Sei sulla pagina 1di 244

O V I D ’ S L O V E R S

Central to Ovid’s elegiac texts and Metamorphoses is his pre-occupation with how desiring subjects interact with and seduce each other. This major study, which shifts the focus in Ovidian criticism from inter tex- tuality to intersubjectivity, explores the relationship between self and other, and in par ticular that between male and female worlds, which lies at the hear t o f Ovid’s vision of poetr y and the imagination. A series of close readings, focusing on both the more celebrated and less studied par ts of the corpus, moves beyond the more often-asked ques- tions of Ovid, such as whether he is ‘for’ or ‘against’ women, in order to explore how gendered subjects converse, complete and co-create. It illustrates how the tale of Medusa, alongside that of Narcissus, re ver- berates throughout Ovid’s oeuvre, becoming a fundamental myth for his poetics. This book offers a compelling, often troubling por trait of Ovid that will appeal to classicists and all those interested in gender and difference.

v i c t o r i a r i m e l l teaches Latin literature a t the University of Rome, La Sapienza. Sh e i s the author of Petronius and the Anatomy of Fiction ( 2002) and a contributor to The Cambridge Companion to Roman Satire ( 2005). Sh e has written numerous ar ticles on Latin literature, especially on the nove l and Ovid.

O V I D ’ S L O V E R S

Desire, Difference and the Poetic Imagination


O V I D ’ S L O V E R S Desire, Difference and the

CAMBRIDGE UNIVERSITY PRESS Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo, Delhi

Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK

Published in the United States of America by Cambridge University Press, New York Information on this title:

© Victoria Rimell 2006

This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press.

First published 2006 This digitally printed version 2009

A catalogue record for this publication is available from the British Library

ISBN 978-0-521-86219-6 hardback ISBN 978-0-521-11780-7 paperback

Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party Internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.


Note on the text

page vi



List of abbreviations


Introduction Na rcissus and Medusa: Desiring subjects and the dialectics of Ovidian erotics


1 Specular logics: Medicamina


2 Double vision: Ar s Amatoria 1, 2 and 3

7 0

3 Seeing seers: Metamorphoses 1011 .84


4 Co-creators: He roides 15


5 What goes around: He roides 16 21


6 Space between: He roides 18 19






Index of passages discussed


General index



Note on the text

In quoting Ovid I have always used the Oxford Classical Text, unless other wise indicated. All translations are my own.



This book was begun in Cambridge fog and finished in the golden light of Rome. I feel extremely lucky to have benefited from the stimulations and ver y different pleasures of both environments over the past fe w years. I’m especially indebted to the Facolta` di Scienze Umanistiche of the University of Rome, La Sapienza, and the Italian Ministr y for Universities (MIUR) for appointing me to my present position. There a re several people I would like to thank: Rober to Antonelli, Francesca Bernardini and Piero Boitani, along with my colleagues Luigi Enrico Rossi, Andrea Cucchiarelli and Maria Broggiato, for welcoming me so warmly at La Sapienza; Alessan- dro Schiesaro, who first suggested I put all my thoughts on Ovid in one place – I couldn’t have done without his wonder ful suppor t; John Hender- son, whose humour helped me retain some small measure o f sanity more than once, for his speed reading and uncannily sound advice; Alessandro Barchiesi, Philip Hardie, Charles Ma r tindale, and the anonymous reader at Cambridge University Press, for encouragement and ver y useful criticisms; and my friend Jane Jones, for her big brain, big hair, and countless hours of illuminating chats. I a m also ver y grateful to Michael Sharp, and to my editors Jackie Warren and Nancy-Jane Rucker. A version of chapter 1 (‘Specular Logics’) appeared in R. Ancona and E. Greene (eds.) ( 2005) Gendered D ynamics in Roman Love Eleg y , reprinted with permission of The Johns Hopkins University Press, while chapter 4 revises Rimell (1999) ‘Epistolar y fictions: authorial identity in Heroides 15 ’, PCPS 45 : 109 35 .




Antike und Abendland


American Jour nal of Philolog y


H. Temporini, Aufstieg und Niedergang der R omischen¨



Berlin 1972 Bulletin of the Institute of Classical Studies


Br yn Mawr Classical Review


Classical Antiquity


Classical Jour nal


Classical Philolog y


Classical Quar terly


Classical Wo rl d


F. Jacoby, Fragmente der griechischen Historike ( 1923 – )


Graz er Beitr age¨


Greece and Rome


Ha r vard Studies in Classical Philolog y


Jour nal of Roman Studies


Liverpool Classical Monthly


Materiali e Discussioni per l’Analisi dei Testi Classici


Oxford Latin Dictionar y


Proceedings of the Cambridge Philological Society


Rheinisches Museum


Transactions of the American Philological A ssociation


Wiener Studien


Introduction Na rcissus and Medusa: Desiring subjects and the dialectics of Ovidian erotics

Ovid’s str ucture i s not merely like a Russian doll, one stor y inside another, i t i s like a snake-pit, in which a pretty indeterminable number of snakes are devouring and being devoured by each other. 1

All love is combat, a wrestling with ghosts. 2

This book explores the gaps in which same and other, male and female can be seen to relate, converse, compete, and co-create in Ovidian poetr y. The chapters included span a large por tion of Ovid’s corpus, star ting with the Medicamina, his little-read treatise on cosmetics in which women are made up in men’s image and vice versa, and ending with the ‘double’ He roides, where heroes and heroines of ancient myth write to and from (ove r and across and with) each other. I am interested here i n the many (flash-) points in Ovidian poetr y where male and female ar tists/lovers are twinned as vying, mutually threatening subjects, and where a narcissistic impulse to collapse other into same/self is rivalled by a more complex dialectic or exchange which seems itself to fire and propel desire. On e of the core aims of this study is to counter some curious imbalances and repressions in recent Ovidian criticism: in par ticular, I discuss the extent to which the dominant model for the Ovidian ar tist, the male vie wer who spurns woman and/or (re)creates her as ar twork and fetish (Narcissus, Orpheus and Pygmalion are key figures) has tended to foreclose investigation of the relationship between gendered creativities in Ovid. Fo r sure, we can all spot competing models of the ar tist – f rom Echo, who turns repetition into originality, pronouncing novissima verba with typical satiric, Ovidian wit, 3



Hofman and Lasdun (1994 ) xii.

See Knoespel (1985 ) 8:

‘ What emerges from Ovid’s account of Echo is the powe r of speech and the ability of Ovid’s own written language to control that speech. Even though Echo is handicapped by Juno’s punishment, her handicap paradoxically emphasizes the adaptability of speech. Ultimately it is the powe r of the written language, Ovid’s own narrative, that emerges from the description of Echo’s language.’

2 Paglia (1990 ) 14.



v i c t o r i a r i m e l l

to spinner Arachne, 4 tor tured embroiderer Philomela, 5 or the daughters of Minyas, stitching rebellion into Metamorphoses 4 – but these are not the characters usually identified with the poet himself , and feminist critics have been more concerned with making such figures visible, rather than (in addition) with scr utinizing how gendered readings and writings contend and overlap. Male ar tists, however, are f requently constr ued as synonymous with the poet. As Segal writes, for example: ‘ Through Orpheus, Ovid

provides a metaphorical reflection of the creative and restorative p ower of his own a r t’. 6 For Anderson, Pygmalion ‘is the creative a r tist par excellence’, 7 or as Rosati puts it, reaffirmed many times over in Ha rdie’s recent book: ‘Ovid

is the poet Na rcissus, the poet bent over in admiration of his ow n vir tuosity,

triumphantly mirroring himself in the astonishment of his public’. 8 For Ha rdie, narcissistic desires (aligned with a bid to conjure up presences) lie at the psychological hear t of Ovid’s poetr y, fuelling an obsession with sameness and doubling (not least, between ar t and nature). 9 Thus the Na rcissus–Echo plot can be seen to ghost-write a string of Ovidian couplings (Ceyx and Alcyone, Leander and Hero, Deucalion and Pyrrha, Pyramus and Thisbe), in which beloveds become mirror images of lovers. 10 Indeed, Narcissus reigns in recent criticism as the figure both for the poet (as he flits between credulity and cynicism, primal magic and urbane irony) and for the desiring, seduced, self-conscious reader. His myth offers a neat allegor y for the move from na¨ıvety to knowingness, nature to a r t celebrated by postmodernism,

a field of thinking owed much of the credit for Ovid’s flight to stardom

at the end of the second millennium. 11 We might even say that Narcissus’ psychodrama has come to define Ovidian poetics as obsessed with linguistic sur faces and passing intensities, with visual display, duplicity and (obvious) feigning. 12








Also see Hinds ( 1998 ) 58 , and Hollander ( 1981) o n Echo as a figure of poetic allusion and as an ironist or satirist (‘Echo’s power is thus one of being able to re veal the implicit’, Hollander writes, 27 ). Echo’s stor y i s one of several Ovidian myths to be appropriated by feminist thought: see Berger ( 1996), Spivak ( 1993).

Fo r a feminist reading of this tale see Miller ( 1988 ).

Se e Joplin ( 1984), and Ma rder ( 1992). 6 Segal ( 1989) 491 .

Anderson (1963) 25 , cf. Solodow (1988 ) 215 , and Hardie ( 2002 a) 23 : ‘ Orpheus and Pygmalion in Metamorphoses 10 are the Ovidian figures for, respectively, the poet and the ar tist in their role as primitive magicians.’

Rosati ( 1983 ) 50 , Hardie (2002 a) 28 et passim . Also see discussion of these models in Elsner and Sharrock ( 1991 ). Leach (1974 ) is among the fe w to vie w Narcissus and his relatives differently: for her, he is a model of ar tistic failure.

Ha rdie ( 2002a), ( 2004). 10 Hardie (2002a) 25882 .

As Hofman and Lasdun put it in their introduction ( 1994 ) xi: ‘there a re many reasons for Ovid’s rene we d appeal. Such qualities as his mischief and cleverness, his deliberate use of shock – not always relished in the past – a re contemporar y values.’

Se e especially recent summar y o f modern reception of Ovid in Ha rdie ( 2002b).

Introduction: Na rcissus and Medusa 3

Ye t the self-love of Na rcissus is only one, limited model for Ovidian desire and Ovidian representation. Ovid stages many different complica- tions, variations or contestations of the Na rcissus plot, and writes desire as a fractured, competitive process. The emphasis on Narcissus as Ovidian

ar tist in recent criticism fascinates me especially because of the way in which

it grounds Ovidian illusionism and fiction in mimetic male desire. This book offers a rather different (and at the same time often complementar y) take on Ovidian ar t and erotics, suggesting ways in which this experimen- tal poet takes his readers far beyond Narcissus’ experience. Much valuable work has been done already on how Ovid writes the decentred self, and blurs or snags what are perceived to be traditional gender categories, espe- cially of masculinity. 13 What I would like to do here, howeve r, is to draw out discussion of what Miller calls the ‘split nature’ of elegiac subjectivity by looking closely, in addition, at how Ovid constr ues intersubjectivity. 14 Broadly, I want to rethink powe r in Ovidian poetr y as relational (rather than hierarchical), and to push against the tendency of Ovidian criticism in the last thir ty years to fall into polar camps. As Miller puts it in his discussion of the Amores :

Depending on whom one reads, elegy is either in league with the Augustan political regime (Kennedy 1993, 35 6 ; Ne wman 1997, 6 ) o r implacably hostile to it and the traditional values it sought to promote; either political allegor y (Edwards 1996 , 24 ) or an apolitical, ludic discourse that gently mocks social custom ( Veyne 1988: 312, 104 8; Kennedy 1993 : 956; Fantham 1996, 108); either exploitative o f women (Kennedy 1993 : 38, 56, 73) o r bent on satirizing Roman misogyny (Greene 1994). 15

Similarly, the inclination of debate on gender in ancient texts has been, almost entirely, t owards analysing either female or male figures, either con- str uctions of femininity or masculinity. 16 This has much to do with a reluctance, deeply engrained in the Western tradition, to enter tain two parallel (same but different) subjects, an aversion magnified by an Anglo- American feminist ideal of a gender-neutral human subject. In this book,

I want to sidestep the kinds of questions that have repeatedly been asked

of Ovid in the last thir ty years, by asking not (simply) about constr ucts of femininity, or of masculinity, o r about whether Ovid can be judged a

13 See e.g. Raval ( 2002), Keith (1999 ) and ( 2000), P. A. Miller ( 2003 ) o r summar y i n Sharrock ( 2002a).

14 P. A. Miller ( 2003 ) offers a ‘symptomatic’ histor y o f Roman erotic elegy, arguing that elegiac poetr y arose from a fundamental split in the nature of subjectivity that occurred in the late first centur y. Hi s book provides a ver y interesting, more historically focused background to my discussions of Ovid’s vision of the self.

15 Miller ( 2003) 30 .

16 Recent exceptions include Keith ( 1999) and Miller ( 2003 ). Se e n. 17 .


v i c t o r i a r i m e l l

anti-, proto- or pseudo-feminist, 17 but instead about relationality, about the desiring subject in Ovidian poetr y as a being-in-relation. 18 Fo r although in the Amores , Ovid’s first published work, erotic rela- tions appear more straightfor wardly formulated (at least, in par ts) as lover/beloved, and Corinna appears to be little more than elegiac subject- matter, with no voice of her own, in the He roides, Ar s Amatoria , Medicamina , and Metamorphoses, we can read varied experiments in juxtaposing canny, rival lovers and in juggling unstable subject/object positions, experiments which postpone any final determination. 19 All these texts, with the excep- tion of Heroides 1 15 , 20 we re written between four and eight years of each other, and make up the backbone of Ovid’s life’s work. The core elegiac texts ( Heroides, Ar s , Medicamina , a s well as the Remedia Amoris, which I dis- cuss only briefly) are all concerned with how desiring subjects interact and seduce each other, and it is this idea of imagining the intersection of male and female worlds which perhaps distinguishes the originality of Ovid’s contribution to Augustan literature, and takes the concept of the uneven, sexy pairing visualized in the elegiac couplet as far as it can go. While elegy, traditionally, has room ‘for one voice only’, tending to reduce ever ything to the persona of the poet–lover, 21 Ovid’s image-conscious poetr y is often focused on dialogue over monologue, moving at the borders of known worlds, both real and imaginar y.

