[Andrew Faulkner] the Homeric Hymns Interpretati(BookZa.org)

THE HOMERIC HYMNS

This page intentionally left blank

The HomericHymns

Interpretative Essays

Edited byANDREW FAULKNER

1

3Great 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 KarachiKuala Lumpur Madrid Melbourne Mexico City Nairobi

New Delhi Shanghai Taipei Toronto

With offices in

Argentina Austria Brazil Chile Czech Republic France GreeceGuatemala Hungary Italy Japan Poland Portugal SingaporeSouth Korea Switzerland Thailand Turkey Ukraine Vietnam

Oxford is a registered trade mark of Oxford University Pressin the UK and in certain other countries

Published in the United Statesby Oxford University Press Inc., New York

# Oxford University Press 2011

The moral rights of the authors have been assertedDatabase right Oxford University Press (maker)

First published 2011

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 reproductionoutside 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 coverand 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, IndiaPrinted in Great Britainon acid-free paper by

MPG Books Group, Bodmin and King’s Lynn

ISBN 978–0–19–958903–6

1 3 5 7 9 10 8 6 4 2

Preface

The Homeric Hymns have by no means been ignored in scholarshipin the past century, but a collection of interpretative essays, whichtreats each of the long narrative Hymns individually and also givesattention to the corpus as a whole, is a desideratum (as expressed tome by several colleagues) that has until now never been realized.I decided to pursue this project in earnest during a short sojourn inLyon, France, in June of 2008, after much stimulating conversationwith the participants in a conference on Greek hymns organized byRichard Bouchon, Pascale Brillet-Dubois, and Nadine Le Meur-Weissman. This confirmed in my mind the timeliness of the presentvolume, whose contents reflect a wealth of important scholarship onthe Homeric Hymns in the past twenty-five years and give voice to theextremely valuable work currently being undertaken on these poems.

I am of course enormously grateful to all of the contributors, whohave supported this volume from the time I first wrote to suggest it inthe summer of 2008 and generously agreed to write original essaysexpressly for inclusion here. Throughout the past two years, they haveall patiently dealt with my requests and queries, and delivered work tome in a timely fashion despite extremely busy schedules. I havelearned a great deal from them all. I owe particular thanks to anumber of individuals. Nicholas Richardson, who has taught me anincalculable amount about the Hymns (and much else as well) sinceI went to Oxford in 2001 to begin my doctoral studies under hissupervision, kindly read my introduction and essay on reception andoffered many valuable comments and criticisms. William Furley andOliver Thomas also read my essay with care and offered insightfulsuggestions for improvement. Athanassios Vergados, Cecelia Nobili,and Ivana Petrovic kindly shared with me work either forthcoming orin progress. Megan Campbell gave me useful feedback on aspects ofthe introduction and Andrea Barrales-Hall efficiently double-checkedreferences. I am indebted to Joy Mellor, whose keen eye as a copy-editor saved me from a number of errors. I would also like to thankmy colleagues at the University of Waterloo for their indefatigableencouragement; of these, Craig Hardiman kindly advised me onseveral matters of art history. As ever, I am grateful for the support

and patience of my wife Kate Timmers, who deserves a laudatoryhymn of her own. Finally, I would like to thank the Delegates ofOxford University Press for undertaking to publish this book and allthose at the Press who have helped to see it to completion, inparticular Hilary O’Shea, Dorothy McCarthy, Jenny Wagstaffe,Emma Barber, and Kathleen Fearn; the anonymous referees for thePress also offered much helpful advice and encouragement.

The illustrations of the Exekias Cup and François Vase are repro-duced by permission of the Staatliche Antikensammlungen undGlyptothek in Munich and Nimatallah / Art Resource in New Yorkrespectively; I am indebted in particular to Dr Annette Hojer,Ms Irene Bösel, Dr Gerhard Gruitrooy, and Ms Tricia Smith fortheir help.

The contributions of Dominique Jaillard (Ch. 7) and ClaudeCalame (Ch. 14) were translated by me from the original French inwhich they were submitted. The authors have confirmed the accuracyof the translations, but any deficiencies in the English style of thesechapters must be attributed to me.

A.F.Waterloo, CanadaJune 2010

vi Preface

Contents

List of Illustrations ixNotes on Contributors xiList of Abbreviations xiii

1. Introduction. Modern Scholarship on the HomericHymns: Foundational Issues 1Andrew Faulkner

PART I

2. The First Homeric Hymn to Dionysus 29Martin West

3. The Homeric Hymn to Demeter: Some Central QuestionsRevisited 44Nicholas Richardson

4. The Homeric Hymn to Apollo: The Question of Unity 59Mike Chappell

5. The Homeric Hymn to Hermes: Humour and Epiphany 82Athanassios Vergados

6. An Erotic Aristeia: The Homeric Hymn to Aphroditeand its Relation to the Iliadic Tradition 105Pascale Brillet-Dubois

7. The Seventh Homeric Hymn to Dionysus: An EpiphanicSketch 133Dominique Jaillard

8. The Homeric Hymn to Pan 151Oliver Thomas

PART II

9. The Collection of Homeric Hymns: From the Seventhto the Third Centuries bc 175Andrew Faulkner

10. Homeric and Un-Homeric Hexameter Hymns:A Question of Type 206William D. Furley

11. The Homeric Hymns as Genre 232Jenny Clay

12. Children of Zeus in the Homeric Hymns: GenerationalSuccession 254Nancy Felson

13. The Earliest Phases in the Reception of the HomericHymns 280Gregory Nagy

14. The Homeric Hymns as Poetic Offerings: Musicaland Ritual Relationships with the Gods 334Claude Calame

Works Cited 359Index Locorum 383General Index 397

viii Contents

List of Illustrations

Fig. 1. François Vase, by Kleitias. By Permission of Nimatallah/ArtResource, New York 32

Fig. 2. Dionysus Cup, by Exekias. By Permission of StaatlicheAntikensammlungen und Glyptothek München 134


Notes on Contributors

Pascale Brillet-Dubois is Maître de conférences in Greek languageand literature at the Université Lumière Lyon 2, France.

Claude Calame is Professor of Greek at the Université de Lausanneand Directeur d’études at the École des Hautes Études en SciencesSociales in Paris, France.

Mike Chappell is Lecturer in the Open University in London, Eng-land.

Jenny Clay is William R. Kenan, Jr., Professor of Classics at theUniversity of Virginia, USA.

Andrew Faulkner is Associate Professor of Classics at the Universityof Waterloo, Canada.

Nancy Felson is Emerita Professor of Classics at the University ofGeorgia, USA.

William Furley is Apl. Professor in the Seminar für Klassische Phil-ologie, Universität Heidelberg, Germany and Senior Research Fellow,Institute of Classical Studies, London, England.

Dominique Jaillard is Maître d’enseignement et de recherche at theUniversité de Lausanne, Switzerland.

Gregory Nagy is Francis Jones Professor of Classical Greek Literatureand Professor of Comparative Literature at Harvard University andDirector of the Harvard Center for Hellenic Studies, Washington,USA.

Nicholas Richardson is Emeritus Fellow of Merton College, Oxfordand former Warden of Greyfriars, Oxford, England.

Oliver Thomas is Christopher Tower Junior Research Fellow inGreek Mythology at Christ Church, Oxford, England.

Athanassios Vergados is Lecturer at the University of Athens, Greece.

Martin West is Emeritus Fellow of All Souls College, Oxford,England.

xii Notes on Contributors

List of Abbreviations

ABV J. D. Beazley, Attic Black-Figure Vase-Painters (Oxford, 1956).

AHS T. W. Allen, W. R. Halliday, and E. E. Sikes, The HomericHymns (Oxford, 1936).

Allen Homeri Opera 5, ed. T. W. Allen (Oxford, 1912).

AS T. W. Allen and E. E. Sikes, The Homeric Hymns (Oxford,1904).

Bernabé Poetae Epici Graeci, ed. A. Bernabé (2nd edn.; Leipzig, 1996–2007).

CA Collectanea Alexandrina, ed. J. U. Powell (Oxford, 1925).

Campbell D. A. Campbell,Greek Lyric III: Stesichorus, Ibycus, Simonides,and Others (Loeb Edition, Cambridge, Mass., 1991).

Càssola F. Càssola, Inni omerici (Milan, 1975).

DELG P. Chantraine, Dictionnaire étymologique de la langue grecque:histoire des mots, revised edition (J. Taillardat, O. Masson,and J.-L. Perpillou, eds.), with a supplement ‘Chroniqued’étymologie grecque 1–10’ by A. Blanc, Ch. de Lamberterie,and Jean-Louis Perpillou (Paris, 2009).

DK Die Fragmente der Vorsokratiker, ed. H. Diels and W. Kranz(6th edn.; Berlin, 1952).

Dr Scholia in Pindarum, ed. A. B. Drachmann, 3 vols. (Lepizig,1903–27).

E.-K. Posidonius: the Fragments, ed. L. Edelstein and I. G. Kidd,vol. 1 (2nd edn.; Cambridge, 1989).

Erbse H. Erbse, Untersuchungen zu den attizistischen Lexika (Berlin,1950).

Farnell L. R. Farnell, Cults of the Greek States, 5 vols. (Oxford,1896–1909).

FGrH Die Fragmente der griechischenHistoriker, ed. F. Jacoby (Berlin/Leiden, 1923–58).

Foerster Libanii Opera, ed. R. Foerster, 12 vols. (Leipzig, 1903–27).

IEG Iambi et Elegi Graeci (ed. altera), ed. M. L. West, 2 vols.(Oxford, 1989–92).

IG II2 Inscriptiones Graecae: Inscriptiones Atticae Euclidis annoposteriores (2nd edn.; Berlin, 1913–40).

Keil Aelii Aristidis Smyrnaei quae supersunt omnia, 2 vols. (Berlin,1898).

LfgrE Lexikon des frühgriechischen Epos, ed. B. Snell et al. (Göttingen,1955–2010).

LIMC Lexicon Iconographicum Mythologiae Classicae, 8 vols. (Zurichand Munich, 1981–99).

Lobel-Page Poetarum Lesbiorum Fragmenta, ed. E. Lobel and D. L. Page(Oxford, 1955).

LSJ H. G. Liddell, R. Scott, H. Stuart Jones, R. McKenzie, andP.G.W.Glare,Greek–English Lexicon, with a Revised Supplement(9th edn.; Oxford, 1996).

M–W Fragmenta Hesiodea, ed. R. Merkelbach and M. L. West(Oxford, 1967).

OC A. Heubeck, S. West, A. Hoekstra, J. B. Hainsworth, J. Russo,and M. Fernández-Galiano, A Commentary on Homer’sOdyssey, 3 vols. (Oxford, 1988–92).

OLD Oxford Latin Dictionary, ed. P. G. W. Glare (Oxford, 1982).

PCG Poetae Comici Graeci, ed. R. Kassel and C. Austin (Berlin,1983– ).

Pertusi Scholia Vetera in Hesiodi Opera et Dies, ed. A. Pertusi (Milan,1955).

Pf. Callimachus, ed. R. Pfeiffer (Oxford, 1949–53).

PMG Poetae Melici Graeci, ed. D. L. Page (Oxford, 1962).

PW H. W. Parke and D. E. W. Wormell, The Delphic Oracle, 2vols. (Oxford, 1956).

Rose Aristotelis qui ferebantur librorum fragmenta, ed. V. Rose(3rd edn.; Leipzig, 1886).

Schober A. Schober, ‘Philodemi De Pietate pars prior’ (CronErc 18:67–125. 1988).

SEG Supplementum Epigraphicum Graecum, ed. J. J. E. Hondiuset al. (Leiden, 1923–75; Amsterdam, 1978–2006; Leiden, 2007–).

Semitelos —Ø��æ�ı �å�ºØÆ —Æ�ØÆŒ�, ed. D. Semitelos (Athens, 1875).

SH Supplementum Hellenisticum, ed. H. Lloyd-Jones and P. Parsons(Berlin, 1983).

SLG Supplementum Lyricis Graecis, ed. D. Page (Oxford, 1974).

xiv List of Abbreviations

SM Bacchylidis Carmina cum Fragmentis, ed. B. Snell andH. Maehler (Leipzig, 1992).

SnM Pindari Carmina cum Fragmentis, ed. B. Snell and H. Maehler(Leipzig, 1987–9).

TrGF Tragicorum Graecorum Fragmenta, Göttingen, ed. B. Snell,R. Kanicht, and S. Radt, 5 vols. (Göttingen, 1971–2004).

Vian F. Vian and E. Delage, Apollonios de Rhodes, 3 vols. (Paris,1974–96).

West M. L. West, Greek Epic Fragments (Loeb Edition, Cambridge,Mass., 2003).

WL M. L. West, The Homeric Hymns, Homeric Apocrypha, Lives ofHomer (Loeb Edition, Cambridge, Mass., 2003).

Abbreviations of ancient authors and their works are for the most part thoseof LSJ and OLD. The Homeric Hymns are abbreviated as Dionysus (1),Demeter (2), Apollo (3), Hermes (4), Aphrodite (5), and then Hymn(s) withnumbers 6–33; or Dion., Dem., Apoll., Herm., Aphr., and then Hy(s). Journaltitles are abbreviated as in L’Année philologique. The spelling of ancientGreek names does not follow one system; where common Latinate formsexist, these have been used, but transliterated Greek forms are also employedin many instances. The choice between bc/ad and bce/ce varies according topreference of the individual author.

List of Abbreviations xv


1

Introduction

Modern Scholarship on the Homeric Hymns:Foundational Issues

Andrew Faulkner

1. BEGINNINGS TO THE EARLYTWENTIETH CENTURY

The Homeric Hymns are a collection of thirty-three hexameterhymns, in antiquity often attributed to Homer but in fact of variedauthorship, provenance, and date.1 Many of the poems are shortcelebrations of a god’s or goddess’s powers and attributes, rangingfrom three to twenty-two lines, but a few (2–5, 7, 19) containextended narrative sections that recount the birth of a deity or animportant event in his or her life and stretch from forty-nine (19, toPan) to almost six hundred lines (4, to Hermes) in length. Modernscholarship on the Hymns has its origins in the fifteenth century,when in 1488 Demetrius Chalcondyles published the first printededition of the Hymns in Florence along with the Iliad and Odyssey.In establishing his text, Chalcondyles compared manuscript readingscritically and often demonstrated admirable editorial skill.2 This

1 For more on the formation of the collection, see in this volume Faulkner(pp. 175–81).

2 This edition seems to have been based upon a manuscript of the f family(probably D), with use made also of a manuscript of the x family (possibly L) aswell as one of the p family. Chalcondyles also appears to have introduced a number ofhis own conjectures. See further Allen (1895a), 154–60, Càssola 612–13.

edition included what are now numbered Hymns 3–33 in the collec-tion, incorporating the long narrative poems to Apollo (3), Hermes(4), and Aphrodite (5). Subsequent scholars gave critical attention tothe language of this corpus in the sixteenth and early seventeenthcenturies, the most significant study being that of the French scholarBernard Martin.3 The next major advance, however, in the study ofthe Hymns came in the eighteenth century. The British scholar JoshuaBarnes published an edition of the Hymns in 1711, but the mostimportant step forward came in the second half of the century whenthe Dutch scholar David Ruhnken turned his critical attention to thetexts and undertook a systematic comparison of manuscript read-ings.4 A significant stimulus for Ruhnken’s work on the Hymns camein the late eighteenth century with the uncovering of the famousmanuscript M in Moscow in 1777 by C. F. Matthaei, who subse-quently sold the codex to Leiden. This manuscript, whose first thirtyfolia are missing, contains at its outset the final twelve lines of what isnow known as the first Hymn to Dionysus, another Hymn thatoriginally contained an extended narrative in its middle section, anda largely intact text of the second long Hymn to Demeter, which isfollowed by Hymns 3–28. 4.5 Apart from a few more recent papyrusfinds and ancient testimonia, M remains the sole surviving witness ofthe first two Hymns.6

Following the rediscovery of these poems, there was a flourish ofscholarship on the Hymns at the turn of the nineteenth century, ascomplete editions together with critical commentaries appearedin quick succession.7 Apart from matters of textual criticism andlanguage, studies of this period gave attention to broader questionssuch as the authorship and performance of the Hymns, as well asthe structure and unity of the longer narrative poems, and many ofthese early views have remained influential to the present day.

3 Martin (1605). For more extensive surveys of early modern editions and philo-logical work on the Hymns before the twentieth century, see AHS cx–cxv, Càssola612–23. The first Latin translation of the Hymns was produced by Georgius Dartonain the sixteenth century.

4 Barnes (1711), Ruhnken (1782). The latter compared manuscripts B, C, and M;see Càssola 619.

5 On the M manuscript, see Gelzer (1994).6 On the text of Dion. see West (2001a), and Ch. 2 in this volume. P. Oxy. 2379 and

P. Berol. 13044 contain fragments of Dem.; see Richardson (1974), 66–7.7 Note in particular Ilgen (1796), Matthiae (1805), Hermann (1806), Wolf (1807).

2 Andrew Faulkner

F. A. Wolf ’s arguments, for example, on the proemial function of theHymns are still widely accepted,8 while Ruhnken’s suggestion that theHymn to Apollo might be composed of originally separate parts pavedthe way for separatist views of that poem, which became dominant inthe nineteenth century in accordance with the broader popularity atthat time of analytical readings of Homer.9

The nineteenth century saw a great deal more work completed onthe manuscript tradition of the Hymns and the establishment of thetext. Subsequent to an important edition of the corpus by A. Bau-meister in 1860, which took into account manuscripts of the ¨ familyfor the first time since Chalcondyles in the fifteenth century, thereappeared in 1886 editions of the Hymns by both E. Abel andA. Gemoll, who considered readings of a number of new manuscriptsof the p family, as well as a study of the manuscript transmission byH. Hollander.10 Then at the end of the century, following his editorialrole in the posthumous publication of A. Goodwin’s Oxford edition,T. W. Allen published a series of important articles on the text of theHymns, work which would lead to the publication of his Oxfordedition with commentary together with E. J. Sikes in the first yearsof the twentieth century.11 This work, which was substantially revisedin 1936 with the collaboration of W. R. Halliday, as well as Allen’sOxford Classical Text, provided the foundation for much of twenti-eth-century scholarship on the Hymns and met no equal until theedition and commentary of Càssola in 1975.12

2. ORAL POETICS

Starting in the early twentieth century, the study of Homeric poetrybegan to undergo a transformation due to an increased appreciationof the role played by oral composition in the development of early

8 See below pp. 17–19.9 On the still debated question of Apoll.’s unity, see Chappell in this volume

(Ch. 4).10 Baumeister (1860), Gemoll (1886), Abel (1886), Hollander (1886).11 Goodwin (1893), Allen (1895a, b; 1897a, b; 1898), AS (1904). Allen was

also largely responsible for the text of the Hymns in Monro (1896).12 Allen (1912), AHS (1936), Càssola. Also of note is the Budé edition of Humbert

(1936).

Introduction 3

Greek hexameter poetry. As is well known to anyone who has workedon Homer, this critical shift was inspired largely by the work ofMilman Parry in the 1920s and 1930s, who demonstrated the for-mulaic, traditionally oral quality of Homeric poetry on the levels ofboth versification and theme.13 The formulaic nature of early hexam-eter poetry was for Parry a sign not just of development from an oraltradition, but of oral composition in performance by Homer, a factwhich he and Albert Lord sought to demonstrate through theirgroundbreaking comparisons with South Slavic epic and other oralpoetic traditions.14

This view naturally had implications also for the Homeric Hymns.Parry himself was aware of the comparative evidence of the Hymnsand the Epic Cycle for the formularism of the Homeric epics, but didnot make extensive use of them for the reason that ‘these poems andhymns belong to different periods and clearly do not all follow thetradition with equal fidelity.’15 Others, however, have subsequentlyargued that some of the Homeric Hymns are oral compositionsbecause of their employment of formular language and style.16 Thisis possible, but the view that the Hymns are entirely oral compositionshas not gone unchallenged.17 In a first instance, it is ultimatelyimpossible to prove from their language that the Hymns, in muchthe same form as we have received them, were entirely oral composi-tions: poets who made use of writing might well have imitated tradi-tional formulaic diction, such that the quantity and quality offormulae in a given composition are no sure indicator of oral com-position, only congruity with oral composition.18 This is true even ofthe Homeric epics themselves, which, although clearly the product ofa poet or poets steeped in the tradition of oral composition andperformance, will have reached the fixed form in which we havethem through the process of being written down.19 One theory is

13 His writings are collected in Parry (1971). For a more detailed overview ofscholarship on Homer and oral poetics, see Foley (1997).

14 See in particular Lord (1960).15 Parry (1971), 4.16 See e.g. Notopoulos (1962), Preziosi (1966), Pavese (1972); cf. Postlethwaite

(1979).17 Kirk (1966) rejected Notopoulos’ method for determining orality in the Hymns

on the grounds that quantity of formulae is not in itself a sign of orality. Kirk’smore subjective analysis is in turn rejected by Lord (1968).

18 See Janko (1982), 18–19, 40–1; cf. Richardson (1974), 31, Clay (1997), 491–2.19 See West (2001b), 3–4. For a defence of the intermediate position that the

Homeric epics are ‘oral-derived traditional texts’, see Foley (1997), 163–4.

4 Andrew Faulkner

that the poems are oral dictated texts recorded by a scribe during oralperformance,20 but the poet or poets of the Homeric epics mightalternatively have taken a more active role in the writing of the poemsover an extended period of years, a process involving changes to thewritten text as it evolved.21 Scholars since the time of Parry have alsorecognized that it is reductive to distinguish sharply between oral andliterate composition, when various intermediate stages between acreative oral tradition and literate composition can be imagined.22

Literate and oral traditions can coexist,23 but one must assume thatwriting a poem down will inherently promote its fixity, while con-tinued creative oral composition will further its multiformity. Thedemonstrable uniformity of a poem in contrast to its multiformity,the latter a sign of continued creative oral composition, therefore haswide implications for how one treats its transmission at differentpoints in antiquity. In the case of the Homeric epics, opinions con-tinue to differ concerning the extent to which genuinely creative oralcomposition persisted in effecting distinctly variant forms of the Iliadand Odyssey. On the one hand, it is maintained that a gradual processof fixing the Homeric epics continued down to the second century BC,until which point oral performance produced multiforms of the Iliadand Odyssey, which are reflected in the variants of the textual tradi-tion.24 On the other hand, it is argued that the Homeric epics seem tohave been largely fixed by the end of the seventh century BC, or atlatest the mid-sixth century BC: in comparison to oral epics of othercultures such as the Chanson de Roland, whose manuscript traditionpreserves alternate versions that vary by thousands of lines, theevidence for variation in the tradition of the Iliad and Odyssey isminimal, for the most part restricted to occasional additional linesand small verbal changes of little narrative import.25 On this view, thecontinuous tradition of oral performance of Homer in the Classical

20 See Janko (1998); as he points out (7), this would not preclude premeditation.21 See West (2000), contra the theory of oral dictated texts.22 See Richardson (1974), 337–8, Janko (1982), 41, Foley (1997), 162–4. Lord (1995),

212–37 himself later accepted an overlap between written and oral compositions.23 See e.g. Thomas (1992).24 See Nagy (1996a), esp. 107–206, with earlier bibliography. He posits five succes-

sive stages, ‘showing progressively less fluidity and more rigidity’ (109–10).25 See Finkelberg (2000), who provides an excellent overview of this debate with

earlier bibliography. On the Doloneia as the one major interpolation in the Iliad,albeit part of the poem by the end of the seventh century BC, see West (2001b), 10–11.

Introduction 5

and Hellenistic periods still produced minor oral variants, as arevisible in the manuscript tradition, but aimed at reproduction of thefixed poems rather than creative oral composition.26

The date at which one considers the Homeric epics to have been inessence fixed naturally affects how one treats the Homeric Hymns andother archaic hexameter poetry. Some, for example, insist that it is notpossible to speak of direct imitation of the Homeric epics in theHymns.27 One must certainly be cautious not to assume direct imita-tion at every turn, but if the Homeric epics were fixed by the seventhcentury BC or earlier and many of the Hymns as we have them seem todate to a period later than this, direct imitation of a fixed poem is adistinct possibility with which one must reckon.

As for the fixity of the Homeric Hymns themselves, approachescontinue to diverge along similar lines. One might not expect theHymns to have been as readily fixed as the Homeric epics. It has beenshown, for example, that the tradition of the Cypria, a poem of theEpic Cycle which deals with the beginning of the Trojan War, displaysgreater multiformity than the Homeric epics,28 a reminder that theIliad and Odyssey held a special place in the tradition. Earlier versionsof hexameter hymns similar to our Hymns are perhaps reflected in thecorpus of hymns by the Lesbian poet Alcaeus at the turn of the sixthcentury BC,29 and Neoanalysis can potentially identify the interactionof the narratives recounted in the Hymns with the Homeric epics atearlier stages of the oral tradition.30 At the same time, there isevidence that several of the longer Hymns became fixed by relativelyearly dates, and it seems reasonable to consider the possibility ofdirect imitation of the Hymns as fixed poems, even amongst theHymns themselves, from the seventh century onward.31 Anotherimportant consideration in the case of the Hymns is that they cannot

26 See Pelliccia (2003).27 See recently the review of my commentary on Aphr. by Bartol (2010), 289–90:

‘his thinking of these poetic products in terms of interrelations between fixed texts isvery difficult to sustain with reference to oral culture’.

28 See Finkelberg (2000), 6–11.29 See West (2002a), 216–17, and in this volume (pp. 39–40); cf. Faulkner

(pp. 200–1).30 See in this volume Brillet-Dubois (Ch. 6). West (Ch. 2) argues for the influence

of Dion. itself on Homer, dating the Hymn possibly to 650 BC.31 For two different, but not mutually exclusive, approaches to the reception and

transmission of the Hymns, which give different emphasis to their fluidity in the sixthand fifth centuries, see in this volume Faulkner (Ch. 9) and Nagy (Ch. 13).

6 Andrew Faulkner

be treated as a unified body. As will be considered in the next section,their use of traditional epic language is not consistent, and it is clearthat some of the poems in the collection date to the fifth century BC orlater.

3 . DATING AND LANGUAGE

The Homeric Hymns have often been assumed to be later composi-tions than both the Iliad and Odyssey and the Hesiodic Theogony andWorks and Days. In many instances, this is clearly the case, but onemust nonetheless be cautious not to assume a priori that the Hymnsare post-Homeric and/or post-Hesiodic. Despite some internal mar-kers that help to establish the dates of the poems, a chief method fordetermining the relative dates of the Hymns has been linguisticcomparison. Several important investigations of the Hymns’ languageundertaken in the past fifty years have in this way attempted to datethe poems in relation to other early hexameter poetry. The modifica-tion of traditional formulae in the Hymns was notably linked toestablished linguistic developments by Hoekstra, who demonstratedon these grounds that the language of Demeter, Apollo, and Aphroditeis at a more advanced stage of the formulaic epic tradition than theIliad and Odyssey, a sign for him also of the poems’ chronologicalrelationship.32 But the most detailed and inclusive study of languageis that of Janko,33 who provides a thorough statistical analysis of awide range of linguistic criteria, including the use of digamma andnu-mobile, as well as a number of alternative older and youngermorphs (such as the older o-stem genitive -�Ø� in comparison to themore advanced contracted form -�ı) in the Homeric epics, theHesiodic corpus, and the Hymns; his statistical results are combinedwith careful evaluation of parallel passages, instances of modification,and other evidence for dating in each of the long Hymns. Just as forHoekstra, his findings place all of the long Hymns later than the

32 Hoekstra (1969), who excludes Herm. His very useful study was unfortunatelyaccompanied by the view that increased modification of formulae indicated inferiority.

33 Janko (1982), who provides (7–16) an excellent review of earlier linguisticstudies and attempts at dating. This remains a widely consulted and indispensabletool for the study of the Hymns.

Introduction 7

Homeric epics, which he dates to the second half of the eighthcentury. Also, with the exception of Aphrodite and the Delian sectionof Apollo, which he cautiously suggests are contemporary with Hes-iod in the first half of the seventh century, Janko dates the longHymns later than the Theogony and Works and Days.

Not everyone, however, has agreed that linguistic criteria are anaccurate means of dating early hexameter poetry. Prior to Janko, thesupposed oral composition of the Hymns led Notopoulos to rejectcomparison of linguistic criteria altogether as a tool for dating.34 Fewwill agree with excluding linguistic criteria as a means of dating to theextent that he proposes, but some of his cautions against such anapproach are nonetheless salutary and reflected in later scholarship.Pavese subsequently argued that differences in language can beexplained by geographical and stylistic variation rather than chron-ological disparity, as part of his contention that there existed distinctmainland and Ionic traditions of early Greek poetry.35 Janko, whosestudy in part responds to Pavese, is well aware of the possibility thatgeographical and stylistic factors may have played a role. He con-siders regional variation while refuting Pavese’s more extreme divi-sions, but argues strongly that linguistic criteria are determinants ofdate. Yet, more recently, arguments have again been made against thechronological significance of linguistic criteria: ‘the major determin-ant of the quantity of younger forms in a given poet is the extent towhich his language diverges from the formulaic, and this depends onmany other factors apart from his date.’36 In the end, it is impossibleto draw firm conclusions on this issue, but one might reasonablyaccept linguistic criteria as a valuable tool for evaluating the Hymns’place in the tradition, without assuming that they provide absoluteindicators of relative date.

Below is a brief survey of the evidence for dating the individuallong Hymns, as well as some of the mid-length and shorter Hymns inthe collection for which it is possible to say something about date.When considering the relative dating of the Hymns, one must keep inmind that the dates given to Homer and Hesiod are themselvesuncertain. The Homeric epics have traditionally been dated to the

34 Notopoulos (1962), 362–5.35 Pavese (1972), 111–65.36 West (1995), 204–5, who, contra Janko, places Hesiod before Homer; see next

paragraph.

8 Andrew Faulkner

second half of the eighth century, prior to Hesiod. In the Works andDays, Hesiod claims to have taken part in the funeral games ofAmphidamas, where he may have performed the Theogony,37 andAmphidamas is said by later sources to have been killed in theLelantine war. The date of this war is uncertain but it will havetaken place sometime in the late eighth or early seventh centuries;this, combined with possible cases of imitatio of Hesiod by early lyricpoets, dates Hesiod at a lower extreme to the first half of the seventhcentury.38 As regards Homer, West has alternatively maintained inrecent years that the Iliad and Odyssey post-date Hesiod and thereforebelong to the middle to second half of the seventh century.39 Inmaking this claim, he refutes earlier attempts to show that Hesiodimitates passages in Homer, and rejects the conclusions of Janko’scomparative linguistic studies. This would push down the dates ofseveral Hymns. However, it remains uncertain whether Homershould in fact be dated later than Hesiod and one must remainopen to the possibility of an eighth century date for the Iliad andOdyssey.40 In turning to the dates of the Hymns, it should be empha-sized from the outset that there is very little certainty in these matters,as is indicated by the variety of views presented below:

First Hymn to Dionysus

The fragmentary nature of this poem, which will originally haveextended to over 400 lines,41 makes dating particularly challenging.Comparison of linguistic criteria is invalidated by the paucity of data.However, it may possibly be one of the oldest Hymns in the collection.Based upon fragments A and D, Allen, Halliday, and Sikes remarkedthat, ‘[t]here is nothing either mythological or linguistic in the frag-ments of this hymn which suggests a late period.’ With the additionof fragment C (P. Oxy. 670), which recounts Hera’s binding byHephaestus, West suggests in this volume that, while the myth

37 Op. 654–9. See West (1966), 44–5.38 See Janko (1982), 94–8, 228–31.39 See in particular West (1966), 40–8, (1995), and (2003), 12 n. 56, for a list of

other scholars in agreement with a seventh-century date.40 For a recent defence of an eighth-century date for Homer, see Lane Fox (2008),

381–4.41 See West (2001a), 1.

Introduction 9

could have existed earlier than our poem, the Hymn may itself haveinfluenced the Iliad and therefore go back to 650 BC or earlier.42 Twopoints of vocabulary in fragment C (P. Oxy. 670) are otherwise notattested before the fourth century,43 although this is by no means acertain indication of a late date. Dihle, however, has argued that bothit and the other fragments of the poem are Hellenistic.44

Hymn to Demeter

This Hymn is in character the most Hesiodic of the corpus. It isprobable that the poem draws directly upon the Theogony, as well asthe Homeric epics.45 A terminus post quem also seems to be estab-lished by the Hymn to Aphrodite, with which the poem shares anumber of parallels that together point to a direct relationship; thebalance of the evidence suggesting that Demeter is secondary.46 If it isaccepted that Demeter was influenced by these poems, the Hymn atthe earliest belongs to the latter half of the seventh century. On theother end, a terminus ante quem of the mid-sixth century, the timewhen Athens took control of, or at least renewed interest in, theMysteries at Eleusis, has reasonably been suggested on the groundsthat Demeter does not mention Athens or show an interest in myth-ology that from this time onward became prominent in Attica.47 Noteveryone agrees that this is a reliable criterion,48 but a date in the lateseventh or early sixth century will not be far wrong.

42 See in this volume West (Ch. 2).43 See Faulkner (2010a). It is possible that P. Oxy. 670 is late and should not be

assigned to Dion., even if the Hymn preserved in the Leiden manuscript was earlyand originally recounted the story of Hera’s binding.

44 Dihle (2002).45 See Richardson (1974), 30–41, Janko (1982), 181–3.46 See Richardson (1974), 42–3, Janko (1982), 163–5, Faulkner (2008), 38–40.47 See Richardson (1974), 5–11. This assumes that the poem was composed for

performance in Eleusis.48 Clinton (1986) has argued that the lack of reference to Athens in the poem is an

indication that the poet was not from Attica and unfamiliar with Eleusis; for criticismof this position, see Richardson in this volume (pp. 51–2). Foley (1994), 169–78, onthe other hand suggests that the Panhellenic nature of the Hymn explains the poem’ssilence concerning Athens. Neither of these, however, argues for a late date.

10 Andrew Faulkner

Hymn to Apollo

The dating of this poem is particularly contentious. The two sectionsof the Hymn, the Delian (1–181) and the Pythian (182–546), haveoften been considered originally separate compositions,49 and there-fore given different dates. Amajority has taken the Delian section to beolder, and Janko argues on the basis of his linguistic analysis that thissection should have a date roughly contemporary with Hesiod.50

Alternatively, West has maintained that the Pythian section is earlierand that the Delian part was modelled upon it in the sixth century.51

The Pythian section has been dated by some to the 580s BC because ofApollo’s prophecy in lines 540–3 that his priests will be ruled by othersif they do not behave. This could refer post eventum to the First SacredWar (c.595–585 BC), which gave control of the Delphic oracle to theAmphictyonic league.52 Amongst the arguments for the Hymn’s ori-ginal disunity is also a theory for its eventual unification in the secondhalf of the sixth century. A scholium to Pindar’s second Nemeaninforms us that a Chian poet Cynaethus composed the Hymn toApollo, and it has been suggested that this Cynaethus may be respon-sible for the combined Hymn, which he performed in 523/2 BC at thefestival of Delian and Pythian Apollo celebrated on Delos by theSamian tyrant Polycrates.53 However, against a date in the late sixthcentury may be the fact that Thucydides (3. 104) speaks of the Deliansection of the Hymn as unique evidence for poetic contests at thefestival in the distant past, which were discontinued in his own time;but he has detailed knowledge of Peisistratus’ purification of Delos andPolycrates’ dedication of Rheneia to Delian Apollo, which couldsuggest that the discontinuation of poetic contests came earlier.54

49 Against Apollo’s original unity, see Chappell in this volume (Ch. 4). For argu-ments in favour of unity, see the next paragraph.

50 Janko (1982), 99–115.51 West (1975), WL 10–12.52 See Janko (1982), 119–21, 127–8, Richardson (2010), 13–15.53 For more on this view, see in this volume esp. Chappell (pp. 71–3), Nagy

(pp. 288–91), with bibliography. Some suggest that Cynaethus combined existinghymns, but West (WL 9–12) has argued that Cynaethus actually wrote the Delianhymn and then combined it with the Pythian hymn for Polycrates’ festival. Heproposes that Cynaethus added the Hera–Typhaon section to the earlier Pythianhymn to please Polycrates, as Hera was an important deity on Samos.

54 See Richardson (2010), 14, also on the lack of any mention of the Pythia in thepoem.

Introduction 11

It must also be emphasized that not all scholars agree that theHymn is composed of two originally separate parts. Both Miller andClay have argued for the Hymn’s original unity on the grounds ofthematic coherence, and Clay’s arguments convinced Janko to changehis mind and adopt a unitarian position in his review of her book.55

Clay also argues against the view that Apollo’s prophecy at 540–3should be linked to the First Sacred War; in this case, the unifiedHymn could go back to the first half of the seventh century, followingJanko’s original dating of the Delian section.56 However, the Hymn’sdescription of the construction of Apollo’s magnificent stone templeat Delphi (285–99) might suggest a date after 650 BC, given that ourcurrent evidence suggests that the first stone temple at Delphi be-longed to the second half of the seventh century.57 If one werealternatively to accept both unity and the link to the First SacredWar, the whole Hymn would date to the 580s.

In the end, one can date Apollo with reasonable confidencebetween the first half of the seventh and the second half of the sixthcentury, but it is difficult to be more precise with any certainty. A datein the early to middle sixth century BC seems attractive; it may benoteworthy that knowledge of the myth recounted in Apollo is notreflected in Alcaeus’ hymn to Apollo, whereas narrative elements ofDionysus, Hermes, and Hymn 33 to the Dioscuri are echoed in hiscorpus of hymns.58 But much depends upon whether one viewsApollo as an originally unified poem, and the weight given to theconnection of Apollo’s prophecy in lines 540–3 to the First SacredWar.

Hymn to Hermes

The poem is universally acknowledged to be the latest of the longHomeric Hymns and its language is certainly the least traditional;59

55 Miller (1979), (1986), Clay (1989), Janko (1991). The Hymn transmitted to uswas at least intended to be received as a unity.

56 See Clay (1989), 87–92, Janko (1991), 13. AHS 184–6 date the poem to the lateeighth century.

57 See Richardson (2010), 14.58 See West (2002a), 216–17, and in this volume West (pp. 39–40), Faulkner

(pp. 200–1).59 See Janko (1982), 133–50.

12 Andrew Faulkner

composition before the sixth century seems unlikely. Nonetheless,there is much debate about the poem’s precise date. A number ofparallels strongly suggest the influence of the unified Apollo onHermes.60 But as we have seen above, the date of this poem is fraughtwith uncertainties and therefore does not provide a secure terminuspost quem. Görgemanns proposed that Hermes’ use of rhetoric,including argumentation based upon probability (265–73), and refer-ences to music as a technē (�å�Å) requiring ‘practice’ (�æ�� ; 447–8),are aligned with sophistic thought and could indicate a date in thefifth century.61 The difficulty with such criteria is the indeterminacyof how early these rhetorical techniques and ideas developed; theirpresence in the Hymn therefore does not exclude a date before thefifth century.62 Artistic evidence has also been used in favour of a datein the sixth century. The syrinx, which is invented by Hermes in lines511–12, is transferred from Hermes to Pan in art after the turn of thesixth century, and the motif of Hermes as an infant cattle thief seemsto have been particularly popular between c. 565 and 490 BC.63 Onemight not, however, wish to posit too strict a correspondence betweenart and literature.64 Finally, it has been suggested that Hermes’reference to Apollo’s Delphic temple as an attractive and challengingmark for theft (177–81) might exclude the period c.548–505 BC, whenthe site is known to have been under construction;65 this is possible,although such an honorific formulation need not exclude contem-porary upheaval. Ultimately, a late-sixth-century date for Hermesseems most attractive, but the Hymn could also belong to the firsthalf of the fifth century.

60 See Richardson (2007), 89–91, (2010), 20–1, Thomas (2009), 289–94.61 Görgemanns (1976). On these and other arguments put forth for dating Herm.

in the fifth century, see Janko (1982), 142, who himself prefers a late-sixth-centurydate.

62 See Richardson (2010), 21. Nobili (2008) argues for a date c.510–500 (sheattempts to tie Herm. to Athens and suggests performance at the Panathenaea orsymposion, in part on the grounds that Herm. privileges the themes of sacrifice,communal banqueting, and music. Performance in Athens is certainly possible, butthe Hymn’s themes of sacrifice, communal banqueting, and music could apply to anumber of performance contexts. On Herm.’s possible links to Olympia, see belowpp. 21–2). Athanassios Vergados, who kindly showed me his commentary on Herm.in advance of publication, also argues that these concepts may have existed earlier andprefers a date in the second half of the sixth century.

63 See Nobili (2008), 184–9, Richardson (2010), 24.64 See Thomas (2009), 28 contra this as a secure tool for dating.65 Thomas (2009), 25.

Introduction 13

Hymn to Aphrodite

Sometimes called the most Homeric of the Hymns, the Hymn toAphrodite is one of the oldest in the collection. Aphrodite’s prophecyto Anchises (197–8) is very similar to Poseidon’s prophecy about thefuture of Aeneas’ race in the Iliad (20. 307–8), and the uniquecharacter of the passages makes a common formulaic source improb-able. Reinhardt argued that the Hymn was in fact composed by thepoet of the Iliad, but its language is almost certainly later than both ofthe Homeric epics, and is very probably also later than both theTheogony and the Works and Days.66 The prophecies nonethelesssuggest a close link between the Iliad and the Hymn: both may havebeen intended to honour a family in the Troad who claimed descentfrom Aeneas, although this theory has not gone unchallenged.67 Onecannot exclude a late eighth-century date, but assuming that theHymn is later than both the Theogony and the Works and Days, itwill most probably have been composed no earlier than the middle ofthe seventh century. As discussed above, a terminus ante quem isprovided by Demeter, which itself seems to belong to the late seventhor early sixth century.

Hymn 7 to Dionysus and Hymn 19 to Pan

The lack of tangible historical references or allusions in the seventhHymn to Dionysus make it difficult to date this Hymn with anyprecision, although its language and straightforward style are consis-tent with a relatively early but post-Homeric date.68 Attempts havebeen made to link the mention of the Tyrsenian pirates (7–8) to theEtruscans and to locate the Hymn in late-seventh-century Corinth.69

However, such associations are at best reflected obliquely in the poemand one might rather emphasize the Panhellenic, temporally generic

66 Reinhardt (1956). On Aphr.’s relationship to Homer and Hesiod, see Faulkner(2008), 26–38. Janko (1982), 165–9 has alternatively suggested that the poem wascomposed between Hesiod’s Th. and Op.

67 See Faulkner (2008), 3–10.68 See AHS 379–80, Janko (1982), 183–4. Càssola 287–8, who reviews earlier

scholarship, concludes that the poem is undatable.69 See recently Nobili (2009).

14 Andrew Faulkner

nature of the narrative.70 The fact that the myth of Dionysus’ abduc-tion is elsewhere first attested in Pindar (fr. 236) has suggested a datein the late sixth or early fifth century, which is perhaps also supportedby the possible reflection of the myth on the Exekias cup of c.530,71

but one should not exclude the possibility of an earlier date. As for thenineteenth Hymn to Pan, which seems to have been influenced byHermes, it is most probable that it does not date before c.500, on thegrounds that the cult of Pan did not expand beyond Arcadia beforethis point.72 One cannot, however, rule out a date in the late sixthcentury; Herodotus (6. 105) reports that Pan’s cult was adopted by theAthenians after the battle of Marathon, but it could have spreadelsewhere earlier than this.73 Also, as we have seen above, Hermesmay itself date to the sixth century.

Other Hymns

The remaining Homeric Hymns, all of which are shorter than thosediscussed above, often do not contain any evidence on which to makea decision about date. Janko analyses certain linguistic features ofHymns 6, 27, 28, 30, 31, 32, and 33, which are all between 16 and 22lines in length,74 but the small data set does not allow for reliableconclusions about date. In most cases, the shorter Hymns can prob-ably be assigned to the period between 700 and 500 BC, but there are afew exceptions which are worth discussing briefly here. An obviousoutsider in the collection is Hymn 8 to Ares. Its style, distinguished byan accumulation of epithets, and its identification of Ares with theplanet Mars suggest Neoplatonic authorship. It has been attributed toProclus in the fifth century AD: this is not certain, but it should in anycase not belong to a time before the third century AD.75 Elsewhere, the

70 See in this volume Jaillard (p. 133 n. 2).71 See WL 16–17, who also compares the juxtaposition of Egypt and the Hyper-

boreans to Pi. I. 6. 23. For the Exekias cup, see in this volume Fig. 2 (Chapter 7,p. 134).

72 For more detailed discussion of Hy. 19’s date and its relation to Herm., seeThomas in this volume (pp. 169–72).

73 See Càssola 361–2 and cf. Janko (1982), 184.74 Janko (1982), 186–7.75 For further discussion and bibliography, see in this volume Faulkner (p. 176

n. 4).

Introduction 15

pair formed by Hymns 31 to Helios and 32 to Selene are distinguishedfrom the rest of the collection by their closing formulae, which stateclearly that they will go on to sing of the deeds of heroes, and haveoften been thought to date to the fifth century or later on stylisticgrounds: a fifth-century date seems reasonable, but an earlier datecannot be ruled out entirely.76 Finally, the concept in Hymn 20 ofHephaestus and Athena leading mankind from a primitive state incaves to civilization through their teaching of crafts could suggest adate for this poem after Protagoras in the second half of the fifthcentury.77

4 . PERFORMANCE AND FUNCTION

At the very end of the Delian section of the Homeric Hymn to Apollo,which describes an Ionian festival in honour of Apollo on Delos(146–78), the poet-narrator speaks directly to the chorus of Delianmaidens. He requests that, should anyone in future ask who is thesweetest singer to visit the island, the maidens identify him as thatman:

��E �’ �s �ºÆ �A�ÆØ ��Œæ��Æ�ŁÆØ Iç�ø ·‘�ıçºe I��æ, �NŒ�E �b ��øfi ��Ø �ÆØ�Æº��fi Å·��F �A�ÆØ ��Ø�Ł�� IæØ��ı�Ø� I�Ø�Æ� .’��E �’ ��æ�� Œº� �Y��, ‹�� K�’ ÆrÆ�I�Łæ��ø� ��æ�ç��ŁÆ ��º�Ø �s �ÆØ��Æ��Æ .�Q �’ K�d �c ��ÆØ, K��d ŒÆd K��ı�� K��Ø�.ÆP�aæ Kªg� �P º��ø �ŒÅ��º�� ººø�Æ

��ø� Iæªıæ��, n� M�Œ�� Œ� ¸Å��. (Apoll. 171–8)

You should all answer in unison, ‘He is a blind man and lives in ruggedChios, and all of his songs are best in perpetuity.’ And we will carry yourfame wherever we should wander between the populated cities of men.And they will believe it since it is indeed true. And I will not stophymning far-shooting Apollo of the silver bow, whom well-tressed Letobore.

76 See Càssola 439–40, 446.77 See WL 18.

16 Andrew Faulkner

These lines portray the poet as the blind Homer himself performingthe Hymn to Apollo at a Delian festival, just as he is said to do also inthe Certamen (315–21), one of many such stops for him throughoutthe Greek world.78 Whether or not this description is typical in itsdetails, it is clear that the Hymns were in part performed by travellingrhapsodes at major festivals. Such performances will often have beencompetitive: in Hymn 6 to Aphrodite, the singer specifically asks thegoddesses ‘to grant victory in this contest’ (�e �’ K� IªH�Ø j ��ŒÅ� �fiH��çæ��ŁÆØ, 19–20), and one can assume that this was the case else-where too.79

Apart from their performance at public religious festivals, Claypoints out that the Hymns could have been performed in more privatecontexts, possibly at symposia or as entertainment accompanyingfeasts in court settings.80 When Odysseus visits the Phaeacians inthe Odyssey, the bard Demodocus sings instalments from the TrojanWar after dinner in the palace of Alcinous (8. 62–96, 499–535), just asPhemius sings of the war at the suitors’ private feast on Ithaca earlierin the poem (1. 325–7). During Odysseus’ visit Demodocus also singsthe well-known tale of Aphrodite’s affair with Ares (8. 266–366), anaccount of embarrassing sexual activity which strongly resembles thenarrative of the Hymn to Aphrodite. Demodocus performs the tale ofthe love affair at a more public gathering organized by the Phaeacians,which includes games and dancing, but one might nonetheless rea-sonably infer from the Phaeacian episode that early hexameter hymnsof some length could, just as epic, have been sung in more privatecourt settings. It has been argued, for example, that the Hymn toAphrodite was composed in honour of a family in the Troad whoclaimed descent from Aeneas, and it is notable that this Hymn, incontrast to some others, shows little to no concern with cult.81

Regardless of their exact performance context, the extent to whichthe Homeric Hymns stood on their own as compositions in thesesettings has been a matter of debate. Since Wolf at the end of theeighteenth century, the Hymns have commonly been thought to serve

78 On the performance of this Hymn, see further in this volume Nagy (Ch. 13).79 See Richardson (2010), 1–2, and in this volume Clay (pp. 236–7), Calame

(p. 355). Th. 3. 104. 5 assumes that Apoll. was performed in a contest (agōn); cf. inthis volume Nagy (pp. 322–3). The term agōn (Iª��) is itself used in its basic sense of‘gathering’ at Apoll. 150; cf. Richardson (2010), 105.

80 Clay (1989), 7, and in this volume (pp. 249–50).81 See Faulkner (2008), 3–10, (forthcoming b).

Introduction 17

as preludes to longer epic recitations. There is some good evidencepointing in this direction. The closing formulae of many Hymnsindicate that the poet will go on to sing another song: most explicitis, ‘Having begun with you I will now pass on to another song’ (�� ’Kªg Iæ�� Æ��ÆØ ¼ºº�� K o��, Aphr. 293, Hys. 9. 9, 18.11), but this introductory function could also be indicated by themore common, ‘And I will remember both you and another song’(ÆP�aæ Kªg ŒÆd ��E� ŒÆd ¼ººÅ ��’ I�Ø�B , Dem. 495, etc.).82

Hymns 31 and 32 uniquely state at their close that they will go onto sing of the deeds of heroes. External sources seem to attest to thisfunction as well. Thucydides, who is the first ancient source to quotefrom the Homeric Hymns, refers to the Hymn to Apollo as a prooi-mion (�æ��Ø��, 3. 104. 4–5), a word which can be taken to mean‘prelude’, and Pindar claims at the beginning of his second NemeanOde that the Homeridae begin their recitations with a prooimion toZeus (˜Øe KŒ �æ��Ø��ı, N. 2. 1–5). As well, in the Homeric epics thebard Demodocus is said to begin his singing with an invocation to agod (› �’ ›æÅŁ�d Ł��F Xæå��, Od. 8. 499).

It is not, however, agreed by everyone that the long Hymns func-tioned as preludes to further epic recitation. One consideration iswhether poems of 300 to 500 lines could actually have served aspreludes to longer recitation when they are substantial pieces them-selves.83 Their size is not necessarily an obstacle, given the scale ofepic and the potential ability of an ancient audience to sit throughlong recitations,84 but Demodocus’ performance of the narrativeabout Ares and Aphrodite discussed above is perhaps an indicationthat narrative Hymns of some length stood on their own in perform-ance. The term prooimion (�æ��Ø��) used by Thucydides of theHymn to Apollo is not decisive evidence either, as it can indicatesimply the first in a series of songs, which may or may not haveincluded epic.85 Even if the term prooimion was originally used ofshorter hymns that preceded epic recitation, it could have beenunderstood in a technical sense by Thucydides without the implica-tion of a subsequent performance of any type.86 There is perhaps no

82 On the exact meaning of the latter, see Richardson (1974), 324–5.83 See AHS xciii–xcv.84 See Richardson (1974), 3–4, (2010), 2.85 See further Clay (1997), 494–8, and in this volume (pp. 237–40).86 AHS xciv–xcv, who compare the modern musical term ‘prélude’.

18 Andrew Faulkner

way to decide with certainty, but the extent to which one views thelonger Hymns as preludes attached to epic poetry in performancepotentially has broader interpretative repercussions. For those whosee the long Hymns as intrinsically linked to epic, the poems will lessreadily be understood to stand on their own as a distinct type or genrewithin the tradition of ancient poetry.87 And whether or not theyshould be thought of as an independent genre of poetry has become acentral discussion in scholarship on theHymns in the past twenty-fiveyears.

5 . GENRE, PANHELLENISM, AND LOCALCONNECTIONS

One of the most influential books published on the Homeric Hymnsin the past fifty years has been J. Clay’s The Politics of Olympus.88 In it,she offers a systematic reading of Demeter, Apollo, Hermes, andAphrodite as complex literary compositions that constitute a distinctgenre within archaic Greek poetry. Linear readings of all four poemsargue that each of the long Hymns ‘describes an epoch-making mo-ment in the mythic chronology of Olympus’:89 in the narratives ofApollo and Hermes, the births of the two gods require a rearrange-ment of the Olympian hierarchy, while Demeter and Aphrodite nar-rate a reordering of relations among gods and between gods andmortals. As such, the Hymns can be seen as overlapping with butdistinct from both Homeric epic and Hesiodic theogonic poetry,which constitute the pre- and post-history of the narratives recountedin the Hymns.

This approach has many positive implications for the study of thelong Hymns and has opened up new avenues for scholarship.90 Somedifferences between the Hymns and epic poetry are obvious at a firstglance: apart from structural divergences, epic focuses chiefly upon

87 See Nagy in this volume (pp. 322–32), who closely links the Hymns and epic.88 Clay (1989).89 Clay (1989), 15; cf. her remarks in this volume (Ch. 11).90 In this volume, see in particular Felson (Ch. 12), who following Clay examines

power relations in the Hymns in light of theogony, and Brillet-Dubois (Ch. 6), but alsopassim.

Introduction 19

the world of mortals, in which the gods sometimes interfere, while thenarratives of the long Hymns focus on deities, who at times enter intothe world of mortals. Nonetheless, the Hymns have often been viewedin the shadow of the Iliad and Odyssey as subordinate, if fascinating,poetry. With Clay, we have an increased understanding of how theHymns are to be viewed as compositions in their own right. Further-more, while the Hymns have frequently been treated as a heteroge-neous body of poetry which, apart from their shared language, shouldbe read largely independently, Clay’s work has pushed scholars toread the Hymns together. As such, her approach also emphasizes thePanhellenic quality of the Hymns in place of individual connectionsto local cults.91

There has been some disagreement with aspects of this approach toreading the Hymns. Each of the Hymns certainly narrates an import-ant moment in the life of a god, but the concern with cosmichierarchy is perhaps not equally strong in all cases.92 One mightwonder to what extent the composers of the Hymns were consistentlyconcerned with theological order rather than entertainment.93 Thereare also consequences open for discussion in assuming a high degreeof cohesion among and within the longer Hymns. Whether or not, forexample, the Hymn to Apollo is composed of two originally independ-ent sections is a question for continued debate.94 That said, Clay’swork has without doubt done much to improve our understanding ofthe Homeric Hymns as unified literary works,95 and the Hymnsshould never again be read in isolation from each other. Whileopinions about particular readings will continue to differ, it is pos-sible to find middle ground. Even if Apollo is a composite of originallyseparate hymns, the poem as we have it has clearly been arranged tobe read as a unity and benefits from being understood in this way.96

Related to these issues is the question of the extent to which theHomeric Hymns should be considered Panhellenic, as opposed topoems attached to particular localities and cults. In some respects,

91 Clay (1989), 8–11.92 See Thalmann (1991), and on Apoll. Chappell in this volume (pp. 74–5). For a

qualification of the cosmic significance of Aphr., see Faulkner (2008), 10–18.93 See in this volume Furley (pp. 208–10).94 See in this volume Chappell (Ch. 4), contra the view of an originally unified

Hymn held by Clay (1989) and Miller (1986).95 See Janko (1991).96 Cf. Richardson (2010), 15.

20 Andrew Faulkner

the Hymns are evidently Panhellenic in outlook. A distinction haswith good reason been drawn between the Homeric Hymns as literaryhymns and the hymns used in cult worship in Greece.97 But it wouldbe wrong to draw an absolute line between the two: the HomericHymns themselves, assuming that they were performed at religiousfestivals, can be understood as genuine offerings to a divinity and inthis way involved in cult worship.98 Nonetheless, the Homeric Hymnsare much less attached to particular ritual contexts, and more stronglylinked to the literary Panhellenic theologies of Homer and Hesiod,than cult hymns.

One might still recognize a difference between the Panhellenism ofthe Homeric epics and the Homeric Hymns, some of which, while notcult hymns per se, show a concern with particular cults not found ineither Homer or Hesiod. On the surface, Apollo and Demeter bothdisplay an interest in local cult: the former narrates the foundation ofApollo’s sanctuaries on Delos and at Delphi, while the latter tells ofDemeter’s foundation of the Mysteries at Eleusis. We are told in theCertamen (315–21) that the Hymn to Apollo was inscribed by theDelians and dedicated in a temple of Artemis, and one might assumea particular connection to worship at Delos and Delphi in the twohalves of the poem.99 With respect to Demeter, certain details ofDemeter’s visit to Eleusis are comparable to what we know of ritualpractice in the Mysteries from other sources, which could suggest arelatively close relationship between the Hymn and Eleusis.100 Else-where in the collection connections to cult are less obvious but havebeen proposed also for both Dionysus and Hermes. The myth ofDionysus freeing Hera from the bonds of Hephaestus has been linkedto the Samian Tonaia, a ritual in which a cult image of Hera wasbound and taken to the shore, and the fragments of the poem seem torefer at least once to the foundation of cult.101 Hermes’ division ofmeat into twelve portions in Hermes (127–9) could associate it

97 See Furley and Bremer (2001), i. ix.98 On the Hymns as a musical offering to the divinity, see in this volume Calame

(Ch. 14); Richardson (p. 53 n. 25) and Nagy (p. 306 n. 78) also question the rigidexclusion of the Hymns from the category of cult hymns.

99 See in this volume Chappell (pp. 66–7).100 See Richardson (1974), 12–30, and in this volume (pp. 50–3). For arguments in

favour of the poem’s Panhellenic outlook see above n. 48.101 See West (2001a), 3–4. Fr. D 1–3 seem to explain the establishment of Diony-

siac trieteric festivals.

Introduction 21

with the cult of the Twelve Gods at Olympia, where Hermes wasworshipped together with Apollo.102 As noted above with respect todating, it has also been argued that the seventh Hymn to Dionysuscould be linked to Corinth.

If accepted, these links may be an indication of original perform-ance context, but could also in some cases be the result of theHymns having developed as literary expansions of cult myth.103 Butnot all of the Hymns evince a connection to cult. In Aphrodite thereis mention of the goddess’s sanctuary on Paphos (58–65), but this isstock epic material and no indication of a particular link to cult;other evidence points to northern Asia Minor as the Hymn’s placeof composition.104 As well, when Aphrodite first arrives beforeAnchises on Mt Ida, in the disguise of a young Phrygian princess,the shepherd’s first reaction is to offer to build her a temple andmake sacrifices to her (100–2). This has been taken by some to referto the foundation of an actual cult on Ida,105 but there is no reasonto assume this here. Anchises’ reaction is a general attempt at pietyin the face of a miraculous stranger, whom (after her suddenappearance on his mountain) he assumes is a goddess; he mentionsno geographical or cultic details, such as we find in Demeter andApollo, and does not even know to which goddess he might bespeaking.106 One must, it seems, accept variation in the extent towhich the Hymns’ narratives connect them to local cults, and itcannot be assumed that the Hymns all developed in the same way.Yet, connections to local cult do not rule out Panhellenic appeal inthe Hymns’ narratives. It is hard to deny this even in the cases ofApollo and Demeter, which, in keeping with the status of Delphiand Eleusis as major Panhellenic sites in Greece, celebrate theuniversal power and significance of Apollo and Demeter. Whetherthey evince local connection or not, the narratives of all the longHymns without doubt transcend both local and temporal bound-aries.

102 See Burkert (2001), 178–88, Johnston (2002), 124–6, WL 13–14, Thomas(2009), 14–20. Larson (1995) argues that the poem contains an aetiology for thecult of the bee maidens at Delphi, and Nobili (2008) connects it to the Panathenaea.

103 See in this volume Furley (Ch. 10).104 See Faulkner (2008), 49–50.105 Podbielski (1971), 44–5, Càssola 549.106 See Faulkner (2008), 179.

22 Andrew Faulkner

6. THE PRESENT VOLUME

The preceding discussion has surveyed a number of essential ques-tions in scholarship on theHomeric Hymns up to the present day. It isnot by any means an exhaustive account, but it is hoped that it willprovide for the reader a foundation for approaching the study of theHymns, and a sense of how the individual essays in this volume,which have been referred to above in the footnotes, fit into thebroader tradition of scholarship on the Hymns.

This book proceeds from a steady flow of interest in the HomericHymns in the past fifteen years. To name a few contributions since theturn of the century: M. L. West published an important article on thereconstruction of Dionysus and his Loeb edition of the Hymns;107

following a gap of more than twenty-five years from the publicationof N. J. Richardson’s full commentary on Demeter in 1974, detailedcommentaries on Apollo, Hermes, and Aphrodite have either beenpublished or are currently in preparation for publication;108 a book-length study of the Hymns’ relationship to Callimachus’ Hymns hasbeen completed;109 and N. J. Richardson has recently published hisown edition and commentary on Apollo, Hermes, and Aphrodite inthe Cambridge Greek and Latin Classics series.110

But despite their recent popularity, there has never been a volumeof collected essays devoted to the corpus of Homeric Hymns. In hisreview of the New Companion to Homer in 1999, Hainsworth notedthe need for a collection of expert studies dedicated to the Hymnsalone.111 It is hoped that the present volume will fill this gap. How-ever, this book is more than just a companion. It aims not only toevaluate the state of scholarship on the Hymns but also to advance it,and it is ultimately hoped that it will prove useful to a wide range ofreaders, from undergraduate students under the direction of an

107 West (2001a), WL. See also the German edition of Pfeiff (2002) and slightlyearlier the Italian edition of Zanetto (1996).

108 On Aphr. Faulkner (2008). In preparation: expanded and revised versions ofthe doctoral theses whose dates are given in brackets, on Apoll. Chappell (1995), onHerm. Vergados (2007) and Thomas (2009). Nobili (2008) is another major study ofHerm. undertaken as a doctoral thesis in recent years. Another commentary on Aphr.and some of the shorter Hymns is also now in preparation by Professor D. Olson.

109 Vamvouri-Ruffy (2004).110 Richardson (2010).111 Hainsworth (1999).

Introduction 23

instructor to experts who turn to the Hymns with an inquisitive eye.In pursuit of this blend, as well as general breadth of vision, thecontributors to the volume are composed of both senior and emer-ging scholars from the United Kingdom, France, Germany, Switzer-land, Greece, the United States, and Canada, who together bringa variety of methodologies and backgrounds to the study of theHomeric Hymns.

The volume, which is made up of thirteen original essays plus thepreceding introduction to scholarship on the Hymns, is dividedstructurally into two parts. In the first part, an essay is devoted toeach of the long narrative Hymns, as well as to the two mid-lengthHymns to Dionysus (7) and Pan (19): Martin West considers thereconstruction of the fragmentary first Hymn to Dionysus and therelationship of its narrative to the Homeric epics; Nicholas Richard-son revisits questions central to the interpretation of the Hymn toDemeter in light of scholarship since the publication of his commen-tary on the poem; Mike Chappell re-examines the question of theHymn to Apollo’s unity; Athanassios Vergados considers the uniquepresentation of epiphany in the Hymn to Hermes; Pascale Brillet-Dubois evaluates the Hymn to Aphrodite’s relationship to the Iliadictradition; Dominique Jaillard examines the epiphanic nature ofHymn 7 to Dionysus in relation to the other Hymns in the collection;and Oliver Thomas offers a systematic reading of Hymn 19 to Panand its relationship to the Hymn to Hermes. In the second part, sixessays give broader attention to the collection as a whole, includingthe shorter Hymns: Andrew Faulkner considers the formation of thecollection and surveys the evidence for the transmission and recep-tion of the Hymns from the seventh to the third centuries BC;WilliamD. Furley explores the relationship of the Hymns to other earlyhexameter hymns and their possible genesis from theogonic hymnsand cult myth; Jenny Clay looks at the generic coherence of theHymns and their development as a genre; Nancy Felson offers areading of intergenerational conflict in Apollo and Hymn 28 toAthena and considers further the Hymns’ relationship to Hesiod;Gregory Nagy examines the earliest phases of transmission andreception of the Hymns from the perspective of oral poetics andperformance, with particular attention to Apollo; and Claude Calameconsiders musical performance in the Homeric Hymns and theirnature as musical offerings to divinities.

24 Andrew Faulkner

All of the contributions stand on their own but not in isolation.Where possible, contributors have taken into account the relevantarguments of others in the volume, and throughout the book pertin-ent discussions elsewhere in the volume (whether in agreement ordisagreement) are signalled in the footnotes as much as is practical.Discussion of particular topics or passages can as well be located byusing the indexes. The consolidated list of works cited at the backprovides a full, if not exhaustive, list of bibliography on the Hymns,which, it is hoped, will also be a valuable tool. In order to accom-modate readers of different backgrounds, quotations in languagesother than English, including Greek and Latin, are together given inthe original and an English translation, except in some cases where itwas not possible to include the Greek original for reasons of space.Transliteration is also used in some instances.

Introduction 25


Part I


2

The First Homeric Hymn to Dionysus

Martin West

For many people mention of ‘the Homeric Hymn to Dionysus’ willbring to mind the charming fifty-nine-line poem, numbered as Hymn7 in modern editions, that tells the story of the god’s capture byEtruscan pirates, his wondrous transformations of himself and theirship, and their mutation into dolphins. There is another brief Hymnto Dionysus, a mere thirteen lines, at number 26. But the collectionoriginally contained a much longer and more important Hymn tohim. Only fragments of it now remain, but they stand under theprestigious heading of Hymn 1. They are sufficient for us to recon-struct the principal myth related in the poem. It will appear that thismyth, and probably this Hymn, was already known to the poets of theIliad and Odyssey and influenced several scenes in those epics.

In the most recent edition of the Hymns the poem is represented byfour separate fragments, containing altogether sixty-three lines orparts of lines.1 In older editions only two of these fragments areprinted. It is not quite certain that the other two belong, but it ishighly probable.

The last twelve lines of the poem are actually preserved in a manu-script of the Hymns, the codex in Leiden (M) that alone preserves theHymn to Demeter. The earlier part of the manuscript contains aportion of the Iliad, after which a quinion and a folio have fallenout; following the gap there appears the end of the Dionysus Hymn,

1 WL 26–31. The four fragments are there designated A–D, with separate line-numbering for each, and that system of reference will be followed here (as throughoutthe volume).

and then the long Demeter Hymn and the rest as far as Hymn 18. 4.On certain assumptions it can be calculated that the Dionysus Hymnwould have been some 411 lines in length.2

The other fragment printed in older editions comes from a quota-tion in Diodorus, who attributes it to ‘the Poet’ (or in another passage‘Homer’) ‘in the Hymns’.3 This must refer to the same collection ofHomeric Hymns that has (mostly) come down to us. As the fragmentconcerns the birth of Dionysus and is clearly from the beginning of ahymn to him, and as it is unlikely that we have lost two separateDionysus hymns from the same corpus, we can accept without moreado, what has been generally assumed since Ruhnken, that Diodorus’fragment is from the same poem as the one in M. In both fragmentsDionysus is addressed by the title ¯NæÆçØH�Æ, ‘Bull god’ (A 3, D 8).

In 1992 André Hurst identified a Geneva papyrus of the second orfirst century BCE as containing the lines quoted by Diodorus and partsof several more. Four of the fragmentary lines, it has since beendiscovered, reappear almost unchanged in the Orphic Argonautica,a poem of the fifth century CE, and can be restored from that source.4

They show that the poet’s description of Mt Nysa, Dionysus’ birth-place, continued beyondDiodorus’ quotation and that he portrayed itas a locus amoenus, cut off from human traffic by high cliffs and a lackof any harbour.

Zeus gave birth to Dionysus there, ‘far from mankind, concealinghim from white-armed Hera’ (��ººe� I�’ I�Łæ��ø�, Œæ��ø�º�ıŒ�º�� æÅ�, A 8). Hera, of course, was ill disposed towards herhusband’s illegitimate children. By the end of the hymn (D) Dionysuswas established as a god, presumably on Olympus, destined to enjoy athriving cult in Greek lands. He no longer had to be kept apart fromHera, who must have become reconciled to him. The main part of thehymn must have contained a mythical narrative that explained howhe progressed from the initial to the final situation.

We can guess how the story went. We know from various sourcesof a myth, evidently well known from sometime before 600 BCE, thattold how both Dionysus and Hephaestus came to be received inOlympus and accepted by Hera, and this was no doubt the tale toldin our hymn. It ran as follows. When Hera gave birth to Hephaestus

2 See West (2001a), 1.3 ‘The Poet’ (› ��ØÅ�� ), Diod. 1. 15. 7; 3. 66. 3; ‘Homer’ (› �OÅæ� ), 4. 2. 4.4 Hurst (1994); West (2001a), 2. Cf. below pp. 42–3.

30 Martin West

she was disgusted at the crippled child she had borne and threw himdown from heaven into the sea. There he stayed for many years withthe sea nymphs, the Nereids. After developing his engineering skillsto the necessary degree, he made a fine golden throne and sent it as apresent to his mother. But in it he had incorporated a secret mechan-ism of invisible bonds, and when she sat down in it, she found herselftrapped. The other gods laughed at her plight. None of them was ableto free her. They debated what to do, and concluded that the onlyrecourse was somehow to induce Hephaestus to come back and undowhat he had done. Ares undertook to go and fetch him by force. Hewent off, but failed to achieve his object because Hephaestus defendedhimself with fire, which Ares could not face. Then Dionysus, who wasnot yet acknowledged as one of the Olympians, went equipped withwine. He made Hephaestus drunk and brought him back to Olympusin a merry mood, riding on a donkey or mule. He released his motherfrom the throne, and she rewarded Dionysus by persuading the otherOlympians to admit him as one of the Twelve Gods.

For a connected account of the myth we have to look to latesources.5 But a series of fragments of Alcaeus shows that he knewand retold the story, and so did Pindar and Epicharmus. The trage-dian Achaeus treated it, or at least touched on it, in a satyr play. Platoalludes to ‘Hera’s binding by her son’ as a well-known poetic myth.6 Itwas a favourite subject with artists.7 The François Vase, a work ofKleitias dating from about 570 BCE (Fig. 1), has a particularly goodrepresentation of it, from which some further details may perhapsbe gleaned. Dionysus and the ithyphallic mule bearing Hephaestusapproach Olympus, attended by a company of Silenoi and nymphs.Behind the throned Zeus and Hera, Athena looks down disdainfullyat the armed but submissively postured Ares, as if mocking the failureof his mission. In front of Zeus, as the first to meet the new arrivals,stands Aphrodite, who makes gestures of dismay. Carl Robert con-jectured that in the version followed by the artist she had beenpromised to Hephaestus as payment for releasing Hera. Wilamowitz

5 Paus. 1. 20. 3; ps.-Libanius, Narr. 30. 1 (viii. 38–9 Foerster); cf. Aristid. Or. 41. 6(ii. 331 Keil); Hyg. Fab. 166; Serv. auct. Ecl. 4. 62.

6 Alc. frs. 349a–e Voigt (cf. Page 1955, 258–61; Libermann 1999, ii. 152–3); Pind.fr. 283; Epicharmus, ˚øÆ��Æd j ῞ çÆØ�� , PCG i. 51; Achaeus TrGF 20 F 16b–17;Pl. R. 378d.

7 LIMC iv(1). 628–9; Gantz (1993), 75–6. For the François Vase see e.g. Arias,Hirmer, and Shefton (1962), plates 41 and 45 and p. 291.

The First Homeric Hymn to Dionysus 31

developed the idea: Aphrodite, he supposed, had been promised toAres if he succeeded in his mission, but in the event it wasHephaestuswho got her.8 Perhaps Hera had promised her to whoever couldinduce Hephaestus to come and set her free, and Dionysus, havingobtained her as his reward, bestowed her on his new friend Hephaes-tus.

The Homeric Hymn was no doubt the classic account that laterpoets and painters had primarily in view. Already in 1872 TheodorBergk, noting the popularity of the myth in archaic poetry and art,remarked that it might have been related in a hymn of the Homerictype, and Wilamowitz postulated a seventh-century hymn to Heph-aestus as the original source of it.9 Bruno Snell made a convincingmodification to the hypothesis, namely that the hymn was not ahymn to Hephaestus but to Dionysus.10 It was Dionysus, after all,who came out of the story with the most credit, and it was hisadmission to Olympus that formed its conclusion. He was, besides,a likelier god than Hephaestus to be honoured with a major hymn.But if such a hymn existed and was widely known in the sixth andfifth centuries, there is some likelihood that it was the same majorhymn to Dionysus as was later incorporated in the corpus of HomericHymns, the one of which we have fragments.

Our third main fragment, C, comes from a papyrus published in1904. The first editors, Grenfell and Hunt, did not recognize the

Fig. 1. François Vase, by Kleitias. By Permission of Nimatallah / Art Re-source, New York

8 Robert (1894), 177 n. 3;Wilamowitz-Moellendorff (1895), 220¼ (1937), 9.Cf. Il.14. 267–8, where Hera promises Hypnos one of the younger Graces if he will help her.

9 Bergk (1872), 761 n. 48; Wilamowitz-Moellendorff (1895), 221–3¼ (1937), 9–12.10 Snell ap. Treu (1952), 138–9 ¼ (1963), 149–50; id. (1966), 102–4; Eisenberger

(1956), 31–2.

32 Martin West

subject matter. R. Ganszyniec in 1922 saw that it contained the storyof Hera’s throne, but he took it to be verse of the late Hellenistic orearly Imperial period. It was not until 1973 that Reinhold Merkelbachproposed assigning it to the archaic hymn that Wilamowitz and Snellhad postulated.11 This is very plausible. The lines are straightforwardin style and do not look like a Hellenistic or later re-telling of themyth. One cannot be certain that they are early,12 but if they are, theymust be from a ‘classic’ work, which cannot be any but the HomericHymn.

About the content of the poem there is not much more that wecan say. In its narrative portion it must have been a fine exampleof the kind of Olympian burlesque represented in the Hymn toHermes, in Demodocus’ song about Ares and Aphrodite in theOdyssey (8. 266–366), and in more than one scene of the Iliad.13

We think of this as a typically Ionian genre, and there is little doubtthat the Dionysus Hymn was an Ionian poem. Wilamowitz connectedthe myth of Hera’s bondage with the ritual of the Samian Tonaiafestival, in which (as may be inferred from an aetiological myth toldby Menodotos, a Hellenistic local historian) the goddess’s woodenimage was taken down to the shore and tied up in withies.14 He alsodraws attention to a story that Hera, on giving birth to Hephaestus,apprenticed him to Kedalion in Naxos to train as a smith; Dionysuswas important on that island.15 Against a Naxian origin for the Hymnis the fact that in fragment A (2–7) Naxos appears in a list of placeswhere, according to the poet, people falsely claim that Dionysus wasborn. The other places in the list are Drakanos, Ikaros, Thebes, andsomewhere by the Alpheios. The mention of Drakanos and Ikaros asclaimants, the first a promontory on Cos, the second a small islandnear Samos, does indicate the southeast Aegean as the poet’s vantage-point.

11 P. Oxy. 670; CA 80–1, cf. 245; Ganszyniec (1922); Merkelbach (1973), 212–15¼(1997), 35–7.

12 Dihle (2002) still maintains that they are late, and Faulkner (2010a) drawsattention to a couple of vocabulary items which, while not intrinsically late information, are not attested in early texts.

13 On humour in Herm., see in this volume Vergados (Ch. 5).14 Wilamowitz-Moellendorff (1895), 234–5 ¼ (1937), 23–5; Menodotos FGrH 541

F 1 ap. Ath. 672a–d.15 SchT Il. 14. 296a; Wilamowitz-Moellendorff (1895), 237 ¼ (1937), 27. Dionysus

and Hephaestus are supposed to have contested the possession of Naxos (sch. Theoc.7. 149/150a).


As to his date, if we accept that he was the model for Alcaeus, wemust put him sometime before 600 BCE. That does not necessarilymean shortly before 600. In several places in the Iliad and Odyssey wefind what look like echoes and adaptations of the Hymn’s centralmyth. The myth might be older than the Hymn. But one has thefeeling that it is not a myth with a firm basis in ancient tradition but anewly invented story; there is a lively freshness about it, as of a storythat has just appeared and is going the rounds, captivating newhearers. It is quite possible that it first appeared in our Hymn, or atany rate that it was our Hymn that first made it widely known. If so,we shall probably have to put the Hymn back at least to 650. It maywell be the earliest in the collection.

Let us look at these apparent Homeric echoes. We may start withthe scene in Iliad 18 where Thetis goes to visit Hephaestus to ask himto make new armour for Achilles. He has a sweet and beautiful wife,one of the Graces (382–3). This conflicts with Demodocus’ tale inOdyssey 8, where Hephaestus is married to Aphrodite. But the Aph-rodite of the Iliad is a pro-Trojan goddess; we have seen her rescuingParis from Menelaos (3. 373–82) and trying to rescue Aeneas fromDiomedes (5. 311–430). The poet could not plausibly have shownher as the mistress of the house in the Thetis–Hephaestus episode.We may see the Grace as a substitute for Aphrodite: the Graces areafter all closely associated with her as spirits of beauty and charm.16

But what is behind the marriage of Hephaestus to Aphrodite?They are an ill-matched pair, and in Demodocus’ tale she is unfaith-ful to him, as if he were not her partner of choice. In the DionysusHymn, as we have seen, there is some reason to suspect that shewas given to him against her will, in fulfilment of Hera’s promise thatif he did what she wanted he would possess the most beautifulgoddess.

When Hephaestus hears that Thetis has come, he is delighted:

q Þ� �� Ø ��Ø�� ŒÆd ÆN��Å Ł�e ��,l ’ K��ø�’, ‹�� ’ ¼ºª� Iç�Œ�� Bº� ��ÆÅ�æe KB N��Å�Ø Œı��Ø�� , l ’ KŁºÅ��Œæ�łÆØ åøºe� K��Æ. ��’ i� ��Ł�� ¼ºª�Æ ŁıfiH,�N � ’ ¯Pæı��Å �� ¨�Ø Ł’ ��Æ�� Œ�º�øfi ,¯Pæı��Å, Łıª��Åæ Ił�ææ��ı ὨŒ�Æ��E�.

16 Il. 5. 338; Od. 8. 364, 18. 193–4; Apoll. 194–5, Aphr. 61; Cypria fr. 5. 1 West.

34 Martin West

�Bfi �Ø ��æ’ �N�� å�ºŒ�ı�� Æ��ÆºÆ ��ºº�,��æ�Æ �� ª�Æ�� Ł’ �ºØŒÆ Œ�ºıŒ� �� ŒÆd ‹æ�ı ,K� ��Bœ ªºÆçıæfiH· ��æd �b Þ�� ὨŒ�Æ��E�IçæfiH �æ�æø� Þ�� ¼�� . �P� �Ø ¼ºº� �Y�� h�� Ł�H� �h�� Ł�Å�H� I�Łæ��ø�,Iººa ¨�Ø �� ŒÆd ¯Pæı��Å Y�Æ�, Æ¥ ’ K��ø�Æ�. (Il. 18. 394–405)

Ah, I have a goddess worthy of fear and respect in my house,the one who saved me when I was in distress after a long fallat the instance of my shameless mother, who wantedto put me out of sight because I was lame.Then I would have suffered woeif Eurynome and Thetis had not received me in their bosom,Eurynome the daughter of back-flowing Oceanus.With them I spent nine years forging many ornaments,pins, bent spirals, rosettes, and necklaces,in their hollow cave, while around it Oceanus’ streamteeming with spume flowed endlessly. No one elseknew, neither of gods nor mortal men;only Thetis and Eurynome knew, the ones who had saved me.

Here is the first half of the myth related in the Dionysus Hymn.Hephaestus leaves it unexplained how he got back to Olympus andtook up his place there; to have continued with the story of the trickthrone would not have been appropriate to the context, where onlyhis indebtedness to Thetis was relevant.

However, we can find a reflection of the throne motif in the DiosApate episode, where Hera seeks assistance from the god Hypnos,Sleep, to overcome Zeus. She goes and finds him in Lemnos (Il. 14.230). Why Lemnos? It is a convenient staging-post on the way fromOlympus to Ida, but not the only available one, and this is hardly asufficient explanation of Hypnos’ being located there. The only godwho would naturally be found there is Hephaestus, for Lemnos is hisfavourite resort.17 It looks as if the poet has a story concerningHephaestus in mind. And when Hera asks Hypnos to put Zeus tosleep for her, she promises to repay him with a fine golden throne thather son Hephaestus will make specially (14. 238–40)—not the mostobvious gift for the god of sleep, who spends his time roaming overthe earth to exercise his office.

17 Cf. Il. 1. 593; Od. 8. 283–4; ‘Hes.’ fr. 148(a).


Hypnos demurs, recalling an earlier occasion when he had putZeus to sleep at Hera’s request, when Heracles was sailing back fromTroy and she had wanted an opportunity to rouse a storm againsthim (243–62). The hero had been blown to Cos in consequence. Thiswas part of the pre-existing Heracles mythology known to the poet ofthe Iliad (cf. 15. 18–30). From it he has borrowed the idea of Hera’sputting Zeus to sleep, with Hypnos’ help, for the Dios Apate. But inreferring to the Heracles story here he adds an inorganic tailpiece thatmay again be spun out of a motif in the Hephaestus story:

n �’ K��ªæ�� åÆº�ÆØ��,ÞØ��Çø� ŒÆ�a �HÆ Ł�� , Kb �’ ��åÆ ��ø�Ç��Ø· ŒÆ� Œ ’ ¼œ�� I�’ ÆNŁæ� ��Æº� ��øfi ,�N c ˝f� ��ØæÆ Ł�H� K��ø�� ŒÆd I��æH�.�c� ƒŒ�Å� ç��ªø�, n �’ K�Æ��Æ�� åø�� æ·–Ç�� ª�æ, c ˝ıŒ�d Ł�Bfi I��Ł�ØÆ �æ��Ø. (Il. 14. 256–61)

And he [Zeus] woke up and was angry,hurling gods about the house; he was lookingfor me [Hypnos] especially,and he would have thrown me out of sight from the sky into the seaif Night had not saved me, she who subdues gods and men.To her I fled; and he gave up, angry though he was,for he was hesitant to do anything displeasing to swift Night.

Hypnos is thrown down from heaven into the sea and saved by afriendly goddess, just as Hephaestus was in the Dionysus Hymn.A variation on the same motif has appeared in Iliad 1. 590–4, whereHephaestus recalls how he was once thrown from heaven, not byHera (whom he is now comforting) but by Zeus; he landed not in thesea but on Lemnos, and he was looked after not by Nereids but by thelocal people, the Sinties.

Hera reassures Hypnos that Zeus does not care so much about theTrojans as he did about his son Heracles, and that he will not be soangry with Hypnos this time. And now she succeeds in persuadinghim by offering him a different reward:

Iºº’ YŁ’, Kªg � Œ ��Ø �Ææ��ø� �Æ� ›�º��æ�ø��ø O�ıØ��ÆØ ŒÆd �c� Œ�ŒºB�ŁÆØ ¼Œ�Ø�Ø�. (Il. 14. 267–8)

But come, I will give you one of the younger Gracesto marry and to be called your wife.

36 Martin West

Hypnos is thrilled with this, and after making Hera confirm the offerwith an oath he sets about doing her bidding. Of course there was notradition of Hypnos being married to one of the Graces, and it makesno mythological sense. It is an invention ad hoc, inspired by Hera’soffer of Aphrodite to Hephaestus. Hypnos could not be offeredAphrodite, so she is replaced by a minor goddess of love and beauty,a Grace, just as in the Thetis–Hephaestus scene in book 18.

Later, when Zeus wakes up and sees how Hera has tricked him, hethreatens her with rough punishment and reminds her of the sameHeracles episode that Hypnos had recalled (Il. 15. 18–30). There weheard of Zeus’ rage against Hypnos; here he speaks of what he haddone to Hera, and again this seems to be a new element in the story,invented for the context.18 He had strung her up with anvils fastenedto her feet and bound her hands with an unbreakable golden ligature.This does not correspond to anything in the Dionysus Hymn, but theadditional detail that ‘the gods were upset and stood round her, butcould not free her’ (Mº�� b Ł��d ŒÆ�a ÆŒæe� Ὄºı��, j ºF�ÆØ �’�PŒ K��Æ�� ÆæÆ��Æ��, 21–2) might be adapted from the storyof when she was trapped in the trick throne. Zeus seized any ofthese sympathizers he could and threw him down to earth (22–4):this corresponds to what we heard previously of his treatment ofHephaestus (Il. 1. 590–4) and Hypnos (14. 257–8).

Turning to the Odyssey, we recall Demodocus’ tale of the adulter-ous affair of Ares and Aphrodite.Here we findHephaestus married toAphrodite, in accord with the story told in the Dionysus Hymn, onlyhere he is represented as having obtained her by paying a bride-priceto her father (8. 318–19). But she despised him for being lame (308–11),just as Hera had done at his birth, and she took Ares as her lover.Initially he seduced her with generous gifts (269), but presently heonly had to suggest a session of lovemaking and she willingly agreed(292–5). But Helios, the all-seeing one, had told Hephaestus what wasgoing on (270–1). The cunning artificer then rigged up his bedroomwith a network of bonds, fine as spider’s silk, so fine as to be invisible,and withdrew to his favourite island, Lemnos (273–84). Seeing thatthe coast was clear, Ares went to the house and took Aphrodite tobed. Down came the bonds, and the lovers found themselves caughtfast, unable to move an arm or a leg. Helios once again reported

18 So Von der Mühll (1952), 227.


proceedings to Hephaestus, who returned from Lemnos, took onelook into the bedroom, and called upon the rest of the gods to comeand see how he had been betrayed. The male gods came (the god-desses were too modest), and they had a good laugh when they sawhow Hephaestus had caught the guilty pair (285–332).

There are two motifs here, besides Hephaestus’ marriage to Aph-rodite, that recall the myth of the Dionysus Hymn. One is Ares’attempt to get the better of Hephaestus, which fails because of thehandicapped god’s mastery of technology. The other is the idea of apiece of furniture that traps those who put themselves on it and holdsthem so fast that no one but its maker, Hephaestus, can release them.The throne that traps Hera seems likely to be the primary instance, atleast if it is the mythical reflection of a seated and bound cult statue.The bed that traps Ares and Aphrodite will be a secondary adaptationof the motif.

At the end of the Dionysus Hymn (fragment D) Zeus proclaimedthe details of the cult with which the god was to be worshipped, andthen gave solemn confirmation that it would be so:

q, ŒÆd ŒıÆ�fi Å�Ø� K�’ Oçæ��Ø ��F�� ˚æ��ø�·I�æ��ØÆØ �’ ¼æÆ åÆE�ÆØ K��ææ��Æ�� ¼�ÆŒ�� ŒæÆ�e ¼�’ IŁÆ��Ø�, ªÆ� �’ KººØ�� Ὄºı��. (D 4–6)

So spoke the son of Kronos, and nodded thereon with his sable brows;and the lord’s ambrosial locks danced upfrom his immortal head, and he sent a tremor through great Olympus.19

These same lines also occur, of course, in the Iliad (1. 528–30), whereZeus confirms to Thetis his promise that he will give the Trojanssuperiority in battle and so ensure Achilles’ restoration to honour.The passage has been celebrated since antiquity as one of the greatestthings in the poem. But how is the coincidence to be interpreted? Arethe lines simply a formula that a poet might use whenever Zeusdecided something?20 The first line does reappear at 17. 209. But in

19 As in certain other places in the Hymns, the manuscript tradition shows acombination of alternative recensions; cf. West (2001a), 9. D 7, S �N�g� K��ı��ŒÆæ�Æ�Ø Å��Æ Z�� (‘So saying, wise Zeus confirmed it with a nod of his head’),gives a shorter replacement for 4–6, and 11–12 are likewise an alternative to 8–10.

20 As suggested by Càssola 465: ‘il cenno di Zeus doveva essere un tema abituale delrepertorio epico, e qualunque rapsodo poteva usarlo’ (‘Zeus’ nod must have been aregular motif in the epic repertoire, and any rhapsode could use it’).

38 Martin West

the Thetis scene the poet certainly does not treat Zeus’ nod as routine;he introduces it as something special:

�N �’ ¼ª� ��Ø Œ�çÆºBfi ŒÆ�Æ��ÆØ, ZçæÆ ��Łfi Å ·��F�� ªaæ K� KŁ�� ª� ��’ IŁÆ��Ø�Ø ªØ��Œøæ· �P ªaæ Ke� �ÆºØ��ªæ�� P�’ I�Æ�Åº��P�’ I��º��Å��, ‹ �� Œ�� Œ�çÆºBfi ŒÆ�Æ��ø. (1. 524–7)

Come, I will nod for you with my head, so that you may rely on it.For this is the greatest sign from me among the immortals:nothing of mine is taken back or delusiveor unaccomplished, that I nod upon with my head.

And when Zeus returns to his halls Hera voices her suspicion that hehas ‘nodded’ definitively for Thetis (558).

It is usually taken for granted, unreflectingly, that the lines areoriginal in the Iliad and that in the Dionysus Hymn they are, if notinterpolated, an imitation. But if we are right in finding so manyechoes and reflections of the Hymn in the Iliad and Odyssey, it will benatural to see the lines as a borrowing from the Hymn by the Iliadpoet. Zeus’ promise to Thetis is of momentous significance for theIliad, and we can see why the poet wanted to emphasize it. Yet from awider viewpoint it may be felt that the Mighty Nod is more appro-priate to an occasion such as that in the Hymn, where another god’shonours and privileges are being solemnized,21 than to a privateagreement with a minor deity for a temporary purpose.

The Dionysus Hymn remained influential for centuries. Alcaeuscomposed one of his own in which he retold the same myth; it wasone of several poems in which he drew on current hexameter hymnsto encapsulate their substance in a more compact lyric format.22 If the

21 Cf. Sappho 44A. 7–8 Voigt.22 In his hymn to Hermes (fr. 308 Voigtþ SLG S264) he related how the infant god

stole Apollo’s cattle and then his bow, the story we know from the Homeric Hymn toHermes. The Homeric Hymn is unlikely to be as early as Alcaeus, but it no doubt hadan older model. Alcaeus’ hymn to the Dioskouroi, fr. 34 Voigt (with which PMG 1012perhaps belongs), had very similar content to Hymn 33. There is also a fragment of ahymn to Artemis, ascribed by some to Alcaeus (304 Lobel–Page) but by others toSappho (44A Voigt), in which the goddess swears an oath of eternal virginity whiletouching the head of her father Zeus, and he assents and grants her her special role asa huntress. In the Homeric Hymn to Aphrodite (25 ff.) there is a similar description,but applied to Hestia instead of Artemis. On Alcaeus’ (and Sappho’s) relationship toearly hexameter hymns and epic poetry cf. West (2002a).


five tiny fragments assigned to it in Voigt’s edition (349a–e) all belongthere, the god was addressed with the Lesbian form of the same title asin the Homeric Hymn, !¯ææÆç�H�Æ ¼ ¯NæÆçØH�Æ (‘bull god’). Thegods’ laughter at the entrapped Hera’s plight was noted, as was Ares’boast that he would fetch Hephaestus by force. Dionysus ended up as‘one of the Twelve’.

Alcaeus and Sappho can already refer to a body of established‘classic’ poems. So do the Attic black-figure vase painters of a decadeor two later, such as Sophilos who depicted the chariot race atPatroclus’ funeral games from the Iliad. On the magnificent volutekrater known as the François Vase Sophilos’ younger contemporaryKleitias painted a whole series of mythological scenes from variouspoems: the Calydonian boar hunt; the chariot race at the games forPatroclus; Theseus’ liberation of the Athenian youths and maidensfrom Crete; the battle of Lapiths and Centaurs; the wedding of Peleusand Thetis; Achilles’ pursuit of the doomed Troilos; Hephaestus’return to Olympus with Dionysus, as described above (Fig. 1); andthe battle of the Pygmies and Cranes. The inclusion of the myth fromthe Dionysus Hymn testifies to its prominence in Athenian con-sciousness at the time.

We know virtually nothing of the treatments by Pindar, Epichar-mus, and Achaeus, but they confirm the continuing appeal of themyth, and the Hymn, which by now will have been current under thename of Homer, will have been well known to these authors, and willhave served as their primary source. ‘Hera’s binding by her son’figures prominently among the poetic topics that Plato would banfrom his ideal city, together with ‘Hephaestus’ casting out by his fatherwhen hemade to assist his mother whowas being beaten [Il. 1. 590–1],and all the battles of gods that Homer has composed’ (R. 378d).23

When Callimachus in his hymn to Zeus asks:

�H ŒÆ� �Ø�, ˜ØŒ�ÆE�� I�� Mb ¸ıŒÆE��;K� ��ØBfi �ºÆ Łı� , K��d ª�� Iç�æØ��.Z�F, �b b� !"�Æ��Ø�Ø� K� �hæ�� çÆ�Ø ª��ŁÆØ,Z�F, �b �’ K� �æŒÆ��fi Å· ��æ�Ø, ��æ, Kł��Æ��;‘˚æB�� I�d ł�F��ÆØ’ . . .K� � �� —ÆææÆ��fi Å #��Å �Œ��. (Call. Hy. 1. 4–10)

23 �HæÆ �b ��f ��e �� ŒÆd �HçÆ��ı Þ�ł�Ø ��e �Æ�æ� , ºº�� Bfi Å�æd�ı��fi Å Iı��E�, ŒÆd Ł��Æå�Æ ‹�Æ �OÅæ� ��ÅŒ�� P �ÆæÆ��Œ�� N �c��ºØ�, �h�’ K� ��ÆØ ��ØÅ�Æ �h�� ¼��ı ��ØH�.

40 Martin West

How shall we sing of him, as Dictaean or Lycaean?I am in two minds, since his birth is disputed.Zeus, they say you were born in the mountains of Ida,but Zeus, they say in Arcadia: which of the two parties, Father, spokefalse?‘Cretans are always liars’ . . .It was in Parrhasia that Rhea bore you.

— we might think he is being the typical Alexandrian, showing off hiserudition, playing off different traditions against each other, andasserting the right to pass judgment on them. In fact, at the outsetof his first hymn, he is following the model of what was probablyalready read as the first Homeric Hymn (A 2–8):24

�Q b� ªaæ ˜æÆŒ��øfi �’, �Q �’ !"Œ�æøfi M��fi ÅçA�’, �Q �! K� ˝��øfi , �E�� ª�� ¯NæÆçØH�Æ,�Q � �’ K�’ �ºç�ØfiH ��ÆfiH �ÆŁı�Ø��Ø{Œı�Æ�Å� $�ºÅ� ��Œ�Ø� ˜Ød ��æ�ØŒ�æÆ��øfi },¼ºº�Ø �’ K� ¨��fi Å�Ø� ¼�Æ� �� ºª�ı�Ø ª��ŁÆØ,ł�ı��Ø· �b �’ ��ØŒ�� Æ�cæ I��æH� �� Ł�H� ��ººe� I�’ I�Łæ��ø� . . . .

For some say it was at Drakanos, some on windy Ikaros,some on Naxos, O scion of Zeus, Bull god,and some at Alpheios the deep-swirling river,while others, Lord, say that it was at Thebes you were born—all false! The father of gods and men gave you birthfar from humankind . . .

A century later the Pergamene scholar Crates of Mallos quotedfrom ‘the ancient Hymns’, meaning no doubt the Homeric Hymns, ahexameter, ‘luxuriant with their own dark grape clusters’ (ÆP�Bfi �Ø��ÆçıºBfi �Ø �ºÆ��fi Å�Ø� Œ��ø�� ) referring presumably to vines. Itis a reasonable assumption that it came from a hymn to Dionysus,and that it was the one Homeric Hymn to Dionysus that we know wehave lost. It is included in my edition as fragment B. The line mighthave stood in the lengthy description of Nysa, where Dionysus wasborn (fr. A). Alternatively it might have described the response ofnature as he made his way with Hephaestus in a bibulous ŒH� (‘revel’) from land to land—a Dionysiac miracle like the one that

24 On the formation of the collection, see Faulkner in this volume (pp. 175–81).


happens on the pirate ship in Hymn 7. 35–42, when a vine laden withgrape clusters spreads along the top of the sail.

Diodorus in the time of Augustus quotes eight lines from fragmentA on the birth of Dionysus.25 The last two he quotes twice again inother passages, and they are also quoted by a scholiast on ApolloniusRhodius. Evidence for knowledge of the Hymn continues in theRoman period. Pausanias, commenting on a painting in the theatreof Dionysus at Athens that showed the god bringing Hephaestus upto heaven, gives a summary of the myth, starting from Hephaestus’birth, that corresponds well with our other evidence for the content ofthe Homeric Hymn, and he introduces it with the words ºª��ÆØ �bŒÆd �� e % Eºº��ø� (‘this also the Greeks say’, 1. 20. 3). Pausaniaswas an avid reader of early poetry, and it is likely that he is drawing onhis direct knowledge of the Hymn. We cannot necessarily assume thisfor the other late sources that relate or allude to the myth. However,the plagiarization of the Hymn by the author of the Orphic Argo-nautica is a striking testimony that cannot be gainsaid. The versesthat he takes over are:

��Ł’ �h �Ø �f� �Å€d] ��æ:[Afi ] �æ��ø� I�Łæ��ø�·�P ª�æ �ƒ ��Ø ºØ]:��, �ÅH� Zå� IçØ�ºØ��ø�,Iºº� �ƒ Mº��Æ]�:�: : ��æÅ ��æØ��æ�� fi Å�łÅº�, �� ŒÆ]º:a ç��Ø ��ØŒÆ ��ºº�.

(A 11–14 � Orph. Arg. 1199–202)

No one of humankind crosses there by ship,for it has no harbour where curly-tipped ships can ride,but a steep cliff encloses it all aboutto a great height. But it grows lovely and delicious things

in abundance.

The later poet applies the lines to the Oceanic island of Demeter fromwhich Pluto carried off Persephone as she picked flowers in the broadgrove. That story is told, of course, at the beginning of the Hymn toDemeter, the Hymn that follows the Dionysus Hymn in the Homericcorpus. It may have been the contiguity of the two Hymns that led thepoet of the Argonautica to think of the topographical description thatstood in the one when he referred to the myth narrated in the other.

25 Diod. 3. 66. 3. A ninth line (A 5) is interpolated in some manuscripts ofDiodorus, but absent from the Geneva papyrus. Cf. above p. 30.

42 Martin West

In any case his knowledge of the passage provides further confirma-tion that the Dionysus Hymn was a standard text, still being read inthe fourth or fifth century of our era.

But then, we know that it survived the end of antiquity togetherwith the rest of the Homeric Hymns; that it survived, indeed, at leastinto the fifteenth century, when Ioannes Eugenikos copied it com-plete in M, probably in Constantinople sometime after 1439. The lossof its greater part occurred at an unknown date between then and1777, when Christian Friedrich Matthaei, professor and Rector inMoscow, discovered the manuscript there. Is it beyond the bounds ofpossibility that the missing leaves might one day turn up, to restore tous one of the earliest of Greek poems and one of the most significantfor literary history?


3

The Homeric Hymn to Demeter

Some Central Questions Revisited

Nicholas Richardson1

1 . INTRODUCTION

This Hymn stands out from the other ones in the collection, becauseof the sustained intensity and dramatic power of its narrative, and theseriousness of its theme, the rape of Demeter’s daughter Persephoneby the lord of the underworld and its consequences. A brief review ofthe poem should illustrate these points.

It is really a hymn to both Demeter and Persephone, deities soclosely linked in both myth and cult that they were often referred tosimply as �g Ł�� (‘the two goddesses’). After announcing that he willsing of Demeter and her daughter the poet immediately moves intothe narrative of Persephone’s violent abduction by Hades, an actionsanctioned directly, as he says, by her father Zeus (1–3). The carefreeinnocence of the scene where she is gathering flowers with thedaughters of Oceanus is undermined by the description of the nar-cissus as a ‘trap’ (��º� , 8) created by Gaia, through the will of Zeus,as a favour to Hades, ‘the host of many’ (��ºı�Œ�fi Å, 9). The wholeuniverse, gods and mortals, heaven, earth, and sea are caught up inwonder at its extraordinary beauty. For Kore (Persephone) it issimply ‘a lovely plaything’ (ŒÆºe� ¼ŁıæÆ, 16), but when she reachedout to pluck it:

1 I should like to thank Andrew Faulkner for his helpful suggestions for improve-ments to this paper. Translations are my own.

å�� b åŁg� �Pæı�ªıØÆ

˝��Ø�� i �� Bfi Zæ�ı�� ¼�Æ� —�ºı�ªø�

¥��Ø IŁÆ��Ø�Ø ˚æ��ı ��ºı��ı� ıƒ� .±æ��Æ �’ IŒ�ı�Æ� K�d åæı��Ø�Ø� Zå�Ø�Ø�qª’ Oº�çıæ��Å�· N�åÅ�� ’ ¼æ’ ZæŁØÆ çø�BfiŒ�Œº��Å �Æ�æÆ ˚æ��Å� o�Æ�� ŒÆd ¼æØ��. (Dem. 16–21)

The broad-wayed earth gaped openOn the Nysian plain, where sprang forth the Lord, the Host of many,With his immortal horses, Kronos’ son who has many names.Seizing her against her will upon his golden chariot,He carried her away lamenting. And she cried out with a shrill voice,Calling upon her father, the son of Kronos, highest and best.

In their dramatic force these opening lines have some similaritywith the introduction to the Hymn to Apollo, which portrays thegod’s terrifying entry to Zeus’ palace and the other deities’ responseto this. But this scene ends with calm restored, and acts as a preludeto the story of Apollo’s birth, whereas in the Hymn to Demeter thenarrative continues with the same intensity, up to the point wherethe mountains and sea re-echo Persephone’s cries and Demeter atlast hears:

O�f � Ø� ŒæÆ��Å� ¼å� �ººÆ��, Içd �b åÆ��ÆØ I�æ��ÆØ Œæ��Æ �Æ&Ç�� å�æ�d ç�ºfi Å�Ø,Œı�� b Œ�ºıÆ ŒÆ�’ Iç��æø� ��º��’ þø�,��Æ�� ’ u �’ �Nø�e K�d �æÆç�æ�� ŒÆd �ªæ��. (Dem. 40–3)

And a sharp grief seized her heart, about her ambrosial locksHer headdress was torn apart by her hands,She cast a dark covering over both her shoulders,And sped like a bird over land and water in her search.

Demeter’s grief at her daughter’s loss is intensified when she learnsfrom the Sun god who her abductor is.Her grief and her anger againstZeus are leading motifs in what follows. Ultimately they lead her tocreate a famine, which continues until Zeus sends Hermes to bringPersephone back to the upper world (305–83). Before this, however,come the central episodes of the poem: Demeter’s visit to Eleusisdisguised as an old woman, her nursing of Demophon, the infant sonof king Celeus, her attempt to immortalize Demophon, which failswhen his mother Metaneira discovers what she is doing, and finally

The Homeric Hymn to Demeter 45

the revelation of her true identity and the building of her temple atEleusis (91–304).

In these scenes the pace of the narrative is more leisurely. In someways they resemble the more domestic episodes in the Odyssey, whoseinfluence they seem to show. There is also a more cheerful atmos-phere, in the description of the daughters of Celeus, whose youthfulinnocence and grace resembles that of Persephone and her compan-ions, their courteous dialogue with the goddess, the jests of Iambewhich cheerDemeter and make her laugh, and the miraculous growthof the baby Demophon. But in counterpoint to this is the continuingtheme of Demeter’s sorrow, and the whole atmosphere is changedwhen Metaneira detects Demeter hiding her son in the fire, and theanger of the goddess at being thwarted is directed towards her.Demeter’s self-revelation (in which she castigates mankind for theirfolly) leads to a dramatic scene of domestic confusion and terror,strongly contrasted with the previous order and harmony of the royalhousehold.

The last part of the narrative (305–489) reverts to the theme ofDemeter’s anger against Zeus and her withdrawal from divine society.But whereas in the opening scenes she was a helpless victim of Zeus’plan, here the tables are turned. The narrative of the events leading upto Persephone’s return progresses in a fairly traditional manner, withmessenger scenes, journeys and speeches (some abbreviated), andwith some epic breadth, as in Persephone’s extended recapitulationof how she was carried off (405–33) and the threefold recurrence ofthe conditions for her future life above and below the earth (398–400,445–7, 463–5).Demeter’s grief and anger eventually give way to joy ather daughter’s recovery, and this is mirrored in the return of life tothe natural world, the two goddesses’ ascent to Olympus, and thepromise of blessings in life and after death to those who enjoy theirfavour. Thus the more positive character of the closing part of theHymn is in marked contrast with the opening scenes. At the sametime, however, the seriousness of death is not avoided. Just as Demo-phon must die, so too Persephone must still spend a third of the yearwith Hades, as ruler of the dead, and those who fail to propitiate herwill suffer punishment for all time to come (364–9). The rites whichDemeter teaches the Eleusinian leaders are strictly secret, and anyonewho is not initiated into these ‘will never have a part in an equaldestiny after death, down in the mouldy gloom’ (�h ��Ł’ ›��ø� Ær�Æ��å�Ø çŁ�� æ ��e Ç�çøfi �Pæ��Ø, 481–2).

46 Nicholas Richardson

In addition to its seriousness and intensity, the Hymn is unusual inthat throughout much of the narrative the focus of attention is on thefemale characters, both divine and human. Hades instigates the ac-tion, but it is Persephone’s cries which echo through the openingscenes. Later on Hades speaks only once, addressing Persephonewhen he has been asked to let her go (360–9). Zeus, whose consentsanctions the drama, is significantly absent when she is carried off anddoes not hear her cries (20–9), and when he later sends a series ofenvoys first to Demeter, then to Hades, and then again to Demeter,we never hear him speak directly: his instructions are reported inindirect speech and in the words of the messengers (314–56, 441–69).In the opening part of the poem Demeter meets Hecate and Helios,both of whom speak to her, Helios at greater length (51–89), andHelios’ attempt to console Demeter by assuring her of the status ofher new son-in-law anticipates Hades’ own words of reassurance tohis bride (82–7, 362–9). But the main function of this scene is totrigger Demeter’s anger and withdrawal from divine society.

In the Eleusinian episode, the main participants are female,Demeterin disguise, Celeus’ daughters, Iambe, and Metaneira. Demophon,although the centre of attention during his divine nursing, can do nomore than cry pitifully when she abandons him (284–91), and kingCeleus appears only at the end, when he carries out Demeter’s instruc-tions to build a temple, again without direct speech (292–300).2

In the final scenes, once Hades has released Persephone, ourattention is focused on the reactions of her and her mother andtheir dialogue. They are joined by Hecate, as Kore’s future attendant,and by Demeter’s mother Rhea, who brings Zeus’ final instructionsand relays them in detail.

In terms of sympathy, the poet encourages us to identify with thefemale characters, especially Demeter and Kore in the portrayal oftheir suffering, and Metaneira in her anxiety and subsequently hergrief for her son. One might see a parallel between this Hymn and thelonger Hymn to Aphrodite, since both goddesses are portrayed asvictims of Zeus’ plans. Whereas Apollo and Hermes in their Hymnsboth perform exploits which proclaim their character and powers,Aphrodite is ironically the helpless victim of her own power of love,

2 On the direct and indirect speeches in the Hymn see especially Beck (2001), andalso Richardson (1974), 41, with his comments on Dem. 172–3, 314–23, and Nünlist(2004), 38–9.


and the characters of Demeter and Kore could almost be defined interms of suffering. In the Hymn to Apollo also Leto is portrayed aswandering helplessly around the Aegean in her search for a birthplacefor Apollo (45–9), and as a victim of the jealousy of Hera (92–106).Later, in the episode about the birth of Typhaon, Hera in turn viewsherself as the suffering victim of Zeus’ disdain, and her angry with-drawal from divine society resembles that of Demeter (305–55).

Demeter’s counterpart, Dionysus, is also regularly portrayed as asuffering god, in this case at the hands of mortals, on whom he takesrevenge (cf. Hy. 7), but he is often seen as an ambivalent figure,androgynous in appearance. By contrast, Athene reveals her mascu-line side in the dramatic description of her birth in full armour fromthe head of Zeus (cf. Hy. 28). Thus, to some extent at least, one mightsay that the feminine in these divine stories is associated with suffer-ing and passion, the masculine with action and achievement.

2. RECENT INTERPRETATIONS OF THE HYMN

Recent studies have considered the poem from a range of differentperspectives.3 These include:

(1) the language and style of the Hymn and its place in the devel-opment of early hexameter poetry;

(2) the relationship of the narrative with the cults of Demeter andPersephone, especially those at Eleusis;

(3) the way in which the Hymn defines (or redefines) the rolesof the two goddesses in the Olympian order of the world;

(4) the aspect of ‘gender conflict’ in the Hymn and the view of therape of Persephone as a possible paradigm for human marriage;

(5) the structure of the narrative, and problems concerning itscoherence.

What follows is a brief review of these various forms of interpretation.

3 This essay is mainly concerned with scholarship on the Hymn since the publica-tion of my edition (1974), reprinted with corrections and additional bibliography in1978.


(1) Place in the tradition of early hexameter poetry

In terms of language, this Hymn draws on a traditional body ofmaterial represented both in the Iliad and Odyssey and also in theHesiodic poems, especially the Theogony and Works and Days. TheHesiodic character of the Hymn is quite marked. In addition towords, phrases, and formulae with Hesiodic parallels, analogies be-tween some parts of the Theogony and the Hymn are close enough tosuggest the possibility of direct influence. Direct knowledge of theIliad and Odyssey is also a distinct possibility.4 There are, moreover,striking points of contact with the longer Hymn to Aphrodite, whichare probably best explained as due to the influence of this poem.5 Atthe same time, the Hymn shows a good deal of individual languageand formulaic developments.6

The dating of all these works remains a subject of debate, but a lateseventh-century date for the Hymn to Demeter would fit with theseassumptions, and with other internal evidence.7 It is possible that thepoem was originally composed for performance at an Eleusinianfestival, but whether the poet was himself from Attica, as has beensuggested, remains uncertain.8

Studies of the language of the Hymn in relation to the earlyformulaic tradition in general have emphasized a greater fullness ofexpression, for example in the use of multiple epithets with names ofdeities,9 and the use of repetition as ‘a major literary device’.10 Segalhas analysed in detail the ways in which repeated language is used toarticulate the narrative, and give emphasis to a series of leadingmotifs, such as grief, wrath, withdrawal, light and darkness, honour,mortality and immortality, fasting and eating, and so on. Porter, indiscussing the use of repetition in the Hymn to Aphrodite, suggestedthat this technique is employed in a way which is more appropriate tohymnic than to epic style.11 The Homeric Hymns describe the nature

4 See Richardson (1974), 30–4, 333–5, Janko (1982), 181–3.5 See Richardson (1974), 42–3, Faulkner (2008), 38–40.6 See Richardson (1974), 43–52.7 Cf. Richardson (1974), 5–11, Janko (1982), 181–3. See in this volume Introduc-

tion (p. 10).8 Cf. Richardson (1974), 12, 52–6, Janko (1982), 183, Clinton (1986).9 Cf. Gaisser (1974), Segal (1981).

10 Cf. Richardson (1974), 59–60, Segal (1981).11 Cf. Porter (1949), esp. 269–72.


of a god or gods, and this gives them a certain timeless or reflectivequality, which is combined with their narrative and dramatic functionsinsofar as they also tell of the actions or exploits of the gods. Repetitioncan help to bring out the permanent character and attributes of a deity,or emphasize the power of the gods. Fullness of expression in relationto the gods is also more suitable for hymnic style, as it helps to dignifyor praise them. For example, in the opening scenes of the Hymn toDemeter the frequent multiple epithets attached toDemeter, Zeus, andHades emphasize their natures, Demeter as solemn and graceful god-dess of fruitfulness, Zeus as supreme ruler of the world, and Hades aslord of the dead. At the same time, the recurrent language also bringsout the motifs of wonder at the miraculous narcissus, the violence ofKore’s rape, and the desperation of her cries for rescue. These featuresreflect the fact, well expressed by Parker, that there is ‘a certain feel ofoxymoron’ to the title Homeric Hymns, since ‘the style and manner ofthese poems are, in a very broad sense, those ofHomer and heroic epic;the content—the attributes, powers, cults and histories of gods—israther that of prayers and hymns’.12

(2) Aetiology and ritual

The central episode of this Hymn culminates in the building ofDemeter’s temple at Eleusis and her promise to teach her rites tothe Eleusinians (270–4, 293–304), a promise fulfilled at the end of thepoem, where these rites are said to be strictly secret (473–82). TheHymn has thus been described as ‘the Mysteries’ foundation myth’.13

It has also been argued that other parts of the narrative of Demeter’squest for Kore can be related to the ritual of the two goddesses.14 Anumber of these passages can be linked to those aspects of the cult ofthe goddesses which did not belong to the central and secret rites ofthe Eleusinian Mysteries: for example, Demeter’s wandering withtorches, her fasting and abstention from washing, sitting at the wellcalled Parthenion, and also sitting in silence on a stool covered by a

12 Cf. Parker (1991), 2.13 Cf. West (2003), 7.14 Cf. Richardson (1974), 20–30, and see also his commentary, especially notes on

Dem. 24–6, 40 ff., 47, 48, 98 ff., 105 ff., 126, 153, 154, 155, 169–88, 176–81, 180 ff.,188–211, 188–90, 192–211, 200, 202 ff., 207, 231–55, 265–7, 292–304, 387–404,438–40, 439, 440, 450, 470–82, 474–6, 489.


fleece inside the palace, the jesting of Iambe, and Demeter’s breakingof her fast by drinking a special potion (the cyceon; ŒıŒ��). Othersconcern the various figures who later received honours in the cult atEleusis.

Not all of these connections are necessarily related solely to thevarious stages of the Mysteries themselves. Iambe’s jesting, for in-stance, is said by Apollodorus (1. 5. 1) to be the origin of the jesting atthe women’s festival of the Thesmophoria, although there is alsoevidence for such jesting, or ‘ritual abuse’, as a preliminary to theMysteries, as well as at other festivals of Demeter.15 In the case ofDemeter’s nursing of Demophon, and her failed attempt to make himimmortal, it is tempting to see a thematic connection with theMysteries, in which the role of Demeter and Kore as divine nursesfor the initiates seems to have been prominent, and the benefitsoffered by initiation could be viewed as a kind of substitute for theimmortality which is denied toDemophon, and by analogy to mortalsin general.16

The extent to which the Hymn is concerned with the cults ofEleusis, and more specifically with the Eleusinian Mysteries, hasbeen questioned by Clinton, who argued strenuously in 1986 that‘the story of the Rape was no doubt a standard one, sung all over theGreek world. Our author gave it a partly Eleusinian setting; he mayhave been an initiate, but he was not deeply interested in Eleusis.’17

This view was countered by Parker, who replied that ‘his argument

15 Cf. Richardson (1974), 213–17.16 Cf. Richardson (1974), 26–30, 231–6, Parker (1991), 8–10.17 Cf. Clinton (1986), 48. He gave the following reasons:

(a) The name Persephone, used in the Hymn, is not found in Greek inscriptionsfrom within the sanctuary at Eleusis. The Hymn also disregards the sacrednames Theos and Thea for Hades and Persephone, and does not refer toHades as Plouton, as inscriptions do.

(b) The Rape is located on the plain of Nysa, rather than Eleusis.(c) Hecate is prominent in the Hymn but not found in Eleusinian inscriptions.(d) Eumolpus is mentioned, but not as the first Eleusinian Hierophant.(e) Keryx, the mythical ancestor of the Attic ª�� of the Kerykes, is not

mentioned.(f ) There is no reference to Athens in the Hymn.(g) Triptolemus is listed among the Eleusinian princes, but the Hymn does not

refer to the story of Demeter’s gift to him of grain.(h) At the end of the poem, Paros and Antron are mentioned as cult centres of

Demeter, as well as Eleusis (Dem. 490–1).


that the poet was ill-informed about details of Eleusinian cult andorganization is much weakened by the chronological gap, probably ofat least a century, between the Hymn and the other, chiefly inscrip-tional, evidence to which he appeals; and the instances of good localknowledge weigh against him more heavily than he allows’. He alsopointed out that Attic poets and vase-painters ‘did not treat the name“Persephone” as taboo in an Eleusinian context’, and that ‘Hecate iscertainly prominent in Eleusinian iconography, even if her cult hap-pens not to be attested’. Parker concluded that ‘if one removes Eleusis,the hymn falls to pieces’.18

Clinton subsequently modified his position, to allow for the greaterprominence of Eleusinian themes, but argued that connections be-tween the Hymn and cult have more to do with the Thesmophoriathan the Eleusinian Mysteries.19 In view of the nature of the evidence,and especially the secrecy attaching to a good many rituals of De-meter and Kore, it may be unwise to be too dogmatic about thisquestion. It is probably reasonable to speak, as Brumfield does, of the‘symbolic complexity and multivalence’ of the Hymn,20 in otherwords that it bears a close relationship to the rituals of the twogoddesses in general, but cannot be treated too literally as a guide toany of these. At the same time it is an attractive idea, as expressed byParker, that ‘the Hymn ends with a kind of poetic epopteia, a viewingof the Mysteries, in which the association of corn with a revelationand a promise about the afterlife is fundamental’. His further sugges-tion is that ‘one might see Demeter’s first mention of the rites at 273as the poem’s equivalent to myēsis [i.e. the first stage of initiation atEleusis], the final revelation of 474 as epopteia [the final stage]’.21

Thus the structure of the narrative would actually mirror the experi-ence of the initiates as they progress from anxiety and fear, as in theclimax of the Demophon episode, to hope and joy in the final visionof a better fate.

If the assumptions which have been made about the closeness ofthis Hymn to cult are correct, one might ask to what extent this is aspecial case among the Homeric Hymns. The first Hymn to Dionysus,

18 Cf. Parker (1991), 6 n. 22, and passim for a detailed review of the evidence.19 Cf. Clinton (1992), esp. 13–37, 96–9, 116–20, and (1993), esp. 110–16. He

argued that the jesting of Iambe, the nursing of Demophon, the roles of Hecate andHermes, and the Callichoron Well (Dem. 272) all fit better with the Thesmophoria.

