132, 137, 139 paidagogoi 53, 72–3, 231, 238 n. 37, 245, 246 n. 65 Palamedes 67, 148, 149 n. 24, 250–8 Panathenaea, musical contests at 21 n. 31 Pandora 73, 185–6 Papirius Carbo, C. (cos. 120) 155 n. 48 Paris 165, 166, 263 parrhesia 129, 141 Pasion (banker) 106 Ł Ø Ł 178–87 pay for civic functions 122, 128 Peisistratus 107 Peitho see Persuasion Pelasgus 73, 74, 90, 97, 103, 107, 110, 113, 114, 133–5, 142 Peleus 69, 80–1, 261, 264, 266–7, 272, 273, 274 worshipped on Icos 266 n. 25 Pelias daughters of, 117 Pelopids, “future woes” of 248 n. 75, 257
General Index Peloponnese, first Athenian campaign in 137 Penelope 251 n. 6 pentakosiomedimnoi 162 n. 77 Pentheus 54, 57, 277–87 Pericleidas 136–7, 139 n. 88 Pericles 125, 127, 128, 129 choregos (472) 126 and theoric distributions 141 Persephone 192 Perseus 112–13 Persian War 126 Persians absolute monarchy of 280 hair-styles 197–8 in Xenophon 57–8 see also Aeschylus (Persians) Persuasion (Peitho), companion of Aphrodite 110, 113, 115 Phaedra 15, 227–8 nurse of 73 Pherecrates Cheiron 199 n. 20 Philocleon 123 Philocles 196 n. 7 Pandionis 7 n. 18, 15 n. 16 Philoctetes 259–60, 263, 275–6 see also Sophocles (Philoctetes) çºÆ 202–8 Phocion 57 n. 34, 58 Phocus (son of Phocion) 58 Phoenix 69, 272, 273 Phrygia 281 Phrynichus (tragic dramatist) 132 Danaid plays 199 n. 21 Dikaioi/Persai/Synthokoi 126 n. 41 Phoenissae 126 n. 41 Phrynis 21 n. 31 Phthia 68–70 Pieria 284 Plato on Aeschylus’ Niobe 70 play titles cited by 13 Young Pioneer Corps 58 pleasures, listing of 285 Pleisthenes 235 pollution 37, 110, 114, 145, 217, 222, 242 n. 48, 268
339
Polybus 214 Polycleitus 268 n. 30 Polydorus (son of Priam) 268 Polygnotus 268 Polyidus 75 Polymestor 269 Polyneices 63, 74, 82–8, 226 n. 7 Polyphrasmon Lycurgeia 7 n. 18, 16 n. 15 Polyxena, sacrifice of 261, 263, 267–9 Pompeius Magnus, Cn. 155 nn. 42, 48 Poseidon 198 n. 14 Pothos see Desire Priam 173 n. 8, 261 n. 8, 263, 266, 268 death of 150 n. 27, 257, 259, 261, 263, 272, 274 proagon 21, 24 n. 36, 29 prologues 43, 66, 98, 103, 111, 173, 183 n. 23, 231–2, 282, 283 double 108 in comedy 24–5 Prometheus 183; see also Aeschylus (Prometheus) binding of 40–4 Proteus 76–9 proxenoi 136 Psamathe (Nereid) 77 Punch 17 n. 18 Pylades 106, 193, 204, 205 nn. 14–15, 230, 245–7 Pyrrhus see Neoptolemus ‘right-wing’, defined 125 n. 35 St John’s College, Oxford 61 n. 1 Salamis, battle of 126 satyr-drama 22, 44–5, 70, 76–80, 203 n. 8, 256 n. 29 Scamandrius (archon 510/09) 159 n. 67 Scipio Aemilianus, L. Cornelius 154–6 Scylla (daughter of Nisus) 150 Semele 277, 278, 282 Semnai Theai 132–3 Sempronia (wife of Scipio Aemilianus) 155 n. 48 Seneca Troades 267 Seriphos 74
340
General Index
Sestos 126 n. 40 Seven against Thebes 54, 57, 74 see also Aeschylus (Seven against Thebes); Euripides (Phoenissae, Suppliants); Sophocles (Antigone, Oedipus at Colonus) sex, never simulated on stage 36–7 Shakespeare, William Hamlet 52, 150 n. 28, 259 Measure for Measure 25 Sherlockismus defined 65 Sicilian Expedition 125 n. 34, 131 siesta 150 n. 28 Sirens 140 skene 31, 46, 137, 138–9, 142, 172 n. 4 Skira (festival) 121 sleep 143–63, 239, 276 Dionysus and 284, 285 Smikrines (in Menander’s Aspis) 207 n. 22 Smikrines (in Menander’s Epitrepontes) 272–3 Socrates 129, 283 Solon 146 n. 19, 200 n. 22 sophia 284–5 Sophocles 202–76 Aias Lokros 33 n. 8, 253, 255–6 Ajax 14, 31 n. 3, 34, 41, 46, 131 n. 56, 210, 244 n. 57 Antigone 5, 202–8, 209–10, 219, 254 Atreus or Mykenaiai 270 connected suites of plays 6, 252–8 early career 117 Electra 53, 55, 212, 230–49, 257 Epigoni 254 n. 19 epigrams on 200 n. 22 Erigone 248 n. 75 Eurypylus 264, 265–6 first victory 117, 127 Herakleiskos 254 n. 19 Hermione 68–70, 264–5, 270–3 Manteis or Polyidus 75, 270 Nauplios Katapleon 250–8 Nauplios Pyrkaeus 250–8 Nausicaa or Plyntriai 270 Niobe 35, 81 no known productions in person abroad 119 Odysseus Mainomenos 253, 256