17 Fo r criticism which sees Ovid pushing against and re versing traditional gender roles see e.g. Hal- lett ( 1990) 193, Jacobson (1974 ) 7 , Curran (1978) 213, Luck (1960 ) 418 , Spentzou (2003 ), James ( 2003 ). For an Ovid who demonstrates the mechanics of male discourses of powe r and domination over/objectification of women, see e.g. Greene ( 1998), Leach ( 1964), Sullivan (1962 ). Fo r more com- plex accounts, in which Ovid does the former to a cer tain extent (or pretends to), while ultimately fulfilling the latter, o r vice versa, see e.g. Raval (2002) and Green ( 197980 ), or Watson (2002). Similarly, i n ‘ resisting’ readings of Ovid (e.g. Liveley 1999 ), there i s often a strong sense that the ‘feminist’ (or almost, here, ‘tactical’) reading is the one which rejects and resists in order exclusively to privilege another vie wpoint, implying that the conventional reading is born of na¨ıve masculin- ism, rather than offering a way to analyse how attitudes and readings compete in the text. Keith (1999 ), along with P. A. Miller (2003 ), is unusual in criticizing the limitations of the above positions, in which the poet is interpreted as either promoting gender subversion and sexual liberation or as reconfiguring a repressive sexuality.

18 This is an Irigarayan term (see especially Irigaray 2000 ): her work calls for the radical reevaluation of the human subject as defined by difference rather than sameness. Many of my thoughts on Ovid in this book have been complicated and enriched by her work, although I am by no means attempting to recuperate Ovid as an utopian French feminist.

19 Miller ( 2003) sees the Roman elegists in general as ‘augurs of instability’, exploring ‘the interstitial space between masculine and feminine, active and passive, for which traditional Roman discourse has no terms’ ( 25); elegy thus becomes ‘a symptom of crisis in the Roman subject’s self-conception’ ( 26 ).

20 The publication date for these poems is uncer tain. Traditionally they have been assigned to the same period as the Amores , and dated at around 15 BC; however some scholars have placed them later, between 10 and 3 BC.

21 As Ba rchiesi puts it: (2001 ) 31 2.

Introduction: Na rcissus and Medusa 5

Thus, while in the Ars Amatoria especially, Ovid surely baits the kind of deliberation which has preoccupied feminist critics of his poetr y ( ‘whose side is he on, anyway?’), 22 this poetr y does ever ything it can to foil our verdict, or to show u p its ow n ingenuousness (which is not to say that men and women ultimately come out quits in the Ar s, only to stress that the tussle of ‘foe against foe’ in this text is without resolution). Ovidian desire often works to break down boundaries, and thus to threaten autonomy, identity, and to collapse difference into incestuous sameness, yet a t the same time it often resists and dodges Na rcissus’ fate, recognizing that connectedness is not synonymous with homogeneity, that the dynamic of relationality is also the vim of creative p rocess, both of writing and reading. Ovidian erotics can be read as a constant battle to transcend a compulsive logic of the same in order to sustain desire, or poetr y itself. In this way, Ovid discerns and wrestles with the fundamental problem of what Irigaray calls the ‘specularisation’ underlying all Western philo- sophical discourse, 23 confronting head-on the perilous implications of a self-perpetuating mode which creates man’s desired object as the reassur- ing negative of his ow n reflection. Similarly, even i n relationships which appear to be self-contained, Ovidian sex depends on multiplications, tri- angulations, substitutions, go-betweens, which inevitably render mirroring interactions much more complex than the Na rcissus–Echo, subject–object (male–female) prototype would suggest. 24 Thus in the Ar s Amatoria , Ovid’s pupils are asked to negotiate a tangle of contrar y advice, to perceive par- allel scenarios through the eyes of men and women, husbands and lovers, Echoes and Na rcissuses, vie wers and vie wed; with Ovid’s l overs, we reach a climax at the end of Ar s 2 only to realize there is one more book to come – yet the Remedia Amoris erases that end, too (and let us not for- get the other appendix to Ars 3 , the Medicamina ). 25 In the double Heroides,

22 In her discussion of the women-focused single He roides, Sharrock (2002a) 99, distills this familiar cross-examination as follows:

‘a cr ucial question is the extent to which we may be able to read a ‘woman’s voice’. What kind of gendered voice is produced by a male author speaking through a female mask, but completely

T he same question arises when we

tr y t o confront more widely the ver y high profile of women in the corpus: is it friendly or not? How far is Ovid implicated in the exposure and objectification of women and denigrating violence towards them, perpetrated in and by his texts?’

23 Irigaray ( 1985 a) explains how logocentrism is incapable of representing femininity/woman other than as the negative o f men’s own reflection. Philosophical meta-discourse, she argues, is only made possible through a logic of the same, a narcissistic process whereby the speculating subject contemplates himself.

24 Just as, Miller argues ( 2003) 24, elegiac women ‘represent less simple identities than complex nexuses of conflicting symbolic norms’.

25 Fo r more detailed discussion of the relationship between reading, sex and counting games in the Ar s Amatoria and Remedia Amoris , see Henderson (for thcoming).

subsuming his masculine authority into the female writing?


v i c t o r i a r i m e l l

similarly, par tnerships swell into love triangles and more messy relationship- maps, while in (the aptly numbered) Amores 2 , Ovidian aspirations and desires stretch and splinter elegiac distichs/diptychs, producing a kaleido- scopic effect: multiple affairs, accomplices, scenarios and readings load the book with epic potential, ye t for the oversexed poet–love r, nothing is ever satis. 26 In this introductor y chapter, I use the Medusa myth (which, as we’ll see, feeds subtly into Ovid’s corpus throughout) as an illustrative way into exploring how a narcissistic logic in Ovidian poetr y is contextualized and challenged, first of all by the troubling existence and (not quite visible) vision of the Other (paradigmatically, Woman), and secondly by an ongoing meeting/clash/interaction of desiring subjects and poetic drives, male and female. This opening discussion is more abstract and theoretical than the close readings of individual texts that follow, and it is also more literal in its exploration of the Medusa and Na rcissus myths, which often (par ticularly in the case of Medusa, a figure who, in more ways than one, can rarely be seen head-on) become shadows, reflections and backgrounds to understanding erotic relationships in Ovidian poetr y itself. Throughout the book, then, I will be probing the limits of a fixation on the myth of Na rcissus in Ovidian criticism: readings of Ovidian desire and poetics as rooted (only) in the paradigm of Na rcissus have tended to reject and quell the Other (the female), underemphasize the extent to which the Ovidian poet is identified, often simultaneously, with other ar tist figures and with other models for individuation, and suppress the horror of self-consciousness, as well as the risk of incredulity (as dramatized in the parallel catoptric myth of Perseus and Medusa told in Metamorphoses 4 and 5 ). Medusa’s presence, insidious but little discussed, gnaws into and rivals the Na rcissus archetype, asser ting two desiring/creative subjects whose inter- course spikes the paradoxes of Ovidian illusionism. Both myths, told in Me t .3 5, a re fundamental models in Ovid for the bir th of poetr y and for the individual’s path to subjectivity. Cr ucially, too, they both make the mirror a symbolic tool for (painful, dangerous) self-realization, a idea Ovid also explores in the Amores, Medicamina and Ar s Amatoria. Narcissus, pet- rified with amazement in Me t. 3, comes of age and becomes a symbol of Ovidian self-consciousness when he understands that he is seeing himself in the reflective pool, while those who see Medusa in Me t .4 and 5 look (their own) death in the face, and are turned into per fect stone statues

26 satis , or rather non satis , i s an impor tant concept in Amores 2 , shor thand for the games of excess played throughout this book. See e.g. 2 . 10.12 , 2 . 10. 22, 2.13 .28, 2 . 14.44.

Introduction: Na rcissus and Medusa 7

(unless, like Perseus, they avoid looking at her straight on by using a mir- ro r – in which Medusa, like Na rcissus, may also see herself ). Medusa is a paradigmatic female vie wer and ar tist, and the myth of her rape, monstrous transformation, and perpetual afterlife as global sculptress is located (along- side Na rcissus’ transformation) at the core of the Metamorphoses. But while Na rcissus looks at and is in love with himself, the myth of Medusa always involves encounters between spectators. Simply put, this is a myth about looked-at woman becoming ultra-powe r ful vie we r (snake-haired Medusa still turns her audience on, and is compulsive vie wing, ye t her audience is ‘castrated’ e ve n a s it is permanently fixed in the state of open-mouthed arousal). But i t is also, cr ucially, about the convergence and collision of gazes, for her narrative culminates at the end of Me t .4 in the moment at which Perseus sees (or does not see) her gr uesome face in his mirror-shield – itself a giant, surrogate eye. Indeed, although she gets only passing mention in one chapter of Fred- erik’s recent The Roman Gaz e ( 2002), and is barely glimpsed in studies of Ovidian spec(tac)ularity and desire, 27 Medusa is ever ywhere i n Ovid, just as she looms large (alongside Na rcissus) in twentieth-centur y philoso- phy and creative writing, in psychoanalysis and French feminist thought. 28 This book attempts to engage with and encompass insights developed in other fields of the humanities that have often tended to be neglected by classicists; just as, in turn, it hopes to show that the classical foundations of mythical archetypes and their treatment in authors such as Ovid can- not be sidestepped or underestimated in the elaboration of critical theor y. Medusa’s stage presence, I’ll argue, ensures that Ovidian desire does not simply re volve around intoxication and restoration, possession, loss and evanescence, but is infused with aggression, re venge, conflict, myster y, sus- pense, rene wal, and above all, fear. And while desire’s mission, as Hardie stresses, is usually to embrace and consume the other (this is, above all, Narcissus’ fantasy), 29 we shouldn’t lose sight of the fact that there a re numerous places in the Ovidian corpus where we can recognize p recisely the opposite conclusion, where the poet is not only hampered in his quest but obsessively concerned with privileging and revelling in the journey itself rather than the destination, to the point that the apparent target is rendered

27 E.g. Rosati ( 1983), Hardie (2002 a), Sharrock (1994 ), (2000 a).

28 Fo r an ove r vie w of Medusa’s impact on Western literature, philosophy and ar t, see Garber and Vickers ( 2003 ).

29 Throughout much of Western philosophical thought, the notion that (male) desire can never be tr uly fulfilled without the ‘possession’ of the other, is all-per vasive. Se e the critique of Levinas, Sa r tre, and Merleau-Ponty in Irigaray ( 2000 ) 1729 .


v i c t o r i a r i m e l l

redundant, cosmetic. The repeated and delusional displacements and sub- stitutions that define Lacan’s desiring subject are often self-consciously celebrated (as the real point, the real victor y) in Ovidian erotics. Perhaps the most flagrant instance of this occurs at the end (which is also a middle and re-beginning) of Amores 2 , where the poet, disgr untled at a rival’s literal and slavish reading of his seduction campaign, blur ts out: si tibi non opus est ser vata, stulte, puella, / a t mihi fac ser ves, quo magis ipse velim (‘If you’re not bothered with guarding the girl for your own sake, cretin, at least guard her for mine, so that I’ll want her all the more’ 2 .19 .1 2). One spiteful Medusan look spells out the magic and cracks the fantasy, e voking Narcissus’ disillusionment: yet the collapse of Narcissus’ deluded lust can coexist with Medusa’s fatal attraction (especially in a book which sets up ‘straight talking’ as a painful trap for its readers, exposing possibilities rather than dealing the bottom line). 30 In other words, it is not only the case, as Hardie emphasizes, that Ovidian poetr y continually yearns to substitute textual ecstasies/fallacies for actual bodily union, resulting in concatenations of absent–presences: Ovid is also concerned, sometimes ve r y obviously, to shun and undermine the drive for possession/unity/symmetr y in order to animate dynamic relationships between subjects. This opening chapter is followed by six close readings which work through the Ovidian corpus, from elegiac didactic to epic to the late ‘double’ epistles, the last Augustan elegy we have. These texts, or bits of texts, are all in different ways concerned (and I want to stress, more concerned than any other of Ovid’s poems) with developing dialectical relationships between desiring subjects. Ovid’s fascination with communication between lovers, and with doubling, interaction, competition and exchange more generally, might be seen to culminate in Her.16 21, when men and women get to write simultaneously and side by side. But I d o not so much want to plot a teleology of Ovidian erotics as to suggest that when we revie w Ovid’s poetr y in the light of these ideas, accentuating par ts of the corpus which have as a whole received less critical attention, 31 ne w or forgotten grains and colours emerge. Together, Ovid’s experiments in par tnering male and female add

30 Se e especially Am .2. 7 and 2 .8 (with Henderson 1991 and 1992 ), where we are caught out and lured to re-read, only to pile up our suspicions. Conversely, Ovid points out in Ar s 1 . 61516 that what was once feigned (being in love) can become tr ue before you even realize i t yourself. Similarly, Medusa’s drop-dead gaze is infuriatingly paradoxical: on one hand she is an anti-Pygmalion, killing off illusionism, while on the other, a s Me t . 5 dramatizes, her stor y is emblematic of tales of wonder, and she punishes a disbelieving audience by turning them into stupefied statues, frozen forever as an object lesson in incredulity.

31 With the exception of the Metamorphoses . The double Heroides and the Medicamina in par ticular remain ve r y understudied.