20 Cf. Brumfield (1981), 240.21 Cf. Parker (1991), 13.


of which only fragments survive, may have told the story of how Herawas bound fast to a throne made by Hephaestus, who was subse-quently persuaded to release her by Dionysus. This myth in turn hasbeen linked to the ritual of the Samian Tonaia, in which the goddess’sstatue was taken down to the seashore and bound with withies.22 Thefinal part of this Hymn, preserved in the Leiden manuscript (M), alsocontains a passage connecting the narrative of the poem with theorigin of the biennial festival ofDionysus (Dion.D 11–12). TheHymnto Apollo is very much concerned with the god’s cults, especially thoseof Delos and Delphi, and in addition to the extended passage aboutthe Delian festival, the later part of the poem is full of references toactions of the god which were commemorated in ritual. It is thusquite heavily aetiological in character.23 The Hymn to Hermes chartsthe origin of a series of inventions created by Hermes, and thesacrifice by Hermes of two of Apollo’s cattle by the river Alpheios isquite probably an ÆY�Ø�� for the cult of the Twelve Gods at Olympia.But in the Hymn to Demeter it remains striking how closely narrativeand cultic elements seem to be interwoven. Some later writers speakof how the story of Kore’s rape and Demeter’s search for her wascommemorated during the Mysteries, and the initiates may well haveseen themselves as to some extent participating in these events.24 Ithas also been suggested that this Hymn in itself is designed as an ‘actof cult’, in view of the close connection at the end between the doubleproclamation of blessing and prosperity for those favoured by thegoddesses (Dem. 480–2, 486–9), and the closing prayer of the poet fortheir favour (490–4): the performance of the poem itself evokes theprosperity which it describes.25

(3) The Hymn and the order of the world

It has been suggested that the Homeric Hymns as a type of earlyhexameter poetry form a complementary genre, which together withthe Homeric epics and the Hesiodic poems charts the nature of the

22 Cf. West (2001a), esp. 2–4, and in this volume (p. 33).23 See also Chappell in this volume (esp. pp. 64–6).24 Cf. Richardson (1974), 25–6, Parker (1991), 4–5.25 Cf. Calame (1997), 130–3. This aspect of ‘cult hymns’ is emphasized by Furley

and Bremer (2001), i. 6–8. In this respect their exclusion of the Homeric Hymns fromthis category may be open to question.


gods and their relationship with mankind. The four longer narrativeHymns, in particular, define their deities’ place within the Olympianpantheon, under the direction of Zeus, and so are concerned with ‘thepolitics of Olympus’.26 The Hymn to Demeter fits into this pattern,since as a result of the events which it describes the cosmic orderundergoes a significant development. At the end of the Hymn the twogoddesses are established firmly as members of the divine society onOlympus (483–6), but Persephone’s situation is unique in that she isHades’ wife, and as such must return to the Underworld for part ofeach year. At the same time the gulf between gods and men, becauseof men’s mortality, is not breached (as Demeter wished to breach itfor Demophon), but instead a new relationship is established formankind after death, thanks to the special favour of Demeter andKore. Moreover, the sufferings of the two goddesses bring them into acloser sympathy with mankind in their plight, since Persephoneherself has been carried off by the lord of the dead, and regularlycrosses the boundary separating the upper and the lower worlds. It isbecause of their closer identification with men, and Kore’s status as anintermediary figure, that Demeter and her daughter can offer mortalstheir special affection and favour, both in life and after they die.27 TheHymn thus has its place from this point of view within the widercontext of early hexameter poetry as a whole.

(4) Marriage and death in the Hymn

Another branch of modern scholarship has seen in the story of Kore’srape a paradigm for human marriage as a form of ‘symbolic deathbefore a symbolic rebirth and reincorporation into a new householdas wife and mother’.28 Other Greek myths of the abduction of a bride,or of a father’s sacrifice of his daughter, have been compared to thisone and interpreted in similar terms. This approach views the story asone of gender conflict, as in some of the cosmological myths inHesiod’s Theogony, and also in Aeschylus’ Oresteia.29 The intenserelationship of mother and daughter is opposed to the masculine

26 Cf. Clay (1989), esp. 15–16, 267–70, and in this volume (Ch. 11). For somequalifications to this approach cf. Furley in this volume (esp. pp. 208–10).

27 Cf. especially Rudhardt (1994), 198–211, and see also Alderink (1982).28 Cf. Foley (1994), 104, and passim.29 Cf. Foley (1994), 112–18.


roles of Zeus, Hades, and also Helios (who supports Hades and Zeus).Demeter’s resistance ultimately forces Zeus to yield to a compromise,whereby Persephone is shared between heaven and underworld,rather as a reconciliation is effected between Zeus and the Erinyesat the end of the Oresteia, and they acquire a new role and honours asbeneficent powers. Given the close analogy and sometimes equationof Demeter and Erinys this parallel is striking, especially as theAeschylean conflict stems originally from Agamemnon’s sacrifice ofhis daughter Iphigeneia, and the anger of her mother Clytemnestra.The Hymn might even be regarded as a kind of dramatic trilogy inminiature, with loss and conflict followed by an intermediary devel-opment, and then final resolution and reconciliation.

The view of Persephone’s rape as a paradigm for marriage ingeneral is more debatable. The imagery and rituals associated withdeath and marriage have elements in common, and this is also true ofmarriage and various mystery cults. But there does not seem to beclear evidence to show that the ancient Greeks in the historical periodviewed the violent abduction of Persephone as a model for marriageas an institution in itself. To maintain it therefore one needs to arguethat the myth is the reflection either of a deeper psychological reality,or of a more primitive stage in Greek society, as some scholars havedone. It is questionable how relevant this is to interpretation of theactual narrative of the Hymn.30

(5) The coherence and significance of the Hymn

The structure of the Hymn has been criticized on the grounds that thesequence of events is not always clearly motivated. In several otherversions of the myth, Demeter comes to Eleusis or another place insearch of her daughter and receives information about the rape, forwhich she rewards the local inhabitants with the gift of agriculture, orestablishes her rites.31 In the Hymn, however, she learns from Heliosthat Hades has abducted Kore. She then withdraws herself fromdivine society and visits Eleusis. No explicit reason is given for this,

30 Cf. the criticisms of Clay (1989), 209–11; also Calame (1997), 124–5, and Kledt(2004), 33–4.

31 Cf. Richardson (1974), 81–2, 174.


although we subsequently learn that she wishes to immortalize Demo-phon. Again, however, we are left to speculate about why she doesthis. Having failed in this attempt she commands a temple to be builtand promises to teach the Eleusinians her rites. But at this point sheagain withdraws, this time within her new temple, and creates thefamine which forces the hand of Zeus. In other myths of a similarkind from Greece and the Near East, the death of crops and vegeta-tion arises more directly from the disappearance or wrath of a power-ful deity, and just as Persephone’s return in the Hymn is followed bythe renewal of life in the fields, so too one might expect that herabduction would have resulted directly in the famine.32

The Hymn presupposes that men already know the art of agricul-ture, whereas from the middle of the sixth century onwards theEleusinian hero Triptolemus receives the gift of corn from Demeterand then teaches it to the rest of the world.33 It was on the RarianPlain, near Eleusis, that the first sowing of crops was said to havetaken place, and it is here in the Hymn that Rhea meets Demeter andgives her Zeus’ instructions, which lead to the rebirth of the crops(Dem. 450–6, 470–3). This suggests that the poet is aware of the formof the story in which agriculture was first given to men after Kore’srape and return, but only alludes to this obliquely.34

It is possible to speculate about the relationship between theHymn and other versions of the myth, in order to explain some ofthe twists and turns of the narrative. But the poet has chosen to tellthe story in a particular sequence and it is only fair to assume that hehas his own reasons for doing so. Moreover the longer HomericHymns differ from Homeric epic in the manner with which theytreat everyday realism, as Parker has observed.35 As he says, ‘theypresent divine myths, stories about the organization of power inthe world, with all the freedom of fantasy that such serious subjectsdemand’. In the Hymn to Hermes, for example, Hermes is ‘hungryfor meat’ (Œæ��ø� KæÆ��Çø�: 64), steals some of Apollo’s cattle, androasts two of them, but then does not eat them despite his hunger(115–37). This may be because he is a god, and gods normally do not

32 Cf. Richardson (1974), 258–60.33 Cf. Richardson (1974), 194–6.34 Cf. Richardson (1974), 297–8.35 Cf. Parker (1991), 4.


consume the main parts of a sacrificial animal, but the poet does notsee the need to explain this. Equally, he does not have to explain allof Demeter’s behaviour to his audience. Part of a god’s power liesprecisely in the fact that their purposes are not always clear to men.At the same time, it is possible to view the actions of Demeter interms of the results they produce.36 The poet wants to give a centralplace to the story of Demophon, with its potential implicationsfor mankind’s lack of immortality, counterbalanced by the specialfavour of the divine nurses for the initiate. Demeter’s visit to Eleusisresults in the institution of her cult there, with its universal import-ance for mortals in general. The motif of the famine is postponeduntil after this central section of the poem. It is also possible, as wehave seen, that the poet wished the structure of his work to reflectthe two main stages of initiation at Eleusis, at the conclusion of thecentral and final parts.

Whatever the truth about such speculations may be, the Hymnsurely gains rather than loses in its effectiveness as a result of themysterious quality of its narrative. Demeter’s visit to Eleusis, forexample, is based on patterns which are typical of early epic narra-tive.37 But into these patterns is woven a sequence of motifs which arequite extraordinary, because this goddess is no ordinary visitor andher cult is one shrouded in awe and mystery. It is the achievement ofthe poet that he has succeeded in evoking so powerfully this sense ofmystery and awe, in a work which uses traditional narrative tech-niques in such an original way.

3. CONCLUSION

The range of interpretations evoked by the Hymn in recent times is atestimony to its appeal, both as a poetic narrative and also as a workwhich sheds light on Greek religious belief and practice. The poet hassucceeded in investing a story which might have had purely local

36 Cf. Parker (1991), 11.37 Cf. Richardson (1974), 179–80, 190, 205–6, 207–8, 226–7, 339–43, etc. See also

Lord (1994), and Foley (1995), 136–80, esp. 160–80.


relevance with a more universal significance, thus giving credibilityand dignity to the claim of Eleusis to be the most important cult-centre of Demeter and Kore. Above all the Hymn in itself could beregarded as an ¼ªÆºÆ, a work of art offered to the goddesses inwhose honour it is composed:

ª�æ�ø� �b Ł��E Œ�ººØ�� I�Ø�Æ�

For songs are the finest of honours for the gods (Theoc., Id. 22. 223).38

38 In addition to the books and articles cited above, the following selection may befound useful: Sowa (1984), Motte (1986), Saintillan (1986), Cheyns (1988), Felson-Rubin and Deal (1994), De Bloois (1997), Nickel (2003).


4

The Homeric Hymn to Apollo

The Question of Unity

Mike Chappell

The Hymn to Apollo is one of the most studied of the HomericHymns. Until recent years scholarship mostly focused on the problemof the hymn’s unity.1 The most important more recent works, thoseof Miller and Clay, have continued to discuss the question of unitybut have taken a more unitarian view than earlier studies, andattempted to demonstrate the Hymn’s unity by treating it as a unifiedwork of art that has a coherent plan and themes linking the twosections. Whether or not one accepts their arguments for the Hymn’sunity, it is a welcome change to read works that do look at the Hymnas a planned poem with definite themes rather than merely as aclumsily stitched together patchwork, and their studies have greatlyincreased our understanding not only of the Hymn to Apollo but allthe Homeric Hymns and their relationship to other archaic Greekpoetry. As their arguments for unity look at the themes of the poem aswell as its structure, after looking at the unity problem I shall combinea discussion of some important themes with analysis of whether thetreatment of these themes does indeed support the idea that the hymnis a well-planned unity. Clay also sees the Hymn’s combination ofDelian and Delphic themes and emphasis on the spread of Apollo’s

1 For a detailed survey of the problem and of earlier opinions on it, see Förstel(1979), 20–62. Important more recent contributions are Janko (1982), Thalmann(1984), Miller (1986), Clay (1989), Stehle (1996). Clay summarizes her views on thehymn in Clay (1994).

worship in terms of the conscious Panhellenism that in her view is afeature of the longer Hymns.2

Whether an original unity or not, theHymn is clearly divided into aDelian and a Pythian section.3 D. (1–181) narrates Apollo’s birth onthe island of Delos. It tells how Leto searched for a place to give birth,being rejected by many places because of their fear that Apollo wouldbe a powerful and dangerous god who might scorn his birthplace(19–48). Eventually, the small rocky island of Delos agrees to be thegod’s birthplace, but only after extorting an oath from Leto thatApollo will honour her and be worshipped there (49–96). Hera triesto delay the birth by keeping back Eileithyia, but after the othergoddesses fetch her, Apollo is born. As soon as the infant god tastesnectar and ambrosia he develops miraculously, bursting his swad-dling bands and declaring that his main attributes will be the bow andlyre, and that he will prophesy to men the will of Zeus (97–132). Thencomes a list of Apollo’s favourite haunts, culminating in the descrip-tion of a festival on Delos and the chorus of Delian maidens, to whomthe poet says farewell. He also asks them to praise him as a blindsinger from Chios (133–78).

P. (182–546) tells of Apollo’s search for a place to found his oracle.His route takes him from Olympus, through Pieria and Iolcus toEuboea, and then up to Delphi through Boeotia and Phocis. He firstchooses the spring Telphousa in Boeotia as a suitable site, but Tel-phousa persuades him that there is a better site on the slopes ofParnassus (244–76). When Apollo arrives there he kills the snakethat was a menace to local people, then lays the foundations of histemple, which is completed by the heroes Trophonius and Agamedesand the people dwelling nearby (277–387). After this, the god decidesto find priests for his temple and chooses a group of Cretans sailing toPylos (388–99). In the form of a dolphin he hijacks their ship andtakes it up the west coast of the Peloponnese to Crisa (400–39). Therehe leaps out of the ship in the form of a star and kindles the fire in histemple. He returns to the Cretans in the form of a vigorous young

2 See below, pp. 64–7. On this issue, cf. the Introduction in this volume (pp. 19–22).3 I shall use the abbreviations D. for the Delian section, and P. for the Pythian.

Unless otherwise indicated, translations of the Hymns are taken from Evelyn-White(1936).

60 Mike Chappell

man, reveals his identity and tells them that henceforth they will serveat his shrine and never return home (440–85). He also commandsthem to found an oracle on the shore to him as Apollo Delphinios,because he appeared to them in the form of a dolphin. He then leadsthem in procession to Delphi, where he tells them that they willlive off the offerings of pilgrims and warns them that if theybehave hubristically they will live under the yoke of new masters(485–544).

The first scholar to question the Hymn’s unity was David Ruhnkenin 1782. He based his argument mainly on lines 165–78, on thegrounds that they contain all the features of a typical hymnic ending.4

He also argued that the first hymn praises the Delian Apollo and thesecond the Pythian Apollo; and that the independence of D. is con-firmed by the external evidence of Thucydides and Aristides.5 In thenineteenth century, scholarship on the Hymn mainly followed theanalytical trend of Homeric scholarship, and some scholars divided itinto more than two original hymns; as early as 1786 Groddeck arguedthat it was made up of five fragments of hymns.

It is not necessary to postulate such a patchwork of fragments.D. and P. are both coherent wholes, except that P. lacks a properbeginning (it has a prologue, but no hymnal opening). As is usual forthe Hymns, the typical present-tense sections come at the beginningand end, and as with the other long Hymns they frame a developednarrative. Each has a coherent theme which is announced at thebeginning and brought to a conclusion by the end. D. has a longerpresent-tense section which describes the occasion of the Hymn’sperformance and refers to the poet himself (146–76). Though itslength and elaboration are untypical, other Hymns do also at theend refer to the performance situation and describe current cultswhich (in the longer Hymns) result from the god’s actions in the

4 Ruhnken (1782). I discuss these features in more detail below (p. 63).5 Thucydides (3. 104) quotes lines 146–50 and 165–72 of the Hymn (with some

variants), and introducing 165–72 says that here the poet finished his praise (K��º��Æ��F K�Æ��ı). Ruhnken and others see this as evidence that Thucydides knew anindependent Delian hymn, but others argue that he was only referring to the end ofthe section praising the Delian maidens. Aristides (Or. 34, 35 Keil) cites 169–72 as theending of a hymn (ŒÆ�Æº�ø� �e �æ��Ø��), but most commentators believe he isdependent on Thucydides here rather than quoting the Hymn first-hand. For theargument that these quotations are evidence of D.’s separate existence, see Jacoby(1933), 689–92; contra, Heubeck (1966). Cf. Clay in this volume (pp. 238–40).

The Homeric Hymn to Apollo 61

narrative (e.g. Dem. 470–95, Hys. 6. 19–21, 13. 3, 24. 4–5, 26. 11–13).Thus the two parts each have a beginning, middle, and end whichmatch those of other Hymns. With the addition of an opening line ortwo to P., both parts would work well as independent Hymns.

Perhaps the most influential separatist theory has been that ofWilamowitz, in his Die Ilias und Homer.6 He argued that the twohymns were not independent fragments later cobbled together; P. wascomposed as an addition to, and imitation of, an originally indepen-dent D. This theory has several attractions. It explains the compositeHymn’s origin as a poetic design rather than an accident of transmis-sion. It accounts for the similarities and parallels between the twoparts, which are too close to be a matter of chance, and for the factthat P. has no proper hymnal beginning (it starts abruptly without theusual announcement of which god will be the subject of the song). Itis also much simpler than some of the complex analytical theoriesabout the Hymn’s genesis. For these reasons something like Wilamo-witz’s theory has remained the most popular among separatists.7

A notable exception is West, who has argued that it was D. that wascomposed later, and modelled on P.8 Later the two hymns were puttogether and some further changes made at that point (notablyaddition of the Typhaon episode at lines 305–54). In his recentLoeb edition,9 West suggests that it was the same poet, Cynaethusof Chios, who first composed the D. section and then later producedthe combined version that we have.

West is not the first to suggest that one hymn was not simply addedto the other, but that the poet who composed the addition also alteredthe original poem. Hermann argued that there was an independentDelian hymn, and a combined hymn that omitted the description oftheDelian festival, as it was only relevant for the original circumstancesof performance (thus consisting of lines 1–139 and 179–546).10 Ascribe who had both versions before him combined them rather thanwrite out lines 1–139 twice. Jacoby combined this theory with thatof Wilamowitz, suggesting that the composer of the P. addition altered

6 Wilamowitz-Moellendorff (1916a), 440–62.7 French scholars have often referred to P. as the ‘suite Pythique’, e.g. Humbert

(1936).8 West (1975).9 WL 9–12.

10 Hermann (1806).

62 Mike Chappell

D., and that traces of the different versions remain in the form ofdoublets in D.11

The heart of the separatist position is that lines 165–78 form, in thewords of Ruhnken, the ‘customary epilogue and ending of a hymn’(solitum hymni epilogum et finem). Certainly they have a strong senseof closure, and have formal elements typical of hymn endings:

(a) Iºº’ ¼ª�Ł’ (‘but come now’) in 165; cf. Dem. 490, Hy. 20. 8.

(b) ¥ºÅŁØ (‘be favourable’) requesting the god’s favour. Forms of¥ºÅŁØ/¥ ºÆÆØ are common in the closing prayer (Dion. D. 8;Hys. 19. 48; 20. 8; 21. 5; 23. 4).

(c) åÆEæ�/åÆ�æ�� (‘farewell’) which occurs in the endings of 27 ofthe 33 Hymns.

(d) the promise to continue singing of the god.

Other features appropriate to the end of a hymn are the shift to thepresent tense, the description of a festival whose origins the narrativehas explained, and the references to the circumstances of perform-ance and the poet himself.12

In favour of the Hymn’s unity, Miller argues that �P º��ø (‘I willnot cease’) in line 177 shows that the poet intends to continue;13 butthis is more naturally understood as the sort of promise to sing ofthe god in the future which is typical of hymnal endings. A preciseparallel is SEG 8. 548. 25 �P º��ø �ª�ºÅ� ��Æ�� ı I��ø� (‘I willnot cease singing of your power’), from the concluding section of ahymn to Isis. At the end of many Homeric Hymns the poet promisesto remember the god as well as to move to another song. It also seemsnatural to take it as parallel to the immediately preceding futures inwhich he promises to spread the fame of the Delian maidens. He saysthat he will praise the maidens (��E �’ ��æ�� Œº� �Y��. 174),but not cease singing of Apollo, just as he will remember the god as hemoves to another subject. If 1–178 existed as an independent hymn,nobody would ever say it seemed incomplete or lacked a properending.

11 Jacoby (1933). His posited doublets are 6/7–9, 10–13/14–18, 72/73–9, 96/98,128/129, 136–8/139 (the last has some manuscript support). Wade-Gery (1936) suggeststhat there are also some doublets in P., notably at 295–9.

12 On the formal aspects of hymnic endings, cf. in this volume Calame (pp. 354–7).13 Miller (1979), (1986), 65–6.


Miller does succeed in showing that the passage has some untypicalfeatures for a hymnic ending. Apollo is not directly addressed, whichis unparalleled in the other Homeric Hymns. It is the Delian maidenswho are addressed by the poet, and the åÆ�æ�� (‘farewell’) is directedto them. As with the gods addressed in other endings, there is arequest for their favour and a promise to sing of them in the future.Miller therefore argues that it is only the Delian maidens (and byextension the Delian theme) of which the poet is taking leave, andthat the passage is a transition to a new theme rather than the end of apoem. These are good points, but I think not enough to counter-balance the many features that give the lines the unmistakable feel ofan ending. It is true that Apollo is not addressed, but the passage stillbegins and ends with him, and we still have a request for his favourand promise to sing of him in future (165, 177–8). I would argue thatthe poet has ingeniously combined his evocation of the circumstancesof performance and address to the maidens with a standard prayer forthe god’s favour and promise to remember him in return.

The other major argument against the Hymn’s unity is the differentgeographical spheres of the two sections. D. is focused only on theAegean and its coasts, and it is only Ionians who visit the great Delianfestival (30–44, 140–64); while P. centres on mainland Greece, not-ably Boeotia and Phocis (especially in 216–99). It could be argued thatthis is just a reflection of their subject matter, but the division neednot have been so clear cut. A poet might have described mainlanderscoming to Delos, or have had Leto travel also over the mainland inher search for a birthplace. Another link would have been to haveApollo go to Delphi after his birth on Delos, as later authors did.14 Inour Hymn, Apollo is said in D. to wander over coasts and islands, anddelight most in Delos; and in P. he is first described (with a presenttense describing typical activities) going to Delphi, and from there toOlympus, from where he begins his journey in search of a site for hisoracle.

On the side of unity, Clay argues that the Hymn treats both Delosand Delphi because its outlook, in common with that of the othermajor Hymns, is consciously Panhellenic.15 Though the two halves do

14 First in E. IT 1234–57, though it appears in fifth-century art (see LIMC s.v.Apollon, nos. 988–97). For later references, cf. e.g. Call. Hy. 2. 103, A. R. 2. 707–8,Hyg. Fab. 140, Serv. ad A. 3. 73, Macr. Sat. 1. 17. 52.

15 Clay (1989), 9–11, 47–9, 92–4, and this volume (p. 246).

64 Mike Chappell

focus on shrines in particular locations, these are shrines that had aPanhellenic and not just local significance. The two sections betweenthem emphasize Apollo’s Panhellenic importance: D. covers theAegean and its coasts, while P. deals first with northern and centralGreece and then the Peloponnese. In both parts the wide geographicalrange of Apollo’s worship is stressed, and his shrines are said toattract pilgrims from far afield, especially at lines 247–53 ¼ 287–93.Though the lines are first spoken when Apollo arrives at Telphousa,this is an aetiological poem and it is the Pythian Apollo and his shrinethat they really glorify by stressing the wide range of pilgrims that theDelphic oracle will attract. This could be understood as an expressionof rivalry with other shrines such as Delos rather than a unifyingPanhellenism.

Clay argues that there would have been no difficulty with a hymndesigned for performance on Delos also narrating the story of Del-phi’s founding, because these were Panhellenic shrines. She states that‘we should not foist off a nineteenth-century parochialism on theGreeks’.16 Yet it is more problematic than she suggests to imaginesuch a combined hymn. Burkert argues that, as the Hymns wereprobably composed for specific local festivals, ‘the Apollo hymn, aswe read it, seems to become an impossibility: the first part in factdescribes the festival and addresses an audience at Delos, but thesecond part presupposes a Boeotian-Phocian setting.’17 It is truethat there was a development of Panhellenic shrines in the archaicperiod, but this does not mean that regional interests and rivalrieswere wholly superseded.18 A hymn performed at Delos would focuson Delian themes, and a Delphic hymn on Delphic.19 Certainly a

16 Clay (1989), 48. Miller (1986), 112–13, expresses similar sentiments. AHS 190–2,Càssola 98, and Förstel (1979), 163, also think it would not have been a problem toperform a combined Delian and Delphic hymn at Delos or Delphi.

17 Burkert (1979), 58–9. This is not an outdated viewpoint. Cf. Graf (2009), 26, ‘Allmajor hymns are intimately connected with a specific local cult about which the texttalks and in which the hymn must have originally been performed’.

18 Panhellenism as a conscious ideology probably only developed after the PersianWars, but from the early archaic period there seems to have been a development ofcultural Panhellenism: cf. Nagy (1979), 6–9, 115–21, Morgan (1993). For a recentsurvey of theories about the rise of Panhellenism, see Mitchell (2007).

19 Apart from D. and Call. Hy. 4 to Delos, no complete Delian hymns havesurvived. Pindar’s Paeans 7b and 12 were probably composed for performance onDelos, and focus on the birth story. Extant Delphic hymns unsurprisingly focus onDelphic themes. The surviving hymns celebrating Delos or Delphi are collected, withdiscussion and commentary, in Furley and Bremer (2001).


Delphic hymn might mention Apollo’s birth on Delos, but not as amajor theme;20 and it is unlikely that a poet performing on Deloswould follow his Delian hymn with a much longer narration of thefounding of the Delphic oracle. Equally, a hymn composed for per-formance at Delphi would hardly state that Delos was Apollo’sfavourite place (as in 146) and describe a Delian festival as if it werethe occasion of the hymn’s performance.

It is reasonable enough to argue that the Homeric gods are Pan-hellenic, in the sense that they are pictured largely on Olympus, andreferences to their local cults are avoided. This is well-established, asClay shows by citing Rohde’s formulation.21 However, the Hymns arenot precisely the same as Homer in this respect. They do use the sameapparatus of the Olympian family, but also show more interest inparticular localities and cults, as one might expect of hymns. Asindicated above, both parts of the Apollo Hymn have a strong senseof place, and the focus is much more on the god at Delos or Delphithan on Olympus, though both parts do open with an Olympianscene to remind us of this background. In the Apollo Hymn thePythian poet shows interest in and knowledge of Boeotian topog-raphy and cult, as well as the focus on the location of Delphi. Apollofounds cults to himself at Telphousa (375–87) and on the shore atCrisa as Delphinios (490–6), and the shrine at Onchestus of Poseidonis also described (229–38). Nor is it correct to say that ‘the hymnpresents the Delian panegyris as a Panhellenic affair rather than alocal festival’,22 since only Ionians are said to be there.

Clay also argues in favour of the Hymn’s Panhellenism that itpresents the foundations of Delos and Delphi as purely Apollinefoundations, with no local predecessors. In both cases the barrennessof the site is stressed (53–5, 529–30), and in P. no mention is made ofany previous owners of the oracle. Thus local traditions are playeddown in favour of the Olympian and Panhellenic. However, the poetmakes no explicit comments to this effect, and the poverty of the sitesof Delos and Delphi can equally be understood in other ways—forexample Miller interprets it as a way of praising Apollo by stressing

20 e.g. Limenius’ Delphic paean describes the birth at 5–10, and Alcaeus’ seems tohave begun with the birth, according to the summary in Himerius (fr. 307c Voigt ap.Him. Or. 48).

21 Clay (1989), 9, citing Rohde (1898), 38–9.22 Clay (1989), 93.

66 Mike Chappell

how he, by his power, has made unpromising locations into famousshrines.23 There is no need to see this praise of Apollo as a promotionof Panhellenism, nor is it a strong argument in favour of the Hymn’sunity.

Stehle rightly argues that D. (which she regards as originallyindependent), with its purely Ionian festival and emphasis on Delosas Apollo’s favourite place, does not have a Panhellenic outlook. Onher view P., which is consciously Panhellenic, was composed as anaddition to D., so it is the combined Hymn that is fully Panhellenic.24

This theory does partially solve a problem facing separatist theories.I have argued that a unified poem celebrating both Delos and Delphiat some length is unlikely, yet, in that case, why would someone havecreated the combined Hymn, and what performance context wouldit suit?25 Someone might have created the combined Hymn with aself-conscious Panhellenic agenda. If this is so, we could say that thewhole Hymn is Panhellenic in celebrating both Delos and Delphi,but its two parts, or at least D., are not—this only makes sense on aseparatist view of the Hymn’s genesis. So if there is Panhellenism inthe Hymn it need not be a unifying theme, but rather an argument forits original disunity.

If the two parts of the Hymn are originally independent, can wedetect linguistic or stylistic differences between them? Linguisticcriteria might provide a more objective basis for deciding the unityquestion. The most important linguistic analysis is that of Janko,26

who does find differences between D. and P. in various linguisticcriteria. The largest variation is in their use of nu-mobile beforeconsonants: D. has 60. 8 occurrences per thousand lines, P. only 13.7. Nu-mobile seems to be a feature largely of Attic-Ionic from an earlyperiod, so this suggests that D. has an Ionian origin, and P. a main-land one, which is what we might expect from their content. Thiscriterion seems to be more reliable than most of Janko’s as anindicator of a poem’s geographical origin, since it does not cut acrossdialectal boundaries in the way that (e.g.) the treatment of pre-vocalic

23 Miller (1986), 22–3, 75.24 Stehle (1996), 177–96. She describes D. as ‘a mimesis of communal poetry and

correspondingly anti-Panhellenic’ (178). Mitchell (2007) similarly argues that D., theearlier poem, is concerned with Ionian identity, and it was the poet of P. who laterjoined his poem to D. and had a more Panhellenic agenda.

25 See below (p. 72) for one proposed answer to this question.26 Janko (1982), 99–132.


digamma does.27 Hesiod’s language is Ionicized in many respects, butnot this one: Janko’s figures for the Theogony and Works and Days are21. 5 and 21. 7 respectively, while for the Iliad and Odyssey they are35. 6 and 37. 2. The Delian section’s higher figure probably reflects itslater date. It is hard to draw conclusions from Janko’s other criteria,since P. varies a great deal in its results relative to other Hymns; butsince this variation is different from the results for D., it is anotherindication of disunity. So the linguistic evidence provides some argu-ments against unity, but has to be used with caution given the short-ness of the Hymn and of D. in particular.28

Many scholars have detected differences in style and ethos betweenthe two parts, although these judgments are inevitably subjective.A common view is that D. is more ‘lyric’ in style, and is also the betterpoem.29 It has been compared to lyric poetry in its swift movement andrapid transition from one topic to another, and the use of apostrophe(though this continues until line 282, well into P.).30 The poet’s refer-ences to himself and his audience also seem more characteristic of lyricthan epic. P. is often more Homeric and leisurely in style, especially inthe episode of the Cretan sailors (388–544), but this could be due to‘epic’ subject matter such as the voyage and landing scene, where theclosest Homeric parallels are found (e.g. 421–7 and Od. 15. 295–300;440–7 and Il. 4. 74–9; 452–5 and Od. 3. 71–4; 503–7 and Il. 1. 433–7,485–6). The Pythian poet also displays a fondness for etymology thatis absent from D. in his explanations of the cult titles Pythios andDelphinios (371–4, 493–6). Miller criticizes ‘the assumption that stylis-tic differences within a literary work can be accounted for only bypositing multiple authorship’.31 This is a fair point, but stylistic differ-ences are more of an argument for disunity than for unity; and more

27 Janko (1982), 64–8. Hoekstra (1965), 78 ff. argued that the frequency of thisusage increased in the course of the epic tradition; Janko shows that this is true at leastfor the Ionian branch.

28 Some examples of Janko’s figures are: % neglect of digamma D. 34.6, P. 22.4; %genitives in -�Ø� D. 34.4, P. 45.6; % genitives in -�ı D. 28.6, P. 52.0; % a-stem acc.plurals before vowels D. 33.3, P. 55.6.

29 D.’s superiority: Wilamowitz-Moellendorff (1916a), 455, Jacoby (1933), 751,Van Groningen (1958), 319–20, Wade-Gery (1936), 56, Càssola 98. Contra: AHS 192.

30 Rapid transitions: 14, 19, 45, 133, 146, and 179 if it is part of D. Apostrophes:19–29, 127, 140–8, 179–81, throughout 207–46, 281–2. In D. the apostrophe is largelyconfined to the opening and closing present-tense sections, whereas in the first part ofP. it is used for the main narrative.

31 Miller (1986), 115.

68 Mike Chappell

subtle linguistic differences, such as the use of nu-mobile, are certainlynot the result of any conscious stylistic variation.

There are nonetheless many parallels in structure and phraseologybetween the two parts. Both open with a scene on Olympus in whichApollo arrives and has an effect on the other gods, and his parentsrejoice in their son (1–13, 186–206). Both have a ‘choice of theme’section introduced by the line �H ��æ �’ ��ø, ��ø �hı��K��Æ (‘How then shall I sing of you who are in all ways a worthytheme of song’, 19/207). Leto wanders through many places lookingfor a place to give birth, as does Apollo in the search for a site for hisoracle, as discussed above.32 Both encounter and have a dialogue witha local goddess (Delos, Telphousa) at the place they choose (inApollo’s case not his final choice). Leto persuades Delos to acceptApollo, while Telphousa persuades Apollo not to choose her. In bothcases there is a promise to build a temple and sacred grove at the site,and here the verbal parallels are close, as are the parallel phrasesreferring to the people who will bring hecatombs to the future shrine,and the founding of the oracle:

ÆN � Œ’ ��ººø�� ŒÆæª�ı �Åe� �åfi Å�ŁÆ,¼�Łæø�� Ø �� IªØ��ı�’ �ŒÆ��Æ K�Ł��’ Iª�Øæ��Ø, Œ��Å � ��Ø ¼�� ÆN��Å�F I�Æ&��Ø, ��Œ��Ø Ł’ �¥ Œ �’ �åø�Ø�å�Øæe ¼�’ Iºº��æ�Å , K��d �h ��Ø �EÆæ ��’ �s�Æ (Apoll. 56–60)

But if you have the temple of far-shooting Apollo, all men will bring youhecatombs and gather here, and incessant savour of rich sacrifice willalways arise, and you will feed those who dwell in you from the hand ofstrangers; for truly your own soil is not rich.

› �’ ¼ººÅ� ªÆEÆ� Iç��ÆØ, fi w Œ�� –�fi Å �ƒ��Æ�ŁÆØ �Å�� ŒÆd ¼º��Æ ��æÅ��Æ· (Apoll. 75–6)

he will go to another land such as will please him, there to make histemple and wooded groves.

Iºº’ �Y �Ø �ºÆ�Å ª�, Ł��, ªÆ� ‹æŒ�� O��ÆØ,K�Ł�� Ø� �æH�� Ø� ��æØŒÆººÆ �Å��

��ÆØ I�Łæ��ø� åæÅ��æØ��, ÆP�aæ ��Ø�Æ��Æ K�’ I�Łæ��ı , K��d q ��ºı��ı� ��ÆØ. (Apoll. 79–82)

32 See above (pp. 60–1).


But if you will dare to swear a great oath, goddess, that here first he willbuild a glorious temple to be an oracle for men, and then [buildtemples] among all mankind; for surely he will be greatly renowned.(trans. adapted from Evelyn-White, 1936)

��B �’ K�d ¸Åº��øfi ��øfi , �� Ø �På –�� ŁıfiH��Æ�ŁÆØ �Å�� ŒÆd ¼º��Æ ��æ��Æ. (Apoll. 220–1)

You stood in the Lelantine plain, but it pleased not your heart to make atemple there and wooded groves.

�B �’ K�d '�ºç��Å · ��ŁØ ��Ø –�� åHæ� I��ø��Æ�ŁÆØ �Å�� ŒÆd ¼º��Æ ��æ��Æ. (Apoll. 244–5)

Then you went to Telphousa; and there the pleasant place pleased youfor making a temple and wooded groves.

'�ºç�F�’, K�Ł�� c çæ��ø ��æØŒÆººÆ �Å��I�Łæ��ø� ��F�ÆØ åæÅ��æØ��, �¥ � �Ø ÆN��K�Ł��’ IªØ��ı�Ø ��ºÅ��Æ �ŒÆ��Æ

(Apoll. 247–9 ~ 258–60, 287–9)

Telphousa, here I intend to make a glorious temple, an oracle for men,who will always bring to me here perfect hecatombs.

These are not standard formulaic phrases, so the parallel passagescannot be independent. Another parallel is the idea that the inhabit-ants of the shrine will feed on the sacrifices brought by worshippers(59–60, 528–36).

These parallels between D. and P. are too great to be coincidental,and show that either the Hymn is a genuine unity or that one poetimitated the other—the two parts can never have been wholly inde-pendent poems joined together by a third person. This means thatthe most economical separatist hypothesis is that one poem wascomposed as an addition to the other, though it is also possible thata poet composed an imitation of one hymn which was not originallyintended to be joined to it. As noted above,33 most separatist criticshave followed Wilamowitz in seeing P. as an addition to a pre-existing D., but West has argued for the contrary.34 His main argu-ment is that some of the parallel passages are less awkward and fittheir context better in P. than in D., and so P. is likely to be theoriginal. This applies especially to the mention of an oracle, and the

33 See above (pp. 62–3).34 West (1975).

70 Mike Chappell

idea that the place’s inhabitants will be able to live off visitors’sacrifices. The first is clearly more natural in P., and West arguesthat Delos does not really express concern about her inhabitants’livelihood, only her own safety and reputation (62–78). In P., onthe other hand, the Cretans naturally express concern about howthey are to live, and it was a common topos that the Delphic priestslived off the offerings of pilgrims (535–7).35 These arguments havesome weight, but are not really conclusive. Förstel argues that themention of Delos feeding her inhabitants from visitors fits the em-phasis on her rockiness and barrenness.36 Janko rejects West’s view aswell, arguing on linguistic grounds that the Delian passages areprobably prior.37

West also points out that P. does not seem designed to follow onfrom D.: it does not pick up on anything in D. but begins abruptlywith Apollo going to Delphi.38 He is right to say that this gives theimpression of an independent poem that has lost its beginning. Onemight then ask: ifD. is designed to precede P., why does it have a clearending? The foundation of the separatist case, as discussed above,39 isthat D. is meant to end at line 178, in which case it was not composedto go before another poem. West’s answer to this is that D. wascomposed with P. in mind, but originally intended as an independentpoem. Later another poet, or possibly the same one, joined the twopoems together.

West further suggests that the author of the original Delian hymnwas the rhapsode Cynaethus of Chios.40 Our information aboutCynaethus comes from a scholium to the first line of Pindar’s secondNemean, which states:41 ‘Homerids they called in ancient times the

35 See the references listed in AHS 264–5.36 Förstel (1979), 293–5 n. 14, lists his objections to West’s arguments.37 Janko (1982), 109–12.38 West (1975), 162.39 Above (p. 63).40 This seems more likely than the view of Wade-Gery (1936) that Cynaethus was

the author of P. If either of the hymns is by this Chian Homerid, it must surely be D.41 Sch. Pi. N. 2, 1c, III 29, 9–18 Dr (¼ FGrH 568 F 5): % ˇÅæ��Æ �º�ª�� e b�

IæåÆE�� f I�e ��F % ˇ�æ�ı ª��ı , �Q ŒÆd �c� ��Å�Ø� ÆP��F KŒ �ØÆ��åB fi q��, ��a�b �ÆF�Æ ŒÆd �ƒ ÞÆłøfi ��d �PŒ�Ø �e ª�� N �OÅæ�� I��ª�� . K�ØçÆ��E �b Kª��ƒ ��æd ˚��ÆØŁ��, �o çÆ�Ø ��ººa �H� K�H� ��Ø��Æ��Æ K�Æº�E� �N �c� % ˇ�æ�ı��Å�Ø�. q� �b › ˚��ÆØŁ� �e ª�� E� , n ŒÆd �H� K�ØªæÆç��ø� % ˇ�æ�ı��ØÅ��ø� �e� �N ��ººø�Æ ª�ªæÆçg o�� I�Æ�Ł�ØŒ�� ÆP�fiH. �y�� s� › ˚��ÆØŁ� �æH�� K� $ıæÆŒ��ÆØ KæÆłfi��Å�� a % ˇ�æ�ı ��Å ŒÆ�a �c� ŒŁ� !Oºı�Ø��Æ, ‰ % I��æÆ�� çÅ�Ø�.


descendants of Homer, who also used to sing his poetry by right ofsuccession; afterwards also the rhapsodes who no longer claimeddescent from Homer (were called Homerids). Prominent becameCynaethus and his like, who are said to have interpolated many versesin the poetry of Homer. Cynaethus was Chian by origin, and hewrote, among the works attributed to Homer, the hymn to Apolloand fathered it on Homer. This Cynaethus was the first to reciteHomer’s verses at Syracuse in the sixty-ninth Olympiad (504/1 BC), asHippostratus says.’42 This date has often been rejected as too late bothfor the hymn’s composition and the first performance of Homer inSyracuse, but it is defended by West and Burkert.43 Hippostratus mayhave taken the information from some list of victors in rhapsodiccontests, and we know that not long before this Hipparchus institutedHomeric recitations at the Athenian Panathenaea.44

Burkert and Janko have independently suggested a possible contextfor the performance of a combined Delian and Pythian hymn, whichalso fits the scholiast’s dating of Cynaethus.45 In 523/2 Polycratescelebrated a festival onDelos. Inquiring of the Delphic oracle whetherhe should call the contests Delia or Pythia, he was told �ÆF�� Ø ŒÆd˜�ºØÆ ŒÆd —�ŁØÆ (‘these are both Delian and Pythian for you’).46

Burkert suggests that this reflects Polycrates asking the oracle toendorse Delian and Pythian games. More recently,47 West combinesthis with his earlier theory, arguing that Cynaethus first composed theindependent D. hymn in the mid-sixth century, modelling it on anearlier P. hymn, and then created the combined Hymn for Polycrates’festival. He also altered the parts of the Hymn when combining them,and the signs of this are visible in our text. In order to please hisSamian patron, he increased Hera’s role by inserting the Typhaonepisode in P. (305–55), and introducing her into lines 94–114 of theD. hymn. Certainly this last passage would run more smoothly with-out the lines referring toHera,48 and the joins at the start and finish of

42 Translation taken from Burkert (1979), 54.43 West (1975), Burkert (1979), 54–7.44 Pl. Hipparch. 228b.45 Burkert (1979), Janko (1982), 112–15.46 Suda s.v. �ÆF�� Ø . . .More detail on the sources in Burkert (1979), 59 n. 31.47 WL 9–12.48 At 95–6 it is awkward that we are told all the goddesses were present except

Hera, and then that Eileithyia also was not present. At 95–9 the repetition of languageis also difficult. Lines 97–8 state that Eileithyia had not perceived Leto’s labour

72 Mike Chappell

the Typhaon episode are awkward.49 If Cynaethus did recite thecombined Hymn at Polycrates’ festival on Delos, then this wouldexplain why he left in the section describing the Delian festival, andomitted the beginning of P. Though West’s theory may not be correctin all its details, it does provide a plausible picture of how the Hymnwe have came into being.

Unitarians deny that the two parts are fully satisfactory as inde-pendent wholes, and suggest ways in which they need each other forthe full completion of their themes. Miller compares the structure ofthe Hymn to programmes of encomiastic rhetoric, in which thesubject’s origin and upbringing are followed by his or her pursuitsand exploits.50 P. represents the praise of the hero’s exploits. Certainlythe unified Hymn’s structure of birth followed by exploits is reason-able, but this is a basic sort of unity that could easily result fromjoining two hymns together. The narration of the god’s birth is astandard theme of hymns, and may or may not be followed bydescription of the god’s upbringing and exploits.51 For Delianhymns the birth of Apollo and/or Artemis was of course the standardsubject, and a hymn narrating the god’s birth would in no sense beincomplete. A common way to follow the narration of the birth mighthave been to describe the current cult on Delos inspired by the birth,as D. does. From a Delian religious perspective the current festival isthe true completion of the birth theme, not Apollo’s exploits else-where.

The D. poet does in any case look forward to Apollo’s maturityby describing his miraculous growth and indicating what his maintimai will be: �YÅ �Ø Œ�ŁÆæ� �� ç�ºÅ ŒÆd ŒÆ��ºÆ ��Æ j åæ��ø �’I�Łæ��Ø�Ø ˜Øe �Å�æ�Æ ��ıº�� (‘The lyre and curved bow shallever be dear to me, and I will declare to men the unfailing will ofZeus’, 131–2). The Hymn also deals in its opening scene with thetheme of a new god’s acceptance on Olympus, which often follows agod’s birth in the Hymns, and describes the favourite haunts of themature god, culminating of course in Delos and his festival there.

because she was sitting in clouds, and then the idea that this was contrived by Heracould easily be an addition.

49 At 305 the digression starts abruptly, and the shifts of subject at 354–6 areawkward.

50 Miller (1986), 6–10.51 On the structure of the Hymns, cf. Janko (1981). On typical themes and their

combinations, cf. Sowa (1984).


Apollo’s statement at lines 131–2 has been used as evidence of theHymn’s unity because of the emphasis it places on Apollo’s oracularrole, which is in the climactic position of the passage and gets a line toitself. Is this preparation for what is in fact the Hymn’s main theme,the founding of the Delphic oracle? The mention of the lyre can alsobe viewed as a preparation for P.’s prologue. This is one of the betterarguments for unity. However, the verses can equally well be ex-plained along the lines already suggested, that in a birth hymn it isonly by such a prophecy that the god’s mature attributes can bedescribed.52 The three attributes mentioned are standard ones forApollo (the same three spheres are mentioned by Callimachus at Hy.2. 24, though he may have had our Hymn in mind) and so need nospecial explanation; but it has to be conceded that we might notexpect so much emphasis on Apollo’s oracular role in a Delianhymn. If one part of the Hymn was composed to join the other, theDelian poet might have emphasized prophecy to link his poem to thePythian one, and the Pythian poet might have picked up on such aline, or even inserted it himself in his combined Hymn. These lastpoints are not arguments in favour of disunity, but show that, if wetake a separatist view on other grounds, links and similarities betweenthe two poems are compatible with this view. Indeed, this is what wemight expect if most separatist theories are right in suggesting onehymn was composed as an addition to the other.

The most ingenious attempt to show thematic connections be-tween the two parts, such that each is incomplete without the other,is that of Clay. She takes lines 67–9 º�Å� ª�æ �Ø�� çÆ�Ø� I��ŁÆº��ººø�Æ j ��ŁÆØ, ªÆ �b �æı�Æ��ı�� IŁÆ��Ø�Ø� (‘They saythat Apollo will be one that is very haughty and will greatly lord itamong gods and men all over the fruitful earth’) to mean that Apollois rumoured to be a threat to Zeus, whose rule on Olympus is not yetfully secure. This theme is also suggested in the opening scene onOlympus, where Apollo appears brandishing his bow and frighteningthe assembled gods (1–13). The theme of a threat to Zeus’ rule is alsofound in P., in the episode of Typhaon’s birth (305–54). Here Heraexplicitly states that she wishes to bear a son who will be greater thanZeus, and Typhaon’s combat with Zeus was a familiar myth, notablyin Hesiod’s Theogony (820–80). Typhaon is linked by the poet to the

52 The god’s attributes can also be mentioned in descriptions of typical activity inthe opening and closing sections of the Hymns—in D. cf. 1–13, 19–24, 140–6.

74 Mike Chappell

Delphic snake killed by Apollo, and hence Apollo is on the side ofZeus and the Olympians against the dangerous chthonic monstersthat threaten his rule. Apollo’s establishment of his oracle is alsounder the aegis of Zeus, and he is the one who conveys Zeus’ will tomortals. Clay links this toHera’s attempts to delay the birth of Apollo,and states that she is also behind Delos’ and other places’ fear of thenew god: ‘the source of the malicious rumour turns out to be noneother than Leto’s persecutor, Hera, no mere jealous wife, but themother of Typho and an avowed enemy of Zeus’ cosmos’.53

As Thalmann points out, there is no evidence at all in the text forthis last statement, or the link it makes between the two parts of theHymn.54 In some other versions of Apollo’s birth (e.g. Callimachus’Hy. 4. 55–204) it is fear of Hera’s anger that makes places reject Leto,but in the Homeric Hymn it is only after Delos agrees to be thebirthplace that Hera’s opposition is mentioned. Delos states that thereason she fears Apollo is that this haughty god may despise a rockyand barren birthplace and harm her (66–78). There is no need to readinto Delos’ fears about Apollo’s character the idea that he will be athreat to Zeus. This would in fact be an unlikely thing to suggestabout Apollo, the most Olympian of gods and loyal son of Zeus, sowould need to be stated more explicitly if it were the poet’s meaning.Delos need not mean any more than that Apollo will be a haughtyand powerful god, and indeed this is how he was often portrayed inGreek art and literature.55 Even if we do understand Delos’ words asmeaning Apollo is a threat to Zeus, the Delian hymn shows that thethreat is negated: in its opening scene Apollo hangs up his bow andaccepts a cup from Zeus, and at line 132 Apollo says that he willprophesy Zeus’ will to men. This is not a theme which is left incom-plete and D. is a satisfactory independent hymn in this as in otherrespects.

For both Clay and Miller, however, it is a central theme of theunified Hymn that Delos’ prophecy is not realized, and that Apollo’sbehaviour in P. shows this. He punishes Telphousa for her pride anddeceitfulness (375–87), kills a monster that was harming local people(366–74), founds his oracle to benefit and instruct mortals (247–53¼

53 Clay (1989), 74. For a reading of intergenerational conflict in Apoll., whichfollows Clay’s approach, see in this volume Felson (Ch. 12).

54 Thalmann (1991), 145. For a more sympathetic review see Janko (1991).55 I will return to this in more detail below (pp. 76–7).


287–93), and finally warns the Cretan priests against hubris or im-piety (537–43). Förstel, who is not a unitarian, agrees with thisdescription of Apollo’s actions in P., but argues that the two partsof the Hymn have different conceptions of Apollo. In P. he is moraland benevolent but in D. he is ambivalent: as the opening sceneshows, he is both frightening and a bringer of joy. Delos fears him,but then rejoices in his birth.56 So for Förstel there is a difference thatsupports a separatist view, while for Clay and Miller the differencebetween the (apparently) dangerous god of D. and the moral god of P.is actually a sign of unity, as it shows a thematic development. Myview is that there is no development of the type posited by Clay andMiller; both sections merely depict Apollo as he is usually portrayed.Apollo is indeed a bringer of both fear and joy in D., but so he is in P.also. His actions can be seen as proud and haughty rather than justand philanthropic. He kills the snake that is an obstacle to his plansand boasts over its corpse (357–74). He punishes Telphousa (andagain boasts over her) not for the sake of any abstract justice, but forattempting to deceive him and keep her shrine to herself (375–87).Heis not more righteous than her, merely more powerful. Moreover, hehijacks the Cretan sailors without any regard for their own wishes,and addresses them disdainfully (531–44).57

This does not mean that Apollo’s behaviour should be condemned:it is simply how a powerful god behaves. It is misguided to see acontradiction between Apollo the Delphic lawgiver, upholder ofmorality and moderation, and the arbitrary power he displays inthe Hymn.58 Gods are not judged by the same standards as men.Apollo’s haughtiness and his warnings to mortals against hubris aretwo sides of the same coin. He is the upholder of boundaries betweenmen and gods. The Delphic maxim ª�HŁØ ��Æı�� (‘know thyself ’) isa reminder to mortals that they are not gods, and it is Apollo whowarns Diomedes in the Iliad:

çæ�Ç��, 'ı��Å, ŒÆd å�Ç��, Å�b Ł��E�Ø�r�’ �Ł�º� çæ��Ø�, K��d �h �� çFº�� ›�E��IŁÆ��ø� �� Ł�H� åÆÆd Kæå��ø� �’ I�Łæ��ø�. (Il. 5. 440–2)

56 Förstel (1979), 279–80.57 Davies (2007), 49, vividly describes Apollo as ‘browbeating a boatload of hapless

Cretan sailors, and in general calling all the shots in a pretty tyrannical fashion’.58 Cf. Davies (1997), esp. 47 and 57–8.

76 Mike Chappell

Take care, give back, son of Tydeus, and strive no longer to makeyourself like the gods in mind, since never the same is the breed ofgods, who are immortal, and men who walk groundling. (trans. Latti-more, 1951)

Similarly he refuses in the Iliad to fight Poseidon over insignificantmortals (21. 461–7) and warns Achilles not to waste his time chasing agod (22. 7–10). He sends the plague on the Achaeans at the start ofthe poem; he descends ‘like the night’ (�ıŒ�d K�ØŒ� , 1. 47) and ‘terriblewas the clash that arose from the bow of silver’ (��Ø�c �b ŒºÆªªc ª��’Iæªıæ�Ø� �Ø�E�, 1. 49). He destroys the Achaean wall easily, like aboy kicking over sandcastles on the seashore (15. 361–6), and strikesPatroclus with his hand, dazing him and shattering his armour(16. 788–817). In the Iliad he is a frightening and powerful god.

For Clay, Apollo’s foundation of the Delphic oracle illustrates bothhis benevolence towards humans and his allegiance to the Olympianorder presided over by Zeus. It shows that ‘as the mouthpiece of Zeus,Apollo will be a mediator between his father and mankind’.59 Theoracle’s foundation is more than just the creation of a new sanctuary:as the first ever oracle, it changes the relationship between men andgods, to some extent bridging the gap between them, and makingthem capable of moral choice.60 The Cretan sailors, in their help-lessness and ignorance of Apollo’s plans, ‘represent the helpless stateof mortals before Apollo inaugurated his oracle’, but now ‘the decreesof the gods are no longer completely hidden from us.’61

These statements are not clearly borne out by the text. If we look atwhat Apollo says about his oracle, he states that it will be rich andmen will bring hecatombs there from all over the Greek world andthat he will give them infallible counsel (lines 247–53 ¼ 287–93). Tothe Cretan sailors he says that they will serve his rich temple andknow the gods’ intentions; that both they and the oracle will behonoured by men for all time; and that they will be able to live onthe meat from the abundant sacrifices brought there (lines 475–85,532–7). He also warns them that if they behave hubristically they willbe subjugated to new masters (lines 538–44). There is nothing in anyof this about mortals now becoming capable of moral choice for thefirst time, an alleviation of their wretched status, or a permanent

59 Clay (1989), 44.60 Clay (1989), 75.61 Clay (1989), 80.


change in relationships between men and gods. The hymn does notstate that Delphi is the first oracle, unless 214–15 (q’ ‰ �e �æH��åæÅ��æØ�� I�Łæ��Ø�Ø� j ÇÅ��ø� ŒÆ�a ªÆEÆ� ��Å , �ŒÆ�Å��º’@��ºº��;) is interpreted as ‘the first oracle’. Most commentatorsand translators prefer to take �e �æH�� as adverbial; e.g. West inhis recent Loeb has ‘how you first went over the earth . . . in search ofa place for your oracle for humankind’.62

Clay further suggests that the human folly and helplessness sung ofby the Muses on Olympus (lines 190–3) will be ‘somewhat alleviatedby the establishment of Apollo’s oracle’,63 but this is a timelessdescription, not one that precedes in time the oracle’s foundation—in fact in this scene Apollo goes up to Olympus from Delphi: ��Ł�� b�æe Ὄºı�� I�e åŁ��e u �� ÅÆ j �r�Ø ˜Øe �æe �HÆ Ł�H��Ł’ ›�ªıæØ� ¼ººø� (‘Thence [from Pytho], swift as thought, hespeeds from earth to Olympus, to the house of Zeus’, 186–7). Thehuman condition the Muses sing of is not something changed by theoracle, it is how things are. To suffer and be ignorant, to grow old anddie, is what it means to be a mortal and not a god. An oracle doesgrant humans some knowledge of the divine will, but many storiesabout oracles stress how cryptic and ambiguous this communicationis. As I have argued above, Apollo is the god who preserves the gapbetween gods and mortals rather than seeking to diminish it. TheCretans are indeed typical humans in their difficulty in recognizingthe god who appears to them and their worries about the future, butthis is not something that has been alleviated because of the oracle’sestablishment—it is normal in stories about a god’s epiphany.

P. does emphasize that the oracle is founded ‘for mankind’ (248,288), but this need not be seen as purely benevolent. There is at leastas much emphasis on the benefits for Apollo himself: humans willcome in crowds bringing him perfect hecatombs (lines 249–50 ¼288–9, 366, 535–6); his temple will be rich and honoured by all men(lines 478–9, 522–3). After punishing Telphousa, Apollo says to herK�Ł�� c ŒÆd Ke� Œº� ��ÆØ, �P�b �e� �YÅ (‘I am going to berenowned here, not you alone’, 381). Olympian gods are not disin-terested benefactors of humankind: they desire to receive honours

62 Miller (1986), 71 n. 182, also argues for the adverbial meaning of �e �æH�� here.The poet emphasizes a concern with the origins of the oracle, not its primacy relativeto other oracles.

63 Clay (1989), 55–6. Similar ideas in Miller (1979), 74, and Förstel (1979), 233.

78 Mike Chappell

from mortals, and the relationship is a reciprocal one. Apollo will tellmen the divine will, but in return he will receive more honour andmore sacrifices.

Clay also stresses that the Hymn depicts Delphi as a purely Olym-pian foundation, and shows Apollo to be the loyal son of Zeus andchampion of the Olympian regime. She argues that the poet sup-presses the myth of the previous owners at Delphi because he wantsto portray it as a purely Apolline and Olympian foundation, andstress the Olympian and male at the expense of the chthonic andfemale.64 This also accounts for the absence of the female Pythia fromthe Hymn, and the depictions of Apollo’s victory over a female snake,and by association over Typhaon and his mother Hera, who is hostileto the patriarchal Olympian order. I have discussed these issues morefully elsewhere,65 and made the objection that on this view it seemsillogical for the poet to suppress the previous (chthonic, female)owners of Delphi; if he wishes to show Apollo’s superiority, whynot depict his victory over them explicitly? It seems more likely thatwe have here an earlier variant of the dragon combat, in which thesnake is merely a menace to the local population, and not a guardianof Earth’s oracle.

Clay regards the longer Hymns as a genre that is unified by contentrather than by structure or language.66 In her view the Hymns fill agap between Hesiod’s Theogony, which depicts the conflicts of theolder gods and Zeus’ rise to power, and the settled Olympianpantheon of Homeric epic, where Zeus’ supremacy is assured andconflicts between gods are confined to squabbling. The Hymns tellhow this ordered Olympian hierarchy was formed, and also howrelations between men and gods came to be as they are. Each Hymntherefore describes a crisis point on Olympus, whose solution leads to‘an epoch-making moment’, a ‘new era in the divine and humancosmos’; and all this is planned and controlled by Zeus, who distrib-utes to the gods their timai. ‘In the final analysis, the politics ofOlympus are the politics of Zeus.’67 In the Hymn to Apollo the crisis

64 As discussed above (p. 77), Clay also takes the absence of previous owners as amark of the Hymn’s Panhellenic orientation, stressing the Olympian over localmythology.

65 Chappell (2006).66 Clay (1989), 6–16, 267–70, and this volume (Ch. 11). For some good criticisms

of this view, see Thalmann (1991).67 Clay (1989), 268.


is the threat the new male god may pose to Zeus, whose loyal son heproves to be. The founding of the oracle demonstrates this loyalty,and also inaugurates a new order between men and gods. In mydiscussion above, I have argued against the view that the Hymndepicts any atmosphere of crisis around Apollo’s birth or ever sug-gests that he is a threat to Zeus, and have also argued that there is nosuggestion in the text that the foundation of Delphi fundamentallychanges the relationship of men and gods.

Zeus’ supremacy and plans are certainly important in the longHymns to Demeter, Hermes, and Aphrodite, but seem less so in theHymn to Apollo. Rutherford has argued that the main differencebetween Pindar’s depiction of Apollo’s birth on Delos and the Hom-eric Hymn’s is the greater role that the former gives to Zeus.68 Nor isthe role of Zeus at Delphi greatly stressed in P., compared to manyDelphic hymns and other poems.69 It is only in D., at line 132, thatApollo explicitly states that he will prophesy to men the ‘infallible willof Zeus’ (˜Øe �Å�æ�Æ ��ıº��). In P. Apollo states that he willdeclare the ‘infallible will’ (�Å�æ�Æ ��ıº��, 252 ¼ 292), but withoutmentioning Zeus. When talking about the oracle to the Cretan sailorshe again makes no mention of Zeus except to introduce himself asZeus’ son Apollo, and he tells the Cretans that they will know the‘gods’ intentions’ (��ıº� �’ IŁÆ��ø�, 484), not the will of Zeusspecifically. Other than this, Zeus is only mentioned in the openingscene, approving of his son (205), and in conventional phrases refer-ring to the weather such as ˜Øe �hæøfi (‘the breeze from Zeus’, 427)and KŒ ˜Øe ÆY�Å (‘by ordinance of Zeus’, 433).70

Overall, the attempts to demonstrate the Hymn to Apollo’s coher-ence and thematic unity are in my opinion unconvincing, and thearguments in favour of its division are stronger. As I have arguedabove, Clay’s argument that the two halves are linked by the theme ofApollo being a possible threat to Zeus who turns out to be his loyalsupporter and combats the chthonic female enemies of Olympus restson assumptions and connections that are not evident in the text(notably her assertion that the Typhaon episode somehow shows

68 Rutherford (1988).69 e.g. Herm. 469–72, 531–40, Alc. fr. 307c Voigt ap. Him., Or. 48 (2. 1 in Furley

and Bremer 2001), A. Eu. 17–19, 616–18.70 Miller (1986), 97 and Clay (1989), 82 view these mentions of Zeus as stressing

his important role in the process, but they are too unemphatic and conventional forthat.

80 Mike Chappell

thatHera was behind the rumour cited byDelos that Apollo would bea violent and lawless god). The two parts have no strong thematicconnection other than the basic idea of a god’s birth followed byexploits, which could easily have resulted from the joining together oftwo originally independent hymns. There is no real evidence forPanhellenism in the two parts; it is true that the combined Hymncan be seen as Panhellenic in dealing with both Delos and Delphi,but I have argued that this Panhellenizing combination of non-Panhellenic parts actually fits better a separatist view of the hymn’sgenesis. Like the other Homeric Hymns, the concerns of D. and P. aremainly aetiological. D. tells how the god himself came into being, andhence the cult on Delos which is depicted in the Hymn’s closingsection; and P. describes the origin of Delphi and its priests, and ofthe cult titles Pythios, Telphousios, and Delphinios.


5

The Homeric Hymn to Hermes

Humour and Epiphany

Athanassios Vergados

Scholars often remark on how the Homeric Hymn to Hermes differsfrom the rest of the major Hymns in its unusual diction and narrativestyle, as well as in its humorous treatment of the gods.1 An unusualfeature that has not yet received sufficient attention is this: contrary tothe rest of the major Homeric Hymns, Hermes lacks a narratedepiphany.

The god’s epiphany seems to be a regular component of therhapsodic hymn and follows a pattern that has been set out in detailby D. Turkeltaub.2 In all the other major Hymns (as well as theseventh Hymn to Dionysus), at some point in the narrative thecelebrated god reveals his or her divine identity to mortals. AfterApollo appears as a dolphin and a star in broad daylight in theHymn to Apollo (400, 440–5), the god assumes the form of an ephebe

1 See for instance WL 12: ‘The Hymn to Hermes (4) is distinctive in other ways. Ofall the earlier Greek hexameter poems, it is without doubt the most amusing. It is alsothe most untraditional in its language, with many late words and expressions, andmany used in slapdash and inaccurate ways; and it is the most incompetent inconstruction, with many narrative inconsistencies and redundancies and no com-mand of the even tempo appropriate to epic storytelling.’ Richardson (2007), 85,relates the poem’s unusual diction and narrative style to its humour. See alsoRichardson (2010), 19–20, 23–4.

2 See Turkeltaub (2003), 16–50, for the general overview of the Homeric epiphanyscene-pattern, and 51–122 for its instantiation in the Homeric Hymns. See alsoKullmann (1956), 83–111, Richardson (1974), 207–8, Bremer (1975), 1–12, Sowa(1984), 236–72.

(449–50) and declares: �Nd �! Kªg ˜Øe ıƒ� ,��ººø� �! �hå�ÆØ �r�ÆØ(‘I am the son of Zeus, and declare that I am Apollo’, 480).3 Similarly,in the seventh Hymn, Dionysus makes his presence manifest, asfragrant wine flows in the pirates’ ship (35–7), a vine and ivy growon the mast (38–41), and the oar-pins are all crowned with garlands(42). In addition, the god transforms himself into a lion (44–5).4 Allthe pirates jump frightened into the sea and are transformed intodolphins, but the god spares the helmsman who was the only one toacknowledge Dionysus’ divine identity from the very beginning,when the pirates’ bonds fell onto the ground (15–24; cf. 46).5 Diony-sus bids the helmsman not to fear him and reveals his divinity (56–7):�Nd �! Kªg ˜Ø��ı�� Kæ��æ�� n� �Œ� ��Åæ j ˚Æ�ÅU $�ºÅ ˜Øe K� çØº��Å�Ø Øª�E�Æ (‘I am loud-roaring Dionysus, whom Cadmus’daughter, Semele, bore, after mingling in love with Zeus’). In Aphro-dite the goddess of love assumes her divine form before Anchises: herhead reaches the roof-beam, and divine beauty radiates from hercheeks (172–9); she then explicitly announces her identity (285–6).Finally, Demeter has two epiphanies in her Hymn. The poet firstnarrates her partial epiphany: when the goddess enters Celeus’ palace,her head reaches the roof-beam while the doors are filled with divineradiance (188–9). But it is later that the audience experience Dem-eter’s full epiphany when she reveals her identity (�Nd �b ˜Å��Åæ�Ø��å� . . . , ‘And I am Demeter, the honoured one . . . ’, 268).Shortly thereafter she assumes her proper form, filling the palacewith radiance and a divine fragrance (268–80).6

Despite the absence of a full epiphany in Hermes, the poet usesmotifs that are typically associated with divine presence, while someof the stages of the typical Homeric epiphany scene do occur in thepoem. After stealing fifty of Apollo’s cows at Pieria, Hermes leads

3 All translations are mine unless otherwise noted.4 On epiphany in Hy. 7, see in this volume Jaillard (Ch. 7). Lines 45–7 mention the

appearance of a bear in the midst of the ship, while the god’s actions are collectivelylabelled ��Æ�Æ. However, this section is possibly interpolated; see Sparshott (1963).For fear as a reaction to a divine epiphany, cf. Hy. 19. 38–9 with Thomas in thisvolume (p. 162).

5 Interestingly, a similar miracle occurs at Herm. 409–14: Apollo attempts to bindHermes with withies, but the bonds fall to the ground, take root, and bind the cowsinstead; see Kuiper (1910) 43–4, Baudy (1989), and Vergados (forthcoming) ad409–14.

6 The epiphany scene in Dem. probably depends on the similar scene in Aphr.; seeFaulkner (2008), 38–40 on the relation between the two poems, and Sowa (1984), 243.

The Homeric Hymn to Hermes 83

them backwards to the Peloponnese. At Onchestus in Boeotia heencounters an Old Man working on his vineyard whom he insultsand to whom he gives instructions (90–3). Later in the Hymn, Apollosomehow becomes aware of the Old Man’s presence and travels toOnchestus to interrogate him about his stolen cattle (184–200).7

Hermes’ shining eyes are mentioned (278, 415–16), as is the divinefragrance and the divine footprint motif (231–2, 237, 342–3),while Hermes’ actions cause helplessness and wonderment to theirobserver.8

It is noteworthy that some of these epiphanic motifs are notperceived by (or focalized through) a mortal, but rather Apollo.Thus it is Apollo who finds himself perplexed by Hermes’ tricks: at196 he admits to the Old Man his wonderment at the fact that thecows disappeared while the bull and the four watchdogs remained inthe field.9 At 219 he exclaims t ��Ø, q ªÆ ŁÆFÆ ��! OçŁÆº�E�Ø�›æHÆØ, ‘Alas, what I see with my eyes is a great wonder indeed’,a phrase elsewhere found in archaic hexameter in contexts wherea mortal observes the effects of a god’s workings.10 At 407Apollo admires Hermes’ power (ŁÆıÆ��ø ŒÆ��Ø�Ł� �e �e� Œæ�� ,‘I wonder at your future power’). After Hermes manages to

7 For the insult motif, see Turkeltaub (2003), 33; for instructions to a mortal as atypical occurrence during divine revelation, see Richardson (1974), 242–4, Turkeltaub(2003), 27. Apollo’s encounter with the Old Man is also to some extent reminiscent ofmotifs typically found in epiphany scenes: the god notices the Old Man and attemptsto convince him to provide information that will help him recover his lost cattle; seeTurkeltaub (2003), 19, 24. The Old Man’s answer to Apollo’s request for informationis also noteworthy. It is clear that he thought his encounter with Hermes extraordin-ary (a talking infant was leading fifty backwards-marching cows!). One might haveexpected the Old Man to realize that he had encountered a god, and perhaps the poetis playing with audience expectations at this point. The Old Man’s failure to perceiveHermes’ divine nature confirms the validity of the god’s insult to the Old Man (seealso below n. 23). It is Apollo who, combining the information provided by the OldMan, the bird omen he receives at 213, and his observation of the footprints at 342–3,perceives the thief ’s divine identity.

8 Herm. 342–3 (�a �! ¼æ! Yå�ØÆ ��Øa �ºøæÆ j �x� �! Iª��Æ�ŁÆØ ŒÆd IªÆı�F�Æ�� æªÆ, ‘the tracks were of two kinds, monstrous, such as to cause amazement,and the deeds of a mighty spirit’). At Il. 13. 68–75 Poseidon’s divinity is recognizedfrom his Yå�ØÆ . . .��H� M�b Œ�Å�ø� (‘the marks of his feet . . . and shins’); cf. Sowa(1984), 247–50.

9 The passivity of the watchdogs may be a further hint at divinity: at Od. 16. 162–3the dogs at Eumaeus’ hut perceive Athena and do not bark at her but flee instead.

10 Cf. Il. 15. 286, 20. 344, Od. 19. 36. Words related to ŁÆFÆ or Ł�� oftenexpress the observer’s reaction to a divine epiphany; cf. Pfister (1924), 317, Bremer(1975), 2 n. 2, Turkeltaub (2003), 31–2, Jaillard (2007), 83.

84 Athanassios Vergados

escape from Apollo’s bonds, the latter ŁÆ�Æ�� IŁæ��Æ (414; ‘[he]saw and wondered). And finally at 455 Apollo expresses his admira-tion for Hermes’ music (ŁÆı�Çø ˜Øe ıƒb ��! ‰ KæÆ�e� ŒØŁÆæ�Ç�Ø ,‘I admire, son of Zeus, how beautifully you sing these things to thelyre’). In his speech to Zeus Apollo remarks on Hermes’ strangefootprints, which appear to him to be �x� �! Iª��Æ�ŁÆØ ŒÆd IªÆı�F�Æ�� æªÆ (‘such as to cause amazement, and the deeds of amighty spirit’, 343). Moreover, the divine fragrance motif is men-tioned when Apollo bursts into Maia’s cave searching for his cattle: anO�c ƒ�æ��Æ (‘a charming fragrance’) wafts around Cyllene (231–2), while, at the sight of Apollo, Hermes dons his ��æªÆ�Æ Łı��Æ(‘his fragrant swaddling clothes’, 237).11 Finally, there are severalreferences to Hermes’ shining eyes during his encounter with Apollo:after his first defence speech, Hermes ‘casts many sparking glancesfrom his brows’ (�ıŒ�e� I�e �º�ç�æø� IÆæ��ø�, 278), while at415–16 he attempts to conceal his radiant look from Apollo (åHæ��º��Å� K�ŒłÆ�� Fæ IÆæ�� j KªŒæ�łÆØ �Æ� , ‘he surveyedthe area with his eyes downcast, eager to hide the darting fire [sc. fromhis eyes]’).12 Yet these motifs do not lead to an epiphanic moment.Indeed neither Hermes nor Apollo makes the slightest attempt todisguise themselves before they appear to the Old Man at Onchestus,even though elsewhere in the Hymns the gods take on disguises (e.g.at Apoll. 400 where the god appears to the Cretan sailors as a dolphin,or Dem. 101 where the goddess arrives at Eleusis with the appearanceof an old woman).

This absence of a proper divine epiphany is particularly striking,given that Hermes conforms to the general pattern found in thelonger Hymns. The Er-Stil that dominates the narrative of rhapsodichymns switches to the Du-Stil that is typical of the hymnal envoi andimplies the addressee’s (¼ the god’s) presence at the end of the poem.In other words, the narrative of the god’s story effects his or herepiphany to the audience and ideally also his or her blessing of that

11 On this motif, see Lohmeyer (1919), 4–14, Deonna (1939), Lilja (1972), 25–30,Meloni (1975), 12–14. In Herm. this motif is used playfully: at 232 the poet adds,��ººa �b BºÆ �Æ�Æ��Æ ��Œ�� Å� (‘and numerous long-shanked sheep weregrazing the grass’). And of course one cannot expect a baby’s diapers to be fragrant!

12 The poet ofHerm. correlates the god’s radiant look with his inventiveness; cf. thesimile at 43–6. On Herm. 415, see Vergados (forthcoming) ad loc. For radiance as amotif associated with divine presence, see Keyssner (1932), 150, Bremer (1975), 5 and7, Dietrich (1983), 68, Turkeltaub (2003), 137.


community.13 Hermes forms no exception to this pattern. After aproem that partly announces the Hymn’s subject, we pass to the parsepica, which presents the god’s birth story and first achievements ineither third person narrative or direct speech. Lines 579–80 switch tothe second person, asking the god to rejoice in the song just com-pleted, and this second person implies Hermes’ presence at theperformance. But why is it that in the other longer Hymns the god’sperceived presence (and consequently the shift from Er- to Du-Stil) iseffected after the god’s epiphany has been narrated in the pars epica,while in the case of Hermes. such narrated epiphany never occurs?14

The peculiar treatment of Hermes’ epiphany in the fourth HomericHymn can be explained if we consider the mode of this god’s pres-entation in early Greek literature. Hermes often appears in situationsthat involve humour and comedy: examples of such presentation canbe found in the Odyssey, the Homeric Hymn to Pan, the fragments ofHipponax, and Old Comedy. When an epiphany of Hermes doesoccur (as for instance his epiphany to Priam in Iliad 24), it is notpresented in the grandiose fashion of the typical hymnic epiphanies.This can be accounted for by Hermes’ nature as a trickster figure whooperates with metis (cunning) and stealth. The poet of Hermes effectsHermes’ epiphany with similar stealth: instead of being explicitlynarrated (as in the other major Hymns), Hermes’ epiphany is enactedin performance. The poet achieves this in two ways. First, by narrat-ing an extremely humorous tale, he elicits from his audience (theHymn’s external audience) the same reaction that Hermes’ actionscause to his own addressees (the Hymn’s internal audience, viz.Apollo and Zeus), i.e. laughter (cf. 281, 389, 420). And, second, bynarrating how the god invented objects and institutions familiar to

13 For the structure of the Homeric Hymns, see Janko (1981), and Calame in thisvolume (Ch. 14) who speaks of evocationes, epicae laudes, and preces. For the Hymnsas staging the god’s epiphany ‘insofar as they represent the performer(s), as well as thegroup for whom he speaks, as “remembering”, or “singing” the god, which is to say,bringing the god into the context of the performance to offer him a verbal agalma’, seeDepew (2000), esp. 73–4; further Danielewicz (1976), 116–19, García (2002), esp.28–34, for whom the ‘exclamation åÆEæ� . . . [is] the singer’s welcome at the arrivinggod’ (33), and in this volume Thomas (p. 164). Bierl (2004), 45, points out that ‘theHomeric Hymns often seem to be songs about the advent of a god.’ See Furley andBremer (2001), i. 61 for the use of second-person imperatives that signal the desire forthe god’s advent.

14 Apoll. does not entirely conform to this norm either, as it moves much morefreely from Du- to Er-Stil than the other longer Homeric Hymns.


the audience from their everyday experience (some of which arenecessary for, or otherwise tied to, the poem’s very performance),the poet intimates Hermes’ constant presence in the audience’s life.15

This last strategy is of course not peculiar to Hermes. But, as I hope toshow in what follows, the poet of Hermes takes this strategy to anextreme by presenting the celebrated god as the inventor par excel-lence, with whom he identifies himself.

HERMES AND HUMOUR

The Homeric Hymn to Hermes is extremely witty and humorous as isappropriate for a poem that celebrates a trickster figure.16 Eventhough the other Homeric Hymns occasionally reveal humorous mo-ments,17 Hermes’ divine persona is especially amusing, and it wouldperhaps not be an exaggeration to speak of a tradition of Hermes’presentation in poetry that involves humour and laughter.18

Epic poetry furnishes examples for this aspect of Hermes’ divinepersona. In the second song of Demodocus (Od. 8. 266–366) the malegods have gathered around Hephaestus’ bed and witness the lamegod’s craftiness in how he was able to trap Ares and Aphrodite.19

Apollo asks Hermes:

% Eæ��Æ, ˜Øe ıƒ, �Ø�Œ��æ�, �H��æ ��ø�q Þ� Œ�� K� ��E�! KŁº�Ø ŒæÆ��æ�E�Ø �Ø��Ł��

�o��Ø� K� ºŒ�æ�Ø�Ø �Ææa åæı�fi Å �çæ��fi Å (Od. 8. 335–7)

15 Hermes’ stealthy presence was well-understood by Callimachus; cf. Vergados(forthcoming), Introduction IV B } 13–17. For the reception of the Homeric Hymns inHellenistic literature, see Faulkner in this volume (Ch. 9).

16 The comical dimension of the poem was already noted by Eitrem (1906), 248,who remarked, ‘At first sight, the entire Hymn appears to contain very few religiouselements; the tone is playfully light, full of humour and jugglery; the divine babe isconceived of as a mischievous Greek youth who deceives his elders, and especially hisclosest kin in order to have fun and pave the way for himself; [Hermes is presented as]divinely gifted and puerile to the extent of scurrility’ (my translation; emphasis in theoriginal). Schneidewin (1848), 663 had spoken of the ‘roguish poet’ who composedthis Hymn. Cf. further Bielohlawek (1930), 203–9, Sikes (1940), 123, and Otto (1987),142, 315, who calls Hermes a Schelm (‘rogue’).

17 See in this volume Clay (pp. 245–6).18 See Nesselrath (2010), 147–9, for a brief overview of Hermes’ presence in earlier

literature.19 For more on this episode, see in this volume Clay (pp. 249–50).


Hermes, son of Zeus, conveyer, giver of good gifts, would you wish to liein bed beside golden Aphrodite, pressed by strong bonds?

To this the divine messenger replies:

ÆD ªaæ ��F�� ª��Ø��, ¼�Æ� �ŒÆ�Å��º! @��ºº��.��d b� �æd ��Ø I��æ�� Içd �å�Ø��,��E �! �N��æ�øfi �� Ł��d �A�Æ� �� ŁÆØ�ÆØ,ÆP�aæ Kªg� �o��ØØ �Ææa åæı�fi Å �çæ��fi Å (Od. 8. 339–42)

I wish that this would happen, lord Far-shooter Apollo. Even if thrice asmany inescapable bonds should hold me, and both you gods and all thegoddesses should look upon me, nevertheless I would lie beside goldenAphrodite.

This answer causes the by-standing gods to laugh (343 S �çÆ�! , K� �bªºø tæ�! IŁÆ��Ø�Ø Ł��E�Ø�; ‘Thus he spoke and laughter aroseamong the immortal gods’).20

In addition, there is a strong connection between Hermes and theiambic poets. The plot of Hermes bears some similarities to the storyabout Archilochus of Paros, transmitted in the Mnesiepes inscription.I quote the relevant passage (E1II. 22–40) with the translation fromClay (2004),21 and underline those words and phrases which resem-ble the story of Hermes:22

ºª�ı�Ø ªaæ �æå�º�å�� Ø ��æ��

Z��Æ ��çŁ��Æ ��e ��F �Æ�æe '�º��ØŒº�ı

�N Iªæ��, �N �e� �B��, n ŒÆº�E�ÆØ ¸�ØH�� ,[u]�� F� ŒÆ�ÆªÆª�E� �N �æA�Ø�, I�Æ��Æ 25�_æøØÆ��æ�� B �ıŒ��

_ , ��º��Å ºÆ��Å ,

[¼]_ª�Ø� �c ��F� �N ��ºØ�· ‰ �! Kª�� ŒÆ�a �e�

[�]��, n ŒÆº�E�ÆØ ¸Ø�� , ��ÆØ ªı�ÆEŒÆ [N]�

_�E� ±Łæ�Æ · ��Æ��Æ �! I�e �H� �æªø� I�Ø�ÆØ

ÆP_�a �N ��ºØ� �æ��ºŁ��Æ �Œ��Ø�. �a �b 30

�__�Æ�ŁÆØ ÆP�e� ��a �

_ÆØ�ØA

_ŒÆd ªºø�� ŒÆd

[K]�_�æø�B�ÆØ, �N �øº��ø� ¼ª�Ø �c ��F�· ç��Æ�� ,

20 For the significance of laughter in this scene, see Brown (1989), 290–1, Halliwell(2008), 77–8. Notice too that in Hy. 19. 44–6, Hermes causes joy to the Olympians byshowing them his newborn son Pan. Cf. also Hermes’ words at Il. 21. 497–501.

21 See Clay (2004), 104, with bibliography on the inscription (SEG 15. 517).22 Some of the similarities between the Mnesiepes inscription and Hermes have

already been pointed out by the first editor, N. Kontoleon (1952), 64–7; cf. Peek(1959) who rejected these similarities and Kambylis (1963), 143, who acknowledgedonly a general resemblance between the two texts. See also Compton (2006), 41–3.


[�N]��E� ‹�Ø ÆP�Æd ��ı�Ø� ÆP�fiH �Øc� I��Æ�·[ÞÅ]Ł��ø� �b

_��ø� ÆP�a b� �P�b �c ��F� �PŒ�Ø

[ç]Æ��æa �r�ÆØ, �æe �H� ��H� �b º�æÆ� ›æA� ÆP��· 35[ŒÆ]�Æ�ºÆª��Æ �b ŒÆd �� Ø�Æ åæ�� ı�[ª]�

_�� ºÆ��E� �a )��Æ �r�ÆØ �a çÆ��Æ

[ŒÆ]d_º�æÆ� ÆP�fiH �øæÅ�Æ�Æ ŒÆd I��º�-

[�]�_�� ÆP�c� ��æ��ŁÆØ �N ��ºØ� ŒÆd �fiH �Æ�æd

[�a] ª_��Æ �ÅºH�ÆØ. 40

They recount that Archilochus, when he was still a young man, was sentby his father Telesikles to the fields, to the district called the Meadows(Leimones), to bring a heifer down for sale. He got up at night beforesunrise, while the moon was still bright, to lead the heifer to the city. Ashe came to the place called Slippery Rocks, they say that he thought hesaw a group of women. And since he thought that they were leavingwork for the city, he approached them and made fun of them. But theygreeted him with good humour and laughter, and asked him if heintended to sell the cow he had in tow. When he answered that hedid, they said that they would give him a good price. But, once they hadsaid this, neither they nor the heifer could be seen, but lying before hisfeet he saw a lyre. He was dumbfounded and, after he had the time toregain his wits, he realized that the women who had appeared to himwere the Muses and it was they who had given him the gift of the lyre.And he picked up the lyre and went to the city and told his father whathad happened.

In both stories, the protagonist is young. The events involve cows:Hermes steals fifty of Apollo’s cows, while Archilochus is sent by hisfather to sell a cow. The events occur at night (cf. Mnesiepes inscrip-tion E1II. 26�Hermes 99 � �b �� Œ��Øc� �æ��Æ�� EÆ $�º��Å,‘the divine Moon had just reached her look-out point’, and 141 ŒÆºe��b ç�ø ŒÆ�ºÆ�� $�º��Å , ‘and the moon’s beautiful light wasshining down upon him’). Both protagonists meet characters whomthey mock,23 while both stories involve the exchange of a cow (or

23 This is termed �Œ��Ø� in the Mnesiepes inscription. We may compareHermes’ cryptic address to the Old Man at 90–3: t ªæ��, ‹ �� çı�a �Œ��Ø K�ØŒÆ��ºÆ ŒAºÆ, j q ��ºı�Ø��Ø �s�! i� �� Æ çæfi Å�Ø·j ŒÆ� �� N�g� c N�g��r�ÆØ ŒÆd Œøçe IŒ��Æ , j ŒÆd �ØªA�, ‹�� Ø ŒÆ�Æ�º��fi Å �e �e� ÆP��F (‘Old manwho digs these curved logs as if they were plants, indeed you will obtain lots of winewhen all these produce [i.e. never!]; thus be blind though you have seen and deafthough you have heard, and be silent since your own affairs are not harmed’). In 90 Iadopt ŒAºÆ with AS and Radermacher (1931). The manuscripts are here divided: *transmits K�ØŒÆ��º� þ�ı while M offers the unmetrical K�ØŒÆ��ºÆ ��ºÆ, where


cows) for a lyre: the price of the lyre is mentioned (in the inscriptionthe lyre is �Øc� I��Æ� [sc. �B �� ], ‘a good price’, while at Herm. 437Apollo exclaims after Hermes’ second song: ��Œ��Æ ��H� I��ØÆ�ÆF�Æ ÅºÆ , ‘this song you have contrived is worthy of fifty cows’).Both stories, furthermore, mention the meadows (º�ØH�� ), andboth protagonists have a relation to the Muses (Archilochus receivesa lyre from them; cf. Herm. 429 where Mnemosyne is said to be thepatron goddess of Hermes).24

This type of tale derives from a folk-tale tradition that is alsoreflected in Hesiod’s Dichterweihe at the outset of the Theogony.25

But Hermes and the Mnesiepes inscription differ from Hesiod’snarrative in that both Hermes and Archilochus’ stories take place atnight instead of high noon, the usual time of a Dichterweihe,26 whilethe lyre in both texts is said to be a worthy price for the animal (oranimals); such a comment is missing from the Hesiodic accountwhere the Muses do not present the poet with a musical instrument.Furthermore, in both texts the narrative is presented by a third person(an ‘objective’ narrator) and not by the character who experienced theencounter with the Muses. And finally, the protagonists poke funat deities related to poetry and song (cf. E1II. 33 and Hermes 75–8,261–80, 294–8, 464–74).

In addition to this general similarity of plot between some ofHermes’ Œºı�a �æªÆ (‘glorious deeds’) in Hermes and Archilochus’encounter with the Muses as preserved in the Mnesiepes inscription,Hermes enjoys a conspicuous presence in the admittedly scantyremains of Hipponax’s poetry.27 The iambic poet prays to Hermes,asking for boots and a cloak to protect himself from cold:

��ºÆ must be a gloss for ŒAºÆ that made it into the text (K�ØŒÆ��ºÆ ŒAºÆ recurs atHes. Op. 427 at a different sedes); see my discussion in Vergados (forthcoming), ad loc.With these words Hermes suggests that the Old Man’s toil is in vain (90–1, he isdigging worthless logs that will never produce anything) and bids him to continue histhoughtless existence (92–3); these words are surely cryptic and manipulate proverbialwisdom; see Clay (1989), 115, and Tzifopoulos (2000), esp. 154–8.

24 For Hermes’ songs, see in this volume Calame (pp. 348–50).25 Dichterweihe denotes the poet’s call to his art by a divinity (or divinities) who

introduce(s) him to poetry and inspire(s) his song. On the Dichterweihe theme, seeKambylis (1965) and West (1966), 159–61.

26 See Kambylis (1965), 59–61.27 Walcot (1979), 347–8, notes some similarities between Hermes and iambic

poetry.


% EæB, ç�º! % EæB, )ÆØÆ��F, ˚ıºº��Ø�,K��å�Æ� ��Ø, Œ�æ�Æ ªaæ ŒÆŒH ÞØªHŒÆd �Æ�Æº�Çø . . .�e åºÆE�Æ� % I��ÆŒ�Ø ŒÆd Œı�Æ��Œ��ŒÆd �Æ�Æº��ŒÆ ŒI�Œ�æ��ŒÆ ŒÆd åæı��F

��Æ�BæÆ ��Œ��Æ ��P�æ�ı ��å�ı (Hippon. fr. 32, IEG)

Hermes, dear Hermes, Maia’s puppy, Cyllenian, I solemnly pray to you;for I tremble rather badly and chatter . . . give a cloak to Hipponax, anda small woollen undergarment and little shoes and little boots and sixtystateres of gold from the other side of the wall.28

This fragment is a parody of a prayer which begins with an emphaticinvocation of the god: Hermes is invoked twice, the second time asç�º� (‘friend’), as if the speaker were addressing an equal and not agod. Furthermore, K��å�ÆØ (‘I solemnly pray’), a word that belongsto high poetry, is followed by a request that concerns rather prosaicmatters and is expressed in low diction.29 The final part of the request(golden stateres ��P�æ�ı ��å�ı, ‘from the other side of the wall’) maystrike one as somewhat unexpected since all the items in the list arepieces of clothing. But it will be remembered thatHermes was also thegod of burglars: in Hermes the god himself declares his intention tobreak into Apollo’s temple at Delphi to steal gold, silver, iron, andother valuable possessions (176–81). Thus it is reasonable for Hippo-nax to ask Hermes for gold that is found on the other side of the wall(i.e. in someone else’s house).30 The persona loquens also invokes

28 On this fragment see De Sousa Medeiros (1961), 16–18, Degani (2007), 99–102.The iambographer deflates motifs typically attested in hymnic invocations: )ÆØÆ��Frecalls the god’s parentage, but its formation is comic as Degani (2007), 100 hasshown: it consists of the suffix -��Å /-(Ø)��Å found in epic patronymics(e.g. —ÅºÅœ��Å , ��æ��Å ) to which the ending -�� is attached, a combinationfound only in nouns derived from animal names (e.g. Iºø��ŒØ�� , ªÆºØ�� ,��ºÆæªØ�� ). Degani accordingly renders ‘cucciolo di Maia’, i.e. Maia’s puppy; seealso Degani (2002), 189–91. Acosta-Hughes (2002), 71–2 n. 6, detects in this fragmentan allusion to the opening of Hermes and draws our attention to the ‘juxtaposition ofgenres’ present in both the Hipponax fragment and Hermes.

29 For K��å��ŁÆØ in high poetry, cf. Il. 5. 119, 16. 829, 17. 35, 17. 450, 21. 121, Od.14. 423, 22. 286, 23. 59, Apoll. 370, Thgn. 358, 944, A. Sept. 276, Ag. 501, Cho. 856, Eu.58, Soph. Trach. 809, OT 249, OC 484, Eur. IT 508 etc. For the low diction ofHipponax’s prosaic request, cf. the onomatopoeic �Æ�Æº�Çø (‘chatter, stammer’)and the diminutives Œı�Æ��Œ�� (‘small woollen undergarment’), �Æ�Æº��ŒÆ (‘littleshoes’), I�Œ�æ��ŒÆ (‘little boots’) which have a colloquial tinge.

30 Degani (2007), 102, rendering ‘nell’altro piatto della bilancia’ (‘in the other ofthe scales’), sees here a reference to Hermes as the god of weights and measures, based


Hermes at Hipponax fr. 3a IEG: % EæB Œı��ªåÆ, fi Å��Ø��d ˚Æ��ÆFºÆ, jçøæH� ��ÆEæ�, ��Fæ� �Ø �ŒÆ�Ææ��F�ÆØ (‘Hermes, dog-choker, Kan-daula in the Maeonian dialect, companion of the thieves, come hitherto help me out’).31 The fragment reflects Hermes’ role as the enemy ofwatchdogs and the leader of thieves (cf. Herm. 175, 292; 194–6).A promise to invoke Hermes occurs at Hipponax fr. 35 IEG (Kæøªaæ �o�ø˙ ‘˚ıºº��Ø� )ÆØ�� % EæB’; ‘for I shall speak thus: “Cylle-nianHermes, son ofMaia” ’). The beginning of this verse is iambic, butits end dactylic (cf. Herm. 408˚ıºº��Ø� )ÆØ�� ıƒ).32 Furthermore,Boupalos is said to have called on Hermes at Hipponax fr. 3 IEG(��ø�� )Æ�Å �ÆE�Æ, ˚ıºº��Å ��ºı�, ‘he called on Maia’s son, theruler of Cyllene’),33 while at fr. 177 IEG the speaker addresses Hermesas the god associated with sleep (% EæB �ŒÆæ, <�f ªÆæ> (Meineke—<n ŒÆd> West) Œ��ı�� r�Æ Kªæ��Ø�, ‘Blessed Hermes, for youknow how to rise someone from his sleep’; cf. Herm. 15, 290).

It is worth stressing once more that in all these fragments ofHipponax Hermes is mentioned in contexts that are reminiscent of

on the interlinear ��F ��æ�æ�ı; but cf. Masson (1962), 124 n. 2, who considers this adeterioration of the text and not a gloss; so De Sousa Medeiros (1961), 20, West(1974), 29. Masson (1962), 124 n. 3, appears sceptical of the idea of ��Øåøæıå�Æ, whileFränkel (1993), 248, translates ‘von der inneren Wand’ (‘from the interior wall’),linking Hipponax’s request to Hermes’ function as the patron of thieves. Hippon. fr.34 (K�d ªaæ �PŒ ��øŒÆ �h� Œø åºÆE�Æ� j �Æ��EÆ� K� å�ØH�Ø ç�æÆŒ�� Þ�ª�� j �h�’I�Œæfi Å�Ø ��f ��Æ �Æ��fi Å�Ø j �ŒæıłÆ , u �Ø c å��ºÆ Þ�ª�ı�ÆØ, ‘for you did notgive me yet a thick cloak to be my defence against cold in the winter, nor did you hidemy feet in thick boots, so that I don’t catch chilblains’) may imply either that Hermesdid not grant Hipponax his wish or may function as a justification for why Hermesought to provide these objects to the poet (i.e. you have not provided them to me inthe past, at least do so now).

31 For �ŒÆ�Ææ��F�ÆØ, see LSJ s.v. �ŒÆ��æ��ø, who refer to Tzetzes ap. Analect.Oxon. 3.351 (where it is glossed �ıÆåB�ÆØ); but cf. Hsch. � 855 �ŒÆ�Ææ��F�ÆØ,º�Ø��æB�ÆØ. There is something distinctly incongruous in the juxtaposition of theiambic poet’s ‘learnedness’ (including a foreign gloss in an invocation of Hermes asthe patron of thieves) in a verse that ends in the rather prosaic �ŒÆ�Ææ��F�ÆØ; seeFränkel (1993), 247.

32 Note that we do find traces of hexameters in Hipponax’s fragments; e.g. possiblyfr. 23 IEG ��f ¼��æÆ ��ı O��Å {�ØÆººØæ�Ø�ØÆ�{, where Masson (1962), ad loc.,and Degani (2007), 84, accept Bergk’s O��Å ! �ØÆº�E ÞØªÅº�. On the presence of dactylsamong the choliambs in Hipponax, see Masson (1962), 26–7. Of course, there arepurely parodic hexameters among Hipponax’s fragments—cf. fr. 128–9a IEG; see alsoWest (1974), 30, Degani (2002), 217 n. 96, Carey (2009), 163–5.

33 For the identity of the speaker, see Degani (2007), 79. Note the exotic term��ºı (¼ king), playing upon ˚ıºº��Å ��Æ; cf. also Masson (1962), 103 withn. 4.


hymnal settings; at frs. 3, 3a, and 32 the god is described by means of areference to his parentage (son of Maia) and an important Kultort, hisbirth-place (Cyllene), while fr. 3a exhibits another hymnal trope,namely the accumulation of some of the god’s main by-names andcharacterizations (Œı��ªåÆ, ˚Æ��ÆFºÆ,34 çøæH� ��ÆEæ�). Hipponax’sinvocations of Hermes, furthermore, are comically incongruous,since the ‘higher’ form of prayer is used to express ‘lowly’ requests.35

To these we should add two more fragments. First, fr. 47 IEG (�Ææ!fiz �f º�ıŒ��º�� æÅ� ��Æ j �æe b� Œı��Ø �e� +ºıÅ��ø�% EæB�), which may have comic intent: as Degani observes,36 thename +ºı��Ø�Ø may be fictitious, derived from çºı�Ç�Ø�; the versewould thus amount to ‘Hermes of the Chatterboxes’.37 Second, fr. 79.9–10 IEG (% EæB �! K % I��ÆŒ�� IŒ�º�ıŁ��Æ j . . . ��F Œı�e �e�çØºÅ��), where West proposes ‘and Hermes having given escort toHipponax’s [had kept safe] the burglar from the dog’.38 Although it isdifficult to interpret these fragments with a high degree of certainty,given that we are lacking the context, it is nevertheless evident thatthe iambic poet presents himself as having a particular relation toHermes.

We observe a similar presentation of Hermes in Old Comedy. Thehero of Aristophanes’ Peace, Trygaios, flies to Olympus, where heencounters Hermes, who is depicted in an undignified manner, remin-iscent of what we find in the fragments of Hipponax. The god is thedoorkeeper of Olympus, left there by the other gods to guard theirpetty household equipment.39 Interestingly, Hermes’ initially hostilereception of Trygaios changes as soon as the mortal offers him a fewpieces of meat.40 It will be remembered that in the Hymn to Hermesthe god is twice said to be ‘desirous of meat’ (Œæ�ØH� KæÆ��Çø�; 64,287) and that he actually attempts to taste some of the meat he hasroasted at the Alpheios river, but is unable to consume his portion.41

34 See Masson (1962), 104–6.35 See West (1974), 29.36 Degani (2007), 107.37 If correct, this may point humorously to Hermes as the patron of oratory.38 See West (1974), 144.39 Ar. Pax 201–2.40 Ar. Pax 191–4.41 In ps.-Apollod. 3. 113 Hermes eats some of the meat. Hermes’ greed is promi-

nent also in Ar. Pl. (1136–8, 1118–32), where the god laments his hunger because ofthe loss of his earlier offerings. Cf. also Telecl. fr. 35 PCG. t ��Ł! % EæB, Œ�� H�ŁıºÅ��ø� (‘Lord Hermes, gulp down some of the offerings’).


The exchanges between Trygaios and Hermes are humorouslyirreverent. For instance, at 400–2 the mortal unscrupulously declaresHermes a thief, stating it in a rather casual manner as if he andHermes were equals who had been on the friendliest terms for along time (note the endearing diminutive % Eæfi ��Ø��, ‘my dear littleHermes’ at 924). The same may be said of 363–5, where Trygaiosexpresses his confidence that Hermes as the god of luck and lot willhelp him. Furthermore, it is only after Trygaios explains to Hermesthe material loss that the collapse of Greece would entail for the godsthat Hermes decides to help him.42 Such a reversed relation betweenHermes and a mortal occurs also at Wealth 1139–70: whereas nor-mally it is the mortal who reminds a god of his previous service tohim (or her) in order to justify his request (da quia dedi), here it isHermes who reminds Karion of his previous services to him andrequests that the humans admit him.43

All this is of course humorous, since the gods (by definition ‘lofty’characters) are depicted in human fashion and implicated in loweractions, which creates comic incongruity.44 This representation of thegods is crucially relevant for the Homeric Hymn to Hermes. Thearchaic hymn recalls and foreshadows in many ways what we havejust observed regarding the presentation of Hermes in iamb andcomedy. To begin with, the praised deity is presented in his infancy:Hermes accomplishes all his Œºı�a �æªÆ (‘glorious deeds’) when he isjust two days old. Of course, the cattle-theft myth is very old, goingback to an Indo-European archetype,45 and Hermes’ cattle-theft iswell-known. But the attribution of this action to the newborn Hermes(probably an innovation of Alcaeus, who also composed a hymn toHermes)46 certainly has a great potential for comedy and even thegrotesque. In fact, Middle Comedy saw a proliferation of the Ł��F

42 His argument is that Helios and Selene, to whom the barbarians sacrifice, havebeen conspiring with the barbarians against the Olympians so that, once the Olym-pians have been removed, they can enjoy the sacrifices they used to offer them. Cf. Ar.Av. 1515–24.

43 Hermes appears also in Phrynichus fr. 61 and fr. 204. Philiscus wrote a playentitled % Eæ�F ŒÆd �çæ��Å ªÆ�Æ� , of which unfortunately not a single fragmentsurvives. Divine satire occurs elsewhere in Comedy; cf. the divine ambassadors in Ar.Av. 1565–693 or the presentation of Dionysus in Ra. (passim).

44 See Nesselrath (1995), 10 on the archaic precedents for such comic incongruity;he cites Arist. Po. 1448a 17 ff. and 1449a 32–4 for the ‘degrading of characters’. For thecomic presentation of the gods in Homer, see below (pp. 97–8).

45 See Lincoln (1976), Walcot (1979), West (2007), 451–2.46 See Cairns (1983), 29–33.


ª��Æ� plays,47 while Hellenistic poetry picked up this theme to a greatextent.48 But this is not all: the infant Hermes, still wrapped in hisswaddling-clothes, delivers defence speeches to Apollo and Zeuswhich are highly polished in terms of their organization and rhet-orical effects (261–77, 368–86).49 Hermes, furthermore, is a tricksterwho outsmarts Apollo, a god who, though both older than him andthe patron of divination, cannot discover his cows’ whereabouts andhas to ask a mortal and his infant brother for information (190–200,254–9). He steals fifty cows from Apollo’s herd and leads them awaybackwards to confuse their owner, while he fabricates a set of won-drous sandals which create footprints that cause Apollo’s amaze-ment.50 As an infant, he constructs a lyre with a view to using it asa means to barter with Apollo for the cattle later (34–5, 47–51).Finally, his actions cause his audience to laugh: at 281 Apollo laughsat Hermes’ defence speech and feigned indifference, while at 389 it isZeus’ turn to laugh at Hermes’ skill in lying.51

Furthermore, there is an indirect reference to the iambic ethosin Hermes. At lines 55–6 Hermes’ improvised hymn to himself iscompared to the playfully teasing words young men direct at eachother at the symposia ( . . . M�� Œ�Fæ�Ø j ��Å�Æd ŁÆº�fi Å�Ø �ÆæÆØ��ºÆŒ�æ��ı�Ø�, ‘just as young men tease each other with indirect attacksat banquets’).52 These Œ�æ��ÆØ were a kind of ‘indirect speech

47 Nesselrath (1995).48 See Ambühl (2005).49 See already Radermacher (1931), 127–8, Görgemanns (1976), 115. Hermes uses

what Kennedy (1994), 14 considers ‘the first specific example of argument fromprobability . . . in Greek’. The poet attempts to represent Hermes’ speech as child-like: in Hermes’ defence speech to Apollo (261–77) we observe repetition in diction(263–5, 265–6, 275–6) and rhythm (eleven of the seventeen lines end with a verb-formin the metrical shape wqq [261–6, 269–71, 274, 277]; of these, six include a trochaicnoun before the verb-form [263, 265, 266, 269, 274, 279], while 268 reproduces thismetrical pattern, though without a verb). Hermes’ speech, furthermore, consists ofshort clauses that sometimes lack connectives (263–4, 266–7, 273). All of this con-tributes to creating the impression of child-talk; see van Nortwick (1974), 94–5.

50 Cf. above n. 8.51 See Halliwell (2008), 100–3, with Furley in this volume (pp. 224–5), who

compares Hermes’ actions in the Hymn to the machinations of a seruus callidus incomedy, at which the audience (both internal and external) has to laugh.

52 For such Œ�æ��ÆØ, cf. IEG 27. 1–6 (Adespota); Alex. fr. 9. 8–10 PCG (withArnott [1996] ad loc.); A.R. 1. 457–9 ��Ø�Æ �’ I�Ø�Æ�d Iºº�º�Ø�Ø� j ıŁ�F�Ł! �x�� ººa ��Ø �Ææa �ÆØ�d ŒÆd �Y�øfi j ��æ��H �łØ�ø��ÆØ, ‹�! ¼Æ�� o�æØ I��Å, ‘thenthey were addressing each other in turns, as young men often pleasantly amusethemselves by food and drink, when insatiate insolence is missing’ (which may be


characteristic of young men at feasts, that involved assertion of statusand could sometimes be playful’.53 This is precisely the implication ofthe simile introducing Hermes’ hymn to himself: the young godattempts to assert his status by means of his improvised song, em-phasizing the duration of Maia and Zeus’ affair and his parents’equal status (note ��ÆØæ��fi Å çØº��Å�Ø, ‘companionable affection’ at 58),as well as praising the wealth in his mother’s cave.

But we can go a step further. J. Clay has demonstrated how themajor Homeric Hymns represent a critical moment in the divinecosmos. The narrator depicts a crisis on Olympus, which threatensto destabilize the world as Zeus has organized it. Zeus then has to actas a mediator, and his involvement leads into a re-definition orre-distribution of divine timai.54 The Hymn to Hermes conforms tothis pattern. Hermes is the last born of the Olympians, which causes aproblem: the divine honours have already been distributed. In orderto acquire his own place in the world of the Olympians, Hermes(whose background is both Olympian through Zeus and Titanic/Promethean through Maia) resorts to stealing what belongs to Apollowith a view to exchanging it later. By doing this, he enacts some of histraditional roles. But the poet adds a twist to this theological scheme.The gods in this Hymn are humanized to the extreme: Hermes, theyoungest brother, covets the possessions of the elder (Apollo). Heeven threatens to resort to (further) thievery if his father does not givehim what he thinks he deserves (174–81). Maia is portrayed as amother who reprimands her son for returning home late (150–61).Her words at 160 (�ææ� ��ºØ�, ‘get lost to where you came from’)indicate her frustration. The quarrelling brothers have to resort totheir father in order to settle their dispute (322–4). Thus a seriousproblem that could destabilize Olympus (what happens when a newgod is added to the already existing pantheon?) is treated in a comic

an echo of Herm.); see further Reitzenstein (1893), 26 n. 2, and MacDowell (1971) onAr. V. 1308–13. On the semantics of Œ�æ��Æ and Œ�æ��E�, see Clay (1999), 620–1with n. 11, Clarke (2001), Lloyd (2004), Gottesman (2008), esp. 6–7. I do not agreewith the suggestion in Compton (2006), 43 n. 5 that Hermes’ song is satirical: Hermes’hymn to himself is playful on account of its possibly risqué topic, yet serious as itshares many points with the opening to the actual Homeric Hymn to Hermes. Whomwould Hermes be satirizing? For the correspondence between the poet’s song anddivine musical performances, see Calame in this volume (p. 348–50).

53 Gottesmann (2008), 11.54 Clay (1989), 15 and in this volume (Ch. 11). Cf. Introduction (pp. 19–20).


way by the Hermes poet, who renders the divine actors in extremelyhuman terms.

Such incongruity occurs also at the linguistic level: Hermes’ languageis deceptively childish.55 Furthermore, lofty epic language describes suchlowly matters as digestive distress. When Apollo seizes Hermes in hisattempt to force him to reveal where he had hidden his cattle, Hermescounters this bie with his metis (note �� . . . çæÆ�� , ‘havingcontrived’ at 294), by ‘letting forth an omen, a wretched labourer ofthe belly, an insolent messenger’ (�Nø�e� �æ�ÅŒ�� . . . j �º��ÆªÆ��æe �æØŁ��, I��ŁÆº�� Iªª�ºØ��Å�, 295–6).56 The combination ofthe word �º�ø� (‘wretched’), generally associated in early poetry withsuch higher concepts as Łı� and łıå� (‘spirit’), with his flatulencewould cause comic deflation, for as Katz (1999), 318 observes, ‘ . . .merely upon hearing �º��Æ, the audience of Hermes would associatethis word with a “lofty” spirit and thus find amusement in the rather lesspositive sort of wind evoked by a ªÆ��æe �æØŁ��.’57

The divine world of this Hymn does not differ significantly fromwhat we encountered in Hipponax and Aristophanes. It may of coursebe objected that Hermes is not the only archaic epic poem that exhibitsdivine comedy. Early epic, after all, offers some examples of such treat-ment of the gods. For instance, at Iliad 1. 571–600,Hephaestus attemptsto preventHera from arguing againstZeus, reminding her howZeus hadthrownhimdown fromOlympuswhenhe attempted to protect her fromhis physical abuse. Themood is lightenedwhenHephaestus offers nectarto Hera and then bustles around in Zeus’ palace pouring nectar to thegods’ cups—an ugly and lame god usurping the role otherwise reservedfor the beautiful and young Ganymede or Hebe. The result is of courselaughter.58 As mentioned above, in Odyssey 8 Ares and Aphrodite aretrapped on Hephaestus’ bed, caught by the invisible net that he hadcreated after Helios informed him about their adulterous affair. This

55 See above, n. 49.56 For Hermes’ ‘fart-omen’, see Katz (1999), esp. 317–19, and Bain (2007), 51–2,

who points out the riddling dimension of this description. See Degani (2007), 9, forsimilar parodic devices in Hipponax’s poetry, which are relevant to Hermes (cunningstratagems, coining of new words, use of lofty epic words or phrases in a lowercontext, detorsio Homeri, and etymological puns).

57 It will be remembered that at Ar. Eq. 639 a fart is also treated as an omen. See theparallels in Katz (1999), 316 n. 3.

58 On the gods’ laughter at Il. 1. 571–600, see Halliwell (2008), 58–64. On p. 63,Halliwell proposes that the gods’ laughter is caused by Hephaestus acting as a ª�ºø��-��Ø� parodying Hebe and Ganymede. This interpretation was already present inscholia bT and A on Il. 1. 584.


too results in laughter, as themale gods gather aroundHephaestus’ bedand look at the trapped adulterers.59

These scenes are humorous, to be sure, but they differ from thecomedy of Hermes. In the Hymn the humour is sustained throughout,while in Homer the brief comic scenes interrupt (and comment on)the main narrative.60 The Hymn poet seems to enjoy reminding usthat Hermes is an infant when he accomplishes his deeds: his swad-dling-clothes are mentioned no fewer than six times (151, 237, 268,301, 306, 388). Furthermore, in both Homeric scenes the gods’ laugh-ter suggests Schadenfreude. In the Iliadic scene the gods laugh at theexpense of a crippled god who acts as the wine-pourer, while in theOdyssey a group of males laughs at the expense of the cuckoldedhusband (even if they admire his craftiness).61

HERMES ’ EPIPHANY

We have thus far examined the amusing aspect of Hermes’ divinepersona, which is manifested both in his presence in works of the epic,iambic, and comic traditions as well as in the way in which the poet ofthe Hymn to Hermes represents the god’s story. Hardly anyone wouldexpect a magnificent epiphany to occur in such a setting. But there isanother set of considerations that may account for the peculiarities ofhis ‘epiphany’ in the Hymn. Hermes is a cunning god, always in

59 To these scenes one may add Aphrodite and Ares’ wounding by Diomedes inIl. 5, the Dios Apate in Il. 14, and the Theomachy of Il. 21. The Theomachy is notessentially different from the divine scenes in Il. 1, 14 and Od. 8: even though it seemsto have a more public character, it appears so only to the poem’s privileged audience.The human actors of the poem are as much unaware of the Theomachy as they are ofHephaestus’ nectar-pouring in Il. 1. For the anthropomorphic presentation of thegods in the Dios apate, see the comments in Golden (1990), 48–50.

60 This is not to say that these scenes are not related to the main story; seeBielohlawek (1930), 195–202, Sikes (1940), Rinon (2006), 211–13.

61 It is worth noting that both Homeric scenes involve an underdog or tricksterfigure who wins either through humour (Il. 1) or through metis (Od. 8). Moreover,just as Hermes’ story presented in Herm., both divine scenes portray the gods in ahumanized fashion, involving problems in the most basic human institution, thefamily. The Iliad presents a father abusing his wife and the son who tries to protecthis mother; the Odyssean tale involves adultery. On the humorous aspects of Homer,see Friedländer (1934); for a different presentation with emphasis on their seriousaspect, see Griffin (1980), 144–204.


motion, closely linked to night and darkness: the cattle-theft takesplace at night, as the poet reminds us many times (67–9, 97–100, 141,155, 341), and the god is also referred to as the companion of thenight (290; cf. 358).62 His nature as a tricky god who operatesstealthily through metis can help us understand the nature of hisepiphanies, when they do occur.

In Iliad 24, Hermes appears to Priam and Idaios at Zeus’ behest tohelp them cross into the Greek encampment and reach Achilles’ tentunnoticed (358–467). Although the god eventually reveals his divineidentity, his epiphany differs significantly from other epiphanies inboth Homer and the Hymns. Instead of a grandiose appearance,Hermes’ self-revelation is subtler and preceded by his enactingsome of his traditional functions.63

To begin with, the form in which Hermes first appears to Priam isthat of a young man on the verge of adulthood (347–8). He isIŒ�ŒÅ�Æ, ‘harmless’ (370–1 Iºº’ Kªg �P�� Þ�ø ŒÆŒ�, ŒÆd � Œ��¼ºº�� j ��F I�Æº��ÆØØ, ‘but I will not harm you at all, and I willward off anyone else from you’). His father’s name is —�º�Œ�øæ (¼‘he who owns many possessions’, 397), which is appropriate for agreedy figure such as Hermes.64 He is the youngest son (399) whowas chosen by lot to join the Trojan expedition.65 At 439 he refers tohis own action of escorting (�PŒ ¼� �� Ø ��e� O�� Æå�ÆØ��, ‘no one would fight against you in scorn at such an escort’;cf. his titles ��ÆE� , łıå�� ); at 440 he is called KæØ��Ø� , anadjective which in antiquity was often understood as ‘beneficial’.66 At445 he casts sleep on the eyes of the Myrmidon guards (thus enacting

62 Note too that Hermes is associated with sudden silence, as a proverbial phrasecited by Plu. Mor. 502–3 reveals: ŒÆŁ��æ ‹�Æ� K� �ıºº�ªøfi �Ø�d �Øø�c ª�Å�ÆØ, �e�% EæB� K��Ø��ºÅºıŁ�ÆØ ºª�ı�Ø� (‘just as when at a gathering there is silence, they saythat Hermes has suddenly come in’).

63 SeeTurkeltaub (2007), 60with n. 28,whopoints out thatHermes’ ‘individualizingpersonality . . . shapes the scene’.

64 Cf. Il. 16. 180–6 for Hermes’ affair with Polymele (¼ ‘she who owns manyflocks’); his son was called Eudoros (‘he who possesses good gifts’).

65 Hermes was the patron god of lot; cf. Suda Œ 1785, Eitrem (1906), 258, Olson(1998) ad Ar. Pa. 365, Jaillard (2007), 126 with n. 143.

66 On the meaning of this epithet, see Reece (1999). Note that four out of sixoccurrences of KæØ��Ø� in the Il. appear in book 24 (360, 440, 457, 679; cf. 20. 34, 20.72). The fact that in this section of the Il. Hermes indeed helps Priam, may suggestthat this is how the poet understood the adjective too, and thus Hermes is enacting the(perceived) meaning of this title.


his role as god associated with sleep), while he opens the gate at 446(thus performing his role as the god of doors, locks, and hinges).67

It is also significant that when Hermes reveals his divine identity toPriam, he does not change his appearance; nor is there any radiance,fragrance, change in size, or any other physical manifestation, incontrast to other divine epiphanies both in the Hymns and inHomer.68 Unlike Dionysus when he is kidnapped by the piratesin Hymn 7, Hermes does not perform any miracles in front of themortals. And Priam and Idaios certainly do not experience any fear,sinceHermes quietly reveals who he is and that he was sent by Zeus tobe the old king’s escort, gives instructions to the old king, and departsfor Olympus. Significantly Priam does not say anything in reply, justas the Old Man in Hermes does not respond to Hermes’ brief address.

Hermes has a similar appearance in Od. 10. 277–9, where heappears to Odysseus. This event is presented in a mediated form inthe hero’s narrative, which shares some of the wording with thedescription of Hermes’ appearance in Iliad 24.69 There are somethematic similarities as well: Hermes helps the hero, he gives himinstructions, but there is no proper epiphany (or if there was one, itwas suppressed by Odysseus’ narrative).70

67 See Farnell (volume 5), 66, on the evidence for Hermes’ cult-names �æ��ºÆØ� ,��ºØ� , ŁıæÆE� , ��æ�çÆE� .

68 On the differences between divine epiphanies in Homer and the HomericHymns, see Dietrich (1983), 70–1, Pucci (2008), 59–60, 81–2, who speaks of ‘blank’epiphanies in Homer, in the sense that we do not know in what form the god appearsto the mortal, and Faulkner (2008), 235. But note that even in the ‘blank’ epiphanies inHomer there are physical manifestations that hint at divinity; e.g. Il. 1. 199–200(Ł��Å�� ! �åØº�� , ��a �! K�æ��! , ÆP��ŒÆ �! �ª�ø j —Æºº��! �ŁÅ�Æ�Å�· ��Ø�g� �ƒ Z�� ç�Æ�Ł��, ‘and Achilles was astonished and turned around and immediatelyrecognized Pallas Athena; the flash of her eyes was terrible’), 3. 396–8 (ŒÆ� Þ! ‰ �s�K��Å�� Ł�A ��æØŒÆººÆ ��Øæc� j ��Ł�Æ �! ƒ�æ��Æ ŒÆd ZÆ�Æ ÆæÆ�æ��Æ, jŁ��Å�� ! ¼æ! ��Ø�Æ . . . ; ‘and when she observed the beautiful neck of the goddessand her breasts and her shining eyes, she was then astonished . . . ’), or Poseidon’s feetand shins at Il. 13. 68–75. Note too that a god’s divine identity may be revealed by hisor her transformation; cf. Od. 3. 371–2 (S ¼æÆ çø��Æ�! I��Å ªºÆıŒH�Ø �Ł��Å jç��fi Å �N��Å· Ł�� ! �º� ��Æ N��Æ , ‘Athena spoke thus and departed in thelikeness of a sea-eagle; and astonishment seized all who saw it’).

69 Cf. Il. 24. 347–8 �B �! N�ÆØ Œ��æøfi ÆN�ı�Å�BæØ K�ØŒg j �æH�� Å��fi Å, ��F ��æåÆæØ��Å X�Å (‘he went forth in the appearance of a young prince at the age when hegrows his first beard, whose youth is most graceful’)� Od. 10. 277–9 ��ŁÆ �Ø % Eæ��Æ åæı��ææÆ�Ø I��ºÅ�� j . . . ��Å��fi Å I��æd K�ØŒ� , j �æH�� Å��fi Å, ��F ��æåÆæØ��Å l�Å (‘then Hermes of the golden wand, met me . . . in the appearance of ayoung man at the age when he grows his first beard, whose youth is most graceful’).

70 See De Jong (2001), 260–1.


This kind of divine presence in which the god enacts some of hisfunctions points to another way of understanding Hermes’ peculiarepiphany in Hermes. Hermes’ presence is subtle and is revealed toboth the internal and external audience of the poem through his �æªÆ(‘deeds’).71 The poet establishes an analogy between the hymnal past(the time when Hermes appeared in the world and introduced hisinventions and cultural institutions) and his own present (whenhumans use these inventions and re-enact the primal performanceof the first ‘hymn to Hermes’). The works and inventions of Hermes,some of which are instrumental for the performance of the veryHymn, establish links between the mythological past and audience’spresent. The lyre which Hermes creates at the beginning of the poemand uses to accompany his own hymn to himself is the same instru-ment the poet uses when performing the Hymn to Hermes which weare hearing.72 Hermes is not only the inventor of the instrument, butalso the founder of the hymnic genre (or at least the composer of thefirst ‘hymn to Hermes’). Furthermore, the �Æd K��Å (a banquet inwhich the food is distributed equitably to the guests) that Hermesestablishes at the banks of Alpheios also forms part of the audience’sexperience, as feasting and dining, both private and public, was animportant cultural institution throughout Greek antiquity.73 In add-ition, if the Hymn was performed at a banquet setting, as has beenproposed,74 then Hermes’ transferring the lyre to Apollo may be aproleptic device. Standing on Hermes’ right, Apollo receives the

71 See Versnel (1987), 50–2, who shows that for the ancient Greeks a god’simmediate presence could be perceived through his or her miraculous deeds, evenwithout the visual perception of his or her actual appearance. Versnel further pointsout that ‘the term epiphaneia denotes two things: the personal appearance of a godand his miraculous deeds . . . [t]o the Greek mind both constituents are merely the twoaspects of one indissoluble unity, of which now the one, then the other may dominatethe description.’ The poet of Hermes emphasizes the god’s Œºı�� or Iæ��ÅÆ �æªÆ(rather than the physical aspect of his divinity) which reflect Hermes’ divine presence.

72 Cf. Clay in this volume (pp. 244–5). The link between the internal performancein a Hymn and the mortal singer’s performance can be found in other Hymns as well,as Calame (Ch. 14) argues in this volume. What distinguishes Hermes from the otherHymns is that the performer re-enacts the god’s original performance (instead ofmerely representing or imitating it).