Oedipus at Colonus 23, 63, 82, 221, 226 n. 7, 254, 257 Oedipus Tyrannus 38–9, 213–23, 254, 257 Palamedes 250–8 Pandora or Sphyrokopoi 270 Peleus 81, 264, 266–7 Phaedra 60, 227 n. 8 Philoctetes 14 n. 10, 53–4, 55, 57, 60, 212, 226, 257, 259–60, 265, 275–6 Phthiotides 67–70; identical with Hermione? 270 Polyxena 264–5, 267–9 Scyrians 264, 265 Telepheia 252–3 Theban plays 254 Trachiniae 106, 228–30, 256–7 so¯phroniste¯s (in charge of ephebes) 58–9 Sourvinou-Inwood, Christiane 61 Sparta agoge 49, 57–8 appeals to Athens (462?) 136–7 Athenian anti-democrats appeal to 161 consults oracle of Ammon 98 earthquake at (465/4) 136 n. 74 Plato and 58 tragedies hostile to 130–1, 133, 139 Spermo (daughter of Anius) 250 n. 3 stasis (civil strife) 161 Stenia (festival) 121 Stesichorus 76–7, 78 n. 67, 191, 235 sticks as emergency weapons 38, 39, 40 Stobaeus 29 striking, avoided in tragedy 37–44 Sun 153 supplication 39, 134–6, 193, 204 Sybaris (daughter of Themistocles) 21 n. 32 Syria 281 Tanagra 156, 161–2 Tantalus 70, 72, 73 Tecmessa 31 n. 3 Teiresias 217–19, 220, 221, 222, 278–9, 282–4 accomplice in Laius’ murder? 213–14 Telecleides 12 n. 3 Amphiktyones 183
General Index Telemachus 173 n. 8, 212, 225 n. 4, 271 Telephus 83 n. 3, 266 Temenus, sons of 117 Tenos 149 n. 24 tetralogies and trilogies 6–7, 15–16, 66–7, 80, 89–117, 136 n. 73, 138, 142, 250–8 see also Aeschylus (Oresteia) Teuthrania 83 n. 3, 176 n. 17 Thasos expedition (465/4) 147 n. 20 Theatre of Dionysus admission charge 122–4, 127–9 capacity of 140 Thebes in Euripides’ Bacchae 277–87 plain of 198 Themistocles 132, 147 n. 20, 162 at Argos 133 choregos (476) 126 daughters of 21 n. 32 Theoclymenus (lover of Ismene) 210 Theoclymenus (son of Proteus) 39–40, 77 Theodectes Lynceus 105, 108 Theognis (elegiac poet) 57 n. 33 Theonoe, slave of (?) 39, 45–6 theorikon 122, 123, 128, 141 Theramenes 56 Thersander (son of Polyneices) 83 n. 3 Theseus 204, 227–8 sons of 269; see also Acamas, Demophon, Hippolytus Thesmophoria 31 Thettalus (son of Cimon) 21 n. 32 Thirty (oligarchy of 404–403) 94 Thrasonides (in Menander’s Misoumenos) 18 n. 22 Thucydides (son of Melesias) 125 Thurii 126 n. 37, 140 Thyestes 150 n. 29, 168 thymele 137 time, treatment of action spread over two days 153–4 Timotheus (lyric poet) 21 n. 31 Titans 183 titles of plays 11–29, 90 n. 4, 97, 116–17, 142, 270
341
double or triple 16–20, 26–8 epithets appended to 14–15 ordinal numbers appended to 14 n. 10 used to mislead or mystify 22–5 torture 30, 39, 159 trilogies see tetralogies Troilus 261 n. 5 Trojan War hardships of 144 unnecessary/foolish 79, 167 willed by Zeus 165, 168 Troy, capture of 147–8, 150 n. 27 n’aura pas lieu? 212, 276 predicted 171, 176–7 Tydeus 74, 83, 210 Tyndareos 271, 273 tyrants Aegisthus (and Clytaemestra) 153, 169, 243 n. 53 Danaus 107, 109, 112, 114, 115, 116, 135, 136 Pentheus 280 Zeus 43 Uranus (Heaven) 115, 179 Vellacott, P.H. 213–4 violence on stage 30–46 voices from offstage 143, 246–7 Vulci 84 Wealth (blind god) 25 White, Sir Thomas 61 n. 1 Wilde, Oscar The Importance of Being Earnest 25 ‘women’s doors’ 151–2 Wooden Horse 262 Xanthippus (father of Pericles) 126 Xerxes 126 n. 40, 271 Zeus 148, 153 n. 34 in Agamemnon 164–88 ‘below’ 192, 194 and the birth of Dionysus 277, 284 brought suffering into the world 182–7 can never satisfy everyone 267 and the Danaids 99–102, 114
342
General Index
Zeus (cont.) defied/deposed in comedy 184 n. 25 Hikesios 100, 134, 135 ‘Hymn to’ 164, 166–7, 175, 178–88 made Pandora a thief 185 omens sent by 171–7 as patron of kings 166
‘pretty rotten and pretty stupid’? 168–70 ‘reveres’ the Athenians 187 and Semele 277, 278, 282 as tyrant 43 uncaring? 257 Xenios 75 n. 49, 165, 166