Introduction: Na rcissus and Medusa


up to an intricate and shifting examination of human (inter)subjectivity which is ever concomitant with parading the origins, nature and scope of poetr y. Chapter 1 looks (while tr ying not to squint) at the Medicamina , and con- siders how the poem relates to Ovid’s advice on make-up and mirrors, and also to the catoptric encounter (rivalling that between Na rcissus and him- self/Echo) between Perseus and Medusa, else where in the corpus – especially the Ars Amatoria . I argue that this is a poem about male scopophilic desire to create woman as same: in a straightfor ward reading of the Medicam- ina , Ovid’s cosmos of cosmetics enacts a specular logic in which women are denied the pleasure o f self-representation and permitted only the hysteria of mimicr y. In holding his ow n mirror up to women in this poem, Ovid turns their tool for self-formation against them, like Perseus assaulting Medusa, and the narcissus bulb face-pack which leaves faces as bright as a mirror only r ubs this in. Yet by stealing cosmetics and their accessories to colour his poet- ics, Ovid also adopts the worries and ambiguities invested in mirrors and masquerade. Mirrors in the ancient world are highly paradoxical: they give women the power t o know and control appearances, but in doing so expose the limits of female individuation – they are her weapon/shield/Achilles’ heel, or Narcissus’ trap. Ovid also exploits the Platonic idea that mirror images share with semblances of all kinds an ambiguous mixture o f being and non-being, challenging the mentality that thinks in terms of here and there, self and not-self. So while the Medicamina strives, like Pygmalion, to forge women in man’s image, as same, as ar twork, what it also does is to turn that scopophilic gaze back on the supreme, imperialist, desiring (male) subject. This is, of course, one of the many instances in Ovidian poetr y where the Na rcissus tragedy gets replayed, albeit in a typically metaphorical way. But here we have two subjects (and two Na rcissuses/Medusas), not one: and those two subjects, man and woman, never totally collapse into unity, a s i n the Narcissus denouement. The mirror, now women and men’s tool for self-formation, confuses subject/object, self/other, o r rather (at the same time), it fuels a battle for subject position. The (near-)sameness of the narcissistic encounter produces the tense energy of desire, while risking killing that desire, or castrating both desiring subject s – a threat which itself adds sparkle to elegiac spor t. Chapter 2, o n Ars Amatoria, tries to unscramble the difficult interactions between the three books, and explores how this text revels in anxieties germane to the ar t o f relationships. Ovid’s textbook is littered with traces of the two primordial mirror-myths, Medusa/Perseus and Na rcissus/Echo, models which are both inter wove n and let loose, so that male and female



v i c t o r i a r i m e l l

lovers/ar tists trade places or play multiple roles. Neat symmetries and chess- board patterns are arranged only to be confounded or messed up, with the result that neither men nor women (ar tists nor readers, lovers nor rivals) can keep the upper hand for long, and the sexes are p ropelled into seemingly endless rounds of competition. While chapter 2 ranges at speed through the Ar s to trace the energetic criss-crossing between lovers and books, chapter 3 magnifies just one slow-motion episode in the Metamorphoses, Orpheus’ lethal backward glance at Eu r ydice on the boundar y between upper and lowe r worlds in Book 10 . This chapter dissects the cr ucial moments of realization and amazement at ar t and beauty that pepper Me t .10. I argue that Orpheus’ song, and the Ovidian narrative that frames it, continually revisit instances of the uncanny, or mirror-stages, in which hierarchies of subject– object are unbalanced or even non-sensical. Once again, the Medusa myth is impor tant here (both in Eu r ydice’s double death and in Orpheus’ murder in Book 11), its theme of threatening confrontation infecting Orpheus’ apparent retreat into narcissism and boy l ove. In parodying the Ars Amatoria (Orpheus gives us dangerous, criminal eros over Ovid’s ‘safe sex’ 32 ), Me t. 10 teaches us how metamorphosis, desire and Ovidian poetr y itself are inspired by points of suspension and movement between two states, identities and subjects. Chapters 4 , 5 and 6 move one step away from ocularcentrism to explore other aspects of Ovid’s interest in the relational subject. In chapter 4 in par ticular, our palette of images will be rather different from that which we have seen so far: in He r. 15, the relation between self and same/self and other, is dramatized less in the contrast between Na rcissus’ and Medusa’s/Perseus’ experiences (although the landscape of Na rcissus’ myth is still ver y much in evidence) than in the slippage between homo- and heterosexual desire, within a complex love triangle created by (once-Lesbian) Sappho, Phaon and Ovid. This strange epistle at the end of the single He roides (in most editions, ‘between’ the single and double He roides), becomes the site for a fascinating per formance of Ovid’s agonistic affair with female predecessor and ‘original’ voice of personal amator y poetr y, Sappho. Our focus is now on a single, hybrid, ar tist–love r, whose letter subsumes several layers of dialogue. Sappho–Ovid’s often cacophonous duet, I’ll argue, takes us one move closer to the lovers’ exchange of letters in Her.1621. In assuming the voice of Sappho, Ovid in many ways writes over and through his female rival (and possible par tner), aggressively asser ting a hierarchy of male over

32 Se e Me t . 10. 152 4, cf. Ar s 1.314. There will be ‘n o crime’ in Ovid’s song, whereas Orpheus will give us ‘unlawful passions’ and the ‘deser ved punishment’ of women, promising also tales of homosexual instead of heterosexual love affairs. This is a point made by Janan ( 1988 b) 116 .

Introduction: Na rcissus and Medusa 11

female, heterosexual over homosexual, writing subject over written object. Yet ultimately that simple reading cannot stand. Ovid’s authorial iden- tity is blatantly and inextricably inter twined with Sappho’s , so that the question becomes not only ‘what is Ovid doing with Sappho/woman?’, but ‘how does Sappho’s voice write Ovidian desire?’, ‘how d o Ovid and Sappho interrelate and to what effect?’ and ‘what does count as a genuine Ovid/Sappho?’ Inter textuality in the letter now becomes a model and vehi- cle for intersubjectivity, and for authorial and sexual interdependency. This is not to say that authorial identities ultimately fuse or blur into one, or that Ovid and Sappho are (just) constr ucted as ‘equals’, but that Ovid’s alliance/overlapping of writing voices hypes up power-play and involves readers in a drama full of suspense. It is in and with this relationality, I argue, that Ovid creates a space for desire and/as creativity, and makes reading seductive. Chapter 5 turns to Her.1621, and focuses on how the double epistles deal with the topic of marriage and coupling (the poems are themselves set out as concatenations of imper fectly mirroring symmetries, constantly playing out not just doublings but triangulations and multiple splittings). Chapter 6 , which in many ways is chapter 5 ’s Siamese twin, plunges into the centre of the collection to pore over the state of Hero ’s and Leander’s relationship. It is interesting that in terms of the recent (feminist, woman- centred) emphases of recent Ovidian criticism, the double Heroides look (or have looked) pretty unsexy. A s Harrison writes in the recent Cambridge Companion to Ovid :

This second collection was perhaps stimulated by the (lost) male replies written by Ovid’s contemporar y Sabinus to the single Heroides ( Amores 2.18 .27 34 , Ex Ponto

4. 16 . 1314 ), but was no doubt also intended to give a n ovel twist to a sequel to

a successful collection. He re we might see the reinser tion of the male as a prime erotic move r a s a reversion in some sense to traditional love elegy, but the generic move upwards from the Amores is maintained overall in the continuing dramatic

and mythological frame work, with material again derived from epic, tragedy and Hellenistic narrative poetr y. 33

Thus putting a male writer back at the helm after the female voices of the single He roides is seen to constitute a move ‘up’ and back to ‘tradi- tion’. Scope for locating a seditious, modern, female voice is now dimin- ished, and like wise anxieties of communication (wrapped up in the satire of inter textual games which so appealed to the postmodern reader) are healed. We know the letter gets to its destination safely, because the lover



v i c t o r i a r i m e l l

replies, while there remains little titillation for the unintended Ovidian reader intercepting the ‘private outpourings’ o f the female psyche. So do the double letters diffuse and resolve ever ything that was foxy and enticing about the single letters? What happens when Ovid writes elegy in dialogue? How do male and female writers interact in the double letters? I suggest in this chapter that Leander and He ro are not simply polarized, as crit- ics have tended to suggest (male/female, power ful/trapped, strong/weak, active/passive, present/absent, public/private, epic/elegiac), but per form a complex double act, each writing from a shoreline looking out to sea, revie wing the same memories, coping with the same vacillations of emo- tion and intent, reflecting, predicting and outwitting each other. Hero and Leander invade each others’ imager y, script and role, while the Hellespont provides the per fect fluid (yet demarcating) space in which this couple’s rivalr y and mutuality can operate, becoming a clogged reser voir of allu- sions, reflections, fantasies and dreams. Both writers, I argue, flit between Na rcissus-like and Echo-like roles, but He ro’s letter, whether she realizes

it or not, is also spiked with the powers of Medusa’s gaze, as her winking

lumen lights Leander’s path. Finally, I should point out that I have tried carefully to avoid imposing too homogenizing a frame on bodies of texts and images which churn up inconsistencies and surprises together with pleasing harmonious patterns. Although the essays in this book all cover and map the same territor y, they never theless pull in different directions, and offer up va r ying rhythms and textures, as do the poems themselves. My writing style, too, is intended to convey as far as possible the aesthetic, emotional and psychological impact

of these texts, together with their (not at all separate) cerebral thrills. I mean to stress the extent to which it is difficult (and beside the point) to retain

a rigid sense of control over the maze of connections evoked in detailed

readings of Ovid’s work. 34 For there is a sense, I think, in which classicists

(unlike modern poets and the general reader of Ovid in translation) 35 have tended to lose touch with the ways in which Ovid powers the imagination

and lets us reve l with him in fantasy that can be as intense and frightening

as it is witty and fun. In par ticular, this book is meant to evoke the complex,

tangled and paradoxical nature o f the Medusa and Na rcissus myths, which to various degrees (as literal references and allusions, but also as wavering, mutating images and metaphors) scaffold my readings throughout. What will emerge, I hope, is not just a jigsaw of precise and detailed arguments,

34 Despite the fact that, to an extent, this is of course what all criticism does.

35 Se e e.g. the collection of free translations of Ovid in Hofman and Lasdun (1994 ): the poets here all in different ways emphasize the sensual beauty and visceral violence of Ovid’s language and narratives.

Introduction: Na rcissus and Medusa


but a por trait of the proliferating and at times be wildering reflections that characterize Ovid’s vision of the self, and of his own literar y career.


d a n g e r o u s l i a i s o n s

I want to begin at the beginning, where desire and poetr y, come into being. The Amores , Ar s Amatoria and Metamorphoses all open with acts of rape (from the stripping and suggestive conquest of Corinna at Am . 1.5 , and the assault of the Sabine women in Romulus’ theatre a t Ar s 1 . 10134 , t o Apollo’s primus amor for Daphne in Me t.1 .452 567). Or rather, they star t with a look, a scopophilia which both objectifies (turns into stone) in the manner of Medusa, and represents the unner ving move from primal narcissism to self-reflective subjectivity that takes place in the encounter with an other (most obviously, woman, Medusa), spurring Perseus’ preemptive strike. With Ovid, we undress Corinna as we look at her ( ecce, Corinna venit / ‘look, here comes Corinna’ Am .1.5.9 ; stetit ante oculos / ‘she stood before my eyes’ 17 ; nil non laudabile vidi / ‘ I saw nothing I didn’t like’ 23 ), just as the Romans spy and mark out their Sabine victims (respiciunt oculisque notant sibi quisque puellam / quam velit / ‘they looked at them, and each one noted with their eyes which girl he fancied’ Ar s 1 .10910 ). Apollo’s hot pursuit of Daphne and Ovid’s sly objectification of Corinna in Am.1 .5 become windows through which we vie w Tereus’ lustful visions of Philomela in Me t .6 (he first rapes her with his eyes: exarsit conspecta virgine / ‘he burned with lust when he looked at her’ 6 .455 ; spectat eam Te reus praecontrectatque videndo / ‘ Te reus gazes at her, and as he looks he feels her already in his arms’ 478 ) i n a tale which weaves a rival beginning and aition of poetr y into the Daedalean middle of Ovid’s epic. 36 Violent male desire, aligned with the r unaway fantasies of a hyperactive poet, energizes a compulsive cycle of competition and retaliation which animates much of Ovidian poetr y, forcing the hand of female creativity and setting the stage for an illimitable opposition/alliance of gendered ar tists/subjects. While in the Amores , poetess Corinna’s foray into adulterous epic at 2. 11 (narrabis multa 49) is fantasized only to be suppressed or forgotten in a ne w triumph of the capta puella at 2 .12, 37 in the Metamorphoses , violated Philomela learns

36 Note the parallelism between the similes comparing the heat of desire t o burning stubble or corn at Met .1 . 492 and 6 .456 , and the echo of ecce Corinna venit (Am .1 .5. 9 ) in ecce venit, referring to Philomela, at Met. 6. 451. This is noted by Hardie (2002 a) 2603 .

37 Cf. Am. 1 .5 (compare ecce Corinna 1 .5. 9, with ecce Corinna 2 . 12.2). In Am. 2 .12. 17 24, Ovid reminds us again how women (Helen, Lavinia, the Sabines) have triggered wars in histor y and myth.



v i c t o r i a r i m e l l

to write and inspires the various spinners and seamstresses who pick up Ovid’s narrative in the epic (as well as in the He roides ). So too, Medusa, raped by Neptune and robbed of her beauty in Me t. 4, becomes a talented sculptress, her ambitious, exper t eye turning all it sees to marble. Her tale begins as a familiar one of erotic pursuit:

excipit unus ex numero procer um quaerens cur sola soror um gesserit alternis immixtos crinibus angues. hospes ait ‘quoniam scitaris digna relatu, accipe quaesiti causam. clarissima forma multor umque fuit spes invidiosa procor um illa, nec in tota conspectior ulla capillis pars fuit; inveni, qui se vidisse referret. hanc pelagi rector templo vitiasse Miner va e dicitur; aversa est et castos aegide vultus nata Iovis texit, neve hoc inpune fuisset, Gorgoneum crinem turpes mutavit in hydros. nunc quoque, ut attonitus formidine terreat hostes, pectore i n adverso, quos fecit, sustinet angues.’

But one of the princes asked to know why just one sister wore snaky braids. The guest replied: ‘Since it is such a fabulous tale, I’ll tell yo u why. She was once a stunning beauty, the jealous hope of many suitors. And of all her assets, her hair was the most striking.

Or so I heard f rom a man who claimed he’d seen her. They say that sea-king Neptune raped her in Miner va’s temple:

Jove ’s daughter turned her face away, hiding chaste eyes behind her aegis, and that the deed be punished as was due, she changed the Gorgon’s locks to ugly snakes. And now, to frighten fear-numbed foes, she wears upon

her breast the serpents she created.

Me t .4 .790 803

Just before this account, at the end of Met.4 , Perseus tells of the ‘wondrous valour’ by which he won Medusa’s snaky head. After stealing the single eye shared by the Graiae, he trekked through rough, trackless woods to where the Gorgons lived, past dozens of men and beasts petrified in the act of looking at Medusa’s face. Armed with a reflective shield, he made his attack:

se tamen horrendae clipei, quam laeva gerebat, aere repercussae formam aspexisse Medusae, dumque gravis somnus colubrasque ipsamque tenebat,

Introduction: Na rcissus and Medusa


eripuisse caput collo; pennisque fugacem Pegason et fratrem matris de sanguine natos.

Bu t he saw Medusa’s horrid face reflected in the bright bronze shield his left hand bore; and while deep sleep disarmed her, and her snakes, he cut her head clean off; the blood gave bir th to

swift-winged Pegasus, and his brother.

4 .782 6

Without pausing to admire this defeat, or extend an account ‘spoiled by its brevity and vagueness’, 38 Me t.5 .1249 continues the saga (the first word of the book, dumque, repeats dumque at 4 .784): the banquet for the wedding of Perseus and Andromeda (whom the hero saved from the sea monster in Met .4 .673 752), becomes the site for an epic battle which replays the scene between Odysseus and the suitors during the feast at the palace of Ithaca. Perseus has to resor t t o using Medusa’s head as a lethal weapon, and by the end of the episode, the banquet hall is lined with corpses and statues.

ve r u m ubi vir tutem turbae succumbere vidit, ‘auxilium’ Perseus, ‘quoniam sic cogitis ipsi’ dixit, ‘a b hoste petam. vultus aver tite vestros, si quis amicus adest!’ et Gorgonis extulit ora. ‘quaere alium, tua quem moveant miracula’ dixit Thescelus, utque manu iaculum fatale parabat mittere, in hoc haesit signum de marmore gestu. proximus huic Ampyx animi plenissima magni pectora Lyncidae gladio petit, inque petendo dextera deriguit nec citra mota nec ultra est. . hi tamen ex merito poenas subiere, sed unus miles erat Persei, pro quo dum pugnat, Aconteus, Gorgone conspecta saxo concrevit obor to. quem ratus Astyages etiamnum vivere, longo ense ferit; sonuit tinnitibus ensis acutis. dum stupet Astyages, naturam traxit eandem marmoreoque manet vultus mirantis in ore. nomina longa mora est media de plebe viror um dicere; bis centum restabant corpora pugnae, Gorgone bis centum riguer unt corpora visa.