73 For the interpretation of the events at the Alpheios as a dais, see Clay (1989),121–2.

74 Clay (1989), 7, and in this volume (p. 233); see also Nobili (2008), 288–91, whoproposes a semi-public symposion as a possible performance venue, and cf. in thisvolume Introduction (pp. 16–19).


instrument and begins to sing. Thereby the two gods enact the K��ØÆ�æªÆ (lit. ‘the [youths’] deeds [at the banquet], going from left toright’, 454) with the lyre.75 Furthermore, Greene has noted the use ofproleptic epithets for Hermes in the Hymn’s proem.76 But suchproleptic characterizations are not confined to Hermes. For instance,at 31 the young god addresses the tortoise in terms that envision it asa musical instrument, using the same double entendre that we willmeet later in the Hymn when Hermes describes the lyre as a hetaira.77

At the same time, this address creates yet another link between thepoem’s mythic time and the audience’s present. The actions ofHermes in this part of the poem have an effect in the audience’sworld since this å�æ�Ø�� (‘[the companion] who beats therhythm’) is present at the poem’s performance.

This analogy (or perhaps temporal collapse) is supported by aetio-logical expressions that emphasize the permanence of the effects ofHermes’ first appearance in the world for humans and gods alike. Atlines 124–6 we learn how Hermes disposed of the hides of theslaughtered cows: ÞØ��f �! K��ı�� ŒÆ�Æ��ıçºøfi K�d ��æfi Å, j ‰ ��Ø �F� �a �Æ��Æ ��ºıåæ��Ø�Ø ��ç�Æ�Ø j �Åæe� �c ��a �ÆF�Æ ŒÆd¼ŒæØ�� (‘and he stretched the cow-hides on a rugged rock, so thatthey are there still now, many years thereafter, a long and incalculabletime after these events’). The number of temporal markers in thesefew lines is remarkably high, and the poet emphasizes the continuityof a ritual practice despite the long time that has passed since Hermesinaugurated it. Likewise at 507–10 the poet underscores the perman-ence of Hermes and Apollo’s reconciliation.78 Here the poet affirmsanother point of continuity between mythic past and his present: thereconciliation of the two divine brothers lasts even today, and this is

75 See Manuwald (2002), 161.76 See Greene (2005), 344, who calls these proleptic epithets of the proem ‘a

metaphorical bank account of traits from which the child can draw as he wins histimai in succeeding episodes’.

77 Cf. 478–86. This eroticized reference to the musical instrument is exploited in acomic context in Pherecrates’ Cheiron (fr. 155 PCG), where the sufferings inflicted onMousike by the poets who practised the so-called ‘New Music’ are also presented inerotic terms.

78 ŒÆd �a b� % EæB j ¸Å��&�Å� Kç�ºÅ�� ØÆ��æ , ‰ ��Ø ŒÆd �F� j ��Æ�! , K��dŒ�ŁÆæØ� b� % EŒÅ��ºøfi Kªªı�ºØ�� j ƒ�æ��, ��Æ� › �! K�øº�Ø�� ŒØŁ�æØÇ�� (‘AndHermes on the one hand, loved Leto’s son continuously, so that there is proof eventoday; for he handed the lyre to the Far-Shooter, and he began to play the lyre skilfullyon his arm’). Note that I do not punctuate after ŒÆd �F�.


proven by the fact that Apollo is still the god of the lyre. This proof isthe sema that indicates to the poet’s audience the truth of his words.

To be sure, the power and presence of the gods is felt through thepermanence of their inventions or deeds elsewhere as well. Theestablishment of the Delphic oracle in Apollo abides in the audience’sexperience. The same can be said of the rituals established by Dem-eter in the long Hymn dedicated to her. And the re-definition ofAphrodite’s power and the ensuing interruption of the generation ofheroes is certainly relevant to the audience’s human condition. Butagain, the Hermes poet takes this theme a step further: the veryperformance of Hermes is a re-enactment of the god’s deeds byusing some of his inventions. By performing the Hymn to Hermesthe poet assumes the role of Hermes by repeating one of the god’smost significant actions, i.e. the composition of the poem.79 Thisidentification does not occur only on a symbolic level as a re-enact-ment, but also in more concrete terms twice in the Hymn. First, inHermes’ hymn to himself (54–63). In this miniature mise en abîmethe poem’s protagonist performs a song of the same genre as theenveloping text (i.e. a ‘hymn to Hermes’). Although Hermes’ ownhymn is introduced by Iç� with the accusative (as opposed to theway the enveloping text begins, i.e. % EæB� o��Ø )�F�Æ . . . , ‘Sing,Muse, of Hermes . . . ’),80 the contents of both versions are never-theless very similar. Both Hermes and the poet’s proems emphasizethe god’s parentage and stress the duration of Zeus and Maia’s love-affair.81 This coincidence in the song’s contents validates the poet’sown account: if the god were to praise himself, he would haveperformed a song similar to the one we are in fact hearing. Second,in Hermes’ words at 475–88, the divine child explains to Apollo howto use the lyre. Apollo should ‘question’ her only with the properpreparation and knowledge if he wants to receive a pleasing answer.But if someone who is inexperienced questions her violently, sheproduces an ill-sounding reply. While these words are spoken bythe character Hermes, they are actually uttered by the poet who is

79 For the identification of the voice of the divinity and the enunciator, see Capponi(2003), 19–20, 29–30 (on Apoll.). Similarly, during a re-performance of Hesiod’sTheogony the bard assumes the persona of Hesiod; cf. Clay (2003), 3 n. 6.

80 The introduction of a Hymn by Iç� þ accusative is found in Hys. 7. 1, 19. 1, 22.1, 33. 1.

81 Note the iterative Tæ�Ç��Œ�� at 58, which is surely meant to remind us of theiterative Ø�ª�Œ�� at 7 and the iterative optative of 8 ŒÆ�� . . . �å�Ø.


impersonating Hermes at this point. We can imagine a professionalbard speaking about his art in the same terms as Hermes.82 The poetthus identifies with his protagonist, Hermes, and this gives the im-pression that the god is present through the performance of a Hymnlike the one he himself had originally composed.83

Hermes’ epiphany is then of a different kind. Instead of beingexplicitly narrated as in the other Homeric Hymns, it is broughtabout indirectly through the narration of the god’s Œºı�a �æªÆ(‘famous deeds’) which are not only described but are also enactedwithin the Hymn’s performance. The humorous aspects of Hermes’divine persona link the Hymn with the similar treatment of the god iniambic poetry (a genre which might have had a ritual-dramaticperformance component)84 and Old Comedy, which precludes afull epiphany such as the ones we meet in the other long HomericHymns or (sometimes) in Homer. Furthermore, in each performanceof Hermes, the bard re-enacts some of the god’s wondrous deeds. Thesinger follows closely the god’s example in praising himself. And justas the singer’s performance partly re-traces Hermes’ own hymn, hisaudience resemble the poem’s internal audience by sharing with thema similar reaction to the god’s story, namely laughter.85 Through thissubtle evocation of the god’s presence, the poet adapts the hymnicconventions to Hermes’ nature as a trickster god, who operates withstealth and metis and whose actions often cause amusement.

82 Cf. Radermacher (1931), 157, who remarked ‘Now it is notHermes speaking anylonger; now the poet himself speaks of the art that is entrusted to him, and he does thiswith warm sensitivity’ (translation mine). For Hermes’ musical performances in theHymn, see also Vergados (forthcoming), introduction, V.

83 Cf. Sinos (1993), 83–5 on the ritual impersonation of gods attested in theepigraphic record. Drawing on parallels from the Divine Liturgy of the EasternOrthodox Church, she speaks of the audience’s ‘playful participation in a shareddrama’, a drama that has the effect of causing the transcending of temporal bound-aries, so that the human audience is elevated to the mythic time, ‘hav[ing] beenremoved from their mundane world and hav[ing] left historic time to experiencethe heroic past, the time of myth, when gods and humans mingled more freely thanthey do in our world’.

84 For the ritual origins of iamb, see West (1974), 22–39. Carey (1986), 63–7,argues contra the ritual component of Archilochus’ poems against Lycambes and hisdaughters; cf. also Bowie (2002), 42–3.

85 For this ‘dramatic’ form of epiphany, see Bierl (2004), 45, who reads E. Ba. andAr. Pax and Pl. as ‘epiphanies in performance’.


6

An Erotic Aristeia

The Homeric Hymn to Aphroditeand its Relation to the Iliadic Tradition

Pascale Brillet-Dubois1

To Jenny Clay

The Hymn to Aphrodite celebrates the power of the love goddess byrecounting her seduction of the mortal Anchises. Scholarship on thisdelightful piece has been discussing for two centuries its complexrelationship with the other Hymns of the collection as well as otherepic poetry, most of all the Iliad with which the Hymn shares anumber of formulae, verses, and thematic sequences.2 Debates haveso far focused on three main questions: (1) When was the Hymncomposed?3 The communis opinio that the poem’s diction and lan-guage—as they appear in the text that we have—are later than theIliad ’s and older than most of the other Hymns’ is now grounded in

1 This chapter is a considerably reworked version of Brillet-Dubois (2006) and ofmy contribution to the Oslo Symposium on ‘Relative Chronology in Greek Epic’organized in 2006 by O. Andersen, D. Haug, and A. Maravela-Solbakk. I thank all thecolleagues who discussed my ideas and helped me to bring them to maturity. S. Scheinand A. Faulkner deserve special mention for their constant support and challengingcritique.

2 The union of Aphrodite and Anchises is mentioned in the Iliad (2. 820–1; 5.311–13) and the Theogony (1008–10).

3 Discussion of the different hypotheses in Janko (1982), 151, Faulkner (2008),47–9, and in this volume Introduction (p. 14).

thorough statistical analyses,4 but no certainty can be reached aboutabsolute dating. (2) Is the Hymn to Aphrodite imitating the Iliad?During the nineteenth century and well into the twentieth, the ex-tensive use of Homeric formulae and vocabulary in the Hymn toAphrodite was considered proof of conscious imitation of Homer bythe Hymn poet.5 With the rise of oral studies following MilmanParry’s work, new interpretations were made of those similarities:they could indicate that the poet was drawing from the same trad-itional stock of formulae as Homer6 and that the Hymn was primarilyan oral composition.7 The debate on the originality of the poet hasnot been resolved to this day, as appears clearly in Faulkner’s intro-duction to his commentary on Aphrodite: although he states in a verycautious way that ‘it is impossible to be certain of direct influence’,8 hebelieves that some sort of imitation, whether between stable oralpoems or written texts, is likely and lists parallels to Homer andHesiod under the label ‘imitatio’.9 To put things simply, such obviousparallels between two hexameter poems as the ones connecting theHymn to Aphrodite to the Iliad now receive two main types ofexplanations: independent composition within the same tradition,or conscious interaction with either the other poem or a commonsource. Oralists often challenge the idea that the poems are quoting oralluding to each other. They insist that, although the poets have beentrained to use certain common formulae, typical scenes and myths,the poems have been composed in different performances and theirmeaning develops independently.10 On the side of interaction,

4 Hoekstra (1969), Janko (1982). Notable exceptions: Reinhardt (1961) andHeitsch (1965) who respectively suggest that the same poet composed the Il.and Aphr., or the Hymn and an interpolated Iliad 20; Freed and Bentman (1955)believe the poem to be Hellenistic.

5 Discussion of the meaning of these parallels has been overshadowed by theendless and disproportionate debate about the existence of Aineiadai as patrons ofone or both texts; see Faulkner (2008), 3–18. I agree with the many scholars who,especially since the survey of the evidence by Smith (1981b), consider the testimoniesconcerning the historical existence of Aineiadai in Troad to be inconclusive andbelieve that the hypothesis of one or two poets composing the Hymn and Il. 20 topraise them is unnecessary to explain the prophecies about Aeneas’ descendance.

6 Porter (1949), 266–7.7 Notopoulos (1962), followed by Preziosi (1966).8 Faulkner (2008), 31, 23–5.9 Faulkner (2008), 31, 35.

10 Nagy (1979), 267, on Il. and Aphr.; Burgess (2001), 134. On this issue,cf. Introduction (pp. 3–7).

106 Pascale Brillet-Dubois

imitatio as a derivative relation between the great Homeric andHesiodic poems and later, minor, or ‘sub-epic’ hexameter poems isgiving way to more complex (and stimulating) analyses of epic poetry.These combine the assumption that the poems are traditional andorally derived with the neo-analyst search for traces of reference topre-existing works. Poems, defined not as texts but as products ofpoetic traditions, can allude to each other in a dynamic way andreference to other extant traditions can be used to convey meaning toan informed audience.11 This does not imply any more that theexemplum is a written text,12 though it is hardly conceivable withoutthe poem or the poetic tradition it belongs to having reached a form,be it oral or written, that is stable enough for an audience to identifyit. In other words, the genealogical model of the relation between thepoems is now being supplemented by a dialogic or even agonistic one.(3) Questions about the hymnic genre also appear central to theinterpretative debate on the Hymn to Aphrodite. While Aphroditeseems to share the other Homeric Hymns’ compositional character-istics,13 it also stands as the least cultic of all the longer Hymns: theconventional mention of Aphrodite’s temple in Paphos and Anchises’promise of offerings (100–2) could hardly be interpreted as an aeti-ology for an extant cult.14 In the analysis of the relationship betweenthe Hymn to Aphrodite and non-hymnic epics, the problem of genrewas first raised as a means to define the hymn’s originality andspecific poetics: Porter, Podbielski, and Lenz insisted that, for all itsparallels in heroic poetry, the poem makes use of traditional materialthat is subordinated to the celebration of Aphrodite’s power. But thepoem as a hymn also met with great interest among scholars who,following upon Vernant’s groundbreaking work on Hesiod, devel-oped anthropological readings of early Greek poetry. That trend ofstudies culminated with J. Clay’s major book The Politics of Olympus,which examines the cosmogonic aspects of Aphrodite’s story in rela-tion mainly to the Theogony and the other Homeric Hymns, but alsoin relation to the Catalogue of Women, the Cypria and the Iliad.15

Clay thus sums up her arguments: ‘while the theogonies outlined the

11 See Rousseau (2001),Currie (2006), Burgess (2009), 56–71.Currie (forthcoming)makes excellent comments on Dem. from this perspective.

12 On intertextuality, see Pucci (1995), and critique in Burgess (2001), 133–4.13 Janko (1981), Lenz (1975). See in this volume Calame (Ch. 14).14 Faulkner (2008) and, in this volume, Introduction (p. 22).15 Clay (1989), Rudhardt (1991).

The Homeric Hymn to Aphrodite 107

genesis of the cosmos and the gods and Homeric epic recounted theexploits of the heroes, the [longer narrative] Hymns narrated criticalmoments in the evolution of the Olympian order and thus filled thegap between the two other genres of epos’.16 The love affair ofAphrodite and Anchises, just as the abduction of Persephone andthe foundation of Eleusis or the birth and establishment of Hermesand Apollo in the other major Hymns, constitutes an epoch-makingmoment in the history of Zeus’ reign. For it was to be, according toZeus’ plan, the last union between a mortal and a god.17 The Hymn’semphasis on the relation between mortals and immortals had beenelucidated previously,18 but Clay places it in the context of a global,consistent, and Panhellenic representation of cosmogonic history.Without clearly addressing the question of epic interaction, herargumentation postulates that the audience of the Hymn and ofGreek epos in general was aware of that cosmogonic backgroundand that the poem’s significance was to be constructed in relation tothat shared knowledge.

The following interpretation is an attempt to supplement Clay’swork and reflect on the relationship between the Hymn to Aphroditeand the Iliad by being more specific about the way the Hymn definesits own place in relation to theogonic and heroic traditions. I hope tocontribute to the understanding of Aphrodite by showing that it notonly refers to general traditional motifs (bathing and dressing, en-counter with a stranger, epiphany) as all epics do,19 but more speci-fically that its narrative structure as a whole links the tale ofAphrodite’s seduction of Anchises to the Achillean aristeia, andthat this dialogue with the Iliadic tradition (which, however subtle,I assume could be identified by at least some members of the audi-ence) is a key element of the Hymn’s poetics and meaning. Theexperience of the immortal goddess mirrors that of the mortal heroand is thus revealed as an exploit paradigmatic of the goddess’sworks20—an ergon both parallel to and other than the heroic one.

16 In the new preface to the reprint of Clay (1989) in 2006, vi.17 Clay (1989), 169–70. See also Brillet-Dubois (2001).18 Segal (1974), Smith (1981a).19 See Graziosi and Haubold (2005). Faulkner (2008) is a precious source for these

parallels.20 On the repetition of �æªÆ (‘works’, 7 occurrences in 21 lines), see AHS 349,

Porter (1949), 251–54, Podbielski (1971), 18–21, Càssola 249, Faulkner (2008), 72–3.


The hymn’s pars epica combines the following traditional scenes:

(1) Divine motivation: Zeus inspires Aphrodite with the love of An-chises (53–7). (2) Preparation: the goddess goes to Cyprus where theGraces bathe, anoint, and dress her (58–65). (3) Journey: she travels toMount Ida (66–9); wild beasts follow in her wake (69–72); she inspiresthem with the desire to mate (72–4). (4) Encounter: she stands beforeAnchises (75–83) who wonders at her appearance (description of theshining clothes and jewellery, 84–90). He greets her as a goddess (91–106); she pretends to be a mortal sent by the gods to wed him (107–43).(5) Intercourse: he gives in to desire, takes her to bed, and undresses her(144–66); they have intercourse and he falls asleep (167–71). (6) After-math: Aphrodite awakens Anchises and reveals herself in her true form(171–80); he reacts with fear (181–90); the goddess comforts him,predicts the birth of Aeneas, announces that she will never boastagain about her power over the gods (191–290). Having ended herspeech, she departs to the sky (291).

On a primary level, this sequence is a typical ‘seduction’ or ‘allure-ment’ narrative with known parallels in Greek epic that we willexamine shortly. Yet one can already notice that the sequence closelyfollows that of the Iliadic aristeia in which the hero (1) is inflamedwith the desire to fight, (2) arms himself, (3)moves towards the front-line, (4) meets his opponent, (5) fights him, and (6) reflects upon hisvictory.21 The main difference lies in the central action: sexual inter-course on the one hand, martial conflict on the other. There are anumber of traditional symbolic and formulaic links between thetwo.22 It is on these grounds, I will argue, that the hymn-poetprogressively develops the narrative of a love aristeia.

Aphrodite’s action begins with a preparation scene which is verysimilar to two famous Homeric passages: Hera preparing herselfbefore seducing Zeus in Iliad 14, and Aphrodite bathing after havingbeen caught in bed with Ares in Odyssey 8. The text of the Hymn waslong thought to be a combination of the Homeric lines.23 It is nowmore generally assumed that the three poems drew independently,

21 See Fenik (1968), Janko (1994).22 Monsacré (1984), 41–94.23 Aphr. 59 � Od. 8. 362; 60 � Il. 14. 169; 61–2 � Od. 8. 364–5; 63 � Il. 14. 172.

Aphr. 64–5, which concern Aphrodite’s dressing, are parallel in content but not inwording to Od. 8. 366 and to Il. 14. 178–85. AHS 357, speak of ‘the writer whocombined the passages in X and q’; cf. Lenz (1975), 118–22.


with different generic agendas, on the same traditional material.24

The hypothesis of a typical preparation scene is strengthened by twoother poems. Fragments 4 and 5 (Bernabé) of the Cypria describe theclothing and adornment of Aphrodite. Fr. 4 focuses on Aphroditedressing in the clothes made by the Horai and the Charites and placesspecific emphasis on the fragrant flowers used to dye the clothes. In fr.5, Aphrodite is singing on Mount Ida among the Nymphs andCharites; they have crowned themselves with garlands made of flow-ers. The purpose of the scene, though, is probably other than todescribe their preparation—the main verb is missing and the adorn-ment is described with a circumstantial participle. Given that fr. 5 isset on Mount Ida and that, according to Athenaeus, fr. 4 belongs inthe first book of the Cypria, it is more than likely that Aphrodite’spreparation scene precedes the Judgement of Paris.25 Hymn 6 alsopresents a comparable scene. The poem celebrates Aphrodite as thesovereign of Cyprus and recalls her arrival on the island. She was firstcarried there by the wind and the waves (3–5) and was welcomedby the Horai who dressed her and adorned her with golden jewellery(5–14). Then she was introduced to the immortals, who greeted hergracefully, and the sight of her inspired the gods with the desire towed her (15–18).

In all cases but the Odyssey, the preparation scene precedes anerotic encounter: Aphrodite exerts her charm on Anchises (Aphr.),the Olympian gods (Hy. 6), and very probably Paris (Cypria), whileHera seduces Zeus (Il. 14). It seems reasonable to think that thedressing motif, followed by journey and encounter, is part of a typicalseduction sequence, just as the arming-motif is part of a fightingsequence. In our examples, seduction might end in intercourse onlyindirectly or implicitly: in the Cypria, Aphrodite substitutes Helen forherself as Paris’ lover,26 while Hymn 6 ends before any of the gods isdesignated to become her husband.

One noticeable feature of our series is that, apart from the Iliad, allsequences involve Aphrodite. A close examination of the Hera scene

24 Porter (1949), 268–9, Janko (1994), 171. Faulkner (2008), 142 does not excludeimitatio.

25 See the beginning of Proclus’ summary. Cf. also E. Hel. 676–8 (bath of thegoddesses precedes the Judgement), IA 1294–8 (judgement takes place next tothe springs of the Nymphs where the goddesses pick flowers). On Euripides and theCypria, see Jouan (1966).

26 See Brillet-Dubois (2001), 258–9.


confirms that, among the many possible tales of seduction, thereexisted a traditionally Aphroditean one, for the Iliad includes elem-ents pointing to a consciously ironic transference of the motif.27 OnceHera has bathed, anointed, and clothed her body, her preparationappears to be still incomplete. The episode then expands to tell of hervisit to Aphrodite: as if her personal seduction were not enough, sheneeds to borrow the girdle in which the love-goddess keeps thecharms (Ł�ºŒ��æØÆ, 14. 215) which constitute her prerogatives andpower over all creatures.28 Using lies that her victim, though familiarwith beguiling persuasion, paradoxically does not detect,29 Hera talkssmiling Aphrodite (çØº��Ø�� , 14. 211) into lending her the gar-ment which is her exclusive property; it allows her to become herself atriumphantly smiling goddess (��Å��, 222; �Ø��Æ�Æ, 223) andensures the success of her erotic trickery against Zeus. It also signalsin a humorous metapoetic way to an ‘epically informed’ audience thetransference of a traditional Aphroditean theme. Therefore, the factthat the narrative seems to wander off an expected course is in itselfmeaningful and draws the attention to the substitution of Hera forAphrodite.30 That the girdle is not mentioned elsewhere is anotherproof that it is only necessary in this case of (poetic and mythological)transfer of prerogatives. Finally, given that Hera, using Aphrodite’sweapons, overcomes Zeus on Mount Ida in order to prevent a Trojanvictory, the re-use of the sequence can be interpreted as a Homericintegration of the cyclic tradition, and as an inverted replica of theJudgement of Paris.31

This illuminates the second digression of the Dios Apate episode:Hera’s visit to Sleep and her promise to give him one of the Graces towed.32 Just as Aphrodite bribed Paris by promising the love of Helen,Hera uses as bait a typicallyAphroditean character, which again points totransference.33 In both narratives, the whole seduction sequence is

27 For theoretical discussions, see Danek (1998), Burgess (2006), (2009), Currie(2006).

28 Compare Th. 201–6.29 Janko (1994), 179.30 Janko (1994), 178–9.31 See Burgess (2001), (2009), 65–6.32 See in this volume West (Ch. 2) on the connection of this episode to the myth

of Hephaestus’ binding of Hera.33 Conversely, it seems significant that Athena, Hera’s ally in war since the Judge-

ment, should be substituted for the Graces in the making of her clothing (14. 178–9).


motivated less by intercourse itself than by the victory at stake. Havingbeen defeated by Aphrodite in the krisis, Hera now ironically has hercontribute to the success of Paris’ opponents. The interaction here israther between Iliadic andCyprian traditions than between the Iliad andAphrodite.

Let us turn back to the Aphroditean seduction sequence. Varia-tions from the pattern point to meaningful poetic choices. In Hymn 6,the bath is replaced by the motif of Aphrodite floating on the sea, ‘insoft foam’ (IçæfiH ��Ø ÆºÆŒfiH, 5).34 The poet thus pointedly choosesthe Hesiodic version of Aphrodite’s birth and sets his description in atheogonic rather than an erotic perspective. The emphasis on thegoddess’s golden jewellery serves as an aetiology of her epithet‘golden-crowned’ (åæı��çÆ�� , 1)35 and manifests her divine nat-ure, while her preparation leads to her introduction into the society ofgods.36 Although the effect she produces on the Olympians is erotic(they all wish to wed her),37 this aspect of the Aphroditean scene issubordinate to the characteristically hymnic representation of heradmission among the immortals.38

Departure from the traditional sequence is also visible in Odyssey 8,where the Aphroditean preparation-scene is placed after sexual inter-course instead of before. Aphrodite’s return to Cyprus is usuallyinterpreted as a ‘natural’39 closure for the episode. After the humilia-tion she just experienced, the goddess is supposed to seek the comfortof her familiar dwelling.40 Instead of this moral interpretation,I would rather suggest that by moving the scene, which traditionallyindicates preparation for love, to the end of the episode, Demodocusprovides an ironical close to his humorous poem: far from beingtamed by humiliation, the goddess is replying by getting provokinglyready for another round.

34 Same pun on aphros/Aphrodite in Th. 191.35 Compounds of åæı�� (‘gold’) appear in lines 1, 5, 8, 9, 11, 12; mention of a

golden crown in 7–8.36 Compare Apoll. 186–8.37 The effect of Aphrodite on the male gods of Olympus might be traditional.

Compare the salacious jokes shared by the gods in Od. 8. 333–43; also Pandora’spresentation to the gods at Th. 588.

38 See Lenz (1975), 16–21.39 See OC i. 371; De Jong (2001), 209.40 Porter (1949), 267.


As for the preparation-scene of the Hymn to Aphrodite, Porterconvincingly argued that it develops traditional themes in its own wayand gives special force to the mentions of fragrance.41 From a narra-tive perspective, the specific treatment of the traditional sequence inthe Hymn consists in separating the description of clothes andjewellery from the preparation motif (58–64, 81–90). This choiceallows a very interesting shift in focalization, since it is not theprimary narrator who describes the effect that Aphrodite producesonce she has bathed and dressed. Rather, it is Anchises, her victim,who mentally surveys her when she appears:

�ªå��Å �’ ›æ�ø� Kçæ�Ç�� ŁÆ�ÆØ�� r�� ª�Ł� �� ŒÆd �¥Æ�Æ �ØªÆº��Æ.��º�� b� ªaæ �� çÆ�Ø��æ�� ıæe ÆPªB ,�rå� �’ K�Øª�Æ��a �ºØŒÆ Œ�ºıŒ� �� çÆ�Ø�� ,‹æ�Ø �’ Iç’ ±�ÆºBfi ��ØæBfi ��æØŒÆºº� q�Æ�,ŒÆº��, åæ��Ø�Ø, �Æ��ŒØº�Ø· ‰ �b ��º��Å��Ł��Ø� Iç’ ±�Æº�E�Ø� Kº��, ŁÆFÆ N��ŁÆØ.�ªå��Å �’ �æ� �xº�� (Aphr. 84–91)

Anchises gazed and took stock of her, wondering at her appearance, herstature and her gleaming garments; for she wore a dress brighter thanfirelight, and she had twisted bracelets and shining earbuds. Round hertender neck there were splendid necklaces of gold, most elaborate, andabout her tender breasts it shone like the moon, a wonder to behold.Anchises was seized by desire.42

The pattern is further modified as the Hymn reduplicates that de-scription when Anchises undresses the goddess:

Œ�� ƒ �æH�� I�e åæ�e �xº� çÆ�Ø��,��æ�Æ �� ª�Æ�� Ł! �ºØŒÆ Œ�ºıŒ� �� ŒÆd ‹æ�ı .ºF�� ƒ Ç��Å� N�b �¥Æ�Æ �ØªÆº��Æ

�Œ�ı� ŒÆd ŒÆ�ŁÅŒ�� K�d Łæ��ı Iæªıæ��º�ı

�ªå��Å . (Aphr. 162–6)

He first removed the shining adornment from her body, the pins andtwisted bracelets and earbuds and necklaces; he undid her girdle, anddivested her of her gleaming garments and laid them on a silver-rivetedchair—he, Anchises.

41 Porter (1949), 267–9.42 Translations of Aphr. are adapted from WL.


Emphasis is thus put not only on Anchises’ reaction to the sight ofAphrodite but also—for the action has no parallel in the otherAphroditean seduction-sequences—on the stripping of her body.These features, I believe, are meant to underline the relation betweenthe Hymn’s sequence and the Achillean aristeia. For if Aphrodite’sdress which is ‘brighter than firelight’ (çÆ�Ø��æ�� ıæe ÆPªB , 86) isthematically similar to Hera’s sunbright veil (Il. 14. 185), lines 86 and87 share their diction not with her preparation-scene but with lines ofIliad 18 describing the objects forged by Hephaistus. The hemistichçÆ�Ø��æ�� ıæe ÆPªB (86) qualifies Achilles’ corslet in the Iliad(18. 610), while line 87 (reduplicated at 163), is a variation of theformula at Il. 18. 400–1:

'B�Ø �Ææ’ �N�� å�ºŒ�ı�� Æ��ÆºÆ ��ºº�,��æ�Æ �� ª�Æ�� Ł’ �ºØŒÆ Œ�ºıŒ� �� ŒÆd ‹æ�ı

I [Hephaestus] spent nine years in the company of [Thetis and Eury-nome] forging many cunning objects, pins and twisted bracelets andearbuds and necklaces.

When the formula first occurs in Aphrodite at line 87, the replace-ment of ‹æ�ı at the end of the verse by the adjective çÆ�Ø�� (‘shining’) qualifying Œ�ºıŒÆ (‘earrings’) allows the Hymn poetboth an insistence on the metallic glow and an extended descriptionof the ‹æ�Ø (‘bracelets’); of their richness, variety, and moonlikebrilliance, which recall the flare of Achilles’ divine weapons in hisarming-scene.43 This emphasis on Aphrodite’s glowing finery resultsfrom a choice comparable to that of Hymn 6, but very different fromthat of frs. 4 and 5 of the Cypria, where the use of natural flowers iswhat characterizes her beauty. Among the possible traditional repre-sentations of the goddess, the description in Aphrodite selects the onethat allows the poem to take on a heroic flavour and through dictionturns the jewellery into a set of weapons comparable to Achilles’armour. From this perspective, it is striking that when shining Aph-rodite appears before Anchises, she is not only intent on seducinghim but anxious not to frighten him (� Ø� �Ææ��Ø��, 83). Fright, aswe know, is the reaction of the Trojan soldiers, then of Priam,Hecuba, and Hector, to the sinister reappearance of Achilles sur-rounded by the light of his bright new armour.44 Both poems make

43 Il. 19. 369–83. See Armstrong (1958), 351–2.44 Il. 20. 44–6; 22. 35–91, 131–6.


it clear that it is the perception of divine power symbolized by anunnatural aura that causes fear, whether it reveals Achilles’ outstand-ing yet temporary superiority or actual divinity: Aphrodite softens theshock she might cause to Anchises by taking the stature and appear-ance of a young girl,45 but as soon as she reveals herself as a goddess,he is terrified (��æ�Å��, 182). There is more to it, I think, than thetraditional expression of the awe that seizes men when confrontedwith the divine: the negative formulation of Aphrodite’s precaution,combined with the Achillean description of her adornment, are dis-creet signals that the narrative model transfered here is the heroicaristeia.46

As for the undressing scene, I said earlier that it had no parallel inthe Aphroditean seduction sequences. Yet it does have two masculinecounterparts. The first one belongs to a masculine seduction scene: atOdyssey 11. 245 Poseidon loosens the girdle of Tyro, a mortal maiden(ºF�� b �ÆæŁ��Å� Ç��Å�). The story shares other elements with thetale of Anchises and Aphrodite: it recounts the union of a god and amortal, and the prediction that Poseidon makes to Tyro after inter-course, promising a splendid offspring, is clearly related to the oneAphrodite makes to her lover. The undressing motif, therefore, mightvery well be part of traditional (not exclusively Aphroditean) seduc-tion scenes. Nevertheless, the only known parallel to Aphrodite’selaborate narrative structure, in which Aphrodite’s dressing scene iscounterbalanced by her undressing and the loosening of her belt(ºF�� ƒ Ç��Å�, 164), features in Iliad 16 where Patroclus’ armingscene (16. 130–44) is pointedly echoed by a disarming scene (16. 793–804). As Apollo strikes him, his helmet rolls to the ground, hisspear breaks in his hands, his shield falls from his shoulders; finally,the god loosens his corslet (ºF�� ƒ Ł�æÅŒÆ, 804).47 This paralleldefines the erotic encounter in the Hymn to Aphrodite in sharpcontrast to single combat: what follows for Patroclus is ultimatedefeat and death, while the stripping of Aphrodite suggests herutter victory as a seducer and reveals the true nature of her sexualpower. Whereas Iliadic heroes strive to overcome their enemies and

45 See Faulkner (2008), 164–5, on this half-successful transformation.46 The episode of Hera’s seduction in the Iliad is also an aristeia, but the use of the

heroic model turns her action into a feminine act of war, whereas I intend to showthat in Aphr. erotic power is praised instead of heroic virtue.

47 The only other occurrence of this formula is at Il. 4. 215 (Machaon loosens thecorslet of an injured Menelaus).


to lay a hand on their spoils, her own feminine, paradoxical heroismaims at intercourse and lies in her ability to induce her masculinevictim to strip her.48

Focusing on the way the Hymn to Aphrodite disrupts the trad-itional order of the seduction sequence has led us to look ahead in thecourse of the narrative and we must take a few steps back. Onceprepared, Aphrodite moves to Mount Ida which is to be her battle-field. There she is first joined by ‘grey wolves and fierce-eyed lions,bears and swift leopards insatiable for deer’ (��ºØ�� º�Œ�Ø åÆæ�� º�� j ¼æŒ��Ø �Ææ��ºØ �� Ł�Æd �æ�Œ��ø� IŒ�æÅ��Ø, 70–1);upon seeing them, she rejoices and casts desire in their breast(72–3). It has been suggested that the motif of the wild animalsfollowing in Aphrodite’s wake might have Asiatic connotations ororigin, while other scholars prefer to underline that the episode illus-trates the goddess’s power as it has been earlier defined (Aphrodite issaid to reign over all creatures including animals, 4–5).49 Neitherexplanation is incompatible with the fact that it also occurs at thepoint where, in the structure of a heroic aristeia, the audience couldexpect a catalogue of minor victims,50 and if the contrast between theferocious nature of the beasts and their sexual activity enhances theoverwhelming strength of desire, these predators also evoke in ahumorous way the heroes who are repeatedly compared with themin Iliadic similes.

Aphrodite then moves on to her great duel with a most challengingopponent, one who matches her excellence, ‘the hero Anchises whohas his beauty from the gods’ (�ªå��Å� læøÆ Ł�H� ¼�� Œ�ºº� �å��Æ,77). The Trojan prince is not only a man whose heroic virtue man-ifests itself in great beauty, like Achilles or Hector: his beauty iswhat makes him a hero in this erotic encounter. His bucolic environ-ment, suggestive of a prosperity based on breeding, and his activity,playing the cithara (80), liken him to Paris and similarly characterizehim as belonging to the Aphroditean world.51 In order not to causehim fear, as we have seen, Aphrodite takes the appearance of an

48 See Schein (forthcoming). Sowa (1984), 47, 74–6, 79–81, sees in the removal ofAphrodite’s clothes the symbol of the fertility-goddess’s death in her ritualmarriage toa mortal. Although suggestive, the comparative perspective of Sowa’s work and itsdefinition of tradition seem to me too broad to be utterly convincing.

49 See Faulkner (2008), 152–3.50 See Fenik (1968), Krischer (1971), 13–89.51 Il. 3. 54. See Faulkner (2008), 160.


inexperienced maiden (�ÆæŁ�� I�� , 82). The disguise is appro-priate in all respects: it allows the goddess to develop the lie about herbeing a mortal sent to Anchises to wed him; it puts her in an inferiorposition in front of Anchises and motivates her apparent reluctanceto have intercourse with him—a very strong incentive indeed for maledesire;52 there also seems to be a traditional association betweenyoung girls and sweet conversation (OÆæ� or OÆæØ�� ),53 which ispart of Aphrodite’s timai along with beguiling persuasion. Thus theform and content of the trickery are paradigmatic of the goddess’sprerogatives as well as the love (�æ� , 91, 144) and lust (¥�æ� , 143)they produce. But they also reverse the traditional formulae compar-ing a hero to Ares,54 a god whose ergon is markedly opposed toAphrodite’s in the Hymn (10). Here a goddess disguises herself as amortal, there epic diction confers a divine persona upon mortals.

The subsequent dialogue is first of all typical of allurement scenesin the epic tradition55 and in itself a splendid illustration of the verbalaspects of Aphrodite’s erga. Yet it is also strongly linked to the part ofan aristeia where heroes, before engaging in fight, question each otherabout their identity and make genealogical claims. ‘Who are youamong mortals?’ ask Diomedes to Glaucus and Achilles to Astero-paios in the Iliad (6. 123 and 21. 150). Diomedes’ more elaboratespeech in Iliad 6 anticipates the possibility that Glaucus might be animmortal, in which case he would renounce the duel in fear of divineanger (6. 128–9). But, says he, if his opponent is a mortal, then theymust proceed to fight. Similarly, Anchises, delaying the satisfaction ofthe eros that has seized him, first asks ‘Who are you among god-desses?’, saluting Aphrodite in a hymnic way (92–9). As will becomeapparent at the time of Aphrodite’s epiphany, he is aware, just asDiomedes, of what dangers threaten the mortal who interacts with animmortal. But when the goddess has convinced him that she is amortal, he immediately proceeds to take her to bed (145–50). Scholarsdebate the sincerity of Anchises’ address:56 has he recognized a

52 Compare Hera’s feigned reluctance at Il. 14. 329–40.53 See Il. 14. 216; Th. 205. Also Il. 22. 127–8.54 The most common one is Ł�fiH I��ºÆ�� @æÅœ, ‘equal to swift Ares’ (eleven

occurrences in the Il.); also �æ��º�ØªfiH r�� @æÅœ, ‘equal to man-destroying Ares’ (sixoccurrences).

55 See Forsyth (1979), Sowa (1984). A. Faulkner kindly let me read his forthcomingstudy ‘Les rencontres érotiques dans l’épopée grecque! (forthcoming, d).

56 See Faulkner (2008), 173–4.


deity in disguise and is he acting as a pious mortal, or is he trying toflatter the girl by pretending to think she is a goddess, thus displayingan ability to deceive that matches hers? I would emphasize that thisambiguity is a fundamental element of the scene’s humour. Theintroductory line (�ªå��Å� �! �æ� �xº��, �� Ø� I�� Åh�Æ,‘love seized Anchises, and he replied to her’, 91) inscribes Anchises’speech in his erotic strategy but leaves the nature of his wordsundefined, as opposed to the line introducing his supplication afterthe goddess’s epiphany (ŒÆ� Ø� ºØ�� Æ ��æ��Æ �æ��Å��Æ,‘and in prayer he said these winged words to her’, 184). The parallel tothe dialogues preceding heroic fights might support the second read-ing. In these exchanges, the fighters boast about their superioritywhile challenging the worth of their opponent. If the heroic modelapplies here, it is reversed to fit the ‘erotic conquest of the stronger bythe weaker’57 and the lovers must not only appear as inferior butmake a claim about it in order to impress their adversary. Anchisesassumes a submissive position, praising the girl he hopes to seduce asif she were divine, promising offerings, praying to her. She in turnreplies with a flattering tale that puts her back in his power, substitut-ing for the man’s reverence towards the goddess an attitude ofmaidenly shyness towards a male. She subtly answers at least partof his prayer by suggesting that, if he weds her, he will gain prestigeand father superb children (127). Then, mirroring the dynamics of hisspeech, Aphrodite voices a supplication (131–8) and promises adowry (140).

In the light of this battle of persuasion and of the poem’s allusiveart, I think it is not too far-fetched to interpret as metapoetic the wayin which Aphrodite, after defining her identity, makes the strangeclaim that she can speak Anchises’ language as well as her own. Shestarts with details on her genealogy and Phrygian origin which canevoke other narratives than the Hymn’s. The adjective O��Œºı�� (‘famous’, 111) applied to Otreus, her alleged father, combined withthe remark that follows the mention of his name, �Y ��ı ±Œ��Ø (‘ifyou ever heard of him’, 111), point to his kleos and to the (epic) wordsthat are conveying it. They are addressed both to the internal andexternal audience.58 It is possible, though unprovable, that the epithet

57 Clay (1989) and (2006), 173.58 Nagy (1979), 265–75, commenting on the speeches of Achilles and Aeneas

before their duel, interprets in this metapoetic way the mention by Aeneas of the


would have sounded ironic to the the Hymn’s listeners: Otreus ismentioned only once in passing in what is left of Greek hexameterpoetry.59 Aphrodite then gives Anchises this puzzling reassurancethat she knows his language, which is, as the text explains immedi-ately, the Trojan language. Commentators are uncomfortable withthis unique allusion to bilingualism: they relate it to the mention ofdifferent languages in the Iliadic catalogue of the Trojans’ allies,60 butdo not convincingly explain why the two characters should notconventionally speak the same language, as in the Iliad the Trojansand Lycians or the Phrygian Asios and his nephew Hector do.61

I think Aphrodite’s remark can receive simultaneously differentinterpretations. On the internal narratological level, it is not moti-vated by any reaction of Anchises concerning her language, yet aspart of the seduction speech, it could be aimed at emphasizing thatthe Phrygian stranger makes a perfect match for the Trojan prince.62

It might also signify that Aphrodite is capable of adopting the samekind of flattering and seductive speech as Anchises. Her utterancecould be transposing the heroic claim made by Aeneas and Hector inthe Iliad that they share the language of blame with Achilles: whenreplying to Achilles’ provocative boasts, both answer him that theyare not to be frightened as if they were children, for they ‘very wellknow, just as he does, how to speak mocking and unseemly words’ (K��d��çÆ �r�Æ ŒÆd ÆP�e j Mb� Œ�æ��Æ M�! ÆY�ıºÆ ıŁ��Æ�ŁÆØ, Il. 20.200–2, 20. 431–3). The diction is close but not identical: ��çÆ �r�Æoccurs at the end of line 113 of Aphrodite, while the claim of theheroes has the hemistich K��d ��çÆ �r�Æ ŒÆd ÆP�e . Finally, on ametapoetic level, the singer of the Hymn is signalling that he knowsvery well of other songs than his, songs about Troy of which the Iliadis an example, and that he has learned to master them. Whileremaining (at least during the performance of the Hymn) an Aphro-ditean poet, he is pointedly borrowing the diction and patterns of theheroic epos.

�æ�Œºı�Æ . . . ��Æ (‘well-known words or songs’, 20. 203–9) that have brought hishalf-divine birth to the knowledge of his opponent.

59 Il. 3. 186.60 See Faulkner (2008), 190.61 Il. 16. 717–18.62 Faulkner (2008), 190, also links Aphrodite’s deceptive use of a foreign language

to Odysseus’ Cretan lies.


Heroes, after delighting in threatening taunts and noble claims, inthe end reject speech impatiently to engage in combat with their enemy;they feel that words, however stimulating, are impeding action.63

Persuasion, on the contrary, is an essential part of the erotic confronta-tion. Just as the sight of Aphrodite was a first love-blow, her beguilingwords hit and wound (143), and Anchises acknowledges that the urgehe feels to make love with Aphrodite derives from her words (145–54).

This brings us back to the undressing scene. It is, at first sight, as wehave said earlier, the prelude to Aphrodite’s sexual victory. Yet thecontrast with Patroclus’ defeat is about to disappear as her successreveals itself to be a delusion planned by Zeus. This reversal isprepared for in several ways. The poet first lingers on the descriptionof Anchises’ bed, which is covered with the hides of the bears andlions he has hunted (157–60). Faulkner synthetizes the multipledegrees of significance of those details:64 they symbolize Anchises’manliness, balancing Aphrodite’s feminine attributes and beauty; bycontrasting the effect of Aphrodite’s presence among the wild beasts,they ‘underline [her] own loss of power in the face of Anchises’momentary strength’. Her set of clothes and jewels become anotherhide, turning her into another hunted animal just about to be killed. Itshould be added that the description brings Anchises closer to theworld of heroes than when Aphrodite first met him playing thecithara, for hunting is not only a peace-time occupation appropriateto warriors but also the principal activity to which fighting andslaying are compared in heroic similes.

As Aphrodite lets the mortal undress her and lead her to bed, thereference to the Iliadic aristeia takes on a more sinister ring, as ifAphrodite could not be pictured as a hero without having, ultimately,to face the reality that lies at the core of heroism: death itself.65The imageof a fatallywoundedwarrior is thus being subtly superimposed on that ofthe goddess, and her intercourse with Anchises becomes an even moreintimate connection to mortality as the poet transfers to her a dictionnormally appropriate to a dying man. Goddess and hero lay side by side‘by the will of gods’ (Ł�H� N��Å�Ø, 166), a formula used once in the Iliad,precisely of the death of Patroclus (Il. 19. 9). At this crucial moment,Aphrodite not only sexually but poetically mingles with a mortal.

63 See e.g. 20. 244–55, 20. 367–70, 22. 126–8, 22. 282.64 Faulkner (2008), 227. See also Schein (forthcoming).65 See Schein (1984).


Sexual intercourse is not deathly and goddesses do not die, thoughin a way Aphrodite’s mortal persona does as she recovers her divineappearance.66 At the same time, the pastoral Aphroditean atmos-phere is restored: mention of the flocks and of their flowery pastures(168–9) serves to dissipate the gloomy shadow cast on the scene byAnchises’ hides and hunting skills. The goddess also regains hersuperiority over Anchises. Her unmistakable divinity causes himto tremble and recoil: now he is the one shyly averting his gaze(182–3).67 Yet the epiphany does not immediately disentangle Aph-rodite from mortality or from heroic poetics. Before she departs, thetwo lovers engage in a dialogue that, despite significant inversions,parallels the last exchange between a victor and his expiring victim.Instead of voicing triumphant taunts as heroes do when their oppo-nent falls to the ground, revealing his vulnerability,68 Anchises, at theepiphany of the immortal, assumes again (this time without anyambiguity) the position of a supplicant. Whereas heroic victorsboast (�hå�ÆØ, K��å�ÆØ) at the sight of their defeated enemy,Anchises takes no responsibility and places it instead on Aphroditewho deceived him (186). He does not rejoice at having eroticallyovercome a goddess, for he knows it to be a dangerous situation: aman might consequently suffer the loss of his menos and be con-demned to renounce a flourishing life (188–90). Following Smith,69 Irely on internal narrative and language to interpret Anchises’ fears asrelated to his wish for a ŁÆº�æe� ª�� (‘a flourishing offspring’, 104)and thus primarily expressing the idea that he might lose sexualpotency and reproductive power. At least this is how Aphroditeunderstands menos: when she invites her lover not to be afraid, shesoothes his worries by announcing that he will have a son (193–6).Later on, she calls this son a Ł�º� (‘offspring’, 278). That in Homermenos has a less specific definition and designates the energy of ahero’s body during fight, or that deprivation of menos indicates deathis but another signal that Aphroditean features are being markedlydefined in relation to and contrast with the heroic epic.70

66 On the epiphany, see Podbielski (1971), 59–64, Lenz (1975), Richardson (1974),208–10.

67 Cf. 156, 182.68 For Hector and Achilles, see Il. 16. 829, 22. 330, but this kind of boasting appears

constantly in the battle-scenes.69 Smith (1981a), 66.70 See Giacomelli (1980), Faulkner (2008), 248–51.


Aphrodite replies to Anchises with a long speech. Clay thus sumsup its triple function: ‘Addressed to Anchises, the goddess’ wordssimultaneously comfort and warn her mortal lover. At the same time,she gives voice to reflections concerning the relations between thehuman and the divine that spring from her forced contact withmortality and her recognition of the complicity of Zeus. Finally,while she refers prophetically to the future, of whose ultimate con-tours she is only dimly aware, she invites our participation andunderstanding of its evolution and necessity.’71 Consideration ofAphrodite’s discourse in the light of heroic epic will lead me toreassess or reinforce Clay’s reading in several ways. Though pro-nounced with the undisguised authority of the goddess, the speechstill belongs to the aristeia, at least to the most complete and elaboratepattern of the Achillean aristeia,72 for it mirrors Patroclus’ andHector’s last words.

Both heroes, as they meet their death, experience a moment of re-and pre-cognition. They suddenly find themselves capable of recog-nizing the gods who brought about their defeat (Il. 16. 844–7, 22.296–303). As for Aphrodite, she does not explicitly acknowledgeZeus’ role in her own misfortune, though she reveals that she isaware of it by assuring Anchises that the gods love him and byalluding to Zeus’ central function in the stories of Ganymedes andTithonus. Lenz has already suggested that, in the hymnic structure ofthe poem, Aphrodite’s speech replaced an epilogue in a divine settingconcluding the plan of Zeus.73 We can now add another structuralargument to his: in the course of Aphrodite’s aristeia, this would bethe moment for the goddess to name Zeus as the agent of her defeat.Her reluctance to do so is a sign of her great shame (ª! Z��Ø�� , 247),of her ‘unutterable’ delusion (�PŒ O��Æ��, 254), but I see it also, ifcompared to her characterization in Iliad 5, where she is paternallyrebuked by Zeus, as participating in the poem’s humour: Aphrodite issomething of a capricious child forced to submit to her father’sauthority but refusing to admit it.

The second important feature of Patroclus’ and Hector’s finalwords is pre-cognition, which takes the form of a warning addressed

71 Clay (1989), 183–4.72 On this aspect of the Achillean aristeia, see Janko (1994), 417–21, Richardson

(1993), 139–44.73 Lenz (1975), 24–5.


to their victorious opponent: death of the triumphant hero willinevitably and soon follow his victim’s. Patroclus foresees Achilles’revenge over Hector (16. 852–4), while Hector predicts that Achilleswill in turn provoke the wrath of the gods and be killed by Paris andApollo (22. 358–60). Now, it is a remarkable lacuna in the goddess’sspeech that she does not explain why she rejects the possibility ofimmortalizing Anchises.74 After all, Zeus has turned his own loverGanymedes into an Olympian, and Aphrodite has no reason toreproduce the mistake of Eos. Instead, she voices regrets:

Iºº! �N b� ��Ø�F�� Kg� �r�� Æ ��

Ç��Ø ��æ� �� Ø Œ�ŒºÅ�� YÅ ,�PŒ i� ��Ø�� ! ¼å� �ıŒØ�a çæ�Æ IçØŒÆº��Ø.�F� � �� b� ��åÆ ªBæÆ ›��Ø�� IçØŒÆº�ł�Ø

�Åº�Ø , �� ! ��Ø�Æ �Ææ��Æ�ÆØ I�Łæ��Ø�Ø�,�Pº�� ŒÆÆ�Åæ��, ‹ �� ıª�ı�Ø Ł�� æ. (Aphr. 241–6)

If you could go on living as you are now in appearance and build, and beknown as my husband, sorrow would not then enfold my subtle mind.But as it is, you will soon be enfolded by hostile, merciless old age, whichattends men in the time to come, accursed, wearisome, abhorred by thegods.

The optatives followed by �F� � (‘but as it is’) express an unfulfilledcondition, but there is nothing explicit to justify why Aphrodite con-siders that condition impossible. Clay postulates ‘on the ground ofboth dramatic and narrative logic’ that ‘Aphrodite knows that herrequest for Anchises’ immortality would meet with scornful rejectionon the part ofZeus, who intended from the first to teach her a lesson’.75

The structural argument derived from the aristeia pattern againsupports her view, for the defeated goddess’s prediction revealsAnchises’ humanity—symbolized, as is shown in the generalizingrelative clause of 245, by ageing—to be the ineluctable consequenceof his victory.

That said, this passage of Aphrodite, however closely framed inheroic diction,76 manages a remarkable displacement of focus. Theerotic duel is a matter of life, not death. Aphrodite is an immortal andcannot be fatally wounded. Anchises will suffer in old age but does

74 See discussion by Clay (1989), 189–91.75 Clay (1989), 190.76 See Faulkner (2008), 278, and his remark on line 243.


not risk death, unless—and Aphrodite is explicit about it—he shouldassume the posture of an actual warrior: should Anchises boast abouthis adventure (K��fi Å ŒÆd K��ÆØ, 286), she predicts divine wrathand death at the hands of Zeus, uttering a truly heroic warning, verysimilar to Hector’s.77 But Anchises has not claimed victory and thehymnic narrative has taken another path.

Aphrodite’s grief (¼å� ) can now be safely interpreted as theinevitable outcome of her own boasting. Let us turn back to thejustification of Zeus’ plan given earlier in the Hymn:

ZçæÆ ��åØ��Æ

Å�! ÆP�c �æ��Å �P�B I��æª�Å �YÅ,ŒÆ� ��! K��ı�Æ�Å �Y�fi Å ��a �A�Ø Ł��E�Ø�

��f ª�º�Ø��Æ�Æ, çØº��Ø�c �çæ��Å,u ÞÆ Ł��f �ı��Ø�� ŒÆ�ÆŁ�Å�Bfi �Ø ªı�ÆØ��,ŒÆ� �� ŒÆ�ÆŁ�Å��f ıƒ�E �Œ�� IŁÆ��Ø�Ø�,u �� Ł�a I�Ø�� ŒÆ�ÆŁ�Å��E I�Łæ��Ø . (Aphr. 46–52)

He wanted to bring it about as soon as possible that not even she was setapart from a mortal bed, to boast among the assembled gods with amerry laugh how she had coupled gods with mortal women, and theyhad borne mortal sons to immortal fathers, and how she had coupledgoddesses with mortal men.

The association of the boasting (quest for individual superiority)78

and of the ‘hitter hit’ motif (experience of a shared vulnerability) is atthe heart of the heroic experience.79 It is in this constant reversal fromgod-like victor to expiring victim that the hero—Achilles more thanany other—learns about the limitations of his nature, but also dis-covers an individual way to transcend them: the conquest of kleos inbeautiful death (Vernant’s belle mort).80 Similarly, Zeus forces Aph-rodite to acknowledge the limitations of divine nature by having hershare the dire grief she imposed on other deities.81 Excessive prox-imity with mortals will cause her to suffer anxieties that will alter herdivine felicity, for she will have to face the inevitable deaths of herlover and son, a situation that the Iliad presents as a threat to cosmic

77 Compare Aphr. 286–90 and Il. 22. 358.78 See Muellner (1976).79 See Schein (1984), 128–67, esp. 148–9.80 See Vernant (1992).81 See Schein (forthcoming).


stability.82 The lesson she is being taught is thus that the superiorityshe exerted on the other gods was temporary and partly illusory; heruniversal power did not exclude her from the community of theimmortals nor allow her to escape the law of cosmic order. As heraristeia ends with a speech assimilating her to expiring heroes, finallytamed by a fate that makes all men equal, she appears to be defini-tively forced to respect the boundaries separating the divine from thehuman. After the closure of the Hymn’s narrative, she will never againengage in coupling immortals and mortals, just as the dead Hectorwill no longer engage in fighting. Thus we come, though from anotherperspective, to the same conclusion as Clay: Zeus’ purpose is cosmo-gonic and aims at preventing unseemly erotic contacts between godsand humans.83 I see no other way to understand these much disputedlines from Aphrodite’s long speech to Anchises:

ÆP�aæ K�d ª! Z��Ø�� K� IŁÆ��Ø�Ø Ł��E�Ø�

��ÆØ XÆ�Æ ��Æ �ØÆ��æb �¥��ŒÆ ��E�,�Q �æd� K�f O�æ�ı ŒÆd ��ØÆ , Æx �� Æ IŁÆ��ı �ı��Ø�Æ ŒÆ�ÆŁ�Å�Bfi �Ø ªı�ÆØ��,��æ��Œ��: ��Æ ªaæ Ke� ��Æ�Œ� ��ÅÆ.�F� �b �c �PŒ�Ø �Ø ��Æ å��ÆØ K��B�ÆØ

��F�� ! IŁÆ��Ø�Ø� (Aphr. 247–53)

I shall suffer great reproach among the gods evermore on your account.Formerly they used to be afraid of my whisperings and wiles, withwhich at one time or another I have coupled all the immortals withmortal women, for my will would overcome them all. But now mymouth will no longer open wide enough to mention this among theimmortals.

Again, the parallel with Hector can shed light on the implications ofthe fear once provoked by Aphrodite. When the Trojan hero expires,the Achaeans, once terrified by his power, now gather to hit and mockhis dead body, in a shocking but pathetic gesture betraying the extentof their relief (Il. 22. 369–75). Until this point, Hector represented forthem the greatest threat of all; but the time when he set fire to theships is over.84 Similarly the immortals will stop dreading Aphrodite’s

82 See Brillet-Dubois (2001), 254–5.83 See discussion in Faulkner (2008), 10–18, who reads the passage differently than

Clay.84 Compare the fear of the Achaeans when Hector sets fire to their ships at Il. 15.

623–8.


power because she will not be a threat any more. And she will neverclaim victory again over the gods because her will shall never againovercome them all, as she claims happened in the past (251). How-ever difficult the text of line 252,85 I take K��B�ÆØ (‘to mention’ or‘clearly name’) to refer to the boasts in which Aphrodite would nameherself as being responsible for the coupling of gods and mortals, justas Å�! O��ÆØ�� (‘do not mention or name’, 290) refers to the boastAnchises could make about seducing Aphrodite. I also take ��F��(‘this’, 253) to refer to the whole sentence ��Æ ªaæ Ke� ��Æ�Œ��ÅÆ (‘for my will would overcome them all’, 251). Fear of thevictims and boasting of the victor both cease as Aphrodite experi-ences ultimate defeat at the hands of Zeus. The birth of Aeneasguarantees that she will never forget the grief that it caused her:

�fiH �b ŒÆd `N��Æ Z��! ��ÆØ, �o��ŒÆ ! ÆN�e��å�� ¼å� , ��ŒÆ �æ��F I�æ� �� P�Bfi (Aphr. 198–9)

His name shall be Aeneas, for an ainon (terrible) sorrow took me,because I fell into a mortal’s bed.

In these lines, the enjambment of 199 and the unusual separation ofthe formula ÆN�e� ¼å� 86 obviously put heavy emphasis on the punbetween `N��Æ and ÆN�� (‘terrible’) at line-end, but also on ¼å� (‘grief ’) which is moved to the following line and stands at thecaesura. Had Aphrodite expressed the causal relation in the reverseorder, starting with the acknowledgement of her feeling to explainsubsequently her son’s name, we would have expected her to come upwith a name beginning with akh—but Akhilles was already taken.Aeneas is thus introduced as another Achilles (his divine motherThetis is a paradigm of Aphrodite’s grieving victims),87 but also asother than Achilles.

For, that Aphrodite shall from now on submit to Zeus and refrainfrom uniting beings who should not be matched (mortals and im-mortals) is only part of the lesson;88 this does not mean she has to

85 The manuscripts read ��Æå��ÆØ which makes no sense. See discussionof the conjectures in Faulkner (2008), 282–3. I follow West (2003) in readingå��ÆØ (‘open wide’).

86 The only two displacements of the formula in the Iliad occur in speeches utteredby Achilles: Il. 16. 55, 19. 307.

87 See Il. 18. 50 ff. On Aphrodite and Thetis, see Schein (forthcoming).88 Contrast the way Hera’s greatness and wisdom match Zeus’ in lines 36–44.


renounce superiority altogether. An aristeia is first and foremost alanguage of praise, which here meets and serves the hymnic celebra-tion of the goddess. Before she finds herself tamed by Anchises andZeus, Aphrodite demonstrates her erotic excellence. And even as shefalls, Zeus’ plan also reveals the proper way for her to exert her uniquepower, just as the heroes’ individual glory balances their inevitabledeath. While she can no longer bring mortals close to immortals insexual intercourse, she can still liken them to the gods by grantingthem a specifically Aphroditean compensation for their mortality:virtually eternal reproduction.89 She is now bearing one of the IªºÆa�Œ�Æ (‘splendid children’) that she promised to Anchises when shewas in disguise (127). Aeneas and his flourishing descendants will bethe eternal reminders not only of Aphrodite’s delusion, but also of hererotic accomplishments:

��d �’ ��ÆØ ç�º� ıƒe , n K� 'æ��Ø� I��ØŒÆd �ÆE�� Æ��Ø �ØÆ��æb KŒª�ª��ÆØ (Aphr. 196–7)

You are to have a dear son who will rule among the Trojans, andchildren to his children always shall be born.

This prophecy has long been at the heart of the discussion of therelationship between the Iliad and Aphrodite, for it echoes the pre-diction made by Poseidon in Iliad 20:

�F� �b �c `N��Æ� ��Å 'æ��Ø� I��Ø

ŒÆd �Æ��ø� �ÆE�� , �� Œ�� Ø�Ł� ª�ø��ÆØ. (Il. 20. 307–8)

The mighty Aeneas shall rule among the Trojans, he and his children’schildren that shall be born hereafter.

I will limit the scope of my comparison to the perspective of compet-ing Aphroditean and Iliadic poetics. Significant differences betweenthe two prophecies point to two contrasting uses of a commontradition concerning Aeneas’ offspring. The lines of the Hymn con-tain a hapax, the verbal form KŒª�ª��ÆØ, which, following P. Chan-traine,90 I consider as a future based on a perfective stem. Because thisis ‘post-Homeric coinage’, Hoekstra, though generally reluctant topostulate borrowings from one epic to another, exceptionallyassumes that the Hymn is here both imitating and modifying the

89 See Smith (1981a), 47–8.90 Chantraine (1935).


Iliad.91 Lamberterie, who suspects the form because of its uniqueness,suggests that the problematic final -��ÆØ of the form was influencedby the Iliadic ª�ø��ÆØ and points to an erroneous quotation.92 I agreethat the Hymn’s version of the prophecy implies the existence of theIliadic diction, but instead of being tempted, as Lamberterie or West,to correct the text of the manuscripts and replace KŒª�ª��ÆØ by aparticiple (KŒª�ª�� or the more fomulaic KŒª�ªÆH�� ), so as tosubordinate the mention of the offspring to the announcement ofAeneas’ rule, I would suggest that the version in Aphrodite is aconscious assessment of erotic values in reference to and against theIliadic definition of heroism. The strange verbal form chosen (maybeeven invented) by the hymn poet is meant to emphasize the differ-ences between the two poetical perspectives. These differences lie inthe syntactical and thus thematic priority that the Iliad gives to thereigning status of Aeneas and his offspring and that the Hymn toAphrodite, adjusting to its addressee, symmetrically gives to the ever-lasting fertility of the lineage.93 Aphrodite’s prophecy answers thewish voiced earlier by Anchises to the divine-looking girl who materi-alized before his eyes:

�� a 'æ��Ø� IæØ�æ��’ ��ÆØ ¼��æÆ,��Ø �’ KØ��ø ŁÆº�æe� ª��, ÆP�aæ �’ ÆP�e��Åæe� KV Ç��Ø� ŒÆd ›æA� ç�� M�º��Ø�

Zº�Ø�� K� ºÆ�E ŒÆd ª�æÆ� �P�e� ƒŒ�ŁÆØ. (Aphr. 103–6)

Grant that I may become a man outstanding among the Trojans, andgive me flourishing offspring for the time to come. Grant me myself tolive long and well, seeing the light of the sun, and enjoying good fortuneamong the people, and to reach the doorstep of old age.

Anchises’ words also echo a famous Iliadic passage, Hector’s prayerfor his son (6. 476–81), in which the Trojan hero asks the gods thatAstyanax be ‘outstanding among the Trojans’ (IæØ�æ��Æ 'æ��Ø�,477) and that he may prove himself as excellent a warrior and king ashis father. In comparison to the Iliadic passages, the same shift offocus appears in the mortal’s wish as in Aphrodite’s prophecy. What

91 Hoekstra (1969), 39–40.92 In a conference given at the Université Lumière-Lyon 2 in 2004.93 See Lenz (1975), 114–15: he explains the difference by the fact that Iliad 20 does

not refer to Aphrodite, whereas the hymnic praise of the goddess logically brings thetopic of her fertility. Podbielski (1971), 74, suspected without elaborating that thehymn-poet was adopting ‘une attitude polémique envers l’épopée héroïque’.


Anchises asks for in order to ensure for himself an outstanding(IæØ�æ�� , 103) position in Trojan society is not royal status, norstrength, nor bloody victory as Hector does, but the privilege of aflourishing offspring. The parallel is completed at lines 278–9: Aph-rodite says that when Anchises first sees his son, ‘[he] will rejoiceas [he] look[s] on him, for he will be quite godlike’ (�e� b� K�c��c �æH�� Y�fi Å Ł�º� OçŁÆº�E�Ø�, j ªÅŁ��Ø ›æ�ø�· �ºÆ ªaæŁ��Œ�º� ��ÆØ); that is to say that Aeneas will be as handsome ashis father, who ‘had his beauty from the gods’ (Ł�H� ¼�� Œ�ºº� �å��Æ,77). In other words, what Anchises hopes to be granted, what he andhis son do obtain, is erotic excellence. Beauty, seduction, fertility arethe qualities that define Aphroditean heroism and constitute not onlythe goddess’s timē, but also her lover’s and son’s. This is how theHymnachieves its programme, stated in the first line of the poem, of cele-brating the ‘deeds’ of golden Aphrodite (�æªÆ ��ºıåæ��ı �çæ��Å ).It might be significant here that the Hymn to Aphrodite does not endwith a prayer as otherHymns do. The one that Anchises expressed hasbeen successfully answered and the whole poem, in a way, invitesAphrodite to repeat her gift by exerting her power over all humans.

As the goddess returns to Olympus (291) and leaves Anchises tohis human world, the poem abruptly abandons heroic diction toregain its hymnic form with the characteristic åÆEæ� Ł�� (‘farewell,goddess’, 292).94 Aphrodite is saluted as the benevolent but distantgoddess she will now forever be.

In order to clarify the relationship between Aphrodite and the Iliadictradition, I will now turn quickly to the Iliad and indicate a fewexamples of the way the heroic poem, in turn, integrates elementspresent in Aphrodite to define by contrast its own thematic andpoetics. At Iliad 5, Aphrodite in person is expelled from the battlefieldby Diomedes. The primary narrator, the hero, and Zeus all voice theidea that she does not belong there,95 for hers are not the works of warbut the works of marriage. Following the same pattern as in the Hymnto Aphrodite’s priamel (7–33), her prerogatives are opposed to Athe-na’s.96 As for the Aphroditean heroes, Anchises, Paris, and Aeneas,they serve as foils to the truly heroicDiomedes, Menelaus, and Achilles.

94 On this feature of the Hymns, cf. in this volume Calame (pp. 354–5).95 Il. 5. 331–3, 5. 348–51, 5. 428–30.96 Il. 5. 428–9. Compare Aphr. 9–11.


The only narrative involving Anchises, apart from the short mentionsof his union with Aphrodite, recounts how he manages, in secret, tomate his mares with Laomedon’s divine horses so that he might breedextraordinary colts. The story, told by Diomedes (5. 263–71), enhanceshis own heroic way of conquering Aeneas’ horses in fighting instead ofusing the means of reproduction for greatness. Paris and Aeneas, evenas they accept the single combat with Menelaus and Diomedes, arerescued and taken away from the battlefield by Aphrodite (3. 373–82, 5.311–17). On a metapoetic level, it is as if we were witnessing theconflict between Aphroditean and Iliadic traditions about who belongsin what poem. I will not dwell on the case of Paris, for it is related moreto the Cypria than to the Hymn to Aphrodite and its examinationwould take me too far. But a few more words are needed to examineAeneas’ position in the Iliad.

Nagy suggests the existence of a traditional Aineid presentingstrong thematic parallels with Achilles’ story.97 His argumentationrests primarily on a passage of Iliad 13 that shows an angered Aeneaswithdrawing from battle because Priam does not honour him(13. 461). The pattern of frustration of timē, wrath, and withdrawalis recognizably Achillean. Another detail of the episode reveals thatthis representation of Aeneas is in fact competing with the hymnicone, for Deiphobus, exhorting Aeneas to join in the fight and defendAlcathous, his brother-in-law, reminds him of the time when he wasa child raised in Alcathous’ house. In this version, it is a hero, notthe Nymphs, who nurses the young son of Aphrodite. Whereas thetransition of Aeneas from the divine to the mortal world in the Hymnis aptly symbolized by the Nymphs, who are moreover traditionalkourotrophoi,98 there is nothing in the Iliad to motivate the entrustingof the care of Aeneas to a male relative, such that the motif appears asdisplaced and secondary compared to the Hymn’s version. Besides,Priam’s attitude towards Aeneas also contrasts with the way hehonours Hector above all men, likening him to the son of a god(24. 258–9). Hector is the only Trojan warrior matching Achilles inthe Iliad and not Aeneas, despite his being actually, not metaphori-cally, the son of a goddess.99 But Hector’s intimate connection to

97 Nagy (1979), 265–6.98 Faulkner (2008), 284–5.99 On the structural opposition between Hector and Aeneas in the Iliad, see Brillet-

Dubois (1998).


Achilles originates in the common fate that awaits them: death andglory. The Iliad, though integrating Aphrodite’s son into the heroicepic and even adjusting his characterization to its own purpose,finally markedly rejects him as unsuited to its narrative. In an extra-ordinary reversal of the Hymn’s poetics, Poseidon suffers ¼å� (‘grief ’) at the thought that Achilles might kill Aeneas (he associates¼å� with Aeneas’ name at 20. 293) and regrets that he should,innocent and pious as he is, experience pain, ‘vainly grieving becauseof the deeds of others’ (ał ��Œ! Iºº��æ�ø� Iåø�, Il. 20. 298).100 Bymotivating in these terms his rescue of Aeneas, Poseidon is clearlyassuming the role that Aphrodite, threatened by the grief of mourn-ing her protégés, assumes elsewhere in the Iliad. But he is alsorepeating, as a powerful fighter among the gods, the earlier exclusionof Aphrodite. However brave (the courage he demonstrates in booktwenty’s duel is unambiguous), Aeneas does not belong in the worldof ¼å� , he does not belong as Achilles’ foe. His moira, his share offate but also his assigned role in epic traditions,101 is to overcomemortality by surviving and fathering a long lineage, and not to con-quer Iliadic glory (he will not reappear in the Iliad after this). In otherwords, he is sent back to poems like the Hymn to Aphrodite. There-fore I do not think it is necessary, as Nagy does, to postulate theexistence of a heroic Aineid: the characterization of Aeneas as anotherAchilles seems to be the result of his integration into the Achilleantradition, just as the erotic nature of Aphrodite’s aristeia results fromits transposition into a song about love.

It appears finally that, while the Hymn to Aphrodite adapts thenarrative structure of the Achillean aristeia to construct its meaningand enhance its cosmogonic and laudatory purposes, the transferenceof Aphroditean elements seems at the same time to be part of theIliad ’s ‘poetic strategies’.102 This strongly suggests that creative andsubtle interaction existed between well-established Aphroditean andIliadic traditions. The transposition of the prophecy concerning Ae-neas, of Hector’s wish for his son, of Patrocles’ disarming, all point tothe borrowing of dictional elements firmly related to specific contexts,themselves included in an elaborate narrative, which makes it difficultto exclude allusion from a strict oralist perspective. That the diction of

100 I take here Iåø� as a participle.101 See Nagy (1979), 268; also 40.102 Burgess (2009), 69.


our Hymn to Aphrodite can be considered in places to be secondary toour Iliad’s might prove that theHymnwas composed or written downlater than the Iliad. It does not in any respect disprove that thehymnic and the heroic traditions developed simultaneously in afruitful dialogue, defining their themes and poetics in relation toeach other. It is very tempting to suppose that the same poets couldadopt alternately Aphroditean or Achillean perspectives, that is,according to the occasion or the mood of the audience, sing in turntales of war or tales of love, glorify the mortal hero or celebrate theworks of the smiling goddess.


7

The Seventh Homeric Hymn to Dionysus

An Epiphanic Sketch

Dominique Jaillard

The seventh Homeric Hymn to Dionysus is among the Hymns whichhave attracted the least critical attention and it is undeniable that, at firstglance, it does not appear to offer much with which to grapple.1 Fromboth a philological and a historical point of view, the poem’s relativebrevity (59 lines) and the absence of precise historical allusions haverendered efforts at dating and contextualization particularly tricky.2 Butit is perhaps the linear and clear character of the Hymn’s narration

1 Note, however, the forthcoming publication of the proceedings of a day-longconference devoted to Hy. 7, organized by A. Andrisano, ‘L’Inno Omerico VII aDioniso e l’iconografia relativa’ (Ferrara, 23 November 2005). Note that Englishtranslations of the Hymns are throughout taken from WL.

2 Conclusions about dating have varied widely: Patroni (1948) attributes Hy. 7 tothe poet of the Iliad and Odyssey, while Gemoll (1886) dates it to the Hellenisticperiod. There is no decisive argument for either a late or an early date. Janko (1982),183–4, concludes that ‘in fact, its diction indicates a post-Homeric date’, and that‘there is nothing to establish the Hymn’s geographical origins, but it may tentativelybe dated within the seventh century’. More cautiously, it is tempting to adopt thenuanced conclusion proposed by AHS 379 (‘There is no reason to deny it to the sixthor seventh century bc’), or the comments of the more recent editors Cassola 288(‘non databile’) and WL 17. In dating a Homeric Hymn it is important not only toconsider isolated elements of the text transmitted in the collection (linguistic featuresor historical allusions), which is at best a variant of a tradition whose degree ofvariability over time is often difficult to establish, but also the tradition itself. In thecase of Hy. 7, any attempt at a solution remains purely speculative, given the currentstate of knowledge. The question also seems inseparable from the equally obscureissue of the performative contexts in which this type of short narrative hymn devel-oped, of which Hys. 7 and 19 are our only surviving examples. The historical problem

which best explains the fact that it has rarely been studied or com-mented upon in and of itself. The very simple framework of the story iseclipsed by the powerful and manifest epiphanic image of the god.Dionysus is captured by pirates and takes possession of their ship,which he transforms by his marvels (ŁÆ�Æ�Æ) into a dionysiac uni-verse similar to that which is depicted by Attic painters: not only thecelebrated image of Dionysus sailing on the Exekias cup (Fig. 2), butalso the image of the oversized god, terrible and fascinating, who risesup on a black-figure amphora of the National Museum of Tarquinia.3

It is these images of the god demonstrating his power and therelationship that they establish between wine and the sea, morethan the Hymn itself, which have attracted the attention of specialists

Fig. 2. Dionysus Cup, by Exekias. By Permission of Staatliche Antiken-sammlungen und Glyptothek München

of the Tyrsenians is discussed below n. 5. It provides very little help for dating andcontextualizing the Hymn.

3 Black-figure kylix, Munich, Antikensammlungen, 2044, found in Vulci, ABV 146.21, v. 530; black-figure neck amphora, Museo Nazionale Tarquiniese 678, LIMC 790,plate 392, v. 500–50.

134 Dominique Jaillard

working on Dionysus and ancient art: these scholars take account ofthe poem in as much as it encapsulates, with an undeniable visualpower, recurrent and highly significant features of dionysiac epi-phany.4 From a different perspective, the Hymn has also been treatedby historians as an important document concerning interaction be-tween the Aegean and the western Mediterranean: they considerprincipally the poem’s mention of Tyrsenian pirates in lines 7–8(ºÅœ��Æd 'ıæ�Å��) and discuss its geography in comparison toother versions of the myth. The Tyrsenians have been taken to referto either Lemnians or Etruscans.5 As legitimate as it might be, thisapproach gives to the few geographical references in the poem aweight which accords poorly with the streamlined character of the

4 See Lissarrague (1987), 16–118, Isler-Kerényi (2007), 171–86.5 See De’ Spagnolis (2004), Nobili (2009), with earlier bibliography. 'ıæ�Å��

(‘Tyrsenian’) can refer to the inhabitants of Lemnos (at least from the time of Thuc.4. 109; see also Philoch. FGrH 308. 100, D. S. 10. 19. 6), who are also taken to bePelasgians (Hdt. 4. 145, 6. 137–40) and Etruscans (Hes. Th. 1015–16, Hecat. FGrH 18,19, etc.). The image of the Tyrsenians, whether they be Lemnians or Etruscans, issufficient to call to mind piracy in the context of the sixth century bc; see Nobili(2009), 11 ff. Taking into consideration the intentionally indeterminate character ofspace in Hy. 7, any attempt to give a more precise historical or geographical value tothe Tyrsenians seems in contradiction to the Hymn’s poetics. The suggestion ofCrusius (1889), often cited since, of an Attic performance context for Hy. 7, specifi-cally at the festivals ofDionysus celebrated at Brauron and as a commemoration of theabduction of young Athenians by Pelasgian or Tyrsenian pirates, is purely speculative.A version of the narrative specifically designed for such a context would haveincreased the number of specific geographical and mythical references. At the most,one could say that there is nothing against a Panhellenic poem such as ours beingperformed in this context, but there is no way to know whether it was. As regards theview that the role played by the Athenians (the capture of Lemnos by Miltiades in 510bc) in the elimination of piracy in the Aegean sea supports an Athenian origin and aconnection with the Anthesteria, during which the god seems to have been carried in afloat shaped like a boat (see the excellent discussion of Parker (2005), 302–3), this istoo general an argument and has no more demonstrative value than the vagueiconographic parallels recently put forward as evidence; see De’ Spagnolis (2004),61–6, criticized by Nobili (2009), 4. The version of ps.-Apollod. 3. 37–8 (whatever itssource) and the monument of Lysicrates suggest that there were Attic variants of thestory—but our version of Hy. 7 is not one of them. The identification of the pirates asEtruscans and Corinth as the place of the Hymn’s production, recently defended byNobili (2009), runs up against the same types of objections: her arguments onlyprovide a context in which versions of the story may have circulated, and to whichEuripides’ version may refer (13). The links she proposes between Hy. 7 and theCorinthian references of the myth of Arion are also no more than suggestive (18–26).However, her article illuminates a real difficulty, by showing that there is no proof thatthe Lemnians were considered Tyrsenians before the second half of the fifth centurybc (the first attestation is by Thucydides).

The Seventh Homeric Hymn to Dionysus 135

narrative, whose framework and emphasis seem on the contrary tobe exceptionally (and perhaps intentionally) detached from all localanchorage, whether it be cultic, mythical, or historical.

For this reason it seems essential foremost to try to understand theinternal logic and the particular poetics of the Hymn’s narrative,whose most striking feature is, as discussed above, the almost exclu-sive focalization on the epiphany of the god. Taking into account itsdisconcerting clarity, it is necessary first of all to compare the seventhHymn with the other Hymns of the existing corpus, in order toevaluate the uniqueness of its formal and thematic characteristics;this is the only somewhat certain ground for comparison, regardlessof the uncertainties surrounding the genesis of the collection. Theseventh Hymn to Dionysus (59 lines) constitutes, in the same manneras the nineteenth Hymn to Pan (49 lines), a short narrative Hymn.As such, it contrasts with both the long ‘narrative’ Hymns andthe short Hymns, which, even when they contain an outline of anarrative (Hys. 6, 27), do not properly speaking develop a narrativeof divine action. This difference of format seems, as much as one canjudge from only two surviving specimens, to involve a more profounddistinction. The narrative sections of the four long Hymns each dealwith a readjustment in the division of timai, the prerogatives andfunctions which define the place or status of everyone in the universe,man or god.6 The four long Hymns all narrate a situation, either thebirth of a ‘new’ god (Hermes, Apollo) or a ‘crisis’ (Demeter, Aphro-dite), involving an adjustment in the articulations or relations ofdivine prerogatives (timai), a refinement or partial redefinition ofthe divisions and interactions between divine powers and betweenmortals and immortals.7

In contrast, however, to the Hesiodic Theogony, a ‘hymn to Zeus’,which enacts the division and redistribution of powers from the pointof view of establishing the basileia (‘dominion’) of the sovereign god,the long Homeric Hymns consider the world to be already organized.The principal articulations are already in place, and are not lackingexcept for the abilities and divine functions which are precisely at theheart of the narrative and are the object of the hymnic celebration.Although they formulate a plan of Zeus, to whom the initiative most

6 See Detienne and Sissa (1989), 161–2, Jaillard (2005), 49–62, (2007), (forth-coming).

7 A feature well-emphasized by Clay (1989), and in this volume (Ch. 11).


often belongs (˜Øe �� K��º�E��),8 the long Hymns narrate thespecific actions of the god which they praise, thereby delineatingcertain areas of competence and prerogatives which belong to thatgod within the divisions of the pantheon. Even if the entirety ofrelations between divine powers is at issue, they are mostly consid-ered in connection to the specific deity celebrated, whom the Hymnby its very nature helps to magnify.9 At some decisive moment, thehymnic narrative represents a fictional situation in which one of thefunctional powers is lacking. The stress on the effects of the god’sabsence helps to define the deity’s field of action.10 Without Hermesthe inventor, Apollo remains ignorant of the lyre, which he claimsas his right just after birth in the Hymn to Apollo (131–2); withoutDemeter, who withdraws in the Hymn to Demeter from the council ofthe gods (��çØ�Ł�E�Æ Ł�H� Iª�æc� ŒÆd ÆŒæe� , ºı��, 92), thetimai of the gods would be reduced to nothing because the sacrificesof men would no longer be accomplished.11

In the Hymn to Hermes, which reflects upon hymnic practice itself,the question of the readjustment of timai is even the explicit subjectof the theogonic song performed by the young god (426–33).12 Thequestion is also treated (in a manner appropriate to Hermes) in boththe terms of division in the exchange with Apollo, as well as in theodd strategies deployed in the abnormal sacrifice accomplished bythe new-born god. All the glorious deeds (Œºı�a �æªÆ) of the youngHermes have an epiphanic value: they show the abilities and the skills(�æªÆ, �å�ÆØ) which at the end of the narrative will be recognized ashis own timai.13 The Hymn to Demeter also deals with timē and thedivision of functions amongst the pantheon. The goddess is deceivedby Zeus, who is above all concerned to ensure regular and regulatedcommunication between the underworld and the upper world (i.e.between Hades and the rest of the gods), without which the inviolableboundary between the living and the dead would create within thepantheon itself an untenable divide. At the end of the crisis, the new

8 Herm. 10. Variants: ˜Øe ��ıºBfi �Ø (‘by the will of Zeus’), Dem. 9; ˚æ��ø�ıŒØ�c� �Øa B�Ø� (‘crafty design of father Zeus’), Dem. 414; see also Aphr. 45–7.

9 See Nagy (1979), 186–7, Jaillard (2010).10 Jaillard (2007), 22–3.11 She and Persephone then return to Olympus at the end of the Hymn (484, �� Þ’

Y�� ˇhºı�� Ł�H� �Ł’ ›�ªıæØ� ¼ººø�), echoing line 92. Cf. Jaillard (2005).12 See Jaillard (forthcoming).13 On epiphany in Herm. see further Vergados in this volume (Ch. 5).


timai that Demeter and Persephone together obtain achieve Zeus’project much better than a marriage which traps Kore in Hadeswould have done: they weave between the different levels of thecosmos a network of interaction which is subtly articulated anddifferentiated. Although she is the queen of Hades, Persephone willparticipate fully in the assembly of the gods on Olympus for two-thirds of the year, by forming an indissoluble functional pairing withher mother. Meanwhile, the power of Demeter extends into Hadesthrough her provision of increased happiness in the afterlife forEleusinian initiates.14 In the Hymn to Aphrodite, Zeus, in turningagainst the goddess the irresistible power of sexual desire, subjugatesan extremely potent cosmic power by her own dominion and deepensthe divisions between mortals and immortals.15 Without resorting toan equally explicit conceptualization of polytheistic ‘theology’, theHymn to Apollo explores the consequences of the god’s birth, whilediscretely delineating the reality before his birth, as evinced, forexample, by the lands not yet cultivated which he passes on route toDelphi.16

It is precisely this reflection upon the structures of the pantheon,the �å�Æ�’ !Oº��ı (‘forms of Olympus’), which Chiron is said tohave taught to humans in a fragment of the cyclic Titanomachy,17 andthe entire theogonic and cosmogonic dimension characteristic of thelong narrative Hymns in the collection (despite their differences)which is absent from the two short narrative Hymns 7 and 19.At least, these short narrative Hymns do not deal explicitly with thepower struggles of the pantheon. The Hymn to Pan tells of the birth ofthe god, but the effects of his birth on the articulation of the pantheonand the ordering of the world are not considered. The narrative

14 See Jaillard (2005).15 Aphrodite’s universal power is emphasized at the outset of Aphr. (1–6). See

Pironti (2007), 116–19. On the separation of mortals and immortals, see Aphr. 45–52,247–55, with the commentary of Clay (1989), 164–70, 191–7, who argues that theHymn narrates the end of unions between gods and mortals. This is a strongpossibility, although the poem is not explicit about a complete division; see Faulkner(2008), 10–18. A diminishment of her powers and the interaction of gods and men isin any case implied.

16 See Detienne (1998), 25–9.17 Fr. 11 Bernabé: �Y �� ØŒÆØ��Å� Ł�Å�H� ª�� XªÆª�, ��Æ j ‹æŒ�ı ŒÆd Łı��Æ

ŒÆºa ŒÆd �å�Æ�’ !Oº��ı (‘And he led the race of mortal men to righteousness,teaching them oaths, good sacrifices, and the forms of Olympus’). Cf. [Hes.] fr. 283(M–W).


emphasizes the dual nature of the god, who is a ‘goat-footed, two-horned, rowdy, merry laugher’ (ÆNª��Å� �ØŒæø�Æ ��º�Œæ��ıªºø�Æ, 37), and the effects that he produces on those to whomhe appears: sight of him causes his terrified mother to flee (38–9), andhe delights the immortals when Hermes shows him to them (45–6).These genealogical references to parents are not an occasion tointroduce a matrimonial strategy aimed at reordering the pantheon,18

but rather serve, as in the shortest Hymns, to explore the features ofthe gods, as do epithets and brief celebrations of functional or de-scriptive traits. If one compares, for example, the first nine lines of thelong Hymn to Hermes, which mention the secret liaisons of Zeus andMaia in the murky cave in Cyllene, with the almost identical lines thatform the main body of the short eighteenth Hymn to Hermes, onlythe presence of a narrative recounting the journey of the newborn godfrom the atimia (‘dishonour’) of the maternal grotto to Olympus, theobject of his desire, confers upon the genealogical references an extrameaning: this is signalled straightaway in line 10 of the long Hymn, atthe start of the narrative proper, by the mention of the ‘purpose’(�� ) of Zeus.

One might nevertheless ask whether the variant in line 4 of Hymn18, naming Maia as the daughter of Atlas, the reprobate, would nothave suggested to a Greek mind, more polytheistic in orientation thanour own and replete with cultural knowledge that we do not have, animplicit equivalent to the absent narrative. One might also considerin this light the very brief introduction of Hymn 7 to Dionysus,‘Of Dionysus, glorious Semele’s son, I will make remembrance: howhe appeared . . . ’ (Içd ˜Ø��ı�� $�ºÅ KæØŒı�� ıƒ�� j ��ÆØ,‰ Kç��Å . . . , 1–2), to which the declaration of Dionysus in line 56,‘I am Dionysus the mighty roarer, born to Cadmus’ daughter Semele’(�rØ �’ Kªg ˜Ø��ı�� Kæ��æ�� , n� �Œ� ��Åæ j ˚Æ�ÅU $�ºÅ)and the closing farewell, ‘I salute you, child of fair Semele’ (åÆEæ�,�Œ� $�ºÅ �P��Ø�� , 58) both respond. As conventional as itseems, this could, by its insistence on maternal descent, be a signifi-cant clue for the analysis of a narrative about the misadventure of agod mistaken for the mortal son of some king (11).

From the very outset, the Hymn speaks of appearance and appar-ition, and their ambiguous nature. With the superior knowledge that

18 On the matrimonial strategies of Zeus, see Bonnafé (1985), 92–6, Clay (2003),29.


is indicated by the formula ��ÆØ (‘I will make remembrance of ’,2), evocative of the powers of Mnemosyne,19 theHymn first celebrateshow the god ‘appeared by the shore of the barren sea’ (‰ Kç��Å �ÆæaŁE�’ ±ºe I�æıª��Ø�, 2). Immediately following this, however, thepoem introduces the deceptive nature of his appearance ‘in the like-ness of a youth’ (��Å��fi Å I��æd K�ØŒ� , 3). The narrative depicts the godon a promontory, clearly visible to those who are ironically not ableto recognize him, and to whom he appears to be a son of a king: ‘theyreckoned he was the son of a princely line’ (ıƒe� ª�æ Ø� �çÆ��Ø��æ�çø� �Æ�Øº�ø�, 11). Perception is therefore from the outsettroubled, as the pirates are both unaware of and at the same timerecognize the affiliation of the young man with Zeus. The ambiguityof the dionysiac manifestation (phanein) begins the narrative andprepares for the victorious epiphany later in the poem. Dionysusreveals himself while also hiding himself, and shows himself whilerendering himself invisible to those who are not capable of seeinghim. If one accepts in the initial ‰ Kç��Å (‘how he appeared’, 2) bothits epiphanic value and its ambiguity, it then follows that it presentsmisrecognition as a constituent part of dionysiac epiphany.20 As in theBacchae of Euripides, it is vision and appearance that bring aboutthe process of misrecognition and recognition.21 The Hymn, therefore,narrates a specific pattern of dionysiac epiphany, which belongs tothe typical representation of the god in Greek culture, within a narra-tive structure and a semantic field that arise from the traditional

19 The formula is a variant, functionally and semantically equivalent to the for-mulae �� )�F�Æ (‘tell Muse’, Hy. 19. 1), �� )�F�ÆØ (‘tell Muses’, Hy. 33. 1),where the name of the god is likewise introduced by the preposition Iç� ; see alsoo��Ø )�F�Æ (‘sing Muse’, Hys. 9. 1, 14. 2), ��E� . . . ¼æå�� )�F�Æ (‘begin to singMuse, Hy. 31. 1), I�� )�F�Æ (‘sing Muse’, Hys. 17. 1, 20. 1), with the name in theaccusative; and ¼æå�’ I��Ø� (‘I begin to sing’, Dem. 1). For a full list of variants, seePavese (1991), 160–2. As Calame (1995), 6, has shown, the evocation of the god in thethird person, as the object of song, is more important than the manner in which thespeaker-narrator introduces it; cf. also in this volume (pp. 337–41). The divine powerat work can be explicitly named, its relationship to the speaker specified, or on thecontrary it can remain implicit, in which case it is contained in the meaning of theverb in the first person. Within the framework of the collection, any attempt toexplain the first person by an evolutionary or chronological schema, or by a distancingfrom the Muse, creates untenable contradictions (for example the early date of Dem.).

20 While the Hymns tend to introduce the narrative by a relative pronoun (‹ , �),the ‰ of Hy. 7 reflects the terms of his apparition, as if the enunciation is favouringthe singularity of his epiphany.

21 See Vernant (1986), 248–50, Segal (1997), 221–2.


knowledge of the poet.22 The Homeric Hymn subtly treats a motif,which the tragic poet later varies and amplifies.

Fettered by the pirates, Dionysus puts on his first display ofmarvels (ŁÆ�Æ�Æ). The bonds do not hold him and the osiers fallaway on their own, while the god sits impassively and smiles with hisdark eyes:

�e� �’ �PŒ Y�åÆ�� , º�ª�Ø �’ I�e �Åº��’ ��Ø��å�ØæH� M�b ��H�, n �b �Ø�Ø�ø� KŒ�ŁÅ��ZÆ�Ø ŒıÆ��Ø�Ø. (Hy. 7. 13–15)

But the bonds would not contain him, the osiers fell clear away from hishands and feet, while he sat there smiling with his dark eyes.

Similarly, the stranger in the Bacchae is liberated from his bonds andsits in an equally tranquil fashion to watch the collapse of Pentheus’palace.23 Upon viewing these wonders, the blindness of the piratesbecomes more profound: according to them, a god has sent them thisyoung man (K��d �E� ��Æº� �Æ�ø�, 31) in order to ensure a goodransom for them (26–31). The blindness of the Theban king is alsoincreased after he sees the deeds of Dionysus: he asks himself howand why the stranger has been freed and reappeared without under-standing.24 Placed against the blindness of the impious king is thesemi-clairvoyance of the tragic messenger,25 which in the Hymnbelongs to the helmsman of the ship. He does not know Dionysus,but he recognizes the divine, even Olympian, nature of the hostagewho does not resemble mortal men:

j ªaæ Z�f ‹�� ª’ K��d� j Iæªıæ�� ººø�Mb —��Ø��ø�, K��d �P Ł�Å��E�Ø �æ��E�Ø��YŒ�º� , Iººa Ł��E �Q !Oº��ØÆ ��Æ�’ �å�ı�Ø� (Hy. 7. 19–21)

This is either Zeus, or silverbow Apollo, or Poseidon; he is not likemortal men, but the gods who dwell on Olympus.

There then follows a second set of marvels (��åÆ � �çØ� KçÆ��ŁÆıÆ�a �æªÆ, ‘But suddenly they began to see miraculous apparitions’,

22 See Detienne and Sissa (1989), 168–71.23 Cf. E. Ba. 614–59; notably 621–2, �ºÅ�� ’ Kªg �Ææg� j X�ıå� Ł��ø�

�º�ı�� (‘And I sat close by watching calmly’).24 Cf. E. Ba. 642–6; esp. 645–6 �H �æ��Ø� j çÆ��fi Å �æe �YŒ�Ø ��E K�E ��ø

�� (‘How have you come outside to appear in front of my palace?’).25 See E. Ba. 770–4.


34), which transforms the vessel into a dionysiac ship, in which theimpious pirates have no more rightful place than Pentheus amongstthe Bacchants. Upon perceiving the fragrant wine that spreads acrossthe black ship, the pirates are seized by astonishment (�Æ��Æ �b ��ç� º�� Æ N��Æ , 37). Then immediately upon seeing (�Q �b N�� ,42) the mast and sails covered by vines teeming with grapes, the suddenappearance of ivy with fruits and flowers, and the oars girded bygarlands, they try in vain to row the ship to shore. Finally, in an ultimatereversal of the relationships of vision, it is the oblique and terrifying lookcast by the god (��Ø�e� ��æÆ N��, 48), now changed in front of theireyes into a roaring lion,26 which definitively places the pirates under thecontrol of fear (Kç��Ł��, 48): having vainly tried to huddle togetherwith the helmsman and having seen their captain seized by the lion (50),they leap into the sea, whereupon they are changed into Dolphins(�� ›H ��Å�Æ�, K��d Y��, �N –ºÆ �EÆ�, 52). Prey to a powerfulblindness that has continued to grow, the pirates are not able to perceiveDionysus when their gaze meets with his, except in his wild and terrify-ing form. By his metamorphosis into a lion and the apparition of ashaggy bear standing erect on his hind paws (46–7),Dionysusmanifestshis power (��Æ�Æ çÆ��ø�, 46) in his most terrible and formidablefacets. Meanwhile, to the helmsman who did not fail to recognize hishidden divinity, he reveals his name (Dionysus Kæ��æ�� , the ‘mightyroarer’, 56) and his genealogy (son of Semele, daughter of Cadmus, andZeus, 56–7) at the end of his epiphany. He also grants him limitlessprosperity and happiness (ŒÆ� Ø� �ŁÅŒ� �Æ��º�Ø��, 54).

The details of the dionysiac epiphany narrated by the Hymn makeevident certain powers of the god and the specific manner of hismanifestation, which could not but suggest to a Greek audienceaccustomed to listening to rhapsodes in the archaic and classicalperiods a broad network of interconnected myths, in which the god,at first not recognized, reveals his power in his most terrible anddestructive aspects, but grants exceptional benefits to those whorecognize him or take part in his rites.27 In both Orchomenus and

26 His roaring (ªÆ �’ ��æÆå��, 45) echoes in epic diction the terrifying groan ofweapons (Il. 4. 420, 12. 396, Hes. Sc. 423) or the earth (Il. 21. 387) and suits the god’sself-identification as the ‘mighty-roarer’ (Kæ��æ�� , 56) to the helmsman.

27 Cf. E. Ba. 859–61: ª��ÆØ �b �e� ˜Øe j ˜Ø��ı��, n �çıŒ�� K� �º�Ø Ł�� , j��Ø��Æ�� , I�Łæ��Ø�Ø �’ M�Ø��Æ�� (‘And he will come to know Dionysus, son ofZeus, who is fully a god, most terrible, although also most gentle to men’). SeeDetienne (1985), esp. 11–43.


Thebes, there is a strict division between the daughters of Minyas orCadmus doomed to religious impurity and the other women, whether‘citizen’ Bacchants destined to ensure the prosperity and happinessof the city or Lydians accompanying the god. However, there areimportant differences between other accounts of dionysiac epiphanyand the narrative of the Hymn. First of all, in the Hymn there is nohuman community or city involved. Everything instead takes placeoffshore, on the ‘wine-faced sea’ (�Y��Æ ��, 7),28 a shimmeringspace that is both uncertain and deceptive. As well, no previousoffence against the god is mentioned, nor any refusal to accept orto recognize him. The god appears on an unnamed promontory, hisdivinity hidden beneath his appearance as a courtly young man,almost as a trap extended to the passing pirates. Moreover, there isno reason given for the profound blindness of the pirates, except forbad fate that leads them on (��f �b qª� ŒÆŒe �æ� , 8).29 At most,the subtle reference to Dionysus’ Theban genealogy (57), could sug-gest the recurrent misrecognition of a figure whose divinity remainsquestionable until he shows his more terrible aspects.

But there are also significant differences between the narrative ofthe Hymn and other versions of the abduction of Dionysus by pirates.In the version given in Euripides’ Cyclops (11–12)—the oldest apartfrom theHymn, if one excludes the brief allusion of Pindar to dolphinswho have not forgotten the life (�Ø��) dear to men30—it is the angerofHera whichmotivates the action of the pirates. She wishes him to besold far away, after he has been freed from themadness with which shepreviously struck him. The ship of Silenus and the satyrs sets off tolook for the god, rounding the Cape of Malea and heading towardsSicily (13–22). In the Library of pseudo-Apollodorus (3. 37–8) the

28 The meaning of the epithet is disputed, but in a context so evidently dionysiac, itwould surely for an audience of the archaic period have suggested the metaphoricalsea crossed by an intoxicated drinker, on which he is tossed around, prey to confusion,and where he risks sinking into a state of mania; see Pi. fr. 124 SnM, Choeril. fr. 9Bernabé, Timae. FGrH 566 F 149, with the comments of Lissarrague (1987), 104–7.On the Exekias cup (Fig. 2, see p. 134), the horizon, the division between sky and sea(the dolphins come up close to the vines that surround the mast), is not indicated,thereby creating a space that is infinite (covered by a coral red background) and open(the depiction is not surrounded by a band as is most common).

29 They flee the same fate when they later throw themselves into the sea (ŒÆŒe��æ�� K�Æº�� , 51).

30 Fr. 236 SnM �ƒ ��ºçE�� (sc. KŒ �H� ºfi Å��H� ª��Ø): çØº��æÆ �’ �PŒ �ºØ��Ø��.


geography is even more precise, as Dionysus hires a ship of Tyrsenianpirates to take him from Icaria, in Attica, to Naxos, places where therewere sanctuaries of the god. Taking into account the method of mytho-graphical rewriting characteristic of the Library, these geographicalreferences in the myth could be at least as old as those of Euripides.The wonders produced byDionysus are not the same (the mast and theoars are changed into snakes, ivy and the sound of flutes spreadthroughout the ship, and the pirates are struck by mania), but theyare enough, as those in theHomeric Hymn, to represent the epiphany ofthe god and the dionysiac transformation of the ship. In Ovid’s Meta-morphoses (3. 597–691),31 the action takes place between Chiosand Naxos. In the Homeric Hymn, however, the only geographicalindications concern the faraway places where the pirates may have tocarry their hostage before he reveals his origins, fromEgypt orCyprus inthe eastern Mediterranean to the uncertain boundaries of the world:‘to theHyperboreans, or beyond’ (K %-��æ��æ�ı j �ŒÆ��æø, 29). Theaction of the poem itself is therefore detached from all local anchorageand has no attributable cause. And under the influence of Dionysus’presence, a somehow reasonable act of piracy is transformed into astrange and almost fantastical journey to the ‘elsewhere’.

It is tempting to interpret these differences as characteristic of thepoetics of the Homeric Hymn, which represent a choice in oppositionto other poetic traditions, or a distinctive trait of the particular versionof the tradition that has entered into the collection. If one acceptsthat the distance from local versions is one of the key indicatorsfor measuring the degree of Panhellenism of a poem or a poetic

31 Other versions or references to the story: Seneca Oed. 449–66; Hygin. Fab. 134,Astr. 2. 17 (¼ Aglaosthenes FGrH 499 F 3, on Naxos), the Tyrsenian pirates leap intothe sea without realizing their actions, under the charm of the song of the babyDionysus’ entourage; AP 9. 82; Luc. Salt. 22 (allusion), Dionysus subdues the pirateswith dance; Philostr. Im. 1. 19, ecphrasis of a tableau in which the ship of theTyrsenian pirates is in the process of attacking the ship of Dionysus and his Maenads,which is burgeoning with vines and ivy—the pirates, seized by madness upon hearingthe dionysiac harmonies, forget to row and some are in the process of transforminginto Dolphins; Opp. H. 1. 650; Nonn. D. 44. 240–9, 45. 105–69, 683–92—otherallusions to the story in Nonn. are discussed by Vian (2000); Serv. A. 1. 57. The friezeof the choragic monument of Lysicrates (334 bc) offers a distinctive version: the scenetakes place on the coast, where Dionysus sits and plays with a panther while his satyrsattack the pirates, whom Dionysus has begun to transform into Dolphins; seeEhrhardt (1993). The connection of the frieze to the narrative of a dithyramb victoryin the Dionysia of 334 bc, which the monument commemorates, or to a satyr drama,is possible but not certain; see further bibliography in Caspo (2003), 80.


tradition,32 the seventh Hymn to Dionysus could be taken as a modelPanhellenic poem, adaptable to many different performance contexts.The process of distillation or delocalization of the narrative may haveproved to be easier since the dionysiac sea is an ‘elsewhere’ that can berepresented more abstractly, as is indicated also by the representationof the winy sea on Attic drinking cups from the middle of the sixthcentury bc. But there are many other ways of distilling and delocaliz-

ing a narrative. For instance, the Arcadia of the nineteenth Hymn toPan is no less Panhellenic: it offers an adequate landscape for the god,but in no way fixes it in a more precise location, nor attaches it to theparticular cults and mythic elaborations of a local community.33

We should also consider for a moment the influence of perform-ance context on the composition of the short narrative hymns. We donot know for certain the constraints imposed upon a rhapsode by aHymn limited to about fifty verses, but the simple fact that thereare Hymns of varying length dedicated to the same deities suggests aresponse to functional demands. How best to celebrate a divinitywithin the predefined framework? How to compose the agalma thatwill most please the rhapsode’s divine audience and most increase histimē? The response of generations of rhapsodes who fashioned thetradition of Homeric Hymns could, in the case of short performances,have been to favour a brief and concentrated narrative sequence,capable of demonstrating the powers of the god within a more orless minor episode, but designed in a fashion so as to encapsulate thedistinctive features of the divinity. The narrative thus constructs apoetic epiphany of the divinity,34 both invoked and evoked, in a scenewhich crystallizes the figure of the god and the details of his activity.Rather than exploring the position of the god within the divisions ofthe pantheon and the cosmos, the Hymn instead takes on a strongiconic value, which, by means of a dense narrative sequence, rendersthe god intensely present in the place of performance.

In this respect, the poetics of the seventh Homeric Hymn toDionysus seem deeply similar to the economical art of painters suchas Exekias, who distils into a synthetic image on his drinking cups the

32 See Nagy (1979), 115–21. On the Panhellenic aspect of the long narrative Hymns,see Clay (1989) and in this volume (pp. 246–7); cf. Introduction (pp. 19–22).

33 On the landscapes of Pan, see Borgeaud (1988), and in this volume Thomas (Ch. 8).34 On the problem of textual epiphanies, see Pucci (1998), 69–80, and for Herm.

Jaillard (2007), 21–2, 69 ff.


epiphany of the dionysiac world, in which the drinker is caught upwhile drinking the wine: the iconic image of the tondo is even possiblyrevealed to him after he has met the fascinating eyes on the exterior ofthe cup. In this context, the actions of the pirates, which in thenarrative logic of the Hymn serve to articulate the particular qualitiesof misrecognition and recognition in the dionysiac epiphany, have noplace. On the cup, there is only the representation of the dionysiacworld in which the drinker participates fully and with happiness, if herespects the rules of the symposion. As has been well-established,35

the iconographic documents of the archaic period do not provideassistance in dating the appearance of the myth of the Tyrsenianpirates, neither the tradition of the story nor the text of the Hymn ashas been transmitted to us. Nonetheless, the connection between theaesthetics of the poetic production and the visual images has anothervirtue, in that it helps in understanding the process of crystallizationand distillation of the narrative: both the painters and the tradition ofthe Homeric Hymn either elaborated or selected the imagery that hasto do with the epiphany of a specifically epiphanic god,36 therebyisolating the motif. As such, both the Hymn and the images expressthe shared knowledge involved in religious and social practices,37

while in return helping, as part of a dynamic process, to give formto that shared knowledge, notably from a Panhellenic point of view.

The differences between the poetics of the two short narrativeHymns, to Dionysus and Pan, depend largely upon specific config-urations of power, whose brief and striking images are elaborated byrhapsodic traditions. The seventh Hymn to Dionysus narrates a pureepiphany, whose context is reduced to a narrative motif required tosignal the initial misrecognition of the god that leads up to hisrevelation: the actions of the Tyrsenian pirates, who seem to comeout of nowhere, take place in a location which has no other functionthan to provide a stage for the manifestation of the god’s power.The epiphanic narrative is constructed in relation to the modes ofappearance and action of the god: the ambiguation of appearance andapparition, the alteration of perception and sight, the freeing of

35 AHS 378, Lissarrague (1987), 118, West (2003), 17. The theory that the repre-sentation of dolphin-men illustrates the metamorphosis of the pirates in Hy. 7 hasrecently been supported by Kossatz-Deissmann (1992), but justly criticized by Ras-mussen and Spivey (1986), Caspo (2003), 79 ff. (with earlier bibliography).

36 See Detienne (1985), 11 ff.37 See Scheid and Svenbro (1996), 3–5.


bonds, and the transformation of the transitory and mobile humanspace of the ship into a purely dionysiac space (as in the GoldenAge,38 there is spontaneous vegetation, but without soil it owesnothing to the cultivating work of Demeter and causes the growingseasons of flowers and fruit to overlap). Plants and dionysiac sub-stances overrun the space: a fragrant wine spreads throughout theship, adding to the visible marvels (ŁÆ�Æ�Æ) a divine scent (O�cI�æ��Å, 36–7), yet another mark of the god’s hold upon the place.39

The quick rhythm of the narrative, which scholars of the nineteenthand early twentieth centuries were quick to condemn as a sign ofpoetic poverty, incoherence, and eccentricity,40 in fact accords withthe accelerated temporality in which the god produces marvels(ŁÆ�Æ�Æ) and phenomena that conflict with the ordinary rhythmof time: the rapid flow of the narrative is underlined by the accumula-tion of adverbs (��åÆ ‘quickly’, ÆP��ŒÆ ‘immediately’, ÆrłÆ ‘rightaway’, K�Æ��Å ‘suddenly’).41 As for the nineteenth Hymn to Pan, itproduces imagery that is just as strongly epiphanic, but does so byintegrating the figure of the god into the landscape in which he carriesout his actions. The evocation is not followed by a single narrativethat treats together certain powers of the god, as is the case in theseventh Hymn to Dionysus; instead, the poem contains a number ofshort, isolated narrative elements (Pan’s dance amongst the shep-herds, Hermes’ arrival in Arcadia, the mother’s fright upon seeing hernewborn’s face, etc.), all of which contribute to the representation ofthe god’s appearance within the pastoral space of the mountain.

In producing a ‘sketch’ of dionysiac epiphany, the seventhHymn toDionysus removes it from the mythical (non-recognition of the godleading to the introduction of his cult throughout Greece)42 or ritual(maenadic possession, symposion) contexts which ordinarily give rise

38 Cf. E. Ba. 141–3, 695–711, with the comments of Segal (1997), 65, Vernant(1986), 254.

39 Cf. the mention of Syrian incense in E. Ba. 144–50, at the height of the trancedescribed in the parodos.

40 Gemoll (1886), 317, AHS 379, Humbert (1936), 169.41 See Detienne (1985), 89–98 for discussion of the features of dionysiac

dunamis. For a parallel analysis of ‘how the scene changes are in syncopationwith steady linguistic features’, see in this volume Thomas (p. 155).

42 Cf. E. Ba. 83–7, Y�� ŒåÆØ, Y�� ŒåÆØ, j Bæ�Ø�� ÆE�Æ Ł�e� Ł��F j ˜Ø��ı��ŒÆ��ª�ı�ÆØ j +æıª�ø� K� Oæø� % Eºº�� N �P- j æıå�æ�ı IªıØ� , �e� Bæ�Ø�� (‘GoBacchae, go, leading Bromius, child of a god, from the Phrygian mountains to thebroad streets of Greece—Bromius’), and see Detienne (1985), 11–42.


to the epiphany (in its many forms) within narrative or ritual practice;the poem thus performs a sort of abstraction, a fact which revealsall the better certain aspects of its internal structure. It is the veryapparition of the god in the form of a young man which produces themisrecognition of the god: as such, Dionysus’ appearance reiteratesthe blurring of lines between mortal and immortal, which is impliedby his double birth as the son of Semele, daughter of Cadmus, andZeus. The ambiguous form in which Dionysus manifests himself onthe promontory, where he appears seemingly without cause, deceivesand lures the Tyrsenian pirates with the promise of a marvellousransom. It is thus the initiative of the god, whose appearance inducesthe wild and impious act of the pirates, which produces the epiphanyof his powers and the dionysiac transformation of the sea.

The known ambiguity of dolphins enamoured with human lifeis significant in that it blurs (or, to be more exact, displaces), in asimilarly ambiguous fashion, another constituent boundary of theworld order—that between men and beasts.43 Through their impietyand their misrecognition of the god, the pirates reveal their lack ofhumanity, while dolphins on the contrary approach towards human-ity. The dolphins’ sudden appearance in the extremely ellipticaldescription of the metamorphosis, with the brief and almost drystatement that the pirates ‘became dolphins’ (��ºçE�� ’ Kª��,53), accomplishes the dionysiac epiphany. As a simple image, it callsto mind, without further description or explicit allusion to anothermyth, the broader imagery of the dolphin familiar to men. Thecompressed image of the dolphin in the Hymn stirs up a densenetwork of other images and stories: dolphins accompanying ships,saving men from drowning, and taking them back to shore;44 dol-phins enamoured with the flute (ç�ºÆıº�Ø)45 and responding to thecall of Arion’s flute with an emotion that remains foreign to thepirates who abduct the poet;46 or dolphins leaping in a circle aroundthe ship like a dancing chorus.47 Moreover, in blurring the boundarybetween humanity and animality, Dionysus also effects another

43 Cf. Pi. fr. 236 SnM, quoted above n. 30.44 Plut. Mor. 984c, Ael. NA 12. 22, etc.; see Stebbins (1929).45 Pi. fr. 140b. 16–17 SnM [��ºçE�� ] �e� � IŒ�� K� ��ı ��º�ª�Ø j ÆPºH�

KŒ��Å�’ KæÆ�e� º� (‘[the dolphin] which the lovely sound of flutes inspires in thewaveless sea’).

46 Hdt. 1. 23–4.47 See Caspo (2003), esp. 78 ff.


displacement: from coarse and savage pirates, he makes musical andbenevolent beasts, whose philanthropy accords with the sweetness ofthe song being performed. This same sweetness is attributed toDionysus in the final invocation of the seventh Hymn:

åÆEæ�, �Œ� $�ºÅ �P��Ø�� • �P� �Å ��Ø��E� ª� ºÅŁ�� ªºıŒ�æc� Œ��B�ÆØ I�Ø�� (Hy. 7. 58–9)

I salute you child of fair Semele; there is no way to adorn sweet singingwhile heedless of you.

It is through his auditory qualities, previously eclipsed by the intensityof the epiphanic vision, that Dionysus reveals himself to the helms-man. He is this time presented as Kæ��æ�� (56), ‘mighty-roarer’ or‘rumbler’, an epithet which conveys the ambiguity of the auditory andmusicalworld ofDionysus, which varies between the terrifying roar ofthe lion, the deafening racket of tympanies mixed with the shrillsounds of flutes, and the sweet and pleasurable sound produced bythese same flutes when they accompany the mountain wandering ofentranced bacchant women.48 By changing the pirates into dolphins,the epiphany of the god transforms the savage cry of the lion into adancing choir and therefore ends by celebrating the sweetness thatDionysus confers upon song.

But why, one might ask, the Tyrsenian pirates? Taking into con-sideration the unique aesthetic of the Hymn, ‘Tyrsenian’ is nothingmore than a convenient signifier in the delocalized space of thedionysiac sea. It is polysemous and connected to various geographicallocations in the Greek world.49 One might hypothesize, in giving fullvalue to the metamorphosis of the pirates into Dolphins, that the seaas a space for piracy is antithetical to the blissful dionysiac seadepicted with such force by Attic painters in the image of Dionysussailing, or figuring in the motif of a voyage across the winy sea that thedrug (ç�æÆŒ��) of wine promises to participants in the symposion. Itis from this dionysiac sea that the god is supposed to come when hearrives by sea in the cities of men. And it is in the same sea that hefinds refuge when he is pursued by his enemies.50

48 See E. Ba. 159–65.49 On 'ıæ�Å�� (‘Tyrsenian’), see above, note 5.50 See the tale of Lycurgus in Il. 6. 123–43.


The seventh Homeric Hymn to Dionysus can be analysed as a docu-ment all the more valuable for our understanding of the epiphanicfigure of Dionysus when one defines more exactly its particular poeticlogic. In order to do so, a double comparative study has been under-taken above: on the one hand, the poem has been compared to theother Hymns of the collection, taking into account the length ofthe narratives and the particular qualities of the deities celebrated;on the other hand, the poem has been compared to the corpus ofnarratives and images dealing with dionysiac epiphany. The seventhHymn, or more probably the tradition that underlies it, has abstractedfrom the framework of the story of Dionysus’ abduction a ‘sketch’ ofdionysiac epiphanic structure. It displays this structure in a narrative-image, whose iconic force lies in the fact that it evokes virtually, inthe absence of any explicit allusion, an entire dionysiac mythology.It recalls a set of narratives, categories, images, cults, and places,which are all possible resonances of dionysiac elements relevant tothe audience. From a poetic point of view, the narrative is configuredaccording to the epiphanic powers of Dionysus. We are thereforedealing not only with the narration of the divine epiphany, but withthe epiphanic structuring of the narrative. In the relationship it createsbetween words and images, the seventh Hymn to Dionysus thus givesprecise meaning to the statements that later tradition attributed toSimonides: ‘the word is the image of things’ (› º�ª� �H� �æÆª��ø��NŒ�� K��Ø�), ‘painting is silent poetry and poetry is a talking painting’(�c� b� ÇøªæÆç�Æ� ��Å�Ø� �Øø�H�Æ� �æ��Æª�æ��Ø, �c� �b ��Å�Ø�ÇøªæÆç�Æ� ºÆº�F�Æ�).51 Other versions of the myth are no lessepiphanic, but by the emphasis they place upon the auditory mani-festations of Dionysus, they create a different relationship betweenthe visual and the auditory. Moreover, through the details that theysupply, other versions predetermine to a greater extent the network ofmyths with which they resonate. Meanwhile, in the long narrativeHomeric Hymns such dazzling epiphanic appearances as that in theseventh Hymn have been subsumed by the complex ‘theological’rationales which order the narrative according to exact configurationsof the pantheon.

51 Simon. T 47 a–b (Campbell).


8

The Homeric Hymn to Pan

Oliver Thomas

The Hymn to Pan begins unexpectedly:

Iç� �Ø % Eæ��Æ� ç�º�� ª�� , )�F�Æ . . . (Hy. 19. 1)

Tell me, Muse, of the dear son of Hermes . . .

Where is the personal name of the divinity which occurs in the firstline of all the other Homeric Hymns, and of the hymnic proem toHesiod’s Theogony?1 The indirection is not strictly ambiguous, sinceonly one son ofHermes is a plausible subject for a hymn. Neverthelessit may, unless the original performance context created an expectationof a hymn to Pan, have required enough cogitation that the next twowords—ÆNª��Å� �ØŒæø�Æ (‘goat-footed, two-horned’)—produceda cognitive ‘epiphany’, in a low-level mimesis of the true epiphanywhich is the notional goal of many Greek hymns.2 But whether or notwe identify Pan immediately, both ç�º�� (‘dear’) and the markedpatronym prime us for the subsequent importance of Hermes’ genea-logical relationship with and affection for Pan.

1 Superficially similar is Hy. 33. 1–3, Içd ˜Øe Œ��æ�ı �ºØŒ��Ø�� )�F�ÆØ,j . . . j ˚��æ� Ł’ ƒ��Æ�� ŒÆd I�Å�� —�ºı��Œ�Æ, ‘Round-eyed(?) Muses, tellof Dios kourous . . . horse-taming Castor and blameless Polydeuces’. However, ˜Øe Œ�Fæ�Ø / Œ��æø had crystallized from a patronymic (‘the sons of Zeus’) to a name in itsown right (‘the Dioscuri’), as its preponderance over apparent alternatives such as˜Øe ıƒ�, Œ��æø ˜Ø� or ˜Øe . . . Œ��æø shows.

2 It is relevant to the possible confusion that Pan’s father is not always Hermes. Themanuscripts offer ÆNª��Å� in verse 2 and ÆNªØ��Å� in 37: either is possible. Forepiphany as a request in cletic hymns, see e.g. Furley and Bremer (2001), i. 61.

Besides ‘goat-footed’ and ‘two-horned’, line 2 calls Pan çØº�Œæ��(‘krotos-loving’), an epithet embellished by its echo of ç�º�� ª�� inthe same colon of line 1: Pan inspires and feels affection with areciprocity which Greek worshippers in particular would appreciate.Here, his affection is directed at krotos, which describes a rangeof percussive noises. Is this the stamp of dancing feet or snap of adancer’s fingers, or the clatter of castanet and tambourine in Cybele’scult? Or is it the krotos of a musician strumming a lyre and theaudience’s applause, in a performance such as Hymn 19 itself ?3 Weneed not decide yet. What follows, we will see, tends to privilege thefirst and last suggestions.

Next comes ‹ �’ (‘who’), the typical hymnic relative moving us onfrom the opening epithets; here ‘epic’ �� (LSJ s.v., B) signals that wewill continue with a description of typical behaviour, rather than witha specific narrative (similarly Apoll. 2, Aphr. 2). Pan

I�a ��Æ

��æ��’ ¼ı�Ø ç�Ø�fi A å�æ��Ł��Ø ��çÆØ (Hy. 19. 2–3)

ranges over the wooded meadows together with nymphs who hauntthe dance.4

As withHermes and his paternal love, we are introduced immediatelyto the nymphs who will play a large role in the Hymn, and to itsparticular focus on their choruses. The poet has varied the line-endformula ��Æ ��Ø��Æ (‘grassy meadows’), so that the ��Æ retaintheir traditional nuance of a habitat for nymphs, but are also adornedwith the trees where dryads reside.5

The clause introduces a location and a type of motion, and fromhere to line 26 the Hymn is dominated by three wild spaces and (fromline 8) a whirl of unrestrained movements through them.6 By

3 Pan snaps his fingers: e.g. LIMC Pan no. 33. Pan in Cybele’s cult: Pi. frs. 95–6SnM, Brommer (1956), 1003. Krotos of lyre: e.g. Posidonius fr. 62 E.-K.; cf. S. fr. 926TrGF, of a harp. The Athenians pleased Pan with the honour of the ‘bare krotos’ ofapplause: Luc. Bis. Acc. 9–10; see Borgeaud (1988), 149.

4 å�æ��ŁÅ (‘who haunt the dance’) is a hapax. Most compounds of -ÅŁÅ express‘with good / bad / familiar / unfamiliar character’. However, compare å�Øæ��ŁÅ ‘accustomed to the hand, tame’ and qŁ� ‘accustomed haunt’ (LfgrE s.v., 1). Schmidtsuggested å�æ�ª�Ł��Ø, ‘who rejoice in the dance’. The nymphs’ erotic relations withPan nearly surface in ç�Ø�fi A (which I translated ‘ranges’): see below on lines 25–6.

5 Nymphs in ��Æ: Il. 20. 8–9, Od. 6. 123–4, Aphr. 98–9; later A. R. 3. 1218.6 In lines 4–7 a song by the nymphs is described: Iwill discuss this later (pp. 159–60).

152 Oliver Thomas

contrast, the lines from 27 on are dominated by Hermes’ purposefuljourneys to Cyllene and to Olympus. Whereas humans and the othergods experience space as a network of set paths and destinations, andwhereas Hermes’ movements as messenger and herald are archetyp-ically directed, Pan moves like the paradigmatically free-ranging goatand the pursuing goatherd, both of whose spirits he embodies,through a more open space beyond civilization’s structures; so, per-haps, his primary association is with Arcadia as a whole rather thanwith a specific city.

ç�Ø�fi A �’ ��ŁÆ ŒÆd ��ŁÆ �Øa Þø��œÆ �ıŒ��,¼ºº�� b� Þ��Łæ�Ø�Ø� Kç�ºŒ�� ÆºÆŒ�E�Ø�,¼ºº�� ’ Æs ��æfi Å�Ø� K� MºØ��Ø�Ø �Ø�Øå��EIŒæ��Å� Œ�æıçc� Åº��Œ�� N�Æ�Æ�Æ��ø�.��ºº�ŒØ �’ IæªØ��Æ �Ø�æÆ�� hæ�Æ ÆŒæ�,��ºº�ŒØ �’ K� Œ�Å�E�Ø �Ø�ºÆ�� ŁBæÆ K�Æ�æø�O�Æ ��æŒ�� • ��b �’ . . . (Hy. 19. 8–14)

He ranges here and there through the thick underwood, sometimesattracted to gentle streams, but sometimes he traverses amid the sheercrags, ascending a very high peak which looks out over the flocks; andhe has frequently run across the tall whitened mountains, and on thelower slopes he has frequently driven and killed beasts, with piercingeyesight. At other times . . . 7

ç�Ø�fi A �’ ��ŁÆ ŒÆd ��ŁÆ �Ø� (‘he ranges here and there through’) recallslines 2–3 ç�Ø�fi A I�� (‘he ranges over’), and contains the first �Ø�(‘through, across’) of an emphatic quartet. After ��ŁÆ ŒÆd ��ŁÆ thevariety of his movements continues to be emphasized by balancingadverbs, all first in their phrases: ¼ºº�� . . . ¼ºº�� ’ Æs (‘some-times . . . but sometimes’) and ��ºº�ŒØ . . .��ºº�ŒØ . . . �� (‘fre-quently . . . frequently . . .At other times’). A consonant effect isachieved in Þ��Łæ�Ø�Ø� Kç�ºŒ�� (‘attracted to streams’), wherethe passive suggests undeliberated, whimsicalwandering. Even withinthe dancing-ground of 22–3, his movements are unrestrained (seebelow).

7 Þø��œÆ may also be clumps of rushes. Switch to aorists denoting a typical actionwhich has happened and continues to happen: Faulkner (2005). Gemoll’s suggestionÅº��Œ�� (so that Pan is looking for the flock) is good; Pan is Åº��Œ�� in thebucolic fragment discussed in Bernsdorff (1999), verse 11. Either way, there is anallusion to the common iconography of Pan peering into the distance after flocks(I��Œ��ø�): see Jucker (1956).

The Homeric Hymn to Pan 153

Verses 2–26 categorize Pan’s world into three broad regions. Thefirst is characterized by meadows, which connote lush vegetation anda water-source, and trees (2–3). Next the nymphs appear higher up,on ‘the crown of the goat-left crag’ (ÆNª�ºØ�� æÅ . . . Œ�æÅ�Æ, 4);this is not, however, left by goat-footed Pan, since he ‘has beenallotted every snowy ridge, the peaks of the mountains, and tracksover crags’ (��Æ º�ç�� Øç��Æ ºº�ªå��, j ŒÆd Œ�æıça Oæø� ŒÆd��æ��Æ Œº�ıŁÆ, 6–7).8 Here water has ceded to snow, and vegeta-tion to rock.9 This second scene is abandoned as we jump downhill towhere we left Pan in lines 2–3: again he ‘ranges’ and again there iswater and vegetation (8–9). The view pans back uphill in lines 10–12,as the ‘snowy ridge’, ‘peak’, and ‘crag’ of 6–7 are carefully revisited inreverse order. Then at line 13 we cut to the Œ�Å�� (literally ‘shins’) ofthe mountains where Pan hunts—a new location as a new character-istic activity is specified.10 But almost immediately Pan climbs againfrom the hunt (14–15), back to the water and vegetation of an uplandmeadow beneath the peaks (20–6). The three scenes could be sche-matized as in the table below:

At the different levels, Pan’s behaviour also varies between the bestialand the humanoid. As a hunter, he is no goat. In the meadows, heplays music and dances like a pastor, but drinks from streams like a

Activity Altitude Characteristics

Pan with nymphs; music and dance Peaks, crags, ridges Snow; bare; no animalsUpland meadow Water; vegetation; animals?

Pan alone; hunting Lower slopes Water; vegetation; animals

8 Œ�æÅ�Æ ‘crown’ is a variant for Œº�ıŁÆ ‘tracks’ which produces a ring-composi-tion with 4 ��æÅ . . . Œ�æÅ�Æ ‘crown of the crag’. But Œº�ıŁÆ avoids a tautology withŒ�æıç� ‘peaks’, and there is still a ring-composition of ideas, if not of words. ��æ��Ø is ‘containing crags’ (LfgrE) not ‘rocky’ (LSJ). Pan and tracks, perhaps qua patron oftranshumant pastors: Preller (1894), 741 n. 3.

9 ‘Goat-left’ was the ancient interpretation of ÆNª�ºØł, whatever the etymology.One explanation was ‘left by goats because bare and pastureless’: Paus. Gr. Æ 38 Erbse.

10 The ‘shins’ are above �� (‘foot-hills’) but below º�ç�Ø (‘ridges, necks’) orŒ�æıçÆ� and Œ�æÅ�Æ (‘peaks, heads’): the metaphors are noted by sch. D Il. 2. 497. LikeArtemis, Pan controls the death of wild animals, sometimes as patron of hunter,sometimes as saviour of hunted; hunting small game is also a herdsman’s pastime. SeeBorgeaud (1988), 62–73.

154 Oliver Thomas

goat. At some points in the peaks he acts like a human pastor (11), butat others he is a supergoat who can even reach the ‘goat-left crag’, or issummoned as a god by the nymphs.

However, any static scheme fails to capture how the scene-changesare in syncopation with steady linguistic figures. The meadow isintroduced in line 8 and expanded with ¼ºº�� (‘sometimes’);then ¼ºº�� ’ Æs (a balancing ‘sometimes’) is not a further expansion,but introduces a scene-change. The crag remains in focus through thefirst ��ºº�ŒØ �’ (‘frequently’, 12), since the line describes the moun-tains as ‘tall’ and ‘whitened’; a new topic—Pan’s running—is intro-duced. By contrast, the ��ºº�ŒØ � in 13 zooms in, specifying the topic(Pan runs in the hunt) and focusing attention rather on the lowerslopes. Meanwhile, 11 Åº��Œ�� (‘which looks over the flocks’) and14 O�Æ ��æŒ�� (‘with piercing eyesight’) form a bridge betweenthe two areas of behaviour, as does 16 �ÆæÆ�æ��Ø (‘would surpass’)after �Ø�æÆ�� (‘has run across’).

At line 14 the prevalence of main verbs expressing motion beginsto wane:

��b �’ ��æ� �ŒºÆª�� r� ¼ªæÅ K�Æ�Ø��, ��Œø� o�� F�Æ� IŁ�æø��ı��· �PŒ i� �� ª� �ÆæÆ�æ��Ø K� �º��Ø�Zæ�Ø , l �’ �Ææ� ��ºıÆ�Ł� K� ��º�Ø�Ø�ŁæB�� K�Ø�æ�å�ı�Æ å�Ø �º�ªÅæı� I�Ø��. (Hy. 19. 14–18)

At other times, when alone at dusk he ascends from the hunt, he makesa cry, amusing himself with sweet music to the accompaniment of reed-pipes: he would not be surpassed in melodies by the bird who amongthe leaves during the blossoming spring pours forth her lament andpours out her honey-voiced song.11

The pace drops for a moment; ‘surpass’ is here, but the contest ishypothetical and musical, rather than the real chase of the hunt (12)or an athletic competition. The shift from motion to music is sig-nalled by the prominent, riddling �ŒºÆª�� (14). Why does the goat-god suddenly ‘make a cry’ like a bird? Line 15 gives a hint: thepanpipes, as a high-pitched wind instrument, approach birdsong;

11 Baumeister’s ��d �’ ��æ�� ‘towards dusk’ avoids the adjectival use of ��æ� as if it were ��æØ� (but compare ��ıå� ¼ K��åØ� ‘at night’). LSJ is wrong about 15K�Æ�Ø��. On 17 l �’ ‘who’, see Germany (2005), 202 n. 47. Although the repetition of‘pour’ in line 18 is slightly troubling, no emendation stands out.


Alcaeus of Messene twice uses the same verb Œº�Ç�Ø� of piping.12 Thismetaphor is then dwelt on for three lines in the Hymn’s only extendedcomparison, of Pan to a nightingale. The comparison is not straight-forward, and I will just sketch its stimulating open-endedness, ratherthan attempting to resolve it into a fixed interpretation.

The nightingale is identified indirectly, by being a paradigm song-bird and a female lamenter, and by the crepuscular solitude of Pan’smusic: nightingales stand out for singing after dusk and with suchmelodic virtuosity and stamina.13 However, the Greek for nightingaleis IÅ��, so the identification is perhaps confirmed by its cognateI�Ø�� (‘song’) at the end of line 18. A further similarity is certainlysweetness: ��ı�� ‘sweet’ is placed with hyperbaton immediatelybefore the comparison, and resumed by �º�ªÅæı� ‘honey-voiced’.Perhaps too, the frenetic, apparently irregular shifts in a nightingale’ssong are matched by the ‘jumping around’ or ‘dancing’ of the pan-piper’s mouth from reed to reed (Ach. Tat. 8. 6. 6–7), and suggest that�ÆæÆ�æ��Ø might mean ‘surpass in tempo’ as much as ‘surpass inquality’.14 However, for all these points of contact, Pan is character-istically ‘amusing himself’ with his music.15 Against male, partlyinstrumental self-entertainment stands female, purely vocal mourn-ing for another, for the nightingale’s song is a threnody poured forthlike honey—the sonic correlate of a wineless libation at a funeralrite.16 This idea of ‘pouring’ sounds has two nuances which coexist intension: smoothness (like honey) or unceasing ‘phonorrhoea’. Thelatter suggests a further difference between Pan and the nightingale:

12 Anth. Plan. 8. 226. For the disputed ªº�Ç�Ø in Pi. fr. 97 SnM, see Lehnus (1979),169–79. The panpipes’ tessitura is inferred from depictions—see Haas (1985)—andfrom the nuances of �ıæ�Çø ‘whistle’ (e.g. LSJ appendix s.v., I 2). A bird’s vocal organ isnow its ‘syrinx’; cf. ‘sandpiper’.

13 The word ‘nightingale’ encompasses both night and song (cf. PIE *ghel- ‘call’).The evening virtuoso is actually a male attracting females. In Greek myth, it wasperhaps specified by pitch to be female (Procne), and from the range of women’ssongs to be lamenting (for her son Itys); similar stories are collected in Epit. Bion.37–43.

14 The panpipes’ sound is ÆN�º� (‘variable’) at E. Ion 499–500. Further evidenceabout Greek panpiping might clarify the picture.

15 In Orph. H. 11. 6, Pan even makes cosmic harmony çØº��Æ�ª��Ø �º�Bfi ‘withplayful music’.

16 For wineless libations, see Graf (1980). Hy. 19 seems to ignore or to be ignorantof the story that Pan plays—rather like the nightingale—in longing for Syrinx. Themyth is extant in Greece and Rome from Vergil on (e.g. Ov. Met. 1. 689–712);Hellenistic sources are assumed. The panpipes’ tone is also mournful at E. Hel. 171.

156 Oliver Thomas

birdsong could, on its own terms, be pleasant while disordered; Pan’smusic, as a projection of human music, requires order. This deviancein the comparison makes this rather different from human musicalcompetition in Greece, including as a trope in bucolic poetry.17 Butthe choice perhaps suggests the inadequacy of alternative compar-anda: clearly no human herdsman’s piping can match Pan’s.

The landscape of the comparison is woods during early spring.Aristotle HA 632b 20–3 casts light on the relevant bird-lore: ‘The(female) nightingale sings continuously for fifteen days and nightswhen the mountain’s foliage is at last returning; afterwards it sings,but no longer continuously.’ In fact, males sing at night in spring untilthey find a mate, then sing almost exclusively during the day.18

Continuity with this landscape helps us switch seamlessly backfrom the change of pace:

�f� � �çØ� �� çÆØ Oæ��Ø�� ºØª��º��Ø

ç�Ø�H�ÆØ ��ŒÆ ��d� K�d Œæ��fi Å �ºÆ��æøfi

º��ÆØ, Œ�æıçc� �b ��æØ��Ø �hæ�� !Hå�.�Æ�ø� �’ ��ŁÆ ŒÆd ��ŁÆ å�æH�, ��b �’ K �� æ�ø��ıŒ�a ��d� �Ø��Ø, ºÆEç� �’ K�d �H�Æ �Æç�Ø��ºıªŒe �å�Ø, ºØªıæBfi �Ø� IªÆºº�� çæ�Æ �º�ÆE K� ÆºÆŒfiH º�ØH�Ø, ��ŁØ Œæ�Œ� M�’ ��ŒØ�Ł� �P��Å ŁÆºŁø� ŒÆ�Æ��ª��ÆØ ¼ŒæØ�Æ ��fi Å. (Hy. 19. 19–26)

With him then the clear-singing mountain nymphs, moving to and frowith rapid footsteps by a dark spring, sing, and Echo moans around themountain’s peak. The divinity, proceeding to this or that part of thechoruses, and sometimes into the centre, makes his way through themon rapid feet, and wears over his back the tawny cloak of a lynx; hisheart swells at the clear singing in a soft meadow where crocus andfragrant blooming hyacinth mingle indistinguishably with the grass.19

We return to a meadow, which contains flowers which bloom early inspring. Pan appears in a ‘cloak’ (ºÆEç� ) which turns out against

17 The genetic relationship between the passage of Hy. 19 and this trope is unclear.18 � �’ IÅ�g� fi ¼��Ø b� �ı��åH �æÆ ŒÆd ��Œ�Æ ��ŒÆ��, ‹�Æ� �e Zæ� X�Å

�Æ��Å�ÆØ• ��a �b �ÆF�Æ fi ¼��Ø �, �ı��åH �’ �PŒ�Ø. See Amrhein, Körner,and Naguib (2002).

19 In line 20, I accept Barnes’s emendation of �ıŒ��, which scans less easily. Echodeserves a capital: nouns in -ø were originally female personifications, and ��æØ��Ø(‘moans around’) is anthropomorphic. For �Ø�ø ‘make one’s way through’ (normally‘marshal’), compare Il. 24. 247, and Kç�ø ‘make one’s way to’, �Ł�ø ‘make one’sway after’.


expectation to be made from . . . lynx. This creates a smooth transi-tion by recalling the hunting of line 13.20 The Eurasian lynx isappropriate both as a denizen of Pan’s remote mountains, and as amaster of subtlety in its reclusive lifestyle and its dappled skin, whichproves the aptness of 14 O�Æ ��æŒ�� (‘with piercing eyesight’).21

Second, Echo, nature’s correlate to Pan’s music, is female and ‘in theminor key’, just as the nightingale was. The detail adds to the pleasantdepiction a tinge of the uneasiness to which piping was the isolatedpastor’s antidote.

As Pan is rejoined by the oreads, the tone rejoins movement andpleasure. Again we find ç�Ø��ø (‘range, go to and fro’), recalling thesituation of the meadow at lines 2–3 and 8–9. The nymphs dance toand fro ��ŒÆ �� ‘with rapid footsteps’, while Pan makes his waythrough them (�Ø��Ø: �Ø� again), equally �ıŒ�a ��; his movementsagain involve the phrase ��ŁÆ ŒÆd ��ŁÆ ‘this way and that’.22 As lines2–3 developed into an inset song, so these lines in a more manneredfashion set up a more elaborate inset song, via the ring-compositionºØª��º��Ø � ºØªıæBfi �Ø� . . .�º�ÆE (‘clear-singing . . . at the clearsongs’), perhaps supported by ��ŒÆ �� ıŒ�a �� (‘withrapid footsteps’). That structure makes lines 25–6 something of acoda, but they are not stranded. They complement the focus onlandscape in the early part of the hymn with a visual and olfactoryclose-up. The sensuous attention to detail is comparable to the Hymnto Demeter 6–14, describing the meadow from where Kore wasabducted. The ravishing meadow recalls the hint of eroticism in line3 ¼ı�Ø ç�Ø�fi A ‘goes to and fro together with’, and ŒÆ�Æ��ª��ÆØ (26,‘mingle’) in particular insinuates what Pan may get up to whenevening turns to night and the dancing ends (see e.g. LfgrE ��ªø 2d).

20 There, ŁBæÆ ‘beasts’ was unclear: it is normally used of large game (and so inthat formula, ŁBæÆ K�Æ�æø� ‘killing beasts’), but Pan more often hunts birds andhares.

21 Lynx remote: X. Cyn. 11. 1. Dappled quality: e.g. E. fr. 863 TrGF, Plu. Mor. 16d.As Nicholas Richardson reminds me, the lynx is itself sharp-sighted (cf. Lynk-eus’supernatural vision), so that it is a match for Pan’s sharp eyesight. In Callimachus’Hymn to Artemis, at 88–9, Pan is butchering a Maenalian lynx for his hunting-dogswhen Artemis arrives. Quite possibly, Callimachus is fleshing out what happened tothe rest of the lynx whose pelt Pan is wearing here. He may furthermore have sensedthe rather similar hunting-then-music structures of Hys. 19 and 27 (to Artemis).

22 Metrical variatio within repetition (as in ��ŒÆ �� . . . �ıŒ�a ��) was atraditional embellishment: see Hopkinson (1982), adding Herm. 77–8 �a �æ��Ł��Z�Ø�Ł��, j �a �’ Z�ØŁ�� æ��Ł��.

158 Oliver Thomas

We are at the end of verse 26, and the Hymn has developed in anunusual way. Unlike the long Hymns or Hymn 7, Hymn 19 fills itshymnic frame by describing at length the god’s typical activities andambit, without a narrative of a specific past episode; the closestparallels are Hymn 27 (22 lines), and the Hymn to Aphrodite, wherethe main narrative does not start until verse 45. The expected narra-tive does now arrive, with the typical contents of Pan’s conception,birth, and introduction to Olympus; but it is subsumed into adescription of an oread song.23 Before considering this, let us returnto the previous nymph-song, since we did not fully explore theramifications of the oreads . . .

Æ¥ �� ŒÆ�’ ÆNª�ºØ�� æÅ ��ı�Ø Œ�æÅ�Æ—A�’ I�ÆŒ�Œº��ÆØ, ��Ø�� Ł��, IªºÆŁ�Øæ��,ÆPå��Ł’, n ��Æ º�ç�� Øç��Æ ºº�ªå��ŒÆd Œ�æıça Oæø� ŒÆd ��æ��Æ Œº�ıŁÆ. (Hy. 19. 4–7)

who tread along the crown of the goat-left crag, invoking [or ‘summon-ing’] Pan, the pastoral god, resplendent in his mane and squalid, whohas as his lot every snowy ridge, the peaks of mountains, and tracksthrough crags.

Following the play in lines 1–2 with the manner of reference to Pan,the expected proper name finally appears here, subsumed into thefirst inset song as the rhapsode’s expected narrative about Pan issubsumed into the second. This final ‘revelation’ of the addressee ispointedly placed, as a cry first in its verse.24 The oreads’ song is achoral hymn, consisting as it does of an invocation with the nameplaced first (as so often in Greek hymns), a trio of epithets, andfurther attributes after a relative pronoun; we do not hear of aparticular prayer, unless I�ÆŒ�Œº��ÆØ is interpreted as ‘summon-ing’. There is in fact a precise parallelism with the encompassingrhapsodic hymn, which began with evocation, followed by threeepithets (two of which were physiognomical, as here), followed bya relative clause, again describing a typical haunt.25 Three furtherpoints about the oreads’ hymn stand out. First, they make an obvious

23 Regarding music within the Homeric Hymns, see Calame, this volume (Ch. 14).24 Sapph. fr. 44. 33 Voigt —��’ O�ŒÆº�� ‘invoking Paean’ has a strikingly

similar effect.25 It is this, and the ring-composition mentioned in n. 8, which imply where the

nymphs’ words stop.


wordplay with ‘Pan’ and ��Æ (‘every’); the rhapsode could not havebrought this out better than with the elided accusative—A�’, since thisis only an apostrophe away frommeaning ‘everything’.26 Second, theyare intimate enough to call him ÆPå��Ł’ (‘dry, squalid’, particularlyin his unwashed hair, 6) without a tone of reproach or horror, andeven to let this follow immediately on IªºÆŁ�Øæ�� (‘resplendent in hismane’), which from a human perspective constitutes a jarring pro-gression. Thirdly, they locate Pan at precisely their own performancelocation, in order to enhance their relationship with their addressee.Returning to the overarching level of the rhapsode’s composition, theuse of ŒÆ�Æ��ı�Ø (‘tread down’ or ‘tread along’, 4) to describe thenymphs’ danced performance is, with its focus on impact, of a piecewith ‘krotos-loving’ just before in line 2. The verb simultaneouslysuggests that Pan’s love of the noise of dancing was important there,and (therefore) that he receives the nymphs’ dances with pleasure.

The nymphs’ second inset song has largely parallel characteristics,but is arranged to take up roughly the second half of the whole hymn.It is performed with Pan’s own participation, so that no cletic func-tion is appropriate. The topic is an example of their various songsabout gods (28–9). We have heard that these songs in general fill Panwith pride and pleasure (24 IªÆºº�� implies both), which sug-gests that they normally contained an element of praise of the god intheir midst, as one might expect. But the genre of this second songis harder to establish from the summary given. Their specificperformance is described with �� (29, ‘tell of ’), which empha-sizes narrative rather than the invocation and prayer which set hymnsapart from other ‘god-poetry’; one could therefore see the song as anextract of theogonic poetry, supplementing Hesiod.27 However, itscontents are typical of the narrative in a hymn to Pan. Alternatively,one might take it more as a hymn toHermes, whose name comes first,

26 Compare Pi. fr. 96 SnM �Æ��Æ� (‘omnifaceted’), Castorion SH 310. 3��ªŒº�Ø�’ (‘all-famed’), PMG 936 several times. Later, and perhaps in the last in-stance—see Wagman (2000)—such wordplay develops into Pan’s interpretation as acosmic ‘Allgott’: so, for example, Orph. H. 11.

27 �� should be compared especially to Hes. Th. 10 ��Eå�� ‘they go’. Theseunaugmented imperfect forms express regular action without any past temporalreference: see West (1989). For the relationship of the Hymns with theogonic poetry,see Furley in this volume (Ch. 10).

160 Oliver Thomas

followed by three clauses beginning ‹ ª� (‘he’), in hymnic Er-Stil.28 Inany case, as with the nymphs’ earlier song, there are clear links towhat the rhapsode chants, which force us to compare and contrast theinset performance with the encompassing one. These enter when Panis born:

�Œ� �’ K� �ª�æ�Ø�Ø�% Eæ��fi Å ç�º�� ıƒ��, ¼çÆæ ��æÆ�ø�e� N��ŁÆØ,ÆNªØ��Å�, �ØŒæø�Æ, ��º�Œæ��, ��ıªºø�Æ. (Hy. 19. 35–7)

She bore in her halls a dear son for Hermes, from the first monstrous tobehold, goat-footed, two-horned, noisy, laughing sweetly.

Compare lines 1–2 % Eæ��Æ� ç�º�� ª�� . . . j ÆNªØ��Å�, �ØŒæø�Æ,çØº�Œæ�� (‘the dear son of Hermes . . . goat-footed, two-horned,noise-loving’). The last adjective has perhaps been changed becausewe are here looking at Pan as at a new quantity, whose likes are stillunknown, though his clatter is immediately perceptible.

One may wonder why it is stated that Pan was monstrous, etc.,‘from the first’ (¼çÆæ, 36). The second inset song here observes atendency in Greek hymns for the god’s character to be brought to afinal form within the narrative’s narrow compass, with the result thattheir personalities do not develop as ours do. This is of a piece withthe characteristic emphasis in Greek hymns on the origins of con-temporary life, which the nymphs’ song certainly shows.29 Thus, at itsclimax it gives an etymology of Pan’s name, instead of the merewordplay of the first song. When Pan was introduced on Olympus,

�� ’ ¼æÆ Łıe� ��æçŁ��IŁ��Æ��Ø, ��æ�ÆººÆ �’ › B�Œå�Ø� ˜Ø��ı�� ,—A�Æ � Ø� ŒÆº��Œ�� ‹�Ø çæ�Æ �A�Ø� ��æł��. (Hy. 19. 45–7)

28 See below (pp. 165–7) on the similarities with Herm. 1–4. Lost hymns to Zeusmay well have described how he fathered other gods, and praised the children too.Conversely, Apoll. includes a hint of a hymn to Leto at 14–18, beginning åÆEæ�,�ŒÆØæ’ t ¸Å��E , K��d �Œ� . . . (‘Be favourable j Hail, Leto, blessed because youbore . . . ’). For Er-Stil see Norden (1913), 163–6; it is rather rare in Greek hymno-graphy, but see the peroration of Aelius Aristides’ To Zeus.

29 Thus Dem. gives an aetiology for the sanctuary and rites at Eleusis, Apoll. forthose at Delos and Delphi, Herm. 24–62 for the tortoiseshell lyre, etc. In other cases,‘world firsts’ are presented without this causative relationship to contemporaryculture: e.g. Herm. 112–41 presents a precursor to (but not the cause of ) Dodecatheoncult at Olympia, and Hermes’ first interest in butchery.


All the immortals took pleasure in their hearts, and especially Dionysusthe Bacchant, and they called him ‘Pan’ because he pleased the mind ofall (pasin).30

This aetiology completes a theme of joy which has pervaded thenymphs’ song, ever since Pan’s birth. First come two lines of thehuman female response:

ç�Fª� �’ I�Æ&�Æ�Æ, º�� ’ ¼æÆ �ÆE�Æ �ØŁ��Å·��E�� ª�æ, ‰ Y�� ZłØ� I��ºØå�� Mßª��Ø��. (Hy. 19. 38–9)

She jumped up and fled, and the nurse left the child. For she was afraidwhen she saw his full-bearded, unappeasable face.

Here Pan’s mother may be the subject of the first words (as of thepreceding clause), or the nurse is the delayed subject of both verbs:the articulation would be clear in performance.31 Thus one or twowomen flee in fright. There is an oblique link here to the commonhuman fright at epiphanies, and also a sense that Pan is causing hisfirst panic attack, though in Greek terms that is normally a specificallymilitary phenomenon.32 The nymphs revel in human misguidedness.The humour is already clear in line 38, especially if the phrase wasfelt to overturn a normative formula çæ�Ø �’ ¼æÆ �ÆE�Æ �ØŁ��Å (‘andthe nurse took the child’) now represented only in Il. 6. 389.33 ThenI��ºØå�� (‘unappeasable’, 39) is ironic: the nymphs, given thepointed disjunction with 37 ��ıªºø�Æ ‘laughing sweetly’ and 24Iª�ºº�� ‘taking pleasure’, are presenting a strong human focal-ization, which they deem to be erroneous.34 39 Mßª��Ø�� (‘full-bearded’) implies that Pan appeared monstrous (cf. the echo 36��æÆ�ø�e� N��ŁÆØ � 39 Y�� ZłØ�) because humans cannot copewith a bearded infant. This horror at Pan’s inhuman body-hair con-trasts with the nymphs of the first song, who happily called himIªºÆŁ�Øæ��, j ÆPå��Ł’ ‘resplendent in his mane and squalid’, and

30 Pan is depicted with Dionysus already in the early fifth century, and verycommonly from the fourth century. See Brommer (1956), 1001; LIMC Pan no. 4,IV G, V E.

31 Some editors (e.g. AHS, WL) instead refer �ØŁ��Å to Pan’s mother. However, theword is normally a nurse as opposed to the mother.

32 Fear at epiphanies: e.g. Richardson (1974), 208. Panic: Borgeaud (1988), 88–102.33 Cantilena (1982), 299.34 For �ØºØª- and friendly laughter, see e.g. PW no. 129, Call. fr. 24. 13 Pf. It is

‘meta-humorous’ that the amusing disjunction centres on the question of Pan’shumour.

162 Oliver Thomas

with the oreads’ focalization of him in ��ıªºø�Æ (‘laughing sweetly’)as in fact more cute than screaming newborn humans.35

The oreads follow this with a balancing couplet containingHermes’ very different response:

�e� �’ Ærł’ % Eæ��Æ KæØ��Ø� K åæÆ ŁBŒ�� , åÆEæ�� b ��øfi ��æØ��ØÆ �Æ�ø�. (Hy. 19. 40–1)

But immediately swift Hermes took him and placed him in his hand,and the divinity rejoiced exceedingly in his mind.

Instead of double abandonment, we have double welcome (takingand holding get separate mentions), and instead of fear we haveexceeding joy, as in 45–7, quoted above. Thus within the humanhymn is an oread song in which, as implicitly in the first inset song,the nymphs can be seen to construct their personal relationship withPan in opposition to his relationship to humans: humans are firstafraid then wary of his society, whereas nymphs ‘get’ him and interactfreely with him (as they are doing during their performance) fromday one. Hymn 19’s view of how Pan interacts with his humanworshippers, and of the differences of mortals and immortals—twoissues which are otherwise important in Greek hymnography—seemspractically limited to this hierarchy pronounced by nymphs.36 How-ever, the representation is ironized and complicated by being that ofsecondary focalizers, under the final poetic control of a human. Theoverarching context of a human hymn to Pan implies that humans donot really think Pan is ‘unappeasable’ (39). Indeed, immediatelyfollowing the second inset song is the rhapsode’s envoi:

ŒÆd �f b� �o�ø åÆEæ�, ¼�Æ�, º��ÆØ � �’ I�Ø�Bfi ,ÆP�aæ Kªg ŒÆd ��E� ŒÆd ¼ººÅ ��’ I�Ø�B . (Hy. 19. 48–9)

35 The Epidaurian Hymn to Pan shows that humans could be more open-mindedabout appearances: PMG 936. 9–11 . . .�Æçıb �øH� �Æ j �På�æ�ı�� ,�P�æ��ø�� j K�æ�ø� �Æ�ŁfiH ª��øfi (‘ . . . a good dancer plying his body of everyform, a good looker conspicuous for his yellow beard’).

36 The oreads have also emphasized at lines 32–3 the normal pecking order: ŒÆdŁ�e þ� . . . K��ı�� j I��æd ��æÆ Ł�Å�fiH (‘although a god, he herded for a mortalman’). The relationship of mortals and immortals is a particular focus throughoutAphrodite. For hymns commenting on a god’s attitudes to his worshippers, see e.g.Apoll. 146 ‘But, Phoebus, it is Delos which pleases your heart the most’, Herm.541–9, 577–8.


And be thus favourable [also ‘with this, goodbye’], lord, I beseech you insong. Meanwhile I will call to mind both you and another song.

Here º��ÆØ ‘I beseech’, and the sense ‘be favourable’ underlyingåÆEæ�, entail that Pan is appeasable.37 The rhapsode’s desired relation-ship of favour-exchange (charis) is not that of spontaneous and mutualpleasure which the nymphs constructed for themselves (though chariscould denote this too), and after all nymphs are of a distinctly superiorstatus to mankind. Nevertheless, it certainly belies their one-sideddepiction of humans. Similarly, this subversion of the nymphs’ viewcomes at a crux of the hymn, where Pan moves from third to secondperson, in a virtual ‘epiphany’ symbolizing the rhapsode’s hope forPan’s presence at his performance. Again, he does not expect Pan to bea fellow-performer, as for nymphs, but he still exposes their tendance indescribing human fear at Pan’s first appearance.

Furthermore, the Homeric Hymns are a genre of communicationbroadcast on two channels: on ‘terrestrial’ is a narrative about anddescription of a god, and on ‘Sky’ is a request directed to the god,for favour. These broadcasts are not sequential, as they might seem,but simultaneous and entangled. The artistry and content of the parsepica, and its public performance, are all displays of reverence whichthe god is to notice, and which are presupposed by the closingrequest.38 Thus the elegance with which our rhapsode handles thetransition into the final address can be part of his strategy for generat-ing gratitude in Pan.39 In the event, the build-up through the nymphs’song about Pan causing joy in the gods, to a request that he should feelfavourable towards mortals, is triply elegant. �o�ø åÆEæ� oscillatesbetween the normal ‘feel charis with this performance’ and, withreference to the preceding verse (47), ‘feel charis as Dionysus rejoicedat you’. Second, this interacts with the very start of the hymn: as weheard in lines 1–2 that Pan is dear toHermes and krotos is dear to Pan,so here we find that Pan is dear toDionysus and that—hopefully—therhapsode’s krotos is as dear to Pan. Thirdly, the interesting partiality ofthe nymphs’ representation of divine-human charis, and its deflation,

37 The point is even clearer with the variant ¥ºÆÆØ (‘I propitiate’). For the sense ofåÆEæ�, see in this volume Calame (pp. 344–5).

38 For the two channels, see Danielewicz (1976), 119; also Calame in this volume(pp. 334–7).

39 The handling of such transitions in some of the other Hymns is discussed byCapponi (2003).

164 Oliver Thomas

self-reflexively become an integral part of how the rhapsode attemptsto engender Pan’s charis for himself.

Let me summarize my discussion of the two inset songs so far. Theyboth contain verbal parallels with the surrounding rhapsodic hymn,and are generically close or identical to choral hymns. Both focus onaspects which one expects in the rhapsodic hymn proper, and indeedsubsume important parts of it: the former focuses on naming (andsubsumes the expected naming of Pan at the beginning of the verse inthe accusative) and the choice of fitting attributes; the latter subsumesa typical hymnic narrative, with its (broadly speaking) ‘aetiological’accent. In both songs, the nymphs construct a close relationshipbetween themselves and Pan with a view to ensuring the efficacity oftheir performance, and in both cases the rhapsode suggests that thesong does find favour. But in the second song, the nymphs’ construc-tion of the human relationship with Pan is undermined by the rhap-sode’s own performance, particularly in the envoi which followsimmediately. These songs thus constitute a detailed meditation onthe contents and performative roles of hymns, using inset choralperformances as a suggestive analogue for rhapsodic performance.The handling of this might profitably be compared and contrastedto the similar role of the Deliades in Apollo 156–75, and the slightlydifferent presentation of the Muses’ songs in the proem of the Theo-gony, though I do not propose to flesh this out here.40

I have postponed my discussion of the opening of the nymphs’second song.

��ı�Ø� �b Ł��f �ŒÆæÆ ŒÆd ÆŒæe� ºı��•

�x�� Ł’ % Eæ��Å� KæØ��Ø�� å�� ¼ººø��, ‰ ‹ ª’ –�Æ�Ø Ł��E Ł�e ¼ªª�º� K��Ø�ŒÆ� Þ’ ‹ ª’ K �æŒÆ��Å� ��ºı��ÆŒÆ, Å�æÆ �ºø�,K��Œ��’· ��ŁÆ � �ƒ �� ˚ıººÅ��ı K��.��Ł’ ‹ ª� ŒÆd Ł�e J� łÆç�æ��æØåÆ Bº’ K��ı��I��æd ��æÆ Ł�Å�fiH• Ł�º� ªaæ ��Ł� �ªæe K��ºŁ��

��çfi Å Kß�º�Œ�øfi ˜æ�� çØº��Å�Ø ØªB�ÆØ.KŒ �’ K�º�� ª�� ŁÆº�æ�� . . . (Hy. 19. 27–35)

They sing of the blessed gods and tall Olympus. For example, they tell ofHermes, beneficent above all others, how he is the quick messenger for

40 Again, see in this volume Nagy (pp. 305–7) and Calame (pp. 351–2). Such acomparison forms part of a forthcoming doctoral dissertation by Sarah Harden(University of Oxford), whom I thank for her comments.


all the gods, and how he came to Arcadia of many springs, the mother offlocks—and his sanctuary as the Cyllenian is there. There, although agod, he herded straggly-haired [or ‘rough-haired’] flocks for a mortalman, since the liquid desire to mingle in love with the beautiful-hairedbride of Dryops assailed him and blossomed. And he accomplished afertile union . . . 41

This is done with notable style. After a chiasmus with alliteration inline 27, the switch in 29 away from formulaic �Æåf ¼ªª�º� (‘fastmessenger’) produces more alliteration, in Ł��E Ł�� ; also present arethe alliterative formulas % EæB KæØ��Ø� and Å�æÆ �ºø�. ThenŁ�º� ��Ł� (‘the desire blossomed’) and ª�� ŁÆº�æ�� (‘a fertileunion’) ring the sentence of lines 33–5. In line 32, łÆç�æ��æØåÆ(‘straggly-haired’) varies and opposes the formulaic and in this caseless interesting ŒÆºº��æØåÆ (‘with beautiful hair’), to bring the flockmore into line with Pan of the unappealing body-hair (6, 39).42

Structurally, the smooth transition into past narrative is noteworthy:the end of the nymphs’ attributive section focuses on Hermes’ travel-ling job, and he is still travelling in K��Œ��’ (‘he came to’, 31).

These lines are important for a quite separate issue: the cluster ofunderlined words and phrases is shared with the beginning of theHymn to Hermes:

% EæB� o��Ø )�F�Æ, ˜Øe ŒÆd )ÆØ�� ıƒ��˚ıºº��Å ��Æ ŒÆd �æŒÆ��Å ��ºı�º�ı,¼ªª�º�� IŁÆ��ø� KæØ��Ø��, n� �Œ� )ÆEÆ��çÅ Kß�º�ŒÆ� ˜Øe K� çØº��Å�Ø Øª�E�Æ. (Herm. 1–4)

Muse, praise Hermes, the son of Zeus and Maia who rules over Cylleneand Arcadia of many flocks, the beneficent messenger of the immortals,

41 The underlining becomes relevant below. WL prints Koechly’s �r�� and makes��å�� qualify ��: ‘It is Hermes alone that they tell of most of all’; but then ‘alone’is practically redundant. The etymology of KæØ��Ø� is probably ‘very swift’: Latte(1955), 192–4. However, ancient lexicography generally interpreted ‘very beneficial’,and it is unlikely that 29 Ł�� is a ‘gloss’ by the poet. For the genitive ˚ıººÅ��ı after�ƒ, compare e.g. Od. 14. 527. ‘Bride’ is the obvious meaning of ��çÅ; for ª�� of sexwithout marriage, see e.g. E. Hel. 190 (Pan rapes a nymph), LSJ ªÆø I 2. I do notthink ��çÅ can mean ‘marriageable daughter’ (cf. LSJ I 2); nor is it ‘nymph’(Villarrubia 1997, 11) since men cannot father nymphs.

42 łÆçÆæ� describes hair in Rufinus AP 5. 27 (a wizened ex-prostitute) and Nonn.D. 4. 363 (a dying man); in the þfourth-century lexicon PSI 8. 892 and Hesychius,łÆçÆæ�åÆ��Å is equated with çıŒØ�åÆ��Å , ‘with hair like/of seaweed’. It is uncom-plimentary, for humans. For ŒÆºº��æØåÆ BºÆ, see Od. 9. 336, 469.

166 Oliver Thomas

whom the beautiful-haired nymph Maia bore after mingling in lovewith Zeus.

The parallels may be even more thoroughgoing, if their apparentdisruption in Hymn 19. 32–3 is removed. Hermes’ infatuated herdingfor the mortal Dryops in order to get close to his wife is only recordedhere, and may well be a motif transferred from a more famous similaroccasion: Apollo’s herding for Admetus, which had initially been anexpiation, was motivated by desire at least by the time of Rhianusfr. 10 (CA) and Callimachus Hy. 2. 49; furthermore, Hermes’ theft ofcattle from Apollo was sometimes—though not in Hermes—locatedduring precisely this service.43 On a different level, the similarity ofHermes’ love for Dryops’ bride and Zeus’ love for Maia picks up onthe ‘like father like son’ characterization of Hermes which is promin-ent through Hermes, but particularly in the nocturnal secrecy of Zeus’affair with Maia described in lines 6–9.

Once we are primed in this way, other links to Hermes gainplausibility. The series of epithets ��æÆ�ø�e� N��ŁÆØ, ÆNªØ��Å�,�ØŒæø�Æ, ��º�Œæ��, ��ıªºø�Æ (36–7) is a rare structure in theHomeric Hymns, whose most elaborate instance is the description ofHermes, again just after his birth (Herm. 13–15).44 That Pan’s carersjump up and leave him (38, especially I�Æ&�Æ�Æ) bears a neat anti-thetical relationship to the newborn Hermes jumping up and leavinghis carers (Herm. 21–2, including I�Æ&�Æ ). Hermes takes Pan toOlympus wrapped in a hare-skin blanket for introduction,as Apollo took the swaddled Hermes to Olympus for introduction.45

Finally, the nymphs’ inset songs overlap with the contents of thesurrounding hymn, as does Hermes’ first song (Herm. 57–8 � 1–9),so that both rhapsodes place themselves in a tradition traceable to

43 The relationship with Vergil’s nymph Dryope who fathers a child with Faunus(A. 10. 551) is tantalizing. See Williams (1978) on the Callimachus passage; in Ant.Lib. 23, Apollo is in love with Admetus’ great-grandson Hymenaeus, and has left hiscows with Admetus’ herd when Hermes steals them. Hermes steals Admetus’ cows:Pherecyd. FGrH 3 F 131, sch. Nic. Alex. 560a; the theft occurs during Apollo’s serviceto Python, a mythic doublet of Admetus, in the hypothesis to Pindar’s Pythians.

44 Strings of epithets become more common in later Greek hymns, e.g. Hy. 8and Orph. H.

45 Internally, the hare-pelt is like Pan’s lynx-pelt of line 24: the less ambitious formof Pan’s hunting is alluded to while his infancy is described. It is perhaps worth notingthat the hare and Hermes’ tortoise provoke epic’s only uses of the genitive Oæ��Œfi��Ø�(Herm. 42, Hy. 19. 43).


superhuman performers.46 One might add that when Pan is depictedplaying the panpipes, this constitutes a point of contact with Hermes511–12 where Hermes invents them, and the contact may be re-inforced by the language of Hymn 19. 15 ��Œø� o�� F�Æ� IŁ�æø�,since the stem IŁıæ- occurs repeatedly in Hermes (152, and withreference to a musical instrument in 32, 40, 52, 485), but rarelyelsewhere in Epos (IŁ�æø in Il. 15. 364; ¼ŁıæÆ x 4).

All these similarities fit neatly with the emphasis which we notedabove: Hymn 19 from the start brings in Hermes as a benevolent andlike-minded father. It therefore might well choose to intertwinethe important relationship of Pan and Hermes with an intertextualrelationship, whereby it is the ‘heir’ to what must have been the mostimportant hexameter hymn to Hermes of its time.47 The hypothe-sized allusion is thus meaningful, and not implausible given a verysimilar relationship—this time of sibling rivalry rather than father–son accord—which I would see between the Hymn to Hermes itselfand the Hymn to Apollo, or between Callimachus’ Hymn to Artemisand the Homeric Hymn to Apollo.48 Such a construction, whereby anintratextual genetic relationship mirrors an intertextual generic one,allows for slippage between the two levels. In particular, Pan pleaseseveryone on Olympus as the poet wishes his hymn to please all hisaudience.

Given this possible intertextual relationship, one might speculateon the link between the comparison with the nightingale (lines 16–18,cited on p. 155) and Odyssey 19. 518–22, where Penelope describesherself crying in bed:

‰ �’ ‹�� —Æ��Ææ�ı Œ��æÅ, åºøæÅU IÅ��,ŒÆºe� I��fi Å�Ø� �Ææ� �� ƒ��Æ��Ø�,��æø� K� ��º�Ø�Ø ŒÆŁ�Ç��Å �ıŒØ��E�Ø�,l �� ŁÆa �æø�H�Æ å�Ø ��ºıÅåÆ çø��,�ÆE�’ Oº�çıæ��Å ! ”�ıº�� . . . (Od. 19. 518–22)

46 The types of correspondence between the inset songs used virtuosically in theproem ofHesiod’s Theogony and its main body are slightly different. Regarding Herm.57–8, see Vergados, this volume (pp. 103–4).

47 By contrast, Herm. can hardly allude to hymns about Pan, in whom it shows nointerest.

48 For parallels between the Hymns, see Dornsieff (1938), 81–4, Richardson (2007),89–91; for interpretation, Thomas (2009), 290–5. On Call. Hy. 3 to Artemis, seeAmbühl (2005) and in this volume Faulkner (pp. 190–1).

168 Oliver Thomas

As when the daughter of Pandareus, the yellow-brown nightingale,sings beautifully at the beginning of spring, sitting among the trees’thick petals, and with rapid modulations pours out her echoing voice,mourning for her son Itylus . . .

The two passages on their own do not require us to posit an allusionwith compression. However, it is at least suggestive that in manyversions Pan’s mother is Penelope.49 The intertext may then be thekind of allusion which evokes something to draw attention to itssuppression, for here Pan’s mother is the bride of Dryops. In betweenthe possible allusions to Penelope and Hermes is verse 21, whereat the sound of the nymphs singing to Pan’s syrinx ‘Echo moansaround the mountain’s peak’ (Œ�æıçc� �b ��æØ��Ø �hæ�� !Hå�).Echo here may well be a symbol, in Pan’s typical location, of Hymn19’s allusivity.50

Pan’s cult began to spread from Arcadia c.500 bc.51 This is a secureterminus post quem for the hymn, since we hear nothing aboutArcadian rhapsodic traditions before this date, and since the nymphsso prominent here seem to have been unusually unimportant inArcadia.52 The possible allusion to theHymn to Hermesmight providea slightly more stringent terminus post: its date is also unclear but inmy view falls between the rebuilding of the temple at Delphi in c.505bc and Sophocles’ Ichneutae.53 It is equally hard to identify a terminusante quem. I have mentioned a possible allusion in Callimachus(above, n. 21). FiliationwithHermes, if accepted, favours a date shortlyafter that Hymn, whose importance for later texts is limited, largely toSophocles’ Ichneutae and Callimachus’Hymns to Zeus and Artemis.54

Hymn 19 shows numerous interactions with traditional formulas, and

49 Roscher (1894).50 There is no strong reason for seeing a further allusion to Pan’s love for Echo, a

story first entailed by Call. fr. 685 Pf. The myth grew naturally from the independentlocation of Echo and randy Pan in mountain wildernesses (for Echo there, seeApollodorus FGrH 244 F 135), which may well be the idea at Hy. 19’s date. Germany(2005) gives a different analysis of Echo’s role in the Hymn.

51 For his importation into Athens c.490 see the famous story in Hdt. 6. 105, withParker (1996), 163–8. Other fifth-century locations: Brommer (1956), 961–8.

52 See Jost (1985), 476.53 Thomas (2009), 20–9, 309–12.54 For the general question of the Hymns’ reception, see in this volume Faulkner

(Ch. 9) and Nagy (Ch. 13).


I see no reason to suppose it was only an amateur’s ‘imitation’ oftraditional diction, rather than professional rhapsodia.55

For alternative dating-evidence, students of early Greek hexametermay turn to stylometry, but Hymn 19’s forty-nine lines are normallytoo few to generate significant statistics.56 Janko (1982), 184–5,suggests one morphological dating criterion which is not statistical,namely the uses of ��ıªºø�Æ and �ØŒæø�Æ. Nouns in -ø wereinitially declined as �-stems (like ÆN�� ), and were only later assimi-lated, at different rates, to �-stems. The process is already identifiablefor åæ� in Iliad 10 and the Odyssey, but the first datable sourcesfor ª�ºø�- are fifth-century and those for Œ�æø�- are Hellenistic.57

Comparable arguments can sometimes be made from semanticdevelopments. Thus �F�Æ� in the sense ‘music’ (15) and ¼ı�Ø asa preposition meaning ‘together with’ (3) are derivative usages, as is��ı�� (‘sweet’) in line 16. But in this case, for varying reasons noneof these improve our terminus post.58

For many other features, the problem is rather that variation mayrest on more than chronological factors. For example, it may havebeen simply the composer’s taste to use compound epithets so fre-quently, or longer strings of epithets, or an unusually dactylic rhythm(particularly in the second foot).59 A corresponding point regardingthe location of the composer’s training is the use of movable nu beforea consonant, to lengthen a princeps syllable (e.g. 38 º�� ’ ¼æÆ).Hymn 19 uses this resource five times, a higher rate than other epic

55 Formularity: Cantilena (1982), Fröhder (1994), 338–45. Hy. 19, if one acceptsthat the myths of Syrinx and Echo are not pertinent, shares the earlier focus on Pan’sattributes and genealogy, rather than on further mythology which gained ground inthe Hellenistic period. But this shift is a broad generalization, not a dating criterion.

56 So for the frequency of observed digamma, articular use of ›-�-��, masculinecaesura, bucolic diaeresis, word-break after the fifth princeps, or the treatment ofMeyer’s laws. For (less than satisfactory) statistics about ›-�-��, see Scott (1921), 90–1.For the metrical terms, see West (1982), 45, 153–5.

57 ª�ºø�-: A. Cho. 447, PW no. 129. Œ�æø�-: Megasthenes FGrH 715 F 27,Euphorion fr. 14 CA.

58 The sense of �F�Æ already occurs in Pratin. fr. 6 TrGF, Pi. N. 3. 28. That of¼ı�Ø is unique, and the prior phase where ¼ı�Ø ¼ adverbial –Æ already occurs atOd. 12. 415. ��ı� was an obscure, obsolete epithet of sleep; by Alcm. fr. 135 PMG,l�ı� ‘sweet’ developed by reanalysis; later, ��ı� was reinstated as an archaizingalternative, and finally (here) transferred to a referent other than sleep. See also Lacore(1997). Unfortunately, these phases cannot be dated.

59 Second foot dactyls x35/49. Compare the figures at Ludwich (1885), 327–9. (Thevery striking contrast between the figures for Il. and Od. deserves wider recognition.)

170 Oliver Thomas

texts, though again the sample is unreliably small. Janko suggestedthat in general the feature was more common in Ionian compositions,but I would allow more weight to factors such as the metrical fasti-diousness or even facility of the individual composer.60

In determining the Hymn’s location, unfortunately external infor-mation about festivals involving Pan is too exiguous to help us.Absences from the Hymn offer a few suggestions. Hymn 19 doesnot mention caves, which figure extremely prominently in Pan’sAttic cult, but hardly at all in Arcadia.61 The absence of allusions toPan’s military engagements might also tell against Attica in the firstfew decades after Marathon, with which his importation to Attica wasclosely allied; Aeschylus, for example, mentions Pan at the site ofPersian slaughter at Psyttaleia at Pers. 448–9. The absence of Cybele /Mother perhaps tells against a Boeotian cult background. Among thecontents which do occur in the hymn, the most indicative is probablyPan’s birth toHermes and the bride ofDryops, at Cyllene.62 Dryops isthe eponym of theDryopes, a race to whom the Greeks assigned threelocal connections: to Arcadia, where Dryops is sometimes located; tothe region which became Doris, from which Heracles expelled thembefore the Dorians moved in; and to the destinations of the subse-quent diaspora, particularly Ambracia, South Euboea, and the EasternArgolid.63 The genealogy goes against the three main local Arcadianversions, including the most popular one that Pan’s mother wasPenelope (to which Hy. 19 may allude).64 Nor does the mention ofCyllene imply an Arcadian cultic background to the Hymn. Rather,Pan is placed there only in literary sources, via Hermes, who isso important to Hymn 19.65 These considerations tell against anArcadian performance context.

60 Janko (1982), 64–8. I feel some unease about the poet’s control when he uses ‹ ª�three times in lines 29–32, switches subject so abruptly in verse 35, and uses K åæÆŁBŒ�� in verse 40 to mean ‘took in hand’ rather than ‘put in another’s hand’.

61 Caves in Attica: e.g. E. Ion 493–4; see Parker (1996), 164–5. They feature in PMG936 (Epidaurus) and prominently in Pindar’s Hymn to Pan (fr. 95. 2 SnM). See furtherLehnus (1979), 116 n. 52.

62 Strictly, line 32 ��Ł’ might mean ‘in Arcadia’ rather than ‘near Cyllene’. But thatwould render the mention of Cyllene less pertinent.

63 For sources on the Dryopes, see Strid (1999).64 Jost (1985), 460–3.65 Pan and Cyllene: also S. Aj. 695. Erycius AP 6. 96 places him there, but also

attributes to two cowherds a heroic bull sacrifice: one cannot say where the fictionstarts.


To sum up: I have argued that the Hymn is probably classical,perhaps from the generation after Hermes. Regarding performancelocation, the discussion is perforce aporetic. I have mentioned someregions which are less likely (Boeotia, Attica, Arcadia) and some suchas the Argolid which are marginally more likely, given our limitedinformation.

The Hymn to Pan is a text which repays examination through severallenses. For historians of religion it is unusually long among literarysources for Pan of (as I have argued) the classical period. It thereforehelps us not only to describe Pan, but potentially to examine thedynamics of a ‘new’ god’s spread across the wider Greek world overthe course of the fifth century. Meanwhile for cultural historiansit helps to map out, particularly through its representation of space,a conceptual world of life in the wilds and margins (K�åÆ�ØÆ�) asimagined by educated Greeks. For students of hymns as a literarygenre, it is a rare example of a mid-length hexameter hymn, and onewhich reflects on a hymn’s functions and the relationship of choraland rhapsodic hymns. Its possible allusion to Hermes is relevant toreconstructing the earliest phase of transmission of the HomericHymns: was Hermes widely known, and by whom and how, or werethe Hymns transmitted from the archives of a rhapsodic school,whose members would have learnt from each other’s work as a matterof apprenticeship; is it coincidental that Hymn 19 ended up followingthe shorter Hymn to Hermes? And in its displacement of such a fixedhymnic element as the god’s proper name in the evocation, its fastintroduction of the whole composition’s essentials, its play both inlines 5–7 and in 27–47 with the poetic use of inset songs, and itsvivid visualization of a god whose hybrid exterior is his most promin-ent characteristic, the Hymn reveals itself as a skilled and ambitiouscomposition. Let us honour the wide-ranging and versatile godappropriately, with a wide-ranging and versatile approach to hisHymn.

172 Oliver Thomas

Part II


9

The Collection of Homeric Hymns

From the Seventh to the Third Centuries bc

Andrew Faulkner

In memory of Penelope Stewart

The medieval manuscript tradition has conveyed a collection of thirty-three Homeric Hymns.1 The twenty-eight manuscripts of the* family,with the exception of three manuscripts of the subgroup z,2 transmitHymns 3–33 (or some portion thereof ) in the same order, adopted bymost modern editions.3 The distinct Leiden codex (M), the sole manu-script witness of Hymns 1 to Dionysus and 2 to Demeter, transposes 10and 11 but otherwise follows the identical sequence until it breaks off at18. 4. It is difficult to know at what point in the poems’ transmission thisparticular order was imposed, but the collectionwill not have been in theexact form we have inherited before the third century ad. The eighthHymn to Ares, characterized by its long list of epithets and planetaryallegory, is markedly different from the other Hymns in the collection

1 All 29 extant manuscripts are late, dated to the fifteenth century. For more on themanuscript tradition, see AHS xi–xliii, Humbert (1936), 12–18, Càssola 593–616,WL 21–2, Faulkner (2008), 525, Richardson (2010), 32–3; on At and M, Wilson(1974), Irigoin (1970), Richardson (1974), 65–7, Gelzer (1994), 113–25.

2 J and K contain Hymns 8–18 followed by 3. 1–186, H Hymns 8–18 followed by 3.1–55. See further AHS xxxix–xlii, Càssola 599–601, West (1970), 304.

3 Humbert (1936), esp. 18–19, groups together the Hymns to the same divinity; forthe order followed in earlier editions see his concordance table (256). Manuscripts ofthe ¨ tradition (and through contamination C) also contain at the end of thecollection a five-line poem �N ��ı , a version of the Homeric epigram transmittedin the pseudo-Herodotean Vita 9 (Allen); see AHS 442–3, Càssola 610—contra itsomission from the collection of Hymns Janko (2004), 283.

and appears Neoplatonic. This poem could have been included inten-tionally in antiquity or inserted sometime thereafter due to an accidentof transmission.4

A compilation of some sort must have been in existence before theinclusion of the eighth Hymn. Both Diodorus and Philodemus seemto have known a collection of Homeric hymns in the first century bc.They quote from the long Hymns to Dionysus, Demeter, and Apollo,referring in a general way to the o��Ø of Homer.5 The source ofPhilodemus’ quotations may go back to Apollodorus’ On the Gods inthe second century, which was an important source for Philodemus.6

The scholia Genevese on Iliad 21. 319 also report that Apollodorusclaimed an initial � (‘s’) in (�)åæÆ�� (‘(s)cherados’ ¼ ‘gravel’) to besuperfluous, ‰ �Ææ’ % ˇ�æøfi �c� ç�æ��Ø�� (‘as in the word “life-giving” (pheresbios) in Homer’): ç�æ��Ø� (‘life-giving’) does notoccur in the Iliad or Odyssey, but several times in the Hymns (Apoll.341,Dem. 450, etc.) and once inHesiod (Th. 693).7 If this does go backto Apollodorus, it assumes that he did not distinguish between the

4 This poem has long been thought later than the rest of the collection and assignedby some to the Orphic corpus; see AHS 384–5, Càssola 297–9. West (1970) proposesthat the Hymn is the work of Proclus, placed in the midst of the Homeric Hymnsthrough a late mishap of transmission; the Hymns are conveyed together with theCallimachean, Proclan, and Orphic hymns in the majority of manuscripts. Gelzer(1987), (1994), 125–9, instead suggests it is the work of an unknown earlier philoso-pher working in the tradition of Plotinus and that it was intentionally inserted in thecollection; the poem’s attribution to Proclus is rejected also by Devlin (1994), 338–42,Saffrey (1994), 75, Berg (2001), 6–7, the latter on the grounds that the concept ofepistrophe is absent in the poem but central to Proclus’ hymns. Others have suggestedthat Proclus compiled the collection; see Càssola lxv.

5 D. S. 1. 15. 7 (��B�ŁÆØ �b �B ˝��Å �e� ��ØÅ�c� K� ��E o��Ø [çÆ��], cit. Dion.A. 9–10), 4. 2. 4 (�e� �OÅæ�� b ��Ø Ææ�ıæB�ÆØ K� ��E o��Ø [çÆ��], cit. Dion. A.9–10), 3. 66. 3 (› ��ØÅ�c K� ��E o��Ø , cit. Dion. A. 2–10); Phld. Piet. p. 87 Schober(�OÅæ� �’ K� [��E o]��Ø , cit. Dem. 440), p. 93 Schober (o[�]�Ø[ �O]Åæ� , cit.Apoll. 91); see Boserup (1971), 109–11, Henrichs (1972), 72–7. Dion. A. 9–10 is alsoquoted by the scholiast on A. R. 2. 1211. Later quotations and references include Paus.1. 20. 3 (possible ref. to Dion.; see West in this volume, p. 42), 1. 38. 2–3 (ref. Dem.154–5), 4. 30. 4 (cit. Dem. 417–20), 10. 37. 4 (ref. Apoll. 269), Ath. 22b (cit. Apoll. 514–6), 653b (possible cit. Dion. B; see West [2001a], 8), and Aristid. Orat. 34. 35 Keil (cit.Apoll. 169–71). Further testimonia are listed in AHS lxxii–lxxviii.

6 See Henrichs (1975a).7 Cf. AHS lxxiii–lxxiv, Richardson (1974), 68. Without indication of its prove-

nance, the word ç�æ��Ø� (‘life-giving’) is connected to a comment of Aristarchus inthe T scholia on Il. 16. 163 ��æØ��ÆØ {�b ªÆ��æ}: �æ��Ææå� I��d ��F‘��æØ��ÆØ’ �Øa �c� �ºÅ��c� ‰ ç�æ�Ø� ç�æ��Ø� (‘the belly is glutted (periste-netai): Aristarchus [reads this] in place of “is distended” (periteinetai) because of theabundance [implied by � (‘s’)], as in pherebios/pheresbios’).

176 Andrew Faulkner

Homeric epics and the Hymns in the phrase �Ææ’ % ˇ�æøfi (‘inHomer’). Opinion about the Hymns’ authorship in ancient sourcesis, as one would expect, inconsistent. An anonymous Life of Homerstates explicitly that theHymns are not byHomer, just as other sourcesspeak of the Hymns as ‘attributed to Homer’ (I�Æç�æ��Ø �N �OÅæ��).8 As well, the scholiast on Pindar Nemean 2 assigns theHymn to Apollo to Cynaethus.9 Other sources, however, speak ofHomer as the author of the Hymns, as did Thucydides when quotingthe Hymn to Apollo.10 The scholiast on Aristophanes’ Birds 575 statesthat, while some scholars claim that the line (in which Iris is said to flylike a timid pigeon) is a playful distortion of Iliad 5. 778, where Heraand Athena are compared to timid pigeons, others claim that the ideaoccurs elsewhere in Homer, ‘because the hymns are also by him’ (�N�dªaæ ÆP��F ŒÆd o��Ø): Iris and Eileithyia are compared to timid pigeonsat Apollo 114.11 At the very least, the Hymns do not seem to have beenset apart from the Homeric epics in the early Hellenistic period asexplicitly as were the hexameter poems of the Epic Cycle, whichAristarchus deemed to have been written by inferior ‘younger’(��æ�Ø) poets.12

In contrast to the view Aristarchus took of the Epic Cycle, there isno certain evidence that Alexandrian scholars from Zenodotus toAristarchus in the third century bc concerned themselves with the

Hymns. Allen, Halliday, and Sikes note that the Homeric scholia on anumber of occasions fail to refer to relevant passages in the Hymns

8 Vita V 20–3 Allen Iººa ��f o��ı ŒÆd �a º�Ø�a �H� �N ÆP�e� I�Æç�æ��ø��ØÅ��ø� �ªÅ�� Iºº��æØÆ ŒÆd �B ç��ø ŒÆd �B �ı��ø ��ŒÆ (‘but the hymnsand the remainder of the poems attributed to him should be thought the work ofothers because of both their nature and their force’), Ar. fr. 56. 26–7 (¼P. Oxy. 2737,second century ad: commentary on Anagyros), sch. Nic. Alex. 130.

9 Sch. Pi. N. 2, 1c, III 29, 9–18 Dr (¼ FGrH 568 F 5), q� �b › ˚��ÆØŁ� �e ª�� E� , n ŒÆd �H� K�ØªæÆç��ø� % ˇ�æ�ı ��ØÅ��ø� �e� �N ��ººø�Æ ª�ªæÆçg o�� I�Æ�Ł�ØŒ�� ÆP�fiH (‘Cynaethus was a Chian by birth, who, of the poems ascribedto Homer, wrote the hymn to Apollo and attributed it to Homer’). See in this volumeChappell (pp. 71–2), Nagy (pp. 288–91).

10 Th. 3. 104 �Åº�E �b �ºØ��Æ �OÅæ� ‹�Ø ��ØÆF�Æ q� K� ��E ��Ø ��E��, – K��Ø�KŒ �æ��Ø��ı ��ººø�� (‘Homer makes this very clear by these verses from the hymnto Apollo’); slightly variant versions of lines 146–50 and 165–72 are quoted. For othersources see above n. 5. On ‘Homer’ used of epic poetry other than the Iliad andOdyssey in the Classical period and earlier, see Burgess (2001), 129–30.

11 Cf. Dunbar (1995), 386.12 See Severyns (1928), 31–61, Burgess (2001), 9.

The Collection of Homeric Hymns 177

when discussing words not known in ‘Homer’.13 This could, as theysuggest, be the result of Hellenistic scholars excluding the Hymnsfrom the canon of authentic ‘Homeric’ texts, but it is neverthelessstriking that the Hymns are not mentioned at all in these contexts.One might consider, for example, that line 51 of Hermes was quotedby Antigonus of Carystus in the third century.14 In P. Oxy. 2737, afragment of a second-century commentary on what is very probablyAristophanes’ Anagyros (fr. 56. 19–27), the commentator remarksthat the transmitted lemma Œ�Œ�� e ��æ�ªø� ��Ø�� (‘the swanaccompanied by its wings such . . . ’) is drawn from the five-line Hymn21. 1+�E��, �b b� ŒÆd Œ�Œ�� e ��æ�ªø� º�ª’ I��Ø (‘Phoebus, theswan also sings of you with a clear voice accompanied by its wings’).Immediately preceding this statement, the commentator reports thatAristarchus thought the lemma to imitate Terpander and that Eu-phorion thought it to derive from Ion. These reported observationsseem to relate to the beginning of the parabatic ode of the Anagyros,which may have been included in a more extensive lemma, of whichonly the end is transmitted in the later commentary.15 If this is thecase, the later commentator’s corrective statement about the source ofthe transmitted lemma need not imply that Aristarchus and Euphor-ion originally made no comment on the similarity of Hymn 21, butthis is at least a strong possibility. The ancient testimonia for theshortest Hymns are even more limited: the scholiast on PindarPythian 3. 14 cites the opening verses of the five-line Hymn 16 asamongst the ‘Homeric hymns’ (K� ��E % ˇÅæØŒ�E o��Ø ), but there isotherwise no quotation of a short Hymn that might be traced back tothe Hellenistic period.16

It is nonetheless hard to believe that the longer Homeric Hymnsescaped the learned activities of the third century bc, when the operaof earlier Greek poets were assembled, arranged, and studied to a

13 AHS lxxix–lxxx. S. West (1967), 32–5, points out that Apoll. may have influ-enced the variant text of Il. 1. 484–94 attested in a Ptolemaic papyrus (P.S.I. 1454; cf.Apoll. 503–7), as well as the text of Od. 15. 295 quoted by Strabo (8. 3. 26, 10. 1. 9) butnot transmitted by the mss. (cf. Apoll. 425).

14 See Vergados (2007), Thomas (2009), 304.15 See Luppe (1973), 281–5, Montana (2006), 169–73, (2009), 43–4, with further

bibliography.16 Cf. Richardson (2010), 3. P. Oxy. 4667, datable to the third century ad, contains

fragments of Hys. 18 and 7, in that order. Although see below p. 197 on linesequivalent to Hy. 18 on a fifth-century Attic lekythos.

178 Andrew Faulkner

greater extent than before. Ordered editions were produced of thelyric poets, in which poems were organized into books according tometre or genre, and within books according to alphabetization anddescending order of importance.17 Callimachus compiled his prodi-gious Pinakes, a structured bibliography of Greek literature in ahundred and twenty books,18 as well as a special chronological tableof the dramatic poets.19 Homeric book-division probably originatedprior to Alexandrian scholarship,20 but the so-called Epic Cycleappears to have been assembled in the Hellenistic period.21 Signs ofeditorial activity on the Hymns are preserved in the medieval manu-scripts at Apollo 136–9.22

There is some evidence that a collection of Homeric Hymns existedin the time of Callimachus. Prominent amongst several possibleallusions to earlier texts in the opening of his hymn to Zeus is theapparent modelling of lines 4–8 on the beginning of the first HomericHymn to Dionysus: apart from the shared hexameter verse and epicKunstsprache, both passages contain the motif of the god’s rivalbirthplaces.23 The hymn to Zeus comes first in the sequence ofCallimachus’ hymns, which does not vary in the extant manuscripts,and it is probable that the transmitted order bears Callimachus’authorial stamp.24 We can therefore conjecture that, in alluding at

17 On the organizational principles used in ancient editions of Greek lyric poets,see Rutherford (2001), 158–62, Pfeiffer (1968), 182–8. Pindar’s K�Ø��ŒØÆ were of coursegrouped by the location of the victory. Alphabetical ordering is also apparent inCallimachus’ Pinakes and was surely a widespread organizational principle; seePfeiffer (1968), 128–30. On Hellenistic books, see also Hutchinson (2008).

18 Frs. 429–453 Pf. On the Pinakes, see further Schmidt (1922), Blum (1977).19 Frs. 454–6 Pf. See Pfeiffer (1968), 131–2. On the collection and study of tragedy

and comedy in the Hellenistic period, see Olson (2007), 26–9, Pfeiffer (1968), 119–20,159–62, 188–96.

20 See Skafte Jensen et al. (1999) for a survey of scholarly opinion on the issue.Some maintain that Homeric book-division was imposed by the Alexandrians, but thedivision into ÞÆłøfi ��ÆØ at least suggests that this goes back to rhapsodic performance;cf. M. L. and S. West in Skafte Jensen et al. (1999), 68–73, Burgess (2001), 31.

21 See Burgess (2001), 7–46, esp. 15–19, with further bibliography.22 See West WL 20–1.23 Call. Hy. 1. 4–8 � Dion. 2–6; cf. West in this volume (pp. 40–1). The first three

lines of the poem call to mind Pi. fr. 89a SnM, a prosodion to Artemis, andCallimachusmay also have intended an allusion to Pindar’s Hymn to Zeus (fr. 29 SnM), whichsimilarly contains the motif of questioning what to sing. On these and other allusions,seeMcLennan (1977), 30, passim, Fuhrer (1988), 60–1,Haslam (1993), 115,Hunter andFuhrer (2002), 171, Fantuzzi and Hunter (2004), 371, Depew (2004), 119.

24 Cf. Hopkinson (1984), 13–17, Bing (1988), 126 n. 57, Cameron (1995), 255, 438,Hunter and Fuhrer (2002).


the beginning of his first hymn to the opening of Dionysus, the first ofthe Homeric Hymns in the Leiden manuscript, Callimachus knew anordered collection in which Dionysus came first.25

It is perhaps also significant that Callimachus’ sixth hymn to De-meter has thematic and structural links to the fifty-nine-line Hymn toDionysus transmitted seventh in our collection: the narratives of bothpoems involve transgression by arrogant mortals against a divinitywho makes an epiphany and metes out punishment; Dionysus is alsonamed and openly compared to Demeter in lines 70–1 as a god whogrows angry for the same reasons as the goddess. As well, whileDionysus’ epiphany in Hymn 7 is accompanied by plenty in theform of wine, fruit, and abundant vegetation (35–42), Demeter in-versely places an unquenchable need for plenty on Erysichthon, whoseappetite is his downfall.26 If Callimachus intended a thematic allusionto this shorterHymn toDionysus in the final hymn in his collection, itcould be seen to echo the allusion to the longer Hymn to Dionysus atthe outset of his collection.27 Without positing one-to-one correspon-dence between each of Callimachus’ hymns and the Homeric Hymns,one might even speculate further about the significance of the allusionfor the early ordering of theHymns. With the exception of the twenty-one-line sixth Hymn to Aphrodite, the first seven Hymns range fromfifty-nine to five hundred and eighty lines and contain developednarratives.28 The shorter sixth Hymn to Aphrodite seems to have beenplaced after the longer Hymn to the same goddess because of similar

25 Cf. Hunter and Fuhrer (2002), 169–3, Fantuzzi and Hunter (2004), 371, Depew(2004), 119, who also point out that the collected Hellenistic edition of Pindar’s worksmay have begun with his hymn to Zeus, to which Callimachus seems to make allusionhere as well (see above n. 23). If this was the case, it would corroborate the idea thatCallimachus is thinking of hymns that begin collections. The order of Pindar’scollected works varies in surviving testimonia: only the Vita Ambrosiana places thehymns first, while P. Oxy. 2438 and Hor. Od. 4. 2 place the dithyrambs first; seefurther Race (1987), Rutherford (2001), 146–8. Nonetheless, Pindar’s hymn to Zeuswill have come first in the book of his hymns (cf. scholia ad Ps.-Luc. Dem. Enc. 19).

26 See Bulloch (1977), 99–101, who discounts specific allusion by Callimachus. Henotes only small verbal similarities between the two hymns (Hy. 7. 42–44 � Call. Hy.6. 59–61, Hy. 7. 50 � Call. Hy. 6. 60).

27 Cf. Depew (2004), 134 on the link to the first hymn suggested by the relation ofDemeter�Isis and Dionysus�Osiris.

28 On the original length of Dion., which can be estimated at over four hundredlines, see West (2001a), 1.

180 Andrew Faulkner

content, but its scale is out of place at this point in the collection,29 andone might entertain the possibility that it was inserted here in a laterarrangement. If this were the case, an early collection of Hymns couldhave begun with the six narrative Hymns to Dionysus, Demeter,Apollo, Hermes, Aphrodite, and Dionysus, in that order. Could Cal-limachus be alluding in his first and sixth hymns to the first and sixthHymns to Dionysus in the collection he knew?

Speculation aside, there can be no doubt that many of the HomericHymns were well-known to early Hellenistic poets. In what follows, Iwill survey the evidence for the transmission and reception of theHymns from the late seventh to the early third centuries bc. Having

just discussed Callimachus, I will first consider the impact of the

Hymns on Hellenistic poetry of the first half of the third century bc,when evidence for the Hymns’ reception is at its greatest. I will thenmove backwards in time to explore the Hymns’ influence from thefourth to the seventh centuries bc. Signs of the Hymns’ impression inthese earlier centuries are fewer than in the Hellenistic period, but apicture of their circulation and impact nonetheless emerges from thesurviving evidence.

As might be expected, Callimachus himself makes extensive verbaland thematic allusion to the Homeric Hymns throughout his collec-tion of hexameter hymns, which both announces his debt to thepoetic tradition and informs his innovation upon these models.Callimachus’ second hymn to Apollo, for example, alludes themat-ically to both of the central narrative components of the Hymn toApollo. First, the god’s birth on Delos, which dominates the Deliansection of Apollo, is obliquely referred to at the outset of the poem:

K��ı�� › ˜�ºØ� �� Ø ç�E�Ø�

K�Æ��Å , › �b Œ�Œ�� K� MæØ ŒÆºe� I��Ø (Call. Hy. 2. 4–5)

The Delian palm tree suddenly nods sweetly,and the swan sings beautifully in the air.

The mention of this Delian palm tree recalls for a learned reader itsrole in the god’s birth in Apollo, in which Leto is said to ‘cast both of

29 See Torres-Guerra (2003), 5, who considers Hy. 6 to be of a different class thanHy. 5, mid-length rather than long, and groups Hy. 7 with the long Hymns because ofits developed narrative.


her arms around a palm tree’ (Içd �b ç��ØŒØ ��º� ��å��, 117) justbefore delivering her son on Delos;30 an allusion which contains ironygiven that Leto’s grasping of the now sweetly nodding tree comes inApollo at the end of an arduous nine days and nine nights of labour(90–1). While mention of the palm tree precedes the god’s birth inApollo, it here announces his epiphany to the celebrants of his cult,signalled explicitly in the opening lines of the poem; in part in thedescription of the palm itself by the unusual application to a tree ofthe verb K�Ø��ø, which is normally used of a divinity giving assent.31

The palm tree is also a detail to which Callimachus returns in hisfourth hymn to Delos, where he does narrate the birth of Apollo. Atlines 209–11, Leto is said to lean back against the trunk of the palmtree with her shoulders, rather than grasp the tree with her arms asshe did in the Homeric Hymn:

º��Æ�� b Ç��Å�, I�e �’ KŒº�ŁÅ ��ÆºØ� þ�Ø ç��ØŒ� ��d �æ�� IÅåÆ��Å ��e ºıªæB

��Øæ��Å· ��Ø� �b �Øa åæ�e �ææ�� ƒ�æ� .

And she loosened her girdle, and leaned back with her shouldersagainst the trunk of a palm tree, distressed by terrible hardship,and a moist sweat flowed over her skin.

It has been suggested that Callimachus’ departure from the earliermodel in this case is meant to reflect his knowledge of Herophilus’medical innovations concerning the best posture for giving birth.32

Whether or not this is behind the change, the passage demonstratesCallimachus’ interest in this feature of Apollo’s birth and strengthensthe case that the image of the palm tree at the outset of his secondhymn is intended to call to mind the god’s nativity, which is replacedin the poem by his cultic epiphany.33

30 See Hunter and Fuhrer (2002), 150, who also point out that line 5, › �b Œ�Œ�� K�MæØ ŒÆºe� I��Ø (‘and the swan sings beautifully in the air’), provides a link to theswans who sing at Apollo’s birth in Call. Hy. 4. 49–55; it is perhaps also intended torecallHy. 21. 1 (cf. above p. 178).Cf. Leto’s grasping of the palm tree also at Thgn. 5–10.

31 See Williams (1978), 19.32 Most (1981).33 This parallelism is perhaps strengthened by the fact that the goddesses attending

Leto’s birth give a ritual shout upon his leap into life at Apoll. 119 KŒ �’ �Ł�æ� �æeç�ø��, Ł�Æd �’ Oº�ºı�Æ� –�Æ�ÆØ (‘He jumped out into the light, and all the goddessesshouted’), as upon his birth at Call. Hy. 4. 258 (a possible echo of Aphr. 19); see furtherFantuzzi and Hunter (2004), 367, Faulkner (2008), 96–7.

182 Andrew Faulkner

Then near the end of his second hymn Callimachus refers to themain narrative thread of the Pythian section of the Hymn to Apollo,the god’s slaying of the dragon Pytho which leads to the foundation ofhis cult in Delphi:

ƒc ƒc �ÆØB�� IŒ��, �o��ŒÆ ��F��˜�ºç� ��Ø �æ��Ø�� Kç��Ø�� oæ�� ºÆ� ,q� �ŒÅ��º�Å� åæı�ø� K��Œ�ı�� ø�.—ıŁ� ��Ø ŒÆ�Ø��Ø �ı�� ÆØ��Ø� Ł�æ,ÆN�e ZçØ . �e� b� �f ŒÆ��Ææ� ¼ºº�� K�’ ¼ººøfi��ººø� TŒf� Oœ��, K�Å��Å�� b ºÆ� ,‘ƒc ƒc �ÆØB��, ¥�Ø �º� .’ �PŁ� �� Åæª��Æ�’ I��Å�BæÆ, �e �’ K��Ø Œ�EŁ�� I��fi Å (Call. Hy. 2. 97–104)

We hear ie ie paieon, because the Delphians firstinvented this song of praise, when you gave proofof the far-shooting of your golden bow.A god-sent beast, a terrible dragon, met with youdescending to Delphi. You slew it, firing swift arrow uponswift arrow, and the people cried out ‘ie ie paieon, shoot an arrow.’Your mother bore you to be a helper straightaway,and you are still sung as such because of this deed.

The short description of the slaying of the dragon here, which endswith another passing reference to Apollo’s birth, is subordinate to theaim of providing an aition for the refrain ie ie paieon, explained bythe similarity of ƒ� (‘hie’) to ¥�Ø (‘shoot’ [hei]) and �ÆØB�� (‘paieon’) toboth �ÆE (‘child’ [pai]) and N�� (‘arrow’ [ion]).34 This acts as anallusive alternative to the aition for the place name Pytho and Apol-lo’s cult title Pythios given in the Hymn to Apollo after the narrationof the same event (363–74): the dragon is there said to rot in the sunafter its death, with forms of the verb ��Ł�Ø� (‘to rot’ [pythein]) usedthree times in twelve lines.35 The etymologies link the two texts, butwhile the much lengthier narrative of the dragon’s death in the

34 See Williams (1978), 85. In Apoll. 514–19 the Cretan sailors are said to sing therefrain on their way to Delphi, led by Apollo, in the manner of the paeans of theCretans inspired by the Muses.

35 On this etymology in later sources see AHS 251–2. Callimachus’ reworking ofthe etymology given in Apoll. is briefly noted by Haslam (1993), 117. Richardson(2010), 132 intriguingly suggests that Apoll. 357 Ne� KçBŒ�� (‘he shot his arrow’)alludes to the etymology of the paean refrain; Callimachus could well have beenthinking of this. On the erudite nature of Callimachean aitia, in contrast to those inthe Homeric Hymns, see Vamvouri-Ruffy (2004), 235.


Homeric Hymn (300–74) is an important phase in Apollo’s founda-tion of his sanctuary at Delphi, the focus of Callimachus’ hymn lieselsewhere. Apollo’s actions explain the origins of the ritual refrain ƒcƒc (‘hie hie’), which has already been heard a number of times in thepoem in different combinations: lines 21 ƒc �ÆØB�� ƒc �ÆØB�� (‘hiepaieon hie paieon’), 25 ƒc ƒc çŁªª��Ł� (‘shout hie hie’), and 80 ƒc ƒc˚Ææ��E� (‘hie hie Carneius’). The last of these occurs within thelengthy section of the poem dedicated to the foundation of Cyrene(65–96), which provides the first palpable indication of the location ofthe ritual performance described in general terms at the outset of thepoem.36 The recollection of the ritual cry strengthens the comparisonbetween these previous passages,37 and reminds the reader that it isCyrene, not Delphi as in the Homeric Hymn, whose foundationis treated most prominently in the poem. In fact, the Cyrene episodeis flanked by shorter narrative accounts, both aetiological,38 related tothe foundation of Apollo’s two most important cult centres treated inthe Homeric Hymn to Apollo: Delos (55–64, his foundation of theHorned Altar together with Artemis), and Delphi (97–104, the slay-ing of the dragon).39 In the first of these foundation myths, therepetition of the word Ł��ºØÆ (‘foundations’, 57, 58, 64) alsoseems intended to recall Apollo 254/294 �ØŁÅŒ� Ł��ºØÆ +�E�� ººø� (‘Phoebus Apollo laid his foundations’).40

In this way, Callimachus recalls the central narrative elements ofthe Homeric Hymn, only to reject them as his principal subjectmatter. Nowhere in his collection of hymns does he narrate in full

36 On the location of the imagined festival, see Bing (1993), 191. The mention ofthe Delian palm in line 4 at first suggests Delos as the location of Apollo’s epiphany,but the point may be to mislead: cf. Williams (1978), 19.

37 See Calame (1993), 45–6, Cheshire (2005), 344–6.38 An aition for Apollo’s title Nomios, explained by his having tended Admetus’

horses (47–9) with an implied link to the verb ��Ø� (‘to pasture’[nemein]), alsoprefigures the Cyrene episode: see Calame (1993), 41, Depew (1993), 74–5.

39 This tripartite division of lines 55–104 is marked by indentation in Pfeiffer’s textand is well-established. Cheshire (2005) proposes a tripartite division of the section onthematic grounds of foundation, celebration, and reciprocation. Cf. Calame (1993),50, who, as part of his analysis of Apollo’s role in construction in the hymn as aprogression towards civilized space, notes that Cyrene ‘is situated between the twogeographical and religious poles of the cult dedicated to Apollo: Delos where the god’sepiphany seems to take place (1–8) and Delphi where his civic function is realised(32–41). These two religious centres are also the two poles of Greek civilization, on themodel given by the Homeric Hymn to Apollo in its two sections!’

40 Cf. Williams (1978), 57, Depew (2004), 122.

184 Andrew Faulkner

the killing of the dragon, a traditionally popular story,41 or thefoundation of Delphi, and when he does narrate the birth of Apolloit is unexpectedly in a hymn dedicated to Delos, and there treatedin an innovative way.42 The Hymn to Apollo was not Callimachus’only model. Callimachus himself treated the myth of Apollo’sslaying of the dragon in his Aetia (fr. 88 Pf.) and may be referringhere to his own poetry,43 while Pindar offered a poetic source forthe foundation of Cyrene.44 The pronounced focus on ritual per-formance in the poem, from the opening scene of epiphany at thegod’s temple to the repetition of the ritual cry, is also reminiscentof cultic hymns such as Aristonous’ fourth-century paean toApollo.45 Even, however, within this context of cultic performance,the Homeric Hymn to Apollo remains an important intertext. Atthe end of the Delian section (156–64), the poet describes a choralperformance of Delian maidens, who know how to imitate per-fectly the voices of others:

��ø� �’ I�Łæ��ø� çø�a ŒÆd �Æ�ÆºØÆ��Ø�E�Ł’ Y�Æ�Ø� çÆ�Å � Œ�� ÆP�e �ŒÆ�� çŁªª��Ł’·�o�ø �çØ� ŒÆºc �ı��æÅæ�� I�Ø��. (Apoll. 162–4)

They know well the voices and the chatter ofall people. Each man would say that he himselfis speaking; their beautiful song is so well fitted.

Bing insightfully proposes that Callimachus’ treatment of the poeticvoice in his second hymn, which blends the voice of choral perform-ance with the voice of the poet-narrator, and ultimately the voices

41 See Vamvouri-Ruffy (2004), 207 n. 161. Brief mention of Delphi and the dragonis also made in Call. Hy. 4. 90–3.

42 See Bing (1988), 91–143: the hymn to Delos is profoundly influenced by Apoll.,but at the same time departs significantly from it; cf. below pp. 199–200 on Pindar’sinspiration for Delos as subject.

43 A similar treatment of the story is given by Apollonius Rhodius (2. 703–13), whomay also be imitating the version in the Aetia. See Pfeiffer ad fr. 88, Williams (1978),82, Albis (1996), 123–5 with further bibliography.

44 SeeCalame (1993), 38–43,Depew (1993), 74–5, Vamvouri-Ruffy (2004), 117–18.45 See text and commentary of this hymn in Furley and Bremer (2001) ii. 45–52,

who note on line 13 çæØŒ�� (‘awe-inspiring’) that, ‘Exactly this mood of excitedawe at the supernatural presence of divinity is invoked by Callimachus.’ The shakinglaurel in lines 10–11 is paralleled at Call. Hy. 2. 1; cf. Williams (1978), 16. See alsoVamvouri-Ruffy (2004), 94–6, 211–15, on Aristonous, and passim on Callimachus’poetic engagement with the form of inscribed cultic hymns.


of those enacting the ritual performance through reading the text,may be looking back to this description of the Delian maidens’remarkable ability to fit their song to the voices of others.46 Thisinterpretation is perhaps strengthened by the fact that, immediatelyfollowing the account of the Delian maidens in the Homeric Hymn,the poet-narrator addresses the chorus directly (165–78): identifyinghimself as a blind man from Chios, he addresses the chorus as a group‘you’ (��E , 171), then switches to the first-person plural to speak ofhow ‘we’ (��E , 174) will carry ‘your’ (��æ��) fame far and wide,before switching once again back to the first-person singular ‘I’ (Kª��,177) to tell of his ceaseless hymning of Apollo. The first-person pluralhere most probably refers to the singer himself (with the plural usedfor the singular) or other rhapsodes.47 But there is nonetheless animplied blending of the singer’s voice with that of the chorus, as bothcelebrate Apollo hymnically.48 In Callimachus’ hymn, the mixing ofvoices throughout the poem explicitly involves movement betweenthe first-person singular and plural. The first-person plural in line 11incorporates both the chorus and the poet-narrator at the outset ofthe poem: Oł��ŁÆ, t % EŒ��æª�, ŒÆd K��Ł’ �h�� ºØ�� (‘We willsee you, O far-shooter, and we will never be lowly’). There is then aswitch to the first-person plural at the beginning of the dragon-slaying episode: ‘we’, which includes Apollo himself, are said to listento the choral refrain ie ie paieon (ƒÅ ƒc �ÆØB�� IŒ��, 97).49 Byincorporating the audience into his complex blend of voices with thisfinal use of the first-person plural in the poem, Callimachus bothrecalls and goes beyond the mixing of voices in the Delian section ofthe Homeric Hymn.

46 Bing (1993), esp. 189. At Apoll. 162 �Æ�ÆºØÆ�� (‘chatter’) appears in only afew mss. in place of the majority reading Œæ��ÆºØÆ��, but fits well with thesubsequent claim that anyone would say he himself is speaking. Bing (196) takesŒæ��ÆºØÆ�� metaphorically also to mean ‘chatter’, but the literal sound ofŒæ�ÆºÆ (musical instruments, perhaps similar to Œæ��ÆºÆ ‘castanets’) could alsowork in the context of a musical chorus. See further the helpful discussion ofRichardson (2010), 107–8, who reads Œæ��ÆºØÆ�� on the grounds that it is thelectio difficilior and that ‘it is hard to see how a word which would most naturallydenote inarticulate or confused sounds can be used in this context’. In either case, themimetic abilities of the Delian Maidens are explicit. Cf. Peponi (2009).

47 See Richardson (2010), 110–11, Aloni (1989), 127–8.48 Cf. Calame in this volume (pp. 351–2).49 See Calame (1993), 47–9.

186 Andrew Faulkner

The explicit identification with the audience at this point also leadsinto the final section of Callimachus’ hymn (105–13),50 a dialoguebetween Apollo and Envy which has long been recognized to reflectupon poetic aesthetic. Envy tells Apollo that he does not honour ‘thesinger who refuses to sing as much as the sea’ (�PŒ ¼ªÆÆØ �e� I�Ø�e�n �P� ‹�Æ �� I��Ø, 106),51 after which Apollo kicks Envy andexplains that the great stream (ªÆ Þ�� , 108) of the Assyrian rivercarries a lot of filth, while bees bring water to Demeter only from thesmall, pure stream of a holy fountain (ŒÆŁÆæ� �� ŒÆd Iåæ�Æ�� . . . j��ÆŒ� K� ƒ�æB Oº�ªÅ ºØ�� , 111–12). The bees (ºØ��ÆØ) functionhere on a number of levels: bees are animals connected to industryand purity, but ‘bee’ is also used metaphorically of both poets and thepriestesses of Demeter.52 Understood as operating together, these twometaphorical uses of ‘bee’ once again identify the speaker (as poet)with the ritual celebrants (the priestesses of Demeter),53 both ofwhomApollo, as divine audience, praises in this concluding section.54

Apollo’s judgment also recalls his praise of the chorus for its singingearlier in the poem (28–9). Much ink has been spilt over the questionof whether the sea (�� , 106) in this final passage should be takento refer to Homeric epic in contrast to the refined, small stream(Oº�ªÅ ºØ�� , 112) which is equated with Callimachus’ poetry.55

Given the thematic allusions and responses that lead into and arerecalled in this final section, it seems most natural to link the image of

50 Cf. Williams (1978), 82, Bassi (1989), esp. 227, who argue that the dragon-slaying episode provides a bridge to the final section; the latter demonstrates thatApollo’s combative role throughout the hymn is linked to his confrontation withEnvy. See also Calame (1993), 50–5 on the links between the epilogue and the Cyreneepisode.

51 I follow Köhnken (1981), 416, and Cameron (1995), 405, in taking �P�’ ‹�Æhere emphatically.

52 See Williams (1978), 92–4. The identification of the bees with Demeter’s priest-esses has been disputed, but there are good grounds for accepting it; see Petrovic(2011c forthcoming). On the possible relationship of this imagery to Philitas’ grove,see Heyworth (2004), 150–1.

53 For this identification, cf. Petrovic (2011c forthcoming), Bing (1993), 192–4.54 Cf. Hunter and Fuhrer (2002), 151–2, who suggest that Apollo’s approval of

Callimachus’ hymn here imitates the poet’s request at Apoll. 165–77 to be remem-bered as the most pleasing of singers (I�cæ X�Ø�� I�Ø�H�, 169).

55 The generic identification of the sea with Homer and qualitative criticism of epicpoetry proposed by Williams (1978), 85–9, is rightly dismissed by Köhnken (1981),Cameron (1995), 403–7. For a list of bibliography on the topic, see Vamvouri-Ruffy(2004), 65–6 nn. 59–62.


the sea specifically to the Hymn to Apollo, rather than generically toHomeric epic.56

An even more pointed departure from the narrative content of oneof the long Homeric Hymns is found in Callimachus’ sixth hymn toDemeter. Like his hymn to Apollo, this poem also opens with theevocation of a cultic setting, in which the voice of the poet-narrator isblended with that of the ritual celebrants, in this case the fasting femalepriestesses of Demeter.57 When the poet-narrator first embarks uponthe narrative, it is to tell the story of Demeter’s search for Persephone(6–16): like the celebrants in the ritual frame, Demeter does not eat,drink, or bathe in her search for Persephone, wandering the earth andsitting by her well in grief until she is persuaded by Hesperus to endher fast. In its broad outline, this is parallel to the narrative recountedin the Homeric Hymn to Demeter, in which Demeter searches for herdaughter and in her anger causes a universal famine for mankind.There are differences of detail (Callimachus has Hesperus convinceDemeter to break her fast, whereas in the Hymn it is Iambe whofunctions in this role),58 but there are structural parallels in Callima-chus’ more condensed account that seem intentionally to recall theHomeric Hymn: as in Demeter (49–50, 200–1), Demeter’s fasting andabstinence from bathing is twice described by Callimachus, first dur-ing the search for her daughter and then when she is seated in grief atthe Callichoron well (11, 15–16).59 The fact that Demeter sits by theCallichoron in line 16 also recalls theHomeric Hymn, in which she sitsby the same well, there called the Parthenion,60 when she arrives at

56 See Cameron (1995), 406:

Having at lines 30–31 announced that his hymn will last for several days, soinexhaustible a theme is Apollo, Callimachus then cuts it short after barely 100lines. At 546 lines, the Homeric hymn to Apollo is five times as long. It is naturalthat Callimachus should make some comment on this conspicuous failure tofulfill not only traditional expectations but his own promise.

Williams (1978), 38 also takes �P�’ . . . Kç’ £� �� qÆæ in Call. Hy. 2. 30 to mean that‘the supposed festival will last several days’, although N. Richardson points out to methat it could refer more generally to hymns sung to Apollo on different occasions. Cf.Traill (1998), who proposes that Envy rejects even Apoll. as a hymn of suitable lengthfor Apollo.

57 On the voice in this poem, see Hopkinson (1984), 3–4, Bing (1995), 37–42.58 See Richardson (1974), 215, Hopkinson (1984), 87, Kledt (2004), 64–5.59 See Bing (1995), 30–1.60 On the identification of these two wells, see Richardson (1974), 326–8.

188 Andrew Faulkner

Eleusis in disguise as an old woman (98–104). Apart from the popu-larity of this myth in antiquity,61 Demeter’s visit to Eleusis in theHymn (91–304), the details of which have been variously linked tocult practice,62 perhaps further suggested to Callimachus the closeassociation of the myth of Persephone’s abduction with the ritualframe he imposes on the hymn.

The impression, however, that Callimachus will go on to narrate infull Demeter’s search for Persephone is short lived. Immediately afterthe second mention of her fasting, the poet-narrator makes an un-expected turn:

c c �ÆF�Æ ºªø� L ��Œæı�� ¼ªÆª� ˜Å�E·Œ�ººØ��, ‰ ��º��Ø� �Æ��Æ �ŁØÆ �HŒ�,Œ�ººØ��, ‰ ŒÆº�Æ� �� ŒÆd ƒ�æa �æ�ªÆ�Æ �æ��ÆI��Æå�ø� I�Œ�ł� ŒÆd K� ��Æ ~�Œ� �Æ�B�ÆØ,±��ŒÆ 'æØ��º�� IªÆŁa� K�Ø��Œ�� å�Æ�·Œ�ººØ��, ‰ (¥�Æ ŒÆ� �Ø ��æ�Æ��Æ IºÅ�ÆØ)�[ÆE�Æ ŒÆŒe� 'æØ��Æ �ŒØ��Ø�Æ ŁBŒ��] N��ŁÆØ.63 (Call. Hy. 6. 17–23)

Let us not speak of these things which brought tears to Demeter,better to tell of how she gave rightful laws to cities.better to tell of how she first cut straw and sacred handfulsof corn and put oxen to the task of threshing them,when Trioptolemus was first taught the good art.better to tell of how (in order that one might avoid transgression)[she made the wicked son of Triopas wraithlike].

Rather than recount the tale of Demeter’s search for her daughter, thepoet-narrator instead tells of how the goddess punishes the hubristicErysichthon for cutting down a tree in her sacred grove by giving himan unquenchable appetite. This rejection of the well-known Perse-phone narrative does not suppose the Homeric Hymn as its onlymodel. The choice of the alternative myth, with the metapoeticimage of the slim Erysichthon, perhaps has in mind (among othermodels) the treatment of both the Persephone and Erysichthonmyths in the Lyde of Antimachus, a figure at the heart of Hellenistic

61 See Richardson (1974), 74–86.62 See Richardson in the present volume (pp. 50–3).63 I here follow the supplement of Wilamowitz; see Hopkinson (1984), 99. Regard-

less of the exact wording, this line will certainly have introduced the punishment ofErysichthon recounted in the subsequent narrative.


debate about Callimachean poetic aesthetic.64 Nonetheless, the nar-rative of the Homeric Hymn remains an important point of referencefor the reader, as signalled by verbal reminiscences at the beginning ofthe Erysichthon tale: the divine tree which Erysichthon cuts is said toreach the sky (ªÆ ��æ�� ÆNŁæØ ŒFæ��, 37) in language similar tothe description of Demeter’s partial epiphany before Metaneira inDemeter, where her head is said to reach the roof-beam of the palace(�º�Łæ�ı j ŒFæ� Œ�æÅ, 188–9);65 the description of Demeter’s actualepiphany before Erysichthon then recalls these same lines in Demeterwhen her head is said to touch Olympus (Œ�çÆºa � �ƒ –łÆ�’!Oº��ø, 58).66 Moreover, the Erysichthon narrative seems to playupon a central element of the myth told in Demeter, namely thefamine which Demeter brings upon mankind in her anger: the sameword (ºØ� ) which in Demeter describes the famine (310–11) isused by Callimachus of the insatiable hunger given to Erysichthon(66–7).67 As in the case of Callimachus’ second hymn to Apollo, thethematic recollection, effected on this occasion through verbal inver-sion, highlights the departure from the earlier model.

Knowledge of both Hermes and Aphrodite is also reflected inCallimachus’ collection of hymns. Influence of Hermes is particularlyevident in the portrayal of precocious young divinities in his hymnsto Zeus and Artemis. In the former, some of the more prominentlinks are as follows:68 Zeus is said to accomplish in the evening thingshe thinks of in the morning (87–8), just as in Hermes the young god issaid to steal the cattle of Apollo in the evening after his birth in themorning (17–18); the placement of Zeus in a ‘golden cradle’ (º�Œ�øfiK�d åæı�øfi , 47) in a cave in Crete recalls Hermes’ repose in a ‘holycradle’ (ƒ�æfiH K�d º�Œ�øfi , 21) in Maia’s cave; Zeus establishes the eagleas the Iªª�ºØ��Å (‘messenger’, 67) of his signs, a word which is

64 See Faulkner (forthcoming a), with further bibliography.65 The intended echo is proposed by Bulloch (1977), 116–21, but treated with

caution by Hopkinson (1984), 113–14. The suggestion of partial epiphany beforeDemeter reveals herself fully as a goddess at lines 57–8 (the same pattern as in Dem.)at least makes the allusion a strong possibility. The language is also used of theepiphany of Aphrodite at Aphr. 173–4; cf. Richardson (1974), 209–10, Faulkner(2008), 238–42.

66 Cf. Hopkinson (1984), 131. Richardson (1974), 69 n. 4, notes other verbalparallels between Dem. and Callimachus.

67 See Haslam (1993, 119 n. 14), Bing (1995, 32). The word is used of Erysichthon’shunger at Hes. fr. 43. 5–7.

68 See further Clauss (1986), 161–6, Thomas (2009), 300–1.

190 Andrew Faulkner

earlier attested only inHermes 296, where it is used of the young god’sportents of flatulence; and both poems contain the theme of rivalryand reconciliation—in Callimachus the claim that Zeus settled thedivision of rule with his brothers Hades and Poseidon through theirrecognition of his strength rather than by lot (55–6), and in Hermesthe god’s reconciliation with his brother Apollo at the end of thepoem (436–578).69 Similar thematic connections to Hermes are foundin Callimachus’ hymn to Artemis:70 like Hermes, the young Artemismoves with precocious speed after her birth to establish her place inthe cosmos, requesting �ØÆ� (‘honours’) from Zeus while sitting onhis knee at the outset of the poem (4–40), perhaps a humorousreworking of Zeus’ role as arbitrator in the struggle between Hermesand Apollo for �ØÆ� in the Hymn; the theme of sibling rivalry withApollo is also again prominent—Artemis proclaims at the outset ofher speech to Zeus that she wishes to be ‘widely celebrated, in orderthat Phoebus should not vie with me’ (��ºıø�ı�Å�, ¥�Æ � �Ø+�E�� Kæ�Çfi Å, 7), while a later scene of arrival on Olympus depictsApollo in a subordinate position welcoming her (140–69), an inver-sion of Apollo’s arrival on Olympus in Apollo (1–13, 186–206);71

Hermes also appears twice in the poem, once in the guise of a Cyclopsto scare disobedient daughters of Oceanus (66–71) and then laterpaired with Apollo to welcome Artemis on Olympus (142–4), inbetween which the description of Artemis’ hunting seems to echoHermes’ theft of the cattle in Hermes.72

Amidst these quite extensive allusions to Hermes and Apollo, theinfluence of Aphrodite on Callimachus’ hymn to Artemis is alsoperceptible. The mention of Artemis’ punishment of a city of unjustmen (Iºº� Ø� �N I��Œø� ��Æº� ��ºØ�, 122) may be a direct imitationof her stated concern for cities of the just in Aphrodite (�ØŒÆ�ø� ��

69 Clauss (1986) suggests that the theme of sibling rivalry and reconciliation inCallimachus’ hymn reflects upon the contemporary strife between Ptolemy Philadel-phus and his older step-brother Ceraunus.

70 See further Ambühl (2005), 288–95, Thomas (2009), 301–2.71 On Artemis’ rivalry with Apollo in Callimachus’ collection, see Hunter

and Fuhrer (2002), 161–4, Fain (2004), Plantinga (2004), 258–64, Ambühl (2005),275–84. On the influence of Apoll. on the Artemis hymn, see also Bornmann (1968),xvi–xvii, Bing and Uhrmeister (1994), 29–30.

72 As Thomas (2009), 302 remarks: ‘The two references to Hermes function asbook-ends to a section of narrative about Artemis’ early hunting whose generalstructure and some of whose details are conditioned by a sequence peculiar to Herm.’


��ºØ I��æH�, 20), a rare attribute of the goddess in literature.73

Another echo is found at the beginning of the poem. The first requestthat Artemis makes to her father is for her virginity (6), which, alongwith her other wishes, is granted by Zeus after she playfully tries topull upon his beard and thereby force him to nod assent:

S � �ÆE �N��F�Æ ª��Ø�� XŁ�º� �Æ�æe

–łÆ�ŁÆØ, ��ººa �b ��Å� K�Æ��Æ�� å�EæÆ ,åæØ ¥ �Æ łÆ��Ø�. �Æ�cæ �’ K��ı�� ª�º��Æ (Call. Hy. 3. 26–8)

Thus speaking, the young goddess wanted tograsp her father’s beard, but to no avail she frequently stretchedforth her hands to touch it. And her father, laughing, nodded assent.

This seems to draw upon a lyric hymn to Artemis attributed to eitherSappho or Alcaeus (Sapph. fr. 44a. 1–11 Voigt), but also the similardescription of Hestia swearing an oath of virginity to Zeus in Aphro-dite (25–32), which includes the detail of her ‘grasping the head of herfather aegis-bearing Zeus’ (±łÆ�Å Œ�çÆºB �Æ�æe ˜Øe ÆNªØ�å�Ø�,27).74 Recollection of Aphrodite in the context of female virginity, animportant theme in Callimachus’ collection,75 is also effected at theoutset of the fifth hymn to Athena. The masculine goddess is in lines19–22 directly contrasted with Aphrodite in not displaying vanity:

�h�’ K Oæ��åÆºŒ�� ª�ºÆ Ł�e �h�� $Ø�F�� º�ł�� Æ� K �ØÆçÆØ��Æ�·

�P�’ �HæÆ· ˚��æØ �b �ØÆıªÆ åÆºŒe� �º�E�Æ��ºº�ŒØ �a� ÆP�a� �d ��ŁÅŒ� Œ�Æ�. (Call. Hy. 5. 19–22)

The great goddess looked neither into a mirror, northe clear eddy of the Simois river;

nor did Hera. But Cypris, taking the translucent bronze,often readjusted the same lock of hair.

Despite the lack of verbal similarity, the negative contrast of the virginAthena at the outset of a hymn calls to mind the priamel at thebeginning of Aphrodite, in which the three virgins Athena, Artemis,and Hestia are presented as exceptions to Aphrodite’s otherwise

73 See Faulkner (2008), 98–101. In this role, Artemis resembles the virginal Dike inHes. Op. 256–62: see Erler (1987), Hunter and Fuhrer (2002), 182–3, Fantuzzi andHunter (2004), 353–55.

74 See Faulkner (2008), 110–12.75 See Depew (2004), 125–35.

192 Andrew Faulkner

universal ability to conquer others in love (7–33): Callimachus invertsthe comparison in a poem that will, like Aphrodite, narrate a poten-tially erotic encounter between a goddess and a mortal man.76

Callimachus’ deep and intricate knowledge of the longer narrativeHomeric Hymns,77 reflected in his innovative reworking of theirlanguage, content, and structure, was not unique amongst poetsof the early third century bc. Amidst a number of possible verbal

echoes,78 several passages in Apollonius’ Argonautica suggest a par-ticular connection with the Hymns. The first lines of the poemcontain a hymnic invocation of Apollo: Iæå�� , +�E��,�ÆºÆØª��ø� ŒºÆ çø�H� j ��ÆØ (‘Beginning from you, Phoebus,I will call to mind the glorious deeds of men of old’, 1–2). Theirlanguage resembles the statement made at the end of several Hymnsthat the singer will ‘call to mind’ (��ÆØ) another song now that hehas begun with the god (cf. Aphr., Hys. 9, 18, 31), and is particularlysimilar to the end of Hymn 32 to Selene: �� ’ Iæå�� ŒºÆ çø�H�j fi ¼��ÆØ �ØŁø� (‘Beginning from you, I will sing the glorious deedsof semi-divine men’, 18–19).79 But this hymnic invocation of Apolloat the outset of the poem, with the placement of ��ÆØ first in theline, is perhaps meant also to recall the first line of Apollo: ��ÆØ�P�b º�ŁøÆØ ��ººø�� Œ��Ø� (‘I will call to mind and notforget far-shooter Apollo’, 1).80 Apollonius delays the expected in-vocation of the Muses until line 22 of his epic,81 and, although��ÆØ here takes as a direct object ŒºÆ çø�H� (‘glorious deedsof men’), the subtle echo of the Hymn would highlight for an edu-cated audience the fact that he, unlike Homer in the Iliad, does notforget to remember Apollo in hymnic fashion at the outset of the

76 See Hunter (1992), 12. Cf. Faulkner (2010b) for the suggestion that the groupingof three hymns to the virgin goddesses Artemis, Delos (who is called the ‘hearth’,ƒ��Å, of islands in Call.Hy. 4. 325), and Athena looks back to this grouping of Athena,Artemis, and Hestia at the beginning of Aphr. There may also be an allusion to Aphr.264–72 at Call. Hy. 4. 82–5: see Faulkner (2008), 288–9.

77 For a possible allusion to Hy. 19 to Pan at Call. Hy. 3. 88–9, see Thomas in thisvolume (p. 158 n. 21).

78 See Campbell (1981), 119–22.79 There is no certainty that Apollonius had Hy. 32 specifically in mind, but this

possibility should not be ruled out; see Clauss (1993), 15–16, Hunter (1993), 129 n.110. For further instances of singers beginning from a god, see Kidd (1997), 161 onArat. 1.

80 Cf. also the inception of Hy. 7. 1–2 Içd ˜Ø��ı�� $�ºÅ KæØŒı�� ıƒ��j ��ÆØ (‘I will call to mind Dionysus, son of glorious Semele’).

81 See Clauss (1993), 17.


poem:82 Apollo plays a significant role in book one of the Iliad, andconcern about this initial hymnic celebration of the god could havebeen stimulated by the existence of an alternative hymnic ‘proem’(�æ��Ø��)83 to Apollo and the Muses at the beginning of the Iliad,mention of which in the prolegomena to the D scholia probably goesback to Crates’ Diorthotika.84 This opening stance of Apollonius alsoprepares the way for the blending of the poet’s voice into Orpheus’hymn to Apollo later in the poem (2. 703–13).85

Other echoes of the Hymns in Apollonius are more straightfor-ward. Parallels between the description of Thetis’ attempt to immor-talize Achilles (4. 869–84) and that of Demeter’s efforts toimmortalize Demophoon in Demeter (237–302) are better explainedby direct influence than a common model.86 A number of passages inthe Argonautica show possible influence of Hermes: the narrative ofthe launch of the Argo and the division of the rowing benchesamongst the heroes (1. 363–401) shares language and structuralfeatures with the description of Hermes’ sacrifice of Apollo’s cattle(111–41)—recollection of the Hymn highlights the shared motifs of ajourney to steal an animal prize (Hermes the cattle, Jason the fleece ofa ram) and reconciliation brought about through the lyre (Hermes’appeasement of Apollo with the lyre at Herm. 417–35, Idas’ disputewith Idmon settled by Orpheus’ song at 1. 460–518);87 Hermes’clandestine theft of cattle and sacrifice of two cattle may also haveinfluenced elements of Jason’s encounter with the bulls in book three

82 See Hunter (1993), 119–29 (1996), 46, on the proem’s establishment of theArgonautica as a rewriting of Homeric epic and the importance of hymnic form in thepoem, which is framed as a ‘hymn’.

83 On this word used of the Hymns, see in this volume Clay (pp. 237–40) andIntroduction (pp. 17–19).

84 See West (2001b), 73; the reference includes the incipit )��Æ I��ø ŒÆd��ººø�Æ Œºı�� (‘I sing of the Muses and Apollo renowned for the bow’).

85 On the mixing of voices in the Orpheus hymn, see Hunter (1993), 150–1: ‘In factit is not possible to distinguish the voices here, as both etymology and aetiology, whichare prominent in these verses, are familiar markers both of the poetic voice ofthe Argonautica and of the voice of hymns.’

86 See Richardson (1974), 237–8, Vian iii 178.87 See the detailed discussion of Clauss (1993), 69–74: the most striking linguistic

parallel is Herm. 128 � A. R. 1. 365 (º��øfi K�d �ºÆ�ÆH�Ø, ‘upon a smooth stone slab’).A further link between the journeys of Jason and Hermes might be found in the factthat both lose their sandal(s) in a river: the former drops one of his (��Øº��) in theriver Anauros, the sign to Pelias that Jason will be his downfall (1. 5–11), whileHermes throws his (��ÆºÆ) into the river Alpheios to cover his tracks (Herm. 139).

194 Andrew Faulkner

of the Argonautica (3. 1191–325);88 finally, Aphrodite’s offer of an¼ŁıæÆ (‘plaything’) to her precocious and tricky son Eros (3. 114–66) and the description of Eros shooting at Medea (3. 275–97) seemto be indebted to Hermes.89 As for Aphrodite, language used of theArgonauts’ erotic encounter with the Lemnian women in book onerecalls the opening of the Hymn: both 1. 803 ˚��æØ�� , l � �çØ�Łı�çŁ�æ�� Æº�� ¼�Å� (‘Cypris, who cast life-destroying delusionupon them [the husbands of the Lemnian women]’) and 1. 850˚��æØ ªaæ K�d ªºıŒf� ¥�æ�� tæ�� (‘For Cypris roused up sweetdesire [in the Argonauts for the Lemnian women]’) echo Aphr. 2˚��æØ�� , � �� Ł��E�Ø� K�d ªºıŒf� ¥�æ�� tæ�� (‘Cypris, who rousesup sweet desire amongst the gods’).

Elsewhere in early Hellenistic poetry, Theocritus clearly rewritesHymn 33 to the Dioscuri in the opening lines of Idyll 22 and struc-tures his poem as a rhapsodic hymn.90 Idyll 17, the encomium ofPtolemy Philadelphus, is also patently fashioned as a hymn (if not asclearly a Hymn as Idyll 22), which interacts with the account ofApollo’s birth on Delos in Apollo, as do Callimachus’ Hymns.91

Also, the reaction of Alcmena to the snakes attacking her infantsons Heracles and Iphicles in Idyll 24. 1–63, which begins in thefashion of a Hymn by placing Heracles’ name as the first word inthe poem,92 seems to recall the Metaneira–Demophoon episode inDemeter,93 a poem which, along with Apollo, seems to have been verypopular in the Hellenistic period. It stood as a principal model forPhilitas’ hymn to Demeter,94 and was probably also a source for

88 See further Thomas (2009), 304, who points out that Jason’s search for the bullscould ‘be an inverse of Herm.: where Apollo looks at confusing trails and cannot findthe cows’ underground stall, Jason succeeds—and he is compared to Apollo (3. 1283).’

89 See Campbell (1983), 18–19, (1994), 116–36 passim, Pace (2004), Thomas(2009), 303–4.

90 See Hunter (1996), 46–57, Sens (1997), 13, 75–95 passim. Alcaeus’ hymn to theDioscuri (fr. 34 Voigt) may also have been a model.

91 See Hunter (2003), 8–9, 142–4, passim, who notes that Aratus (SH 109) may alsohave drawn on Apollo.He also (145) signals a possible echo of Hy. 26. 3–4 at Id. 17. 59.

92 Cf. Hy. 15 to Heracles and see Hunter (1996), 11. On this common feature of theHymns, cf. Faulkner (2008), 70–1.

93 See Hunter (1996), 12, with further bibliography. Thomas (2009), 305 tenta-tively compares elements of Heracles’ questioning of a cowherd of Augeas in Id. 25 toApollo’s questioning of the old man of Onchestus in Herm. 189–212.

94 See Spanoudakis (2002), 239–40, passim.


Philicus’ hymn to the goddess.95 Aratus may have had Hermes inmind in Phaenomena 268–9, when he tells of Hermes’ invention ofthe lyre.96 More examples might be adduced, but these are enough todemonstrate the widespread and pervasive influence of the Hymns onearly Hellenistic poetry. As Hunter points out, this popularity can belinked to the general prevalence of hymnic praise evident in Hellen-istic literature, which is tied to the changing dynamics of patronage inan age when rulers had themselves become divine: ‘The “Homerichymn”, which identified the areas of a god’s power and placed him orher within the overall scheme of the divine, seems in retrospect anobvious vehicle for describing these shifting boundaries of power.’97

Prior to the Hellenistic period, the demonstrable influence of theHymns is less, but their currency from the seventh to the fourthcenturies bc nonetheless appears to have been considerable. In ex-

ploring the transmission and reception of the Hymns before the thirdcentury in particular, one must always keep in mind that the con-tinued tradition of oral rhapsodic performance produced variants tothe texts transmitted in our collection. The quotation of the Hymn toApollo by Thucydides shows that at least minor variants of the longerHymns existed in his time,98 and there were undoubtedly manydifferent versions of short hexameter hymns circulating in the fifthcentury bc and earlier. Nonetheless, there are indications, as will be

explored below, that Apollo at any rate was a relatively fixed text bythe end of the sixth century, such that one can speak tangibly of itsdirect influence on later texts. Indeed, even while assuming ongoingoral performance, one must consider the possibility that the longHymns transmitted in our collection were fairly fixed texts, in somecases from at least the late seventh or early sixth century bc onward,

95 SH 676–80; see Richardson (1974), 70. In his recent edition and study of thepoem Furley (2010) proposes that the unknown goddess with whom Demeter isspeaking in fr. 680 is Aphrodite, who on this reading refers to her son Aeneas asforced upon her by Zeus: this is highly speculative, but would provide a link to Aphr.,which contains the only account of Zeus as the instigator of Aphrodite’s affair withAnchises.

96 See Kidd (1997), 281–3; he also (278) suggests that Arat. 263 ��ØÆ )ÆEÆ(‘distinguished Maia’) recalls Herm. 19, 183, the only other occurrences of the phrase.

97 Hunter (1996), 47. Petrovic (2011b forthcoming) argues that the Hymns weremodels for Hellenistic epyllia.

98 See above n. 10. On the variants of Thucydides quotations and the reception andtransmission of the Hymns within the oral tradition, see Nagy in this volume (Ch. 13).

196 Andrew Faulkner

and were (despite minor variants) models for later literature (includ-

ing even for other Hymns).99 In some instances, the accumulation,uniqueness, or context of parallels is particularly suggestive, but it willbe useful also to adduce some more speculative cases, particularly inliterature of the fourth and fifth centuries, with the admission thatone can in many cases not be certain.

Plato, despite his concern with the classification of hymns (o��Ø:one of the acceptable forms of poetry in his ideal city)100 and frequentquotation of Homer, does not cite the Hymns.101 But the quotation ofApollo by Thucydides, just mentioned above, suggests that the Hymnswere established in the literary landscape at least by the late fifthcentury. Thucydides refers to the verses he cites as from the ‘prooimionof Apollo’ (�æ��Ø�� F ��ººø�� , 3. 104); in the absence of anymention of the Pythian section of the Hymn or further poetic praise ofthe god, the term prooimion (‘prelude’) is best taken here as a genericterm for the Hymns, which seem to have served as preludes to longerepic recitations.102 Fragments of the Thebaid and Lyde of Antimachusof Colophon, active at the end of the fifth century and an importantprecursor to later Hellenistic authors, also indicate knowledge of theHymns.103 Some of the shorter Hymns must as well have been circulat-ing in the fifth century. An Attic lekythos dated to c.470 bc shows a

young boy holding a papyrus roll, on which are inscribed the first lines

of our Hymn 18 to Hermes, % EæB� I��ø (‘I sing of Hermes’), anindication that hexameter hymns of this type served as school texts inthe fifth century.104

There is good reason as well to think that dramatists of the fifthcentury were familiar with the Hymns. Exemplary of the difficulty

99 On the fixity of early hexameter texts, see Introduction (pp. 3–7).100 Cf. R. 10. 607a, Lg. 700b. See Furley and Bremer (2001) i. 8–12.101 Garvey (2008) argues that Plato looks back to the Hymns in the fashioning of

his Atlantis myth as a hymn to Athena, but the connections he notes are extremelygeneral. The rejection of Hephaestus’ binding of Hera as a suitable myth forhis ideal city at R. 378 may reflect knowledge of Dion.; cf. West in this volume (p. 40).

102 See, however, Clay in this volume (pp. 237–40) and cf. Introduction(pp. 17–19).

103 See Richardson (1974), 69, Matthews (1996), 26, 38–9, passim; the latter rightlypoints out that the extent of influence of the Hymns on the Lyde proposed by DelCorno (1962), 89–90, who supposed Antimachus to be reaffirming their Homericpaternity, is not supported by the fragmentary remains of the poem.