Bu t Perseus saw he was outnumbered, and cried, ‘Since yo u force me to it, I’ll go to my own enemy for aid:

if any friend of mine is here, be sure t o turn away!’ And with that he raised the Gorgon’s head up high.



38 Anderson ( 1997) 494.



v i c t o r i a r i m e l l

‘ Tr y scaring someone else with your so-called magic tricks’, yelled Thescelus, aiming his deadly spear, about to throw; but in that ve r y pose he stuck, a statue made of marble. Ampyx, next, tried stabbing great-souled Perseus in the hear t; but as he thr ust, his hand went rigid – froze n i n the act. .

Those men got punished as deser ved, but there was one, Aconteus – on Perseus’ side – who looked upon the Gorgon’s face and hardened into stone. Thinking he was still alive, Astyages pounced with long sword drawn; the blade clanged as it str uck. As Astyages stood be wildered, the same uncanny powe r took its hold, and there he stood, his marble face amazed. It would take an age to list the ordinar y victims; two hundred men sur vived the fight; two hundred

saw the Gorgon and we re petrified.

Met . 5 .177 86, 200 09



Unlike her close relative Narcissus, who, as we ’ve seen, has been elevated as the supreme model for poetic illusionism and for the insatiability and

emptiness of (Ovidian) desire, Medusa is a dispassionate statue-maker and Pygmalion-nemesis who intends to stamp out life and presence, without remorse. 39 Like Narcissus, she enthralls, stupefies, and is finally killed (con- ceivably – we can never quite see ) by her ow n glance, yet she does not waste away into the landscape: the blood from her decapitated head spawns ne w life in the shape of Chr ysanor and Pegasus, from whose hoofprints gushed the springs of the Muses, Aganippe and Hippocrene. Indeed, as an aition of poetr y t o rival that of Narcissus, Medusa’s myth is knotted with contradictions: the (once) beautiful, hideous, irresistible/repulsive Gorgon both excites and castrates, paralysing men in permanent arousal/death. As de Vo s writes, echoing Clair and many other twentieth-centur y critics and

ar tists, ‘Medusa represents what may not be represented

she wanders

in a territor y that belongs to both death and life, male and female, order and chaos, the visible and the invisible.’ 40 While she is, as ‘the look’ itself, an essential figure within the specular economies of Ovidian poetics, she is also a blind spot, a myster y we can’t look at directly, existing, almost, ‘behind the mirror’. 41 For while we see and feel her effect, and glimpse par ts

39 Although in Met. 5.2036 , a s we have just seen, Astyages mistakes one of her petrified victims, Aconteus, for a living man. On this occasion the effect of Na rcissus’ and Medusa’s creativity is the same, but its motivation ver y different.

40 De Vos ( 2003).

41 That is, Medusa might be a figure for male fear of and need to conquer the other via specular duplication, but arguably she also fits Irigaray’s analysis of woman and especially the female genitals as in some sense outside of any specular representation. Sh e i s not simply the horror of ‘nothing to see’ (Irigaray 1985 a, 47), but a complex, phallic-vulvic myster y that is never tr uly seen.

Introduction: Na rcissus and Medusa 17

of her, especially her hair, the spotlight throughout Ovid is not on her but on the faces of her astonished victims. 42 We find her loitering in a twi- light zone at the entrance of Hades, 43 the only mor tal Gorgon, 44 ye t still not quite human (the snake is an ear thbound creature, but can magically shed its old skin and appear to recapture its youth), 45 her vulvic head a reminder of the gate to all mor tal existence, the ontological threshold. Her look is the model for all human interaction, especially between men and women, but at the same time she’s the most blatantly fantastical monster of all (Tr . 4 . 7 .11 20). She is both the sign (after Freud) of the possibility of phallic womanhood, and a p re-castrated freak, an emblem of the inevitable punishment of female powe r. 46 Hence Medusa has appeared in twentieth-centur y feminist thought as both a figure for the silencing of women, for women’s self-hatred, and for the free-rein given under patriarchy to sadistic fantasies of women, 47 but also as an icon of resistance and rage at female subjugation, or even, faced fearlessly, as a source of vibrant creativity. 48 Meanwhile, in popular culture since the early 1990s (and since ‘that dress’ that launched Elizabeth Hurley’s career), the Versace brand, with its omnipresent Medusa-head logo, has defined the paradoxe s of a controversial, ne w-generation feminism which celebrates empowerment in the act of attracting and manipulating a male gaze. 49 Similarly, when Medusa’s terrifying head is transferred to Athena’s shield at Met . 4 .802 3 , i t becomes an apotropaic emblem and source of civic and mar tial strength. As Garber and Vickers summarize i n their recent

42 This is of course precisely her seductive appeal: like Te reus eyeing up Philomela in Me t .6 , we have to imagine what we can’t see (fingit quae nondum vidit 6. 492).

43 Snakes in general are threshold figures, especially in Ovid, where they are often metapoetic markers of the ends of books (as the snake recoils, so we wind up the papyr us). Se e Barchiesi ( 1997) 1903 .

44 See Hesiod, Theogony 27083 .

45 See e.g. He rcules at Ovid, Me t. 9 .266 ( utque novus serpens posita cum pelle senecta / luxuriare solet squamaque nitere recenti, / sic, ubi mor tales Ti r ynthius exuit ar tus, / par te sui meliore viget maiorque videri / coepit et augusta fieri gravitate ve rendus ). Also Ar s 3. 77, Lucretius 3. 612 14, Tibullus 1 . 4 .35 , Lucan 9. 717 18, Pliny, NH 8.41 . 99 .

46 As Kofman (1985 ) puts it, women’s genitals ‘at once awaken and appease castration anxiety’. Se e extract in Garber and Vickers (2003 ) 167 .

47 See Cixous ( 1976 ).

48 See e.g. ‘ The muse as Medusa’ (Sar ton 1974 ), or ‘A navy-blue afro’ (McElroy 1975), also published in Garber and Vickers ( 2003).

49 Madonna led the way in this during the 1980s and 1990 s, and her original ‘Girl Power’ was inherited by ‘daughters’ Britney, Christina, and the Spice Girls (themselves a craftily engineered, Medusan blend of ‘scar y’ and ‘baby’, pouting Posh and girning Ginger). Se e discussion in Sharrock ( 2002 ) and de Lauretis ( 1984 ), alongside the extended critiques of ocularcentrism in Irigaray (especially 1974) and Cixous (e.g. 1984 ). Elizabeth Hurley once said, laying claim to the fashionable feminist angle of Versace’s vision, ‘Unlike many other designers, Versace designs clothes to celebrate the female form rather than eliminate it.’ Of course, the sexual charge of the red-carpet Versace dress also hinges on and exploits the ‘scariness’ of the beautiful woman.



v i c t o r i a r i m e l l

collection, the Medusa myth seems to include, from its ve r y beginnings,

a mass of conflicting and confusing elements. 50 In par ticular, critics have

long debated the status of the mirroring moment that immediately pre- cedes petrification. As Goodman puts it: ‘what or who, exactly, i s turned into stone by Medusa? Is it those who dare to look at her, o r those at whom she looks?’ 51 The identities of Perseus and Medusa overlap as well as collide:

in their first meeting, when Perseus uses a mirrored sur face to look upon the head of his enemy, h e risks coming face to face with his own Medusan reflection, that of the bulging-eyed, snarling, enraged warrior r ushing to attack. 52 Then, in the aftermath of her decapitation, he uses her head as

a mask-like weapon, and becomes her lethal gaze. 53 As Freud, Lacan and

Sa r t re have all recognized, there lies within the Medusa myth a kind of identification between seer and seen, a meeting of looks rather than (or as well as) a simple subject–object relation. 54 For both Medusa and Perseus, this encounter functions as a Lacanian mirror-stage, in which gazer and gazed-upon are looked at in a dialectic of mutual reflection. 55 Ye t a t th e same time, their clash reminds us of the problematics of per fect specularity, the cracks that disr upt the operation of Lacan’s model. As Derrida posits in Of Grammatolog y, the reflective interplay of apparently identical images is always based on an inevitable disunity that already defines the first image, creating an invisible otherness or ‘surplus’ that disr upts specular harmony. 56 In the first chapter of this book, I discuss how this dazzling, enigmatic, drawn-out flash of mirroring becomes the nub of the Medicamina , when the woman Ovid makes over washes off her face-pack and emerges not just as a female Na rcissus, but as an embodiment of the mirror itself, a gorgeously reflective face ready to stun men (or itself, a s it gazes admiringly

50 Garber and Vickers (2003 ) 7 .

51 Goodman (1996 ). See extract in Garber and Vickers (2003 ) 272 .

52 Vernant (1991 ) 118, notes the affinities between the mask of the Gorgon and the facial mimicr y of the beserk warrior.

53 Me t. 5 . 178 9 captures this paradox: auxilium

54 The architectonics of the Freudian and Lacanian subject depend fundamentally on seeing the self exteriorized, for which the Perseus–Medusa encounter is a cr ucial model. For Lacan in par ticular, this encounter replicates the instantaneous identification, the specularity and symmetr y that characterize the Imaginar y o rder. For Sar t re , Medusa is ‘the look’: his entire discussion of human relations is

developed within the context of the subject–object conflict which the look initiates. Se e especially Sa r t re ( 1966) 555: ‘the profound meaning of the myth of Medusa is the petrification of Being-for-itself in the Being-in-itself by the other’s look.’ Also see Laing (1970 ), and discussion in Barnes (1974)

a b hoste petam .

22 8.

55 Lacan defines the essence of the Imaginar y as a dual relationship, a reduplication in the mirror, a n

immediate opposition between consciousness and its other in which each term becomes its opposite and is lost in the play of reflections.

56 Derrida (1976 ) 36.

Introduction: Na rcissus and Medusa 19

in the looking glass), as well as show them their own image (quaecumque afficiet tali medicamine vultum, / fulgebit speculo levior illa suo / ‘whoever shall treat her face with such a medicamina, will shine smoother than her ow n mirror’ 678). We never quite forget that (this) woman, though cosmetically preened (and pr uned) was once wild: a landscape of devouring brambles, a bitter fr uit (Me d. 3 5 ), a Medusan glare. In the Ars Amatoria (as I discuss in chapter 2 ), the look is ever ything, a complex concentration of written and spoken expression ( saepe tacens vocem verbaque vultus habet / ‘a silent face often communicates voice and words’ Ars 1.574 ), 57 and despite Ovid’s many prescriptions of the order seduction should follow (the man should make the first move , h e should send the first letter, etc.), the contention of male and female gazes destabilizes hierarchy and linearity in a way reminiscent of the uncer tainties of positioning germane to the Perseus–Medusa confrontation. Thus in the two parallel passages about mutual mirroring at Ars 2 .197 202 ( arguit: arguito; quicquid probat illa, probato 199 ) and Ar s 3 . 513 14 (spectantem specta; ridenti mollia ride 513), we lose track of whose gaze i s dominant, and who looks first. 58 In the Ar s , one can be active i n passivity, vulnerable in the attack, a par t of the spectacle one obser ves, 59 just as in looking (aggressively, objectifyingly) at Medusa, in meeting the eyes of an other, we realize that we have become object to another’s subject. This is precisely, a s I suggest in chapter 3, what Orpheus learns in Met.10 , when his backwards look kills Eu r ydice for a second time, but also (figuratively, in the simile at 10 .6471 ) turns him into stone – a scenario which repeats Perseus’ first sighting of Andromeda chained to the rock in Me t. 4 . 6737 . In the face of Narcissus’ (failed) desire, which achieves unity only in tragic sameness, Medusa posits a radical alterity, and catalyses desire a s a disorganizing and dislocating force that is both cancerous and creative. In embodying fear of the other, of the monstrous, the unlike-self, Medusa thr usts unfamiliarity into our ver y I, and figures the dialectical relation between same and other. Thus Ha rdie’s point that Ovid (like his readers) plays Na rcissus at successive stages of awareness, as both lover–poet

57 Cf. Am. 1. 4 . 19 : verba superciliis sine voce loquentia dicam / ‘with my eyebrow s I’ll tell yo u words that speak without sound’.

58 Cf. Am . 1. 4 .1718 : me specta nutusque meos vultumque loquacem; / e xcipe fur tivas et refer ipsa notas / ‘keep your eyes on me, my nods, the language of my face. Catch my secret signs, and send me back your own.’

59 Ar s 1. 2778 presents a typical paradox: conveniat maribus, ne quam nos ante rogemus, / femina iam par tes victa rogantis aget / ‘if it suited us men not to ask a woman first, then she, already won, would play the asker’. Compare the advice of Dipsas, a rather Medusan figure with flashing eyes (oculis quoque pupula duplex / fulminat 1 . 8.15 16) at Am . 1. 8: et quasi laesa prior nonnumquam irascere laeso / vanescit culpa culpa repensa tua / ‘sometimes too, when yo u have hur t him, get angr y, as if injured first – charge met by counter-charge will cancel itself out’ 79 80 .

2 0

v i c t o r i a r i m e l l

nostalgic for a t r ue, primordial feigning and pomo ironist delighting in imperial ar tifice, might be recast as a tension between (gendered, same but different) Na rcissistic and Medusan creativities, between ar t a s insubstan- tial (oral/mental) invention, and ar t a s monumental permanence marking representation as death. 60 Ye t Medusa fur ther complicates this oscillation between eros and thanatos , speech and writing, nature and culture, by incarnating and deconstr ucting its tensions: hers is a healing/poisonous blood (much like Ovid’s phar makon in the Remedia Amoris ) 61 and the fons sacra born of her son Pegasus, which rivals and reclaims the fount in which Na rcissus’ fictions are reflected, is also the bir thing pool for Fama, the arch illusionist and relentless gossip who whips up epic plots throughout the Metamorphoses . Indeed, as Ha rdie points out, the only way anyone can ever really ‘see’ this miraculous spring is through fama, literar y tradition. 62 Pallas Athena repor ts in Met. 5.256 9 :

fama novi fontis nostras per venit ad aures, dura Medusaei quem praepetis ungula r upit. is mihi causa viae. volui mirabile factum cernere; vidi ipsum materno sanguine nasci.

I’ve just heard about a brand ne w spring, which gushed from the hard hoof of Medusa’s flying horse. This is the reason for my journey; I wanted to set eyes on this mar vel; I saw the same horse being born of his mother’s blood.