104 See Beazley (1948), (1950), 318–19. Cf. Richardson (2010), 153 ad Herm. 1.


in distinguishing between cult and literary hymns, the choruses oftragedy often enact religious choral hymns within the context of adramatic plot.105 It is unsurprising that playwrights may at times havedrawn on the Hymns in such instances and it is worth consideringsome possibilities, however uncertain these may be. The second stasi-mon of Euripides’ Helen (1301–68) recounts the tale of Demeter’ssearch for Persephone. It departs from the tale told in Demeter (e.g.in its syncretism of Demeter and Cybele), but there are a number ofsimilarities with the Hymn that together suggest influence: Demeterrushesmadly over land andwater in her search (1301–7�Dem. 43–4);Demeter is in a state of ‘longing’ (��Łøfi ) for her daughter (1306–7 �201, 304); Demeter is accompanied by Athena and Artemis, who arecompanions of Persephone at the time of her abduction in the Hymn(1314–16 � 424); Zeus is involved in the abduction (1317–18 �passim); Demeter causes a famine upon the earth, bringing an end tosacrifices (1327–37 � Dem. 305–12); Zeus sends divine messengers,albeit different ones, to try to appease Demeter (1338–49 � 313–28);the goddess laughs in both accounts and turns from her anger (1349–52�Dem. 202–4), although in theHymn this is caused by Iambe and isprior to the famine, whereas in Euripides the envoy of gods is success-ful in cheering Demeter.106 One wonders whether Euripides mightelsewhere have had the first lines of the firstHymn toDionysus inmindwhen describingDionysus’ birth in his hymn to that god in the parodosof the Bacchae (64–169: see 71 ˜Ø��ı�� ø ‘I will hymn Diony-sus’): similar language describes Zeus’ concealment of Dionysus fromHera in the context of the double birth from Zeus’ thigh (98 Œæı��e�Iç’ �HæÆ ‘hiding [him] from Hera’ � Dion. A 8 Œæ��ø� º�ıŒ�º��HæÅ� ‘hiding [you Dionysus] from white-armed Hera’). In thehymn to Aphrodite at the beginning of the fourth stasimon of theHippolytus, Eros is said to tame all creatures of land, sea, and sky(1268–81); the structure is traditional for expressing ‘all creation’,107

but its occurrence in a hymn brings to mind the opening lines ofAphrodite (3–5). More extensively, a healthy body of shared languageand details of plot seem clearly to point to the influence of Hermes onSophocles’ Ichneutae, which likewise told of Hermes’ invention of the

105 On the classification of dramatic hymns, see Furley and Bremer (2001), i. 273–9.106 See Richardson (1974), 69, Allan (2008), 292–310, passim.107 See Barrett (1964), 394.

198 Andrew Faulkner

lyre and theft of Apollo’s cattle.108 There are, however, few signs of theHymns’ influence inAeschylus.109 At least, thePythia’s account of howApollo comes to occupy Delphi in the opening lines of the Eumenides(1–19) has the site pass peacefully from Gaia, to Themis, to Phoebe,and then as a birthday gift to Apollo; there is no mention of Apollo’sviolent conquest of Delphi as recounted in Apollo (300–74).110 Turn-ing to comedy, we have seen above that Aristophanes’ Birds 575 mayechoApollo 114,111 and it is probable that knowledge of the same poemis reflected in Knights: Apollo shouts prophetically from his temple inlanguage that recalls his entrance to theDelphic temple inApollo (1016YÆå�� K� I��Ø� �Øa �æØ��ø� KæØ��ø� ‘[Apollo] shouted from hissanctuary through the prized tripods’ � Apoll. 443 K �’ ¼�ı��ŒÆ��ı�� Øa �æØ��ø� KæØ��ø� ‘[Apollo] entered his sanctuarythrough the prized tripods’). The repeated attribution of the phrase)�ı��ø� Ł�æ��ø� O�æÅæ� (‘busy servant of the Muses’) to Homer(ŒÆ�a �e� �OÅæ��) in Birds 909–10 and 913–14 might also reflectknowledge of Hy. 32. 20, where singers are called the ‘servants of theMuses’ ()�ı��ø� Ł�æ�� ), although the expression occurs as wellin the cyclic Margites (fr. 1. 2 Bernabé) and in Hesiod (Th. 100).112

Lyric poetry likewise provides evidence of the Hymns’ impact andshows that by the early fifth century Apollo at least was established inthe canon of hexameter poetry circulating in Greece. A fragmentaryPaean (7b) of Pindar that treated the origins of Delos-Asteria notonly reflects knowledge of the poem but also sets the subject ofPindar’s own choral poetry against it as a measure:

Œ�ºÆ��ÆŁ’ o��ı ,% ˇ�æ�ı [�Œa ¼�æØ]�:�� ŒÆ�’ IÆ�Ø��N�� I: [�d �PŒ Iº]º��æ�ÆØ I�’ ¥��Ø K��d ÆP[��d �e ��]�Æ�e� –æÆ)�Ø�Æ[E�� KºÆ��]��. (Pa. 7b. 10–14)113

108 See the detailed discussion of Thomas (2009), 309–12, Richardson (2010), 25.Athanassios Vergados, whose commentary on Herm. is forthcoming, also discussesthe connection.

109 Cf. Wecklein (1920), 3–4, whose general discussion of the relationship of theHymns and tragedy remains useful, if dated.

110 See Clay (1989), 62–3 on the uniqueness of Apoll.’s narrative. Cf. Rutherford(2001), 396–7.

111 Above p. 177. Cf. also above p. 178 for a possible recollection of Hy. 21 in theAnagyros.

112 Cf. Dunbar (1995), 529.113 This text follows the supplements of D’Alessio (1995).


Sing hymns,journeying [far from] Homer alwaysupon an [un]trodden wagon-track,not on the mares of others,since we drive the winged chariot of the Muses.

This rejection of ‘Homer’ in a poem concerned withDelos can only bean allusion to the Hymn to Apollo (Apollo and Leto are addressed inlines 1–3). There has been some debate whether the fragmentarypapyrus text necessarily signals a departure from Homer, but thediscrepancies between the narratives of the two poems (Asteria isnot a subject in the Hymn), along with differences in the accounts ofApollo’s birth in the Hymn and another Pindaric fragment (Pa. 12),make this interpretation by far the most probable.114 Pindar as chorusleader also states a few lines later (18–20) that ‘the minds of men areblind’ (�]ıçºÆ: [d ªa]æ I��æH� çæ�� ) who pursue the ‘road of wisdom’(��ç�Æ ›�� ) without the Heliconian Muses: this is possibly a refer-ence to the poet-narrator’s self-identification as a blind bard fromChios in Apollo (165–78) and is indicative not only of Pindar’sdeparture from the narrative content of the hexameter poem butalso his problematization of the authority of the Hymn’s poeticvoice in relation to his own choral voice.115 Elsewhere, Bacchylidesseems to imitate the journey of Hades and Persephone in Demeter(375–83) in an eagle simile (Ep. 5. 16–30).116

Earlier still, West notes several possible points of contact betweenthe Hymns and the Lesbian poetry of Sappho and Alcaeus in the lateseventh and early sixth centuries:117 the similarity of Hestia’s oath ofvirginity in Aphrodite (25–32) and Artemis’ oath of virginity in ahymn attributed to either Sappho or Alcaeus (Sapph. fr. 44a. 1–11Voigt), both discussed above as models for Callimachus,118 provides

114 See Rutherford (2001), 243–52, 364–72, D’Alessio (1995) with furtherbibliography, contra Di Benedetto’s % ˇ�æ�ı [��º��æØ]�:�� ŒÆ�’ IÆ�Ø�� j N�� (‘Journeying upon the well-trodden wagon track of Homer’).

115 See Bing (1988), 103–10, Fearn (2007), 9–16; the latter also links Pa. 7b. 42 ��Æ[Ø and Apoll. 176 ��ÆØ. On the possible blending of choral and rhapsodicvoices in Apoll. and Callimachus’ later engagement with this aspect of the Hymn, seeabove pp. 185–8.

116 See Maehler (2004), 113. Cf. Richardson (1974), 75 for a possible reflection ofDemeter’s claim to arrive at Eleusis from Crete (Dem. 123–4) in B. fr. 47 SM.

117 West (2002a), 216–17, and in this volume (pp. 39–40) on the links betweenDion. and Alcaeus.

118 See above pp. 191–2.

200 Andrew Faulkner

one such case; broadly similar to Hermes, Alcaeus’ hymn to Hermes(fr. 308 Voigt and SLG S264) told of the birth of Hermes and his theftof Apollo’s cattle and arrows;119 his hymn to Dionysus (fr. 349a–eVoigt) recounted the birth of the god and the myth of the binding ofHera, as in Dionysus; and his hymn to the Dioscuri (fr. 34 Voigt)treats much of the same material as Hymn 33 to the pair.120 None ofthese correspondences need be the result of direct influence of theHymns transmitted to us (at least in their exact form). As West pointsout, this is certainly possible in the cases of Dionysus and Hymn 33 tothe Dioscuri, but the probable late sixth- or early fifth-century date ofHermes rules out its influence on Alcaeus,121 while the oath ofvirginity in Aphrodite by Hestia, a goddess rarely personified, has asecondary look and might be linked to Alcaeus’ hymn by a commonhexameter model.122 If one assumes with him that Alcaeus wasdrawing upon a body of hexameter hymns, of which similar versionsmay be included in our collection, it is striking that there is no sign ofdirect interaction with the narrative of Apollo in Alcaeus’ own hymnto the god (fr. 307 Voigt).123 This could be a sign that the centralnarratives of the Hymn’s Delian and Pythian sections are largelyepicized embellishments of cult myth composed principally in thesixth century;124 they were at least not circulating with the body ofepic hexameter hymns available to Alcaeus and Sappho.

The earliest possible influence of the Hymns (or at least earlierversions of the hexameter hymns transmitted to us) on Homer istreated elsewhere in the volume.125 It remains to consider briefly theinteraction of the Hymns themselves, which extends back at least to

119 Cf. Page (1955), 252–8. Hor. Carm. 1. 10. 9–12 drew on Alcaeus’ hymn.120 For a list of parallels, see Page (1955), 267.121 Thomas (2009), 292–3, notes that mention of Apollo’s fear that Hermes will

steal his ŒÆ��ºÆ ��Æ at Herm. 515, which the trickster god subsequently promisesnot to do, ‘simultaneously alludes to and eschews the version of reconciliation firstextant in Alcaeus’ Hymn to Hermes’.

122 Cf. Faulkner (2008), 110–11 and 45–50, on the links between Aphr. and theLesbian poets, which may be explained by the Hymn’s origin in northern Asia Minor.

123 On the distance between Alcaeus’ hymn and Apoll., see Page (1955), 249–50.The former recounts Apollo’s return to Delphi from the Hyperboreans, who weresignificant players in traditional hymns to Apollo ascribed in antiquity to Olen; onthese, see Furley and Bremer (2001), i. 146–51.

124 See in this volume Furley (pp. 221–4).125 See in this volume West (Ch. 2) and Brillet-Dubois (Ch. 6).


the beginning of the sixth century. The Hymn to Aphrodite, datingprobably to the middle or late seventh century bc and one of the olderpoems in the collection, seems to have given inspiration to the poet

of Demeter. Numerous verbal parallels, some of them unique withinextant early epic, link the two directly—e.g. �Ø��å� (‘honoured’,Aphr. 31–2 � Dem. 268–9), �º�Łæ�ı j ŒFæ� Œ�æÅ (‘her head reachedthe roof-beam’, Aphr. 173–4 � Dem. 188–9 in the context of epi-phany)—and the balance of the evidence suggests that Demeter issecondary.126 Assuming that Demeter was itself composed in the lateseventh or early sixth century, Aphrodite will have gained a reason-ably wide circulation by this time. Elsewhere, Hermes seems to fash-ion itself in relation to Apollo, two poems that are connected in ageneral fashion by Apollo’s role as a principal protagonist inHermes:127 on the level of detail, Hermes’ encounter with the oldman in Boeotian Onchestus, a substantial detour for a god on his wayto Elis from Cyllene, may have been suggested by Apollo’s sojournthere during his search for a suitable spot to found his oracle in Apollo(229–38), while both gods precociously ‘jump’ into action at theirbirths (Herm. 20 Ł�æ� � Apoll. 119 �ŒŁ�æ�)—the theme of competi-tion and reconciliation between Hermes and Apollo in Hermes sets itin competition with Apollo on the metapoetic level. Furthermore,instances of allusion to both theDelian and Pythian sections of Apollosuggest that Hermes responds to a unified poem that had achievedwide enough popularity to make the intertexual references in Hermesperceptible to an audience. If one subscribes to the view thatCynaethus combined the two sections of Apollo for performance ata festival of Polycrates in 523/2,128 this would provide a terminus postquem of the late sixth century for Hermes,129 but this is not certain.130

As we have seen, the canonical status of Apollo is similarly reflected inPindar. Later still, it is likely, as Thomas argues in this volume, that

126 See Janko (1982), 163–5, Richardson (1974), 42–3, Faulkner (2008), 38–40.127 On the links between these two poems see Richardson (2007), 89–91, (2010),

20–1, and the detailed discussion of Thomas (2009), 290–5, with further bibliography:Thomas (296–7) also points out a number of parallels which could suggest that thepoet of Herm. knew Dem.

128 See in this volume Chappell (pp. 71–3) and Nagy (pp. 288–91).129 See Thomas (2009), 20–9.130 On the dating of the Hymn, in this volume Introduction (pp. 12–13).

202 Andrew Faulkner

Hymn 19 to Pan makes substantial allusion to Hermes, a case ofmetapoetic genealogy.131

Turning to the smaller Hymns, 31 to Helios and 32 to Selene forman obvious pair (both celebrate celestial bodies and uniquely mentionthe deeds of heroes in their closing formulae). Otherwise, several ofthe smaller Hymns share similarities with longer Hymns in the collec-tion: Hymn 13 (three lines) to Demeter and Persephone is a variant ofthe first two lines of Demeter; Hymn 17 (five lines) to the Dioscurishares language with Hymn 33 (Hy. 17. 3–4 � Hy. 33. 4–5; Hy. 17. 5� 18); Hymn 18 to Hermes (twelve lines) is a variant of the first ninelines of Hermes; in Hymn 27 to Artemis (twenty-two lines), thedescription of the goddess’s pleasure in the bow and hunt, as well asher subsequent return to Apollo’s house in Delphi, recalls the scenesof her brother’s return to Olympus in Apollo (Hy. 27. 4–20 � Apoll.2–9, 186–206);132 and Hymn 30 to Earth Mother of All (nineteenlines) shares language with Demeter (Hy. 30. 7–8 � Dem. 486–7; Hy.30. 18-19 � Dem. 494-5). Of these, the links between Apollo andHymn 27 are enough to suggest direct influence of the former,although one cannot be certain of this: Artemis’ movement on thehunt causes the mountains to tremble (�æ��Ø �b Œ�æÅ�Æ j �łÅºH�Oæø�, Hy. 27. 6–7), a possible adaptation of the gods’ trembling atApollo’s approach to Olympus (‹� [��ººø�Æ] �� Ł��d ŒÆ�a �HÆ˜Øe �æ��ı�Ø� N��Æ, Apoll. 2); both Artemis and Apollo draw theirbows (��Æ �Ø�Æ��Ø, Hy. 27. 5 and Apoll. 4); Artemis unstrings andhangs up her bow upon arrival at Delphi (Hy. 27. 11–17), both ofwhich Leto does for her son upon his arrival on Olympus (Apoll. 5–9);and Artemis leads the chorus of the Muses and Graces at Delphi (Hy.27. 13–20), in which she joins after her brother’s arrival on Olympus(Apoll. 197–9). If there is a direct connection, Artemis’ dancing atDelphi looks secondary to the scene on Olympus, which is describedin much greater detail and is particularly appropriate for the god ofmusic, who travels from Delphi to Olympus while playing the lyre

131 In this volume (pp. 166–8). It is notable that the short Hy. 18 to Hermesprecedes Hy. 19 to his son Pan, although this could be due to the hand of a later editorrather than indicating an earlier link between the two.

132 AHS 242 also compare Hy. 28. 15–16 (to Athena eighteen lines) to Apoll. 7 and12, but the link seems negligible. However, shared language between Hys. 27 and 28(Hy. 28. 3 � 27. 2; Hy. 28. 10 � Hy. 27. 8) could suggest a direct relationship betweenthe two (Gemoll supposed they were composed by the same person): Athena’s birthcauses existence to shake and cry forth, as does Artemis’ hunting.


(Apoll. 182–5). It is also significant that the similarities are withthe opening scenes of both the Delian and Pythian sections of Apollo.As regards the other similarities between shorter and longer Hymnsnoted above, it is possible that some of the Hymns derived fromothers. Hymn 25 to the Muses and Apollo, for example, has oftenbeen thought to derive from the longer hymn to the Muses at theoutset ofHesiod’s Theogony.133 One must also consider the possibilitythat the longer narrative Hymns in fact developed as expansionsof shorter hexameter hymns to gods, variants of which have beentransmitted in our collection. In the end, however, such similarities,even as extensive as those between Hymn 18 and the opening ofHermes, are easily explained by the Hymns’ development in a com-mon tradition and do not necessarily suppose direct influence of theHymns transmitted in our collection.

The eclecticism of time has undoubtedly effaced much which couldbe added to the parallels and possible allusions discussed above.Nonetheless, the surviving evidence provides a broad picture of thetransmission, reception, and status of the Hymns from the late sev-enth to the third centuries bc. From the earliest stage, we can imagine

the Hymns being circulated in Greece by rhapsodes, as describedin the poet’s self-referential comments in Apollo (165–78), and func-tioning oftentimes as preludes to recitations of longer epic poetry.By the late seventh century, knowledge of the Hymns, or at least abody of hexameter hymns in the same tradition, seems to be reflectedin the poetry of Sappho and Alcaeus. The direct influence of theHymns on each other can also be traced to the late seventh or earlysixth century, as indicated by the close relationship of Aphrodite andDemeter. By the last half of the sixth or early fifth century, Apollo atleast had achieved canonical status, as reflected in Pindar andHermes.The Hymns were familiar to both lyric and dramatic poets and mayhave played some role in the fifth-century school curriculum. Thucy-dides’ famous quotation of Apollo reflects that poem’s continuedpopularity at the end of the fifth century, although the evidencedoes not allow us to posit a collection of Hymns at this point. Thetraces of their transmission in the fourth century are scanty, but bythe early third century they were widely emulated by the earliest

133 See AHS 420, West (1966), 186, but contra Càssola 401–2.

204 Andrew Faulkner

Hellenistic poets. The numerous and intricate allusions to the Hymnsby Callimachus in particular allow for the supposition that a collec-tion of some sort was formed in the early third century bc, at a time

when Alexandrian scholars were collecting together the literature of

the past and the Hymns’ popularity was at its greatest. Scholarship onthis topic has made enormous strides in the past century. In compari-son to theHomeric epics theHymns acquired little notice in antiquity,but their substantial impact on the Greek poetic tradition must none-theless be acknowledged.


10

Homeric and Un-HomericHexameter Hymns

A Question of Type

William D. Furley

It is easy to forget that the Homeric Hymns were not the only kind ofhexameter hymn to gods in circulation in the archaic to early classicalperiod in Greece. Because they, solely, are visible to us, one tends tothink that hymns started with compositions like the Homeric Hymnsand developed from there. The purpose of this contribution will be toexamine what we can reconstruct about other types of hexameterhymns of this period in order to compare their characteristics withthose of the Homeric Hymns.1 In that way, I hope, we will learn moreabout the Homeric Hymns by seeing how they differ from othercompositions. It might be as well at this point to sketch some com-monly held opinions about the Homeric Hymns as a poetic type.

The thirty-three hymns in our collection vary greatly in lengthfrom several hundred verses (approximately equivalent to a ‘book’ ofHomeric epic) to only a handful of lines.2 They begin, almost

1 I take the main period of currency of compositions such as the extant HomericHymns to have been the seventh–fifth century BC. Precise dating of individual hymnsin the collection is notoriously difficult; see Introduction (pp. 7–16). Janko (1982) isstill fundamental.

2 This is to omit the thirty-fourth Hymn, entitled ‘To Strangers’ (�N ��ı ), as it isnot a hymn to one or more gods. It does, however, mention the birth of River Hermosto Zeus. Pfeiff (2002), 196–7, makes the plausible suggestion that a (or ‘the’) compilerof the collection of Homeric Hymns may have inserted these lines from the pseudo-Herodotean Life of Homer (9 Allen) as a kind of poetic sphragis. The lines are

universally, with a third-person invocation of the god to beaddressed,3 and then launch almost immediately into a descriptionof the god’s nature and his mythical origin and deeds. They close,typically, with a salute to the god and a self-exhortation to strike upanother song following this hymn. The variable length is explained byvarying degrees of expansion of the middle section in which the god’sbehaviour and career are elaborated upon.4 Here, we feel, length was amatter of the author’s discretion. A long hymn required narrativeelaboration; a short hymn could be reduced to a brief sketch of thesubject. In antiquity the term prooimion is used of the HomericHymns, indicating some kind of prelude.5 A combination of evidencepoints to the conclusion that the Homeric Hymns served as preludesto epic recitations by rhapsodes, who introduced their recital ofHomeric and cyclic epic with a bow to the gods in the same languageand metre as Ionic epic.6 We know that rhapsodic competitions tookplace at various panegyreis in Greece;7 one can well imagine how arhapsode would want to gain favour at the commencement of hisperformance both with the local audience and the local gods bysinging a hymn of praise to the patron god of the occasion. Thesinger of the Apollo, for example, is clearly in Delos when he singsof the birth of Delian Apollo and engages in a kind of dialogue with alocal chorus of Delian women.8 Not that the view of the Hymnsas rhapsodic preludes is completely universal: some have imaginedother settings for the performance of the Hymns: large formal ban-quets, for example, might provide an alternative setting, such as is

conventionally transmitted as Epigram 1. Càssola and WL both omit it from theireditions of the Hymns.

3 Calame uses the term evocatio to distinguish this from direct appeal to a god tocome or appear (invocatio); see in this volume (Ch. 14). Clay in this volume(pp. 235–6) points out that the Homeric Hymns typically move from an openingaddress to the deity in the third person to a farewell at the close in the first person.Thus the intervening narrative has established a line of communication with the godaddressed, which the poet acknowledges with his closing salute.

4 Surely Ausfeld’s description of the central part of a hymn as pars epica derivedfrom the Homeric Hymns: Ausfeld (1903), 505.

5 See esp. Böhme (1937); Aloni (1980). Cf. in this volume Introduction (pp. 17–19)and Clay (pp. 237–46).

6 The conclusion of Hy. 32 to Selene says explicitly that the singer wishes to pass onto recitation of the deeds of mortals and half-gods; this seems equivalent to the kleaandrōn of epic.

7 Herington (1985), ch. 1; Kirchhoff (1893).8 More on this below pp. 221–4.

Homeric and Un-Homeric Hexameter Hymns 207

illustrated in Homer himself when Demodocus sings of the adulteryof Ares and Aphrodite before the assembled Phaeacians.9

The conspicuous and ‘Homeric’ feature of the Hymns is, at least inthe long Hymns, their extended narrative. They tell, at length, what‘happened’ in mythical time when the chosen god first came intoexistence or first came to prominence. True, Janko distinguishes such‘mythical’ hymns in the collection from ‘attributive’ hymns, which tellonly of the gods’ timeless attributes and power,10 but the memorablequality of the major Hymns, and some of the medium length Hymns,is their dramatic narrative, recreating for the listener a key moment inthe rise to prominence of the divine subject. Not all the myths told inthe Homeric Hymns concern the birth and youthful development ofthe god, although that is an important element of the Hymns toApollo, Hermes, and probably the fragmentary first Hymn to Diony-sus.11 Others concentrate on a crisis or turning-point in the god’scareer which changed things for ever. Demeter loses her daughter tothe god of the Underworld, Hades, but establishes her main cult onearth, the Eleusinian Mysteries, in the course of her unhappy searchfor her daughter. Aphrodite in her Hymn is made by Zeus to couplewith a human, Anchises, in order to end her playful tyranny over theother gods in making them stoop to unions with men and women.Interestingly, Janko sees a common element in both birth myths and‘crisis myths’ (my term): the ‘apparition’ (his term) of the god eitherliterally through appearing in the world at birth, or through epiphanyat the climactic moment of a drama on earth.12 Accepting this, wecould see the main point of the extended myths in the Hymns asdocumenting the god’s dazzling appearance among the company ofOlympians as a force to be reckoned with, and worshipped by peoplefor this reason.

Clay has famously argued that the major Hymns focus on the‘politics of Olympus’; that is, on the inter-personal power relationsbetween gods under the supremacy of Zeus.13 The myths can be seento dramatize key moments in the acquisition or distribution of poweror prerogatives among individual gods: Hermes vies with Apollo;

9 See in particular Clay (1989), and in this volume (pp. 249–50).10 Janko (1981).11 Cf. West (2001a) and in this volume (Ch. 2).12 Janko (1981), 13–14.13 Clay (1989), and in this volume (Ch. 11).

208 William D. Furley

Zeus subjugates Aphrodite to his will; Demeter and Hades becomereconciled after the rape of Kore. Once the conflict has been resolved,usually by Zeus, and the antagonists reconciled, a divine order isestablished which provides the framework for cult and worship.Thus narrative action in the past tense may be seen to lead to, andjustify, a lasting order in the world. The insight is valuable, but it mayattribute a greater theological authority to the Homeric Hymns thanthey themselves intended. Their composers, if they were intent onwinning a competition in epic recital, probably wished to show offtheir narrative and recitatory prowess rather than lay down a newtheological order for the audience.14 No doubt the myths they drewon and ‘epicized’ were traditional and well-known in one form oranother to the audience. Of course, a good epic-style narrative of atraditional myth would lend popularity and hence authority to thepoet’s version, but the rhapsodes who composed these overtures toepic were probably not great innovators.15 One might point to thedebate surrounding the Hymn to Demeter. Clearly it describes thefounding of the Eleusinian Mysteries and at some points it refers toknown features of cult at Eleusis. However, as has been pointed out byClinton, there are many discrepancies and omissions in the Hymn’saccount of Eleusis compared with what we know of historical cultthere from other sources which make it unlikely that the author is a

14 The more arbitrary nature of the theology of these compositions comes out inPlu. Mor. 1133c (De mus. 6), �a ªaæ �æe ��f Ł��f ‰ ��º��ÆØ Iç��Øø��ØK��ÆØ�� (sc. ŒØŁÆæøfi ��d) �PŁf K�� c� % ˇ�æ�ı ŒÆd �H� ¼ººø� ��Å�Ø�· �Bº�� b��F�! ��Ø �Øa �H� '�æ��æ�ı �æ��Ø�ø�, ‘[sc. kitharodes] first dispatch their religiousduty to the gods as they see fit and then proceed directly to the poetry of Homer andothers. This is obvious from the prooimia of Terpander.’ One would like to knowmore about these prooimia of Terpander; cf. id. 1132d with Böhme (1937), 15–16.They seem to have been the precursor of the nome (�� ŒØŁÆæøfi �ØŒ� ); cf. Proclus ap.Phot. Bibl. 320a 35 and Rutherford (1995). In another passage of Plu. Mor. 1132d (Demus. 3), Terpander is said to have ‘set his own and Homer’s epic compositions tomusic’ (��E �Æı��F ŒÆd ��E % ˇ�æ�ı ºÅ ��æØ�ØŁ��Æ fi ¼��Ø� K� ��E IªH�Ø). It seemswe must distinguish rhapsodic recital of Homeric works from their musical, kithar-oedic performance; Pausanias vehemently denies that Hesiod, for example, was akitharode, but recited ‘to the staff ’ (9. 30. 3 K�d Þ��ı ��ç�Å ). Cf. Böhme (1937), 25and 49; and Càssola xxii–xxv.

15 This is not to exclude the possibility of some original invention by the authors.Faulkner (2008), for example, points to elements in Aphrodite which may be innova-tive (Zeus’ revenge on Aphrodite and the pursuit of Hestia by Poseidon and Apollo:see his notes on lines 53–199 and 24–5 respectively).


‘high priest’ or even ‘founding father’ of the Eleusinian cult.16 I believethe truth is more likely to be that these Homeric Hymns are a specialtype of hymn adapted by their authors, the Homerids and rhapsodes,from traditional hymnic stock for their own immediate purpose andaudience: epic recitation, whether in competition with other rhap-sodes at an agon mousikos or at another festive occasion.17 What weshould do at this point is try to establish what this hymnic stockmight have been at the time when the Homeric poems were spreadinglike wildfire around Greece (late seventh and sixth centuries BC).18

A good starting point, I think, is Hesiod. It has often been observedthat both his major works, Theogony and Works and Days, begin withintroductory sections addressed to divinities, the Muses and Zeusrespectively.19 This seems to confirm our belief that hymns such asthe Homeric Hymns served as preludes to longer epic compositions.The proem of the Theogony is over a hundred lines long, and com-plex; that of the Works is shorter, addressed to Zeus. In the TheogonyHesiod begins by invoking the Heliconian Muses and describing theirtypical activity: they dance and sing, wash in a spring on Mt Helicon,then proceed to hymn the Olympian gods (1–21). On one occasion,Hesiod relates, they appeared to him while he was herding his flocks,and ordered him to become a poet. He should sing, they said, ofthings future and past, and of the race of immortal gods. In doing so,he should commence and finish his song by addressing the Muses(22–34). Then, in a second take, Hesiod revisits the relation betweenhis poetic undertaking and the Muses.20 He describes how, on Olym-pus now, the Muses sing, as they had instructed him, of thingspresent, future, and past and please Zeus’ ear by singing in dulcettones of the generations of gods preceding the present Olympiandynasty. They trace the gods’ lineage from Gaia and Ouranos in thebeginning, through their descendants, ‘givers of good things’, to Zeus,

16 Clinton (1986), esp. 48. In later work Clinton (1992) proposed that thehymn bore closer affinities to the Thesmophoria ritual. For criticism of Clinton’sposition, see Richardson in this volume (pp. 50–3).

17 Most ancient sources cite—and rather sparsely at that—from ‘Homer’s hymns’or ‘the poet’s’. Ath. 22b wonders whether Apoll. was by Homer or ‘one of the Home-ridai’; % ˇÅæ��ÆØ is the term used by Pi. N. 2. 1. Cf. in this volume Faulkner (p. 176n. 5).

18 Cf. West (1995).19 Among many works on these opening sections see, for example, Schwabl (1963),

Friedländer (1966), Von Fritz (1956), Minton (1970), Ziegler (1911).20 Janko (1981) speaks of two separate proems, in fact.


father of gods and men, most mighty (35-51). It is clear that this songof the Muses is closely parallel to what Hesiod intends in the Theo-gony. At the end of the proem, after elaborating on the birth and giftsof the Muses, Hesiod invokes them again to come now and celebrate‘the sacred race of immortal gods who descended from Ge and starryOuranos’ (105-6). The typical performance of the Muses on Olym-pus, then, is depicted as a model or precedent of what Hesiod intendsnow. Inspired by the Muses he wishes to emulate their example byhymning the gods, in particular, their generations down to the pres-ent day rule of Zeus.21

The traditional way of taking the proem of the Theogony, I think, isto see it as a hymnic prelude to a didactic poem. As the Homeridsprefaced epic with a ‘Homeric’ hymn, so Hesiod prefaced his poemabout the birth of the gods with a salute and tribute to the Muses.22

No doubt this is partly true. But there is another way of looking at it,which may explain rather better the nature of the proem itself.Hymns commonly start either by direct invocation in the secondperson (or third) of the god(s) addressed, or by calling on collectiveMuses to visit the poet with inspiration to hymn the deity. The proemto the Works and Days is definitely of the latter type: the PierianMuses are called upon to hymn Zeus. It is hardly necessary to listexamples generally from Greek hymnody here. The pattern is toocommon.23 Perhaps we can also discern the influence of this modelon the concept of the proem to Hesiod’s Theogony. The ‘hymn to theMuses’may be seen from this vantage point as an extended version ofthe traditional appeal to the Muses for inspiration on the threshold ofdelivering a hymn. In this proem Hesiod calls first on the Heliconian(1–51), then the Pierian Muses to inspire his song (52–114). True, theextended proem has many of the attributes of a fully-fledged hymn inits own right. The Muses are invoked; their characteristic behaviour is

21 On musical performance by the Muses and other gods in the Hymns mirroringthe musical activity of the poet, see in this volume Calame (Ch. 14).

22 Cf. Càssola xxi, ‘la Teogonia ci offre un esempio (l’unico esempio sopravvissuto)della sequenza prooimion-oime’ (‘the Theogony offers an example (the only survivingexample) of the sequence prooimion-oime’).

23 Cf. Wheeler (2002). Note also the alternative beginning of the Iliad recorded inAnecd. Osanni (Nauck 1867, 273) consisting of a short hymnic invocation of theMuses and Apollo. Cf. Böhme (1937), 13, West (2001b), 73. And the opening line ofEumelos’ hymn (fr. 16 Bernabé) )�Å��Å ŒÆd ZÅ�e !Oºı��ı K��Æ Œ�FæÆØ. OnEumelos cf. West (2002b).


described; their lineage is named. But the function of this appeal tothe Muses is the same as in more curtailed appeals for assistance fromthe Muses which we find in many other hymns. According to thismodel, the Theogony itself becomes the hymn addressed to theOlympian gods; the initial appeal to the Muses changes its statussomewhat, from prooimion with separate subject matter (as we im-agine the relationship between a Homeric Hymn and following recita-tion of heroic epic to have been), to subordinate part of the mainendeavour: a long hexameter narrative hymn whose subject is thegenealogy of the gods from the first beginnings to the present dayworld order. It suits this view that Hesiod himself may refer to theTheogony as the hymnos with which he won the prize in the funeralgames of Amphidamas.24

If this is right, we have gained considerable insight into whatconstituted a good hymn in the eyes of Hesiod and, presumably, hiscontemporary audience. Theogony was good material for hymns.What the gods wanted to hear was how they came into being, howthey came to power. Hesiod’s Theogony goes on to tell of the genera-tions of gods who preceded Zeus and Hera, beginning with darkChaos (115 ff.). He takes the story up to the ascendancy of Zeus, hisbattles against challengers, and his establishment of the present orderof Olympians.

Hesiod’s Theogony was by no means the only hymnos we hearabout which had theogony as its main subject. A number of nebulousfigures such as Musaeus, Orpheus, Pamphos, and Olen are accreditedwith hymnic compositions explaining cosmogony in terms of suc-ceeding generations of gods. Most of these works are completely ornearly completely lost. Some few fragments, preserved in laterauthors, hint at the nature of the lost texts.25 Two surviving texts,however, one a recent discovery, and one a comic lampoon, help usbetter to appreciate the nature of these theogonic cosmologies. Thediscovery, and partial decipherment, of the so called Derveni Papyrus

24 As argued by West (1966), 44–5. Cf. Hes. Op. 656–7: ��ŁÆ çÅØ j o�øfi�ØŒ��Æ��Æ çæ�Ø� �æ��! T��Æ. Of course, hymnos here may bear the more generalsense of ‘song’ or ‘song of praise’ which it commonly has in epic, and not referspecifically to a song about the gods. Cf. Furley and Bremer (2001), i. 9–10 with notes,Ford (2002), 12 n. 27.

25 For a brief survey of these texts see below (pp. 215–17).


shows us a text with two aspects.26 It appears to be written by asophistic commentator explaining an older poetic text which heattributes to Orpheus.27 His method is to quote a line or two of thepoetic original, then to offer an ‘explanation’ of the poet’s real mean-ing. For the commentator, the underlying, ‘secret’ meaning of thepoem is the physical and conceptual constitution of the universe. Inthis, he seems to be a physikos like the pre-Socratics, who soughtuniversal theories of matter. Some have suggested that the commen-tator is in fact trying to update the message of Orpheus by showingthat the original divine personifications in the poem may be inter-preted as symbolic physics and ethics. If that were true the authormight have been an Orphic himself; or at least, not a sophistic‘deconstructor’ of the Orphic gospel.28 Assuming that the versesquoted by the author stem (though not exclusively) from the samepoem by ‘Orpheus’ and in the order quoted,29 we may assemble themto gain an impression of the original hexameter composition:30

. . .Who were born from Zeus the almighty king . . .When Zeus took from his father the ordained ruleand strength in his hands and the famed divinity . . .thus he (Kronos?) prophesied all that was allowed for him to doso that he might rule on the fine throne of snowy Olympus.And when Zeus heard his father’s prophecy . . .he swallowed the reverend one, who first entered the air . . .Ouranos, son of Night, who first ruled as king.From him in turn Kronos, then Zeus wise-counsellor,who possessed Wit and royal honour among gods . . .of the reverend first-born King, and in him (Zeus) allimmortal and blessed gods and goddesses were joinedand rivers and lovely springs and everything else

26 Decades after the discovery of the Derveni papyrus in 1962 near Thessalonica wenow finally have an authoritative edition, with good photographs, in Kouremenos-Parássoglou-Tsantsanoglou (2006). For an account of the find, and gradual, some-times surreptitious, circulation of early versions of the text see Betegh (2004).

27 Cols. 7. 5 (restored) and 18. 2 and 7, for explicit attribution to Orpheus.28 This is not the place for an examination, or even a summary, of this perplexing

question. Some fundamental discussions of the question are to be found in Burkert(1968), Laks and Most (1997), Betegh (2004), esp. ch. 5.

29 This is roughly the premise of Betegh (2004), ch. 3.30 Cf. West (1983), esp. 114–15 for an exempli gratia reconstruction of the portion

of the original commented on by the Derveni author; Laks and Most (1997); Betegh(2004); Bernabé pars ii. i, 2–32 for text of the original verses. Detailed discussion of theissues of reconstruction here are beyond the scope of this chapter.


which had existed then; and he became the one and only.Now he is King of all and will be forever more . . .Zeus is king, Zeus the ruler over all with the shining bolt . . .And he placed therein the sinews of silver-eddying Acheloos . . .. . .which shines for many mortals on the measureless earth . . .And when Zeus’ mind had devised all the works . . .31

Although many details of this text elude our grasp, the generalnature of the composition is clear enough. The Orphic poem whichthe Derveni commentator is discussing is a hymnic theogony tracingthe rise to sovereignty of Zeus following previous generations of gods.The commentator calls the composition a ‘hymn’ in col. 7. 2.32 Atanother point he quotes a statement by Orpheus ‘in the hymns’ (22.11 K� ��E % .��Ø ), which sounds like the title of a collection ofhymns by Orpheus.33 Whether the theogony commented upon indetail by the Derveni author was one of this collection is open toquestion. In the quoted verses we hear of a succession myth as inHesiod: Ouranos–Kronos–Zeus, finally, who assumes all gods andbeings within himself (16. 3–4 �fiH �! ¼æÆ �� j IŁ��Æ:��Ø �æ��çı)by swallowing the first-born deity of creation.34 Since the Dervenipapyrus itself is dated to the fourth century BC, and its commentatorto the late fifth or early fourth, the Orphic poem quoted must, bysimple logic, belong to the fifth century or earlier. The Derveni

31 I cite the Greek text from Bernabé. Numbering is by column and line-number:�]Q ˜Øe K��ª: : �� [��æ��]� �Æ�ØºB:� (8. 2) j Z�f b� K��d �c �:Æ�æe ��F ��æÆŁ[�]çÆ�� Iæåc� j [I]ºŒ�� ! Kª å��æ��Ø {�}[º]�:�[�� Œ]Æ: [d] �Æ��: [Æ] Œı�æ�� (8. 4–5) j[‰ �b] �åæÅ�� –�Æ��Æ, �� ƒ Ł[Ø q� I��Æ�]Ł:ÆØ (11. 10) (S �! in. Furley: m �bBetegh, al.) j ‰ i�: �: [å�Ø ŒÆ]�a ŒÆºe� ��:� �Øç�� !Oº��ı (12. 2) j Z�f b� K��d�c: �:Æ�æe ��F ��æÆ: [Ł]:�çÆ�! IŒ��Æ[ ] (13. 1) j ÆN��E�ª ŒÆ�: �Ø��, n ÆNŁæÆ �åŁ�æ: ��æH�� (13. 4) j ˇPæÆ�e ¯Pçæ��Å , n �æ��Ø��: : �Æ��º�ı�� (14. 6) j KŒ ��F �c˚æ�� Æ:s�Ø , ��Ø�Æ �b Å��Æ Z�� (15. 6) j B�Øª ŒÆd: [ÆŒ�æø� ŒÆ�å]ø:�Æ�ØºÅ��Æ �Ø: [�� (15. 13) j �æø��ª��ı �Æ�Øºø ÆN��ı, �fiH �! ¼æÆ �� j IŁ��Æ�:�Ø�æ��çı �ŒÆæ� Ł��d M�: b ŁÆØ�ÆØ j ŒÆd ��Æ�d ŒÆd ŒæB�ÆØ K��æÆ��Ø ¼º:ºÆ �� Æ, j–:��Æ ��! qª ª�ªÆH�! · ÆP�e �! ¼æÆ �F: �� ª��. (16. 3–6) j [�F� �! K��d]� �Æ�Øº�f :��: [øª ŒÆ� �’ ��! ��]�Ø�Æ (16. 14) j Z�f : �Æ�Øº�� , Z�f �! Iæåe ±��:ø: �IæªØŒæÆı�� (19. 10) j r �Æ �! KªŒÆ:�º��! �å�º�Ø�ı Iæªı:æ��: �:ø (23. 11) j m ��ºº�E çÆ��Ø �æ��! {Ø} K�! I��æ��Æ ªÆEÆ� (24. 3) j [ÆP�]a:æ [K]�: �d �[c ��]�:Æ ˜Øe[ çæc�]�Æ�: [� �]æ:ª:Æ: (25. 14).

32 [o]��: [�ª]Ø:B ŒÆd Ł�[Ø]�:a ºª�[��Æ, ‘a hymn containing pure and unobjection-able matter’.

33 See below p. 216. It was, of course, distinct from the surviving collectionof Orphic Hymns, which are later; cf. West (1983), 28–9.

34 Phanes or Protogonos in other Orphic theogonic cosmogonies. This is not theplace to examine details of the Derveni theogony, for which see West (1983), ch. 3.


commentator treats the Orphic text as a venerable theological orhieratic text, hardly as a contemporary poetic work. That the poemgoes back to the sixth century, that is, to the main period of circula-tion of rhapsodic ‘Homeric hymns’ is hardly a flight of fancy.35

The second text is Aristophanes’ mock bird theogony in Birds(685–703).36 The chorus of birds first invoke the ‘dearest of birds’,the nightingale, to accompany the ‘hymns’ which they are about tosing (678–9 �� H� KH� o�ø�). The role of the nightingale iscomparable to that of the Muse(s) in conventional hymnology, in-voked to assist and inspire the song. Then the bird chorus begin theanapaestic section of the play’s parabasis. They introduce themselvesas the ‘immortal, eternal gods, ethereal, ageless and unfathomablywise’, and then present an account of their origin (688–702). Theo-gony began, they say, when black-winged Night conceived a ‘wind-egg’in the deep caverns of Erebos, from which winged Eros hatched. Inhis gleaming plumage he mated with another winged partner, Chaos,from which union the race of immortal birds arose (694–9). Thechorus claim they preceded the Olympian gods in the theogonicorder. They are the true children of Eros, winged and gifted withflight, companions of lovers (704). The song is a playful lampoon onOrphic theogonies, designed, the birds say, as riposte to the sophistProdicus (692 —æ��Œøfi �ÆW K�F Œº��Ø� �Y�Å�� e º�Ø��, ‘tell Pro-dicus from me to get lost in future’). We gather from this thatProdicus, too, specialized in the interpretation of Orphic hieroilogoi, which the birds aim to match with their playfully subversivetheogony.37

At this point we might glance at some fragmentary remains of the‘pre-historic’ poets mentioned before. Olen was said to have been themost ancient composer of hymns.38 A Lycian himself, he is said to

35 Bernabé ii. 1, p. 10 ‘theogoniam Orphicam saec. VI a. Ch. in papyro laudatam’.36 For pertinent scholarly literature cf. Bernabé ii. 1, pp. 73–5 (fr. 64 V).37 Cf. Guthrie (1971), 238, 274–80. Our sources do not mention a specific theog-

ony propounded by Prodicus, nor a learned work on ‘Orphic’ or other theogonies;they do accredit him, however, with a theory about the origin of religion: men came toworship useful and life-sustaining things and called them by gods’ names, e.g.Demeter for bread. See S. E. M. 9. 18 and Phld. Piet. ¼ P. Herc. 1428 fr. 19 (Henrichs,1975b).

38 Boio, a Delphic poetess quoted by Paus. 10. 5. 8: Ὠº�� Ł! , n ª�� æH�� +��Ø� �æÆç��Æ , j �æH�� ! IæåÆ�ø� o�ø� ��Œ��Æ�! I�Ø�� (‘Olen, who was thefirst divine speaker of Phoibos [Apollo], and first composed the strain of ancienthymns’). For Olen cf. AHS lv–lvi, Furley and Bremer (2001), i. 146–51.


have composed hymns for the Delian gods. In particular, his compos-itions celebrated the birth of Apollo and Artemis and the advent ofother goddesses to Delos (Eileithyia and Hera) to attend the birth.39

This will become relevant in the discussion of the Homeric Hymn toApollo below. The Attic Pamphos is said to have composed hymns forthe Attic cult of Demeter and Kore at Phlya.40 A hymn to Demeter iscited repeatedly by Pausanias.41 The nature of his hymn(s) is indi-cated by a remark in Philostratus to the effect that ‘Zeus is life-producing; all things emerge from the earth through him’.42 Thispoints clearly to a cosmogonic function of Zeus in the hexameterverses quoted, in a manner comparable with the Derveni theogony.

The collection of early ‘Orphic Hymns’ cited at one point by theDerveni author is also mentioned by Pausanias (9. 30. 12)43 and bythe fourth-century Atthidographer Philochoros, who cites the ‘hymnsof Orpheus’ for corroboration of the identity of Demeter and Hes-tia.44 The passage in the hymn may have been theogonic, but that isnot certain. Pausanias characterizes these ‘Orphic Hymns’ as ‘less wellembellished by words than the Homeric Hymns but more devotionalthan these by virtue of their sanctity’.45 It is a comparison we will bereturning to.

Musaeus is sometimes depicted as a kind of disciple of Orpheus.Pausanias says that his hymn to Demeter contained the informationthat Attic Phlyos was the son of Ge;46 a work attributed to Musaeus

39 Note also Pi. Pa. 12. 17 where Eileithyia and Lachesis are said to have cried aloudwhen Leto bore the twins; the mention of Lachesis may tie in with Pausanias’ remark(8. 21. 3) that Olen called Eileithyia �hºØ�� , ‘the good spinner’, obviously because sheis identical to Fate. Otherwise we learn from Pausanias that Olen called Eileithyia‘older than Kronos’ and ‘Mother of Eros’ (9. 27. 2).

40 For cult and archaeology cf. Marinatos (1951); Loucas and Loucas (1986),396–9; Parker (1996), 305.

41 1. 38. 3; 1. 39. 1; 8. 37. 9; 9. 31. 9 et al.42 Her. 693 ‹�Ø Z�f �YÅ �e ÇøØ�ª��F� ŒÆd �Ø! �y I��Æ�ÆØ �a KŒ �B ªB ��Æ.43 They were sung, Pausanias says, as accompaniment to the ritual of the Lykomi-

dai at Phlya in Attica.44 Cf. Obbink (1994): Œi� �:�E % .��Ø �! !O:æ:ç: [�f ] �Ææa +Øº�å�æøØ ªB� [Œ]Æd

˜�Å�æÆ �c� ÆP�c� % E��ÆØ (‘and in the Hymns Orpheus in Philochoros [says]that Demeter is the same as Hestia’). The fragment of Philochoros is from Phld.Piet. (p. 103 Schober) ¼ FGrH 328 F185. The mention of Orpheus’ hymns in theDerveni papyrus is in col. 22. 11–12.

45 9. 30. 12 Œ��øfi b� �c �H� K�H� ��ı��æ�EÆ çæ�Ø�� i� �� ª� % ˇ�æ�ı ��f o��ı , �ØB �b KŒ ��F Ł��ı ŒÆd K �º�� KŒ��ø� lŒ�ı�Ø. Cf. AHS lvii.

46 Paus. 4. 1. 5. For the minuscule surviving records about his hymns and hisrelationship to Orpheus, see Bernabé ii. 3, frs. 57–61.


told that (Eleusinian) Triptolemus was the son of Ocean and Ge.47

The Cretan Epimenides is certainly a historical figure, although hisprowess as prophet and healer became legendary.48 He is creditedwith both a Theogony (frs. 46–53 Bernabé) and a Birth of the Kuretesand Korybantes (Bernabé ii. 3, p. 159).

This brief survey is sufficient to show that, without exception, themythical (and semi-legendary) poets were held to have composedhexameter hymns for the gods, whose character we may in most casesdiscern as being theogonic. Most of the citations in later sources fromthese works tend to be of the form ‘according to X, a certain god/goddess was the son/daughter of xy divine parents’.

Diels–Kranz collected the fragments of the ‘mythical poets’ at thebeginning of their compendium of pre-Socratic texts. Hesiod, too, hasoften been termed ‘proto-philosophical’ in that his theogonic poemoffers a mythical account of cosmic, that is, physical, history. Theconnection is apparent in the Derveni author, too. His work focuseson the Orphic theogonic poem precisely because it offers an accountof the origins of the world, albeit couched in theological terms(ƒ�æ�º�ª�E�ÆØ, ‘he theologizes’, in e.g. col. 7. 7) which the authorreinterprets in accordance with his own ‘enlightened’ world view.The theogonic hymn, in short, may be seen as incipient philosophy.In some of the pre-Socratics we find comparable hexameter hymnsserving as preludes to philosophy. Empedocles seems to have com-posed a ‘hymn to Apollo’, which Zuntz has interpreted as germinal inhis philosophical system.49 Parmenides prefaces his Way to Truthwith a proem in which cosmic personifications take on mythicalstatus. The legendary figure Linos, too, is said to have composedcosmogonic hexameter poems on the origin of the universe and thecreation of living creatures.50

The Homeric Hymns should, in my opinion, be seen against thisbackground. Since only they among epic hymns of the archaic periodsurvived intact in the manuscript tradition, it is easy to assume thatthey typify, not to say constitute, the genre. I hope the precedingparagraphs are sufficient to show that this was not the case. On the

47 Paus. 1. 14. 3 ¼ fr. 60 F Bernabé, where Pausanias expresses uncertainty aboutthe author of these particular hexameters.

48 Testimonia and fragments are collected in Bernabé ii. 3, 105–68.49 Zuntz (2005), 1–25; cf. Solmsen (1980).50 Bernabé frs. 78–93.


contrary, the dominant type of hexametric hymn to the gods in thisperiod is likely to have been a theogonic poem narrating the geneal-ogy of gods and how they came to assume present powers. Interest-ingly, Clay has argued that this is precisely the deeper significance ofthe Homeric Hymns: they chart the emergence of the Olympian godsunder Zeus’ dominion, showing how powers and privileges wereapportioned among them.51 In fact it was the theogonic hymnwhich aimed to show a god’s ancestry, place in a genealogical scheme,and family relations within a divine pantheon. Divine genealogy wasthe way of establishing a god’s credentials, his or her place in thecosmic pecking order. We will now turn to the Homeric Hymns andsee how they adapted and modified this traditional scheme no doubtin response to different aesthetic and performative considerations.

The genealogical, theogonic model is by no means absent from theHomeric Hymns. A count shows that over half of the extant collectioncontained an account of the god or goddess’s birth.52 Even thisaccount, however, differs from that given in theogonic hymns. Onemay compare Hesiod’s account of the birth of Aphrodite with thesixth Hymn to the same goddess. Hesiod is interested in placingAphrodite’s birth in the cosmic scheme of things.53 Gaia castratesOuranos and flings his severed organ into the sea. The foaming organgerminates the goddess Aphrodite who is washed up on the coast ofCyprus. She steps onto land and grass grows under her feet. Hesiodexplains that this myth accounts for two appellations of Aphrodite,Cythereia (from Cythera, where she beached), and Cyprogeneia(from Cyprus). He adds that Eros and Himeros became her com-panions and that she acquired from the beginning the power over‘maidenly talk, smiles and deceits, pleasure and sweet and enjoyablelove’ (205–6, �ÆæŁ��ı �’ O�æ�ı �Ø��Æ�� ’ K�Æ��Æ �� j �æłØ�� ªºıŒ�æc� çØº��Å�� ØºØå�Å� ��). Hesiod has told us precisely

51 Clay (1989), and in this volume (Ch. 11).52 I count eighteen hymns out of thirty-three which tell of the birth or parentage of

the god/goddess hymned; in nine of these eighteen the birth myth is a major narrativemotif: Dion., Apoll., Herm., Hys. 6 (Aphrodite), 16 (Asclepius), 17 (Dioscuri), 19(Pan), 28 (Athena), 31 (Sun). Only one, the last mentioned to Helios, is what onemight call a ‘genealogical hymn’: the author is concerned to trace Helios’ lineage backthree generations. Of the eighteen Hymns with a god’s birth one (32 Selene) mentionsthe birth of another god, not the Hymn’s recipient but her own offspring Pandeia byZeus. Cf. Clay in this volume (pp. 241–2) for divine birth as a prominent motif of theHymns.

53 Hes. Th. 173–206.


Aphrodite’s parentage, the circumstances of her birth, her companiongods and her area of potency.

When we compare the sixth Hymn with this we see that the authorpicks up on Hesiod’s account when he narrates that the West windblew Aphrodite onto the coast of Cyprus ‘amid soft foam’ (5). Heomits all mention of how this foam was generated. Instead helaunches into an ecphrasis of the goddess’s dressing and adornmentby the Horai when she emerged on land (5–15).54 The passage isessentially an elaboration of Hesiod’s line in which he says that grassgrew under the nascent goddess’s feet. The Homeric author thennarrates how the Horai led Aphrodite up to the assembled companyof gods, who admired the sight of her and each male god vowed hewould like her as wife. Then the author signs off, with an explicit wishthat Aphrodite should grant him ‘victory in this competition’ by‘supporting my song’ (19–20 �e �! K� IªH�Ø j ��ŒÅ� �fiH�� çæ��ŁÆØ,Kc� �! ��ı�� I�Ø��).55 The author has concentrated on the aes-thetic quality of Aphrodite’s appearance, both at the moment of herbirth and when she was introduced to the gods. He omits all mentionof her parentage and her divine powers. True, these were known tothe audience. The point I am making is that we can see the author’sprocess of selection at work here in his composition. His strengths areecphrastic embellishment of a narrative moment. One might call thetwo scenes he evokes—Aphrodite’s first appearance on land, then herappearance among the gods—tableaux. The second also contains thegerm of dialogue, an essential feature of epic,56 when the male godsmutter under their breath in admiration of Aphrodite’s beauty. Thescene is reminiscent of Demodocus’ lay in Odyssey 8 when the godsstand around looking at the spectacle of Aphrodite trapped in bedwith Ares (321–32). In short, the author disassociates his narrativefrom the cosmic order established in Hesiod’s Theogony and concen-trates on episodic embellishment. Theogony and cosmogony havebeen truncated leaving only a pretty scene.

54 Cf. Boedeker (1974).55 For the endings of the Hymns generally, see Capponi (2003). On the reference to

a competition in this hymn cf. in this volume Clay (p. 236), Nagy (pp. 322–5), andCalame (p. 355). West (1992), 19 n. 25 suggests the panegyris at Old Paphos as apossible setting.

56 Cf. J. Griffin (1986); on speech in the Hymns, see Faulkner (forthcoming c).


This analysis can be extended to the major fifth Hymn to Aphro-dite.57 Here the author abandons theogony altogether, concentratinginstead on the power over love which Aphrodite has received as herlot. His treatment focuses, however, on the reduction of Aphrodite’sabsolute power over gods, men, and animals to make them all coupleat her will, sometimes with degrading partners. The narrative, onemight say, aims at refinement of the plain theogonic statement thatAphrodite has received sexual love as her divine power. It turns outthat the uncontrolled wielding of this power is an irritant to Zeus’dominion. He himself is made to fall in love with mortal women, tothe vexation of Hera, and other gods too engage in demeaning unionswith mortals, while Aphrodite herself remains immune. The authortells a story reflecting a plan by Zeus to limit Aphrodite’s power. Sheis made to fall in love with a mortal herself, the Trojan Anchises, andgive birth to a mortal child, Aineias (45–52). Her distress at thishumiliation means that she will in future no longer be able to wreakhavoc among the gods by making them pair with mortals indiscrimin-ately (247–55). In view of this, Clay and others have pointed out thatthe hymn’s narrative dovetails with the Hesiodic scheme in that theage of heroes came to an end according to Zeus’ will when godsceased joining with humans in love.58 The taming of Aphrodite inthis Hymn seems to support a cosmogonic narrative according towhich intercourse between gods and humans ceased at a given point,ushering in the grim age of iron. The view has not gone unchallenged,however. As Faulkner argues, the Hymn indeed narrates a humilia-tion of Aphrodite leading to greater circumspection on her part in thefuture. Whether that amounts to cessation of relations between godsand humans for ever after is, however, not said. There is a differencebetween a mildly chastised Aphrodite and one whose powers to causeexistential anarchy have been drastically curtailed.59

According to my analysis, the Hymn’s narrative is secondary totheogonic hymns. Aphrodite’s power to cause love is the datum of thestory; its relativization is the author’s theme. From representing acosmic force capable of moving the hearts of gods, humans, and beastsat will, Aphrodite becomes a fallen queen. The Hymn’s author picks

57 On this Hymn see now Faulkner (2008) and in this volume Brillet-Dubois(Ch. 6). See also Bergren (1989); Pirenne-Delforge (1989).

58 Clay (1989), 167–9.59 Faulkner (2008), 10–18.


up a minor mythical motif—the birth of Aeneas—which in turn islikely to be secondary to the other great romance on Mt Ida (thejudgement of Paris), in order to show Aphrodite involved in an (alltoo) human drama, a sordid affair almost. The narrative containsmany epic elements—travel and adventure, dialogue with deceitfulspeech, human susceptibility to divine will, the nature of mortality—indeed the Hymn has been called ‘the most Homeric’ in the collec-tion.60 Comparison with the model of theogonic hymns shows justhow far the author has departed from the hymnode’s usual remit.Instead of glorifying the goddess, he has shown how Zeus shamedher. Instead of establishing her primeval power, he has described itslimitation. He has completely omitted her descent. Schmid’s observa-tion that Aphrodite in the Hymn is like a grand Ionian lady stoopingto folly with a social inferior is not so wide of the mark.61 At any rate,the composition is as far from cultic praise of the goddess as it couldbe.62 The appeal is to the emotionality of the piquant union betweenAphrodite disguised and the good-hearted but helpless Anchises.

Particularly relevant to my discussion is the third Hymn toApollo.63 Here the Hymn composer indeed takes up the theme ofthe god’s birth, but, as I will show, he does this in characteristically‘Homeric’ fashion, distinct from the standard genealogical form ofpraise. Apollo’s birth is, in fact, one of the ‘great’ hymnic subjects. Inmany extant compositions and fragments we hear of Leto’s deliveryof the divine twins on Delos, an event which is accompanied by a kindof pantheistic celebration on earth. The elements hold their breath, asit were, while the twins are born; nature blossoms and earth and itsrivers gleam golden and silver. The birth moment is celebrated as anepiphany of two mighty gods, children of Zeus himself and Leto.64

The author of the Homeric Hymn embellishes this traditional mythin multiple, revealing ways. In a nutshell, he extracts the birth mythfrom its theogonic context—children of Zeus outside his ‘prime’

60 e.g. in Gottfried Hermann’s Leipzig edition (1806): Homeri nomine dignissi-mum.

61 Schmid (1929), I 239–40.62 It is interesting, however, to note the author’s hymnic enthusiasm when describ-

ing Aphrodite’s arrival at Mt Ida (69–74): all nature falls in with her wanton spiritwith literal enthusiasmos.

63 The literature on this Hymn is vast; see in this volume Chappell (Ch. 4).64 See Furley and Bremer (2001), nos. 2. 1, 2. 6, 3. 3 with further literature;

Rutherford (1988).


union with Hera—and amplifies it into a complex drama told interms of human emotions and rivalry. The emphasis turns fromsimple worship of the child god in his astonishing infantile glory toa narrative whose human interest will draw in the assembled (human)audience. The narrative is built around two different confrontations:that between Leto and Delos and that between Hera and the othergoddesses. The exchanges between these paired antagonists involvetravel, persuasive speech, and trickery (Iris tricks Hera by negotiatingwith Eileithyia privately)—all elements in the epic poet’s repertoire.Leto must persuade Delos to accommodate Apollo’s birth althoughthe island has heard that Apollo will be a wild and unruly god (49 ff.);the other goddesses must outwit Hera who, through jealousy of Zeus’strapping son, alone impedes the birth by restraining Eileithyia�ª��Œ� , divine aide in childbirth (97–101). The emotionsinvolved in the first exchange are Delos’ and other lands’ trepidation(47 �æ�ø, ‘tremble’) at the prospect of harbouring Apollo, over-come in Delos’ case by the promise of rich reward from Apollo’sworshippers who will flock to the island if it is home to his favouritecult (56–60). The author’s main invention here concerns Delos her-self, who becomes a living being capable of trepidation and kerdosunē,profit-lust. The theme is successful as literary narrative, lending as itdoes human interest and suspense to the question of where Apollowill be born. But it distracts from the genealogical core of the story:Delos moves into centre-stage instead of Leto, the mother, and,indeed, Zeus, the father. The second drama—that played out betweenthe other goddesses and Hera—similarly raises suspense but distractsfrom the Olympian status of Apollo. Hera is against his birth throughreasons of human jealousy and must be tricked by Iris into releasingEileithyia. The narrative as a whole is light years from Hesiodic orOrphic theogony. A simple account of who bore whom to whom isembellished by an extended cast, rapid movements, canny speech,faction, and deceit. This is surely a tale told for human entertainment(and edification) rather than cultic worship.

Uniquely, the Hymn appears to take time to describe its intendedaudience and occasion. Having established that Delos is Apollo’sfavourite island, the author goes on to describe the assembly ofIonians who meet to celebrate his cult with an agōn including boxing,dancing, and song (149). He describes the fine appearance of thecongregation; how, on this day, they appear ‘godlike and ageless’(151). The remark fits the Greek convention whereby the charis of


worship, its charm and beauty, is seen as a means of obtaining divinecharis, grace or favour, by reciprocal exchange.65 The author thendescribes one particular performance at the Ionian festival: the songdance of the Deliades, Delian Women, who sing hymns to Apollo,Artemis, and Leto, to long-dead previous generations of men, in thecourse of which they imitate the voices and ‘hubbub’ of all races ofmen. This last is likely to be a reference to a song describing thecoming of the Hyperboreans to Delos, a legendary northern people,who, according to myth, travelled through all countries on theirpilgrimage to Delos to bring the sacra to the god.66

The author of the Hymn then addresses the Deliades personally(165–78). He asks them to remember him in the future; when askedby travellers who their favourite itinerant bard is, they are to answer: ablind man from Chios whose songs always win the prize. Clearly theallusion is to Homer. Less clear is the position taken by the hymnodehimself: he may only be inferring that this song is a work of Homer,not that he, the itinerant bard, is Homer. An ancient tradition heldthat Cynaethus of Chios forged works of Homer, among them theHymn to Apollo itself.67 It is a nearly compelling conclusion that theHymn’s author (whoever he was) composed the hymn to DelianApollo for performance of the pan-Ionia at Delos.68 He recommendshimself and the works of the ‘blind bard’ to the Deliades just afterpraising their own performance at the Delian Games for Apollo.Moreover, the praise of the appearance of the Ionians at their majorDelian festival is surely intended as a captatio benevolentiae. Giventhat, we may see in the Homeric Hymn to Delian Apollo a competi-tion piece for performance at Apollo’s festival before a large audienceintent on multiple sources of entertainment: sport, dance, and song. It

65 Cf. Furley and Bremer (2001), 1. 61–3, Parker (1998), Bremer (1998), Scott(1983).

66 Cf. Hdt. 4. 35; Paus. 5. 7. 8; 9. 27. 2; 10. 5. 7. Cf. Sale (1961), Furley and Bremer(2001), 1. 142–51; ibid. 151 for the suggestion that the Œæ��ÆºØÆ�� , lit. ‘rattlingwith castanets’, or �Æ�ÆºØÆ�� (a variant reading in the mss.), ‘chattering’, of theDelian women reflected the foreign dialects encountered along the ‘via Hyperborea’.

67 Sch. Pi. N. 2. 1c III 29, 9–18. Dr. Burkert (1979) defends the content ofthe scholion to the extent that Cynaethus may indeed have popularized theHomeric poems in Sicily as late as the sixth century BC. Cf. West (1975). See furtherin this volume Chappell (pp. 71–3), Nagy (pp. 288–91).

68 Burkert (1979) suggests that the piece was composed for Polycrates of Samos’Delia kai Pythia festival at Delos in 522 BC; cf. in this volume Nagy (p. 286). For adiscussion of the unity of the composition, cf. Chappell in this volume (Ch. 4).


is precisely the audience and occasion which the embellishment ofcult myth I have pointed to in the Hymn would lead us to expect.69

We see a marked departure from traditional theogonic poetry to caterfor the sophisticated taste of an Ionian audience brought up onHomer. Incidentally, the Hymn will have found an interesting foilin the traditional hymns to Apollo, Artemis, and Leto—ascribed toOlen, no doubt—recited by the Deliades. Whilst paying due respect tothe patron god of the festival, Apollo, the author of the Hymn shoneby his narrative inventiveness, his personalized treatment of myth, inshort its ‘Homerification’.

The best example of what one might call ‘mythical expansion’ of atheme, in this case the birth of a god, comes in Hermes.70 Here we findthe basic mythical occurrence—the birth of Hermes by a nymphMaiato Zeus71—amplified into a lengthy drama with many twists andturns, humour, irony, invention, direct speech, and forensic appeal.The birth itself is narrated in Homeric terms as a ‘Seitensprung’72 byZeus with Maia who lived in a remote cave out of sight of Olympus.Zeus visited her while Hera slept (8), a reversal of the Dios apatē in Il.14. The author salutes Hermes’ arrival in the world with a spateof epithets all emphasizing his tricky, deceitful, and wily character(13–16). The epithets read almost like a satire of traditional strings ofepiklēses with their honorific titles. On birth, Hermes immediatelysprang into action. The author is concerned to show the amazinginfantile energy, ambition, and resourcefulness of the god. On thevery day of his birth he has invented the kithara by midday and stolenthe cattle of Apollo by evening (17–19). Instead of remaining quietlyin his crib, he immediately sets out to look for his big brother’s cattle.On setting foot outside Maia’s cave he encounters a tortoise amblingpast the entrance (22–4). The encounter might be described as aplayful skit on high epic style.73 Hermes addresses the creature asone might a respected comrade in arms, only to ‘excise his lifespan’

69 Cf. Barner (1977) for some theoretical remarks on the role of audience expecta-tions for poetic form and content. One of his examples is Demodocus’ song of Aresand Aphrodite in the Odyssey.

70 On this Hymn, see in this volume Vergados (Ch. 5).71 Hesiod’s Theogony devotes a mere couplet to the event, containing precisely

these bare bones of information, plus Hermes’ job as messenger of the gods (938–9).72 Engl. ‘extramarital affair’ but with connotations of casual and short-term plea-

sure.73 On this episode, see Shelmerdine (1984).


(ÆNH�! K��æÅ��, 42) abruptly in order to turn his shell into a lyre.Hermes’ speech to the tortoise is full of gleeful but also sadistic wit.‘Lovely creature,’ he addresses the tortoise, ‘chorus-member, compan-ion of the feast’ (åÆEæ�, çıc� Kæ��Æ, å�æ�Ø�� ÆØ�e ��Æ�æÅ, jI��Æ��Å �æ�çÆ��E�Æ, 31–2): the terms are maliciously ironical as thetortoise will not enjoy future dancing and feasting but will forfeit hisshell as an instrument. Hermes points out to him that it is better to bedead than alive as, living, he will function only as a charm againstsorcery, whilst dead he will resound beautifully. The killing itself isdescribed with a startling aesthetic oxymoron: Hermes carries the‘lovely plaything’ indoors, then ‘gouges out its life with a knife of greysteel’ (KæÆ��Ø�e� ¼ŁıæÆ, 40; ªºıç��øØ ��ºØ�E� �Ø��æ�ı j ÆNH�!K��æÅ��, 41–2). The contrast is between Hermes’ appreciative speechand the lumbering speechless creature, as well as with the cheerfulcallousness of the actual killing. My point here is that the narrativehas leapt, like Hermes himself, from the god’s incunabula to a miniatureepic encounter with a very unheroic opponent. Not only has the authorembellished the birth myth with a lively and piquant narrative in epicstyle, he has even capped epic klea andrōn with parody. It is no greatstep from here to the mock-Homeric Batrachomyomachia.74

It is true that the subsequent narrative of Hermes’ theft of Apollo’scattle and the adjudication before Zeus results in a reconciliationbetween Hermes and his older brother, which may be construed as astep in the establishment of the ‘politics of Olympus’. But the tone of thenarrative is throughout witty and tongue-in-cheek. It is as if the authorwere playing on tradition for literary effect rather than legislating forrelations on Olympus. The listener or reader, like Zeus himself (389),has to smile at the god’s youthful antics and deceitfulness as one smiles,for example, at the machinations of a servus callidus in comedy. Theaesthetic effect is a long way from awe and respect for the god’s numen.

Finally, narrative amplification of a god’s birth myth is well-illustrated by the surviving opening section of the first Hymn in thecollection, to Dionysus. A Geneva papyrus augments to some extentlines quoted by Diodorus.75 The poet begins with a kind of priamel,

74 On the sophistication of the language of this hymn as an index of fifth-centurycomposition see Görgemanns (1976); cf. in this volume Introduction (pp. 12–13).

75 D. S. 3. 66. 3, citing from › ��ØÅ�c K� ��E % .��Ø , ‘the poet in the Hymns’;Hurst (1994). On the reconstruction of the text see West (2001a) and in this volume(Ch. 2).


stating that there are rival accounts of where Dionysus was born:‘Some say in Drakanos, some in windy Ikaros, others in Naxos, Ochild of Zeus, Eiraphiotas, some beside the River Alpheios with itsdeep eddies; others, Lord, say you were born in Thebes, but they arewrong’ (Dion. A 1–6 �Q b� ªaæ ˜æÆŒ��øfi �’, �Q �’ !"Œ�æøfi M��fi Å jç��’, �Q �’ K� ˝��øfi , �E�� ª�� ¯NæÆçØH�Æ, j �Q � �’ K�’ �ºç�ØfiH��ÆfiH �ÆŁı�Ø��Ø j ¼ºº�Ø �’ K� ¨��fi Å�Ø� ¼�Æ� �� ºª�ı�Ø ª��ŁÆØj ł�ı��Ø).76 The author immediately states his preference for aversion which places Dionysus’ birth from Zeus on a rocky island‘far from Phoenicia and close to the streams of Egypt’ (A 9–10). Theline with which the author launches into the birth narrative remindsone of a number of Odyssean descriptions of rocky islands rising fromthe sea.77 There follows quite an elaborate ecphrasis of the un-approachable island—it has no harbour for ships, but is a high, sheerrock all round. Plant life is abundant.78 At this point the papyrusbecomes scrappy although it is clear that the description continuesfor at least nine more lines. Some individual words may be made out:something ‘extends’ (A 16 ��ı[��ÆØ; cf. Od. 9. 116), probably theisland itself at a distance from the mainland; some place on it isprobably ‘away from the roar of the sea’ (A 18 I�e çº��: [��Ø�ŁÆº��Å ); in A 21 there may be mention of nymphs as ‘attendants’(IçØ]�:�:º:ø�); in A 23 we hear of ‘pleasant meadows’ (�:æÆ��Ø ��:�: [�)and A 24 may refer to the shade ‘under thickets’ (��e �æı:�: E: : : ��e�æı:�: Ø:� West). A line quoted by Crates (of Mallos?) referring to the‘lush growth of vines’ might also come from this section of theHymn.79 What survives is enough to show that the author took care

76 Text in West (2001a), 10, omitting line 5, Œı�Æ�Å� $�ºÅ� ��Œ�Ø� ˜Ød��æ�ØŒ�æÆ��øfi , ‘[they say] that Semele having become pregnant bore to Zeus thun-der-wielder’; the Geneva papyrus and some mss. of Diodorus omit this line and it doesnot suit the context well, as the author goes on immediately to say that Zeus boreDionysus (sc. from his thigh).

77 Line 9 ��Ø � �Ø ˝��Å o�Æ�� Zæ� , I�Ł�� oºfi Å . . . ‘there is a high and verdantmountain, Nysa’; cf. e.g. Od. 4. 844 ��Ø � �Ø �B�� fi Å ±ºd ��æ��Æ, j ��Åªf !"Ł�ŒÅ �� $��Ø� �� ÆØ�Æº��Å , ‘there is a rocky isle in mid-ocean, lying betweenIthaca and rocky Samos’.

78 Lines 11–14; supplements to the papyrus are made possible by an almostverbatim copy of the lines in Orph. Arg. 1199–202.

79 Dion. B ¼ Crates ap. Ath. 653B, K� ��E % .��Ø ��E IæåÆ��Ø : ÆP�Bfi �Ø ��ÆçıºBfi �Ø�ºÆ��fi Å�Ø� Œ��ø�� , ‘in the Ancient Hymns: “laden with their bunches of darkgrapes” ’; cf. West (2001a), 8.


to describe the scene of Dionysus’ clandestine birth away from Hera’sjealous eye with considerable Homeric enargeia.80

Only the end of the Hymn is reliably transmitted in anotherquotation by Diodorus. Zeus acknowledges the inauguration of Dio-nysus’ trieteric cult on earth. He nods his head in assent and Olympusquakes. The poet then takes leave of Dionysus with a reverent bow(Dion. D 6–12). Between the god’s birth, however, and his instate-ment on Olympus a narrative of several hundred lines has been lost.81

West, following Wilamowitz and, before him, Bergk, has argued thatthe myth was most likely that of Dionysus’ success in persuadingHephaestus to return to Olympus in order to free his mother from therigged throne which he had sent her.82 This myth involved a web ofintrigue and rassentiment among the Olympians. Hera ejected herchild Hephaestus from heaven for his deformity. He fashioned athrone for her which trapped her in chains when she sat on it. Aresfailed to force Hephaestus to relent. Dionysus succeeded in the mis-sion by giving his boon to Hephaestus, wine, and escorting him,drunk, on a donkey up to Olympus. There, it seems, Aphroditeawaited Hephaestus as bride. The scene is memorably depicted onthe François Vase by Kleitias (Fig. 1, see Chapter 2, p. 32). Thisnarrative, then, took as its starting point the clandestine birth ofDionysus on a remote rocky isle called Nysa and followed the god’srise to power through a truly Olympian upheaval: the chaining, andrelease, of Hera. One can see how Hera’s jealousy at Dionysus’ birth(from Zeus’ thigh, without Hera’s agency) will have been appeased bythe good offices of Dionysus in persuading Hephaestus to releaseHera. The author apparently succeeded in tying in his version of thegod’s birth with a major myth describing Dionysus’ adult prowessand his instatement as a major Olympian. A primitive story of thewine god’s birth from a father’s thigh has grown into a full-scale

80 Dion. A 8 Œæ��ø� º�ıŒ�º�� æÅ�, ‘secretly from pale-armed Hera’.81 West (2001a), 1 calculates the whole length of the hymn as 411 lines.82 Cf. P. R. 378d and Euthphr. 8b4; the reference there to something which ‘pleases

Hephaestus but displeases Hera’ must also be to the ‘chains of Hera’. It is not certainthat P.Oxy. 670 contains a section of this narrative in which Zeus addresses Hera,although West’s reconstruction of text and sense is persuasive; see in this volume(Ch. 2). For a possible objection, see Faulkner (2010a).


family drama,83 one which inspired a series of later poetic treatmentsand painted pictures.84

It is, as Pausanias said, that the Homeric Hymns have more ‘orna-ment of words’ than Orphic (and other) hymns, which have more‘devotional piety’. In the case of Dionysus’ birth myth we are in aposition to compare the ‘Homeric’ treatment with the Orphic. Clem-ent of Alexandria and a number of other authors give a version of anOrphic hieros logos which no doubt found expression in a hexameterpoem attributed to Orpheus, i.e. a ‘hymn’.85 Dionysus was held tohave been born of a strange incestuous union, namely that betweenZeus and Persephone, who herself was the daughter of Zeus and hismother Demeter. So, according to this version, Persephone was Zeus’wife, daughter, and half-sister, making Dionysus his son, grandson,and nephew. The Orphics were not content with this lurid drama,however, reminiscent of some recent incestuous goings-on in anisolated mid-Western farmstead, but inflicted a second birth onDionysus more traumatic even than the ‘orthodox’ birth from Zeus’thigh.86 The Titans enticed Dionysus within their grasp by means oftoys. Then they dismembered him and cooked the body parts first in acauldron then on spits held over flames, but not before Athena madeoff with the still beating heart. Zeus blasted the Titans for theirmisbehaviour and gave the dismembered Dionysus to Apollo forburial. Apollo buried the parts on Parnassus but the god was broughtback to life by Rhea,87 no doubt using the heart which Athena hadsaved.88 It is a pity we do not have the Derveni author’s allegoricalinterpretation of this Orphic myth. One wonders what cosmic prin-ciples the dismemberment of Dionysus might have symbolized. As itstands, however, the narrative contrasts starkly with the ‘Homeric’birth-myth of Dionysus in the first Hymn. Orpheus operates with a

83 It occurs to me: in the manner of a vine spliced onto a vine-stock in a standardviticultural process.

84 Alc. fr. 349a–e Voigt; Pi. fr. 283 SnM; Epich. ˚øÆ��Æd j �AçÆØ�� ; AchaiosTrGF 20 F 16b–17.

85 Clement Protr. 2. 17. 2, quotes two hexameters from Orpheus’ '�º�� in thisconnection.

86 For sources, see Bernabé ii. 1, pp. 50–68.87 Where ‘Dionysus’ tomb’ was shown in Apollo’s sanctuary: Plut. Mor. (De Is. et

Os.), 365a3. The myth is memorably treated at the beginning of Jane Harrison’sThemis.

88 Note the ritual ‘waking’ of the god performed annually at Delphi: Plu. Mor.(De Is. et Os.), 365a6.


gruesome set of symbolic actions constituting, one might say, dis-tinctly adult viewing. It is not the subtleties of human relations whichcharacterize these divine agents but rather a ghoulish manipulation ofbody-parts more reminiscent of Frankenstein’s monster. The differ-ence seems to me to reflect clearly the different audiences for whichthe myths were intended: the ‘Homeric’ version is an aestheticallypleasing narrative designed to give its human audience insight intothe world of the mighty Olympians. The Orphic narrative seems to beintended for an audience of initiates who will understand the sig-nificance of these arcana.89

In this chapter I have been arguing that archaic Greek hexameterhymns to the gods commonly took as their basic material a mythicalaccount of the gods’ genealogy (= theogony). To attribute birth andparentage to a god was to establish both his/her existence and theirrelations to preceding and present generations of divinities. It was toidentify a particular divine power within a complex of divinities gov-erning the world using the model of human genealogy. The gods areconceived as a family, with ancestors, a pater familias, his first lady, hischildren from this union and others, and their children. Herodotus atone point says that the hymn (literally ‘incantation’) sung by a Persianmagos at sacrifice consists of nothing but theogony.90 No doubt Her-odotus records this detail because the custom contrasts with Greekusage, in which hymns, including those sung during worship, do notconsist solely of genealogical lists but contain praise and embellishmentof various sorts. It shows, however, that, at its most basic, a song ofworship could consist of the plain recital of gods’ ancestry.

I began by showing that for Hesiod, the gods are most pleased tohear a hymn about their own birth and ancestry. His Theogonyproceeds to satisfy that requirement. Similarly the ancient collectionof Orphic hymns (quite distinct from the extant collection) appears tohave consisted of similar theogonic hexameter verse, albeit with adifferent constellation and sequence of divinities. My examples werethe hymn of ‘Orpheus’ quoted by the Derveni author, a collection of

89 Note the distinction in P. Prt. 316d7–9 that Homer and Hesiod practisedpoetry whilst Orpheus, Musaeus et al. practised ‘initiations and prophecies’ (��º��a ŒÆd åæÅ�øØ��Æ ).

90 1. 132: �ª� I�cæ �Ææ��g K�Æ��Ø Ł��ª��Å�, �¥Å� �c KŒ�E��Ø ºª�ı�Ø �r�ÆØ�c� K�Æ�Ø�� (‘a priest standing by sings an incantatory theogony, as those peopleexplain their incantation to be’).


(lost) Orphic hymns to which a number of ancient authors refer,Aristophanes’ parody of an Orphic theogonic hymn in Birds as well asnumerous other lost hexameter poems by legendary poets.91 Againstthis background the Homeric Hymns can be seen to have successfullycontrived new departures. They, too, frequently take up the theme ofa god’s birth, but their telling of the birth receives a different emphasisand focus of interest, and the birth myth itself is often the springboardfor a longer narrative of how the god developed after his/her birth.From bare theogony the Homeric Hymns transform the gods’ arrivalin the world into a kind of Bildungsroman,92 in which the young godsfind their place in the pantheon by undergoing trials and tests inhuman fashion. This humanizing of the gods’ early careers makesthem figures with whom a human audience can readily identify. Who,by contrast, could identify with the strange gods of Orphic theogo-nies? They served rather to shock and jolt as a tool in that disorienta-tion which seems to have formed part of initiation into a mystery cult.The humanization of the gods in the Homeric Hymns, on the otherhand, made Persephone a figure with whom daughters wrenchedfrom their parental homes by marriage at a young age could iden-tify.93 Hermes’ youthful antics will have appealed to many youngGreeks keen to challenge the authority of brothers and fathers. Like-wise the myth (probably) told about Dionysus in the first Hymn and(certainly) in the seventh Hymn contained actions towards which ayoung male might aspire: rescue of a mother, outwitting pirates. TheHomeric Hymns demystify and familiarize old and traditional divin-ities. The narratives popularize and humanize their stories so as toreduce the distance between the god and worshipper. This familiar-ization works in two directions: the gods are made human andhumans conversely godlike. When the poet of the third Homeric Hymn

91 Clay in this volume (pp. 244–5) distinguishes between hymnic and theogonicpoetry in that the latter treats ‘the pre-history of the Olympian order’. I do notthink this is a possible objection to my comparison in this chapter of ‘Homeric’with theogonic hymns, as I am using ‘theogony’ in its broad sense of ‘birth story ofa god’ without restricting the gods concerned to the pre-Olympians. Nor does Hesiodin his Theogony. He takes the story from the very beginning to the birth ofthe Olympians and some of their descendants.

92 Engl. ‘educational tale’ with emphasis on the hero’s development to maturity asa result of trials and experiences.

93 See Foley (1994), passim.


to Apollo describes the congregation assembled on Delos to worshipApollo he says that their beauty makes them appear godlike andageless (çÆ�Å Œ! IŁÆ��ı ŒÆd Iª�æø ��ÆØ ÆN��, 151). The sterntheogony of distant and unapproachable divinities still present inHesiod has become a company of gods temptingly like ourselves.


11

The Homeric Hymns as Genre

Jenny Clay

The thirty-three hymns, collectively known as the Homeric Hymns,form a highly heterogeneous collection of hexameter poems dedi-cated to various divinities. They range from three to five hundred andeighty lines and vary in date from the eighth to the second century BCE

and even later.1 Given the great diversity of their length and theirapparent temporal span, can we even speak of them as constituting agenre? Our ignorance of their provenance and circumstances ofperformance make it difficult to define them generically as a functionof occasion. Attempts to tie individual Hymns to particular cults orfestivals have met with only limited success.2 Even apparently obviouscases like Apollo and Demeter with their connections to specific cult-places and their founding have, on closer examination, proved prob-lematic. To be sure, the former includes a vivid description of theDelian festival, but it also—if one believes the poem is a unity—describes the equally important establishment of the god’s Delphicsanctuary.3 Demeter too links Eleusinian cult to a broader Panhellenic

This chapter revisits my earlier discussions in Clay (1989) and (1997) of the HomericHymns. While it summarizes previous conclusions, it also elaborates on some differ-ent aspects of that earlier work, especially the possible evolution of the genre. Forgeneral introductions to the Hymns, see now Richardson (2010) and WL 3–25.

1 For an overview on the dating of the Hymns, see the Introduction (pp. 7–16), thisvolume.

2 But cf. in this volume Richardson (pp. 50–3) and Chappell (pp. 64–7). It is worthnoting that Janko (1991), 12, has been converted to a unitarian view of Apollo.

3 See Burkert (1979) for an attempt to link Apollo with both Cynaethus andPolycrates’ Delian festival. On the Hymn’s disputed unity, see also the discussion ofChappell, this volume (Ch. 4).

framework and does not require performance in Eleusis. Some of thedivinities honoured within our corpus (e.g. Sun, Moon) do not evenhave cults; others are dedicated to heroes (Heracles, Asclepius, andthe Dioscuri). Various performance contexts have been proposed forour Hymns: rhapsodic contests at festivals, symposia, and entertain-ments in royal or aristocratic settings.4 All are credible, but no singlevenue seems to offer a generic definition.

Despite their label in our manuscripts, our collection is neither byHomer nor strictly speaking hymns, if we remember that in thearchaic period o�� simply means ‘song’.5 Nor do they qualify ifwe adopt the definition of Furley and Bremer in their collection ofGreek Hymns (‘songs used by the ancient Greeks in worship’), wherethe authors deliberately omit what they call ‘literary hymns’ includingthe corpus of the Homeric Hymns.6 In celebrating divinities, theirsubject-matter does indeed ally them with cult hymns, but in theirlanguage and diction, and their Panhellenism—precisely in their notbeing tied down to one locale or occasion—they form a subgroup ofhexameter epos. As a recent scholar puts it: ‘We should surely won-der, at the least, whether the Hymns, works designed to entertain andneeding no pious devotion to render them palatable, were necessarilyany more occasional or context-bound than was epic itself.’7 How-ever, certain shared characteristics, both formal and thematic (open-ing and closing formulae, relative predication, and description ornarrative) may point to collective generic expectations and, in anycase, led to their being collected and transmitted together, presum-ably during the Hellenistic period.8 Divorced from local cultic con-cerns, these compositions may have had diverse functions and diverseperformance settings that may have changed over time. Both internaland external evidence points to such an evolution while certain

4 For attempts to nail Hermes down to a specific occasion, see Johnston (2002),who connects the hymn to various Hermaia festivals; and Nobili (2008), who arguesfor a Panathenaic setting, which has also been suggested as a venue for many of theother Hymns. Yet if Athens was a focal point for performance of hexameter hymns, itis striking that we have no long hymn to Athena in our collection.

5 Cf. Ford (2002), 12 and n. 27, for a discussion of hymnos, along with examples.6 Furley and Bremer (2001), i. ix; cf. 43: ‘they are not cult hymns in any real sense.’

García (2002) attempts to define our hymns as ‘religious’ but not cultic. But cf. Clay(forthcoming).

7 Parker (1991), 1–2.8 For the formation of the collection, see in this volume Faulkner (Ch. 9).

The Homeric Hymns as Genre 233

characteristic features indicate a common conceptual space, which wewill attempt to define.

Let us begin from some rather basic observations concerning genreand ancient genres in particular.9 Early Greek poetic genres may bedefined according to various criteria: formal (e.g. metre, dialect),performance mode (e.g. monodic, choral, etc.), occasion (victorycelebration, symposium, religious festival, etc.), content (praise,blame, martial epic, stories about the gods). Clearly, these categoriesmay overlap in sundry ways, and audience expectations of the kind ofperformance they are going to witness may be reinforced or under-mined. Indications of genre, for instance occasion and performancemodes, may be internal, or evidence may involve external testimonia.Yet whatever system of classification we set up, whether tidy ormessy, must not be viewed as static. So along with a synchronicscheme of genres, we cannot ignore their diachronic evolution,which may move in tandem with developments in other genres.

The collection known as the Homeric Hymns shows unmistakablesigns of such an evolution, first of all in the huge disparity in length.No one, I think, would deny that scale is a generic marker. It is furtheragreed that at least some of the shorter Hymns may be abbreviationsof the long ones or that the long Hymns may constitute rhapsodicexpansions of the short ones.10 Yet we must remain open to thepossibility that some shorter Hymns are not truncated versions oflonger ones and that the latter may form an autonomous genre.11 Inany case, their variety in length offers an apparent indication ofchanging performance circumstances and needs.

Whether one begins from the long Hymns or the shorter compos-itions will inevitably shape the discussion and have a major impacton generic definitions.12 It is probably safe to assume that the

9 For general discussions of early Greek genres, see Harvey (1955), Rossi (1971),Ford (1992), 13–56, and (1997). On the Homeric Hymns in particular: Keyssner(1932), Lenz (1975), Janko (1981), Clay (1989), 6–16, 267–70, and (1997), Fröhder(1994), 1–90, Paz de Hos (1998), and Richardson (2010), 1–9.

10 Cf. AHS xcv, Càssola xii–xxi; Koller (1956), and Paz de Hos (1998) on the longHymns as expansions of the short ones; cf. Parker (1991), 14 nn. 2 and 3 for a list ofwhat he considers ‘abbreviated’ Hymns.

11 On the mid-length Hymns, see Fröhder (1994). I include the so-called mid-length Hymns among the shorter ones, even though one (Hy. 7 to Dionysus) offersa fairly extensive narrative as does Hy. 19 to Pan. But the shortest of the longHymns (Aphr.) with 293 lines is five times as long as the longest of the mid-lengthHymns (Hy. 7) with 59 lines.

12 Cf. above n. 10.

234 Jenny Clay

Alexandrian scholars who collected these compositions and createdthe corpus, ascribing them to Homer, were in large part motivated bytheir common formal characteristics (epic style, hexameter). Onceagain, they may not have been aware of their development and theirvaried functions or their temporal frameworks.

Another aspect of the dynamic character of genre—and one thathas rightly captured scholarly attention lately—is the issue of re-performance. The same composition can be performed in differentcontexts, and it can even build into itself the possibilities of suchvarying venues as well as evolving performance modes.13 Songscomposed for public festivals could be—and were—sung at intimatesymposia. Finally, freestanding genres might be swallowed up intolarger genres, while subordinate genres can cut loose and becomeindependent. We need only think of the monumental Homeric epicsthat manage to incorporate catalogues, gooi, and folk-tales amongothers.14 With these generalities in mind and acknowledging thatmuch here is inevitably speculative, let us try to use some of thevaried, even contradictory, evidence, internal and external, concern-ing the Hymns to define their generic integrity as well as make anattempt to explain the various stages of their evolution.

In addition to metre and the use of the Epic Kunstsprache, theopening and closing formulae are the most characteristic features ofthe Hymns.15 Many open with the expression ‘I begin to sing’. Thepresence of the speaking—or rather the singing—‘I’ at the beginningand the farewell to the divinity in the second person which brings thecomposition to a close differentiate the Hymns from epic recitationwhere the Muse is asked to sing and the speaker appears to submergeor meld his own voice with hers.16 What sets the Hymns apart fromprayers and most cultic hymns is the absence of an opening address to

13 Cf. Nagy (1994). For Pindar, see, among others, Morrison (2007) and Morgan(1993).

14 For instance, the Catalogue of Heroines in the Nekyia incorporates the Ehoiatradition in the Hesiodic Catalogue. For gooi in the Iliad, see Tsagalis (2004); for folk-tale motifs, especially in the Odyssey, see Page (1973) and Hölscher (1988).

15 In fact, the shortest of these compositions, the three-line Hy. 13 to Demeter,consists of nothing else. See Calame, this volume (pp. 334–6), and (2005), 22–4, wholabels these opening formulae evocationes: ‘Such an evocation tends to establishbetween the divinity and the speaker . . . a mediated relationship. This relationship isput in place through the act of singing itself ’ (p. 24).

16 While the Muse or Muses may indeed be invoked (eight times) in the epicmanner, the subject of their song is always a divinity. On three occasions, the god to be


the divinity in the second person. Formally like an epic invocation,the god who is to be the subject of the song is named in the thirdperson. The presence of the hymnist’s ‘I’ thus indicates a certaindistance between the speaker and his subject matter in the thirdperson, which via a relative pronoun17 introduces a brief descriptionor a narrative vignette in the short and mid-length Hymns, or a fullfledged extended narrative in the long ones. By the end of the Hymn,however, the distance has been bridged, and the god has become fullypresent; the bard, whose account has somehow precipitated an epi-phany of the god, can now bid the god farewell with the formulaicåÆEæ� and perhaps also offer a prayer or promise future praise. Whatseems to distinguish the Hymns from prayers and cult songs is thismovement from the distant third-person announcement of the open-ing to the proximity of the final greeting.18 The vehicle of thisdynamic progression is their most distinctive feature, the interveningnarrative in the epic style. The Hymns’ efficacy is thus demonstratedby first drawing attention to the gulf between the mortal singerand his divine subject and finally revealing the Hymns’ successfulbridging of the distance that normally obtains between god and man.If epic makes the heroic past present, the Hymns make the divinepresent.19

The closing exhortation (åÆEæ�), which simultaneously bids the godfarewell and asks that he take pleasure in the song just completed, is atypical feature of Greek hymns, and by no means limited to theHomeric Hymns; in our Hymns, a brief prayer for favour, generalprosperity, or pleasing song often follows. In one case (Hy. 6) a Hymnends with a request for victory ‘in this contest’ (K� IªH�Ø �fiH��,19–20);20 another presents a request ‘to return here joyfully nextyear and for many years to come’ (�e �’ �A åÆ�æ��Æ K uæÆ Æs�Ø ƒŒ�ŁÆØ, j KŒ �’ ÆsŁ’ ‰æ�ø� �N ��f ��ºº�f K�ØÆı�� , Hy. 26.12–13), which suggests an annual festival; both appear to allude to thesetting of the Hymns’ performance. On three occasions, a promiseconcludes: ‘having begun from you, I shall pass on to another hymnos’

celebrated is invoked directly. Aphr. is an exception since the subject is announced notas the goddess herself, but her works (�æªÆ).

17 On the ‘hymnic relative’, see the classic study of Norden (1913), 168–76.18 Calame (2005), 19–35.19 Cf. Bakker (2002), García (2002), Depew (2000).20 West (1992), 19 n. 25, argues that the occasion of Hy. 6 ‘was one of the Cyprian

festivals of Aphrodite, such as the panegyris in Old Paphos’.

236 Jenny Clay

(��Æ��ÆØ ¼ºº�� K o��); in the two late Hymns to the Sun (31)and Moon (32), the following song is explicitly described as heroicepic. These closing formulae point to performance in agonistic con-texts, possibly in conjunction with recurring festivals, and as part of asequence of songs—either more hymns or heroic epics. But it bearsemphasizing that such statements occur either in short compositionsor those recognized to be late; moreover, a majority of the Hymnsmake no mention of a song to follow.21

Occurring twelve times, the most common closing formula—‘I willremember you and another song’ (ÆP�aæ Kªg ŒÆd ��E� ŒÆd ¼ººÅ

��’ I�Ø�B )—is usually interpreted as equivalent to the��Æ��ÆØ formula, that is, as a transition to another song thatfollows immediately. But some scholars interpret ¼ºº�� . . . o�� and¼ººÅ . . . I�Ø�B as ‘the rest of my song’, in other words, that theinvocation to the god is the first part or a prelude to a composition,which is identified with the kitharodic nomos, and not an independ-ent genre at all.22 But the phrase can just as plausibly be understood asa promise of future hymnic praise to the same divinity; the presenthymn may be completed, but the task of praising the god is never-ending and may be continued either immediately or on some futureoccasion.23

These closing formulae constitute the basis for the proem theory,which goes back toWolf (1795) and is still widely accepted, that holdsthat the Homeric Hymns were performed as preludes to epic recita-tions. The scale of the longer Hymns, viewed as rhapsodic expansionsof the shorter ones, it is argued, does not preclude their serving aspreludes to epics of the magnitude of our Iliad and Odyssey at major

21 Cf. Paz de Hos (1998), 63 who argues that the short Hymns could be used on allkinds of occasions and do not fundamentally differ from ‘lyric’ hymns in theircontents.

22 See Koller (1956), followed by Nagy (1990a), 353–60. But consider Certamen97–8 ()�F�’ ¼ª� �Ø �� ’ K��Æ �� ’ K��Æ �æ� �’ K��Æ j �H� b� Å�b� ¼�Ø��, �f �’¼ººÅ �B�ÆØ I�Ø�B , ‘Muse, come now, the things that are and those that will be andwere—of these sing nothing; but you bring to mind another song’), where theexpression must mean ‘another song’. In the longer Hymns Koller believes thatthe formula becomes a kind of vestige of an earlier practice. Böhme (1937) tracesthe origins of the longer Hymns to the kitharodic nomos itself.

23 In Od. 8. 492–3, after complimenting Demodocus for his accuracy in singing themisfortunes of the Achaeans, Odysseus asksDemodocus: ‘come now, pass on and singof the wooden horse’ Iºº’ ¼ª� �c ��ÅŁØ ŒÆd ¥��ı Œ�� ¼�Ø�� j ��ıæÆ��ı. Whatis striking is that over a hundred lines have intervened since the bard sang and over400 since he sang about the Trojan War.


festivals like the Panathenaea.24 But Iwill try to make the case that thesheer length of the long Hymns is by no means their only commonfeature. A possible further reference to the practice of beginning anepic recitation with a prelude to a god has been detected at Odyssey 8.499 where Demodocus is said to ‘begin from the god’ or to ‘be rousedby a god’ (› �’ ›æÅŁ�d Ł��F Xæå��, çÆE�� ’ I�Ø��) before helaunches into his account of the wooden horse. But the Greek isambiguous: even if he did ‘begin from the god’, did he begin with ahymnic praise of the god (which one?) or did he, like the poet of theIliad, begin his epic narrative with a divinity who instigates the action(�� . . . Ł�H�)? Moreover, the use of the genitive of the god to becelebrated with ¼æå�ÆØ does not seem hymnic.25 The testimonia ofPindar and Thucydides have also been invoked to support the proemtheory, but they too are not unproblematic. Pindar opens the secondNemean with a reference to the Homeridae (‘Sons of Homer’, pre-sumably a guild of rhapsodes): ‘Even as the Sons of Homer, singers ofsewn songs, begin for the most part from a prooimion to Zeus, so toohas this man received a down payment of victory in the sacred gamesfor the first time in the much-hymned grove of Nemean Zeus’ (N. 2.1–5).26 Yet if this was a pervasive custom cultivated by the Homer-idae, it may refer only to a brief nod to Zeus; at any rate, it is notreflected in our collection of the Homeric Hymns, where Zeus receivesonly a brief four-line composition (Hy. 23).27 Thucydides’ references(3. 104. 4–5) also raise questions; as evidence for both athletic andmusical competitions on Delos, he cites lines 146–50 from the Hymnto Apollo (with some variants), as ‘from the prooimion of Apollo’;immediately afterwards he goes on to quote Apoll. 165–72 ‘from the

24 A proem of only one line to the Iliad, similar to our hymnic openings, is attested,but its authenticity is suspect: ‘I sing the Muses and Apollo of the famous bow’()��Æ I��ø ŒÆd ��ººø�Æ Œºı��). Cf. Kirk (1985), 52 and more recentlyWest (2001b), 73. On the proem theory, cf. Introduction (pp. 17–19), this volume.

25 The exception, Hy. 25 to the Muses and Apollo, is obviously an abbreviationdrawn from the opening of the Theogony. For the meaning of ¼æåøÆØ there, see Clay(2003), 50–3.

26 �OŁ�� æ ŒÆd % ˇÅæ��ÆØ j ÞÆ��H� K�ø� �a ��ºº’ I�Ø�� j ¼æå��ÆØ, ˜Øe KŒ�æ��Ø��ı, ŒÆd ‹�’ I��æ j ŒÆ�Æ��ºa� ƒ�æH� Iª��ø� �ØŒÆç�æ�Æ ��Œ�ÆØ �æH�� j˝��Æ��ı K� ��ºıß��øfi ˜Øe ¼º��Ø.

27 Pindar’s analogy between ‘down payment’ and a prooimion would militateagainst Koller’s (1956) interpretation that the term refers to the first part of acomposition, since future payments need not follow immediately.

238 Jenny Clay

same prooimion’.28 If we understand prooimion as a prelude toanother song, did an epic recitation follow? Possibly, but Delos wasfamous throughout the Greek world, as the Hymn itself acknow-ledges, for its choral competitions. If the Delian section was originallya stand-alone composition, then Thucydides’ use of the term prooi-mion would be peculiar since the Delian Hymn would not be aprelude to anything.29 Or did it continue with the Pythian narrative,as in our texts? If it did, Thucydides might have thought the Delianhymn to be a prelude to the longer piece describing the god’s found-ing of his Delphic sanctuary.30 Indeed, other evidence also suggestsprooimion may refer to the first of a series of songs.31 Paradoxically,Stesichorus is quoted to have said: ‘I will pass on to another prooi-mion’ (��ØØ �b K�d ��æ�� æ��Ø��, fr. 64 PMG). In some manu-scripts, the long Hymn to Aphrodite is linked to the following shortHy. 6; and the beginning of Hesiod’s Theogony has been interpretedas a sequence of two hymns to the Muses (1–34, Heliconian, 35–104,Olympian) followed by the exordium proper (105–15).32 In Apollo,the Muses are said to sing ‘the deathless gifts of the gods and thesufferings of human beings’ (��ı�� ÞÆ Ł�H� �Hæ’ ¼�æ��Æ M�’I�Łæ��ø� j �ºÅ��Æ , 190–1); likewise in the Theogony the Musesentertain Zeus by singing about the race of the gods followed by therace of human beings and Giants (44–50). Such a sequencing of songswith compositions about gods preceding narratives about heroeswould thus reflect Greek hierarchical notions embracing gods—

28 KŒ �æ��Ø��ı ��ººø�� (3. 104. 4. 3); KŒ ��F ÆP��F �æ��Ø��ı (3. 104. 5. 2). For asurvey of the use of the term, see Constantini and Lallot (1987).

29 Paz de Hos (1998), 62, notes that ‘in the unique reference to an extant hymn inwhich the term prooimion appears (i.e. in Thucydides), verses are quoted from a longhymn, and it is precisely in those that the function as proem is most disputable’(translation mine): ‘en la única referencia a un himno concreto en que aparece eltérmino “proemio” (la de Tucidedes), se citan versos de un himno largo, y esprecisamente en éstos donde la función de proemio parece más discutibile’). For theview that Apoll. did lead into a longer epic recitation, see Nagy, this volume (Ch. 13).

30 For separatist arguments, see Chappell, this volume (Ch. 4).31 Cf. Càssola xix: ‘The proem is not an exordium of a rhapsodic song, but rather

the first song in a series, and as such, it must have its own exordium’ (translationmine): ‘Il proemio non è l’esordio di un rapsodia, bensì la prima rapsodia di una seria,e come tale deve contenere un proprio esordio.’ Note Pindar’s expression �æ��Ø�ø�I��º� (P. 1. 4).

32 However, Hesiod’s proem is integral to the Theogony that follows rather than afree-floating prelude that could be attached to any poem; cf. Clay (2003), 53–72. Onthe link between prooimion and oimê, see Aloni (1980).


men—beasts. But the crucial point remains that the term prooimion isa functional term: it means the first in a sequence of songs; it does notintrinsically indicate a mode of performance, whether choral ormonodic, melic or rhapsodic.33 Perhaps over time and through cus-tom, its content became restricted to songs about the gods; perhapstoo the term could become further narrowed to the hexameter com-positions that make up our Hymns and thus became a generic termfor our Hymns.34 We will return to these issues when we attempt toreconstruct the evolution of the genre and review additional externalevidence for performance contexts. The history and pre-history of thecompositions that have come down to us as the Homeric Hymns may,then, not be monolithic, but complex and multiple with varied func-tions at different periods, adapted to changing situations and occa-sions of performance.

While acknowledging that different premises may lead to otherconclusions, I start from certain basic assumptions that inevitablyshape the argument that follows. If the shorter Hymns—and I includethe mid-length Hymns—were indeed exordia to other songs, theymay never have formed a completely independent genre. Yet manydiscussions of the Hymns begin by positing the shorter or middle-length compositions as the norm with the longer Hymns a subsequentdevelopment.35 If, however, the long narrative Hymns do not consti-tute rhapsodic expansions of the short ones, they may represent anautonomous type of hexameter epos, developing alongside, and com-plementary to, heroic epic. It must, however, be emphasized that it isnot merely the length of these poems that distinguishes them fromtheir shorter counterparts. Rather, they share certain features in theirnarrative content and its elaboration that should likewise be consid-ered generic characteristics. Thus I posit that their style, content, andnarrative purpose offer the generic coordinates for this kind of com-position and should provide the starting point for a discussion oftheir genre. I will further argue that within hexameter epos thelong narrative Hymns form a separate genre, a genre with its own

33 Cf. Aloni (1998), 117–38, who, however, believes that prooimia were tied tospecific occasions rather than compositions that could be re-performed. This wouldnot seem to be the case with our Hymns. See below on Panhellenism.

34 For the shift from functional criteria for genres in the archaic period to moreformalist characteristics in the Classical and Hellenistic periods, see Ford (2002),11–12.

35 E.g. Fröhder (1994), 15. See also Clay (1997), 495–97, and cf. above n. 10.

240 Jenny Clay

pre-history and evolution, utilizing traditional diction and techniqueswhich may parallel and sometimes overlap with, but also divergefrom, those of heroic epic. If the tales of Troy were already traditionalfor Homer and his audience, they must have been equally familiarwith the story of Persephone’s abduction or Apollo’s combat with thePythian dragon. That is, that by discovering the commonalities in thelonger Hymns, we can begin to define the Homeric Hymns as a genre.

The purely formal similarities (metre, diction, opening and closingformulae, etc.) that are shared by the short and the long Hymns aredwarfed by the main feature that defines and distinguishes the latter:the highly developed central mythological narrative. Moreover, thesecompositions reveal a shared conceptual framework that is notmerely formal, but thematic.36 Each presents a fully developed nar-rative whose main actors are divinities, gods of the Olympianpantheon. The story line serves to define and present dramaticallythe character and prerogatives of the divinity to be praised, both inrelation to other gods and to human beings, that may involve thefounding of cult or some other innovation or invention. Birth narra-tives provide ideal scripts as they portray the newborn gods as theyacquire their individual timai and spheres of influence, often over-coming various kinds of opposition, to become fully integratedmembers of the Olympian pantheon. Thus Apollo, Hermes, andpresumably the fragmentary Dionysus, which stood first in the collec-tion.37 But there are also other episodes in the lives of the gods thatmay offer suitable hymnic material: incidents that are, in one way oranother, defining moments in the history of the gods and theirrelations with mankind. While discrete episodes, they seem to con-tribute to a larger whole that recounts the evolution of the Olympianorder and thus fills the gap between its pre-history that culminates inthe triumph of Zeus, which forms the subject-matter of theogonic

36 Sowa (1984) traces some thematic similarities such as type scenes found in boththe Hymns and Homeric epic. García (2002) finds the theme of recognition central tothe major Hymns.

37 See West (2001a) for an attempted reconstruction and his comment (p. 2): ‘Atthe start of the hymn, Dionysos is born in a remote region . . .By the end, he isestablished as a god, presumably on Olympus, destined to enjoy a thriving cult inGreek lands. The main part of the hymn must have contained a narrative thatexplained how he progressed from the initial to the final situation’. Also, his chapter(2) in this volume. But see Faulkner (2010a).


epos, and heroic epic which presupposes the stability of Zeus’ regimeand the final configuration of the pantheon under his tutelage.

This temporal mythological framework explains the importance ofZeus in these narratives. Although sometimes discreetly positioned inthe background of the action and rarely intervening directly in thenarrative, the supreme god is often the crucial wellspring of the actionor the one who manipulates its course or is required for its denoue-ment.38 It is also his interventions that guarantee the innovations oradjustments in the relations of power between the gods and sanctiontheir spheres of influence. In Apollo, immediately after his birth,Apollo declares his allegiance to Zeus (132), thus revealing himselfnot as a threat to his father, but a loyal upholder of the Olympianorder. In Hermes, Zeus’ recognition of his illegitimate son (389–95)leads to the reconciliation of the wily trickster god with his moreaugust and powerful, but nevertheless easily outwitted, older brotherApollo. The integration of Hermes into the pantheon allows formediation and exchange, the characteristic prerogatives of the newgod, between divine spheres of influence and established prerogatives.In Demeter, Zeus’ acquiescence to his daughter’s rape by Hades (3)sets the plot in motion, one that results in Zeus’ incorporation of theunderworld within his domain;39 moreover, his assent to Perseph-one’s alternating sojourns below and above the earth (460–9) pro-vides the compromise that reintegrates Demeter into the Olympianorder.

If Zeus is in some sense the architect of the narratives of theHymns, and their outcomes rearrange and adjust the powers andprerogatives of the gods under his regime, their plots also inauguratefundamental changes in the relations between gods and mortals.Thus, through the founding of his oracle, Apollo becomes themeans of curing human ignorance of the divine by making Zeus’will known to mankind (Apoll. 132). Hermes’ arrival on the cosmicstage also signals a change in human existence; the god’s manyinventions (e.g. lyre, Herm. 41–51; fire sticks, Herm. 111; exchange,Herm. 496–9) raise mankind from its brutishness and ease the humanlot through, among other things, technology and commerce. Evenwhile narrating a sexual union between the goddess and a mortal,Aphrodite at the same time insists on the unbridgeable gulf between

38 See Faulkner (forthcoming c).39 Rudhardt (1994); Clay (1989), 211–14.

242 Jenny Clay

human beings and the immortal gods (Aphr. 239–46). Although alsoemphasizing this fundamental divide, Demeter nevertheless mitigatesthe absolutes of human mortality throughDemeter’s establishment ofher Eleusinian cult that offers to her initiates some sort of hope for abetter life after death (Dem. 480–2).

If, as I have suggested, each Hymn can be considered a discreteepisode within the larger story of the evolution and stabilization ofthe Olympian order, these compositions can even be ordered chron-ologically as distinct phases within that development. Thus Apollobelongs to an early phase of Zeus’ ascension to supreme power; itplays out within the shadow of the Succession Myth as the new god isrumoured to become a violent ruler over gods and men, and Hera’sopposition is reminiscent of earlier revolutions in heaven.40 But in thecourse of the Hymn, Apollo not only defeats the enemies of Zeus’order, but also becomes, through the founding of the Delphic oracle,not his father’s adversary, but his prophet. As disseminator of hisfather’s will, he extends Zeus’ dominion over mankind and opensa new venue for communication between gods and men. Hermesrecounts the birth of the last-born of the Olympians, whose divinestatus first remains in doubt, and who must wrest his privileges andestablish his timai through his hallmark modus operandi, trickery andtrade. Doubtless, the fragmentary Dionysus that stands first in ourcollection also fitted into such a chronological framework. Of our twofragments traditionally ascribed to the poem, one (A) begins with thegod’s birth and, if the second (D) comes from the Hymn’s end, thepoem closes with Zeus’ confirmatory nod admitting the god and hismother into Olympus. But an additional fragment (B) suggests thatthe main narrative may have involved the binding of Hera and thereturn of a drunken Hephaestus to Olympus in the entourage ofDionysus, a favourite theme of early Greek vase painting.41 Heretoo one can perhaps reconstruct a larger context of cosmogonicsignificance, involving not only some of the final workings-out ofthe Succession Myth in the reconciliation of Zeus and Hera via theirchildren,42 but also, on a more abstract level, the necessary combin-ation of craft and inspiration, art and desire.43 Aphrodite clearly is set

40 On Apoll.’s relationship to Hesiod, see also in this volume Felson (pp. 271–9).41 See West (2001a), and this volume (Ch. 2), Merkelbach (1973), Wilamowitz-

Moellendorff (1895¼1937). Cf. above n. 37.42 See Bonnafé (1985), 87–101, and Clay (2003), 28–30.43 I am indebted to my student Sarah Herbert for this insight.


in the era just before the Trojan War in which Aeneas, the son thatthe goddess bears to Anchises, will become a leading figure. With theTrojan War traditionally identified as the conflict that brings theheroic age to an end,44 Aeneas’ birth coincides with the end of divineand human intimacy. Finally, Demeter depicts a world very muchlike our own in which initiation into the goddess’s Mysteries holdsout the possibility of some happier state for mankind in the afterlife.

The chapters in the history of the Olympian pantheon representedin the Hymns have clear links to the end of Hesiod’s Theogony whichlikewise sketches an outline of the aftermath of Zeus’ accession topower.45 It is, I suggest, no accident that the plot of Demeter can beviewed as an expansion of the three lines Hesiod consecrates to thegoddess:

ÆP�aæ › ˜�Å�æ� ��ºıç�æ�Å K ºå� qºŁ��,m �Œ� —�æ��ç��Å� º�ıŒ�º��, m� �Ø�ø��f læ�Æ�� w �Ææa Å�æ� , ��øŒ� �b Å��Æ Z�� . (Th. 912–14).

Then he [Zeus] entered the bed of much-nurturing Demeter,who gave birth to Persephone, whom the Lord of the Deadsnatched from her mother, but cunning Zeus had given her.

In fact, I have argued that the Hymns elaborate on the lines that formthe climax of the Theogony: ‘The blessed gods . . . then urged Olym-pian wide-seeing Zeus . . . to be king and lord of the immortals; and hedivided up their timai well’ (881–5). Hermes provides an exemplar ofthe temporal continuity and generic complementarity between hym-nic and theogonic poetry—by which I mean the pre-history of theOlympian order—in the young god’s two performances on his newlyinvented lyre: the first song he improvises is a ‘Hymn to Hermes’, amirror image of the composition we are in fact hearing, that beginsfrom his birth (57–62).46 His second performance is a theogony thatdescribes how ‘the immortal gods and the dark earth first came into

44 Cf. Il. 12. 13–33, Hes. Op. 158–66, fr. 204 (M–W).45 See Furley, this volume (Ch. 10), on the relationship between the Hymns and

theogonic poetry. Iwould, however, emphasize that it is precisely when it comes to thegenealogies of the Olympians that Hesiod’s Theogony becomes terse. See Clay (1989),15, where I characterize the Homeric Hymns as filling the gap between theogonicpoetry and heroic epic.

46 Clay (1989), 108–10. See Calame, this volume (Ch. 14), on musical performancewithin the Hymns; onHermes’ performance with his newly invented lyre, cf. Vergados(pp. 101–2).

244 Jenny Clay

being, and each received his share . . . according to their seniority,recounting all in due order’ (427–8, 431–3). Hermes’ theogony willnaturally end with his own birth, a hymn to Hermes, this one perhapsmore complete than the first, as he is now well on his way to acquiringthe timai that will become his share. We should also note that thegod’s performance begins with a proem dedicated to Mnemosyne,‘the mother of the Muses, whom he celebrated first of the gods withhis song’ (429–30).47 Thus the episodic narrative character of theHymns functions within a larger mythico-temporal continuum thatspans the history of the cosmos from its first beginnings through thetriumph of the Olympians and the ordering of the pantheon, includ-ing the heroic age and even beyond to our own diminished times.48

While duly recognizing the cosmic dimension of each of the majorHymns, we must not overlook another feature: their entertainingcharacter, most prominent, to be sure, in Hermes, the wily kid whooutfoxes and both repels and charms his older and more pompousbrother.49 Aphrodite too exploits the comic potential of the goddess’sseduction of the unwitting but lusty Anchises, which culminatesin his terrified recognition of her identity and his cowering underthe blankets while clutching the bedclothes (Aphr. 181–3). Apolloincludes a charming talking island, Delos, whose poverty obligesher to look after her self interests (Apoll. 63–73), and the crafty springTelphousa, who gets her comeuppance for attempting to deceive thegod (Apoll. 256–74, 375–87); Apollo himself shows up as a monstrousdolphin heaving himself on board ship and terrifying the ignorantsailors who worry more about their next meal than the grandiose fatethat awaits them (Apoll. 399–404, 528–30). Even the august Dem-eter’s world-destroying grief and anger is relieved by laughter,50 andher sublime attempts to immortalize a human child are punctured byan anxious mother’s fears (Dem. 198–204, 281–91). These lightermoments do not, however, undermine the cosmic import andgrandeur of these stories, and they are of course not confined to

47 Clay (1989), 138–40.48 I intend to outline this mythico-historical continuum in my forthcoming His-

tory of the World, Part One.49 On humour in Herm., see in this volume Vergados (Ch. 5).50 This is of course not the overtly sexual story of Baubo preserved in some

‘Orphic’ versions. Cf. Richardson (1974), 213–17, on Iambe and jesting in Demeter,with discussion also of Baubo and the later Orphic version with reference to laughterand its significance in Eleusinian cult.


the hymnic narratives, but are to a certain extent inherent in theanthropomorphism of the Homeric gods.51 Yet in the case of ourHymns, their disconnection from an immediate cultic context allowssimultaneously for moments of light-hearted entertainment as well asfor serious theological speculation. While for us the comic elementsdetract from the reverence owed the gods, the Greeks apparentlycould combine the two without discomfort.

Finally, the Hymns are characterized by a distinct Panhellenictendency, that is, a strategy of avoiding versions or associations thattie them down to specific local traditions, which, to be sure, may notpreclude indications of their origins. Such avoidance is most clearlyexemplified in the fragmentary first Hymn to Dionysus, which rejectsa number of local claims to be the god’s birthplace.52 Even Apollo andDemeter, which focus on the founding of cults, do not depict sites orrituals of merely local interest; the accent throughout is on thePanhellenic dimension of these sanctuaries. Moreover, in unitinghis celebration of both Delian and Pythian Apollo and includinggeographical catalogues that encompass the wider Greek world, thepoet of Apollo insists on the Panhellenic status of the god. Demeterlikewise seems to diverge in important ways from both local Eleusin-ian and other traditional versions and instead offers a unique syn-thesis designed to reinforce its Panhellenic message.53 In the case ofAphrodite the delicious narrative of the goddess’s seduction of theTrojan prince Anchises and the subsequent birth of the hero Aeneas,which some scholars view as a composition first intended to enhancethe ancestral claims of a local Trojan aristocratic family, is trans-formed into a universal story by the emphasis accorded the actions ofZeus. He plays no role in other versions,54 but his plan frames theHymn within a world-historical context that signals the end of an erawhen gods and mortals united to produce the demi-gods calledheroes and therefore precipitates the conclusion of the heroic age.55

51 One thinks of the laughter provoked by Hephaestus in Il. 1. 599. Cf. Griffin(1980), 183–5.

52 But see West, this volume (p. 33).53 On the relationship of Dem. to Eleusinian cult, see Richardson in this volume

(pp. 50–3).54 Cf. Faulkner (2008), 138.55 For the view presented here, see Clay (1989), 165–201; Van der Ben (1981),

69–107; and Brillet-Dubois, this volume (pp. 123–6). Contra, see most recentlyFaulkner (2008), 3–18, with earlier literature.

246 Jenny Clay

In a different manner, Hermes, while incorporating various locationsand traditions associated with the god, shuns any easy identificationof provenance in order to insist on the young god’s Panhellenicnature and his unique contributions to the Olympian pantheon.The pervasively Panhellenic colouring of the Hymns points to theirdiffusion throughout all parts of the Greek world; no matter whereeach composition arose and under whatever local conditions orimpulses, like Homeric epic, it could be performed in Sparta as wellas Athens, in Metapontum as well as Lesbos. Panhellenism, in turn,suggests a certain degree of mobility for the Hymn poets and theircompositions; whatever the locus of ‘original’ performance and audi-ence, re-performance in different venues continued to shape theHymns and reinforce their Panhellenic character, creating a kind oftheological lingua franca.56

Having now outlined a descriptive definition of the major HomericHymns as a genre, I propose to sketch out a possible scenario,inevitably speculative, for their evolution. I begin from the Homericpoems, which as I have said swallowed up and incorporated varioussubgenres. As such, they may give us intriguing glimpses into the pre-history of the genre while attesting to our earliest examples of rhap-sodic hymns, modified, to be sure, to fit their epic contexts. Severalscenes in the Iliad would seem to fulfil the criteria we have laid outand offer possible candidates for such generic incorporation: I speakfirst of the conflict between Zeus and Poseidon in Iliad 15 (158–217),which may replay in a lighter key a far more serious earlier competi-tion between the two gods for divine supremacy. There Poseidon’ssupport for the Greeks conflicts with Zeus’ promise to Thetis to givemomentary victory to the Trojans. When Zeus threatens him bypulling rank, Poseidon appeals to the ancient division of honoursamong the three sons of Kronos (15. 187–93). Before backing down,he invokes the possibility of a widespread rebellion among the gods.Conflict between the Olympians in tandem with the distribution andadjustments of their prerogatives under the supreme hegemony ofZeus are, as we have seen, dominant themes in the Hymns. What inthe Iliad becomes a quarrel quickly resolved could have originated ina cosmic struggle for control of the Olympian pantheon. Such cosmicthemes form the subtext ofDemeterwhere the plot springs from Zeus’

56 On the Panhellenic nature of Hy. 7, see in this volume Jaillard (Ch. 7);cf. generally on the issue of Panhellenism, Introduction (pp. 19–22).


design to annex the nether regions to his domain through the mar-riage of his daughter Persephone. The oddly interrupted theomachyin Iliad 20 (4–75) and 21 (385–514), which frames Achilles’ murder-ous rampage and serves to amuse Zeus, offers similar material,perhaps from a similar source.57

Even more germane to the present discussion is the seduction ofZeus by Hera in Iliad Book 14 that also suggests an instance of theepic’s appropriation of hymnic material. Like the Homeric Hymns instyle and diction, it is also full of intimations of cosmogony. Scholarshave long pointed out its parallels to the Hymn to Aphrodite as bothgoddesses arm themselves for seduction by donning their finery.58

Like Aphrodite, the Iliadic episode incorporates the high comedy offeminine wiles overcoming witless males in the throes of desire withmore serious hints of the consequences of attacking the patriarchalsystem and Zeus’ supremacy. Moreover, it manages to unite divinehorseplay with the sublime image of the hieros gamos that guaranteesuniversal fecundity (Il. 14. 346–51). The seduction episode thus nicelyfits the contextual criteria for the Homeric Hymns. In its presentposition within the Iliad, it mainly serves to retard the inevitableprogress of the plot by diverting Zeus from his plan to grant theTrojans temporary superiority—and to give the audience a breakfrom the battlefield. It could, however, be appropriate to multiplemythological frameworks where, for her own nefarious purposes,Hera manages to distract Zeus—even though only for a while—toundermine his intentions. If one were to speculate, the story might fitnicely into a Hymn to Heracles orDionysus,59 or even some other taleof Zeus’ amorous adventures that have wide-ranging consequencesand incite Hera’s opposition; Apollo surely provides a far morethreatening example of Hera’s power. In any case, both the ribaldcharacter of the Iliad ’s divine seduction as well as its cosmogonicovertones find parallels in narrative elements of our Hymns. In factWest, in this volume (Ch. 2), argues for the adaptation and incor-poration of motifs from the fragmentary first Hymn to Dionysus intoboth the Iliad and the Odyssey in precisely the episodes we have been

57 See Il. 1. 396–406 and Slatkin (1991).58 For the suggestion that Hera’s preparations are modelled on Aphrodite in the

Cypria, see Janko (1994), 185. Cf. Faulkner (2008), 32–3, 141–8, and Brillet-Dubois,this volume (pp. 109–12), for further parallels.

59 Cf. West, this volume (pp. 35–7).

248 Jenny Clay

discussing: the Dios apatê and the Lay of Ares and Aphrodite. Hym-nic motifs may also lie behind Hephaestus’ attempt to effect a recon-ciliation between Hera and Zeus that brings the first book of the Iliad(571–600) to a close. Other passing references in the Iliad to Thetis’rescue of Zeus (1. 396), Hera’s anvils (15. 18–22), and Zeus’ threats ofa tug of war with an insubordinate pantheon (8. 18–27) may wellincorporate hymnic material. Moreover, the description of an angryHera descending to Tartarus and the seat of the Titans (8. 477–83)has obvious parallels with her actions in the Hymn to Apollo(332–39), where she enlists the aid of the Titans to produceTyphoeus.

Perhaps even more relevant to a reconstruction of the pre-historyof the Homeric Hymns and helpful for the understanding of theirperformance context is the Lay of Ares and Aphrodite, the second ofthe songs performed by Demodocus, the resident bard in Phaeacia(8. 266–366), which has often been compared to the HomericHymns.60 Wilamowitz thought it a model for our Hymn to Aphro-dite, but also a decadent and interpolated sequel to a no longerextant Homeric Hymn to Hephaestus, although its origins and pointmay lie elsewhere.61 To be sure, the singing or chanting of thePhaeacian bard to the lyre differs from the unaccompanied recita-tion of the rhapsodes (Pindar’s ‘singers of stitched songs’). WhileHomer’s depiction may well be archaizing, there is nevertheless noreason to doubt that it may depict an earlier stage of performance,well suited to the presentation of compositions of limited length inwhich the singer chanted to the accompaniment of the lyre.62 In thisnotorious episode, Demodocus recounts how the beauteous Aphro-dite and her equally handsome lover Ares were caught in flagranteby her ugly and lame husband Hephaestus who draped their bedwith chains. The cuckolded husband invites the gods to witness

60 Cf. AHS lxxxvii–lxxxviii who remark that only the invocation, i.e. proem, andthe farewell are missing.

61 Wilamowitz-Moellendorff (1895), 12–14; Merkelbach (1973), following Snell(1966), suggested that the story of Dionysus’ restoring Hephaestus to Olympusbelongs more properly to a hymn to Dionysus, and West (2001a) identifiedits beginning and end with our first Hymn to Dionysus. See above n. 37.

62 Nagy (1990a), 21, speaks of ‘diachronic skewing’ in the depiction of the Homericbards. This performance mode of accompanied recitative may be the commonancestor of both epic recitation and the more musical kitharodia. Cf. the perform-ances of Hermes at Herm. 57–62 and 425–33, and see Nobili (forthcoming).


the trapped lovers, a sight that precipitates gales of laughter amongthe (male) divinities. Within its context, Demodocus’ song offers alight-hearted diversion and inversion of heroic epic themes, possiblytruncated and tailored to fit the dominant Odyssean motifs ofinfidelity and its consequences as well as the triumph of metis. Theaffinity of the episode with our hymnic narratives has long beenrecognized.63 Here too one can perhaps reconstruct a larger contextof cosmogonic significance in the conflict between Hephaestus andAres, the two legitimate male offspring of Zeus and Hera.64 In theIliad Ares is always problematic, and Hephaestus for all his skills isdefective and always appears under his mother’s thumb; Ares, on theother hand, is called the �åŁØ�� Ł�H� (‘most hated of the gods’, 5.890) by his father and treated with similar contempt by Athena.Underneath the lascivious Odyssean tale, which has some of the funof Aphrodite, lies a nuanced version of the final working out of theSuccession myth: the competition between Ares and Hephaestusembodies the tension between beauty and brawn (bie) and craftand guile (metis), where metis triumphs over bie; yet neither ofthese divine sons can rival their father Zeus who combines both.65

If, as I have suggested, the divine narratives of the long Hymns canbe viewed as parts of a larger account that constitutes a history of theOlympian order, then early Greek epos provides us with a closeanalogy in the traditions of the Epic Cycle where individual episodesfrom the story of the Trojan War, while self-contained, neverthelessform part of a larger narrative tapestry.66 Similarly, while each Hymnis an autonomous composition, its plot simultaneously functions as

63 Note the use of Iç� at Od. 8. 267, which occurs in Hy. 7 to Dionysus, 19 to Pan,22 to Poseidon, and 33 to the Dioscuri, and is identified as a common opening of thecitharodic nomos. But in the Hymns, the name of the god follows in the accusative,while Demodocus sings ‘about the love of Ares and Aphrodite’.

64 In Demodocus’ song (as in the Iliad), Hephaestus is the son of both Hera andZeus, rather than the product of her lame effort to bring forth a son without him.Clearly, the Homeric poems know several traditions involving Hephaestus; in theIliad he is married to Charis.

65 I intend to elaborate further on the incorporation of hymnic material in aseparate study.

66 Cf. Burgess (2004), 3: ‘it is apparent that mythological material was thought tocohere into certain units; a Heroic Age with chronologically arranged groupings ofmythological tales, like the Theban wars and the Trojan war, is assumed in Homericand Hesiodic poetry.’ Burgess (ibid.), 15 cites Apoll. along with Hes. fr. 375 M–W asan example of ‘discontinuous performance’ with more than one singer; but he alsosimilarly characterizes Demodocus’ two episodes from the Trojan War (p. 17).

250 Jenny Clay

part of a larger narrative that defines the Olympian pantheon. And Ihave argued that the Hymns form part of an even more comprehen-sive master narrative that embraces three strands of epos: theogonic,Olympian, and heroic.67 The three songs of Demodocus in the Odys-sey with the two episodes from the Trojan War flanking the hymn-like narrative of Ares and Aphrodite are suggestive and may help todefine performance occasions for both kinds of compositions. Firstand last, ‘The Quarrel between Achilles and Odysseus’ (8. 73–82) andthe ‘Trojan Horse and the Fall of Troy’ (8. 499–520) are thematicallylinked by the motif of the triumph of Odyssean metis, but they do notform a continuous narrative; but instead they belong to the beginningand the end of the war, thus offering bookends that allude to thewhole narrative. Both are performed after dinner in Alcinous’ court.The hymnic narrative in which Hephaestus traps Ares and Aphroditein flagrante is framed by these two heroic episodes and is performedoutdoors at an ad hoc public (but not religious) festival with thePhaeacians looking on; Demodocus’ song may have been mimed insome way, and it is certainly followed by other kinds of dancing. Butperhaps we should not make too sharp a distinction between publicand private here; if heroic episodes could be presented at festivals,why should the Hymns not be sung after dinner or at symposia?68

The scenario that I am suggesting has a parallel widely accepted byPindarists who recognize that an original choral presentation andoccasion of an ode can be repeatedly re-performed and disseminatedvia a solo performance in a variety of settings.69 Indeed, the Hymn toApollo points to a certain interchangeability in performance modesand occasions: the Delian Maidens offer a choral and perhaps mi-metic performance in the context of a Panhellenic (or Panionian)festival.70 The hymnist who identifies himself as the ‘Blind Man fromChios’ first describes the festivities on Delos and the Maidens’ per-formance and then proposes what seems to be a reciprocal arrange-ment: the girls are to advertise his excellence to all those who frequentthe island; he in turn promises to spread the fame (kleos) of theirperformance throughout the world as he traverses ‘the well-inhabited

67 See Clay (1989), 268–70.68 Cf. Faulkner (forthcoming b).69 For instance, Morgan (1993) and Morrison (2007).70 Cf. Clay (1989), 50–2. A similar interchangeability obtains in the Theogony

where Hesiod purveys a monodic (although not necessarily unaccompanied by thelyre) version of the Muses’ choral song.


cities of men’ (147–76). The text implies that choral and soloperformances are to a certain extent interchangeable. The poet’sversion, whether it is envisaged as rhapsodic or accompanied by theApolline lyre, constitutes a portable and solo rendering equivalent tothe lyric presentation on Delos.

By way of conclusion, let me now offer a possible outline for thedevelopment of the Homeric Hymns that attempts to integrate theoften contradictory internal and external evidence into a diachronicscheme. Originally, the major hexameter hymns were independentcompositions usually not attached to a single location or cult, butconsciously Panhellenic, usually focused on one of the Olympiangods, and which define and exemplify the divinity’s role within thepantheon and in relation to mortals. Their length made them appro-priate both to an evening’s entertainment or to a public festival. Thelatter context may have given rise to competitions, such as thatenvisioned in the story of a duel between Homer and Hesiod inwhich each presented a hymn to Apollo.71 Passages in both Hesiodand the Hymn to Apollo suggest that if several relatively shortpoems—say under 500 lines—were performed, it may have becomecustomary to arrange them in a sequence so that songs in which thegods were the chief actors would precede episodes drawn from theexploits of heroes. Unlike the heroic narratives of similar lengthpresented on similar occasions, the stories of the gods remainedepisodic compositions and were not amalgamated into the cyclicepics that set out to present the Trojan War from beginning toend. The lack of a straightforward narrative continuity between thedivine narratives precluded any simple linkage. But with the rise ofthe monumental Iliad and Odyssey, these originally independentcompositions could become drawn into the ambit of those epicperformances, becoming the first in a series of songs, a pro-oimionproper. Some poets might, however, still continue to offer the trad-itional brief salute to a god that introduced heroic narrative—equiva-lent to our short Hymns—but others might have preferred to beginwith a more extensive narrative where the gods were the protagonists.Both might be called prooimia, but the labelmasks a divergent historyand development. Nevertheless, the longer Hymns maintained theirindependent existence and may have continued to be performed as

71 Cf. in this volume Nagy (pp. 294–304).

252 Jenny Clay

autonomous compositions at symposia and festivals both private andpublic. They also maintained their generic coherence as narrativesembodying theological speculation that complemented the Pan-hellenic epic and theogonic poetry of Homer and Hesiod. To con-clude, the longer Homeric Hymns in both their scale and episodiccharacter find an analogue in the oimai of the Epic Cycle, discreteincidents constituting an evening’s entertainment, but neverthelesspart of a master narrative of the exploits of the heroes. The mon-umentalization ofHomeric epic finally precipitated the disappearanceof these earlier forms, and hymnic material could eventually beswallowed up in these huge productions. At a certain point, theancestors of our Hymns may have acquired a new function, as pre-ludes to heroic epics; but they nevertheless continued to be performedin their previous venues, exercising their earlier functions. An analy-sis of our Hymns and their generic characteristics may thus provide aprecious glimpse of, and an illuminating contribution to, the pre-history of hexameter epos.


12

Children of Zeus in the Homeric Hymns

Generational Succession

Nancy Felson

This chapter explores how Zeus’ displacement of his father Kronos inthe succession myth of Hesiod’s Theogony is reflected in HomericHymns honouring his offspring. It examines how the triangularstructure of the mother and son united against the father, reiteratedin three generations of the Theogony and leading to a ‘revolution’ inthe first two cases—Kronos and Gaia against Ouranos, Zeus, andRhea/Gaia against Kronos—finds expression in Hymns to two ofZeus’ offspring. In principle, unless the father and son reach amutually beneficial accommodation, ‘son’ always entails a suppres-sion of ‘father’.1 Both Athena and Apollo (in Hymn 28 and Hymn 3,and other archaic Greek texts) invert their father’s narrative and endup as his staunch and reliable allies. Yet they both retain traces of an‘as if ’ narrative that reduplicates (at least in part) the life-story of Zeusas told primarily in Hesiod’s Theogony.

At two moments in typical Greek hero tales and in the story ofZeus’ coming to power a son confronts his father and either yields tohim or challenges and often supplants him. One is at birth and theother at the peak of youth, which the Greeks call l�Å (hēbē). TheHomeric formula, ‘when he reached the metron hēbēs’, marks thissecond critical moment, while an oracle (predicting a confrontation,or the father’s defeat at the hands of his son, or hatred of the one

1 Pucci (1992), drawing on Lacan, is particularly insightful on this point.

toward the other, or simply excess energy in either) marks the first.The following schema sets forth the plot sequence typical of tales ofintergenerational strife; it includes variants at discrete moments inthe chain of events:

1. The father sires a series of sons.2. He hates and/or fears some of them

a. because of their excessive manhood, size, or monstrosity(hybridity).

b. because he has learned from an oracle that one of them willdisplace him.

3. He tries to obliterate those dread sonsa. by repressing them.b. by swallowing them as each is born.c. by swallowing the pregnant mother.

4. With this act he dishonours/violates their mother.5. He also dishonours/violates the sons.6. The mother is enraged and in pain.7. She betrays her spouse

a. by enlisting the aid of their youngest son.b. by rescuing the threatened son and instructing him to

retaliate later, at the peak of youth (hēbē).8. The youngest son embraces her plan and punishes (¼ dis-

places) the fathera. by castrating him.b. by defeating him (or his generation) in battle.c. he may receive aid from an ancestress (mother or grand-

mother).

9. The victorious son becomes king, takes a bride, and sires sons.10. The defeated father may curse or threaten his youngest son or

all his sons.11. The victorious son exiles or murders his father (¼ parricide).

Alternatively, the second half of the sequence can develop as follows:

7b. The mother betrays her spouse by saving a threatened son andinstructing him to retaliate later at the peak of youth (hēbē).

8b. The rescued son grows up in hiding, away from his natal home.9b. At hēbē he returns to carry out his vengeance.

10b. He receives aid from an ancestress (mother or grandmother).

Children of Zeus in the Homeric Hymns 255

11b. He defeats his father (and his allies) in a war or other contest.12b. He exiles or murders his father (¼ parricide).13b. The victorious son gains the kingdom and a bride as queen.

Another alternative ending, instead of 10b–13b, is:

9c. The father defeats his son and eliminates him as a threat(through murder or exile).

10c. The father retains his kingdom and his wife as queen.

The plot sequence extracted from the succession myth in Hesiod’stext reveals the two critical moments of tension between father andson: at the child’s birth and at the peak of his youth. Zeus, forexample, outstrips Kronos twice in his rise to power: at birth, when(aided by Gaia and Rhea and the trick of the swaddled stone) heavoids being ingested like his siblings, and at the peak of his youth,when he returns from Crete, wins the intergenerational Titanomachyand banishes the vanquished Titans, including his own father, toTartarus. Once Zeus is victorious, all the blessed gods (the Olympianvictors), through the plans of Gaia, urge him to become king and torule (Th. 881–5; cf. in the proem 112–13 and 72–4).

On his pathway to kingship, and even after his election as king,Zeus faces a series of challengers who would have usurped his king-ship. These include: (1) Prometheus; (2) Typhoeus; and (3) a child ofMetis.2 In the ‘life-history’ of these last two challengers—Typhoeusand the unnamed, unborn son—their birth (or anticipated birth) ismarked as menacing to Zeus: Typhoeus immediately challenges Zeus,while Metis’ son never gets that opportunity (since he is never con-ceived, much less born).

In the Homeric Hymns to Athena (28) and Apollo (3) these twocritical moments of potential conflict are collapsed into one, themoment of ‘arrival’. In both cases, tensions rise and are marked bydivine or cosmic disorder. While Apollo’s parents, Leto and Zeus,diffuse the aggression of their bow-armed son, Athena dispels tensionby disarming herself and thus making her father Zeus rejoice.In Apollo’s case, the aggressive energy or impulse to attack is trans-lated into his struggles against females and monsters, while Athenachannels her energy through heroes she supports and, in Hymn 11,through citizens at war.

2 On the Prometheus and Typhoeus episodes as part of the theogonic tale, seeVergados (2007), 51–4.

256 Nancy Felson

Among Zeus’ divine offspring, Athena and Apollo ‘inherit’ therevolutionary trait from their progenitors. Each one could, in prin-ciple, undermine cosmic order. Yet in different ways, and for differentreasons, each overcomes or redirects any such tendencies, becomingin the end a supporter of the cosmic order over which Zeus willcontinue to preside. In each case, the avoidance or resolution ofconflict allows the fourth-generation offspring to take up his or herplace in the now fully established order of things and to assume anappropriate identity without seriously challenging the position of theone whose signal epithet, ‘father of gods and men’ (�Æ�cæ I��æH� ��Ł�H� ��), denotes his supreme patriarchal authority.

Gaia, as primal mother, has a decisive role in the succession mythof the Theogony even after she sanctions the rule of Zeus, andthis maternal role provides a model for one subset of mothers ofpotentially rebellious sons. Gaia is a bivalent figure, who functions intwo incompatible capacities, as both an enemy and a friend of order.3

She has her own personal goals and plans, according to which she actsas a helper or an obstructer, combining disorderly traits with a needfor security and stability. On the one hand, she has monstrousqualities, like several of her Ancient Near Eastern prototypes.4 Gaiaexpresses her monstrosity through the products of her womb, mostnotably the Cyclopes, the Hundred-Handers, and finally Typhoeus.She is frequently called ‘monstrous Earth’ ( ÆEÆ ��º�æÅ: Th. 159, 173,479, 821, 858; cf. 505, 731, 861) and the implement with which shehas Kronos castrate Ouranos shares that epithet (179: ��º�æØ�� . . .–æ�Å�, ‘monstrous sickle’). At the same time, Gaia’s order-loving sideinforms her actions for most of the poem, until Zeus defeats her lastmonstrous offspring, Typhoeus. For one thing, Gaia knows the fu-ture: she is the source of prophecies, especially those having to dowith future kings. Moreover, right after she comes into being, Gaia isthe stable seat for two constituencies, the celestial (Olympian) and thechthonic gods:

. . . ÆP�aæ ��Ø�ÆˆÆE’ �Pæ��æ�� , ��ø� �� I�çÆºb ÆN�d

3 Mondi (1984), 334, comments that it is ‘useless to try to make synchronic senseout of the shifting allegiances of Gaia in the various episodes of the Theogony’.

4 On Gaia’s relation to her Mesopotamian precursors, see West (1966), 25–30, andPenglase (1994), 103–4 and 189–90.


IŁÆ��ø� �Q �å�ı�Ø Œ�æÅ �Øç�� !Oº��ı,'�æ�Ææ� �’ M�æ��Æ ıåfiH åŁ��e �Pæı��Å (Th. 116–19)

But then Gaia the broad-breasted, the ever unshakable seat of allthe immortals who possess the peaks of snowy Olymposand murky Tartarus in the recess of the broad-pathed earth . . .5

The phrase ‘ever unshakable seat’ (�� I�çÆºb ÆN��) is a metaphorfor an evolved, stable state of the cosmos. As she evolves from ‘then’to ‘now’ time, Gaia is fundamentally tied to cosmic stability: she isboth committed to it and particularly sensitive to disturbances of it.In a sense, then, her purpose and the direction of the poem are oneand the same. She is a kind of temporal glue persisting across gen-erational orders.

As time moves forward in the succession plot, Gaia relinquishessome of her initial power. When she parthenogenetically producesOuranos equal to herself, in her first act of creation, there is a gapbetween her intention, expressed in two sequential purpose clauses,and what almost immediately transpires:

ˆÆEÆ � ��Ø �æH�� b� Kª��Æ�� r�� øı�Bfi

ˇPæÆ�e� I��æ��Ł’, ¥�Æ Ø� ��æd ��Æ ŒÆº��Ø,Zçæ’ �YÅ ÆŒ�æ��Ø Ł��E �� I�çÆºb ÆN�� (Th. 126–8)

Gaia first of all bore, equal to herself,starry Ouranos, so that he might cover her on every side,in order that he/there would be an ever unshakable seat for the blessed

gods.

The first purpose clause, ‘so that he might cover her on every side’(¥�Æ Ø� ��æd ��Æ ŒÆº��Ø), indicates Gaia’s intention, capturingher focalization (what she sees or visualizes or imagines) as sheimplements her plan: she produces Ouranos expecting that they willbe commensurate (symmetrical and equal) and knowing that, on herown, she cannot maintain cosmic order. Yet the connotations of theverb ŒÆº��Ø exceed Gaia’s vision: to ‘cover’ can also mean to ‘bury’,to ‘eclipse’. Thus she is mistaken in believing that the presence ofOuranos equal to herself will stabilize the cosmos. As it turns out,the fact that Ouranos is ‘equal to herself ’ is the source of theireventual conflict. On a physical level, the image of Sky covering or

5 Translations of Hesiod are my own. Those of the Hymns are by Rayor (2004).

258 Nancy Felson

roofing over Earth, both having identical dimensions, invokes a stableform. It is also an anthropomorphic image of coitus, with the trad-itional ‘male on top’, thus implying male domination and femalesubordination.6

The translation of the second purpose clause is problematic: is thesubject of �YÅ (‘would be’) Ouranos, which seems most natural, orGaia herself;7 or could �YÅ even be existential, ‘so that there would bean ever immovable seat for the blessed gods’? If the subject is Our-anos, then Gaia is relegating to her son and first mate half of herinitial prerogative to be the ever immovable seat for Olympian andchthonic gods. Now he is to be that stable seat for the blessed godswho inhabit Olympus—the ÆŒ�æ��Ø Ł��E —while she retains thatprerogative for the chthonic gods who inhabit Tartarus.8

Gaia’s belief that stability will require a male consort accountsnot only for her production of Ouranos equal to herself but also forher successive support of one male entity after another to be the rulerof gods and men. She consistently expects the right king to maintainorder and stability. With this in mind, she acts, time and again, totip the balance in favour of one potential male ruler or another. Shealso encounters one disappointment after another. As a characterin the story (and not as the first prophetess), Gaia is neither clairvoy-ant nor perspicacious. She herself evolves over time, at first producingthe Cyclopes and Hundred-Handers with no particular forethought,but later giving birth to Typhoeus out of exasperation: she is enragedbecause Zeus’ lightning and thunderbolts have devastated her anddisturbed the other elements within the universe, the sky and the seas.

Although hatred between generations is most intensely expressedas hatred between father and son,9 in Hesiod’s Theogony ‘monstrous

6 Cf. Lakoff and Johnson (1980), 14–21, on ‘orientation metaphors’, such asup–down, in–out, front–back, on–off, deep–shallow, and central–peripheral. Theyall have to do with spatial orientation.

7 To translate 128, ‘so that she would be . . . ’, as in Most (2006), requires a difficultchange of subject; a more natural subject of �YÅ is Ouranos, the subject of thepreceding verb ŒÆº��Ø (see West [1966], 198). But an existential translation, asI have proposed, makes sense as well, since Gaia’s capacity to remain a stable seatforever is intimately tied to the existence of such a stable seat. The goal of ensuringsuch stability is projected forward into the ‘now’ time of the poem.

8 This would mark a split between the Olympian and the Chthonian realms. In anycase, Gaia expects to retain to the end of time (ÆN��) the value she has from her birth,not on her own but through the anticipated stabilizing existence of Ouranos.

9 The emotion of hatred permeates tales of intergenerational conflict. Hatred of afather motivates a son’s act of aggression—as in the case of Kronos’ castration ofOuranos or Zeus’ defeat of Kronos in the Titanomachy. A father’s hatred for his


Gaia’ helps create the conditions for such mutual hatred—in part byproducing dread children (��Ø�a �Œ�Æ), in part by fuelling competi-tion between an existing male power, such as Ouranos, and newcreations who might vie for the position of king of gods and men.For example, enraged at Ouranos for keeping some of their offspringin her recesses, she incites the young Kronos to castrate his father.Here, as often, hatred between father and son may already haveexisted, but Gaia helps ignite it.

In the Theogony, ��Ø�� (‘dread’) marks a child (�Œ��) as mena-cing to its parents. The adjective reflects the focalization of thevulnerable and replaceable father, to whom, in particular, an off-spring appears to be dread.10 The superlative ‘most dread of children’(��Ø��Æ�� /-�Ø �Æ��ø�) depicts Kronos at 138 and certain offspringof Gaia and Ouranos at 154 (‹��Ø . . .ˆÆ�Å �� ŒÆd ˇPæÆ��FK��ª��).11 Along with synonyms such as ¼�ºÅ�� (‘unapproach-able’), this derogatory adjective is regularly used as well for monsterswithin and outside the poem, oversized, often deformed creaturesthat threaten cosmic order; in particular, the descendants of Phorcysand Ceto (Th. 270–336). The use of ��Ø�� to describe potentialusurpers as well as genuine monsters aligns both types of disruptersof order in the same paradigmatic set. Some ‘dread children’ (��Ø�a�Œ�Æ) are literally monstrous (the Hundred-Handers, Typhoeus,hybrid creatures); for others their ‘monstrosity’ indicates their

offspring is often proleptic: he especially hates the son predicted to dislodge him, theone he expects to usurp his throne. Aggression between fathers and sons ofteninvolves damaging their respective bodies or banishing and relocating the loser inthe contest. The struggle determines who will occupy (and monopolize) the seat ofpower—who, as king, will wield the sceptre, assign tasks and prerogatives to hissubordinates, and mediate quarrels, and who will have superior bodily strength/energy (IºŒ�/�� ). Zeus’ body is augmented by the accoutrements of thunder andlightning—metonymic extensions of his embodied self.

10 The use of ��Ø��Æ�� /-�Ø (‘dread’) for rebellious or overly powerful sons under-scores their potential to harm the father’s domain or realm or indeed (as for Ouranos)his very body. These designations align despised or despicable sons with the brood ofmonsters, deformed enemies of cosmic order for whom the adjective ��Ø�� (‘dread’) isregularly used, often accompanied by a string of adjectives.

11 The relative ‹��Ø (‘as many as’) probably designates only the Hundred-Handersand the Cyclopes but not the first brood of Titans. Yet this interpretation remainsproblematic, especially if one locates the hiding place (º�å� ) from which Kronosambushes and castrates his father (174, 178) inside the recesses or womb of motherEarth.

260 Nancy Felson

anticipated rebellion against their father or against the king. Thedesignation of Kronos and the second and third of Ouranos’ broodsas ‘most dread’ points to their capacity to wreak havoc, destabilize thecosmos, and usurp whatever power (or licence) is held by their fatherOuranos.

The last of the challengers of Zeus in the third generation of thesuccession myth is also a ‘dread’ (��Ø�� ) creature—Typhoeus—theyoungest (›�º��Æ�� ) and final offspring of Gaia after she mateswith Tartarus through golden Aphrodite (820–52). Like his olderhalf-siblings, especially the Hundred-Handers, Typhoeus is mon-strous in his deformed body, with a hundred heads of a dread serpent(��Ø��E� �æ�Œ�� ) coming from his shoulders (824–5). Moreover,the hypothetical statement, ‘he would have ruled over the mortalsand immortals, had not the father of men and gods taken note’ (ŒÆ�Œ�� ‹ ª� Ł�Å��E�Ø ŒÆd IŁÆ��Ø�Ø� ¼�Æ��, j �N c ¼æ’ O�f ��Å�� Æ�cæI��æH� �� Ł�H� ��, 837–8), designates Typhoeus as an ‘almost-usur-per’. In this respect he resembles and is in the same set as Kronosand Zeus.

Gaia, in the end, values not random disorder but political andcosmic stability, which explains her active role in the successionmyth: ultimately, at her urging, the gods elect Zeus king.12 Shecomes to fully support him in the belief, never stated outright, that,under him, there will be stability. Before that, Gaia in the Theogony isthe prototypical mother who supports a son against a powerful andunjust father.

Theogonic elements lurking beneath the surface of the HomericHymns to Athena and Apollo help underscore the vast gap betweenthe early days of Olympian ‘history’ and its later denouement, a gapthat Clay articulates when she writes: ‘Between theogonic poetry andepic there remains a gap, one that is filled by the Olympian narrativesof the longer hymns. The major hymns, then, serve to complete the

12 This election precedes Gaia’s production of Typhoeus, which, in a synchronicreading, is difficult to interpret. Was Gaia having second thoughts about Zeus’capacity to rule or, perhaps, was she wishing to present Zeus with one last challengeto overcome? Hesiod exerts great effort to legitimize Zeus’ kingship and affirm theprinciple that Zeus will never be overthrown. The perpetuation of Zeus’ hegemonymust have had an important political function in archaic Greek society. On Gaia’smotivations in the Theogony, see Clay (2003), 26–7 with n. 43.


Olympian agenda and provide the clearest account of . . . the politicsof Olympus.’13

In the case of Hymn 28 to Athena, as Càssola puts it in hiscommentary, ‘the tradition received by the rhapsode contaminatesthe myth of Athena with a myth of divine succession’.14

1 . ATHENA: HYMN 28

Athena’s birth, as recounted in Hymn 28 and in a few other ancientsources, notably Pindar’s Olympian 7 (35–8), is a disruptive cosmicevent.15 She is fourth in the patriline that extends from Ouranos toKronos to Zeus. As such, she shares a set of parents with the son whowould have supplanted Zeus as king of the cosmos, had he been born(Th. 897–9). In this section, I will first demonstrate the presence ofthe usurper trait in Athena and then explore the manifold ways inwhich she differs from other divine offspring that threaten the cos-mos. Using selected passages from Hesiod’s Theogony and Homer’sIliad, Book 8, I will show how deeply Hymn 28 is imbued withcosmogonic and theogonic conflicts and how Athena evolves in thecourse of that short Hymn to become her father’s powerful ally. Hymn11 captures her as she has evolved into a ‘defender of cities’(Kæı��ºØ ). In that capacity, she will channel her extraordinaryforce towards positive and orderly ends for the host of citizens sheprotects as they come and go (Hy. 11. 4: Kææ��Æ�� ºÆe� N��Æ ��Ø�� ). This civic role is in line with her traditional supportof heroes against their (often monstrous) enemies.16

13 Clay (1989), 15, argues plausibly that ‘the hymns fill a gap between Hesiod’sTheogony, which depicts the conflicts of the older gods and Zeus’s rise to power, andthe settled Olympian pantheon of Homeric epic, where Zeus’s supremacy is assuredand conflicts between gods are confined to squabbling’.

14 Càssola 419: ‘la tradizione accolta dal rapsodo contamina il mito di Atena conun mito di successione divina.’

15 On Athena’s birth in full panoply, see AHS 424–5; Penglase (1994), 230–6; andDeacy (2008), 21–5, and cf. Pindar’s account (O. 7. 38), where Hephaestus delivers thegoddess, and Sky and Mother Earth react to her birth with voiceless agitation andabsolute immobility.

16 Athena supports winners, like Heracles, Theseus, and Bellerophon, againstmonstrous adversaries, according to Willcock (1970), 6.

262 Nancy Felson

Athena’s power for good develops precisely from the fact that, hadnot Zeus swallowed her pregnant mother and given birth to her as hisown, and had she not exhibited self-restraint, she would have under-mined his power. In her unrealized narrative, she would have repli-cated what, in previous generations, Zeus did to Kronos and Kronosto Ouranos: displaced her father.

In Hesiod’s account Zeus has been elected king of the gods when hetakes his first wife, Metis:

Z�f �b Ł�H� �Æ�Øº�f �æ��Å� ¼º�å�� Ł�� )B�Ø�,�º�E��Æ Ł�H� �N�ıEÆ� N�b Ł�Å�H� I�Łæ��ø�.Iºº’ ‹�� c ¼æ’ ��ºº� Ł�a� ªºÆıŒH�Ø� �Ł��Å��ŁÆØ, ��’ ��Ø�Æ ��ºøfi çæ�Æ K�Æ�Æ��Æ Æƒıº��Ø�Ø º�ª�Ø�Ø� �c� K�Œ��Ł�� Å��,ˆÆ�Å çæÆ��fi Å�Ø ŒÆd ˇPæÆ��F I��æ�� ·�g ª�æ �ƒ çæÆ��Å�, ¥�Æ c �Æ�ØºÅ��Æ �Øc�¼ºº� �å�Ø ˜Øe I��d Ł�H� ÆN�Øª��ø�.KŒ ªaæ �B �¥Ææ�� æ�çæ��Æ �Œ�Æ ª��ŁÆØ·�æ��Å� b� Œ��æÅ� ªºÆıŒ��Ø�Æ 'æØ��ª��ØÆ�,r�� å�ı�Æ� �Æ�æd �� ŒÆd K��çæ��Æ ��ıº��,ÆP�aæ ��Ø�’ ¼æÆ �ÆE�Æ Ł�H� �Æ�ØºBÆ ŒÆd I��æH�X�ºº�� ŁÆØ, ��æ�Ø�� q��æ �å��Æ·Iºº’ ¼æÆ Ø� Z�f �æ��Ł�� c� K�Œ��Ł�� Å��,u �ƒ �ıçæ��ÆØ�� Ł�a IªÆŁ�� ŒÆŒ�� . (Th. 886–900)

Zeus, king of the gods, took Metis (Wisdom) as his first wife,she being the most wise of gods and mortal men.But right when she was about to give birth to the goddess, bright-eyed

Athena, then, after deceiving her mind through cunningand with guileful words, he put her into his belly,by the crafty plans of Earth and of starry Sky;for this was how they had declared it to him, lestanother of the eternally living gods should have the kingly honour

instead of Zeus.For it was destined that exceedingly wise children would be born of her:first she would give birth to a maiden, bright-eyed Tritogeneia,owning strength equal to her father and sound counsel,and then she would give birth to a son, a king of gods and of men,with an extremely violent heart.But before that could happen Zeus put her into his belly,so that the goddess would advise him about good and evil.

Here Hesiod highlights the joint role of Gaia and Ouranos incounselling Zeus, with the use of çæÆ��fi Å�Ø (‘by their cunning


plans’, 891) and its cognate çæÆ��Å� (‘the two of them declared/advised’, 892). The purpose clause introduced by ¥ �Æ (‘lest some otherone of the eternally living gods hold the kingly honour instead ofZeus’, 892–3) indicates that, at this juncture, they both support Zeus’remaining in power. The prophecy itself is given in indirect discourse.It pairs Athena and her potential younger brother, designating themboth as exceedingly wise children (��æ�çæ��Æ �Œ�Æ). That they areboth in the prophecy and that the prophecy motivates Zeus toswallow Metis pregnant with Athena mark Athena too as a potentialthreat, like other children (�Œ�Æ) in prophecies. Moreover, the posi-tive term ��æ�çæ��Æ (‘wise’) applies to the brother as well, who, as wesoon learn (897), is destined to become king. At 895–6 the two aredifferentiated: the maiden Tritogeneia, first born, will possess ‘mightequal to her father’ and sound counsel, while her brother, whomMetis ‘was going to bear as king of gods and men’, will possess ‘avery violent heart’, like the Cyclopes at 139 (��æ�Ø�� q��æ �å��Æ ,‘having overweening hearts’) and the Hundred-Handers at 149(��æ�çÆ�Æ �Œ�Æ, ‘overbearing children’). Though the two participialclauses, introduced by �å�ı�Æ� (896) and �å��Æ (898) respectively,distinguish the siblings sharply from one another, the equation ofAthena’s strength (�� , 896) with her father is surprising andarresting, especially if we recall that Gaia produced Ouranos ‘equalto herself ’. This ‘equality’ seems to invite competition. Should weimagine that, despite being female, being born from Zeus’ head andhaving sound counsel (K��çæ��Æ ��ıº��), Athena could have held thekingly honour instead of Zeus? This (and this alone) would explainwhy Zeus swallows Metis as she is about to give birth to Athena andwhy he feels the need to usurp the prerogative of the female womb.

Moments of theogonic tension in Homeric epic look back to Zeus’victories over his predecessors, which must have been part of theextensive oral poetic tradition. The Iliad, in particular, preservestraces of dissent which threaten to disrupt the cosmic order nowconsolidated under Zeus’ kingship, despite the general truth that ‘theHomeric poems show us the fully perfected and stable Olympianpantheon in its interaction with the heroes’, while ‘the Theogonyreveals the genesis of the Olympian order and ends with the trium-phal accession to power of Zeus’.17 A brief overview of these traces

17 Clay (1989), 15.

264 Nancy Felson

reveals four main points that illuminate Hymn 28: (1) the treatmentof Athena as if she were Zeus’ son and heir; (2) the collusion againstZeus by Athena and Hera and Zeus’ angry reaction, which includes athreat to hurl any disobedient Olympian to gloomy Tartarus; (3)Zeus’ appeal to his might and his boast of being superior to all theother gods put together; and (4) Zeus’ reference to Athena’s sword inhis final threat to his daughter as ��º�æØ��, ‘huge, monstrous’ (Il. 8.424), the same epithet that Hesiod applies to Kronos’ implement, thesickle, with which he castrated his father (Th. 179: ��º�æØ�� . . .–æ�Å�), and regularly to Gaia.

As Book 8 begins, Zeus commands all the gods to stay out of the frayand to refrain from protecting the Achaeans against Hector (8. 5–27),re-asserting his supreme power. He threatens to hurl all who disobeyhis command to gloomy Tartarus (13–16)—a traditional motif.18

Athena is the first to object. After acknowledging his strength (31–2), she says simply, ‘We (i.e. Hera and I) pity the Danaans’ (˜Æ�ÆH�Oº�çıæ��Ł’, 33). Zeus, smiling, tries to placate his dear child (ç�º��Œ� , 39) and claims: ‘I want to be gentle to you’ (KŁºø � ��Ø X�Ø� �r�ÆØ, 40). The epithet X�Ø� (‘gentle’) is used especially for a father, ora king like a gentle father,19 as opposed to the reckless father(I��ŁÆº� �Æ��æ) of Hesiod’s succession myth, as we shall see.

When the quarrel resumes at 8. 350–484, Hera and Athena form analliance that structurally parallels Gaia’s alliance with Kronos. AtTheogony 164–6 Gaia addresses her offspring as ‘sons of mine andof a wicked father’ (�ÆE�� K�d ŒÆd �Æ�æe I�Æ�Ł�º�ı). She asks themto obey her and together avenge their father’s evil outrage, since hewas the first to devise unseemly deeds (I�ØŒÆ ��Æ�� æªÆ). Simi-larly, at Il. 8. 350–6 Hera approaches Athena to urge that the two ofthem join forces because of the evil works of Hector, thus blaming ahuman, not Zeus, for her revenge. The two goddesses, Zeus’ wife anddaughter, act as partners in an act of disobedience: together, they willdefy Zeus’ earlier command and try to enter the fray.

18 Harrell (1991), 307–18, argues that the Iliad poet and Hesiod took this tradi-tional ‘rhipto motif ’ from a common tradition. She cites two Hesiodic fragments: fr.54a M–W (¼ P. Oxy. XXVIII 2495 fr. 1a) and fr. 30 M–W (¼ P.Oxy. XXVIII 2481 fr.1. 15–23). The former, though poorly preserved, seems to suggest that Zeus, in angerat Apollo, was about to hurl him to Tartarus and would have killed him, presumablyfor attacking the Cyclopes (cf. ps.-Apollod. 3. 10. 4).

19 See Felson (2000), 89–98, and (2002), 189–200, on these two divergent para-digms of fatherhood.


Athena’s response to Hera at Iliad 8. 358–80 resembles that ofKronos to Gaia at Theogony 170–2, when he accepts her challengeand tells her of his disregard for Ouranos. Athena characterizes Zeusas mad and evil and a thwarter of her intents:

Iººa �Æ�cæ ��e çæ��d Æ��ÆØ �PŒ IªÆŁBfi �Ø

�å�ºØ� , ÆNb� IºØ�æ� , KH� ��ø� I��æø�� · (Il. 8. 360–1)

But my father rages in his evil mind.Hard, forever wicked, he is the thwarter of my impulses.

Then she indirectly aligns herself with Hera through her oppositionto Thetis. She claims that, although Zeus is in her debt, he does notremember how she protected Heracles (362–9); i.e., he is ungratefulto her. Then, in strong language, she adds that he hates her (��ıª�Ø,370) and has accepted the plans of Thetis, but that ‘there will be a daywhen he will again call her his dear bright-eyed one’ (��ÆØ �� ‹�’ i�Æs�� ç�ºÅ� ªºÆıŒ��Ø�Æ �Y�fi Å, 373). Her rivalry with Thetis for Zeus’attention recalls Hera’s own rivalry when she detects that Zeus hasconspired with Thetis to honour Achilles at the Achaeans’ expense(1. 539–43 and 552–9). Even though Athena aligns herself with herfather’s wife, the quarrel does not escalate: Hera and Athena even-tually back down and comply with Zeus’ command.

Collusion between Athena and Hera against Zeus is a familiarmotif in the Iliad: the two heckle Zeus at critical moments in thepoem when he acts, or threatens to act, unilaterally, as when heexpresses his desire to rescue his son Sarpedon and Hera objectsand reminds him that not all the other gods may approve of him(16. 431–9) or when he expresses his desire to rescue Hector andAthena reacts with similar language (22. 178–81).

In Book 8, after railing against Zeus, Athena urges Hera to arm forbattle (374) and she herself takes off her peplos and dons the chiton ofcloud-gathering Zeus (385–8), implying (by wearing his garment)that she is taking his place. When Zeus learns of their defiance,instead of carrying out his earlier threat to hurl them to Tartarus,he sends Iris to deliver further threats: that he will lame their horses,cast them from their chariots, and shatter their chariots, using histhunderbolt, ‘so that the bright-eyed one may come to understandwhat happens when she fights with her father’ (Zçæ! �N�Bfi ªºÆıŒH�Ø ‹�! i� fiz �Æ�æd �åÅ�ÆØ, 406).

266 Nancy Felson

In the Theogony, mother–child collaboration and father–childalienation precede the violent crossroads confrontation, wherebythe son replaces the father. In Iliad 8, though, Zeus never comes toblows with Athena and Hera; the menacing speeches they exchangeexpress a weaker form of the kind of sedition that we find in thetheogonic mother/son conspiracies to unseat or unman a father andking. Moreover, the subversive pact between the hyper-masculineAthena and her disgruntled stepmother Hera parallels the interge-nerational alliances in the Theogony between Kronos and Gaia, Zeusand Rhea, Zeus and Gaia, and Typhoeus and Gaia.

Against this background, I now turn to Hymn 28 to Athena. Thishymn encapsulates an essential feature of the representation of thegoddess, the juxtaposition of her two distinctive traits of rebellious-ness and self-restraint. The former trait appears at the first of the twopossible moments of intergenerational conflict: right after birth butnot at hēbē. I have divided the Hymn into eight segments, as follows(line numbers in brackets):

A. conventional beginning, including a string of traits (1–4)B. relative clause (4–6)C. reaction of immortals (6–7)D. event 1 (7–9)E. reaction of cosmos (9–14)F. event 2 (14–16)G. reaction of Zeus (16)H. conventional couplet ending (17–18)

A. [—Æºº��’ �ŁÅ�Æ�Å�, Œı�æc� Ł��, ¼æå�’ I��Ø�ªºÆıŒH�Ø� ��º�Å�Ø� I��ºØå�� q��æ �å�ı�Æ�

�ÆæŁ�� ÆN��Å� Kæı��ºØ� IºŒ��Æ�

'æØ��ª��B,] B. [�c� ÆP�e Kª��Æ�� Å��Æ Z�� B KŒ Œ�çÆºB , ��º��œÆ ��å�’ �å�ı�Æ�åæ��Æ �ÆçÆ��ø��Æ·] C. [��Æ �’ �å� ��Æ ›æH��Æ IŁÆ��ı ·] D. [m �b �æ��Ł�� ˜Øe ÆNªØ�å�Ø�K��ı�ø þæ�ı�� I�’ IŁÆ��Ø� ŒÆæ��ı��Æ�’ O�f� ¼Œ��Æ·] E. [ªÆ �’ Kº�º�Ç��’ ! …ºı�� Ø�e� ��e �æ�Å ˆºÆıŒ��Ø�� , Içd �b ªÆEÆ��æ�Æº�� N�åÅ��, KŒØ��ŁÅ �’ ¼æÆ �� Œ�Æ�Ø ��æçıæ�Ø�Ø ŒıŒ�� , ��å�� ’ –ºÅK�Æ��Å · ��B�� ’ % -��æ�� IªºÆe ıƒ� ¥��ı TŒ��Æ �Åæe� åæ��] F. [�N�� Œ��æÅ


�¥º��’ I�’ IŁÆ��ø� þø� Ł��Œ�ºÆ ��åÅ—Æººa �ŁÅ�Æ�Å·] G. [ª�ŁÅ�� b Å��Æ Z�� .]H. [ŒÆd �f b� �o�ø åÆEæ�, ˜Øe �Œ� ÆNªØ�å�Ø�·ÆP�aæ Kªg ŒÆd ��E� ŒÆd ¼ººÅ ��’ I�Ø�B .] (Hy. 28. 1–18)

A. [I sing the glorious goddess Pallas Athena,owl-eyed deity with crafty wisdom and steady heart,revered virgin, stalwart guardian of the city,Tritogeneia.] B. [From his august head, cunning Zeushimself gave birth to her, born in warlike armourof gleaming gold.] C. [Awe seized all the gods watching.]D. [She sprang quickly from his immortal headand stood in front of Zeus who bears the aegis,shaking her sharp spear.] E. [Great Olympus reeledviolently beneath the might of her shining eyes,the earth let out an awful cry, and the deep shifted,churning with purple waves. Suddenly the seaheld still and the shining son of Hyperion haltedhis swift horses a long while] F.[until the maidenPallas Athena lifted the godlike armourfrom her divine shoulders,] G. [and wise Zeus rejoiced.]H. [Hail, child of aegis-bearing Zeus—but I will remember you and the rest of the song.]

The string of Athena’s attributes is enumerated within the invoca-tion (A), with cumulative force, a rhetorical strategy that underscoresboth the hybridity and the excess of monstrous offspring.20 Thegoddess has an implacable heart (I��ºØå�� q��æ �å�ı�Æ�, 2); in theTheogony she has a life-force equal to her father’s and prudentcounsel (r�� å�ı�Æ� �Æ�æd �� ŒÆd K��çæ��Æ ��ıº��, 896), whileher brother has an excessively violent spirit (��æ�Ø�� q��æ �å��Æ,898). Athena in the Hymn is a chaste virgin, a mighty protector ofcities, Triton-born. In the relative clause at B, she is the one whomZeus, after swallowing Metis, bore from his august head in fullpanoply; a second string of attributes describes her weapons (5–6).A series of three reactions ensues. First, at C, awe (��Æ , 6) grips theimmortal gods as they witness the birth. The narration of the birthevent (D) tells what they see even as it reiterates B, adding details.One detail in particular suggests aggressive belligerence: brandishing

20 Cf. Th. 146 (Cyclopes), 148 and 153 (Hundred-Handers), 297–300 (Echidna),307 (Typhoeus), 312 (Cerberus) and 320 (Chimera).

268 Nancy Felson

the sharp javelin (��Æ�’ O�f� ¼Œ��Æ, 9). To all this—the birth infull panoply and the brandishing of the javelin—the cosmos re-sponds with anxious anticipation (E).21 At this pivotal moment inthe story, the listener/reader wonders what will happen next. Fromthe focalization of not only the immortal gods (C) but all the naturalelements which see Athena emerge in her full panoply, this particulargoddess, so much like her father, has the potential to disrupt cosmicorder.

The cosmic turbulence in the Hymn (9–14: E), is brought on by thebirth of a fully armed Athena brandishing her javelin. With greatcompression, three elements—sky, earth, and sea—experience com-motion. Suddenly, the sea and the sun reverse the situation, each by asurprise move that seems to get Athena’s attention. Her response is toremove her armour from her shoulders, and this act immediatelyintroduces calm. Before that, noisiness and commotion are evident inEarth’s letting out an awful cry (Içd �b ªÆEÆ j ��æ�Æº�� N�åÅ��,10–11) and in the churning motion of the deep (KŒØ��ŁÅ �’ ¼æÆ �� j Œ�Æ�Ø ��æçıæ�Ø�Ø ŒıŒ�� , 11–12).

Section E shares themes and formulaic diction with the upheavals,respectively, of the Titanomachy (Th. 629–721) and especially theTyphonomachy (Th. 820–900). In the former, the first cosmic battle,noise and widespread conflagration signal the scale of the upheavaland the collapse of order: ‘All around, the life-giving earth roared as itburned, and all around the great immense forest crackled; the wholeearth boiled, and the streams of Ocean and the barren sea’ (Içd�b ªÆEÆ ç�æ��Ø� K�Ææ�ªØÇ� j ŒÆØ��Å, º�Œ� �! Içd �ıæd �ª�º!¼�� oºÅ. j �Ç�� b åŁg� �A�Æ ŒÆd ! 0Œ�Æ��E� Þ�ŁæÆ j �� !I�æ�ª�� ·, 693–6). Moreover, noise provides the ground for thecomparison in the striking simile, which invokes an image either ofthe first coupling of Earth and Sky—a regression to the time whenearth and sky were not yet separate—or of a cataclysmic collapse ofSky on Earth as a result of strife:22

21 For other sources on the birth in full panoply, see Càssola 419–21.22 I read the Titanomachy and Typhonomachy synchronically, while acknow-

ledging the ways in which they reduplicate one another and the problematic joins:cf. especially Solmsen (1982) and, on the possibility of a synchronic rather than adiachronic reading of the Theogony and of a middle ground between a Unitarian and aneoanalytic approach to the poem, Mondi (1984), 325–44. For Most (2006), 59 n. 38,the simile implies that Zeus’ actions in this epic intergenerational battle are undoingthe union with which Gaia instigated change: ‘the analogy is not to some cataclysmic


‰ �N ÆEÆ ŒÆd ˇPæÆ�e �Pæf o��æŁ��

��º�Æ��· ��E� ª�æ Œ� ªÆ ��e ��F�� Oæ�æ�Ø�B b� Kæ�Ø��Å , ��F �! �ł�Ł�� K��æØ�� ·�� F�� ª�� Ł�H� �æØ�Ø �ı�Ø��ø�. (Th. 702–5)

It seemed just as if Gaia and broad Ouranos up abovewere approaching one another:23 for such a great thud rises upas she is dashed down and as he dashes her down from on high;so great a thud was produced as the gods ran together in strife.

The reeling of Sky, the awful cry of Earth, and the churning of thedeep in the Hymn resonate with the turbulence of the same threeelements in the Typhonomachy, where parallel diction and theme(even without the destruction by fire, caused by Zeus’ lightning andthunderbolt in the Typhonomachy) assure us that here too we have acase of cosmic unrest.24

At F in the Hymn, ‘after a long while’ (�Åæe� åæ��) Athena liftsher armour from her shoulders. The duration of time allows Athenato take time considering her next move. The conjunction �N��

(‘until’), introducing F, implies that this cosmic unrest ends withAthena’s act of self-restraint, whereby the goddess breaks the patternof conflict and next-generation (usually male) usurpation. Moreover,because she is a virgin, she will not replicate the female propensity inthe succession myth toward using the womb as an instrument toretaliate against male brutality by producing a challenger. Thusneither as a ‘male’ nor as a female will she attempt to unseat herfather and threaten the cosmic order. The verb ª�ŁÅ�� (‘he rejoiced’,16) in G gives us access to Zeus’ focalization not so much on the birthin full panoply of his daughter as on her decision to disarm.25 The

final collapse of the sky onto the earth, but instead to the primordial sexual unionbetween Sky and Earth.’ I would add that the imagery of Ouranos dominating Gaiafrom above suggests a violent and conflictual sexual union, perhaps even drawing onthe image of a (male) victor raping a vanquished (female) city.

23 The verb ��º�Æ�� (‘approach’) can have a hostile valence, as here, or a positiveone.

24 Cf. especially how, in the Typhonomachy, ‘great Olympus was shaken . . . andthe earth was groaning’ (ªÆ ��º��Ç��’ ! …ºı�� j . . . K��åØÇ� �b ªÆEÆ, Th.842–3), ‘all the earth was seething, and the Sky and Sea’ (�Ç�� b åŁg� �A�Æ ŒÆd�PæÆ�e M�b Ł�ºÆ��Æ·, Th. 847), and ‘huge/monstrous Gaia was groaning’ (��åØÇ��b ªÆEÆ ��º�æÅ, Th. 858). The turbulence that arises from this clash between Zeus andTyphoeus is the very opposite of the ‘stable seat forever’ (�� I�çÆºb ÆN��) of theearlier purpose clause (Th. 128).

25 On ª�ŁÅ�� (‘he rejoiced’) as relief cf. Odysseus at Od. 13. 250, when he realizeshe has come home.

270 Nancy Felson

placement of Zeus’ rejoicing right after Athena’s removal of herweapons marks it as a sign of relief and not simply an indicator of afather’s pride in his offspring and joy at her birth. As noted earlier,births are intrinsically unsettling: in the Theogony, anxious fathersgreet with apprehension the birth of sons (especially those that arehuge, monstrous, and over-manly). Here, the hymnist defers theaccount of Zeus’ reaction to Athena’s birth until the new-borndaughter has herself eliminated the possibility that she will becomea fourth-generation usurper. In the last couplet of the Hymn (17–18),as the hymnist bids farewell to Athena, he surrounds the child (�Œ� )by two genitives that describe her father (˜Øe and ÆNªØ�å�Ø�, ‘ofaegis-bearing Zeus’).26 This arrangement points to Athena’s destiny,to be her father’s obedient and unrebellious child.27

The Hymn thus encapsulates both Athena’s potential to challengeher father and establish her own rule and her voluntary subordinationto her father’s cosmic order. It celebrates her signature quality, thepractice of self-restraint: she will use her energy (�� ) ‘equal’ to herfather’s to uphold the cosmos over which he presides as king. And shewill channel this energy to protect cities against their enemies andhelp citizens as they come and go (Hy. 11), as she traditionallyenhances both the �� (‘energy’) and IºŒ� (‘might’) and often theB�Ø (‘cunning’) of the victorious heroes whom she favours.

2 . HYMN 3 TO APOLLO

The Hymn to Apollo recounts the distinctive ways in which Apollo’spotential for rebellion is tamed and redirected. Even though Apolloreduplicates (and thus threatens) his father in multiple ways, he turnsout to be reliable and orderly and comes to assume a legitimate placein his father’s household. Though Apollo starts out as a potentiallythreatening child, he soon becomes a supporter of cosmic stabilityunder his father. The Hymn tells the story of how (and why) hedirects his formidable energy against unruly forces, such as Pythoand Telphousa. Apollo, then, belongs to the class of heroes who

26 On the aegis, especially in Homeric epic, cf. Gantz (1993), 84–5.27 Cf. Athena’s claim, not attested till the fifth century (A. Eu. 827–8), to know

where Zeus keeps the thunderbolt implies that Zeus has confidence in his dutifuldaughter.


return home at hēbē to assume their legitimate role in their father’shousehold and kingdom. His story, as recounted in the Hymn, be-longs to the tale-type of divine succession myths, a large category thatincorporates (with a difference) hero-tales of exile and return, like thestories of Jason, Theseus, and Bellerophon.

The proem of Hymn 3 encapsulates both Apollo’s capacity todisrupt order and his entry into the divine community:

��ÆØ �P�b º�ŁøÆØ ��ººø�� Œ��Ø�,‹� �� Ł��d ŒÆ�a �HÆ ˜Øe �æ��ı�Ø� N��Æ·ŒÆ� Þ� �! I�Æ&��ı�Ø� K�d �å��e� Kæå��Ø�

�� Iç! ��æ�ø�, ‹�� çÆ��ØÆ ��Æ �Ø�Æ��Ø.¸Å�g �! �YÅ �� ÆæÆd ˜Ød ��æ�ØŒ�æÆ��øfi ,l ÞÆ �Ø�� ! Kå�ºÆ�� ŒÆd KŒº�œ�� çÆæ�æÅ�,ŒÆ� �ƒ I�! NçŁ�ø� þø� å��æ��Ø� �º�F�Æ

�� Æ��ŒæÆ�� æe Œ��Æ �Æ�æe ��E�

�Æ��º�ı KŒ åæı��ı· �e� �! K Łæ�� x�� ¼ª�ı�Æ.�fiH �! ¼æÆ �Œ�Ææ ��øŒ� �Æ�cæ ��Æœ åæı��øfi

��ØŒ�� ç�º�� ıƒ��· ��Ø�Æ �b �Æ�� ¼ºº�Ø��ŁÆ ŒÆŁ�Ç�ı�Ø�. åÆ�æ�Ø � �� ØÆ ¸Å��,�o��ŒÆ ��ç�æ�� ŒÆd ŒÆæ��æe� ıƒe� ��ØŒ��. (Apoll. 1–13)

I will remember and not forget far-shooting Apollo.Gods tremble as he approaches the home of Zeus:all rise from their seats as he draws nearwhen he stretches his gleaming bow.Only Leto stays beside Zeus who delights in thunder.She unstrings Apollo’s bow, closes his quiver,lifts the bow from his mighty shoulders,hangs it from a golden peg on a pillar near his father,leads him to his throne and bids him sit.His father hands him nectar in a golden cup,welcoming his dear son—then the other godsreturn to their seats. Queen Leto rejoicesthat she bore a strong son, an archer.

In the proem, which chronologically takes place at Apollo’s hēbē, theyoung god approaches Olympus with bow drawn taut, as if he is onthe attack.28 The gods tremble at his arrival and only take their seatsafter Leto has removed his weapons and Zeus has offered nectar,

28 For a full discussion of the tenses in the proem, see Clay (1989), 23–9, whoemphasizes the timeless quality of the aorists, and especially Bakker (2002), 65–7,76–7, who treats the present tenses like the comparans of a simile, framing the series ofaorists. The latter are not temporal but ‘perceptual aorists’. My focus on Apollo’s

272 Nancy Felson

greeting his son with a welcoming toast that incorporates the younggod into the divine community and into his father’s home. Together,these benign parental acts eliminate any threat the youth might haveposed. This opening scene on Olympus, where Apollo appears brand-ishing his bow and frightening the assembled gods, gives substance tothe rumour Delos quotes at lines 67–9, when she explains to Leto herreluctance to become the birthplace of Apollo.

The depiction of Apollo’s arrival on Olympus captures one of hissalient features: he is the quintessential kouros (‘young man’), as hisregular epithet, IŒ�æ��Œ�Å (‘of the unshorn hair’, 134), indicates.29

In terms of the story pattern for intergenerational rivalry presentedabove, Apollo arrives home (i.e., to his father’s house) at hēbē,returning like countless heroes (and like Zeus at Th. 492–3) once heis of age, with his bow fully drawn and his mood aggressive. For theyoung Apollo, as for the ��Ø�a �Œ�Æ (‘dread children’) in the Theog-ony, there are two critical moments of danger: at his birth and uponhis arrival at Olympus.

In these first thirteen lines of the Hymn, the behaviour of each ofApollo’s parents differs markedly from that of succession-myth par-ents in the Theogony. In fact, all the family members in the Hymn(mother, father, and son) treat one another in a manner opposite totheir theogonic counterparts. Leto, who disarms Apollo, contrastswith Gaia, the mother of Kronos, a most dreadful (��Ø��Æ�� ) off-spring, whom she arms and provides with a plan, a hiding place, andan implement for castrating Ouranos (Th. 161–2, 179–81). Zeus inthe Hymn welcomes Apollo into the Olympian community, whileOuranos banishes his children, at least the second and third broods,relocating them in the bowels of Gaia (Th. 156–8) and binding themin Tartarus (Th. 501–6 the Cyclopes, 617–23 the Hundred-Handers),and Kronos swallows his as each emerges from Rhea’s womb(Th. 459–62). While theogonic fathers fear they will be overthrownby their sons, Zeus in the Hymn, by now secure in his kingly power,

arrival at hēbē does not preclude their interpretations. The arrival at hēbē can also beseen as a timeless, exemplary arrival.

29 Cf. his role as the one who guides Telemachus to manhood at Od. 19. 86–8,where the beggar-Odysseus reassures Penelope that, even if Odysseus himself hasperished, ‘here is Telemachus, his son, by grace of Apollo grown such a man’ (Iºº’ X�Å�ÆE ��E� ��ººø�� ª� �ŒÅ�Ø, j 'ÅºÆå� ). Whatever else this passage suggestsabout the bow contest at the festival of Apollo (see Austin 1975, 245), it also marksApollo as the god who escorts a young boy to manhood.


harbours no such fear with respect to Apollo. Benign gestures fromhis parents defuse Apollo’s desire to take Olympus by storm.30 Laterin the Hymn (186–8) Apollo is completely and harmoniously incor-porated into the community of the gods, when he leads them in thedance.31

When Delos responds to Leto’s request for a birthplace by openlyexpressing the reasons for her hesitation to grant it, she not onlyarticulates the fear of all the lands previously visited by Leto butintroduces a negative characterization of the god (even while dis-claiming any responsibility for it):

º�Å� ª�æ �Ø�� çÆ�Ø� I��ŁÆº�� ººø�Æ

��ŁÆØ, ªÆ �b �æı�Æ��ı�� IŁÆ��Ø�Ø�ŒÆd Ł�Å��E�Ø �æ��E�Ø� K�d Ç��øæ�� ¼æ�ıæÆ�. (Apoll. 67–9)

They say that Apollo will be someone exceedingly recklessand will lord it greatly over immortalsand mortal men along the life-sustaining field.

Delos’ use of the term I��ŁÆº�� (‘reckless’) to explain why she isreluctant to provide a birthplace for Apollo places him in a paradig-matic set with theogonic usurpers (as well as ‘atasthalic’ figures inHomeric epic, like Aegisthus and the Suitors).32 By attributing thecharge of recklessness and haughtiness to rumour (‘they say’, ç��Ø�),the island avoids incurring the god’s wrath. Yet her fears remind theHymn’s audience of Apollo’s aggressive behaviour in the proem and,in a sense, reinforce the notion that I�Æ�ŁÆº�Æ (‘recklessness) isintegral to his character.33

30 Cf. how, in the Odyssey, Odysseus includes Telemachus in his plans and how, atthe Bow Contest, Telemachus desists from attempting to string the bow in obedienceto a signal from his father.

31 Cf. Il. 1. 472–4. On the relation between this later arrival scene and the arrivalon Olympus in the proem, see Bakker (2002), 80–1.

32 On I�Æ�ŁÆº�Æ as a mark of adolescent excess, see Felson (2000), 89–98. As thequintessential term for a disruptive individual in archaic poetry, ‘recklessness’ isinappropriate for an Olympian god. In the context of cosmic evolution and cosmicstability the terms ‘reckless’ (I��ŁÆº� ) and ‘dread’ (��Ø�� ) tend to characterize thesame entities.

33 Moreover, Delos’ trembling (Iººa �� æ�ø, 66) not only reiterates thetrembling and fear of the previously visited lands (K�æ�� ŒÆd K��Ø�Æ�, 47) butre-invokes the trembling of all the gods except Leto upon Apollo’s arrival at Olympus(�æ��ı�Ø�, 2).

274 Nancy Felson

Apollo’s rumoured recklessness is an indicator of his potential tobe the son who unseats his father; it identifies the newborn, evenbefore his birth, as destined to terrify and usurp, and is isomorphicwith (or performs an analogous function to) all the theogonic pro-phecies that predict a son’s overthrow of the father. As it turns out,however, females and humans, not father Zeus, have the most to fearfrom Apollo.

As Apollo comes of age, he directs his destructive energy against aserpent that resists him (Pytho) and a land/spring that defies (lies to)him (Telphousa). He also orders and threatens humans who mightdisobey him (the Cretan servants). Like his father, he uses violence toimplement cosmic order.

Apollo’s management of his violent attributes and tendencies ce-ments his bond with his father, rather than undermining their rela-tion. The hymnist accentuates the resemblance between father andson by incorporating a long digression on Hera’s rage at Zeus forgiving birth by himself and her active choice to produce Typhaon asZeus’ rival (300–54). Whether or not this episode was part of animagined ‘original’ hymn to Apollo, or has been interpolated at a latertime (perhaps when the Delian and Pythian portions were joined), isbeyond the scope of this paper.34 The important point about the Heraepisode is how very theogonic the goddess’s complaint, plan, andimplementations are and how much they form a continuation ofthe theme of the myth of divine succession, with its goal of ensuringa stable seat forever.

The key role of the mother in the succession plot is evident in theHymn to Apollo not only in the actions of the gentle mediator, Leto,but in those of Hera, her polar opposite. In the Hera episode, begin-ning at line 300 and focalized entirely by the goddess, Hera blamesZeus for starting the contest by producing Athena from his head apartfrom her (ŒÆd �F� ��çØ� K�E� �Œ� ªºÆıŒH�Ø� �Ł��Å�, j m �A�Ø�ÆŒ�æ��Ø ��Æ�æ��Ø IŁÆ��Ø�Ø�, 314–15). Athena’s preeminence(��Æ�æ��Ø) exacerbates the offence and contributes to Hera’s claimthat cloud-gathering Zeus is the first to dishonour her (‰ �!I�Ø�Ç�Ø� ¼æå�Ø ��ç�ºÅª�æ�Æ Z�� , 312). The offence is furthercompounded by the deformity of the child whom she bore alone,

34 See Richardson (2010), 126–31, on the Typhaon episode; he argues againstconsidering it to be an addition to the original version of the Hymn (126), as manyhave suggested; cf. in this volume Chappell (pp. 72–3).


Hephaestus, a cripple (ÞØŒ�e ��Æ , 317). When she hurls Hephaes-tus in anger from Olympus, Thetis rescues him and cares for him,thereby thwarting Hera’s destructive plan to destroy (or at least evict)her child. Such an opposition between the two goddesses and rivalscan also be seen in Book 1 of Homer’s Iliad (493–611).

From her public complaint to the rest of the Olympians Hera nowturns to Zeus, whom she reviles in direct address, using language thatcorresponds to his (perceived) slight to her. Most strikingly, shethreatens to devise some new evil thing in answer to his offence:

�å�ºØ�, ��ØŒØº�B�Æ, �� F� Å��ÆØ ¼ºº�;�H ��ºÅ �r� ��Œ�Ø� ªºÆıŒH�Ø� �Ł��Å�;�PŒ i� Kªg ��Œ�Å�; ŒÆd �c Œ�ŒºÅ�Å ��Å qÆ Þ! K� IŁÆ��Ø�Ø�, �Q �PæÆ�e� �Pæf� �å�ı�Ø�.Içæ�Ç�� F�, � �� Ø ŒÆŒe� Å��! O��ø.ŒÆd �F� � ��Ø Kªg ��å��ÆØ, u Œ� ª�Å�ÆØ�ÆE K� , ‹ Œ� Ł��E�Ø ��Æ�æ��Ø IŁÆ��Ø�Ø�,�h�� e� ÆN�å��Æ�! ƒ�æe� ºå� �h�! Ke� ÆP�B ·�P� ��Ø �N �P�c� �øº��ÆØ, Iºº! I�e ��E��Åº�Ł! K�F�Æ Ł��E�Ø ��ÆØ IŁÆ��Ø�Ø�. (Apoll. 322–30)

Cruel, cunning trickster, what else will you plan?How dare you bear owl-eyed Athena on your own?Could I not have borne her? I was still called yoursamong the immortals who live in wide heaven.Watch out that I do not devise some evil in return.I will scheme to bear a child who will be preeminent among the

immortal gods.I, at least, will not shame our holy marriage,but I will not approach your bed. Being far awayfrom you, I will still be among the undying gods!

Thereafter, Hera prays to Earth and wide Heaven above and theTitans to ‘grant a child apart from Zeus, in no way weaker in strengththan he, a child greater than Zeus by as much as Zeus is greater thanKronos’ (ŒÆd �� ÆE�Æ j ��çØ ˜Ø� , Å�� Ø ��Å� K�Ø��ıÆ Œ��ı: jIºº! ‹ ª� çæ��æ� ��ø, ‹�� ˚æ��ı �Pæ��Æ Z�� , 337–9). Then shedramatically enacts her threat by striking the earth with her massivehand (¥Æ�� åŁ��Æ å�Øæd �Æå��fi Å, 340). The earth shifts in response(ŒØ��ŁÅ �’ ¼æÆ ªÆEÆ ç�æ��Ø� , 341) and Hera rejoices in the sight,believing her prayer will be fulfilled (m �b N��F�Æ j �æ�� n� ŒÆ�aŁı��, 341–2).

276 Nancy Felson

In this long digressive passage, Hera quarrels with Zeus andattempts to thwart his will. They contend over who has the preroga-tive of childbirth, based on who produces the superior child. Hera,enraged, behaves toward Zeus in ways reminiscent of her fore-mothers, Gaia and Rhea, in the Theogony. She reacts like Gaia,when Ouranos interferes with her birthing and represses several ofher broods of offspring (pressing them back into her recesses) andlike Rhea, when Kronos undoes her birthing by ingesting each of heroffspring. Hera first threatens to retaliate and bear a child who wouldbe preeminent among the immortal gods (ŒÆd �F� � ��Ø Kªg��å��ÆØ u Œ� ª�Å�ÆØ j �ÆE K� , ‹ Œ� Ł��E�Ø ��Æ�æ��ØIŁÆ��Ø�Ø�, 326–7). Then she affirms her intention to undermineZeus’ kingship by producing a usurper (�Œ��): ‘let him be as muchstronger than Zeus as Zeus is stronger than Kronos’ (çæ��æ� . . .‹�� ˚æ��ı �Pæ��Æ Z�� , 339). In her speech, Hera aligns herselfwith Gaia of the Theogony. Like Gaia, she is using her womb as aninstrument of revenge. Moreover, by her action of striking the earthwith her palm, she involves Gaia in the production of her chthonicchild. Hera’s quarrel with Zeus, though not of the magnitude of Gaia’swith Ouranos or Rhea’s with Kronos, is in the same paradigmatic setas part of the succession myth. All these quarrels are cosmogonic.

To be sure, the lengthy Hera passage has the feel of an insertioninto the text of a pre-existing episode, like the Typhoeus episode inthe Theogony. It has, however, come down to us embedded in a Hymnin which Hera, in the Delian portion, detains Eileithyia in order toobstruct the birth of the dragon-slayer Apollo. Therefore, it is appro-priate to read the episode synchronically as undergirding the deeppolarization of male versus female in the Hymn.

The Hymn to Apollo enacts the transformation of the rebellious soneager to dominate into the obedient son eager to emulate his fatherwithout ever replacing him. The role of the gentle mother in uphold-ing family harmony rather than fuelling the natural antagonismbetween father and child is enacted by Leto in the proem, whileZeus’ extension of hospitality to his feisty son, when he returnshome at his metron hēbēs, is the gesture that invites filial obedience.For Apollo, the son, obedience to his father guarantees that he willhave a legitimate place in the universe and will have a number ofrealms in which he can be preeminent. Thus the tone of the Hymnthat recounts the birth and exploits of the god Apollo draws on the


narrative pattern of the succession myth to depict, by way of contrast,the harmonious, stable, evolved hierarchic positioning of the newchild within the cosmos.

The Hymn celebrates the young god as a figure who, instead ofrebelling, attacks and eliminates two female entities, Pytho (357–62)and Telphousa (382–7), and thereby consolidates his masculinity.35

In his defeat of these two, Apollo replicates his father’s victory overTyphoeus in the Theogony (853–68). He will, in addition, disseminatethe word of Zeus the father through his own oracle at Delphi (132).

By the end of the poem, Apollo is incorporated and tamed: he willnever undermine his father, though he may, like other youths, ex-perience ‘benign regression’.36 He has evolved from a potential chal-lenger to his father’s staunch ally. He dominates a series of potentiallydangerous females much as Zeus in the Theogony incorporates Metis:he acquires the name and locale first of Delos, then of Telphousa, andfinally of Pytho. He specifically resembles his father not only in hisaggressive arrival on Olympus, ready for combat, but in his slaying ofthe dragon that guards Delphi, Pytho, which corresponds to Zeus’slaying of Typhoeus.

The Homeric Hymns to Athena and Apollo present each god as anally of Zeus who might have been, or might have remained a rival.Both Hymns partake of the theme of intergenerational conflict and inboth, Zeus’ position as king of gods and men is strengthened onceeach god is incorporated into his regime. Given the scholarly opinionon how the major Hymns relate to the Theogony and to the politics ofOlympus,37 it is fascinating to find Hymn 28 functioning in a mannersimilar to the long Hymns, including theHymn to Apollo. In this shortbut nonetheless dramatic poem, Athena’s self-restraint, when sheremoves her armour, brings joy and relief to her enthroned father;in Hymn 3, though his capacity for rebellion may remain, Apollo

35 Cf. Felson (1994), 86–7, on Telemachus’ motivation for hanging the twelvemaidens.

36 Felson (1994), 72, 167–8 n. 22. Cf. Apollo’s ‘benign regression’ in Hymn 4 (toHermes), where he vies for a place in the cosmic hierarchy with his newborn brotherand their father, Zeus, good-humouredly mediates the quarrel. On their sibling rivalryand its resolution, see Harrell (1991), 307–18, and Vergados (2007).

37 See esp. Clay (1989), and in this volume (Ch. 11). For a contrary view regardingApoll., see Chappell in this volume (pp. 73–81).

278 Nancy Felson

channels his potentially subversive energy against female menaces toorder. The actions of each child of Zeus illustrate how rapport can beestablished and tension resolved in time between a potentially mena-cing (monstrous) offspring and a potentially hostile (dread) father. Ifthe rapprochement is successful, the offspring will never upset cosmicorder by unseating the father. Consequently, as Gaia wishes in theTheogony (128), the ‘seat’ (�� ) of the cosmos will remain, indeed,‘stable forever’ (I�çÆºb ÆN��).


13

The Earliest Phases in the Receptionof the Homeric Hymns

Gregory Nagy

INTRODUCTION

It has been argued that Hesiodic poetry, like Homeric poetry, con-tains references to four aspects of oral poetry: composition, perfor-mance, reception, and transmission.1 In the present project, I arguethat the poetry of the Homeric Hymns, as oral poetry, similarlycontains references to all four of these aspects. In making this argu-ment, I will concentrate on the reception of the Hymns, analysing theinternal evidence of references to the Hymns within the Hymnsthemselves along with the external evidence of early references wefind elsewhere, especially in the comments of Thucydides on theHomeric Hymn to Apollo. In the course of making my argument,I will refer to interconnecting arguments I have published elsewherein the pursuit of related projects.2

1 The argument is made in Nagy, ‘Hesiod and the Ancient Biographical Traditions’(2009a), hereafter abbreviated as HB, with reference to Most (2006), xix-xx, whomentions all four aspects. The argumentation of Most differs, however, from my own(see especially HB 273).

2 These projects are represented especially by twin books that I abbreviate asfollows: HC ¼ Homer the Classic (2008/9) and HPC ¼ Homer the Preclassic (2009/10). I refer also to these related books: BA ¼ Best of the Achaeans (1979; 2nd edn.1999), PH ¼ Pindar’s Homer (1990a), GM ¼ Greek Mythology and Poetics (1990b),PP ¼ Poetry as Performance (1996a), HQ ¼ Homeric Questions (1996b), PR ¼ Plato’sRhapsody and Homer’s Music (2002), HR ¼ Homeric Responses (2003), HTL ¼Homer’s Text and Language (2004c).

I use the term reception here not in the narrow sense that applies instudies of literature, where this term conventionally refers to what-ever happens after a given piece of literature is composed for trans-mission to the public. A broader sense of the term is needed if we aredealing with oral traditions.3

The phenomenon of reception in oral traditions is connected withthe transmission of any composition by way of performance. In anyoral tradition, the process of composition is linked to the process ofperformance, and any given composition can be recomposed eachtime it is performed. The performer who recomposes the compositionin performance may be the same performer who composed it earlier,or it may be a new performer, even a succession of new performers.The point is, such recomposition-in-performance is the essence oftransmission in oral traditions. And this kind of transmission is thekey to a broader understanding of reception.4

Unlike what happens in literature, where reception by the publichappens only after a piece of literature is transmitted, reception inoral traditions happens during as well as after transmission. That isbecause the process of composition in oral traditions allows forrecomposition on each new occasion of performance for a publicthat sees and hears the performer. In oral traditions, there is anorganic link between reception and performance, since no perform-ance can succeed without a successful reception by the public thatsees and hears the performer or performers.

The link between reception and performance affects the actualcontent of the composition performed before a given public. That isbecause the performance of a given composition can speak aboutitself. For example, the performance can say things about the contextof performance or even about the performer or performers. Whatis said, however, will be subject to change from performance toperformance, and such change can actually affect the content of thecomposition by way of recomposition-in-performance.5

3 The next three paragraphs derive from HB 282–3.4 For an example of concrete applications of theories about transmission, I cite my

work on the oral poetics of the ancient Greek symposium Nagy (2004b).5 PP 207–25.

Reception of the Homeric Hymns 281

CONTEXTS OF RECEPTION: THE CASE OF THEHOMERIC HYMN TO APOLLO

On the basis of external evidence about one of the Homeric Hymns,combined with the internal evidence of that Hymn and of otherHymns, we can say in general that the Homeric Hymns were per-formed at festivals. The primary external evidence comes from Thu-cydides, who quotes two passages from a composition that we knowas the Homeric Hymn to Apollo. In what follows, I give a translation ofthe text of those two passages, in its entirety, together with a transla-tion of the framing text of Thucydides (3. 104. 2–6):6

{3.104.2} [The island of] Rheneia is so close to Delos that Polycrates,tyrant of the people of [the island-state of] Samos, who had supremenaval power for a period of time and who had imperial rule [¼ arkhein]over the islands, including Rheneia, dedicated Rheneia, having capturedit, to the Delian Apollo by binding it to Delos with a chain. After thepurification [katharsis], the Athenians at that point made for the firsttime the quadrennial festival [pentetēris] known as the Delia. {3.104.3}And, even in the remote past, there had been at Delos a great comingtogether of Ionians and neighbouring islanders [nēsiōtai], and they werecelebrating [¼making theōria] along with their wives and children, justas the Ionians in our own times come together [¼ at Ephesus] for [thefestival of] the Ephesia; and a competition [agōn] was held there [¼ inDelos], both athletic [gumnikos] and musical [mousikos], and the citiesbrought [anagein] song-and-dance groups [khoroi]. {3.104.4} Homermakes it most clear that such was the case in the following verses [eposplural], which come from a prooemium [prooimion] having to do withApollo:

But when in Delos, Phoebus, more than anywhere else, you delight[terpesthai] in your heart [thumos],

there the Ionians, with chitons trailing, gatherwith their children and their wives, along the causeway [aguia],and there with boxing and dancing and songthey have you in mind and delight [terpein] you, whenever they set up a

competition [agōn].

6 As for the text of the medieval transmission of the Hymn, I argue in HPC I}26 (inthe footnotes) that most if not all of the textual differences between the medieval andthe Thucydidean transmission are a matter of authentic formulaic variations.

282 Gregory Nagy

{3.104.5} That there was also a competition [agōn] in the craft of music[mousikē (tekhnē)], in which the Ionians went to engage in competition[agōnizesthai], again is made clear by him [¼ Homer] in the followingverses, taken from the same prooemium [prooimion]. After making thesubject of his humnos the Delian song-and-dance group [khoros] ofwomen, he was drawing toward the completion [telos] of his words ofpraise [epainos], drawing toward the following verses [epos plural], inwhich he also makes mention [epi-mnēsthēnai] of himself:

165 But come now, may Apollo be gracious, along with Artemis;and you all also, hail [khairete] and take pleasure, all of you

[Maidens of Delos]. Keep me, even in the future,in your mind, whenever someone, out of the whole mass of

earthbound humanity,while here [in Delos], after arduous wandering, someone else,

asks this question:‘O Maidens, who is for you the most pleasurable of singers

170 that wanders here? In whom do you take the most delight[terpesthai]?’

Then you, all of you [Maidens of Delos], must very properlyrespond [hupokrinasthai], without naming names:7

‘It is a blind man, and he dwells in Chios, a rugged land.’

{3.104.6} So much for the evidence given by Homer concerning the factthat there was even in the remote past a great coming together andfestival [heortē] at Delos; later on, the islanders [nēsiōtai] and theAthenians continued to send song-and-dance groups [khoroi], alongwith sacrificial offerings [hiera], but various misfortunes evidentlycaused the discontinuation of the things concerning the competitions[agōnes] and most other things—that is, up to the time in question[¼ the time of the purification] when the Athenians set up the competi-tion [agōn], including chariot races [hippodromiai], which had nottaken place before then.

7 According to the Thucydidean transmission of the Hymn, we read here Iç�ø ,which I interpret to mean ‘without naming names’. The medieval transmission showsother variants besides Iç�ø , including Iç’ �ø� ‘about me’. In HC 2}27 n. 25, Iconnect the adverb Iç�ø to the adjective ¼çÅ� , which was understood to be asynonym of I��ıŁ� (as we see in the scholia to Arat. 1. 270. 2). This word I��ıŁ� isused in the sense of ‘without information’, as in Od. 3. 88 and 184. When the DelianMaidens are asked to respond to the question ‘who is the singer?’, they respondwithout naming names, that is, without giving information about the singer’s name.


As we can see from this account by Thucydides, the setting for theHomeric Hymn to Apollo was the festival of the Delia on the island ofDelos (3. 104. 2). This festival is viewed here in the historical contextof a major reorganization of the Delia, undertaken by the city-state ofAthens (3. 104. 6). At the time of that reorganization, which can bedated to the winter of 426 bce, this festival of theDelia was the centralevent that gave meaning to the Delian League, a political and culturalconfederation of Ionian city-states that eventually came under thecontrol of what we now know as the Athenian maritime empire.8

In the passage I just quoted, Thucydides explicitly compares theDelia, as a seasonally recurring Panionian festival celebrated at Delos,with the Ephesia, a seasonally recurring Panionian festival celebratedon the mainland of Asia Minor, in Ephesus (3. 104. 3). The Ephesia,as a festival, had been a rival of the Panionia, which had once been thepremier Panionian festival in Asia Minor.9 Clearly, the festival of theDelia rivalled those other two festivals as an expression of Ionianidentity, and this identity was ultimately re-defined by the city-stateof Athens in its claimed role as the metropolis or ‘mother city’ of allIonian city-states. Such a role was foundational for the genesis of theAthenian maritime empire.10

In the context of referring to the Panionian festival of the Delia ascelebrated on the island of Delos, Thucydides links this festival to thepolitics of empire. He is taking for granted the immediate politicalreality, which was the Athenian maritime empire, and he is high-lighting an earlier reality, which was an earlier maritime empire thattook shape in the sixth century bce under the rule of Polycrates ofSamos: as we have seen in the passage I quoted, Thucydides describesPolycrates as ‘tyrant of the people of [the island-state of] Samos, whohad supreme naval power for a period of time and who had imperialrule [¼ arkhein] over the islands’ (3. 104. 2).11 For Thucydides, themaritime empire of Polycrates in an earlier time was a predecessor ofthe Athenian maritime empire in the historian’s own time.12

8 Hornblower (1991), 527, Rhodes (1994), 258–9.9 HPC II}}237–8, 246–61. On the festival of the Panionia, I cite the foundational

work of Frame (2009), especially ch. 11.10 HPC I}}15–23, II}}19–23.11 —�ºıŒæ��Å › $Æ�ø� ��æÆ�� N�å��Æ �Ø�a åæ�� Æı�ØŒfiH ŒÆd �H� �� ¼ººø�

��ø� ¼æ�Æ .12 HPC II}}239–44, 252, 254.

284 Gregory Nagy

The wealth, power, and prestige of the Athenian empire are re-flected in the magnificent spectacle of the Delian festival as celebratedin the era described by Thucydides, after the reorganization of 426bce. Such magnificence, as we can infer even from the historian’sunderstated description, is given a full-blown treatment in the repor-tage of Plutarch in his Life of Nicias (3. 5–7). At a later point, we willconsider the relevance of Plutarch’s reportage. For now, however,I concentrate on the perspective of Thucydides as he proceeds toreconstruct earlier phases of the festival of the Delia on the basis ofthe two passages he quotes from the Homeric Hymn to Apollo.

In the Hymn as quoted by Thucydides, Delos is described as afestive centre where representatives of all Ionian cities converge ina grand assembly to validate their common origin by celebrating aPanionian festival (146–50). The historian is saying that this descrip-tion of the festival indicates a prototype of the Delia, to be contrastedwith the reorganized version of the festival, which as we have seen canbe dated to 426 bce. After that date, the festival of the Delia wascelebrated on a quadrennial basis.13

Thucydides says that there had been also an earlier Athenian kathar-sis (‘purification’) of Delos, and that it took place at the initiative ofPeisistratus of Athens (3. 104. 1). This earlier katharsis signals an earlierAthenian inauguration of the same festival of the Delia at Delos.14

This earlier moment in the history of the Delia goes back to somepoint within a period of time that lasted from 546 to 510 bce, whenPeisistratus and his sons were ‘tyrants’ of Athens. Thucydides viewsthis earlier moment as a precedent for the later moment, in 426 bce,when this festival was reorganized by Athens in the era of theAthenian democracy. In terms of what Thucydides is saying, therewas already a prototype of the Athenian maritime empire in the era ofthe Peisistratidai in the sixth century bce, and there was already a

13 As Thucydides says (3. 104. 2), ‘after the purification [katharsis], the Atheniansat that point made for the first time the quadrennial festival known as the Delia’ (ŒÆd�c� ��Åæ��Æ �� æH�� a �c� Œ�ŁÆæ�Ø� K��Å�Æ� �ƒ �ŁÅ�ÆE�Ø �a ˜�ºØÆ). Thewording prōton ‘for the first time’ here refers to the first time that this festival wascelebrated as a pentetēris or ‘five-year festival’ (by way of inclusive counting), that is,on a quadrennial basis (to restate by way of non-inclusive counting). Clearly, thewording does not refer to the first time that this festival was ever celebrated.

14 Besides Thucydides, Herodotus too (1. 64. 2) refers to this earlier katharsis(‘purification’), and he specifies that it was initiated by Peisistratus. See Hornblower(1991), 527.


version of the festival of the Delia that reflected the realities of thatearlier empire.15

In the era of the Peisistratidai, we find a major complication in thehistory of Delos and the Delia. As we can see from the account thatI have quoted from Thucydides, control of Delos and the Delia by thePeisistratidai of Athens must have shifted to Polycrates of Samos for abrief period in the late 520s bce, though it must have shifted back tothe Peisistratidai after the death of Polycrates in 522, which had led tothe collapse of his Panionian empire.16

This complication has left its mark in the Homeric Hymn to Apolloas we know it. As we can see from the mention of Polycrates in theaccount I have quoted from Thucydides, this Hymn was once per-formed in the period when control of Delos and the Delia shiftedtemporarily from the Peisistratidai to Polycrates.17 Thucydides signalsthe occasion of such a performance when he tells about the chaining ofthe island of Rheneia to the island of Delos by Polycrates. On thisoccasion, according to later sources, Polycrates organized an event thatresembled a combination of two festivals, the Delia and the Pythia, foran ad hoc celebration on the island of Delos.18 The Homeric Hymn toApollo, which offers praise for both the Delian and the Pythian aspectsof the god Apollo, fits the occasion. And this occasion has been dated to522 bce.19 Soon thereafter, in the same year, Polycrates was overthrownand killed by the Persians. Peisistratus had died earlier, in 528/7.

AN ANCIENT ATHENOCENTRICVIEW OF HOMER

The indications about a performance of the Homeric Hymn to Apolloat the festival of the Delia on the island of Delos in 522 bce need to be

15 Hornblower (1991), 520.16 For a sketch of the relative chronology, involving Naxos as well as Samos and

Athens, see Aloni (1989), 46–7, 54–5, 62–3, 122–3.17 HPC I}33.18 Zenobius of Athos 1. 62; Suda s.v. tauta kai Puthia kai Dēlia; for a fuller

collection of sources, see Aloni (1989), 35 n. 2, 83 n. 1.19 See Burkert (1979), 59–60, Janko (1982), 112–13; West (1999), 369–70 n. 17

argues for 523, but his dating criteria depend on whether or not we posit a perfectmatch between the datable events narrated by Herodotus and Thucydides. For moreon possible performance of Apoll. at Polycrates’ festival, see in this volume Chappell(pp. 71–3) and cf. Introduction (pp. 11–12).

286 Gregory Nagy

correlated with the idea of a prototypical performance of the Hymnby Homer himself. In order to understand this idea, we must firstconsider how Homer was viewed in Athens at the time when it wasruled by the Peisistratidai (546–510 bce). Within that period theserulers appropriated not only Delos and the Panionian festival of theDelia. They appropriated Homer as well. And they did so by makingthe Homeric Iliad and Odyssey the primary repertoire for perform-ance at the most important festival in that city, the Panathenaea.20

This appropriation did not happen all at once. Still, there was amost decisive moment that led to it, and we can see that moment inthe following description of an initiative taken by one of the sons ofPeisistratus, Hipparchus, when he ‘brought over’ (komizein) theHomeric Iliad and Odyssey to Athens for performance there at theseasonally recurring festival of the Panathenaea:

Hipparchus . . .who publicly enacted many and beautiful things tomanifest his expertise [sophia], especially by being the first to bringover [komizein] to this land [¼ Athens] the verses [epos plural] ofHomer, and he required the rhapsodes [rhapsōidoi] at the Panathenaeato go through [diienai] these verses in sequence [ephexēs], by relay [exhupolēpseōs], just as they [¼ the rhapsodes] do even nowadays. ([Pl.]Hipparch. 228b–c)

Partly on the basis of this passage, Martin West has argued thatHipparchus arranged for the first complete performance of theHomeric Iliad and Odyssey by rhapsōidoi ‘rhapsodes’ competing atthe festival of the Panathenaea that was celebrated in 522 bce (start-ing on 19 August).21 I in turn have argued that these epics had been‘brought over’ (komizein) to Athens from the island state of Chios,along with Homer’s notional descendants, called the Homēridai, whowere the official performers of these epics in Chios.22

20 HPC I}25.21 West (1999), 382.22 HPC I}}53–5, 141–56. The use of the word komizein in [Pl.] Hipparch. 228b, in

expressing the idea that Hipparchus ‘brought over’ to Athens the epē (¼ epos plural)‘verses’ of Homer, is parallel to the use of the same word komizein in expressing theidea that Hipparchus also ‘brought over’ to Athens the poet Anacreon—on a stateship, from the island of Samos (228c). These two initiatives are relevant to my overallargumentation about the shaping of the Panathenaea by the Peisistratidai (HQ 81n. 50). I find it relevant to highlight here a passing reference in Plato to a garland ofgold that the rhapsode Ion of Ephesus expects to win in competition for first prize in


In this light, we can see more clearly the significance of the fact thatthe speaker in the Homeric Hymn to Apollo says explicitly that hishome is the island of Chios (171–3). Such a Chiote signature on thepart of the speaker was for Athenians a sign of their claim tothe ownership of Homer. When Thucydides recognizes the speakerof the Hymn to Apollo as Homer (3. 104. 4–6), he is expressing anAthenocentric point of view by equating Homer, poet of the Iliad andOdyssey as performed at the Panathenaea in Athens, with Homer ofChios, the notional ancestor of the Homēridai, who is envisioned asperforming the Homeric Hymn to Apollo at the Delia in Delos.23

From an Athenocentric point of view, then, the Hymn to Apolloand the Homeric Iliad and Odyssey are all composed by Homer,ancestor of the Homēridai. And all this poetry is composed in ametre known as the dactylic hexameter, which as we will see is theprimary medium of performance for rhapsodes in general and for theHomēridai in particular.

AN ANCIENT POST-ATHENOCENTRICVIEW OF HOMER

From a post-Athenocentric point of view, by contrast, the man fromChios who speaks in the Hymn to Apollo is not Homer but a rhapsodeby the name of Cynaethus, who may or may not be one of theHomēridai. This Cynaethus may be speaking in Homeric hexametersas if he were Homer, but he is not Homer. A passage that repre-sents this alternative point of view comes from the scholia for Pin-dar’s Nemean 2, and I offer here my translation:

Homēridai was the name given in ancient times to those who weredescended from the lineage of Homer and who also sang his poetry[poiēsis] in succession [ek diadokhēs]. In later times, (it was the name

rhapsodic performance of Homer at the feast of the Panathenaea: Ion is quoted assaying that this garland will be awarded to him by the Homēridai (Ion 530d). Thisreference to the Homēridai shows that the Athenians in the late fifth century (which isthe dramatic time of the Ion) recognized the Homēridai of Chios as the officialregulators of rhapsodic competitions in performing the Homeric Iliad and Odysseyat the Panathenaea in Athens. See HC 3}36.

23 HPC I}}138–40.

288 Gregory Nagy

given also to) rhapsodes [rhapsōidoi] who could no longer trace theirlineage back to Homer. Of these, Cynaethus and his association becamevery prominent. It is said that they are the ones who made many of theverses [epos plural] of Homer and inserted [en-ballein] them into his [¼Homer’s] poetry [poiēsis]. Cynaethus was a Chiote by lineage, and, ofthe poetic creations [poiēmata] of Homer that are ascribed to him[epigraphein] as his [¼ Homer’s], it was he [¼ Cynaethus] who wrote[graphein] the humnos to Apollo and attributed it to him [¼ Homer].24

And this Cynaethus was the first to perform rhapsodically [rhapsōideîn]in Syracuse the verses [epos plural] of Homer, in the sixty-ninth Olym-piad [¼ 504–501 bce], as Hippostratus says. (Schol. Pi. N. 2, 1c, III 29,9–18 Drachmann [¼ FGrH 568 F 5])

It can be argued that the ultimate source for this compressed andelliptic account transmitted in the scholia for Pindar is Aristarchus ofSamothrace, head of the Library of Alexandria in the second centurybce.25 As for the dating of Cynaethus by Hippostratus, the time-frame of 504–501 bce for his rhapsodic performance in Syracuseis comparable with the date proposed for the performance of theHomeric Hymn to Apollo on the occasion arranged by Polycrates ofSamos: as we have seen, that date could be 522 bce. So the performerfor that occasion could have been Cynaethus himself.

The scholia for Pindar have more to say about this Cynaethus, andagain I give a translation:

Philochorus [FGH 328 F 212] says that they [¼ rhapsōidoi] were calledthat [¼ rhapsōidoi] on the basis of the idea of composing, that is,stitching together, the song. Proof for this comes from Hesiod, whosays [¼ Hes. fr. 357 M–W]:

InDelos, back then at the very beginning, I andHomer, singers [aoidoi],sang- and-danced [melpein],26 stitching together [rhaptein] a song innew humnoi, making Phoebus Apollo the subject of our song, the onewith the golden weapon, the one born of Leto.

24 Martin (2000), 419 n. 58 suggests that the phrasing here could mean instead:‘and dedicated it to him [¼ Apollo]’. See also Collins (2004), 184.

25 See HTL 28–9 n. 14, where I argue that Aristarchus is the basic source for thestatement up to the portion mentioning the testimony of Hippostratus concerning thedate of a rhapsodic performance by Cynaethus in Syracuse.

26 The verb melpein / melpesthai and the noun molpē convey the combination ofsinging and dancing: PH 12}29 n. 62 (¼ p. 350) and n. 64 (¼ p. 351).


Nicocles [FGrH 376 F 8] says that Hesiod was the first to performrhapsodically [rhapsōideîn]. The investigations of Menaechmusindicate that rhapsodes [rhapsōidoi] were called verse singers [stikhōi-doi] because verses [stikhoi] were called staffs [rhabdoi] by some people.Here is another version: the Homēridai were in former times thedescendants of Homer, but then, in later times, they were a groupcomprised of Cynaethus and his associates, who were called rhabdōidoi[staff-singers]. For these [¼ Cynaethus and his associates] are the oneswho used to bring back to memory and to perform the poetry [poiēsis]of Homer, which had been scattered. But they mistreated [lumai-nesthai] it [¼ the poetry]. And they [¼ the Homēridai] always startedwith a prooemium [prooimion], making mostly Zeus their point ofdeparture and occasionally the Muses. (Sch., Pi. N. 2. 1d–e, III 31,7–20 Dr)

In the first part of the relevant Pindar scholia, as quoted earlier, itwas claimed that Cynaethus and his associates ‘interpolated’ (en-ballein) additional hexameter verses and ascribed to Homer variousother compositions. Supposedly, rhapsodes like Cynaethus illegiti-mately interpolated additional verses to augment the original versesof Homer.27

In the second part of these same scholia, as quoted just now,it is claimed that Cynaethus and his group ‘mistreated’ the body ofHomeric poetry. As we can see from the claims in the first part, thisalleged mistreatment involved the adding of hexameter verses thatwere not originally Homeric.

To test the supposition that Cynaethus added verses to Homer’sown verses, let us consider the structure of the Homeric Hymn toApollo as we have it. This Hymn appears, at least on the surface, to bea combination of two originally separate hymns,28 and so it seemsreasonable to understand the Pindaric scholia to mean that Cy-naethus did add verses to an earlier hymn composed by Homer. Interms of such an understanding, the verses supposedly added byCynaethus could be described as Hesiodic rather than Homeric.Here is why. These verses constitute the part of the Homeric Hymnto Apollo that celebrates the god Apollo as he was worshipped atDelphi. In other words, the referent of these verses was the Pythian

27 HPC I}160.28 For this view, see Chappell in this volume (Ch. 4) and cf. Introduction

(pp. 11–12).

290 Gregory Nagy

Apollo, not the Delian Apollo who was worshipped at Delos. And,as Richard Martin has shown, the verses of the Homeric Hymn toApollo celebrating the Pythian Apollo are distinctly Hesiodic in style,whereas the verses celebrating the Delian Apollo are distinctlyHomeric.29 By the term ‘Hesiodic’ he means the style that is char-acteristic of the Theogony and Works and Days; by ‘Homeric’ hemeans the style that is characteristic of the Iliad and Odyssey.30

If Cynaethus himself really performed theHomeric Hymn to Apolloat the festival of the Delia at Delos in 522 bce, then it follows that thiswould-be descendant of Homer could have conflated a Homerichymn to Apollo with a rival Hesiodic hymn, treating the Hesiodicversion as an aspect of an overall Homeric tradition.31

In terms of such an ancient post-Athenocentric point of view, then,the real Homer would have performed a hymn to Apollo that wasmore simple than the augmented Hymn as performed by Cynaethusof Chios, and the supposedly original Homeric hymn would havecelebrated only the Delian Apollo at Delos, not the Pythian Apollo atDelphi.

OTHER ANCIENT VIEWS OF HOMER

Such an ancient view ofHomer must have been based on the evidenceof an ancient text of the Homeric Hymn to Apollo that featured boththe Delian and the Pythian portions as they have survived to this day.To be contrasted are other ancient views that must have been basedsimply on the hearing of a performance, not on the reading of a text.

For audiences who heard a rhapsode perform the Hymn on a givenoccasion at the festival of the Delia, what would matter is not theidentity of the rhapsode but the identity of the poet who was being re-enacted by the rhapsode. Similarly, for audiences who heard rhap-sodes perform the Homeric Iliad or Odyssey on a given occasion at

29 Martin (2000).30 HPC I}174. See also Petrovic (2011a and b forthcoming).31 HPC I}175. As for why Apoll. survives as aHomeric and not as aHesiodic hymn,

I offer an explanation in HB 302. For further discussion of the Delian and the Delphicparts of this Hymn, see Chappell (Ch. 4), with alternative explanations and relevantbibliography.


the festival of the Panathenaea, what would matter is the simple factthat the rhapsodes re-enact Homer.

Here and hereafter, I equate the concept of ‘re-enactment’ with theconcept conveyed by the Greek word mimēsis, which I hereafter writesimply as mimesis. When the Homeric rhapsode performs the wordsthat say ‘tell me, Muses!’ (Il. 2. 484) or ‘tell me, Muse!’ (Od. 1. 1), this‘I’ is not just a representation ofHomer: it isHomer.32 In other words,the rhapsode is re-enacting Homer by performing Homer, and he isHomer so long as the mimesis stays in effect, so long as the perform-ance lasts. From the standpoint of the mimesis, the rhapsode is arecomposed performer: he becomes recomposed into Homer everytime he performs Homer.33

From this point of view, the rhapsode who adds verses in the act ofperformance does not necessarily disqualify himself as a re-enactor ofHomer. The adding of verses is an act of interpolation only in termsof a text, not in terms of a performance.

EVIDENCE FROM LIFE OF HOMER NARRATIVES

What I just said about the adding of verses is a theme that is actuallyattested in a text that belongs to a set of texts known as the Life ofHomer narratives.34 Before I show the attestation, however, I offer ageneral formulation about the significance of such narratives:35

The Life of Homer traditions represent the reception of Homeric poetryby narrating a series of events featuring purportedly ‘live’ performancesby Homer himself. In the narratives of the Lives, Homeric compositionis consistently being situated in contexts of Homeric performance. Ineffect, the Lives explore the shaping power of positive and even negativeresponses by the audiences of Homeric poetry in ad hoc situations ofperformance. To put it another way, the narrative strategy of the Lives isa staging of Homer’s reception.

My describing the Life of Homer traditions as a staging convergeswith my aim to show that the narratives of these Lives are myths, not

32 PP 61.33 Again, PP 61. See also Papadopoulou (2004) and Petrovic (2011b forthcoming).34 For background on these narratives, see HPC I}}55–136 (also Nagy 2004a).35 HPC I}}57–8.

292 Gregory Nagy

historical facts, about Homer. To say that we are dealing with myths,however, is not at all to say that there is no history to be learned fromthe Lives. Even though the various Homers of the various Lives areevidently mythical constructs, the actual constructing of myths aboutHomer can be seen as historical fact.

And the point I have just made about Life of Homer narratives appliesalso to Life of Hesiod narratives.36

That said, I am ready to show the attestation I have found in theLife of Homer narratives where even the figure of Homer himselfis viewed in the act of adding verses to his own compositions.It happens in the Herodotean Life of Homer, which is composed inthe Ionic dialect and which reflects a political and cultural perspectivethat is favourable to the island state of Chios.37 In this narrative,we catch the figure of Homer himself in the act of ‘interpolating’(en-poieîn) by adding verses glorifying Athens while composing theIliad and Odyssey in Chios (379–99).38 Homer eventually leavesChios and sails off for Athens, expecting to perform there his aug-mented version of the Iliad and Odyssey after a stopover in Samos(483–4); when he sets sail from Samos, however, Homer makes themistake of making another stopover, on the island of Ios, and he diesthere before he can reach his final destination, which would have beenAthens (484–516).39

As we see from this example taken from the Life of Homer trad-itions, Homer can be viewed as a prototypical performer whosecompositions come to life in his performances.

Unlike the Herodotean Life of Homer, where Homer composes forperformance in Athens though he never reaches Athens, another Lifeof Homer narrative shows another outcome. This narrative, known asthe Contest of Homer and Hesiod (or Certamen), is derived from thework of Alcidamas, an Athenian intellectual who flourished inthe fourth century bce. According to the narrative of the Contest,

36 HB 278–80.37 I here and below cite the Herodotean Life of Homer and the Contest of Homer

and Hesiod from Allen.38 HPC I}}94–8. At I}153, I offer an explanation for a complication that we see here

in the Herodotean Life of Homer: while this narrative accepts the idea that the Iliadand Odyssey as composed in Chios by Homer were later performed at the Panathe-naea in Athens, it differs from other such narratives by undermining the genealogy ofthe Homēridai of Chios.

39 Commentary in HPC I}}91–9, 141, 144, 154.


the figure of Homer does in fact get to perform in Athens (276–85),and then he goes on to perform in other major cities of the Helladicmainland, especially Corinth (286–7) and Argos (288–314). Homer’slong-awaited songmaking tour of the Helladic mainland, which failedto take place in the Herodotean Life of Homer, is realized in theContest of Homer and Hesiod.40

HOMER AS PERFORMER OF THE HYMN TOAPOLLO IN A LIFE OF HOMER NARRATIVE

Just as Thucydides attributes the Homeric Hymn to Apollo directly toHomer, so too does the Athenocentric narrative of the Contest ofHomer and Hesiod:

After he [¼ Homer] stayed a while in the city [of Argos], he sailed overto Delos to the festival [panēguris] there. And, standing on the Altar ofHorn he speaks the humnos to Apollo, the beginning of which is:

I will keep in mind and not leave out of mind Apollo, who makesthings work from afar. [This verse matches the first verse of Apoll.]

Then, after the humnos was spoken, the Ionians made him [¼Homer] their common citizen [koinos politēs]. And the people ofDelos, writing down his verses [epos plural] on a white tablet [leukōma],dedicated them in the sacred space of Artemis. Then, after the festival[panēguris] was declared to be finished, the Poet [poiētēs] sailed to Ios tomeet Kreophylos. (Contest 315–22)

The myth about Homer in this narrative amounts to an aetiology ofthe festival of the Delia.41 Also, this myth about Homer in Delosmotivates the institutional reality of Homeric reception, just as surelyas it motivates the institutional reality of the festival that definesHomer as the model performer at that festival. And this receptiondepends on the basic idea that Homer comes to life in perform-ance.42

40 Commentary in HPC E}53.41 HPC I}}124–36. By ‘aetiology’, I mean a myth that motivates an institutional

reality, especially a ritual. See BA 16}2n2 (¼ p. 279).42 HPC I}130.

294 Gregory Nagy

HOMER AND HESIOD AS PERFORMERS OFTHE HYMN TO APOLLO IN ANOTHER

LIFE OF HOMER NARRATIVE

In the myth we have just seen about the performance of Homer at thefestival of the Delia, there is no mention of anyone else performing atthat festival. So we do not expect to see anyone other than Homerperforming at the Delia. But there exists an alternative myth thatshows another performer performing along with Homer at that samefestival, and that other performer turns out to be Hesiod. This alter-native myth is embedded in the verses of the Hesiodic fragment thatI quoted earlier (fr. 357 M–W).43 In terms of the myth embeddedin these verses, Homer and Hesiod collaborated with each otherby producing two hymns, that is, two humnoi—one performed byHomer and the other performed by Hesiod.

MYTHS ABOUT CONTESTS OF HOMERAND HESIOD

For the moment, I will concentrate on the idea of these two perform-ances, not on the actual mode of performing. In the myth that we arenow considering, the reference to performances by both Homer andHesiod at Delos can be viewed as part of a vast complex of variantmyths centring on the myth of a competition between the figures ofHomer and Hesiod in a contest that will determine which of the twois the better poet. In the example at hand, the idea of an actualcompetition is only implicit, since the myth confines itself to showingthe two poets in the act of producing two Hymns together. As we areabout to see, however, the idea of competition is explicit in otherexamples.

I begin with two references to such a competition in the overallnarrative of the Contest of Homer and Hesiod. In the first of these tworeferences, there is a passing mention of a contest of Homer andHesiod at Aulis in Boeotia (54–5). In the second reference, we find the

43 See above pp. 289–90.


beginning of an extended narrative about another contest betweenthese two figures, which takes place at Chalcis in Euboea (68) on theoccasion of funeral games commemorating the death of a king namedAmphidamas (63–4). In the course of this extended narrative, whichtakes up a huge portion of the overall narrative of the Contest ofHomer and Hesiod, Homer is ultimately defeated by Hesiod in thepoetic competition held at Chalcis, though the narrative attributes thevictory of Hesiod to the favouritism of a contemporary king whodeclares Hesiod as the winner and who thus overrules the favourablereception that Homer is accorded by the populace attending thecontest (62–214).

It has been argued that the myth of the Contest of Homer andHesiod as mediated by Alcidamas was his own invention, based on apassing reference in Hesiodic poetry to a poetic contest that Hesiodhad won in Chalcis (Hes. Op. 650–9).44 But the argument that thismyth was invented by Alcidamas in the fourth century bce is unten-able. Even in terms of this argument, it has to be admitted that theverses quoted in the text of the Contest as the verses of Homer andHesiod themselves ‘were already current in the fifth century’.45 Andthere is ample evidence to show that the myths embedded in the textof the Contest must be far earlier than even the fifth century bce.46

As I will now argue, one version of the myth of a contest betweenHomer and Hesiod is embedded even in the Hesiodic Works andDays. And, as I will argue thereafter, another version is embedded inthe Homeric Hymn to Apollo.

AN EMBEDDED MYTH ABOUT THE CONTESTOF HOMER AND HESIOD IN THE HESIODIC

WORKS AND DAYS

In the Hesiodic Works and Days, we see such a trace in a variant versereported by the Hesiodic scholia. In this variant verse, Hesiod isdeclaring that his adversary in the poetic contest that he won inChalcis was Homer himself:

44 WL 299.45 WL 299.46 HB 300.

296 Gregory Nagy

defeating god-like Homer with [my] humnos, at Chalcis (Sch. Hes. Op.657a Pertusi)47

instead ofwinning with [my] humnos, [I say that I] carried away [as a prize] atripod with handles on it. (Hes. Op. 657)48

The variant verse mentioning Homer is also attested in the Contest ofHomer and Hesiod (213–14), where it is part of an epigram ascribedto Hesiod, who reputedly composed it in celebration of his victoryover Homer.49

AN EMBEDDED MYTH ABOUT THE CONTESTOF HOMER AND HESIOD IN THE HOMERIC

HYMN TO APOLLO

In the Homeric Hymn to Apollo, there is no explicit reference to thenames of either Homer orHesiod, and there is no explicit reference toa competition between the two. But there is an implicit competition inthe form of a competing set of two parts of one Hymn. One of thesetwo parts of the Hymn, as we have seen, is addressed to Apollo asworshipped at the festival of the Delia at Delos and the other isaddressed to Apollo as worshipped at the festival of the Pythia atDelphi. The competition between these two parts of one Hymn, as wehave also seen, is all-pervasive, if Martin is right that theDelian part iscomposed in a Homeric style while the Pythian part is composed in aHesiodic style.50

47 o�øfi �ØŒ��Æ��’ K� �ÆºŒ��Ø Ł�E�� ῞ˇÅæ��.48 o�øfi �ØŒ��Æ��Æ çæ�Ø� �æ��’ T��Æ.49 All this is not to question whether the variant verse as attested in the medieval

manuscript tradition of Hes. Op. 657 is genuine to the Hesiodic tradition. This versementioning the tripod itself may be just as genuine as the verse mentioning Homer.And each one of these two alternative verses is composed in an epigrammatic stylethat characterizes the verses that follow (658–9) as also the verse that precedes (656).Though both of these two variant verses may be genuine, the one that directlymentions Homer must have been phased out in the course of the poem’s crystal-lization into a fixed text. See HB 304, with reference to further argumentation inGM 78.

50 Martin (2000). See also Petrovic (2011b forthcoming).


And the winner of this implicit competition between Homer andHesiod must be in this case Homer, not Hesiod, since the Delian partof the Homeric Hymn to Apollo predicts the superiority of Homer toall other poets every time he returns to the Delia. In response to thequestion addressed to the Maidens of Delos, ‘Who is the mostpleasurable of singers that wanders here, in whom do you (Maidens)take the most delight [terpesthai]?’ (169–70), the Delian Maidensmust declare that the greatest singer of them all is the blind manwho dwells in Chios (172). This blind man, ancestor of theHomēridaiof Chios, must be Homer himself, who keeps eternally returning tothe seasonally recurring festival of the Delia to ask the same eternalquestion that calls for the same eternal response.51

It remains to be seen how the identity of Homer depends on hisinteraction here with the Delian Maidens in the Homeric Hymn toApollo. But first, we must consider further the theme of Homer’seternal return at the festival of the Delia.

HOMER AND HESIOD AS RHAPSODES

When the Homeric Hymn to Apollo says that Homer the wandererreturns eternally to the Delia, what is really being said is that Homergets to be re-enacted year after year by wandering rhapsodes whocompete at each new seasonally recurring occasion of the festival. Theidea is that Homer gets to be re-enacted eternally by rhapsodes. AsI have already argued, the ‘I’ or ‘me’ in rhapsodic performances ofHomeric poetry is notionally Homer himself, not the rhapsode. AndI must now add that the same principle applies to Homer’s eternalcompetitor, Hesiod.

51 HC 2}}38–9. The ‘someone’ who asks theDelian Maidens the question that leadsto their response is described as a wanderer who comes to the Delia and will be hostedthere (167–8). In the version as transmitted in the medieval manuscript tradition, theuse of the wording tis ‘someone’ (167) leaves open the option of imagining the samesinger returning again and again, in an eternal loop, with each seasonal recurrence ofthe festival of the Delia. As for the version as transmitted in the quotation made byThucydides (3. 104. 5), the variant wording tis . . . allos ‘someone else’ (verses 167–8)leaves open the same option: this seemingly ‘other’ person becomes the same personas the speaker—once the response of the Delian Maidens to that ‘other’ person isactually quoted by the speaker.

298 Gregory Nagy

Not only are Homer and Hesiod re-enacted by rhapsodes. They areimagined as performing like rhapsodes in their own right. The variantmyths about a primal contest of Homer and Hesiod stem from theidea that these two poets performed like rhapsodes, and this idea inturn stems from variant rhapsodic traditions in the transmission ofHomeric and Hesiodic poetry. The existence of such rhapsodic trad-itions is acknowledged by Plato, as when Socrates imagines Homerand Hesiod themselves in the act of ‘performing in the manner ofrhapsodes’ (rhapsōideîn) as they ‘go wandering around’ (perier-khesthai) from city to city (Pl. R. 10. 600d–e).52

The wording we find here is strikingly similar to the wording wefind in the Contest of Homer and Hesiod, where Homer is explicitlypictured as a rhapsode who wanders from city to city: ��Ø��Æ��Æ ªaæ�e� )Ææª��Å� �OÅæ�� æØæå��ŁÆØ ŒÆ�a ��ºØ� ÞÆłøfi ��F��Æ, ‘havingmade [poieîn] the [comic poem] Margites [in the city of Colophon inAsia Minor], Homer went wandering around [perierkhesthai] fromcity to city, performing in the manner of rhapsodes [rhapsōideîn]’(55–6).53 After his poetic competition with Hesiod at Chalcis(72–211), Homer is said to continue his life as a wandering rhapsode,and the wording echoes the earlier wording we saw just a momentago: ‘as he went wandering around [perierkhesthai], he was telling hispoetic creations [poiēmata]’ (255).54

RHAPSODIC COMPETITION ANDCOLLABORATION

I focus on the symmetrical image, as verbalized by Plato’s Socrates, ofHomer andHesiod as itinerant professional rhapsōidoi ‘rhapsodes’whoperform at urban festivals (Pl. R. 10. 600d–e). A prime example of suchperformances is the tradition of rhapsodic competitions in performingHomeric and other epic poetry at the festival of the Panathenaea in thecity of Athens. And, as we can see most clearly from the survivingevidence about that tradition, rhapsodic performances at such a festivalwere not only competitive: they were also collaborative.

52 ÞÆłøfi ��E� . . . ��æØØ��Æ .53 I back up this interpretation of the text in HPC I}}103–7.54 ��æØ�æå�� º�ª� �a ��Ø�Æ�Æ.


At the festival of the Panathenaea, already in the era of the Peisis-tratidai, competing rhapsodes were required to take turns in perform-ing by relay the Homeric Iliad and Odyssey, as we saw in the passage Iquoted earlier from the Platonic Hipparchus (228b–c). In such rhap-sodic performance by relay, collaboration was fused with competi-tion, since the rhapsodes competing at the Panathenaea would beexpected to collaborate with each other in the process of performing,by relay, successive parts of integral compositions like the HomericIliad and Odyssey.55

The regulating of rhapsodes in their relay performances, as we seeit at work in the description we read in the Platonic Hipparchus, didnot originate at the festival of the Panathenaea at Athens, even ifHipparchus can be credited with the initiative of bringing overto Athens, for the Panathenaea of August 522 bce, the Homēridaiof Chios for the purpose of regulating the rhapsodic performancesof the Homeric Iliad and Odyssey at that Athenian festival. As thework of Douglas Frame has shown, such regulation of rhapsodicperformance by relay, as represented by the Homēridai, was alreadyoperational in the late eighth and early seventh centuries at thefestival of the Panionia held at the Panionion of the Ionian Dodeca-polis in Asia Minor.56 Still, since the practice of regulating the relayperformances of rhapsodes can be seen most clearly in the case of thePanathenaea in Athens, I refer to this practice as the PanathenaicRegulation.57

The model I have developed so far, where rhapsodes collaborate aswell as compete in the process of performing, by relay, successiveparts of integral compositions like theHomeric Iliad and Odyssey, canbe used to explain the unity of these epics as they evolved over time.58

And now it can also be used to explain the unity of the HomericHymn to Apollo as a fully integrated composite of two hymns. The

55 PR 36–9. For a comparative perspective on the concept of competition-in-collaboration, see PP 18.

56 Frame (2009), ch. 11. Also, I agree with Frame’s argument (583–4) that theHomeric Iliad and Odyssey, as epic traditions, ‘reached Athens almost immediatelyafter they took root on Chios, and that even earlier they may have begun to be knownin Athens directly from the Panionia’. In terms of this argument, Hipparchus can becredited only with the actual authorization of the Homēridai as regulators of epicperformances by rhapsodes at the Panathenaea in Athens.

57 More on the Panathenaic Regulation in HC 2}297 and HPC I}}37–43, 47, 54,144, 166–8, 229–31; II}}109–10; E}69.

58 PR 42–7.

300 Gregory Nagy

prototypical event of performing such a composite of a Delianhymn composed by Homer and of a Pythian hymn composed byHesiod is dramatized in the Hesiodic fragment that I quoted earlier(fr. 357 M–W). The word rhaptein ‘stitch together’ as used in theverses in this fragment is a reference to the performances of rhap-sodes, since the word rhapsōidos means, etymologically, ‘he whostitches together [rhaptein] songs [aoidai]’.59 As for the word melpein‘sing-and-dance’ as also used here, I need to postpone my analysisuntil we reach a later point in the argumentation.

As we can see from these verses, a rhapsodic performance of aHymn to Apollo at the Delia in Delos was not necessarily envisionedas one performance by one rhapsode. There could be successiverhapsodic performances by successive rhapsodes, even by successiveHomēridai, who could be re-enacting Homer and Hesiod in succes-sion. And such succession would require collaboration as well ascompetition.