Similarly, Pindar, Pythian 12 identifies Medusa and the Gorgons as the ori- gins of a popular music entitled ‘the many-headed tune’. 63 The Gorgons’ world, as Vernant puts it, is ‘one of disquieting noise’. 64 The ar t o f the flute was invented by Athena precisely to simulate the shrill sounds she heard escaping from the Gorgons’ mouths, ye t interestingly, this innova- tion triggers a repetition of Medusa’s defeat and another round of poetic competition: for when she plays the flute, her face re d and puffed up, Athena looks at her reflection in water and realizes she has become Medusa, whereupon she throw s away the instr ument, which is then taken up by Marsyas, who ends up flayed by Apollo for daring to challenge him in a musical contest. The figure of Medusa (especially as she is bound up in a mirroring relationship with Narcissus) has multiple faces, and her ability

60 Se e Hardie ( 2002a) passim, and Rosati ( 1983 ).

61 Se e Euripides, Io n 1005 . Rem .323 (et mala sunt vicina bonis / ‘there’s such a fine line between faults and charms’) might ser ve as the slogan of Ovid’s double-edged text.

62 Ha rdie ( 2002a) 238.

63 Pindar, Pyth.12. 1223.

64 Vernant (1991) 124.

Introduction: Na rcissus and Medusa 21

to dr um up strife and opposition while also incapacitating the dualistic logic that turns her into death itself, makes her a power ful figure for Ovid- ian metamorphosis and for the elegiac rhythms which snake through all Ovid’s texts. 65 As the Gorgon’s petrified subjects also look (or symbolically, according to Freud and Ferenczi, are) 66 permanently animated/aroused, so the vast majority of Ovid’s literal metamorphoses are tor tures which pun- ish by suspending victims in a state of painful semi-death (at Me t.10.487 , for example, Myrrha pleads, mutataeque mihi vitamque necemque negate / ‘change me and refuse me both life and death!’). And each of Ovid’s elegiac couplets, as visualized in the first poem of his debut work (Am. 1. 1 .3 4 , 17 18) suspends difference in repetition, wedding tumescence with castration, deflation with re-creation, in compulsive limping rhythm. I will be suggesting throughout this book that the crackle of Medusan magic in the Ovidian woman helps stage desire not just as a bland, incessant game of absence and lack but also as a collision of creative energies and convictions. Medusa shadows Ovidian erotics, playing Cupid’s evil sidekick (just as Na rcissus is hounded by his hellish double Tantalus), hurling snakes in place of arrow s t o fi re lethal ardour in her victims. He r gaze i s embodied in both the elegiac puella, with her flashing, enrapturing look ( o facies oculos nata tenere meos Am.2. 17 .12), 67 and in the ar tifying glare of the poet inspired by ‘traces of the Medusan steed’ (Fasti 5 .8 ; cf. Ex P. 4 . 8 . 80) who even rejects the potential self-destr uctiveness of his ow n creative might (Medusa’s powers would fail on him, he claims at Ex P. 1. 2.35 6). 68 The imaginative leaps of the exile poetr y a re arguably already present as sparks

65 Although it is beyond the scope of this book to discuss this in detail, Melanie Klein’s work on the creativity of envy and aggression is perhaps useful here for understanding Medusa’s maternal hybridity and paradoxical nature: Klein challenges Freud’s late writings, in which he comes to see all human beings as living under the destr uctive force of a terrifying death-drive, arguing instead that aggression, destr uctiveness, envy and mourning are cr ucial to the process of individuation,

and to creative development generally. Violent, murderous fantasies harboured by the child towards the mother facilitate the containment of destr uctive drives and allow for a c reative development of projective and introjective p rocesses, deemed vital by Klein for successful self-organization. Se e especially Klein (1957).

66 See Freud ( 1922) and Ferenczi ( 1926).

by you r e yes , that have taken mine

67 Cf. p a rc e

perqu e tuos oculos, qui rapuere meos / ‘spare me

captive’ ( Am .3 .11 b.45 8).

68 ipsa Medusa oculis veniat licet obvia nostris, / amittet vires ipsa Medusa suas / ‘if Medusa herself came before m y eyes, even Medusa would lose her power’. In affirming, at once, both Ovid’s tragic powerlessness in exile and his successful sur vival as an ar tist in the most testing of circumstances, these lines encapsulate the ner vous paradoxe s r unning through the Tristia and Ex Ponto. Kristeva (2003 ) 43, talks of a ‘g en´ ealogie´ secr ete` entre le pouvoir des Gorgones et l’exp erience´ esth etique.´ Elle nous fait comprendre que si l’ar tiste par vient a` eviter´ d’etrˆ e la vittime de M eduse,´ c’est parce-qu’il la re fl ete` tout en etant´ une transubstantiation de son sang.’

2 2

v i c t o r i a

r i m e l l

in Ovid’s first elegiac verse, and I a m conscious of the fact that, at either end of a poetic career, the Amores and Epistulae Ex Ponto border and extend this book’s central concerns. At Am. 3. 12 , Ovid boasts: per nos Scylla patri caros furata capillos / pube premit rabidos inguinibusque canes; / nos pedibus pinnas dedimus, nos crinibus angues / ‘it was we poets who made Scylla steal her father’s t reasured locks, and hide savage dogs in her groin. It is we who have given wings to feet, and mingled snakes with hair’ (213 ). Medusa (as the mythic figure we can never quite set eyes on) stands for the most wondrous and magical of tales. Equally, Ovidian poetr y can (like Medusa herself ) control nature, destroy snakes (those quintessential gazers) after pulling out their fangs (car mine dissiliunt abr uptis faucibus angues Am .2 . 1.25 ). And when in Am.3 .6 these phallic powers (ostensibly) falter, and the poet is unable to cross the stream to meet with his domina (just as he fails to get an erection in the following poem, 3 . 7, where h e lies, a mere species , o r spectacle, on the bed 69 ), he wishes he had the wings Perseus wore ‘when he carried off that head thick with dreadful snakes’

( terribili densum cum tulit angue caput 3 .6.14 ). The Ar s Amatoria develops

this anxious, double-edged poetic ego fur ther, when it makes Perseus its leading example of a successful love r (see Ar s 1. 53 5 ) – though at this point

the condition for winning Andromeda (conquering Medusa, and the snake- like sea monster) 70 is repressed, only to come back to haunt pupils in Ars 2 , when it is revealed that women who appear to be ‘mild and gentle’ to their lovers may in reality be ‘more violent than grim Medusa’ (tor va violentior illa Medusa 2.309 ). In the Amores, Corinna’s hair is the focus for a stretching of tensions between objectified puella and petrifying domina, between phallic and impotent authors, narcissistic and Medusan desires. We can glimpse here the beginnings of Ovid’s career-long interest in the possibilities of female creativity and in the bitter thrill of male–female rivalr y, or the relation- ship between two elements in a couple more generally. 71 While at the star t of the collection Ovid looks for ward t o acquiring some well-turned out subject-matter, aut puer aut longas compta puella comas (‘either a b oy or a girl with long and tidy hair’ Am.1 .1 .20 ), and Corinna enters at Am.1 .5 with

69 Am. 3 . 7. 15.

70 In his fight with the sea monster about to attack Andromeda, Perseus is compared to an eagle pouncing on a serpent in a field ( Me t .4. 71420).

71 I.e., in Ovid erotic relationships are always mirrored/enacted in poetic ones. In the Amores , the poet’s relationship with Corinna and other girls gets dramatized on a small scale in the elegiac couplet and on a larger scale in the diptych poems, e.g. 2.11 and 2 . 12, 2 . 7 and 2. 8 , which turn reading into a game of comparing and contrasting, undulating back and for th, making connections between two (or more) poems, a process which has us almost literally dancing to elegiac rhythms.

Introduction: Na rcissus and Medusa 23

tresses combed in a centre par ting (candida dividua colla tegente coma 10 ), he still finds something to avenge (like malus ultor Orestes) at Am .1 .7, and tears her digestos capillos ( 11) 72 commenting that her ‘disordered locks’ we re in fact ‘most becoming’ ( nec dominam motae dedecuere comae 12 ), before advising her to rearrange them ( pone recompositas in statione comas! 68 ). 73 At Am. 1 . 11 , he gets Nape, adept at ordering incer tas crinos, on side, ye t Am . 1. 14 visualizes Corinna’s transformation from (coiffured) Na rcissus to (snaky, frizzy, poison-haired) Medusa, and from the inspiration for fine- spun, subtly coloured deductum car men (see lines 510 ), to a fashion model for tor tured, imperial, post-Callimachean kitsch (at lines 45 50 , she’ll have to wear a gaudy blonde wig created from the shorn heads of German pris- oners of war – a triumph that brings no glor y for originality). Her hair used to be malleable, passive, picture-per fect, the ideal elegiac materia ( dociles et centum flexibus apti 13), so beautiful Bacchus and Apollo could have lusted after it:

formosae periere comae, quas vellet Apollo, quas vellet capiti Bacchus inesse suo;

That luscious hair is dead – hair even Apollo, or Bacchus, would have found desirable.

Am .1. 14. 31 2

This couplet evokes the description of Na rcissus at Me t. 3 .4201 (specta t

dignos Baccho, dignos et Apolline crines / ‘he gazed

or Bacchus’), at the point at which he is na¨ıvely in love with himself. But now she has singed it into sinuous ringlets, and dyed it too, applying the ‘mingled poison’ to her ow n head (ipsa dabas capiti mixta venena tuo 44). Like Medusa, Corinna has been punished for being ‘conquered’ by Ovid in Amores 1 . 5 by having her best asset r uined and turned into venomous snakes, whereupon the poet, Perseus-like, holds a mirror up to her metamorphosed face:

a t hair wor thy of Apollo


72 Just as he can’t restrain himself from tearing the hair of Corinna’s Medusan praeceptor amoris in Am . 1. 8. 11011.

73 Cf. Apollo in Me t. 1 , whose unsatisfiable, narcissistic lust for Daphne translates as a desire t o o rder her messy hair (spectat inor natos collo pendere capillos / et ‘quid si comantur?’ ait / ‘he looks at her hair hanging down her nock in disarray and says, “what if it we re properly styled?” ’ 4978). Note also that Daphne’s gleaming eyes which shine like stars as Apollo looks at them (videt igne micantes / sideribus similes oculos / ‘he looks at her eyes gleaming like stars’ 498 9 ) capture her hybridity as at once Na rcissus, object of narcissistic fantasy, and wild-haired, lightning-eyed Medusa, the virgin r unning in vain from rape (Narcissus’ eyes are also twin stars at 3 . 420). Apollo’s undressing of Daphne with his eyes, and the meeting and conquering of her gaze in lines 498 501 , reminiscent of the primus amor of Am . 1 .5, also looks for ward t o the duel of Medusa and Perseus at the end of Met .4 .

2 4

v i c t o r i a r i m e l l

quid male dispositos quereris periisse capillos? quid speculum maesta ponis inepta manu? non bene consuetis a te spectaris ocellis:

Why cr y ove r the loss of that thatch of hair? Why are you so sad to put that mirror down, yo u silly girl? You’re gazed upon by eyes not used to such a sight.

Am . 1 .14 .35 7

As I discuss in chapters 1 and 2, this is precisely the move Ovid makes in the Medicamina , when he warns women that ‘the time will come, when you’ll hate to look in the mirror, and grief will prove a second cause of wrinkles’ (tempus erit, quo vos speculum vidisse pigebit, / e t veniet r ugis altera causa dolor Me d.47 8 ), and also, even more over tly, a t Ars 3. 499508, when Ovid imagines his pupils holding a mirror up to her face, Athena-like, in the middle of a temper tantr um:

vos quoque si media speculum spectetis in ira cognoscat faciem vix satis ulla suam.

If yo u took a look in the mirror mid-passion, hardly one of yo u would recognise herself.

Ars 3 .507 8

Ye t a t the same time, like Na rcissus, Corinna is reproved for her ow n vanity, and her use of the mirror parallels Na rcissus’ (failed) self-realization. In Latin literature, the notion that Medusa is ever the modern, drop-dead beauty 74 is spelt out most clearly in Lucan, Bellum Civile 9, where she sets the trend for all Rome’s fashion-savvy women in an image reminiscent of Ovid’s Medusa-Narcissuses in his Medicamina , or of Versace’s eternal fondness for ‘big hair’ in catwalk shows driven by his Medusa-head symbol: 75

ipsa flagellabant gaudentis colla Medusae,


femineae cui more comae per terga solutae


surgunt adversa subrectae fronte colubrae, vipereumque fluit depexo crine venenum.


Medusa love d i t when the snakes whiplashed her neck. Just like women wear their hair today, the vipers hang down her back, all loose, but rear up from her forehead at the front. And the poison flows when the hair is combed.

Bellum Civile 9.632 5

74 Benjamin (1999 ) imagines Medusa as ‘the face of modernity’, blasting us with the beauty of her ‘immemorial gaze’. See extract in Garber and Vickers ( 2003 ) 89.