In making this point, I draw on the relevant wording in the scholiafor Pindar’s Nemean 2 (1c) translated above, focusing on the expres-sion ek diadokhēs with reference to the relay singing of the Homēridaiof Chios: ‘and they [¼ the Homēridai] also sang his [¼ Homer’s]poetry [poiēsis] by taking turns [ek diadokhēs]’.60 This expression ekdiadokhēs is conventionally translated as ‘in succession’.61 But such atranslation leaves it open whether the ‘succession’ is from ancestor todescendant or from one participant to another while taking turns.I have collected examples of this expression where we see both senses,succession by heritage and succession by relay.62

59 PP 61–9. See the fragment above p. 289.60 �Q ŒÆd �c� ��Å�Ø� ÆP��F KŒ �ØÆ��åB fi q��. See the scholia above pp. 288–9.61 So West (1999), 368.62 HPC I}}163–4. We find an example of the first sense in the scholia for Pi. O. 6

(158a, I 191 Dr), where Hieron is said to have inherited a priesthood ek diadokhēs ‘insuccession’ from one of his ancestors. We find an example of the second sense in thescholia for Pi. P. 12 (25 Semitelos p. 121), where the three Graiai are said to share oneeye and one tooth, using them ek diadokhēs ‘in relay’, that is, by taking turns. There isanother example in Arist. Ph. 5. 227a 28–9: ŒÆd �x�� ºÆ�a <�> KŒ �ØÆ��åB ç�æaKå��Å, �ı��åc �’ �h (‘and just as the torch race by relay [ek diadokhēs] is locomotionthat is consecutive but not continuous . . . ’). Moreover, the expression ek diadokhēscan mean ‘taking turns’ in contexts where it is used together with allēlois ‘with eachother’. A case in point is a passage from Arist. fr. 347. 15 Rose (¼ Ael. VH 1. 15) wherehe describes how a mother bird and a father bird warm the eggs in their nest by takingturns (ek diadokhēs) with each other (allēlois). I end this list with a particularlystriking example from the scholia D for Iliad I (604), where ek diadokhēs refers to


On the basis of the examples I have collected, I offer this formula-tion about rhapsodic traditions of relay performance, applying theterms synchronic and diachronic as pioneered by Saussure:63

In order to represent this diachronic succession of generations, therehas to be a synchronic grouping of these generations as a corporation ofpractitioners. That corporation is named as the Homēridai, the ‘des-cendants of Homer’. In the act of performance, the descendants are allsynchronized as one corporation who incorporate the ancestor bytaking turns in re-enacting him.64

In the scholia for Pindar’s Nemean 2 (1c), the idea that members ofthis corporation of the Homēridai sing the poetry of Homer in relay,taking turns, is then followed up by the idea of generational succession.But the legitimacy of this succession is questioned. From an Aristarch-ean perspective, the successors of Homer are not genuinely doing whattheir predecessor had done, and so they are not genuine. So they areillegitimate. This supposedly illegitimate corporation described as ‘Cy-naethus and his association’65 engage in various poetic activities like‘interpolating’ (en-ballein) additional verses to augment the supposedlygenuine verses of Homer or ‘ascribing’ (epi-graphein) to Homer ahumnos that they composed on their own. Even though the statementas recorded in the scholia for Pindar rejects the poetic activities of‘Cynaethus and his associates’ as illegitimate, typical of those who aremore recent than the genuine Homer, it nevertheless sets up a parallelbetween them and the Homēridai—as associations of performers. Byimplication, just as the Homēridai sing Homer as a group, taking turns,so too ‘Cynaethus and his associates’ sing Homer as a group, taking

the relay singing of the Muses: ŒÆd Æy�ÆØ ��ººø�� ŒØŁÆæ�Ç�� KŒ �ØÆ��åB �Ææaæ� fi q�� (‘and they, while Apollo was playing the kithara, were singing in relay[para meros], by taking turns [ek diadokhēs]’).

63 For my use of the words ‘synchronic’ and ‘diachronic’, see HR 1, with referenceto De Saussure (1916), 117.

64 See HPC I}164. There I go on to argue that the same can be said about, say, themother bird and the father bird that feed their young in relay: that principle of relay isthe model for the idea that each new generation has to follow the practice of theprevious generation in feeding the young. Or again, the principle of the relay in theAthenian torch race is a ritualized way of expressing the continuity of the tradition oftorch racing in and of itself. We may compare the idea of the Asklēpiadai, notionaldescendants of the prototypical physician Asklēpios, who are figured as a corporationof physicians who practise medicine by continuing the practice of their ancestor. Soalso the Homēridai, notional descendants of Homēros, are figured as a corporation ofsingers who continue the practice of singing Homer. For them to sing in relay is asynchronic ritualization of the diachronic continuity.

65 �ƒ ��æd ˚��ÆØŁ��.

302 Gregory Nagy

turns. Whatever it is that ‘Cynaethus and his associates’ may do, thestatement is explicit about what is done by the Homēridai: they sing thepoetry of Homer ‘by taking turns’ (ek diadokhēs).66

This formulation applies primarily to the model of the Homeric Iliadand Odyssey as relay poetry performed by the Homēridai, but it canapply secondarily to the Homeric Hymn to Apollo, in the sense thatHomer and Hesiod can be imagined as performing by relay. And, inthis kind of relay performance, Homer and Hesiod would have to becollaborating with each other while they are competing with each other.

There are traces of this theme of rhapsodic relay even in thenarrative of the Contest of Homer and Hesiod, the setting of whichis Chalcis. In this case, however, the balance between competition andcollaboration is markedly uneven, and the narrative makes it evidentthat the unevenness stems in part from the faulty and even unfairregulation of this contest at Chalcis by the king who presides. In thecourse of the unfairly regulated competition between Homer andHesiod, Hesiod has the advantage of testing Homer unilaterally bychallenging his rival to perform a variety of feats in poetic improvisa-tion (72–204).67 Still, even though the terms of this particular

66 HPC I}165.67 At one point in the narrative of the Contest, after Homer has already countered

successfully the initial poetic challenges thrown at him by Hesiod, he is challenged byHesiod with what is meant to be an impossible poetic task (Contest 94–101): )�F�’¼ª� �Ø �� ’ K��Æ �� ’ K��Æ �æ� �’ K��Æ j �H� b� Å�b� ¼�Ø��, �f �’ ¼ººÅ �B�ÆØ I�Ø�B (‘Come Muse, sing for me things that are, and that will be, and thathave been once upon a time, j yes sing them, and, of all of these things, sing nothing,but you must have in mind the rest of the song’); Homer responds, �P� ��’ Içd˜Øe ��øfi ŒÆ�Æå�� ¥��Ø j –æÆ�Æ �ı��æ�ł�ı�Ø� Kæ�Ç�� æd ��ŒÅ (‘Neveraround a tomb of Zeus will horses with clattering hooves j smash up the chariots theydraw as they compete for victory’). I interpret the expression ¼ººÅ . . . I�Ø�B in thewording of Hesiod’s challenge to Homer to mean ‘the rest of the song’, not ‘anothersong’. At a later point, I will examine further contexts of this and other relatedexpressions, showing that these expression refer to the poetics of rhapsodic relay,where one rhapsode takes over from the previous rhapsode in performing whatbecomes one song by way of the relay. In the present context, the challenge in theriddle is this: how can you sing ‘the rest of the song’ in competitive relay with theprevious singer if you cannot sing anything about the past, present, and future?Homer rises to the challenge by singing, in competitive relay, about an event that issurely exempt from the past, present, and future because it can never happen, and thatevent is the death of Zeus, to be commemorated by funeral games held in the god’shonour. If we chose the alternative interpretation, ‘another song’, then the point ofHesiod’s challenge would be blunted, because there would be no difficulty in respond-ing to the challenge of simply performing ‘another song’, which would not need to bea distinctly different song and could even be any other song at all.


competition are shown to be unfair, the formal characteristics ofHesiod’s quoted poetic challenges and of Homer’s quoted poeticresponses are clearly traditional.68 Such formal characteristics canalso be found in ancient reportage concerning rhapsodic competi-tions that actually took place in the historical period extending fromthe sixth into the fourth century bce.69

The parallelism of Homer and Hesiod as symmetrically matchedpoetic opponents is formalized only at the very end of the Contest ofHomer and Hesiod, when the storytelling reaches the final phase ofthe contest (176–210): ‘It is only when each poet is asked to recite thefinest piece of his poetry that their abilities can be weighed against oneanother.’70

This last part of the Contest is what comes closest to what we see inthe Homeric Hymn to Apollo, which combines the best of Homer as amaster of this form of poetry with the best ofHesiod as a rivalmaster.71

But Hesiod, unlike Homer, is a recessive figure in this Hymn, whichafter all represents the legacy of the Homēridai. In the end, the Hymnto Apollo survives as a Homeric hymn, not as a Hesiodic one. Thefigure of Hesiod could perhaps have defeated Homer at the festival ofthe Pythia in Delphi, just as he defeats him at Chalcis, but Hesiodcannot defeatHomer at the festival of theDelia inDelos, whereHomeras the ancestor of the Homēridai must have his own victory.72

68 Another version of this competitive exchange is provided by Plutarch’s Banquetof the Seven Sages (153f–4a), who says that it was another poet, Lesches of Lesbos, whochallenged Hesiod by way of quoting his own verses about saying nothing abouteverything (Lesches Little Iliad fr. 23 Allen), to which Hesiod responded by wayof quoting verses of his own about the impossible event of a funeral for Zeus(Wilamowitz-Moellendorff [1916b], p. 56. 2–3). In the version of the myth as wehave just seen it in the Contest, by contrast, the verses that match closely the versesquoted by Lesches are attributed to Hesiod (97–8), whereas the verses that matchclosely the verses quoted by Hesiod are attributed to Homer (100–1). See HB 303,where further analogies are analysed.

69 Collins (2004), 177–8, 185.70 WL 300.71 Such a combination of hymns, I should note, could have been performed solo as

well as by relay. Even a solo performance, however, would have accommodated a re-enactment of a relay from Homer to Hesiod, so long as the principle of performingby relay was in any case embedded in the conflated composition.

72 On a potential competition between Homer and Hesiod at Delphi, see HB300–3. A related topic, as I show there, is the appropriation of Hesiodic poetry bythe Peisistratidai of Athens.

304 Gregory Nagy

A DISTINCTION BETWEEN RHAPSODICAND CHORAL PERFORMANCE

The theme of a victory by Homer over Hesiod at the Delia in Delos islinked with the theme of Homer’s interaction with the Delian Maid-ens in the Hymn to Apollo. This interaction, as the text of the Hymnmakes clear, leads to Homer’s status as a Panhellenic celebrity: onceHomer leaves Delos behind after his performance there, he willwander from city to city throughout the Hellenic world (174–6),spreading the poetic kleos ‘fame’ of the Delian Maidens (174), andthis fame will be a reciprocation of the fame of his own poetry, whichis thus universalized (173). So this fame, stemming from Homer’sinteraction with the Delian Maidens at the Delia, is essential for hisvictory at that festival. But there is a distinction to be noted here:although Homer wins by way of performing as a rhapsode, the DelianMaidens with whom he interacts are performing in a mode that is notrhapsodic but choral. This distinction between rhapsodic and choralperformance turns out to be vital for an understanding of the Homer-ic Hymn to Apollo and of all the Homeric Hymns.73

I begin by elaborating on the point I just made, that the perform-ance of the Delian Maidens in the Homeric Hymn to Apollo ispictured as choral. Wording that indicates such performance is evi-dent from the references in the Hymn to their singing (161 o��I��ı�Ø�, 164 I�Ø��) and to their dancing to the rhythm of percussiveinstruments (162 Œæ��ÆºØÆ��).74

In this context of choral performance, I highlight the fact that theDelian Maidens in the Homeric Hymn to Apollo are described asmasters of mimesis or ‘re-enactment’ (verb mimeîsthai at verse163).75 This reference is saying something that is fundamentally

73 Before we proceed, I must note that I use the words ‘choral’ and ‘chorus’ in theancient sense of the Greek word khoros, which means ‘song-and-dance group’ or‘choral group of singers-dancers’, to be contrasted with the everyday meaning of themodern English word ‘chorus’, which fails to convey the aspect of dancing thatcomplemented the aspect of singing in the ancient Greek word khoros. See PH1}8 (¼ p. 20).

74 Peponi (2009). She notes the iconographical evidence showing Muses in the actof dancing to the rhythm of these instruments. The parallelism of the Muses and theDelian Maidens in this regard is relevant to what I have to say later about other suchparallelisms. On the reading Œæ��ÆºØÆ��, cf. in this volume Faulkner (p. 186 n. 46).

75 Commentary in HC 2}27 n. 22.


true about choral performance in general, which as we know from thesurviving textual evidence is highly mimetic. A most striking examplein general is the extant body of choral lyric poetry composed by suchmasters as Simonides and Pindar.76

What I just said about choral performance applies to rhapsodicperformance as well: this medium too is highly mimetic. A moststriking example is the interaction ofHomer with the Delian Maidensin the Homeric Hymn to Apollo. As we have seen, Homer re-enactsthe Maidens by quoting what they say, which is said not in their ownchoral medium but in the rhapsodic medium of the Hymn.77 So themedium of rhapsodic performance shows that it can make a mimesisof the medium of choral performance as exemplified by the DelianMaidens, who are the absolute masters of choral mimesis. This way,Homer demonstrates that he is the absolute master of rhapsodicmimesis.

As absolute masters of choral mimesis, the Delian Maidens in theHomeric Hymn to Apollo are parallel to the Muses in the HesiodicTheogony, who are pictured as an idealized chorus (7 å�æ�� , 63å�æ��) as they sing (10/43/65/67 Z��Æ� ƒ�E�ÆØ, 68 O�d ŒÆºBfi ) anddance (3–4 ��’ ±�Æº�E�Ø� j Oæå�F��ÆØ, 8 K��ææ��Æ�� b ��,70 KæÆ�e �b ��H� o�� F�� Oæ�æ�Ø). And, further, the figure ofHomer as an absolute master of rhapsodic mimesis is parallel to thefigure ofHesiod in the Theogony, who re-enacts the Muses by quotingwhat they say, but who says it not in the choral medium of the Musesbut in the rhapsodic medium of the Theogony (99–101 ÆP�aæ I�Ø�e j . . . Œº�EÆ �æ��æø� I�Łæ��ø� j ��Ø).78

76 In the case of Pindar’s victory odes, for example, the speaking ‘I’ can make amimesis of everyone and anyone who may be relevant to the act of praising the victor:the laudator, the laudandus, the kōmos, an optional khoros embedded within thekōmos, the ancestor, the athlete, the hero, and so on. Pindar’s odes also make mimesisof a wide variety of poetic functions, including what Bundy (1972) describes as a‘hymnal’ function (55–7). As Bundy shows, Pindar’s odes can even make mimesis of‘the actual process of thought in arriving at its goal’ (59 n. 59; see also 61–2).

77 HC 2}40.78 I use the term ‘rhapsodic’ here in a sense that resembles the usage of Meyer

(1933), though I distance myself from the idea that ‘rhapsodic’ is not ‘cultic’. This idealeads to the further idea of a dichotomy between ‘rhapsodic’ hymns and ‘cult’ hymns.Both these ideas shape the thinking of Bundy (1972), 73; also Bundy (1986), 46; morein Miller (1986), 1–6. I find it important that Miller (4) entertains the view that ‘therhapsodic and the “cult” hymn share a common origin’. See Furley in this volume(Ch. 10) on rhapsodic hymns as epicized elaborations of cult hymns.

306 Gregory Nagy

On the basis of these and other parallels between the DelianMaidens of the Hymn to Apollo and the Muses of the HesiodicTheogony, I have argued that the Maidens of the Hymn are in factlocal Muses of Delos.79 There is a symmetry between the poet of theHymn to Apollo and Hesiod as the poet of the Theogony. I focus onthe fact that the Delian Maidens are addressed with the hymnicsalutation khairete in the Homeric Hymn to Apollo (166) just as theOlympian Muses are addressed with the hymnic salutation khairetein the Theogony (104). This symmetry indicates that the poet of theHomeric Hymn is in effect addressing the local Muses of Delos, whoare divine in their own right.80

That said, I come back to my argument that the Delian Maidens inthe Homeric Hymn to Apollo make mimesis by way of choral perform-ance, while Homer makes mimesis by way of rhapsodic performance.On the basis of the parallelisms we have just seen between the HomericHymn to Apollo and the Hesiodic Theogony, I argue that the DelianMaidens, as Homer’s local Muses in the Hymn, are a divine model forthe performance ofhumnoi byHomer just as theMuses in theTheogonyare a divine model for the performance of humnoi by Hesiod.81

THUCYDIDES ON CHORAL PERFORMANCE BYTHE DELIAN MAIDENS AT THE DELIA

As we saw in the passage I quoted from Thucydides (3. 104. 5) at theoutset, the historian refers to theDelian Maidens as gunaikes ‘women’

79 See also HB 284. In Apoll., the Delian Maidens are described as the therapnai‘attendants’ of the god Apollo (157), and they are addressed by the poet of the Hymnwith the hymnic salutation khairete ‘hail and take pleasure’ (166), in conjunction withthe god Apollo (165). With his salutation of khairete (166), the poet is asking theDelian Maidens to accept the kharis ‘favour’ of his song and to give him their ‘favour’,their kharis, in return. The hymnic salutation khaire / khairete is used in the HomericHymns to address the given god / gods presiding over the performance of each givenHymn. Similarly in the Hesiodic Theogony, the figure of Hesiod addresses the Museswith the hymnic salutation khairete (104) in the context of naming them, in conjunc-tion with Apollo, as the divine sources of poetic power (94–5). On the salutationkhaire cf. in this volume Calame (p. 354).

80 On the transformation of the Hesiodic Muses in the Theogony from localHeliconian goddesses to Panhellenic Olympian goddesses, see HB 285.

81 HC 2}40. On musical performance embedded in the Hymns, see in this volumeCalame (Ch. 14).


who perform in a ‘song-and-dance group’, that is, in a khoros(�e� . . .˜ÅºØÆŒe� å�æe� �H� ªı�ÆØŒH�). I will start with this referenceto choral performance at the Delia as I examine further the testimonyof Thucydides on the relationship of the Delian Maidens and Homer.

When Thucydides refers to this group as a khoros of local ‘women’,not Muses, he is simply following his usual practice of downplayingdetails of myth and ritual. But I argue that the Delian Maidens whoperform at the festival of the Delia in Delos are simultaneously Musesas well as real-life performers participating in a chorus of local girls orwomen, since the role of divinity can be appropriated by participantsin a chorus during choral performance.82 That is to say, the DelianMaidens as a local female chorus can re-enact the Delian Maidens aslocal Muses.83

What is essential for now is that even Thucydides accepts the ideathat the chorus of Delian Maidens with whom Homer himself inter-acted in the Homeric Hymn to Apollo is somehow a prototype for thechoruses that perform at the Delia in historical times. Relevant is thewording of Thucydides where he makes these further remarks aboutthe performances of choruses at the Delia (3. 104. 6): ‘So much for theevidence given byHomer concerning the fact that there was even in theremote past a great coming together and festival [heortē] at Delos; lateron, the islanders [nēsiōtai] and the Athenians continued to send song-and-dance groups [khoroi], along with sacrificial offerings [hiera].’84

For city-states to send their local choruses to perform at festivals likethe Delia was considered to be part of a ‘sacrifice’, as conveyed by theword hiera ‘sacrificial offerings’ in the description I just quoted fromThucydides.85 In fact, as I show in a study of epigraphical and literarysources dating especially from the fourth century bce, a traditionalword for ‘festival’ was thusia, which literally means ‘sacrifice’.86

There is a most relevant attestation of both these words, hiera‘sacrificial offerings’ and thusia ‘sacrifice, festival’ in an account

82 HB 284–5; also Nagy (2006), 317, 319–20.83 See also Calame (2001) 30, 104, 110.84 ��ÆF�Æ b� �OÅæ� K��ŒÅæ�ø�� ‹�Ø q� ŒÆd �e ��ºÆØ �ª�ºÅ �� ŒÆd ��æ�c

K� �Bfi ˜�ºøfi · o��æ�� b ��f b� å�æ�f �ƒ �Å�ØH�ÆØ ŒÆd �ƒ �ŁÅ�ÆE�Ø �Ł’ ƒ�æH��.

85 Relevant also is the word anagein ‘bring’ in Thucydides 3. 104. 3, since this verb,with khoroi as its direct object, ‘seems to describe a typically religious “contribution” ’.So Kowalzig (2007), 71.

86 PR 40–1, 48–9, 52–3, 83.

308 Gregory Nagy

concerning the celebration of the Delia at Delos in the glory days ofthe Athenian empire. This account, which gives a full-blown descrip-tion of the magnificent spectacle of choral performances at the Delia,comes from Plutarch’s Life of Nicias (3. 5–7). This description byPlutarch, composed half a millennium after the events described, stillfeatures the essential ritual concepts of hier(ei)a ‘sacrificial offerings’and thusia ‘sacrifice, festival’.87

And we know that the practice of sending choral groups to performat Delos, as attested in Plutarch’s narrative about events taking placein the second half of the fifth century, was in full bloom already in thefirst half of the fifth century bce, when choruses sang and dancedchoral songs composed by such master poets as Simonides andPindar. The primary evidence comes from surviving fragments oftheir choral songs.88

87 Here is my translation of the relevant passage in Plutarch’s Life of Nicias(3. 5–7), showing in context the two key words hier(ei)a ‘sacrificial offerings’ andthusia ‘sacrifice, festival’:

Nicias is remembered for his ambitious accomplishments with regard to Delos—accomplishments most spectacular in all their splendour and most worthy of thegods in all their magnificence. Here is an example. The choral groups [khoroi]that cities used to send (toDelos) for the performances of songs sacred to the god(Apollo) used to sail in (to the harbour of Delos) in a haphazard fashion, and thecrowds that would gather to greet the ship used to start right away to call on theperformers to start singing their song. There was no coordination, since theperformers were still in the process of disembarking in a rushed and disorganizedway, and they were still putting on their garlands and changing into theircostumes. But when he [¼ Nicias] was in charge of the sacred voyage [theōria](to Delos), he first took a side trip to the island of Rheneia, bringing with him thechoral group [khoros] and the sacrificial offerings [hiereia] and all the rest of theequipment. And he brought with him a bridge that had been made in advance,back in Athens, to fit the present occasion, and this bridge was most splendidlyadorned with golden fixtures, with dyed colours, with garlands, with tapestries.Overnight, he took this bridge and spanned with it the strait between Rheneiaand Delos—not a very great distance. Then, come daylight, he led the processionin honour of the god and brought across the bridge to their destination theperformers of the choral group [khoros], who were outfitted most magnificentlyand were all along performing their song. Then, after the sacrifice [thusia] andafter the competition [agōn] and after the feasting, he set up as a dedication tothe god the (famous) bronze palm tree . . .

I should note about this passage that the initiative of Nicias in spanning the straitbetween the islands of Rheneia and Delos must have been interpreted as the ritualequivalent of the earlier chaining together of the two islands by Polycrates, as narratedby Thucydides 3. 104. 2.

88 This evidence has been surveyed and analysed by Kowalzig (2007), ch. 2.


In one of these choral songs, composed by Simonides to be per-formed at the Delia in Delos, the Delian Maidens themselves areactually called upon to shout a choral cry (PMG 519 fr. 55a 3):[Oº�º�]�Æ�� ˜Æº�ø� Ł�ªÆ�æ� (‘shout the cry of olologē, daughtersof the Delians’).89

The majority of these choral songs performed in honour of Apollocan be identified as paianes ‘paeans’.90 And there is an explicitreference in Euripides to such a paean as chorally performed by theDelian Maidens themselves at Delos:

A paean do the Delian Maidenssing as a humnos around the temple gates,singing (Apollo) the true child of Letoas they swirl, and they have such a beautiful khoros.I too, singing paeans at your palace,aged singer that I am, like a swan,from my greybearded throat,will send forth a cry. For whatever is realhas a place to stay in my humnoi. (E. HF 687–95)91

As we can see from this reference in a choral song composed byEuripides, the Delian Maidens are described as performing the kindof choral song that is known as the paian ‘paean’, which is equatedhere with the performing of a humnos.92

The equation is made clear in the syntax of the wording, which canbe analysed in two steps. First, the verb humneîn ‘sing a humnos’ takesas its inner object the song that is sung as a humnos, and this song isin fact a paian ‘paean’. Second, the same verb humneîn takes as itsouter object the name of the god Apollo, who is both the object ofpraise and the subject of the song that is the humnos. When I use theexpression subject of the song here, I mean the subject matter, notthe grammatical subject. In the grammar of a humnos as a song, thedivinity that figures as the subject of the song is in fact the grammat-ical object of the verb of singing the song.

89 Kowalzig (2007), 64–6.90 Kowalzig (2007), 57.91 Here are the original Greek lines that correspond to my translation: �ÆØA�Æ b�

˜ÅºØ�� j <�ÆH�> ��F�’ Içd ��ºÆ j �e� ¸Æ��F �h�ÆØ�Æ ª��, j �ƒº��ı�ÆØŒÆºº�å�æ�Ø·j �ÆØA�Æ �’ K�d ��E �º�Łæ�Ø j Œ�Œ�� S ªæø� I�Ø�e j ��ºØA� KŒ ª��ø�j Œ�ºÆ��ø· �e ªaæ �s j ��E o��Ø�Ø� ��æå�Ø.

92 See Kowalzig (2007), 66, and, in general, Henrichs (1996).

310 Gregory Nagy

The ritual context of such choral performance is made evident ina paean (Paean 12 ¼ fr. 52m SnM) composed by Pindar to beperformed at Delos by a chorus sent there from the island state ofNaxos (6, [˝Æ]��Ł��), and the context of the actual performance of thepaean is highlighted as a thusia ‘sacrifice’ (also line 6, Łı��[Æfi ]).93 Inthis same paean by Pindar, performers belonging to a local femalechorus, who are described as enkhōriai ‘the local ones’, are said to beshouting a ritual cry (19, [..]�çŁª�Æ�� ’ Kªå�æØÆØ). It has beenargued, plausibly, that this local female chorus can be identified withthe Delian Maidens.94 And I note with great interest that this localfemale chorus is evidently interacting with the visiting male chorussent from the island state of Naxos. The precise nature of such choralinteraction is most likely to be mimetic: for example, the visitingmale chorus may be re-enacting the performance of the local DelianMaidens,95 much as the visiting rhapsode Homer is re-enacting theperformance of the Maidens in the Homeric Hymn to Apollo.96

THUCYDIDES ON RHAPSODIC PERFORMANCEBY HOMER AT THE DELIA

From the evidence, then, of choral songs actually performed at thefestival of the Delia in Delos, it is clear that the Delian Maidens areenvisaged as performers of choral songs such as paeans.97 And it isalso clear that the verb humneîn, which I have been translating as‘sing a humnos’, expresses in those songs the ideal of choral

93 Kowalzig (2007), 60.94 Kowalzig (2007), 66–7.95 Kowalzig (2007), 71 argues that ‘the Naxians performing Paean 12 . . . them-

selves become, as it were, the Deliades’. I disagree, however, with her idea (67) thatthese Delian Maidens are ‘local women in myth’ as well as ‘local women in ritual’.

96 On the re-enacting of mythical performances of female khoroi by way ofthe ritual performances of male khoroi, see Power (2000) on B. Ode 13; also Nagy(forthcoming).

97 In this context, I return to a point I made earlier, when I noted the parallelismbetween the Panionian festival of the Delia at Delos and the Panionian festival of theEphesia at Ephesus. I think that this parallelism is relevant to another parallelism—between the choral performances of the Delian Maidens at Delos and the choralperformances of the so-called Lydian Maidens at Ephesus. For ancient references tothe Lydian Maidens, and for a brief analysis of these references, I cite PH 10}31 (¼ pp.298–9).


performances by the Delian Maidens. From the wording of Thucy-dides, however, it is just as clear that the historian understands thissame verb humneîn as a word that applies to the performing of ahumnos by Homer himself.98 And the problem is, the Homeric formof performance is not choral but rhapsodic.

We can see the problemmore clearly whenwe consider the wordingused by Thucydides in referring to the Delian Maidens (3. 104. 5),�e� . . .˜ÅºØÆŒe� å�æe� �H� ªı�ÆØŒH� ��Æ , which I have beentranslating this way: ‘after making the subject of his humnos theDeliansong-and-dance group [khoros] of women . . . ’. Thucydides may wellhave imagined the Delian Maidens as simply a chorus of local womenwhom Homer is praising, but the historian’s use of the verb humneînsays something different, showing that the Delian Maidens are notonly the object of Homer’s praise but also the subject of a humnosaddressed to divinities and telling all about these divinities.

The use of the verb humneîn here by Thucydides is in fact anaccurate syntactical replica for what is actually being said to divinitiesand about divinities in the Homeric Hymns in general, not only in theHomeric Hymn to Apollo in particular. For example, in the Hymn toHermes (verse 1), the object of humneîn and the main subject to bepraised is the god Hermes. We see comparable uses of humneîn at thebeginnings of Homeric Hymns 9 (Artemis), 14 (Mother of the gods),31 (Helios): o��Ø )�F�Æ at verses 1, 1, 2 of Hymns 4, 9, 14 and��E� . . . ¼æå�� )�F�Æ at verse 1 of Hymn 31. So also in the Hymn toApollo, as we see at verses 177–8, the object of humneîn ‘singing thehumnos’ and the main subject to be praised is the god Apollo: ÆP�aæKªg� �P º��ø �ŒÅ��º�� ººø�Æ j ��ø� Iæªıæ��, n� M�Œ�� Œ� ¸Å�� (‘as for me, Iwill not stop making Apollo the subject of myhumnos, the one who shoots from afar, the one with the silver bow,the one who was born of Leto with the beautiful hair’). In this lastexample, the performer who sings the praises of Apollo is figured asHomer himself.

In the Delian part of the Hymn to Apollo, the figure of Homer issinging the praises of the god Apollo and, secondarily, of the DelianMaidens as the local Muses of the god. As we have already seen, theseMaidens are described as the therapnai ‘attendants’ of the god Apollo

98 The word humnos is sometimes translated simply as ‘song’. As my analysisproceeds, I hope to show that this translation is too general and that humnos is used inspecific contexts referring to the linking of songs in performance.

312 Gregory Nagy

(157), and the figure of Homer addresses them with the hymnicsalutation khairete ‘hail and take pleasure’ (166), in conjunctionwith the god Apollo (165).

Here I repeat a relevant fact, that the hymnic salutation khaire /khairete is used in the Homeric Hymns to address the given divinity ordivinities presiding over the performance of each given Hymn. Simi-larly in the Hesiodic Theogony, the figure of Hesiod addresses theMuses with the hymnic salutation khairete (104) in the context ofnaming them, in conjunction with Apollo, as the divine sources ofpoetic power (94–5). So the Delian Maidens, just like the Muses,receive the same form of hymnic salutation that all the divinities ofthe Homeric Hymns are eligible to receive.

At the very moment when Thucydides is getting ready to quote theverses in which the figure of Homer gives the Delian Maidens hishymnic salutation, the historian describes as an act of epainos ‘praise’what Homer is about to do (3. 104. 5): ‘after making the subject of hishumnos the Delian song-and-dance group [khoros] of women, he wasdrawing toward the completion [telos] of his words of praise [epai-nos], drawing toward the following verses [epos plural]’.99 From thiswording of Thucydides, we see that the verses in which Homeraddresses the Delian Maidens are morphologically parallel to versesthat address divinities like the Muses or even Apollo himself in theHomeric Hymns.

THE PERFORMING OF HUMNOI BY THE DELIANMAIDENS AND BY THE MUSES

We have already seen that the figure of Homer ‘sings a humnos’, asexpressed by the verb humneîn, in the Hymn to Apollo at verses177–8, where the object of humneîn and the main subject to bepraised is the god Apollo himself (ÆP�aæ Kªg� �P º��ø �ŒÅ��º��

��ººø�Æ j ��ø� Iæªıæ��, n� M�Œ�� Œ� ¸Å��). But nowwe will see that the Delian Maidens themselves can be shown in theact of ‘singing a humnos’, as expressed by the same verb humneîn: it

99 �e� ªaæ ˜ÅºØÆŒe� å�æe� �H� ªı�ÆØŒH� ��Æ K��º��Æ ��F K�Æ��ı K �� a��Å.


happens earlier on in the same Hymn, at verses 158–9, and onceagain the object of humneîn and the main subject to be praised isthe god Apollo himself, though he is accompanied this time by hisdivine mother and sister, Leto and Artemis (Æ¥ �’ K��d iæ �æH�� b��ººø�’ ��ø�Ø�, j Æs�Ø �’ Æs ¸Å�� ŒÆd @æ��Ø� N�åÆØæÆ�).

The fact that the Delian Maidens are shown here in the act of‘singing the humnos’, as expressed by humneîn, shows that they areacting as Muses: we have already seen the same verb humneîn incontexts where the Muses themselves are called on to sing the praisesof a given divinity who presides over a given humnos, as at thebeginnings of Homeric Hymns I listed above.

Here I need to review what I said before about the context ofhumneîn in the Heracles of Euripides (687–9), where the agents ofpraise for Apollo are envisaged once again as the Delian Maidens: inthis context, the verb humneîn means not just ‘sing a humnos’ butalso, more specifically, ‘sing a humnos to and about a divinity’, takingas its inner object the song that is sung as a humnos and as its outerobject the name of the divinity who is the subject of the song. I repeatmy formula: in the grammar of a humnos as a song, the divinity whofigures as the subject of the song is in fact the grammatical object ofthe verb of singing the song.

All this is not to say that the Delian Maidens, or the Muses, for thatmatter, sing only the praises of gods. To make this point, I start bynoting that the Delian Maidens also sing about ‘men and women ofancient times’, as we see in the full version of the description that tellsabout the performance of the Maidens in the Hymn to Apollo:

And after they [¼ the Delian Maidens] make Apollo, first and foremost,the subject of their humnos [humneîn],

and, right after him, Leto and Artemis, the one who delights in herarrows,

then they [¼ the Delian Maidens] take note of men and women of thepast

as they [¼ the Delian Maidens] sing their humnos, enchanting all thetribes of humans. (Apoll. 158–61)100

100 Here are the original Greek lines that correspond to my translation: Æ¥ �’ K��diæ �æH�� b� ��ººø�’ ��ø�Ø�, j Æs�Ø �’ Æs ¸Å�� ŒÆd @æ��Ø� N�åÆØæÆ�, j�Å��ÆØ I��æH� �� ÆºÆØH� M�b ªı�ÆØŒH� j o�� I��ı�Ø�, Łºª�ı�Ø �b çFº’I�Łæ��ø�.

314 Gregory Nagy

The noun humnos here (161) applies evidently to subjects other thanthe subject of gods as we see them being praised in the HomericHymns.

Similarly, the Muses in the Hymn to Apollo include subjects thatare other than their main subject, Apollo, in their own singing of thehumnos, as we can see in this description of the Muses as they ascendto the heights of Olympus:101

In the meantime (as the Muses are ascending to Olympus), they aretaking turns, all of them, as they sing with their beautiful voice,

making as the subject of their humnos [humneîn] the immortalizinggifts of the gods (yes, I can see it all102) and, when it comes tohumans,

(they make as the subject of their humnos) the things that humans haveto endure—all the things they have to deal with, since humans areunder the power of the immortal gods,

and they live out their lives without the right thoughts and without theright solutions. They are unable

to discover the remedy for death and the prevention of old age. (Apoll.189–93)103

As we can see in this passage, a humnos that begins with a divinity ordivinities as its opening subject can then move on to stories of thehuman condition as its next subject. The surviving evidence aboutsuch stories takes the form of poetry composed in hexameters andtransmitted in rhapsodic performances.104

And what kinds of stories could be told in such rhapsodic perform-ances? Closest to home, such stories could be about humans in the eraof heroes, as we find them in some of the larger Homeric Hymns (as inthe case of Anchises in the Homeric Hymn to Aphrodite). Orthey could be about heroes as we find them in epic poetry like theHomeric Iliad and Odyssey. Such a subject is announced in Hymn 32

101 On the ‘politics of Olympus’, see GM 48–9, Clay (1989), 9–10, and this volume(Ch. 11); also HB 295 n. 64.

102 The particle ¼æÆ / ÞÆ / ¼æ ‘so, then’ has an ‘evidentiary’ force, indicating that thespeaker notionally sees what is simultaneously being spoken. See Bakker (2005) 12 n.32, 80, 84, 97–100, 104, 146, 172 n. 33.

103 Here are the original Greek lines that correspond to my translation:)�F�ÆØ �Ł’ –Æ �A�ÆØ I�Ø��ÆØ O�d ŒÆºBfi j ��F�� ÞÆ Ł�H� �Hæ’ ¼�æ��Æ M�’ I�Łæ��ø� j�ºÅ��Æ , ‹�’ �å�� ’ IŁÆ��Ø�Ø Ł��E�Ø� j Ç��ı�’ IçæÆ�� ŒÆd I�åÆ��Ø, �P�b��Æ��ÆØ j ��æ��ÆØ ŁÆ��Ø� �’ ¼Œ� ŒÆd ª�æÆ� ¼ºŒÆæ·

104 HC 2}}110–17.


to Selene (17–20).105 Or again, they could be about heroes in genea-logical poetry. Such a subject is announced in Hymn 31 to Helios(17–19).106 Such a subject is also announced in the part of theHesiodic Theogony that tells of mortals begotten or conceived byimmortals (starting at verse 965). In the case of the Theogony, thepart of the narrative that tells about the genesis of heroes comes afterthe initial part of the composition, where the figure of Hesiod per-forms a humnos (33 ��E�) that praises the Muses (34), who arethemselves the models for the performing of a humnos (11/37/51��F�ÆØ, 70 ��ÆØ ).107

A PARTING OF WAYS BETWEEN CHORAL ANDRHAPSODIC PERFORMANCES OF THE HUMNOS

In what we have seen so far, the performing of the humnos in theHomeric Hymns is visualizedmostly as a choral activity. And the choralaspects of performing the humnos are shown most clearly in descrip-tions of the Muses, who are the idealized performers of the Hymns andwho are represented as singing and dancing in an ideal form of choralperformance. But the fact remains that the real form of performinghumnoi, when these humnoi are Homeric Hymns, must be rhapsodic,not choral, since these Hymns are composed in hexameter, which is adecidedly rhapsodic form of performance. So we see here a parting ofways between rhapsodic and choral performances of the humnos.

The rhapsodic form of performing theHomeric Hymns affects eventhe idealized descriptions of the Muses in these Hymns, since theseMuses reveal some traits of rhapsodic as well as choral performers.In the last quotation I showed from the Homeric Hymn to Apollo, forexample, we can see an allusion to the poetics of rhapsodic relay in thedescription of theMuses in the act of performance: these goddesses aresaid to be ‘taking turns, all of them, as they sing with their beautifulvoice’ ()�F�ÆØ � Ł’ –Æ �A�ÆØ I�Ø��ÆØ O�d ŒÆºBfi , 189).108

105 HC 2}110.106 Again, HC 2}110.107 HC 2}}112–14.108 Relevant is a reference I cited earlier from the scholia D for Iliad I (604), where

the phrase ek diadokhēs refers to relay singing by the Muses: ŒÆd Æy�ÆØ ��ººø��

316 Gregory Nagy

Aside from such sporadic examples, however, most references tothe performing Muses in the Homeric Hymns show them in a purelychoral setting. By contrast, the references to the performing Homerin the Hymn to Apollo show him in the act of rhapsodic perform-ance.

It can happen, however, that the idealized choral form of perform-ing the Homeric Hymns affects even the otherwise realistic descrip-tions of rhapsodic performance. A case in point is the Hesiodicfragment, quoted above (fr. 357 M–W), in which we have found anembedded version of the myth about a contest between Homer andHesiod. In this fragment, as we have seen, the verb melpein ‘sing-and-dance’ explicitly conveys a combination of singing and dancing.109 Sothis example illustrates once again the point I made earlier, thatrhapsodic as well as choral performances are highly mimetic.

Still, the fact remains that there is a parting of ways betweenrhapsodic and choral performances of the humnos that is thisHymn. And, as we will now see, Thucydides is aware of the gapproduced by such a parting of ways.

THUCYDIDES ON THE GAP BETWEEN CHORALAND RHAPSODIC PERFORMANCES OF

THE HOMERIC HYMN TO APOLLO

As we saw earlier, Thucydides says at one point that the HomericHymn to Apollo actually refers to the practice of sending choralperformers to the Delia (3. 104. 3–4). At a later point he adds that,while this practice was continued up to the time when the Atheniansreorganized the festival in 426 bce, other related practices hadalready been discontinued (3. 104. 6): ‘but various misfortunes evi-dently caused the discontinuation of the things concerning the com-petitions [agōnes] and most other things—that is, up to the time inquestion [¼ the time of the purification] when the Athenians madethe competition [agōn], including chariot races [hippodromiai],

ŒØŁÆæ�Ç�� KŒ �ØÆ��åB �Ææa æ� fi q�� (‘and they [¼ the Muses], while Apollo wasplaying the kithara, were singing in relay [para meros], by taking turns [ek dia-dokhēs]’).

109 PH 12}29 n. 62 (¼ p. 350) and n. 64 (¼ p. 351). See above pp. 289–90.


which had not taken place before then’.110 As I will now argue, one ofthe aspects of the festival of the Delia that had been ‘discontinued’, interms of the reconstruction offered by Thucydides, was the practiceof performing rhapsodically the Homeric Hymn to Apollo, to becontrasted with the continuing practice of sending choral performersto that festival.

As the wording of Thucydides indicates here, what had beendiscontinued at the Delian festival were agōnes ‘competitions’ otherthan the competitions in choral performance. Earlier in the passageI have quoted, Thucydides describes such non-choral competitions interms of two categories (3. 104. 3): ‘and a competition [agōn] was heldthere [¼ in Delos], both athletic [gumnikos] and musical [mousi-kos]’.111 The first kind of agōn ‘competition’ mentioned here byThucydides is self-evident: it is gumnikos or ‘athletic’, as exemplifiedby the contests of boxing mentioned in the Hymn to Apollo (149). Asfor the second kind of agōn ‘competition’mentioned here, however, itcan easily be misunderstood: it is an agōn that is mousikos. So far,I have translated this word mousikos as ‘musical’, but I must nowpoint out that a mousikos agōn is a competition that involves morethan what we mean when we say ‘musical’. More generally, as we areabout to see, a mousikos agōn is a competition in recitative as well assung forms of performance. Such recitative forms of performance, aswe are also about to see, are basically rhapsodic, as exemplified by theHomeric Hymns.

In the same extended passage I quoted at the beginning fromThucydides, he refers to this same kind of competition also as agōnmousikēs. Here again is his wording (3. 104. 5): ‘that there was also acompetition [agōn] in the craft of music [mousikē (tekhnē)], in whichthe Ionians went to engage in competition [agōnizesthai], again ismade clear by him [¼Homer] in the following verses, taken from thesame prooemium [prooimion]’.112 Here again, though I have so fartranslated agōn mousikēs as ‘competition in the craft of music’, I mustpoint out that the adjective mousikē, which describes the understoodnoun tekhnē, refers to a craft that is more than simply ‘musical’. More

110 �a �b ��æd ��f IªH�Æ ŒÆd �a �º�E��Æ ŒÆ��º�ŁÅ ��e �ıç�æH�, ‰ �NŒ� , �æd��c �ƒ �ŁÅ�ÆE�Ø �� e� IªH�Æ K��Å�Æ� ŒÆd ƒ��æ��Æ , n �æ��æ�� PŒ q�.

111 ŒÆd Iªg� K��Ø�E�� ÆP��ŁØ ŒÆd ªı�ØŒe ŒÆd �ı�ØŒ� .112 ‹�Ø �b ŒÆd �ı�ØŒB Iªg� q� ŒÆd Iªø�Ø��Ø Kç��ø� K� ��E�� Æs �Åº�E , –

K��Ø� KŒ ��F ÆP��F �æ��Ø��ı.

318 Gregory Nagy

generally, as we are about to see, the tekhnē or ‘craft’ of mousikē refersto forms of rhapsodic as well as sung performance.

Decisive comparative evidence comes from the terminology con-cerning the competitions at the festival of the Panathenaea inAthens. In the Aristotelian Constitution of the Athenians (60. 1),there is an explicit reference to the ‘competition [agōn] in mousikē’(�e� IªH�Æ �B �ı�ØŒB ) at the Panathenaea. The author goes on tosay that ten magistrates called athlothetai ‘arrangers of the contests[athloi]’ were selected by lot every four years to organize the festivalof the Panathenaea, and one of their primary tasks was the manage-ment of this ‘competition [agōn] in mousikē’. According to Plu-tarch’s Pericles (13. 9–11), the Athenian statesman Pericles reformedthis competition in mousikē when he was elected as one of theathlothetai.113

In such contexts, the word mousikē is a shorthand way of sayingmousikē tekhnē, meaning ‘craft of the Muses’, that is, ‘musical craft’.As I am now arguing, however, it is misleading to think of ancientGreek mousikē simply in the modern sense of music, since thecategories of ‘musical’ performers at the Panathenaea included notonly kitharōidoi ‘kithara-singers’ and kitharistai ‘kithara-players’ andaulōidoi ‘aulos-singers’ and aulētai ‘aulos-players’ but also rhapsōidoi‘rhapsodes’.114And the medium of rhapsodes performing at thePanathenaea was recitative and thus not ‘musical’ in the modernsense of the word. By recitative I mean (1) performed without singingand (2) performed without the instrumental accompaniment of thekithara or the aulos.115

In the historical period, as we see especially in the evidence fromthe fourth century bce, the competitive performances by rhapsodes atthe Panathenaea were limited to the Homeric Iliad and Odyssey.These performances were ‘musical’ only in an etymological sense,and the medium of the rhapsode was actually closer to what we call‘poetry’ and farther from what we call ‘music’ in the modern sense ofthe word. Still, the fact remains that the performances of rhapsodesbelonged to what is called the ‘competition [agōn] in mousikē’ (�e�

113 Rhodes (1981), 670–1.114 PR 38, surveying the evidence from inscriptions (especially IG II2 2311) and

from literary sources (especially Pl. Ion and Lg.).115 PR 36, 41–2.


IªH�Æ �B �ı�ØŒB ), just like the performances of citharodes, cithar-ists, aulodes, auletes, and so on.116

We find in Plato’s Ion an explicit reference to the recitation ofHomeric poetry by rhapsodes competing in ‘the competition [agōn]in mousikē’ at the Panathenaea (Ion 530a).117 And there is anothersuch explicit reference in Speech 5 of Isocrates, the Panegyricus (159),where the speaker describes as an ancestral Athenian institution theperforming of Homeric poetry in competitions at the Panathenaea,referring to these competitions as athloi ‘contests’ in mousikē (�� E �B �ı�ØŒB ¼Łº�Ø ).

As I have already argued, the Homeric Hymn to Apollo, which iscomposed in hexameter, just like the Homeric Iliad and Odyssey, was arhapsodicmedium.And so, just like theHomeric Iliad andOdyssey, thisHomeric Hymn would have been performed in the context of an agōn‘competition’ in the craft of mousikē. This is understood by Thucydideswhen he says (3. 104. 3), ‘and a competition [agōn]was held there [¼ inDelos], both athletic [gumnikos] and musical [mousikos], and the citiesbrought song-and-dance groups [khoroi]’,118 and when he says later (3.104. 5), ‘that there was also a competition [agōn] in the craft of music[mousikē (tekhnē)], in which the Ionians went to engage in competition[agōnizesthai], again is made clear by him [¼ Homer] in the followingverses, taken from the same prooemium [prooimion]’.119

In the wording of Thucydides here (3. 104. 5), I highlight the ‘also’(ŒÆ�) in the expression ‘that there was also a competition [agōn] inthe craft of music’ (‹�Ø �b ŒÆd �ı�ØŒB Iªg� q�). Since Thucydides isreferring to whatHomer is about to say about his own performance inthe verses about to be quoted, it is clear that mousikē must be under-stood here as including not only the performances of choral groupsbut also the performance of the Homeric Hymn, which is a rhapsodicmedium composed in hexameter.

So the agōn ‘competition’ in mousikē at the Delia in Delos, accord-ing to the understanding of Thucydides, had once included theperformance of the Homeric Hymn to Apollo. And of course it alsoincluded the performances of khoroi ‘choruses’ as part of an overall

116 HC 3}29.117 I have analysed this passage in PR 22, 37–8, 99.118 ŒÆd Iªg� K��Ø�E�� ÆP��ŁØ ŒÆd ªı�ØŒe ŒÆd �ı�ØŒ� , å�æ�� I�Bª�� Æƒ ��º�Ø .119 ‹�Ø �b ŒÆd �ı�ØŒB Iªg� q� ŒÆd Iªø�Ø��Ø Kç��ø� K� ��E�� Æs �Åº�E , –

K��Ø� KŒ ��F ÆP��F �æ��Ø��ı.

320 Gregory Nagy

mousikos agōn.120 And the mousikos agōn matched an overall gum-nikos agōn or athletic competition. In the verses that Thucydidesquotes as supporting evidence from the Hymn to Apollo, there isreference made to boxing, dancing, and song, as we have alreadyseen (verse 149).

But then, sometime after the Homeric Hymn to Apollo was firstperformed, according to the understanding of Thucydides, most as-pects of both the mousikos agōn and the gumnikos agōn were ‘discon-tinued’ until the Athenians re-established them in the year 426 bce. Irepeat the relevant wording of the historian (3. 104. 6): ‘later on, theislanders [nēsiōtai] and the Athenians continued to send song-and-dance groups [khoroi], along with sacrificial offerings [hiera], butvarious misfortunes evidently caused the discontinuation of the thingsconcerning the competitions [agōnes] and most other things—that is,up to the time in question [¼ the time of the purification] when theAthenians set up the competition [agōn], including chariot races[hippodromiai], which had not taken place before then’.121

This statement about the continuity of the practice of sendingkhoroi ‘choruses’ to perform competitively at the Delia in Delosmatches the evidence we have already considered about the perform-ances of choral songs like paeans at the Delia in the era of Simonidesand Pindar. And the corresponding statement about discontinuitiesmatches our lack of evidence for any other continuity. Evidently,among the practices that had lapsed before 426 bcewas the seasonallyrecurring performance of the Homeric Hymn to Apollo, despite theprediction we hear within the Hymn about the eternal continuity ofthis practice. And it is not at all clear whether the Athenians restoredthis practice in 426 bce. What they did restore, however, is a set ofpractices that match the practices then current at their own greatfestival, the quadrennial Panathenaea, as we see from the referencemade by Thucydides to a new quadrennial format and the inclusionof such new events as chariot racing.122

120 I analyse in HC 3}}77–94 the meaning of mousikē in Plato’s Laws. In thatanalysis, I should have mentioned also Strabo 10. 3. 9–10. See Petrovic (2011a forth-coming). I note with special interest what Strabo says about Plato and the Pythagor-eans at the beginning of 10. 3. 10.

121 o��æ�� b ��f b� å�æ�f �ƒ �Å�ØH�ÆØ ŒÆd �ƒ �ŁÅ�ÆE�Ø �Ł’ ƒ�æH� ��, �a�b ��æd ��f IªH�Æ ŒÆd �a �º�E��Æ ŒÆ��º�ŁÅ ��e �ıç�æH�, ‰ �NŒ� , �æd� �c�ƒ �ŁÅ�ÆE�Ø �� e� IªH�Æ K��Å�Æ� ŒÆd ƒ��æ��Æ , n �æ��æ�� PŒ q�.

122 See Rhodes (1994), 260.


The word hiera ‘sacrificial offerings’, applied by Thucydides in thecontext of the Delia (3. 104. 6), is surely applicable in the context ofthe Panathenaea as well. The same can be said about the word thusiain the sense of ‘festival’, as we have seen it used by Plutarch along withthe word hier(ei)a in referring to the Delia (Nic. 3. 5–7). A primeexample of such an overall sacrificial context is the great Athenianfestival of the Panathenaea, to which Plato’s Socrates refers as a thusia(Ti. 26e).123

The Athenocentrism inherent in the reorganization of the festival oftheDelia atDelos is made explicit in themyth that connects Theseus, asculture hero of Athens, with the Delia: he is described as the founder ofan agōn ‘competition’ there (Plu. Thes. 21. 3).124 Though the Atheno-centrism is left implicit in the account of Thucydides, the underlyingidea is unmistakable: ‘Thucydides claims, as Athenian propagandamust have done at the time, that the Athenians were not creatingsomething new but reviving an ancient Ionian festival.’125

The setting of the festival of the Delia as reorganized in 426 bce isparallel to the setting of the Homeric Hymn to Apollo itself. In termsof the Hymn, as quoted and interpreted by Thucydides, the speaker inthis setting is the speaker of the Hymn. And Thucydides recognizesthe speaker of the Hymn as Homer (3. 104. 4, 5, 6). The ancienthistorian thinks he is quoting the words of Homer himself as hequotes from the Hymn the verses we recognize from the medievalmanuscript transmission of Homer (Thucydides 3. 104. 4 / 5: Hymnto Apollo 146–50 / 165–72). This thinking of Thucydides is a mostvaluable piece of evidence about ancient ideas ofHomer. It goes to theroot of the conventional Athenian idea of Homer.126

AN OCCASION FOR PERFORMINGA HOMERIC HYMN

We may conclude from the testimony of Thucydides that the festivalof the Delia had undergone many changes over the years leading up

123 PR 53.124 Hornblower (1991), 523; Rhodes (1994), 258.125 Rhodes (1994), 259.126 See also Graziosi (2002), 222–6, who adduces Choricius Laudatio Marciani 2. 3.

322 Gregory Nagy

to 426 bce. But one thing clearly remained a constant, and it iscaptured by the word agōn, which means literally ‘competition’.When Thucydides refers to the festival of the Delia in the passagesI quoted above, he uses this noun not once but several times.127 In thetext of the Homeric Hymn itself, the same noun agōn is used inreferring to the festival (150). And what kind of ‘competition’ takesplace at this festival? In the words of the Hymn as quoted by Thucy-dides, there is ‘singing and dancing’ as well as ‘boxing’: ‘and therewith boxing and dancing and song j they have you in mind anddelight [terpein] you, whenever they set up a competition [agōn]’(149–50).128 So, evidently, the contests were ‘musical’ as well as‘athletic’. In using these words ‘musical’ and ‘athletic’, I am followingdirectly the wording used by Thucydides in considering the evidenceof the verses he quotes from the Hymn (in this case, 149–50). Hereagain is his wording (3. 104. 3–4): ‘a competition [agōn] was heldthere [¼ in Delos], both athletic [gumnikos] and musical [mousikos],and the cities brought song-and-dance groups [khoroi]. Homermakes it most clear that such was the case in the following verses[epos plural], which come from a prooemium [prooimion] having todo with Apollo’.129

In another one of the Homeric Hymns we see an explicit referenceto such a festival that served as a setting for the competitive perform-ance of a Hymn. It happens in Hymn 6 to Aphrodite (19–20), whichrefers explicitly to the occasion of the festival where the performanceof this Hymn is taking place: ‘I pray to you [¼ Aphrodite] to grantthat in the competition [agōn] that is at hand I may win victory.Arrange my song.’130

It seems to me unjustifiable to assume that a performer could winsuch a victory by performing a Homeric Hymn in competition withother performers performing other such Hymns. The competition,I will now argue, is not between Hymns. Each Hymn has something

127 Once at 3. 104. 3, once at 3. 104. 5 [along with the verb derived from the noun],twice at 3. 104. 6.

128 ��ŁÆ �� ıªÆå�fi Å �� ŒÆd OæåÅ��ıE ŒÆd I�Ø�Bfi j �Å��Ø �æ��ı�Ø�, ‹�Æ�ŒÆŁ�ø�Ø� IªH�Æ.

129 ŒÆd Iªg� K��Ø�E�� ÆP��ŁØ ŒÆd ªı�ØŒe ŒÆd �ı�ØŒ� , å�æ�� I�Bª�� Æƒ ��º�Ø .�Åº�E �b �ºØ��Æ �OÅæ� ‹�Ø ��ØÆF�Æ q� K� ��E ��Ø ��E��, – K��Ø� KŒ �æ��Ø��ı��ººø�� .

130 �e �’ K� IªH�Ø j ��ŒÅ� �fiH�� çæ��ŁÆØ, Kc� �’ ��ı�� I�Ø��.


additional that is part of the same Hymn. And the competitionincludes not only each Hymn but also that additional somethingthat is part of each Hymn.131 In making this argument, I introducea word that Thucydides uses twice in referring to the Homeric Hymnto Apollo: that word is prooimion, which I have until now translatedsimply by way of a Latin term borrowed from the Greek, ‘prooe-mium’. As far as Thucydides is concerned, the Homeric Hymn toApollo is ‘the prooimion of Apollo’, and it is from this prooimion thathe quotes the words of Homer (3. 104. 4/5 KŒ �æ��Ø��ı ��ººø�� jKŒ ��F ÆP��F �æ��Ø��ı).

The essential characteristic of a prooimion is that it starts aperformance. To back up this formulation, I return to an ancientparaphrase that I have already quoted in a larger context but havenot yet highlighted. This paraphrase concerns the Homēridai, and wefind it in the scholia for Pindar’s Nemean 2 (1d): ‘and they [¼ theHomēridai] always started with a prooimion, making mostly Zeustheir point of departure and occasionally the Muses.’132 This word-ing paraphrases the genuine wording of Pindar, which is a mimeticrewording of a genuine ‘prooimion of Zeus’ as performed by theHomēridai:

(Starting) from the point where [hothen] the Homēridai, singers, mostof the time begin [arkhesthai] their stitched-together words, from theprooimion of Zeus . . . (Pi. N. 2. 1–3)133

Besides Pindar’s imitation, there is an actual attestation of aHomeric Hymn to Zeus. It is highly compressed, consisting of onlyfour verses:

I will sing Zeus as my subject, best of the gods, and most great,whose sound reaches far and wide, the ruler, the one who brings things

to their outcome [telos], the one who has Themisattentively seated at his side, and he keeps her company with regular

frequency.

131 For a different view on the Hymns as prooimia, see Clay in thisvolume (pp. 237–40); cf. Introduction (pp. 17–19).

132 ÆN�d �s� �c� Iæåc� ‰ K�d �e �º�E�� KŒ ˜Øe K��Ø�F�� æ��ØØÆÇ��Ø, K��b ŒÆd )�ı�H�.

133 ῞ˇŁ�� æ ŒÆd % ˇÅæ��ÆØ j ÞÆ��H� K�ø� �a ��ºº! I�Ø�� j ¼æå��ÆØ, ˜Øe KŒ�æ��Ø��ı.

324 Gregory Nagy

Be propitious, you whose sound reaches far and wide, son of Kronos,you who are most resplendent and most great. (Hy. 23)134

Zeus is figured as a transcendent hymnic subject: he can presideover a humnos even if that humnos is being performed at a festivalsacred to another god. This way, Zeus gets to preside over a humnosthat leads to a transcendent form of epic, which is a hymnicconsequent of the ‘prooimion of Zeus’. And that epic form is thepoetic legacy inherited by the ‘descendants of Homer’, theHomēridai.135

TRANSITION

Here I will shift gears in order to give a working definition of thetechnical terms (1) hymnic subject and (2) hymnic consequent. And,in the process of defining these two terms, I will address the basicmeanings of the Greek words humnos and prooimion.

THE HYMNIC SUBJECT

By hymnic subject, I mean the divinity who is invoked in a humnos or‘hymn’, as in a Homeric Hymn.

Up to now, I have avoided using the word hymn with a lower-case‘h’ as a translation of humnos. Only with reference to the HomericHymns, which do qualify as humnoi in the Greek language, haveI equated humnos with hymn. From here on, however, I will use theEnglish word hymn as the equivalent of Greek humnos in general,since we are about to see convergences between the ordinary uses ofEnglish hymn and the programmatic uses of Greek humnos in choralas well as rhapsodic forms of performance. What I say here about the

134 Here are the original Greek lines that correspond to my translation: ZB�Æ Ł�H��e� ¼æØ�� I��ÆØ M�b ªØ�� j �Pæ��Æ Œæ��Æ ��º��ç�æ��, ‹ �� ¨Ø��Ø jKªŒºØ�e� �Ç��fi Å �ıŒØ��f O�æ�ı OÆæ�Ç�Ø. j ¥ºÅŁ’ �Pæ��Æ ˚æ��Å Œ��Ø�� ªØ��.

135 HPC I}}242–9.


noun hymn applies also to the adjective hymnic. That said, I am readyto explain further what I mean by the term hymnic subject.

In a hymn, the invoked divinity who presides over a given festivalis the subject of the hymn. In the grammar of the Homeric Hymns,for example, the divinity who figures as the subject of any hymn isnormally the grammatical object of the verb of singing the hymn.136

In the logic of the Homeric Hymns, the divinity that presides over theoccasion of performance becomes continuous with the occasion andthus becomes the occasion.137

And this occasion of the humnos is notionally perfect because thegod who is the occasion is perfect. Such perfection is expressed by wayof the word eu-humnos (�hı�� ) ‘good for hymning’, as in thissublime aporetic question:138

For how shall I hymn you, you who are so absolutely [pantōs] good forhymning [eu-humnos]? (Apoll. 19, 207)139

The theology, as it were, of this aporetic question can be formulatedthis way:140

Faced with the absoluteness of the god, the performer experiences arhetorical hesitation: how can I make the subject of my humnos some-thing that is perfect, absolute? The absoluteness of this hymnic subject issignalled by the programmatic adverb pantōs ‘absolutely’, which modi-fies not only the adjective eu-humnos ‘good for hymning’ but also theentire phrasing about the absoluteness of the subject.141 The absolute-ness of the god Apollo is continuous with the absoluteness of thehumnos that makes Apollo its subject. This Homeric Hymn is sayingabout itself that it is the perfect and absolute humnos. As such, it is notonly the beginning of a composition but also the totality of the compos-ition, authorizing everything that follows it, because it was begun so

136 As at the beginnings of Hys. 2, 4, 6, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 20, 21, 23.26, 27, 28, 30, 31, 32.

137 HC 2}83.138 On the term ‘aporetic question’, see Bundy (1972) 47. On ‘apologetic’ and

‘aporetic’ rhetorical strategies, see Bundy (1972), 59 n. 59; also 60–1, 65.139 �H ª�æ �! ��ø, ��ø �hı�� K��Æ. At Apoll. 19, the manuscript

reading is ª�æ, while at 207, it is �’ ¼æ.140 HC 2}24.141 On the syntax of pantōs ‘absolutely’ as an overall modifier of absolute phrase-

ology, see for example Solon F 4.16 ed. West and the commentary in Nagy (1985),59–60, PH 9}7n38 (¼ p. 256).

326 Gregory Nagy

perfectly. And the source of the perfection is the god as the subject ofthe humnos.

The naming of the divinity as the subject of the humnos, togetherwith the initial describing of the divinity, is the notionally perfectbeginning of the humnos, and this beginning is the prooimion.Whereas the word prooimion refers only to the start of the con-tinuum, the word humnos refers to both the start of the continuumand the continuum itself. The Homeric Hymn to Apollo is a perfectexample: it refers to itself in terms of a humnos (verses 158, 161,178), while Thucydides refers to it explicitly as a prooimion (3. 104.4, 5).142

The etymology of the compound noun prooimion ‘prooemium’indicates that the word is a metaphor referring to the world of patternweaving.143As for the etymology of the simplex noun humnos, it hasin common with the compound noun prooimion a metaphoricalreference to fabric work, but in this case the fabric work is notspecified as pattern weaving. If I am right in explaining the etymologyof humnos in terms of the root huph- as in huphainein ‘weave’, thebasic idea conveyed by this noun is weaving in general, not patternweaving in particular.144

THE HYMNIC CONSEQUENT

By hymnic consequent, I mean a performance, epic or otherwise, thatfollows the performance of a humnos. There is a formal relationshipbetween the concept of humnos and the concept of epic as a hymnicconsequent.145 Here I highlight only one aspect of that relationship,which is the device of metabasis. Three most telling examples fromthe Homeric Hymns show this identical wording:

142 HC 2}89. See also Petrovic (2011a forthcoming).143 HC 2}92. The word prooimion means literally the ‘initial threading’ of a song,

parallel to the etymology of Latin exordium, which likewise means ‘prooemium’ inpoetic and rhetorical contexts and which can likewise be traced back to the basic ideaof an ‘initial threading’ or, to say it more technically in terms of fabric work, exastis or‘selvedge’. See also PP 63 n. 20 and PR 72, 81, with reference to Latin ex-ordium as asemantic equivalent of Greek pro-oimion.

144 HC 2}91, with bibliography on alternative etymological solutions.145 HC 2}}97, 109, 113–14, 116.


But, having started off from you, I will move ahead and shift forward[metabainein] to the rest of the humnos. (Hys. 5. 293, 9. 9, 18. 11)146

The transition in all three cases, as marked by metabainein ‘moveahead and shift forward’, is predicated on the idea of a perfectbeginning. The idea is, ‘I begin, starting from the god’. The processof transition or metabasis, signalled by the verb metabainein, isactivated by the hymnic salutation khaire / khairete, which I interpretas ‘hail and take pleasure’. In each of the three Homeric Hymns we areconsidering, the salutation khaire takes place in the verse precedingthe verse of metabasis. Implicit in this imperative form khaire of theverb khairein is the meaning of the related noun kharis, whichconveys the idea of a ‘favour’ achieved by reciprocating the pleasureof beauty.147 What drives the performative gesture of khaire / khaireteis the fundamental idea that the reciprocal favour of kharis is thesame beautiful thing as the pleasure that it gives. To give suchpleasure, I argue, is the key to success in reception.148

After the signal khaire / khairete in the Homeric Hymns, the actualprocess of metabasis can be activated. This process is made explicitin the expression I have just highlighted, ��Æ��ÆØ ¼ºº�� K o��‘I will move ahead and shift forward [metabainein] to the rest of thehumnos’ (Hys. 5. 292–3, 9. 7–9, 18. 10–12). The word humnos inthe wording ¼ºº�� K o�� in the Homeric Hymns marks the wholeperformance, so that ¼ºº�� K o�� means not ‘extending intoanother performance’ but ‘extending into the rest of the perform-ance’.149 So also the expression ¼ººÅ . . . I�Ø�B in other HomericHymns means not ‘another song’ but ‘the rest of the song’, as in thefollowing example:

146 ��F �’ Kªg Iæ�� Æ��ÆØ ¼ºº�� K o��.147 Making this idea explicit, I have formulated this paraphrase of khaire / khairete

in the context of all its occurrences in the Homeric Hymns (HC 2}99): Now, at thisprecise moment, with all this said, I greet you, god (or gods) presiding over the festiveoccasion, calling on you to show favour [kharis] in return for the beauty and thepleasure of this, my performance. See also Bundy (1972), 44, 49. For more on therhetoric of seeking the pleasure of the gods, see his p. 62 n. 65.

148 With reference to the use of kharis in Hy. 24 to Hestia (5), Bundy (1972), 83,speaks of a ‘concern for the pleasure of a critical audience as well as for that of thegod’.

149 PH 12}33 (¼ pp. 353–4), following Koller (1956), 174–82; see also Bakker(2005), 144, disagreeing with Clay (1997), 493. Further discussion in Petrovic (2011bforthcoming).

328 Gregory Nagy

You [¼ Demeter and Persephone] be favourably disposed, granting mea livelihood that fits my heart’s desire, in return for my song.

As for me, I will keep you in mind along with the rest of the song. (Dem.494–5)150

Here is the way I can summarize my findings about these transi-tions: ‘Metabasis is a device that signals a shift from the subject of thegod with whom the song started—what I have been calling thehymnic subject—and then proceeds to a different subject—in whatmust remain notionally the same song.’151

I bring this section to a close by quoting the relevant words of ElroyBundy: ‘Beginnings, middles, and ends: the meaning of literatureresides in its transitions.’152

THE STAGING OF A HUMNOS INTHE HOMERIC ODYSSEY

My formulation of the meaning of humnos is relevant to an ongoingperformance by the singer Demodocus as represented in Rhapsody

150 Here are the original Greek lines that correspond to my translation: �æ�çæ�� I��’ fiT�B �� Łı�æ�’ O��Ç�Ø�. j ÆP�aæ Kªg ŒÆd ��E� ŒÆd ¼ººÅ ��’ I�Ø�B .Other examples of this type include Apoll. 545-6, Herm. 579–80, Hys. 6. 19–21, 10.4–6, 19. 48–9, 25. 6–7, 27. 21–2, 28. 17–18, 29. 13–14, 30. 17–19, and 33. 18–19. Theseand other such examples of the expression ¼ººÅ . . . I�Ø�B have been described interms of a ‘break-off formula’ by Bundy (1972), 52–3, even though he recognizes the‘transitional’ function of this formula (52). I find the term ‘break-off ’ misleadingbecause it blunts the idea of ‘transitional’ (for more on Bundy’s use of the term‘transitional’ see his p. 87).

151 HC 2}109. There I go on to say:

Ideally, the shift from subject to different subject will be smooth. Ideally, thedifferent subject will be consequential, so that the consequent of what was startedin the humnos may remain part of the humnos. This way, the transition will leadseamlessly to what is being called ‘the rest of the song’. In other words, theconcept of humnos is the concept of maintaining the song as the notionally samesong by way of successfully executing a metabasis from the initial subject to thenext subject. The initial subject of the god and the next subject are linked as onesong by the humnos in general and by the device of hymnic metabasis inparticular. What comes before the metabasis is the prooimion, the beginning ofthe humnos. What comes after the metabasis is no longer the prooimion—but itcan still be considered the humnos.152 Bundy (1972), 59 n. 58.


8 of the Odyssey. The ongoing performance is signalled by the wordhumnos at verse 429. And the occasion for this ongoing humnos isan ongoing festival, as signalled by the word dais ‘feast’ at the sameverse 429.153 This ongoing performance of Demodocus is envisagedas an alternative to and rival of the epic about to be performed byOdysseus in Rhapsodies 9 through 12 of the Odyssey. And why is theperformance of Demodocus so different from the later performanceof Odysseus? As I argue, it is because Demodocus is represented asperforming forms of song that resemble (1) the epic form of the epicCycle, in the case of his first and third songs, which are about theTrojan War, and (2) the hymnic form of the Homeric Hymns, in thecase of his second song, which is about the love affair of Ares andAphrodite.154 By contrast, Odysseus is represented as performinga form of song that resembles the epic form of the Homeric Odysseyitself.

Here are ten relevant points, presented here in the form of anoutline:

(1) Each of the three songs of Demodocus in Rhapsody 8 of theOdyssey starts with a new hymnic prooimion, and each one of thesethree new prooimia is followed by a new hymnic consequent.155

(2) In the case of the first and the third songs, the hymnic con-sequent is epic poetry about the Trojan War. In the case of the secondsong, the hymnic consequent is a choral song and dance that makes amimesis of the love affair of Ares and Aphrodite (370–80; supple-mented by 262–5); in the overall narration, the performance of choralsong and dance is preceded by an embedded narration of this loveaffair, performed by Demodocus and quoted by the rhapsodic me-dium of the Homeric Odyssey (266–366).156

(3) Just as the Homeric Hymns have hymnic prooimia and allowfor metabasis to follow, so also the third song of Demodocus hasa hymnic prooimion followed by a metabasis, which is performedby Demodocus after the disguised Odysseus challenges him tometabainein ‘move ahead and shift forward’ to the story of the

153 HPC I}223. On this passage as evidence for the performance of the Hymns, seeClay in this volume (pp. 249–50).

154 HPC I}}188–223.155 HPC I}}242.156 HPC I}}207–8. In effect, the rhapsodic medium is making a mimesis of the

mimesis made by the choral medium.

330 Gregory Nagy

Wooden Horse in the epic narration of the Trojan War (492��ÅŁØ).157

(4) Although the metabasis as successfully executed in the thirdsong of Demodocus may be typical of the general epic form of theCycle, it is antithetical to the specific epic form of the Homeric Iliadand Odyssey, which was shaped by what I have already described asthe Panathenaic Regulation of rhapsodic relay.158 This regulation, aswe have seen, requires each successive performer of Homeric poetryto continue the epic performance at exactly the point where theanterior performance left off. And so, unlike the Homeric Hymnsand unlike the third song of Demodocus, the Homeric Iliad andOdyssey allow for no metabasis—except for the metabasis performedby Demodocus himself as a rival of Odysseus.159

(5) Just as the Homeric Iliad and Odyssey allow for no metabasis,these two epics allow for no hymnic prooimia either. Such a prooimionis missing at the start of both the Iliad and theOdyssey.160 And it is alsomissing at the start of Rhapsody 9 of the Odyssey, which is the start ofan embedded rhapsodic performance by Odysseus himself. To becontrasted with this regulated rhapsodic performance of Odysseus,starting with Rhapsody 9 and extending all the way to the end ofRhapsody 12, are the three unregulated performances of Demodocusin Rhapsody 8, each of which is preceded by a distinctly hymnicprooimion.161

(6) The god invoked in each one of the three hymnic prooimiaperformed by Demodocus in Rhapsody 8 turns out to be one andthe same god, but the identity of this god is revealed only in Rhapsody13, after both Demodocus and Odysseus have finished their rivalperformances. Retrospectively, we can see that both these rival

157 HPC I}}225–6. This poetic act of moving ahead makes it possible to shift thenarrative forward from what had been narrated in the first song of Demodocus, whichhad started his ongoing epic narration of the Trojan War, and thus the third song canjump over what had been narrated in the second song. Such a shifting form of epicnarration, as embedded in Od. 8, is analogous to the forms of epic that could havebeen introduced by the Homeric Hymns. And such a form of epic is typical of the epicCycle. See HC 2}307.

158 HC 2}}297, 304.159 HPC I}240.160 There survives, however, a variant Iliadic prooimion, addressed to the Muses

and Apollo: see Wilamowitz-Moellendorff (1916b), 32. 16–20; I analyse this variant inHPC I}260.

161 HPC I}240.


performances took place in the context of an ongoing festival, whichis stylized as a dais ‘feast’ in Rhapsody 8 (429; also at 61).162

(7) The word agōn ‘competition’ in Rhapsody 8 (259, 260, 380)points to the festivities that have been ongoing at this festival eversince it started with an animal sacrifice (59–61), which inauguratesthe dais ‘feast’ (61). Here I recall the Hymn to Apollo, where we haveseen this same word agōn ‘competition’ used with reference to therecurrent festival of Apollo on the island of Delos (150). In theOdyssey, the feasting and the competition that start in Rhapsody 8continue all the way through the narrative performed by Odysseus inRhapsodies 9 through 12, lasting all night. Then, when dawn finallyarrives in Rhapsody 13 (23), there is another animal sacrifice (24),and this time the divine recipient of the sacrifice is mentioned byname: he is Zeus himself (25). This god, as I argue, is the transcendenthymnic subject of the Homēridai.163

(8) In the course of time, the prooimion of the Homēridai getsdetached from its epic consequent, which can be either the HomericIliad or the Homeric Odyssey, and these two epics can then developtheir own abbreviated prooimia, which are non-hymnic—that is,without the naming of a hymnic subject.164

(9) In the case of the Homeric Iliad and Odyssey, the ancient recep-tion of these epics featuring prooimia that are non-hymnic goes back tothe age ofCallimachus. In that age, the old poetic form of the prooimionas represented by the Homeric Hymns was rethought as a new genre,separable from the old poetic form of the epic consequent.165

(10) The Hymns of Callimachus, which preclude an epic conse-quent, are the clearest examples of this new genre, which is stillknown to this day as a hymn.166

WAS THE HYMN A GENRE IN ITS EARLIESTATTESTED PHASES?

I argue that the answer to this question is no. As we have seen in theearliest textual evidence, the humnos in the sense of ‘hymn’ can take

162 HPC I}298.163 Again, HPC I}298.164 HPC I}}284–9.165 HC 2}}118–22.166 HPC I}287. See also Petrovic (2011b forthcoming).

332 Gregory Nagy

on a variety of different forms and functions, both choral and rhap-sodic.167 So the hymn transcends the concept of genre—at least, untilthe era of Callimachus. At best we can apply to the hymn the largerconcept of a supergenre, just as Richard Martin applies this largerconcept to epic.168

167 I note with interest that the stylized hymn to Apollo as described in Il. 1. 472–4figures itself as a paean (�ÆØ��Æ, 473); see Bundy (1972), 74.

168 Martin (2005), 17. Also Martin (1997). For a different view of the Hymns as agenre, see Clay in this volume (Ch. 11).


14

The Homeric Hymns as Poetic Offerings

Musical and Ritual Relationships with the Gods

Claude Calame

˜�Å�æ’ M�Œ�� c� Ł�e� ¼æå�’ I��Ø�,ÆP�c� ŒÆd Œ��æÅ� ��æØŒÆººÆ —�æ��ç��ØÆ�.åÆEæ�, Ł��, ŒÆd �� ı ��ºØ�, ¼æå� �’ I�Ø�B .

Of Demeter the lovely-haired, the august goddess first I sing,of her and her daughter beautiful Persephone. I salute you,goddess: keep this city safe, and give my song its beginning.

(Hy. 13)1

There is in the corpus of Homeric Hymns no shorter compositionthan this poem addressed to Demeter. Nevertheless, despite theHymn’s extreme conciseness, these three verses not only have thetripartite structure which is characteristic of every Homeric Hymn,but above all they fulfil the essential pragmatic aspects of the Hymns.On the one hand the goddess hymned is evoked in the third person bythe ‘I’ of the persona cantans; Demeter is then defined very briefly byher engendering of Persephone, before she is invoked directly by thespeaker and narrator ‘I’, who addresses a special request to her:evocatio, epica laus, preces.2 The central descriptive and narrative

1 English translations of the Hymns are drawn from WL.2 The term evocatio is used instead of invocatio because in the Homeric Hymns the

god is evoked in the third person (and not invoked in the second); epica laus is usedbecause the central part of the Hymns, which is narrative and descriptive, correspondsto the praise of a divinity; preces is used because the final prayer is adressed in thesecond person to the divinity.

section generally introduced by the so-called hymnic relative, whichin the longer Homeric Hymns leads to narrative accounts of dozens ofverses, is here reduced to just one line; overlapping with the evocatio,the epica laus introduced by ŒÆ� mentions only the existence ofPersephone as daughter of Demeter.3 From an enunciative point ofview, the final prayer is introduced by the expression which concludesnumerous addresses to the gods: åÆEæ� (‘rejoice’). This resonant åÆEæ�(‘rejoice’) corresponds not to a simple form of salutation, but refers tothe rejoicing experienced by the goddess upon hearing the sung praisewhich has just been addressed to her.4 Within the principle of do utdes, the wish is accompanied by a prayer concerning the well-being ofthe city. From then on, with the reciprocity which the prayer impliesfor the relationship with the god, the poem is presented as a musicaloffering to the divinity: she rejoices in it and, in exchange, ensures thesafety of the city.

But whoever speaks the utterance also speaks something prag-matic. On the one hand, the poetic ‘I’ of the persona loquens engagesin a double act of speaking: first in undertaking, as ‘I’, the beginning ofthe song (¼æå�’ I��Ø�, ‘I begin to sing’), thereby choosing the mostperformative of the three forms of opening known in Homericpoetry;5 then in asking the goddess to begin the song itself (¼æå� �’I�Ø�B , ‘begin the song’) in a request that responds to the wish,addressed to the divinity, that she rejoice in the song. This secondpetition also confirms the function of the Homeric Hymns as proemsto aedic songs and rhapsodic recitations:6 first the evocation and thepraise of the divinity with the final petition addressed to her, then thenarrative song itself placed under the inspiration and the authority ofthe divinity celebrated, by way of hymn as prelude to a longer poem.On the other hand, the request for safeguarding the city is accom-panied by an act of verbal deixis, which consists of �� ı ��ºØ�

3 Cf. Calame (2005), 19–22, following several studies of structure, among which seeesp. Bremer (1981), 195–203.

4 Cf. Calame (2005), 26–9, with further bibliography on the meaning of åÆ�æ�Ø� inn. 57; cf. Nagy in this volume (pp. 327–9). See also below n. 39 on theDelian section ofthe Homeric Hymn to Apollo.

5 Cf. Calame (2000), 59–69. The two other forms are: ‘Sing, Muse, of Artemis . . . ’,and ‘Muse, tell me of the deeds of Aphrodite . . . ’. The connection between the form ofhymns and prayers is explored notably by Aubriot-Sévin (1992), 172–93.

6 On I�Ø�� as ‘creative singer’ and ÞÆłøfi �� rather as ‘reciter’ see, among others,Nagy (1990a), 19–26, (2003), 41–3.

The Homeric Hymns as Poetic Offerings 335

‘save this city’. The self-referential act of singing (aedic-rhapsodic) islocated in both time and space: it takes place hic et nunc. Whatever itslength, the Homeric Hymn, in its role as a hymnic song addressed to adivinity as a musical offering, has the function of introducing therhapsodic recitation into the series of ritual acts honouring thisdivinity. The hymn therefore makes the performance of the rhapsodicsong itself a ritual act and an activity of cult.7 By its function as aproem on a cultic level, the Homeric Hymn in a way transforms thenarrative song into melic song. The narrative poem is sometimesintroduced by a performative formula: �� ’ Kªg Iæ��

��Æ��ÆØ ¼ºº�� K o�� (‘having begun with you, I will pass onto another song’), and I will do it now.8 It is well-known that thedifferent forms of melos are full of these performative futures, refer-ring in a self-referential manner to the act of singing.9

It is observable that, by different enunciative means of the categoryof speech acts, the Homeric Hymns try to establish a practical rela-tionship with the divinity, whom the aoidos-rhapsode is praising. It isduring the performance, sung and presented as such, that the poem istransformed into a musical offering: as a prelude, this hymnic poemis destined to make the aedic-rhapsodic recitation that it introduces acultic act; as such it is inserted among the different ritual acts hon-ouring a divinity on the occasion of his or her celebration in a city or aparticular sanctuary. It is from this perspective of ritualized musicalperformance that, in what follows, I will examine the poetic meansemployed in the Homeric Hymns for establishing and maintaining arelationship of a cultic nature with the divinity hymned: naming anddefinition of the divinity in the introductory verses of the evocatio;

7 On the Homeric Hymns as rhapsodic proems, according to the sense of the termgiven by Th. 3. 104, see Nagy in this volume (pp. 322–5), as well as my references andnotes in Calame (2005), 193 n. 5; for some reservations about this view for the longHymns, see in this volume Clay (pp. 237–40) and cf. Introduction (pp. 17–19). For themovement from sung aedic creation to rhapsodic repetition or recitation, see Nagy(1990a), 20–5.

8 This formula concludes Aphr. (293) and Hy. 9. 9, thus in long Hymns as well asshort. See also the common closing formula ÆP�aæ Kªg ŒÆd ��E� ŒÆd ¼ººÅ ��’I�Ø�B (‘but I will remember you and another song’): Apoll. 546, Hy. 6. 21, etc.Cf. below n. 51.

9 On the forms of the present and the ‘performative future’, which in general makemelic poems speech acts and consequently acts of singing, see Calame (2005), 195 n. 25with further bibliography, and D’Alessio (2004), 274–84. For the narrative category of‘melic’ instead of modern ‘lyric’ poetry, see Calame (2008a), 85–106.

336 Claude Calame

musical scenes recounted in the descriptive and narrative section ofthe epica laus; and the ‘performative’ enunciation in the concludingsection of the preces.

1 . AEDIC AND RHAPSODIC DESIGNATIONOF THE DIVINITY (EVOCATIONES)

% EæB� I��ø ıºº��Ø�� æª�Øç��Å�,ıºº��Å ��Æ ŒÆd �æŒÆ��Å ��ºı�º�ı,¼ªª�º�� IŁÆ��ø� KæØ��Ø��, n� �Œ� )ÆEÆ@�ºÆ�� Łıª��Åæ ˜Øe K� çØº��Å�Ø Øª�E�Æ

ÆN��Å·

Of Hermes I sing, the Cyllenian Argus-slayer, the lord of Cyllene andArcadia rich in flocks, the immortals coursing messenger, whom mod-est Maia bore, the daughter of Atlas, in shared intimacy with Zeus.

(Hy. 18. 1–5)

Every Homeric Hymn begins with the naming of the god about to beevoked in the initial speech act, here represented by I��ø (‘I sing’:aoidos then rhapsode). One will remember that for Herodotus inparticular, ‘to name’ (O��Ç�Ø�) a god is not simply to give him aname, but also to identify him; it is to confer qualities upon him,which in a polytheistic system are placed in relation to the functionspractised in a specific field of action. Indeed, from the perspectiveheld by Plato’s Cratylus, the name of a divinity cannot be distin-guished from his substance; in the absence of a clear distinctionbetween signifier and signified, the name of a god reflects hisessence—Eros is for instance connected to the verb Þ�E� (‘to flow’)because desire flows into the soul from outside (420a–b). And, froman aetiological perspective, the identification of a divinity by a propername is accompanied in general by the institution of sacrificialpractices, if not by the foundation of a sanctuary. In short, thisprocess of substantial naming of the gods corresponds to that ofwhich many Homeric Hymns are the vehicle: the narrative part ofthe hymnic poem often consists in defining and awarding to the godbeing celebrated his timē (�Ø�). The word timē refers to the portionwhich rightfully belongs to a particular god, as opposed to the otherdivinities of a pantheon that in Greece is always moving and changing


in historic time and civic space.10 Thus, well before the proposals ofGeorges Dumézil for the study of gods in a polytheistic system, theGreek gods were defined, in an indigenous way, by a name and aprofile which reflect functions and modes of operation for a particulardomain and aspect of the social life of mortals.11

But such naming is regularly accompanied by qualifications withthe help of terms which, already in Pausanias, are identified asepiclēseis: simple surnames which, in the case of the gods, could eitherhave been created by the poets, or correspond to local names or toepiclēseis shared by all. Without being at all systematic, the distinctionproposed by Pausanias concerning Poseidon venerated in Arcadiasuggests a differentiation of operational order between theonymsproper, poetic epithets, local cultic epiclēseis, and pan-Hellenic epi-clēseis, in categories which often overlap.12 Because of their narrativecharacter, the Homeric Hymns mostly contain poetic qualifications:for example, ¼ªª�º� IŁÆ��ø� KæØ��Ø� (‘swift messenger of theimmortals’) in Hymn 18 to Hermes is a qualification followed im-mediately by the hymnic relative, which introduces the narrative partof the poem; this is also the case in the opening of the long Hymn toHermes, through the use of the same formulaic verse (3). Otherwise, if�æª�Øç��Å (‘Argus-slayer’) can generally be considered an alterna-tive theonym toHermes in Homeric poetry, the initial qualification ofthe god hymned and described often refers to his place of origin or tohis space of privileged action:13 Hermes of Cyllene rules over MountCyllene in Arcadia, where he is born (also in Herm. 1–2).

This geographical parameter is so important for the relationshipwhich the naming and the qualification create with the god that it can

10 Cf. Hdt. 2. 49–50, with my comments on the meaning of the naming of a god inCalame (forthcoming); concerning the process of awarding timai, which takes placein the Homeric Hymns, see notably Clay (1989), 15–16, passim, and in this volume(Ch. 11).

11 See my clarification of this correspondence between the indigenous perspectiveand the scholarly point of view on the organization of the Greek polytheistic system inCalame (2006), 210–25 with full references, particularly to Dumézil (1974), 186, 239.

12 Paus. 7. 21. 7: ‘At the harbour is a temple of Poseidon with a standing image ofstone. Besides the names (O��Æ�Æ) given by poets to Poseidon to adorn their verses,and in addition to his local names (K�Øå�æØÆ), all men give him the followingsurnames (K�ØŒº��Ø )—Marine, Giver of Safety, God of Horses’ (trans. Jones 1933).

13 This observation applies generally to the hymnic form: cf. Furley and Bremer(2001), i. 52–6. For �æª�Øç��Å as a ‘synonym’ of % EæB , see for example Il. 2. 103–4and Od. 1. 84, as well as in the Hymns themselves Hy. 29. 7 (Hestia); cf. Jaillard (2007),211.

338 Claude Calame

be considered to constitute the leading thread of the narrative. Start-ing from the enunciative and performative question of how oneshould hymn a god so well-celebrated (�H ��æ �’ ��ø, ��ø �hı�� K��Æ—‘How shall I hymn you, fit subject as you are in everyrespect’, 19) the narrative of the Hymn to Apollo goes through all ofthe places which could accept the god’s birth, before situating himfirst in Delos (after a long list of cities on the Greek islands and thecoast of the Aegean), then in Delphi (at the end of a new geographicalvoyage, which this time focuses on mainland Greece).14 The aedicnarration thus becomes the aetiological ‘myth’ which explains themusical honours given to the founding god in his sanctuary on Delosand the power of his oracular voice at the site of Delphi.

To this spatial locating, with its at times aetiological function, isgenerally added an indication of genealogy; it is often by this temporallocating that the transition is effected, in general by way of a hymnicrelative, from the evocatio to the epica laus: the passage from the‘discourse’ (marked by the use of the first person in the present or theperformative future) to the ‘narrative’ (marked by the third personand aorist, to use the linguistic categories proposed by Émile Benve-niste, who speaks of ‘discours’ on one hand, and of ‘histoire’ or ‘récit’on the other).15 It is on Mount Cyllene that Maia gives birth toHermes after her union with Zeus, in both Hy. 18 and the longHymn to Hermes. Likewise, for example, it is in the foothills ofMount Taygetus that Leda gives birth to the Dioscuri, presentedat the same time as Tyndaridai and sons of Zeus (Hy. 17. 1–6; cf.33. 1–5); Coronis bears Asclepius, the son of Apollo, on the Dotianplain in Thessaly (Hy. 16. 1–3); and it is in the valleys of Mount Nysa(whatever its precise location) that the nymphs rear the infant Dio-nysus, born from the union of Zeus and Semele (Hy. 26. 1–5).16

Following his enunciative naming, therefore, accompanied by somequalifications more poetic than cultic, the divinity is readily providedin the initial section of the evocatio with a geographic-genealogicalidentity.

14 Apoll. 19; this rhetorical question with its future performative form is repeated atthe outset of the Pythian section (207); cf. Miller (1986), 20–6, Nagy (2009b), 17–19.See in this volume Chappell (p. 64) on the geography of the Delian and Pythiansections.

15 Cf. Calame (2005), 22–4.16 Cf. Furley and Bremer (2001), i. 54–5 on the importance of the god’s location in

various hymnic forms.


Now, from the point of view of syntax, the enumeration of thepoetic qualities and the possible epiclēseis attributed to the hymnedgod is characterized by parataxis in asyndeton. Uncommon in Greek,this device is connected to the expression of strong emotion. It marksin particular the appeals to the divinity which initiate and punctuatecultic hymns, such as the so-called paean of Philodamos:17

[ . . . . . . .] ˜ØŁ�æÆ��, B�Œå’�[hØ�, 'ÆFæ� Œ]Ø�:�:�:åÆE-�Æ, Bæ�Ø’, MæØ�Æ[E ƒŒ�F�ÆE��’] ƒ�æÆE K� uæÆØ (1–4)

Come here, Lord Dithyrambos, Bakchos, god of jubilation, Bull, with acrown of ivy in your hair, Roarer, oh come in this holy season of spring.(trans. Furley-Bremer, 2001)

Then in another metrical rhythm:

¯P�E t Ne [B�Œå’, t Nb —ÆØ�]� (5)

Euhoi, o io Bakchos, o ie Paian

The initial address to the god and the refrain repeated after everystrophe of this hymnic song do not contain any connective particles.Through invocatio (in the second person) and not simply evocatio (inthe third person), the direct address to the divinity aims to provoke,by the musical performance of the song, his epiphany. The device ispushed to the extreme in the Orphic Hymns, the majority of whichenumerate the qualities of the hymned god in asyndetic parataxis.It corresponds at the same time to the invocatory offering to thehymned divinity which the Orphic Hymn represents and to thetheological desire, belonging to the Orphic movement, to superim-pose the divinities of the traditional pantheon in order to merge themin unity with Zeus.18 In the Homeric Hymns, the evocation in asyn-deton of the name of the god and his qualities is combined with theact of singing represented by the self-referential forms in the first

17 Text and translation from Furley and Bremer (2001), no. 2. 5; see their excellentcommentary at ii. 58–60 and for parataxis in the definition of the god invoked orevoked, cf. i. 53–4.

18 Cf. Hopman-Govers (2001), 37–49, Morand (2001), 59–76. It is in this respectthatHy. 8 to Ares, which is composed of an accumulation of epithets in asyndeton andwhich assumes the hymnic form of a prayer, can be considered to be of Orphicinspiration; cf. Càssola 297–9.

340 Claude Calame

person; the asyndetic parataxis of the epithets and the verbal perform-ative forms give to the inaugural naming of the divinity its pragmaticdimension. This naming does not necessarily aim at the epiphany ofthe hymned divinity, but in any case at his praise; that is, it aims tohonour him with a musical offering, which will be continued inthe ritualized aedic-rhapsodic recitation that the Homeric Hymnsintroduce as a poetic genre.19

2. MUSICAL RELATIONSHIPS BETWEENMEN AND GODS (EPICAE LAUDES)

In the long Hymns, as in the short, the narrative section of the aedic-rhapsodic composition generally further specifies the genealogicaland geographical identity of the god and places the god’s identity inrelation to humans. If one takes into consideration the most devel-oped narratives, in the Hymn to Apollo the god Apollo is settled fromhis birth in Delos itself, where the Ionians come together to pay himmusical homage (to which I will return later); then the narrativeplaces the young god at Delphi, where he founds his oracle, whichreveals the plans of the gods, in a temple ‘honoured by many men’(��ºº�E�Ø ��Ø�� I�Łæ��Ø�Ø�, 479). At the end of the Hymn toDemeter, Demeter establishes her daughter Persephone in cult atEleusis, between Olympus and Hades, with the aid of Zeus and theruling family, and she founds the different rites and mysteries whichwill ensure prosperity for men in the present life and the afterlife.If theHymn to Hermes settles the newborn god in the heart of Arcadiain the valleys of Mount Cyllene, the end of the poem makes him theguide to the underworld and the messenger of mortals, whom he doesnot hesitate to deceive. Finally, the Hymn to Aphrodite places thegoddess between Cyprus and Cythera (as elsewhere do the two shortHymns to the goddess of seduction, 6 and 10), and ends with theprobable separation between mortals and immortals.20 Although it isnot the case for Apollo at Delphi, the place hymned usually corres-ponds to the birthplace of the divinity who, in one way or another,

19 Cf. above n. 7 and, for hymns in general, Aubriot-Sévin (1992), 182–93.20 On the question of the implied separation of men from the gods in Aphr.,

see Faulkner (2008), 8–10.


helps to organize the relations between gods and mortals, betweenOlympus and Hades. From the geographical point of view, nothing issaid concerning the place of performance of the poem, except forgeneric designations such as ‘this city’ (�� ºØ�).21 This lack ofspatial determination in the act of verbal deixis leaves open, on thepragmatic level, the re-performance of the hymnic poem in differentlocations and ritual circumstances of enunciation:22 it also ensures forthem the possibility of Panhellenic diffusion, just as for the Homericpoems introduced by the hymnic prelude.

As the promise made to Persephone by Hades in the Hymn toDemeter shows in particular, the notion of timē implies a relationshipbetween gods and mortals. The daughter of Demeter, in her role asthe young bride of the god of the underworld, reigns in the afterlifeover all mortals, but she also enjoys, amongst the immortals, the‘greatest honours’ (�Øa �ª��Æ , 366); whoever should committhe injustice of not consenting to the offering, which the accomplish-ment of the sacred rites constitutes, will inevitably be punished.23

Involving the definition of a mode of action and the delimitation of afield of intervention, the ‘honours’ (�ØÆ�) which the long HomericHymns establish by way of the aetiological narrative engage thedivinity concerned in the reciprocal relationship between gods andmortals mentioned above.

Taking into consideration the importance of descriptive and nar-rative expansion in the hymnic form,24 I will now focus my attentionon the role of the Muses’ arts in the attribution of an ‘honour’ (timē)to a god, and by consequence the attribution to him of a particularquality, ability, or function. Indeed, music, understood in an indigen-ous way as a combination of poetic singing, choreography, andinstrumental melody, plays a central role in the establishment andmaintenance of ritual communication between mortal men and the

21 Cf. Hy. 13. 3. Hy. 24. 4 to Hestia asks the goddess of the hearth to intervene in‘this house’ (�æå�� I�a �rŒ��). This type of verbal deixis, frequent in melic poetryand in tragedy with its strong pragmatic dimension, is treated in Calame (2004a),which should be read together with the other studies on this subject collected in thesame special volume of Arethusa 37 (2004).

22 On the re-performance of the Homeric Hymns, see in this volume Clay (p. 235).23 Dem. 345–69; see the commentary of Richardson (1974), 269–75 and in this

volume (pp. 50–3) concerning the allusion which these verses contain to the institu-tion of the mystery rites at Eleusis. Also Clay (1989), 251–4.

24 Following the excellent suggestion formulated and developed by Vamvouri-Ruffy (2004), 27–36.

342 Claude Calame

immortal gods. The mise-en-scène of the different arts of the Muses inthe narrative section of the Homeric Hymns takes on, in relation tothe function of the poem itself as a musical offering, a reflexive aspectthat requires elucidation. Of concern in this respect are not only thedivinities involved in the field of music (in the Greek sense of theterm) by their individual qualities and abilities, but also other deitiesof the Greek polytheistic system.

Thus in Hymn 30 dedicated to Gaia, the Homeric aoidos-rhapsodesubstitutes in the narrative, in place of a biographical episode of thegoddess, a list of the benefits that the protection of Mother Earthaccords to all beings: prosperous children, abundant harvests, flour-ishing flocks, a house filled with riches, a city of beautiful womenmarked by justice and prosperity—in one word life (�� , 6) formortal men. For young men, as well as for young women who per-form choral dances in flowery fields, this means good cheer(�Pçæ��Å, 13–15), which is also that of the gods. Such are thehonours (emphasized by the repeated use of the verb �ØA� to enclosethe list of honours in lines 7–16 in ring composition) the augustgoddess provides to mortals in exchange for the present song, whichis offered to her in the final section of the preces. The joining of thisfinal prayer of the form do ut des with the central part of the Hymnand the reciprocal relationship between the immortal divinity andmortal men are both underlined by the exceptional use of the secondperson in the epica laus, which is still, as is usual, introduced by ahymnic relative (m çæ��Ø . . . , 2). By the immediate passage from ‘shewho’ to ‘you’, and by the introduction of a form of beatification(ÆŒÆæØ�� : ‘happy is he whom you are going to honour with abeneficent heart’, n �’ Zº�Ø� , ‹� Œ� �f ŁıfiH j �æ�çæø� �Ø��fi Å , 7–8),the relationship between immortals and mortals is situated on theenunciative plane; it coincides with the relationship that the aoidos-rhapsode establishes with the goddess celebrated by his poem, hic etnunc, and the choral rejoicing evoked in the city that has benefitedfrom the prosperity granted by Gaia.25

Musical performance is also staged on at least two occasions in thelong narrative that forms the central part of the Hymn to Aphrodite.In the first instance, when the goddess approaches Anchises tendinghis cattle on Mount Ida, the young, divinely beautiful hero devotes

25 The final part of Dem. (480–2 and 486–9) is marked by a double ÆŒÆæØ�� ;cf. Richardson (1974), 310–14 and 316–21, as well as Calame (2008a), 68–71.


himself to music; like Achilles in the Iliad, he plays his lyre by himself,apart from his fellow herdsmen (75–80). In a sort of parallelism,the goddess presents herself to the young shepherd in disguise as an‘untamed young girl’ (�ÆæŁ�� I��Å , 82). She pretends to bemortal, seized by Hermes while she was dancing in a chorus ofArtemis with her companions (comprised of both nymphs andyoung girls); she is like the young maidens seized by the king ofSparta Aristomenes while dancing in a chorus for Artemis at Caryae(117–25).26 To these two scenes which follow the paradigm of musi-cal activity reserved for unmarried heroes, young men on the onehand, young women on the other, one can also add the description ofthe rearing of Aeneas, the child born from the union of an immortalgoddess and a young mortal hero; his destiny is to be raised onMountIda by the nymphs, who, being neither mortals nor immortals, areassociated with the beautiful choral dances of the gods (256–63).27 Toparticipate in choral celebration is therefore for mortal men to sharein the life of the gods.

Now, if for Aphrodite choral activity is only a mask which concealsher status as a young woman ready to unite in love even with a mortalman, the same certainly cannot be said of Artemis. The twenty-two-line Hymn 27 dedicated to her confirms the musical gifts to which theHymn to Aphrodite makes allusion. The central section of the Hymn,which describes in the third person but in the present tense theparticular abilities of the virgin goddess, presents two of the privilegedactivities of the sister of Apollo, one after the other in a double scene:after indulging in the hunt for savage beasts, which howl in the shadyforest on the peaks of mountains, the archer goddess unstrings herbow in order to go to Delphi and take her place next to her brotherApollo (4–14); assuming once again her graceful habits, she takes thelead in the beautiful chorus formed by the Muses and the Charites(15–20). It is in taking on this role as chorus-leader that the goddessinitiates the choral poems which sing of how Leto gave birth to hertwo children, Artemis and Apollo. In this narrative section which,through the use of the present tense, makes the double identity of

26 Cf. Calame (2001), 149–53, and Faulkner (2008), 163–5 concerning the status ofthe parthenos (compared to that of the woman) whose identity Aphrodite adopts, andthe status attached to choral activity; also 192–5 on the the Homeric case of Hermes’rape of Polymele while she is dancing in a choral group (Il. 16. 179–83).

27 On the ambiguous status of the nymphs, see the commentary of Faulkner(2008), 185–6.

344 Claude Calame

Artemis as both huntress and choral-leader a permanent quality, thegenealogy of the goddess is projected from the Homeric Hymn uponthe choral song of the Muses and the Graces; led by Artemis, thischoral song assumes a reflexive character.28

Artemis is therefore in the category of gods who are active in thefield of the Muses’ arts. The same is naturally true of Apollo, asdemonstrated in the five-line Hymn 21 dedicated to him.29 Thispoem, which is of an atypical enunciative form, has song as itstheme from start to finish:

+�E��, �b b� ŒÆd Œ�Œ�� e ��æ�ªø� º�ª’ I��ØZåŁfi Å K�ØŁæfi��Œø� ��Æe� ��æÆ �Ø��Æ

—Å��Ø�� b �’ I�Ø�e �åø� ç�æØªªÆ º�ª�ØÆ��ı��c �æH�� ŒÆd o��Æ�� ÆNb� I��Ø.ŒÆd �f b� �o�ø åÆEæ�, ¼�Æ�· ¥ºÆÆØ � �’ I�Ø�Bfi . (Hy. 21)

Phoibos, of you the swan too sings in clear tone from its wings as italights on the banks beside the eddying river Peneios; and of you thebard with his clear-toned lyre and sweet verse ever sings in first placeand last. So I salute you, lord, and seek your favour with my singing.

In an evocatio, which remarkably assumes the form of an invoca-tion through the direct address to Phoebus, the persona cantans turnsover his voice, altogether exceptionally, to the beating wings of theswan; the bird is charged with hymning the god (I��Ø), in a melo-dious voice, on the river Peneius in Thessaly (1–3). Then in a secondaddress in the second person (�) and by way of an epica laus, thespeaker delegates his voice to an anonymous singer (I�Ø�� ) with asweet voice, who hymns the god in the present tense (3–4). Finally, inthe verse of the preces, the meaning of the traditional request to thegod to rejoice in the song (�o�ø åÆEæ�) is made explicit: ¥ºÆÆØ � �’I�Ø�Bfi (‘I seek your favour with my singing’). The song which spansthe whole poem (and which the poem itself is) becomes a sacrificialoffering which, within the principle of do ut des, is intended to win thefavours of the hymned god.

28 On this double function of Artemis, cf. Calame (2001), 52–3, 91–101. See alsoAphr. 19 with the relevant commentary by Faulkner (2008), 95–7, on the relationshipof Artemis with music and choral dance.

29 Càssola 375 puts forward the hypothesis that these five verses are the finalsection (preces) of a longer hymn, the rest of which is now lost; also 578 for thequalities which the Greeks ascribed to the song of the swan.


As far as Hymns full of musical activity are concerned, there is alsonaturally the short seven-line Hymn 25 addressed at the same time toApollo, the Muses, and Zeus:

)�ı��ø� ¼æåøÆØ ��ººø�� ˜Ø� ��·KŒ ªaæ )�ı��ø� ŒÆd �ŒÅ��º�ı ��ººø��

¼��æ� I�Ø��d �Æ�Ø� K�d åŁ��d ŒÆd ŒØŁÆæØ��Æ� ,KŒ �b ˜Øe �Æ�ØºB� · n �’ Zº�Ø� , ‹� �Ø�Æ )�F�ÆØç�ºø��ÆØ· ªºıŒ�æ� �ƒ I�e ��Æ�� Þ�Ø ÆP��.åÆ�æ��, �Œ�Æ ˜Ø� , ŒÆd Kc� �Ø��Æ�’ I�Ø��·

ÆP�aæ Kªg� �ø� �� ŒÆd ¼ººÅ ��’ I�Ø�B . (Hy. 25)

From the Muses let me begin, and Apollo and Zeus. For from the Musesand far-shooting Apollo men are singers and lyre-players on earth, andfrom Zeus they are kings. He is fortunate whom the Muses love: thevoice flows sweet from his lips. I salute you, children of Zeus; honourmy singing. And I will take heed both for you and for other singing.

In a more traditional form, this Hymn is from the outset under-taken by the ‘I’ of the aoidos-rhapsode, who begins with a briefformula of evocatio and goes on to tell (by way of a laus or parsepica) of the dependence of singers and lyre-players upon the Musesand Apollo, in contrast with kings who depend upon Zeus (with asubtle genealogical allusion). From this point, the principle of do utdes of the concluding prayer can only relate to song: in exchange forthe musical celebration offered to the children of Zeus, the poetic ‘I’asks these divinities to honour his own song and finishes with thecommon formula of introduction to rhapsodic song: ÆP�aæ Kªg��ø� �� ŒÆd ¼ººÅ ��’ I�Ø�B (‘And I will take heed both foryou and for other singing’, 7); with a probable etymological playon the name of the Muses, which is repeated three times in six verses(1, 2, 4), ��ÆØ recalls the name of their mother Mnemosyne, theincarnation of poetic memory.30 The thematic coincidence betweenthe three constituent parts of the Hymn is underlined, from theenunciative point of view, not only by the use of the present in thenarrative section, but above all by the introduction of a formula ofbeatification (ÆŒÆæØ�� ): n �’ Zº�Ø� , ‹� �Ø�Æ )�F�ÆØ j ç�ºø��ÆØªºıŒ�æ� �ƒ I�e ��Æ�� Þ�Ø ÆP�� (‘He is happy whom the Museslove; from his mouth flows a sweet voice’, 4–5). Through the

30 Hy. 25 (Muses, Apollo, Zeus); on the possible etymological relationship between)�F�Æ and )�Å��Å, see the bibliography below in n. 52.

346 Claude Calame

enunciative formulation of these four verses, which appear identicallyin the prelude of the Theogony and without doubt have a formulaicquality, the ‘narrative’ level is confused with that of the ‘discourse’.31

The dependence of singers and lyre-players on the Muses and Apolloand the happiness they can gain from the love of the Muses in factbelong to the poetic ‘I’, the persona cantans. Especially when the‘argument’, of the narrative and descriptive part is focused on song,the aoidos-rhapsode becomes, through the song as musical offering,the intermediary between men and gods; in the final prayer hebecomes the direct protagonist of the hymnic relationship betweenmortals and immortals.

But, in a polytheistic system, several divinities can intervene in agiven field, each according to his or her own qualities, abilities, andfunctions. As concerns the field of musical arts, the corpus of HomericHymns demonstrates this. In the shortest of the Hymns dedicated tohim (Hy. 26), Dionysus is straightaway described by the epithetKæ��æ�� (‘mighty roarer’, 1) which refers to the resounding clamourthat marks his appearance. Following this, the narrative of his rearingby the nymphs concludes with the mention of the ‘noise’ (�æ�� , 10)which fills the valleys of Mount Nysa in a revel; followed by thenymphs, who act as would a choral group, the young god so ‘oftenhymned’ (��º�ı�� , 7) runs through the forest crowned with ivy andlaurel like a chorus-leader.32 On the other hand, the Mother of godsand men (venerated in different forms in cults attested in many Greekcities of Asia Minor) is the object of the six-line Hymn 14; its centralpart is entirely focused on the musical signs indicating the presence ofthe goddess: the sounds of cymbals and tambourines, the bellow ofoboes, the howling of wolves and lions resounding in the valleysof mountains.33 Finally, it is no surprise to see Pan, in the epic sectionof the relatively long Hymn 19 dedicated to him, gliding through thelandscape of the highest peaks and returning from the hunt playing

31 Verses 2–5 correspond to Hes. Th. 94–7, without evincing a direct relationship:see the remarks of Càssola 401–2, 580, and on this Hesiodic passage Brillante (2009),61–4, Pucci (2007), 111–21. On the distinction between ‘narrative’ and ‘discourse’ andthe use one can make of it for the study of the pragmatics of Greek poetry, see Calame(2005), 1–16 and (2008b), 124–9.

32 Dionysus is also qualified as Kæ��æ�� at Hy. 7. 56; cf. in this volume Jaillard(p. 149).

33 In his commentary, Càssola 327–30, mentions the different places where theMother of the Gods was worshipped, under different toponymic epiclēseis.


his reed pipes; in his hands the pipes produce more melodious soundsthan the sweetest songs of the birds in spring (2–18). At the outset,the god ‘who loves rhythmical sound’ (çØº�Œæ��, 2), following ahymnic relative, dances among the nymphs ‘devoted to choral dance’(å�æ��Ł��Ø, 3; corrected by Schmidt to å�æ�ªÅŁ�Ø ‘dance-merry’)who summon the presence of the god (2–5). It is upon his returnfrom the snowy peaks that the mountain nymphs, with the beating oftheir feet and the sound of their clear voices, cause to resound a softmeadow of grass mixed with crocus and hyacinth; it is similar to theerotic meadow where Persephone is seized in the opening scene of theHymn to Demeter. The god joins these choral dances, only to becometheir leader (12–26).34 In addition, in the second part of this devel-oped descriptive-narrative section, it is the task of the gods them-selves, in unison with Olympus, to sing (��F�Ø, 27) of the love affairbetween Hermes of Cyllene and the nymph-daughter of Dryops, aswell as the birth of the deformed god who succeeds in bringing joy tothe immortals, including Dionysus.35

There remain the long Homeric Hymns composed in honour ofHermes, the inventor of the lyre for Apollo, and in honour of Apollohimself, who is installed musically in Delos and then in Delphi. Thesetwo Hymns receive detailed attention in individual chapters earlier inthis volume (Chappell Ch. 4 and Vergados Ch. 5), and I will hereconfine myself to looking at aspects of musical activity in the poems.First, I will consider the ring-composition of the narrative in theHymn to Hermes, which at the beginning of the Hymn places theinvention of the lyre on the same day as the birth of the god, and atthe end recounts at length the transmission of the enchanting instru-ment to Apollo. There are therefore two musical scenes, the first ofwhich, placed under the control of the immortals, allows us to bepresent at the creation of the ‘tortoise singer’ (åºı I�Ø�� , 25),which is the lyre. By killing the tortoise in order to appropriate itsshell, the infant god in a way immortalizes the tortoise by making ofhim a ‘symbol of great benefit’ (��º��, 30), which sings with abeautiful voice (I��Ø , 38). He then takes hold of the recently createdinstrument, with its seven strings in harmony, and strikes some chordswhich give rise to his beautiful song (��e ŒÆºe� ¼�Ø��, 54). Hermes,

34 The nature and function of erotic meadows are illustrated by Dem. 2–11; seeparallels and commentary in Calame (2002), 173–85.

35 On music in Hy. 19 cf. Thomas in this volume (Ch. 8).

348 Claude Calame

compared to adolescents who compete in poetry during festive cele-brations (55–6), subsequently sings of the love affair of Zeus and Maia,of which he is himself the product. Thus, after having delegated thesinging of Hermes’ genealogy to the Muse in the initial invocation(o��Ø, 1), and after having treated the same theme right at thebeginning of the narrative section introduced by the hymnic relative,the speaker-narrator finally turns over to the god himself the task ofsinging his divine and mountainous origin.36

As concerns the transmission of the marvellous instrument toApollo, the long scene preceding the concluding phase of the narra-tive (which is called in narratology the ‘phase of sanction’) making upthe central part of the Homeric Hymn to Hermes, is punctuated bymusical moments. To begin with, there is the melody which Hermesplays on the recently created lyre in order to appease Apollo andinspire in him the erotic sweet desire to which the music gives rise.Then it is the turn of the song, pronounced in a voice with a similarlyerotic effect (KæÆ�c çø��, 426), in which the god speaks of theprimordial times when each of the gods received his or her ownportion (�EæÆ�, 428); special mention is made in this passage ofMnemosyne, the mother of the Muses and the incarnation of aedicmemory. Then follow Apollo’s questions about the origin of thismusical technique (�� å�Å, �� F�Æ, 447), which inspires anoverpowering anxiety, as well as good cheer (�Pçæ��Å, 449), eroticdesire, and sleep; it is even more powerful than the choral songswhich the god sings with the Olympian Muses to the sounds of thedouble oboe (450–3). Finally, there is the scene of exchange in which,in return for his function as blessed guide, Hermes gives over toApollo his instrument and an art which the musician god will beable to employ at thriving banquets, in choral dances, and on theoccasion of ritual processions, providing widespread ‘good cheer’(�Pçæ��Å, 482).

Thus, with the lyre, which he takes hold of in order to lead the oxenstolen by Hermes back to Olympus with music, Apollo increases hissphere of musical ability, while Hermes is still the inventor of the lyre.In exchange, the divine guide of the immortals receives, by way oftimē (516), the privilege of promoting commercial transactionsamongst men; with the gift of the caduceus, a golden staff signifying

36 Cf. Jaillard (2007), 222–6, and Calame (2008b), 136–9, concerning the enun-ciative polyphony in Pindar.


wealth and prosperity, Apollo provides to Hermes the sign of thisparticular ability, which carries with it the power to fulfil the princi-ples ensuring just words and good acts. The caduceus is therefore a‘symbol’ (��º��, 527), which corresponds to the ‘symbol’ of thelyre at the beginning of the poem. And just as Apollo can either helpor hinder mortals with prophecy, a skill which he reserves for himself,Hermes will in the future be able to help or deceive the ‘tribes ofmortal men’ (çFºÆ Ł�Å�H� I�Łæ��ø�, 578).37 The inventions of thelyre and the caduceus serve to enrich verbal communication betweenmen and gods, but only through the interpretation and seductioninherent in the sung or recited word, which can be deceptive.

It is obviously through his powers of music and oracular voice thatApollo is associated with Delos and Delphi respectively, places towhich the long Hymn dedicated to the archer god pays homage.Whether it be a true rhapsodic composition, put together through a‘stitching’ of a Delian and a Pythian section, or a unified hymn, thedouble narrative section which forms the poem’s centre treats thebiography of the god from his birth on Delos and his installation in asanctuary marked by choral music to the foundation of his oracularsanctuary at Delphi.38 In a narrative in the aorist tense (althoughpunctuated by direct addresses in the second person, first to Apollo’smother and then to the god himself ), Apollo, recently born andnourished with nectar and ambrosia, himself defines the attributescorresponding to his future abilities: the lyre, the curved bow, and theoracle for revealing the will of Zeus to men. It is from this perspectivethat, on his own initiative, he settles himself inDelos and fromMountCynthus watches the gymnastic contests and the competitions ofmusic and dance organized in his honour by the people from neigh-bouring Ionia. The narrative movement from the aorist to the pre-sent, accompanied by a new direct address to Phoebus, makes thislong musical scene coincide with the nunc (and perhaps the hic) ofthe recitation of the Hymn. To this extent, the reciprocity which thenarrative establishes between the joy of the god witnessing the hon-ours rendered to him and the rejoicing of the Ionians in the

37 Cf. Jaillard (2007), 215–29, Càssola 535–41 on the objects of exchange, the termsof which are explored in Leduc (2001). For the performance of Herm. as a ‘re-enactment’ of the deeds of the god, see Vergados in this volume (pp. 103–4).

38 On the question of Apoll.’s unity, see in this volume Chappell (Ch. 4). The issueis also usefully treated by Càssola 97–102; on the circumstances of composition andperformance of the poem, such as it has come down to us, see Aloni (1989), 107–31.

350 Claude Calame

performance of these musical offerings anticipates the close connec-tion which at the end of his hymnic composition the aoidos-speakeror rhapsode-speaker establishes with the god; all the more so giventhat the musical activity and the pleasure of choral performance makethe assembled Ionians the equals of the gods, forever liberated fromcares and old age (151–2).39

It is then that the Delian women, the young servants of the far-shooting god, enter the scene. They sing a hymn which enchantshumans, in praise of Apollo, Artemis, and Leto, as well as the menand women of old (��ø�Ø�, 158; o�� I��ı�Ø�, 161). Musically,the young women of Delos in the service of Apollo sing what theaoidos-rhapsode sings in his hymnic composition.40 There follows adirect address to the Delian maidens, which reinforces this strongrelationship between immortal divinities and mortal men by way ofthe shared choral celebration. It also reinforces the enunciative rela-tionship between the speaker-‘I’ and the performer of the HomericHymn. In a significant switch from the third to the second person, theDelian maidens are first asked to rejoice (åÆ�æ��, 166), then tocelebrate the persona poetica himself (who is identified as I�cæl�Ø�� I�Ø�H�, 169; ‘the sweetest of singers’). In a remarkable re-flexive movement it is the poetic ‘I’, distinguished by the bewitchingsweetness of his voice, who becomes the object of the Delian maidens’song: through this technique of delegating his own aedic-rhapsodicvoice to the choral voice of the young maidens, whom he himselfplaces on stage, the aoidos-rhapsode introduces (in the third personand once again marked by poetic terpein) a form of ‘signature’(sphragis): the chorus of Delian maidens sing, ‘he is a blind man, helives in rocky Chios, all of his songs remain supreme afterwards’(�ıçºe I��æ, �NŒ�E �b ��øfi ��Ø �ÆØ�Æº��fi Å j ��F �A�ÆØ ��Ø�Ł��IæØ��ı�Ø� I�Ø�Æ� , 172–3). Beyond the question of the identity ofthis singer, who is marked by a blindness that reflects his poeticinspiration, everything happens as if the choral voice of the youngmaidens created by the speaker-singer were colouring his own voice;indeed, following the ‘signature’, the ‘I’ of the persona cantans adopts

39 Apoll. 127–32, 138–55. On the nature of the offering pleasing to the god, whichrepresents the hymn in general, see the excellent discussion of Aubriot-Sévin (1992),182–93; on the relationship of the bow and the lyre, as defined from an anthropolo-gical point of view, see Mombrun (2001), 67–78.

40 On the status of the choral group of Delian women, cf. Calame (2001), 104–10.The function of their song is recently examined by Peponi (2009), 39–51.


the plural ‘we’: it is this poetic ‘we’ which, reciprocally with the praiseof the singer by the young dancers, will spread Œº� to other cities,the poetic glory of ‘you’ the Delian maidens.41

Regardless of any rhapsodic ‘stitching’, related to which there isgreat textual uncertainty,42 the Pythian section of the Hymn to Apollobegins with a double musical progression. In a first movement to-wards rocky Pytho, the son of Leto, dressed in a divinely perfumedcloak, accompanies himself on a lyre (ç�æØª�, 184), from which hebrings forth sounds that give rise to desire. Then, after a change oflocation to Olympus, he is received by the song of the Muses, to thesound of the lyre (Œ�ŁÆæØ , 188); in their beautiful voice, accompaniedby the dances of the Graces, the Hours, Harmony, Hebe, and Aph-rodite herself, the Muses sing (��F�Ø�, 190) of the privileges of theimmortals and the misfortunes of humans who suffer from old ageand death. The Muses are followed in the dance by Artemis, while herbrother plays the lyre, as well as by Ares and Hermes; at this sight oftheir two children dancing and singing, wise Zeus and Leto of thegolden locks rejoice. In response, the model song of the young Apollois taken up by the poetic ‘I’ who, as at the beginning of the Delphicsection, poses the rhetorical question, ‘How shall I hymn you, fitsubject as you are in every respect?’ (�H ��æ �’ ��ø, ��ø �hı�� K��Æ, 207). The use of the future performative form ��ø

confirms that this in fact coincides with the beginning of the narrativepraise offered by the speaker-‘I’, who in this context once again usesthe present form I��ø (‘I sing’, 208).43 This is yet another corres-pondence between musical performance amongst the gods and mu-sical recitation undertaken by the mortal singer, in the context of aritual and cultic sequence addressed to the divinity.