75 Se e Garber and Vickers (2003) for a selection of Versace adver tising images which play on the Medusa–Perseus myth.

Introduction: Na rcissus and Medusa 25

As I discuss in chapter 6, e ven the apparently picture-per fect symmetr y o f Hero’s and Leander’s letters in He r .18 19 is stained with drops of Medusan blood: although Leander declares that he is not another Perseus, and wants to avoid looking at Andromeda at lines 1513 (Andromedan alius spectet 18. 151), the aliud lumen (Hero’s lamp/eyes, 18 .155 ) that guides him over the reflective waves is to prove lethal, and his wish for Daedalus’ winged sandals (18 . 49) reminds us of Perseus’ flying shoes (a key prop in his plot against the Gorgon), for which Ovid yearns in the cr ucial parallel scene in Am . 3.6 . In her letter, meanwhile, Hero asks for Neptune’s understanding on account of his own experience in love with a string of women star ting with Medusa: when she says the Hellespont is dangerous for pretty young girls, she is thinking not only of Helle but also of Medusa, the beautiful maiden transformed into a colubrine monster after being raped by the god

of ocean ( He r .19. 12941 ). Yet still, an embittered He ro can’t resist stirring the waters and voicing her wish to meet Leander mid-strait ( 19 . 1678 ): her letter blends dreamy yearning and passivity with bristling insecurity and threat. In He r. 1819, the still, mirroring waters of decorous elegy (as of Nar- cissus’ pool), enjoye d by Leander on his previous swim across the Helle- spont (18 .7780 ) a re now stirred up into dark, menacing, and unpredictable waves. As I touch on fur ther in chapter 6, the psychological territor y o f the double Heroides , where Ovid writes from different positions simultaneously, almost merges into that of the Tristia and Ex Ponto: 76 as the poet sails into exile, he (like Leander) faces the storms of his own torment, of Augustan ire, and of past epics, and writes (from) foreign lands beyond Rome, beyond pax Augusta, and beyond the conventions of Roman elegy. In Tr .4 .7, the unimaginable emotional chaos that threatens to engulf him as he is driven into the ‘enemy territor y’ and made to look upon Rome’s imperial might as a vulnerable outsider is equated with the existence of the monstrous Gorgon:

credam prius ora Medusae / Gorgonis anguinis cincta fuisse comis

carissime, credam / mutatum curam deposuisse mei / ‘I’d sooner believe that

the Gorgon Medusa’s face was garlanded with snaky locks

you, my dearest, had changed and stopped caring about me’ (Tr .4 .7 . 1112 , 1920). The exiled Ovid is paralysed by ‘empty fears’ and ‘e xcessive d read’

nimium timor Ex P. 2 .7 .6, 7 ), immersed in Medusan

battles where ‘slender arrow s are dipped in serpent’s gall’ and ‘the human head becomes an offering’ (Ex P .4 .9. 814 ). At the same time, he is disturbed by visions of Rome’s d readful, aggressive face – whether the flashing tor va

( super vacui metus

than that

quam te,

76 Also see Rahn (1958 ) and Rosenmeyer (1997 ).

2 6

v i c t o r i a r i m e l l

for ma of a silver Augustus figurine at Ex P.2 .8 .21 2, or the hastily banished vision of the femina princeps as vicious Gorgon at Ex P .3 .1 . 124. 77 This is where Ovid tries to convince his wife to take his case to Augustus’ wife Livia:

quid trepidas et adire times? non inpia Procne filiave Aeetae voce movenda tua est, nec nur us Aegypti, nec saeva Agamemnonis uxor, Scyllaque, quae Siculas inguine terret aquas, Telegonive parens ve r tendis nata figuris, nexaque nodosas angue Medusa comas, femina sed princep s

Why are you so scared to meet her? It’s not evil Procne or the daughter of Aeetes you need to charm, nor Aegyptus’ daughter in law, o r Agamemnon’s wife, nor Scylla, terrorizing seas of Sicily with her groin. It ’s not Telegonus’ mum, with her inborn metamorphic skill, or Medusa, her hair tied up and bound with snakes,

but the First Lad y

Ex P. 3 . 1.119 26

The creative work required to compose a love letter to a belove d or suitor in a relationship which is being blocked, is on the rocks, or is as ye t unconsummated (as displayed in He r.121), is precisely Ovid’s task in the exile works, especially in the epistolar y Tristia , poems which at once taunt and seduce, addressing a deeply familiar audience and landscape which at the same time has become strangely alien ( a terra terra remota mea / ‘a land far removed f rom my own’ Tr . 1 .1 . 128). The poetics of displacement and anxiety developed in He roides 16 21, Tristia and Ex Ponto reach their apex in the riddle of Ibis, a Tarantino-esque acer vation of vendettas against a nameless enemy which summons up all of Medusa’s foreign, Eastern, othering intensity: saxificae videas infelix ora Medusae / Cephenum multos quae dedit una neci / ‘I hope yo u see the face of Medusa that turns to stone, and single-handedly decimated all those Cephenians’ ( 551 2 ). 78 As

77 Ha rdie discusses Ex P. 4 .4 . 4350 , i n which Ovid realizes that as he imagines Rome, he cannot see himself in the crowds, as an ‘inversion of Narcissus’ tragic recognition’ (Hardie 2002a, 314). But this is just one aspect, and one stage, of a more complex revisiting and development of Ovidian visuality and specularity in the exile poetr y, in which Ovid’s Orphic, imaginative powers are liable to backfire as they churn up ne w horrors and fears, ne w anxieties about the self in relation to outside/object/other/past.

78 Unfor tunately it is beyond the scope of a single book to explore i n detail how the ideas discussed here operate in all Ovidian poetr y, and I have chosen to confine my (hopefully suggestive) comments on the Amores , exile poetr y, and much of the Metamorphoses , to the introduction and conclusion.

Introduction: Na rcissus and Medusa 27

Williams and Hinds have noted, while the exile poetr y obsessively re writes Ovid’s oeuvre, it also seems to stain the works that precede it, appearing as an blowout of neurosis and rage that has been bre wing in some form or other since the Amores , f rom the comic suffering of the exclusus amator and the self-destr uctive backstabbing of the Ar s and Remedia , to the suspended tor ture and semi-death that is Ovidian metamorphosis. 79

e p i c v e n o m

Ovid’s longest and most ambitious work, the epic Metamorphoses, i s full of Medusa’s offspring – vicious baby snakes born from the gore o f her head. The Gorgon’s lookalikes – Salmacis, Andromeda, Argus, Juno, Envy, Miner va , Diana, Medea among them – all stare opponents down in this text, and charge them up for battle. 80 When in Me t.1, Juno sets hundred- eyed Argus to watch over rival Io (another raped woman rendered mute and forced to write in silence), Jupiter hits back by sending stor yteller Mercur y to conquer the beast: Argus is be witched by the tale of Pan and Syrinx, and when all his eyes are closed in sleep, Mercur y slashes his nodding head just where it joins the neck, sending it rolling down the cliff and defiling the landscape with blood ( 1.71321 ). Argus’ murder previe ws the cunning beheading of Medusa by winged Perseus in Me t.4 81 (which is also a stor y o f rival ar tists ), but the Gorgon is just beginning to infect this narrative, for as one darkness fills Argus’ multiple eyes, so Juno (like Miner va seizing the Gorgon’s snaky tresses at Me t. 4. 8023) takes them and places them on the feathers of her peacock wings, a Narcissus–Medusa re velling in the beauty of her super-charged gaze: 82

79 Williams ( 2002 ) 244, Hinds ( 1985). As Williams outlines, the obsessive tendencies of Ovid’s persona in the Ibis can also be found allegorized in the Metamorphoses, e.g. in the myth of Aglaur us, 2 .798832, while his ‘post-transformation grief’ in the Tristia and Ex Ponto resembles that of Niobe, Met. 6 . 310

12 . Note also that the final line of Ex P.4 , where Ovid claims again that he is a living corpse and

says that ‘there is n o space now for a n e w wound’ (non habet in nobis iam nova plaga locum 4 .15 . 52 ), relives the terrible fate of Actaeon, who dares to look upon naked Diana and is torn apar t by his

ow n hunting dogs until iam loca vulneribus desunt , and takes us back also to Tr . 2. 103 5, where Ovid compares his own unwitting crimen to that of Actaeon.

80 See Keith ( 1999 ) 222 , o n the similarities between Medusa and Andromeda in the Me t : ‘Andromeda’s beauty attracts Perseus’ prolonged gaze [Met. 4.6727, cf. 5 . 22 and 4 .205 ], much as the petrifying sight of Medusa’s snaky head compels the eternal gaze o f those who look directly upon her [4. 780 1

Both Medusa and Andromeda, as mesmerizing objects of the gaze, endanger the

cf. 5. 177 209 ]

men who look at them.’

81 Lucan, Bellum Civile 9 .663 makes this literal (the sword Perseus uses to behead Medusa is still stained

with the blood of ‘another monster’, Argus).

82 In the Medicamina and Ar s Amatoria, Juno, or her bird, the peacock, epitomizes the narcissistic female: see Me d. 33 4 , and Ar s .1 .627.

2 8

v i c t o r i a r i m e l l

excipit hos volucrisque suae Saturnia pennis conlocat et gemmis caudam stellantibus implet.

Saturnia takes those eyes and sets them on the feathers of her bird,

loading its tail with starr y je wels.

Met .1 . 722 3

The blood that pours from wounded and slashed serpents is the malignant life-energy of the Metamorphoses. The primal attack on Olympus by snake- footed giants and the mixing of lower and upper worlds at Me t.1 . 1824 inaugurates a bellum which will never, i n poetr y, be resolved. Jupiter resolves to cut away the infection with a knife when he kills the Python born of mother ear th (1 .1901 ), yet it continues to spread, sustaining and sustained by his ow n t roubled, faithless marriage with vindictive Juno, and by mirroring conflicts between resentful lovers and rivals (in par ticular, note that the adjective tor vus, used several times in Ovid for Medusa’s baleful gaze, e.g. tor va colubriferi lumina monstri Me t . 5.241 , is also the buzzword for Juno’s stare, e.g. Satur nia tor va Me t.4 .464 ). 83 Simultaneously, Ovid’s golden, Augustan epic explodes into action at the end of Book 1 with a climactic meeting of eyes, when Phaethon’s mother Clymene looks at the sun, the blazing, cyclopic eye of the cosmos (spectansque ad lumina solis 1. 767 ). This scene introduces the powe r of the visual to stupefy readers and scorch the globe in Book 2 , and looks for ward t o the passionate, vengeful glares of Perseus and Medusa in Books 4 and 5 (a stor y which begins when Perseus steals the single lumina of the Graiae, 4. 775 ), and to the burning up of Polyphemus’ sun-like eye, both in his lust for Galatea and in his blinding by Odysseus’ giant poker, in Book 13 . 84 The je welled yoke of Phoebus’ horses, which reflects the sun’s rays at 2 .110 (clara repercusso reddebant lumina Phoebo) becomes Perseus’ sun-like, eye-like, helioscopic shield which reflects the head of Medusa (se tamen horrendae clipei, quem laeva gerebat, / aere repercussae for mam adspexisse Medusae 4.782 3). We

83 tor vus is a fairly rare adjective i n Ovid, used a total of twenty-two times, of which four directly describe either Medusa, or Miner va taking on the terrifying qualities of Medusa that are implicitly or explicitly contained in her aegis (e.g. Met .2.752 , 6 . 34). It is always associated with the angr y, intense, and aggressive gaze, and over whelmingly with aggressive, dangerous animals (the boar and the bull), and with women, especially Juno, Medusa, Miner va and Diana, at times when they are enraged and intent on violent revenge. It also indicates a face contor ted horribly in pain or grief (e.g. Neptune at Me t .2 . 270 and Hecuba at Me t .13 . 542), and is associated with frowning, shaggy- haired tragedy at Am .3.1. 12. Medusa, as the single figure most associated with the adjective, seems to epitomize this kind of gaze a t its most threatening, and her myth also synthesizes all the slightly different meanings of tor vus in Ovidian poetr y (she is terrifying, angr y, monstrous, vengeful, but also tragically deformed, an image of grief and death).

84 Polyphemus, like the sun, has one eye ( unum est in media lumen mihi fronte 13 . 851 ; soli tamen unicus orbis 13. 853 ).

Introduction: Na rcissus and Medusa


can compare Narcissus’ repercussae imaginis umbra at Me t. 3 . 434, a s well as the shiny waves of the Hellespont at Her . 18.77: unda repercussae radiabat imagine lunae. Phaethon is blinded by the eye of the sun ( suntque oculis tenebrae per tantum lumen abor tae / ‘darkness came over his eyes from such an excess of light’ Met .2.181 ), and almost becomes a double for Medusa’s victims in Book 5, a s well as an Icar us who fails to maintain a p roper distance from consuming lumina , unlike Daedalus and his close relative Perseus. 85 Similarly, narcissistic Salmacis tests out the powe r of her ow n reflection in a pool at Me t.4. 312, before unleashing it on Hermaphroditus, becoming at once a snake-like Medusa (ut serpens 4 .362 ) 86 and an overheated sun (or its laser-like reflection):

flagrant quoque lumina nymphae, non aliter quam cum puro nitidissimus orbe opposita speculi referitur imagine Phoebus.

her nymph eyes gleamed like when a mirror held against the sun reflects its shiny, dazzling orb. Me t . 4. 347 9

We see Medusa (or her creations) ever ywhere i n the universe of Ovid’s Metamorphoses: her sun-eye illuminates the ear th, her giant, rocky Atlas holds up the entire heaven ( 4 .661 2), the ve r y sea-bed prickles with coral that is a petrified representation of her tangled hair ( 4 .750 2), while Medu- san rivers hiss through wondrous landscapes, setting challenges for intrepid heroes (e.g. Achelous at 8 .881 , 9.63 5 ), 87 and ever y wide-eyed reaction to death, metamorphosis, or fantastic fiction replicates the recoil of her victims (the verb stupeo , and the adjective attonitus , litter this epic). 88 In Metamorphoses 3 , the founding of Thebes and the Dionysiac dou- blings of tragedy and civil war begin when Cadmus’ men are stunned and then slaughtered by a snake they encounter in a primeval forest. He roic Cadmus kills the snake (like Apollo in Book 1, o r Perseus in Book 4 ) and just as he begins to be paralysed by fascination for his prey ( dum spatium victor victi considerat hostis / ‘while the conqueror stands staring at the

85 Both are master ar tists who use fake wings to achieve their greatest feat.

86 Cf. snake-slayer Cadmus, who becomes a snake in Met. 4 ( ut serpens 4 . 576 ).

87 We might also note that Nileus’ shield, engraved with the image of seven-forked Nile in gleaming silver and gold (5. 1879 ) i s a double for Athena’s shield, car ved with the snake-head of Medusa.

88 I d o not want to claim, necessarily, that ever y single occurrence of these words ‘alludes’ to the sight of Medusa, more that the experience of seeing Medusa, as dramatized in Me t. 4 and 5 , i s t he most shocking and wondrous sight of all in Ovidian poetr y (not only in Me t .), and as such becomes an impor tant paradigm, often shadowing similar encounters.


v i c t o r i a r i m e l l

bulk of his conquered enemy’ 3. 95 ), and hears a voice condemning him ‘to b e gazed upon like a serpent’ (et tu spectabere serpens 98), he is shielded by Athena (presumably with her Gorgon shield), 89 who advises him to se w the monster’s teeth in the ear th. The sur vivors of the suicidal army born of these seeds become the Bacchus-worshiping anguigenae (‘snake- people’ 3. 531) who together with their creator Cadmus found the ne w city. Throughout the ‘ Theban books’ (Me t .3 and 4), problems of self-identity and self-knowledge faced by this ne w race are bound up in the cracked- mirror clash of demon-eyed serpent(s) and serpent-killers, for which the Medusa–Perseus encounter (and the painful path to subjectivity it models) is the prime representation. The tale of Tiresias, which encompasses the prophecy of Na rcissus’ fate, begins with his interr uption of two mating serpents in the forest ( 3 .324 38 ): it is from these snakes that he acquires the gender-bending power and split identity that also defines shape-shifting Bacchus, the puer aeter nus with the head of a maiden (Met. 4 .18, 20). 90 Just as Cadmus comes close to experiencing the Dionysiac, Medusan logic that turns all spectators into spectacles (as I’ve stressed, looking at the Gor- gon always involves seeing oneself being seen, becoming object as well as subject), so his grandson Actaeon and Echion’s grandson Pentheus are vic- timized for their voyeurism in the most violent way possible, at 3 . 155 252 , and 3. 511 733 . 91 Diana, who first blushes like a cloud lit up by the rays of the sun ( 3 .183 4 ), and then casts back those rays when she ora retro flexit ( 3 .187 8), is a Medusa double and Saturnian avenger whose loose hair as she bathes (3 .169) is copied in the streaming tresses of Bacchic Agave a s she dismembers Pentheus ( 3 .7267). Pentheus above all is guilty of not believ- ing Acoetes’ account of Bacchus’ serpentine power, his ability to envelop ever ything in intoxicating vines ( inpediunt hederae remos nexuque recur vo / serpunt et gravidis distinguunt vela cor ymbis / ‘but ivy entwines and clings to the oars, coils and snakes upwards, decking the sails in heavy, hanging clusters’ 3 .664 5 ). So while, as Ha rdie puts it, ‘the city’s royal family finds its reflection in Na rcissus’ pool’, 92 Theban identity is also refracted and fractured in

89 Pallas adest 3 . 102, cf. Athena coming to Perseus’ aid at Met .5. 46 7 (bellica Pallas adest et protegit aegide fratrem / datque animos / ‘then came warlike Pallas, protecting her brother with her shield, and filling him with courage’).