In accordance with other enunciative modalities, an analogousphenomenon is repeated at the end of the Pythian section and

41 On this scene as a mimetic staging of the singer’s inspiration by a chorus ofMuses, cf. Nagy (2009b), 19–30; on a certain ‘inter-changeability’ in the monodic orchoral mode of performance of the Hymns, see Clay in this volume (pp. 250–2). Forthe authorial meaning of the sphragis device, see Calame (2004b), 13–19, with furtherbibliography; and as far as the (erotic) charm produced by musical poetry is con-cerned, see the numerous parallels and bibliographical references quoted in Calame(2002), 58–63. On the relationship with the ritual, see Taddei (2007), 89–93.

42 On this ‘stitching’ cf. Miller (1986), 66–9. On the unity of the Hymn see also thereferences given above in n. 38.

43 Cf. Apoll. 19. The narrative incoherence has suggested to some a lacuna;cf. Càssola 499–51.

352 Claude Calame

therefore at the end of the unified long Hymn dedicated to Apollo.The god is compelled to reveal his identity to the Cretan sailors whomhe intends to lead to Delphi (now liberated from the monsterTyphaon) to become his first priests: ‘For I am Zeus’ son, I declaremyself Apollo’ (�rØ �’ Kªg ˜Øe ıƒ� , ��ººø� �’ �hå�ÆØ �r�ÆØ, 480).As for the foreign Cretans, they will be the guardians of the god’ssanctuary which is honoured by all men. Within a narrative thatunfolds very normally in the aorist and the third person, the authori-tative speech of Apollo in direct discourse in the present and futuretenses brings about the coincidence of past heroic time and space withthe time and space of the Hymn’s recitation. Equally, the differentindications given by Apollo in direct discourse of the ritual actionsthat the Cretan sailors must accomplish before they reach Delphi areplaced in connection, from an enunciative point of view, with the hicet nunc of the performance of the Hymn. This is the case in particularwith the paean that the Cretans sing (NÅ�ÆØ��’ I��Ø�, ‘sing IePaieon’, 500) while following the god, who therefore becomes theirchorus leader. As is often the case in epic narrative, the accomplish-ment of the different ritual acts prescribed by the god in directdiscourse is subsequently integrated into the narrative in the aorist.

The musical scene plays a leading role in the narrative: it portraysthe musical accompaniment of the chorus-leader Apollo on the lyreand the procession of the Cretans singing Ie Paieon, just like theCretans whose paeans are inspired by the Muses (517–18). When hehas arrived within the confines of Parnassus, it only remains forApollo to confer to the young Cretans the administration of hissanctuary, including the function of intermediary between the menwho visit Delphi and the god of prophecy and justice. According tothe will of the god, this function will be taken up by ‘interpreters’(�Å��æ� , 542), in concurrence with the role that Heraclitus attrib-utes to the Delphic oracle: not to ‘speak’ (ºª�Ø�), not to ‘conceal’(Œæ��Ø�), but to ‘indicate’ (�ÅÆ��Ø�).44 Commentators have iden-tified the Delphic interpreters with the administrators of the Delphicsanctuary who became the representatives of the amphictyony.45

44 Heraclit. fr. 22 B 93 Diels-Kranz. The ritual acts prescribed by Apollo in thedirect discourse of lines 475–501 (see the initial vocative address ��E��Ø) are then takenup again in the narrative mode in lines 503–23. Detienne (1998), 134–44, 169–72describes in vivid detail the profile and function of this god, who is at the same timethe founder, informer, and interpreter of the sanctuary.

45 For the creation of the amphictyony and the administrative circle of the sanc-tuary, cf. Càssola 91–2, 515–16.


3. POETIC CONTRACTS BETWEEN GODSAND MORTALS (PRECES)

The formulaic phrasing of the concluding verses of each HomericHymn aims to achieve, in and by the poem itself, a relationshipbetween mortal men and the divinity. In a previous morphologicalstudy, I have tried to illuminate the different poetic and verbalmethods used in the Homeric Hymns to make the hymnic compos-ition a musical offering to the invoked divinity and to involve the godin the reciprocal relationship proper to prayer.46 This poetic contractof do ut des generally begins with a direct address to the divinity; theevocatio of the initial section becomes the invocatio, just as in everyother hymnic form. Under the grammatical form ‘you’, the god takespart in a direct relationship, achieved in the present poem, with thepoetic ‘I’ of the aoidos-rhapsode who is the representative of thepeople witnessing the cultic ceremony for which the Homeric recita-tion is intended.

In the great majority of Homeric Hymns included in the corpustransmitted by the manuscript tradition, the final section, termedpreces, opens with a formula of ‘salutation’ (åÆ�æ�Ø�); it is appropriateto interpret this åÆ�æ�Ø� (which begins the prayer section in 27 of the33 Hymns) in the etymological sense of the verb, as an invitation tothe divinity to rejoice.47 As I indicated at the outset of this chapter, insome cases the object of rejoicing is mentioned: it is the song. Thus, inthe brief Hymn 9 to Artemis, the goddess is invited, with all otherfeminine divinities, to ‘rejoice in the song’ (åÆEæ� . . . I�Ø�Bfi , 7); this is aformulation that one finds, for example, also at the end of Hymn 25 tothe Mother of gods and men, discussed above. Also, in a third of theHymns which conclude with a request to the divinity to rejoice,the imperative åÆEæ� is accompanied by �o�ø (‘thus’), which linksthe pleasure to the preceding laudatory section.48 The role of thecentral narrative and descriptive part of the Hymn is thus confirmed;

46 The essential elements of my study of morphology of the preces section found inCalame (2005), 24–30 are therefore again taken up here.

47 For bibliography on the literal sense of åÆ�æ�Ø�, see above n. 3.48 Dion.D 11, Apoll. 545,Herm. 579,Hys. 9. 7, 14. 6, 16. 5, 18. 10, 19. 48, 21. 5, 26. 11,

28. 17.

354 Claude Calame

the narrative and descriptive praise is introduced into the logic of thereciprocity between man and god.

The final address to the divinity often leads into a prayer, the objectof which is explicit. Thus, Hymn 11 dedicated to Athena asks thegoddess to ‘grant us’ (�� ¼Ø, 5) good fortune and prosperity, whilein the short Hymn 10 to Aphrodite the aoidos-rhapsode-‘I’ asks thegoddess to grant a song full of charm that evokes amorous desire(ƒ�æ��Æ� I�Ø��, 5). This same reciprocal relationship is made clearin Hymn 30 to Gaia, already discussed above, by the participation ofthe young maidens in choral celebrations which signify the prosperitythat Gaia gives to just cities: ‘Rejoice, mother of the gods, consort ofstarry Heaven; be favourable and grant comfortable livelihood in returnfor my singing’ (åÆEæ�, Ł�H� ��Åæ, ¼º�å’ ˇPæÆ��F I��æ�� j�æ�çæø� �’ ¼��’ fiT�B �� Łı�æ�’ Z�ÆÇ�, 17–18). Just like theepica laus section, the concluding verses are focused upon aedicsong. Elsewhere, the request is sometimes accompanied by an act ofverbal deixis, which connects the request to the hic et nunc of thehymnic prayer: in the brief Hymn 13 to Demeter cited at the begin-ning of this chapter, the speaker-singer asks the goddess to keep safe‘this city’ (�� ºØ�, 3) while beginning the song (¼æå� I�Ø�B , 3);and at the end of Hymn 6 to Aphrodite the poetic persona explicitlyasks the goddess not only to order his song, but above all to grant himvictory in ‘this competition’ (K� IªH�Ø �fiH��, 19–20). This contest canbe little other than a competition of musical recitation, such as thatwhich formed an integral part of the Panathenaic festival.49

A third distinctive trait of the final section of the Homeric Hymns isthat many of the concluding sequences establish a contract betweenthe persona cantans (including those for whom he is the spokes-person) and the divinity evoked, then invoked. The aedic-rhapsodichymn, which itself constitutes a musical offering by its performance,often places on the level of musical celebration the contractual rela-tionship of reciprocity between the divinity and the poetic-‘I’ (thelatter readily expanded to a poetic-‘we’ that reflects the communityassociated with the cultic ceremony). In exchange for the pleasureprovoked by the song which is in the process of being offered to him,

49 On the Iª�� as a competition of rhapsodic recitation, cf. Shapiro (1992) and inthis volume Nagy (pp. 322–5). According to West (1992), 19 n. 25, Hy. 6 was meant tobe performed at ‘one of the Cyprian festivals of Aphrodite, such as the panegyris ofOld Paphos (Strab. 14. 6. 3)’.


Dionysus is called upon in Hymn 26 to grant to his audience-‘we’future rejoicing at the due time each year. And it is no surprise to seethe Muses and Apollo, who have just been lauded for the support theygive to singers and lyre-players, engaged in a reciprocal poetic rela-tionship in Hymn 25: the Muses and Apollo are invited to rejoice inexchange for the glory that they are requested to confer upon the songof the poetic-‘I’ (Kc� �Ø��Æ�’ I�Ø��; ‘honour my song’, 6). Themodel for this reciprocal rejoicing in musical activity is furnished bythe narrative of the Delian scene in the first part of the long Hymn toApollo.

On the level of syntax, the reciprocity thus established betweengods and mortals is underlined, from an enunciative point of view, bythe proximity of the forms of ‘you’ and ‘I’. This is, for example, thecase also at the end of the prelude ofHesiod’s Works and Days.50 Thismorphological and syntactical proximity is particularly noticeable inthe formulas that conclude a certain number of the Hymns by an-nouncing, in a performative manner, the movement to another song:the line ÆP�aæ Kªg� �ø� �� ŒÆd ¼ººÅ ��’ I�Ø�B (‘And I willtake heed both for you and for other singing’) concludes Hymn 30 toGaia and, as we have seen above, Hymn 25 to the Muses and Apollo.The same formula is found also at the end of Hermes, Demeter, andHymn 6 to Aphrodite, not to mention the Hymn to Apollo.51

Finally, the form ��ÆØ refers to the function of memory, whichis the foundation of poetic epic recitation and also central to otherpoetic genres. Embodied in the figure of Mnemosyne, the mother ofthe Muses (both names being etymologically linked to memory), thisfunction of memory corresponds to the inspired ability of the aoidos-rhapsode to draw from the long tradition of epic poetry the resourcesrequired to narrate the great deeds of the heroes who were still closeto the gods: creative memory, which depends upon a tradition ofmemory to maintain and revive the common recollection of a foun-dational poetic past.52 The function of memory finds fulfilment at theclose of many Homeric Hymns in the form of the performative future,

50 Cf. Hes. Op. 10, with parallels in Calame (2005), 40–1.51 On this and the formulae below as an indication of passage to another song,

cf. in this volume Nagy (pp. 322–5); for an alternative interpretation, see Clay(pp. 237–40).

52 This poetic function of aedic memory has been widely explored: see the refer-ences given in Calame (2005), 195 n. 26, to which one should add Brillante (2009),43–6.

356 Claude Calame

which expresses the intention of the poetic-‘I’ in the very completionof his act of singing. The performance of the present Hymn is an actof praise and verbal offering to the divinity, which is destined to leadinto a new act of poetic memory.

This performative movement to another song is also made explicitin another concluding formula, which likewise contains a form of theperformative future: ��F �’ Kªg Iæ�� Æ��ÆØ ¼ºº�� K

o�� (‘having begun with you, I will pass on to another song’).This formula concludes the Hymn to Aphrodite, as well as both theshort Hy. 18 to Hermes (with repetition of the invitation to rejoiceafter the announcement of movement to another song) and the evenshorter Hy. 9 to Artemis. In the last case, after the invitation to rejoicein the song (I�Ø�Bfi , 7), the first part of the concluding formula isanticipated by an expression which recalls an opening formula: ÆP�aæKªg � �� æH�Æ ŒÆd KŒ �Ł�� ¼æå�’ I��Ø� (‘Of you and from youfirst I sing’, 8).53 In addition to the new formal and syntactic connec-tion made between ‘I’ and ‘you’ and the terminal use of a performa-tive future, which confirms the function of this Hymn as a proem tofurther poetic performance, this concluding line establishes the se-mantic equivalence often noted between the terms o�� and I�Ø��.

To conclude, understood in a generic manner as a song, the HomericHymn is an offering made by mortals to a god, which leads intoanother offering, itself also poetic. In this ritual play of gift andcounter-gift, the Homeric Hymn establishes between the god and acommunity of mortals a poetic contract concerning the performanceof the song as a ritual sacrifice.

The polytheistic culture of ancient Greece did not have a monopolyon this verbal and musical contract between a mortal and a divinity.One finds an analogous form in the Hebrew Psalms in this invocatoryprelude (141, 1–2):

Lord I call upon you: make haste unto to me,Give ear to my supplication when I call upon you.May my prayer be set before you like incense,And the raising up of my hands be an evening sacrifice.

53 Other Hymns contain more subtle indications of performance: cf. Calame(2005), 196 n. 28.



Works Cited

Abel, E. (1886), Homeri hymni, epigrammata, Batrachomyomachia (Leipzig).Acosta-Hughes, B. (2002), Polyeideia: The Iambi of Callimachus and the

Archaic Iambic Tradition (Berkeley).Albis, R. V. (1996), Poet and Audience in the Argonautica of Apollonius

(Lanham).Alderink, L. J. (1982), ‘Mythological and Cosmological Structure in the

Homeric Hymn to Demeter’, Numen 29: 1–16.Allan, W. (2008), Euripides: Helen (Cambridge).Allen, T. W. (1895a), ‘The Text of the Homeric Hymns. I’, JHS 15: 136–83.——(1895b), ‘The Text of the Homeric Hymns. II’, JHS 15: 251–313.——(1897a), ‘The Text of the Homeric Hymns. III’, JHS 17: 45–62.——(1897b), ‘The Text of the Homeric Hymns. IV’, JHS 17: 241–67.——(1898), ‘The Text of the Homeric Hymns. V’, JHS 18: 23–32.Aloni, A. (1980), ‘Prooimia, Hymnoi, Elio Aristide e i cugini bastardi’, QUCC

33: 23–40.——(1989), L’aedo e i tiranni: ricerche sull’ inno omerico a Apollo (Rome).——(1998), Cantare glorie di eroi (Turin).Ambühl, A. (2005), Kinder und junge Helden: innovative Aspekte des

Umgangs mit der literarischen Tradition bei Kallimachos (Leuven).Amrhein, V., Körner, P., and Naguib, M. (2002), ‘Nocturnal and Diurnal

Singing Activity in the Nightingale: Correlations with Mating Status andBreeding Cycle’, Animal Behaviour 64: 939–44.

Arias, P. E., Hirmer, M., and Shefton, B. B. (1962), A History of Greek VasePainting (London).

Armstrong, J. I. (1958), ‘The Arming Motif in the Iliad’, AJPh 79: 337–54.Arnott, W. G. (1996), Alexis: The Fragments (Cambridge).Aubriot-Sévin, D. (1992), Prière et conceptions religieuses en Grèce ancienne

jusqu’à la fin du Ve siècle av. J.-C. (Lyon).Ausfeld, K. (1903), ‘De Graecorum precationibus quaestionibus’, in Jahrbuch

für classische Philologie, Supplement Band 28: 502–47.Austin, N. (1975), Archery at the Dark of the Moon: Poetic Problems in

Homer’s Odyssey (Berkeley).Bain, D. (2007), ‘Low Words in High Places: Sex, Bodily Functions, and Body

Parts in Homeric Epic and Other Higher Genres’, in Finglass, Collard, andRichardson (2007), 40–57.

Bakker, E. J. (2002), ‘Remembering the God’s Arrival’, Arethusa 35: 63–81.

——(2005), Pointing at the Past: From Formula to Performance in HomericPoetics (Washington).

Barner, W. (1977), ‘Neuphilologische Rezeptionsforschung und dieMöglichkeiten der klassischen Philologie’, Poetica 9: 499–521.

Barnes, J. (1711), Homeri Ilias et Odyssea: accedunt Batrachomyomachia,hymni et epigrammata (Cambridge).

Barrett, W. S. (1964), Euripides: Hippolytus (Oxford).Bartol, K. (2010), Review of Faulkner (2008), Gnomon 82: 289–92.Bassi, K. (1989), ‘The Poetics of Exclusion in Callimachus’ Hymn to Apollo’,

TAPhA 119: 219–31.Baudy, D. (1989), ‘Das Keuschlamm-Wunder des Hermes (Hom. h. Merc.

409–13): ein möglicher Schlüssel zum Verständnis kultischer Fesselung?’,GB 16: 1–28.

Baumeister, A. (1860), Hymni Homerici (Leipzig).Beazley, J. D. (1948), ‘Hymn to Hermes’, AJA 52: 336–40.——(1950), ‘Some Inscriptions on Vases’, AJA 54: 310–22.Beck, D. (2001), ‘Direct and Indirect Speech in the Homeric Hymn to

Demeter’, TAPhA 131: 53–74.Berg, R. M. van den (2001), Proclus’ Hymns: Essays, Translations,

Commentary (Leiden).Bergk, T. (1872), Griechische Literaturgeschichte i (Berlin).Bergren, A. (1989), ‘The Homeric Hymn to Aphrodite. Tradition and

Rhetoric, Praise and Blame’, CA 8: 1–41.Bernsdorff, H. (1999), Das Fragmentum Bucolicum Vindobonense

(P. Vindob. Rainer 29801): Einleitung, Text und Kommentar (Göttingen).Betegh, G. (2004), The Derveni Papyrus. Cosmology, Theology and

Interpretation (Cambridge).Bielohlawek, K. (1930), ‘Komische Motive in der homerischen Gestaltung

des griechischen Göttermythus’, ARW 28: 185–211.Bierl, A. (2004), ‘“Turn on the Light!” Epiphany, the God-Like Hero

Odysseus, and the Golden Lamp of Athena in Homer’s Odyssey (esp.19.1–43)’, ICS 29: 43–61.

Bing, P. (1988), The Well-Read Muse: Present and Past in Callimachus andthe Hellenistic Poets (Göttingen).

——(1993), ‘Impersonation of Voice in Callimachus’ Hymn to Apollo’,TAPhA 123: 181–98.

——(1995), ‘Callimachus and the Hymn to Demeter’, Syllecta Classica 6:29–42.

——and Uhrmeister, V. (1994), ‘The Unity of Callimachus’ Hymn toArtemis’, JHS 114: 19–34.

Blum, R. (1977), Kallimachos und die Literaturverzeichnung bei denGriechen: Untersuchungen zur Geschichte der Bibliographie (Frankfurt).

360 Works Cited

Boedeker, D. (1974), Aphrodite’s Entry into Greek Epic, MnemosyneSupplement 31 (Leiden).

Böhme, R. (1937), Das Prooimion. Eine Form sakraler Dichtung der Griechen(Diss., Heidelberg).

Bonnafé, A. (1985), Éros et Éris: Mariages divins et mythe de succession chezHésiode (Lyon).

Borgeaud, P. (1988), The Cult of Pan in Ancient Greece (English trans. ofRecherches sur le dieu Pan [1979, Rome] by K. Atlass and J. Redfield)(Chicago).

Bornmann, F. (1968), Callimachi Hymnus in Dianam (Florence).Boserup, I. (1971), ‘Zu Philodems De pietate und Heraklit B 80’, ZPE 8:

109–15.Bouchon, R., Brillet-Dubois, P., and Le Meur-Weissman, N. (forthcoming),

Hymnes de la Grèce antique: Entre littérature et histoire (Lyon).Bowie, E. (2002), ‘Ionian Iambos and Attic Komoidia: Father and Daughter,

or Just Cousins?’, in A. Willi (ed.), The Language of Greek Comedy(Oxford), 33–50.

Bremer, D. (1975), ‘Die Epiphanie des Gottes in den homerischen Hymnenund Platons Gottesbegriff ’, ZRGG 27: 1–21.

Bremer, J. M. (1981), ‘Greek Hymns’, in H. S. Versnel (ed.), Faith, Hope andWorship. Aspects of Religious Mentality in the Ancient World (Leiden),193–215.

——(1998), ‘The Reciprocity of Giving and Thanksgiving in GreekWorship’,in Gill, Postlethwaite, and Seaford (1998), 127–39.

Brillante, C. (2009), Il cantore e la Musa. Poesia e modelli culturali nellaGrecia arcaica (Pisa).

Brillet-Dubois, P. (1998), ‘Généalogie et géographie troyennes dans lediscours d’Enée au chant XX de l’Iliade’, in D. Auger and S. Saïd (eds.),Généalogies mythiques (Nanterre), 189–201.

——(2001), ‘Les liaisons dangereuses: Dieux et mortels dans l’Hymnehomérique à Aphrodite’, Europe 865: 250–60.

——(2006), ‘La revanche d’Aphrodite: L’Hymne homérique à Aphrodite etl’idéal héroïque de l’Iliade’, in P. Brillet-Dubois and E. Parmentier (eds.),+Øº�º�ª�Æ. Mélanges en l’honneur de M. Casevitz (Lyon), 67–75.

Brommer, F. (1956), ‘Pan’, RE suppl. 8, 949–1008.Brown, C. G. (1989), ‘Ares, Aphrodite, and the Laughter of the Gods’,

Phoenix 43: 283–93.Brulé, P. and Vendriès, Ch. (eds.) (2001), Chanter les dieux. Musique et

religion dans l’Antiquité grecque et romaine (Rennes).Brumfield, A. C. (1981), The Attic Festivals of Demeter and their Relation to

the Agricultural Year (Salem).Bulloch, A. W. (1977), ‘Callimachus’ Erysichthon, Homer and Apollonius

Rhodius’, AJPh 98: 97–123.

Works Cited 361

Bundy, E. L. (1972), ‘The “Quarrel between Kallimachos and Apollonios”Part I: The Epilogue of Kallimachos’s “Hymn to Apollo” ’, CaliforniaStudies in Classical Antiquity 5: 39–94.

——(1986), Studia Pindarica (Berkeley).Burgess, J. S. (2001), The Tradition of the Trojan War in Homer and the Epic

Cycle (Baltimore).——(2004), ‘Performance and the Epic Cycle’, CJ 100: 1–24.——(2006), ‘Neoanalysis, Orality and Intertextuality: An Examination of

Homeric Motif Transference’, Oral Tradition, 21.1: 148–89.——(2009), The Death and Afterlife of Achilles (Baltimore).Burkert, W. (1968), ‘Orpheus und die Vorsokratiker. Bemerkungen zum

Derveni-Papyrus und zur pythagoreischen Zahlenlehre’, A&A 14: 93–114.——(1979), ‘Kynaithos, Polycrates, and the Homeric Hymn to Apollo’, in

G. W. Bowersock, W. Burkert, and M. C. J. Putnam (eds.), Arktouros:Hellenic Studies Presented to Bernard M. W. Knox (Berlin), 53–6 ¼Burkert (2001), 189–97.

——(2001), Kleine Schriften, i: Homerica, ed. C. Riedweg (Göttingen).Cairns, F. (1983), ‘Alcaeus’ Hymn to Hermes, P.Oxy. 2734 fr.1, and Horace,

Odes 1, 10’, QUCC 42: 29–35.Calame, C. (1993), ‘Legendary Narration and Poetic Procedure in

Callimachus’ Hymn to Apollo’, in M. A. Harder, R. F. Regtuit, and G. C.Wakker (eds.), Hellenistica Groningana: Callimachus I (Leuven), 37–55.

——(1995), ‘Variations énonciatives, relations avec les dieux et fonctionspoétiques dans les Hymnes homériques’, MH 52: 1–19. [Revised Englishtrans. in Calame (2005), 19–35.]

——(1997), ‘L’Hymne homérique à Déméter comme offrande: regardrétrospectif sur quelques catégories de l’anthropologie de la religiongrecque’, Kernos, 10: 111–33 ¼ Calame (2008a), 63–83.

——(2000), Le récit en Grèce ancienne. Énonciations et représentations depoètes, 2nd edn. (Paris). [English trans. of 1st edn. The Craft of PoeticSpeech in Ancient Greece (1995, Ithaca) by D. Collins and J. Orion.]

——(2001), Choruses of Young Women in Ancient Greece: Their Morphology,Religious Role, and Social Function. [English trans. of Les Choeurs dejeunes filles en Grèce archaïque, 2 vols. (Rome, 1977) by D. Collins andJ. Orion], 2nd edn. (Lanham).

——(2002), L’Éros dans la Grèce antique, 2nd edn. (Paris). [English trans. of1st edn. The Poetics of Eros in Ancient Greece (1999, Princeton) byJ. Lloyd.]

——(2004a), ‘Deictic Ambiguity and Auto-Referentiality: Some Examplesfrom Greek Poetics’, Arethusa 37: 415–43.

——(2004b), ‘Identités d’auteur à l’exemple de la Grèce classique: signatures,énonciations, citations’, in C. Calame and R. Chartier (eds.), Identitésd’auteur dans l’Antiquité et la tradition européenne (Grenoble), 11–39.

362 Works Cited

——(2005), Masks of Authority: Fiction and Pragmatics in Ancient GreekPoetics [English trans. of Masques d’autorité. Fiction et pragmatique dansla poétique grecque antique (2005, Paris) by P. M. Burk] (Ithaca).

——(2006), ‘L’histoire comparée des religions et la construction d’objetsdifférenciés: entre polythéisme gréco-romain et protestantismeallemand’, in M. Burger and C. Calame (eds.), Comparer lescomparatismes. Perspectives sur l’histoire et les sciences des religions(Paris), 209–35.

——(2008a), Sentiers transversaux. Entre poétiques grecques et politiquescontemporaines (Grenoble).

——(2008b), ‘Entre récit héroïque et poésie rituelle: le sujet poétique quichante le mythe’, in S. Parizet (ed.), Mythe et littérature: Poétiquescomparatistes (Paris), 123–41.

——(forthcoming), ‘Les “noms” des dieux: les pouvoirs de la dénominationentre polythéisme athénien classique et théologie orphique’ (Paris).

Cameron, A. (1995), Callimachus and his Critics (Princeton).Campbell, M. (1981), Echoes and Imitations of Early Epic in Apollonius

Rhodius (Leiden).——(1983), Studies in the Third Book of Apollonius Rhodius’ Argonautica

(Hildesheim).——(1994), A Commentary on Apollonius Rhodius’ Argonautica III 1–471

(Leiden).Cantilena, M. (1982), Ricerche sulla dizione epica, i (Rome).Capponi, M. (2003), ‘Fins d’hymns et sphragis énonciatives’,QUCC 75: 9–35.Carey, C. (1986), ‘Archilochus and Lycambes’, CQ² 36: 60–7.——(2009), ‘Iambos’, in F. Budelmann (ed.), The Cambridge Companion to

Greek Lyric Poetry (Cambridge), 149–67.Caspo, E. (2003), ‘The Dolphins of Dionysus’, in E. Caspo and M. Miller

(eds.), Poetry, Theory, Praxis. The Social Life of Myth, Word, and Image inAncient Greece. Essays in Honour of William J. Slater (Oxford), 69–98.

Chantraine, P. (1935), ‘Grec KŒª�ª��ÆØ (Hymne homérique à Aphrodite,197)’, BSL 36: 131–2.

Chappell, M. (1995), ‘A Commentary on the Homeric Hymn to DelianApollo, with Prolegomena’ (Ph.D. thesis, London).

——(2006), ‘Delphi and the Homeric Hymn to Apollo’, CQ² 56: 331–48.Cheshire, K. (2005), ‘Thematic Progression and Unity in Callimachus’Hymn

to Apollo’, CJ 100: 331–48.Cheyns, A. (1988), ‘La structure du récit dans l’Iliade et l’Hymne homérique

à Déméter’, RBPh 66: 32–67.Clarke, M. (2001), ‘Heart-Cutting Talk: Homeric Œ�æ��ø and Related

Words’, CQ² 51: 329–38.Clauss, J. (1986), ‘Lies and Allusions: The Addressee and Date of

Callimachus’ Hymn to Zeus’, ClAnt 5: 155–70.

Works Cited 363

——(1993), The Best of the Argonauts (Berkeley).Clay, D. (2004), Archilochos Heros. The Cult of Poets in the Greek Polis

(Washington).Clay, J. S. (1989), The Politics of Olympus (Princeton) [a 2nd edn. was

published in 2006].——(1994), ‘Tendenz and Olympian Propaganda in the Homeric Hymn to

Apollo’, in J. Solomon (ed.), Apollo: Origins and Influences (London),23–36.

——(1997), ‘The Homeric Hymns’, in Morris and Powell (1997), 489–507.——(1999), ‘Il. 24.269 and the Semantics of Œ�æ��ø’, CQ² 49: 618–21.——(2003), Hesiod’s Cosmos (Cambridge).——(forthcoming), ‘Theology and Religion in the Homeric Hymns:

Some Preliminary Thoughts’, in Bouchon, Brillet-Dubois, and LeMeur-Weissman (forthcoming).

Clinton, K. (1986), ‘The Author of the Homeric Hymn to Demeter’, OAth 16:43–9.

——(1992), Myth and Cult: The Iconography of the Eleusinian Mysteries(Stockholm).

——(1993), ‘The Sanctuary of Demeter and Kore at Eleusis’, in Marinatosand Hägg (1993), 110–24.

Collins, D. (2004), Master of the Game: Competition and Performance inGreek Poetry (Washington, DC, and Cambridge, Mass.).

Compton, T. M. (2006), Victim of the Muses. Poet as Scapegoat, Warrior, andHero in the Greco-Roman and Indo-European Myth and History(Washington, DC).

Costantini, M. and Lallot, J. (1987), ‘Le �æ��Ø�� est-il un proème?’, inP. Hoffmann, J. Lallot, and A. le Boulluec (eds.), Le Texte et sesrepresentations. Études de littérature ancienne 3 (Paris), 13–27.

Crusius, O. (1889), ‘Die homerische Dionysushymnus und die Legender derVerwandlung der Tyrsener’, Philologus 48: 193–228.

Currie, B. (2006), ‘Homer and the Early Epic Tradition’, in M. J. Clarke,B. G. F. Currie, and R. O. A. M. Lyne (eds.), Epic Interactions. Perspectiveson Homer, Virgil, and the Epic Tradition (Oxford), 1–46.

——(forthcoming), ‘Perspectives on Neoanalysis from the Archaic Hymns toDemeter’, in �. Andersen and D. Haug (eds.), Relative Chronology inEarly Greek Epic Poetry (Cambridge).

D’Alessio, G. B. (1995), ‘Una via lontana dal cammino degli uomini (Parm.frr. 1 þ 6 D.-K.; Pind. Ol. VI 22–7; pae. VIIb 10–20)’, SFIC 13: 143–81.

——(2004), ‘Past Future and Present Past: Temporal Deixis in Greek ArchaicLyric’, Arethusa 37: 267–94.

Danek, G. (1998), Epos und Zitat: Studien zu den Quellen der Odyssee(Vienna).

364 Works Cited

Danielewicz, J. (1976), Morfologia hymnu antycznego (Poznań) [contains anEnglish summary].

Davies, J. K. (1997), ‘The Moral Dimension of Pythian Apollo’, in A. B. Lloyd(ed.), What is a God? Studies in the Nature of Greek Divinity (London),43–64.

——(2007), ‘The Origins of the Festivals, especially Delphi and the Pythia’,in S. Hornblower and C. Morgan (eds.), Pindar’s Poetry, Patrons andFestivals (Oxford), 47–71.

Deacy, Susan (2008), Athena: Gods and Heroes of the Ancient World(London and New York).

De Bloois, N. (1997), ‘Rape, Marriage or Death? Gender Perspectives in theHomeric Hymn to Demeter’, PhQ 76: 245–62.

Degani, E. (2002), Studi su Ipponatte (Hildesheim).——(2007), Ipponatte. Frammenti (Bologna).De Jong, I. J. F. (2001), A Narratological Commentary on the Odyssey

(Cambridge).Del Corno, D. (1962), ‘Ricerche intorno alla Lyde di Antimaco’, Acme 15:

57–95.Deonna, W. (1939), ‘¯Pø��Æ. Croyances antiques et modernes: l’odeur suave

des dieux et des élus’, Genava 17: 167–263.Depew, M. (1993), ‘Legendary Narration and Poetic Procedure in

Callimachus’ Hymn to Apollo’, in M. A. Harder, R. F. Regtuit, andG. C. Wakker (eds.), Hellenistica Groningana: Callimachus I (Leuven),37–55.

——(2000), ‘Enacted and Represented Dedications: Genre and GreekHymns’, in M. Depew and D. Obbink (eds.), Matrices of Genre. Authors,Canons, and Society (Cambridge, Mass.), 59–80.

——(2004), ‘Mimesis and Aetiology in Callimachus’ Hymns’, in M. A.Harder, R. F. Regtuit, and G. C. Wakker (eds.), Hellenistica Groningana:Callimachus II (Leuven), 57–77.

De Saussure, F. (1916), Cours de linguistique générale [critical edn. 1972 byT. de Mauro] (Paris).

De Sousa Medeiros, W. (1961), Hipónax de Éfeso, Vol. 1: Fragmentos dosiambos (Coimbra).

De’ Spagnolis M. (2004), Il mito omerico di Dionysos ed i pirati tirreni in undocumento de Nuceria alfaterna (Rome).

Detienne, M. (1985), Dionysos à ciel ouvert (Paris).——(1998), Apollon le couteau à la main. Une approche expérimentale du

polythéisme grec (Paris).——and Sissa, G. (1989), La vie quotidienne des dieux grecs (Paris).Devlin, N. G. (1994), ‘The Hymn in Greek Literature’ (D.Phil. thesis,

Oxford).Dietrich, B. C. (1983), ‘Divine Epiphanies in Homer’, Numen 30: 53–79.

Works Cited 365

Dihle, A. (2002), ‘Zu den Fragmenten eines Dionysos-Hymnus’, RhM 145:427–30.

Dornsieff, F. (1938), ‘Zum homerischen Hermeshymnos’, RhM 87: 80–4.Dumézil, G. (1974), La religion romaine archaïque, 2nd edn. (Paris).Dunbar, N. (1995), Aristophanes: Birds (Oxford).Ehrhardt, W. (1993), ‘Die Fries des Lysikratesmonuments’, Antike Plastik 22:

7–67.Eisenberger, H. (1956), ‘Der Mythos in der äolischen Lyrik’ (Diss. Frankfurt).Eitrem, S. (1906), ‘Der homerische Hymnus an Hermes’, Philologus 65:

248–82.Erler, M. (1987), ‘Das Recht (˜"˚˙) als Segensbringerin für die Polis’, SIFC

80: 5–36.Evelyn-White, H. G. (1936), Hesiod, Homeric Hymns, Homerica, revised edn.

(Cambridge, Mass.).Fain, G. L. (2004), ‘Callimachus’ Hymn to Artemis and the Tradition of the

Rhapsodic Hymn’, BICS 47: 45–56.Fantuzzi, M. and Hunter, R. (2004), Tradition and Innovation in Hellenistic

Poetry (Cambridge).Faulkner, A. (2005), ‘Aphrodite’s Aorists: Attributive Sections in the

Homeric Hymns’, Glotta 81: 60–79.——(2008), The Homeric Hymn to Aphrodite (Oxford).——(2010a), ‘The Homeric Hymn to Dionysus: P. Oxy. 670’, ZPE 172: 1–2.——(2010b), ‘Callimachus and his Allusive Virgins: Delos, Hestia, and the

Homeric Hymn to Aphrodite’, HSCPh 105: 53–63.——(forthcoming a), ‘Fast, Famine, and Feast: Food for Thought in

Callimachus’ Sixth Hymn to Demeter’, HSCPh 106.——(forthcoming b), ‘The Performance of the Homeric Hymn to Aphrodite’,

in Bouchon, Brillet-Dubois, and Le Meur-Weissman (forthcoming).——(forthcoming c), ‘The Silence of Zeus: Speech in the Homeric Hymns’,

proceedings of Lampeter conference on hymns (May, 2009).——(forthcoming d), Les rencontres érotiques dans l’épopée grecque, Gaia

(2011).Fearn, D. (2007), Bacchylides: Politics, Performance, Poetic Tradition

(Oxford).Felson, N. (1994), Regarding Penelope: From Character to Poetics in the

Odyssey (Princeton).——(2000), ‘Paradigms of Paternity: A Study of the Gentle Father in

Homer’s Odyssey’, in A. Rengakos and J. N. Kazazis (eds.), Euphrosyne:Studies in Ancient Epic and its Legacy in Honor of D. N. Maronitis(Stuttgart), 89–98.

——(2002), ‘Threptra and “Invisible Hands”: the Father/Son Relation in Iliad24’, in C. Higbie and M. Malamud (eds.), Epos and Mythos, Arethusa 35:35–50 and 189–200.

366 Works Cited

Felson-Rubin, N. and Deal, H. M. (1994), ‘Some Functions of the DemophonEpisode in the Homeric Hymn to Demeter’, in Foley (1994), 190–7.

Fenik, B. (1968), Typical Battle Scenes in the Iliad (Wiesbaden).Finglass, P. J., Collard, C., and Richardson, N. J. (eds.) (2007), Hesperos.

Studies in Ancient Greek Poetry Presented to M. L. West on His SeventiethBirthday (Oxford).

Finkelberg, M. (2000), ‘The Cypria, the Iliad, and the Problem ofMultiformity in Oral and Written Tradition’, CPh 95: 1–11.

Förstel, K. (1979), Untersuchungen zum homerischen Apollonhymnos(Bochum).

Foley, H. P. (1994), The Homeric Hymn to Demeter: Translation,Commentary, and Interpretive Essays (Princeton).

Foley, J. M. (1995), The Singer of Tales in Performance (Bloomington).——(1997), ‘Oral Tradition and its Implications’, in Morris and Powell

(1997), 146–73.Ford, A. (1992), Homer: The Poetry of the Past (Ithaca).——(1997), ‘The Epic as Genre’, in Morris and Powell (1997), 396–414.——(2002), The Origins of Criticism: Literary Culture and Poetic Theory in

Classical Greece (Princeton).Forsyth, N. (1979), ‘The Allurement Scene: A Typical Pattern in Greek Oral

Epic’, CSCA 12: 107–20.Frame, D. (2009), Hippota Nestor (Washington, DC).Fränkel, H. (1993), Dichtung und Philosophie des frühen Griechentums. Eine

Geschichte der griechischen Literatur von Homer bis Pindar (Munich).Freed, G. and Bentman, R. (1955), ‘The Homeric Hymn to Aphrodite’, CJ 50:

153–9.Friedländer, P. (1934), ‘Lachende Götter’, Die Antike 10: 209–26.——(1966), ‘Das Proömium von Hesiods Theogonie’, Hermes 49: 1–16 ¼

E. Heitsch (ed.), Hesiod (Darmstadt), 277–94.Fröhder, D. (1994), Die dichterische Form der Homerischen Hymnen:

Untersucht am Typus der mittelgrossen Preislieder (Hildesheim).Fuhrer, T. (1988), ‘A Pindaric Feature in the Poems of Callimachus’, AJPh

109: 53–68.Furley, W. D. (2010), ‘Philikos’ Hymn to Demeter’, Paideia 64: 483–508.——and Bremer, J. (2001), Greek Hymns, Vols. i–ii (Tübingen).Gaisser, J.H. (1974), ‘Noun–Epithet Combinations in the Homeric Hymn to

Demeter’, TAPhA 104: 113–37.Ganszyniec, R. (1922), ‘˙#`$ ¸-$"$’, Archiv für Religionswissenschaft 21:

498–9.Gantz, T. (1993), Early Greek Myth (Baltimore and London).García, J. F. (2002), ‘Symbolic Action in the Homeric Hymns: The Theme of

Recognition’, CA 21: 5–39.

Works Cited 367

Garvey, T. (2008), ‘Plato’s Atlantis Story: A Prose Hymn to Athena’, GRBS48: 381–92.

Gelzer, T. (1987), ‘Bemerkungen zu Homerische Ares-Hymnus (Hom.Hy. 8)’, MH 44: 150–67.

——(1994), ‘Zum Codex Mosquensis und zur Sammlung der HomerischenHymnen’, Hyperboreus 1: 113–37.

Gemoll, A. (1886), Die homerischen Hymnen (Leipzig).Germany, R. (2005), ‘The Figure of Echo in the Homeric Hymn to Pan’, AJPh

126: 187–208.Giacomelli, A. (1980), ‘Aphrodite and After’, Phoenix 34: 1–19.Gill, C., Postlethwaite, N., and Seaford, R. (eds.) (1998), Reciprocity in

Ancient Greece (Oxford).Golden, L. (1990), ‘'e ª�º�E�� in the Iliad’, HSCPh 93: 45–57.Goodwin, A. (1893), Hymni Homerici (Oxford).Görgemanns, H. (1976), ‘Rhetorik und Poetik im homerischen

Hermeshymnus’, in H. Görgemanns and E. A. Schmidt (eds.), Studienzum antiken Epos (Meisenheim am Glan), 113–28.

Gottesman, A. (2008), ‘The Pragmatics of Homeric kertomia’, CQ² 58: 1–12.Graf, F. (1980), ‘Milch, Honig und Wein’, in G. Piccaluga (ed.), Perennitas:

studi in onore di Angelo Brelich (Rome), 209–21.——(2009), Apollo (London).Graziosi, B. (2002), Inventing Homer: The Early Reception of Epic

(Cambridge).——and Haubold, J. (2005), Homer: The Resonance of Epic (London).Greene, E. (2005), ‘Revising Illegitimacy: The Use of Epithets in the Homeric

Hymn to Hermes’, CQ² 55: 343–9.Griffin, J. (1980), Homer on Life and Death (Oxford).——(1986), ‘Words and Speakers in Homer’, JHS 106: 36–67.Groningen, B. A. van (1958), La Composition littéraire archaïque grecque

(Amsterdam).Guthrie, W. K. C. (1971), The Sophists (Cambridge).Haas, G. (1985), Die Syrinx in der griechischen Bildkunst (Vienna).Hainsworth, J. B. (1999), review of Morris and Powell (1997), JHS 119: 177–9.Halliwell, S. (2008), Greek Laughter: A Study of Cultural Psychology from

Homer to Early Christianity (Cambridge).Harrell, S. E. (1991), ‘Apollo’s Fraternal Threats: Language of Succession and

Domination in the Homeric Hymn to Hermes’, GRBS 32: 307–29.Harvey, A. E. (1955), ‘The Classification of Greek Lyric Poetry’, CQ² 5: 157–75.Haslam, M. W. (1993), ‘Callimachus’ Hymns’, in M. A. Harder, R. F. Regtuit,

and G. C. Wakker (eds.), Hellenistica Groningana: Callimachus I (Leuven),111–25.

Heitsch, E. (1965), Aphroditehymnos, Aeneas und Homer: SprachlicheUntersuchungen zum Homerproblem, Hypomnemata 15 (Göttingen).

368 Works Cited

Henrichs, A. (1972), ‘Toward a New Edition of Philodemus’ Treatise OnPiety’, GRBS 13: 67–98.

——(1975a), ‘Philodems De pietate als mythographische Quelle’, CronErc 5:5–38.

——(1975b), ‘Two Doxographical Notes: Democritus and Prodicus onReligion’, HSCPh 79: 93–123.

——(1996), ‘Dancing in Athens, Dancing in Delos: Some Patterns of ChoralProjection in Euripides’, Philologus 140: 48–62.

Herington, J. (1985), Poetry into Drama. Early Tragedy and the Greek PoeticTradition, Vol. 49, Sather Classical Lectures (Berkeley).

Hermann, G. (1806), Homeri hymni et epigrammata (Leipzig).Heubeck, A. (1966), ‘Thukidides II 104’, WS 79: 148–57.Heyworth, S. (2004), ‘Looking into the River: Literary History and

Interpretation in Callimachus, Hymns 5 and 6’, in M. A. Harder, R. F.Regtuit, and G. C. Wakker (eds.), Hellenistica Groningana: Callimachus II(Leuven), 139–59.

Hoekstra, A. (1965), Homeric Modifications of Formulaic Prototypes: Studiesin the Development of Greek Epic Diction (Amsterdam).

——(1969), The Sub-Epic Stage of the Formulaic Tradition (Amsterdam).Hollander, H. (1886), Die handschriftliche Überlieferung der homerischen

Hymnen (Osnabrück).Hölscher, U. (1988), Die Odyssee: Epos zwischen Märchen und Roman

(Munich).Hopkinson, N. (1982), ‘Juxtaposed Prosodic Variants in Greek and Latin

Poetry’, Glotta 60: 162–77.——(1984), Callimachus: Hymn to Demeter (Cambridge).Hopman-Govers, M. (2001), ‘Le jeu des épithètes dans les Hymnes

orphiques’, Kernos 14: 35–49.Hornblower, S. (1991), A Commentary on Thucydides. 1, Books 1–3

(Oxford).Humbert, J. (1936), Homère: Hymnes (Paris).Hunter, R. (1992), ‘Writing the God: Form and Meaning in Callimachus,

Hymn to Athena’, MD 29: 9–34.——(1993), The Argonautica of Apollonius: Literary Studies (Cambridge).——(1996), Theocritus and the Archaeology of Greek Poetry (Cambridge).——(2003), Theocritus: Encomium of Ptolemy Philadelphus (Berkeley).——and Fuhrer, T. (2002), ‘Imaginary Gods? Poetic Theology in the Hymns

of Callimachus’, in F. Montanari and L. Lehnus (eds.), Entretiens de laFondation Hardt: Callimaque (Geneva), 143–87.

Hurst, A. (1994), ‘Un nouveau papyrus du premier Hymne homérique: lepapyrus de Genève 432’, in A. Bülow-Jacobsen (ed.), Proceedings of the20th International Congress of Papyrologists, Copenhagen, 23–29 August,1992 (Copenhagen), 317–21 with plates 27–8.

Works Cited 369

Hutchinson, G. O. (2008), Talking Books: Readings in Hellenistic and RomanBooks of Poetry (Oxford).

Ilgen, D. (1796), Hymni Homerici cum reliquis carminibus minoribusHomero tribui solitis et Batrachomyomachia (Halle).

Irigoin, J. (1970), ‘Les manuscrits grec I: quelques catalogues récents’, REG83: 500–29.

Isler-Kerényi C. (2007), Dionysos in Archaic Greece: An UnderstandingThrough Images (Leiden).

Jacoby, F. (1933), ‘Der homerische Apollonhymnos’, Sitzungsberichte derpreussischen Akademie der Wissenschaften, Phil.-hist. Klasse 15: 682–751.

Jaillard, D. (2005), ‘Mises en place du panthéon dans les Hymnes homériques.L’exemple de l’Hymne à Déméter’, Gaia 9: 49–62.

——(2007), Configurations d’Hermès. Une Théogonie hermaïque, KernosSupplément 17 (Liège).

——(2010), ‘˚æÆ��ø� IŁÆ��ı �� Ł��f ŒÆd ªÆEÆ� Kæ��. “Il réalisa lesdieux immortels et la terre ténébreuse” (Hymne homérique à Hermès 427)’,in E. Bono and M. Curnis (eds.), Linguaggi del potere, poteri del linguaggio(Alessandria), 51–66.

——(forthcoming), ‘En matière de timé, j’obtiendrai la même hosíequ’Apollon. L’Hymne homérique à Hermès comme réajustement dupanthéon’, in Bouchon, Brillet-Dubois, and Le Meur-Weissman(forthcoming).

Janko, R. (1981), ‘The Structure of the Homeric Hymns: A Study in Genre’,Hermes 109: 9–24.

——(1982), Homer, Hesiod and the Hymns: Diachronic Development in EpicDiction (Cambridge).

——(1991), ‘Review of Clay (1989)’, CR 41: 12–13.——(1994), The Iliad: A Commentary. Books 13–16 (Cambridge).——(1998), ‘The Homeric Poems as Oral Dictated Texts’, CQ² 48: 1–13.——(2004), ‘Homeric Hymns and Apocrypha’ (review of M. L. West,

Homeric Hymns. Homeric Apocrypha. Lives of Homer), CR 54: 283–6.Johnston, S. (2002), ‘Myth, Festival and Poet: The “Homeric Hymn to

Hermes” and Its Performative Context’, CP 97: 109–32.Jones, W. H. S. (1933), Pausanias: Description of Greece, Vol. 3 Books 6–8. 21

(Cambridge, Mass.).Jost, M. (1985), Sanctuaires et cultes d’Arcadie (Paris).Jouan, F. (1966), Euripide et les légendes des chants cypriens: des origines de la

guerre de Troie à l’Iliade (Paris).——(1986), Mort et fécondité dans les mythologies (Paris).Jucker, I. (1956), Der Gestus des Aposkopein: ein Beitrag zur Gebärdensprache

in der antiken Kunst (Zürich).Kambylis, A. (1963), ‘Zur Dichterweihe des Archilochos’, Hermes 91:

129–50.

370 Works Cited

——(1965), Die Dichterweihe und ihre Symbolik. Untersuchungen zuHesiodos, Kallimachos, Properz und Ennius (Heidelberg).

Katz, J. (1999), ‘Homeric Hymn to Hermes 296: �º��Æ ªÆ��æe �æØŁ��’, CQ²49: 315–19.

Kennedy, G. A. (1994), A New History of Classical Rhetoric (Princeton).Keyssner, K. (1932), Göttervorstellung und Lebensauffassung im griechischen

Hymnus (Stuttgart).Kidd, D. (1997), Aratus: Phaenomena (Cambridge).Kirchhoff, J. W. A. (1893), ‘Beiträge zur Geschichte der griechischen

Rhapsodik’, in Sitzungsberichte der Königlich Preussischen Akademie derWissenschaften zu Berlin, Vol. 2 (Berlin), 893–918.

Kirk, G. S. (1966), ‘Formular Language and Oral Quality’, YCS 20: 153–74.——(1985), The Iliad: A Commentary: Books 1–4 (Cambridge).Kledt, A. (2004), Die Entführung Kores. Studien zur athenisch-Eleusinischen

Demeterreligion (Stuttgart).Köhnken, A. (1981), ‘Apollo’s Retort to Envy’s Criticism’, AJPh 102: 411–22.Koller, H. (1956), ‘Das kitharodische Prooimion’, Philologus 100: 159–206.Kontoleon, N. M. (1952), ‘˝ÆØ K�ØªæÆçÆd ��æd ��F �æåØº�å�ı KŒ —�æ�ı’,�æåÆØ�º�ªØŒc ¯çÅ�æ� 91: 32–95.

Kossatz, T. and Kossatz-Deissmann, A. (1992), ‘Martin von Wagner,Dionysos und die Seeraüber’, in H. Fronig, T. Hölscher, and H. Mielsch(eds.), Kotinos. Festschrift für Erika Simon (Mainz), 469–78.

Kouremenos, Th., Parássoglou, G. M., and Tsantsanoglou, K. (eds.) (2006),The Derveni Papyrus (Florence).

Kowalzig, B. (2007), Singing for the Gods: Performances of Myth and Ritual inArchaic and Classical Greece (Oxford).

Krischer, T. (1971), Formale Konventionen der homerischen Epik (Munich).Kuiper, K. (1910), ‘De discrepantiis hymni homerici in Mercurium’,

Mnemosyne 38: 1–50.Kullmann, W. (1956), Das Wirken der Götter in der Ilias. Untersuchungen

zur Frage der Entstehung des homerischen ‘Götterapparats’ (Berlin).Lacore, M. (1997), ‘˝��ı� o�� ’, Gaia 1: 13–40.Lakoff, G. and Johnson, M. (1980), Metaphors We Live By (Chicago).Laks, A. and Most, G. W. (eds.) (1997), Studies in the Derveni Papyrus

(Oxford).Lane Fox, R. (2008), Travelling Heroes: Greeks and Their Myths in the Epic

Age of Heroes (London).Larson, J. (1995), ‘The Corcyian Nymphs and the Bee Maidens of the

Homeric Hymn to Hermes’, GRBS 36: 341–58.Latte, K. (1955), ‘Zur griechischen Wortforschung II’, Glotta 34: 190–202.Lattimore, R. (1951), The Iliad of Homer (Chicago).Leduc, C. (2001), ‘Cinquante vaches pour une lyre! Musique, échange et

théologie dans l’Hymne à Hermès I’, in Brulé and Vendriès (2001), 19–36.

Works Cited 371

Lehnus, L. (1979), L’inno a Pan di Pindaro (Milan).Lenz, L. H. (1975), Der homerische Aphroditehymnus und die Aristie des

Aeneas in der Ilias (Bonn).Libermann, G. (1999), Alcée. Fragments (Paris).Lilja, S. (1972), The Treatment of Odours in the Poetry of Antiquity

(Helsinki).Lincoln, B. (1976), ‘The Indo-European Cattle-Raiding Myth’, HR 16: 42–65.Lissarrague F. (1987), Un flot d’images. Une esthétique du banquet grec

(Paris).Lloyd, M. (2004), ‘The Politeness of Achilles: Off-Record Conversation

Strategies in Homer and the Meaning of Kertomia’, JHS 124: 75–89.Lohmeyer, E. (1919), ‘Vom göttlichen Wohlgeruch’, SHAW 9.Lord, A. (1960), The Singer of Tales (Cambridge, Mass.).——(1968), ‘Homer as Oral Poet’, HSCPh 72: 1–46.——(1995), The Singer Resumes The Tale, ed. M. L. Lord (Ithaca).Lord, M. L. (1994), ‘Withdrawal and Return: An Epic Story Pattern in the

Homeric Hymn to Demeter and the Homeric Poems’, in Foley (1994),181–9.

Loucas, E. and Loucas, I. (1986), ‘Un autel de Rhéa-Cybèle et la GrandeDéesse de Phlya’, Latomus 45: 392–404.

Ludwich, K. H. A. (1885), Aristarchs homerische Textkritik nach denFragmenten des Didymos, ii (Leipzig).

Luppe, W. (1973), ‘“Anagyros”: Oder nicht? Zur Identifizierung von Pap.Oxy. 2737’, ZPE 11: 275–88.

MacDowell, D. M. (1971), Aristophanes: Wasps (Oxford).McLennan, G. R. (1977), Callimachus: Hymn to Zeus (Rome).Maehler, H. (2004), Bacchylides: A Selection (Cambridge).Manuwald, G. (2002), ‘Das Singen des kleinen Hermes und des Silen: zum

homerischen Hermeshymnos und zu Vergils Sechster Ekloge’, RhM 145:150–75.

Marinatos, N. and Hägg, R. (1993), Greek Sanctuaries: New Approaches(London).

Marinatos, S. (1951), Platon 3: 228–42.Martin, B. (1605), Variarum lectionum libri quattuor (Paris).Martin, R. P. (1997), ‘Similes and Performance’, in E. Bakker and A. Kahane

(eds.), Written Voices, Spoken Signs (Cambridge, Mass.), 138–66.——(2000), ‘Synchronic Aspects of Homeric Performance: The Evidence of

the Hymn to Apollo’, in A. M. González de Tobia (ed.), Una nueva visiónde la cultura griega antigua hacia el fin del milenio (La Plata), 403–32.

——(2005), ‘Epic as Genre’, in J. M. Foley (ed.), A Companion to AncientEpic (Oxford), 9–19.

Masson, O. (1962), Les fragments du poète Hipponax (Paris).Matthews, V. (1996), Antimachus of Colophon (Leiden).

372 Works Cited

Matthiae, A. (1805), Homeri Hymni et Batrachomyomachia (Lepizig).Meloni, P. (1975), Il profumo dell’immortalità. L’interpretazione patristica di

Cantico 1, 3 (Rome).Merkelbach, R. (1973), ‘Ein Fragment des homerischen Dionysos-Hymnus’,

ZPE 12: 212–15.——(1997), Philologica. Ausgewählte kleine Schriften (Stuttgart and Leipzig).Meyer, H. (1933), Hymnische Stilelemente in der frühgriechischen Dichtung

(Diss., Köln).Miller, A. M. (1979), ‘The “Address to the Delian Maidens” in the Homeric

Hymn to Apollo: Epilogue or Transition?’, TAPA 109: 173–86.——(1986), From Delos to Delphi: A Literary Study of the Homeric Hymn to

Apollo (Leiden).Minton, W. W. (1970), ‘The Proem-hymn of Hesiod’s Theogony’, TAPhA

101: 357–77.Mitchell, L. G. (2007), Panhellenism and the Barbarian in Archaic and

Classical Greece (Swansea).Mombrun, Ph. (2001), ‘Apollon: de l’arc à la lyre’, in Brulé and Vendriès

(2001), 59–96.Mondi, R. (1984), ‘The Ascension of Zeus and the Composition of Hesiod’s

Theogony’, GRBS 25: 325–44.Monro, D. B. (1896), Homeri Opera, Vol. 5 (Oxford).Monsacré, H. (1984), Les larmes d’Achille: Le héro, la femme et la souffrance

dans la poésie d’Homère (Paris).Montana, F. (2006), ‘Aristophanes’, in G. Bastianini, M. Haslam, H. Maehler,

F. Montanari, and C. Römer (eds.), Commentaria et lexica Graeca inpapyris reperta (CLGP), I. 1. 4, Munich.

——(2009), ‘Terpandrean Hypotexts in Aristophanes’, Trends in Classics 1:36–54.

Morand, A.-F. (2001), Études sur les Hymnes orphiques (Leiden).Morgan, C. (1993), ‘The Origins of Pan-Hellenism’, in Marinatos and Hägg

(1993), 18–44.Morgan, K. A. (1993), ‘Pindar the Professional and the Rhetoric of the˚0)ˇ$’, CP 88: 1–15.

Morris, I. and Powell, B. (1997), A New Companion to Homer (Leiden).Morrison, A. D. (2007), Performance and Audience in Pindar’s Sicilian

Victory Odes (London).Most, G. (1981), ‘Callimachus and Herophilus’, Hermes 109: 188–96.——(2006), Hesiod: Theogony, Works and Days, Testimonia (Cambridge,

Mass.).Motte, A. (1986), ‘Mort et renaissance dans les mystères d’Éleusis’, in Jouan

(1986), 71–82.Muellner, L. (1976), The Meaning of Homeric Eukhomai through its

Formulas (Innsbruck).

Works Cited 373

Nagy, G. (1979), The Best of the Achaeans (Baltimore).——(1985), ‘Theognis and Megara: A Poet’s Vision of His City’, in

T. J. Figueira and G. Nagy (eds.), Theognis of Megara: Poetry and thePolis (Baltimore), 22–81.

——(1990a), Pindar’s Homer: The Lyric Possession of an Epic Past(Baltimore). Electronic edition <http://www.press.jhu.edu/books/nagy/PHTL/toc.html> (1997).

——(1990b), Greek Mythology and Poetics (Ithaca).——(1994), ‘Genre and Occasion’, Métis 9–10: 11–25.——(1996a), Poetry as Performance: Homer and Beyond (Cambridge).——(1996b), Homeric Questions (Austin).——(2002), Plato’s Rhapsody and Homer’s Music: The Poetics of the

Panathenaic Festival in Classical Athens (Cambridge).——(2003), Homeric Responses (Austin).——(2004a), ‘L’aède épique en auteur: la tradition des Vies d’Homère’, in

C. Calame and R. Chartier (eds.), Identités d’auteur dans l’Antiquité et latradition européenne (Grenoble), 41–67.

——(2004b), Homer’s Text and Language (Urbana).——(2006), ‘Homer’s Name Revisited’, in G.-J. Pinault and D. Petit (eds.), La

langue poétique indo-européenne: Actes du Colloque de travail de la Sociétédes Études Indo-Européennes Paris, 22–24 octobre 2003 (Leuven), 317–30.

——(2008/2009), Homer the Classic (Washington, DC, and Cambridge,Mass.). The 2008 online version is available at <chs.harvard.edu>. The2009 printed version is distributed by Harvard University Press.

——(2009a), ‘Hesiod and the Ancient Biographical Traditions’, in F.Montanari, A. Rengakos, and Ch. Tsagalis (eds.), Brill Companion toHesiod (Leiden), 271–311.

——(2009b), ‘Perfecting the Hymn in the Homeric Hymn to Apollo’, inL. Athanassaki, R. P. Martin, and J. F. Miller, Apolline Politics and Poetics(Athens), 17–44.

——(2009/2010), Homer the Preclassic (Washington, DC, and Berkeley). The2009 online version is available at <chs.harvard.edu>. The 2010 printedversion is published by the University of California Press.

——(forthcoming), ‘A Second Look at the Poetics of Re-Enactment in Ode13 of Bacchylides’.

Nauck, A. (1867), Lexicon Vindobonense (Petersburg).Nesselrath, H.-G. (1995), ‘Myth, Parody, and Comic Plots: The Birth of Gods

and Middle Comedy’, in G. W. Dobrov (ed.), Beyond Aristophanes.Transition and Diversity in Greek Comedy (Atlanta), 1–27.

——(2010), ‘Vom kleinen Meisterdieb zum vielgeplagten Götterboten.Hermes in den Göttergesprächen Lukians’, in C. Schmitz (ed.), Mythosim Alltag—Alltag imMythos. Die Banalität des Alltags in Unterschiedlichenliterarischen Verwendungskontexten (Munich), 147–59.

374 Works Cited

http://www.press.jhu.edu/books/nagy/PHTL/toc.html

http://www.press.jhu.edu/books/nagy/PHTL/toc.html

Nickel, R. (2003), ‘The Wrath of Demeter: Story Pattern in the Hymn toDemeter’, QUCC 73, 1: 59–82.

Nobili, C. (2008), ‘L’inno omerico a Hermes e le tradizioni poetiche locali’(thesis, Milan).

——(2009), ‘L’inno omerico a Dioniso (Hymn. Hom. VII) e Corinto’, ACME62. 3: 3–35.

——(forthcoming), ‘The Songs of Hermes and the Genre of the HomericHymns: A Chapter in the History of Literary Criticism’.

Norden, E. (1913), Agnostos Theos: Untersuchungen zur Formengeschichtereligiöse Rede (Leipzig).

Notopoulos, J. A. (1962), ‘The Homeric Hymns as Oral Poetry: A Study ofthe Post-Homeric Oral Tradition’, AJPh 83: 337–68.

Nünlist, R. (2004), ‘The Homeric Hymns’, in I. de Jong, R. Nünlist, andA. Bowie (eds.), Narrators, Narratees and Narratives in Ancient GreekLiterature (Leiden), 35–42.

Obbink, D. (1994), ‘A Quotation of the Derveni Papyrus in Philodemus’“On Piety” ’, CronErc 24: 111–35.

Olson, D. (1998), Aristophanes: Peace (Oxford).——(2007), Broken Laughter: Select Fragments of Greek Comedy (Oxford).Otto, W. (1987), Die Götter Griechenlands (Frankfurt am Main).Pace, C. (2004), ‘Il tizzone sotto la cenere: Apol. Rh. 3. 275 ss. e l’Inno

omerico a Hermes’, in R. Pretagostini and E. Dettori (eds.), La culturaellenistica: l’opera letteraria e l’esegesi antica (Rome), 95–111.

Page, D. L. (1955), Sappho and Alcaeus (Oxford).——(1973), Folktales in Homer’s Odyssey (Cambridge).Papadopoulou, I. (2004), ‘Poètes et (Philo)sophoi: Pour une archéologie de la

mimesis’, RPha 24: 3–16.Parker, R. (1991), ‘The Hymn to Demeter and the Homeric Hymns’, G&R 38:

1–17.——(1996), Athenian Religion: A History (Oxford).——(1998), ‘Pleasing Thighs: Reciprocity in Greek Religion’, in Gill,

Postlethwaite, and Seaford (1998), 105–25.——(2005), Greek Polytheism at Athens (Oxford).Parry, A. (1971), The Making of Homeric Verse (Oxford).Patroni, G. (1948), ‘L’inno omerico VI a Dioniso’, Athenaeum 26: 65–75.Pavese, C. (1972), Tradizioni e generi poetici della Grecia arcaica (Rome).——(1991), ‘L’inno rapsodico: analisi tematica degli inni omerici’, in A. C.

Cassio and G. Cerri (eds.), L’inno tra rituale e letteratura nel mondoantico. Atti di un colloquio Napoli 21–24 ottobre 1991 (Rome), 155–78.

Paz de Hos, M. (1998), ‘Los himnos homéricos cortos y las plegariascultuales’, Emérita 66: 49–66.

Peek, W. (1959), ‘Neues von Archilochos’, Philologus 99: 4–50.

Works Cited 375

Pelliccia, H. (2003), ‘Two Points About Rhapsodes’, in M. Finkelberg andG. G. Stroumsa (eds.), Homer, the Bible, and Beyond. Literary andReligious Canons in the Ancient World (Leiden), 98–116.

Penglase, C. (1994), Greek Myths and Mesopotamia (London).Peponi, A. E. (2009), ‘Choreia and Aesthetics in theHomeric Hymn to Apollo:

The Performance of the Delian Maidens (lines 156–64)’, CA 28: 39–70.Petrovic, I. (2011a, forthcoming), ‘Greek Hymns: The True Never-Ending

Stories’.——(2011b, forthcoming), ‘Rhapsodic Hymns and Epyllia’.——(2011c, forthcoming), ‘Callimachus’Hymn to Apollo and Greek Metrical

Sacred Regulations’, in Hellenistica Groningana (Leuven).Pfeiff, K. A. (2002), Homerische Hymnen. Ubertragung, Einführung und

Erläuterungen, Herausgegeben von G. von der Gönna and E. Simon(Tübingen).

Pfeiffer, R. (1968), History of Classical Scholarship: From the Beginning to theEnd of the Hellenistic Age (Oxford).

Pfister, F. (1924), ‘Epiphanie’, RE Suppl. 4: 277–323.Pirenne-Delforge, V. (1989), ‘Conceptions et manifestations du sacré dans

l’Hymne homérique à Aphrodite’, Kernos 2: 187–97.Pironti, G. (2007), Entre ciel et guerre. Figures d’Aphrodite en Grèce ancienne,

Kernos Supplément 18 (Liège).Plantinga, M. (2004), ‘A Parade of Learning: Callimachus’ Hymn to Artemis

(lines 170–268)’, in M. A. Harder, R. F. Regtuit, and G. C. Wakker (eds.),Hellenistica Groningana: Callimachus II (Leuven), 257–77.

Podbielski, H. (1971), La structure de l’hymne homèrique à Aphrodite à lalumière de la tradition littéraire (Wrocław).

Porter, H. N. (1949), ‘Repetition in the Homeric Hymn to Aphrodite’, AJPh70: 249–72.

Postlethwaite, N. (1979), ‘Formula and Formulaic: Some Evidence from theHomeric Hymns’, Phoenix 33: 1–18.

Power, T. (2000), ‘The Parthenoi of Bacchylides 13’, HSCPh 100: 67–81.Preller, L. (1894), Griechische Mythologie, i, 4th edn., rev. C. Robert (Berlin).Preziosi, P. G. (1966), ‘The Homeric Hymn to Aphrodite: An Oral Analysis’,

HSCPh 71: 171–204.Pucci, P. (1992), Oedipus and the Fabrication of the Father: Oedipus

Tyrannus in Modern Criticism and Philosophy (Baltimore and London).——(1995), Ulysse Polutropos: Lectures intertextuelles de l’Iliade et de

l’Odyssée (Lille).——(1998), The Song of the Sirens. Essays on Homer (Lanham).——(2007), Inno alle Muse. Esiodo, Teogonia, 1–115 (Pisa).Race, W. H. (1987), ‘P. Oxy. 2438 and the Order of Pindar’s Works’, RhM

130: 407–10.Radermacher, L. (1931), Der homerische Hermeshymnus (Vienna).

376 Works Cited

Rasmussen, T. and Spivey, N. J. (1986), ‘Dioniso e i pirati nel ToledoMuseum’, Prospettiva 44: 2–8.

Rayor, D. (2004), The Homeric Hymns: A Translation, with Introduction andNotes (Berkeley).

Reece, S. (1999), ‘�HŒ� KæØ��Ø� % EæB (Iliad 20.72): The Modification ofa Traditional Formula’, Glotta 75: 85–106.

Reinhardt, K. (1956), ‘Zum homerischen Aphroditehymnus’, in Festschrift fürBruno Snell zum 60. Geburtstag (Munich), 1–14.

——(1961), Die Ilias und ihr Dichter (Göttingen).Reitzenstein, R. (1893), Epigramm und Skolion. Ein Beitrag zur Geschichte

der alexandrinischen Dichtung (Gießen).Rhodes, P. J. (1981), A Commentary on the Aristotelian Athenaion Politeia

(Oxford).——(1994), Thucydides: History Book III (Warminster).Richardson, N. J. (1974), The Homeric Hymn to Demeter (Oxford).——(1993), The Iliad: A Commentary. Books 21–4 (Cambridge).——(2007), ‘The Homeric Hymn to Hermes’, in Finglass, Collard, and

Richardson (2007), 83–91.——(2010), Three Homeric Hymns: Apollo, Hermes, and Aphrodite

(Cambridge).Rinon, Y. (2006), ‘Mise en abîme and Tragic Signification in the Odyssey’,

Mnemosyne 59: 208–25.Robert, C. (1894), Griechische Mythologie von L. Preller, i: Theogonie und

Götter, 4. Aufl. (Berlin).Rohde, E. (1898), Psyche (Freiburg).Roscher, W. H. (1894), ‘Die Sagen von der Geburt des Pan’, Philologus 53:

362–77.Rossi, L. E. (1971), ‘I generi letterari e le loro leggi scritte e non scritte nella

letteratura classiche’, BICS 18: 69–94.Rousseau, P. (2001), ‘L’intrigue de Zeus’, Europe 865: 120–58.Rudhardt, J. (1991), ‘L’hymne homérique à Aphrodite. Essai d’interprétation’,

MH 48: 8–20.——(1994), ‘Concerning the Homeric Hymn to Demeter’ (translated from

the French ‘A propos de l’hymne homérique à Déméter’ [MH 35: 1–17] byL. Lorch and H. P. Foley), in Foley (1994), 198–211.

Ruhnken, D. (1782), Epistola critica I. in Homeridarum Hymnos etHesiodum, ad L. C. Valckenarium, 2nd edn. (Leiden).

Rutherford, I. (1988), ‘Pindar on the Birth of Apollo’, CQ² 38: 65–75.——(1995), ‘Apollo’s Other Genre: Proclus on NOMOS and His Source’,

CP 90: 354–61.——(2001), Pindar’s Paeans: A Reading of the Fragments with a Survey of the

Genre (Oxford).Saffrey, H. D. (1994), Proclus: hymnes et prières (Paris).

Works Cited 377

Saintillan, D. (1986), ‘Fécondité, mort et mariage. A propos des mythesd’Ouranos et de Déméter’, in Jouan (1986), 51–70.

Sale, W. (1961), ‘The Hyperborean Maidens on Delos’, HThR 54: 75–89.Scheid, J. and Svenbro, J. (1996), The Craft of Zeus. Myths of Weaving and

Fabric (Cambridge, Mass.).Schein, S. L. (1984), The Mortal Hero: An Introduction to Homer’s Iliad

(Berkeley).——(forthcoming), ‘Divine and Human in the Homeric Hymn to Aphrodite’,

in Bouchon, Brillet-Dubois, and Le Meur-Weissman (forthcoming).Schmid, W. (1929), Geschichte der griechischen Literatur (Munich).Schmidt, F. (1922), Die Pinakes des Kallimachos (Berlin).Schneidewin, F. W. (1848), ‘Anmerkungen zum Hymnus auf Hermes’,

Philologus 3: 659–700.Schwabl, H. (1963), ‘Aufbau und Struktur des Prooimions der hesiodischen

Theogonie’, Hermes 91: 385–415.Scott, J. A. (1921), The Unity of Homer (Berkeley).Scott, M. (1983), ‘Charis in the Homeric Hymns’, AClass 26: 1–13.Segal, C. (1974), ‘The Homeric Hymn to Aphrodite: A Structuralist

Approach’, CW 67: 205–12.——(1981), ‘Orality, Repetition and Formulaic Artistry in the Homeric“Hymn to Demeter” ’, in C. Brillante, M. Cantilena, and C. O. Pavese(eds.), I Poemi Epici Rapsodici non Omerici e la Tradizione Orale (Padua),107–62.

——(1997), Dionysiac Poetics and Euripides’ Bacchae, 2nd edn. (Princeton).Sens, A. (1997), Theocritus: Dioscuri (Idyll 22), Hypomnemata 114

(Göttingen).Severyns, A. (1928), Le cycle épique dans l’école d’Aristarque (Paris).Shapiro, H. A. (1992), ‘“Mousikoi Agones”: Music and Poetry at the

Panathenaia’, in J. Neils (ed.), Goddesses and Polis: The PanathenaicFestival in Ancient Athens (Princeton), 53–75.

Shelmerdine, S. C. (1984), ‘Hermes and the Tortoise. A Prelude to Cult’,GRBS 25: 201–8.

Sikes, E. E. (1940), ‘The Humour of Homer’, CR 54: 121–7.Sinos, R. H. (1993), ‘Divine Selection. Epiphany and Politics in Archaic

Greece’, in C. Dougherty and L. Kurke (eds.), Cultural Poetics in ArchaicGreece (Cambridge), 73–91.

Skafte Jensen, M. et al. (1999), ‘Dividing Homer’, SO 74: 5–91.Slatkin, L. (1991), The Power of Thetis (Berkeley).Smith, P. M. (1981a), Nursling of Mortality. A Study of the Homeric Hymn to

Aphrodite (Frankfurt).——(1981b), ‘Aineiadai as Patrons of Iliad XX and the Homeric Hymn to

Aphrodite’, HSCPh 85: 17–58.

378 Works Cited

Snell, B. (1966), ‘Dionysos oder Hephaistos?’, in Gesammelte Schriften(Göttingen), 102–4.

Solmsen, F. (1980), ‘Empedocles’ Hymn to Apollo’, Phronesis 25: 219–27.——(1982), ‘The Earliest Stages in the History of Hesiod’s Text’, HSCPh 86:

1–31.Sowa, C. A. (1984), Traditional Themes and the Homeric Hymns (Chicago).Spanoudakis, K. (2002), Philitas of Cos (Leiden).Sparshott, F. E. (1963), ‘Homeric Hymn 7.44–8’, CR 13: 1–2.Stebbins, E. B. (1929), The Dolphin in the Literature and Art of Greece and

Rome (Baltimore).Stehle, E. (1996), Performance and Gender in Ancient Greece: Nondramatic

Poetry in its Setting (Princeton).Strid, O. (1999), Die Dryoper: eine Untersuchung der Überlieferung

(Uppsala).Taddei, A. (2007), ‘Mneme e terpsis. Rammemorazione e rituale nell’ Inno

omerico ad Apollo’, in L. Marrucci and A. Taddei (eds.), Polivalenzeepiche. Contributi di antropologia storica (Pisa), 79–93.

Thalmann, W. G. (1984), Conventions of Form and Thought in Early GreekEpic Poetry (Baltimore).

——(1991), ‘Review of Clay (1989)’, CP 86: 144–7.Thomas, O. R. H. (2009), ‘A Commentary on the Homeric Hymn to Hermes

184–396’ (D.Phil. thesis, Oxford).Thomas, R. (1992), Literacy and Orality in Ancient Greece (Cambridge).Torres-Guerra, J. B. (2003), ‘Die Anordnung der homerischen Hymnen’,

Philologus 147: 3–12.Traill, D. (1998), ‘Callimachus’ Singing Sea (Hymn 2. 106)’, CP 93: 215–22.Treu, M. (1952), Alkaios, 1. Aufl. (Munich; 2. Aufl.: 1963).Tsagalis, C. (2004), Epic Grief: Personal Laments in Homer’s Iliad (Berlin).Turkeltaub, D. W. (2003), ‘The God’s Radiance Manifest: An Examination of

the Narrative Pattern Underlying the Homeric Divine Epiphany Scenes’(Diss., Cornell).

——(2007), ‘Perceiving Iliadic Gods’, HSCPh 103: 51–82.Tzifopoulos, Y. (2000), ‘Hermes and Apollo at Onchestos in the Homeric

Hymn to Hermes: The Poetics and Performance of PoeticCommunication’, Mnemosyne 53: 148–63.

Vamvouri-Ruffy, M. (2004), La fabrique du divin (Liège).Van der Ben, N. (1981), ‘De Homerische Aphrodite-hymne 2: Een

interpretatie van het gedicht’, Lampas 14: 69–107.Van Nortwick, T. (1975), ‘The “Homeric Hymn to Hermes”. A Study in

Early Greek Hexameter Style’ (Diss., Stanford).Vergados, A. (2007), ‘A Commentary on the Homeric Hymn to Hermes’

(Diss., Virginia).

Works Cited 379

——(forthcoming), A Commentary on the Homeric Hymn to Hermes (DeGruyter).

Vernant J.-P. (1986), ‘Le Dionysos masqué des Bacchantes d’Euripide’, inJ.-P. Vernant and P. Vidal Naquet (eds.), Mythe et tragédie en Grèceancienne II (Paris), 237–70.

——(1992), ‘La belle mort et le cadavre outragé’, in G. Gnoli and J.-P.Vernant (eds.), La mort, les morts dans les sociétés anciennes (Paris),45–76.

Versnel, H. S. (1987), ‘What Did Ancient Man See When He Saw a God?Some Reflections on Greco-Roman Epiphany’, in D. van der Plas (ed.),Effigies Dei. Essays on the History of Religions (Leiden), 42–55.

Vian, F. (2000), ‘Dionysos et les pirates tyrrhéniens chez Nonnos’, inM. Cannatà Fera and S. Grandolini (eds.), Poesia e religione in Grecia(Perugia), 683–92.

Villarrubia, A. (1997), ‘Una lectura del Himno Homérico a Pan’, Habis 28:7–13.

Von der Mühll, P. (1952), Kritisches Hypomnema zur Ilias (Basel).Von Fritz, K. (1956), ‘Das Proömium der hesiodischen Theogonie’, in

Festschrift für Bruno Snell zum 60. Geburtstag (Munich), 29–45 ¼E. Heitsch (ed.), Hesiod (Darmstadt), 295–315.

Wade-Gery, H. T. (1936), ‘Kynaithos’, in Greek Poetry and Life: EssaysPresented to Gilbert Murray on his Seventieth Birthday (Oxford), 56–78.

Wagman, R. (2000), L’inno epidaurico a Pan: il culto di Pan a Epidauro(Pisa).

Walcot, P. (1979), ‘Cattle Raiding, Heroic Tradition, and Ritual: The GreekEvidence’, HR 18: 326–51.

Wecklein, N. (1920), Die Homerischen Hymnen und die griechischenTragiker (Munich).

West, M. L. (1966), Hesiod: Theogony (Oxford).——(1970), ‘The Eighth Homeric Hymn and Proclus’, CQ² 20: 300–4.——(1974), Studies in Greek Elegy and Iambus (Berlin).——(1975), ‘Cynaethus’ Hymn to Apollo’, CQ² 25: 161–70.——(1982), Greek Metre (Oxford).——(1983), The Orphic Poems (Oxford).——(1989), ‘An Unrecognized Injunctive Usage in Greek’, Glotta 67: 135–8.——(1992), Ancient Greek Music (Oxford).——(1995), ‘The Date of the “Iliad” ’, MH 52: 203–19.——(1999), ‘The Invention of Homer’, CQ² 49: 364–82.——(2000), ‘The Gardens of Alcinous and the Oral Dictated Text Theory’,

Acta Antiqua Academiae Hungaricae, 479–88.——(2001a), ‘The Fragmentary Homeric Hymn to Dionysus’, ZPE 134: 1–11.——(2001b), Studies in the Text and Transmission of the Iliad (Munich).

380 Works Cited

——(2002a), ‘The View from Lesbos’, in M. Reichel and A. Rengakos (eds.),Epea Pteroenta: Beiträge zur Homerforschung (Stuttgart), 207–19.

——(2002b), ‘“Eumelos”: A Corinthian Epic Cycle?’, JHS 122: 109–53.——(2003), ‘“Iliad” and “Aethiopis” ’, CQ² 53: 1–14.——(2007), Indo-European Poetry and Myth (Oxford).West, S. (1967), The Ptolemaic Papyri of Homer (Cologne).Wheeler, G. (2002), ‘Sing, Muse: The Introit from Homer to Apollonius’,

CQ² 52: 33–49.Wilamowitz-Moellendorff, U. von (1895), ‘Hephaistos’, NGG 1895. 217–45¼ Kleine Schriften v(2) (Berlin 1937), 5–35.

——(1916a), Die Ilias und Homer (Berlin).——(1916b), Vitae Homeri et Hesiodi (Bonn).Willcock, M. (1970), ‘Some Aspects of the Gods in the Iliad’, BICS 17: 1–10.Williams, F. (1978), Callimachus: Hymn to Apollo (Oxford).Wilson, N. G. (1974), ‘A Puzzle in Stemmatic Theory Solved’, RHT 4:

139–42.Wolf, F. A. (1807), Homeri et Homeridarum Opera et Reliquiae, Vol. 5

(Leipzig).Zanetto, G. (1996), Inni Omerici (Milan).Ziegler, K. (1911), ‘Das Proömium der Werke und Tage Hesiods’, Archiv für

Religionswissenschaft 14: 393–405.Zuntz, G. (2005), Griechische philosophische Hymnen, Studien und Texte zu

Antike und Christentum, Vol. 35 (Tübingen).

Works Cited 381


Index Locorum

ACHAEUS (TrGF)fr.20 F 16b–17: 31, 228 n. 84

ACHILLES TATIUS

8. 6. 6–7: 156

AELIAN

NA12. 22: 148 n. 44

AESCHYLUS

Ag.501: 91 n. 29Cho.447: 170856: 91 n. 29Eu.1–19: 19917–19: 80 n. 6958: 91 n. 29616–18: 80 n. 69827–8: 271 n. 27Pers.448–9: 171Sept.276: 91 n. 29

AGLAOSTHENES (FGrH)fr.499 F 3: 144 n. 31

ALCAEUS (Voigt)fr.34: 39 n. 22, 195 n. 90, 201307c: 66 n. 20, 80 n. 69, 201308: 39 n. 22, 201349a–e: 31, 40, 228 n. 84

ALCMAN (PMG)fr.135: 170 n. 58

ALEXIS (PCG)fr.9. 8–10: 95 n. 52

ANTHOLOGIA GRAECA

6. 96: 171 n. 65 (Erycius)8. 226: 156 n. 129. 82: 144 n. 31

ANTONINUS LIBERALIS

23: 167 n. 43

APOLLODORUS (FGrH)fr.244 F 135: 169 n. 50

PS.-APOLLODORUS

1. 5. 1: 513. 10. 4: 265 n. 183. 37–8: 135 n. 5, 1433. 113: 93 n. 41

APOLLONIUS RHODIUS

1. 1–2: 1931. 803: 1951. 363–401: 1941. 457–9: 95 n. 521. 460–518: 1942. 703–13: 185 n. 43, 1942. 707–8: 64 n. 142. 1211: 176 n. 53. 114–66: 1953. 275–97: 1953. 1191–325: 1953. 1218: 152 n. 54. 869–84: 194

ARATUS

Ph.263: 196 n. 96268–9: 196fr. (SH)109: 195 n. 91

ARISTIDES

Or.34–5: 61, 176 n. 541. 6: 31 n. 5

ARISTOPHANES

Av.575: 177, 199909–14: 1991515–24: 94 n. 421565–693: 94 n. 43Pax.363–5: 94365: 99 n. 65191–4: 93201–2: 93400–2: 94924: 94Eq.639: 97 n. 561016: 199Pl.1118–32: 93 n. 411136–8: 93 n. 411139–70: 94V.1308–13: 96 n. 52fr. (PCG)56. 19–27: 17856. 26–7: 177 n. 8

ARISTOTLE

HA632b 20–3: 157Ph.227a 28–9: 301 n. 62Po.1448a: 94 n. 441449a: 94 n. 44fr. 347. 15: 301 n. 62

ATHENAEUS

22b: 176 n. 5, 210 n. 17653b: 176 n. 5

BACCHYLIDES

fr.5. 16–30: 200

CALLIMACHUS

Hy.1. 4–8: 1791. 4–10: 40–11. 47: 1901. 55–6: 1911. 67: 1901. 87–8: 1902. 4–5: 181–2

2. 11: 1862. 21: 1842. 24: 742. 25: 1842. 30: 188 n. 562. 47–9: 184 n. 382. 49: 1672. 65–96: 1842. 80: 1842. 97: 1862. 97–104: 183–52. 103: 64 n. 142. 105–13: 187–82. 209–11: 1823. 4–40: 1913. 26–8: 1923. 66–71: 1913. 88–9: 158 n. 213. 122: 1913. 140–69: 1914. 49–55: 182 n. 304. 55–204: 754. 82–5: 1934. 90–3: 185 n. 414. 258: 182 n. 334. 325: 193 n. 765. 19–22: 192–36. 6–16: 188–96. 17–23: 189–906. 37: 1906. 58: 1906. 59–61: 180 n. 256. 66–7: 1906. 70–1: 180fr. (Pf.)24. 13: 162 n. 3488: 185429–6 (Pinakes): 179685: 169 n. 50

CASTORION (SH)fr.310. 3: 160 n. 26

CERTAMEN (Allen)54–214: 295–6, 29972–204: 30394–101: 303 n. 6797–8: 237 n. 22176–210: 304255: 299276–314: 294315–22: 17, 21, 294

384 Index Locorum

CHOERILUS (Bernabé)fr.9: 143 n. 28

CLEMENT OF ALEXANDRIA

Protr.2. 17. 2: 228

CYPRIA (Bernabé)fr.4–5: 110, 114

DIODORUS SICULUS

1. 15. 7: 30, 176 n. 53. 66. 3: 30, 42 n. 25, 176 n. 5, 2254. 2. 4: 30, 176 n. 510. 19. 6: 135 n. 5

EPICHARMUS (PCG i)fr.51: 31 n. 6, 228 n. 84

ERYCIUS (Anthologia Graeca)6. 96: 171 n. 65

EUMELUS (Bernabé)fr.16: 211 n. 23

EUPHORION (CA)fr.14: 170 n. 57

EURIPIDES

Ba.64–169: 19883–7: 147 n. 42141–3: 147 n. 38144–50: 147 n. 39159–65: 149 n. 48614–59: 141 n. 23642–6: 141 n. 24695–711: 147 n. 38770–4: 141 n. 25859–61: 142 n. 27Cyc.11–32: 143Hel.171: 156 n. 16190: 166 n. 41676–8: 110 n. 251301–68: 198HF687–9: 314687–95: 310

Hipp.1268–81: 198IA1294–8: 110 n. 25Ion493–4: 171 n. 61499–500: 156 n. 14IT508: 91 n. 291234–57: 64 n. 14fr. (TrGF)863: 158 n. 21

HERODOTUS

1. 23–4: 148 n. 461. 64. 2: 285 n. 141. 132: 2292. 49–50: 338 n. 104. 35: 223 n. 664. 145: 135 n. 56. 105: 15, 169 n. 516. 137–40: 135 n. 5

HECATAEUS (FGrH)fr.18–19: 135 n. 5

HERACLITUS (D.–K.)fr.22 B 93: 353

HESIOD

Op.158–66: 244 n. 44256–62: 192 n. 73427: 90 n. 23650–9: 9 n. 37, 296656–7: 212 n. 24657: 297Th.1–115: 210–12, 239, 306–710: 160 n. 2744–50: 23972–4: 25694–5: 31394–7: 347 n. 31100: 199104: 313112–13: 256116–19: 257–8126–7: 258–9128: 270 n. 24, 279138: 260

Index Locorum 385

HESIOD (cont.)139: 264149: 264154: 260156–8: 273159: 257161–2: 273164: 265170–2: 266179: 257179–81: 273191: 112 n. 34201–6: 111 n. 28205: 117 n. 53205–6: 218270–336: 260459–62: 273492–3: 273501–6: 273617–23: 273629–721: 269693: 176693–6: 269702–5: 270820–52: 261820–80: 74820–900: 269837–8: 261842–58: 270 n. 24853–68: 278881–5: 244, 256886–900: 263–4896: 268897–9: 262898: 268912–14: 244938–9: 224 n. 71965: 3161008–10: 105 n. 21015–16: 135 n. 5

PS.-HESIOD

fr.30: 265 n. 1854a: 265 n. 18148(a): 35 n. 17204: 244283: 138 n. 17357: 289, 295, 301, 317375: 250 n. 66Sc.423: 142 n. 26

HIMERIUS

Or.48: 66, 80 n. 69

HIPPONAX (IEG)fr.3: 92–33a: 92–323: 92 n. 3232: 91, 9334: 92 n. 3035: 9247: 9379. 9–10: 93128–9a: 92 n. 32177: 91

HOMER

Il.1. 47: 771. 49: 771. 199–200: 100 n. 681. 396: 2491. 433–7: 681. 472–4: 274 n. 311. 484–94: 178 n. 131. 524–7: 391. 528–30: 381. 539–59: 2661. 571–600: 97, 2491. 590–4: 36–71. 593: 35 n. 171. 599: 246 n. 512. 103–4: 338 n. 132. 484: 2922. 820–1: 105 n. 23. 54: 116 n. 513. 136: 119 n. 593. 373–82: 34, 1303. 396–8: 100 n. 684. 74–9: 684. 215: 115 n. 474. 420: 142 n. 265. 119: 91 n. 295. 263–71: 1305. 311–430: 34, 129–305. 440–2: 76–75. 778: 1775. 890: 2506. 123: 1176. 123–43: 149 n. 506. 128–9: 1176. 389: 162

386 Index Locorum

6. 476–81: 128–98. 5–40: 2658. 18–27: 2498. 350–484: 265–68. 424: 2658. 477–83: 24912. 13–33: 244 n. 4412. 396: 142 n. 2613. 68–75: 84 n. 8, 100 n. 6813. 461: 13014. 172: 109 n. 2314. 178–9: 111 n. 3314. 178–85: 109 n. 2314. 185: 11414. 211: 11114. 215: 11114. 216: 11714. 222–3: 11114. 230: 3514. 238–40: 3514. 256–61: 36–714. 267–8: 36–714. 329–4014. 346–51: 24815. 18–22: 24915. 18–30: 3715. 158–217: 24715. 286: 84 n. 1015. 361–6: 7715. 364: 16816. 55: 126 n. 8616. 130–44: 11516. 163: 176 n. 716. 179–83: 344 n. 2616. 180–6: 99 n. 6416. 431–9: 26616. 717–18: 119 n. 6116. 788–817: 7716. 793–804: 11516. 829: 91 n. 29, 121 n. 6816. 844–7: 12216. 852–4: 12317. 35: 91 n. 2917. 209: 3817. 450: 91 n. 2918. 382–3: 3418. 394–405: 34–518. 400–1: 11418. 610: 11419. 9: 12019. 307: 126 n. 8619. 369–83: 114 n. 43

20. 4–75: 24820. 8–9: 152 n. 520. 44–6: 114 n. 4420. 200–9: 11920. 293: 13120. 298: 13120. 307–8: 13, 12720. 344: 84 n. 1020. 431–3: 11921. 121: 91 n. 2921. 150: 11721. 319: 17621. 385–514: 24821. 387: 142 n. 2621. 461–7: 7721. 497–501: 88 n. 2022. 7–1022. 35–91: 114 n. 4422. 127–8: 117 n. 5322. 131–6: 114 n. 4422. 178–81: 26622. 296–303: 12222. 330: 121 n. 6822. 358–60: 12322. 369–75: 12524. 247: 157 n. 1924. 258–9: 13024. 347–8: 99, 100 n. 6924. 358–467: 99–10024. 370–1: 9924. 397: 9924. 439–40: 9924. 445: 99

Od.1. 1: 2921. 325–7: 173. 71–4: 683. 88: 283 n. 73. 184: 283 n. 73. 371–2: 100 n. 684. 844: 2266. 123–4: 152 n. 58. 62–96: 178. 73–82: 2518. 266–366: 33, 37–8, 87–8, 97–8,

249–50, 330–28. 283–4: 35 n. 178. 321–32: 2198. 333–43: 112 n. 378. 335–7: 87–88. 339–43: 88

Index Locorum 387

HOMER (cont.)8. 362: 109 n. 238. 364–5: 109 n. 238. 366: 109 n. 238. 429: 3308. 492–3: 237 n. 238. 499: 18, 2388. 499–535: 17, 2519. 116: 2269. 336: 166 n. 429. 469: 166 n. 4210. 277–9: 10011. 245: 11512. 14: 170 n. 5813. 250: 270 n. 2514. 423: 91 n. 2914. 527: 166 n. 4115. 295: 178 n. 1315. 295–300: 6816. 162–3: 84 n. 919. 36: 84 n. 1019. 86–8: 273 n. 2919. 518–22: 168–922. 286: 91 n. 2923. 59: 91 n. 29

HOMERIC HYMNS

Hy. 1 (Dionysus)A 1–10: 176 n. 5A 1–24: 226A 2–8: 41A 3: 30A 8: 30, 198A 11–14: 42D 4–6: 38D 6–12: 227D 7: 38 n. 19D 8: 30, 63D 11–12: 53

Hy. 2 (Demeter)1: 140 n. 191–21: 442–11: 348 n. 343: 2429: 137 n. 814: 137 n. 820–9: 4740–3: 4543–4: 19849–50: 18851–89: 4782–7: 47

91–304: 45–4692: 137101: 85188–9: 83, 190, 202198–204: 245200–1: 188202–4: 198237–302: 194268: 83268–9: 202270–4: 50281–91: 245284–300: 47293–304: 50304: 198305–12: 198345–69: 342366: 342424: 198460–90: 242473–82: 50480–2: 243480–9: 343 n. 25484: 137 n. 11486–7: 203494–5: 203, 329495: 18305–489: 45–6310–11: 190314–56: 47360–9: 47441–69: 47450: 176450–6: 56470–3: 56470–95: 62480–94: 53481–2: 46490: 63

Hy. 3 (Apollo)1: 1931–13: 45, 69, 74, 272–41–139: 621–181: 60186–8: 2742: 1522–9: 2033: 1844–9: 20314–18: 161 n. 2819: 69, 326, 339

388 Index Locorum

30–4: 6445–9: 4847: 22253–5: 6655–64: 18456–60: 69, 22259–60: 7062–78: 7163–73: 24566–78: 7567–9: 74, 273–475–6: 6979–82: 69–792–106: 4895–9: 72 n. 4897–104: 184, 222114: 177, 199117: 182119: 182 n. 33, 202127–55: 350–1131–2: 73–4, 137132: 75, 80, 242, 278136–9: 179140–64: 64, 191146: 163 n. 36146–76: 61, 238–9, 252, 322, 351–2149: 318, 321149–50: 323151: 231156–75: 165158–9: 314158–61: 314–15162: 186 n. 46162–4: 185–6162–78: 61, 63, 186–7, 200, 204, 223–4,

305–7169: 187169–72: 298171–8: 16–17174: 63177: 63177–8: 312–14179–546: 62, 352–3182–5: 204182–546: 60–1186–7: 78186–8: 112 n. 36186–206: 69, 203189: 316189–93: 315190–1: 239190–3: 78

197–8: 203205: 80214–15: 78216–99: 64220–1: 70229–38: 66, 202244–5: 68247–9: 70247–53: 65, 75, 77248: 78249–50: 78252: 80256–74: 245285–99: 12288: 78300–74: 184, 199, 275–6305–55: 48, 62, 73 n. 49, 74322–30: 276326–7: 277332–9: 249337–42: 276341: 176357: 183 n. 35357–74: 76, 278366–74: 75, 183370: 91 n. 29371–4: 68375–87: 66, 75, 245381: 78382–7: 278388–544: 68399–404: 245400: 82, 85421–7: 68425: 178 n. 13427: 80433: 80440–5: 82440–70: 68443: 199449–50: 83452–5: 68475–85: 77478–9: 78479: 341480: 83484: 80490–6: 66493–6: 68503–7: 68, 178 n. 13514–19: 183 n. 34522–3: 78

Index Locorum 389

HOMERIC HYMNS (cont.)528–30: 245528–36: 70529–30: 66531–44: 76532–7: 77535–70: 71537–43: 76538–44: 77540–3: 12546: 336 n. 8

Hy. 4 (Hermes)1: 1031–4: 161 n. 28, 3381–54: 348–96–9: 1678: 22410: 137, 13913–15: 16713–24: 22415: 9217–18: 19019: 196 n. 9620: 20221: 19021–2: 16724–62: 161 n. 2931: 10234–5: 9540–2: 22541–51: 24242: 167 n. 45, 22543–6: 85 n. 1247–51: 9554–63: 103–455–6: 9557–8: 16757–62: 244, 249 n. 6264: 56, 9367–9: 9977–8: 15890–3: 84, 89 n. 2497–100: 9999: 89111: 242112–41: 161 n. 29, 194115–37: 56124–6: 102127–9: 21141: 89, 99150–61: 96

152: 168155: 99160: 96174–81: 96175: 92176–81: 13, 91184–200: 84189–212: 195 n. 93190–200: 95194–6: 91196: 84219: 84231–2: 84–5237: 84–5242–3: 84254–9: 95261–77: 95265–73: 13278: 84–5281: 86287: 93290: 92, 97292: 91294–6: 96296: 191322–4: 96341: 99342–3: 84 n. 8343: 85368–86: 95389: 86, 225389–95: 242407: 84408: 92409–14: 83 n. 5414: 85415–16: 84–5417–35: 194420: 86425–33: 245, 249 n. 62426–578: 349–50429: 89436–578: 191437: 89447–8: 13454: 102455: 85469–72: 80 n. 69475–88: 103496–9: 242507–10: 102511–12: 13, 168

390 Index Locorum

515: 201 n. 121531–40: 80 n. 69541–9: 163 n. 36577–8: 163 n. 36

Hy. 5 (Aphrodite)1–6: 138 n. 152: 152, 1953–5: 1987–33: 129, 19319: 182 n. 3320: 19225–32: 39 n. 22, 192, 20031–2: 20236–44: 126 n. 8845–7: 137 n. 845–52: 124, 138 n. 15, 22058–65: 22, 109 n. 23, 11369–74 221 n. 6270–3: 11675–82: 34477: 116, 12981–90: 113–1482: 11784–91: 11391: 117–1892–9: 117–1898–9: 152 n. 5100–2: 22103–6: 128–9104: 121111: 118113: 119117–25: 344127: 127127–40: 118143: 117, 120144: 117145–54: 117, 120157–60: 120162–6: 113164: 115166: 120168–9: 121172–9: 83173–4: 190 n. 65, 202181–3: 245182: 115182–3: 121184: 118186–90: 121193–6: 121

196–7: 127197–8: 14198–9: 126239–46: 123, 243247: 122247–55: 125–6, 138 n. 15, 220254: 122264–72: 193 n. 76278: 121278–9: 129285–6: 83286: 124290: 126292: 129293: 18, 328, 336 n. 8

Hy. 6 (Aphrodite)1: 1125: 1125–15: 21919–21: 17, 62, 219, 236, 336, 355

Hy. 7 (Dionysus)1: 103 n. 801–2: 139, 193 n. 802: 1403: 1407–8: 14, 14311: 14013–15: 14115–24: 8319–21: 14129: 14431: 14135–45: 83, 18036–7: 14737–57: 14242–50: 180 n. 2651 143 n. 2853: 14856: 139, 347 n. 3256–7: 8357: 14358: 13958–9: 149

Hy. 9 (Artemis)1: 140 n. 197–9: 3579: 18, 328, 336 n. 8

Hy. 10 (Aphrodite)5: 355

Index Locorum 391

HOMERIC HYMNS (cont.)Hy. 11 (Athena)4: 2625: 355

Hy. 13 (Demeter)1–3: 334–73: 62, 342 n. 21

Hy. 14 (Mother of the Gods)2: 140 n. 19

Hy. 16 (Asclepius)1–3: 178, 339

Hy. 17 (Dioscuri)1: 140 n. 191–6: 3393–5: 203

Hy. 18 (Hermes)1: 1971–5: 33711: 18, 328

Hy. 19 (Pan)1: 103 n. 80, 140 n. 19, 1511–2: 161, 1641–9: 1672–3: 152–3, 1582–27: 154–5, 3483: 1704–7: 159–608–9: 1588–14: 15314–18: 155–615: 168, 17016–18: 168–919–26: 157–832–3: 163 n. 36, 16735–7: 16136–7: 16737: 13938: 167, 17038–9: 162–339: 16343: 167 n. 4545–7: 161–247: 16424: 16027–35: 165–628–9: 16038–9: 83 n. 4, 13940–1: 163

44–6: 88 n. 20, 13948: 6348–9: 163–4

Hy. 20 (Hephaestus)1: 140 n. 198: 63

Hy. 21 (Apollo)1–5: 3451: 178, 182 n. 305: 63

Hy. 22 (Poseidon)1: 103 n. 80

Hy. 23 (Zeus)1–4: 3254: 63

Hy. 24 (Hestia)4–5: 61, 342

Hy. 25 (Apollo, the Muses, Zeus)1–7: 346–76: 356

Hy. 26 (Dionysus)1–5: 3391–10: 3473–4: 195 n. 9111–13: 6212–13: 236

Hy. 27 (Artemis)4–20: 203, 344–5

Hy. 28 (Athena)1–18: 267–713–16: 203 n. 132

Hy. 29 (Hestia)7: 338 n. 13

Hy. 30 (Mother of All)2–16: 3437–8: 20317–18: 35518–19: 20319: 356

Hy. 31 (Helios)1: 140 n. 1918–20: 237, 316

Hy. 32 (Selene)18–19: 193, 207 n. 6, 237, 31620: 199

392 Index Locorum

Hy. 33 (Dioscuri)1: 103 n. 80, 140 n. 191–3: 151 n. 11–5: 3394–5: 203

HORACE

Carm.1. 10. 9–12: 201 n. 119Od.4. 2: 180 n. 25

HYGINUS

Astr.2. 17: 144 n. 31Fab.134: 144 n. 31140: 64 n. 14166: 31 n. 5IEG (Adespota)fr.27. 1–6: 95 n. 52

ISOCRATES

Paneg.159: 320

PS.-LIBANIUS

Narr.30. 1 (viii. 38–9 Foerster): 31 n. 5

LINUS (Bernabé)fr.78–93: 217

LUCIAN

Salt.22: 144 n. 31

PS.-LUCIAN

Dem. Enc.19: 180 n. 25

LXXPsalm 141: 357

MACROBIUS

Sat.1. 17. 52: 64 n. 14

MARGITES (Bernabé)1. 2: 199

MEGASTHENES (FGrH)

fr.715 F 27: 170 n. 57

MENODOTUS (FGrH)fr.541 F 1: 33

MUSAEUS (Bernabé)fr.46–53: 21760 F: 217

NICANDER

Alex.130: 177 n. 8560: 167 n. 43

NICOCLES (FGrH)376 F 8: 290

NONNUS

D.4. 363: 166 n. 4244. 240–9: 144 n. 3145. 105–69: 144 n. 3145. 683–92: 144 n. 31

OPPIAN

H.1. 650: 144 n. 31

ORPHICA

A.1199–202: 43, 226Hy.11. 6: 156 n. 15fr. (Bernabé)8–25: 212–15

OVID

Met.1. 689–712: 156 n. 163. 597–691: 144

PAUSANIAS

1. 20. 3: 31 n. 5, 42, 176 n. 51. 38. 2–3: 176 n. 5, 2161. 39. 1: 2164. 1. 5: 2164. 30. 4: 176 n. 55. 7. 8: 223 n. 667. 21. 7: 3388. 21. 3: 216 n. 398. 37. 9: 2169. 27. 2: 216 n. 39, 223 n. 66

Index Locorum 393

PAUSANIAS (cont.)9. 30. 12: 2169. 31. 9: 21610. 5. 7: 223 n. 6610. 5. 8: 215 n. 3810. 37. 4: 176 n. 5

PHERECYDES (FGrH)fr.3 F 131: 167 n. 43

PHERECRATES (PCG)fr.155 (Cheiron): 102 n. 77

PHILICUS (SH)fr.676–80: 196

PHILOCHORUS (FGrH)fr.308. 100: 135 n. 5328 F 185: 216328 F 212: 289

PHILODAMUS (Furley-Bremer)1–5: 340

PHILODEMUS

Piet. p. 87 Schober: 176 n. 5Piet. fr. 19 (Henrichs, 1975b)

PHILOSTRATUS

Her.693: 216

PHILOSTRATUS JUNIOR

Im.1. 19: 144 n. 31

PINDAR

N.2. 1–5: 18, 210 n. 17, 238, 3243. 14: 1783. 28: 170O.6: 301 n. 627. 35–8: 262P.1. 4: 239 n. 3112: 301 n. 62fr.7b: 199–20029: 179 n. 23

52m: 31189a: 179 n. 2395. 2: 171 n. 6195–6: 152 n. 396: 160 n. 2697: 156 n. 12124: 143 n. 28140b 16–17: 148 n. 45236: 15, 143, 148 n. 43283: 31, 228 n. 84Scholia ad N. 2: 71–2, 177 n. 9, 223 n. 67,

284–91, 301–3

PLATO

Cra.420–a–b: 337Euthphr.8b 4: 227Ion530a: 320530d: 288 n. 22Lg.700b: 197Prt.316d 7–9: 229 n. 89R.378d: 31, 40, 197, 227 n. 82600d–e: 299607a: : 197

PS.-PLATO

Hipparch.228b–c: 287, 300

PLUTARCH

Mor.16d: 158 n. 21365a 6: 228 n. 87502–3: 99 n. 62984c: 148 n. 441132d: 209 n. 141133c: 209 n. 14Nic.3. 5–7: 285, 309, 322Per.13. 9–11: 319Thes.21. 3: 322

PMG936: 160 n. 26, 163 n. 35,

171 n. 61

394 Index Locorum

1012: 39 n. 22

POSIDONIUS (E.–K.)fr.62: 152 n. 3

PRATINUS (TrGF)fr.6: 170 n. 58

RHIANUS (CA)fr.10: 167

SAPPHO (Voigt)fr.44. 33: 159 n. 2444A. 7–8: 3944A. 1–11: 192, 200–1

SEXTUS EMPIRICUS

M.9. 18: 215 n. 37

SENECA

Oed.449–66: 144 n. 31

SERVIUS

A.1. 57: 144 n. 313. 73: 64 n. 14Ecl.4. 62: 31 n. 5

SLGS264: 39 n. 22, 201

SIMONIDES

fr.T 47 a–b (Campbell): 150519 (PMG): 310

SOPHOCLES

Aj.695: 171 n. 61OC484: 91 n. 29OT249: 91 n. 29Trach.809: 91 n. 29fr. (TrGF)926: 152 n. 3

STESICHORUS (PMG)fr.64: 239

STRABO

8. 3. 26: 178 n. 1310. 1. 9: 178 n. 1314. 6. 3: 355 n. 49

TELECLIDES (PCG)fr.35: 93 n. 41

THEOCRITUS

Id.17: 19522: 19522. 223: 5824. 1–63: 195

THEOGNIS

358: 91 n. 29944: 91 n. 29

THUCYDIDES

3. 104: 11, 18. 61, 177 n. 8, 196–7, 238,282–6, 288, 307–9, 312–13, 317–25,327

4. 109: 135 n. 5

TIMAEUS (FGrH)566 F 149: 143 n. 28

TITANOMACHY (Bernabé)fr.11: 138

VERGIL

A.10. 551: 167 n. 43

VITA HOMERI HERODOTEA (Allen)379–516: 293

XENOPHON

Cyn.11. 1: 158 n. 21

INSCRIPTIONS

SEG8. 548. 2515. 517 (Mnesiepes Inscription): 88–90

PAPYRUS FRAGMENTS

P. Herc.

Index Locorum 395

PAPYRUS FRAGMENTS (cont.)1428: 215 n. 37P. OXY.670: 33, 227 n. 822438: 180 n. 252737: 177 n. 8, 178

4667: 178. n. 13

PAP. GENAV.432: 30

P.S.I.1454: 176 n. 13

396 Index Locorum

General Index

References to specific passages of ancient authors discussed in the volume can befound in the Index Locorum

Achilles 105–32, 114–17, 119, 121 n. 68,123–4, 126 n. 86, 130–1

Aeneadae 17, 106 n. 5Aeneas 14, 118 n. 58, 126–8, 130 n. 99,

131, 244, 246name 126birth 221, 344

Aeschylus 54–5, 171, 199; see alsoHomeric Hymns

Agamemnon 55Alcaeus 6, 12, 39, 156, 192, 200–1Amphidamas, funeral games 9, 212Amphyctionic League 11Anchises 14, 22, 83, 105–32, 343Antimachus 189, 197Aphrodite

affair with Ares 17, 32–3, 37–8, 97,219, 249–50, 330

animals, power over 116, 120, 220birth 218–19epiphany 83, 121, 190 n. 65Hephaestus, marriage to 34loss of power 124–7, 208, 220Paphos, sanctuary at 22seduction, type scene 109–10,

112–15, 120Apollo

birth 60, 73, 138, 181–2, 221–2, 339child of Zeus 256–7, 265 n. 18,

271–8epiphany 184 n. 36old man at Onchestus, encounter

with 84oracle/prophecy 11–12, 60–1, 70–1,

74, 77–9Apollonius Rhodius 193–5; see also

Homeric HymnsArcadia 145, 147, 151–72, 338, 341Archilochus 88–90Ares 117; see also AphroditeArion 135 n. 5, 148Aristarchus 177–8

Aristophanes 93–4, 97, 177–8, 215;see also Homeric Hymns

Athenabirth from Zeus 48, 256–7, 262–71,

275

Bacchylides 200; see also HomericHymns

bilingualism 119Brauron 135 n. 5

Cadmus 139, 142–3, 148Callimachus 179–93; see also Homeric

HymnsChalcondyles, D. 1–2Chanson de Roland 5Chios 144, 186, 223, 287–8, 293, 298Chiron 138Corinth 14, 22, 135 n. 5, 294cosmogony 212, 216, 219Crates of Mallos 41, 226Cretan sailors 60–1, 76–8, 353cult hymns 20–2, 50–3, 66, 107–8, 185,

197–8, 206–31, 232–4, 306 n. 78Cybele 152, 171, 198Cyllene 93, 139, 153, 171, 338, 339, 341Cynaethus 11, 62, 71–3, 177, 202, 223,

288–91Cyprus 110, 112, 144, 218, 341Cyrene 184–5

dating; see Homeric HymnsDelia 284–6Delos 21, 60, 64–7, 69, 71–3, 75–6, 81,

207, 216, 221–3, 231, 273–4, 278,285–6, 308–9

Demeteranger at Zeus 45–6bringer of famine 45, 55–6epiphany 83search for abucted daughter; see

Persephone

Demodocus 17–8, 34, 37, 112, 208, 219,237 n. 23, 238, 249–52, 302, 329–31

Derveni papyrus 212–16Dichterweihe 90Diomedes 117, 129–30Dionysus

abduction by pirates 100, 133–50admission to Olympus 30–3birth 30, 148, 226–9, 339counterpart of Demeter 48epiphany 83, 133–50trieteric festival 21 n. 101, 53wine 134, 142, 146–7, 149

Doloneia 5 n. 25Drakanos 33Dryops 167, 169, 171, 348

Echo 157 n. 19, 158, 169, 170 n. 55Eleusinian Mysteries 46, 50–3, 138,

208–10Empedocles 217Eos (Dawn) 123Epic Cycle 4, 6, 177, 250, 330, 331 n. 157epica laus (pars epica) 341–53Epimenides 217epiphany 83, 85–6, 98–104, 108, 121,

135–6, 140, 142–50, 164, 190Erinys 55Erysichthon 189–90Etruscans 135etymology 68, 154 n. 9, 161, 166 n. 41,

183–4, 327Euripides 140, 143, 198, 310, 314; see

also Homeric Hymnsevocatio 207 n. 3, 337–41Exekias 134, 143 n. 28, 145

First Sacred War 11–12François Vase 31–2, 40, 227

Gaia 218, 257–67, 273, 277, 279,343, 355

Ganymede 97, 123genealogy 118, 142, 170 n. 55, 171, 212,

218, 229, 339genre 19–20, 53–4, 79–80, 101, 107–8,

160, 164, 206–31, 232–53, 332–3Glaucus 117Graces 34, 111, 352

Hebe 97, 352Hector 114, 116, 121–5, 130, 265–6

Helen 110Hephaestus 97

binding of Hera 9–10, 30–3, 40, 227birth 30–1, 275–6marriage 34, 37–8

Hera; see also Hephaestusanger 30, 48, 74–5, 97, 143, 222, 227,

265–7, 275–7Samian Tonaia 21, 33, 53seduction of Zeus 35–7, 109–12, 248

Heracles 36–7, 195, 266Hermes; see also Pan

birth 108, 224, 339epiphany 85–6, 98–104humour 87–98, 225meat, sacrifice of 21–5, 56–7, 93–4thief 13, 83–4, 92, 94–5trickster 86, 224, 242

Herophilus 182Hesiod 160, 210–12, 217–18, 261 n. 12,

265 n. 18, 290, 295–9Hipparchus 72, 287Hipponax 90–3, 97Hippostratus 72, 289Homer 288–94Homeric Hymns

collection 41, 175–81, 234–5dating 7–16editions 1–3Hellenistic poetry, influence on 181–96language 7–8, 49–50, 312performance 16–19, 86–7, 101–4,

145–6, 165, 207–9, 219, 223–4,232–4, 251–2, 280–333

structure 61–4, 85–6, 206–8, 233–7,334–57

Hymn 1 to DionysusCallimachus, influence on 179–81dating 9–10, 34, 39Euripides, influence on 198origin 33

Hymn 2 to DemeterApollonius, influence on 194Bacchylides, influence on 200Callimachus, influence on 188–90dating 10Eleusis, connection to 7, 10, 21, 50–3,

161 n. 29, 209–10, 232–3Euripides, influence on 198female characters 47Hesiodic character 49language and style 49

398 General Index

structure 55–6Theocritus, influence on 195tone 44–8

Hymn 3 to ApolloAeschylus, influence on 199Apollonius, influence on 193Aristophanes, influence on 199Callimachus, influence on 181–8cults, connection to 53dating 7, 11–12geography 64, 339inscribed by Delians 21language and style 67–71performance 223–4, 282–6Pindar, influence on 199–200Theocritus, influence on 195unity and structure 11–12, 59–81,

290–1Hymn 4 to Hermes; see also Hymn 19

to PanApollonius, influence on 194–5Callimachus, influence on 190–1cults, connection to 21–2, 53dating 12–13humour 87–98, 225language and style 82, 97performance 13 n. 62, 86–7, 101–4,

233 n. 4rhetoric 13, 95Sophocles, influence on 198

Hymn 5 to Aphrodite; see also AeneadaeApollonius, influence on 195Callimachus, influence on 191–3cults, connection to 17, 22dating 7, 14, 49, 105–6, 202Euripides, influence on 198humour 118, 122imitation of Homer 106–7immortality vs. mortality 108, 126–7,

163 n. 36, 341, 352language 127–8, 195, 198structure 105–32Theocritus, influence on 195

Hymn 7 to DionysusCallimachus, influence on 180–1dating 14–15, 133 n. 2performance 145–6structure 136style 146–7

Hymn 8 to Ares 15, 175–6, 340 n. 18, 227Hymn 19 to Pan

dating 15, 169–72

humour 162–3language 169–72performance 165, 169–72relationship to Hermes 166–8style 146–7, 151–72

Hymn 20 to Hephaestusdating 16

Hymn 31 to Helios 203dating 16

Hymn 32 to Selene 203Apollonius, influence on 193dating 16

Hymn 33 to the Dioscuri 39 n. 22,201, 339

Theocritus, influence on 195Hyperboreans 144, 201 n. 123, 223Hypnos (Sleep) 35–7, 111

Iambe 51, 93, 188, 198, 245 n. 50Ida 22, 110, 116, 221, 344Ioannes Eugenikos 43Ikaros 33

Judgement of Paris 110–11, 221

Karion 40, 94Kleitias 31, 40kleos 118, 124, 305Kore; see PersephoneKronos 214, 254–79krotos 152, 160, 164

Lelantine War 9Lemnos/Lemnians 35, 37, 135 n. 5Leto 48, 60, 64, 69, 181–2, 221–2,

256, 272–6Linus 217Lycurgus 149 n. 50Lynx 158, 167 n. 45lyre 74, 101–2, 194, 225, 344–53Lysicrates 135 n. 5, 144 n. 31

manuscripts 1–2, 29–30, 43,89 n. 22, 175

Maia 96, 139, 167, 224, 339Menelaos 34, 115 n. 47Metis 256, 263–4, 278Mnemosyne 88–90, 140, 245, 356–7Mnesiepes inscription 88–90Musaeus 212, 216Muses 90, 210–12, 305–8, 314–17,

342–9, 352

General Index 399

music 154, 156–9, 318–20, 334–57Mysteries, Eleusinian 21, 50–3, 208–9

Naxos 33, 144, 143, 311Neoanalysis 6, 107nightengale 156–7, 168–9nymphs 152, 154, 159–60, 162–6, 344,

347–8

Odysseus 66, 100, 270 n. 25, 273 n. 29,274 n. 30, 330–2

Olen 201 n. 123, 212, 215–16Onchestus 84–5oral poetics 3–9, 106–7, 280–1Orchomenos 142Orpheus 194, 212–15, 228–9Orthodox Church (Christian) 104

n. 81Ouranos 214, 218, 258–61

Pamphos 212, 216Pan 151–72

birth 138cult 15, 169name 152, 161relationship to Hermes 13, 151 n. 2,

166–8Panathenaea 13 n. 62, 22 n. 102, 72, 287,

300, 319Panhellenism 14, 20–22, 64–7, 81, 108,

135–6, 144–6, 232–4, 246–7Parry, M. 4Peisistratus 11, 285–7Penelope 168–9, 171, 273 n. 29Pentheus 142performance; see Homeric HymnsPersephone

abduction and marriage 42, 44–6,54–5, 108, 137–8, 188–9, 189, 209,241, 248

Phemius 17Pindar 18, 31, 65 n. 19, 80, 143, 167 n.

43, 171 n. 61, 199–200, 238–9, 249,251, 306, 309, 311, 324; see alsoHomeric Hymns

Polycrates 11, 72–3, 202, 223, 284, 286Politics of Olympus 79, 96–7, 107–8,

135–9, 208–9, 225, 242–5Poseidon 115, 127, 131, 247, 338preces 86 n. 13, 354–7Priam 99–100, 114, 130Prometheus 256

prooimion (prelude) 17–19, 197, 207–8,237–40, 290, 324–5, 327, 336

repetition 49–50, 95, 108 n. 20, 155 n.11, 158 n. 22, 184, 185, 336 n. 7

Rhea 56, 228, 256, 267, 273, 277Ruhnken, D. 2, 30, 61, 63

Schadenfreude 98Semele 139, 148seruus callidus 95 n. 51, 225sexual potency 121Silenus 143Sophocles 169, 198; see also Homeric

Hymnssymposium 17, 146, 149, 251syrinx 13

Tartarus 259, 261, 265, 266, 273Telphousa 60, 69, 75–6, 271, 275, 278Thebes 143Thetis 34–5, 38–9, 126, 194, 266–7Theocritus 195; see also Homeric

Hymnstheogony 212, 214–22, 229–31Thesmophoria 50–3, 210 n. 16Thucydides 11, 18, 61, 135 n. 5, 177,

196–7, 204, 238–9, 282–6, 288,307–9, 312–13, 317–25, 327

timai 73, 79, 96, 102 n. 74, 117, 136–8,191, 241, 337–8

Titanomachy 256, 269Triptolemus 56, 217Troy 119Trygaios 93–4Twelve Gods, cult at Olympia 22, 53,

161 n. 29Typhaon 73, 74–5, 275 n. 34Typhoeus 256, 257, 259, 261, 267, 278Tyro 115Tyrsenian pirates 14, 133–50

virginity 192, 200, 201

Wilamowitz-Moellendorff, U. von 31–2,62, 227, 249

Wolf, F. A. 3, 17–18, 237

Zeus; see also Herarole in Hymns 47, 79–80, 96, 108,

122–7, 138–9, 208–9, 214, 220–1,225, 227, 242, 254–79, 325

400 General Index

Documents

[Andrew Faulkner] the Homeric Hymns Interpretati(BookZa.org)