90 Cadmus and his wife are finally metamorphosed into serpents, mirroring the pair of snakes Tiresias encounters in Book 3 (see 4. 576 603). Suitably, the metamorphosis occurs when Cadmus and his family are revie wing (or re-reading, relegunt 570) the misfor tunes of their house, and the transformed couple ‘remember what they once were ’ ( 603).

91 Se e Feldherr ( 1997) and Ha rdie (2002 a, 16772) o n the shift from spectator to spectated as a recurrent motif in Met. 3.

92 Ha rdie ( 2002a) 166 .

Introduction: Na rcissus and Medusa 31

Medusa’s enraged, ecstatic face – o r rather, in its multiple, imper fect rep- resentations. The fates of Actaeon and Pentheus, while operating in over t parallel with the tragedy of Na rcissus, 93 also expand the Dionysiac poetics of this book beyond the male autoeroticism and non-violent collapse of other into self that is the core of Narcissus’ narrative: androgynous Bacchus and bisexual Tiresias preside over a layered competition and compound of male and female gazes which throw identities into turmoil and leave behind a string of blinded, mutilated and tormented victims. The visual battle between genders and fissured subjects in Books 3 and 4 is imagined poetically in patterns of elegiac doubling and mirroring (most obviously in the Na rcissus episode, as Rosati has shown), 94 and becomes an allegor y for and enactment of civil strife, the conflict of brother against brother, father against son, which suspends alterity in sameness and vice versa. 95 After Book 5 , the Medusa–Perseus duel/par tnership raises its ugly head again in Me t.6 with the myth of Tereus, Procne and Philomela, and in Me t.7, in the shape of Medea and Jason. In the first tale in a string of conjugal couplings spanning Me t.6 . 401 11 .795 , the Gorgon’s rage hamstrings narcis- sistic ambitions: Tereus’ solipsistic fr ustrations and projection of his ow n desires onto silenced, Echo-like Philomela are counteracted and punished by sisters who become rival ar tists, chthonic Furies who force the narcissis- tic rapist to know himself, in the form of his beautiful miniature, Itys. Their violation teaches him what his lust for possession is really all about, and writes a drama which replays Na rcissus’ realization that other is contained in himself (inque suam sua viscera congerit alvum / ‘he stuffs his ow n guts into his belly’ 6.651 ). Like a Perseus–Medusa hybrid, with her streaming, bloody hair and speechless, mask-like visage (6.6567 ), Philomela hurls Itys’ grisly Medusan head into his father’s face:

prosiluit It yosque caput Philomela cr uentum misit in ora patris

Philomela jumps out and thr usts the gor y head of Itys

right into his father’s fac e

Me t .6 .6589

93 See especially Actaeon’s realization of his changed self in the pool at 3. 200 ; c f. Narcissus’ re velation that he is his reflection at 3 .463 .

94 Rosati ( 1983 ).

95 Compare e.g. vocat illa vocantem / ‘she calls him calling’ 3. 382; qui probat, ipse probatur, / dumque petit petitur / ‘he praises, and is himself what he praises; while he seeks, he himself is sought’ 3 . 4256 , with the fratricidal scrap at 8. 441 2 (Toxea quid faciat dubium pariterque volentem / ulcisci fratrem frater naque fata timentem / ‘then Toxeus stood hesitating what to do, wanting to avenge his brother but at the same time fearing he would share his brother’s fate’), or at 9 . 4079 ( ultusque parente parentem / natus erit facto pius et sceleratus eodem / ‘and his son shall avenge parent on parent, filial and wicked in the same act’).


v i c t o r i a r i m e l l

In his agony, Tereus calls on those (already present) snaky sisters ( vipereas sorores 662), vowing in his metamorphosed state to hit back with his own armed look ( facies ar mata videtur 674 ). The rape of Philomela and the castration of her serpent-like tongue 96 at the hands of a Perseus–Tereus, 97 followed by Procne’s Bacchic fur y and retribution (her head and face draped with snaking, black ivy) synthesize the plot of Medusa’s creation, destr uc- tion and afterlife, setting it in tandem/conflict with the tragedy of Na rcissus. Medea, meanwhile, is a Medusan serpent-slayer whose magic song can break snakes’ jaws (7.203 ) and turns the army born (like Cadmus’) of ser- pent’s teeth against themselves ( 212 ). She joins forces with Jason to anaes-

thetize the per vigilis dracon , guardian of the golden fleece ( 149), and uses her potions to rene w her father-in-law’s youth (like her own snakes shedding old skin, at 237 ), but finally leaves an audience attonitus at her crimes (426 ), escaping on a car drawn by winged dragons ( 350 , 398). In Book 8 , the boar of Calydon is a ser vant of outraged Diana which, like Diana in Book 3 , or laser-eyed Salmacis in Book 4, i s possessed of a flashing gaze ( igne micant

flamma 8 . 356 ): note that vengeful women

are compared to savage boars at Ar s Amatoria 2 .373 . 98 When the beast is finally killed, it commands a fascinated audience ( mirantes spectant 423), and Meleager’s rash presentation of its head to Atalanta (the beheading is a Medusan ‘castration’ in tandem with Scylla’s misjudged ‘castration’ o f her father’s hair at the beginning of the book) triggers a civil war, taking us back to the battles that follow Perseus’ beheading of Medusa in Book 5. In Book 9 , the duel of snake-killer Hercules and river Achelous (who can transform into a snake at will) reenacts the Medusa–Perseus conflict once more, emphasizing in its narrative twists the extent to which snake-venom corr upts seamless, unified mirroring or opposition, infecting ever ything and ever yone it touches. For while He rcules, tr ue to form, defeats his

oculi 8 . 284 ; emicat ex oculis

96 Note also that the scene in which Philomela’s amputated tongue twitches on the ground, still

tr ying to plead with Te reus, is previe wed in the battle between Perseus’ army and his enemies in

Me t. 5 (compare luctantem loqui

linguam / abstulit 6 . 5567 , with the still-complaining tongue of

beheaded Emathion at 5 . 1056 : atque ibi semianimi verba exsecrantia lingua / edidit ).

97 In a shor t note Freud (1922) posits that Medusa’s decapitation visualizes the castration complex. When the little boy faces female genitals for the first time (with their snaky hair, experienced subconsciously as both multiple, threatening, and castrated penises), he is shocked by her lack of a penis, and realizes his ow n cannot be taken for granted. Medusa’s head, similarly, has a devastating effect on the obser ve r – the subject is str uck dumb, paralysed, petrified, but also (permanently)

aroused. Note that Philomela’s tongue is compared to a snake’s tail, making a Freudian connection between head and lower-body, mouth and genitals.

98 sed neque fulvus aper media tam saevus in ira est

/ femina quam socii deprensa paelice lecti: / ardet et

in vultu pignora mentis habet / ‘but neither is the re d boar so savage at the height of his fur y is a woman when a rival is taken in the bed she shares: she burns and her face betrays her ever y emotion’ Ar s 2 .373 , 3778.

a s

Introduction: Na rcissus and Medusa 33

slipper y opponent, tearing off his bull-horns in another replica of Scylla’s castration of her father, h e i s finally killed by the Hydra’s poison delivered

in Deianira’s cloak, and takes on a snake-like form when (like Aeson) he casts off his mor tal frame, achieving immor tality ‘just like a serpent, its old age sloughed off with its skin’ (utque novus serpens posita cum pelle senecta

9 .266 ). In Books 10 and 11 similarly, as I discuss in chapter 3 , Orpheus’

confrontation with Eu r ydice as he returns from the under world has strong Medusan connotations, and the last we see of the poet in the upper world

is his bloody head resting on a sandy shoreline, just like the ora Medusae at

Me t .4 . 741 3 . A s the epic closes in on the historical past in the final books,

it culminates in the self-realization of Actaeon’s, or Na rcissus’ unpunished

relative, Cipus ( cum vidit Cipus in unda / cor nua 15 . 5656; cf. ut vero vultus et cor nua vidit in unda 3 .200 99 ) and in Aesculapius’ command of the snake’s

healing, apotropaic powers in his foundational voyage, in snake-form, to the Medusan caput re r um that is Rome. In Aesculapius, the giant, chthonic

Python, or the Furies’ hair of the early books of the Metamorphoses , becomes

a golden manifestation of sun-god Apollo’s child, and is no longer to be feared, even when looked upon directly. The god instr ucts:

‘pone metus; veniam simulacraque nostra relinquam. hunc modo serpentem, baculum qui nexibus ambit, perspice et usque nota, visum ut cognoscere possis.’

‘don’t b e scared! I will come and leave m y shrine. Just look upon this snake, which twines around my staff, and fix it with your gaze, so yo u can recognize the sight.’

Me t .15 .658 60

The oculi micantes of the monstrous boar of Calydon, the vision of one’s bestial self in reflective water, the devastating backwards glance of Diana and Medusa ( ora retro / flexit 3 .187 8 ; retro versus protulit ora 4 .656) – all this is now without painful consequence, and translates into a beautiful

vision ( oculos circumtulit igne micantes 15 . 674 ; caer uleas despectat aquas 699 ; oraque retro / flectit 15. 6856 ). As the snake enters Rome, unfolding its scaly, epic, papyr us-like coils ( perque sinus crebros et magna volumina labens 721),

it becomes an obvious closural symbol for this end-book, 100 simultaneously

‘putting an end to’ the pestilence that has plagued the city and the epic

as a whole (finem

/ luctibus inposuit venitque salutifer Urbi 7434).

99 I retain this line, which is deleted by Heinsius and Tarrant (on the grounds that it is repeated from

1 .640 1), but kept by Anderson ( 1982 ). The repetition from 1 .6401 (the scene in which Io is also terrified by her ow n reflection), is significant rather than gratuitous. 100 Se e Barchiesi (1997) 190 3 .


v i c t o r i a r i m e l l

The r un-down of Caesar’s imperial achievements in the lines that follow affirm the extent to which Rome’s p ower and identity are contingent on conquering the other, o n attempting to embody and project snake-like, monumentalizing Medusan dread, whilst limiting the tricky, boomeranging potential of the mirror in which that caput , in order to be represented at all, is necessarily reflected. Ye t i n doing so, the Metamorphoses must enfold the destr uction of Troy and Virgil’s Aeneid into its final volumina , a s there i s no esche wing the binding of powe r with trauma, rebir th with death, in Medusan poetics.

As she foresees Caesar’s murder, Venus recalls, fear fully, the antiqua damna of her race (774 6 ): the ‘falling walls of ill-defended Troy ’ over whelm her

( 770 ). Ovid’s shiny, existential epic casts back and reshapes the identity-

formations of the Aeneid, a text also inaugurated and energized by Juno’s Medusan furor , which blows Aeneas towards the dream of Rome on seas churned up by chthonic winds and sexual jealousies. As Knox outlines in his famous ar ticle, the tragic drama of Aeneid 2 hinges on the paradoxical symbolism of the snake/snake-flame, which promises rene wal in the midst of destr uction, a destr uction to be both repeated and overcome/refracted in

the fall of Car thage (and in Dido’s Medean-Medusan promise of re venge), and also in the war in Latium, provoked by twin poison snakes plucked from Allecto’s hair in Book 7. A s the serpents strangle Laocoon 101 and flicker throughout Book two, taking human form in sinuous Sinon and hissing Pyrrhus ( 2.4715 ), Aeneas’ narrative (fixated subconsciously, no doubt, on winning/conquering his Car thaginian ‘other’, Dido) 102 hints at the Trojan capacity to appropriate and counteract Greek snakiness. In Ovid, inci- dentally, Perseus continues this role when he slaughters the snake-like sea monster which ploughs towards the rocks like a swift ship at Met. 4.706, fus- ing the parallel images of twin water-serpents and the Greek fleet heading

for Troy’s shores in Aen .2.205/256 : ad litora tendunt

The turning point is first the death of Androgeos at Ae n .2 .370 85 , who recoils as if he has stepped on a venomous snake when he realizes that the ‘Greek’ allies he finds himself among are really Trojans in disguise. Then, at 2 . 67986 , Iulus’ hair catches fire with a snake-tongued flame that no longer harbours bad omens (it is innoxia , 683, just like the innoxius serpent

which appears from Anchises’ tomb at 5 . 92). As Knox writes:

litora nota petens. 103

101 Note also that after killing Laocoon and his sons, the twin snakes take refuge in Miner va ’s citadel under her shield (which, we remember, is decorated with Medusa’s snakes): Ae n .2 . 2257 .

102 On the simultaneous fantasy of the Dido and Aeneas relationship as incestuous, brother–sister union, see Hardie (for thcoming).

103 Also compare ecce , Met .4 . 706 , with ecce , Ae n.2. 203 .

Introduction: Na rcissus and Medusa


in this flame the pattern of the dominant metaphor is complete

cast its old skin. All previous significances of the serpent are here by implication

summed up and rejected in favour of the ne

significance, but he is joyful ( laetus 687 ) and prays that the omen be confirmed ( 691). The confirmation comes in the form of thunder on the left, a falling star, and finally Creusa’s p rophecy to Aeneas of a n e w kingdom in the West. 104

In Book 8, twin serpents sent by Juno are strangled by Hercules (8 .289), and in the battle of Actium, on Aeneas’ shield, Augustus stands high on his stern, twin flames rising from his blessed brow (8 .6801 ), while queen Cleopatra is pictured on the verge of defeat, hounded by twin snakes of death ( 8. 697 ), before the eye is drawn to Caesar’s triumphal march through the gates of Rome. The Aeneid’s civilizing project is predicated on the (barely disguised) repression, recuperation and refraction of otherness, cul- minating in the arming of Aeneas with the shield in Book 8, which will allow him to deflect Dido’s/Juno’s vengeful glare, 105 and to bounce back the snake-infested trauma of Troy onto ne w enemies. Hi s chosen bride Lavinia appears in the text as a sexless absence, whose demure gaze will never infringe upon the narcissistic duels and ambitions of war. Notice that the Cyclopes at Aeneid 8 .43540 are o rdered to stop forging the snaky aegis of Pallas Athene in order to focus their attentions on Aeneas’ similar shield:

the serpent has

Anchises does not realize its full

par te alia Ma r ti curr umque rotasque volucris instabant, quibus ille viros, quibus excitat urbes; aegidaque horriferam, turbatae Pallidis arma, cer tatim squamis serpentum auroque polibant conexosque anguis ipsamque in pectore divae Gorgona desecto ve r tentem lumina collo.

Else where they worked flat out on Mars’ chariot and its flying wheels, with which he rouses men and cities; and eagerly with golden scales of serpents they polished up the dreaded aegis, armour of wrathful Pallas, the inter wove n snakes, and on the goddess’ breast the Gorgon herself – neck severed, eyes rolling.

Ae n. 8.4338

104 B. M. W. Knox ( 1950) 3578 .

105 Philip Ha rdie points out to me that when Aeneas meets Dido in the under world at Aen.6 .450 ff., he strives to ‘soothe’ the ‘fierce-eyed queen’ (talibus Aeneas ardentem et tor va tuentem / lenibat 4678 ) with the result that she lowers her gaze and is metaphorically turned into stone ( 471 ). Note again the use of the adjective tor vus . Aeneas here plays a Perseus figure who in (unintentionally) ‘conquering’ Medusan Dido, himself takes on and uses aspects of her specular powe r. This scene is also typical of the way in which the Perseus–Medusa confrontation is recast by Ovid as a lovers’ showdown.


v i c t o r i a r i m e l l

Not only is Aeneas’ shield set in parallel with Athena’s apotropaic armour; it also itself contains the image of Catiline trembling at the sight of the Furies ( 8 .6989 ), as well as Miner va defending herself (we presume, with the Gorgon breastplate) at 8.699, a superhuman shield within a shield. Ye t as critics have long recognized, this is also an epic in which the unconscious clamours for expression: the anxiety surrounding Lavinia’s untapped sexuality catches light at 7.71 7, when her long hair bursts into flame and sends black smoke billowing through the palace, per ver ting the parallel omen of Iulus’ fier y head in Book 2 (although this time, cr ucially, the serpentine connotations of flames are repressed), while we know all too well that what Aeneas sees on Pallas’ swordbelt at 12 .945 is the uncanny vision of the Danaids slaughtering/castrating their husbands on the night before their wedding. In the inter textual web that braces the final lines of the Aeneid, the lost Pallas is intimately connected with the ghost of (Pallas) Athene; when Aeneas says Pallas te hoc vulnere, Pallas / immolat / ‘Pallas wounds you, Pallas sacrifices you with this spear’ (12 .9489 ), he not only also hides, metaphorically, behind Athene’s shield as he steps into the shoes of Achilles, who when he kills Hector cries, ‘Pallas Athene will conquer yo u

with my spear’ ( Iliad .22. 2702 ), but absorbs and at once realizes/fr ustrates the re venge impulses of all the Aeneid ’s Medusan women: Juno, Amata, Allecto, even Dido. For the epic began with Juno’s wish that she could kill Aeneas in a reenactment of Pallas Athene’s punishment of Ajax (1. 3941 ), but when Aeneas plunges his sword into Turnus’ chest he, too, is furiis accensus (12 .946 ), next to Juno, accensa at 1.29 , and Dido, furiis incensa at

4 .376 .

t h e l o o k o f i m p e r i a l i s m

In her analysis of the haunting, ever-changing appearance of Medusa throughout Western ar t, Jean Clair calls the clever indirectness of Perseus’ attack on the Gorgon (that is, his use of the mirror-shield) ‘the lesson of culture’ . 106 The detour of the ‘figurative’ enables him/us to make the dread- ful and frightening more concrete, and to detach himself/ourselves from it. Thus the other is caught without ever being seen. 107 Those especially skilled at this act of ‘exo rcizing’ a re called ‘heroes’ or ‘ar tists’, and succeed in transforming Medusa into Na rcissus by making her confront herself in

106 Clair ( 1989 ) 65.

107 Again, we might compare Irigaray’s analysis of patriarchy ( 1985 a) in which she suggests that the feminine is never defined on its ow n terms in Western culture, but always as mirror, reflection or object.

Introduction: Na rcissus and Medusa 37

the mirror-shield. Triumph over Medusa, Clair argues, entails the master y of fear and the founding of a n e w order and regularity. Thus Athena, quasi- masculine and eternal virgin, protector of cities, helper of male heroes, a child without a mother, hides herself behind the face of otherness, con- ve r ting the uncanny Gorgon into an apotropaic symbol, a breastplate or shield. I have suggested that we can trace this Persean, or quasi-Persean con- quest through Vergil’s Aeneid and Ovid’s Metamorphoses , a s well as through his amator y texts. Ovidian poetr y often stages the (male, epic) imperialist project as a mission to behead/castrate and tame Medusa/the ‘other’ (as if, at the same time, to rein in or disguise empire’s own annihilative might), making narcissism the condition of its success and setting Na rcissus against Medusa. Indeed, this basic contest is a cr ucial aspect of and metaphor for the ongoing, energizing conflict between genders and lovers in Ovidian poetr y. Clair also suggests that the most disturbing, ugly representations of Medusa coincide with times of upheaval and revolution in which human knowledge and the belief in ratio are being questioned, whereas she often re ve r ts to, or takes on the guise of a charming, attractive and ve r y much humaniz ed woman in eras of comparative stability and classical restraint. The two faces of Medusa (the apotropaic mask and beautiful sign of the anthropomorphization of nature on one hand, incarnation of chaos, fear, and disorientation of the self on the other) co-exist throughout Western histor y, but opposing elements in her myth are highlighted in a cyclic, fluc- tuating movement. Howeve r, one of the things that emerges from my read- ings of Ovid in this book is the extent to which Ovid allows a simultaneous melting/ contention/interaction of Medusa figures (or, Medusa–Narcissus figures) to explore, effectively, the neurotic tensions and nostalgias of high imperialist culture. As we will see fur ther in chapters 1 and 2, Ovid’s seduc- tive women in the Medicamina, Ars Amatoria and Remedia Amoris seem often to alternate between being, or seeming to be, pretty, manipulable models in line with Pygmalion’s fantasies, and unpredictable, frightening and ungovernable figures who might lash out or regain the upper hand when yo u least expect it. The same edginess is often par t o f the appeal of Ovid’s He roides , too, as we will see in the later chapters (46). As I’ve stressed throughout this introduction, and as Ovidian poetr y shows us, Medusa is a complex, deeply contradictor y figure, whose image and asso- ciation is always difficult to fix (let alone see). We have also explored the extent to which Medusa and Na rcissus are close cousins, two sides of the same coin: both are imperial architects and visionaries, turning landscapes into marble, and make individuation and the rise of technology contingent


v i c t o r i a r i m e l l

on violence (in Medusa’s case her gaze is aggressive, deadly to enemies, including herself ), and on tragedy (in the Na rcissus myth, ar tification is connected either with na¨ıve delusion or with the realization that what one creates is after all super ficial and empty). In Ovidian poetr y, the consistent inter twining of (echoes of ) Narcissus’ and Medusa’s myths ensures that subjects’ relationships with the mirror, and with their ow n reflected, projected selves, are i n different ways always fraught and risky. Ovid’s beautiful, composed, cosmetically enhanced self, like Rome’s stunning architectural landscape, is always both obviously ar ti- ficial and prone to collapse: ‘ The time will come when you’ll hate to look in the mirror’, Ovid tells his audience in the Medicamina . While Augustan Rome, especially in the Medicamina and Ar s , is conceived as a narcissist’s city, a hall of luminous mirrors, shiny metals and glossy marble (‘ours is tr uly an age of gold’, as Ovid puts it at Ar s 2.277 ), we are never allowed t o forget that there is black ear th festering under that marble face-lift (the por- tico of Octavia is super ficially rich in its marble coating, exter no mar more dives opus Ar s 1.70 ; cf. nigra sub imposito mar more terra latet / ‘black ear th lies hidden under marble buildings’ Me d .8). So too, there i s a limit to how far Ovidian style, as it calls attention ‘to the linguistic sur face of texts’ 108 can (or wants to) impose order on a primal, pre-linguistic (pre/extra-imperial) world that is also the domain of the Eastern, the female, the Dionysiac. In other words, although Ovidian poetics strongly identifies with an imperialistic imagination, making narcissistic poet and ar tist figures tri- umphant Perseuses at war with a disturbing and exterior ‘other’, at the same time it is inspired by and often almost synonymous with that same terrifying ‘enemy’, an enemy that is no longer necessarily ‘outside’ empire, but to be found, like a Trojan horse, a discomfor ting mirror-image, within the walls of one’s own city and psyche. 109 While in Ar s 1, for example, Augustus’ popular reenactments of Roman conquests in the East attract a cosmopolitan crowd (ingens orbis in Urbe fuit / ‘the whole world was in our city’ Ar s 1. 174), this symbolic, captive audience also mixes identities and positions, so that many a local is now ‘over throw n’ by a ‘foreign love ’

( quam multos advena torsit amor! 1 .175 ), just as even a barbarian can play the

108 Ha rdie (2002b) 5 .

109 As Habinek reminds us (2002 ) 467:

‘Because Ovid’s position as both subject and object of the imperial gaze i n many ways resembles our own, exploration of his politics invites uncomfor table self-scr utiny on the par t o f the critic – a consideration that may explain why most studies of Ovidian politics limit themselves to examining the degree to which the poet distances himself from the princeps , rather than considering the extent to which his writing is implicated in Roman imperialism.’

Introduction: Na rcissus and Medusa 39

imperialist narcissist if only he is rich enough ( dummodo sit dives, barbar us ipse placet 2. 276 ). Ovid himself, we have seen, is inspired by Medusa’s horse, Pegasus, born from her severed neck ( Fasti 5.8), and the fearsome Gorgon represents the limits and triumph of the poetic imagination (see again, Am. 3. 12 . 21 33). In the exile poetr y, moreover, as I mentioned earlier, the dread and aggression associated with snaky Medusa is evoked to describe both the tribes of Tomis (e.g. in Ex P.4.9. 814) and Augustus’ wife Livia back in Rome, whom Ovid’s coniunx is scared to approach on his behalf ( Ex P.3 . 1.124). The pluralism and narcissistic luxuriance that characterizes (Ovid’s vision of ) Augustan Rome, as well as Ovidian erotic poetr y, never quite smother either the horror of difference, or the threat of (apparent) sameness. In Ovidian mythology, I suggest, it is the figure of Medusa who intr udes into the imperial Na rcissus plot to excavate the black beneath the marble, the weakness beneath the pomposity, forcing us to confront the terrifying otherness of the external world within ourselves. This terror of the primi- tive and of (female) ear th, moreover, is fused throughout Roman imperial literature with desire and nostalgia for the wild woman, the bit of rough, for a r ural landscape cove red i n unpr uned brambles and thick, dark woodland. When yo u control her and get her on your side, then, petrifying Medusa is an icon for stylistic decadence, for the imposition of beauty and form on the flux of experience, for the triumph of monumentalizing, imperialist ambition. But she also reminds us of the complexities of that ambition (rooted in, and nostalgic for a simpler, f reer, more violent r ural past) and embodies its painful, repressed implications for the self. Medusa’s rage and creative p ower, never reigned in or contained enough, spikes not only the Ovidian puella , but also (inevitably) Ovidian poetics. Her entr y into the Na rcissus myth does much to unsettle Ha rdie’s recent analysis of Ovid’s mythic role-models and negotiation of the past:

Ovid’s fascination with the permeability of the boundar y between image and reality reflects not just a sense of his own powers as ar tist, but also registers a reaction to contemporar y political and cultural circumstances. What might be characterized as a regressive and mystifying poetics, expressing itself in tales of animated statues and the like, compensates for the hyper-sophistication of a late

Alexandrian ar tistic culture with a nostalgic attempt to reconnect with a primitive, archaic, shamanist and Orphic model of the poet as magician, whose car mina

Within a specifically

Roman context, Ovid makes his own distinctive contribution to the triumviral and Augustan vates -concept, in which the poet reinvigorates his poetic authority by appeal to numinous models of poetic production from the past, and by which the

(‘songs’/ ‘spells’) can directly control the external worl d

4 0

v i c t o r i a r i m e l l

poet mimics the princeps ’ attempt to revive political authority through a transfusion into the present of the charisma of the great Romans of the past, and also through a claim to privileged access to the divine. Ovid’s reenchantment of his poetic world should also be seen in the context of the visual representation of imperial powe r. 110

Fo r Hardie, then, as I outlined at the beginning of this introduction, the fig- ures of Na rcissus, Orpheus and Pygmalion, the great illusionists and magi- cians who can bring statues and reflections to life, are the key archetypes for the Ovidian ar tist. Yet a s I have stressed, when Ovid delves into mytho- logical figurations and aetiologies of magic and make-believe, he confronts Medusa’s metamorphosis, death and rebir th (and all the chaos, fur y and creativity she represents) alongside the enchantment and grief of Na rcissus (or of his ar tist-relatives Pygmalion and Orpheus). In Medusa, moreover, antiquity is not (just) a sweet and safe escape, not least because in her the past is immanent in the present, regression inherent in progress, cynicism in sorcer y, death in life. This is not to say that Ovid is more o r less in tune with an imperial agenda, howe ve r we might measure it, only to stress that his poetr y encompasses and toys with desires and fears which cannot adequately be contained within the banks of Na rcissus’ pool.

110 Ha rdie (2002a) 192.

c h a p t e r 1

Specular logics: Medicamina

A mutual gaze without speech holds too much emotion, becomes awkward, even hostile. When two animals look in mutual gaze, either the submissive one looks away or the more p ower ful may attack. 1

Me n a re real. Women are made-up. 2

The Medicamina, which advises women on how t o per fect their complex- ions, can be read as a distilled contemplation, a thumbnail sketch, of what I will be discussing throughout this book: Ovid’s p robing of the relationship between self and other in the context of a high imperial culture which, like the dressing table mirror, seems to breed self-awareness. Throughout Ovid’s oeuvre, illusionism and the mapping of subject–object relations are linked closely to the ar ts of mirroring and making-up: the Medicamina makes this point most over tly, hence the logic of beginning here. Mo reover, antiquity’s central (and, as we ’ve seen, related) myths of catoptric thanatosis, Perseus’ killing of Medusa and the tragedy of Na rcissus, become animating subtexts in the Medicamina both for Ovid’s imagining of the self-indulgent puella at her toilette, and for the poet’s/readers’ ex